Springer Series in Materials Science 254

V.A.G. Rivera
Danilo Manzani Editors

Technological
Advances
in Tellurite
Glasses
Properties, Processing, and Applications
Springer Series in Materials Science

Volume 254

Series editors
Robert Hull, Charlottesville, USA
Chennupati Jagadish, Canberra, Australia
Yoshiyuki Kawazoe, Sendai, Japan
Richard M. Osgood, New York, USA
Jürgen Parisi, Oldenburg, Germany
Tae-Yeon Seong, Seoul, Korea (Republic of)
Shin-ichi Uchida, Tokyo, Japan
Zhiming M. Wang, Chengdu, China

More information about this series at http://www.springer.com/series/856

The Springer Series in Materials Science covers the complete spectrum of materials
physics, including fundamental principles, physical properties, materials theory and
design. Recognizing the increasing importance of materials science in future device
technologies, the book titles in this series reflect the state-of-the-art in understand-
ing and controlling the structure and properties of all important classes of materials.
V.A.G. Rivera • Danilo Manzani
Editors

Technological Advances
in Tellurite Glasses
Properties, Processing, and Applications
Editors
V.A.G. Rivera Danilo Manzani
Facultad de Ciencias Fı́sicas Department of Chemistry
Universidad Nacional Mayor State University of Londrina - UEL
de San Marcos Londrina, PR, Brazil
Lima, Peru

ISSN 0933-033X ISSN 2196-2812 (electronic)

Springer Series in Materials Science
ISBN 978-3-319-53036-9 ISBN 978-3-319-53038-3 (eBook)
DOI 10.1007/978-3-319-53038-3

Library of Congress Control Number: 2017935099

© Springer International Publishing AG 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with
regard to jurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Contents

1 Introduction to Tellurite Glasses . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Raouf El-Mallawany
2 Linear and Nonlinear Optical Properties
of Some Tellurium Oxide Glasses . . . . . . . . . . . . . . . . . . . . . . . . . 15
Cid B. de Araújo and Luciana R.P. Kassab
3 Trivalent Lanthanides in Tellurite Glass . . . . . . . . . . . . . . . . . . . . 41
V.A.G. Rivera and L.A.O. Nunes
4 Nonoxide Tellurium-Based Glasses . . . . . . . . . . . . . . . . . . . . . . . . 59
Mathieu Hubert, Catherine Boussard-Plédel,
Bruno Bureau, and Pierre Lucas
5 Photonic Applications of Tellurite Glasses . . . . . . . . . . . . . . . . . . 93
Luiz Carlos Barbosa, Cicero Omegna Filho, and Enver
Fernandez Chillcce
6 Lasers Utilising Tellurite Glass-Based Gain Media . . . . . . . . . . . . 101
Billy D.O. Richards and Animesh Jha
7 Tellurite Glasses for Optical Amplifiers . . . . . . . . . . . . . . . . . . . . 131
Luiz Carlos Barbosa, Cicero Omegna Filho,
and Enver Fernandez Chillcce
8 Broadband Emission in Tellurite Glasses . . . . . . . . . . . . . . . . . . . 155
Kummara Venkata Krishnaiah, Jose Marques-Hueso,
and Raman Kashyap
9 Tellurite Glass Fibers for Mid-infrared Nonlinear
Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
Xian Feng, Peter Horak, and Francesco Poletti

v
vi Contents

10 Tellurite Thin Films Produced by RF Sputtering for Optical

Waveguides and Memory Device Applications . . . . . . . . . . . . . . . . 241
Luciana R.P. Kassab, Davinson M. da Silva, Vanessa D. del Cacho,
Leonardo Bontempo, Sebasti~ao G. dos Santos Filho, and Marco Isaias
Alayo Chavez
11 Laser Writing in Tellurite Glasses . . . . . . . . . . . . . . . . . . . . . . . . . 259
T. Toney Fernandez, S.M. Eaton, G. Jose, R. Osellame,
P. Laporta, and J. Solis
12 Supercontinuum Generation in Tellurite Optical Fibers . . . . . . . . . 277
Frédéric Désévédavy, Gregory Gadret, Jean-Charles Jules,
Bertrand Kibler, and Frédéric Smektala
13 Tellurite Glasses for Plasmonics . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
V.A.G. Rivera, F.A. Ferri, and E. Marega Jr.

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
Contributors

Cid B. de Araújo Departamento de Fı́sica, Universidade Federal de Pernambuco,

Recife, PE, Brazil
Luiz Carlos Barbosa Departamento de Eletrônica Qu^antica – Instituto de Fı́sica
Gleb Wataghin, UNICAMP – Universidade Estadual de Campinas, Campinas, SP,
Brazil
Leonardo Bontempo Faculdade de Tecnologia de S~ao Paulo/CEETEPS, S~ao
Paulo, SP, Brazil
Escola Politécnica da Universidade de S~ao Paulo, S~ao Paulo, SP, Brazil
Catherine Boussard-Plédel Equipe Verres et Céramiques, UMR 6226, Université
de Rennes 1, Rennes, France
Bruno Bureau Equipe Verres et Céramiques, UMR 6226, Université de Rennes 1,
Rennes, France
Vanessa D. del Cacho Faculdade de Tecnologia de S~ao Paulo/CEETEPS, S~ao
Paulo, SP, Brazil
Marco Isaias Alayo Chavez Escola Politécnica da Universidade de S~ao Paulo,
S~ao Paulo, SP, Brazil
Enver Fernandez Chillcce Departamento de Eletrônica Qu^antica, Instituto de
Fı́sica Gleb Wataghin, UNICAMP – Universidade Estadual de Campinas,
Campinas, SP, Brazil
Frédéric Désévédavy Laboratoire Interdisciplinaire Carnot de Bourgogne ICB,
UMR 6303 CNRS-Université de Bourgogne Franche-Comté, Dijon, France
S.M. Eaton Dipartimento di Fisica, Politecnico di Milano, Milano, Italy
Istituto di Fotonica e Nanotecnologie-Consiglio Nazionale delle Ricerche
(IFN-CNR), Milano, Italy

vii
viii Contributors

Raouf El-Mallawany Department of Physics, Faculty of Science, Menofia

University, Shebeen El-Kom, Menofia, Egypt
Xian Feng Institute of Laser Engineering, Beijing University of Technology,
Beijing, China
T. Toney Fernandez Dipartimento di Fisica, Politecnico di Milano, Milano, Italy
F.A. Ferri Departamento de Fı́sica, Universidade Federal de S~ao Carlos,
S~ao Carlos, SP, Brazil
Cicero Omegna Filho Luxtec Optical System, Campinas, SP, Brazil
Gregory Gadret Laboratoire Interdisciplinaire Carnot de Bourgogne ICB, UMR
6303 CNRS-Université de Bourgogne Franche-Comté, Dijon, France
Peter Horak Optoelectronics Research Centre, University of Southampton,
Highfield, Southampton, UK
Mathieu Hubert CelSian Glass & Solar, Eindhoven, The Netherlands
Animesh Jha School of Chemical and Process Engineering, University of Leeds,
Leeds, UK
G. Jose Faculty of Engineering, Institute for Materials Research, School of Chem-
ical and Process Engineering, University of Leeds, Leeds, UK
Jean-Charles Jules Laboratoire Interdisciplinaire Carnot de Bourgogne ICB,
UMR 6303 CNRS-Université de Bourgogne Franche-Comté, Dijon, France
Raman Kashyap Department of Engineering Physics, Polytechnique Montréal,
Montréal, QC, Canada
Department of Electrical Engineering, Polytechnique Montréal, Montréal, QC,
Canada
Luciana R.P. Kassab Faculdade de Tecnologia de S~ao Paulo/CEETEPS, S~ao
Paulo, SP, Brazil
Bertrand Kibler Laboratoire Interdisciplinaire Carnot de Bourgogne ICB, UMR
6303 CNRS-Université de Bourgogne Franche-Comté, Dijon, France
Kummara Venkata Krishnaiah Department of Engineering Physics,
Polytechnique Montréal, Montréal, QC, Canada
P. Laporta Dipartimento di Fisica, Politecnico di Milano, Milano, Italy
Istituto di Fotonica e Nanotecnologie-Consiglio Nazionale delle Ricerche
(IFN-CNR), Milano, Italy
Pierre Lucas Department of Materials Science and Engineering, University of
Arizona, Tucson, AZ, USA
Contributors ix

E. Marega Jr. Instituto de Fı́sica de S~ao Carlos–INOF, USP, S~ao Carlos, SP,
Brazil
Jose Marques-Hueso Institute of Sensors, Signals and Systems, School of Engi-
neering and Physical Sciences, Heriot-Watt University, Edinburgh, UK
L.A.O. Nunes Instituto de Fı́sica de S~ao Carlos—INOF, University of Sao Paulo,
S~ao Carlos, SP, Brazil
R. Osellame Istituto di Fotonica e Nanotecnologie-Consiglio Nazionale delle
Ricerche (IFN-CNR), Milano, Italy
Francesco Poletti Optoelectronics Research Centre, University of Southampton,
Highfield, Southampton, UK
Billy D.O. Richards School of Chemical and Process Engineering, University of
Leeds, Leeds, UK
V.A.G. Rivera Facultad de Ciencias Fı́sicas, Universidad Nacional Mayor de San
Marcos, Lima, Peru
Sebasti~ao G. dos Santos Filho Escola Politécnica da Universidade de S~ao Paulo,
S~ao Paulo, SP, Brazil
Davinson M. da Silva Faculdade de Tecnologia de S~ao Paulo/CEETEPS, S~ao
Paulo, SP, Brazil
Frédéric Smektala Laboratoire Interdisciplinaire Carnot de Bourgogne ICB,
UMR 6303 CNRS-Université de Bourgogne Franche-Comté, Dijon, France
J. Solis Laser Processing Group, Instituto de Optica, CSIC, Madrid, Spain
Chapter 1
Introduction to Tellurite Glasses

Raouf El-Mallawany

1.1 Smart Materials

The interdisciplinary field of materials science, commonly known as materials

science and engineering, involves the discovery and design of new materials,
with an emphasis on solids. Materials scientists emphasize understanding how the
history of a material (its processing) influences its structure, and thus the material’s
properties and performance. The understanding of processing-structure-properties
relationships is called the “materials triangle,” as shown in Fig. 1.1. Materials
scientists seek to tune the structure and synthesize materials with properties that
provide optimum performance in every type of materials application.
Smart materials are materials designed to have one or more properties that can
be significantly changed in a controlled fashion by external stimuli, such as stress,
temperature, moisture, pH, electric or magnetic fields. Sometimes a smart material
can be categorized as a shape memory material (SMM) or a shape memory
technology (SMT). Some examples of smart materials are:
• Piezoelectric materials
• Shape-memory materials
• Magnetic shape memory materials
• Smart inorganic polymers
• Temperature-responsive polymers
• Halochromic materials
• Chromogenic materials
• Photomechanical materials
• Self-heating materials

R. El-Mallawany
Department of Physics, Faculty of Science, Menofia University, Shebeen El-Kom,
Menofia, Egypt
e-mail: [email protected]

© Springer International Publishing AG 2017 1

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_1
2 R. El-Mallawany

Fig. 1.1 The structure– Performance

properties–performance
triangle

Structure Properties

• Magnetocaloric materials
• Thermoelectric materials
• Self-cleaning glasses
There are a wide range of different smart materials and each offers different
properties that can be changed. Some materials are very good indeed and cover a
huge range of scales. Previously, especially since 2001–2011, contributions have
opened new horizons in physical aspects of semiconducting tellurite glasses
research [1–3]. The success in gathering and demonstrating the physical properties
of tellurite glasses has opened a new area of glass science [1–3].
Prior to 1984 [4], tellurite glasses were considered as intermediate glass forms
and attracted no international attention, but now, after 30 years they are receiving
focus due to their unique physical properties when modified with transition metals
or rare earth ions to get unique properties (2016) [5]. Characteristic temperatures
and microhardness of (ZnO)x–(AlF3)y–(TeO2)z glass systems have been measured
[5]. Pure TeO2 glass, which was prepared in 1984 [4], contained 1.6 mole % Al2O3
which came from the walls of the aluminum crucible. Tellurite glasses have many
potential applications in areas of optoelectronics such as laser technology and fiber
optics. The interference and independence of the simultaneously occurring physical
processes are the key problems in the field of material science. Since solving the
theoretical difficulties with amorphous solids, tellurite glasses have received much
international research attention, as shown in Table 1.1.

1.2 Some Unique Physical Properties of Tellurite Glasses

Since the early 1990s, optical properties of tellurite glasses have been explored by
calculating the number of ions per unit volume and the polarizability of ions
[6]. Sellmeier coefficients and chromatic dispersions for some tellurite glasses [7]
have been calculated. In 2004, preparation and measurements of the luminescence
properties of Er3þ ions in new tellurite glasses were achieved [8]. A broad emission
spectrum of width 121 nm, peaking at a wavelength of 1550 nm, has been obtained.
The maximum observed decay time of 4I13/2 to 4I15/2 transition at wavelength
1550 nm is about 4.5 ms. Therefore, tellurite–titanium–tungsten glasses were
proposed to be good candidates for a broadband erbium-doped fiber amplifier
(EDFA). The values of full-width half-maximum (FWHM) ranging from 93 to
1 Introduction to Tellurite Glasses 3

Table 1.1 2857 articles for Number of Number of

tellurite glasses from 1977– Year articles Year articles
Jan.2017 by using research
2016 þ 2017 175 1996 27
engine “Scopus”
2015 178 1995 35
2014 206 1994 35
2013 210 1993 17
2012 208 1992 17
2011 205 1991 10
2010 167 1990 20
2009 157 1989 17
2008 161 1988 8
2007 154 1987 5
2006 136 1986 12
2005 133 1985 9
2004 114 1984 4
2003 79 1983 6
2002 67 1982 4
2001 57 1981 4
2000 43 1980 2
1999 42 1979 5
1998 26 1978 7
1997 36 1977 7

121 nm for 0.01 to 2.5 mol % Er2O3 were observed. A significant increase in the
overall intensity of upconversion fluorescence was observed with an increasing
concentration of Er3þ ions. The relative increase in intensity of red emission with
respect to the intensity of green emission was observed with an increasing concen-
tration of erbium. It has been concluded that [8] that Er3þ-doped TeO2-WO3-TiO2
glasses would be promising photonic materials for infrared amplifiers, as well as for
the green and red upconversion emissions.
In 2008, researchers developed new tellurite glasses, 80TeO2–5TiO2–(15-x)
WO3–xAnOm, where AnOm was Nb2O5, Nd2O3 and Er2O3, 0 < x < 7 mol% and
measured their optical properties in the visible and UV regions [9]. In the range of
the visible spectrum the linear refractive indices (nλ) for the studied three glass
series were found to be among the highest values of any known glasses. The
refractive index has been increased to 2.22 of the modifier 7 mol% Nd2O3. High
refractive indices of these glass samples were attributed to the high polarization of
the host material TeO2. Moreover, the incorporation of the transition metal cation
Nb5þ and rare earth cations Nd3þ and Er3þ at the expense of W4þ cation led to
further higher polarization of the glass and consequently an increase in the refrac-
tive index. As a measure of dispersion, the Abbe numbers were calculated for these
glass series and found to decrease to 16.02 for 7 mol% Nd2O3. In the ultraviolet
range, position of the fundamental absorption edge shifts to higher wavelength with
an increase of Nb2O5, Nd2O3, but shifts to a lower wavelength with an increase of
4 R. El-Mallawany

Table 1.2 Article type of Article type Number of articles

tellurite glasses
Article 20542106
Conference paper 674673
Conference review 33
Article in press 2116
Review 1718
Letter 5
Erratum 4
Book chapter 1
Note 1
Total 28102857

Er2O3 content. The shifts of the absorption edge are most likely related to structural
rearrangements of the glass network and modifier. The optical energy gap (Eopt)
was found to increase to 3.59 eV for 7 mol% Nd2O3. Also, applications of the new
tellurite glasses have been studied by measuring absorption/emission analysis of
RE3þ in lithium–boro–tellurite glasses [10]. Luminescence spectra, optical proper-
ties of TeO2–WO3–Li2O glasses doped with Nd, Sm and Er rare earths and the
effects of pre-readout annealing treatments on TL mechanism at therapeutic radi-
ation doses have been measured [11, 12].
Some of the correlations were described for the first time in [1] and combined in
novel ways to suggest new research directions. In 2005 the “International Materials
Institute” IMI-USA for New Functionality in Glass recorded 5 Video seminars [13]
for international educational purposes entitled “An Introduction to Tellurite
Glasses.” Now, tellurite glasses are receiving international attention for their
potential applications in areas of optoelectronics such as laser technology and
fiber optics. Table 1.1 shows the number of published articles per year for the
period 1977–2016 by using the “Scopus” research engine. Tables 1.2 and 1.3 show
the types and the subject areas, respectively, of the articles published through
October 2016. It is very easy to see the huge increase in international attention
for tellurite glasses in 2016 compared to the work published in 1977, as shown in
Table 1.1.

1.3 Recent Processing, Properties and Applications

of Tellurite Glasses

In 2014, researchers successfully achieved Er3þ–Yb3þ co-doped TeO2–PbF2

oxyhalide tellurite glasses for amorphous silicon solar cells [14]. When this glass
was applied at the back of amorphous silicon solar cells in combination with a rear
reflector, maximum external quantum efficiency and luminescence quantum effi-
ciency of 0.27 and 1.35%, respectively, were achieved at 300 mW excitation. In
addition, the effect of Pb–Te–O glasses on Ag thick film contact in the crystalline
1 Introduction to Tellurite Glasses 5

Table 1.3 Article subject area of tellurite glasses

Subject area Number of articles
Physics and astronomy 17191777
Materials science 15901638
Engineering 853885
Chemistry 507515
Computer science 262275
Mathematics 126136
Biochemistry, genetics and molecular biology 9291
Chemical engineering 5758
Energy 1920
Multidisciplinary 18
Medicine 1613
Earth and planetary sciences 1117
Social Sciences 9
Business, management and accounting 4
Environmental science 4
Undefined 2
Agricultural and biological sciences 1
Arts and humanities 1
Immunology and microbiology 1
Pharmacology, toxicology and pharmaceutics 12
Total 28102857

silicon solar cells has been studied [15]. The observed microstructures, I–V elec-
trical performance and contact resistance maps led to a better understanding of the
relationship between glass properties and contact resistance [15].
Dy3þ-doped lithium tellurofluoro borate glasses were prepared with the chem-
ical composition (40–x) B2O3–15TeO2–15Li2O–15LiF–15NaF–xDy2O3, where
x ¼ 0.05, 0.1,0.25,0.5,1.0 and 2.0 wt%) by the conventional melt quenching
technique and characterized through FTIR, optical absorption, luminescence and
life time measurements [16]. The yellow to blue luminescence intensity ratios and
color chromaticity coordinates of the title glasses have also been estimated to
evaluate the white light emission as a function of Dy3þ ion concentration
[16]. Also, the energy transfer mechanism has been investigated for Er3þ-doped
germinate–tellurite glass for mid-infrared 2.7 μm fiber laser material; the spectro-
scopic characteristics as well as its good thermal properties indicate that this kind of
glass is an attractive host for developing mid-infrared fiber lasers [17]. In addition,
luminescence spectral analyses of Tm3þ ion-doped lead tungsten tellurite glasses
for visible red and NIR applications have been studied, concluding that 0.5 mol %
of Tm3þ ion--doped glass is most suitable for generating bright visible Red and NIR
lasers to operate at 650 and 800 nm, respectively [18]. Ligand field and Judd-Ofelt
(JO) intensity parameters of samarium-doped tellurite glass have been studied
[19]. The large value of its spectroscopic quality factor (greater than unity) makes
6 R. El-Mallawany

the proposed glass system interesting for various optical devices fabrication
[19]. The effect of silver nanoparticles on the 1.53 μm fluorescence in Er3þ/Yb3þ
co-doped tellurite glasses has been studied [20]. The enhanced 1.53 mm band
fluorescence indicates that the Er3/Yb3 co-doped tellurite glass with an appropriate
amount of Ag nanoparticles is a promising candidate for the development of Er3-
doped fiber amplifiers (EDFAs) applied in the WDM systems [20]. Also, the effects
of Gd3þ: Ag co-doping on structural and magnetic properties of lead tellurite glass
ceramics has been studied, and it was found that the values obtained for μeff are
lower than the magnetic moment of Gd3þ ions in the free ion state [21]. A
maximum laser out power of 34 mW at 2040 nm was achieved with a 9 cm long Ho
3þ
-doped tellurite glass fiber, which was first reported on the 2-μm laser output from
a Ho3þ-doped tellurite oxide glass fiber [22]. A positive value for the gain cross-
sections has been found for a population inversion factor c of 0.4 in the spectral
range from 1060 to 1110 nm, which suggests the potential use of Nd3þ-doped
TeO2–PbF2–AlF3 glasses as a laser host and applications [23]. Also, the effects of
Nd2O3 concentration in the laser emission of TeO2–ZnO glasses established the
Nd2O3 concentration range for laser action operation in TeO2–ZnO glasses, guiding
the correct method for the preparation of these glasses for solid state laser applica-
tions [24]. A blue to NIR quantum cutting glass can be fabricated as a luminescence
layer for enhancing the efficiency of silicon solar cells by studying photon down
shifting in strong NIR-emitting Er3þ–Yb3þ embedded tungsten tellurite glass
[25]. Also, concentration effects on the optical spectroscopic properties of Er3þ-
doped TeO2-Nb2O5-ZnO tellurite glass system can be used as an efficient host for
laser operation and broadband optical amplifiers [26].
The Er3þ/Tm3þ co-doped tellurite glass containing silver nanoparticles is a
promising host material applied for the 1.85 mm band solid state lasers and
amplifiers at 1.85 μm [27]. There is a considerable improvement in the insulating
performance of (60–x)TeO2–10K2O–10P2O5–10B2O3–10ZnF2–xEr2O3 (where
x ¼ 0.0, 0.05, 0.1, 0.3, 0.5 and 1.0 mol%) when the Er2O3 concentration is about
0.5 mol% in the glassy system; glasses prepared in this way may be effective
materials for device fabrication to be used in various applications [28]. Moreover, a
novel tellurite glass based hybrid-core photonic crystal fiber (HCPCF) and can find
significant applications in optical communication, signal processing, and sensing
systems [29]. The laser stimulated effects in 70TeO2–10ZnO–10WO3–5TiO2–
5Na2O (mol%) glasses doped with Bi2O3 and Pr2O3, respectively, has been studied
[30]. The photo induced processes were performed using two coherent beams of
532 nm doubled frequency Nd:YAG pulsed laser at angles varying within 28 –26 .
The low-power 532 nm beam has served as a probe for detection of photo induced
changes. A crucial dependence on the Pr3þ and Bi dopants was observed that allows
using these compounds for laser-operated optical devices [30]. Previously, rare-
earth ion-doped TeO2 and GeO2 glasses had been classified as laser materials
[31]. These two types of glasses exhibit unique combinations of optical and
spectroscopic properties, together with their attractive environmental resistance
and mechanical properties. Engineering such a combination of structural, optical
and spectroscopic properties is feasible only as a result of structural variability in
1 Introduction to Tellurite Glasses 7

these two types of glasses, since more than one structural unit (TeO4 bi-pyramid,
TeO3trigonal pyramid, and TeO3þd polyhedra) in tellurite and (GeO4 tetrahedron,
GeO3 octahedron) in GeO2 based glasses may exist, depending on composition.
Moreover, tellurite-based fibers have been achieved and found to have applications
to optical communication networks due to the gain characteristics of erbium-doped
tellurite fiber amplifiers (EDTFAs) [32]. The efficiency of a commercial solar cell
of silicon covered with a tellurite based glass was increased by 7% when the glass
was doped with 1% of Tb3þ as compared to an un-doped glass as the cover top
[33]. Quaternary glass system xSiO2–(80–x) Bi2O3–15B2O3–5TeO2 samples may
be considered as potential candidates for nonlinear optical applications [34]. A
solid-core tellurite glass fiber with 1.8 dB/m loss at 1.55 μm has been made by using
the built-in casting preform fabrication method and rod-in-tube fiber drawing
technique [35]. Also, pumping a 10 cm fiber piece with picosecond pulses of
3–5  1012 W/cm2 and 0.1% of the fundamental power limited by the coherence
length of 0.3–5 μm, has exhibited conversion into visible third-harmonic power
tunable over a broad near-IR wavelength ranging from 1500 to 1680 nm [35].
This book will focus on:
• Introduction to tellurite glasses
• Linear and nonlinear optical properties of some tellurium oxide glasses
• Trivalent lanthanides in tellurite glass
• Non-oxide tellurium-based glasses
• Photonic applications of tellurite glasses
• Lasers utilizing tellurite glass based gain media
• Tellurite glasses for optical amplifiers
• Broadband emission in tellurite glasses
• Tellurite glass fibers for mid-infrared nonlinear applications
• Tellurite thin films produced by RF sputtering for optical waveguides and
memory device applications
• Laser writing in tellurite glasses
• Supercontinuum generation in tellurite optical fibers
• Tellurite glasses for plasmonics
Chapter 2, “Linear and Nonlinear Optical Properties of Some Tellurium Oxide
Glasses” focuses on the methods used for fabrication and characterization of the
TeO2 based glasses. Stokes and anti-Stokes photoluminescence are discussed for
frequency upconversion assisted by phonon annihilation, enhanced frequency
upconversion, energy transfer in the presence of metallic nanoparticles and rare-
earth photoluminescence in the presence of silicon nanocrystals. Third-order
nonlinear optical properties are discussed. Also, Chap. 2 explains the interest in
these glasses due to their large transmittance from the visible to the near-infrared,
small phonon energies, large resistance to atmospheric moisture, large acceptance
of rare-earth (RE) ion- doping and the high nonlinear (NL) optical response. These
characteristics allow the successful use of tellurium oxide glasses (TOG) for color
displays, sensing, optical limiting, and all-optical switching, among other applica-
tions. In this chapter, the authors review some of their previous works performed
8 R. El-Mallawany

with samples based on different (TOG) compositions. Photoluminescence

(PL) properties of (RE) ion-doped (TOG) and the influence of embedded metallic
and semiconductor nanoparticles on the PL efficiency are discussed. A short review
of the third-order nonlinearity of one (TOG) composition in the visible and near-
infrared is also presented in this chapter. The (NL) refraction indices and
(NL) absorption coefficients of the glass are evaluated as well as their figure-of-
merit for all-optical switching and optical limiting. The possible mechanisms
contributing for the optical response of the samples are discussed.
Chapter 3, “Trivalent Lanthanides in Tellurite Glass” analyzes one of the most
discussed fields of spectroscopy of REIs-doped tellurite glasses, some of its appli-
cations in frontline research, emphasized in telecommunications and other fields
with technologic applications, by giving a brief account on the status of this subject.
Also, this chapter focuses on:
• Rare-earth ions in a static crystal fields
• Classification of crystal fields stated and selection rules
• Intensities of optical transitions
• Optical properties of the REIs in tellurite glasses
Further efforts and systematic studies must be realized in order to offer new
developments to extend the applications field of REI-doped tellurite glasses and for
the optimization and nanophotonic applications.
Chapter 4, “Nonoxide Tellurium-Based Glasses” concentrates on the chalcogen
element tellurium, and constitutes a particular class of materials used in numerous
technological applications by:
• Structure and properties of tellurite glasses
• Synthesis of tellurite glasses
• Applications of tellurite glasses, fibers for mid- and far-Infrared spectroscopies,
opto-electrophoretic sensing and thermoelectricity.
Also, in this chapter, the fundamentals of the tellurite glasses and of their
structure are presented. An overview of the different families of tellurite glasses,
as well as their specific properties, is given in this chapter. The considerations
related to their relatively complex fabrication processes are also described.
Chapter 5, “Photonic Applications of Tellurite Glasses” summarizes:
• Photonic applications in tellurite glasses,
• Photonics field applications: Nano wires, Planar and cannel Waveguide,
• Fiber on glass waveguide and Localized temperature sensing.
Chapter 6, “Lasers Utilizing Tellurite Glass Based Gain Media” reviews the
laser sources based on tellurium oxide (TeO2) glass hosts reported to date, whether
in the form of bulk glass, fiber or microspheres. The majority of laser sources
reported using tellurite glass as host material are based on rare-earth ion (Nd3þ, Er
3þ
, Tm3þ and Ho3þ) dopants; however, there are also reports on supercontinuum
generation and Raman lasing in highly non-linear tellurite glass fibers. All of the
tellurite glass based lasers discussed in this chapter operate in the infrared spectral
1 Introduction to Tellurite Glasses 9

region with laser wavelengths around 1, 1.5, 1.9 and 2.1 μm for Nd3þ, Er3þ, Tm3þ
and Ho3þ doping, respectively, while supercontinuum and Raman laser sources
emit in the ranges 0.8–4.9 and 1.5–2.65 μm, respectively. The maximum optical
output power reported to date from a tellurite glass laser is 1.12 W using cladding
pumped fiber. Lasers operating in continuous wave, Q-switched and mode-locked
regimes have also been demonstrated using rare-earth-doped tellurite glass hosts.
The future prospects for lasers based on tellurite glasses are also discussed.
Chapter 7, “Tellurite Glasses for Optical Amplifiers” explores tellurite glasses
for optical amplifiers plus the fabrication of core and clad tellurite optical fibers, of
the Er3þ-doped tellurite micro-structured optical fiber, tellurite rod, tube and
capillaries, tellurite micro-structured preform, and tellurite micro-structured optical
fiber. Also, optical attenuation of tellurite fibers, amplified spontaneous emission,
gain and laser generation, broadband optical amplification, and rare earth level
lifetimes in tellurite glasses have been achieved.
In Chap. 8, “Broadband Emission in Tellurite Glasses” explains the prospects of
broadband absorbing/emitting lanthanide and transition metal doped glasses and
glass-ceramics, attending to their emission ranges and bandwidth with particular
emphasis on lanthanide-doped tellurite glasses. The most common emitting ions are
presented one by one. Next, it focuses on the spectroscopy and the Judd-Ofelt
parameters and their related radiative properties exclusively for Er3þ-doped
tellurite glasses, as well as a brief description of the other lanthanide ions. The
different parameters that affect the broadband absorption/emission in tellurite
glasses are then reviewed. In addition, Chap. 8 summarizes the dopants that can
exhibit broadband emission (not only in the near infra-red region but also in the
visible region) and their bandwidth along with their energy level structures. Finally,
the different applications of tellurite glasses are explored, including lanthanide and
transition metal ions fiber amplifiers, Raman amplifiers, superbroad band emission
for telecommunications, supercontinuum generation, solar devices and possible
applications in laser induced cooling.
In Chap. 9, “Tellurite Glass Fibers for Mid-infrared Nonlinear Applications”
presents the recent progress of using tellurite glass nonlinear optical fibers for mid-
infrarednonlinear applications by:
• Various fabrication approaches producing conventional solid core/cladding
tellurite glass preforms and structured tellurite glass preforms
• Two technical difficulties found during the early stage of using small-core
tellurite glass fiber for generating nonlinear supercontinuum into the
mid-infrared region
“Tellurite Thin Films Produced by RF Sputtering for Optical Waveguides and
Memory Device Applications” is the focus of Chap. 10. Results of pedestal
waveguides fabricated using optical lithography followed by plasma etching, for
integrated photonics and integrated optical sensors applications are reported. Opti-
cal amplifiers have been obtained based on Yb3þ/Er3þ co-doped Bi2O3-WO3-TeO2
thin films. Pedestal waveguides, using Bi2O3-WO3-TeO2 thin films, as core layer,
have been employed for optical devices applications, as Mach-Zehnder
10 R. El-Mallawany

Interferometers; low propagation losses were obtained at 633 and 1050 nm for
waveguides in the 20–100 μm width range. Also, a bistable current mechanism in
Au nanoparticles embedded in TeO2-ZnO thin films has been demonstrated. The
nucleation of the metallic nanoparticles was performed with an adequate heat
treatment and the influence of the film thickness and of the gold nanoparticles
size on the bistable properties is also reviewed, showing the feasibility of using
TeO2-ZnO composition for memory device applications.
Chapter 11, “Laser Writing in Tellurite Glasses” reports the next main points:
• Femtosecond laser waveguide writing
• Nonlinear absorption
• Relaxation and material modification
• Repetition rate
• Linear and nonlinear propagation
• Laser writing in tellurite glasses, Near Infrared and Mid-Infrared and
applications
Chapter 12,“ Supercontinuum Generation in Tellurite Optical Fibers” summa-
rizes the bibliographic data from the literature. Then, it briefly considers on tellurite
glass compositions for supercontinuum generation in optical fibers. Thirdly, the
different fibers structures (microstructured, suspended core or step index fibers,
tapered fibers) are presented. Fourthly, the chapter discusses the pumping sources in
relation with the fibers chromatic dispersion as well as coupling issues, the
supercontinuum performances are presented. The fifth and sixth parts deal with
modelization and nonlinear propagation phenomena in relation with
supercontinuum generation.
Chapter 13, “Tellurite Glasses for Plasmonics” is the last chapter. In this chapter,
we devote attention to tellurite glasses doped with rare-earth ions (REIs) and their
optical properties that has led to their wide use in optical devices. For instance,
many authors have suggested increasing the propagation length of the surface
plasmon polariton (SPP) by producing a population inversion in the dielectric
medium adjacent to the metallic nanostructure or via energy transfer between
REIs and the metallic nanostructure. It is worth mentioning that the main purpose
of this chapter is to show some of the concrete concepts of plasmonic nanostructure
interaction with REIs exhibiting a field enhancement in a transparent medium with
such nanostructures via a localized surface plasmon resonance, such as:
• Localized surface plasmon resonance and plasmonic nanostructures
• Tellurite glasses doped with REIS
• Plasmonic nanostructure coupled with REIS
• Potential applications
Moreover, the physical properties of tellurite glasses still have a high research
potential. In 2017, physical, structural, thermal, and optical spectroscopy studies of
the homogeneous tellurite glasses TeO2–B2O3–MoO3–ZnO–R2O (R ¼ Li, Na, and
K)/MO (M ¼ Mg, Ca, and Pb) have been studied [36]. The conclusion in reference
[36] is that prepared tellurite glasses with high thermal and glass stabilities could be
1 Introduction to Tellurite Glasses 11

suggested as potential materials to realize highly efficient fiber Raman amplifiers

and promising low phonon hosts for rare earth elements doping that can become
potential candidates for the development of solid-state lasers, optical fibers, and
optical displays. Also, in 2017, variation of energy absorption and exposure buildup
factors with incident photon energy and penetration depth for boro-tellurite (B2O3–
TeO2) glasses has been achieved [37]. Boro–tellurite glass with 5 mol% of B2O3,
the energy absorption buildup factors (EABF) and exposure buildup factors (EBF)
values are found to be lowest in low-to-intermediate energy smaller than 3 MeV,
thus exhibiting superior gamma-ray shielding properties [37]. The obtained results
[37] have scientific value for developing excellent shielding properties of materials
and synthesizing new materials for gamma ray shielding applications. Moreover, in
2017 silicate modified bismuth–borate–tellurite glasses may be considered as
potential candidates for nonlinear optical applications [38].

References

1. R. El-Mallawany, Tellurite Glasses Hand Book, 1st edn. (CRC Press, Bota Racon, FL, 2002.)
www.crcpress.co. ISBN:0849303680, 540 Page
2. H. Soorage, Physics and Chemistry of Rare-Earth Ions Doped Glasses (TTP Publishers, Stafa-
Zurich, Switzerland, 2008)
3. R. El-Mallawany, Tellurite Glasses Hand Book, 2nd edn. (CRC Press, Bota Racon, FL, 2011.)
www.crcpress.co. ISBN 9781439849835, 532 Pages
4. E.F. Lambson, G.A. Saunders, B. Bridge, R.A. El-Mallawany, The elastic behavior of TeO2
glass under uniaxial and hydrostatic pressure. J. Non-Cryst. Solids 69, 117–133 (1984)
5. H.A.A. Sidek, R. EL-Mallawany, S. Rosmawati, A.K. Yahya, Characteristic temperatures and
microhardness of (ZnO)x–(AlF3)y–(TeO2)z tellurite glass systems. Chalcogenide Lett. 13,
169–176 (2016)
6. R. El-Mallawany, The optical properties of tellurite glasses. J. Appl. Phys. 72, 1774–1777
(1992)
7. G. Ghosh, Sellmeier coefficients and chromatic dispersions for some Tellurite glasses. J.Am.
Ceramic. Soc. 78, 2828–2830 (1995)
8. R. El-Mallawany, A. Patra, C.S. Friend, R. Kapoor, P.N. Prasad, Study of luminescence
properties of Er 3þ -ions in new tellurite glasses. Opt. Mater. 26, 267–270 (2004)
9. R. El-Mallawany, M.D. Abdalla, I.A. Ahmed, New tellurite glass: optical properties. Mater.
Chem. Phys. 109, 291–296 (2008)
10. N.S. Hussain, G. Hungerford, R. El-Mallawany, M.J.M. Gomes, M.A. Lopes, N. Ali,
J.D. Santos, S. Buddhudu, Absorption and emission analysis of RE3þ(Sm3þ and Dy3þ):
Lithium Boro Tellurite glasses. J. Nanosci. Nanotechnol. 8, 1–6 (2008)
11. I.Z. Hager, R. El-Mallawany, A. Bulou, Luminescence spectra and optical properties of TeO2-
WO3-Li2O glasses doped with Nd, Sm and Er rare earth ions. Phys. B Condens. Matter 406,
972–980 (2011)
12. R. El-Mallawany, H.M. Diab, Effect of pre-readout annealing treatments on TL mechanism in
Tellurite glasses at therapeutic radiation doses level. Measurement 46, 1722–1725 (2013)
13. “An Introduction to Tellurite Glasses”, Five lecture series presented by Prof. Raouf
El-Mallawany, Minufiya University, Egypt for the International Materials Institute for New
Functionality in Glass at Lehigh University, Bethlehem, PA, Dec. 6, 2005. Module 1 - Tellurite
Glass Introduction, to run rm video:http://rm1.cc.lehigh.edu:8080/dept/IMI/Module_
1Tellurite_ass_Intro/index.html Module 2 - Tellurite Glass History, to run rm video: http://
12 R. El-Mallawany

rm1.cc.lehigh.edu:8080/dept/IMI/Module_2-Tellurite_Glass_Intro-History/index.html Mod-
ule 3 - Elastic & Thermal Properties, to run rm video: http://rm1.cc.lehigh.edu:8080/dept/
IMI/Module_3-Elastic_and_Thermal_Properties/index.html Module 4 - Electrical and Dielec-
tric Properties, to run rm video: http://rm1.cc.lehigh.edu:8080/dept/IMI/Module Module 5-
Optical Properties, to run rm video: http://rm1.cc.lehigh.edu:8080/dept/IMI/Module_5-Opti
cal_Properties/index.html.
14. F. Yang, C. Liu, D. Wei, Y. Chen, J. Lu, S.-e. Yang, Er3þ–Yb3þ co doped TeO2-PbF2
oxyhalide tellurite glasses for silicon solar cells. Opt. Mater. 36, 1040–1043 (2014)
15. J. Qin, W. Zhang, S. Bai, Z. Liu, Effect of Pb-Te-O glasses on Ag thick-film contact in
crystalline silicon solar cells. Sol. Energy Mater. Sol. Cells 144, 256–263 (2016)
16. V. Uma, K. Maheshvaran, K. Marimuthu, G. Muralidharan, Structural and optical investiga-
tions on Dy3þ doped lithium tellurofluoro borate glasses for white light applications. J. Lumin.
176, 15–24 (2016)
17. Y. Lu, M. Cai, R. Cao, S. Qian, S. Xu, J. Zhang, Er3þ doped germanate–telluriteglassformid-
infrared2.7 μm fiber laser material. J. Quant. Spectrosc. Radiat. Transf. 171, 73–81 (2016)
18. M. Venkateswarlu, S.K. Mahamuda, K. Swapna, A. Srinivasa Rao, A. Mohan Babu,
S. Shakya, D. Haranath, G. Vijaya Prakash, Luminescence spectral studies of Tm 3þ ions
doped Lead Tungsten Tellurite glasses for visible Red and NIR applications. J. Lumin. 175,
225–231 (2016)
19. Y. Tanko, S. Ghoshal, M.R. Sahar, Ligand field and Judd-Ofelt intensity parameters of
samarium doped tellurite glass. J. Mol. Struct. 1117, 64–68 (2016)
20. L. Wu, Y. Zhou, Z. Zhou, P. Cheng, B. Huang, F. Yang, J. Li, Effect of silver nanoparticles on
the 1.53 μm fluorescence in Er3þ /Yb3þ codoped tellurite glasses. Opt. Mater. 57, 185–192
(2016)
21. M. Boscaa, L. Pop, L. Bolundut, N. Tothazan, G. Borodi, I. Vida-Simiti, R. Stefan, A. Popa,
E. Culea, P. Pascuta, Effects of Gd3þ: Ag co-doping on structural and magnetic properties of
lead tellurite glass ceramics. Ceram. Int. 42, 1169–1117 (2016)
22. D. Li, W. Xu, P. Kuan, W. Li, Z. Lin, X. Wang, L. Zhang, C. Yu, K. Li, H. Lili,
Spectroscopicand laser properties of Ho3þ doped lanthanum-tungsten-tellurite glass and
fiber. Ceram. Int. 42, 10493–10497 (2016)
23. E.A. Lalla, U.R. Rodrı́guez-Mendoza, A.D. Lozano-Gorrı́n, A. Sanz-Arranz, F. Rull, V. Lavı́n,
Nd3þ-doped TeO2–PbF2–AlF3 glasses for laser applications. Opt. Mater. 51, 35–41 (2016)
24. L.M. Moreira, V. Anjos, M.J.V. Bell, C.A.R. Ramos, L.R.P. Kassab, D.J.L. Doualan, P. Camy,
R. Moncorge, Nd3þ-doped TeO2–PbF2–AlF3. Opt. Mater. 58, 84–88 (2016)
25. P. Anurag, R.E. Kroon, V. Kumar, H.C. Swart, Photon down shifting in strong NIR emitting Er
3þ
-Yb3þ embedded tungsten tellurite glass. J. Alloys Compd. 657, 32–36 (2016)
26. A. Maaoui, M. Haouari, Z. Zaaboub, I. Fraj, F. Saidi, H. Ben Ouada, Concentration effects on
the optical spectroscopic properties of Er3þ doped TeO2-Nb2O5-ZnO vitreous system.
J. Alloys Compd. 663, 395–406 (2016)
27. B. Huang, Y. Zhou, C. Pan, Z. Zhou, J. Li, G. Yang, The 1.85 μm spectroscopic properties of
Er3þ /Tm3þ co-doped tellurite glasses containing silver nanoparticles. J. Alloys Compd. 686,
785–792 (2016)
28. M.S. Sajna, T. Sunil, C. Jayakrishnan, J. Cyriac, P.R. Biju, N.V. Unnikrishnan, NIR emission
studies and dielectric properties of Er3þ-doped multicomponent tellurite glasses. Spectrochim.
Acta A Mol. Biomol. Spectrosc. 161, 130–137 (2016)
29. J. Han, S. Li, T. Zhang, Design on a novel hybrid-core photonic crystal fiber with large
birefringence and high nonlinearity. Optical and Quantum Electronics. 48, 371–376 (2016)
30. G. Lakshminarayana, K.M. Kaky, J. Jedryka, A.M. El-Naggar, A.A. Albassam, G. Myronchuk,
M.A. Mahdi, Laser induced elastooptics in novel Bi2O3, and Pr2O3 doped tellurite rich glasses.
Mater. Lett. 183, 322–324 (2016)
31. A. Jha, B. Richards, G. Jose, T. Teddy-Fernandez, P. Joshi, X. Jiang, J. Lousteau, Rare-earth
ion doped TeO2 and GeO2 glasses as laser materials. Prog. Mater. Sci. 57, 1426–1491 (2012)
1 Introduction to Tellurite Glasses 13

32. A. Mori, Tellurite-based fibers and their applications to optical communication networks.
J. Ceram. Soc. Jpn. 116, 1040–1051 (2008)
33. L.d.A. Florêncio, L.A. Gómez-Malagón, B.C. Lima, A.S.L. Gomes, J.A.M. Garcia,
L.R.P. Kassab, Efficiency enhancement in solar cells using photon down-conversion in Tb/
Yb-doped tellurite glass. Sol. Energy Mater. Sol. Cells 157, 468–475 (2016)
34. N. Berwal, S. Dhankhar, P. Sharma, R.S. Kundu, R. Punia, N. Kishore, Physical, structural and
optical characterization of silicate modified bismuth borate-tellurite glasses. J. Mol. Struct.
1127, 636–644 (2017)
35. A. Lin, A. Ryasnyanskiy, J. Toulouse, Tunable third-harmonic generation in a solid-core
tellurite glass fiber. Opt. Lett. 36, 3437–3439 (2011)
36. G. Lakshminarayana, K.M. Kaky, S.O. Baki, A. Lira, P. Nayar, I.V. Kityk, M.A. Mahdi,
Physical, structural, thermal, and optical spectroscopy studies of TeO2-B2O3-MoO3-ZnO-R2O
(R ¼ Li, Na, and K)/MO (M ¼ Mg, Ca, and Pb) glasses. J. Alloys Compd. 690, 799–816
(2017)
37. M.I. Sayyed, H. Elhouichet, Variation of energy absorption and exposure build up factors with
incident photon energy and penetration depth for boro-tellurite (B2O3-TeO2) glasses. Radiat.
Phys. Chem. 130, 335–342 (2017)
38. N. Berwal, S. Dhankhar, S. Preeti, R.S. Kundu, R. Punia, N. Kishore, Physical, structural and
optical characterization of silicate modified bismuth-borate-tellurite glasses. J. Mol. Struct.
1127, 636–644 (2017)
Chapter 2
Linear and Nonlinear Optical Properties
of Some Tellurium Oxide Glasses

Cid B. de Araújo and Luciana R.P. Kassab

2.1 Introduction

A large number of heavy-metal oxide (HMO) glasses doped with rare-earth

(RE) ions has been studied by several decades and still attracting interest from
the basic point of view as well as for applications in lasers, amplifiers, displays,
modulators, switches, optical limiters, and sensors, among other applications.
Indeed, a variety of HMO glasses (e.g., tellurites [1], germanates [2, 3], and
antimony [2, 4]) still deserves numerous studies motivated by their large transmit-
tance from the visible to the near-infrared, small phonon energies, large chemical
stability, large acceptance of RE ions doping, and high nonlinear (NL) optical
response. These characteristics allowed the successful use of HMO glasses for
studies performed by many groups (see, e.g., refs. [1, 2, 5, 6]).
Among the HMO used for photonics, the tellurium oxide glasses (TOG) show
several advantages when compared with other glasses. Beside the facile methods of
fabrication of bulk TOG samples, these glasses are particularly important because
of their thermal, mechanical, and chemical properties that allow processing the
glass to fabricate films of good optical quality, synthesizing glass-ceramics, and
manufacturing optical fibers and rectangular waveguides. In general, TOG present
linear refractive indices of
2, large transmittance window (360–1000 nm), low
cutoff phonon energy (
700/cm), large mechanical resistance, high chemical
durability, high vitreous stability, and high solubility of RE ions doping species.

C.B. de Araújo (*)

Departamento de Fı́sica, Universidade Federal de Pernambuco, 50670-901 Recife, PE, Brazil
e-mail: [email protected]
L.R.P. Kassab
Faculdade de Tecnologia de S~ao Paulo/CEETEPS, 01124-060 S~ao Paulo, SP, Brazil

© Springer International Publishing AG 2017 15

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_2
16 C.B. de Araújo and L.R.P. Kassab

In this chapter, we review some of our previous results obtained with samples
having different TOG compositions. In Sect. 2.2, we will describe the methods used
for the sample’s fabrication and their basic characterization. In Sect. 2.3, we
describe studies related to the phenomena of Stokes and anti-Stokes
photoluminescence (PL) in TOG pure and doped with RE. Also samples containing
silver nanoparticles (Ag-NPs) or silicon nanocrystals (Si-NCs) were studied, and
some results are reviewed. In Sect. 2.4, we report NL optical experiments where the
NL refraction and NL absorption of TOG samples were measured aiming their
evaluation for application in all-optical devices. Finally, in Sect. 2.5, a summary of
the results and further comments on the potential of TOG for photonics are given.

2.2 Methods Used for Fabrication and Characterization

of the TeO2-Based Glasses

The samples used were prepared by the conventional melt-quenching technique.

All reagents and doping species were oxide powders obtained commercially. The
high pure reagents (~99.999%) were melted in platinum crucibles, at different
temperatures (750–800  C) depending on the composition, during 20–120 min,
quenched in air, in a heated brass mold, annealed for 2 h at 270–360  C to avoid
internal stress, and then cooled to room temperature inside the furnace. After
cooling, the glass samples were cut and polished for the optical experiments. For
the experiments with glasses having Ag-NPs, the samples with compositions
including AgNO3 were submitted to additional heat treatment (HT), at the
annealing temperature, during different periods of time, to reduce the Agþ ions to
Ag0 and to nucleate Ag-NPs; samples without Ag-NPs were also produced to be
used for comparison with the samples having NPs. The samples doped with RE ions
containing Si-NCs were also prepared by the melt-quenching method, and more
details of their fabrication are given in Sect. 2.3.3. In all PL experiments, we
observed that the emitted intensities do not change after HT of the samples without
AgNO3 (or without Si-NCs), but they change as a function of the Ag-NPs (or -
Si-NCs) amount that was controlled by the HT duration in the samples containing
AgNO3 (or Si-NCs).
To determine the size and shape of the Ag-NPs and Si-NCs, a high-resolution
transmission electron microscope (HR-TEM) operating at 300 kV was used. A
spectrophotometer that operates in the visible and near-infrared regions was used to
measure the absorption spectra of the samples from 300 to 1000 nm, and the linear
refractive index was measured using the M-line technique.
PL spectra were measured with different excitation sources, depending on the
doping species and the goals of each experiment. For continuous-wave (CW)
excitation in the visible range, a xenon lamp coupled to a 0.25 m monochromator
was used. For excitation in the infrared, the light sources used were CW diode lasers
operating either at 980 nm or 805 nm. The PL signals obtained with the CW sources
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 17

were analyzed by a phase fluorometer equipped with a 0.25 m spectrometer coupled

to a CCD. For the experiments with pulsed lasers, two sources were used. An
optical parametric oscillator (400 kW, pulses of 5 ns, repetition rate of 20 Hz)
pumped by a Q-switched Nd:YAG laser was used for the PL studies with basis on
phonon-assisted transitions. In these experiments, the signals were recorded using
photomultipliers coupled to digital oscilloscopes and computers. For the NL optical
experiments, a Q-switched and mode-locked laser (1064 nm, 17 ps, 10 Hz) was
used. The experiments were based on the Z-scan technique integrated in a
4f-system, as described in [7, 8]. The image receiver at the output of the
4f-system is a 1000  1018 pixels cooled CCD camera (30  C) operating with
a fixed gain. The sample is moved in the focal region along the beam propagation
direction (Z axis). Open- and closed-aperture Z-scan normalized transmittances are
numerically processed from the acquired images by integrating over all the pixels in
the first case and over a circular numerical filter in the second one (corresponding to
a linear aperture transmittance S ¼ 0.73 in order to optimize simultaneously the
sensitivity and the signal-to-noise ratio). The incident intensity was adjusted by a
polarizing system at the entry of the setup.
A Ti:sapphire laser (800 nm, 1.56 eV, 76 MHz, 150 fs) was used for measure-
ments based on the Kerr gate technique [9]. The laser beam was split in two beams:
probe and pump beams with intensities of 508 MW/cm2 and 52 MW/cm2, respec-
tively. The sample was positioned between two crossed polarizers. The probe beam
propagates along the Z axis, and the pump beam propagates along a direction
forming a small angle (
1 ) with the Z axis. The angle between the electric fields
of the pump and probe beams was set to 45 . When the pulses of the two beams
overlap spatially and temporally on the sample, the pump beam induces an NL
birefringence, and the polarization of the probe beam rotates while propagating
through the sample. Then, a fraction of the probe beam is transmitted through a
polarizer/analyzer, which is oriented perpendicularly to the electric field of the
incident probe beam. The signal, collected by a photodiode, is analyzed as a
function of the delay time between the pump and probe pulses.
Most of the experiments were performed at room temperature, but in the case of
Sect. 2.3.1, the sample’s temperature was varied from 100 to 540 K.

2.3 Stokes and Anti-Stokes Photoluminescence

The PL experiments were performed using samples with various compositions

having different trivalent RE ions. The main goal was the investigation of the
contribution of phonon transitions in a TOG sample (Sect. 2.3.1) and the contribu-
tion of the Ag-NPs (Sect. 2.3.2) and Si-NCs (Sect. 2.3.3) for the PL enhancement.
18 C.B. de Araújo and L.R.P. Kassab

2.3.1 Frequency Upconversion Assisted by Phonon

Annihilation

Frequency upconversion (UC) is a phenomenon in which PL with wavelength

smaller than the excitation wavelength is emitted by a physical system. The UC
process is also referred as anti-Stokes PL, while the emission with wavelength
longer than the excitation wavelength is called Stokes PL. The UC process finds
many applications, such as infrared-pumped compact lasers operating in the visible
range [10], infrared-to-visible converters and light harvesting, important for solar
cells [11], temperature sensors [12], and colored displays [13], among others.
Normally, UC is a NL process in which two or more photons from the excitation
beam are absorbed to generate one UC photon. Changing the sample’s temperature
is an interesting possibility for controlling the UC spectrum exploiting phonon-
assisted transitions in RE-doped materials. The experiment reviewed in this section
was performed to investigate a thermally activated UC emission in an Nd3þ-doped
TeO2-ZnO glass (labeled as TZO glass), prepared with 2.0 wt% of Nd2O3 according
to the procedure summarized in Sect. 2.2 and described in details in ref. [14]. The
sample was studied by resonant excitation of the Nd3þ transition 4I9/2 ! 4F5/2 at
805 nm. Figure 2.1a, b shows the room temperature PL spectrum from 340 to
800 nm. The UC emission centered at 754 nm, due to transition 4F7/2 ! 4I9/2,
presented linear intensity dependence versus the laser intensity and exhibited
670-fold enhancement when the sample’s temperature changed from 200 K to
535 K; this is the largest value reported for thermally assisted one-photon-induced
UC emission. The other emissions, associated to other electronic transitions, show
NL dependence with the laser intensity. They were identified in comparison with
previous measurements in various Nd3þ-doped glasses [14]. Figure 2.2a, b illus-
trates the behavior of the emission at 754 nm versus the laser intensity and
temperature. The slope in Fig. 2.2a indicates that only one laser photon at 805 nm
is involved in the generation of each 754 nm UC photon. Considering the excitation

a b
4 4
2 4 F7/2 I9/2
1.0 P3/2 I11/2; 4D3/2 4
I13/2 1.0
Upconversion emission intensity

2 4
I9/2;4D3/2 4
0.8 P3/2 I11/2 0.8
I15/2

I13/2
I9/2
4

4
4

I9/2

0.6 0.6
D3/2
I11/2

G9/2
G3/2

I11/2
4
4

G7/2, K13/2

2
4

I9/2

I15/2
4
D5/2,2P1/2

0.4 0.4
4

I13/2
2

K15/2

4
H11/2

G9/2

4
2

4
D3/2 I9/2
4

0.2 0.2
G7/2
2

2
4

0.0 0.0
340 360 380 400 420 440 460 480 500 520 560 600 640 680 720 760
Wavelength (nm) Wavelength (nm)

Fig. 2.1 Room temperature UC emission spectrum of Nd3þ-doped TeO2-ZnO glass for pulsed
excitation at 850 nm [14]. The bands amplitudes were normalized to the bands at 415 nm (a) and
754 nm (b)
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 19

UC emission at 750 nm (arb. units)

2200
2000 a 12000 b
1800
1600 10000
1400
8000
1200

1000 6000

800 4000

2000
600
0
20 30 40 50 60 190 240 290 340 390 440 490 540
Laser intensity (arb. units) Temperature (K)

Fig. 2.2 (a) Dependence of the UC intensity at 754 nm versus the laser intensity at room
temperature. (b) Comparison between the experimental data (circles) and the model results for
different temperatures (EPM ¼ 700/cm, black solid line; EPM ¼ 400/cm, green dashed line;
EPM ¼ 850/cm, red dashed line; and EPM ¼ 600/cm, blue dashed line) [14]

Fig. 2.3 Simplified energy levels scheme of Nd3þ representing the pathway to generate UC
emission at 754 nm by excitation at 805 nm. Solid (dashed lines) upward arrows represent laser
excitation from the fundamental (excited) level; dotted lines represent phonon-assisted tempera-
ture-dependent process, and the downward arrows represent radiative transitions. The fundamental
and excited states are represented by |0>, |1>, and |2>, respectively [14]

conditions and the linear behavior of the UC emission versus the laser intensity, we
may conclude that the signal is originated in a process which starts with the resonant
absorption 4I9/2 ! 4F5/2 followed by annihilation of phonons with promotion of the
excited ions to the thermally coupled 4F7/2 level from where the PL at 754 nm
originates. This UC pathway was verified measuring the temperature dependence of
the PL intensity. Figure 2.2b shows the 670-fold PL enhancement related to the
temperature change from 200 K to 535 K. A rate-equation model for Nd3þ
population densities based on the levels scheme of Fig. 2.3, including tempera-
ture-dependent absorption cross sections and nonradiative transition rates,
describes the thermal behavior of the UC process. The temperature dependence
of the one-photon-excited UC emission is well described considering the so-called
20 C.B. de Araújo and L.R.P. Kassab

effective phonon mode (EPM) [15], with energy of 700/cm, illustrated in Fig. 2.3b
by the black solid line that shows a good match with the experimental data. The
numerical results considering other different EPM values, 400/cm (green dashed
lines), 850/cm (red dashed lines), and 600/cm (blue dashed lines), are also shown.
From the results shown in Fig. 2.2b, we concluded that it is not the matrix cutoff
phonon mode the one dominating the phonon-assisted UC transition but the EPM,
which represents a statistical average that takes into account the phonon’s energies
and their occupation phonon number [15]. We recall that the observation of the
phonon-assisted UC phenomenon in a fluoroindate glass [16] revealed a large
sensitivity to the sample’s conditions such as humidity and mechanical resistance.
Normally, TOG, either in bulk or as optical fibers, are much more stable than
fluoroindate glasses; for the TZO glass, the reproducibility of the results is less
affected by the external conditions than fluoroindate glasses. Another important
point to be considered is the fact that the UC process reported for TZO requires only
one EPM and then it is more efficient than the previously reported cases that require
participation of multiphonon transitions.

2.3.2 Enhanced Frequency Upconversion and Energy

Transfer in the Presence of Metallic Nanoparticles

PL enhancement in rare-earth (RE)-doped HMO glasses containing metallic NPs

has been investigated by many authors after the pioneer paper by Malta et al.
[17]. The influence of the NPs on the PL efficiency of the RE ions is larger when
the incident light and/or the PL wavelengths are near resonant with the resonance
frequency of the localized surface plasmons (LSP), the quanta of coherent free-
electron oscillations in the NPs. The NPs’ dielectric function, their shape and size
distribution, the host environment, and the relative distances between the ions and
the NPs are important parameters that influence the PL efficiency. PL quenching
due to ET from the RE ion to the NP is dominant when the distance NP-RE ion is
small [18].
The influence of metal NPs on the PL of glasses with various compositions was
also studied by various groups (see, e.g., [19–25]).
One successful way to grow silver or gold NPs inside HMO glasses is based on
the melt-quenching method. The appropriate concentration of NPs and their aver-
age size and shapes depend on the glass viscosity, and the growth process is
controlled by the diffusion of the metal atoms and ions in the melt that also depends
on the procedure used for the heat treatment. Therefore, one important step to
obtain large PL enhancement is the efficient nucleation of the metallic NPs through
controlled HT of the samples. The selection of materials with appropriate param-
eters is a hard task when one is performing experiments with NPs ensembles such as
the ones performed with glasses. Also, the selection of glasses with large transpar-
ency window and small cutoff phonon energy is very important to reduce the
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 21

probability of nonradiative relaxation of the RE ions. As mentioned in the previous

sections, TGO glasses are excellent materials for PL experiments because they
present chemical and physical characteristics that contribute for the fabrication of
good samples. In all cases studied by us, the presence of NPs contributed for the PL
efficiency either due to the influence of the large local field on the RE ions
positioned in the vicinity of the NPs or due to the energy transfer (ET) from the
NPs to the RE ions. PL quenching was observed when the NPs concentration was
above certain values because the proximity between the RE ions and the NPs
increases the probability of ET from the excited RE ions to the NPs.
We demonstrated the nucleation of metallic NPs in various TOG compositions
[21, 26–35] containing Ag-NPs and Au-NPs. The presence of Pb2þ clusters in
TeO2-PbO-GeO2 glasses with Ag-NPs led to enhanced PL in the visible range [28],
whereas for Pr3þ-doped TeO2-PbO-GeO2 glass, enhanced Stokes and anti-Stokes
emission were successfully observed and analyzed [21, 29]. Further experiments
with TeO2-PbO-GeO2 glass doped with Eu3þ and containing Au-NPs [30] and Tb3þ
-doped TeO2-ZnO-Na2O-PbO glass with Ag-NPs [31] were also reported. In all
cases, large increase in the samples’ luminescence was observed. The role of ET
processes was also exploited in TOG containing metallic NPs and co-doped with
two different RE species.
Some selected examples of our work with tellurium oxide-based glasses are
described below:

2.3.2.1 Tb3þ/Eu3þ-Co-Doped TZNP Glasses with Ag-NPs

The first PL study on the influence of Ag-NPs in Tb3þ/Eu3þ-co-doped TeO2-ZnO-

Na2O-PbO glasses (labeled as TZNP glasses) was reported in [32]. The samples
were prepared with 2.0 wt% of Tb4O7, 1.0 wt% of Eu2O3, and 4.0 wt% of AgNO3.
In the experiments, the samples were excited using light with frequency larger than
the frequency bandgap of the glass, at 355 nm, which is possibly in resonance with
Tb3þ and Eu3þ transitions originating from the ground state. However, absorption
by Ag-NPs may also occur. Luminescence bands from 480 to 700 nm were
observed due to radiative transitions associated to Eu3þ and Tb3þ. The contribution
of ET processes and the intensified local field due to the Ag-NPs allowed obtaining
enhanced PL in the orange-red spectral region.
Figure 2.4 shows the absorption spectra of the samples from 350 to 700 nm.
Transitions originating from the ion ground state are observed at 480 nm (Tb3þ: 7
F6 ! D4), 465 nm (Eu3þ: 7F0 ! 5D2), and 395 nm (Eu3þ: 7F0 ! 5 L6). The
broadband centered at
490 nm, observed in the samples heat-treated for times
longer than 2 h, is attributed to LSP resonances. The absorption band presents an
inhomogeneous broadening due to the various shapes and sizes of Ag-NPs as well
as aggregates in the samples. The band amplitude increases for longer HT times due
to the increase of the Ag-NPs’ volume fraction.
PL measurements were performed using a 15 W xenon lamp (pulses of 3 μs at
80 Hz), followed by a 0.2 m monochromator to select the wavelength at 355 nm.
22 C.B. de Araújo and L.R.P. Kassab

Fig. 2.4 Absorption

spectra in Tb3þ/Eu3þ-doped
TZNP glasses with Ag-NPs,
heat-treated during various
times [36]

Fig. 2.5 PL spectra in in Tb

3þ
/Eu3þ-doped TZNP
glasses with Ag-NPs for HT
during different times; the
results for the sample
without Ag-NPs are shown
for reference (excitation at
355 nm) [32]

The luminescence spectra in Fig. 2.5 exhibit bands due to 4f-4f transitions associ-
ated to Tb3þ and Eu3þ. The results for different HT times show that the PL intensity
in the orange-red region is enhanced while increasing the volume fraction occupied
by the NPs. No PL signal in this spectral range was detected when the samples
containing only Eu3þ were excited under the same conditions. So the simultaneous
presence of Tb3þ and Eu3þ is essential to observe the strong PL signal in the
orange-red region; the emissions originating from the Eu3þ levels grow with the
increase of the HT time reaching an enhancement of ~100%. As can be observed
comparing Figs. 2.4 and 2.5, the LSP band overlaps with the 5D0 level, and then an
increase in the Eu3þ luminescence is expected due to the enhanced local field in the
proximity of the NPs. As the excitation wavelength is not in resonance with the LSP
absorption band, ET from excited Ag-NPs to the RE may occur, but it is not the
dominant process. The spectra of Fig. 2.5 can be understood by analyzing the Tb3þ/
Eu3þ energy level scheme shown in Fig. 2.6.
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 23

32 L
5D K
28 2 ET
5D
J
5L
Energy (x 103 cm–1)

3 6
24 5D
3
5D
5D 2
20 4
CR 5D
1
16 5D
355 nm
545 nm

580 nm

590 nm

614 nm

650 nm

695 nm
12
7F
8 0 7F
56
7F 7F
4 4 34
7F 7F
5 12
0 7F
6
7F
0

Tb3+ Eu3+
Fig. 2.6 Energy levels scheme of Tb3þ and Eu3þ. The solid lines represent radiative transitions,
and dotted lines represent phonon relaxation process. Dashed lines represent CR and ET processes
[32]

One possible UC pathway is due to ET from the 5D3 (Tbþ3) level to the energy
level 5L6 (Eu3þ); from this level, after nonradiative relaxation, the Eu3þ excitation
reaches level 5D0 from where radiative transitions to Eu3þ levels 7Fj (J ¼ 0–4) may
occur. Another energy pathway starts with nonradiative relaxation from level 5D3
(Tb3þ) to 5D4 (Tb3þ) providing radiative relaxations from 5D4 (Tb3þ) to the Tb3þ
levels 7Fj (J ¼ 0–6) related to emissions in the blue-red spectral region. Also quasi-
resonant cross-relaxation (CR) to Eu3þ levels 5Dj (J ¼ 0, 1, 2) may occur, as
indicated in Fig. 2.6. Efficient CR processes are very probable considering the
large concentration of RE ions. Following the CR, radiative decay corresponding to
Eu3þ transitions 5D0 ! 7FJ (J ¼ 0–4) takes place. We note that the Tb3þ emissions
at 485 and 545 nm are weak because of the ET to Eu3þ.
Figure 2.7 summarizes the relative increase of the luminescence bands at
590 and 614 nm as a function of the HT time. The results indicate that a large
number of Eu3þ are properly located nearby the Ag-NPs. There is always an ideal
distance that favors enhancement of the PL, and if some ions are not in adequate
positions, this may be the cause for not obtaining a large PL enhancement. In the
present case, the contribution of ET processes and the intensified local field due to
the Ag-NPs allowed obtaining enhanced PL in the orange-red spectral region, but
unfortunately the method of sample fabrication does not allow control of the
relative RE-NP distance.
24 C.B. de Araújo and L.R.P. Kassab

Fig. 2.7 Integrated PL

intensity for the Eu3þ
transitions at 590 nm and
614 nm versus the HT time
[32]

Fig. 2.8 TEM image of the

Tm3þ-doped TZO glass
containing Ag-NPs, heat-
treated during 48 h [33]

2.3.2.2 Tm3þ-Doped TZO Glass with Ag-NPs

Infrared-to-visible and infrared-to-infrared UC processes in Tm3þ-doped TZO

glass containing Ag-NPs were studied in another TGO glass with large potential
for photonic applications [33]. The choice of Tm3þ was motivated by the large
infrared-to-visible UC efficiency reported for other TGO glassy hosts and due to the
several applications already demonstrated for Tm3þ-doped glasses. In the present
case, the doping species were 0.5 Tm2O3 and 2.0 AgNO3 (in wt%). One TEM
image of the samples studied is shown in Fig. 2.8 for the sample heat-treated during
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 25

48 h at 325  C. Isolated Ag-NPs with diameters from
10 nm to
50 nm and

aggregates with various shapes can be seen. Figures 2.9 and 2.10, obtained with
excitation at 1050 nm, show that the PL bands centered in
477 nm,
650 nm, and

800 nm, corresponding to the transitions 1G4 ! 3H6, 1G4 ! 3F4, and 3H4 ! 3H6,
respectively, change their amplitude for different HT times. Notice that all PL
bands increase with the HT up to 48 h. One order of magnitude enhancement is
observed for the whole PL spectra that is a remarkable result in comparison with our
previous reports for other RE ions [21, 29–32]. The spectra corresponding to HT
during 72 h show a partial quenching of the PL intensity. Again the results can be
understood recalling that heat treating the samples for very long time leads to large
Ag-NPs concentration and the relative distances between the Tm3þ and the NPs
become very small. Then the excited Tm3þ transfer the energy absorbed from the

Fig. 2.9 Emission spectra

of Tm3þ-doped TZO glass
containing Ag-NPs, for
different HT times.
Excitation at 1050 nm [33]
(a) visible luminescence
(b) near-infrared
luminescence
26 C.B. de Araújo and L.R.P. Kassab

Fig. 2.10 Emission spectra

of Tm3þ/Yb3þ-doped TZO
glass containing Ag-NPs,
for different HT times.
Excitation at 980 nm [26]

laser beam to the NPs which dissipate the energy by heat. The dependence of the
UC signals with the laser intensity was analyzed to identify the possible routes
corresponding to each UC emission. Log-log plots of the UC intensities versus the
laser intensity for the transitions 2G4 ! 3H6, 1G4 ! 3F4, and 3H4 ! 3H6 present
slopes of
2.7,
2.8, and
1.8, respectively, indicating that the PL bands at
477 nm and 650 nm are due to the absorption of three laser photons, while the
transition 3H4 ! 3H6 is due to the absorption of two photons. The UC processes
occur because of the intermediate phonon-assisted steps. Even the excited state
absorption 3F4 ! 3F2,3 which is resonant is followed by emission of phonons due to
the decay from level 3F2,3 to the level 3H4.
The present results demonstrate the large influence of Ag-NPs in the infrared-to-
visible and infrared-to-infrared UC luminescence of Tm3þ-doped TZO glasses.
Enhancement of the UC emission in the heat-treated samples is attributed to the
increase of the local field on the Tm3þ located in the vicinity of the NPs as the
frequency of the incident light beam is not close to the LSP resonance frequency of
the Ag-NPs. The tenfold enhancement observed for the whole PL spectrum is an
important result that illustrates the potential of using metal-dielectric composites to
improve the performance of luminescent materials containing RE ions.

2.3.2.3 Tm3þ/Yb3þ-Doped TZO with Ag-NPs

For several years, Tm3þ-doped TGO glasses have attracted PL studies because
there are metastable levels of Tm3þ suitable for UC luminescence in the blue and
green spectral region. Addition of Yb3þ enhances the efficiency of infrared-to-
visible conversion due to the large ET rate from excited Yb3þ to Tm3þ. However,
although efficient infrared-to-visible conversion was obtained for Yb3þ/Tm3þ-co-
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 27

doped materials [37, 38], obtaining larger UC enhancement is still of great interest.
The results presented in this section show the possibility of finding new approaches
to reach larger UC enhancement by nucleation of Ag-NPs inside of a Yb3þ/Tm3þ
TGO nanocomposite prepared with 0.5 wt% of Tm2O3, 3.0 wt% of Yb2O3, and
4.0 wt% of AgNO3. The HT procedure described in Sect. 2.2 was used for the
Ag-NPs nucleation during various time intervals. PL bands corresponding to Tm3þ
transitions were observed at 480 nm, 650 nm, and 800 nm due to the ET from Yb3þ
to Tm3þ, by excitation with a diode laser operating at 980 nm, in resonance with the
Yb3þ transition 2F7/2 ! 2F5/2. UC emissions centered at 480 nm, 650 nm, and
800 nm, due to the Tm3þ transitions (1G4 ! 3H6, 1G4 ! 3F4, and 3H4 ! 3H6), were
measured, and large enhancement (
300%) for the UC luminescence was observed
in the heat-treated samples in comparison with samples without Ag-NPs as shown
in Fig. 2.10. As in the cases presented before, the growth of the PL bands correlates
with the increase of the volume fraction occupied by the Ag-NPs. TEM images
obtained for the samples heat-treated during 24 and 72 h are shown in Fig. 2.11.
Particles with average diameters varying from 30 nm (for the sample heat-treated
during 24 h) to 10 nm (sample heat-treated during 72 h) were observed together

Fig. 2.11 TEM image of

the Tm3þ/Yb3þ-doped TZO
glass containing Ag-NPs,
for HT during 24 h (a) and
72 h (b) [26]
28 C.B. de Araújo and L.R.P. Kassab

Fig. 2.12 Normalized

integrated intensity of Tm3þ
/Yb3þ-doped TZO glass
containing Ag-NPs [26]

with aggregates of NPs with dimensions of
80 nm nm that can also be seen. This
result shows that longer HT produces fragmentation of the large particles and
aggregates formed during the initial 24 h period; the growth of the PL bands
correlates with the increase of the Ag-NPs concentration. The integrated intensity
of the each PL band shown in Fig. 2.12 grows considerably for increasing HT times,
and the bands at 480 and 650 nm, having wavelengths near the LSP resonances of
the isolated NPs, present larger derivative as a function of the HT time. The PL
band at 800 nm is also enhanced due to the presence of aggregates. The large PL
enhancement of the Tm3þ transitions is attributed to the resonance of the laser
wavelength with the Yb3þ transition that is more intense than the 4f-4f transitions
of Tm3þ and the Yb3þ concentration that is six times larger than the Tm3þ
concentration. The quadratic dependence of the PL band centered at 800 nm,
corresponding to the transition 3H4 ! 3H6, with the laser power shown in
Fig. 2.13, indicates that two excited Yb3þ participate in the generation process
transferring their energies to one Tm3þ. The transitions 1G4 ! 3H6 (480 nm) and 1
G4 ! 3F4 (650 nm) present slopes equal to
3 indicating that three excited Yb3þ
are participating in the UC process. The PL growth due to the HT of the samples
demonstrates that Ag-NPs play an essential role in the UC processes that involve
triads and quartets of the RE ions. The radiative transitions corresponding to visible
light, close to resonance with the LSP, are more influenced by the isolated NPs. A
comparison between the present results and the results for Yb3þ/Tm3þ-doped
germanate glasses [39] shows larger UC enhancement for the TZO glass even for
a three-time smaller amount of silver in the starting glass composition but with a
more appropriate concentration ratio between Tm3þ and Yb3þ.
These results represented the first report showing the PL enhancement in the
presence of silver nanostructures involving triads and quartets of RE ions in a TGO
nanocomposite.
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 29

Fig. 2.13 UC intensity as a

function of the laser power
for Tm3þ/Yb3þ-doped TZO
glass containing Ag-NPs
[26]

2.3.2.4 Er3þ-Doped TWB Glasses with Ag-NPs

Experiments performed with Er3þ-doped TeO2-WO3-Bi2O3 glasses (labeled as

TWB glasses) illustrate well the contribution of Ag-NPs to increase the PL effi-
ciency of TGO glasses doped with RE ions [34]. The effect of the NPs on the Er3þ
luminescence was controlled by appropriate HT of the samples prepared with
1.0 wt% of Er2O3 and 2.0 wt% of AgNO3. Enhancement up to 700% was obtained
for the upconverted emissions at 527 nm, 550 nm, and 660 nm, when the sample
excitation is made at 980 nm. Since the laser frequency is far from the NPs’ LSP
resonance frequency, the PL enhancement was attributed to the local field increase
in the proximity of the NPs and not to ET from the NPs to the emitters. This was the
first time that the effect was investigated for tellurite-tungstate-bismutate glasses,
and the enhancement observed is the largest reported for a TGO glass. Figure 2.14
shows a TEM image of a sample heat-treated during 24 h, with Ag-NPs having an
average size of 35 nm; the electron diffraction pattern in the inset of Fig. 2.14 shows
the crystalline structure of the NPs. The LSP absorption band is not clearly
observed in the spectra of Fig. 2.15 because of the small amount of Ag-NPs and
the strong absorption of the TWB glasses in the blue region and because large NPs
present broadband resonances with small amplitudes. However, the spectra reveal
the contribution of the LSP absorption band as a tail in the blue-green region. The
LSP resonance wavelength is estimated to be between 420 and 500 nm considering
the dielectric function of silver [40] and the glass refractive index (~2), in agree-
ment with the tail observed in Fig. 2.15.
Figure 2.16 shows the UC spectra of the samples. The PL bands centered at
527 nm, 550 nm, and 660 nm correspond to the 2H11/2 ! 4I15/2, 4S3/2 ! 4I15/2, and 4
F9/2 ! 4I15/2 transitions of Er3þ, respectively. Notice that the largest enhancement
of the emission at 550 nm was obtained for HT during 24 h, whereas for 48 h and
72 h, the UC signal is reduced, probably because the concentration of Ag-NPs
became large and then the average distance NP-RE ion became smaller than the
30 C.B. de Araújo and L.R.P. Kassab

Fig. 2.14 Transmission

electron microscope image
of Er3þ-doped TWB glass
with Ag-NPs heat-treated
during 24 h [34]

Fig. 2.15 Absorption

spectra of Er3þ-doped TWB
glass heat-treated during
various time intervals. The
spectrum of a sample
without Ag-NPs is also
shown for comparison with
the other samples [34]

optimum distance for PL enhancement. Again, since the 980 nm excitation wave-
length is far from the LSP resonance wavelength, the probability of direct excitation
of the LSP band is small, and ET from the NPs to the Er3þ is negligible. So the
intensity enhancement observed is attributed to the increased local field in the
vicinities of the Ag-NPs. The quadratic dependence of the UC intensities versus
the laser intensity, shown in Fig. 2.17, indicates that two laser photons are contrib-
uting to generate each UC photon. The excitation pathway is the same identified in
previous experiments with lead germanate glasses corresponding to two steps of
one-photon absorption, according to 4I15/2 ! 4I13/2 ! 4F7/2 [41]. The same intensity
behavior was observed for the other samples.
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 31

Fig. 2.16 UC
luminescence spectra of Er
3þ
-doped TZO glasses heat-
treated during various time
intervals; the spectrum of
the sample without Ag-NPs
is also shown for
comparison with the other
samples [34]

Fig. 2.17 Dependence of

the UC signals at 550 and
660 nm versus the laser
intensity for the sample
heat-treated during 24 h
[34]

2.3.3 Rare-Earth Photoluminescence in the Presence

of Silicon Nanocrystals

Another way to obtain PL enhancement of RE ions has been demonstrated using

semiconductor nanocrystals (NCs) [36, 42–51]. The absorption cross sections of the
NCs are usually larger than the RE cross sections, and then ET from the NCs to the
RE ions may be an efficient mechanism to achieve growth of the RE luminescence.
In this case, the size of the NCs may be adjusted to match their energy gap and the
energy of the desired RE emitting states.
32 C.B. de Araújo and L.R.P. Kassab

In this section, we review recent results of the enhanced PL in TZO glass doped
with Er3þ and containing Si-NCs. This was the first study of a TGO glass in which
the nucleation of Si-NCs was obtained using the melting-quenching technique. The
previous reports of enhanced Er3þ luminescence were based on silicate glass
containing Si-NCs prepared by more complex techniques such as ion implantation,
lithography, or chemical vapor deposition [43, 48–53]. For TZO glasses doped with
Er3þ and containing different concentration of Si-NCs, the PL increase was attrib-
uted to ET from excited Si-NCs to the Er3þ located in the vicinity of the NCs. In this
case the samples were prepared with 1.0% of Er2O3 and different concentrations of
silicon powder (0.01, 0.1, and 0.4 wt%). Enhancement of
300% was observed in
the visible and in the near-infrared regions. In particular, the fourfold enhancement
observed for the broad emission centered at
1530 nm, corresponding to the Er3þ
transition 4I3/2 ! 4I15/2, indicates large potential of the composite material for
interface with existing telecommunication devices. Figure 2.18a–d shows TEM
images of Si-NCs with average sizes of 20 nm and 50 nm, obtained with melting
temperatures of 800  C and 900  C, respectively. Figure 2.19a, b shows the PL
spectra of the samples melted at 800  C, while Fig. 2.19c, d shows the results for the
samples melted at 900  C (PL spectra for samples without Si-NCs are also shown
for comparison). The bands observed correspond to the transitions 2H11/2 ! 4I15/2
(
525 nm), 4S3/2 ! 4I15/2 (
545 nm), 4F9/2 ! 4I15/2 (
680 nm), and 4I13/2 ! 4I15/2
(
1530 nm). The signals at 545 nm and 1530 nm are enhanced by
200% for the

Fig. 2.18 Electron

microscope images of
Si-NCs in Er3þ-doped TZO
glasses. (a) and (b), 0.1 wt%
and 0.4 wt% Si samples
melted at 800  C; (c) and
(d), 0.4 wt% Si sample
melted at 900  C [35]
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 33

0.3
0.6
Intensity (arbitrary units)
1% Er2O3
1% Er2O3/0.1% Si
1% Er2O3/0.4% Si
0.2
0.4

0.1 0.2

550 660 690 1400 1500 1600 1700

Wavelength (nm) Wavelength (nm)
(a) (b)

1% Er2O3
Intensity (arbitrary units)

0.8
1% Er2O3/0.1% Si
0.012
1% Er2O3/0.4% Si 0.6

0.4
0.006
0.2

0.000 0.0
500 550 600 650 1450 1500 1550 1600 1650
Wavelength (nm) Wavelength (nm)
(c) (d)
Fig. 2.19 Infrared-to-visible frequency UC and downconversion in Er3þ-doped TZO glass with
Si-NCs (excitation wavelength: 980 nm). (a) and (b), samples melted at 800  C; (c) and (d),
samples melted at 900  C [35]

samples melted at 800  C, in comparison with the samples without Si-NCs. In the
samples prepared with 0.4 wt% of Si powder and melted at 900  C, PL quenching of
the visible emission and large enhancement of
300% for the 1530 nm emission
are observed. These results show that the Si-NCs play an important role on the PL
behavior of the samples. The PL quenching in the visible range, attributed to the
process of back ET from excited Er3þ to the Si-NCs, is larger in the samples with
higher Si concentration as well as in the samples with the larger Si-NCs sizes. This
result together with the larger enhancement observed in the sample melted at
900  C is due to the higher absorption cross section of the larger Si-NCs and also
indicates that more Er3þ are located in the vicinities of the Si-NCs than in the
samples melted at 800  C.
A study of the PL intensity versus the laser intensity for the signals at 545 nm
and 1530 nm, emitted by the samples melted at 800  C and 900  C, was performed
and showed quadratic (linear) dependence of the PL intensity at 545 nm
(at 1530 nm) versus the laser intensity indicating that two laser photons generate
one photon in the visible range, while only one laser photon is involved in the
34 C.B. de Araújo and L.R.P. Kassab

generation of each photon at 1530 nm. These processes are similar to the ones
observed in [54] for germanate glasses. Because the large Si-NCs have indirect
bandgap, the electron-hole recombination rate is small, and then the ET process
from the Si-NCs to Er3þ is favored. The PL enhancement in the green range is
attributed to the interaction between two excited Si-NCs and one Er3þ, while the
enhancement of the emission at 1530 nm is due to ET from one excited Si-NC to
one Er3þ. The simple technique described here to embbed Si-NCs inside TGO
glasses opens new routes for further studies. While in the case of metal NPs, large
losses may occur due to the heating produced by the electron oscillations, in the
case of Si-NCs this process is not relevant. However, the studies of PL enhancement
of RE ions in TGO glasses are starting, and a deeper knowledge of all the processes
involved in the samples preparation and control of the Si-NCs sizes and shapes are
still necessary in order to explore this new route for obtaining more efficient PL
enhancement.

2.4 Third-Order Nonlinear Optical Properties

Beside the large applicability of TOG for UC, these glasses are promising materials
for infrared technologies, laser devices, and NL photonics. Their NL properties
were studied by various authors who reported NL refractive index n2~1015 cm2/W
and NL absorption coefficient α2~101 cm/GW in the near-infrared [55–61]. The
large nonlinearity of TOG is due to the high polarizability of the Te-O bonds and
the electron lone pair of the Te2þ ion [62, 63]. The NL susceptibility of TOG may
be enhanced by including in the glass composition compounds containing ions with
lone pairs such as Bi3þ and Pb2þ or ions with unoccupied d orbitals such as W6þ or
Nb5þ. From the results reported in the literature, it is clear that addition of Bi2O3 or
Nb2O5 contributes to increase the value of n2. However, in many cases, an increase
of α2 is also observed because of the optical bandgap reduction and/or introduction
of new localized states inside the bandgap due to the non-bridging oxygen ion
content. The increase in the α2 value makes the glass composition not attractive for
all-optical switching.
In this section, we review the third-order NL properties of a multicomponent
TOG with composition in mol %, (80x) TeO2-15GeO2-5K2O-xBi2O3 for x ¼ 5
(sample TGKB5), x ¼ 10 (sample TGKB10), and x ¼ 15 (sample TGKB15), at
800 nm and 1064 nm. The different amounts of TeO2 and Bi2O3 were used to
investigate their relative contribution for the nonlinearity. The samples were pre-
pared by melting the raw materials, tellurium oxide (TeO2), germanium oxide
(GeO2), bismuth oxide (Bi2O3), and potassium carbonate (K2CO3), previously
stoichiometrically weighted in order to obtain 7 g of bulk glass. The starting
powdered materials were mixed and loaded in a gold crucible. Then, the batch
was melted at 760  C for 1 h to ensure the complete elimination of CO2 from the
decomposition of the carbonate and a good homogenization and fining. Finally, the
melt was cooled inside a stainless mold preheated at 20  C below the glass
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 35

transition temperature, Tg , annealed at this temperature for 2 h and slowly cooled

down to room temperature to minimize residual internal stress. The values of Tg and
the crystallization temperature, Tx, for the three samples are given in [64].
The glass compositions investigated were previously used to fabricate
microstructured optical fibers as well as co-doped optical fibers for white light
generation and IR emission [65, 66].
Table 2.1 presents the linear refractive index, n0, and linear absorption coeffi-
cient, α0, of the samples studied. The values of n0, α0, and the optical bandgap, Eg,
do not change much when the relative concentration of TeO2 and Bi2O3 is changed.
The large value of n0
2 is due to the contribution of the electron lone pairs of Te2þ
and Bi3þ and to their large polarizability.
As mentioned in Sect. 2.2, the NL refractive indices were measured using the
Z-scan integrated in a 4f-system technique [7, 8]. Figure 2.20 shows the Z-scan
profiles for the closed-aperture experiments indicating positive n2 values for the
three samples. The open-aperture Z-scan experiment exhibited a small NL absorp-
tion signal only for the TGB5 sample. For the other samples, the signal was smaller
than the minimum value that our setup could detect (0.003 cm/GW).
The results obtained for n2, summarized in Table 2.2 for the three samples, do
not differ much because the hyperpolarizabilities of TeO2 and Bi2O3 have the same
order of magnitude. The large hyperpolarizability of TeO2 is due to the empty 5d
orbitals and to the electron lone pair of the Te2þ ion; Bi2O3 also contributes for the
NL susceptibility by analogous reasons.

Table 2.1 Index of refraction, n0, absorption coefficient, α0, and optical bandgap, Eg
n0 α0 (cm1)
Sample 800 nm 1064 nm 1550 nm 800 nm 1064 nm Eg (eV)
TGKB5 2.10 2.08 2.02 0.11 0.13 3.16
TGKB10 2.12 2.10 2.03 0.12 0.11 3.08
TGKB15 2.14 2.11 2.04 0.11 0.14 3.06

Fig. 2.20 Closed-aperture

Z-scan profiles for
excitation at 1064 nm. Laser
intensity: 7 GW/cm2
(sample TGKB5), 3 GW/cm
2
(sample TGKB10), 2.9
GW/cm2 (sample TGKB15)
[64]
36 C.B. de Araújo and L.R.P. Kassab

Table 2.2 Nonlinear n2(1016cm2/W)

refractive indices of the
Sample 800 nm 1064 nm
samples
TGKB5 6.5  0.7 46.5  11.2
TGKB10 8.0  0.8 56.3  9.0
TGKB15 7.1  0.7 48.8  8.8

Fig. 2.21 Kerr gate signal

at 800 nm for different
probe intensities (sample
TGKB15). Pump intensity:
508 MW/cm2 [64]

The NL response at 800 nm is illustrated by Fig. 2.21 which shows the behavior
of the Kerr gate signal as a function of the delay time between the pump and probe
pulses. The setup for this experiment is described in Sect. 2.2.
The symmetrical signals of Fig. 2.21 indicate that the NL response of the
samples is faster than 150 fs, the laser pulse duration. From these measurements,
we determined 6.5  1016 < n2 < 8.0  1016 cm2/W. As for 1064 nm, the NL
absorption coefficients of the samples were smaller than the minimum value that
our apparatus could measure (0.003 cm/GW). The results are summarized in
Table 2.2.
It is important to note that the values of α2 for both wavelengths are at least one
order of magnitude smaller than the results reported for other TeO2-based glasses.
Then, with basis on the present measurements, we calculated the figures of merit for
all-optical switching, F ¼ n2/λα2, and we obtained F > 1 for all samples that
indicate the possibility of using the materials for all-optical switching [67].

2.5 Conclusion

In this chapter, the frequency upconversion PL properties and the NL optical

response of some tellurium oxide glasses were reviewed. Enhancement of the PL
properties due to the contribution of phonon-assisted transitions in neodymium-
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 37

doped glasses or due to the presence of silver or silicon NPs in samples singly- and
co-doped with different rare-earth ions (terbium, thulium, erbium, europium, and
ytterbium) was described, for light excitation in the visible and in the near-infrared.
Energy transfer from excited silver NPs to the rare-earth ions and growth of the
local electromagnetic field due to difference between the dielectric function of the
NPs and the host material are the mechanisms contributing for PL enhancements
that in some cases reach more than one order of magnitude. In the samples with
silicon NPs, the mechanism contributing for the increase in the PL efficiency is the
energy transfer from excited silicon NPs to the ions that are promoted from the
ground state to excited states from where they emit light in the visible and in
the infrared. Measurements of the NL refractive index and NL absorption coeffi-
cients for tellurium oxide glasses containing bismuth oxide in their compositions
were also reviewed. The experiments discussed demonstrate that the NL parameters
in the near-infrared are equally influenced by the TeO2 and Bi2O3 components
present in the samples. The glasses studied exhibit large NL refractive indices, but
small NL absorption coefficients and their values indicate the possibility of using
the samples for all-optical switching devices at the near-infrared in the
sub-picosecond regime.
In conclusion, the works reviewed here demonstrated the large potential of new
tellurium oxide glass compositions for luminescent devices and ultrafast all-optical
switching.

Acknowledgments We acknowledge financial support from the Brazilian Agencies: Conselho

Nacional de Desenvolvimento Cientı́fico e Tecnológico (CNPq) and Fundaç~ao de Amparo a
Ciência e Tecnologia do Estado de Pernambuco (FACEPE). The work reported here was
performed in the framework of the National Institute of Photonics (INCT de Fotônica) and
PRONEX-CNPq/FACEPE projects. We acknowledge the Nanotechnology National Laboratory
(LNNano), CNPEM Campinas, Brazil, and Laboratório de Microscopia Eletrônica/IFUSP for
TEM measurements.

References

1. R.A.H. El-Mallawany, Tellurite Glass Handbook: Physical Properties and Data (CRC, Boca
Raton, FL, 2001)
2. NIIR Board of Consultants and Engineers, The Complete Book on Glass and Ceramics
Technology (Asia Pacific Business Press, Inc., India, 2005)
3. A. Margaryan, M. A. Piliavin (eds.), Germanate Glasses: Structure, Spectroscopy and Prop-
erties (Artech House Inc., Boston, 1993)
4. M. Razeghi (ed.), Antimony: Characteristics, Compounds and Applications (Nova Science
Publishers, New York, 2012)
5. M. Yamane, Y. Asahara, Glasses for Photonics (Cambridge University Press, Cambridge,
2000)
6. L.R.P. Kassab, C.B. de Araújo, in Photonics Research Development, ed. by V.P. Nilsson
(Nova Science Publishers, Inc., New York, 2008)
7. K. Fedus, G. Boudebs, Opt. Commun. 292, 140 (2013)
8. K. Fedus, G. Boudebs, Opt. Commun. 284, 1057 (2011)
38 C.B. de Araújo and L.R.P. Kassab

9. R.J. Sutherland, Handbook of Nonlinear Optics (Wiley, New York, 1996)

10. R. Scheps, Prog. Quantum Electron. 20, 271 (1996)
11. H.-Q. Wang, M. Batentschuk, A. Osvet, L. Pinna, C.J. Brabec, Adv. Mater. 23, 2675 (2011)
12. G.S. Maciel, L.d.S. Menezes, A.S.L. Gomes, C.B. de Araújo, Y. Messaddeq, A. Florez,
M.A. Aegerter, IEEE Photonics Technol. Lett. 7, 1474 (1995)
13. E. Downing, L. Hesselink, J. Ralston, R. Macfarlane, Science 273, 1185 (1996)
14. M.S. Marques, L.d.S. Menezes, W. Lozano B., L.R.P. Kassab, C.B. de Araújo, J. Appl. Phys.
113, 053102 (2013)
15. F. Auzel, Phys. Rev. B 13, 2809 (1976)
16. L.d.S. Menezes, G.S. Maciel, C.B. de Araújo, Y. Messaddeq, J. Appl. Phys. 90, 4498 (2001)
17. O.L. Malta, P.A.S. Cruz, G.F. de Sá, F. Auzel, JOL 33, 261 (1985)
18. A. Pillonnet, A. Berthelot, A. Pereira, O. Benamara, S. Derom, G.C. des Francs, A.M. Jurdyc,
Appl. Phys. Lett. 100, 153115 (2012)
19. T. Hayakawa, S.T. Selvan, M. Nogami, Appl. Phys. Lett. 74, 1513 (1999)
20. L.P. Naranjo, C.B. de Araújo, O.L. Malta, P.A.S. Cruz, L.R.P. Kassab, Appl. Phys. Lett. 87,
24194 (2005)
21. V.K. Rai, L.d.S. Menezes, C.B. de Araújo, L.R.P. Kassab, D.M. da Silva, R.A. Kobayashi,
J. Appl. Phys. 103, 093526 (2008)
22. L.R.P. Kassab, D.S. da Silva, R. de Almeida, C.B. de Araújo, Appl. Phys. Lett. 94, 101912
(2009)
23. T. Som, B. Karmakar, J. Appl. Phys. 105, 013102 (2009)
24. V.A.G. Rivera, S.P.A. Osorio, Y. Ledemi, D. Manzani, Y. Messaddeq, L.A.O. Nunes,
E. Marega Jr., Opt. Express 18, 25321 (2010)
25. R.J. Amjad, M.R. Sahar, M.R. Dousti, S.K. Ghoshal, M.N.A. Jamaludin, Opt. Express 21,
14282 (2013)
26. L.R.P. Kassab, L.F. Freitas, T.A.A. Assumpç~ao, D.M. da Silva, C.B. de Araújo, Appl. Phys. B
104, 1029 (2011)
27. C.B. de Araújo, D.S. da Silva, T.A.A. de Assumpç~ao, L.R.P. Kassab, D.M. da Silva, Sci.
World J. 2013, 385193 (2013)
28. C.B. de Araújo, L.R.P. Kassab, R.A. Kobayashi, L.P. Naranjo, P.A.S. Cruz, J. Appl. Phys. 99,
123522 (2006)
29. L.R.P. Kassab, C.B. de Araújo, R.A. Kobayashi, R.d.A. Pinto, D.M. da Silva, J. Appl. Phys.
102, 103515 (2007)
30. R. de Almeida, D.M. da Silva, L.R.P. Kassab, C.B. de Araújo, Opt. Commun. 281, 108 (2008)
31. L.R.P. Kassab, R. de Almeida, D.M. da Silva, C.B. de Araújo, J. Appl. Phys. 104, 093531
(2008)
32. L.R.P. Kassab, R. de Almeida, D.M. da Silva, T.A.A. de Assumpç~ao, C.B. de Araújo, J. Appl.
Phys. 105, 103505 (2009)
33. T.A.A. de Assumpç~ao, D.M. da Silva, M.E. Camilo, L.R.P. Kassab, A.S.L. Gomes, C.B. de
Araújo, N.U. Weter, J. Alloys Compd. 586, S516 (2014)
34. V.P.P. de Campos, L.R.P. Kassab, T.A.A. de Assumpç~ao, D.S. da Silva, C.B. de Araújo,
J. Appl. Phys. 112, 063519 (2012)
35. D.S. da Silva, T.A.A. de Assumpç~ao, G.B.C. de Simone, L.R.P. Kassab, C.B. de Araújo, Appl.
Phys. B 121, 117 (2015)
36. L. Yu, H. Liu, J. Nanomater. 2010, 461309 (2010)
37. G. Wang, W. Qin, L. Wang, G. Wei, P. Zhu, R. Kim, Opt. Express 16, 11907 (2008)
38. R.H. Page, K.I. Schaffers, P.A. Waide, J.B. Tassano, S.A. Payne, W.F. Krupke, W.K. Bischel,
J. Opt. Soc. Am. B 15, 996 (1998)
39. T.A.A. Assumpç~ao, D.M. da Silva, L.R.P. da Silva, C.B. de Araújo, J. Appl. Phys. 106, 063522
(2009)
40. P.B. Johnson, R.W. Christy, Phys. Rev. B 6, 4370 (1972)
41. D.M. da Silva, L.R.P. Kassab, S.R. Luthi, C.B. de Araújo, A.S.L. Gomes, M.J.V. Bell, Appl.
Phys. Lett. 90, 081913 (2007)
2 Linear and Nonlinear Optical Properties of Some Tellurium Oxide Glasses 39

42. A.A. Bol, R. van Beek, A. Meijerink, Chem. Mater. 14, 1121 (2002)
43. D.J. Lockwood, J. Mater. Sci. Mater. Electron. 20, S235 (2009)
44. M.J.A. de Dood, J. Knoester, A. Tip, A. Polman, Phys. Rev. B 71, 115102 (2005)
45. X.L. Wu, Y.F. Mei, G.G. Siu, K.L. Wong, K. Moulding, M.J. Stokes, C.L. Fu, X.M. Bao, Phys.
Rev. Lett. 86, 3000 (2001)
46. R.A. Senter, C. Pantea, Y. Wang, H. Liu, T.W. Zerda, J.L. Coffer, Phys. Rev. Lett. 93, 175502
(2004)
47. E.F. Pecora, T.I. Murphy, L.D. Negro, Appl. Phys. Lett. 101, 191115 (2012)
48. N. Koshida (ed.), Device Applications of Silicon Nanocrystals and Nanostructures. Nano-
structure Science and Technology (Springer, New York, 2009)
49. A.J. Kenyon, Prog. Quantum Electron. 26, 225 (2002)
50. F. Priolo, G. Franzò, D. Pacifici, V. Vinciguerra, F. Iacona, A. Irrera, J. Appl. Phys. 89,
264 (2001)
51. A. Irrera, F. Iacona, G. Franzò, M. Miritello, R.L. Savio, M.E. Castagna, S. Coffa, F. Priolo,
J. Appl. Phys. 107, 054302 (2010)
52. C. Meier, A. Gondorf, S. Lüttjohann, A. Lorke, H. Wiggers, J. Appl. Phys. 101, 103112 (2007)
53. C. Delarue, G. Allan, M. Lannoo, Phys. Rev. B 48, 11024 (1993)
54. D.S. da Silva, L.P. Naranjo, L.R.P. Kassab, C.B. de Araújo, Appl. Phys. B 106, 1015 (2012)
55. Y. Chen, Q. Nie, T. Xu, S. Dai, X. Wang, X. Shen, J. Non Cryst. Solids 354, 3468 (2008)
56. V.K. Rai, L.d.S. Menezes, C.B. de Araújo, Appl. Phys. A 91, 441 (2008)
57. E. Yousef, M. Hotzel, C. Rüssel, J. Non Cryst. Solids 342, 82 (2004)
58. E. Yousef, M. Hotzel, C. Rüssel, J. Non Cryst. Solids 353, 333 (2007)
59. L. Canioni, M.O. Martin, B. Bousquet, L. Sarger, Opt. Commun. 151, 241 (1998)
60. S. Kim, T. Yoko, J. Am. Ceram. Soc. 78, 1061 (1995)
61. H. Nasu, T. Uchigaki, K. Kamiya, H. Kanbara, Jpn. J. Appl. Phys. 31, 3899 (1992)
62. M. Lines, Phys. Rev. B 43, 11978 (1991)
63. A.P. Mirgorodsky, M. Soulis, P. Thomas, T. Merle-Méjean, M. Smirnov, Phys. Rev. B 73,
134206 (2006)
64. T.A. Oliveira, D. Manzani, E.L. Falc~ao-Filho, Y. Messaddeq, G. Boudebs, K. Fedus, C.B. de
Araújo, Appl. Phys. B 116, 1 (2014)
65. D. Manzani, J.L. Ferrari, F.C. Polachini, Y. Messaddeq, S.J.L. Ribeiro, J. Mater. Chem. 22,
16540 (2012)
66. D. Manzani, Y. Ledemi, I. Skripachev, Y. Messaddeq, S.J.L. Ribeiro, R.E.P. de Oliveira,
C.J.S. de Matos, Opt. Mater. Express 1, 1515 (2011)
67. G.I. Stegeman, in Nonlinear Optics of Organic Molecules and Polymers, ed. by H.S. Nalva,
S. Miyata (CRC, Boca Raton, FL, 1997), p. 799
Chapter 3
Trivalent Lanthanides in Tellurite Glass

V.A.G. Rivera and L.A.O. Nunes

3.1 Introduction

Trivalent lanthanides, also called rare-earth ions (REIs), show light emissions in
almost all the near infrared and visible spectrum. Such luminescent properties are
often related to potential applications in a variety of fields since sharp emission
lines, meta-stable states, the ability of converting long/short-wavelengths
(up/down-conversion process), energy transfer processes (resonant and
non-resonant) and high quantum efficiency make them appropriate for the manu-
facture of many optical devices.
Tellurite glasses are excellent candidates for telecommunication applications
due to their higher solubility for REIs, low phonon energy, the vibrational modes
characteristic of their composition and the arrangement of their structure bonds. In
this framework, photons emitted from REIs can interact with amorphous structures
in a number of ways. Photon interaction types depend on the sites, which can be
expressed in electronic terms for convenience, rather than the chemical point view.
In this sense, the technological challenges in REI spectroscopy are to understand
opto-magnetic interactions and several other problems that are yet to be solved.
The present chapter analyses one of the most discussed fields of spectroscopy of
REIs: doped tellurite glasses and some of their applications in frontline research.
Since it is practically impossible to cover all the theoretical and experimental
aspects of REI spectroscopy in a single chapter, we emphasize applications in

V.A.G. Rivera (*)

Facultad de Ciencias Fı́sicas, Universidad Nacional Mayor de San Marcos, Lima, Peru
e-mail: [email protected]
L.A.O. Nunes
Instituto de Fı́sica de S~ao Carlos—INOF, University of Sao Paulo, Caixa Postal 369,
13560-970 S~ao Carlos, SP, Brazil

© Springer International Publishing AG 2017 41

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_3
42 V.A.G. Rivera and L.A.O. Nunes

telecommunications and related technological fields by giving a brief account on

the status of this subject.

3.2 Free Ions

To understand the energy levels of REIs in a crystal it is necessary to examine them

in detail in the free atom. Since the spectral lines originate from transitions within
the 4f manifold (intra-4f transitions, contrary to what is usually observed in atomic
spectra), only the energy levels of the 4f electrons in the free atom or ion can be
determined with good accuracy from a Hamiltonian that contains Coulomb repul-
sion and spin-orbit interaction. For the 4f electrons these two interaction energies
are of roughly the same order of magnitude. The free ion energy levels can be
obtained from first principles, a procedure that provides a good free ion eigenvalue
as well as eigenfunctions, which can then be used to describe the energy levels
observed in solids.
The REIs in solids are either divalent (4fN5s25p6) or trivalent (4fN–15s25p6). The
most common valence state of the REIs in solids is the trivalent one. The 4f
electrons are shielded from external fields by two electron shells with larger radial
extension (5s25p6), which explains the atomic nature of their spectra. Therefore, the
4f electrons are only weakly perturbed by the charges of the surrounding ligands
(host matrix), since the position of this electronic shell is below the top of the
valence band. In this manner, the crystal environment constitutes a small perturba-
tion on the atomic energy levels of the REI; therefore, spectroscopic properties can
be understood from consideration of the free ions. In addition, the wavefunctions of
the free ions represent a good zero order approximation for a description of solid
state properties.
The energy levels for the trivalent free REIs Pr3+, Gd3+, and Er3+ have been
analyzed in arc spectra by J. Sugar [1], Kielkopf and Crosswhite [2], and
Crosswhite and Moos [3], respectively.
The energy levels of a free REI are generally deduced by considering only
interactions between the 4f electrons themselves; since all the other electron shells
are spherically symmetric, their effect on all the terms of a 4f configuration is of the
first order. However, if we consider that the aspherical 4f shell can alter the closed
spherical shell, one can tolerate a small interaction with 4f shell, which often can be
neglected, allowing one to solve the angular part of the Schrodinger equation. If the
symmetry of the problem is known, we can obtain information on the number of
energy levels and relative ordering expected, as well as the selection rules for
transitions of various types of radiation. To solve of the radial part of the
wavefunction, we require knowledge of the exact magnitude of the wavefunctions,
which often is harder to determine. Therefore, we can write the Hamiltonian that
determines the 4f levels as [37]:
3 Trivalent Lanthanides in Tellurite Glass 43

h2 XN XN
Z ∗ e2 X N
e2 X N
H¼ Δi  þ þ ζ ðr i Þsi  li ð3:1Þ
2m i¼1 i¼1
ri r
i<j ij i¼1
|fflfflffl{zfflfflffl} |fflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflffl}
Hc H SO

where N ¼ 1, . . ., 14 is the number of 4f electrons, Z*e is the screened charge of the

nucleus because the closed electron shells have been neglected, and ζ(ri) is the
spin–orbit coupling function, defined as [37]:

h2 dUðr i Þ
ζ ðr i Þ ¼ ð3:2Þ
2m2 c2 r i dr i

U(ri) is the potential in which the electron i is moving. The first term in Eq. (3.1) is
the kinetic energy of the 4f electrons, and the second term is the Coulomb
interaction with the nucleus. This term represents the interaction of the nucleus
with the closed shells, where the charge fluctuation at any one point has
non-negligible effects, hence, it modifies only this term’s magnitude but not the
symmetry of the closed shells. Thus, it replaces the real charge of the nucleus with a
screened charge. These two terms of Eq. (3.1) are spherically symmetric and
therefore do not remove any of the degeneracies on the configuration of the 4f
electrons. The third and fourth terms are the mutual Coulomb interaction of 4f
electrons (Hc) and their spin-orbit interaction (Hso), respectively, which are respon-
sible for the energy level structure of the 4f electrons. There are two limiting cases
for the relative size of these two interactions: (1) when Hc  Hso, we have the
so-called Rusell-Saunders coupling and the energy level structure is determined
from the diagonalization of Hc. (2) When Hc  Hso, we have the so-called j–j
coupling framework. In REIs these terms are of about equal magnitude and,
therefore, calculations of energy levels must consider both terms; this situation is
called intermediate coupling. Details of the matrix element calculations of Hc and
Hso can be found in [4, 5]. For instance, in a configuration of two f electrons, the
Coulomb interaction matrix elements have the form:
 
2  e2  2 0 0 0 0 
f SLMS ML  f S L M S M L ð3:3Þ
r 12

where S, L, Ms., and ML are the quantum numbers total spin angular momentum,
orbital angular momentum, magnetic quantum number, and electron spin number,
respectively, of the two f electrons. Employing the Legendre polynomials, we can
P  r<k
develop r12 ¼
1
rkþ1
Pk ð cos ω12 Þ, where r< and r> are the smaller and larger of
>
k
P ∗
the radii of the two electrons, and Pk ð cos ω12 Þ ¼ 2kþ1
4π
Y kq ðθ1 ; φ1 ÞY kq ðθ2 ; φ2 Þ,
q
where Ykq represents spherical harmonics. Now we define the tensor operator as:
44 V.A.G. Rivera and L.A.O. Nunes

rffiffiffiffiffiffiffiffiffiffiffiffiffi
4π
Ckq ¼ Y kq ð3:4Þ
2k þ 1
P
with the result Pk ð cos ω12 Þ ¼ ð1Þq Ckq ð1Þ  Ckq ð2Þ ¼ Ck ð1Þ  Ck ð2Þ, therefore:
q

 
  r <k  
f 2 SLMS ML e2 kþ1
Ck ð1Þ  Ck ð2Þf 2 S0 L0 MS 0 ML 0 ð3:5Þ
r>

We define the scalar product of two tensor operator as T k  U k ¼

P
ð1Þq T kq U kq , and let us assume j1 and j2 are the momenta of the two systems
q
coupled to J, the momentum of the total system. Then the dependence on the
projection quantum number j1j2 can be determined by the Wigner-Eckart theorem
[6]:
( )
j1 0 j2 0 J
0 0 0 0 j1 0 þj2 þJ
hγj1 j2 JMJ jðT k  U k Þjγ 0 j1 j2 J MJ i ¼ ð1Þ δJJ0 δMJ MJ0 
j2 j1 k
X
hγj1 jjT k jjγ 00 j1 0 ihγ 00 j2 jjU k jjγ 0 j2 0 i ð3:6Þ
γ 00

where (kk) is a so-called reduced matrix element and the {} factor is a 3j symbol
[4, 7], and γ stands for all quantum numbers on which the tensor operator does not
act. From Eq. (3.6) the tensor operator Tk acts only on part 1 in the coupled system
hγj1j2JMJ| and the operator Uk acts only on part |γ 0 j1 0 j2 0 J 0 MJ0 i in the same system.
Now, we are in a position to obtain the matrix elements of the Coulomb and spin-
orbit interactions. As Coulomb interaction is an electric interaction, it does not act
on the spin; therefore we obtain:
 
2  e2  2 0 0 0 0 
f SLMS ML  f S L M S M L
X r 12    
¼ f k ðr Þ f 2 :::Ck ð1ÞCk ð2Þf 2 :::0
k
X l l k ð3:7Þ
¼ f k ðr Þð1Þ2lþL δLL0 δML ML 0 δSS0 δMS MS 0 ðhlkCk kliÞ2
l l L
X
k
 2
L 3 k 3 3 3 k
¼ f k ðr Þð1Þ δLL0 δML ML 0 δSS0 δMS MS 0
k
0 0 0 3 3 L
3 Trivalent Lanthanides in Tellurite Glass 45

For the radial integral we used the definition

ð
r <k 2
f k ðr Þ ¼ e 2 R4f ðr 1 ÞR24f ðr 2 Þdr 1 dr 2 ð3:8Þ
r kþ1
>

In fact, we can calculate the Coulomb matrix element for every fN configuration
by tensor operator techniques; however, for N > 2 this becomes a problem.
Nevertheless, there are conventions that enable us to apply tables directly [8].
The matrix elements of the spin-orbit interaction can be evaluated by using the
P
N
tensor formalism, since the spin-orbit coupling Hamiltonian ζ ðr i Þðsi  li Þ is an
i¼1
example of a product tensor operator. The spin-orbit interaction matrix elements
have the form:

 X
N  
f N SLMJ  ζ ðr i Þðsi  li Þf N S0 L0 MJ 0 ¼
i¼1
pffiffiffiffiffi 0 S S0 1 N   
84ζ 4f ð1ÞJþLþS δJJ0 δMJ MJ0 f SLjV 11 jf N S0 L0 ð3:9Þ
L0 L J

The matrix elements for the operator V11 are calculated in [8]. Additionally,
from the assumed equivalence of the 4f electrons, the spin-orbit coupling parameter
is equal for all 4f electrons, and is defined as:
ð
ζ 4f ¼ R24f ζ ðr Þdr ð3:10Þ

The matrix elements of the spin-orbit coupling are responsible for the mixing of
states with the same J but different L and S values, while the diagonal matrix
elements are obtained by:
 N   N 
f :::H so f ::: ¼ ζ ðλ=2Þ½J ðJ þ 1Þ  LðL þ 1Þ  SðS þ 1Þ ð3:11Þ

Where λ is a constant that depends on L and S.

As mentioned above, in REIs the Coulomb and spin-orbit interactions are of the
same order of magnitude, but the splitting effect produced by spin-orbit interaction
between two levels strongly depends on given state (J, L and S). For instance, the
splitting between the 2F7/2 and 5F5/2 terms is 2000 cm–1 for Ce3+ and 10,000 cm–1
for Yb3+; such a separation is produced only by the spin-orbit interaction.
The interactions considered so far are the most important ones describing REIs
free energy levels. In comparing experimental and theoretical results, however,
discrepancies of the order of 100 cm–1 remain, and a number of researchers have
tried to reduce these discrepancies by taking into account further interactions.
46 V.A.G. Rivera and L.A.O. Nunes

Fig. 3.1 Energy levels of the triply charged lanthanide ions. The most technologically important
radiative transitions are labeled

Trivalent REIs in solids for many of the energy levels below 30,000 cm–1 have been
analyzed, see Fig. 3.1. In addition, there are many free ion energy levels very close
to each other, so that in many instances their crystal field levels overlap, which
makes it difficult to analyze the spectra in terms of discrete J states. The energies
that are the centers of gravity of the crystal field multiplets do not vary much for
different hosts, meaning that their position in a typical host material is representa-
tive of that in all other crystals. A compilation of the energy levels found in the
trivalent REIs appears in Fig. 3.1, which is a great guide for the analysis of an
unknown spectrum.
G.H. Dieke shows the approximate positions of the next excited configurations
with respect to the 4fN configuration [9]. Where the 4fN configuration is relatively
isolated, the next excited configuration is 4fN-15d, which starts about 50,000 cm–1
above the ground term of the 4fN configuration for Ce3 + and about 100,000 cm–1
above it for Yb3+. These data are the free ions. On the other hand, the 4fN ! 4fN-
15d transitions are very intense because they are no longer forbidden by the parity
selection rule for electric dipole radiation. In the near-infrared region they may
obscure 4fN ! 4fN transitions.
Figure 3.2 shows the radial distribution functions of the 4f, 5s, 5p, and 6s
electrons for Gd+ as obtained from the Hartree-Fock calculation by Freeman and
Watson [10]. This graph shows that the 4f electrons are indeed inner electrons with
relatively small hrni values. The crystal field interaction is proportional to these hrni
values, and thus such an interaction will be small in REIs. In addition, the 4f
wavefunctions do not extend very far beyond the 5s and 5p shells.
3 Trivalent Lanthanides in Tellurite Glass 47

1.4
4f 4f
5s
1.2 5p
6s
1.0 5s
6s2
5p
5d0-1
0.8 5s25p6
P 2(r)

4f2-14
1s2-4d10
0.6
core
0.4 unfilled shell
6s
shielding electrons
unfilled shell
bonding electrons
0.2

0
0 0.6 1.4 2.2 3.0 3.8 4.6 5.4 6.2 7.0
r (a.u.)

Fig. 3.2 Square of the radial function times the squared radius for the 4f, 5s, 5p, and 6s electrons
in Gd+ [11, 12]. Picture inset schematically shows that the 4f orbital is within the 6s, 5p and 5s
levels

3.3 Rare-Earth Ions in a Static Crystal Field

The most notable feature about the spectra of REIs in ionic crystals is the sharpness
of many lines in the absorption and emission spectra. The spectra suggest that the
various groups of lines are due to transitions between free ion states split by the
interaction of the REIs with the surrounding charges of the lattice. This gives a
measure of the magnitude of the interaction of 4f electrons with the crystal field of
100 cm1. Such an interaction is usually described by the charges of the crystal field
at the site of the 4f electrons, also so-called fields of the crystal field. However, the
4f transitions are forbidden for electric dipole radiation by the parity selection rule.
The question arises, what then makes these transitions be possible? In 1937, van
Vleck [10] noted that the interaction with the crystal field or with the lattice
vibrations can mix states of different parity into the 4f states, and that these
admixtures of other parity wave functions make the transitions possible.
An REI within a crystal will experience an inhomogeneous electrostatic field,
the so-called crystal field, producing a distortion in the closed shells of the REI,
which is produced by the charge distribution in the crystal. Consequently, it pro-
duces changes in the energy level. This crystal filed splitting can be obtained
by applying the potential produced by the crystal to the REI 4f wave functions.
48 V.A.G. Rivera and L.A.O. Nunes

The existence of splitting tells us: (1) The development of the crystal potential into
spherical harmonics contains a second order term (related to the crystal symmetry)
namely, the number of levels into which an REI J terms is split in a crystal field of a
given symmetry. (2) From the size of the splitting the size of the crystal field at the
site of the electrons can be deduced.
A realistic description of this interaction must take into account that each
individual ion of the crystal has spatial charge clouds which can penetrate each
other, inclusive the REI, and thereby interact. Therefore, let us assume that:
• An overlap of the charge distributions of the neighboring ions and the 4f
electrons is produced due to the extended charge distribution of the crystal.
• The lattice vibrations and their effect on the energy levels of the REI are
neglected, i.e., the REIs are considered static in the crystal.
• The interaction of 4f electrons of adjacent REIs is neglected.
• In addition, the 4f electrons of one ion are considered independent of each other.
From these assumptions, we obtain the crystal potential [13]:
X
V¼ Bkq Ckq ðθi ; φi Þ ð3:12Þ
k , q, i

The crystal field parameter Bkq is defined as:

ð  k 
q r<
Bqk ¼ e ð1Þ ρðRÞCkq ðθ; φÞ kþ1 dτ ð3:13Þ
r>

where the crystal field has charge density ρ(R) and the 4f electrons have radius ri.
From Eq. (3.12) we can determine the angular part of the crystal field matrix
elements employing the tensor operator technique:
 N    X k N   
f SLJMJ V f N S0 L0 J 0 MJ 0 ¼ Bq f SLJMJ U kq f N S0 L0 J 0 MJ 0 hf kCk kf i
kq

ð3:14Þ

V describes the electrical interaction, therefore it cannot act on the spin, thus
S ¼ S0 . The matrix elements are diagonal only in S, and Eq. (3.14) makes it possible
to evaluate matrix elements of the crystal field Hamiltonian between different states
of J. If in a matrix element an operator of rank k connects angular momenta l1 and l2
so that the matrix elements is nonzero, then l1 + l2  k, and for f electrons k  6
always holds. The crystal field splitting is energy, the matrix elements have to be
real, therefore, k is always even. The nonzero terms (k, q) are restricted by the point
symmetry at the site of the REI of the crystal.
3 Trivalent Lanthanides in Tellurite Glass 49

3.4 Classification of Crystal Field States

and Selection Rules

The selection rules in crystals are connected with the symmetry of the crystal under
consideration, such as shown the Eqs. (3.7–3.14), i.e., principles physics-mathematical
capable of dealing such symmetry properties. (For a brief introduction see [He76].)
Here we do not intend to review group theory, but we want to outline those
elements that are needed to apply the theory to crystal spectroscopy. For more
details, see [Ti64, La74]. Additionally, the crystal field splitting of REIs is derived
from the Kramers theorem (Kramers degeneracy), which states that for ions with an
odd number of electrons all crystal field levels are at least doubly degenerate, or that
nondegenerate levels in a crystal field can be found only for ions with an even
number of electrons. Let us consider an atom with average charge distribution that
is independent of the direction of motion of the electron, i.e., independent of the
direction of the spin electron. In this case, the interaction energy must be indepen-
dent of the sign of the directional quantum number Mj. If we consider a state with
J1 ¼ ½ and one with J2 ¼ 1, this means that the electric field cannot split the J1 state
any further, whereas the J2 state will be split into an MJ ¼ 0 and an MJ ¼ 1 state.
The site symmetry of REIs in an amorphous lattice also imposes limits on the
crystal field operator; therefore, our interest is to identify the non-vanishing com-
ponents of crystal field operators and their matrix elements. From Eq. (3.13) we can
write:
k
Bq ¼ ð1Þq Bqk ∗ ð3:15Þ

where the number of non-zero Bqk depends on the site symmetry. For low site
symmetry this number may be great. When analyzing experimental data the Bqk are
usually determined by the least squares fit. This is often an ambiguous procedure
with several equally likely solutions.
Group theory is a general tool in physics and is therefore widely employed and
can be applied to analyze crystal spectra, since almost all the necessary information
is available in table form [13]. Several authors have carried out calculations of
crystal field energies and corresponding crystal field parameters for REIs in diverse
compounds [14].
Theoretical models, more or less phenomenological based on the symmetry of
REIs, are essential in providing a clear theoretical understanding of electronic
interactions of REIs in solids. Not only can theoretical models generate the
phenomelogical crystal field of f-element ions in solids, but they can also reveal
fundamental aspects of the ion-ligand interactions. However, the calculated crystal
field parameters are expected to be more or less different from those determined by
experiments.
50 V.A.G. Rivera and L.A.O. Nunes

Therefore, we can write the effective atomic Hamiltonian as the free atomic ion
interaction (Eq. (3.1)) and the contribution of the crystal field (Eq. (3.12)) as:

h2 XN XN
Z ∗ e2 X N
e2 X N
H¼ Δi  þ þ ζ ðr i Þsi  li þ V ð3:16Þ
2m i¼1 i¼1
ri r
i<j ij i¼1

The crystal field produce a splitting of energy levels due to the orbital
wavefunction of the REI will increase an electron’s energy when the REI is located
in a region of high electron density. That is, the crystal field is responsible for
multiplet splitting, which can be observed in spectral line shapes.

3.5 Intensities of Optical Transitions

Equation (3.16) is time-independent; it does not describe stationary states of the

system. States are treated using time-dependent perturbation theory resulting in
transitions between the states established by static interactions. In luminescent
devices, the most important term from Eq. (3.16) is V, which gives rise to the
emission and absorption of photons from decay radiation of the REI. This involves
the interaction between the electron charge and the electric field and the interaction
between the electron spin and the magnetic field.
The experimental data on spectra of REIs show that the radiation is mostly
electric dipole (ED) in nature, though some cases of the magnetic dipole radiation
are also observed. Since the optical transitions take place between levels of a
particular 4fN configuration, the electric dipole radiation is forbidden in the first
order, because the electric dipole operator has uneven parity and the transition
matrix element must have even parity (Laporte selection rules).
Van Vleck [54] pointed out that the dipole electric radiation can only occur
because the 4fN states have admixtures of 4fN-1nl configuration (nl will be mostly
5d), thus the 4fN-1nl has to be chosen such that it has opposite parity from 4fN. The
admixture is produced via interactions that have odd parity, and this depends on the
host matrix of the RE ions. However, we can have four dominant sources of optical
radiation in RE ion spectra:
1. Forced electric dipole induced by odd terms of the crystal field
2. Forced electric dipole radiation induced by lattices radiation
3. Allowed magnetic dipole transition
4. Allowed electric quadrupole radiation
In free atoms, the magnetic dipole is about six orders of magnitude weaker than
the electric dipole radiation. This latter occurs only as consequence of a perturba-
tion; both kinds of radiation show up in the RE spectra with about equal intensity.
Quadrupole radiation is less probable still, in comparison with the magnetic dipole.
3 Trivalent Lanthanides in Tellurite Glass 51

There are different types of transitions between levels (called electric dipole,
electric quadrupole, and magnetic dipole) and therefore the transitions are divided
into allowed ones (with a high probability) and so-called forbidden transitions (for
transitions where the electric dipole transition is quantum mechanically forbidden).
Because ED transitions ar induced by the crystal field, the free atom selection
rules break down almost completely, i.e., the selection rules with regard to the L, S
and J quantum numbers are now essentially governed by the crystal field
interaction.
The Judd-Ofelt theory is a good approximation for the determination of the
intensities in RE crystal spectra [55, 56], where the intensities are expressed in
terms of oscillator strengths, where the total oscillator strength describes a transi-
tion from an energy level a to an energy level b. This Judd-Ofelt formalism has been
applied to the analysis of a number of systems, but in those analyses the crystal field
splitting of the terms is neglected; therefore, the total absorption intensities between
the ground term and the excited terms are analyzed with only three empirical
parameters [4, 9, 15].
For more information about REI spectroscopy, including energy level cal-
culation, strength and other spectroscopic characterizations techniques, see Refs.
[16–19].

3.6 Optical Properties of the REIs in Tellurite Glasses

Perhaps the most demanding commercial products are cellphone, television, inter-
net and communication systems. In such systems, signal transmission through an
optical fiber is crucial. However, long-distance data transport operates in several
electromagnetic spectral windows, so bandwidth is a challenge, as witnessed by a
series of developments starting from optical fiber transmission at low attenuation
[20, 21]. In this sense, the field of luminescence from REI has been one of steady
growth during the past two decades, mainly owing to the ever-increasing demand
for optical sources and amplifiers operating at wavelengths compatible with optical
communication technology.
For instance, most of the interest in luminescent REI has concentrated on trivalent
erbium (Er3+), in particular its emission band around 1.55 μm (4I13/2 !4I15/2
transition) that allows the transmission and/or amplification of signals in this region.
Such Er3+ properties allowed the development of the erbium-doped fiber amplifier
(EDFA) in 1980s [22–24] for the C and L bands, regions without the necessity for
expensive optical to electrical conversion [25], as well as thulium-doped fiber
amplifiers (TDFA) for the S-band and praseodymium-doped fiber amplifiers
(PDFA) for the 1.3 μm band [26–28]. We note that the REIs maintain the majority
of their individual properties in different glass matrices. Today, there continues to
exist the same demand for increasing bandwidth, in hopes of extending the spectral
range of fiber amplifiers and associated sources using other luminescent REIs and
novel host materials. The explosive growth of information in the twenty-first
52 V.A.G. Rivera and L.A.O. Nunes

century, and the rapid increase in demand and the exchange of information, have
led to an urgent need for information on ultra-high speed and ultra-fast response and
large capacity processing.
The most common tellurite glass preparation techniques are the melt quench
method, chemical vapor deposition and the sol gel method. The melt quenching
technique was the first glass preparation technique used in the glass industry and in
the research field, before chemical vapor deposition and the sol gel technique
[29, 30, 64]. According to Mallawany, tellurite glass formation occurs, for example,
as covalent, ionic, molecular, metallic and hydrogen bonded materials.
Tellurite glasses (multicomponent oxide glasses in which the major component
is TeO2) possess nonlinear susceptibilities on the order of 10 compared to those
found in silica, and therefore offer an attractive option for nonlinear applications
[31, 32]. Moreover, they offer the additional benefit of a broad optical transmission
window extending from 350 nm to 5 μm, offering the option of mid IR
transmittance [31, 33, 34]. Complete spectroscopic analysis of Yb3+, Pr3+, Nd3+,
Er3+, Tm3+ and Ho3+ doped tellurite glasses has revealed their potential for laser
[35] and amplifier devices for optical communication wavelengths [36–40], and
optical fiber [31, 41–46].
Thermal properties of tellurite glasses depend on the amorphous structure, since
they depend on connectivity of structural units and strength of interatomic bonds.
Furthermore, it is important to consider the influence of modifiers in determining
the structure of glass and its properties, since the preparation of glasses can define
their optical transparency [47], viscosity and other physical properties [48].
Characterization techniques, because they are crucial to researchers as basic
tools in identifying the structure and thereby properties of synthesized materials,
which in turn determines whether the designed tellurite glasses are suitable for
particular applications.
Erbium-doped tellurite glasses offer the advantage of increased bandwidth over
silica glass [38, 49–51], and devices with gain-flattened bandwidths of up to 80 nm
have been reported [52] or even more [53]. We can produce waveguides by ion
exchange [54–56], which provides good depth control and a uniform concentration
profile through the waveguide.
Thulium-doped tellurite glass been investigated for application in optical ampli-
fiers in the 1.4 μm region [57].
A. Jha studied the effects of compositional modification on the thermal and
viscous flow properties on tellurite glasses, and compared them with the resistance
as a function of devitrification. In addition, Jha also reports on the characterization
of OH impurities in tellurite glasses.
Nonlinear optical materials show a variety of frequency conversion mechanisms
with a wide range of applications to photonics, information, medical, industrial and
military technologies. In general, glasses are known for their ease of fabrication,
chemical composition, flexibility and dopants solubility (e.g., REI). Among the
oxide glasses, tungsten-tellurite glasses (TTG) have shown several advantages,
such as a wide transmission window (0.4–5.0 μm), good thermal and mechanical
stability, and intermediate cut-off phonon energy (900–950 cm1) [57, 58].
3 Trivalent Lanthanides in Tellurite Glass 53

Tellurite glass has the highest refractive index among oxide glasses, both at the
visible and near IR regions, so TeO2 based glass systems are considered to be
promising materials in photonic and optoelectronic devices [59]. Their relatively
high refractive index is a result of high polarizability of tellurium ions [60]. Addi-
tionally, they also possess large third order nonlinear optical susceptibilities, which
are approximately 100 times higher than in case of traditional silicate glass [61, 62].
Large solubility of REI to the contrary, e.g., as reported by Polman and
co-workers [63], here we see the effect of implantation damage on the tendency
of erbium ions to cluster in soda-lime glass films. Tellurite glass has particularly
high linear and nonlinear refractive indices (e.g., 1.8–2.3 and 20–50  1020 m2/W,
respectively) [64]. Absorption coefficients below the absorption edge vary expo-
nentially with photon energy indicating the presence of Urbach’s tail. The origin of
the Urbach energy is associated with phonon-assisted indirect transitions and
structural changes [33]. In this case, the maximum energy in the optical absorption
associated with the tail is the band gap of the tellurite glass.
Tellurite glass supports particularly high concentrations of REI without tending
to form precipitates. Nevertheless, the tendency to exhibit concentration quenching
varies considerably with others forms of the glass.
These properties allow diverse technology applications, for instance, second
harmonic generation [65, 66], parametric oscillation, high gain as a good medium
for high-power level laser and four-wave mixing, and white light generation
[67, 68]. Furthermore, we can obtain in tellurite glasses the ion–ion interactions
(due to multipolar interactions between neighboring) which have been studied for
some time; the first detailed treatment being that due to Forster [69] and Dexter
[70]. Other process also can be found in tellurite glasses, such as co-operative
upconversion (enabling the tuning colors) [71–73], energy migration, cross relax-
ation, and excited state absorption (ESA).
On the other hand, the incorporation of metallic nanoparticles or the interface of
metallic nanostructure in/with REI-doped tellurite glasses are interesting methods
to enhance their optical properties [74, 75]. There are promising methods to
improve the optical properties of REI-doped tellurite glasses for diverse applica-
tions such as amplifiers, solid-state lasers, sensors, etc.
In recent years, REI-doped tellurite glass is being potentially utilized as a
spectral converter to improve the photovoltaic conversion of c-Si solar cells
[33, 76].

3.7 Conclusion

This chapter present results in which it was demonstrated that simultaneous with
exploitation over the last decades of tellurite glasses doped with rare-earth ions,
considerable progress has been made on obtaining efficient luminescent composites
from REI-doped tellurite glasses compatible with optical communication
54 V.A.G. Rivera and L.A.O. Nunes

technology, optical displays, lasers and optical memory devices, including biomed-
ical imaging and energy conversion.
The success of new applications of REI-doped tellurite glasses depends on
improvement in the understanding of the properties of ion–ion interactions,
co-operative up-conversion, energy migration, cross relaxation, excited state
absorption and the incorporation of metallic nanoparticles or the interface of
metallic nanostructure in/with REI-doped tellurite glasses. Further efforts and
systematic studies must be realized in order to offer new developments to optimize
and extend the applications field of REI-doped tellurite glasses into areas such as
nanophotonic applications [74].
We hope that this chapter stimulated our readers to further the development of
theoretical and experimental work on tellurite glasses and rare earths ions.

Acknowledgement This work was financially supported by the Brazilian agencies FAPESP,
CNPq and CEPOF/INOF. V.A.G. Rivera extends thanks FAPESP for financial support (project
2009/08978-4 and 2011/21293-0) that allowed my post-doctoral and my gratitude to Dr. Luiz
Antonio Nunes of the Instituto de Fı́sica de S~ao Carlos—University S~ao Paulo—Brazil and the
Dr. Yannick Ledemi and the Dr. Younnes Messaddeq of the Centre d’Optique, Photonique et
laser—University Laval- Canada for the discussions on this issue.

References

1. J. Sugar, Phys. Rev. Lett. 14, 731 (1965)

2. J.F. Kielkopf, H.M. Crosswhite, J. Opt. Soc. Am. 60, 347 (1970)
3. H.M. Crosswhite, H.W. Moos, Conf Opt Prop Ions in Crys (Wiley, New York, 1967)
4. B.R. Judd, Operator Techniques in Atomic Spectroscopy (McGraw-Hill, New York, 1963)
5. R.C. Powell, Symmetry, Group Theory and the Physical Properties of Crystals (Springer, New
York, 2010)
6. J.J. Sakurai, Modern Quantum Mechanics (Addison Wesley, Boston, 1994). isbn:ISBN
0-201-53929-2
7. A.R. Edmonds, Angular Momentum in Quantum Mechanics (Princeton University Press,
Princenton, NJ, 1960)
8. C.W. Nielson, G.F. Koster, Spectroscopic Coefficients for pn, dn, and fn Configurations (MIT
Press, Cambridge, MA, 1964)
9. G.H. Dieke, Spectra and Energy Levels of Rare-Earth Ions in Crystals (Wiley, New York,
1968)
10. J.H. van Vleck, J. Phys. Chem. 41, 67 (1937)
11. A.J. Freeman, R.E. Watson, Phys. Rev. 127, 2058 (1962)
12. V.A.G. Rivera, F.A. Ferri, E. Marega Jr., in Plasmonics - Principles and Applications, ed. by
K. Y. Kim. Localized Surface Plasmon Resonances: Noble Metal Nanoparticle Interaction
with Rare-Earth Ions (InTech, Croatia, 2012). doi:10.5772/50753
13. S. Hufner, Optical Spectra or Transparent Rare Earth Compounds (Academic Press, London,
1978)
14. G. Liu, B. Jacquier (eds.), Spectroscopic Properties of Rare Earths in Optical Materials
(Springer, Berlin, 2005)
15. O.L. Malta, S.J.L. Ribeiro, M. Faucher, P. Porcher, Theoretical intensities of 4f-4f transitions
between stark levels of the Eu3+ ion in crystals. J. Phys. Chem. Solids 52, 587–593 (1991)
3 Trivalent Lanthanides in Tellurite Glass 55

16. G.H. Dieke, Spectra and Energy Levels of Rare Earth Ions in Crystals (Interscience Pub-
lishers, New York, 1968)
17. B.M. Walsh, Advances in Spectroscopy for Lasers and Sensing, ed. by B Di Bartolo and
O. Forte (Springer, Berlin, 2006)
18. X. Chen, L. Liu, G. Liu, J. Nanosci. Nanotechnol. 8, 1126 (2008)
19. W. Demtrder, Laser Spectroscopy: Basic Concepts and Instrumentation, 3rd edn. (Springer,
New Delhi, India, 2004)
20. G. Rajan (ed.), Optical Fiber Sensors: Advanced Techniques and Applications (CRC Press,
Boca Raton, 2015)
21. S. W. Harun, H. Arof (eds.), Current Developments in Optical Fiber Technology (InTech,
Croatia, 2013). isbn:ISBN 978–953–51-1148-1
22. P.J. Mears, L. Reekie, I.M. Jauncey, D.N. Payne, Electron. Lett. 23, 1026 (1987)
23. E. Desurvire, R.J. Simpson, P.C. Becker, Opt. Lett. 12, 888 (1987)
24. M. Dejneka, B. Samson, MRS Bull. 24, 39 (1999)
25. E. Desurvire, The golden age of optical fiber amplifiers. Phys. Today 47, 20–27 (1994)
26. E. Desurvire, Erbium-Doped Fiber Amplifier (John Wiley, New York, 1994)
27. S. Sudo, Optical Fiber Amplifiers: Materials, Devices, and Applications (Artech House,
Boston, 1997)
28. M.J.F. Digonnet, Rare Earth Doped Fiber Lasers and Amplifiers (Marcel Dekker, New York,
1993)
29. V. Kokorina, Glasses for Infrared Optics (CRC Press, Boca Raton, FL, 1996)
30. M. Yamane, Y. Asahara, Glasses for Photonics (Cambridge University Press, Cambridge,
2000.) ISBN 0521580536
31. J.S. Wang, E.M. Vogel, E. Snitzer, Tellurite glass: a new candidate for fiber devices. Opt.
Mater. 3, 187–203 (1994)
32. M. Shubert, B. Wilhelmi, Nonlinear Optics and Quantum Electronics (Wiley, New York,
1986)
33. V.A.G. Rivera, Y. Ledemi, M. Pereira-da-Silva, Y. Messaddeq, E. Marega Jr., Sci. Rep. 6,
18464 (2016)
34. A. Mori, Tellurite-based fibers and their applications to optical communication networks.
Nippon Seramikkusu Kyokai Gakujutsu Ronbunshi/J Cera Soc Jpn 116, 1040–1051 (2008)
35. R. Morea, J. Fernandez, R. Balda, J. Gonzalo, Pulsed laser deposition of rare-earth-doped
glasses: a step toward lightwave circuits. Proc. SPIE 9744, 974402 (2016)
36. V.A.G. Rivera, D. Manzani, Y. Messaddeq, L.A.O. Nunes, E. Marega Jr., Study of Er3+
fluorescence on tellurite glasses containing Ag nanoparticles. J. Phys. Conf. Ser. 274, 12123
(2011)
37. R. El-Mallawanya, Amitava Patraa, Christopher S. Frienda, R. Kapoora, P.N. Prasad, Study of
luminescence properties of Er3+-ions in new tellurite glasses, Opt. Mat. 26, 267–270 (2004)
38. V.A.G. Rivera, L.C. Barbosa, Spectroscopic properties of Er3+-doped sodium-modified
tellurite glasses for use as optical amplifiers at 1540 nm. J. Lumin. 156, 116–123 (2014)
39. R. Rolli, K. Gatterer, M. Wachtier, M. Bettinelli, A. Speghini, D. Ajò, Optical spectroscopy of
lanthanide ions in ZnO-TeO2 glasses. Spectrochimica Acta-Part A: Mol Bio Spect 57,
2009–2017 (2001)
40. Y. Lu, M. Cai, R. Cao, Y. Tian, F. Huang, S. Xu, J. Zhang, Enhanced effect of Er3+ ions on 2.0
and 2.85 μm emission of Ho3+/Yb3+ doped germanate-tellurite glass. Opt. Mat. 60, 252–257
(2016)
41. A. Mori, Y. Ohishi, S. Sudo, Erbium-doped tellurite glass fibre laser and amplifier. Electron.
Lett. 33, 863–864 (1997)
42. E. Snitzer, E.M. Vogel, J.-S. Wang, U.S. Patent 5251062 A, Tellurite glass and fiber amplifier,
Bell Communications Research, Inc.
43. V.V. Ravi Kanth Kumar, A.K. George, J.C. Knight, P.S.J. Russell, Tellurite photonic crystal
fiber. Opt. Express 11, 2641–2645 (2003)
56 V.A.G. Rivera and L.A.O. Nunes

44. A. Jha, B. Richards, G. Jose, T. Teddy-Fernandez, P. Joshi, X. Jiang, J. Lousteau, Rare-earth

ion doped TeO2 and GeO2 glasses as laser materials. Progr. Mater. Sci. 57, 1426–1491 (2012)
45. L. Huang, A. Jha, S. Shen, X. Liu, Broadband emission in Er3+-Tm3+ codoped tellurite fibre.
Opt. Express 12, 2429–2434 (2004)
46. H. Ebendorff-Heidepriem, K. Kuan, M.R. Oermann, K. Knight, T.M. Monro, Extruded
tellurite glass and fibers with low OH content for mid-infrared applications. Opt. Mat. Express
2(432) (2012)
47. N. Mott, Electrons in Glass, Nobel Lecture (Cavendish Laboratory, Cambridge, England,
1977)
48. R.A.H. El-Mallawany, Tellurite Glasses Hand Book-Physical Properties and Data (CRC
Press, Boca Raton, FL, 2002)
49. C.E. Chryssou, F. Di Pasquale, C.W. Pitt, IEEE J. Selec. Topics Quantum Electron. 6,
114 (2000)
50. A.P. Lopez-Barbero, W.A. Arellano-Espinoza, H.L. Fragnito, H.E. Hernandez-Figueroa,
Micro-wave Opt. Tech. Lett. 25, 103 (2000)
51. V.A.G. Rivera, M. El-Amraoui, Y. Ledemi, Y. Messaddeq, E. Marega, Expanding broadband
emission in the near-IR via energy transfer between Er3+–Tm3+ co-doped tellurite-glasses.
J. Lumin. 145, 787–792 (2014)
52. C.E. Chryssou, Opt. Commun. 184, 375 (2000)
53. V.A.G. Rivera, M. El-Amraoui, Y. Ledemi, Y. Messaddeq, E. Marega, Control of the radiative
properties via photon-plasmon interaction in Er3+-Tm3+-codoped tellurite glasses in the near
infrared region. Opt. Exp. 22, 21122–21136 (2014)
54. V.A.G. Rivera, E.F. Chillcce, E. Rodriguez, C.L. Cesar, L.C. Barbosa, Planar waveguides by
ion exchange in Er3+-doped tellurite glass. J. Non-Cryst. Solids 352, 363–367 (2006)
55. S. Marjanovic, J. Toulouse, H. Jain, C. Sandmann, V. Dierolf, A.R. Kortan, N. Kopylov, R.G.
Ahrens, Characterization of new erbium-doped tellurite glasses and fibers. J. Non-Cryst. Solids
322, 311–318 (2003)
56. V.A.G. Rivera, E. Rodriguez, E.F. Chillcce, C.L. Cesar, L.C. Barbosa, Waveguide produced
by fiber on glass method using Er3+-doped tellurite glass. J. Non-Crys. Solids 353, 339–343
(2007)
57. S. Tanabe, in: Shibin Jiang (ed.), Rare-earth doped materials and devices V, Proc. SPIE, 4282,
85 (2001)
58. W. Miniscalco, IEEE J. Lightwave Technol. 9, 234 (1991)
59. P. Hagenmuller (ed.), Inorganic Solid Fluorides, Chemistry and Physics (Academic Press,
New York, 1985)
60. V. Dimitrov, S. Sakka, Electronic oxide polarizability and optical basicity of simple oxides.
J. Appl. Phys. 79, 1736 (1996)
61. G.H. Beall, L.R. Pinckney, Nanophase glass-ceramics. J. Am. Ceramic Soc. 82, 5–16 (1999)
62. J. Wasylak, K. Ozga, I.V. Kityk, J. Kucharski, IR optical limiting in europium and thulium
doped oxide glasses. Infrared Phys. Technol. 45, 253–263 (2004)
63. E. Snoeks, P.J. Kik, A. Polman, Opt. Mater. 5, 159 (1996)
64. R. El-Mallawany, Tellurite Glasses Handbook—Physical Properties and Data (CRC Press,
London, 2001)
65. A. Narazaki, K. Tanaka, K. Hirao, N. Soga, J. Appl. Phys. 85, 2046 (1999)
66. A. Narazaki, K. Tanaka, K. Hirao, N. Soga, Effect of poling temperature on optical second-
harmonic intensity of lithium sodium tellurite glass. J. Am. Ceram. Soc. 81(2735–2737) (1998)
67. K. Damak, E.S. Yousef, C. Rüssel, R. Ma^alej, White light generation from Dy3+ doped tellurite
glass. J. Quant. Spectrosc. Radiat. Transf. 134(55–63) (2014)
68. N.K. Giri, D.K. Rai, S.B. Raia, White light upconversion emissions from Tm3+-Ho3+-Yb3+
codoped tellurite and germanate glasses on excitation with 798 nm radiation. J. Appl. Phys.
104, 113107 (2008)
69. T. Forster, Ann. Phys. 2, 55 (1948)
70. D.L. Dexter, J. Chem. Phys. 21, 836 (1953)
3 Trivalent Lanthanides in Tellurite Glass 57

71. H.T. Amorim, M.V.D. Vermelho, A.S. Gouveia-Neto, F.C. Cassanjes, S.J.L. Ribeiro,
Y. Messaddeq, Red–green–blue upconversion emission and energy-transfer between Tm3+
and Er3+ ions in tellurite glasses excited at 1.064 μm. J. Sol. State Chem. 171, 278–281 (2003)
72. V.A.G. Rivera, Y. Ledemi, M. El-Amraoui, Y. Messaddeq, E. Marega, Green-to-red light
tuning by up-conversion emission via energy transfer in Er3+–Tm3+-codoped germanium–
tellurite glasses. J. Non-Cryst. Solids 392, 45–50 (2014)
73. N. Rakov, G.S. Maciel, M.L. Sundheimer, L. De, A.S.L. Gomes, Y. Messaddeq,
F.C. Cassanjes, G. Poirier, S.J.L. Ribeiro, Blue upconversion enhancement by a factor of
200 in Tm3+-doped tellurite glass by codoping with Nd3+ ions. J. Appl. Physiol. 92, 6337–6339
(2002)
74. V.A.G. Rivera, O.B. Silva, Y. Ledemi, Y. Messaddeq, E. Marega Jr., Collective Plasmon-
Modes in Gain Media: Quantum Emitters and Plasmonics Nanostructures, Springer Brief in
Physics (Springer International Publishing, Cham, 2015). doi:10.1007/978-3-319-09525-7
75. B. Huang, Y. Zhou, F. Yang, L. Wu, Y. Qi, J. Li, The 1.53 μm spectroscopic properties of Er3+/
Ce3+/Yb3+ tri-doped tellurite glasses containing silver nanoparticles. Opt. Mat. 51, 9–17
(2016)
76. X. Zhou, J. Shen, Y. Wang, Z. Feng, R. Wang, L. Li, S. Jiang, X. Luo, An efficient dual-mode
solar spectral modification for c-si solar cells in Tm3+/Yb3+ Codoped Tellurite Glasses.
J. Amer. Soc. Cer. Soc. 99, 2300–2305 (2016)
Chapter 4
Nonoxide Tellurium-Based Glasses

Mathieu Hubert, Catherine Boussard-Plédel, Bruno Bureau,

and Pierre Lucas

4.1 Introduction

Telluride glasses, i.e., nonoxide tellurium-based glasses, are a specific class of

tellurium-based glasses, with peculiar structural features and outstanding proper-
ties, unmatched by other types of glasses, regarding notably optical transmission
and electrical properties. These make them of great interest for several applications,
but the nature of these glasses also brings specific requirements in terms of
preparation.
The first part of this chapter will be dedicated to the structural aspects of
different families of telluride glasses and an overview of their main properties.
After briefly describing the synthesis process of chalcogenide glasses, the second
part of the chapter will focus on the latest developments in three of the most
important applications of these glasses, i.e., mid- and far-infrared spectroscopies
for space exploration and sensing/biosensing, materials for thermoelectricity, and
data storage.
The telluride glasses considered in this chapter are those for which tellurium is
the main constituent and which do not contain oxygen. Glasses for which tellurium
is not the main constituent (in terms of molar percentage), but in which tellurium

M. Hubert (*)
CelSian Glass & Solar, Eindhoven, The Netherlands
e-mail: [email protected]
C. Boussard-Plédel • B. Bureau
Equipe Verres et Céramiques, UMR 6226, Université de Rennes 1, Rennes, France
e-mail: [email protected]; [email protected]
P. Lucas
Department of Materials Science and Engineering, University of Arizona, Tucson, AZ, USA
e-mail: [email protected]

© Springer International Publishing AG 2017 59

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_4
60 M. Hubert et al.

has a significant role on the structure and properties of the glass, such as glasses
from the Te-As-Se system, are also included.
This chapter focuses on the stable telluride glasses, i.e., those for which bulk
samples can be obtained using conventional chalcogenide glass synthesis process
(as described in Sect. 4.3), and telluride glasses used for phase change applications.
These include chalcohalide glasses and glasses based on the Ge-Te and As-Te
systems, with addition of other elements such as selenium (Se), gallium (Ga),
antimony (Sb), or iodine (I). Glasses based on silicium-tellurium compositions
such as Si20Te80 (investigated notably for acousto-optic modulation [1]) and glasses
in the Si-(Ge)-As-Te and Si-P-Te systems, relatively less studied as compared to
other telluride glasses, are not covered. More information on these glasses can be
found in the literature, notably in [1–4].

4.2 Structure and Properties of Telluride Glasses

4.2.1 General Aspects

The chalcogen elements are those located under the oxygen in the periodic table of
the elements: sulfur (S), selenium (Se), and tellurium (Te). These elements are well
known as major components in some glasses, namely, the chalcogenide glasses,
especially for sulfur and selenium, which are good glass formers [3, 5]. On the other
hand, despite its similarity with sulfur and selenium in terms of electronic config-
uration and of structure, the behavior of tellurium is totally different as far as its
ability to form glassy material is concerned [6–8].
Indeed, the three chalcogen elements are characterized by an outer electronic
shell embedding six electrons s2 px2 py1 pz1. Only two of them give rise to chemical
bonds, and, most of the time, chalcogen elements are twofold coordinated in solid-
state materials [9, 10]. In a glass, they form a kind of polymeric floppy network in
which the lone pairs play mainly a stereochemical role. Nevertheless, whereas for
sulfur and selenium the bonds are strongly covalent, the situation is much more
complex for tellurium, which presents a semi-metallic behavior.
These differences are already observable in the crystalline phases of Se and
Te. Although their crystallographic structure is exactly the same, their physical
properties are clearly different. Indeed the crystalline network of both Se and Te is
constituted of mono-dimensional, aligned, and parallel helicoid chains. These
chains are weakly linked to each other by Van der Waals bonds due to dipole-
dipole interactions. For selenium, these weak bonds are easy to break by thermal
agitation, leading to a low melting temperature of Tm ¼ 216  C. Above this melting
point, elementary selenium becomes a viscous liquid which vitrifies easily as a
floppy network constituted of interlaced long covalent chains. In the case of
crystalline Te, the metallic character is much stronger, which yields an electrical
conductivity several orders of magnitude higher than that of Se. The interatomic
4 Nonoxide Tellurium-Based Glasses 61

Fig. 4.1 Sketch of the

crystalline phase structure
of pure tellurium and
photography of a
crystal bulk

Fig. 4.2 Typical transmission range for several families of glass in the infrared range. The insert
shows a typical example of a telluride glass sample

distances show that a π electronic delocalized bonding reinforces the intra-chains as

well as the inter-chains cohesion of Te atoms (see Fig. 4.1). This results in a higher
melting point for tellurium as compared to selenium, i.e., Tm ¼ 422  C, and the
formation of a more fluidic liquid where atomic associations are very weak,
comparable to that of a molten metal. This makes it complicated to vitrify a pure
tellurium melt, unless using extremely fast quenching rates of 1010  C/s (e.g., using
splat cooling) to obtain small chips of pure vitreous Te.
Though pure tellurium glass cannot be obtained for practical applications, a
large number or telluride glass compositions have been developed and present an
interest for numerous applications. One of the main interests of telluride glasses is
their high infrared transparency. Due to the lack of light elements in their structure,
chalcogenide glasses (i.e., glasses based on one of the chalcogen elements, viz.,
sulfur, selenium, or tellurium) possess a wide range of transmission in the mid- and
far-infrared. As illustrated in Fig. 4.2, their transmission range extends much further
than oxide glasses (such as silica) or fluoride glasses, which structures are based on
the lighter elements oxygen and fluorine, respectively.
62 M. Hubert et al.

In particular, tellurium being the heaviest of the chalcogen elements, telluride

glasses can present a transmission range extending up to >25 μm, while that of
sulfide and selenide glasses typically extends up to 11 μm and 16 μm, respectively.
These outstanding optical properties make them of great interest for numerous
applications requiring a large transmission in the infrared. It has to be noted that
the onset of transmission of telluride glasses is close to 2 μm, hence making these
glasses opaque to visible light and showing a dark, metallic aspect, as shown in the
insert of Fig. 4.2.
Their refractive index is relatively high, ranging from approx. 2.5 to 3.5
depending on the glass composition [11, 12]. In addition, some compositions
show low signal losses, making them of interest for spectroscopic applications in
the infrared, as described later in this chapter.
Due to the complexity of their synthesis process and despite their interest for
optical application due to their low phonon character, telluride glasses have only
been developed recently for far-infrared optics. On the other hand, a lot of attention
has been paid to the fast and reversible glass to crystal transformation for optical
storage applications. Indeed, the initial interest for telluride glasses arose from their
electrical properties, first investigated by Northover and Pearson in the early 1960s
[13, 14] and soon identified by Ovshinsky [15, 16] as being promising as phase-
change materials for data storage, due to this fast-phase reversibility. These features
are described in more details in Sects. 4.2.2.5 and 4.4.3.
It should be noted that, as compared to other types of glasses, the thermal
properties of telluride glasses are relatively weak. Indeed, they usually possess
glass transition temperatures (Tg) below 200  C, sometimes as low as below 50  C
for some chalcohalide glasses, and their crystallization temperature is comprised
between 200  C and 350  C, typically, depending on the glass composition [3, 4,
17–19].

4.2.2 Different Families of Telluride Glasses

Telluride glasses can be subdivided into several families, depending on their

composition. These different families possess specific structural features as
described below.

4.2.2.1 Chalcohalide Glasses: TeX Glasses

To fight against the tendency of tellurium to nucleate crystallites, the objective is to

trap the free electrons provided by tellurium itself in the batch. In the middle of the
1980s, first achievements have been obtained by introducing halogen elements such
as X ¼ Cl, Br, and I in the glass composition, to produce the so-called TeX glasses
[17–19]. The halogens are interesting for the following reasons:
4 Nonoxide Tellurium-Based Glasses 63

Fig. 4.3 Structure of the pure tellurium crystalline phase and of the Te3Cl2 phase showing the
grafting of chlorine on the tellurium chains

1. As electronegative elements, they tend to attract the electrons.

2. As heavy elements, they enable to keep transmission far in the mid-infrared.
3. They are onefold coordinated and thus maintain the polymeric network open and
floppy.
To illustrate this situation, Te3Cl2 is a very interesting model [20] (see Fig. 4.3).
This crystalline phase is built up with helix chains where the chlorine atoms are
grafted on the tellurium and tend to cut the interchains π metallic bonding.
Free rotation along the chains is permitted around the σ bonds, inducing a loss of
periodicity and consequently an easy glass-forming ability. Among this glass
family, the Te/I system is surely the most interesting for mid-infrared application
since Te and I are heavy, close neighbor elements in the periodic table. The optical
transmission of these glasses is the widest ever measured in the far-infrared and
extends up to 28 μm. However, due to their very low dimensionality, their thermo-
mechanical properties are very poor, with glass transition temperatures limited to
Tg  80  C for the most stable Te-I glass. Despite their outstanding optical
properties, the limited thermal properties of the chalcohalide glasses limit their
use for commercial applications.

4.2.2.2 Highly Reticulated Telluride Glasses

Te-based glasses in the Ge/Te system have been widely studied. Germanium is a
fourfold element below carbon (C) and silicium (Si) in the periodic table. The
glassy domain for the germanium-tellurium system is located around the GeTe4 and
64 M. Hubert et al.

Fig. 4.4 (left) Structure of the Ge15Te85 glass (Ge in black, Te in brown) and (right) structure of
the Ge20Te73I7 glass with iodine atoms (purple) opening the network and making it more floppy

the Ge15Te85 stoichiometric compositions, and their structure is expected to be built

with GeTe4 tetrahedra bonded by Te-Te dimers, by analogy with GeSe4 and GeS4
glasses [21] (see Fig. 4.4). This reticulated network thus enables to get higher glass
transition temperatures (Tg), which can reach up to about Tg ¼ 140  C in this glassy
system. However, these glasses remain complicated to synthesize and need very
fast quenching, making it complicated to obtain as bulk material.
Addition of gallium (Ga) to chalcogenide glasses can be very interesting for bulk
infrared optic applications. Indeed, previous work on optical storage glasses clearly
indicated that the presence of Ga increases significantly the stabilization of the
amorphous phases [6]. The role played by the Ga atom can be understood in view of
its electronic structure. It possesses three bonding electrons and naturally forms
three covalent bonds. Nevertheless, this electron paired deficient element easily
tends to change toward a tetrahedral, fourfold coordination by capturing lone pair
electrons from its neighbors, such as Te. By trapping the electrons from the
tellurium, the gallium atoms prevent the formation of Te metallic crystallites and
increase the glass-forming process by enhancing the covalent bonding tendency.
The mixture of the three elements Ge, Te, and Ga proved to be a very interesting
combination to obtain glasses stable against crystallization on the one hand and
exhibiting high Tg on the other hand.
An interesting glass composition in the Ga-Ge-Te system is Ga10Ge15Te75,
characterized by a glass transition temperature of Tg ¼ 172  C, a crystallization
temperature of Tx ¼ 285  C and ΔT ¼ 113  C (with ΔT ¼ Tx–Tg) [22]. This ΔT
value is the best criterion to assess the stability of a glass [23]. It is considered that
this gap should be at least higher than 100  C so that the glass can be shaped into
devices by heating above Tg without crystallization (even higher than 120  C for
drawing of fibers). It has to be noted that even if this composition Ga10Ge15Te75
possesses a ΔT > 100  C and is one of the most stable telluride compositions, the
fabrication of optical devices by molding or drawing remains very challenging.
An alternative way to obtain telluride glasses with high Tg consists in introduc-
ing low amounts of iodine to the Ge-Te system. The beneficial role played by iodine
4 Nonoxide Tellurium-Based Glasses 65

is similar to that played in chalcohalide glasses (described above). In order to

achieve a highly reticulated network in the Ge-Te-based glass, no more than
10 mol% of iodine should be introduced. The best known compromise in the Ge-
Te-I system corresponds to the composition Ge20Te73I7 (see structure in Fig. 4.4)
which is characterized by Tg ¼ 150  C, Tx ¼ 274  C, and ΔT ¼ 124  C [24]. By
taking advantage of this larger glass stability (higher ΔT value), it is possible to
obtain optical fibers by drawing this glass above Tg. It has to be emphasized that
iodine is a very difficult element to manage during glass synthesis because of its
high volatility. In particular, control of the actual concentration of iodine in the final
sample is extremely challenging.
Recently, the synthesis of a new family of telluride glasses based on the
combination of the Ge-Te covalent systems with the ionic salt silver iodine (AgI)
has been reported [25, 26], opening the way towards the development of a new
generation of infrared systems. Indeed, as mentioned above, the introduction of the
elementary electronegative iodine into the GeTe4 vitreous stoichiometry enables
trapping the free electrons from tellurium, thereby helping to form more stable
glasses [24].
Moreover, the heavy atomic weight of iodine allows for maintaining the low
phonon character of the glassy matrix, therefore retaining the far-IR transparency
while providing improved rheological properties. However, high-purity synthesis
reproducibility is difficult since elementary iodine is a volatile compound. The
addition of Ag is expected to stabilize the glass. Indeed, Ramesh et al. showed that
silver addition in Ge-Te vitreous systems could improve the glass-forming ability,
as Ag possesses its own substructure constraining the glassy network [27].
A recent structural study confirmed the beneficial role played by Ag and I in
tellurium-based glass compositions (GeTe4-AgI; see Fig. 4.5) [28]. These new
tellurium glasses exhibit several promising properties: a large optical window
extending from 2 μm to 20 μm (on bulk glass) and no observable crystallization
peak on thermal analysis curves (see Fig. 4.5), which is a real challenge for a
tellurium glass. Thanks to their exceptional properties, some tapered optical fibers

Fig. 4.5 Structure of the GeTe4-AgI glasses together with their thermal analysis curves showing
no crystallization peak
66 M. Hubert et al.

have been recently fabricated from these glasses [29], exhibiting the lowest optical
losses ever obtained in telluride glasses and a transparency ranging from 2 to 16 μm,
supplanting the currently in-service optical fibers made from selenium-based
glasses (see Sect. 4.4.1.1).

4.2.2.3 Stabilizing Telluride Glasses by Addition of Selenium

Among the chalcogen elements, selenium is the best glass former. Therefore, the
addition of selenium to a tellurium-based glass composition can be expected to
enhance glass stability. The glass region of interest corresponds to vitreous mate-
rials having low selenium contents, as addition of Se will be detrimental to phonon
absorption and an excess of Se is known to provoke phase separation [30].
Starting from a GeTe4 matrix, the gradual replacement of tellurium by selenium
(up to 5 mol%) leads to an increase in thermal stability, improving the fiber drawing
ability of the glasses.
Particularly, compositions with 3, 4, and 5 mol% of Se added to a GeTe4 glass
present ΔT values of 106  C, 106  C, and 112  C, respectively, while the ΔT value
for the initial GeTe4 glass composition is only 80  C. A decrease in thermal stability
of the glasses is observed for addition of 10 mol% Se (Te70Ge20Se10 composition)
and beyond until the non-mixing zone, as illustrated in Fig. 4.6.
The substitution of Te by 3% of Se in a GeTe4 glass, yielding a glass with a
composition of Te76Ge21Se3, does not affect significantly the IR cutoff. This
composition thus offers a good compromise for any application in the
mid-infrared. Starting from such compositions, optical fibers (both single and
double index) transmitting light in the mid-infrared far enough to detect the CO2
absorption line located around 15–16 μm have been designed [31, 32]. This
achievement has been carried out in the framework of the Darwin mission of the
European Space Agency (ESA), described in more detail in Sect. 4.4.1.1 [33].

Fig. 4.6 Ge-Te-Se ternary diagram showing a non-mixing zone midway between GeTe4 and
GeSe4 and As-Se-Te ternary diagram with the Te2As3Se5 glass composition selected for Fiber
Evanescent Wave Spectroscopy experiments
4 Nonoxide Tellurium-Based Glasses 67

On the other hand, a significant amount of research has been conducted over the
past 15 years on glasses belonging to the Te-As-Se (TAS) system (see Fig. 4.6), for
mid-infrared spectroscopy in the range 2–12 μm [34–36]. In this system, the glass
composition Te2As3Se5 is very stable against crystallization and against any chem-
ical attack due to its strong covalent network. It is possible to shape sophisticated
tapered optical fibers from this glass, especially sensitive for infrared spectroscopy,
as described later in the chapter. However, the high Se content in this glass, as
compared to pure telluride glasses, makes their transparency limited to 12 μm in the
mid-infrared, against 16 μm for the GeTe4-AgI glass fibers mentioned previously.

4.2.2.4 Copper-Containing Telluride Glasses for Electrical

Applications

Beyond its interest for far-infrared optics, tellurium is also an atypical element
because of its semiconducting behavior [37]. Telluride glasses, as amorphous
materials, are poor thermal conductors, and as semiconducting materials, they are
among the glasses showing the highest electrical conductivity. The combination of
both properties makes them interesting materials for applications such as opto-
electrophoretic sensing and thermoelectric materials.
The best initial glasses for such application belong to the Ge-As-Te system as it
is possible to obtain, from this glassy system, glass compositions with very high
tellurium contents, such as the Ge10As15Te75 glass (see structure in Fig. 4.7a),
which possesses electrical conductivity as high as σ ¼ 102 S/m.
In order to improve the electrical conductivity of this glassy system, new
compositions including metallic elements such as copper (Cu) have been developed
[37, 38]. The increase in conductivity obtained with increased copper content can
be seen in Fig. 4.7b. For some telluride glasses, especially those partially stabilized
with addition of Se, more than 30% of Cu can be added. Currently, the most
interesting telluride glass compositions for thermoelectric applications include

Fig. 4.7 (a) Structure of the Ge10As15Te75 initial glass and (b) Evolution of the glass conductivity
with the addition of copper to this glass
68 M. Hubert et al.

As25Te38Se7Cu30, As15Te54Se6Cu25, and As24Te50Se6Cu20, for which the introduc-

tion of copper has increased the electric conductivities by four orders of magnitude,
reaching values close to σ ¼ 102 S/m. In such glasses, Cu clearly acts as a glass
former in the network. Further studies are being carried out to get deeper under-
standing of the structural and electronic role played by Cu in these glasses.

4.2.2.5 Telluride Glasses as Phase-Change Materials

Tellurium-based alloys, especially those in the ternary system Ge-Sb-Te, are the
most appropriate to observe phase transformation under laser beam irradiation
[39–41]. This feature makes them of great interest for data storage, described in
more details in Sect. 4.4.3. Intensive research has been conducted to identify the
most suitable compositions for phase-change materials (PCM), and one the most
interesting alloys belongs to the Ge-Sb-Te system, the reference composition being
Ge22Sb22Te56 (or Ge2Sb2Te5).
Alternative glasses have been identified, e.g., by substituting Sb or germanium
by gallium (Ga) or tin (Sn). An overview of the different telluride materials
developed throughout the years, from the early PCM materials based on Te85Ge15
glasses doped with Sb, S, or P [15, 16] to the more complex alloys developed since,
is shown schematically in Fig. 4.8 [42].
While the structural features of phase-change telluride materials in their amor-
phous state is relatively unknown, the structure of the crystalline phase is much
better understood. The most widespread telluride PCM such as Ge-Te, Ge-Sb-Te
(GST), or X-Sb-Te (with X ¼ In, Ag, Au) materials form rock salt crystalline

Fig. 4.8 Ternary phase diagram depicting different phase-change alloys, their year of discovery
as phase-change alloy, and their use in different optical storage products. Reprinted with permis-
sion from Macmillan Publishers Ltd.: Nature, Ref. [42], Copyright 2007
4 Nonoxide Tellurium-Based Glasses 69

Fig. 4.9 Schematic view of

the GST crystal. Te atoms
form an f.c.c. sublattice,
while Sb, Ge, and vacancies
form the other sublattice.
Reprinted from [43],
Copyright (2006), with
permission from Elsevier

structures [42], as illustrated in Fig. 4.9 [43]. Other systems such as Au-Ge-Sn-Te
or Ag-In-Sb-Te may form cubic crystalline phases [42].

4.3 Synthesis of Telluride Glasses

Regardless of the targeted application, telluride glasses generally require high

levels of purity. In particular, all traces of oxygen should be avoided during their
synthesis, as like for other chalcogenide glasses, the presence of oxygen notably
leads to the formation of oxide bonds, highly detrimental to the infrared transmis-
sion properties of these glasses. To achieve the desired level of purity, the synthesis
of telluride (non-oxide) glasses generally has to be carried out in silica setups under
vacuum, in order to avoid any contamination due to oxygen, hydrogen, or other
oxides. The procedure is quite common to any chalcogenide glass [3] and will only
be briefly covered in this chapter. A typical setup for preparation of telluride glasses
is represented schematically in Fig. 4.10.
The initial pure raw elements are weighted in the adequate proportions and
introduced in a silica tube, subsequently connected to a vacuum and sealed, to
obtain an ampoule under vacuum. This ampoule containing the raw materials is
then placed in a furnace at relatively high temperature (generally below 900  C) for
melting and homogenization, before quenching (time and temperature of mixing
and quenching depend on the glass compositions). After quenching, and as for most
types of glasses produced, an annealing step is necessary to relax the structure and
avoid mechanical stresses in the bulks.
To obtain extremely high glass purities (e.g., for fabrication of fibers), additional
purification of the raw elements employed may be necessary, e.g., by removing the
70 M. Hubert et al.

Fig. 4.10 Typical experimental setup used for preparation of telluride glasses

oxide layer at their surface by heating up under vacuum. Elemental tellurium itself
may be purified by dipping it in hydrogen bromide (HBr) and subsequently cleaning
it with alcohol. Oxygen getters, like aluminum or magnesium, may also be intro-
duced (typically several hundreds of ppm) in the batch to trap residual traces of
oxygen in the mixture. In this case, an additional distillation step is required to
remove the oxide particles (Al2O3 or MgO) from the glasses. Tellurium tetrachlo-
ride (TeCl4) may also be used to capture the remaining traces of hydrogen (gaseous
HCl is formed and must be evacuated from the batch).
It has to be noted that due to the relatively low glass-forming ability of telluride
glasses, or in other words, their relatively high tendency to crystallize during the
synthesis, combined with the limited quenching rate that can be achieved using this
synthesis process, the size of the samples that can be obtained is usually limited to
some grams. Other synthesis methods may also include ultrafast quenching [37],
employed for the most unstable glasses and yielding samples of smaller size, or
mechanosynthesis for the production of powders [44].

4.4 Applications of Telluride Glasses

4.4.1 Fibers for Mid- and Far-Infrared Spectroscopies

4.4.1.1 Far-Infrared Exploration

In recent years, space programs for detecting new extrasolar terrestrial planets
(or exoplanets) have been initiated. One of the main interests of these programs is
to answer the question whether life exists on these exoplanets. So, National
Aeronautics and Space Administration (NASA) and the European Space Agency
(ESA) have launched two important programs, the “Terrestrial Planet Finder” and
“Darwin” programs [45, 46], respectively, with the aim of detecting signs of life,
expressed by the simultaneous presence of water, ozone, and carbon dioxide in the
4 Nonoxide Tellurium-Based Glasses 71

Fig. 4.11 Scheme of nulling principle for the detection of signal from a planet

atmosphere of the planet. These three molecules exhibit absorptions bands located
in the mid- and far-infrared regions at 6.3 μm, 9 μm, and 15 μm, respectively. Their
detection therefore requires use of material exhibiting high transmission in this
spectral range, making telluride glasses candidates of choice for this application.
In order to characterize the surface and atmospheres of the exoplanets and detect
the chemical signatures of life (H2O, O3 and CO2), one of the challenges is to
suppress the light from the parent star in order to detect its much smaller planets. To
achieve this goal, a telescope flotilla operating in the infrared spectral region is
required. Indeed, in the visible spectrum, an Earth-like planet is outshone by its star
by a factor of a billion, while the difference is lower by a few orders of magnitude in
the infrared.
The method used is the nulling interferometry, which consists in creating
destructive interferences to suppress the invasive signal of stars and detect the
very low signal from its planet (Fig. 4.11). Moreover, the wave fronts of light must
be free of distortions. This can only be achieved by using single-mode waveguides.
For the Darwin mission, the ESA defined two operational wavelength ranges:
6–12 μm for the detection of H2O and O3 and 12–20 μm for the detection of CO2.
For the short-wavelength band (6–12 μm), the so-called TAS glass (from the Te-As-
Se glassy system) has been chosen due to its ability to be shaped into fibers and its
transmission window extending from 2 to 12 μm. For the second range, telluride
glasses having an extended transmission window and which can be drawn into
optical fibers have been specifically developed.
Single-Mode TAS Fiber for the 6–12 μm Spectral Range.
In order to elaborate single-mode fibers, two glass compositions with close
refractive indices are needed for the core and the clad. From the Te-As-Se system,
the glass compositions Te20As30Se50 and Te20.5As30Se49.5 have been chosen as
clad glass and core glass, respectively. The refractive index values of selected
glasses in the TAS system, obtained by the minimum of the deviation method on
three glass prisms having three compositions, and the evolution of the refractive
index as function of the mol% of tellurium in the glasses are shown in Fig. 4.12 [47].
72 M. Hubert et al.

Fig. 4.12 Evolution of the refractive index values versus wavelength and versus the percentage of
tellurium in the glassy composition

Fig. 4.13 2D and 3D representations of the far-field intensity distribution of a TAS single-mode
fiber

For the elaboration of the TAS single-mode fibers, the classical rod-in-tube
method combined with the collapsing of the core and the clad under vacuum was
used [48], with cladding tubes prepared by the traditional rotational casting method
(int. diam. 3 mm and ext. diam. 10 mm) and a core glass (diam. 2 mm) obtained by
drawing a 10 mm diam. rod on a fibering tower. The core rod is then inserted in the
clad tube to form a preform which is then drawn into the single-mode fiber with an
ext. diam. of 525 μm and a core diam. of 22 μm.
Figure 4.13 shows the representation 2D and 3D of the far-field intensity
distribution of such TAS fiber, measured with a CO2 laser operating at 10.6 μm
and a 2D-array micro-bolometer IR CCD camera, on a fiber coated with an
absorbing gallium coating to eliminate the propagation of higher-order optical
modes in the cladding (cladding modes). The transmitted signal has only one
peak showing a Gaussian shape. The image is circular, and no large contribution
from higher-order modes is visible, confirming the production of single-mode fibers
with exceptional transmission in the far-infrared region up to 12 μm.
Higher Tellurium-Containing Glasses for the 12–20 μm Spectral Range.
The TAS glasses presented above contain approx. 50 mol% selenium, which
limits their transmission to about 12–13 μm (transmission measured on fiber). In
order to develop fibers with a transmission extended further in the infrared up to
20 μm (to allow for detection of the spectral signature of CO2 at 15 μm),
4 Nonoxide Tellurium-Based Glasses 73

Fig. 4.14 IR transmission spectra of the most stable glass compositions in the Te-Ge-AgI, Te-Ge-
I, Te-Ge-Ga, and Te-Ge-Se systems (thickness of the samples indicated in legend)

chalcogenide glass compositions with higher tellurium content had to be developed,

tellurium being the heaviest chalcogen element. As pure tellurium exhibits a strong
metallic character and its number of free electrons is several orders of magnitude
higher than for Se, it is quite impossible to vitrify even using fast quenching.
As described previously, a way to decrease the number of free electrons is to mix
Te with other elements or compounds such as Ge, Ga, I, AgI or very small amounts
of Se. The transmittance spectra of bulk samples from the compositions showing
the largest known transmission window in the Te-Ge-Ga (TGG), Te-Ge-I (TGI),
Te-Ge-Se (TGSe), and Te-Ge-AgI (TGAgI) glassy systems are displayed in
Fig. 4.14. For all these glasses, a transmission extending to more than 20 μm, and
even 25 μm for TGAgI glasses notably, is obtained.
The Te-Ge-Ga ternary system was the first Te-based glass system presenting
glasses containing 70–80% of Te reported in the literature [22]. The bulk glasses,
and especially the Te75Ge15Ga10 composition, present an exceptional optical trans-
mission window, between 2 and 28 μm. The lowest optical loss calculated from
bulk transmittance is close to 0.6 dB/cm in the 6–20 μm wavelength range
[49]. However, as the maximum ΔT is limited to 115  C, this composition shows
a high tendency to crystallize during fiber drawing process. Indeed, to shape a
tellurium glass into optical fibers with a limited risk of nucleation of metallic Te
nanoparticles, the glass thermal stability criteria ΔT should be higher than 120  C.
Although this condition is met for the most stable glasses in the Te-Ge-I system,
it is quite impossible to consider the elaboration of single-mode fibers from glasses
74 M. Hubert et al.

in this system. Indeed, the slightest deviation in the composition for glasses used in
the core or the clad, and thus variation in their refractive index values, can prevent
the single-mode propagation of the signal within the fiber. The presence of the
volatile iodine in these glasses makes the control of the composition very challeng-
ing and limits their use for such high-precision optical applications.
In order to overcome those issues, more recently, the Te-Ge-Se and Te-Ge-AgI
glass systems have been explored, and selected compositions exhibiting broad
infrared transmissions and superior stabilities (ΔT > 120  C) have been identified
[25, 30, 32, 33, 50]. The introduction of small amounts of Se in GeTe4 has no
significant impact on the transmission but permits to stabilize the glass. As a first
step toward the elaboration of single-mode fibers, multimode step-index fibers were
elaborated by traditional rod-in-tube method. The refractive index is easily modi-
fied by the substitution of Te by Se, and two compositions have been selected for
core and clad composition, namely, Te76Ge21Se3 and Te71Ge21Se8 [31]. Neverthe-
less, this method is not suitable for the elaboration of fibers having very small cores,
since several steps of casting and drawing are required, which are sources of risks of
crystallization. To overcome that issue, fiber elaboration methods based on preform
casting are under investigation.
The introduction of iodine as a salt, silver iodide, in the glasses from the Te-Ge-
AgI system, greatly improves the stability of the glass by stabilizing the iodine
content in the glass. Thanks to a distillation-purification process, single index fibers
with a minimum of attenuation of 3 dB/m at 10 μm have been obtained for the glass
(GeTe4)90(AgI)10, corresponding to the minimum value ever obtained for a pure
telluride fiber reported up to date [29].
The comparison of the attenuation curves for fibers before and after the polishing
process of the glass rod used during the fiber elaboration process is presented in
Fig. 4.15. After polishing, the losses are significantly decreased, showing the
importance of the surface quality of the starting rod.
Figure 4.16 shows the result of carbon dioxide detection in the infrared using
(GeTe4)90(AgI)10 fibers in chamber filled with CO2 [25]. The collected spectrum
shows a clear, broad CO2 absorption bands around 15 μm. This observation
validates the (GeTe4)90(AgI)10 glass fibers as potential candidates for designing
optical waveguide operating until 16 μm as required by the ESA or NASA, for
space application.
These fibers have also shown their ability to be implemented in Fiber Evanescent
Wave Spectroscopy (FEWS) for the detection of molecules having absorptions
bands above 12 μm.
As an example, the absorbance spectra of pure chloroform (2N), recorded for
different sensing zone diameters and compared to the spectrum recorded with a
TAS fiber, are shown in Fig. 4.17. The large transmission range of the fiber is
highlighted by the presence of all the absorption bands of chloroform from 3 to
15 μm. Especially, the strong C-Cl stretching and C-Cl bending absorption peaks
located at 13.3 μm and 14.9 μm, respectively, are well identified. Moreover, as the
fiber diameter at sensing zone decreases, the sensitivity increases significantly due
to the much larger number of internal reflections.
4 Nonoxide Tellurium-Based Glasses 75

Fig. 4.15 The optical losses of (GeTe4)90(AgI)10 single index fibers obtained from the preforms
with (black line) and without (red dotted line) optical polishing

Fig. 4.16 Gaseous CO2 infrared spectrum recorded thanks to a black body source signal trans-
mitted through a (GeTe4)90(AgI)10 glass fiber. The inset shows the (GeTe4)90(AgI)10 glass fiber
76 M. Hubert et al.

Fig. 4.17 Chloroform absorbance spectra recorded by FEWS using a Te-Ge-AgI glass fiber. The
inset is chloroform absorbance registered with a Te20As30Se50 (TAS) fiber for comparison. By
decreasing the sensing zone diameter, the fiber sensitivity could be greatly enhanced

More details on the use of telluride glasses for evanescent wave spectroscopy are
given in the next section.

4.4.1.2 Fiber Evanescent Wave Spectroscopy

Virtually all molecules and biomolecules exhibit highly specific vibrational signa-
tures in the mid-IR domain from 2 to 16 μm. Infrared spectroscopy therefore
provides a highly selective mean of detecting hazardous molecules for environ-
mental monitoring or scrutinizing biomolecules for medical diagnostic. Fourier
transform infrared spectroscopy (FTIR) is indeed one the most well-established
analytical technique for molecular detection. In a conventional spectroscopic
experiment, the sample of interest must be introduced between the source and the
detector, which significantly limits the application of FTIR for in situ monitoring or
requires removal of biological specimen from its environment (biopsy) for medical
applications. Instead, the Fiber Evanescent Wave Spectroscopy (FEWS) technique
permits to send and collect an infrared signal remotely using an infrared-transparent
optical fiber.
In this technique the light from a FTIR source or a tunable IR laser is coupled
into one end of the fiber and collected at the other end on a detector. The sample of
interest interacts with the evanescent wave along a tapered section of the fiber
which acts as the sensing zone (see Fig. 4.18). Along the sensing zone, the light
4 Nonoxide Tellurium-Based Glasses 77

Fig. 4.18 Schematic of a Fiber Evanescent Wave Spectroscopy experiment showing the tapered
section of the infrared fiber which serves as the sensing zone

wave confined in the fiber extends up to ~1 μm above the surface and enables strong
coupling with any substance in direct contact with the fiber [51, 52]. Thinner tapers
lead to increased interactions as depicted in Fig. 4.18 and result in improved
detection limits.
The resulting absorption spectrum is mainly identical to that of a transmission
measurement or attenuated total reflectance (ATR) measurement with only minimal
wavelength dependence. This technique has therefore been broadly used for optical
sensing applications ranging from detection of pollutants in water reservoirs to
monitoring the effect of toxicants on live human cells [53, 54].
This technique is particularly well suited for the study of biological molecules
due to the highly specific vibrational signature located in the fingerprint region
between 5 and 12 μm. As depicted in Fig. 4.19, the rich spectral features of
biological molecules can be associated with the various components of the micro-
organism such as amides, amino acids, nucleic acids, phospholipids, and poly-
saccharides. This vibrational signature permits to establish the identity of
microorganisms and to identify their strains [55–57] or to distinguish between
healthy and diseased tissues [58, 59].
However, effective spectral analysis requires collection of high quality infrared
signals with optimal intensity and minimal noise. In that respect, telluride glasses
are highly desirable due to their excellent transparency in the spectral region of
interest as shown in Fig. 4.19.
The low phonon energy associated with heavy tellurium atoms yields multi-
phonon cut-offs extending beyond 20 μm for bulk glasses [24]. Proper composi-
tional engineering allows for the tuning of the optical window of these materials
while retaining good rheological properties for fiber drawing [21, 25, 60–62]. Good
glass formation properties are necessary to avoid the formation of crystallites which
leads to high scattering losses in fibers. The addition of small amounts of selenium
to the telluride glass provides an effective mean of stabilizing the glass structure
78 M. Hubert et al.

Fig. 4.19 Top: vibrational signature of human lung cells collected using a tapered chalcogenide
fiber sensor. The signal of each cell component can be identified, including A, amide I and II; B,
fatty acids and amino acids; C, phospholipids and nucleic acids; and D, polysaccharides. Bottom:
optical transmission spectrum of a Ge20As20Te45Se15 glass fiber

and realizing low attenuation fibers while retaining sufficiently wide transparency
windows for effective biosensing as depicted in Fig. 4.19 [31, 32].
Indeed, the high quality infrared signals conferred by telluride fibers sensors
allows for the possibility of performing complex statistical spectral treatments such
as principal component analysis [52, 63]. This is of particular interest for food
safety where the identification of particular bacterial strains is required.
Figure 4.20 shows an example of PCA analysis which illustrates the possibility
of differentiating between listeria, salmonella, staphylococcus, and endogenous
flora in a raw cheese sample.
Similar approaches can be applied for medical diagnostics in animals as well as
humans, such as cancerous tissue analysis in rodents [64] or profiling of human
serum for multiple metabolic dysfunctions [65]. This technology has now reached
maturity in the field of medical sensing and is currently being developed
industrially [63].

4.4.1.3 Opto-Electrophoretic Sensing

In addition to their wide infrared transparency, telluride glasses possess unusually

high electrical conductivity that can reach many orders of magnitude above that of
4 Nonoxide Tellurium-Based Glasses 79

0.3
0.2
0.1
0.0

PC4
–0.1
–0.2
–0.3
0.4
–0.4 0.2
–0.5 0.0
–0.4
–0.2 –0.2

3
0.0

PC
0.2 –0.4
E. coli PC 0.4
2 –0.6
S. aureus 0.6

Fig. 4.20 Selective detection of bacterial spores using spectral analysis (PCA) in telluride glass
sensors based on (a) Fiber Evanescent Wave Spectroscopy, (b) ATR opto-electrophoretic sensor

Fig. 4.21 A telluride ATR element and a schematic drawing of the electrophoretic sensors used
for collection and analysis of charged biological molecules

other chalcogenide glasses [66]. This has led to the possibility of using these
materials as both optical sensors and electrodes for the electrophoretic capture
and simultaneous optical detection of biomolecules. Indeed, most biomolecules
and microorganisms such as proteins, bacterial spores, and viruses carry a net
surface charge that can be utilized to force their migration under an electric field.
This has enabled the development of opto-electrophoretic sensors that permit to
enrich the optical sensing zone with a biomolecule of interest for its detection and
identification [21, 66, 67]. The sensor is composed of a telluride ATR base that
serves as the positive electrode and an indium tin oxide (ITO) counter electrode that
is negatively charged to drive the migration of biomolecules as depicted in
Fig. 4.21.
Under the application of modest voltages, the biomolecules undergo a rapid
downward motion and get electrodeposited on the surface of the ATR element,
where the infrared evanescent wave reaches high intensities. The resulting spectral
feature is depicted in Fig. 4.22a for the example of a bovine serum albumin
solution.
The characteristic amide I and II signals at 1550 cm1 and 1650 cm1, respec-
tively, show a rapid increase as soon as the voltage is applied, thereby indicating
80 M. Hubert et al.

Fig. 4.22 (a) Amide I and II signal of a BSA samples electrodeposited on a Ge-As-Te ATR
element. (b) Increased voltage showing greater electrodeposition kinetic

Fig. 4.23 Selective PCA identification of E. coli and Staphylococcus bacterial spores using a
telluride ATR opto-electrophoretic sensor

migration of the BSA molecules. Moreover, an increase of the applied voltage

results in faster deposition kinetics is obtained, as shown in Fig. 4.22b. It was also
shown that reversing the voltage permits to partially remove the molecules from the
surface of the ATR base, therefore demonstrating the reversible nature of the
process [67].
The high intensity signal resulting from the migration of the target molecules
within the evanescent wave enables effective PCA analysis for the identification of
hazardous species such as bacteria [21, 67]. Figure 4.22b and Fig. 4.23 show the
selective PCA identification of E. coli and Staphylococcus aureus in different
spectral regions. The lower signal-to-noise ratio observed for E. coli in Fig. 4.23
results in much wider scatter in the PCA plot and emphasizes the sensitivity of the
technique on high quality optical signals. Compositional engineering of the tellu-
ride glass permits to obtain an optimal trade-off between electrical properties and
high infrared transparency for maximizing the signal intensity.
4 Nonoxide Tellurium-Based Glasses 81

Fig. 4.24 Schematic of electron migration under application of a thermal gradient in a thermo-
electric material

4.4.2 Thermoelectricity

Thermoelectric materials are a category of solids that exhibits migration of charge

carriers (e- or holes) under application of a temperature gradient ΔT, as illustrated
in Fig. 4.24. This allows to directly converting heat into electricity or alternatively
converting electricity into cooling. These materials have raised renewed interest
due to their potential as sustainable energy sources but also due to the lack of
moving parts which makes thermoelectric devices highly reliable for sensitive
applications such as spacecraft generators.
The efficiency of a thermoelectric material is quantified by a dimensionless
figure of merit ZT which is defined as ZT ¼ S2σT/(κ L þ κE) where S is the
thermopower or Seebeck coefficient in (V/K ), σ is the electrical conductivity in
(S/m), κL and κE are the lattice and electronic component of the thermal conduc-
tivity in (W/mK), and T is the temperature in (K). The Seebeck coefficient S
quantifies the extent of charge carriers migration in response to a temperature
gradient and is defined as the resulting voltage V divided by the temperature
gradient ΔT (S ¼ V/ΔT ).
For near-room-temperature applications, such as refrigeration and waste heat
recovery up to 200  C, the best commercially available materials are Bi2Te3 alloys
with a figure of merit of ZT ¼ 1. Most of the recent gains in ZT have been obtained
by nano-structuring crystalline materials in an effort to reduce κ L while maintaining
high electrical conductivity.
This approach has been referred to as the “phonon glass, electron crystal
(PGEC)” approach for materials that can efficiently scatter phonons like glasses
(and lower κL) while maintaining crystal-like conductivity and maximize the power
factor S2σ, thereby effectively raising ZT. So far this approach has led to Bi-Sb-Te
materials with a ZT ¼ 1.2 at 300 K [68].
An alternative approach to the PGEC paradigm involves taking advantage of the
intrinsically low thermal conductivity κ L of glassy materials and engineering their
electrical properties to maximize the power factor S2σ. This method has recently
received much attention and has potential to produce materials with ZT > 2 (see
Fig. 4.25) [37, 38, 69–71].
Telluride glasses have particularly low thermal conductivity because of the
disorder of their structure as well as a low phonon energy resulting from heavy
atoms like Te. The disordered structure induces scattering of heat-carrying phonons
[72], while the high atomic weight of the tellurium reduces the speed of sound and
thereby reduces the thermal conductivity [73].
82 M. Hubert et al.

Fig. 4.25 Evolution of the Seebeck coefficient with conductivity for CuGeTe and CuAsSeTe
glasses with increasing Cu content. The ZT values shown are calculated from a κ L value ~101 W/
Km and room temperature (T ¼ 300 K)

Typical thermal conductivities for telluride glasses are in the range ~ 101 W/
Km (0.1 W/Km for Ge-Te [38], 0.12 W/Km for Ge-Te-Se [74], 0.35 W/Km for
Pb-Ge-Se [75], 0.29 for As-Te-Se). This represents some of the lowest conductiv-
ities known for solids. This range of conductivity is about ten times lower than
conventional thermoelectric materials, thereby raising the possibility of obtaining
high ZT from these telluride glass compositions.
Glassy materials typically have low electrical conductivities, which leads to low
power factor and low ZT. However, it has been shown that alloying of telluride
glasses with transition metals such as Cu leads to dramatic increases in electrical
conductivity by up to 8 orders of magnitudes [37, 38]. Fig. 4.25 shows the evolution
of the Seebeck coefficient as a function of conductivity for a series of CuGeTe and
CuAsTe glasses. The predicted values of ZT considering a glassy thermal conduc-
tivity κ L ~ 0.1 W/ Km are indicated on the graph and show that some outstanding
values near ZT ~ 2 could be obtained for these glasses despite an expected decrease
in Seebeck down to S ¼ 200300.
Active research is ongoing to increase the conductivity of telluride glasses
within the 104–105 S/m range. Many strategies are under consideration, including
doping with various transition metals, modifying the glassy host matrix, and partial
crystallization to improve mobility [69].
4 Nonoxide Tellurium-Based Glasses 83

4.4.3 Data Storage

Given the significance of the applications of telluride glasses for data storage, a
considerable amount of literature can be found on this topic. Comprehensive
reviews on phase-change telluride materials and on data storage strategies based
on them can notably be found in [42, 76–84]. This section will only provide an
overview on the importance of data storage based on phase-change materials
(PCM) and the role of telluride glasses in these applications.
With the constantly and exponentially growing amounts of data generated in our
modern society, the need for reliable and durable data storage solutions is con-
stantly increasing. Different data storage technologies exist. Non-volatile memories
(NVM), which do not require an external power supply to retain the stored
information, can be distinguished from volatile memories, for which the informa-
tion is lost when the power source is disconnected. Among NVM, several
approaches can be found, depending on the type of materials employed and the
way the information is stored.
As of today, NVM devices are dominated by silicon-based memories, for which
the digital information is stored as electric charges on MOSFET (metal-oxide-
semiconductor field-effect transistor), i.e., stored as bits “0” and “1.” According
to the so-called Moore’s law, memory densities will double with each smaller
generation, every 2 years. While improvement of the technologies have allowed
for a downscaling of the memory devices while maintaining an increase in their
capacities, size limits may be reached in a near future, due to limitations of the
materials themselves.
An alternative to these transistor-type memories relies on the use of phase-
change materials, for which the digital information is stored as different states of
the material itself, exhibiting significantly different properties. The information is
stored in a binary form, i.e., bits “0” and “1,” represented by the amorphous and a
crystalline state of the material, respectively. A controlled process, typically via
laser, electrical or electromagnetic pulses, allows for switching from one state to the
other (amorphous to crystalline or crystalline to amorphous), depending on the
intensity and the duration of the pulse.
As illustrated in Fig. 4.26 for the case of a thin film of PCM deposited as active
layer on a DVD, the initial glassy material is heated by a laser beam to reach the
crystallization temperature Tc which has to be very close to the glass transition Tg.
This initialization phase makes the DVD ready to be recorded by a more powerful
laser beam which allows to reach locally the melting point Tm of the alloys and then
to produce a local glassy spot. Then, the measurement is based on the contrast of
reflectivity between the glassy spot and the crystallized phase.
In order to qualify as good PCM material for NVM devices, the transition
between the two states (i.e., between amorphous and crystalline state in the case
of chalcogenide PCM) must be controllable, reversible, and occur at very short time
scales to allow for fast data writing. Ultrafast crystallization rates are thus required.
84 M. Hubert et al.

Fig. 4.26 Principle of the phase change process from glass to crystal

In the meantime, industrial standard generally requires the memory devices to

permit storage of the data over periods of more than 10 years, at a temperature of
100  C. These boundary conditions imply that the material, and especially the
amorphous, “metastable” state, should remain stable and not be subject to sponta-
neous crystallization over at least this required period. In other words, the base glass
should have a low stability allowing for very fast crystallization, but in the mean-
time the stability should not be too low as to avoid spontaneous, undesired crystal-
lization over time. The combination of these two properties makes the development
of PCM materials very challenging, and only a few types of materials known to date
fulfill these requirements.
Tellurium-based chalcogenide glasses are the most promising material for such
PCM-based NVM devices and have had a significant place in the development of
this technology. Notably, compositions from the Ge-Sb-Te system have been
playing a key role in the development of NVM devices. The interest for these
tellurium-based glasses arises from the large difference in the optical and electrical
properties of the material when in glassy or in crystalline form. While the crystal-
line form has low electrical conductivity and low reflectivity, the corresponding
amorphous phase shows much larger values. This difference is exploited to store the
information as bits “0” (glassy phase) and “1” (amorphous phase).
Laser-induced switching of a telluride glass with composition Te81Ge15Sb2S2
was first demonstrated in 1971 by Feinleib et al. [39], based on the pioneering work
on telluride alloys electrical properties by Northover et al. [13, 14] and threshold
4 Nonoxide Tellurium-Based Glasses 85

switching behavior by Ovshinsky [15, 16]. Further investigations, notably by

Yamada et al. [41, 85], identified pseudo-binary amorphous GeTe-Sb2Te3 alloy
films as excellent candidates for PCM. These materials were first used commer-
cially in 1990. Ge-Sb-Te telluride PCM have since been widely used for large-scale
commercial products such as DVD-RAM and Blu-ray disks, while Sb-Te-based
materials have been exploited commercially for CD-RW, DVD-RW, and Blu-ray
disks, as illustrated in Fig. 4.8 [42]. New types of tellurium-based materials showing
promising potential as PCM for data storage continue to be investigated, such as
titanium-doped Sb2Te3 glasses (Ti-Sb-Te system) [86].
For production of memories based on PCM, scaling is an important aspect to
consider. The scaling of PC materials depends on several factors, including crys-
tallization and amorphization time, melting temperature, thermal stability of the
amorphous phase, threshold voltage, and smallest thickness/unit size at which
phase change may occur. This latter point is very important. Indeed, with reducing
the scale of the devices, the intrinsic behavior of the material itself may be
dependent on its size [77, 81].
It has to be noted that evolution in data storage efficiency does not only depend
on the PCM material itself. Advances in data storage are also correlated to advances
in the data writing technology. This is notably illustrated by the evolution of the
storage capacity of disks. With evolution of the laser technology and lens systems
used for engraving, the recording density and capacity have constantly increased,
from CD-RW to DVD-RW to Blu-ray disks, as shown in Fig. 4.27 [42].
The development of new phase-change materials and understanding of their
properties and behavior are now more and more relying on simulation tools. The
accuracy and power of these simulation tools are constantly increasing, and a
considerable effort is put in constantly improving them. These tools are probably
the key for further progress in understanding PCM and developing new, faster,
smaller, and even more reliable materials. Several examples of the use of simula-
tion for phase-change materials can be found in recent literature [87–91].

Fig. 4.27 Comparison of optics, recording densities, recording capacities, and disk structures
used for CD, DVD, and BD. Changes in the lens system are shown in the upper portion; the disk
structure is depicted beneath. The change in beam cross section is displayed at the bottom.
Reprinted with permission from Macmillan Publishers Ltd.: Nature, Ref. [42], Copyright 2007
86 M. Hubert et al.

4.5 Summary

Telluride glasses, i.e., oxygen-free tellurium-based glasses, exhibit outstanding

optical and electrical properties, making them materials of choice for several highly
technical applications. First developed in the 1960s, these glasses rapidly triggered
a high interest for their capacity of being used as phase-change materials for data
storage. A considerable amount of research has been made in this field, notably on
glasses from the Ge-Sb-Te system, particularly appropriate for this application.
These glasses have been used for massive commercial applications such as
CR-RW, DVD-RW, and Blu-ray, emphasizing the utmost importance of these
materials.
Despite their exceptional transparency in the mid- and far-infrared up to
>20–25 μm (depending on the composition), one of the main limitations in the
use of telluride glasses resides in their relatively low thermal stability. This brings
significant challenges in the production of bulk materials for optics. Only recently
have telluride glasses been developed presenting a thermal stability high enough to
allow production of optical fibers without crystallization. This achievement,
obtained through development of new compositions stabilized by small additions
of selenium to the composition notably, has opened the way to several new fields of
applications. Such highly infrared-transmitting fibers are now being used for
applications such as sensing/biosensing by Fiber Wave Evanescent Spectroscopy
(FEWS) and are promising candidates for space exploration.
The recent development of bulk telluride glasses containing high amounts of
copper, improving further their electrical properties, has led to an increasing
interest for these materials for thermoelectric applications, which are of particular
importance in terms of development of sustainable energy sources.
The different features of telluride glasses presented in this chapter emphasized
their crucial role in our society. Their impact is already significant, notably in terms
of their utilization for data storage, and is expecting to grow in view of their
development for spectroscopic and energetic applications, for which active research
is undergoing.

References

1. L.A. Kulakova, B.A. Matveev, B.T. Melekh, Si–Te acousto-optic modulator for the
1.7–10.6 μm IR region. J. Non-Cryst. Solids 266–269(Part 2), 969–972 (2000)
2. L.A. Kulakova, V.K. Kudoyarova, B.T. Melekh, V.I. Bakharev, Synthesis and physical
properties of Si(Ge)–Se–Te glasses. J. Non Cryst. Solids 352(9–20), 1555–1559 (2006)
3. M.A. Popescu, Non-Crystalline Chalcogenides (Springer, Netherlands, 2001)
4. R.A.H. El-Mallawany, Tellurite Glasses Handbook-Physical Properties and Data (CRC Press,
Boca Raton, 2011)
5. V.F. Kokorina, Glasses for Infrared Optics (Taylor & Francis, 1996)
6. B. Bureau, S. Danto, H.L. Ma, C. Boussard-Plédel, X.H. Zhang, J. Lucas, Tellurium based
glasses a ruthless glass to crystal competition. Solid State Sci. 10, 427–433 (2008)
4 Nonoxide Tellurium-Based Glasses 87

7. B. Bureau, C. Boussard-Pledel, P. Lucas, X.-H. Zhang, J. Lucas, Forming glasses from Se and
Te. Molecules 14(11), 4337–4350 (2009)
8. X.H. Zhang, B. Bureau, P. Lucas, C. Boussard-Pledel, J. Lucas, Glasses for seeing beyond
visible. Chemistry (Weinheim an der Bergstrasse, Germany) 14(2), 432–442 (2008)
9. J.A. Savage, Glass formation and D.S.C. data in the Ge-Te and As-Te memory glass systems.
J. Non Cryst. Solids 11(2), 121–130 (1972)
10. C. Adenis, V. Langer, O. Lindqvist, Reinvestigation of the structure of tellurium. Acta
Crystallogr. C 45(6), 941–942 (1989)
11. A.R. Hilton, Optical properties of chalcogenide glasses. J. Non Cryst. Solids 2, 28–39 (1970)
12. A.R. Hilton, S. Kemp, Chalcogenide Glasses for Infrared Optics (McGraw-Hill, New York,
2010)
13. W.R. Northover, A.D. Pearson, U.S. Patent 3,117,013 A—Glass composition, (1964)
14. J.F. Dewald, W.R. Northover, A.D. Pearson, U.S. Patent 3,241,009 A—Multiple resistance
semiconductor elements, 1966
15. S.R. Ovshinsky, U.S. Patent 3,271,591 A—Symmetrical current controlling device, 1966
16. S.R. Ovshinsky, Reversible electrical switching phenomena in disordered structures. Phys.
Rev. Lett. 21(20), 1450–1453 (1968)
17. J. Lucas, X.H. Zhang, Les verres de chalcohalogenes, une nouvelle famille de materiaux pour
l’infrarouge. Les verres des systemes Te-Cl et Te-Cl-S. Mater. Res. Bull. 21(7), 871–876
(1986)
18. J. Lucas, X.H. Zhang, The tellurium halide glasses. J. Non Cryst. Solids 125(1–2), 1–16 (1990)
19. X.H. Zhang, G. Fonteneau, J. Lucas, The tellurium bromide glasses: new I.R. transmitting
materials. Mater. Res. Bull. 23(1), 59–64 (1988)
20. R. Kniep, D. Mootz, A. Rabenau, Strukturuntersuchungen an niederen Tellurhalogeniden:
Modifizierte Tellur-Strukturen. Angew. Chem. 85(11), 504–505 (1973)
21. P. Lucas, Z. Yang, M.K. Fah, T. Luo, S. Jiang, C. Boussard-Pledel, M.-L. Anne, B. Bureau,
Telluride glasses for far infrared photonic applications. Opt. Mater. Express 3(8), 1049–1058
(2013)
22. S. Danto, P. Houizot, C. Boussard-Pledel, X.H. Zhang, F. Smektala, J. Lucas, A family of far
infrared transmitting glasses in the Ga-Ge-Te system for space optics. Adv. Funct. Mater. 16,
1847 (2006)
23. B. Frumarová, M. Frumar, J. Málek, New halide glasses [the system CdCl2-PbCl2-KI-PbO].
Mater. Res. Bull. 29(10), 1035–1040 (1994)
24. A.A. Wilhelm, C. Boussard-Pledel, Q. Coulombier, J. Lucas, B. Bureau, P. Lucas, Develop-
ment of far-infrared-transmitting Te based glasses suitable for carbon dioxide detection and
space optics. Adv. Mater. 19, 3796–3800 (2007)
25. C. Conseil, J.-C. Bastien, C. Boussard-Pledel, X.-H. Zhang, P. Lucas, S. Dai, J. Lucas,
B. Bureau, Te-based chalcohalide glasses for far-infrared optical fiber. Opt. Mater. Express
2(11), 1470–1477 (2012)
26. X. Wang, Q. Nie, G. Wang, J. Sun, B. Song, S. Dai, X. Zhang, B. Bureau, C. Boussard,
C. Conseil, H. Ma, Investigations of Ge–Te–AgI chalcogenide glass for far-infrared applica-
tion. Spectrochim. Acta A 86, 586–589 (2012)
27. K. Ramesh, S. Asokan, K.S. Sangunni, E.S.R. Gopal, Glass formation in germanium telluride
glasses containing metallic additives. J. Phys. Chem. Solid 61(1), 95–101 (1999)
28. P. Jóvári, S. Cui, V. Nazabal, I. Kaban, B. Beuneu, M. Dussauze, C. Boussard-Plédel,
B. Bureau, Network rearrangement in AgI-doped GeTe4 glasses. J. Am. Ceram. Soc. 98(3),
1034–1039 (2015)
29. S. Cui, C. Boussard-Pledel, J. Lucas, B. Bureau, Te-based glass fiber for far-infrared biochem-
ical sensing up to 16 μm. Opt. Express 22(18), 21253–21262 (2014)
30. S. Maurugeon, B. Bureau, C. Boussard-Plédel, A.J. Faber, X.H. Zhang, W. Geliesen, J. Lucas,
Te-rich Ge–Te–Se glass for the CO2 infrared detection at 15 μm. J. Non Cryst. Solids 355
(37–42), 2074–2078 (2009)
88 M. Hubert et al.

31. S. Maurugeon, B. Bureau, C. Boussard-Plédel, J. Troles, A.J. Faber, P. Lucas, X.H. Zhang,
J. Lucas, Telluride glass step index fiber for the far infrared. J. Lightwave Technol. 28,
3358–3363 (2010)
32. S. Maurugeon, B. Bureau, C. Boussard-Plédel, A.J. Faber, P. Lucas, X.H. Zhang, J. Lucas,
Selenium modified GeTe4 based glasses optical fibers for far-infrared sensing. Opt. Mater. 33,
660–663 (2011)
33. C. Conseil, V.S. Shiryaev, S. Cui, C. Boussard-Pledel, J. Troles, A.P. Velmuzhov,
A.M. Potapov, A.I. Suchkov, M.F. Churbanov, B. Bureau, Preparation of high purity
Te-Rich Ge-Te-Se fibers for 5–15 μm infrared range. J. Lightwave Technol. 31(11),
1703–1707 (2013)
34. D. Le Coq, K. Michel, J. Keirsse, C. Boussard-Pledel, G. Fonteneau, B. Bureau, J.-M.
Le Quéré, O. Sire, J. Lucas, Infrared glass fibers for in-situ sensing, chemical and biochemical
reactions. C. R. Chim. 5(12), 907–913 (2002)
35. B. Bureau, X.H. Zhang, F. Smektala, J.-L. Adam, J. Troles, H.-L. Ma, C. Boussard-Pledel,
J. Lucas, P. Lucas, D. Le Coq, M.R. Riley, J.H. Simmons, Recent advances in chalcogenide
glasses. J. Non Cryst. Solids 345–346, 276–283 (2004)
36. G. Delaizir, M. Dussauze, V. Nazabal, P. Lecante, M. Dolle, P. Rozier, E.I. Kamitsos,
P. Jovari, B. Bureau, Structural characterizations of As-Se-Te glasses. J. Alloys Compd. 509
(3), 831–836 (2011)
37. P. Lucas, C. Conseil, Z. Yang, Q. Hao, S. Cui, C. Boussard-Pledel, B. Bureau, F. Gascoin,
C. Caillaud, O. Gulbiten, T. Guizouarn, P. Baruah, Q. Li, J. Lucas, Thermoelectric bulk glasses
based on the Cu-As-Te-Se system. J. Mater. Chem. A 1(31), 8917–8925 (2013)
38. A. Pereira Goncalves, E. Branco Lopes, O. Rouleau, C. Godart, Conducting glasses as new
potential thermoelectric materials: the Cu-Ge-Te case. J. Mater. Chem. 20(8), 1516–1521
(2010)
39. J. Feinleib, J. deNeufville, S.C. Moss, S.R. Ovshinsky, Rapid reversible light-induced crystal-
lization of amorphous semiconductors. Appl. Phys. Lett. 18, 254–257 (1971)
40. G.-F. Zhou, Materials aspects in phase change optical recording. Mater. Sci. Eng. A 304–306,
73–80 (2001)
41. N. Yamada, E. Ohno, K. Nishiuchi, N. Akahira, M. Takao, Rapid-phase transitions of GeTe-
Sb2Te3 pseudobinary amorphous thin films for an optical disk memory. J. Appl. Phys. 69(5),
2849–2856 (1991)
42. M. Wuttig, N. Yamada, Phase-change materials for rewriteable data storage. Nat. Mater. 6(11),
824–832 (2007)
43. A.L. Lacaita, Phase change memories: state-of-the-art, challenges and perspectives. Solid
State Electron. 50(1), 24–31 (2006)
44. M. Hubert, E. Petracovschi, X.-H. Zhang, L. Calvez, Synthesis of germanium–gallium–
tellurium (Ge–Ga–Te) ceramics by ball-milling and sintering. J. Am. Ceram. Soc. 96(5),
1444–1449 (2013)
45. A. Léger, Strategies for remote detection of life—DARWIN-IRSI and TPF missions. Adv.
Space Res. 25(11), 2209–2223 (2000)
46. A. Léger, J.M. Mariotti, B. Mennesson, M. Ollivier, J.L. Puget, D. Rouan, J. Schneider, Could
we search for primitive life on extrasolar planets in the near future? Icarus 123(2), 249–255
(1996)
47. C. Conseil, Q. Coulombier, C. Boussard-Plédel, J. Troles, L. Brilland, G. Renversez,
D. Mechin, B. Bureau, J.L. Adam, J. Lucas, Chalcogenide step index and microstructured
single mode fibers. J. Non Cryst. Solids 357(11–13), 2480–2483 (2011)
48. P. Houizot, C. Boussard-Plédel, A.J. Faber, L.K. Cheng, B. Bureau, P.A. Van Nijnatten,
W.L.M. Gielesen, J. Pereira do Carmo, J. Lucas, Infrared single mode chalcogenide glass
fiber for space. Opt. Express 15(19), 12529–12538 (2007)
49. S. Zhang, Z. X-h, M. Barillot, L. Calvez, C. Boussard, B. Bureau, J. Lucas, V. Kirschner,
G. Parent, Purification of Te75Ga10Ge15 glass for far infrared transmitting optics for space
application. Opt. Mater. 32(9), 1055–1059 (2010)
4 Nonoxide Tellurium-Based Glasses 89

50. V.S. Shiryaev, A.P. Velmuzhov, M.F. Churbanov, A.D. Plekhovich, C. Boussard-Plédel,
J. Troles, C. Conseil, V.G. Plotnichenko, Preparation and investigation of high purity Ge–
Te–AgI glasses for optical application. J. Non Cryst. Solids 377, 1–7 (2013)
51. P. Lucas, M.R. Riley, C. Boussard-Pledel, B. Bureau, Advances in chalcogenide fiber evanes-
cent wave biochemical sensing. Anal. Biochem. 351, 1–10 (2006)
52. P. Lucas, B. Bureau, in Advanced infrared glasses for biochemical sensing, eds. by
C.M. Pradier, Y.J. Chabal, Biointerface Characterization, (Advanced IR Spectroscopy,
Elsevier, 2011) p. 217–243
53. P. Lucas, D. LeCoq, C. Juncker, J. Collier, D.E. Boesewetter, C. Boussard-Pledel, B. Bureau,
M.R. Riley, Evaluation of toxic agent effects on lung cells by fiber evanescent wave spectros-
copy. Appl. Spectrosc. 59, 1–9 (2005)
54. H. Steiner, M. Jakusch, M. Kraft, M. Karlowatz, T. Baumann, R. Niessner, W. Konz,
A. Brandenburg, K. Michel, C. Boussard-Pledel, B. Bureau, J. Lucas, Y. Reichlin, A. Katzir,
N. Fleischmann, K. Staubmann, R. Allabashi, J.M. Bayona, B. Mizaikoff, In situ sensing of
volatile organic compounds in groundwater: First field tests of a mid-infrared fiber-optic
sensing system. Appl. Spectrosc. 57, 607–613 (2003)
55. D. Naumann, Microbiological characterizations by FT-IR spectroscopy. Nature 351, 81 (1991)
56. M. Diem, S. Boydston-White, L. Chiriboga, Infrared spectroscopy of cells and tissues: shining
light onto a novel subject. Appl. Spectrosc. 53(4), 148A–161A (1999)
57. D. Naumann, in Infrared spectroscopy in microbiology, ed. by R.A. Meyers. Encyclopedia of
Analytical Chemistry, (John Wiley & Sons Ltd, 2000) p. 102
58. M. Diem, N. Laver, K. Bedrossian, J. Schubert, K. Papamarkakis, B. Bird, M. Miljkovic, in
Detection of Viral Infection in Epithelial Cells by Infrared Spectral Cytopathology, ed. by
Jurgen Popp, Handbook of Biophotonics, vol. 2, (Wiley-VCH, 2011) p. 251–258
59. M. Diem, K. Papamarkakis, J. Schubert, B. Bird, M.J. Romeo, M. Miljkovic, The infrared
spectral signatures of disease: extracting the distinguishing spectral features between normal
and diseased states. Appl. Spectrosc. 63(11), 307A–318A (2009)
60. P. Lucas, C. Boussard-Pledel, A. Wilhelm, S. Danto, X.-H. Zhang, P. Houizot, S. Maurugeon,
C. Conseil, B. Bureau, The development of advanced optical fibers for long-wave infrared
transmission. Fibers 1, 110–118 (2013)
61. Z. Yang, O. Gulbiten, P. Lucas, T. Luo, S. Jiang, Long-wave infrared-transmitting optical
fibers. J. Am. Ceram. Soc. 94(6), 1761–1765 (2011)
62. Z. Yang, T. Luo, S. Jiang, J. Geng, P. Lucas, Single-mode low-loss optical fibers for long-wave
infrared transmission. Opt. Lett. 35, 3360–3362 (2010)
63. B. Bureau, C. Boussard, S. Cui, R. Chahal, M.L. Anne, V. Nazabal, O. Sire, O. Loreal,
P. Lucas, V. Monbet, J.-L. Doualan, P. Camy, H. Tariel, F. Charpentier, L. Quetel, J.-L.
Adam, J. Lucas, Chalcogenide optical fibers for mid-infrared sensing. Opt. Eng. 53(2),
027101/027101–027101/027107 (2014)
64. S. Hocde, O. Loreal, O. Sire, C. Boussard-Pledel, B. Bureau, B. Turlin, J. Keirsse, P. Leroyer,
J. Lucas, Metabolic imaging of tissues by infrared fiber-optic spectroscopy: an efficient tool for
medical diagnosis. J. Biomed. Opt. 9(2), 404–407 (2004)
65. M.L. Anne, C. Le Lan, V. Monbet, C. Boussard-Pledel, M. Ropert, O. Sire, M. Pouchard,
C. Jard, J. Lucas, J.L. Adam, P. Brissot, B. Bureau, O. Loreal, Fiber evanescent wave
spectroscopy using the mid-infrared provides useful fingerprints for metabolic profiling in
humans. J. Biomed. Opt. 14(5), 054033 (2009)
66. Z. Yang, A.A. Wilhelm, P. Lucas, High-conductivity tellurium-based infrared transmitting
glasses and their suitability for bio-optical detection. J. Am. Ceram. Soc. 93(7), 1941–1944
(2010)
67. Z. Yang, M.K. Fah, K.A. Reynolds, J.D. Sexton, M.R. Riley, M.-L. Anne, B. Bureau, P. Lucas,
Opto-electrophoretic detection of bio-molecules using conducting chalcogenide glass sensors.
Opt. Express 18, 26754–26759 (2010)
90 M. Hubert et al.

68. B. Poudel, Q. Hao, Y. Ma, Y. Lan, A. Minnich, B. Yu, X. Yan, D. Wang, A. Muto, D. Vashaee,
X. Chen, J. Liu, M.S. Dresselhaus, G. Chen, Z. Ren, High-Thermoelectric Performance of
Nanostructured Bismuth Antimony Telluride Bulk Alloys. Science 320, 634–638 (2008)
69. J.B. Vaney, G. Delaizir, E. Alleno, O. Rouleau, A. Piarristeguy, J. Monnier, C. Godart,
M. Ribes, R. Escalier, A. Pradel, A.P. Goncalves, E.B. Lopes, G.J. Cuello, P. Ziolkowski,
E. Mueller, C. Candolfi, A. Dauscher, B. Lenoir, A comprehensive study of the crystallization
of Cu-As-Te glasses: microstructure and thermoelectric properties. J. Mater. Chem. A 1(28),
8190–8200 (2013)
70. A.P. Goncalves, E.B. Lopes, G. Delaizir, J.B. Vaney, B. Lenoir, A. Piarristeguy, A. Pradel,
J. Monnier, P. Ochin, C. Godart, Semiconducting glasses. A new class of thermoelectric
materials? J. Solid State Chem. 193, 26–30 (2012)
71. J.B. Vaney, A. Piarristeguy, A. Pradel, E. Alleno, B. Lenoir, C. Candolfi, A. Dauscher,
A.P. Goncalves, E.B. Lopes, G. Delaizir, J. Monnier, M. Ribes, C. Godart, Thermal stability
and thermoelectric properties of CuxAs40-xTe60-ySey semiconducting glasses. J. Solid State
Chem. 203, 212–217 (2013)
72. D.G. Cahill, R.O. Pohl, Lattice Vibration and Heat Transport in Crystalls and Glasses. Annu.
Rev. Phys. Chem. 39, 93–121 (1998)
73. A. Majumdar, Thermoelectricity in Semiconductor Nanostructures. Science 303, 777–778
(2004)
74. S.N. Zhang, J. He, T.J. Zhu, X.B. Zhao, T.M. Tritt, Thermal conductivity and specific heat of
bulk amorphous chalcogenides Ge20Te8-xSex (x ¼ 0,1,2,8). J. Non Cryst. Solids 355, 79–83
(2009)
75. J. Philip, R. Rajesh, C. Preethy Menon, Carrier-type reversal in Pb–Ge–Se glasses:
Photopyroelectric measurements of thermal conductivity and heat capacity. Appl. Phys.
Lett. 78, 745–747 (2001)
76. M. Wuttig, Phase-change materials: Towards a universal memory? Nat. Mater. 4(4), 265–266
(2005)
77. S. Raoux, F. Xiong, M. Wuttig, E. Pop, Phase change materials and phase change memory.
MRS Bull. 39(08), 703–710 (2014)
78. Phase Change Materials—Science and Applications (2009). Springer US
79. G.W. Burr, M.J. Breitwisch, M. Franceschini, D. Garetto, K. Gopalakrishnan, B. Jackson,
B. Kurdi, C. Lam, L.A. Lastras, A. Padilla, B. Rajendran, S. Raoux, R.S. Shenoy, Phase change
memory technology. J. Vac. Sci. Technol. B 28(2), 223–262 (2010)
80. S. Raoux, Phase Change Materials. Annu. Rev. Mat. Res. 39(1), 25–48 (2009)
81. S. Raoux, G.W. Burr, M.J. Breitwisch, C.T. Rettner, Y.C. Chen, R.M. Shelby, M. Salinga,
D. Krebs, S.H. Chen, H.L. Lung, C.H. Lam, Phase-change random access memory: A scalable
technology. IBM J. Res. Dev. 52(4), 465–479 (2008)
82. S. Raoux, T.J. Ibm, in Phase change memory (PCM) Materials and Devices. ed. by Y. Nishi.
Advances in Non-Volatile Memory and Storage Technology, (Woodhead Publishing, Amster-
dam, 2014) p. 161–199
83. S. Raoux, W. Wełnic, D. Ielmini, Phase Change Materials and Their Application to Nonvol-
atile Memories. Chem. Rev. 110(1), 240–267 (2010)
84. W. Wełnic, M. Wuttig, Reversible switching in phase-change materials. Mater. Today 11(6),
20–27 (2008)
85. N. Yamada, E. Ohno, N. Akahira, N. Ki, N. Ki, M. Takao, High Speed Overwritable Phase
Change Optical Disk Material. Jpn. J. Appl. Phys. 26(S4), 61 (1987)
86. M. Xia, M. Zhu, Y. Wang, Z. Song, F. Rao, L. Wu, Y. Cheng, S. Song, Ti–Sb–Te alloy: A
candidate for fast and long-life phase-change memory. ACS Appl. Mater. Interfaces 7(14),
7627–7634 (2015)
87. Molecular Dynamics Simulations of Disordered Materials—From Network Glasses to Phase-
Change Memory Alloys (2015), vol 2015. Springer Series in Materials Science. Springer
International Publishing
4 Nonoxide Tellurium-Based Glasses 91

88. D. Ielmini, in Phase Change Memory Device Modeling, eds. by S. Raoux, M. Wuttig, Phase
Change Materials, (Springer, New York, 2009) p. 299–329
89. S.R. Elliott, Chalcogenide Phase-Change Materials: Past and Future. Int. J. Appl. Glass Sci. 6
(1), 15–18 (2015)
90. M. Boniardi, D. Ielmini, I. Tortorelli, A. Redaelli, A. Pirovano, M. Allegra, M. Magistretti,
C. Bresolin, D. Erbetta, A. Modelli, E. Varesi, F. Pellizzer, A.L. Lacaita, R. Bez, Impact of
Ge–Sb–Te compound engineering on the set operation performance in phase-change memo-
ries. Solid State Electron. 58(1), 11–16 (2011)
91. S. Senkader, C.D. Wright, Models for phase-change of Ge2Sb2Te5 in optical and electrical
memory devices. J. Appl. Phys. 95(2), 504–511 (2004)
Chapter 5
Photonic Applications of Tellurite Glasses

Luiz Carlos Barbosa, Cicero Omegna Filho, and Enver Fernandez Chillcce

5.1 Introduction

Tellurite glasses have been studied since the 1990s, when researchers found that
these materials could have great potential for applications in the field of photonics,
due to properties such as high rates of linear and nonlinear refraction, medium-
frequency phonons, high solubility to ions of rare earths and low melting points.
E. Vogel from Bell Labs first demonstrated applications of these glasses in the form
of fiber as optical amplifiers achieving high gain using the rare earths Erbium and
praseodymium. Later, Oishi from NTT (Japan), demonstrated fiber amplifiers and
amplifiers with high gain. Applications in the field of photonics achieved break-
throughs combining Oishi’s work with mixed microstructured optical fiber tellurite
glasses as supercontinuum fiber devices. Microstructured optical fibers such as
glass quantum communication devices were demonstrated. Micro lenses for appli-
cations in telecommunications and medicine for endoscopy were also developed.
Amplifiers in the 1550 nm range have been manufactured by the ion exchange
method for use with both planar and channel lasers.

L.C. Barbosa (*) • E.F. Chillcce

Departamento de Eletrônica Qu^antica, Instituto de Fı́sica Gleb Wataghin, UNICAMP –
Universidade Estadual de Campinas, Rua Sérgio Buarque de Holanda, 777 Cidade
Universitária, 13083-859 Campinas, SP, Brazil
e-mail: [email protected]; [email protected]
C.O. Filho
Luxtec Optical Systems, Rua Dr. Souza Brito, 126/483, 13040-012 Campinas, SP, Brazil
e-mail: [email protected]

© Springer International Publishing AG 2017 93

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_5
94 L.C. Barbosa et al.

5.2 Photonic Applications of Tellurite Glasses

Tellurite glasses present high linear and nonlinear refractive indexes, so they lend
themselves to many new applications and the development of new products in the
photonics field. The first report on tellurium dioxide glass (TeO2), or tellurite glass
was made in 1956 by G. Barady [1]. His research showed that stable tellurite-based
glasses could be obtained by adding a small amount of modifiers such as Li2O into
fused tellurium-dioxide (TeO2) [2]. The nonlinear refractive index (n2) of
Tellurite glass was reported by A. Traore [3], using a new measurement
technique called Induced Grading Autocorrelation (IGA) with a result of
n2 ¼ ð4:48  0:25 Þ1015 cm W
2 . This n2 value is ten times more than that of silica

glass, n2 ¼ ð2:49  0:07 Þ1016 cm W

2 , and indicated tellurite glasses as potential

candidates for the implementation of optical switches.

Highly nonlinear optical materials show a variety of frequency conversion
mechanisms with a wide range of applications to photonics, information, medical,
industrial, and military technologies. V. A. G. Rivera et al. [4] reported on the
interaction between 2F5/2 ! 2F7/2 radiative transition from Yb3þ ions and localized
surface Plasmon resonance (from gold/silver nanoparticles) in a tungsten-tellurite
glass. Plasmonics has played a revolutionary role in the quantum–photon interac-
tion and opened a wide range of practical applications that involve the manipulation
of light on the nanoscale.
M. Liao et al. [5] showed supercontinuum (SC) generation in highly nonlinear
tellurite microstructured fibers pumped by a continuous wave (cw)/quasi-cw laser.
For fibers with a constant core diameter, when pumped in the anomalous dispersion
region, the SC is symmetric in a fiber that has a zero dispersion wavelength close to
the pump wavelength. For fibers with a longitudinally varying diameter, the
calculated phase-matching conditions show that they have a broad wavelength
range of dispersive waves, and therefore the measured SC spectrum can be broader
than one octave.
A broadband near-infrared emission in Pr3þ single doped and Pr3þ/Yb3þ
codoped tellurite–tungstate glasses for optical amplification was reported by
M. P. Belançon et al. [6]. In the Pr3þ single doped samples pumped at 474 nm,
the emission band at 1480 nm presents a full-width at half maximum (FWHM) of
~140 nm. The Yb3þ addition produces a FWHM broadening up to ~155 nm,
achieved through the 1G4!3H5 transition (Fig. 5.1a). The emission spectra when
the codoped sample is pumped at 980 nm shows efficient energy transfer from Yb3þ
to Pr3þ, resulting in an intense Pr3þ emission around 1330 nm. This shows that by
changing the pump wavelength it is possible to select the 1D2!1G4 or the 1G4!3H5
transition, displacing the emission band from ~1480 to ~1330 nm (Fig. 5.1b).
Thanks to the high polarizability of Te4þ ions and the high transparency from
visible wavelengths up to the mid-IR range, TeO2-based glasses exhibit the
required characteristics for the development of efficient diamagnetic magneto-
optic fiber components. Y. Shiyu et al. [7] have reported Verdet constant measure-
ments in Tellurite fibers. The Verdet constant of tellurite fiber was found to be
5 Photonic Applications of Tellurite Glasses 95

Fig. 5.1 (a) Emission spectra of Pr3þ doped and Pr3þ/Yb3þ codoped tellurite–tungstate glasses
under 474 nm pumping and (b) schematic energy diagrams for the Pr3þ/Yb3þ codoped sample
containing the main energy-transfer process involved in the NIR emission [6]. (From
M.P. Belançon, J.D. Marconi, M.F. Ando and L.C. Barbosa, Opt. Mater. 36, 1020–1026 (2014))

28.1  0.5 rad (T m)1. This result suggests that tellurite fibers can be used for
magneto-optical sensing and monitoring applications.

5.3 Photonics Field Application of Tellurite Glasses

5.3.1 Nanowires

Photonic nanowires were fabricated by L. Tong et al. using a technique that drew
directly from bulk glasses instead of from glass fibers [8]. The produced nanowires
are shown in Fig. 5.2. When manufactured using the direct draw technique, the
nanowires not only show low optical losses, but also offer favorable properties such
96 L.C. Barbosa et al.

Fig. 5.2 Electron microscopic characterizations of as-drawn glass nanowires. (a) SEM image of a
100-nm diameter tellurite glass nanowire. (b) SEM image of an elastically bent 320-nm diameter
silicate glass nanowire. (c) SEM image of the cross section of a 400-nm diameter tellurite glass
nanowire. (d) SEM image of a spiral plastic bend of an 80-nm diameter phosphate glass nanowire.
(e) SEM image of a 170-nm diameter tellurite glass nanowire with sharp plastic bends. (f) TEM
examination of the sidewall of a 210-nm diameter phosphate glass nanowire. (From L. Tong,
L. Hu, J. Zhang, J. Qiu, Q. Yang, J. Lou, Y. Shen, J. He and Z. Ye, Opt. Express 14(1), 82–87
(2006) [8])

as high index for tight optical confinement and photoluminescence for active
devices, making them very promising building blocks for future micro- and nano-
scale photonic circuits and devices.

5.3.2 Planar and Channel Waveguide

Planar waveguides were produced by an Agþ!Naþion exchange technique in Er3þ

-doped tellurite glass with a composition of 75TeO2–2GeO2–10Na2O–12ZnO–
1Er2O3 (mol%) [9]. A tellurite channel waveguide (see Fig. 5.3) was fabricated
by using Agþ!Naþ ion exchange and photolithography as demonstrate by V. A.
G. Rivera [10].
Channels waveguides were fabricated in tellurite glass using laser-written pro-
cedures by P. Nandi et al. [11]. Figure 5.4a shows the top view of three channels in a
typical glass sample photographed using an optical microscope. Fig. 5.4b shows the
waveguides facets. Fig. 5.4c shows a mode-profile of the guided 1310 nm laser
beam written through the channel laser. Fig.5.4d shows the 3-D intensity profile of
this channel.
5 Photonic Applications of Tellurite Glasses 97

Fig. 5.3 Channel waveguide printed on Tellurite glass by using Photolithography. (From
V.A.G. Rivera, M.Sc. thesis, Unicamp University, Brazil, 150 (2005) [10])

Fig. 5.4 (a) Laser-written channels observed through an optical microscope at writing speeds of
0.01 cm/s (top channel), 0.02 cm/s (middle channel), and 0.03 cm/s (bottom channel). Channels as
long as 1 cm were written. (b) Cross-sectional side images of the waveguides shown in (a). (c)
Mode-profile image of channel-transmitted 1310 nm laser light. (d) Intensity profile of the channel
output depicted in (c). Horizontal lines in (a, b, and c) denote the indicated scale lengths. (From
P. Nandi, G. Jose, C. Jayakrishnan, S. Debbarma, K. Chalapathi, K. Alti, A. K. Dharmadhikari,
J. A. Dharmadhikari, D. Mathur, Opt. Express 14(25), 12145–12150 (2006) [11])
98 L.C. Barbosa et al.

Fig. 5.5 (a) Plain-view SEM micrograph showing the ‘FOG’ waveguide. (b) Cross-section
micro-photograph of the waveguide fabricated using the FOG method. (From V.A.G. Rivera,
E. Rodriguez, E.F. Chillcce, C.L. Cesar, L.C. Barbosa, J. Non-Crystal. Solids 353, 339–343
(2007))

5.3.3 Fiber on Glass Waveguide

Taking advantage of a Thermal Mechanical Analyzer (TMA), a new type of

waveguide was produced by. V.A.G. Rivera by coupling an erbium doped fiber
core onto a planar glass substrate [12]. Both optical fiber core and substrate were
fabricated from tellurite glass. This method is named fiber on glass (FOG) and
resulted in a new photonic device. Fig. 5.5a shows a plain-view of a Scanning
Electron Microscope (SEM) micrograph of the fabricated waveguide device.
Arrows in the figure indicate the fiber core and substrate, respectively. Fig. 5.5b
shows a cross-section micro-photograph of the same device.

5.3.4 Localized Temperature Sensing

E. P. Schartner et al. [13] reported the development of a point temperature sensor,

based on monitoring upconversion emission from erbium:ytterbium-doped tellurite
coatings on the tips of optical fibers. The dip coating technique allowed multiple
sensors to be fabricated simultaneously, while confining the temperature-sensitive
region to a localized region on the end-face of the fiber. The strong response of the
rare earth ions to changing temperature allows a resolution of 0.1–0.3  C to be
recorded over the biologically relevant range of temperatures from 23–39  C.
Sensors were fabricated from commercially available multi-mode fiber, with a
core diameter of 62.5 μm and an outer diameter of 125 μm. For temperature sensing
measurements, the glass host material chosen was sodium zinc tellurite glass, doped
with 1 mol% erbium and 9 mol% ytterbium. The coating method used to fabricate
the sensing region on the fiber tips is shown in Fig. 5.6. Tellurite glass was used for
5 Photonic Applications of Tellurite Glasses 99

Optical fibres

Immerse 10 secs Cool and

Tip remove

Rare Earth doped molten glass

Fig. 5.6 Fiber coating method, in which the fiber tip is briefly immersed within the molten glass to
form a thin layer of active temperature-sensitive tellurite glass on the end face of the silica fiber.
(From E.P. Schartner, T.M. Monro, Sensors 14, 21693–21701 (2014) [13])

the fabrication of the tips, as the melt temperature used here (850  C) is significantly
below the softening point of the silica glass fibers (1600  C), so no deformation of
the fibers is observed.

References

1. G.W. Brady, X-Ray study of tellurium oxide glass. J. Chem. Phys. 24, 477 (1956)
2. G.W. Brady, Structure of tellurium oxide glass. J. Chem.Phys. 27, 300 (1957)
3. A. Traore, Measurement of the nonlinear refractive index of Tellurite glass Fiber by using
induced grating autocorrelation technique”, Ph.D. thesis, University of Maryland, 117, 2011
4. V.A.G. Rivera, Y. Ledemi, M.A. Pereira-da-Silva, Y. Messaddeq, E. Marega Jr., Plasmon-
photon conversion to near-infrared emission from Yb3þ:(Au/Ag-nanoparticles) in tungsten
tellurite glasses. Sci. Rep. 6, 18464 (2016)
5. M. Liao, W. Gao, Z. Duan, X. Yan, T. Suzuki, Y. Ohishi, Supercontinuum generation in short
tellurite microstructured fibers pumped by a quasi-cw laser. Opt. Lett. 37(11), 2127–2129
(2012)
6. M.P. Belançon, J.D. Marconi, M.F. Ando, L.C. Barbosa, Near-IR emission in Pr3þsingle
doped and tunable near-IR emission in Pr3þ/Yb3þ codoped tellurite tungstate glasses for
broadband optical amplifiers. Opt. Mater. 36, 1020–1026 (2014)
7. Y. Shiyu, J. Lousteau, M. Olivero, M. Merlo, N. Boetti, S. Abrate, Q. Chen, Q. Chen,
D. Milanese, Analysis of Faraday effect in multimode tellurite glass optical fiber for
magneto-optical sensing and monitoring applications. Appl. Opt. 51(19), 4542–4546 (2012)
8. L. Tong, L. Hu, J. Zhang, J. Qiu, Q. Yang, J. Lou, Y. Shen, J. He, Z. Ye, Photonic nanowires
directly drawn from bulk glasses. Opt. Express 14(1), 82–87 (2006)
9. V.A.G. Rivera, E.F. Chillcce, E. Rodriguez, C.L. Cesar, L.C. Barbosa, Planar waveguides by
ion exchange in Er3þ-doped tellurite glass. J. Non-Cryst. Solids 352, 363–367 (2006)
10. V.A.G. Rivera, Fabricaç~ao e Caracterizaç~ao de Guias de Onda paraAmplificadores Ópticos
Curtos com Vidros Germano-Teluritos, M.Sc. thesis, Unicamp University, Brazil, 150, 2005
100 L.C. Barbosa et al.

11. P. Nandi, G. Jose, C. Jayakrishnan, S. Debbarma, K. Chalapathi, K. Alti, A.K. Dharmadhikari,

J.A. Dharmadhikari, D. Mathur, Femtosecond laser written channel waveguides in tellurite
glass. Opt. Express 14(25), 12145–12150 (2006)
12. V.A.G. Rivera, E. Rodriguez, E.F. Chillcce, C.L. Cesar, L.C. Barbosa, Waveguide produced
by fiber on glass method using Er3þ-doped tellurite glass. J. Non-Cryst. Solids 353, 339–343
(2007)
13. E.P. Schartner, T.M. Monro, Fibre tip sensors for localised temperature sensing based on rare
earth-doped glass coatings. Sensors 14, 21693–21701 (2014)
Chapter 6
Lasers Utilising Tellurite Glass-Based Gain
Media

Billy D.O. Richards and Animesh Jha

6.1 Introduction

One of the alluring properties of tellurite glass is its relatively wide infrared
transmission window, being transparent up to around 5–6 μm due to its relatively
low maximum phonon energy which may vary between 650 cm
1 and 800 cm
1,
depending on specific composition [1]. Figure 6.1 shows an example of the absorp-
tion coefficient and transmission (in inset) spectra for a 4.4 mm thick barium-
lanthanum-tellurite glass sample. The spectra have not been corrected for the ~10%
Fresnel reflection losses at both glass-air interfaces, occurring due to the relatively
high refractive index of tellurite glass which is typically in the range 1.8–2.3. The
transmission spectrum in the Fig. 6.1 inset has a maximum value of 60% transmis-
sion at a wavelength of 2.7 μm and drops to half that value at a wavelength of
5.9 μm. The absorption peaks at 3.3 μm and 4.4 μm are due to hydroxyl (OH
) ion
impurities [2] in the glass network which may originate predominantly from two
sources: water in the raw materials and glass-melting environment. These funda-
mental OH
absorption bands obscure the mid-infrared transparency of the glass,
and their overtone absorption bands in the 1.5 μm range overlap with the tail of the
Er3þ: 4I15/2!4I13/2 ground state absorption and may thereby facilitate a resonant
energy transfer from the 4I13/2 state to neighbouring OH
ions, resulting in radiation
quenching via strong molecular absorption. Hydroxyl ion contamination in tellurite
glass may be reduced by introducing fluorides into the batch, bubbling the melt with
reactive gases and melting in a pure and dry oxygen atmosphere [3, 4].
Another property of tellurite glass, which makes it an interesting host material
with which to study solid-state lasers, is its ability to dissolve relatively high
concentrations of rare-earth ions (compared to silica) without clustering, resulting

B.D.O. Richards (*) • A. Jha

School of Chemical and Process Engineering, University of Leeds, Leeds LS2 9JT, UK
e-mail: [email protected]; [email protected]

© Springer International Publishing AG 2017 101

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_6
102 B.D.O. Richards and A. Jha

Fig. 6.1 Absorption coefficient spectrum and (inset) transmission spectrum of a 4.4 mm thick
barium-lanthanum-tellurite glass sample. The spectra have not been corrected for reflection losses

in the potential to realise very compact laser devices [5]. The high nonlinearity of
tellurite glass has also made it a popular choice for the demonstration of
supercontinuum generation, with broadband laser emission spanning several
micrometres wavelength range. The high refractive index of tellurite glasses (n ffi
1.8–2.3 [6]) means that they can offer high stimulated emission cross section (S),
expressed by S / (n2þ2)/9n, over a broad emission bandwidth [7].
Lasers are characterised by certain basic parameters which describe their effi-
ciency such as the laser threshold and the laser slope efficiency. In order for laser
action to be realised, the gain must be higher than the round-trip cavity losses at the
laser wavelength. If the losses are higher than the gain, then the intensity of the light
will be reduced by the losses more than it is increased by the amplification medium,
and there will be no overall gain. At threshold, the gain is equal to the round-trip
cavity losses of the laser which means that for a given system, the laser threshold
will be higher for lower values of output coupler reflectivity, for example, Fresnel
reflection. For a three-level laser system, the absorbed pump power required to
reach threshold Pth is given by Eq. (6.1)

α þ δ0 1
Pth ¼ ð6:1Þ
2 κξth

where α is the unsaturated signal GSA term, δ0 is the round-trip laser cavity loss,
κ is the gain term and ξth is a function of pump power and takes into account the
ground-state depletion and pump excited-state absorption (ESA) effects and is
equal to unity if these effects are absent. For a four-level laser, there is no signal
ground-state reabsorption; the α term in Eq. (6.1) can be ignored [8]. However, the
6 Lasers Utilising Tellurite Glass-Based Gain Media 103

output power Pout and slope efficiency s of the laser are proportional to the
transmission coefficient of the output coupler, so although a higher transmission
output coupler results in higher cavity losses, it can potentially lead to higher slope
efficiencies. The output power and slope efficiency of a three- or four-level laser
which has a quantum efficiency of 100% and is not influenced by ESA processes are
given by Eqs. (6.2) and (6.3), respectively,

T 1 hνs
Pout ¼ ðPabs
Pth Þ ð6:2Þ
δ0 hυp
T 1 hυs
s¼ ð6:3Þ
δ0 hυp

where T1 is the power transmission coefficient of the output coupler, hνs and hνp
are the signal and pump photon energies, respectively, and Pabs is the absorbed
pump power [8]. Because the transmission of the output coupler makes up only a
part of the total cavity losses, increasing the value of T1 will also increase the T1/
δ0 ratio and therefore the slope efficiency.
The slope efficiency of a laser is defined as the output power divided by the
pump power after threshold has been reached [8]. This would normally be
expressed in terms of the power of the pump that is actually absorbed in the gain
medium; however, it is not always possible to measure this value, and therefore the
incident or launched pump power values are sometimes used instead.
In most cases, and for all of the laser examples discussed here, the laser
wavelength is longer than the pump wavelength, meaning that the energy of the
laser photons is lower than the pump photons and that even if every pump photon
resulted in a laser photon (100% quantum efficiency), the output laser power will be
lower than the pump laser power. This difference is known as the quantum defect
and is often specified as a proportion of the pump photon energy. For example, a
2 μm laser pumped with 1 μm light has a quantum defect of λpump/λlaser ¼ 0.5,
resulting in a Stokes efficiency limit of 50%. In certain circumstances it is possible
for a laser to exceed the Stokes efficiency limit by achieving an internal quantum
efficiency which is greater than 100%. One example of this is the “two-for-one”
energy cross-relaxation process which can occur in Tm3þ doped laser gain media
when pumped at around 800 nm. The laser output wavelength is in the range
1.9–2.0 μm, resulting in a quantum defect and Stokes efficiency limit of around
40%. However, the “two-for-one” cross-relaxation process can potentially produce
two laser photons for every pump photon and therefore a quantum efficiency of up
to 200% [9]. In practice, an internal quantum efficiency of 180% (and slope
efficiency of 68%) has been demonstrated using this pumping scheme in a Tm3þ
doped germanate fibre laser [10].
Lasers can be operated in either continuous wave (CW) or pulsed modes. Two
common methods for producing pulsed laser emission are Q-switching and mode-
locking. With Q-switching, the loss of the laser cavity is modulated between a high
loss and low loss state. The quality factor (Q) of a laser cavity is defined as the ratio
104 B.D.O. Richards and A. Jha

of energy stored in the cavity to that lost per round trip. While pumping an active
medium, when the Q of the cavity is low, the loss in the cavity is too high for the
laser to reach threshold, and the population in the upper laser level builds up to a
higher level than would be achieved in CW mode. When the Q switches to a high
value and the cavity losses are reduced, the light is amplified until the gain is
saturated and the threshold condition is reached, and a short pulse of intense laser
light is emitted [11].
The time duration between the loss of the cavity switching to a low value and the
laser pulse being emitted is called the pulse build-up time ts and is given by
Eq. (6.4)
m
ts ¼ ½nL L þ nC ðd
LÞ ð6:4Þ
c

where m is the number of passes the developing beam makes through the
amplifier, d is the distance between the mirrors, L is the length of the gain medium
and nL and nC are the refractive indices of the gain medium and the space in the
cavity that does not contain gain medium, respectively [12]. During the pulse build-
up time, the energy must be stored in order to produce a pulse; therefore, the upper
laser level must have a lifetime which is longer than the build-up time, τu > ts. In
rare-earth-doped solid-state lasers, the upper laser level lifetimes are often several
milliseconds (ms) long, and as a result such lasers are capable of producing a pulse
with a duration usually in the 1 ns to 1 μs region. Longer upper laser level lifetime
offers greater energy storage and potentially larger pulse energy if the repetition
rate of the Q-switch is optimised. Another important requirement for Q-switched
operation is that the pumping flux duration Tp is longer than the pulse build-up time
and at least as long as the upper level lifetime, Tp  τu [12]. This applies to solid-
state lasers which are often flash-lamp pumped; however, Q-switched fibre lasers
are usually continuously pumped [13], and therefore this criterion is met. The pulse
energy of a Q-switched laser can be reduced if the laser system suffers from
parasitic lasing between pulses when the cavity losses are high. The Q-switching
can be achieved actively using mechanical techniques, such as an optical chopper or
rotating mirror or prism, or electronically using an acousto-optic modulator (AOM)
or electro-optic modulator (EOM), and passive Q-switching can be achieved using
a saturable absorber, for example [14].
The minimum pulse duration that can be achieved by Q-switching is usually in
the order of a nanosecond; therefore, to produce laser pulses shorter than is quantum
mechanically allowable via Q-switching, another technique called mode-locking
may be employed. Using mode-locking, pulse durations of picoseconds (ps) and
femtoseconds (fs) are possible, dependent on the cavity material, dispersion prop-
erties of optical cavity and saturable absorber characteristics. For a laser gain
medium, if the amplification bandwidth is wide enough, then it is possible for
more than one laser mode to oscillate simultaneously. A laser cavity can support
many longitudinal modes of oscillations provided an integer number of wave-
lengths matches the distance between the cavity mirrors, and the separation
6 Lasers Utilising Tellurite Glass-Based Gain Media 105

frequency between the laser modes depends on the refractive index of the gain
medium n and the distance between the mirrors d such that Δνsep ¼ c/2nd. If several
modes are made to be in phase with each other and are combined, then the resulting
output intensity is characterised by repetitive ultrashort pulses. The more modes
that are combined, the shorter the pulse will be; therefore, having an amplification
medium with a very wide gain bandwidth, such as Ti:sapphire, for example, is
desirable for short-pulse generation. The pulse width ΔtP is given by ΔtP ¼ 1/gain
bandwidth [12].
Mode-locking of a laser cavity can be achieved by using an active or passive
shutter positioned at one end of the laser cavity next to one of the mirrors. An active
shutter must be designed to open at precisely the correct time and for a period of
time short enough to allow a single pulse to pass, which is determined by the
particular gain medium in use. A passive shutter will automatically open upon the
arrival of an intense pulse. The opening of the shutter results in the electric fields of
all the modes being maximised at the same time and therefore being in phase.
Active mode-locking techniques include using an AOM or a short-pulse pump
source to switch the gain in the amplification medium. Several passive mode-
locking techniques are available such as colliding pulse mode-locking which uses
a saturable absorber, additive pulse mode-locking which uses two coupled cavities
and Kerr lens mode-locking which uses the self-focussing effect due to the
nonlinearity of a material [12].

6.2 Rare-Earth-Doped Tellurite Glass Lasers

6.2.1 Neodymium: Nd3þ

As explained above, the lower phonon energy of tellurite glass (650–800 cm
1),
compared with silicates (1050–1120 cm
1) and phosphates (1100–1200 cm
1),
significantly reduces the phonon-mediated relaxation rates. The high refractive
index of the glass increases the radiative transition probability which together
with lower multiphonon decay rates increases the radiative quantum efficiency of
the Nd3þ: 4F3/2–4I11/2 transition at 1.06 μm with a commensurate increase in laser
slope efficiency. The higher refractive index of tellurite glasses (>1.8) compared to
silicate and phosphate hosts also favours greater Stark splitting and yields larger
tuneable laser bandwidth. As a result, the stimulated emission cross-sections of Nd
3þ
in tellurite glass are the largest of all pure oxide glasses, as determined by Judd-
Ofelt analysis [6]. The peak fluorescence wavelength of Nd3þ in tellurite glass is
shifted to longer wavelengths (1061–1063 nm) compared to phosphate glasses
(1052–1057 nm) due to the nephelauxetic shift and greater covalency of the
Nd-O bonds in tellurite glass [15] and/or larger refractive index of tellurite glass
[16]. This apparent red-shift effect has also been observed in Er3þ [17]- and Tm3þ
[16]-doped tellurite glasses.
106 B.D.O. Richards and A. Jha

6.2.1.1 Bulk Glass

The first report of a tellurite glass-based laser was in 1978 using Nd3þ-doped bulk
glass and a 514.5 nm Ar:ion laser to pump the 2G9/2 level [18]. Since then, the
preferred wavelength for pumping Nd3þ-doped tellurite bulk glass lasers has been
at around 800 nm into the 2H9/2, 4F5/2 levels due to the availability of high-power,
low-cost diode lasers [6, 19–21], or at 870 nm to directly pump the 4F3/2 upper laser
level [22]. The longer pump wavelengths also result in reduced quantum defect.
Lei et al. used a host glass composition of 86.6TeO2–8.4BaO–4.0Na2O–1.0ZnO
(wt%) as a laser host for Nd3þ ions. The 2 mm-thick Nd3þ-doped tellurite glass disc
was pumped at 804.3 nm using a pulsed Ti:sapphire laser, with an 8 mm long cavity
constructed using a highly reflective (HR) dichroic mirror and 4%, 12% and 21%
transmissive output couplers (OCs). The maximum output energy and slope effi-
ciency of 1.3 mJ and 14.7%, respectively, were achieved in a quasi-continuous
wave (QCW) regime. Figure 6.2 shows the laser output energy with respect to pump
energy for the three different output couplers, where thresholds of 2.20 mJ, 2.75 mJ
and 4.20 mJ and slope efficiencies of 5.5%, 11.9% and 14.7% were achieved using
the 4%, 12% and 21% transmission output couplers, respectively [6].
More recently, a 1 wt% Nd3þ-doped 85 TeO2–15 ZnO (wt%) bulk tellurite glass
CW 1062 nm laser was reported with maximum output power and slope efficiency
of 45 mW and 21% (with respect to absorbed pump power), respectively, when
pumped at 806 nm. Figure 6.3 illustrates the laser output power as a function of
absorbed pump power for a range of output coupler transmissions of 0.8%, 2.7%
and 8%, showing that 2.7% yielded the highest slope efficiency and output
power [21].

Fig. 6.2 Output energy 1.40

(a)
versus input energy for a Ti:
sapphire pumped Nd: (b)
tellurite glass laser, output
coupler transmission
(a) T ¼ 4%, (b) T ¼ 12%
Output energy (mJ)

and (c) T ¼ 21% [6]

0.70
(c)

0.00
0.00 3.50 7.00 10.50 14.00
Pump energy (mJ)
6 Lasers Utilising Tellurite Glass-Based Gain Media 107

Fig. 6.3 Laser output versus absorbed pump power curves obtained after pumping the TZO glass
sample at around 806 nm, for the output coupler transmissions of 0.8%, 2.7% and 8% [21]

6.2.1.2 Fibre

The first report of a tellurite fibre laser was by Wang et al. in 1994 using a Nd3þ-
doped single-mode fibre. The composition of tellurite glass used was 76.9TeO2–
6.0Na2O–15.5ZnO–1.5Bi2O3–0.1Nd2O3 and 75TeO2–5Na2O–20ZnO (mol%) for
the core and cladding glasses, respectively. The fibre had a 3 μm  6.5 μm elliptical
core and a numerical aperture (NA) of 0.21. The laser cavity was constructed
without the use of mirrors, relying only on the 11.9% Fresnel reflections from the
fibre/air interfaces at the fibre ends. The Fresnel reflection value for tellurite fibre is
significantly higher than in silica and fluoride fibres (~4%) due to its high refractive
index of around 2. When pumped using an 818 nm Ti:sapphire laser, the 0.6 m long
tellurite fibre laser, centred at 1061 nm, achieved an output power of 8 mW with a
slope efficiency of 46% with respect to the absorbed pump power (assuming equal
laser emission from both fibre ends) and a threshold of 27 mW of absorbed pump
power. Figure 6.4 shows the fibre laser output from one end of the fibre [23].

6.2.1.3 Microcavity

The first demonstration of a tellurite microsphere laser was performed by Sasagawa

et al. in 2002. Microspheres doped with 1 wt% Nd3þ were fabricated from tellurite
glass with the composition 70 TeO2–20 ZnO–10Li2O. Laser emission at 1.06 μm
was observed when pumped at 800 nm using a Ti:sapphire laser, which was coupled
108 B.D.O. Richards and A. Jha

4 L = 0.6 m
Laser Power Output (mW)

3 R1 R2

Slope = 23%
1
(output measured
from one end only)

0
0 10 20 30 40 50 60
Absorbed Pump Power (mW)

Fig. 6.4 Laser output power from one end at 1.061 μm versus pump power (0.818 μm) for Nd3þ-
doped tellurite fibre, where L ¼ 0.6 m and R1 ¼ R2 ¼ 11.9%. The total laser efficiency above
threshold is 46% [23]

into 140 μm and 201 μm diameter microspheres using a 20 microscope objective,
while the 1.06 μm laser radiation was collected using a multimode fibre for analysis
using an optical spectrum analyser (OSA). The 140 μm diameter microsphere laser
exhibited an incident pump threshold of 81 mW; however, coupling of pump light
into the microsphere was not optimised [24].

6.2.2 Erbium: Er3þ

Er3þ-doped lasers are of great interest for their potential use in the 1.5 μm tele-
communications band, and tellurite glasses are of particular interest as they can
offer high rare-earth ion solubility as well as broadband gain [7]. Er3þ-doped
tellurite glass was shown to have 1.3 times higher stimulated cross section than
silica and fluoride glass at 1530–1580 nm and >2 times higher at 1600 nm [7].

6.2.2.1 Fibre

The second ever report of a tellurite fibre laser was in 1997 by Mori et al. using an
Er3þ-doped single-mode fibre. The drawn fibre had a core diameter of 3 μm, a Δn of
0.7% and transmission loss of 3 dB/m at 1300 nm. The laser cavity was constructed
using a 0.85 m long fibre and 12.3% Fresnel reflections from both fibre ends.
6 Lasers Utilising Tellurite Glass-Based Gain Media 109

Fig. 6.5 Output laser 3

power (from one end of the
fibre) at 1560 nm versus
launched pump power [7]

laser power, mW
0.65 %

0
0 200 400 600
pump power, mW 268/3

The 1560 nm laser output reached a maximum of 5 mW at a slope efficiency of

1.3% with respect to launched 978 nm pump power, assuming equal emission from
both fibre ends, as shown in Fig. 6.5. The launched pump threshold was
120 mW [7].
A tunable Er3þ/Ce3þ-doped tellurite fibre ring laser (Fig. 6.6) was demonstrated
by Dong et al. in 2011 using glass compositions 80TeO2–10ZnO–10Na2O and
75TeO2–15ZnO–10Na2O (mol%) for the fibre core and cladding, respectively.
The tellurite fibre ring laser was dual pumped using 980 nm and 1480 nm laser
diodes to excite the 4I11/2 and 4I13/2 energy levels of Er3þ, respectively. The laser
was tuned from 1527 nm to 1610 nm (Fig. 6.7) with the peak power occurring at a
wavelength of 1558 nm. The addition of the 1480 nm pump yielded an improve-
ment of around 10 dBm in output power compared to 980 nm pumping only. The
addition of Ce3þ co-doping with Er3þ is common as there is a resonant energy
transfer from Er3þ: 4I11/2!4I13/2 to Ce3þ: 2F5/2!2F7/2 which reduces ESA from the
Er3þ: 4I11/2 level and improves the ~1.5 μm gain characteristics [25]. Another
co-dopant which might be used for reducing the Er3þ ESA is Eu3þ, as discussed
by Shen et al. [26]. Without co-doping to reduce pump-induced ESA which results
in green light in Er3þ-doped glasses, the performance of lasers and amplifiers
operating in the 1530–1610 nm region suffers adversely in bulk glass, fibres and
waveguides [27].
110 B.D.O. Richards and A. Jha

980nm 1480nm
pump LD1 pump LD2
Er/Ce codoped fibre

WDM coupler WDM coupler

Isolator

Tunable filter PC
9:1 coupler

Output

Fig. 6.6 Schematic diagram of the Er3þ/Ce3þ-co-doped tellurite fibre ring [25]

20
20
10

0 0

−10
−20
Power (dB m)

Net gain (dB)

−20

−30 −40

−40
−60
−50
−80
−60

−70 −100
1520 1540 1560 1568 1600

Wavelength (nm)
3þ 3þ
Fig. 6.7 Output spectra of the Er /Ce -co-doped tellurite fibre ring laser [25]

6.2.2.2 Microcavity

The first Er3þ-doped tellurite glass microsphere lasers were reported in 2003 by
Peng et al. [28, 29]. Tellurite glass was chosen for investigation due to its high
emission cross section resulting in higher gain. 975 nm pump light was delivered to
6 Lasers Utilising Tellurite Glass-Based Gain Media 111

the 33 μm diameter microspheres using a 1.3 μm diameter tapered fibre and the
laser signal collected using the same fibre. The laser wavelength was in the range
1560–1610 nm, with wavelength tuning and multi-/single-mode operation achiev-
able by adjusting the pump power and fibre taper position, respectively. A typical
feature of microsphere lasers is very low pump threshold power, which in this
demonstration was <2 mW [28, 29].

6.2.3 Thulium: Tm3þ

An area that has received significant attention during recent years is that of Tm3þ-
doped lasers operating in the 1.9–2.0 μm wavelength region. This spectral range is
important from the point of view of atmospheric science, remote sensing, medicine
and as a source which enables mid-IR light sources via nonlinear processes. There
have been many reports of high-power Tm3þ silica fibre lasers; however, Tm3þ-
doped tellurite bulk glass and fibre lasers are also being increasingly researched.
The Tm3þ: 3F4–3H6 ~2 μm laser transition is one of the broadest of all of the rare-
earth ions and, therefore, offers scope for realising widely tuneable lasers and broad
bandwidth gain material for the production of ultrashort pulses [30].

6.2.3.1 Bulk Glass

CW 1.9 μm tellurite bulk glass lasers using zinc-sodium-tellurite (TZN) and zinc-
sodium-germanium-tellurite (TZNG) glass compositions doped with Tm3þ were
demonstrated when pumped at 793 nm using a Ti:sapphire laser [31]. The bulk
glass samples were Brewster cut to eliminate reflection losses with sample lengths
of 7 mm and 5 mm to achieve 80% pump absorption in the TZN and TZNG samples
doped with 1.5 wt% and 2 wt% Tm3þ, respectively. The cooled samples were
placed into an asymmetric astigmatic-compensated Z-folded laser cavity. The
TZNG sample achieved a maximum output power of 124 mW and slope efficiency
of 28% with respect to the absorbed pump power. Figure 6.8 shows the detailed
results of the output power versus absorbed pump power for the TZNG glass with
various output coupler transmittances, with a 6.1% transmission output coupler
yielding the highest output power and slope efficiency values. The tuning range of
the two tellurite glass gain media was measured by inserting a prism into the laser
cavity, and a 195 nm tuning range of 1830–2025 nm (1875–2010 nm FWHM) was
demonstrated using the TZNG glass, illustrated in Fig. 6.9 in comparison with the
TZN glass sample [31].
The same Tm3þ-doped TZNG bulk glass laser was also studied under 1211 nm
pumping using a semiconductor disc laser to pump the peak of the 3H5 absorption
band of Tm3þ. Using an astigmatically compensated 3-mirror V-type cavity, this
laser produced a slope efficiency of 22%, with respect to absorbed pump power, and
112 B.D.O. Richards and A. Jha

Fig. 6.8 The Tm3þ:TZNG laser output power versus absorbed pump power [31]

Fig. 6.9 The Tm3þ:TZN and the Tm3þ:TZNG glass laser tunability measured using a 0.8%
output coupler and with a fused silica prism as a tuning element when pumped at 793 nm using a
Ti:sapphire laser. The inset provides the tunability values (FWHM) for the two media [31]

an output of 60 mW, limited only by the available pump power, while the FWHM
tuning range was 115 nm (1850–2040 nm) (Fig. 6.10) compared to 135 nm when
pumped at 793 nm (Fig. 6.9). The slightly inferior efficiency of the 1211 nm
pumped laser was attributed to the better beam quality of the Ti:sapphire laser [32].
6 Lasers Utilising Tellurite Glass-Based Gain Media 113

Fig. 6.10 The Tm3þ:TZNG normalised tuning plot obtained with a birefringent filter, the FWHM
for the 0.8% OC is highlighted. The pump source was a 1211 nm semiconductor disc laser [32]

6.2.3.2 Fibre

The first report of Tm3þ laser emission from a tellurite fibre was by Richards et al.
in 2007 [33]. In this instance, the fibre was co-doped with Yb3þ and pumped using
the 1088 nm output from a Yb3þ silica fibre laser which simultaneously pumped the
tails of the Yb3þ: 2F5/2 and Tm3þ: 3H5 absorption bands. This pumping scheme
resulted in strong two- and three-photon pump excited state absorption (ESA) and
subsequent intense blue fluorescence from the Tm3þ: 1G4 level. Pump ESA and
visible fluorescence reduce the population inversion at the 3F4 upper laser level and
therefore result in reduced output power and slope efficiency. This laser achieved a
maximum output power of 67 mW and slope efficiency of 10% with respect to
absorbed pump power before the 80TeO2–10ZnO–10Na2O (mol%) fibre suffered
thermal damage at the pump end. Figure 6.11 shows how the laser slope efficiency
varied as a function of output coupler reflectivity, with a 50% reflective output
coupler yielding the highest value of 10%. At the time of its publication, this work
represented the highest output power ever reported from a tellurite fibre laser [34].
By employing a 1.6 μm pump source (e.g. using an Er/Yb silica fibre laser), it is
possible to directly in-band pump the upper laser level of Tm3þ and alleviate the
problems of pump ESA and fibre damage [35]. Due to the lack of pump ESA and
very low quantum defect associated with in-band pumping, a Tm3þ-doped tellurite
fibre laser with 280 mW output power and 76% slope efficiency was achieved,
which was close to the Stokes efficiency limit of ~84% (internal quantum efficiency
of ~90%), from a 32 cm long fibre. Figure 6.12 compares the slope efficiency of this
laser with different values of reflectivity of output couplers, while the inset shows a
 114 B.D.O. Richards and A. Jha

0.06
50% R output coupler. Slope = 10%
70% R output coupler. Slope = 9%
0.05 90% R output coupler. Slope = 2%
Laser output power (W)

0.04

0.03

0.02

0.01

0.00
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Absorbed pump power (W)

Fig. 6.11 Laser output power with respect to absorbed pump power for ~15 cm-long tellurite fibre
with varying output coupler reflectivities [34]

160

0.30 140 Slope = 75.8%

120
−2 μm laser output power (W)

0.25
Intensity (a.u.)

100

80

0.20 60

40

0.15
20

1940 1945 1950 1955 1960 1965 1970 1975 1980

Wavelength (nm)
0.10 Slope = 60.4%

0.05
99%-12% cavity
99%-50% cavity
0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55
Launched pump power (W)

Fig. 6.12 Output power of the 2 μm laser with respect to launched pump power for a 32 cm-long
fibre with 99%–12% and 99%–50% cavities. The inset graph shows a typical laser spectrum from a
32 cm-long fibre and a 99%–12% cavity [35]
6 Lasers Utilising Tellurite Glass-Based Gain Media 115

typical laser output spectrum. Lasing was observed in fibres as short as 9 cm

[35]. This result represented the highest power tellurite laser at the time. While
the well-known two-for-one cross-relaxation process (3H4þ3H6)!(3F4þ3F4) asso-
ciated with pumping Tm3þ lasers at around 800 nm [36] can theoretically result in a
quantum efficiency of up to 200%, the Stokes efficiency limit is low at around 40%,
meaning that a maximum slope efficiency up to 80% may be possible. In practice,
the highest slope efficiency reported using this pump scheme is 68%, (representing
a quantum efficiency of 180%), from an 800 nm pumped Tm3þ-doped germanate
fibre laser [10], suggesting that in-band pumping offers more efficient pumping of
Tm3þ, with slope efficiencies of 71%, 76% and 84% reported from silica [37],
tellurite [35] and fluoride [38] fibres, respectively. The higher slope efficiency is
possible in fluoride fibres due to lower multiphonon relaxation rate, compared with
tellurite and silica fibres. A further consideration, however, is the efficiency of the
pump source itself, and 800 nm diode laser technology is significantly more mature
than 1.6 μm diode laser technology. With a NA of 0.29 at 2 μm and a core diameter
of 14 μm, this fibre was heavily multimode having a V number (V ¼ (2π/λ)aNA,
where a is the fibre core radius) of 6.4. If V < 2.405; only a single mode can be
supported by the fibre.
For in-band pumping where the laser emission occurs from the same level that is
being pumped, the optimum pump wavelength should be the wavelength at which
the ratio of absorption to emission cross section (σ a/σ e) of that level is the highest,
assuming there are no ESA processes to depopulate the level [38]. This wavelength
was determined to be 1.606 μm in Tm3þ-doped tellurite glass, as can be seen in
Fig. 6.13 [39].
By employing a double-clad fibre design based on high Tg tungsten-tellurite
glass and high-power 803 nm diode-cladding pumping, Li et al. demonstrated a
~1.9 μm Tm3þ tellurite fibre laser with a CW output power of 1.12 W, the highest
power from a tellurite laser at the time, and importantly, the first report of a directly
diode-pumped tellurite bulk or fibre laser. The relatively high 2.9 dB/m background
loss of the fibre resulted in a laser slope efficiency of just 20% with respect to
absorbed pump power for a 40 cm long fibre; however, this increased to 31% when
the fibre was shortened to 12.5 cm [40]. The same group followed on from this work
by demonstrating single-mode laser output from Tm3þ and Tm3þ/Ho3þ-doped
tellurite fibres when pumped using an 800 nm laser diode. The same glass compo-
sition of 60TeO2–30WO3–10La2O3 was used to fabricate the double-clad fibres
which had a core diameter of 9 μm and NA of 0.14 [41]. By comparison, the fibres
developed for the high power Tm3þ tellurite fibre lasers were multimode with a
core diameter of 18 μm and a NA of 0.14 [40]. A 20 cm long Tm3þ tellurite single-
mode fibre yielded 494 mW of 1.9 μm laser output with a slope efficiency of 26%
with respect to absorbed 793 nm pump power. Single-mode laser output was
confirmed by profiling the beam shape which had an M2 beam quality factor of
1.09 [41].
A 2 μm Tm3þ fibre laser with an output power of 0.75 W and slope efficiency of
28.7% with respect to absorbed 793 nm pump power was demonstrated using a
glass composition based on the glass-forming oxides of tellurium and germanium.
116 B.D.O. Richards and A. Jha

Peak at 1.606 μm
5

4
σa/σe

0
1500 1600 1700 1800 1900 2000
Wavelength (nm)

Fig. 6.13 Absorption/emission cross section ratio (σa/σe) for Tm3þ-doped tellurite glass. The
emission cross section was calculated using the F-L equation. The peak value is at a wavelength of
1.606 μm [39]

The motivation behind this approach was to combine the favourable properties of
tellurite glass, such as high rare-earth ion solubility, low phonon energy and higher
radiative rate, with the improved thermo-mechanical properties of germanate glass.
The glass composition used to fabricate the double-clad, single-mode fibre was
45GeO2–25TeO2–15PbO–10(La2O3þAl2O3)–5(CaOþSrOþLi2O)–1Tm2O3 (mol
%). The resulting glass had a glass transition temperature of 470  C which is higher
than tungsten-tellurite glass (450  C) [42] and zinc-sodium-tellurite glass (~290  C)
[43] and similar to lead-sodium-germanate glass (475  C) [44], confirming the
improvement in the thermal properties of the tellurite glass through the addition of
GeO2. The emission cross section of the Tm3þ: 3F4–3F6 transmission in the
tellurium-germanate glass peaked at 7  10
21 cm2 at 1.9 μm [42], which compares
to 15  10
21 cm2 at 1.85 μm in zinc-sodium-tellurite glass [45] and 5.8  10
21
cm2 at 1.88 μm in germanate glass [46], suggesting that the spectroscopic properties
of Tm3þ in tellurium-germanate glass occupy a range between those of individual
tellurite and germanate glasses.

6.2.3.3 Microcavity

Tm3þ is most commonly used to produce laser emission in the 1.9–2.1 μm spectral
region from the 3F4–3H6 bottom level transition; however, it is also possible to
realise gain at 1.4–1.5 μm from the 3H4–3F4 transition. This is of particular interest
6 Lasers Utilising Tellurite Glass-Based Gain Media 117

as a method of extending the gain bandwidth of fibre amplifiers beyond the range
available from standard erbium-doped fibre amplifiers (EDFAs) for optical tele-
communication applications. Producing a laser using this transition is complicated
by the fact that the lifetime of the upper 3H4 level is around five times shorter than
the lower 3F4 level in most glass hosts, resulting in a “self-terminating” transition
[47]. It is therefore necessary to depopulate the lower laser level. One method for
achieving this is to co-dope the glass with another rare-earth ion which has a
resonant energy transfer with the Tm3þ: 3F4 level; however, another option is to
simultaneously lase the 3F4–3H6 transition, resulting in what is known as a cascade
laser. A tellurite glass microsphere is one of the host glasses in which this Tm3þ
cascade laser architecture has been demonstrated [48]. The low phonon energy of
tellurite glass helps to prolong the lifetime of upper laser levels by reducing the
multiphonon decay rate. In order to achieve high efficiency ~2 μm lasing using
~800 nm pumped Tm3þ glass, it can be beneficial to increase the Tm3þ concentra-
tion in order to enhance the two-for-one cross-relaxation process; however, in order
to achieve lasing at ~1.5 μm, it is preferable to reduce the cross-relaxation in order
to keep the 3H5 upper laser level lifetime as long as possible; therefore, optimising
the Tm3þ doping concentration is critical. It was found that when the Tm3þ
concentration was increased above 0.5 wt%, the 3H5 lifetime reduced from
180 μs to 120 μs or less [48]. The 793 nm pump light was launched into the tellurite
glass microsphere using a fibre taper, and laser emission at 1.5 μm and 1.9 μm was
observed. The laser threshold was lower for the 1.9 μm transition; however, once
the 1.5 μm threshold pump power was reached and began lasing, the slope effi-
ciency of the 1.9 μm laser output increased significantly due to the increase in
population inversion to the 3F4 level due to stimulated emission of the 3H4–3F4
transition [48].

6.2.4 Holmium: Ho3þ

Another rare-earth dopant ion commonly used to produce laser radiation in the 2 μm
spectral region is Ho3þ, from its 5I7 upper laser level. Compared to Tm3þ, Ho3þ
emits at the slightly longer wavelength of around 2.1 μm. Unlike Nd3þ, Yb3þ, Er3þ
and Tm3þ, Ho3þ does not have any absorption bands which coincide with readily
available and cost-effective diode lasers such as those emitting at 800 nm and
980 nm. Ho3þ is therefore commonly co-doped with a sensitising ion to provide
useful absorption bands, and the most common is Tm3þ which can be pumped at
around 800 nm where the two-for-one cross-relaxation process can be exploited,
and the energy transfer from the Tm3þ: 3F4 level to the Ho3þ: 5I7 upper laser level
can be very efficient, at more than 82% in tellurite glass due to the resonant
nonradiative energy transfer [49].
118 B.D.O. Richards and A. Jha

6.2.4.1 Bulk Glass

Tm3þ/Ho3þ-co-doped TZN (80-10-10 wt%) bulk tellurite glass has been demon-
strated to lase in both CW and fs mode-locked operation. In Tm3þ/Ho3þ-co-doped
gain media, the optimisation of the dopant concentrations is critical in order to
balance various energy transfer processes such as upconversion and ESA, two-for-
one cross-relaxation and Tm3þ to Ho3þ energy transfer. Of the samples investigated
in [50], the composition with 2 wt% Tm2O3 and 0.1 wt% Ho2O3 was found to
contain the optimum concentrations of dopants in the host glass. In CW mode, the
tellurite bulk glass laser produced 74 mW (pump power limited) of 2012 nm laser
radiation with a slope efficiency of 26% with respect to the absorbed 793 nm pump
power. When the output coupler reflectivity was decreased from 98% to 96%
(representing an increase in cavity loss), the laser wavelength reduced to
1944 nm suggesting that the laser emission was from the Tm3þ: 3F4–3H6 transition.
The tuning range of the laser was measured to extend from 1870 nm to 2080 nm,
with a FWHM of 125 nm [50].

6.2.4.2 Fibre

A Ho3þ-doped tellurite fibre laser was first demonstrated in 2008 by Tsang et al.,
using 1.6 μm Er3þ/Yb3þ [49] and 1.088 μm Yb3þ [51] silica fibre laser pump
sources. In the case of 1.6 μm pumping, Tm3þ was added as a sensitising ion due
to the efficient excitation into the Tm3þ: 3F4 level. 160 mW of 2.1 μm laser output
power was realised with a slope efficiency of 62% with respect to launched pump
power, which is 82% (internal quantum efficiency) of the Stokes efficiency limit
of 76%. The optimum fibre length in this experiment was in the range 40–78 cm;
however, lasing was observed in fibres as short as 10 cm [49]. High-power Tm3þ/
Ho3þ-doped silica fibre lasers have been demonstrated; however, the increased
background loss of silica at 2.1 μm coupled with the long fibres required to
overcome the low pump absorption of rare-earth ions in silica fibre has generally
resulted in high pump threshold values of several Watts and modest slope
efficiencies [52, 53]. The absorption of 1.088 μm light in Ho3þ is quite low;
therefore, Tm3þ and Yb3þ sensitising was employed to increase the pump
absorption [51]. At the time these works were published; they represented the
longest wavelength tellurite glass laser and the first demonstration of a pulsed
tellurite fibre laser.
This same 1.6 μm pumped Tm3þ/Ho3þ-co-doped tellurite fibre laser was also
operated in Q-switched mode using a mechanical chopper as a Q-switch with a
repetition rate of 19.4 kHz [49]. The non-optimised laser cavity produced laser
pulses with energies up to 4.36 μJ, or peak powers of up to 40 W, with durations in
the range 109–162 ns from a 79 cm long fibre. In this arrangement, at least three
pulses were observed per Q-switch opening, suggesting that pulse energy could be
increased by optimising the cavity and Q-switch repetition rate [45]. This is
6 Lasers Utilising Tellurite Glass-Based Gain Media 119

Fig. 6.14 Typical output spectra of Tm3þ/Yb3þ- and Tm3þ/Ho3þ/Yb3þ-co-doped tellurite fibre
lasers [45]

believed to be the first demonstration of a Q-switched laser based on a tellurite glass

host. Figure 6.14 compares the output laser spectra of Tm3þ and Tm3þ/Ho3þ-doped
tellurite fibre lasers operating in the ~1.9 μm and ~2.1 μm regions, respectively.
As with Tm3þ, it is possible to directly in-band pump the Ho3þ: 5I7 upper laser
level in order to reduce the quantum defect and therefore increase the Stokes
efficiency limit. This concept has previously been demonstrated by pumping a Ho
3þ
-doped silica fibre laser with the 2.046 μm output of a Tm3þ-doped silica fibre
laser to produce laser radiation at 2.1 μm with a slope efficiency of 82% and an
internal quantum efficiency of 84% [54]. The highest power demonstration of a
resonantly pumped Ho3þ-doped silica fibre laser was achieved using two 1.95 μm
Tm3þ-doped silica fibre lasers to in-band, cladding pump the fibre, achieving
140 W of 2.12–2.15 μm laser power, with a slope efficiency of 59% with respect
to launched pump power [55]. This concept has also been demonstrated using Ho3þ
-doped tellurite fibre when pumped using a Tm3þ-doped silica fibre laser. The
output wavelength of the pump laser was adjusted in the range 1.98–2.06 μm by
altering the fibre length and mirror reflectivities, and laser emission from the Ho3þ:
tellurite fibre was detected in the range 2.056–2.104 μm. A full analysis of the laser
performance was not possible at the time, but the results indicate that Stokes
efficiency limits of up to 98% are possible with this pumping scheme based on
the pump and laser wavelengths recorded during laser experiments and displayed in
Fig. 6.15 [45]. As diode lasers operating at around 2 μm become more readily
120 B.D.O. Richards and A. Jha

Fig. 6.15 The pump and laser spectra for Tm3þ-doped silica fibre laser pumped Tm3þ/Ho3þ/Yb3þ
co-doped tellurite fibre lasers [45]

available, this in-band pumping scheme may prove to be a method for achieving
highly efficient and high-power 2.1 μm laser emission.
Yao et al. have demonstrated a 2.077 μm Ho3þ laser based on 70TeO2–20BaF2–
9.25Y2O3–0.75Ho2O3 (TBYH) microsctructured fluorotellurite fibre which was
in-band pumped using a 1.992 μm Tm3þ-doped silica fibre laser. A maximum
output power of 161 mW was achieved with a slope efficiency of 67.4% with
respect to the launched pump power from a 27 cm long fibre. By adding BaF2 to the
glass composition, the absorption coefficient due to hydroxyl (OH
) contamination
was reduced to around 0.1 cm
1 (0.43 dB/cm) in a 1 mm thick bulk glass sample.
Figure 6.16 shows the transmission spectrum of the 1 mm thick fluorotellurite glass
sample which exhibits around 80% transmission in the wavelength range
0.47–5.14 μm [56].
In-band pumping of Ho3þ ions using ~2 μm pump lasers is a promising method
for achieving high efficiency ~2.1 μm lasers, however, the concept has only been
demonstrated to date using ~2 μm Tm3þ fibre lasers which are in turn pumped
using ~800 nm diode lasers, reducing the overall system efficiency. Pumping Ho3þ
using ~2 μm laser diodes would potentially result in higher overall system effi-
ciency; however, these diodes are not yet readily available with high powers.
Another option is to directly pump Ho3þ using a 793 nm laser diode by sensitising
with Tm3þ ions. This approach has been demonstrated by Li et al. using a 7 cm
6 Lasers Utilising Tellurite Glass-Based Gain Media 121

Fig. 6.16 Transmission

spectrum of the 1 mm-thick
undoped barium-yttrium-
fluorotellurite glass sample
[56]

long, single-mode, double-clad 60TeO2–30WO3–10La2O3 tellurite fibre, yielding

35 mW of laser output at 2046 nm. A disadvantage of this pumping scheme is that it
results in intense blue and green upconversion emission from the Tm3þ: 1G4!3H6
and Ho3þ: (5S2,5F4)!5I8 transitions, respectively, to the detriment of the ~2 μm
laser performance [41].

6.3 Other Tellurite Glass Light Sources

6.3.1 Raman Lasers

In 2003, Mori et al. reported the first tellurite fibre Raman amplifier [57]. The larger
Raman gain coefficient and Stokes shift of the tellurite fibre compared to silica fibre
allowed the use of a shorter fibre to achieve wide-band amplification [57]. Bulk
tellurite glass with the composition 85TeO2–15WO3 was shown to have a peak
absolute Raman gain coefficient 30 times higher than that of bulk silica glass [58],
and BaO/Bi2O3 and NbO6/MoO6 containing tellurite glasses have been shown to
have broadened Raman gain spectra without sacrificing the peak gain coefficient
[59]. Stimulated Raman scattering was observed in photonic crystal fibres drawn
from tellurite glass with the composition 5Na2O–20ZnO–75TeO2 [60].
Qin et al. have subsequently demonstrated a ring-cavity tellurite fibre Raman
laser with a tuning range covering 1495–1600 nm (or free running wavelength of
1665 nm) using a single-mode tellurite fibre with high Raman gain coefficient of
55 W
1km
1 and large Raman shift of 22.3 THz (183 nm). In comparison, silica
fibre has a Raman gain coefficient which is approximately 16 times lower than the
122 B.D.O. Richards and A. Jha

tellurite glass composition used and a Raman shift of only 13.2 THz. A higher value
of Raman gain coefficient results in lower Raman laser threshold power. The
Raman fibre laser was pumped using a 5 W, 1480 nm laser source. The gain
medium was a 150 m long single-mode tellurite fibre with a core diameter of
2.2 μm and a background loss of 0.02 dB/m [61].
Raman soliton self-frequency shifting (SSFS) has been demonstrated in a highly
nonlinear microstructured 72TeO2–24WO3–4La2O3 (mol%) tellurite fibre,
resulting in 100 fs laser pulses tunable in the range 1.6–2.65 μm. This source was
operated in two pump regimes, firstly using 1.6 μm, 2 nJ, 150 fs pulses, and
secondly using 2 μm, 3 nJ, 150 fs pulses. The data suggests that output beyond
3 μm should be possible with higher pump energies and use of fibre with lower
mid-IR losses [62].

6.3.2 Supercontinuum Sources

Tellurite glass fibres have been successfully used to generate supercontinuum light
due to their high nonlinearity and ability to be formed into photonic crystal fibres
(PCFs) [63–68]. High-power supercontinuum sources are readily available com-
mercially; however, being based on silica fibre, their bandwidth is limited to the
400–2400 nm range. There is therefore significant interest in producing
supercontinuum generation further into the infrared using low phonon energy
glasses such as tellurites, fluorides and chalcogenides [66]. An 8 mm-long tellurite
PCF was pumped using a 1550 nm laser with 100 fs-long, 1.9 nJ pulses to produce
broadband supercontinuum generation spanning 4080 nm, from 789 nm to
4870 nm, with an average power of 90 mW. Supercontinuum sources offer the
advantages of high bandwidth, brightness and coherence and are therefore of
significant interest for mid-IR spectroscopy such as molecular fingerprinting. The
high optical nonlinearity of tellurite glass is augmented by the enhanced waveguide
nonlinearity of the PCF structure. An advantage of tellurite glass compared to some
other glasses is that its high nonlinearity allows the use of shorter waveguides which
has the benefit of a smoother supercontinuum spectrum and lower total background
material loss. The glass composition used to fabricate the PCF was 75TeO2–
12ZnO–5PbO–3PbF2–5Nb2O5, and the material and waveguide dispersions
resulted in a zero dispersion wavelength (ZDW) of 1380 nm [63]. The chromatic
dispersions of a range of binary tellurite glasses were determined using the wave-
length dependence of their refractive indices and Sellmeier fitting, showing that the
ZDWs of the glasses ranged from 2086 nm to 2324 nm, depending on the glass
network modifier used. For comparison, the ZDWs of pure TeO2 and SiO2 are
1688 nm and 1273 nm, respectively [69].
6.4 Summary of the Performance of Rare-Earth-Doped Tellurite Glass-Based Lasers

Gain Laser Pump laser Max output Max slope

medium wavelength CW/ wavelength power/ efficiency Pump
Ref Glass composition Dopant form (μm) pulsed (nm) energy (%) threshold
[6] 86.6TeO2–8.4BaO– Nd3+ 2 mm thick 1.066 Gain 804.3 Ti:sap- 1.3 mJ 14.7 4.20 mJ
4.0Na2O–1.0ZnO (wt%) disc switched phire laser
[18] 95.5TeO2–4.5Li2O Nd3+ Bulk glass 1.065 – 514.5 Ar:ion – 14 20 mW
laser
[19] 80TeO2–20WO3 (mol%) Nd3+ Bulk glass 1.065 Gain 805 Ti:sap- 11 μJ 12 16 μJ
(0.5–1.0 mol switched phire laser (incident)
%)
[20] 74.6TeO2–8.8ZnO– Nd3+ (NdF3) Bulk glass 1.056–1.064 QCW 790–810 Ti: – – 27 mJ
16.6ZnF2 (mol%) (1, 2 wt%) sapphire laser
[21] 85TeO2–15ZnO (wt%) Nd3+ (1 wt%) Bulk glass 1.062 CW 806 45 mW 21 17 mW
6 Lasers Utilising Tellurite Glass-Based Gain Media

(absorbed) (absorbed)
[23] 76.9TeO2–6.0Na2O– Nd3+ (0.1 mol Fibre 1.061 CW 818 Ti:sap- 8 mW 46 27 mW
15.5ZnO–1.5Bi2O3– %) phire laser (absorbed) (absorbed)
0.1Nd2O3 (mol%)
[24] 70TeO2–20ZnO–10Li2O Nd3+ (1 wt%) Microsphere 1.06 CW 800 Ti:sap- – – 81 mW
phire laser (incident)
[25] 80TeO2–10ZnO– Er3+ (0.9  Fibre 1.525–1.610 CW 980 nm & 10 dBm – –
10Na2O (mol%) 1020/cm3) 1480 nm laser
diode
Ce3+ (2.1 
1020/cm3)
[28] – Er3+ Microsphere 1.56–1.61 – 975 <1 μW – 2 mW
[29] – Er3+ Microsphere 1.606 – 1480 41 μW – 2 mW
(continued)
123
 124

(Continued)

Gain Laser Pump laser Max output Max slope

medium wavelength CW/ wavelength power/ efficiency Pump
Ref Glass composition Dopant form (μm) pulsed (nm) energy (%) threshold
[31] 80TeO2–10ZnO–10Na2O Tm3+ (1.5 wt Bulk glass 1.876–2.017 CW 793 – – –
(mol%) %)
75TeO2–10ZnO–10Na2O– Tm3+ (2 wt%) 1.830–2.025 124 mW 28 150 mW
5GeO2 (mol%) (absorbed) (absorbed)
[32] 75TeO2–10ZnO–10Na2O– Tm3+ (2 wt%) Bulk glass 1.85–2.04 CW 1211 semicon- 60 mW 22.4 150 mW
5GeO2 (mol%) ductor disc (absorbed) (incident)
laser
[34] 80TeO2–10ZnO–10Na2O Tm3+/Yb3+ Fibre 1.910–1.994 CW 1088 Yb3+ 67 mW 10 114 mW
(mol%) (1.0/1.5 wt%) fibre laser (absorbed) (absorbed)
[35] 80TeO2–10ZnO–10Na2O Tm3+/Yb3+ Fibre 1.88–1.99 CW 1570–1610 Er 280 mW 76 160 mW
3+
(mol%) (1.0/1.5 wt%) /Yb3+ fibre
laser
[40] 60TeO2–30WO3–10La2O3 Tm3+ (1 mol Fibre 1.937 CW 800 diode laser 1.12 W 20 1.46 W
(mol%) %) (absorbed) (absorbed)
[41] 60TeO2–30WO3–10La2O3 Tm3+ Fibre 1.900 CW 793 diode laser 494 mW 26 0.5 W
(mol%) (absorbed) (absorbed)
Tm3+/Ho3+ 2.046 35 mW – –
[42] 45GeO2–25TeO2–15PbO– Tm3+ (1 mol Fibre 1.968 CW 793 diode laser 0.75 W 28.7 0.56 W
10(La2O3+Al2O3)–5(CaO %) (absorbed) (absorbed)
+SrO+Li2O) (mol%)
[48] – Tm3+ (1 mol Microsphere 1.5 & 1.9 CW 793 Ti:sap- – – 2.7 &
%) phire laser 0.7 mW
B.D.O. Richards and A. Jha
[49] 80TeO2–10ZnO–10Na2O Yb3+/Tm3+/ Fibre 2.086 CW 1570–1610 0.16 W 62 0.1 W
(mol%) Ho3+ (1.5/1.0/ (launched) (launched)
1.0 wt%) Q- Er3+/Yb3+ fibre 26 mW – –
switched laser (average)
(160 ns
pulses) 0.65 μJ
[50] 80TeO2–10ZnO–10Na2O Tm3+/Ho3+ Bulk glass 2.012–2.048 CW 792 Ti:sap- 74 mW 26 100 mW
(mol%) (2.0/0.1 wt%) phire laser (absorbed)
Mode- 38 mW 11 275 mW
locked (average (absorbed) (absorbed)
(630 fs power)
pulses)
[51] 80TeO2–10ZnO–10Na2O Yb3+/Tm3+/ Fibre 2.1 CW 1088 Yb3+ 60 mW 25 15 mW
(mol%) Ho3+ (1.5/1.0/ fibre laser (launched) (launched)
1.0 wt%)
[56] 70 TeO2–20 BaF2–9.25 Ho3+ (0.75%) Fibre 2.077 CW 1992 Tm3+ 161 mW 67.4 57 mW
Y2O3 fibre laser (launched) (launched)
6 Lasers Utilising Tellurite Glass-Based Gain Media
125
126 B.D.O. Richards and A. Jha

6.5 Future Prospects for Tellurite Glass-Based Laser

Sources

Due to the relatively low phonon energy of tellurite glass and its resulting infrared
transmission range up to around 5 μm, considerable effort has been focussed on
developing rare-earth-doped tellurite glass which exhibits fluorescence in the mid-
infrared spectral region at wavelengths greater than 2.1 μm which may be viable for
laser operation. To date, rare-earth-doped glass-based solid-state lasers operating at
wavelengths greater than 2.1 μm have been restricted to the use of fluoride glasses
such as ZBLAN glass which has a maximum phonon energy of around 550 cm
1
and an infrared transmission of 90% up to around 6 μm [30, 70]. Mid-infrared lasers
using ZBLAN glass as a host material have been demonstrated several times using
Er3þ (λ ¼ 2.8 μm, [71]), Dy3þ (λ ¼ 2.9 μm, [72]) and Ho3þ (λ ¼ 2.9 μm, [73],
λ ¼ 3.9 μm, [74]) ions; however, ZBLAN glass suffers from relatively poor
chemical, mechanical and thermal stability. Silica glass is extremely robust and
capable of supporting laser powers of many tens of kilowatts, but cannot support
laser wavelengths greater than around 2.2–2.5 μm due to its relatively high maxi-
mum phonon energy of 1100 cm
1 [30, 75, 76].
Several reports demonstrate 2.7 μm fluorescence from Er3þ-doped [77, 78] and
Er /Nd3þ-co-doped [79] fluorotellurite glasses and ~3 μm fluorescence from Dy3þ
3þ

-doped tellurite and fluorotellurite glasses [80, 81]. It was found, however, that in
Dy3þ-doped zinc-sodium tellurite and fluorotellurite glasses, the bottom level 6H13/
2! H15/2 2.9 μm decay transition favours multiphonon absorption over radiative
6

absorption and also suffers from quenching due to residual OH
ions [81]. In
tellurite glasses with maximum phonon energies up to around 800 cm
1, five
phonons are required to bridge the Dy3þ: 6H13/2!6H15/2 energy gap, compared to
seven in ZBLAN glass, resulting in larger nonradiative rate in tellurite glass
compared to ZBLAN glass. Numerical modelling suggested that population inver-
sion at 3042 nm may be reached in 1 mol. % Dy3þ-doped TZN glass when the
1300 nm pump threshold of 15 kW/cm2 is exceeded [81].
Since tellurite glass is able to dissolve high concentrations of rare-earth, it is
ideally suited for integrated optics, as was recently demonstrated by depositing
phosphate-modified tellurite glasses in thin-film waveguide form onto a GaAs
substrate. The integrated device was pumped with an on-chip 980 nm source, and
the amplification gain was reported to be ~2 dB/cm in the 1530 nm region, showing
promise for integrated waveguide lasers and amplifiers [82]. Another aspect of
integrated device development was to demonstrate the efficacy of Er3þ-doped
tellurite glass-polymer waveguides for 1.5 μm lasing and amplification. The inte-
gration of tellurite glass with polymer yields a more structurally compatible wave-
guide, because of the smaller difference in the coefficient of thermal
expansion [83].
6 Lasers Utilising Tellurite Glass-Based Gain Media 127

6.6 Conclusions

Several demonstrations of lasers utilising tellurite glass as the gain medium host in
the form of bulk glass, fibre and microspheres have been reported when doped with
the rare-earth ions Nd3þ, Er3þ, Tm3þ and Ho3þ, operating in the 1 μm, 1.5 μm and
2 μm wavelength regions, in CW, Q-switched and mode-locked operation. To date,
the maximum output power that has been achieved using a tellurite glass laser is
1.12 W using a double-clad tellurite fibre. Being a relatively soft glass with a low
melting temperature, usually in the range 700–1000  C, tellurite glass is unable to
support the laser power levels of silica glass; however, their ability to be heavily
doped offers flexibility in terms of doping schemes and dimensions. The high
nonlinearity of tellurite glass has been exploited to produce fibre Raman and
supercontinuum laser sources offering tunability in the 1495–2650 nm and
789–4870 nm wavelength ranges, respectively.

References

1. R.A.H. El-Mallawany, Tellurite Glasses Handbook (CRC Press, Boca Raton, FL, 2000)
2. P. Joshi, B. Richards, A. Jha, Reduction of OH
ions in tellurite glasses using chlorine and
oxygen gases. J. Mater. Res. 28(23), 3226–3233 (2013)
3. J. Massera et al., Processing of tellurite-based glass with low OH content. J. Am. Ceram. Soc.
94(1), 71–77 (2011)
4. H. Ebendorff-Heidepriem et al., Extruded tellurite glass and fibers with low OH content for
mid-infrared applications. Opt. Mater. Express 2(4), 432–442 (2012)
5. A. Jha, S. Shen, M. Naftaly, Structural origin of spectral broadening of 1.5-μm emission in
Er3þ-doped tellurite glasses. Phys. Rev. B 62(10), 6215–6227 (2000)
6. N. Lei, B. Xu, Z.H. Jiang, Ti:sapphire laser pumped Nd:tellurite glass laser. Opt. Commun.
127(4–6), 263–265 (1996)
7. A. Mori, Y. Ohishi, S. Sudo, Erbium-doped tellurite glass fibre laser and amplifier. Electron.
Lett. 33(10), 863–864 (1997)
8. M.J.F. Digonnet, in Rare-Earth-Doped Fiber Lasers and Amplifiers: Second Edition, Revised
and Expanded, ed. by M.J.F. Digonnet. Continuous-Wave Silica Fiber Lasers (Marcel Dekker,
Inc., New York, 2001)
9. J.F. Wu et al., Efficient thulium-doped 2-μm germanate fiber laser. IEEE Photonics Technol.
Lett. 18(1–4), 334–336 (2006)
10. J.F. Wu et al., Highly efficient high-power thulium-doped germanate glass fiber laser. Opt.
Lett. 32(6), 638–640 (2007)
11. P. France, Optical Fibre Lasers and Amplifiers (Blackie and Son Ltd., Glasgow, London,
1991)
12. W.T. Silfvast, Laser Fundamentals, 2nd edn. (Cambridge University Press, Cambridge, 2004)
13. F.Z. Qamar, T.A. King, Self-induced pulsations, Q-switching and mode-locking in Tm-silica
fibre lasers. J. Mod. Opt. 52(7), 1031–1043 (2005)
14. M. Morin, R. Larose, F. Brunet, in Rare Earth Doped Fiber Lasers and Amplifiers: Second
Edition, Revised and Expanded, ed. by M.J.F. Digonnet. Q-Switched Fiber Lasers (Mercel
Dekker Inc., New York, 2001)
15. M.J. Weber, J.D. Myers, D.H. Blackburn, Optical properties of Nd3þ in tellurite and
phosphotellurite glasses. J. Appl. Phys. 52(4), 2944–2949 (1981)
128 B.D.O. Richards and A. Jha

16. M. Naftaly, S. Shen, A. Jha, Tm3þ-doped tellurite glass for a broadband amplifier at 1.47 μm.
Appl. Optics 39(27), 4979–4984 (2000)
17. R. El-Mallawany et al., Study of luminescence properties of Er3þ-ions in new tellurite glasses.
Opt. Mater. 26(3), 267–270 (2004)
18. J.C. Michel, D. Morin, F. Auzel, Spectroscopic properties and laser effect of tellurite and
phosphate glasses strongly doped with neodyme. Rev. Phys. Appl. 13(12), 859–866 (1978)
19. H. Kalaycioglu et al., Lasing at 1065 nm in bulk Nd3þ-doped telluride-tungstate glass. Opt.
Commun. 281(24), 6056–6060 (2008)
20. A. Miguel et al., Spectral study of the stimulated emission of Nd3þ in fluorotellurite bulk glass.
Opt. Express 21(8), 9298–9307 (2013)
21. M.J.V. Bell et al., Laser emission of a Nd-doped mixed tellurite and zinc oxide glass. J. Opt.
Soc. Am. B 31(7), 1590–1594 (2014)
22. I. Iparraguirre et al., Laser action and upconversion of Nd3þ in tellurite bulk glass. J. Non
Cryst. Solids 353(8–10), 990–992 (2007)
23. J.S. Wang et al., Neodymium-doped tellurite single-mode fiber laser. Opt. Lett. 19(18),
1448–1449 (1994)
24. K. Sasagawa et al., Nd-doped tellurite glass microsphere laser. Electron. Lett. 38(22),
1355–1357 (2002)
25. J.J. Dong et al., Dual-pumped tellurite fiber amplifier and tunable laser using Er3þ/Ce3þ
codoping scheme. IEEE Photonics Technol. Lett. 23(11), 736–738 (2011)
26. S. Shen, B. Richards, A. Jha, Enhancement in pump inversion efficiency at 980 nm in Er3þ,
Er3þ/Eu3þ and Er3þ/Ce3þ doped tellurite glass fibers. Opt. Express 14(12), 5050–5054
(2006)
27. M. Irannejad et al., Erbium-ion-doped tellurite glass fibers and waveguides-devices and future
prospective: part II. Int. J. Appl. Glass Sci. 4(3), 202–213 (2013)
28. X. Peng et al., Fiber-taper-coupled L-band Er3þ-doped tellurite glass microsphere laser. Appl.
Phys. Lett. 82(10), 1497–1499 (2003)
29. X. Peng et al., Temperature dependence of the wavelength and threshold of fiber-taper-coupled
L-band Er3þ-doped tellurite glass microsphere laser. Appl. Phys. Lett. 83(26), 5380–5382
(2003)
30. M. Pollnau, S.D. Jackson, in Solid-State Mid-Infrared Laser Sources, ed. by I.T. Sorokina,
K.L. Vodopyanov. Mid-Infrared Fiber Lasers (Springer-Verlag, Berlin, 2003)
31. F. Fusari et al., Spectroscopic and lasing performance of Tm3þ-doped bulk TZN and TZNG
tellurite glasses operating around 1.9 μm. Opt. Express 16(23), 19146–19151 (2008)
32. F. Fusari et al., Tunable laser operation of a Tm3þ-doped tellurite glass laser near 2 μm
pumped by a 1211 nm semiconductor disk laser. Opt. Mater. 32(9), 1007–1010 (2010)
33. B. Richards et al., Infrared emission and energy transfer in Tm3þ, Tm3þ-Ho3þ and Tm3þ-Yb3þ
-doped tellurite fibre. Opt. Express 15(11), 6546–6551 (2007)
34. B. Richards et al., ~2 μm Tm3þ/Yb3þ-doped tellurite fibre laser. J. Mater. Sci. Mater. Electron.
20, 317–320 (2009)
35. B. Richards et al., Efficient ~2 μm Tm3þ-doped tellurite fiber laser. Opt. Lett. 33(4), 402–404
(2008)
36. S.D. Jackson, Cross relaxation and energy transfer upconversion processes relevant to the
functioning of 2 μm Tm3þ-doped silica fibre lasers. Opt. Commun. 230(1–3), 197–203 (2004)
37. T. Yamamoto, Y. Miyajima, T. Komukai, 1.9-μm Tm-doped silica fiber laser-pumped at 1.57-
μm. Electron. Lett. 30(3), 220–221 (1994)
38. R.M. Percival et al., A 1.6-μm pumped 1.9-μm thulium-doped fluoride fiber laser and amplifier
of very high-efficiency. IEEE J. Quantum Electron. 31(3), 489–493 (1995)
39. B. Richards, Tellurite glass mid-infrared (1.9–2.1 um) fibre lasers (Institute for Materials
Research, University of Leeds, 2008)
40. K.F. Li, G.A. Zhang, L.L. Hu, Watt-level ~2 μm laser output in Tm3þ-doped tungsten tellurite
glass double-cladding fiber. Opt. Lett. 35(24), 4136–4138 (2010)
6 Lasers Utilising Tellurite Glass-Based Gain Media 129

41. K.F. Li et al., Tm3þ and Tm3þ-Ho3þ co-doped tungsten tellurite glass single mode fiber laser.
Opt. Express 20(9), 10115–10121 (2012)
42. S. Gao et al., ~2 μm single-mode laser output in Tm3þ-doped tellurium germanate double-
cladding fiber. IEEE Photonics Technol. Lett. PP(99), 1–1 (2015)
43. C.J. Hill, A. Jha, Development of novel ternary tellurite glasses for high temperature fiber optic
mid-IR chemical sensing. J. Non Cryst. Solids 353(13–15), 1372–1376 (2007)
44. X. Jiang et al., Investigation on germanium oxide-based glasses for infrared optical fibre
development. Opt. Mater. 31(11), 1701–1706 (2009)
45. B. Richards et al., Tellurite glass lasers operating close to 2 mm. Laser Phys. Lett. 7(3),
177–193 (2010)
46. B.M. Walsh et al., Optical properties of Tm3þ ions in alkali germanate glass. J. Non Cryst.
Solids 352(50–51), 5344–5352 (2006)
47. S. Shen et al., The effect of composition on the emission spectra of Tm3þ-doped tellurite
glasses for an S-band amplifier operating at 1.46 mm. J. Non Cryst. Solids 326, 510–514
(2003)
48. J.F. Wu, S.B. Jiang, N. Peyghambarian, 1.5-μm-band thulium-doped microsphere laser orig-
inating from self- terminating transition. Opt. Express 13(25), 10129–10133 (2005)
49. Y. Tsang et al., Tm3þ/Ho3þ codoped tellurite fiber laser. Opt. Lett. 33(11), 1282–1284 (2008)
50. F. Fusari et al., Femtosecond mode-locked Tm3þ and Tm3þ-Ho3þ doped 2 μm glass lasers.
Opt. Express 18(21), 22090–22098 (2010)
51. Y. Tsang et al., A Yb3þ/Tm3þ/Ho3þ triply-doped tellurite fibre laser. Opt. Express 16(14),
10690–10695 (2008)
52. S.D. Jackson, Recent advances in high-power silica-fibre lasers operating at 2 μm. SPIE 6801
(2008)
53. S.D. Jackson, F. Bugge, G. Erbert, High-power and highly efficient diode-cladding-pumped
Ho3þ-doped silica fiber lasers. Opt. Lett. 32(22), 3349–3351 (2007)
54. S.D. Jackson, Midinfrared holmium fiber lasers. IEEE J. Quantum Electron. 42(1–2), 187–191
(2006)
55. A. Hemming et al., High power operation of cladding pumped holmium-doped silica fibre
lasers. Opt. Express 21(4), 4560–4566 (2013)
56. C. Yao et al., Holmium-doped fluorotellurite microstructured fibers for 2.1 mm lasing. Opt.
Lett. 40(20), 4695–4698 (2015)
57. A. Mori et al., Ultra-wide-band tellurite-based fiber Raman amplifier. J. Lightwave Technol.
21(5), 1300–1306 (2003)
58. R. Stegeman et al., Tellurite glasses with peak absolute Raman gain coefficients up to 30 times
that of fused silica. Opt. Lett. 28(13), 1126–1128 (2003)
59. R. Jose et al., Tailoring of Raman gain bandwidth of tellurite glasses for designing gain-
flattened fiber Raman amplifiers. J. Opt. Soc. Am. B 25(3), 373–382 (2008)
60. V.V.R.K. Kumar et al., Tellurite photonic crystal fiber. Opt. Express 11(20), 2641–2645
(2003)
61. G.S. Qin et al., Widely tunable ring-cavity tellurite fiber Raman laser. Opt. Lett. 33(17),
2014–2016 (2008)
62. M.Y. Koptev et al., Widely tunable mid-infrared fiber laser source based on soliton self-
frequency shift in microstructured tellurite fiber. Opt. Lett. 40(17), 4094–4097 (2015)
63. P. Domachuk et al., Over 4000 nm bandwidth of mid-IR supercontinuum generation in
sub-centimeter segments of highly nonlinear tellurite PCFs. Opt. Express 16(10), 7161–7168
(2008)
64. M.S. Liao et al., Tellurite microstructure fibers with small hexagonal core for supercontinuum
generation. Opt. Express 17(14), 12174–12182 (2009)
65. M.S. Liao et al., A highly non-linear tellurite microstructure fiber with multi-ring holes for
supercontinuum generation. Opt. Express 17(18), 15481–15490 (2009)
66. D. Buccoliero et al., Thulium pumped high power supercontinuum in loss-determined opti-
mum lengths of tellurite photonic crystal fiber. Appl. Phys. Lett. 97(6), 061106–(1–3) (2010)
130 B.D.O. Richards and A. Jha

67. M.S. Liao et al., Five-order SRSs and supercontinuum generation from a tapered tellurite
microstructured fiber with longitudinally varying dispersion. Opt. Express 19(16),
15389–15396 (2011)
68. M.S. Liao et al., Supercontinuum generation in short tellurite microstructured fibers pumped
by a quasi-CW laser. Opt. Lett. 37(11), 2127–2129 (2012)
69. G. Ghosh, Sellmeier coefficients and chromatic dispersions for some tellurite glasses. J. Am.
Ceram. Soc. 78(10), 2828–2830 (1995)
70. V. Ta’eed et al., Ultrafast all-optical chalcogenide glass photonic circuits. Opt. Express 15(15),
9205–9221 (2007)
71. X. Zhu, R. Jain, 10-W-level diode-pumped compact 2.78 μm ZBLAN fiber laser. Opt. Lett. 32
(1), 26–28 (2007)
72. S.D. Jackson, Continuous wave 2.9 μm dysprosium-doped fluoride fiber laser. Appl. Phys.
Lett. 83(7), 1316–1318 (2003)
73. T. Sumiyoshi et al., High-power continuous-wave 3- and 2-μm cascade Ho3þ:ZBLAN fiber
laser and its medical applications. IEEE J. Sel. Top Quantum Electron. 5(4), 936–943 (1999)
74. J. Schneide, C. Carbonnier, U.B. Unrau, Characterization of a Ho3þ-doped fluoride fiber laser
with a 3.9-μm emission wavelength. Appl. Optics 36(33), 8595–8600 (1997)
75. S.D. Jackson, T.A. King, CW operation of a 1.064-μm pumped Tm-Ho-doped silica fiber laser.
IEEE J. Quantum Electron. 34(9), 1578–1587 (1998)
76. H. Jiang, L. Zhang, Y. Feng, Silica-based fiber Raman laser at > 2.4 mm. Opt. Lett. 40(14),
3249–3252 (2015)
77. Y. Guo et al., Er3þ-doped fluoro-tellurite glass: A new choice for 2.7 μm lasers. Mater. Lett.
80, 56–58 (2012)
78. H. Zhan et al., Intense 2.7 mm emission of Er3þ-doped water-free fluorotellurite glasses. Opt.
Lett. 37(16), 3408–3410 (2012)
79. H. Zhong et al., 2.7 μm emission of Nd3þ, Er3þ codoped tellurite glass. J. Appl. Phys. 106(8),
083114 (2009)
80. B.D.O. Richards et al., Mid-IR (3–4 μm) fluorescence and ASE studies in Dy3þ doped tellurite
and germanate glasses and a fs laser inscribed waveguide. Laser Phys. Lett. 10(8), 085802
(2013)
81. L. Gomes et al., Spectroscopy of mid-infrared (2.9 mm) fluorescence and energy transfer in Dy
3þ
-doped tellurite glasses. J. Opt. Soc. Am. B 31(3), 429–435 (2014)
82. M. Irannejad et al., Active glass waveguide amplifier on GaAs by UV-pulsed laser deposition
and femtosecond laser inscription. Laser Phys. Lett. 9(5), 329 (2012)
83. Z. Zhanxiang et al., Active glass–polymer superlattice structure for photonic integration.
Nanotechnology 23(22), 225302 (2012)
Chapter 7
Tellurite Glasses for Optical Amplifiers

Luiz Carlos Barbosa, Cicero Omegna Filho, and Enver Fernandez Chillcce

7.1 Introduction

Tellurite glasses present interesting physical and optical properties, such as high
rates of linear and nonlinear refraction, low melting points, low and medium
frequencies of phonons, and especially high solubility for rare-earth ions. Tellurite
glasses became potentially important for the manufacture of optical amplifiers with
high bandwidth for optical communications in the region of infrared spectrum. The
high solubility property for rare-earth ions, especially the erbium ions, allows the
manufacture of optical fiber amplifiers as small as possible. There are a lot of
studies published with almost all rare-earth ions, such as tellurite glass doped with
erbium, ytterbium, praseodymium, thulium, holmium ions, and others, and were
manufactured and put in optical amplifier trade with large broadband around
wavelength of 1550 nm involving tellurite glasses co-doped with erbium and
thulium ions. Of great importance was the manufacturing method, mainly rod in
tube method of fiber with a double shell. Lately, micro-structured optical fiber
amplifiers doped with erbium ions have been developed.

L.C. Barbosa (*) • E.F. Chillcce

Departamento de Eletrônica Qu^antica, Instituto de Fı́sica Gleb Wataghin, UNICAMP –
Universidade Estadual de Campinas, Rua Sêrgio Buarque de Holanda, 777 Cidade
Universitária, 13083-859 Campinas, SP, Brazil
e-mail: [email protected]; [email protected]
C.O. Filho
Luxtec Optical System, Rua Dr. Souza Brito, 126 / 483, 13040-012 Campinas, SP, Brazil
e-mail: [email protected]

© Springer International Publishing AG 2017 131

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_7
132 L.C. Barbosa et al.

7.2 Tellurite Glasses for Optical Amplifiers

Tellurite glasses have some properties that make them attractive for manufacture of
optical devices based on rare-earth-doped optical fibers such as optical amplifiers
and fiber lasers [1–7]. The reason is that when tellurite glasses are compared with
silica glasses, the tellurite glasses are better hostesses for the rare-earth ions (e.g.,
erbium). Tellurite glasses present a high solubility of rare earths which is favorable
to avoid the undesirable effects as clustering of the ions or quenching effects during
the emission process [8]. Such special characteristics of tellurite glasses have
attracted attention because of its potential use in broadband optical amplifiers for
the telecom industry [6].
Fiber lasers and optical amplifiers have been performed with conventional core-
cladding configuration rare-earth-doped tellurite fibers. The first demonstration of
optical gain and laser generation, based on Er3+-doped tellurite core/clad fibers, was
realized by Mori et al. [1].
Micro-structured optical fibers (MOF) are an interesting option to such conven-
tional fibers because they have important constructive advantages. Since the light
confinement within the fiber core is obtained through the distribution of air holes
surrounding the core, it is not necessary to use two optically different (different
linear refractive indices) but thermo-mechanically similar materials to fabricate the
fiber. This is important because problems as glass crystallization during the drawing
process are easier to avoid when the core and the clad of the fiber are made of the
same glass material.
Tellurite glasses are also adequate for manufacture of optical fiber devices based
on the nonlinear response of the material [9–16] because they have enhanced
nonlinear properties in comparison with those of silica glasses, reaching a nonlinear
refractive index that is approximately 30 times higher [8, 17].
Tellurite fibers present some drawbacks related to optical attenuation and
mechanical fragility [18, 19]. Much effort has been made to improve these two
aspects. The optical attenuation of tellurite fibers may be influenced by the material
purity and the fabrication methods (rod-in-tube, extrusion, stack-and-draw, casting,
and others) employed. Several works have reported the production of tellurite fibers
with an optical attenuation (in the spectral region around 1550 nm) between 1 and
27 dB/m [12, 20–23]. Moreover, a tellurite fiber (used in nonlinear applications)
with a low loss of 20 dB/km was also reported [10, 24].
The fact that tellurite glasses allow the doping with high concentration of rare-
earth ions enables the manufacture of optical devices with short fibers. Tellurite
glass doped with 7500 ppm Er2O3 guarantees the maximum erbium concentration
which allows the reduction of the fiber length without reducing the 4I13/2 level
lifetime [25]. This aspect makes up for the high losses presented by tellurite fibers
in comparison with silica fibers.
7 Tellurite Glasses for Optical Amplifiers 133

Fig. 7.1 (a) Schematic representation of the suction process. (1) Vacuum pump, (2) valve for air
flow control, (3–4) flexible hoses, (5) a stainless steel Die with 160 μm hole diameter to avoid
turbulent flow, (6) silica tube, and (7) Pt crucible containing the molten tellurite glass sample. (b)
Tellurite rod and tube produced by the suction method. (c) Image of a cane (OD ¼ 2 mm) produced
from a rod (OD ¼ 7 mm) using the drawing process. (d) Tellurite fiber preform produced by the
rod-in-tube technique (From R. Narro-Garcı́a, E.F. Chillcce, L.C. Barbosa, E. de Posada,
M. Arronte, E. Rodriguez, Journal of Luminescence 134, 528–532 (2013))

7.2.1 Fabrication of Core and Clad Tellurite Optical Fibers

For fabrication of tellurite glass preforms, so far several techniques based on the
glass melt-quenching method have been reported, and each one of them owns its
merits and drawbacks, respectively. The double-crucible technique was important
in the past, because it was a continuous process and therefore offers potentially very
low-cost operation. However, it has not been able to deliver the required precision
in control of fiber dimensions and characteristics [20]. This problem can be solved
by using the rod-in-tube technique which allows a better control of the fiber
dimensions.
The fabrication of tellurite glass preforms and fibers by using the rod-in-tube
technique was reported by Vogel et al. [26]. Recently, Narro-Garcı́a et al. [27] have
produced tellurite rods and tubes by using the suction method schematically
described in Fig. 7.1a. This method consists of sucking the melted glass into a
silica tube with an inner diameter (ID) ¼ 7 mm and outer diameter (OD) ¼ 10 mm
using a vacuum pump. Tellurite rods or tubes were produced depending only on the
viscosity of the molten glass sample during the suction process. To produce the
rods, the molten glass was sucked into the silica tube at a higher viscosity than that
used to fabricate the tubes. Figure 7.1b shows an image of the rods and tubes
produced by the suction method. The rod (OD ffi 7 mm) and the tube (OD ffi 7 mm,
ID ffi 3.5 mm) are approximately 15 cm long.
As described above, the suction method is an easy and quick way to obtain rods
and tubes from melted tellurite glasses. This method allows an excellent control of
the rod diameter and the external diameter of the tubes. Nevertheless, the main
problem of the suction method is the low control in the internal diameter of the
fabricated tubes. Even with this drawback, the tubes fabricated by this method are
suitable for the fabrication of conventional tellurite fibers, since the air gap formed
134 L.C. Barbosa et al.

in the interface between the core and cladding glass of the preform must be easily
removed with the application of a negative pressure (by using a vacuum pump) to
the preform during the fiber fabrication. This solution also leads to an improvement
of the interface between the core and cladding glass, which significantly reduces the
optical attenuation of the fiber.
For the production of the tellurite fibers by using the rod-in-tube technique,
Narro-Garcı́a et al. [22] have realized the following steps:
1. A 2 mm cane was drawn from the 7 mm rod (Fig. 7.1c).
2. The 2 mm cane was inserted into a tube (OD ¼ 7 mm and ID ¼ 3.5 mm) to
produce a “rod-in-tube” preform (Fig. 7.1d).
3. This preform is then drawn to generate a new 2 mm cane. Despite the processes
(1)–(3) contribute to reduce the glass stress and surface roughness, the sample
obtained after the third fabrication step must be annealed at 250  C for 24 h or
more to further improve the glass properties.
4. The new 2 mm cane is inserted into another tube (similar to step 2).
5. This new “rod-in-tube” preform is drawn to produce core and clad tellurite
fibers. Both the canes and the core and clad tellurite optical fibers (named
step-index tellurite fiber—SITF) can be produced using a soft glass drawn tower.
To produce tellurite fibers, it is very important to consider the main physical and
optical properties, such as mechanical fragility, rugosity, geometric shape, numer-
ical aperture, modal field confinement, and dispersion. The thermal stability (ΔT)
and the Hruby number (HR) are relevant thermomechanical parameters in the
tellurite fiber engineering. These parameters have direct influence in the mechanical
fragility of tellurite fibers caused by the crystallization processes during glass
formation or fiber fabrication.
Fabrication of five different SITFs was reported, by Narro-Garcı́a et al. [27],
using rod-in-tube technique and tellurite glass compositions showed in Table 7.1.
The core of the homemade SITFs is doped and co-doped with erbium and ytterbium
ions (“A” to “E” glass types). The tellurite fiber clad is the same for all five fibers,
and it is composed from the glass type “H.” The five SITFs are composed of core
and clad glasses: SITF-A (A-core/H-clad), SITF-B (B-core/H-clad), SITF-C
(C-core/H-clad), SITF-D (D-core/H-clad), and SITF-E (E-core/H-clad). The
SITFs were produced at temperatures ranging from 465 to 490  C, a preform
feeding speed of 3 mm/min, and a fiber drawing speed of 4 m/min. The high
thermal stability (>130  C) and an HR  0.96 (Table 7.1) were parameters which
favored the tellurite fiber drawing process as pointed by Narro-Garcia.
Figure 7.2 shows the cross section image of a core/clad tellurite fiber and the
light guided by the same fiber. Figure 7.2a, b show the cross section of a tellurite
fiber type SITF-A [22], approximately 180 μm external diameter and 8 μm core
diameter, produced by using rod-in-tube technique. Figure 7.2c, d show the 2D and
3D images, respectively, of a supercontinuum light guided by the SITF-A. The
supercontinuum light ranges from 600 to 1750 nm approximately. The 2D and 3D
images were obtained by using an Infrared camera.
Table 7.1 Glass composition and optical and physical properties of the studied glass samples
Glass composition Thermal, physical, and optical properties
Er2O3 n(λ [nm])
Sample Glass base (ppm) Yb2O3 (ppm) Tg ( C) Tx ( C) Tm ( C) ΔT ¼ (Tx  Tg) HR Density (g/cm3) n(λ1) n(λ2) n(λ3)
A core 10,000 0 377 510 648 133 0.96 5.93 2.1441 2.0842 2.0769
7 Tellurite Glasses for Optical Amplifiers

B core 5000 20,000 372 537 676 165 1.19 6.07 2.1393 2.08 2.0729
C core 10,000 20,000 389 550 680 161 1.24 6.06 2.1463 2.0851 2.0782
D core 15,000 20,000 381 520 644 139 1.12 5.99 2.1309 2.0734 2.0669
E core 10,000 40,000 384 540 676 156 1.15 6.14 2.1345 2.0759 2.069
H Clad 0 0 348 498 598 150 1.5 5.85 2.1276 2.0693 2.0629
Core glass base (A–E): [69TeO2-24WO3-3Nb2O5-4Na2O](mol%) +2000 ppmAl2O3 (wt%). Clad glass (H): [72.5TeO2-20WO3-1.5Nb2O5-6Na2O] (mol%)
+2000 ppm Al2O3 (wt%). Measured thermal parameters: Tg—glass transition temperature, Tx— temperature for crystallization onset, and Tm—melting
temperature. The refractive index was measured at λ1 ¼ 632.8 nm, λ2 ¼ 1305.4 nm, and λ3 ¼ 1536.0 nm
Source: R. Narro-Garcı́a, E.F. Chillcce, L.C. Barbosa, E. de Posada, M. Arronte, E. Rodriguez, Journal of Luminescence 134, 528–532 (2013)
135
136 L.C. Barbosa et al.

Fig. 7.2 (a) cross section image of an SITF-A obtained by using an optical microscope. (b) Core
region amplified of SITF-A. (c) 2D image and (d) 3D images of the core and clad optical modes
guided by the SITF-A (From R. Narro-Garcia, H. Desirena, E. F. Chillcce, L. C. Barbosa,
E. Rodriguez and E. De La Rosa, Opt. Comm 93–101 (2014))

7.2.2 Fabrication of the Er3+-Doped Tellurite

Micro-Structured Optical Fiber
7.2.2.1 Tellurite Rod, Tube, and Capillary Fabrication

Tellurite rods and tubes can be produced by the centrifugation method as pointed by
Chillcce et al. [28]. Firstly, the melted glass (in a platinum crucible) is sucked inside
a silica tube by means of a vacuum pump, and this quickly solidifies as a tube or a
rod (Fig. 7.1). In both cases, it presents low optical quality, with the presence of
frozen bubbles. In the case of a tube, it also gives a varying internal diameter. The
second stage consists of holding the silica tube (with the tellurite glass inside it) in a
drilling chuck (Fig. 7.3a). A flame torch with planar and uniform temperature
distribution is slowly, and precisely, positioned close to the silica tube, which is
then heated while rotating at high speed. The initial formation process of a tellurite
glass tube is illustrated in Fig. 7.3a, for a temperature above soft point (Ts) and a low
centrifugation speed (approximately 750 rpm). It is possible to see (arrow) the
internal hole being formed; note two horizontal lines on the left-hand side of the
tube. After the process is initiated, the rotation speed is increased (approximately
1800 rpm) in order to allow the completion of the tube formation, as observed in
Fig. 7.3b.
7 Tellurite Glasses for Optical Amplifiers 137

Fig. 7.3 (a) Initial

formation process of the
tellurite glass tube. Note the
two horizontal lines on the
left-hand side of the tube.
(b) Tellurite glass tube.
Note that the two horizontal
lines are now all the way
through the tube (From E. F.
Chillcce, C. M. B. Cordeiro,
L. C. Barbosa, and C. H.
Brito Cruz, J. Non-Cryst.
Solids 352(32–35),
3423–3428 (2006))

This method enables the thickness of the tube to be controlled via the maximum
rotation speed. As the thermal expansion coefficients of both tubes (the outside
silica tube and the inner tellurite tube) are different, it is possible to remove the
internal formed tube without cracking it. The process relies on removing the silica
plus tellurite tube fast enough from the flame to allow the tellurite to contract and
release from the silica, but not so fast that it breaks it due to thermal shock.
Figure 7.4 shows images of tellurite tubes (7 and 10 mm outside diameters)
prepared through suction technique (Fig. 7.4a) and centrifugation technique
(Fig. 7.4b). The tellurite tubes (7 mm outside diameter) made by using suction
give a varying internal diameter, bubbles, and rugose surface. On the other hand,
the tellurite tubes (7 and 10 mm outside diameters) made by using the centrifuga-
tion technique show high optical and geometrical quality: smooth surface, no frozen
bubbles, and uniform external and internal diameter.
The next stage is to draw canes and capillaries from the rod and formed tubes
(Fig. 7.5). Figure 7.5a, b show some tellurite capillaries that were drawn (700 μm
diameter) from a tube made by centrifugation method. Firstly, they need to be fixed
in a handle to increase their length (Fig. 7.5c). The handle consists of a silica tube
and a brass connector to hold the tellurite tube.
The rods and capillaries are produced by using a soft glass optical fiber tower at
draw temperature being around 520  C. In this case, the tellurite tube is based on the
composition 71TeO2-22.5WO3-5Na2O-1.5Nb2O5 (% mol). As in the first stage
(tube formation), no crystallization problems occur (because HR > 2.3 for this
tellurite tube), and the capillaries show the same quality as the original tubes.
138 L.C. Barbosa et al.

Fig. 7.4 Tellurite tubes

made by using suction
(a) and centrifugation
(b) techniques

Fig. 7.5 (a) High-quality tellurite glass tube (10 mm diameter, 120 mm long) produced by
centrifugation method and several capillaries; (b) front view shows tube, capillaries, and rod;
(c) system to hold a short tellurite tube inside the furnace (From E. F. Chillcce, C. M. B. Cordeiro,
L. C. Barbosa, and C. H. Brito Cruz, J. Non-Cryst. Solids 352(32–35), 3423–3428 (2006))

7.2.2.2 Tellurite Micro-Structured Preform Fabrication

The tellurite micro-structured optical fibers (MOF) can be produced by using the
stack-and-draw technique. This technique is commonly used to produce silica
photonic crystal fibers (PCF) [29]. This technique consists of the following: First,
the capillaries and rods are hand stacked to form regular (or periodic) structure.
7 Tellurite Glasses for Optical Amplifiers 139

Fig. 7.6 Step sequence of hand stacking of capillaries and rods to produce a soft glass photonic
crystal fiber of five capillaries ring around the solid core

Fig. 7.7 Micro-structured fiber preforms made of silica (a–d), borosilicate (e), and tellurite
glasses (f)

Figure 7.6 shows a step sequence to produce a solid core photonic crystal fiber
preform.
Second, the whole set (the stacked capillaries and rods) is inserted in a jacket
tube as observed in Fig. 7.7. This filled jacket is commonly named as preform.
Figure 7.7 shows diverse photonic preforms fabricated by hand stacking of capil-
laries and rods. Usually preforms can be made from glasses such as silica
(Fig. 7.7a–d), borosilicate (Fig. 7.7e), and tellurite (Fig. 7.7f).
140 L.C. Barbosa et al.

Finally, this preform is then drawn directly to a micro-structured fiber by using a

fiber drawn tower.

7.2.2.3 Tellurite Micro-Structured Optical Fiber Fabrication

Tellurite micro-structured optical fibers can be fabricated extrusion and stack-and-

draw techniques. In the extrusion technique, the melted glass is forced against a dye
with proper geometry [20]. Contamination problems, the difficulty of preparing
fibers with different glasses and/or complex geometries, and the importance of
properly designing the dye are just some of the drawbacks of extrusion technique.
To overcome these problems, the stack-and-draw technique was proposed to
“homemade” glasses, particularly to tellurite glasses based on the composition of
71TeO2-22.5WO3-5Na2O-1.5Nb2O5.
Figure 7.8 shows the cross section images of the preforms and the MOF
produced by the stack-and-draw technique [30].

Fig. 7.8 Cross section images of tellurite MOF preforms and fibers. (a) Preform with the air
regions indicated by the letter “a”. (b) Preform similar to that of (a) but after the vacuum process is
applied. Note that the air regions are now full of solid tellurite glass. (c) SEM image of the cross
section of an MOF. (d) Detailed SEM image of the core region of an MOF (From E. F. Chillcce,
R. Narro-Garcı́a, J. W. Menezes, E. Rodriguez, D. Marconi, H. L. Fragnito, and L. C. Barbosa,
Proc. SPIE vol. 8257, 82570B-1/82570B-10 (2012))
7 Tellurite Glasses for Optical Amplifiers 141

The MOF can be fabricated by the following two steps: First, a structured
preform cane with a similar structure to that of the MOF must be produced.
Figure 7.8a shows the cross section image of an Er3+-doped tellurite MOF preform
with its core rod surrounded by 18 undoped capillaries, whole stacked inside a
tellurite jacket tube (defined as the first clad). In this preform cane (external
diameter ~1 mm), there are air gaps between the adjacent capillaries, between
capillaries and the central rod (fiber core intentionally doped with 7500 ppm of
Er2O3), and between capillaries and the first clad. These air regions are indicated in
Fig. 7.1a by the letter “a.” Figure 7.8b shows the cross section image of a preform,
similar to that of Fig. 7.8a, after the air regions were evacuated by using a vacuum
pump during the drawing process, as described in ref. [28]. The preform obtained
after this process also has an external diameter of ~1 mm.
Second, this preform cane must be placed inside another tellurite tube (defined as
the second clad, with their external and internal diameters of ~7 and 2 mm,
respectively). To add this second clad to the preform without any discontinuity in
the material, but preserving the structure intact, two pressure processes are applied
simultaneously. On the one hand, the MOF preform holes are pressurized with
nitrogen gas (maximum pressure of 150 mbar). On the other hand, in the region
between the first clad of the MOF perform and the second clad, a vacuum is applied.
Both pressure processes are executed simultaneously at ~520  C. The pressure in
the holes prevented them from collapsing when the vacuum is applied in the region
between the first clad of the MOF preform and the second clad, making the second
clad bind to the external surface of the MOF preform. Finally, this hole (the external
tube and the preform cane “inside it”) is drawn to obtain the final tellurite MOF.
Figure 7.8c shows a scanning electron microscope (SEM) image of the trans-
versal section of the structured region of the fiber, composed as expected of
18 holes and a solid core. Figure 7.8d shows an image with details of the transversal
section that corresponds to the region of the MOF core. This last image shows that
the MOF core remains suspended by thin membranes (thickness <1 μm) which
extend along the fiber. The obtained tellurite MOF has ~350 μm of external
diameter and ~5 μm of core diameter.
Figure 7.9 shows images of the light intensity distribution (taken with a near-
infrared optical camera) in different regions of the transversal section of the tellurite
MOF. The light source has a wavelength of 980 nm. It was coupled to the MOF
solid core by using a silica single mode fiber (SMF, core diameter ~8 μm).
Figure 7.9a shows the light intensity distribution when the spot of the light source
covers the entire transversal section of the MOF. Note that the region that surrounds
the hollow structure, which was defined as first clad, guides the light. The most
likely reason for such a behavior is the mechanical stress concentrated in this
region. This stress produces a slight increase in the refractive index of the material
of the first clad in relation to the surrounding material, allowing the light guidance.
It is also possible to observe that the interstitial regions among the holes act as small
cores that can confine and guide the light. Figure 7.9b shows the light intensity
142 L.C. Barbosa et al.

Fig. 7.9 Light intensity distribution images of the transversal section of a tellurite MOF. (a) When
the light covers the whole transversal section of the MOF. (b) When the light is efficiently coupled
to the MOF core. (c) 3D representation of the light intensity distribution in the MOF core shown
in (b)

distribution in the MOF core region, when the light was efficiently coupled to the
core. In such a situation, neither light intensity distribution in the first clad nor in
the interstitial regions was observed. Finally, Fig. 7.9c shows the Gaussian shape, of
the light intensity distribution in the core region, in a three-dimensional plot.

7.3 Optical Attenuation of Tellurite Fibers

Optical attenuation of tellurite fibers has been reported, in several works, to be

between 1 and 27 dB/m (in the spectral region from around 1310, 1400, and
1550 nm) [12, 20–23]. Moreover, a tellurite fiber (used in nonlinear applications)
with a low loss of 20 dB/km was also reported [10, 24]. Figure 7.10a shows the
attenuation spectra of a core/clad tellurite fiber [31]. Its core diameter is 7 μm and
the cladding diameter is 200 μm. The fiber was designed with a core composition of
69TeO2-24WO3-4Na2O-3Nb2O5 + 2000ppmAl2O3 and the cladding 72.5TeO2-
20WO3-6Na2O-1.5Nb2O5 + 2000 ppm Al2O3. We can notice the high loss of the
TeO2 fiber, 7 dB/m or 7000 dB/km compared with the loss of the silica glass fiber
SMF-28 which is about 0.35 dB/km around 1342 nm. The loss in the tellurite fiber
in that region is about 20,000 times higher. Figure 7.10b shows the attenuation
spectrum (ranging from 600 to 1750 nm) of a 12 cm long Er3+-doped tellurite MOF
shown in Fig. 7.9 [30]. The absorption bands, 4I15/2 ! 4F9/2(655 nm), 4I15/2 ! 4I9/
2(796 nm), I15/2 ! I11/2(977 nm) and I15/2 ! I13/2(1526 nm), correspond to Er
4 4 4 4 3+

ion transitions. The minimum of the attenuation, around 1350 nm, is approximately
3 dB. On the other hand, the maximum of the attenuation, in the region around
1530 nm, is approximately 25 dB.
7 Tellurite Glasses for Optical Amplifiers 143

Fig. 7.10 Tellurite optical fiber attenuation spectra. (a) Core/clad fiber made of TeO2-WO3-
Na2O-Nb2O5 glass [31]. (b) A 12 cm-long Er3+-doped tellurite MOF [30] (From A. Traore, PhD
Thesis, University of Maryland, Baltimore County, USA (2011) and E. F. Chillcce, R. Narro-
Garcı́a, J. W. Menezes, E. Rodriguez, D. Marconi, H. L. Fragnito, and L. C. Barbosa, Proc. SPIE
vol. 8257, 82570B-1/82570B-10 (2012))

7.4 Amplified Spontaneous Emission, Gain,

and Laser Generation

To produce amplified spontaneous emission, gain, and laser, it is common to use a

two-pump scheme in a co-propagating/counter-propagating configuration, typically
a setup such as shown in Fig. 7.11a [30, 32]. In the first one, both pumps can be
980 nm lasers, with maximum optical power (for the co-propagating/counter-
propagating directions). In the second one, the co-propagating pump can be a
980 nm laser, and the counter-propagating pump a 1480 nm laser, with maximum
optical powers.
In both pump laser configurations, a wavelength division multiplexing coupler
(WDM 980/1550 nm) can be used to couple the co-propagating pump laser and the
probe signal to the tellurite MOF end. This probe signal is used only for the gain
measurements. At the other tellurite MOF end, another WDM coupler (980/1550 or
1480/1550 nm) is connected. The 980 (or 1480) nm arm is used to couple the
counter-propagating pump laser. The 1550 nm arm is used to allow the output
spectra measurements to be taken by using an optical spectrum analyzer (OSA).
The WDM couplers connectorized with SC/APC or FC/APC connectors are
used, making it necessary to use connectorized patch cords of standard fiber (SF) or
single mode fiber (SMF) to guide the light up to the tellurite MOF.
Figure 7.11b shows the image of an 11 cm tellurite MOF (it is the same of
Fig. 7.9, core diameters ~5 μm) mounted on the MPS and excited with the 980 nm
pump lasers. From this image, it is possible to observe an intense green emission
along the tellurite MOF that was generated by the up-conversion process. As it is
well known, the difference between the core diameters of the tellurite MOF (~5 μm)
and the SF (~9 μm) produces elevated coupling losses. To reduce such losses, a 1 m
highly nonlinear fiber (HNLF) with a core diameter of ~4 μm is used as an
144 L.C. Barbosa et al.

Fig. 7.11 (a) Schematic representation of the experimental setup used for measuring the gain
spectra and the laser generation. (b) Image showing the microposition stages and the green
emission produced, in the MOF, by the up-conversion process. (c) Amplified image of the
coupling between the HNLF and tellurite MOF when both fiber cores are matched

intermediate fiber. Two microposition stages (MPS) are used to match the cores of
the HNLF and of the tellurite MOF. With this arrangement, it is possible to reduce
the coupling losses, in each fiber end (HNLF/MOF or MOF/HNLF), to 2 or 3 dB.
Figure 7.11c shows a zoom of the connection between the HNLF and the
tellurite MOF ends, obtained by using an optical microscope. This image clearly
shows the green emission generated along the tellurite MOF as it is excited with the
980 nm (or 1480 nm) sources. The ends of the MOF are cleaved by using a hand
diamond cleaver, perpendicular (or almost) to the fiber axis as shown in
Fig. 7.11c. These straight facets allow obtaining the reflection cavity for the laser
generation.
Figure 7.12 shows the amplified spontaneous emission (ASE) power spectra of
the Er3+-doped tellurite MOFs [30]. The pump powers, in the co-propagating/
counter-propagating directions (Fig. 7.11a), are intentionally adjusted during the
emission measurements in order to avoid laser generation. The ASE power spectra
of Er3+-doped tellurite MOFs (fiber lengths ranging from 5 to 16 cm) are obtained
by using an OSA (bandwidth resolution (BWR) ¼ 10 nm) and a 980 nm/980 nm
7 Tellurite Glasses for Optical Amplifiers 145

Fig. 7.12 The ASE power

spectra of Er3+-doped
tellurite MOFs. The tellurite
MOF lengths range from
5 to 16 cm (From E. F.
Chillcce, R. Narro-Garcı́a,
J. W. Menezes,
E. Rodriguez, D. Marconi,
H. L. Fragnito, and L. C.
Barbosa, Proc. SPIE vol.
8257, 82570B-1/82570B-10
(2012))

-40 -34
(a) (b)
-36
-42
Power (dB)
Power (dB)

-38
-40
-44
-42

-46 -44
-46
-48 -48
1515 1530 1545 1560 1575 1590 1605 1515 1530 1545 1560 1575 1590
Wavelength (nm) Wavelength (nm)

Fig. 7.13 The ASE spectra of (a) and (b) correspond to the 980 nm/980 nm and 980 nm/1480 nm
pump schemes, respectively (From E. F. Chillcce, R. Narro-Garcı́a, J. D. Marconi, J. W. Menezes,
M. F. Ando, E. Rodriguez, H. L. Fragnito, and L. C. Barbosa, Latin America Optics and Photonics
Technical Digest, OSA, LM1C.4 (2012))

pump power scheme. From these spectra, it is possible to observe that the ASE
power maxima vary on functions of the MOF lengths and of the wavelength. For
MOFs with lengths of 5 and 8 cm, the ASE power maxima occur in the 1535 nm
wavelength region, but for MOFs with lengths of 12 and 16 cm, those maxima
occur in the 1557 nm wavelength region. This feature related to the infrared
emission shift, due to the fiber length increases, may be caused by the Er3+ ion
emission and reabsorption process along the fiber.
Figure 7.13a, b show the ASE power spectra of an 11 cm MOF, obtained by
using an OSA (BWR ¼ 1 nm) and the 980 nm/980 nm and 980 nm/1480 nm pump
schemes, respectively [32]. For both pump configurations, the ASE spectra were
restricted to the 1550 nm region. In both cases, the bandwidths (at 3 dB of the
maximum ASE power) go from 1530 up to 1570 nm, with a maximum ASE power
around 1557 nm. But the ASE spectrum corresponding to the case of the 980 nm/
1480 nm pump scheme is flatter in comparison to that of 980/980 nm pump scheme.
146 L.C. Barbosa et al.

Fig. 7.14 Gain spectra for pump schemes of (a) 980/980 nm, (b) 980/1480 nm, and (c) example of
signal gain (λsignal ¼ 1557.3 nm, 980/980 nm pump). The obtained gain is 22.6 dB (From E. F.
Chillcce, R. Narro-Garcı́a, J. D. Marconi, J. W. Menezes, M. F. Ando, E. Rodriguez, H. L.
Fragnito, and L. C. Barbosa, Latin America Optics and Photonics Technical Digest, OSA,
LM1C.4 (2012))

Fig. 7.15 (a) Laser emission spectra of 5, 8, and 12 cm-long Er-doped MOFs. (b) Laser
generation at ~1558 nm (OSNR ~42 dB)

Figure 7.14a, b show the gain spectra obtained for Er-doped tellurite MOF
(~11 cm length) [32]. In Fig. 7.14a, both pumps are 980 nm lasers with powers of
~17 dBm and ~16 dBm for the co-propagating and the counter-propagating direc-
tions, respectively. In this case, the average gain value is ~13 dB and the maximum
gain is ~22 dB. In Fig. 7.14b, the co-propagating pump is a 980 nm laser with an
optical power of ~17 dBm, and the counter-propagating pump is a 1480 nm laser
with an optical power of ~16 dBm. In this latter case, the average and the maximum
gain values are ~15 dB and ~27 dB, respectively. In both cases, the high-gain region
goes from ~1530 up to ~1570 nm. The 980/980 nm pump scheme also presents a
slightly worse gain performance than that of the 980/1480 nm pump scheme.
Figure 7.14c shows an example of signal gain (λsignal ¼ 1557.3 nm, 980/980 nm
pump). The obtained gain is 22.6 dB.
Figure 7.15 shows the laser emission spectra of Er3+-doped tellurite MOFs,
obtained by using an OSA (BWR ¼ 0.2 nm) in the range 1500–1600 nm.
7 Tellurite Glasses for Optical Amplifiers 147

Figure 7.15a shows the laser emission spectra of Er3+-doped tellurite MOFs with
lengths of 5, 8, and 12 cm. First, for a 5 cm-long MOF, the laser generation occurs
preferentially in the region around 1535 nm. Second, for an 8 cm-long MOF, the
laser generation occurs in regions around 1535 nm and 1558 nm. Finally, for a
12 cm-long MOF, the laser generation occurs preferentially in the region around
1558 nm. The laser generation has a similar behavior to that of ASE as a function on
the MOF length.
Figure 7.15b shows the laser emission obtained with a 11 cm long tellurite MOF
and the 980/980 nm pump (optical power ~20 dBm for both the co-propagating and
the counter-propagating pumps).
The emission wavelength is located at ~1558 nm, which coincides with the
wavelength of the maximum peak of the spontaneous emission spectrum shown in
Fig. 7.12. This is expected because the medium gain has the greatest chance of
being higher than the cavity losses around this region. The optical signal-to-noise
ratio (OSNR) obtained is ~42 dB. This value can be increased by using an improved
pump configuration as the one shown in [7]. Such a configuration would also allow
the tuning of the laser along the entire gain region.

7.5 Broadband Optical Amplification

Tellurite glasses present considerably potential applications for broadband ampli-

fication in the optical communication window (region around 1550 nm wave-
length). In this context, tellurite glasses doped with Er3+ ions show broadband
emission in comparison to that of silica or fluoride glasses doped with Er3+ ions.
Figure 7.16 shows emission cross section spectra of the 4I13/2 ! 4I15/2 transition in
Er3+-doped (Al/P silica, fluoride, and tellurite), in Er3+-doped oxy-fluoride tellurite
(with 5 and 30 mol% of ZnF2), and in Er3+-doped oxy-fluoride tellurite-silica
(OFTS) glasses [33]. The emission cross section spectra in the OFTS and in the
oxy-fluoride tellurite (containing low ZnF2) glasses are broader in comparison with
those in the Al/P silica and in the fluoride glasses. Such increases in bandwidth can
be related to the Er environment in the host glasses. These Er environment are
associated with the Er-F ionic and/or the Er-O covalent bonds as well as the Er sites
of the TeO4, TeO3+1, and TeO3 structural units. In contrast, oxy-fluoride tellurite
glasses with high ZnF2 concentrations present narrowing bands as a consequence of
the ionic behavior of the glass. As a result, the oxy-fluoride tellurite glass
containing 30 mol% of ZnF2 present a narrow emission spectrum that is very
similar to that of the fluoride glass.
On the other hand, it is possible to increase more the emission bandwidth of Er3+
-doped tellurite glasses. Broadband emission spectra of Er3+-Tm3+-co-doped
tellurite fibers with bandwidths more than 160 nm (around the 1550 nm band)
were reported by several researchers [5, 34].
Figure 7.17 shows the amplified spontaneous emission (ASE) spectra of Er3+-
doped fibers and Er3+-Tm3+ co-doped fibers. Figure 7.17a shows ASE spectra of
148 L.C. Barbosa et al.

Fig. 7.16 The emission cross section spectra of the 4I13/2 !4I15/2 transition in Er3+-doped Al/P
silica, fluoride, tellurite, oxy-fluoride tellurite (with 5 and 30 mol% of ZnF2), and oxy-fluoride
tellurite-silicate (OFTS) glasses. The spectra indicated by asterisk are from reference [1] (From
E.F. Chillcce, I.O. Mazali, O.L. Alves, L.C. Barbosa, Optical Materials 33, 389–396 (2011))

Fig. 7.17 (a) The ASE power spectra for Er2O3-Tm2O3-co-doped fibers with 10 cm-long fiber.
The 980 nm (120 mW) diode laser when co-propagating scheme is used. (b) The ASE intensity
spectra for Er2O3-Tm2O3-co-doped fibers with 15 cm-long fiber. The 790 nm (400 mW) Ti:
sapphire pump laser in co-propagating scheme is used (From E.F. Chillcce, E. Rodriguez,
A.A.R. Neves, W.C. Moreira, C.L.César, L.C. Barbosa, Opt. Fiber Technol. 12, 185 (2006))

7500 ppm Er2O3-doped fiber (curve (a)), 7500 ppm Er2O3-2500 Tm2O3 ppm
co-doped fiber (curve (b)), 7500 ppm Er2O3-5000 ppm Tm2O3 co-doped fiber
(curve (c)), 7500 ppm Er2O3-7500 Tm2O3 ppm co-doped fiber (curve (d)),
7500 ppm Er2O3-10,000 ppm Tm2O3 co-doped fiber (curve (e)), 7500 ppm
Er2O3-15,000 ppm Tm2O3 co-doped fiber (curve (f)), pumped with 980 nm diode
laser in the co-propagating scheme. The ASE spectra of Fig. 7.17a shows only the (4
I13/2!4I15/2) transitions for Er3+-doped fiber (curve (a)) and the (4I13/2!4I15/2) and (
3
F4!3H6) transitions for the Er3+-Tm3+ co-doped fibers (curves (b), (c), (d), (e),
7 Tellurite Glasses for Optical Amplifiers 149

and (f)). The (3F4!3H6) transition that appears with the addition of Tm3+ ions is
excited by the energy transfer (ET1) process.
Figure 7.17b shows the ASE spectra for 7500 ppm Er2O3 and several Tm3+
concentrations. The ASE intensity spectra shown to correspond for (a1) 7500 ppm
Er2O3-doped tellurite fiber, (b1) 7500 ppm Er2O3-2500 ppm Tm2O3-co-doped fiber,
(c1) 7500 ppm Er2O3-5000 ppm Tm2O3-co-doped tellurite fiber, and (d1) 7500 ppm
Er2O3-7500 ppm Tm2O3-co-doped tellurite fiber. In this case, all fibers are 15 cm
long and are pumped with a 790 nm (400 mW) Ti:sapphire laser in the
co-propagating scheme. Only the (4I13/2 ! 4I15/2) transition for Er3+-doped tellurite
fibers (curve (a1)) is observed again, but all the (4I13/2 ! 4I15/2), (3H4 ! 3F4), and (3
F4 ! 3H6) transitions are present in the Er3+-Tm3+ co-doped tellurite fiber (curves
(b1), (c1), and (d1)). The 1440–1530 nm band ASE power increases with the
Tm2O3 concentration up to 5000 ppm and then drops for higher concentrations.
This can be explained by the fact that the energy transfer efficiencies strongly
depend on the ion separation, which diminish with concentration. Consequently, the
ET1 process becomes more efficient with higher Tm2O3 concentration which tends
to populate the lower 3F4 Tm3+ level and to depopulate the 4I13/2 Er3+ level. This
ET1 processes would decrease the whole 1470–1550 nm band ASE power.

7.6 Rare-Earth Level Lifetimes in Tellurite Glasses

The lifetime of the excited states of rare earths is a function radiative and
nonradiative decay processes. The radiative decay process can be calculated by
using the Judd-Ofelt parameters. The nonradiative decay process covers the multi-
phonon decay, the energy transfer (ET), as well as the inhomogeneous broadening
(characteristic of the glass host) [5]. In the Er3+-doped glasses, the free OH is one of
the dominant quenching centers, especially for the lifetimes of the 4I11/2 and 4I13/2
levels, from where just two or three phonons are required for nonradiative
de-excitation. However, for highly concentrated Er3+-Tm3+ co-doped glasses, the
ET process is also an efficient process and important quenching center for the
lifetime of the 4I13/2 level. The 4I13/2 level lifetime of Er3+ ions hosted in silica,
phosphate, and fluoride glasses is high (more than 10 ms) in comparison to that of
Er3+ ions hosted in tellurite glasses (around 5 ms) [5, 21, 25].
Figure 7.18a shows the 4I13/2 level lifetime of the Er3+ ions hosted in a 70TeO2-
19WO3-7Na2O-4Nb2O5 (%mol) glass [25]. The 4I13/2 level lifetime shows clearly
the Er3+ ion concentration dependence. The curve labeled as “calculated” is
obtained from the radiative decay probability by using the Judd-Ofelt parameters,
and the curve labeled as “experimental” is the measured lifetime. Figure 7.18b
shows the 4I13/2 level lifetime of Er3+ ions hosted in oxy-fluoride tellurite glasses
with compositions 7500 ppm Er2O3-(80-x)TeO2-xZnF2-20ZnO, where x ¼ 5,
10, 15, 20, 25, and 30 mol% [33]. The lifetime increases from 3 to 5.5 ms when
the ZnF2 concentration raises from 5 to 30 mol%. However, small differences
150 L.C. Barbosa et al.

Fig. 7.18 The 4I13/2 level lifetime of Er3+ ions. (a) Lifetime as function of the Er3+ ion concen-
tration. (b) Lifetime as function of the ZnF2 concentration in oxy-fluoride tellurite glasses (From
E. F. Chillcce, S. P. A. Osório, E. Rodriguez, C. L. César, L. C. Barbosa, Proc. SPIE 5723,
248–252 (2005) and E.F. Chillcce, I.O. Mazali, O.L. Alves, L.C. Barbosa, Optical Materials
33, 389–396 (2011))

Fig. 7.19 (a) The 4I13/2 level lifetime, of Er3+ ions, as a function of the Tm3+ ion concentration.
(b) Dependence of the 4I13/2 level lifetime on the fiber length (δ) (From E.F. Chillcce,
E. Rodriguez, A.A.R. Neves, W.C. Moreira, C.L.César, L.C. Barbosa, Opt. Fiber Technol.
12, 185 (2006) and R. Narro-Garcia, H. Desirena, E. F. Chillcce, L. C. Barbosa, E. Rodriguez
and E. De La Rosa, Opt. Comm 93–101 (2014))

(<6%) in absolute value are observed when the lifetime is measured at 1530 band
(dot curve) or at 1550 nm band (triangular curve).
Figure 7.19a shows the 4I13/2 level lifetime (of Er3+ ions) as a function of the
Tm2O3 concentration [5]. The lifetimes are measured from luminescence that
merges at 90 degrees from the Er3+-Tm3+-co-doped tellurite fiber. The lumines-
cence is produced by pumping the tellurite fiber core with 980 nm diode (or 790 nm
Ti:sapphire) laser. The core/clad tellurite fiber is made from 70TeO2-19WO3-
7Na2O-4Nb2O5 (mol%). The tellurite fibers core are co-doped with 7500 ppm
7 Tellurite Glasses for Optical Amplifiers 151

Er2O3 and (0–15,000 ppm) Tm2O3 and intentionally doped with 80,000 ppm of
Nb2O5 to increase the numerical aperture. The lifetime represented at the left axis of
Fig. 7.19a decreases as the Tm3+ concentration increases because the ET1 process
becomes more efficient. The 980 nm pumping lifetimes are larger than 790 nm
pumping ones for low concentration and tend to become equal for concentrations
above 7500 ppm. This fact is due to the excited state absorption (4I13/2 ! 2H11/2) at
790 nm that depopulates the 4I13/2 level. At higher concentrations, however, the
ET1 probability (WET1), represented at the right axis of Fig. 7.19a becomes dom-
inant, and both lifetimes tend to the same value.
Figure 7.19b shows lifetimes of Er3+-Yb3+-co-doped tellurite fibers [22]. The
fibers have 2 mm outer diameter approximately and fiber length from 0.5 to 10 mm.
For the tellurite fibers co-doped with 10,000 ppm Er2O3 and 40,000 ppm Yb2O3
((Er/Yb ¼ 1/4) proportion), the lifetime increases from 3.5 to 4.6 ms when the
length fiber increases from 0.5 to 10 mm. For the tellurite fibers co-doped with
10,000 ppm Er2O3 and 20,000 ppm Yb2O3 ((Er/Yb ¼ ½) proportion), the lifetime
increases from 3.4 to 4.6 ms when the length fiber increases from 0.5 to 10 mm. The
reason why the decay time increases with the length is due to radiation trapping. In
this phenomenon, photons spontaneously relaxed from 4I13/2 level are reabsorbed
by the neighboring ions in the ground state (4I15/2 level). This process of
reabsorption and reemission is repeated several times and the overall result is an
increase in the lifetime. The effect is more pronounced with the sample size,
refractive index, and spectral overlap of the fluorescence and absorption. Since
the glass length is large and the refractive index is high (2.08), it is expected an
increment of the decay time.

References

1. A. Mori, Y. Ohishi, S. Sudo, Erbium-doped tellurite glass fibre laser and amplifier. Electron.
Lett. 33(10), 863–864 (1997)
2. M. Yamada, A. Mori, K. Kobayashi, H. Ono, T. Kanamori, K. Oikawa, Y. Nishida, Y. Ohishi,
Gain-flattened tellurite-based EDFA with a flat amplification bandwidth of 76 nm. IEEE
Photon. Technol. Lett. 10(9), 1244–1246 (1998)
3. H. Ono, T. Sakamoto, A. Mori, J. Kani, M. Fukutoku, An Erbium-doped tellurite fiber
amplifier for WDM systems with dispersion-shifted fibers. IEEE Photon. Technol. Lett. 14
(8), 1070–1072 (2002)
4. S. Shen, L. Huang, P. Joshi, A. Jha, Gain characteristics of Er3+/Ce3+ codoped tellurite short
fibre amplifier pumped at 980 nm. Electron. Lett. 39(25), 1797–1799 (2003)
5. E.F. Chillcce, E. Rodriguez, A.A.R. Neves, W.C. Moreira, C.L. Cesar, L.C. Barbosa, Er3+-Tm
3+
co-doped tellurite fibers for broadband optical fiber amplifier around 1550 nm band. Opt.
Fiber Technol. 12(2), 185–195 (2006)
6. A. Mori, Tellurite-based fibers and their applications to optical communication networks.
J. Cerma. Soc. Jpn. 116(10), 1040–1051 (2008)
7. J. Dong, Y.Q. Wei, A. Wonfor, R.V. Penty, I.H. White, J. Lousteau, G. Jose, A. Jha, Dual-
pumped tellurite fiber amplifier and tunable laser using Er3+/Ce3+ codoping scheme. IEEE
Photon. Technol. Lett. 23(11), 736–738 (2011)
152 L.C. Barbosa et al.

8. J. Toulouse, A. Lin, A. Ryaznyanskiy, A. Belwalkar, C. Guintrand, C. Lafontaine,

W. Misiolek, and I. Biaggio, Development of new tellurite fibers for mid-IR applications, in
Proceedings of IEEE Winter Topicals (Institute of Electrical and Electronics Engineering,
2011), pp. 185–186.
9. H. Masuda, A. Mori, K. Shikano, M. Shimizu, Design and spectral characteristics of gain-
flattened tellurite-based fiber Raman amplifiers. J. Lightwave Technol. 24(1), 504–515 (2006)
10. G. Qin, H. Sotobayashi, M. Tsuchiya, A. Mori, T. Suzuki, Y. Ohishi, Stimulated Brillouin
scattering in a single-mode tellurite fiber for amplification, lasing, and slow light generation.
J. Lightwave Technol. 26(5), 492–498 (2008)
11. P. Domachuk, N.A. Wolchover, M. Cronin-Golomb, A. Wang, A.K. George, C.M.B. Cordeiro,
J.C. Knight, F.G. Omenetto, Over 4000 nm bandwidth of mid-IR supercontinuum generation
in sub-centimeter segments of highly nonlinear tellurite PCFs. Opt. Express 16(10),
7161–7168 (2008)
12. X. Feng, W.H. Loh, J.C. Flanagan, A. Camerlingo, S. Dasgupta, P. Petropoulos, P. Horak,
K.E. Frampton, N.M. White, J.H.V. Price, H.N. Rutt, D.J. Richardson, Single-mode tellurite
glass holey fiber with extremely large mode area for infrared nonlinear applications. Opt.
Express 16(18), 13651–13656 (2008)
13. A. Lin, A. Zhang, E.J. Bushong, J. Toulouse, Solid-core tellurite glass fiber for infrared and
nonlinear applications. Opt. Express 17(19), 16716–16721 (2009)
14. M. Liao, X. Yan, G. Qin, C. Chaudhari, T. Suzuki, Y. Ohishi, A highly non-linear tellurite
microstructured fiber with multi-ring holes for supercontinuum generation. Opt. Express 17
(18), 15481–15490 (2009)
15. G.S. Qin, X. Yan, M. Liao, A. Mori, T. Suzuki, Y. Ohishi, Wideband supercontinuum
generation in tapered tellurite microstructured fibers. Laser Phys. 21(6), 1115–1121 (2011)
16. I. Savelii, J.C. Jules, G. Gadret, B. Kibler, J. Fatome, M. El-Amraoui, N. Manikandan,
X. Zheng, F. Désévédavy, J.M. Dudley, J. Troles, L. Brilland, G. Renversez, F. Smektala,
Suspended core tellurite glass optical fibers for infrared supercontinuum generation. Opt.
Mater. 33(11), 1661–1666 (2011)
17. E. Yousef, M. Hotzel, C. Russel, Linear and non-linear refractive indices of tellurite glasses in
the system TeO2–WO3–ZnF2. J. Non Cryst. Solids 342(1–3), 82–88 (2004)
18. J. Wang, S. Prasad, K. Kiang, R.K. Pattnaik, J. Toulouse, H. Jain, Source of optical loss in
tellurite glass fibers. J. Non Cryst. Solids 352(6–7 Spec), 510–513 (2006)
19. R. Stepien, R. Buczynski, D. Pysz, I. Kujawa, A. Filipkowski, M. Mirkowska, R. Diduszko,
Development of thermally stable tellurite glasses designed for fabrication of microstructured
optical fibers. J. Non Cryst. Solids 357(3), 873–883 (2011)
20. V.V. Ravi-Kanth-Kumar, A.K. George, J.C. Knight, P.J. Russell, Tellurite photonic crystal
fiber. Opt Express 11(20), 2641–2645 (2003)
21. S. Shen, A. Jha, E. Zhang, S.J. Wilson, Compositional effects and spectroscopy of rare earths
(Er3+, Tm3+, and Nd3+) in tellurite glasses. C. R. Chim. 5(12), 921–938 (2002)
22. R. Narro-Garcia, H. Desirena, E.F. Chillcce, L.C. Barbosa, E. Rodriguez, E. De La Rosa,
Optical and spectroscopy characterization of Er3+-Yb3+ co-doped tellurite glasses and fibers.
Opt. Commun. 317, 93–101 (2014)
23. E.F. Chillcce, E. Rodriguez, O.L. Alves, C.L. Cesar, I.O. Mazali, L.C. Barbosa, Fabrication of
photonic optical fibers from soft glasses. J. Am. Ceram. Soc. 93(2), 456–460 (2010)
24. A. Mori, H. Masuda, K. Shikano, M. Shimizu, Ultra-wide-band tellurite-based fiber Raman
amplifier. J. Lightwave Technol. 21(5), 1300–1306 (2003)
25. E.F. Chillcce, S.P.A. Osório, E. Rodriguez, C.L. César, L.C. Barbosa, Lifetime of the 4I13/2
excited level of the Er3+-ion in the glass system TeO2-WO3-Na2O-Nb2O5. Proc. SPIE 5723,
248–252 (2005)
26. J.S. Wang, E.M. Vogel, E. Snitzer, Tellurite glass: a new candidate for fiber devices. Opt.
Mater. 3(3), 187–203 (1994)
7 Tellurite Glasses for Optical Amplifiers 153

27. R. Narro-Garcı́a, E.F. Chillcce, L.C. Barbosa, E. de Posada, M. Arronte, E. Rodriguez,

Fabrication and upconversion luminescence of Er3+/Yb3+ codoped TeO2-WO3-Na2O-Nb2O5-
Al2O3 glass fiber. JOL 134, 528–532 (2013)
28. E.F. Chillcce, C.M.B. Cordeiro, L.C. Barbosa, C.H. Brito Cruz, Tellurite photonic crystal fiber
made by a stack-and-draw technique. J. Non Cryst. Solids 352(32–35), 3423–3428 (2006)
29. C. Knight, T.A. Birks, P.S.J. Russel, D.M. Atkin, All-silica single mode optical fiber with
photonic crystal cladding. Opt. Lett. 21(19), 1547–1549 (1996)
30. E.F. Chillcce, R. Narro-Garcı́a, J.W. Menezes, E. Rodriguez, D. Marconi, H.L. Fragnito,
L.C. Barbosa, Er3+- doped Tellurite Micro-structured Fiber: Laser Generation and Optical
Gain. Proc. SPIE 8257, 82570B1–8257010 (2012)
31. A. Traore, Measurement of the nonlinear refractive index of Tellurite glass fiber by using
induced grating autocorrelation technique, PhD thesis, University of Maryland, Baltimore
County, 102 (2011).
32. E.F. Chillcce, R. Narro-Garcı́a, J.D. Marconi, J.W. Menezes, M.F. Ando, E. Rodriguez,
H.L. Fragnito, L.C. Barbosa, Optical amplifier based on a Er3+-doped tellurite microstructured
optical fiber. Lat. Am. Opt. Photon. Tech. Digest OSA, LM1C.4. (2012) (2012)
33. E.F. Chillcce, I.O. Mazali, O.L. Alves, L.C. Barbosa, Optical and physical properties of Er3+-
doped oxy-fluoride tellurite glasses. Opt. Mater. 33, 389–396 (2011)
34. L.H. Huang, A. Jha, S. Shen, X. Liu, Broadband emission in Er3+–Tm3+ codoped tellurite fiber.
Opt. Express 12, 2429 (2004)
Chapter 8
Broadband Emission in Tellurite Glasses

Kummara Venkata Krishnaiah, Jose Marques-Hueso, and Raman Kashyap

Tellurite glasses are excellent hosts for broadband emission. This chapter details the
prospective of broadband absorbing/emitting lanthanide and transition metal-doped
glasses and glass-ceramics, attending to their emission ranges and bandwidth, with
particular emphasis on lanthanide-doped tellurite glasses. The most common emit-
ting ions are presented one by one. Next, it focuses on the spectroscopy and the
Judd–Ofelt parameters and their related radiative properties exclusively for Er3+-
doped tellurite glasses, as well as a brief description of the other lanthanide ions.
The different parameters that affect the broadband absorption/emission in tellurite
glasses are then reviewed. In addition, we summarize the dopants that can exhibit
broadband emission (not only in the near-infra-red region but also in the visible
region) and their bandwidth along with their energy level structures. Finally, the
different applications of tellurite glasses are explored, including lanthanide and
transition metal ions, fiber amplifiers, Raman amplifies, superbroadband emission
for telecommunications, supercontinuum generation, solar devices, and possible
applications in laser-induced cooling.

K.V. Krishnaiah (*)

Department of Engineering Physics, Polytechnique Montréal, P.O. Box 6079, Montréal, QC,
Canada, H3C 3A7
e-mail: [email protected]
J. Marques-Hueso
Institute of Sensors, Signals and Systems, School of Engineering and Physical Sciences,
Heriot-Watt University, Edinburgh EH14 4AS, UK
R. Kashyap
Department of Engineering Physics, Polytechnique Montréal, P.O. Box 6079, Montréal, QC,
Canada, H3C 3A7
Department of Electrical Engineering, Polytechnique Montréal, P.O. Box 6079, Montréal, QC,
Canada, H3C 3A7

© Springer International Publishing AG 2017 155

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_8
156 K.V. Krishnaiah et al.

8.1 Introduction

Broadband is a word which has permeated into everyday usage, referring to radio,
television, and the Internet. In optical communication, broadband is synonymous
with as large an optical window as possible for the transfer of signals in the form of
“photons” through an optical fiber. As a rule, the most cost-effective way of
transmitting signals has been by the multiplexing of as many channels as possible
within a given bandwidth. If a signal is attenuated, it is amplified in a section of
lanthanide (Ln)-doped transmission medium by optically pumping it at a suitable
wavelength. So far, broadband emission has been observed in different Ln and
transition metal (TM) ion-doped multicomponent glasses, which found applications
in white light-emitting diodes, lasers, and optical amplifiers in the visible, near-
infrared (NIR), mid-infrared (MIR), and infrared (IR) regions.
The absorption/emission spectra of Ln ion-doped glass are often characterized
by the position of the individual Stark levels of the manifold often masked by the
inhomogeneous linewidths of the levels. Since glasses are amorphous, the absorb-
ing/emitting ions are exposed to different environments, which origins of the
broadening of the absorption bands when compared to crystals [1]. Hence, they
show a higher efficiency of absorption from broadband sources, and they have the
capability of high peak power generation due to their high saturation fluence and
broad emission bandwidth.
Two main families are used as emitting impurities: the lanthanides and transition
metals. For the comparison, the absorption spectra for one of the Ln and TM ions—
taking an example of Nd3+ and Cr3+ ions, respectively—are shown in Fig. 8.1. The
absorbance of both families is different due to characteristic electronic structure of
each group. Generally, the absorption bands are broad in the case of TMs, while
they are sharp for Ln ion. This is due to the presence of a number of electrons at the

Fig. 8.1 Absorption

spectra of Cr3+ (TM) and
Nd3+ (Ln) ions in glasses
8 Broadband Emission in Tellurite Glasses 157

outermost orbital and their surrounding orbital in the respective ions. The electronic
configurations of Cr3+ and Nd3+ ions are represented as

Cr3+: 1s2 2s22p6 3s2 3p6 3d3.

Nd3+: 1s2 2s22p6 3s2 3p6 3d10 4s2 4p6 4d10 4f3 5s25p6.

In the case of Ln ion, the electrons of the partially filled 4f shell are shielded from
interactions with external forces by the overlying 5s2 and 5p6 shells. Shielding
means that the Ln radiative transitions in solid hosts resemble those of the free ions
besides weak electron–phonon coupling. As a result, the Ln energy levels are only
slightly affected by the environment surrounding the Ln ions and remain practically
invariant for a given ion in various hosts. This eases the engineering of optical
materials with tailored properties.
Regarding Ln ion, the outer most 4f orbital shielded by the occupied 5s25p6
orbital leads to sharp absorption bands, whereas the TM ions that occupy the outer
most 3d orbital do not have the shielding and are therefore able to participate
directly in the interaction/bonding with the host matrix. When compared with
TM-doped glasses, the Ln-doped glasses possess more or less similar properties
in the entire matrix, with several excited states suitable for optical pumping, longer
fluorescence lifetime, and high photoluminescence quantum yield. Hence, the
Ln-doped materials are perfect for applications in photonics that deals with the
manipulation of the active medium and transfer and storage of information. More-
over, the ligand field (LF) theory [2] is more appropriate to the systems of TM ions
as they are not shielded from the surrounding environment, whereas the crystal field
(CF) theory is then a precise approximation when applied to Ln ions. The
TM-doped systems are highly colored and exhibit paramagnetism, due to the
presence of d electrons. The broad optical absorption spectra of different TM ions
in different host matrices that provide different ligand environments offer informa-
tion about the determination of cation and their sites and an estimation of the degree
of covalency of the metal ligand bonds.
Broadband absorption/emission is strongly dependent on the structure of the
glass, which influences the features such as the spectral profile, lifetimes, and
photoluminescence quantum yields (PLQY). All these properties are highly essen-
tial for the materials that are used in optical amplifier applications. In some cases,
the spectral bands from the glasses are broadened, and they are overlapped by
forming continuous bands, due to the variations in the local crystal field. The
inhomogeneous nature of the Ln spectra in a glass is beneficial for optical pumping
with broadband sources such as xenon flash lamps because most of the spectrum is
absorbed. In addition, the effective cross section is low for amplified spontaneous
emission (ASE) which limits the excited state population. Both of these features
provide a higher storage of energy in glass rather than in crystals.
In this chapter, we outline the broadband emission of Ln and TM ion-doped
tellurite glasses, with particular emphasis on the spectroscopic and broadband
emission properties of Er3+ ions. The parameters that influence broadband emission
are discussed, including the dopant concentration, chemical composition, thickness
158 K.V. Krishnaiah et al.

of the sample, heat treatment, codoping with other impurities, and finally the
density of OH
ion. The influence of reabsorption on the broadening of the spectral
profile is also discussed. In addition, broadband emission in selective TM-doped
glasses is outlined. The broadband emission and bandwidth for Ln and TM
ion-doped tellurite glasses are compared; their advantages and disadvantages in
optical telecommunication and solar cell applications are discussed.

8.2 Energy Level Structure of Broadband Emitting Ln

and TM Ions

The most commonly emitting Ln ions are Ce3+, Pr3+, Nd3+, Dy3+, Ho3+, Er3+, Tm3+,
and Yb3+ while the TM ions are Ti3+, Cr3+, Ni2+, and Bi+. Their applications are
determined by utilizing their energy level structure.

8.2.1 Lanthanides

• Erbium (Er3+): Among the Ln elements, Er3+ (4f11) ion, which emits NIR
emission at around 1550 nm for the 4I13/2 ! 4I15/2 transition and shown in
Fig. 8.2, has stupendous applications in fiber-optic communication as an erbium-
doped fiber amplifier (EDFA). It also exhibits NIR to visible upconversion
(UC) around green and red emissions, obtained by the transitions, 4S3/2 ! 4I15/2

4F
7/2
20 2H
4S 11/2
3/2
ESA

NR
Energy (x 103 cm–1)

16 4F
9/2
4I
12 9/2
4I
11/2
2.7 μm
0.98 mm GSA

8
4I
13/2
1.5 μm

548 nm

671 nm
488 nm

525 nm

0 4I
15/2

Er3+
Fig. 8.2 Partial energy level diagram of Er3+ ions in TeO2–WO3–ZrO2 glasses. Ground state
absorption (GSA), excited state absorption (ESA), and non-radiative (NR) channels are also shown
(From ref. [3], copyright 2015, American Scientific Publishers)
8 Broadband Emission in Tellurite Glasses 159

and 4F9/2 ! 4I15/2, respectively, and which are useful in color displays,
sensor, etc. [3–5]. On the other hand, the emission at 0.98 μm that originates
from the 4I11/2 ! 4I15/2 transition can be efficient for enhancing the effi-
ciency of c-Si solar cells [6].
• Ytterbium (Yb3+) ion has attracted as a dopant in ultrafast laser materials [7–9]
and laser-induced cooling of solids [10, 11] to name but few applications. It has
also been used as Yb3+-doped fiber amplifiers (YDFA) and lasers for femtosec-
ond laser waveguide inscription in the transparent material to design integrated
circuits [12, 13]. The studies in high-field lasers for nuclear fusion reaction are
indicated that the Yb3+-doped materials, especially glasses, are the best host
materials for efficient energy storage [14]. Yb3+ (4f13) ion has a very simple
energy level scheme, as shown in Fig. 8.3, which consists of only two manifolds,
namely, the ground 2F7/2 and excited 2F5/2 states, which are separated by energy
of 10,000 cm
1. Yb3+-doped laser materials are efficiently pumped by high-
power diode lasers in the wavelength range of 0.9–1.1 μm, and laser operation
takes place in the wavelength range of 1.0–1.1 μm, including the 1.06 μm laser
line of the Nd3+ ion. Efficient lasing is possible from Yb3+-doped materials
because of the low quantum defect (the energy difference between the pump and
the lasing photons) which minimizes the heating in the system and provides
cooling as well. Thus, the laser materials with low quantum defects have lower
heating, and, in general, it is about 3–4 times lower in the case of Yb3+ ion when
compared to Nd3+ ion [16]. In addition, as there are no intermediate levels, the
quantum efficiency is not decreased by the processes of non-radiative relaxa-
tions, concentration quenching (cross-relaxation) and ESA. The solubility is
high in the case of Yb3+ ions when compared with other Ln-doped solid-state
laser materials without significant reduction in the excited state lifetime.
• Cerium (Ce3+)-doped glasses are attractive materials for multitude of applica-
tions, particularly for fabricating white light-emitting diodes (w-LEDs). Ce3+
also exhibits a broad absorption band due to 4f ! 5d transition, shown in
Fig. 8.4, in the near ultraviolet (UV) region and emits in the visible region in
the range of 450–700 nm when excited with a blue-LED [18]. Most of the
w-LEDs are fabricated by combining a blue-LED chip with yellow luminescence
phosphors, i.e., Ce3+-doped yttrium–aluminum–garnet (YAG) phosphors [19].

Fig. 8.3 Absorption and 10923

2
Energy (x 103 cm-1)

emission channels of Yb3+ 10438 F5/2

ion in lead–phosphate glass
10 10256
(Copied from ref. [15], 8
1003 nm
1058 nm

copyright 2014 Elsevier)

975 nm

6
975 nm
958 nm
927 nm

998 nm
913 nm

4
2 805
453 2
0 286 F7/2
Yb3+ 0
160 K.V. Krishnaiah et al.

Fig. 8.4 Energy level

diagram of Ce3+ ion in
CaAl2O4/Ce3+, bandgap of
the host, and the trapping
mechanism. After
excitation to the conduction
band of lower 5d levels
where trapping occurs
through tunneling (Redrawn
from ref. [17])

Moreover, w-LEDs have been the subject of increasing interest due to their
potential for lasers, indicators, automobile headlights, and general illumination
[20]. On the other hand, Ce3+ ion also acts as a sensitizer in laser and photo-
thermo-refractive glasses, as they are often used to decrease photodarkening in
optical fibers or as decolorizing agent.
• Praseodymium (Pr3+)-doped glasses are promising hosts for optical amplifiers
in the second telecommunication window. Pr3+ has the 4f2 electronic configura-
tion with 3H4 ground state. The broadband emission which originates from the
1
G4 ! 3H5 (1.33 μm) and 1G4 ! 3H4 (1.01 μm) are shown in Fig. 8.5.
The zero-dispersion wavelength in silica fiber varies the region between
1310 and 1320 nm, and these wavelengths are particularly important for
telecommunications [22]. The PLQY and gain of Pr3+-doped amplifiers at
1.3 μm are restricted by the short lifetime of the excited level. Theoretical
predictions indicate that a longer lifetime could be possible in low phonon
energy glasses [23]. The Pr3+ ion has an emission centered at 1.6 μm which
originates from the (3F3, 3F4) ! 3H4 transition with an energy gap of
~1350 cm
1. This radiative transition is virtually forbidden in most glasses,
and has so far not been detected in glasses even in sulfides, with the exception
of LaCl3 crystal [24].
• Neodymium (Nd3+)-doped phosphate and silicate glasses have been used for the
development of high-energy and high peak power laser applications in the fusion
reactions. These glasses can exhibit properties of high gain, high-energy storage,
large extraction efficiency, and high laser damage threshold based on the glass
composition, P2O5–K2O/Li2O–MO–Al2O3 (MO ¼ MgO, SrO, CaO, or BaO)
[25], and are reported for large laser amplifier applications. In general, three NIR
emission bands that are identified at 900 nm (4F3/2 ! 4I9/2), 1064 nm (4F3/2 !
4
I11/2), and 1340 nm 4F3/2 ! 4I13/2 wavelengths are shown in Fig. 8.6. More-
over, broadband emission at 1.3 μm can be useful for the optical amplifiers as it
exists in the zero-dispersion window. In addition, a broad bandwidth and
8 Broadband Emission in Tellurite Glasses 161

Fig. 8.5 Partial energy 1G

levels of Pr3+ ion in low 4
phonon energy halide-,
fluoride-, and sulfide-based
glasses (Reproduced from
non-radiative
ref. [21], copyright, 1998,
American Institute of 3F
4
Physics)
3F
3

3F
2

3H
6

1.3 mm amplifier

3H
5

1.01 mm pump

3H
4

Fig. 8.6 The NIR emission

15000
spectra and partial energy λexc = 532 nm
level diagram (inset) of Nd 4I
3+ 4
ion in TeO2–ZnO–Na2O– 11/2 F3/2
Energy (cm–1)

4F 10000
Li2O–Nb2O5 glass (Re-used 3/2
Intensity (arb. units)

with permission from ref.

[26], copyright, 2010, 4
I15/2
5000
Elsevier) 4
I13/2
4
I11/2
4
0 I9/2
Nd3+
4I 4I
9/2 13/2

900 1000 1100 1200 1300 1400

Wavelength (nm)

tunable laser emission, is witnessed by varying the Nd3+ concentration and

pump wavelength, due to a high CF inhomogeneity of glass matrix [27].
• Dysprosium (Dy3+) has practical importance as it exhibits prominent emissions
in the yellow (~572 nm) and blue (~481 nm) which are assigned to 4F9/2 ! 6H13/2
and 4F9/2 ! 6H15/2 transitions, respectively. Furthermore, a broadband emission at
162 K.V. Krishnaiah et al.

Fig. 8.7 Partial energy

level diagram of Dy3+ ions
in oxyfluoride glass and
glass-ceramics showing
various visible and NIR
emissions along with
excitation wavelengths at
451 and 884 nm. Cross-
relaxation channels in the
NIR region (CR1 and CR2)
are also presented (dotted
arrows) (Re-printed with
permission from ref. [28],
copyright, 2009, American
Institute of Physics)

1.32 μm originating from the 4F11/2 ! 6H9/2 level of the Dy3+ ion is an interesting
for fiber amplifier application in optical transmission systems. The energy level
structure and the emission channels are shown in Fig. 8.7.
• Holmium (Ho3+) ion is one of the wide range of emitting ions for spectroscopic
studies as it exhibits several electronic transitions with a relatively long-lived 5I7
level and large peak stimulated emission cross sections [29, 30] in the visible,
NIR, MIR, and IR regions. An optical amplification at 1.2 μm due to the 5I6 ! 5I8
transition is highly desirable as it is located in the minimum wavelength dispersion
of silica fiber. It has been demonstrated that laser emission of the Ho3+ ion is in the
range of 1.2–4.9 μm [31, 32]. In addition, the Ho3+ ion provides eye-safe potential
laser emission around 2.0 μm region with a low threshold which is attractive for
applications in atmospheric communication systems. The excitation and emission
channels of Ho3+ ion are shown in Fig. 8.8.
• Thulium (Tm3+) ion exhibits emissions at a variety of wavelengths starting
from the UV to the NIR of the electromagnetic spectrum. The emission band
centered at around 1460 nm that is ascribed to the 3H4 ! 3F4 transition is shown
in Fig. 8.9, which is potential for developing the Tm3+-doped fiber amplifiers
(TDFAs) in telecommunication systems. A low-loss optical fiber transmission
window in the range of 1400–1600 nm allows a band extension in the spectral
range corresponding to the S-band amplifier region. Moreover, S-band is on the
short wavelength side of the conventional EDFA of C-band at 1530–1570 nm. In
addition, a band centered at 1800 nm that is attributed to 3F4 ! 3H6 is suitable
for developing lasers for medical applications and atmospheric and chemical
sensing [34–36], existing at a longer wavelength side of the EDFA. The TDFA at
1.46 and 1.8 μm has been utilized for extending the transmission bandwidth of
optical fibers beyond that possible with the EDFAs [37].
8 Broadband Emission in Tellurite Glasses 163

Fig. 8.8 Partial energy

level diagram of Ho3+ ion
along with their broadband
emission channels in Ho3+-
codoped glass under
473 and 532 nm excitation

Fig. 8.9 Partial energy

levels of Tm3+ in germanate
glass. EM energy migration,
Wmp multi-phonon
relaxation rate, CR cross
relaxation (Reprinted with
permission from ref. [33]
copyright, American
Chemical Society, 2011)

8.2.2 Transition Metal Ions

• The titanium (Ti3+)–sapphire (Al2O3) laser is a TM ion laser that provides

continuous and pulsed modes due to the interaction between electronic levels
of 3d electrons with phonon vibrations that lead to broadband emission. The Ti3+
ion has only 1d electron that existed in two energy levels, and with the laser
operation within a vibronic transition, it appears like a four-level laser system. A
wide tuning range of the laser operation is possible due to a large energy
difference between the E3/2 excited state and the 2T2g ground state, as shown
in Fig. 8.10. The upper and lower level of the 2Eg state can be split due to CF
164 K.V. Krishnaiah et al.

Fig. 8.10 Energy level

structure of Ti3+ ions in
Ti3+:Al2O3 crystals. NR
non-radiative relaxation

effect. The Ti3+-doped tunable laser has been the subject of interest since its first
breakthrough by Moulton in 1982 [38], which is useful for many applications in
all the fields whereever the light–matter interaction is needed at near-infrared
region. A very broad infrared emission band in the range of 1200–1800 nm (peak
at 810 nm) was found in Ti3+-doped silicate glass, when the excitation was
performed between 400 nm and 600 nm (T2g ! Eg transition in an approxi-
mately cubic field) [39].
• Nickel (Ni2+)-doped glasses (Li2O–Ga2O3–SiO2, ZnO–Al2O3–SiO2) and trans-
parent glass-ceramics (GCs) have been investigated extensively for their appli-
cations in broadband amplifiers, fibers, and tunable lasers in recent years. This
kind of Ni2+-doped multicomponent materials exhibit broad emission in the long
wavelength optical communication window when they excited with most robust
and cost-effective 980 nm laser diode [40–42]. The electronic structure of the Ni2+
ions in silicate glass is shown in Fig. 8.11. The broadband emission of Ni3+ ion,
covering the wavelength band 1050–1700 nm region, increases enormously
when codoped with Yb3+ ions in silicate GCs due to an efficient energy transfer,
Yb3+ ! Ni2+. Moreover, structural and optical studies of Ni2+ ions are enhanced
significantly due to their existence within the low phonon nanocrystals of trans-
parent GCs and are responsible for the enhanced broadband NIR emissions
[43]. The quantum efficiency of broadband emission is found to be higher than
95% for MgGa2O4 and LiGa5O8 glass-ceramics, estimated by using Struck–
Fonger relation [44]. The broadband emission of Ni2+ in the wavelength region
from the O-band to L-band (1260–1625 nm) in optical communication systems
suggests that the Ni2+-doped glasses and GCs appear promising for
superbroadband gain media when a net gain is observable over the emission
bandwidth.
8 Broadband Emission in Tellurite Glasses 165

25
Energy (× 103 cm–1)

20 4T
2 3T (3F) 3
A2 (3T1)

Energy, x103 cm–1

2E 1
15 ET
10
3
10 3T (3F) B2 (3T2)
532nm
960nm 2
5 1.2-1.6 μm
5 1200nm
4A 3A (3F) 0 3
0 2 2 B1 (3A2)
4+
Cr
Octahedral Cr3+ Octahedral Ni2+

Fig. 8.11 Energy level diagram of Cr3+ and Ni2+ (left) ions in Cr3+/Ni2+-codoped SiO2–Al2O3–
ZnO–TiO2–ZrO2 glass and GCs showing the Cr3+ ! Ni2+ energy transfer (ET). Possible emission
lines were represented by solid and dash arrow for radiative processes and non-radiative pro-
cesses, respectively. The 2E energy level is embedded into the 4T2 energy level [43]. The Cr4+ ion
energy level is also shown at the right side

• Chromium (Cr4+, Cr3+)-doped glasses exhibit broadband emission covering the

whole range of spectrum with large bandwidth by surpassing the traditional Pr3+-,
Tm3+-, and Er3+-doped fiber amplifier [45] bandwidths together. The broadband
emission profile and bandwidth depend on the valence state of Cr. On the other
hand, the broadband emission intensity of Ni2+ at 1200 nm in transparent ZnO–
Al2O3–SiO2 GCs containing ZnAl2O4 nanocrystals improves by 8-times higher
intensity with Cr3+ codoping when excited at a wavelength of 532 nm. This can be
attributed to an efficient energy transfer from Cr3+ to Ni2+, which has been
confirmed by time-resolved spectroscopy [46].
• Bismuth ion is one of the most thoroughly investigated in the main groups of
TM ions, designated as a wonder and exceptional metal because of its diverse
oxidation states and formation of bismuth clusters as well, which result from an
easy movement of electrons in the p orbital to form different chemical combi-
nations. This allows to behave as a smart optically active center in a variety of
host materials and found applications in broadband optical amplifiers and fiber
lasers [47]. Broadband NIR emission has been demonstrated in the range of
1100–1400 nm from the 3P1 ! 3P0 transition of Bi+ ions in Bi-doped glasses
[48] including borate, phosphate, aluminsilicate and chalcogenide glasses, as
shown in Fig. 8.12. The figure of merits for bandwidth and gain of the glasses are
better than those for Er3+-doped silicate glasses and Ti3+:Al2O3 and make
Bi-doped glasses the most promising gain medium for the broadband amplifiers
and tunable laser sources with a FWHM bandwidth larger than 250 nm and a
lifetime longer than 200 μs [50].
• Multicomponent: There has been a significant interest on the synthesis and
investigation of broadband NIR luminescent multicomponent materials by the
exploration of novel dopants of the main group of TM ions (Ti3+, Cr3+, Ni2+, Bi+
ions), because their electronic transitions in the outermost d or p orbital are
strongly influenced by their ligands. Thus their emission bands exhibit a wide
range of characteristics [39, 46, 51].
166 K.V. Krishnaiah et al.

Fig. 8.12 Energy level diagram of Bi+ ion in phosphate and germinate glasses, proposed based on
energy-matching conditions (NIR emission at 1300 nm) at different excitations. [49]

8.3 Tellurite Glasses and Their Background

Tellurite glasses have been intensively investigated due to their special properties,
which allows them to be utilized as a nonlinear optical material. The properties
include a large transmission window, high linear and nonlinear refractive indices,
high resistance against devitrification, high Ln ion solubility, slow corrosion rate,
large amount of Ln dopants, and high dielectric constant when compared to other
traditional oxide (silicate, borate, and phosphate) glasses, which allows their utili-
zation as a nonlinear optical material. An important characteristic of these materials
is the relatively low cutoff phonon energy compared with other traditional oxide
glasses. This property is essential to enhance the optical and fluorescence properties
of Ln-doped glasses, and thus these glasses are revealed to be strong candidates for
the development of photonic devices, particularly, optical amplifiers. The detailed
description of optical, dielectric, thermal, structural, and nonlinear properties of
tellurite glasses is presented, in previous chapters of this book. The first report on
the fabrication of tellurite glasses by Berzelious in 1834 pointed to transparent and
colorless glasses from the tetratellurites of the alkali metals and barium [52]. After
more than a century, a group of tellurite glasses (TeO2–PbO–Li2O/Na2O/BaO/
WO3/B2O3/ZnF2) were prepared, and their optical, dielectric, and thermal proper-
ties are investigated by J.E. Stanworth in 1952, based on the electronegativity
concept [53]. Since then these glasses have found applications in lasers, amplifiers,
optical switches, sensors, and supercontinuum sources which cover a large range of
spectral window between 400 and 5000 nm [54, 55]. The tellurite glass family is
divided into three categories: alkali, alkali earth, and heavy metal modified glasses
[56]. Although pure tellurium oxide glass has not yet been obtained, the TeO2 is
8 Broadband Emission in Tellurite Glasses 167

classified as an intermediate oxide according to Dietzel’s field strength concept

[57]. Specific interest in lithium–tellurite glasses was first reported by Brady et al.
[58] for the structural studies, and since has been proposed for use as a solid
electrolyte [59].
During the nineteenth and twentieth centuries, tellurite glasses were extensively
investigated as they possess nonlinear susceptibilities, tenfold higher than those
found for silicate glass and therefore they offer many opportunities for nonlinear
applications [60]. Several families from tellurite glasses have been used for the
development of optical amplifiers, lasers, nonlinear effects [61], and Raman ampli-
fiers (TeO2–Bi2O3–BaO) [62]. Moreover, the most prominent composition Na2O–
ZnO–TeO2 among the ternary families for fabricating efficient fibers for the
development of optical amplifiers. First, binary (TeO2–PbO and TeO2–BaO) [53]
and then ternary [63–65] followed by multi-component Na2O–ZnO–Bi2O3–TeO2
[66] and Li2O–ZnO–Bi2O3–TeO2 [67] glass systems were invented. In addition,
tellurite multicomponent glasses, TeO2–BaO–SrO–Nb2O5–WO3 and TeO2–BaO–
SrO–Nb2O5–WO3–P2O5, have been successfully drawn into optical fibers which
were developed for their remarkable Raman gain sources [68]. Furthermore, the
ternary family of TeO2–Na2O–MgO, TeO2–Na2O–ZnO and TeO2–Na2O–BaO
glasses has been used to produce bulk glass billets of sufficient size for the
fabrication of amplifier fiber.

8.4 Broadband Emission in Tellurite Glasses

Optical communication technology turned a significant corner in the 1990s with the
development of optical fiber amplifiers, multiplexers, and de-multiplexers and
advanced toward exploiting its potential of large capacity for data transfer at
10Tbit/s rates in long-haul systems. This extended range with large capacity of
information transmission, combined with light routing functions, allowed the
transition from point to point communications to truly flexible network. Hence,
the use of the largest bandwidth amplifiers for networks is highly desirable. Doped
fiber amplifiers offer the capacity of inline amplification; however, there are many
choices for the dopant for any given specific application. The broadband emission
of some of Ln and TM ions in glasses, their spectral transitions, full width at half
maximum (FWHM), peak wavelengths (λp), and the span of their emission band-
widths and potential applications are summarized in Table 8.1.
There is a great interest in Er3+-doped tellurite glass which is considered as one
of the finest candidates for 1.5 μm wavelength broadband optical amplification.
They have been the focus of interest for fiber amplifiers worldwide, since the optical
and spectroscopic properties of Ln-doped bulk glasses and optical fibers were
reported by Wang and coauthors [60, 74, 75]. Broadband amplification in the
L-band has also been reported in Er3+-doped tellurite fibers in 1997, which presents
a gain bandwidth of 80 nm [76]. Together with the analysis of spectroscopic and
optical properties of Ln-doped tellurite glasses, Wang et al. have also emphasized
168 K.V. Krishnaiah et al.

Table 8.1 Comparison of emission, FWHM, application, peak wavelength, and their transitions
of Ln and TM ions
Ln/ Emission Peak
TM wavelength FWHMTelecommunication wavelength
ion (nm) (nm) window/application (nm) Transition Ref.
Ce 450–750 120 White-LED 550 2
D3/2(5d ) ! 2F5/2,7/ [18]
3+
2
Pr3+ 1200–1700 100 O-band 1300 nm 1
G4 ! 3H5 [22]
Nd 1100–1200 50 O-Band 1.2 5
I6 ! 5I8 [69]
3+
40 High energy high 1.45 5
I4 ! 5I6
power lasers
Dy 1250–1450 82 U-band/w-LED/ 1330, 1670 6
F11/2(6H9/2) ! 6 [28]
3+
1600–1800 lasers H15/2, 6H11/2 ! 6
H15/2
Ho 1150–1250, 160 E-band 1209, 1487, 5
I6 ! 5I8, 5I5 ! 5I6, [70]
3+
1450–1520, 2050 5
I7 ! 5I8
1900–2200
Er3+ 1400–1600 105 S, C/C + L-band, 1530 I13/2 ! 4I15/2
4
[3]
laser cooling of
solids
Tm 1100–1300 105 E, S-band, laser 1200 H4 ! 3F4
3
[71]
3+
1400–1600 cooling of solids 1460
Yb 900–1100 60 YDFA, high energy 975 F5/2 ! 2F7/2
2
[15]
3+
high power lasers,
laser cooling of
solids
Bi+ 1200–1700 300 O,E,S,C,L,U 1200 3
P1 ! 3P0 [72]
Ti3+ 500–860 190 nmTunable laser 637 2
E ! 2 T2 [73]
Cr 1100–1700 O,E,S,C,L,U 1400 2
E ! 4A2 [45]
4+

Ni 1120–1800 290 O,E,S,C,L,U 1300 T2 ! 3A2

3
[43]
2+

the importance of tellurite glass fibers to engineer superbroadband amplifiers, not

possible with traditional Ln-doped silica fibers. This enhanced bandwidth of
tellurite glasses for optical communication devices is a major driver for research
on this subject area since 1995. In 1997, the first demonstration of Er3+-doped
tellurite glass fiber amplifier for multichannel amplification, at the NTT laboratory
in Japan [76], further accelerated worldwide interest [77, 78] for device engineer-
ing. The combination of high solubility of the Ln ion, low transmission loss, large
optical nonlinear coefficient, and good glass formation has provided a strong base
for developing the glass for integrated optical devices in the optical transmission
region from 400 to 5000 nm [55].
Erbium-doped fiber is a central component for all optical amplifiers; fiber
preamplifier and power and distributed amplifiers; and repeaters. Most of the fibers
are produced from silica by the variations of the standard fiber fabrication
8 Broadband Emission in Tellurite Glasses 169

technology, which has some disadvantages, such as low solubility of the Ln ions
and narrow bandwidth. These are carefully addressed and solutions have been
found, to overcome the aforementioned drawbacks. The fabrication methods,
selection of particular fibers for specific applications with high photoluminescence
quantum yield (PLQY), have been reported [79]. In addition, wavelength division
multiplexing (WDM) is an essential tool for realizing ultrahigh-capacity optical
fiber communication systems by the advent of wavelength division fiber amplifiers
(WDFAs) in 1987 [80]. In order to increase the transmission volume of the WDM
system, it is essential to flatten the gain spectrum and broaden the bandwidth of
conventional EDFAs. One of the ways to meet this requirement is by selecting
tellurite-based EDFA’s (EDTFAs) which can provide a wide amplification band-
width in the range of 1400–1600 nm region for the 4I13/2 ! 4I11/2 transition, due to
their higher excited state lifetime and emission cross-sections [76]. These are
considered as one of the best materials to realize broadband amplification, even
though they have low glass thermal stability and strong upconversion [60]. The
non-radiative processes include both the characteristics of the glass host and the Ln
ion concentration. The control of these non-radiative processes is challenging for
the development of efficient EDFAs. Moreover, EDFAs can also offer a broad
bandwidth amplification and a higher radiative transition efficiencies. Hence, they
are promising candidates for fabricating novel optical amplifiers in the C-band and
L-band as a means of extending the transmission bandwidth beyond the range
available from conventional silica-based EDFAs [76, 81].
Conventional EDFAs can be produced from Er3+-doped alumino–silicate glass
that offers not only a superior performance over nonsilicate ones in terms of the
long-term stability but also an efficient splicing to single-mode fibers. However, the
intrinsic gain spectrum of conventional EDFAs is not uniform, and the gain
bandwidth is limited to ~30 nm with a low stimulated emission cross sections
[82]. On the other hand, fiber amplifiers based on Er3+-doped fluoride and tellurite
glasses usually exhibit gain bandwidths wider than 50 and 70 nm, respectively
[74, 83]. Particularly, the EDTFAs are now a well-known candidate for broadband
optical amplifiers in WDM systems, which exhibit a wide effective gain bandwidth
by virtue of a high stimulated emission cross section for the 1.5 μm emission
compared to the conventional silicate-based counterparts [84].
An emerging demand for broadband fiber amplifiers in optical communications
has changed to Tm3+-doped glasses. The thulium-doped fiber amplifiers (TDFAs)
operate in the ~1.45 (3H4 ! 3F4) and ~1.65 μm (3F4 ! 3H6) bands, which are
potentially useful for supplementary service channels for home security monitoring
and network maintenance channels, respectively [85]. A broadband emission cen-
tered at 810 nm is responsible for the Tm3+:3H4 ! 3H6 transition effectively
overlapped via reabsorption with the absorption transition of Dy3+:6H15/2 ! 6F7/2
in tellurite glasses. It is well known that TDFAs in the 1450 nm band suffer from
ASE at around 810 nm. Instead of Tm3+-doped layer, a Dy3+-doped layer is
introduced in an optical fiber that absorbs the 810 nm emission effectively and
consequently suppresses the ASE. When pumped at 980 nm, a broad ASE is
observed, ~100 nm wide in the range of 1450–1650 nm band around 1530 nm.
170 K.V. Krishnaiah et al.

It was found that the emission spectrum from EDTFs is almost twice as broad as the
corresponding spectrum in Er3+-doped tellurite glass. The changes in ASE with
regard to fiber length and pumping power were reported in Er3+/Yb3+-codoped
tellurite fiber [86]. A demand for a high transmission capacity continues to increase
with data, voice, and video signals via an internet and other emerging multimedia
applications. This can be addressed by the recent availability of optical fibers
(AllWave™) with low loss between 1280 and 1700 nm. The use of Tm3+ ions as
a dopant will make possible to extend the band in the spectral range from 1450 to
1500 nm, the S-band amplification region. In order to achieve both high gain and
flat gain over the entire spectral region, the choice of a suitable host glass matrix for
Tm3+ dopant is of importance [71].
Weber et al. [87] reported the broadband emission and excited state lifetimes in
the NIR region at 1.34 μm (effective bandwidth of 54 nm) and 1.06 μm (effective
bandwidth of 28 nm) in Nd3+-doped tellurite glasses containing various alkali and
higher-valence-state cations and in a series of new phospho-tellurite glasses. Later,
in 1994, Wang et al. [88] reported the broadband emission at 1.3-μm of Nd3+- and
Pr3+-doped tellurite glasses of the compositions, TeO2–ZnO–Li2O/Na2O/K2O/
Rb2O/Cs2O. They described glass synthesis, formation, waveguide fabrication,
optical properties, and suitability for fiber drawing, based on TeO2-ZnO-Na2O
glasses doped with Ln ions, as their phonon energy is lower than that of silicate
glasses. As a result, Pr3+- and Nd3+-doped tellurite glasses have been considered for
fiber amplifiers at 1.3 μm.

8.4.1 Spectroscopy of Broadband Emission in Tellurite

Glasses

The field of spectroscopy of Ln3+-doped glasses is of great importance for achiev-

ing highly efficient and improved optical quality materials and lowering cost. This
is because of their utilization in lasers, waveguides, and luminescent devices in the
field of photonics. It is observed that although the absorption and emission band
positions of these ions are not very sensitive to their environment, their radiative
and non-radiative transition probabilities can be modified with the host material.
These optical properties are important for evaluating the lasing characteristics of
any material. The effect of the host matrix on the local environment of a given Ln
ion with its first nearest neighboring anions such as oxygen can be explained by
using Judd–Ofelt (JO) [89, 90] theory and its intensity parameters Ω2, Ω4, and Ω6.
In addition, the absorption and emission spectra of the Ln ion depends on the nature
of the electric-induced dipole–dipole interaction.
In 1962, two independent researchers, Judd and Ofelt, published the work on
electric dipole transitions of 4f–4f intensities of Ln ions in crystals and glasses.
These two articles have been cited in different publications more than 5000 times
on the 50th anniversary of their publication. The methodologies proposed in their
8 Broadband Emission in Tellurite Glasses 171

articles are tools to predict the optical and spectroscopic properties of Ln ions in a
liquid or solid environment (e.g., spontaneous radiative emission rates, branching
ratios, radiative lifetimes, etc.). The physical mechanisms including CF, electron
correlation effects based on the statistical and dynamic models, and spin-orbit
interactions, have been discussed with the JO theory [91]. According to the JO
theory, the oscillator strength for the absorption band that corresponds to an
electronic transition from the ground state ΨJ to an excited state Ψ0 J0 can be
evaluated from the following relation:
2
8π 2 mcυ ðn2 þ 2Þ 8π 2 mcυ
f exp ¼ f ed þ f md ¼ Sed þ nSmd ð8:1Þ
3he ð2J þ 1Þ
2 9n 3he2 ð2J þ 1Þ

where m is the mass of the electron, c is the speed of light in vacuum, ν is the
wavenumber (cm
1) associated with the transition, h is the Planck constant, e is the
ðn2 þ2Þ
2

electronic charge, n is the refractive index of the host lattice, 9n is the local field
correction factor for the effective field induced by the host lattice, J is the total
angular momentum of the ground state, (2 J + 1) is the degeneracy, and 2S+1LJ and
S is the line strength of the electronic transition involved. In addition, the magnetic
dipole transitions with selection rules ΔS ¼ ΔL ¼ 0 and ΔJ ¼ 0, 1 can contribute
to the oscillator strength of Ln ions.
In the case of electric dipole transitions,
X    2
Sed ¼ e2 Ωλ ΨJ U λ Ψ0 J 0 ð8:2Þ
λ¼2, 4, 6

Regarding to magnetic dipole transitions,

e2 h2
ðΨJ kL þ 2SkΨ0 J 0 Þ
2
Smd ¼ ð8:3Þ
16π 2 m2 c2

where Sed and Smd, Ω2, Ω4, and Ω6 are called JO intensity parameters which are
independent of the CF and depend only on the specific Ln ion. The kU(λ)k2 are the
doubly reduced matrix elements for a transition Ψ J ! Ψ 0 J0 and are evaluated from
the intermediate coupling approximation.
The JO parameters along with refractive index (n) of the material are used to
predict the radiative properties of excited states of Ln3+ ion. The radiative transition
probability (A) for the Ψ J ! Ψ 0 J0 transition is evaluated by using [92]:
2
64π 4 υ3 nðn2 þ 2Þ 64π 4 υ3
AðΨJ; Ψ0 J 0 Þ ¼ Sed þ n3 Smd ð8:4Þ
3hð2J þ 1Þ 9 3hð2J þ 1Þ
0 0
The total radiative transition probability (AT) is the sum of the A(ΨJ, Ψ J ) terms
calculated from all the existing states of an excited state
172 K.V. Krishnaiah et al.

X
AT ðΨJ Þ ¼ AðΨJ; Ψ0 J 0 Þ ð8:5Þ

AT is total radiative lifetime (τrad) of an excited state given by

1
τrad ðΨJ Þ ¼ ð8:6Þ
AT ðΨJ Þ
0 0
The peak stimulated emission cross section, σ(λp)(ΨJ , Ψ J ), between the states
0 0 0 0
ΨJand Ψ J having a probability of A(ΨJ, Ψ J ) can be expressed as

  λp 4
σ λp ðΨJ; Ψ0 J 0 Þ ¼ AðΨJ; Ψ0 J 0 Þ ð8:7Þ
8πcn2 △λeff

where λpis the transition peak wavelength and Δλeff is an effective bandwidth, given
by dividing the area of the emission band by height of the band. The laser transitions
are characterized by stimulated emission cross sections which are as high as
possible for a given laser material. Another interesting spectroscopic parameter,
to characterize the Ln ion luminescence, is the effective bandwidth (Δλeff) and
expressed as
Z
I ðλÞdλ
△λeff ¼ ð8:8Þ
Imax

where I(λ) is the intensity at the wavelength λ and Imax is the intensity at the peak
wavelength λp.
In 1960, McCumber at Bell Laboratories has invented a detailed theory [93],
now known as the famous McCumber (MC) theory, which refers a quantitative
correlation between the absorption and the emission cross sections of dopants
(Ln or TM) in a given solid-state laser gain media. He used the thermodynamic
principles by following the earlier theoretical investigations by Albert Einstein. MC
theory is particularly more useful for evaluating spectroscopic properties of quasi-
three-level gain media, e.g., of Ln-doped type. Hence, the MC relation for evalu-
ating the emission cross section (σ em(λ)) is given by
 
Z1 EZL
hcλ
1
σ em ðλÞ ¼ σ ab ðλÞ exp ð8:9Þ
Zu kT

where Zl and Zu are related to the ground and excited state partition functions,
respectively; EZL the zero-line energy, is equal to the energy separation between the
lowest energy Stark components of the ground and excited states of Ln ion; and h is
Planck’s and k is Boltzmann’s constants. The absorption cross section, σ ab(λ), of a
particular transition can be evaluated by using the following equation:
8 Broadband Emission in Tellurite Glasses 173

I 0 
2:303log
σ ab ¼ I
ð8:10Þ
Cl

where C is the Ln ion concentration and l is thickness of the glass sample.

The experimental lifetime of any luminescent level is given by [94, 95]

1 1
¼ þ W MPR þ W ET þ W OH ð8:11Þ
τf τrad

where τf is experimental lifetime and τrad is radiative lifetimes, WMPR is the

multiphonon relaxation (MPR) rate, WET WETis the energy transfer rate among
the Ln3+ ions and other impurities, and WOH is the energy transfer rate between Ln
ion and OH
radicals.
The experimental lifetime and the stimulated emission cross section are also
crucial parameters for an optical amplification process. In order to test the feasibil-
ity of laser glasses for the development of EDFA devices, the products of
τexp  FWHM and the σ em  FWHM represent the figures of merit for gain and
bandwidth, respectively [96], since a high value of these parameters is mandatory
for optical amplification.
Optical gain cross section is an important factor to evaluate the performance of a
gain media. It has been suggested that the laser glass should have higher gain cross
section, σ g, to achieve a higher gain than loss, for laser applications. The
wavelength-dependent gain cross section is expressed as [97]

σ g ðλÞ ¼ γσ em ðλÞ
ð1
γ Þσ ab ðλÞ ð8:12Þ

where γ (between 0 and 1 in steps of 0.2) represents the ratio of population-inverted

ions to the total Ln ion concentration.
JO intensity parameters are evaluated from the experimental oscillator strengths
of various absorption bands and refractive index of glass. In this chapter, the
environment around the Ln ion in tellurite glass family (binary, ternary, and
multi-component) is considered by taking only the Er3+ ion (serving as a EDFA)
for the broadband emission (4I13/2 ! 4I15/2), as an example. The comparison of
three intensity parameters, spectroscopic quality factor, and their trend for the Er3+:
tellurite glasses is listed in Table 8.2. In general, Ω2 is sensitive to the asymmetry in
the area of the Ln ions with short-range effects, and the Ω4 and Ω6 are correlated
with the long-range effects of glass host [129]. In a similar fashion, Ω2 is found to
be affected by the covalence between the Ln ions and the ligand anions, while Ω4
and Ω6 reflect the bulk properties of the host such as rigidity and density [130]. The
spectroscopic quality factor (χ) is an important predictor for stimulated emission in
a laser active medium [131], initially represented for Nd3+ ion [132] and later used
for evaluating all of the lasing ions. The value of χ was found to be as high as 5 for
TeO2–P2O5–Al2O3–La2O3 [98] glass and as low as 0.18 for TeO2–PbF2–B2O3 [99]
glasses, which means that the tellurite glass modified with P2O5 is suitable for the
laser emission. The trend of the JO parameters Ω2 > Ω4 > Ω6 indicates the
174 K.V. Krishnaiah et al.

Table 8.2 Comparison of refractive index (n, at 1550 nm), JO intensity parameters (Ωλ, λ ¼ 2,
4 and 6, 10
20 cm2), trend of JO parameters and spectroscopic quality factor (χ) with different Er
3+
systems
JO parameters
χ ¼ Ω4/
System n Ω2 Ω4 Ω6 Ω6 Trend
TeO2–WO3–ZrO2 [3] 2.103 1.72 0.93 0.79 1.17 Ω2 > Ω4 > Ω6
TeO2–P2O5–Al2O3–La2O3 [98] 2.0* 3.4 1.0 0.20 5.00 Ω2 > Ω4 > Ω6
TeO2–PbF2–B2O3 [99] – 5.69 0.31 1.74 0.18 Ω2 > Ω6 > Ω4
TeO2–ZnO–Na2O–P2O5 [100] – 4.54 1.12 1.34 0.84 Ω2 > Ω6 > Ω4
TeO2–Bi2O3–ZnO–Nb2O5 [101] 2.15 4.41 1.07 1.57 0.68 Ω2 > Ω6 > Ω4
TeO2–ZnO–Nb2O5–TiO2 [102] 1.945 2.07 0.50 0.32 1.52 Ω2 > Ω4 > Ω6
TeO2–ZnO–ZnF2 (TZF16) [103] 2.002 4.76 1.57 1.13 1.38 Ω2 > Ω4 > Ω6
TeO2–ZnO–ZnF2 (TZF35) [103] 1.833 3.02 1.28 1.14 1.81 Ω2 > Ω4 > Ω6
Na2O–Nb2O5–TeO2 (NNT) [104] 2.058 6.86 1.53 1.12 1.36 Ω2 > Ω4 > Ω6
TeO2–ZnO–ZnF2–Na2O (ETZ) [105] 2.033# 5.60 1.87 1.03 1.81 Ω2 > Ω4 > Ω6
TeO2–ZnO–PbO–PbF2–Na2O (TZPPN) 2.040 4.28 1.68 1.38 1.21 Ω2 > Ω4 > Ω6
[106]
TeO2–BaF2–Na2O (TBNE–1e) [107] 1.747 5.63 1.42 1.17 1.21 Ω2 > Ω4 > Ω6
B2O3–ZnO–TeO2 (1ErTB) [108] 2.409a 4.96 2.02 2.39 0.84 Ω2 > Ω6 > Ω4
TeO2–AlF3–YF3–CaF2–BaF2–SrF2– – 2.59 1.27 0.89 1.42 Ω2 > Ω4 > Ω6
MgF2 [109]
TeO2–AgO–BaO–Nb2O5 [110] 2.071 4.76 1.06 0.89 1.19 Ω2 > Ω4 > Ω6
TeO2–Bi2O3–BaO–Li2O [111] 2.31 4.61 1.43 0.73 1.95 Ω2 > Ω4 > Ω6
TeO2–Ga2O3 [112] 1.9729 6.46 1.64 1.47 1.11 Ω2 > Ω4 > Ω6
TeO2–ZnO–LiF [113] 2.01 6.07 2.48 0.72 3.44 Ω2 > Ω4 > Ω6
TeO2–PbF2–AlF3 [114] 2.11 5.93 1.99 1.11 1.79 Ω2 > Ω4 > Ω6
TeO2–BaF2–BaO–La2O5 [115] – 5.97 1.64 1.38 1.18 Ω2 > Ω4 > Ω6
TeO2–GeO2–Na2O [116] 2.01 4.22 1.39 0.38 3.65 Ω2 > Ω4 > Ω6
TeO2–TiO2–Al2O3 [117] 2.15 1.60 0.67 0.43 1.55 Ω2 > Ω4 > Ω6
TeO2–ZnO–PbO–TiO2–Na2O [118] 2.04 1.67 0.95 0.86 1.10 Ω2 > Ω4 > Ω6
TeO2–ZnO–Nb2O5–GeO2 [119] 2.01* 5.62 1.50 1.18 1.27 Ω2 > Ω4 > Ω6
TeO2–B2O3–Li2O [120] 1.65* 7.36 2.79 2.31 1.20 Ω2 > Ω4 > Ω6
TeO2–WO3–PbO–BaF2 [121] 2.058 5.83 1.67 0.87 1.91 Ω2 > Ω4 > Ω6
TeO2–TiO2–Nb2O5–BaF2 [121] 2.150 6.45 1.53 0.90 1.71 Ω2 > Ω4 > Ω6
TeO2–TiO2–Nb2O5 [121] 2.191 5.38 1.36 0.89 1.52 Ω2 > Ω4 > Ω6
TeO2–ZnO–ZnF2 [122] 1.84 3.02 1.28 1.14 1.12 Ω2 > Ω4 > Ω6
TeO2–WO3–Bi2O3 [123] 2.29& 6.06 1.57 0.95 1.65 Ω2 > Ω4 > Ω6
TeO2–ZnO–Li2O [124] – 6.56 1.60 1.44 1.11 Ω2 > Ω4 > Ω6
TeO2– B2O3–Na2O–ZnO–Li2O–CaO 1.688 6.69 1.37 1.09 1.26 Ω2 > Ω4 > Ω6
[125]
TeO2–B2O3–SrO–BaO–Li2O–LiF [126] 1.689 7.63 1.49 1.80 0.82 Ω2 > Ω4 > Ω6
TeO2–ZnO–WO3–TiO2–Na2O [127] – 0.66 0.35 0.26 1.34 Ω2 > Ω4 > Ω6
AlF3–YF3–CaF2–BaF2–SrF2–MgF2– – 2.74 1.12 0.96 1.17 Ω2 > Ω4 > Ω6
TeO2 [128]
Asterisk indicates ‘n’ at 633 nm, ampersand indicates ‘n’ from minimum deviation method and
hash indicates ‘n’ at 1064 nm
a
Evaluated from Dimtrov and Sakka relation
8 Broadband Emission in Tellurite Glasses 175

existence of a higher degree of covalency or/and asymmetry around the Ln ions; all
the tellurite glasses exhibit the same trend except for B2O3–ZnO–TeO2 (1ErTB)
[108], TeO2–ZnO–Na2O–P2O5 [100], TeO2–PbF2–B2O3 [100], and TeO2–Bi2O3–
ZnO-Nb2O5 [101] glasses. This may be due to the modification of glass network
with glass formers, B2O3, P2O5, and Bi2O3, which leads to increase the mechanical
stability of the glass.
Stimulated emission cross section (σ em), full width half maximum (FWHM),
figure of merit for the bandwidth (σ em  FWHM), radiative lifetime (τR), fluores-
cence lifetime (τf), figure of merit for the gain (τf  FWHM), quantum efficiency
(η), and photoluminescence quantum efficiency (PLQY) of the Er3+:tellurite glasses
for the 4I13/2 ! 4I15/2 transition are presented in Table 8.3. For lasers and amplifiers,
the stimulated emission cross section should be as high as possible. From Table 8.3,
the σ em is found to be as high as 11.96 cm
2 for TeO2–ZnO–Nb2O5–TiO2 [102] and
as low as 1.01 cm
2 for TeO2-WO3-Bi2O3 [123] glass. For optical amplification,
the Er3+:glass fiber should exhibit a large bandwidth to make use of sending/
receiving the signal in the specified region of the spectrum. The FWHM reported
for the Er3+-doped fiber system is as large as 100 nm for both TeO2–WO3–ZrO2 [3]
and TeO2–ZnO–WO3–TiO2–Na2O [127] and as low as for TeO2–P2O5–Al2O3–
La2O3 [98] glasses. A large bandwidth for the tungstate–tellurite glass may be due
to the presence of WO3 in both the glasses. Hence, tellurite glasses are highly
suitable, particularly tungstate–tellurite glasses, for wideband optical amplification
applications. In a similar fashion, the figure of merit of the glass should be as high as
possible for both laser and amplifier applications. The value of figure of merit for
the bandwidth has been found to be as high as 913 cm3 for TeO2–WO3–ZrO2 [3]
and as low as 74 cm3 for TeO2–WO3–Bi2O3 [123], comparing all reported glasses.
The FWHM, σ em and σ em  FWHM for the 4I13/2 ! 4I15/2 transition of Er3+-doped
different glasses are presented in Table 8.4. It is interesting to note that tellurite
glasses have large bandwidths when compared to other traditional phosphate,
silicate, bismuth, and germinate glasses. To make an efficient amplifier requires a
long metastable state lifetime. The glass with a long lifetime of 8.2 ms is TeO2–
ZnO–ZnF2 (TZF35) [103] glass, which is the largest value when compared to all the
tellurite glasses. In contrast, the quantum efficiency, estimated via the τf/τR, is
found to be as high as 184% for TeO2–ZnO–ZnF2 (TZF35) [103] glass. It is
observed that the PLQY is reported for the TeO2–WO3–ZrO2 [3] glass and found
to be as high as 20%, by the absolute method with an integrating sphere measure-
ment. The research can be focused in this direction for establishing the better
quantitative and qualitative properties to develop the fiber lasers and amplifiers.
The wavelength-dependent gain cross-section spectra of Er3+-doped tellurite
glass for different population inversions (γ) are shown in Fig. 8.13. From the
Fig. 8.13, the gain cross section increases, and the gain profile encompasses at the
shorter wavelength side of the spectra with increasing population inversion. As can
be seen from Fig. 8.13, gain cross section is positive at γ ¼ 0.5 in the spectral range
from 1531.5 to 1650 which covers the C + L bands. At higher population inversion,
γ ¼ 1, one can expect a broader bandwidth in the spectral range of 1450–1650 nm
and a higher gain cross section at 1550 nm. In Er3+-doped tellurite glasses, this
 176

Table 8.3 Comparison of emission cross section (σ em, 10
21 cm2), full width and half maximum (FWHM, nm), figure of merit for the bandwidth
(σ em  FWHM, 10
28 cm3) and gain (τf  FWHM, 10
23 cm2 s), radiative (τR, ms) and fluorescence (τf, ms) lifetimes and quantum efficiency (η, %),
absolute photoluminescence quantum yield (PLQY, %) from the integrating sphere method for the excited levels
System σ em FWHM σ em  FWHM τR τf τf  FWHMa η PLQY
Er3+-doped glasses: 4I13/2 level
TeO2–WO3–ZrO2 [3] 8.7 105 913 2.99 3.23 3.39 108 20
TeO2–P2O5–Al2O3–La2O3 [98] 6.1 41 250 7.2 4.1 1.68 56 –
TeO2–PbF2–B2O3 [99] 5.37 86 461 7.31 – – – –
TeO2–ZnO–Na2O–P2O5 [100] 6.79 78 529 4.21 3.25 2.53 77 –
TeO2–Bi2O3–ZnO–Nb2O5 [101] 6.81 96 653 2.64 3.84 3.68 145a –
TeO2–ZnO–Nb2O5–TiO2 [102] 11.96 55 657 5.52 3.73 2.05 70 –
TeO2–ZnO–ZnF2 (TZF16) [103] 8.8 73.4 645 3.23 5.9 4.33 182a –
TeO2–ZnO–ZnF2 (TZF35) [103] 7.2 68.8 516 4.45 8.2 5.62 184a –
Na2O–Nb2O5–TeO2 (NNT) [104] 10.2 – – 3.02 2.9 – 96 –
TeO2–AgO–BaO–Nb2O5 [110] 7.93 49.32 391 4.8 3.46 1.70 72 –
TeO2–Bi2O3–BaO–Li2O [111] 8.69 58 504 3.21 2.64 1.53 82 –
TeO2–Ga2O3 [112] 8.54 58 495 2.90 2.0 1.16 69 –
TeO2–PbF2–AlF3 [114] 8.1 84 689 2.78 1.76 1.44 63 –
TeO2–BaF2–BaO–La2O5 [115] 6.82 91 621 3.70 2.69 1.84 72 –
TeO2–GeO2–Na2O [116] 7.79 81 630 8.8 7.5 6.07 85 –
TeO2–TiO2–Al2O3 [117] 8.6 – – 4.1 2.2 – 54 –
TeO2–ZnO–PbO–TiO2–Na2O [118] 6.3 70 441 4.04 – – – –
TeO2–ZnO–Nb2O5–GeO2 [119] 7.85 66 518 3.48 2.9 1.91 83 –
TeO2–B2O3–Li2O [120] 7.10 85 603 4.52 0.55 0.46 12 –
TeO2–WO3–PbO–BaF2 [121] 6.3 78 491 3.5 3.8 2.96 108 –
K.V. Krishnaiah et al.
TeO2–TiO2–Nb2O5–BaF2 [121] 6.9 76 522 3.0 4.6 3.49 155a –
TeO2–TiO2–Nb2O5 [121] 6.5 80 520 2.9 4.1 3.28 141a –
TeO2–ZnO–ZnF2 [122] 7.8 75 585 7.6 5.5 4.12 72 –
TeO2–WO3–Bi2O3 [123] 1.03 77 79 2.3 3 2.31 130a –
TeO2–ZnO–Li2O [124] 9.71 70 679 3.79 6.89 4.82 181a –
TeO2–B2O3–Na2O–ZnO–Li2O–CaO [125] 7.38 96 708 4.98 1.93 1.85 39
TeO2–B2O3–SrO–BaO–Li2O–LiF [126] 8.73 54 471 4.2 2.18 1.17 52 –
TeO2–ZnO–WO3–TiO2–Na2O [127] 8.64 100 864 7.24 – – – –
TeO2–AlF3–YF3–CaF2–BaF2–SrF2–MgF2 [128] 8.3 55 456 9.52 5.23 2.87 55 –
TeO2–ZnO [133] 8.81 82 722 3.33 2.23 1.82 66 –
TeO2–Na2O [134] 8 65 – – 5 3.25 – –
TeO2–WO3 [134] 8 85 680 – 5 4.25 – –
8 Broadband Emission in Tellurite Glasses

Tm3+-doped glasses: 3H4 level

TeO2–ZnO–Na2O [135] – 110 – 0.43 0.38 0.42 88 –
TeO2–K2O–La2O3 [136] 3.54 124 441 – – – – –
TeO2–ZnO–Na2O [137] 0.36 114 40 0.35 0.32 0.38 100 –
TeO2–TiO2–Nb2O5 [138] 4 112 448 0.29 0.29 0.32 100 –
TeO2–WO3–PbO (TWP5) [139] 3.8 101 383 0.3 0.3 0.30 100
TeO2–WO3–PbO (TWP25) [139] 4 102 408 0.27 0.27 0.27 100
a
Evaluated from the reported values
177
178 K.V. Krishnaiah et al.

Table 8.4 Comparison of FWHM for the 4I13/2 ! 4I15/2 transition of Er3+ ion in different glassy
systems
System σ em FWHM σ em  FWHM
TeO2–WO3–ZrO2 [3] 8.7 105 913
TeO2–BaF2–BaO–La2O3 [115] 6.82 91 621
Bismuth [140] 7 79 554
Silicate [141] 5.5 40 220
Phosphate [142] 6.4 37 237
Germanate [143] 5.7 42 239

10
Optical gain cross-section (10–21 cm2)

8
6
4

2
0
Emission
–2 γ = 0.2
–4 γ = 0.4
γ = 0.5
–6 γ = 0.6
–8 γ = 0.8
Absorption
–10
1400 1450 1500 1550 1600 1650 1700
Wavelength (nm)
Fig. 8.13 Gain cross-section spectra of Er3+ ion for the 4I13/2 ! 4I15/2 transition as a function of
different population inversions (Reprinted with permission from ref. [122], copyright, 2013,
Elsevier)

value is found to be as high as 7.1 cm
1 for TeO2–ZnO–ZnF2 [122] when compared
with 1.8 cm
1 for TeO2–ZnO–Na2O–BaO–La2O3 [144], 3.4 cm
1 for TeO2–
Nb2O5–ZnO–LiF [145], and 4.54 cm
1 for TeO2–ZnO–PbO–PbF2–Na2O [106].

8.4.2 Influence of Ln Concentration on Broadband Emission

Broadband emission is highly influenced by the Er3+ ion concentration, shown in

Fig. 8.14. The inhomogeneously broadened emission may be due to site-to-site
variations [134, 146, 147] which leads to modification of the peak emission
8 Broadband Emission in Tellurite Glasses 179

Fig. 8.14 Normalized

broadband emission spectra
of Er3+ ions in TeO2–WO3–
ZrO2 glasses for various
concentrations, and also
absorption spectrum is
represented with dotted line
(Reprinted from ref. [3],
copyright, 2015, American
Scientific Publishers)

wavelengths, profiles, and FWHMs with increasing Er3+ concentration. The inten-
sity of broadband emission decreases after an initial increase, and the emission peak
shifts to a longer wavelength, whereas the emission band at the longer wavelength
decays with increasing Er3+ concentration. On the other hand, FWHM increases till
0.5 mol% with a maximum FWHM of 106 nm and thereafter decreases with
increasing Er3+ concentration. This broadening is due to the reabsorption which
usually appears in a three-level system whenever the absorption and emission
spectra overlap each other. The factors that directly influence the broadening of
emission spectra are many—sample thickness, Ln ion concentration, and optical
path length [3, 133, 148]—which lead to difference in broadband emission profile,
excited state lifetime, and FWHM for the 4I13/2 ! 4I15/2 transition of Er3+ ion
[149]. With increasing Er3+ concentration, the spectral profile shows broadening at
the lower energy tail of the spectra due to the presence of two network formers,
TeO2 and WO3, and thus existence of two types of sites, compared to other tellurite
glasses. The resultant spectrum is inhomogeneously broadened by the combination
of contributions from all the sites [134, 146, 147]. When the absorption and
emission peaks are normalized, the main features, peak positions of absorption,
and emission lines remain the same; however, the large overlap of absorption and
emission spectra at a peak wavelength (λp) broadens the 1.5 μm emission with
increasing Er3+ concentration. Moreover, absorption analysis shows no significant
change in line shape and bandwidth of the absorption bands even at higher Er3+
concentrations [133]. Babu et al. reported that the FWHM of the 4I13/2 ! 4I15/2
transition increases from 67 to 93 nm when Er3+ ion concentration is increased
between 0.1 and 2.0 mol% and then decreases to 77 nm for 3.0 mol% [150].
The broadband emission can be enhanced by energy transfer process in codoped
systems. The process of energy transfer in Er3+/Tm3+-codoped tellurite (TeO2–
ZnO–Na2O–GeO2) glasses has allowed the expansion of the broadband emission to
cover the entire telecommunication window S, C + L and U bands in the range of
180 K.V. Krishnaiah et al.

Fig. 8.15 Normalized 1.2

0.3Er 4 4
emission spectra of Er3+- I13/2 I15/2
0.3Er-0.3Tm 3
and Er3+/Tm3+-doped 1.0 0.3Er-0.5Tm 4
H 3
F4
fluorotellurite glasses by

Intensity (arb. units)

0.3Er-0.7Tm
excitation at 792 nm 0.3Er-1Tm
0.8
wavelength laser (Reprinted
with permission from ref. λexc = 792 nm
[152], copyright, 2014, 0.6 3
F4 3
H6
Elsevier)
0.4

S C+ L U
0.2

0.0
1300 1400 1500 1600 1700
Wavelength (nm)

1400–1900 nm with a bandwidth of 134 nm at 1530 and 283 nm at 1790 nm for the
combination of Tm3+(0.05 mol%)/Er3+(1 mol%) ions. This type of broadening is
accomplished from the combination of both the ion emission through the energy
transfer processes 4I11/2(Er3+) ! 3H5(Tm3+), 4I13/2(Er3+) ! 3H4(Tm3+), and 4I13/2
(Er3+) ! 3F4(Tm3+). With increasing Tm3+ ion concentration, the broadband
intensity and lifetime of the Er3+ ion for the 4I13/2 ! 4I15/2 transition decreased [151].
Miguel et al. [152] reported a similar effect on the broadband emission of Er3+/Tm3+-
doped fluorotellurite glasses (TeO2-ZnO-ZnF2), and the spectra are shown in Fig. 8.15.
The FWHM was found to be 50–150 nm for the Er3+ (0.3 wt%) and Er3+(0.3 wt%)/Tm3+
(1 wt%)-doped glasses and for the 3H4 ! 3F4 and 4I13/2 ! 4I15/2 transitions, respectively.
In a similar fashion, broadband emission expands in Er3+/Tm3+-codoped TeO2–K2O–
Nb2O5–Gd2O3 glasses, in the range of 1.3–1.8 μm with a FWHM of 185 nm, observed
under 808 nm wavelength laser excitation. Furthermore, the intensity ratios of Er3+ and
Tm3+ emission change significantly with the Tm3+ ion concentration due to the energy
transfer and back transfer processes. The FWHM of tellurite glasses in Table 8.4 are
better than those reported for silicate glass (40 nm) [141], phosphate glass (37 nm) [142],
germanate glass (42 nm) [143], silica fiber (90 nm) [153], and LGS-E7 phosphate glass
(46 nm) [154]. On the other hand, the FWHM was reported to be 58, 127, 87, and 35 nm
for the Er3+/Yb3+-, Tm3+/Yb3+-, Er3+/Tm3+- and Nd3+-doped zinc-boro-tellurite
glasses, respectively [155].
The broadband intensity, bandwidth, quantum efficiency, and excited state
lifetime of Er3+ ion for the 4I13/2 ! 4I15/2 transition enhanced with increasing
sensitizer (Ce3+ and Yb3+) concentrations, in TeO2–ZnO–Nb2O5–Na2O glasses
when compared with single Er3+-doped glass. This is due to the efficient energy
transfer from Ce3+/Yb3+ ions to the Er3+ ion under 980 nm wavelength excitation
[156]. The relative gain at 1530 nm enhances by nearly 50% when a fiber is
codoped with Ce3+ ion, also the effective gain bandwidth for the laser operation
in the wavelength range of 1490–1620 nm, when the fiber is pumped with a 980 nm
wavelength laser [157]. The FWHM for the 4I13/2 ! 4I15/2 transition also increases
8 Broadband Emission in Tellurite Glasses 181

from 62 to 69 nm with increasing WO3 content in Er3+/Ce3+/Yb3+-codoped TeO2–

Bi2O3–TiO2 glass. The broadening of the spectra is explained by the presence of
different structural units, TeO3 trigonal pyramid and TeO4 trigonal bipyramid units,
BiO3 pyramidal and BiO6 octahedral units, and WO4 tetrahedral and WO6 octahe-
dral units. All these structural units have different spatial distributions and also the
LF strengths. As a consequence, the intensities of the Er3+ emission bands (due to
the presence of individual Stark transitions) are different in all the structural units,
which leads to inhomogeneous broadening in their emission profile [158]. The
FWHM further enhances to 290 nm in Tm3+-doped and 450 nm for the Tm3+/Ho
3+
ion-codoped multicomponent tellurite (TeO2–GeO2–PbO–PbF2–BaO–Nb2O5–
LaF3) glasses [159]. The MIR broadband emission of Dy3+ ion at 2.9 μm for the 6
H13∕2 ! 6H15∕2 transition with a bandwidth of 290 nm from Dy3+:TeO2–ZnO–Na2O
and TeO2–ZnO–Na2O–ZnF glasses, is significantly larger than the bandwidth
(225 nm) in ZBLAN fluoride glass [160].

8.4.3 Effect of Glass Formers and Modifiers on Broadband

Emission

The effects of glass formers (B2O3, P2O5, and SiO2) and modifiers (alkali, alkali
earth, and heavy metal oxides) on the broadband emission intensity, FWHM, and
excited state lifetimes, are summarized. The FWHM is 65 nm for the 1.53 μm
emission in erbium, which is slightly affected by the addition of P2O5 (phonon
energy ~1200 cm
1). It can be seen that the integrated intensity ratio of 1.53 μm
band and the 0.98 μm bands increases with P2O5 content. The addition of P2O5 into
the Er3+-doped tungsten–tellurite glass speeds up population by feeding the emis-
sion at 1.5 μm, through non-radiative 4I11/2 ! 4I13/2 relaxation, and simultaneously
suppresses UC without affecting the emission efficiency of the 4I13/2 level. The
non-radiative transition rate enhanced by a factor of 3.3 for the 6 mol% P2O5
content is obtained using upconversion spectroscopy, which is in good agreement
with that determined from the NIR emission spectra [161]. On the other hand, the
addition of B2O3, which has a phonon energy of 1500 cm
1, enhances the popula-
tion of the 4I13/2 level via increased multiphonon relaxation rate in Er3+-doped
tellurite fibers. The transition rate for a 4I11/2 ! 4I13/2 transition dramatically
increases by incorporating 15 mol% B2O3 without sacrificing the spectroscopic
properties of glass [162, 163]. With the addition of small amount of both WO3 and
B2O3 to the tellurite glass matrix (TeO2-ZnO-Na2O), the broadband emission
intensity increases by 15%, and profile broadening of about 18 nm has been
observed. [164].
Effect of cationic field strength of modifiers (Li2O, Na2O, K2O, MgO, CaO, SrO,
BaO) on the spectroscopic properties of the 1.5 μm broadband emission in Er3+-
doped tellurite glasses has been reported by Yu et al. [165]. The JO parameter, Ω6,
diminishes with decreasing of the cationic field strength, z/a2, of the modifiers. As
182 K.V. Krishnaiah et al.

the cationic field strength decreases, the polarization effect of the LF increases
around the Er3+ ion in these glasses, which leads to decreasing fluorescence peak
intensity and bandwidth. The glass modified with MgO exhibits higher excited state
lifetime, larger bandwidth, higher emission intensity, and larger figure of merit for
the bandwidth and gain compared to other alkali and alkali earth modified glasses.
This may be due to a higher value of cationic field strength for MgO-modified glass,
which decreases with an increase in atomic number of alkali earth elements in
tellurite glasses. Based on these studies, the modifier, MgO, might be a promising
candidate for large bandwidth amplifier and also for tunable ultrashort pulse laser
materials [165].
Shen et al. [166] described the composition effect on the broadband emission of
Tm3+-doped TeO2–Na2O–ZnO glass for application in S-band optical amplifica-
tion. The emission becomes broader, as the emission intensity at 1520 nm continues
to increase and shifts to longer wavelengths with increasing Na2O content. On the
other hand, the spectral profile and FWHM for the Tm3+:3H4 ! 3F4 transition in
TeO2–ZnO–Na2O–MoO2 show no significant change, whereas the lifetime
decreases with increasing MoO2 content [135]. Yang et al. [167] reported the effect
of codoping Ce3+(Eu3+,Tb3+) on the broadband Er3+ emission, as both the 4I11/2 and
4
I13/2 levels are depopulated with the addition of Eu3+ and/or Tb3+ ions, where as
the opposite trend is observed for the Ce3+ doping. Hence, Er3+/Ce3+-codoped
tellurite glasses have been seen to be efficient for broadband EDFA applications.
Nazabal et al. [168] reported the influence of fluorine substitution on the
broadband emission and lifetime of Er3+ ion in oxyfluorotellurite glasses. The
broadening of the emission peak is related to the changes in the glass matrix
structure and the surrounding coordination of the Er3+ ions due to the presence of
cations and anions. Three different types of structural units in the network former
provides an important site-to-site variation of the LF that is favored by the intro-
duction of fluorine to the oxide glass in a controlled proportion. In a similar fashion,
the broadband emission for the 4I11/2 and 4I13/2 transitions of Er3+ ion is found to
change with the ZnF2 content in TeO2–ZnO–ZnF2. The emission intensity of
the 4I11/2 transition increases when the ZnF2 concentration increases, while that
of the 4I13/2 transition only increases when a concentration of ZnF2 is greater than
10 mol%. However, the intensity of the 4I11/2 transition is only greater than that of
the 4I13/2 transition in glasses containing higher than 20 mol% of ZnF2 [169].

8.4.4 Influence of Heat Treatment on Broadband Emission

8.4.4.1 Glass-Ceramics Containing Nanocrystals

The broadband emission at 1.5 μm enhances significantly after heat treatment in

Er3+-doped glasses, as shown in Fig. 8.16. When the parent glasses (PG) are heat
treated above the glass transition temperature (Tg ¼ 345  C), a considerable
change occurs in the emission intensities of the three glass-ceramic
8 Broadband Emission in Tellurite Glasses 183

Fig. 8.16 Broadband

emission of Er3+-doped
glass and glass-ceramics
(Reprinted with permission
from ref. [170], copyright,
2007, Elsevier)

(GC) samples from their PG samples. On the other hand, the GCB sample heat
treated at 375  C for 3 h has stronger emission intensity than the GCA sample
which was heat treated at 375  C for 2 h. If the glass samples heat treated for
longer duration, a large crystalline environment can be created in glass sample,
leads to more Er3+ ions accommodated in the low phonon crystalline phase of
Er2Te5O13. Moreover, the GCC sample that was heated for 3 h at a higher
temperature of 380  C has a lower intensity than GCB due to its inhomogeneity
and reduced in transmittance, due to an increase in the size of the crystals at higher
temperatures, i.e. through increased scatter. The lifetime of the 4I13/2 level is
found to be 4.96 ms for PG, 5.36 ms for GCA, 5.52 ms for GCB, and 5.36 ms for
GCC samples. On the whole, the lifetime increases after heat treatment, which
further indicates that some Er3+ ions are successfully embedded within the low
phonon cubic crystal phase of Y2Te6O15 or Er2Te5O13. Although the lumines-
cence intensities of the oxyfluorotellurite glass samples are improved after heat
treatment, the FWHM of the samples are found to be 66, 65, 66, and 71 nm for PG,
GCA, GCB, and GCC, respectively. Both the emission and excited state lifetime
for the Er3+:4I13/2 ! 4I15/2 transition decrease for the GC sample when compared
to the PG sample. This is explained by the presence in part of the Er3+ ions in the
glass matrix, leading to a decrease in the distance between the Er3+ ions and an
increase in the energy transfer. The crystalline nature is confirmed by the CF
splitting of the absorption and emission spectra for the GC samples [171].
Jha et al. [172] reported that the broadening of broadband emission of Er3+ ion is
affected by the variation of nanoscale CF with the crystalline phase of Na2Te8O17 in
TeO2–Na2O glass. The FWHM and excited state lifetime of the 4I13/2 level are
found to be increased from 57 to 137 nm and 3.8–5.4 ms, respectively, with an
increase in the heat treatment temperature to 380  C, due to an increase in the size
of the nanocrystals. On the other hand, phonon energy of the matrix decreases with
increase in heat treatment temperature, which leads an increased lifetime and
FWHM of Er3+ ion.
184 K.V. Krishnaiah et al.

8.4.4.2 Metal Nanoparticles

The effect of the heat-treated silver nanoparticles (Ag NPs) on the spectroscopic
properties (JO parameters, excited state lifetime, broadband intensity, and band-
width) of Er3+ ion has been investigated. The surface plasmon resonance (SPR)
band of Ag NPs within the 510–550 nm range identified from the absorption
spectrum in addition to the distinctive absorption bands of Er3+ ions. The JO
parameters suggest that Er3+ ions are successfully partitioned into Ag NPs, which
enhances the electromagnetic field surrounding by the Er3+ ions. Hence, the pres-
ence of silver NPs nucleated and grown during the annealing process for 10 h above
the Tg improves both of broadband emission intensity and the excited state lifetime
relative for the 4I13/2 ! 4I15/2 transition. This is mainly attributed to the strong local
electric field induced by SPR of metal NPs and also due to energy transfer from the
silver NPs to Er3+ ions. The quenching of broadband emission due to the back
energy transfer from Er3+ ions to Ag NPs. The details of the nucleation and crystal
growth kinetics and their combined effect in tellurite glasses are described in
Chap. 23 of this book [110].

8.4.5 Broadband Emission in Thin Films

Tellurite glasses also have the advantage that they can be deposited as thin films. Er
3+
-doped phospho-tellurite thin films have been synthesized by pulsed laser depo-
sition (PLD) technique on silica substrates using an ArF laser at 193 nm wavelength
with 20 ns pulses and a laser fluence of 3.2 J/cm2, at various substrate temperatures,
oxygen pressures, and substrate-to-target distances. The propagation losses of the
phospho-tellurite thin films have been measured and found to be in the range of
0.13–0.75 dB/cm, and the refractive indices of the films are in the range of
1.623–1.649. The FWHM of the thin film, found to be 44 nm, is comparable to
that of the parent glass, 42 nm. The broadband spectra of the thin films deposited at
10 and 20 Hz repletion rates are shown in Fig. 8.17. The broadband emission
intensity decreases, and the lifetime increases by increasing the substrate temper-
ature from 100 to 200  C. This enhancement in lifetime is due to a decrease of OH

concentration at higher deposition temperatures. The results of the Er3+-doped
phospho-tellurite thin films have significant applications for integrated optic
devices using pulsed laser deposition (PLD) technique [174].

8.4.6 Quenching of Broadband Emission from OH
Groups

The free OH
groups are treated as a luminescence quencher due to their easy
movement and directly coupling with the Ln ion emissions. Primary sources for the
OH
impurities may come from the starting materials and by atmospheric moisture
8 Broadband Emission in Tellurite Glasses 185

Fig. 8.17 Emission spectra

of Er3+-doped parent glass
and the same glass material
of thin films deposited by
PLD technique at 10 and
20 Hz laser repetition rates
(Reprinted with permission
from ref. [173], copyright,
2007, SPIE)

during the melting. The effect of OH
concentration on the emission properties of
Er3+:4I13/2 ! 4I15/2 transition in tellurite glasses has been reported. The emission
intensity and excited state lifetime of the 4I13/2 level for tellurite glasses can be
improved substantially through an oxygen-bubbling process, which decreases the
OH
content with increasing bubbling time, or by doing the synthesis in a
dry-oxygen atmosphere. The absorption coefficient of OH
decreases after an
initial increase with increasing Er3+ ion concentration. Quenching of the emission
and lifetime occurs due to a small energy gap of ~6500 cm
1 of the 4I13/2 ! 4I15/2
transition in Er3+ matches with the energy of the second harmonic of the OH

stretching vibration. If an Er3+ ion is coupled to an OH
ion, non-radiative
relaxation of the 4I13/2 level can occur by exciting two OH
vibrational quanta,
which leads to a decrease in the broadband emission [173, 175].

8.4.7 Broadband Emission in Low OH
Content Tellurite

Glasses

Tellurite glasses with extremely low OH
content have been investigated for their
broadband MIR applications such as lasers, especially fiber lasers at around 3 μm,
that have numerous applications in a wide range of fields including military,
medical, and ranging for atmospheric and chemical sensing and monitoring. MIR
fluoride glasses (e.g., ZBLAN glass) and crystal materials (e.g., CaGa2S4) are
usually chosen as host materials to incorporate Er3+ ions because of their low
phonon energy, which have some disadvantages, such as a complex synthesizing
route, fragility, hygroscopic in nature, and extreme expense [176]. The glass
composition of TeO2-ZnF2-Na2O, identified and synthesized by traditional
melting-quenching method, was optimized to remove physical H2O molecules
and chemical OH
groups to reach extremely a low OH
content with a physical
186 K.V. Krishnaiah et al.

and chemical dehydration (PCDH) technique at high pressure in ultradry O2

atmosphere within a semi-closed steel-chamber furnace, for the first time [177].
The broadband MIR emission at 2.7 μm of an Er3+ ion related to 4I11∕2 ! 4I13∕2
transition, shown in Fig. 8.18, is closely linked to the above applications and
therefore has been intensely investigated [107, 178–180]. The effect of OH

absorption coefficient (αOH
) on the FWHM in different tellurite glasses is
represented in Table 8.5. The best glass is found to be TeO2–ZnF2–NaF with a
large FWHM as broad as 165 nm and αOH
as low as 0.035 cm-1. On the other hand,
the maximum doping concentration of Er3+ in fluorotellurite glass was only 1.25 wt
% (less than 0.5 mol%), due to the crystallization caused by the addition of ZnF2
into the tellurite glass system. Such low doping concentration of Er3+ ions is not
sufficient for efficient lasing at 2.7 μm. Therefore, heavily (maximum 19 mol%) Er
3+
-doped fluorotellurite glasses were prepared by a melt-quenching technique in
corundum crucibles at 900  C for 35 min under ultradry O2 atmosphere and studied
for their broadband emission at 2.7 μm [107]. Higher broadband intensity and
bandwidth (150 nm) at 2.7 μm were obtained when codoped with Er3+/Pr3+ in
low OH
content tellurite glasses with 978 nm wavelength laser excitation [181]
compared to single Er3+-doped tellurite glass.

Fig. 8.18 Broadband

emission of Er3+: TeO2–
ZnF2–NaF glass samples at
2.7 μm by 978 nm
excitation (Reproduced
with permission from ref.
[178], copyright, Opt. Soc.
Am., 2012)

Table 8.5 Comparison of FWHM and absorption coefficient at 2.7 μm for the Er3+:tellurite
glasses
Glass host FWHM at 2.7 μm αOH
1 Ref.
TeO2–ZnF2–NaF 163 0.035 [179]
TeO2–ZnO–Na2O–Li2O – 0.27 [180]
ZBLAN glass 113 – [107]
TeO2–BaF2–Na2O 150 0.0702 [181]
8 Broadband Emission in Tellurite Glasses 187

The MIR emission intensity increases with increasing Er3+ concentration (till
1.25 mol%) in oxyfluorotellurite glasses [178]. On the other hand, the introducing of
alkaline or alkaline earth modifiers into the glass matrix, drying of the fluorine-based
raw materials before melting, and melting of the glass in dry-oxygen atmosphere lead
to a significant reduction of OH
content in tellurite glasses. Employing these steps for
the synthesis of TeO2–Bi2O3–ZnO glass results in a decrease in 93% of OH
content
as it is compared with a glass prepared at ambient atmosphere. The reduction in OH

content is verified by a decrease of the absorption band centered at 3000 cm
1
[182]. Recently, Savelii et al. [183] reported on the modification of OH
absorption
which leads to generation of supercontinuum lasers at this region (around 2.7 μm) in
tellurite glasses. When the fibers are fabricated with suspended core microstructure
and then pumped by femtosecond pulses with nJ pulse energy, the resulting
supercontinuum has a bandwidth of more than 2000 nm (1000–3000 nm range).

8.4.8 Superbroadband Emission in Tellurite Glasses

The superbroadband luminescence devices that include broadband optical ampli-

fiers and tunable lasers, which cover the low-loss telecommunication region
(1.2–1.7 μm) [184, 185], have been developed from the Ln/TM systems. In the
previous sections, broadband emission was explained in single TM- and Ln-doped
luminescent systems. Different possibilities to further enhance the broadband
emission and quantum efficiency of the doped systems have been found in the
literature, such as codoping [186, 187] with Ln and TM ions, section of low phonon
energy systems [188], or nanostructured materials [189].
Broadband NIR emission is at around 1.20 μm from the 1G4 ! 3H4 transition in
Tm3+ ions reported in tellurite glasses [190]. This wavelength region is very near to
the zero dispersion (~1.3 μm) of silica fibers, and optical amplification capacity in
this range has been increased considerably by using the dense wavelength division
multiplexing (DWDM) [69]. There is a wavelength gap around 1.3 μm when doped
with Tm3+ and Tm3+–Er3+ systems, which can be occupied with dopants of Pr3+:
1
G4 ! 3H5, Nd3+: 4F3/2 ! 4I13/2, or Dy3+:(6H9/2,6F11/2) ! 3H5. But all these
transitions have very low efficiency in silicate glasses because of the strongest
non-radiative rate and the higher phonon energy that limit their applications for
optoelectronic devices. This can be overcome by selecting tellurite glasses
fabricated under controlled synthesis conditions [191].
Superbroadband emission from Tm3+–Bi3+ codoped GeO2–Ga2O3–Na2O
glasses has been described in the wavelength range of 1.0–1.7 μm, when excited
with a 793 nm laser. An efficient energy transfer process (67.7%) takes place from
Bi3+ to Tm3+ ions, which is beneficial for producing a flat broadband profile. The
larger emission cross-section and lifetime confirm that the potential of codopants,
Tm/Bi, for the generation of superbroadband optical amplifiers and tunable lasers
[192]. In addition, broadband emission covering a wavelength range from 1.30 to
1.67 μm was observed in Pr3+-doped fluorotellurite glasses, pumped at 488 and
188 K.V. Krishnaiah et al.

590 nm wavelengths, due to the radiative transition from the 1D2 ! 1G4 (Pr3+). The
emission profiles, bandwidth, and excited state lifetime were tuned by the Pr3+
concentration. The quantum efficiency obtained was as high as 73.7% for
0.05 mol% Pr3+ concentration [193]. On the other hand, upon excitation with a
793 nm wavelength laser, broadband emission from 1.3 to 1.7 μm was observed in
Nd3+/Tm3+/Er3+-doped fluorotellurite glasses. Hence, all the studies indicate that
the emission bandwidth is better than the TM ion-doped glasses [194].

8.5 Broadband Absorption for Solar Cells

Solar energy harvesting is one of the applications where a broad bandwidth becomes
relevant. Approximately 90% of the present-day solar cells are produced from
crystalline silicon (c-Si) because of its moderate price and high efficiency when
compared with organic or amorphous materials. This high efficiency originates from
its broadband absorption in comparison to other materials with narrower bandwidth,
such as thin films of CdTe and CuInGaSe2. The most advantageous property of
silicon for photovoltaics is a band gap energy (1.12 eV) which is close to the optimal
band gap that better matches with the solar spectrum (1.2 eV). The Shockley–
Queisser theory thus restricts the efficiency of silicon solar cells to about 30%
[195]. There are two main losses that limit the solar cell efficiency: first, the
low-energy photons than the semiconductor bandgap energy are not absorbed.
Second, the higher-energy photons than the bandgap produce a single electron-hole
pair, and the excess of energy is lost by thermalization losses, which in turn heats the
solar cell and decreases its performance. In an attempt to solve this problem,
multijunction solar cells were developed with different bandgaps and materials.
However, the higher price, difficulty in fabrication, and other problems such as the
current matching between the layers have restricted these cells to specialist applica-
tions [196]. A possibility to increase the region of absorption of a single junction solar
cell is to use energy conversion methods to make use of unutilized photons into useful
ones by all-optical processes. These processes are upconversion, downconversion,
and luminescent down-shifting. Broadband absorption is a crucial for the success of
these optical processes, as well be shown in the following sections.
Alkali and mixed-alkali effect on the broadband absorption, emission profiles,
and bandwidths in Er3+-doped tellurite glasses (TeO2–ZnO–Na2O/K2O/Li2O–NaF)
has been reported [197]. The effective absorption bandwidth for the 4I15/2 ! 4I13/2
and 4I15/2 ! 4I11/2 transitions is calculated to be the integrated absorption coeffi-
cient over wavelength and divided by the peak intensity. This effective absorption
bandwidth as a function of alkali ions concentration is shown in Fig. 8.19. As can be
seen from Fig. 8.19, the highest effective bandwidth is obtained for glass samples
with an alkali ion ratio of R2O/M2O ¼ 1 for all three series of glasses (R and M
stand for alkali ions). The effective bandwidth of the alkali glasses is 10%, higher
for glasses with mixed-alkali ions when compared with single alkali glasses.
However, the changes in effective absorption bandwidth may be due to
8 Broadband Emission in Tellurite Glasses 189

Effective absorption bandwidth (nm)

Fig. 8.19 Variation of
effective absorption 74
bandwidth as a function of
alkali ion concentration in 72
all three series of TeO2– 70
ZnO–NaF glass modified
with triangle: xNa2O/(4.5- 68
x)Li2O; filled square:
xNa2O/(4.5-x)K2O; 66
diamond: xK2O/(4.5-x)Li2O 64
ratios (Reproduced with
permission from ref. [197], 62
copyright, 1999, SPIE) 0 1 2 3 4
X

modification by the local field effect of the Er3+ ion. In general, the bandwidth
broadening of Er3+ ions in glasses is inhomogeneous broadening that is caused by
the differences in the LF around Er3+ ions. Hence, the difference in LF around Er3+
ions is predominant in mixed-alkali-modified glasses when compared to that of
single alkali-modified glasses, which may be due to an increase of cation field
strength (in the case of mixed-alkali effect) which was discussed in Sect. 4.3.

8.5.1 Upconversion

UC is a process in which two or more low-energy photons produce one high-energy

photon [198]. The Ln ions are ideal emitters in such a conversion processes due to
their ladderlike electronic energies [199]. In 2002 Trupke and coworkers at UNSW
calculated that the use of an upconverter placed behind a c-Si solar cell could boost
the efficiency limit from 30 to 37.4% under one-sun illumination [200]. Two main
Ln systems have been studied for this purpose. The first one is the Er3+/Yb3+ ion
pair, where the Yb3+ provides a high cross section for absorption at 980 nm and the
Er3+ provides the emissive levels at red, green, and blue wavelengths. The first
experiment of upconversion for photovoltaics was pathway in 1996 using this
couple of ions. Gibart et al. [201] showed the response of a GaAs solar cell under
980 nm illumination by using a upconverting Er3+/Yb3+-doped vitroceramics.
Later, other solar cell technologies with higher bandgap were tested with this
system (it should be noted that c-Si cannot benefit from this system because it
already absorbs up to 1100 nm). In 2010, Wild et al. [202] applied β-NaYF4:Yb3+/
Er3+ to a-Si solar cells under 980 nm illumination. The second most studied system
for UC photovoltaics relies solely in the Er3+ energy diagram, which allows
absorption at 1470–1580 nm (4I15/2 ! 4I13/2) with emission at 980 nm (4I11/2 !
4
I15/2). In 2005 Shalav et al. [203] reported the response of a c-Si solar cell under
1520 nm illumination, with an external quantum efficiency of 2.5% at 1520 nm by
using NaYF4:Yb3+/Er3+, which has shown to be an efficient upconverter [204].
190 K.V. Krishnaiah et al.

The broadband absorption of the Ln ion is a crucial point for the applicability of
this technology. Since the upconversion efficiency depends nonlinearly on the
incident power, and it increases exponentially with the irradiance [205]. Hence,
there is a possibility to gather a larger fraction of the solar spectrum resulting in an
increase in the useful irradiance, leading to a higher upconversion rate [206]. The
benefit is clear when considering that the early experiments were run under narrow-
bandwidth 1523 nm laser excitation, while the absorption of Er3+ spans from 1470
to 1480 nm, approximately.
Figure 8.20 shows the spectral irradiance of the sun with a maximal fraction of
its energy that can be used for electron-hole production by a c-Si solar cell. In red is
shown the absorption of NaYF4:Er3+ due to the 4I15/2 ! 4I13/2 transition, and in blue
is its 980 nm emission due to the 4I11/2 ! 4I15/2 transition. It can be seen that a large
portion of the spectrum remains to be absorbed. Different strategies have been
proposed to broaden the absorption bandwidth by upconversion. The Er3+/Dy3+-
codoped systems have been studied for 1300 nm absorption [208] and Tm3+-doped
glass-ceramics for 1220 nm absorption and 808 nm emission [209]. The use of
organic antennas has been recently proposed [210], although its use is limited for
sensitization of nanocrystals, which are by nature less efficient than bulk phosphors
[211]. A different strategy has been proposed based on the use of NIR quantum dots
to harvest the photons between 1100 and 1350 nm and use their downshifted
luminescence to feed the Er3+-based upconverter (with absorbance at
1470–1580 nm) [212, 213], as shown by the blue arrows of Fig. 8.20. It has been

Fig. 8.20 Solar irradiance and the fraction that can be utilized for electron-hole production in a
c-Si cell. The red area represents absorption of Er3+ in NaYF4:Er3+ due to the 4I15/2 ! 4I13/2
transition, while the blue one represents its 980 nm emission due to the 4I11/2 ! 4I15/2 transition
(Reprinted with consent from ref. [207], copy right, 2015, Wiley publications)
8 Broadband Emission in Tellurite Glasses 191

shown recently that the use of photonic crystals can help to modify the emission
from quantum dots and thus to overcome some challenges like its small Stokes shift
or the mismatch between the quantum dots emission and the Er3+ absorption
[207]. This would result in a possible utilization of all the energy in the wavelength
range of 1100–1580 nm.
Novel Ln-doped materials with NIR broadband absorption could make an
important contribution to this field. The upconverter material should fulfill a few
specific properties, such as absorption of wavelengths longer than 1100 nm
(Eg < 1.12 eV), response at low excitation intensities (in the range of W/cm2),
and high UC efficiency and high transmittance of the up-converted light, that are
essential for the application in silicon solar cells.
Recently, an enhancement in the photocurrent of Si solar cell through UC has
been investigated in Er3+/Yb3+-codoped oxyhalide tellurite glasses [6] and
fluoroindate [214] glasses, excitation at 980 and 1480 nm wavelengths, respec-
tively. When this UC tellurite glass is coupled to a-Si solar cells, photocurrent
enhances significantly under 980 nm wavelength laser excitation due to a high
recombination rate. Exciting with both AM1.5G (Air Mass 1.5, G represents global
and includes both direct and diffuse radiation) and 400 mW of 980 nm wavelength
laser results in 0.45% improvement in efficiency of the solar cell [6]. A photograph
of the solar cell with the upconverter glass is shown in Fig. 8.21. The influence of
active layers on the enhancement of conversion efficiency of broadband NIR
radiation has also been reported by utilizing UC in Yb3+/Er3+:tellurite glass.
Under 976 nm wavelength laser illumination, the open circuit voltage increased
by 5.5% for codoped Yb3+/Er3+: sample, compared to 2.7% for the single-doped
Er3+ sample [215].

Fig. 8.21 A digital photograph of the Er3+/Yb3+-codoped UC glass with the Si solar cell.
Excitation was performed from the side of the glass with a 980 nm laser. The incident light is
perpendicular to the solar cell and the glass (Reproduced with the permission from ref. [6], copy
right, 2014, Elsevier)
192 K.V. Krishnaiah et al.

8.5.2 Downconversion/Quantum Cutting

The thermalization of absorbed high-energy photons (300–500 nm) is one of the

major mechanisms for energy loss in silicon-based solar cells. Photons with high
energy only produce a single electron-hole pair, while the excess of energy of the
electron is delivered to the semiconductor in the form of heat, which in turn
increases the temperature and reduces the efficiency of the solar cell.
One of the promising solutions to this problem is to develop the materials with
high/efficient quantum cutting (QC) or downconversion of one high-energy photon
into two or more low-energy photons. This process was theoretically proposed for
the first time in 1957, when quantum yields higher than unity were proposed
[216]. Later it was experimentally observed in the Pr3+ ion by two different groups
[217, 218]. QC can occur by different mechanisms. The firstly observation was
based on the energy levels of a single ion. In this case all the energy leaps happen in
the same ion, which requires a middle energetic level in order to radiatively relax
twice: first from the excited state to an intermediate level, with the emission of one
photon, and then to the ground state by emitting a second photon. Another mech-
anism is based on the existence of two different ions to provide the energy leaps or
even using energetic levels from the lattice [219]. When the QC is achieved by
using ions of two different spices, different routes can be taken. The first ion, called
the sensitizer, will always absorb the high-energy photon. From here, it can emit the
first photon and excite the second ion, called the activator, by energy transfer. The
activator will then produce a second photon. Another possibility is that the sensi-
tizer cedes part of its initial energy to the activator by cross-relaxation and both emit
one photon each. Obviously this process can be produced by two identical ions. In
this way, the process is realized by photon cascade emission from a single and
codoped Ln systems through the mechanisms of energy transfer (through cooper-
ative DC) [220, 221] and multistep photon emissions [222, 223].
QC process can be very efficient. In 1999, this process with 200% quantum yield
(two emitted photons per each absorbed photon) was reported by using a pair of Ln
ions Gd3+ and Eu3+ [224]. Since then, it has been investigated intensively in
different materials. It has been postulated that the combination of a silicon solar
cell and an overlaying DC layer could reach an overall limit in efficiency of 38.6%
[225]. Hence, broadband QC has attracted much for its potential application in
relieving the thermalization of the charge carriers in c-Si solar cells [225].
A large list of downconverting combinations enables many possibilities for PVs
[226, 227]: compounds codoped with Ce3+/Yb3+ for absorption at 300–400 nm,
Pr3+/Yb3+ for the 430–500 nm range, Tm3+/Yb3+ for 455–480 nm absorption,
and Tb3+/Yb3+ for 475–495 nm. It can be observed that Yb3+ is often selected
for its high intense emission at ~1 μm which falls well in the highest spectral
response of the c-Si solar cells. This can be observed in Fig. 8.22, which shows
the solar spectrum (AM1.5G, background with blue), the spectral response of
Si solar cells (line a), the excitation spectrum of Yb3+ ion by monitoring at
980 nm (line b), and the broadband emission obtained at 405 nm wavelength
8 Broadband Emission in Tellurite Glasses 193

Fig. 8.22 Blue background

is the terrestrial solar
spectrum (AM1.5G) in the
visible and near-infra-red
regions. Blue-shaded area
is the maximum available
portion for the DC process.
The spectral response of the
solar cell (a), excitation
spectrum monitored at
980 nm (b), and emission
spectrum of Yb3+ ion under
405 nm wavelength laser
excitation (c) (Reproduced
with permission from the
ref. [228], Copy right, 2013,
Elsevier)

excitation (line c). Other factors have to be taken into account when engineer-
ing the QC layer, like the thickness of the QC material, which greatly influ-
ences the performance of the solar cell. The thickness of the layer has to be
optimized to minimize the impact of self-absorption [229].
QC emission has been observed in Pr3+/Yb3+ codoped TeO2–ZnO–Na2O glasses
via a two-step photon energy transfer process. With a 499 nm wavelength excita-
tion, a broadband emission at 1001 nm was observed attributed to the 2F2/5 ! 2F2/7
transition of Yb3+ ions. The population of Yb3+ ions is due to the two-step energy
transfer from Pr3+ through 1G4 multiplet by emission and excitation spectra of
respective ions. The quantum efficiency was found to be 136.7% for the 5 mol% Yb
3
-doped tellurite glasses [230], and it can find application for enhancing the
conversion efficiency of c-Si-based solar cells.

8.6 Bi-Doped Fiber Amplifiers

Ln-doped fiber lasers with emitting wavelengths in the NIR region are widely used
for applications in optical communication, material processing, and medicine,
which have been discussed in the previous section. However, within a spectral
region of 1150–1500 nm, no efficient Ln-doped fiber lasers exist. Moreover, there is
a great demand for new active materials that are suitable to generate fiber lasers and
optical amplifiers in this spectral region. TM-doped materials to develop efficient
devices have been unsuccessful [231]. In 2001, Fujimoto described the broadband
near-IR luminescence in Bi-doped Al2O3-SiO2 glass [48, 51] The luminescence
band is centered at 1250 nm with a bandwidth of 300 nm, which is prime interest as
it can be used for developing tunable lasers and broadband optical amplifiers in the
spectral region described above.
194 K.V. Krishnaiah et al.

The origin of the infrared emission from Bi-doped glasses remains unclear and
has been attributed to Bi+ [232], Bi5+ [51, 233], Bi metal clusters [234], and also
negatively charged Bi2
dimers [235]. Bismuth is a polyvalent element with
oxidation states: Bi5+, Bi3+, Bi2+, Bi2
, and Bi+. Two processes take place while
molten Bi-doped glass: one is oxidation (at a higher valence state) and the other one
is reduction (at a lower valence state) [236]. Detailed explanations of the valance
state of Bi and their applications in lasers and amplifiers have been reported in refs.
[233, 237]. Bi-doped GeO2–Al2O3–Na2O glasses were reported for the broadband
luminescence at 1220 nm in the wavelength range of 1000–1600 nm, observed
when excited with 800 nm wavelength. The broadband emission intensity increased
with the addition of Bi5+ in NaBiO3 glass matrix by 5.6-fold, and also the FWHM
changed from 195 to 275 nm, when compared with the Bi3+ in Bi2O3 glass
matrix [233].
Femtosecond (fs) laser irradiation was performed to investigate the valence state
of the Bi active centers that lead to broadband emission in bismuth-containing
glasses. It is observed that a local absorption in the visible spectral region can be
modified in a transparent bismuthate glasses after interaction with fs laser. This
leads to change in absorption centers which exhibit the well-known broadband
emission at ~1250 nm upon excitation with 785 nm wavelength. The absorption and
emission intensities change drastically with increasing pulse energy of an incident
laser. These changes may be due to photoinduced reduction reaction Bi3+ ! Bi+,
while it was previously discussed that the formation of Bi-clusters by the diffusion
of ions was previously discussed, due to a very short period of time interaction of
laser with the material. Therefore, the Bi+ centers are proposed as the major origin
of broadband emission in Bi-doped glasses [238].

8.7 Raman Amplifiers

Raman amplifiers have gained much attention because of their significant advan-
tages which include a broad amplification bandwidth and tunability of Stokes
scattering by simply changing the pump wavelength. Gain flattening by pumping
with tunable wavelengths is also possible for WDM systems, which has an advan-
tage over traditional EDFAs. Raman amplification in silica-based glasses was
demonstrated in optical waveguides in 1972 by Stolen et al. [239]. Raman ampli-
fiers are better than EDFAs, as the amplification occurs at discrete locations of the
EDFAs which depend on the Er3+ ion concentration, whereas the Raman amplifi-
cation arises entire length of the fiber. Furthermore, the amplification in EDFAs is
limited to discrete wavelengths (depends on the energy levels of Er3+ ion), whereas
Raman amplification does not depend on pump wavelength; therefore the gain
bandwidth should not be affected.
Raman amplification has been utilized for enhancing the optical transmission
performance as it improves the optical signal-to-noise ratio (OSNR)
[240, 241]. Raman amplifiers exhibit a large OSNR at longer wavelengths than at
8 Broadband Emission in Tellurite Glasses 195

shorter wavelengths, in the case of backward pumping [242]. This limitation can be
overcome by using an additional forward-pump source. The co-propagating Raman
pumping scheme offers a low noise figure, and low interference noise which is
induced by signal double-Rayleigh scattering when combined with counter-
propagating pumping [243].
Distributed fiber Raman amplifiers (DFRAs) were extensively investigated for
optimizing a flat gain and noise performance in a long-haul WDM system. It is also
important to note that the noise performance of the Raman amplification can be
enhanced significantly by EDFA post-amplification. In addition, the OSNR of
Raman amplifier is boosted if a signal power is minimal along the length of the
fiber for a fixed nonlinearity [244]. The performance of the DFRAs is enhanced
enormously by pumping with incoherent sources rather than the coherent
sources [245].
Optical amplification is essential whenever a signal is attenuated. Raman scat-
tering has become the preferred approach for optical amplification as it is only
restricted by the pump wavelength. However, Raman active modes of the gain
medium may not be very strong, limiting amplification. To enhance the gain
bandwidth of tellurite glasses and to make the spectra more continuous, glass
systems have been developed by adding WO3 and P2O5 in a TeO2-BaO-SrO-
Nb2O5. The Raman scattering experiments indicate that these glasses have higher
Raman gain coefficients and bandwidths than conventional tellurite and the silica
glasses. The Raman gain coefficients of the glasses were obtained from spontane-
ous Raman scattering measurements by excitation at 633 nm wavelength laser. The
results showed the widest gain bandwidths while maintaining higher gain coeffi-
cients, compared with other reported tellurite glasses [246].
The effect of modifiers, WO3, MoO3 and P2O5, on the Raman gain coefficient
and bandwidth in tellurite TeO2–ZnO–Na2O–Nb2O5 glasses has been investigated
[247]. The gain coefficient is found to be 1.7 times higher and 1.9 times higher for
MoO2 and P2O5 modified tellurite glasses, respectively, whereas the Raman band-
width is 37 and 39 times for the respective glasses when compared to silicate glass.
This indicates that the P2O5 modified tellurite glasses are promising candidates as a
gain media for broadband Raman fiber amplifier. Raman gain coefficient and
amplification bandwidth was found to be 40 and 1.7 times higher than in silicate
glass, respectively. The TeO2-ZnO-Nb2O5-MoO3 glass is a potential material for
photonic devices, particularly, fiber Raman amplifiers [248]. Raman gain band-
width is found to be higher for phospho-tellurite glasses and thin films compared to
other tellurite glasses [249].

8.8 Supercontinuum Generation

The MIR supercontinuum (SC) light sources have also gained attention for monitor-
ing plasma-etching processes, sensing of trace gases, molecular fingerprinting, laser
surgery, IR counters, and biomedical diagnosis applications. SC can be generated by
196 K.V. Krishnaiah et al.

Fig. 8.23 SC spectrum of

undoped (the solid curve)
and Tm3+-doped (the
dashed curve) TMFs under
excitation with amplified
laser at 274 mW pump
power and pulse width of
290 fs (Reprinted with
permission from ref. [251],
copyright, 2013, SPIE)

pumping a nonlinear fiber with high intense mode-locked fiber lasers leading to
broadband emission with an extended bandwidth many times wider than the spectrum
of the pumped pulses. The very first report of SC generation was in 1966, in which
broadening of the bandwidth was observed when Q-switched ruby laser pulses passed
through CS2 liquid [250]. Very recently, SC generation was reported in a 20 cm long
Tm3+-doped tellurite microstructured fiber (TMFs) when excited at 1.56 μm wave-
length of femtosecond fiber laser with 274 mW power and a pulse width of 290 fs, as
shown in Fig. 8.23. SC generation spanning from 1.35 to 2.34 μm was obtained in
undoped and TMFs. The output power of SC light was 48 mW with conversion
efficiency of 44% by considering the coupling efficiency (~40%) of the pump light
[252]. Another work reported on SC generation at 1–5 μm with W-type tellurite fiber
pumped with 3 W power (20 ps pulses and 30 MHz repetition rate), achieving a
conversion efficiency as high as 60% when pumped with 2 W power.

8.9 White LED’s

White LEDs (w-LEDs) have found tremendous applications in automobile headlights

and general illumination as their emission is homogeneous and environmental
friendly, much better than incandescent and fluorescent lamps [20, 253]. Usually,
w-LEDs are fabricated by the combination of yellow luminescence phosphors on the
top of the blue light-emitting diode (blue-LED) chip [19]. However, it could cause a
chromatic aberration and poor white luminescence. In addition, they have high
luminescence efficiency, longer lifetime, reliability, safety, and environmental
friendly characteristics [254]. In recent days, Ln-doped materials like glasses and
glass-ceramics are reported to be better alternatives for the phosphors, since they can
be mold in any shape for a direct utilization of the UV chip [255]. The other
advantages such as homogenous light emission, and easy industrial production and
8 Broadband Emission in Tellurite Glasses 197

they are potential candidates for replacing conventional incandescent and fluorescent
lamps [256]. Among several host matricies, oxyfluoride glasses and glass-ceramics
are excellent candidates for applications in optoelectronic owing to their advanta-
geous properties including chemical, thermal, and mechanical stability besides their
low phonon energy [257].
White light emission can be obtained by downshifting [258, 259] and
upconversion [260, 261] processes in Ln-doped tellurite glasses. In down-shifting,
the Ce3+- and Dy3+-doped glasses have been investigated for the generation of
white light emission which can be tuned by varying the Ln ion concentration
[259, 262, 263], glass composition [264], and pump wavelength [18, 265]. In
upconversion, Er3+∕Tm3+∕Yb3+-codoped [260], Er3+∕Pr3+∕Yb3+-codoped [261]
tellurite glasses, and Ho3+∕Tm3+∕Yb3+-doped [266] tellurite glasses have been
investigated for the generation of white light emission upon excitation at 980 nm
laser by varying the Ln ion concentration and the pump power. Multicolor lumi-
nescence comprises of primary colors green, red, and blue upconversion emissions:
luminescence color can be tuned from multicolor to broadband white color by
changing the pump power of the excitation source. The UC process is only possible
in tellurite glasses due to their low phonon energy among the other traditional oxide
glasses.

8.10 Laser Induced Cooling of Solids

It is well known that an intense laser beam can be used to cut, damage, and probe or
ablate materials. Perhaps, a lesser known fact is the utilization of laser with a
suitable wavelength to cool a low phonon transparent dielectric suitably doped with
a Ln-doped material [267] or a semiconductor [268]. This field of research is termed
as laser cooling of solids or optical refrigeration or laser refrigeration or anti-Stokes
fluorescence cooling, for the development of optical refrigerators/cryocoolers [269]
and radiation-balanced lasers [270]. These devices have been treated as a better to
mechanical refrigerators due to their light weight, absence of vibrations, cooling to
cryogenic temperatures, and long lifetime.
In laser cooling an anti-Stokes photon is emitted after absorption of a pump photon
and a phonon. This phenomenon is explained by taking a simple energy level diagram
of Yb3+ ion and is shown in Fig. 8.24. When a low-energy laser photon is incident on
the Yb3+ system, electrons get excited to the lowest energy level of the upper
manifolds. The excited electrons reach equilibrium by absorbing phonons from the
host lattice. Hence, electrons populate to higher-energy states with the help of
phonons within the manifold and decay radiatively to the ground state by emitting
a photon spontaneously at the mean fluorescence wavelength (λf) shorter than the
absorbed pump photon. The emitted photon energy is higher than the pump energy
(λp). Finally, the electron is again scattered to a higher level thermodynamically by
absorbing a phonon of kBT energy within the manifold. If the removal process of
these phonons from the system in a repeated cycle is successful, and competing
198 K.V. Krishnaiah et al.

Fig. 8.24 Mechanism of

laser cooling in Yb3+-doped
systems

non-radiative effects are absents cooling of the sample occurs. This process was first
proposed by Pringsheim in 1929 [271] and demonstrated experimentally by Epstein
in 1995 [268]. It is not as possible to cool all solids due to very stringent requirements.
There are few properties that should be exhibited by the cooled material which
include a low maximum phonon energy, low background absorption, and high
PLQY close to unity. There are only a few broadband emitting ions, i.e., Yb3+, Er3+,
and Tm3+, which as dopants in certain host materials (glasses and crystals) are
exhibiting these properties, in which cooling has already been demonstrated
[268–299]. These are summarized in Table 8.6. Galina et al. [272] predicted that the
glass-ceramics containing nanocrystals may cool less than 100 K if most of the Ln ions
are located within the low phonon energy crystalline phase and also the Ln ions should
exhibit PLQY near to unity. Very recently, from the same group, Saores et al. [273]
reported that Yb3+-doped glass-ceramics containing cubic nanocrystals of PbF2
(phonon energy, ~250 cm
1) exhibit the properties of low phonon energy, higher
PLQY, and lower background absorption than their parent glasses which were prepared
under ambient atmosphere without further purification of raw materials. The glass-
ceramics have been proposed to the laser cooling community, as they offer a better
alternative to the single crystals, which have the advantages of scaling in a short span of
time; Ln ions are located within the low phonon energy nanocrystals, have low
production cost, and are easy to dope with larger Ln ion concentration
[274, 275]. Glass-ceramics may be cooled to a temperature of
13.2 K from a sample
chamber temperature if their PLQY improves to 0.03 and background absorption
reduced by a factor of 50 [275].
Laser cooling process based on anti-Stokes fluorescence in Ln-doped systems is
illustrated in Fig. 8.24. The cooling efficiency of a solid sample can be described as

Pcool λp
ηcool ¼ ¼ ηabs ηext
1, ð8:13Þ
Pabs λf
αr
ηabs ¼ ð8:14Þ
ðαr þ αb Þ
Table 8.6 Comparison of cooled materials with dopant ions, concentrations, phonon energy, PLQY, background absorption (αb), mean fluorescence
wavelength (λf), minimum achievable temperature (MAT) measured using different techniques, thermocouple (TC), photothermal deflection (PTD), infrared
(IR) camera, differential luminescence thermometry (DLT), optical calorimetry (OC) by using FBG interrogation, and laser-induced temperature modulation
spectrum (LITMoS)
Phonon
Dopant ion, energy (cm
1 αb, (10 λf MAT from RT, K

4
concentration System ) PLQY ) m
1 (nm) λp (nm) (temperature drop) Method
Yb3+, 1 wt% ZrF4–BaF2–LaF3–AlF3–NaF–PbF2 506 0.99 – 995 1010 289.7 (0.3 K drop) PTD, IR
[267]
Yb3+, 1 mol% BaF2–InF3–GaF3–ZnF2–LuF3–GdF3 ~450 0.996 0.1 1005 1010 – PTD
[284–286] and
1.016
Yb3+, 05  10 AlF3–BaF2–CaF2–YF3–SrF2 [287] ~500 – – – – 292.7 (drop of 0.13) IR
20
ions/cm3
8 Broadband Emission in Tellurite Glasses

Yb3+, 1 mol% CdF2–CdCl2–NaF–BaF2–BaCl2– ~450 0.996 0.1 – 1010 – PTD

ZnF2 [285, 286] and
1.016
Yb3+, 3 at% Y3Al5O12 [288] 630 0.991 0.029 1010 1029 285 (8) OC
1026 208 (FBG)
Yb3+, 2 mol% SiO2–Al2O3–PbF2–CdF2–YF3– ~930 glass, 0.97 0.79 1000 1025 – OC
YbF3, Oxyfluoride GCs [273] 250 crystal (FBG)
Yb3+, 10 at% YLiF4 [280] ~300 0.996 0.2 1009 1020 114 (189) LITMoS
DLT
Yb3+, 3.5 at% KGdWO5 [289] ~400 – – 992 1020 – PTD
Yb3+, 2.5at% BaY2F8 [290] 360 – – 995 1025 289 (4) IR
Yb3+, 5  1019 KPb2Cl5 [291] 200 1.0 and – 986 1005 150 PTD
ions/cm3 1.002
Yb3+, 2 wt% LuLiF4 [292] 460 0.995 0.42 981.5 1020 145 (148) DLT
Tm3+, TeO2–Bi2O3–PbO fiber [293] 700 – – – 1949 264.15 (30  C)
Tm3+, 1 wt% ZrF4–BaF2–LaF3–AlF3–NaF [294] 506 0.2 0.99 1820 1825–1970 291.8 (1.2) IR
199

(continued)
Table 8.6 (continued)
200

Phonon
Dopant ion, energy (cm
1 αb, (10 λf MAT from RT, K

4
concentration System ) PLQY ) m
1 (nm) λp (nm) (temperature drop) Method
Tm3+, 1.2 at% BaY2F8 [295] 360 0.98 2 1793 1855 290 (3  C) IR
Er3+, 0.5 mol% CdF2–CdCl2–NaF–BaF2–BaCl2– 450 0.997 – 852.5 860 292.5 (0.5) IR
ZnF2 [296, 297] and
1.007
Er3+, 0.5 mol% KPb2Cl5 [285, 286, 298] 200 0.997 – 852.5 870 292.3 (0.7  C) PTR, IR
and
1.007
Er3+, 3.6  10 KPb2Cl5 [299] 200 – – 1540 1568 – TC
19
ions/cm3
K.V. Krishnaiah et al.
8 Broadband Emission in Tellurite Glasses 201

is the absorption efficiency, which includes the resonant absorption, αr, and
background absorption, αb. The external PLQY of the system by including all the
non-radiative losses can be expressed as

ηe W r
ηext ¼ ð8:15Þ
ðηe W þ W nr Þ

where ηe is the fluorescence escape efficiency [276]. This also depends on the
refractive index of the sample. Wr is the radiative decay rate and Wnr is
non-radiative decay rate. The fluorescence escape efficiency depends not only on
the refractive index but also on the shape of the sample. As can be seen in
(Eq. 8.13), the Ln hosts alone can satisfy the unequality Wnr  Wr and are suitable
for laser cooling by anti-Stokes fluorescence. The mean fluorescence wavelength,
λF, can be calculated as
R
λI f ðλÞdλ
λf ¼ R ð8:16Þ
I f ðλÞdλ

where If(λ) is the emission spectrum.

Purification and synthesis of Ln-doped novel cooled materials have been
described in ref. [277]. It is a good choice to select Yb3+ ion for achieving laser
cooling because of their simple energy level structure and other advantages,
explained in Sect. 2.1. Yb3+-doped heavy metal fluoride glasses are cooled to less
than 190 K when excited with 1020 nm laser [278]. The cooling efficiency is
enhanced by selecting broadband emitting ions at longer wavelengths [279]. The
background absorption decreases, while maintaining the same PLQY for all the
concentrations, with increasing Yb3+ concentration [280]. This linear behavior
occurs for only some extent of Yb3+ concentration and may be changed to a
nonlinear behavior at higher concentration due to concentration quenching, which
depends on the impurity level of the material [281]. On the other hand, the mean
fluorescence wavelength increases with increasing Ln ion concentration [277], and
it exhibits a redshift with decreasing temperature [278]. The former one is due to
reabsorption effects and the later one is due to Boltzmann distribution. From
Table 8.6, today’s minimum achievable temperature (MAT) is less than 114 K in
Yb:YLF single crystal which has of phonon energy as low as 300 cm
1, background
absorption as low as 0.2  10
4 m
1, and PLQY as high as 0.997 close to unity
[281]. Hence, the best sample is Ln:YLF single crystal for achieving cooling at
cryogenic, liquid nitrogen temperatures.
Heavy metal oxide (Bi2O3, PbO)-modified tellurite glass offers a high transmis-
sion window (0.4–6.5 μm) and low phonon energy (~700 cm
1). Cooling can be
achieved in heavy metal oxide modified tellurite glass fiber by purifying the starting
materials and by eliminating the transition metal impurities, such as Fe2+ and Cu2+.
These impurities have been reduced from a value of >8 to 0.6 ppm, and from >2 to
0.2 ppm, respectively, and OH
impurities have been reduced significantly
202 K.V. Krishnaiah et al.

[282]. The Tm3+-doped tellurite fiber (from NP Photonics) drops a temperature of

30  C from room temperature when the fiber is pumped with a power of 2.5 and
1.2 W for 1949 and 1943 nm Tm3+-doped fiber lasers, respectively. The cooling
power density of the fiber is found to be 5 W/cm3 with an input power of 2 W.
Moreover, the fiber-based cooling offers the engineers to transfer heat to a remote
location by utilizing a fiber and also separates sensitive instrumentation from heat
and vibration sources [283].

8.11 Summary

The importance of broadband emission of lanthanide- and transition metal-doped

glasses, particularly tellurite glasses, has been outlined. From the Judd–Ofelt
intensity parameters, one can estimate the environment around the lanthanide ion
in Er3+-doped tellurite glasses. The radiative properties such as excited state
lifetimes, quantum efficiency, stimulated emission cross section, and figure of
merit for the bandwidth and gain are compared with those of the tellurite glasses,
reported in the literature. The full width at half maximum, excited state lifetime,
and intensity of the broadband emission for the 4I13/2 ! 4I15/2 transition of Er3+ ion
are tuned with the lanthanide ion concentration, heat treatment temperature, OH

ion concentration, modifiers, and glass formers. Extended bandwidth has been
achieved by codoping of lanthanide and transition metal ions in tellurite glasses.
The utilization of up- and downconversion emitting Er3+- and Yb3+-codoped
tellurite glasses for the enhancement of c-Si cell (Eg ~ 1.12 eV) efficiency has
been discussed. Special attention was given to exploring the broadband emitting Yb
3+
, Er3+, and Tm3+ ion-doped low phonon energy hosts (fluoride glasses and single
crystals), including tellurite glasses, for laser cooling of solids. Overall the
Ln-doped tellurite glasses have many advantages over traditional silica-based
fiber amplifiers, because of their extended bandwidth, long excited state lifetime,
high transmission range and larger stimulated emission cross section compared to
traditional silicate and phosphate glasses.
Broadband emission has been investigated extensively in transition metal ions,
Cr3+-, Cr4+-, Ni3+-, and Bi3+-doped glasses because their emission covers the entire
telecommunication window with extended bandwidth. However, their applications
are limited because of the short excited state lifetime (in the order of μs). Otherwise,
these are the best alternative for the lanthanide-doped traditional fiber amplifiers,
and they may turn out to be the choice for future fiber amplifiers. Overall, the
Bi-doped glasses have the highest bandwidth around ~300 nm when compared with
those of other lanthanide- and transition metal-doped glasses.

Acknowledgment Dr. K.V. Krishnaiah is highly grateful to the Department of Engineering

Physics, Polytechnique Montreal, for providing the necessary facilities to investigate various
glasses and nano-glass-ceramics.
8 Broadband Emission in Tellurite Glasses 203

References

1. S. Huefner, Optical spectra of transparent rare earth compounds (Academic, New York,
1978), p. 247
2. B.N. Figgis, Comprehensive coordination chemistry, Ch. 6 (1987) p. 214–276.
3. K. Venkata Krishnaiah, G. Venkataiah, J. Marques-Hueso, P. Dharmaiah, C.K. Jayasankar,
B.S. Richards, Sci. Adv. Mater. 7, 345–353 (2015)
4. H. Liu, D.D. Chen, Q.Y. Zhang, Z.H. Jiang, Opt. Mater. 28, 302 (2006)
5. K. Sardar, C.H. Coeckelenbergh, R.M. Yow, J.B. Gruber, T.H. Allik, J. Appl. Phys. 98,
033535 (2005)
6. F. Yang, C. Liu, D. Wei, Y. Chen, J. Lu, Shi-e Yang. Opt. Mater. 36, 1040 (2014)
7. V. Petrov, U. Griebner, D. Ehrt, W. Seeber, Opt. Lett. 22, 408 (1997)
8. C. H€onninger, R. Paschtta, M. Graf, F. Morier-Genoud, G. Zhang, M. Moser, S. Biswal,
J. Nees, A. Braun, G.A. Mourou, I. Johannsen, A. Giesen, W. Seeber, U. Keller, Appl. Phys.
B 69, 3 (1999)
9. S. Ohara, Y. Kuroiwa, Opt. Express 17, 14104 (2009)
10. R.I. Epstein, M.I. Buchwald, B.C. Edwards, T.R. Gosnell, C.E. Mungan, Nature (London)
377, 500 (1995)
11. E.S. de Lima Filho, K. Venkata, Y. Ledemi, Y.J. Yu, Y. Messaddeq, G. Nemova, R. Kashyap,
Opt. Express 23, 4630–4640 (2015)
12. C.B. Schaffer, A. Broadeur, J.F. Garcia, E. Mazur, Opt. Lett. 26, 93 (2009)
13. J. Limpert, T. Scheiber, T. Clusnitzer, K. Zollner, H.J. Fuchs, E.B. Kley, H. Zellmer,
A. Tunnermann, Opt. Express 10, 628 (2002)
14. J. Nies, S. Biswal, F. Druon, J. Faure, M. Nantel, G.A. Mourou, A. Nishimura, H. Takuma,
J. Itatani, J.C. Chateloup, C. Honninger, IEEE J. Sel. Top. Quantum Electron. QE-4,
372 (1998)
15. K. Venkata Krishnaiah, R. Rajeswari, K. Upendra Kumar, S. Surendra Babu, I.R. Martı́n,
C.K. Jayasankar, J. Quant. Spectrosc. Radiat. Transfer 140, 37–47 (2014)
16. Y. Fan, IEEE J. Quantum Electron. QE-29, 1457 (1993)
17. D. Jia, W.M. Yen, J. Electrochem. Soc. 150, H61–H65 (2003)
18. L.H.C. Andrade, S.M. Lima, A. Novatski, A. Steimacher, J.H. Rohling, A.N. Medina,
A.C. Bento, M.L. Baesso, Y. Guyot, G. Boulon, Appl. Phys. Lett. 95, 081104 (2009)
19. J.K. Park, M.A. Lim, C.H. Kim, H.D. Park, J.T. Park, S.Y. Choi, Appl. Phys. Lett. 82,
683 (2003)
20. S. Nakamura, G. Fasol, The blue laser diode: GaN based light emitters and lasers (Springer,
Berlin, 1997)
21. M. Naftaly, A. Jha, W.G. Jordan, J. Appl. Phys. 84, 1800 (1998)
22. S.Q. Man, E.Y.B. Pun, P.S. Chung, Opt. Commun. 168, 369–373 (1999)
23. G.F. Imbush, in Advances in Nonradiative Processes in Solids, ed by B. Di Bartolo. (Plenum,
New York, 1991), p. 261
24. S.R. Bowman, J. Ganem, B.J. Feldman, A.W. Kueny, IEEE J. Quantum Electron. 30, 2925
(1994)
25. J.H. Campbell, T.I. Suratwala, J. Non Cryst. Solids 263/264, 318 (2000)
26. S. Surendra Babu, R. Rajeswari, K. Jang, C.E. Jin, K.H. Jang, H.J. Seo, C.K. Jayasankar, JOL
130, 1021–1025 (2010)
27. I. Iparraguire, J. Azkargorta, R. Balda, K. Venkata Krishnaiah, C.K. Jayasankar, M. Al-Saleh,
J. Fernández, Opt. Express 19, 19440 (2011)
28. P. Babu, K.H. Jang, E.S. Kim, L. Shi, H.J. Seo, F. Rivera-López, U.R. Rodrı́guez-Mendoza,
V. Lavı́n, R. Vijaya, C.K. Jayasankar, L. Rama Moorthy, J. Appl. Phys. 105, 013516 (2009)
29. L. Feng, J. Wang, Q. Tang, L.F. Liang, H.B. Liang, Q. Su, JOL 124, 187 (2007)
30. C.H. Qi, X.R. Zhang, H.F. Hu, Acta Opt Sinica 14, 583 (1994)
31. D. Lande, S.S. Orloy, A. Akella, K. Inoue, Opt. Lett. 22, 1722 (1997)
32. C.J. He, X.B. Chen, Y.G. Sun, L. Chen, Chin. J. Quantum. Electron. 19(2), 109 (2002)
204 K.V. Krishnaiah et al.

33. R. Xu, Y. Tian, L. Hu, J. Zhang, J Phys. Chem. A. 115, 6488–6492 (2011)
34. J.Y. Allain, M. Monerie, H. Poignant, Electron. Lett. 25, 1660 (1989)
35. S. Tanabe, X. Feng, T. Hanada, Opt. Lett. 25, 817 (2000)
36. J. Wu, Z. Yao, J. Zong, S. Jiang, Opt. Lett. 32, 638 (2007)
37. M. Yamada, A. Mori, K. Kobayashi, H. Ono, T. Kanamori, K. Oikawa, Y. Nishida, Y. Ohishi,
IEEE Photon. Technol. Lett. 10, 1244 (1998)
38. P. Moulton., Solid State Res. Rep., Lincoln Lab., MIT, pp. 15–21 (1982).
39. B. Peng, T. Izumitani, Rev. Laser Eng. 22, 823 (1993)
40. B.N. Samson, L.R. Pinckney, J. Wang, G.H. Beall, N.F. Borrelli, Opt. Lett. 27, 1309–1311
(2002)
41. T. Suzuki, G.S. Murugan, Y. Ohishi, Appl. Phys. Lett. 86, 131903 (2005)
42. T. Suzuki, K. Horibuchi, Y. Ohishi, J. Non Cryst. Solids 351, 2304–2309 (2005)
43. B. Wu, J. Ruan, J. Ren, D. Chen, C. Zhu, S. Zhou, J. Qiu, Appl. Phys. Lett. 92, 041110 (2008)
44. T. Suzuki, Y. Arai, Y. Ohishi, JOL 128, 603–609 (2008)
45. S. Tanabe, X. Feng, Appl. Phys. Lett. 77, 818 (2000)
46. B. Wu, S. Zhou, J. Ruan, Y. Qiao, D. Chen, C. Zhu, J. Qiu, Opt. Express 16, 2508 (2008)
47. H.-T. Sun, J. Zhou, J. Qiu, Prog. Mater. Sci. 64, 1–72 (2014)
48. Y. Fujimoto and M. Nakatsuka, Jpn. J. Appl. Phys., Part 1 40, L279 (2001).
49. X. Meng, J. Qiu, M. Peng, D. Chen, Q. Zhao, X. Jiang, C. Zhu, Opt. Express 13, 1628 (2005)
50. M. Peng, B. Wu, N. Da, C. Wang, D. Chen, C. Zhu, J. Qiu, J. Non-Cryst, J. Non Cryst. Solids
354, 1221–1225 (2008)
51. Y. Fujimoto and M. Nakatsuka, Jpn. J. Appl. Phys. 40(Part 2, No. 3B), L279–L281 (2001).
52. J.J. Berzelious, Ann, Phys. Chem 32, 577 (1834)
53. J.E. Stanworth, Nature 169, 581 (1952)
54. R.A.H. El-Mallawan, Tellurite Glasses Handbook: Physical Properties and Data, 2nd edn.
(2012)
55. A. Jha, B.D.O. Richards, G. Jose, T. Toney Fernandez, C.J. Hill, J. Lousteau, P. Joshi, Int.
Mater. Rev. 57, 357–382 (2012)
56. P. Balaya, C.S. Sunandana, J. Non Cryst. Solids 175, 51 (1994)
57. W. Vogel, Glass Chemistry, 2nd edn. (Springer, Berlin, 1992), p. 46, 166, 172
58. G.W. Brady, J. Chem. Phys. 27, 300 (1957)
59. C.S. Sunandana, T. Kumaraswani, J. Non Cryst. Solids 85, 247 (1986)
60. J.S. Wang, E.M. Vogel, E. Snitzer, Opt. Mater. 3, 187–203 (1994)
61. H. Zhang, Z. Zhou, A. Lin, J. Cheng, H. Liu, J. Si, F. Chen, X. Hou, J. Opt. 14,
065201–065206 (2012)
62. J. Jackson, C. Smith, J. Massera, C. Rivero-Baleine, C. Bungay, L. Petit, K. Richardson, Opt.
Express 17, 9071 (2009)
63. Y. Tsang, B. Richards, D. Binks, J. Lousteau, A. Jha, Opt. Express 16, 10690–10695 (2008)
64. A. Lin, A. Zhang, E.J. Bushong, J. Toulouse, Opt. Express 17, 16716–16721 (2009)
65. M.R. Oermann, H. Ebendorff-Heidepriem, Y. Li, T. Foo, T.M. Monro, Opt. Express 17,
15578–15584 (2009)
66. A. Ryasnyanskiy, A. Lin, C. Guintrand, I. Biaggio, and J. Toulouse, Optical Fibre Commu-
nication Conference, OSA Technical Digest (CD) (Optical Society of America, 2011),
pp. 1–3.
67. M. Liao, X. Yan, G. Qin, C. Chaudhari, T. Suzuki, Y. Ohishi, Opt. Express 17, 15481–15490
(2009)
68. Y. Ohishi, G. Qin, M. Liao, X. Yan, and T. Suzuki, in Optical Fiber Communication (OFC),
Collocated National Fiber Optic Engineers Conference, 2010 Conference on (OFC/NFOEC),
(2010), pp. 1–3
69. K. Driesen, V.K. Tikhomirov, C. G€ orller-Walrand, V.D. Rodriguez, A.B. Seddon, Appl.
Phys. Lett. 88, 073111 (2006)
70. S. Balaji, A.D. Sontakke, R. Sen, K. Annapurna, Opt. Mater. Express 1, 138 (2011)
71. E.R. Taylor, L. Ng, N.P. Sessions, H. Buerger, J. Appl. Phys. 92, 112–117 (2002)
8 Broadband Emission in Tellurite Glasses 205

72. M.A. Hughes, T. Akada, T. Suzuki, Y. Ohishi, D.W. Hewak, Opt. Express 17, 19345 (2009)
73. S.M. Lima, J.R. Silva, L.H.C. Andrade, A. Novatski, A.N. Medina, A.C. Bento, M.L. Baesso,
Opt. Lett. 35, 1055 (2010)
74. A. Mori, T. Sakamot, K. Kobayshi, J. Light Technol. 20, 822 (2002)
75. X. Feng, S. Tanabe, T. Hanada, J. Am. Ceram. Soc. 84, 165 (2001)
76. A. Mori, Y. Ohishi, S. Sudo, Electron. Lett. 33, 863 (1997)
77. S.X. Shen, A. Jha, X.B. Liu, M. Naftaly, K. Bindra, H.J. Bookey, A.K. Kar, J. Am. Ceram.
Soc. 85(6), 1391–1395 (2002)
78. S.X. Shen, A. Jha, E. Zhang, S.J. Wilson, C. R. Chim. 5(12), 921–938 (2002)
79. A.B. James, J. Lightwave Teachnol. 9, 220–227 (2002)
80. R.J. Mears, L. Reekie, I.M. Jauncey, D.N. Payne, High-gain rare earth-doped fiber amplifier
at 1.54 mm, in: Optical Fiber Communication Conference, 1987 OSA Technical Digest
Series, Optical Society of America, 1987, p. 167
81. H. Ono, A. Mori, K. Shikano, M. Shimizu, IEEE Photon. Technol. Lett. 14, 1070 (2002)
82. W.J. Miniscalco, J. Lightwave Technol. 9, 234 (1991)
83. B. Clesca, D. Ronarc’h, D. Bayart, Y. Sorel, L. Hamon, M. Guibert, J.L. Beylat, J.F. Kerdiles,
M. Semenkoff, IEEE Photon. Technol. Lett. 6(509), 3 (1994)
84. Y. Ohishi, A. Mori, M. Yamada, H. Ono, Y. Nishida, K. Oikawa, Opt. Lett. 23, 274 (1998)
85. T. Sakamoto, M. Shimizu, M. Yamada, T. Kanamori, Y. Ohishi, Y. Terunuma, S. Sudo, IEEE
Photon. Technol. Lett. 8, 349 (1996)
86. J. Zhang, S. Dai, G. Wang, H. Sun, L. Zhang, L. Hu, JOL 115, 45–52 (2005)
87. M. Weber, J. Myers, D. Blackburn, J. Appl. Phys. 52, 2944 (1981)
88. J. Wang, E. Vogel, E. Snitzer, J. Jackel, V. Silva, Y. Silberg, J. Non Cryst. Solids 178,
109 (1994)
89. B.R. Judd, Phys. Rev. 127, 750–761 (1962)
90. G.S. Ofelt, J. Chem. Phys. 37, 511–520 (1962)
91. L. Smentek, B.G. Wybourne, B.A. Hess Jr., J. Alloys Compd. 323–324, 645–648 (2001)
92. J. Garcı́a Solé, L.E. Bausa, D. Jaque, An Introduction to the Optical Spectroscopy of
Inorganic Solids (John Wiley and Sons Ltd., London, 2005), pp. 224–233
93. D.E. McCumber, Phys. Rev. A 136, 954 (1964)
94. S.A. Payne, L.L. Chase, L.K. Smith, W.L. Kway, W.F. Krupke, IEEE J. Quantum Electron.
QE-28, 2619 (1992)
95. C. Gorller-Warland, K. Binnimans, in Handbook on the Physics and Chemistry of Rare
Earths, ed by K. A. Gschneidner Jr., L. Eyring, vol 25 (North Holland, Amsterdam, 1998)
96. M.J.F. Digonnet, C.J. Gaeta, Appl. Optics 24, 333 (1985)
97. S. Taccheo, P. Laporta, C. Svelto, Appl. Phys. Lett. 68, 2521 (1996)
98. P. Nandi, G. Jose, C. Jayakrishnan, S. Debbarma, K. Chalapathi, K. Alti,
A.K. Dharmadhikari, J.A. Dharmadhikari, D. Mathur, Opt. Express 14, 12145–12150 (2006)
99. M.V. Vijaya Kumar, K. Rama Gopal, R.R. Reddy, G.V. Lokeswara Reddy, B.C. Jamalaiah,
JOL 142, 128–134 (2013)
100. I. Jlassi, H. Elhouichet, M. Ferid, C. Barthou, JOL 130, 2394 (2010)
101. M. Seshadri, E.F. Chillcce, J.D. Marconi, F.A. Sigoli, Y.C. Ratnakaram, L.C. Barbosa, J. Non
Cryst. Solids 402, 141–148 (2014)
102. K. Venkata Krishnaiah, J. Marques-Hueso, K. Suresh, G. Venkataiah, B.S. Richards,
C.K. Jayasankar, JOL 169, 270–276 (2016)
103. A. Miguel, R. Morea, J. Gonzalo, M.A. Arriandiaga, J. Fernandez, R. Balda, JOL 140, 38–44
(2013)
104. H. Lin, S. Jiang, J. Wu, F. Song, N. Peyghambarian, E.Y.B. Pun, J. Phys. D Appl. Phys. 36,
812–817 (2003)
105. V. Nazabal, S. Todoroki, S. Inoue, T. Matsumoto, S. Suehara, T. Hondo, T. Araki,
T. Cardinal, J. Non Cryst. Solids 326-327, 359–363 (2003)
106. K. Damak, E. Yousef, S. AlFaify, C. Rüssel, R. Ma^alej, Opt. Mater. Express. 4, 597–612
(2014)
206 K.V. Krishnaiah et al.

107. R. Wang, X. Meng, F. Yin, Y. Feng, G. Qin, W. Qin, Opt. Mater. Express. 3, 1127–1136
(2013)
108. Z. Ashur, S. Mahraz, M.R. Sahar, S.K. Ghoshal, M. Reza Dousti, JOL 144, 139–145 (2013)
109. F. Huang, X. Liu, L. Hu, D. Chen, J. Appl. Phys. 116, 033106 (2014)
110. H. Fares, H. Elhouichet, B. Gelloz, M. Ferid, J. Appl. Phys. 116, 123504 (2014)
111. H. Lin, S. Tanabe, L. Lin, Y.Y. Hou, K. Liu, D.L. Yang, T.C. Ma, J.Y. Yu, E.Y.B. Pun, JOL
124, 167–172 (2007)
112. H. Lin, K. Liu, E.Y.B. Pun, T.C. Ma, X. Peng, Q.D. An, J.Y. Yu, S.B. Jiang, Chem. Phys.
Lett. 398, 146–150 (2004)
113. P. Babu, H.J. Seo, C.R. Kesavulu, K.H. Jang, C.K. Jayasankar, JOL 129, 444–448 (2009)
114. U.R. Rodriguez-Mendoza, E.A. Lalla, J.M. Caceres, F. Rivera-Lopez, S.F. Leon-Luis,
V. Lavin, JOL 131, 1239–1248 (2011)
115. M. Pokhrel, G.A. Kumar, S. Balaji, R. Debnath, D.K. Sardar, JOL 132, 1910–1916 (2012)
116. V.A.G. Rivera, L.C. Barbosa, JOL 156, 116–123 (2014)
117. S. Brovelli, A. Galli, R. Lorenzi, F. Meinardi, G. Spinolo, S. Tavazzi, V. Sigaev, S. Sukhov,
P. Pernice, A. Aronne, E. Fanelli, A. Paleari, J. Non Cryst. Solids 353, 2150–2156 (2007)
118. S.O. Baki, L.S. Tan, C.S. Kan, H.M. Kamari, A.S.M. Noor, M.A. Mahdi, J. Non Cryst. Solids
362, 156–161 (2013)
119. R. Rolli, M. Montagna, S. Chaussedent, A. Monteil, V.K. Tikhomirov, M. Ferrari, Opt.
Mater. 21, 743–748 (2003)
120. J. Coelho, J. Azevedo, G. Hungerford, N.S. Hussain, Opt. Mater. 33, 1167–1173 (2011)
121. R. Balda, M. Al-Saleh, A. Miguel, J.M. Fdez-Navarro, J. Fernández, Opt. Mater. 34, 481–486
(2011)
122. A. Miguel, M. Al-Saleh, J. Azkargorta, R. Morea, J. Gonzalo, M.A. Arriandiaga,
J. Fernandez, R. Balda, Opt. Mater. 35, 2039–2044 (2013)
123. X. Shen, Q.H. Nie, T.F. Xu, Y. Gao, Spectrochim. Acta A 61, 2827–2831 (2005)
124. H. Desirena, E De la Rosa, A Shulzgen, S Shabet and N Peyghambarian. J. Phys. D Appl.
Phys. 41, 095102 (2008)
125. K. Selvaraju, N. Vijaya, K. Marimuthu, V. Lavin, Phys. Status Solidi A 210(3), 607–615
(2013)
126. K. Maheshvaran, S. Arunkumar, K. Venkata Krishnaiah, K. Marimuthu, J. Mol. Struct. 1079,
130–138 (2015)
127. G. Lakshminarayana, J. Qiu, M.G. Brik, G.A. Kumar, I.V. Kityk, J. Phys. Condens. Matter
20, 375101 (2008)
128. F. Huang, X. Liu, Y. Ma, S. Kang, L. Hu, D. Chen, Sci. Rep. 5(8233), 1–6 (2015)
129. S. Tanabe, T. Ohyagi, N. Soga, T. Hanada, Phys. Rev. B 46, 3305 (1992)
130. L.M. Fortes, L.F. Santos, M. Clara Goncalves, R.M. Almeida, M. Mattarelli, M. Montagna,
A. Chiasera, M. Ferrari, A. Monteil, S. Chaussedent, G.C. Righini, Opt. Mater. 29, 503 (2007)
131. D.K. Sardar, C.C. Russell III, R.M. Yow, J. Appl. Phys. 95, 1180–1184 (2004)
132. X. Qiao, X. Fan, M. Wang, X. Zhang, Opt. Mater. 27, 597–603 (2004)
133. N. Jaba, H. BenMansour, A. Kanoun, A. Brenier, B. Champagnon, JOL 129, 270–276 (2009)
134. S. Shen, M. Naftaly, A. Jha, Opt. Commun. 205, 101 (2002)
135. W.J. Chung, Y.G. Choi, JOL 130, 2175–2179 (2010)
136. D. Zhou, R. Wang, Z. Yang, Z. Song, Z. Yin, J. Qiu, J. Non Cryst. Solids 357, 2409–2412
(2011)
137. M. Naftaly, S. Shen, A. Jha, Appl. Optics 39, 4979 (2000)
138. R. Balda, J. Fernández, S. Garcı́a-Revilla, J.M. Fernández-Navarro, Opt. Express 15, 6750
(2007)
139. R. Balda, L.M. Lacha, J. Fernández, M.A. Arriandiaga, J.M. Fernández-Navarro, D. Munoz-
Martin, Opt. Express 16, 11836 (2008)
140. J.H. Yang, S.X. Dai, Y.F. Zhou, L. Wen, L.L. Hu, Z.H. Jiang, J. Appl. Phys. 93, 977 (2003)
141. X. Zou, T. Izumitani, J. Non Cryst. Solids 162, 68 (1993)
8 Broadband Emission in Tellurite Glasses 207

142. S. Jiang, T. Luo, B.C. Hwang, F. Smekatala, K. Seneschal, J. Lucas, N. Payghambarian,

J. Non Cryst. Solids 263–264, 364 (2000)
143. H. Lin, E.Y.B. Pun, S.Q. Manans, X.R. Liu, J. Opt. Soc. Am. B 18, 602 (2001)
144. Y. Benmadani, A. Kermaoui, M. Chalal, W. Khemici, A. Kellou, F. Pelle, Opt. Mater. 35,
2234–2240 (2013)
145. A.A. Assadi, K. Damak, R. Lachheb, A. Herrmann, E. Yousef, C. Russel, R. Maalej, J. Alloys
Compd. 620, 129–136 (2015)
146. Y. Luo, J. Zhang, J. Sun, S. Lu, X. Wang, Opt. Mater. 28, 255 (2006)
147. N.G. Boetti, J. Lousteau, A. Chiasera, M. Ferrari, E. Mura, G.C. Scarpignato, S. Abrate,
D. Milanese, JOL 132, 1265–1269 (2012)
148. S. Dai, C. Yu, G. Zhou, J. Zhang, G. Wang, L. Hu, JOL 117, 39–45 (2006)
149. M. Mattarelli, M. Montagna, L. Zampedri, A. Chiasera, M. Ferrari, G.C. Righini,
L.M. Fortes, M.C. Goncalves, L.F. Santos, R.M. Almeida, Europhys. Lett. 71(3), 394–399
(2005)
150. P. Babu, H.J. Seo, K.H. Jang, K. Upendra Kumar, C.K. Jayasankar, Chem. Phys. Lett. 445,
162–166 (2007)
151. V.A.G. Rivera, M. El-Amraoui, Y. Ledemi, Y. Messaddeq, E. Marega Jr., JOL 145, 787–792
(2014)
152. A. Miguel, M.A. Arriandiaga, R. Morea, J. Fernandez, J. Gonzalo, R. Balda, JOL 154,
136–141 (2014)
153. H. Jeong, K. Oh, S.R. Han, T.F. Morse, Chem. Phys. Lett. 367, 507 (2003)
154. V.P. Gapontsev, S.M. Matitrsin, A.A. Isineev, V.B. Kravchenko, Opt. Laser Technol. 14,
189 (1982)
155. G. Lakshminarayana, R. Vidya Sagar, S. Buddhudu, JOL 128, 690–695 (2008)
156. W. Stambouli, H. Elhouichet, C. Barthou, M. Férid, J. Alloys, J. Alloys Compd. 580,
310–315 (2013)
157. S. Shen, L. Huang, P. Joshi, A. Jha, Electron. Lett. 39, 1797 (2003)
158. D. Yin, S. Peng, Y. Qi, S. Zheng, Y. Zhou, X. Wang, J. Alloys Compd. 581, 534–541 (2013)
159. J. Żmojda, M. Kochanowicz, D. Dorosz, Photon. Lett. Poland 6(2), 56–58 (2014)
160. L. Gomes, J. Lousteau, D. Milanese, E. Mura, S.D. Jackson, J. Opt. Soc. Am. B 31,
429 (2014)
161. Y. Luo, J. Zhang, X. Zhang, X. Wang, J. Appl. Phys. 103, 063107 (2008)
162. P. Joshi, S. Shen, A. Jha, J. Appl. Phys. 103, 083543 (2008)
163. X. Shen, Q. Nie, T. Xu, S. Dai, X. Wang, Spectrochim. Acta A 66, 389–393 (2007)
164. S. Zheng, Y. Zhou, D. Yin, Y. Qi, S. Peng, X. Xu, Mater. Lett. 106, 341–343 (2013)
165. C. Yu, D. He, G. Wang, J. Zhang, L. Hu, J. Non Cryst. Solids 355, 2250–2253 (2009)
166. S. Shen, A. Jha, S.J. Wilson, E. Zhang, J. Non Cryst. Solids 326–327, 510–514 (2003)
167. J. Yang, Z. Liyan, L. Wen, S. Dai, L. Hu, Z. Jiang, Chem. Phys. Lett. 384, 295–298 (2004)
168. V. Nazabal, S. Todoroki, A. Nukui, T. Matsumoto, S. Suehara, T. Hondo, T. Araki, S. Inoue,
C. Rivero, T. Cardinal, J. Non Cryst. Solids 325, 85–102 (2003)
169. E.F. Chillcce, I.O. Mazali, O.L. Alves, L.C. Barbosa, Opt. Mater. 33, 389–396 (2011)
170. C. Yu, J. Zhang, L. Wen, Z. Jiang, Mater. Lett. 61, 3644–3646 (2007)
171. A. Miguel, R. Moreab, M.A. Arriandiaga, M. Hernandez, F.J. Ferrer, C. Domingo,
J.M. Fernandez-Navarro, J. Gonzalo, J. Fernandez, R. Balda, J. Eur. Ceram. Soc. 34,
3959–3968 (2014)
172. A. Jha, P. Joshi, S. Shen, Opt. Express 16, 13526 (2008)
173. S. Dai, C. Yu, G. Zhou, J. Zhang, G. Wang, J. Non Cryst. Solids 354, 1357–1360 (2008)
174. M. Irannejad, G. Jose, A. Jha, D.P. Steenson, Opt. Mater. 33, 215–219 (2010)
175. I. Jlassi, H. Elhouichet, M. Ferid, R. Chtourou, M. Oueslati, Opt. Mater. 32, 743–747 (2010)
176. B.D.O. Richards, A. Jha, G. Jose, X. Jiang, Proc. SPIE 8039, 80390R (2011)
177. A. Lin, A. Ryasnyanskiy, J. Toulouse, Opt. Lett. 36, 740 (2011)
178. H. Zhan, Z. Zhou, J. He, A. Lin, Opt. Lett. 37, 3408 (2012)
208 K.V. Krishnaiah et al.

179. L. Gomes, M. Oermann, H. Ebendorff-Heidepriem, D. Ottaway, T. Monro, A.F.H. Librantz,

S.D. Jackson, J. Appl. Phys. 110, 083111 (2011)
180. B. Wang, L. Cheng, H. Zhong, J. Sun, Y. Tian, X. Zhang, B. Chen, Opt. Mater. 31, 1658
(2009)
181. H. Zhan, A. Zhang, J. He, Z. Zhou, L. Li, T. Shi, X. Xia, J. Si, A. Lin, J. Alloys, J. Alloys
Compd. 582, 742–746 (2014)
182. J. Massera, A. Haldeman, J. Jackson, C. Rivero-Baleine, L. Petit, K. Richardson, J. Am.
Ceram. Soc. 94(1), 130–136 (2011)
183. I. Savelii, F. Desevedavy, J.-C. Jules, G. Gadret, J. Fatome, B. Kibler, H. Kawashima,
Y. Ohishi, F. Smektala, Opt. Mater. 35, 1595–1599 (2013)
184. G.A. Thomas, B.I. Shraiman, P.F. Glodis, M.J. Stephen, Nature 404(6775), 262–264 (2000)
185. S. Kasap, in The Optics Encyclopedia, ed by T. Brown, K. Creath, H. Kogelnik, M. A. Kriss,
J. Schmit, M. J. Weber. Optoelectronics, vol 4 (Wiley-VCH, Weihein, 2004), pp. 2237–2284
186. J. Dong, Y.Q. Wei, A. Wonfor, R.V. Penty, I.H. White, J. Lousteau, G. Jose, A. Jha, IEEE
Photon. Technol. Lett. 23(11), 736–738 (2011)
187. L. Huang, A. Jha, S. Shen, X. Liu, Opt. Express 12(11), 2429–2434 (2004)
188. H. Takebe, K. Yoshino, T. Murata, K. Morinaga, J. Hector, W.S. Brocklesby, D.W. Hewak,
J. Wang, D.N. Payne, Appl. Optics 36(24), 5839–5843 (1997)
189. C. Strohh€ofer, A. Polman, Appl. Phys. Lett. 81(8), 1414–1416 (2002)
190. B. Zhou, H. Lin, E.Y.B. Pun, Opt. Express 18(18), 18805–18810 (2010)
191. S. Tanabe, C. R. Chim. 5(12), 815–824 (2002)
192. B. Zhou, H. Lin, B. Chen, E.Y.B. Pun, Opt. Express 19, 6514 (2011)
193. B. Zhou, L. Tao, Y.H. Tsang, W. Jin, E.Y.B. Pu, Opt. Express 20, 3803 (2012)
194. B. Zhou, L. Tao, Y.H. Tsang, W.J. Tsang, E.Y. Pun, Photon. Technol. Lett. 2, 924 (2012)
195. W. Shockley, H.J. Queisser, J. Appl. Phys. 32, 510–519 (1961)
196. H. Cotal, C. Fetzer, J. Boisvert, G. Kinsey, R. King, P. Hebert, et al., Energ. Environ. Sci. 2,
174–192 (2009)
197. L.L. Neindre, T. Luo, B.-C. Hwang, J. Watson, S. Jiang, N. Peyghambarian, Proc. SPIE 3622,
58–65 (1999)
198. X. Liu, Y. Chun-Hua, J.A. Capobianco, Chem. Soc. Rev. 44, 1299–1301 (2015)
199. G.H. Dieke, H.M. Crosswhite, Appl. Optics 2, 675–686 (1963)
200. T. Trupke, M.A. Green, P. Wurfel, J. Appl. Phys. 92, 4117 (2002)
201. P. Gibart, F. Auzel, J.C. Guillaume, K. Zahraman, Jpn. J. Appl. Phys. 35, 4401–4402 (1996)
202. J. de Wild, J.K. Rath, A. Meijerink, W.G.J.H.M. van Sark, R.E.I. Schropp, Sol. Energ. Mat.
Sol. Cells 94, 2395 (2010)
203. A. Shalav, B.S. Richards, T. Trupke, K.W. Krämer, H.U. Güdel, Appl. Phys. Lett. 86, 013505
(2005)
204. A. Ivaturi, S.K.W. MacDougall, R. Martı́n-Rodrı́guez, M. Quintanilla, J. Marques-Hueso,
A. Meijerink, K.W. Krämer, B.S. Richards, J. Appl. Phys. 114, 013505 (2013)
205. M. Pollnau, D.R. Gamelin, S.R. Lüthi, H.U. Güdel, M.P. Hehlen, Phys. Rev. B 61, 3337
(2000)
206. S.K.W. MacDougall, A. Ivaturi, J. Marques-Hueso, K.W. Krämer, B.S. Richards, Opt.
Express 20(S6), A879 (2012)
207. J. Marques-Hueso, R. Peretti, R. Abargues, B.S. Richards, C. Seassal, J.P. Martı́nez-Pastor,
Adv. Opt. Mater. 3, 568–574 (2015)
208. C. Strümpel, M. McCann, C. del Ca~ nizo, I. Tobias, P. Fath, presented at 20th EPVSEC,
Barcelona, Spain 2005.
209. H. Lin, J. Marqués-Hueso, D. Chen, Y. Wang, B.S. Richards, Mater. Res. Bull. 47, 4433
(2012)
210. W. Zou, C. Visser, J.A. Maduro, M.S. Pshenichnikov, J.C. Hummelen, Nat. Photon. 6,
560 (2012)
211. J.-C. Boyer, F.C.J.M. van Veggel, Nanoscale 2, 1417 (2010)
212. A. Shalav, Ph. D. Thesis, University of New South Wales, Sydney 2006.
8 Broadband Emission in Tellurite Glasses 209

213. J. C. Goldschmidt, P. Loper, S. Fischer, S. Janz, M. Peters, S. W. Glunz, G. Willeke,

E. Lifshitz, K. Kramer, D. Biner, presented at Optoelectronic and Microelectronic Materials
and Devices (COMMAD), Sydney, 2008.
214. M.A. Hernández-Rodrı́guez, M.H. Imanieh, L.L. Martı́n, I.R. Martı́n, Sol. Energ. Mat. Sol.
Cells 116, 171–175 (2013)
215. M. Wyrwas, J. Żmojda, D. Dorosz, M. Kochanowicz, K. Iwanowicz, T. Ragin, W. Mazerski,
J. Dorosz Proc, Proc. SPIE 8454, 845415–845411 (2012)
216. D.L. Dexter, Phys. Rev. 108, 630–633 (1957)
217. W.W. Piper, J.A. DeLuca, F.S. Ham, JOL 8, 344–348 (1974)
218. J.L. Sommerdijk, A. Bril, A.W. De Jager, JOL 8, 341–343 (1974)
219. J.K. Berkowitz, J.A. Olsen, JOL 50, 111–121 (1991)
220. W. W. Piper and F. S. Ham, J. Lumin., 1974, 8, 344–348, P. Vergeer, T.J.H. Vlugt,
M.H.F. Kox, M.I. den Hertog, J.P.J.M. van der Eerden, A. Meijerink, Phys. Rev. B
71 (2005) 014119.
221. H. Lin, S. Zhou, H. Teng, Y. Li, W. Li, X. Hou, T. Jia, J. Appl. Phys. 107, 043107 (2010)
222. D.Q. Chen, Y.L. Yu, H. Lin, P. Huang, Z.F. Shan, Y.S. Wang, Opt. Lett. 35(2), 220–222
(2010)
223. J.J. Eilers, D. Biner, J.T. van Wijngaarden, K. Krämer, H.-U. Güdel, A. Meijerink, Appl.
Phys. Lett. 96, 151106 (2010)
224. R.T. Wegh, H. Donker, K.D. Oskam, A. Meijerink, Science 283, 664 (1999)
225. T. Trupke, M.A. Green, P. Wurfel, J. Appl. Phys. 92(3), 1668–1674 (2002)
226. Q.Y. Zhang, X.Y. Huang, Progr. Mater. Sci. 55, 353–427 (2010)
227. G. Lakshminarayana, J. Qiu, J. Alloys Compd. 481, 582–589 (2009)
228. Z. Liu, J. Li, L. Yang, Q. Chen, Y. Chu, N. Dai, Sol. Energ. Mat. Sol. Cells 122, 46–50 (2014)
229. A. Boccolini, J. Marques-Hueso, D. Chen, Y. Wang, B.S. Richards, Sol. Energ. Mat. Sol.
Cells 122, 8–14 (2014)
230. X. Zhou, G. Wang, K. Zhou, Q. Li, Opt. Mater. 35, 600–603 (2013)
231. E.M. Dianov, Light Sci. Appl. 1, e12 (2012)
232. X.G. Meng, J.R. Qiu, M.Y. Peng, D.P. Chen, Q.Z. Zhao, X.W. Jiang, C.S. Zhu, Opt. Express
13(5), 1635–1642 (2005)
233. X. Wang, H. Xia, Opt. Commun. 268(1), 75–78 (2006)
234. M. Peng, J. Qiu, D. Chen, X. Meng, C. Zhu, Opt. Lett. 30(18), 2433–2435 (2005)
235. S. Khonthon, S. Morimoto, Y. Arai, Y. Ohishi, J. Cerma. Soc. Jpn. 115(1340), 259–263
(2007)
236. S. Zhou, H. Dong, H. Zeng, G. Feng, H. Yang, B. Zhu, J. Qiu, Appl. Phys. Lett. 91, 061919-
1–061919-3 (2007)
237. E.M. Diano, J. Non Cryst. Solids 355, 1861–1864 (2009)
238. M. Peng, Q. Zhao, J. Qiu, L. Wondraczek, J. AM. Ceram. Soc. 92(2), 542–544 (2009)
239. R.H. Stolen, E.P. Ippen, A.R. Tynes, Appl. Phys. Lett. 20, 62–64 (1972)
240. M. Islam, IEEE J. Sel. Top. Quantum Electron. 8, 548–559 (2002)
241. J. Bromage, J. Lightwave, Technol. 22, 79–93 (2004)
242. J. Hu, B.S. Marks, C.R. Menyuk, J. Lightwave Technol. 22, 1519–1522 (2004)
243. G. Bolognini, S. Sugliani, F. Pasquale, IEEE Photon. Technol. Lett. 16, 1286–1288 (2004)
244. V.E. Perlin, H.G. Winful, IEEE Photon. Technol. Lett. 14, 1199–1121 (2002)
245. T. Zhang, X. Zhang, G. Zhang, IEEE Photon. Technol. Lett. 17, 1175–1177 (2005)
246. R. Jose, Y. Ohishi, Appl. Phys. Lett. 89, 121122 (2006)
247. J. Yuan, Q. Yang, D.D. Chen, Q. Qian, S.X. Shen, Q.Y. Zhang, Z.H. Jiang, J. Appl. Phys. 111,
103511 (2012)
248. R. Jose, Y. Ohishi, Appl. Phys. Lett. 90, 211104 (2007)
249. M. Irannejad, G. Jose, A. Jha, P. Steenson, Opt. Commun. 285, 2646–2649 (2012)
250. N. Bloembergen, P. Lallemand, Phys. Rev. Lett. 16, 81–84 (1966)
210 K.V. Krishnaiah et al.

251. R. Thapa, D. Rhonehouse, D. Nguyen, K. Wiersma, C. Smith, J. Zong, A. Chavez-Pirson, in:

David H. Titterton, Mark A. Richardson, Robert J. Grasso, Harro Ackermann, Proc. SPIE,
889808 (2013).
252. Z.-X. Jia, L. Liu, C.-F. Yao, G.-S. Qin, Y. Ohishi, W.-P. Qin, J. Appl. Phys. 115, 063106
(2014)
253. E.F. Schubert, Light-Emitting Diodes, 2nd edn. (Cambridge University Press, New York,
2006), p. 292
254. C. Lin, R.S. Liu, Y.S. Tang, S.F. Hu, J. Electrochem. Soc. 155, J248 (2008)
255. S. Tanabe, S. Fujita, S. Yoshihara, A. Sakamoto, S. Yamamoto, Proc. SPIE 5941, 594112
(2005)
256. Q. Xu, Z.M. Yang, S.X. Dai, J. Yang, L. Hu, Z.J. Jiang, J. Alloys Compd. 361, 313 (2003)
257. S. Fujita, S. Yoshihara, A. Sakamoto, S. Yamamoto, S. Tanabe, Proc. SPIE 5941, 594111
(2005)
258. K. Boonin, P. Yasaka, J. Kaewkhao, Appl. Mech. Mater 431, 27 (2013)
259. G. Venkataiah, C.K. Jayasankar, J. Mol. Struct. 1084, 182 (2015)
260. Y. Dwivedi, A. Rai, S.B. Rai, J. Appl. Phys. 104, 043509 (2008)
261. H. Desirena, E De la Rosa, P Salas and O Meza. J. Phys. D Appl. Phys. 44, 455308 (2011)
262. W. He, X. Wang, J. Zheng, X. Yan, J. Am. Ceram. Soc. 97, 1750 (2014)
263. F. Zhang, Z. Xiao, L. Yan, F. Zhu, A. Huang, Proc. SPIE 7518, 751817 (2009)
264. J. Pisarska, Opt. Mater. 31, 1784–1786 (2009)
265. K. Venkata Krishnaiah, K. Upendra Kumar, C.K. Jayasankar, Mater. Express 3, 61 (2013)
266. L.X. Sun, H. Gong, B.J. Chen, H. Lin, E.Y.B. Pun, J. Appl. Phys. 105, 106109 (2009)
267. R.I. Epstein, M.I. Buchwald, B.C. Edwards, T.R. Gosnell, C.E. Mungan, Nature 377(6549),
500–503 (1995)
268. J. Zhang, D. Li, R. Chen, Q. Xiong, Nature 493, 504 (2013)
269. B.C. Edwards, M.I. Buchwald, R.I. Epstein, Rev. Sci. Instrum. 69, 2050 (1998)
270. C.E. Mungan, J. Opt. Soc. Am. B 20, 1075 (2003)
271. P. Pringsheim, Z. Phys. A-Hadron Nucl. 57, 739–746 (1929)
272. G. Nemova, R. Kashyap, J. Opt. Soc. Am. B. 29, 3034 (2012)
273. E.S. Filho, K.V. Krishnaiah, Y. Ledemi, Y.-J. Yu, Y. Messaddeq, G. Nemova, R. Kashyap,
Opt. Express 23, 4630 (2015)
274. K. Venkata Krishnaiah, Y. Ledemi, E.S. Filho, Y. Messaddeq, R. Kashyap, Proc. SPIE 9380,
93800P-1-8 (2015)
275. E.S. Filho, K.V. Krishnaiah, Y.-J. Yu, Y. Ledemi, Y. Messaddeq, R. Kashyap, Proc. SPIE
9380, 93800R-1-9 (2015)
276. B. Heeg, P.A. DeBarber, G. Rumbles, Appl. Optics 44, 3117–3124 (2005)
277. M.P. Hehlen, in Optical Refrigeration: Science and Applications of Laser Cooling of Solids,
ed by R. Epstein, M. Sheik-Bahae. Design and fabrication of rare-earth-doped laser cooling
materials (Wiley-VCH, Weihein, 2009), pp. 33–74
278. D.V. Seletskiy, S.D. Melgaard, R.I. Epstein, A. Di Lieto, M. Tonelli, M. Sheik-Bahae, JOL
133, 5–9 (2013)
279. R. Epstein, M. Sheik-Bahae, Optical Refrigeration: Science and Applications of Laser
Cooling of Solids (Wiley-VCH, Weihein, 2009)
280. S. Melgaard, D. Seletskiy, V. Polyak, Y. Asmerom, M. Sheik-Bahae, Opt. Express 22, 7756
(2014)
281. G. Boulon, Y. Guyot, M. Ito, A. Bensalah, C. Goutaudier, G. Panczer, J.C. G^acon, Mol. Phys.
102, 1119–1132 (2004)
282. D. Nguyen, J. Zong, D. Rhonehouse, A. Miller, Z. Yao, Arturo Chavez-Pirson, G. Hardesty,
N. Kwong, R. Binder. Proc. SPIE 8275, 8275–8205 (2012)
283. D. L Rhonehouse, J. Zong, D. Nguyen, R. Thapa, K. Wiersma, C. Smith, A. Chavez-Pirson,
Proc. SPIE Vol. 8898 (2013) 88980D-1-8.
284. M.T. Murtagh, G.H. Sigel Jr., J.C. Fajardo, B.C. Edwards, J. Non Cryst. Solids 253, 50–57
(1999)
8 Broadband Emission in Tellurite Glasses 211

285. J. Fernandez, A. Mendioroz, R. Balda, M. Voda, M. Al-Saleh, A.J. Garcia-Adeva, J.L. Adam,
J. Lucas, Proc. SPIE 4645, 135 (2002)
286. J. Fernandez, A. Mendioroz, A.J. Garcia, R. Balda, J.L. Adam, Phys. Rev. B 62, 3213 (2000)
287. J.V. Guiheen, C.D. Haines, G.H. Sigel, R.I. Epstein, J. Thiede, W.M. Patterson, Phys. Chem.
Glasses 47, 167–176 (2006)
288. E.S. Filho, G. Nemova, S. Loranger, R. Kashyap, Opt. Express 21, 24711 (2013)
289. S.R. Bowman, C.E. Mungan, Appl. Phys. B B71, 807 (2000)
290. S. Bigotta, D. Parisi, L. Bonelli, A. Toncelli, M. Tonelli, A.D. Lieto, J. Appl. Phys. 100,
013109 (2006)
291. A. Mendioroz, J. Fernández, M. Voda, M. Al-Saleh, R. Balda, A.J. Garcı́a-Adeva, Opt. Lett.
15, 1525 (2002)
292. B. Zhong, J. Yin, Y. Jia, L. Chen, Y. Hang, J. Yin, Opt. Express 39, 2747 (2014)
293. D.L. Rhonehouse, J. Zong, D. Nguyen, R. Thapa, K. Wiersma, C. Smith, A. Chavez-Pirson,
Proc. SPIE 8898, 88980D-1-8 (2013)
294. C.W. Hoyt, M. Sheik-Bahae, R.I. Epstein, B.C. Edwards, J.E. Anderson, Phys. Rev. Lett. 85,
3600 (2000)
295. W. Patterson, S. Bigotta, M. Sheik-Bahae, D. Parisi, M. Tonelli, R. Epstein, Opt. Express 16,
1704 (2008)
296. J. Fernandez, A.J. Garcia-Adeva, R. Balda, Phys. Rev. Lett. 97, 033001 (2006)
297. A.J. Garcia-Adeva, R. Balda, J. Fernandez, Proc. SPIE 6461, 646102 (2007)
298. A.J. Garcia-Adeva, R. Balda, J. Fernandez, Opt. Mater. 31, 1075–1081 (2009)
299. N.J. Condon, S.R. Bowman, S.P. O’Connor, R.S. Quimby, C.E. Mungan, Opt. Express 17,
5466 (2009)
Chapter 9
Tellurite Glass Fibers for Mid-infrared
Nonlinear Applications

Xian Feng, Peter Horak, and Francesco Poletti

9.1 Introduction

The mid-infrared (mid-IR) 2–5 μm and 8–13 μm regions are the two atmospheric
transmission windows, where the Earth’s atmosphere is relatively transparent.
These two windows are important for using remote laser sensing for the atmo-
spheric, security, and industrial applications such as detecting remote explosives,
countermeasures against heat-seeking missiles, and covert communication systems
[1]. A broadband or tunable laser source with medium- or high-average power level
(100 mW to 10 W) ranging between 2 and 5 μm is thus highly demanded.
Unfortunately, the commonly used active laser sources, e.g., transition metal
ion-doped chalcogenide crystal lasers, quantum cascade lasers, and rare-earth-
doped fluoride fiber lasers, are not able to completely cover the whole range. For
example, the state-of-the-art Cr(II):ZnSe tunable laser can only cover the range of
1.97–3.35 μm [2], and the state-of-the-art Fe(II):ZnSe tunable laser can only cover
the range of 3.95–5.05 μm [3]; for quantum cascade laser, 3 μm is the short
wavelength limit due to the material limitations such as conduction band offset
[4, 5]; mid-IR rare-earth-doped fluoride fiber lasers can provide relatively narrow-
line emission discrete spectra in the range of 2–4 μm, because the phonon energy
(500–600 cm
1) of fluoride glass hosts is still not low enough for lasing at longer
wavelengths and the emission wavelengths because 4f-4f electron transitions of a
certain type of rare-earth ions are not very sensitive to the glass host [6, 7].

X. Feng (*)
Institute of Laser Engineering, Beijing University of Technology, 100 Pingleyuan, Chaoyang
District, Beijing 100124, China
e-mail: [email protected]
P. Horak • F. Poletti
Optoelectronics Research Centre, University of Southampton, Highfield, Southampton SO17
1BJ, UK

© Springer International Publishing AG 2017 213

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_9
214 X. Feng et al.

Instead, χ(2) nonlinear crystal optical parametric oscillator (OPO) is one of the
most commonly used approaches for generating widely tunable output covering the
whole 2–5 μm range [8]. But for achieving medium- or high-average power level
(100 mW to 10 W) and wide wavelength tunability, the volume of an OPO is
normally large, and complicated optical configurations are required.
On the other hand, the recent progress in tailored low-dispersion highly
nonlinear photonic crystal fibers (PCFs, also called as microstructured optical
fibers) [9] has shown that fiber-based χ(3) nonlinear laser sources, such as the
supercontinuum [10], fiber OPO [11], or frequency comb [12], can also fulfil this
task. What is more, fiber lasers show significant advantages over other solid state
lasers as an effective approach to provide economic, compact, and flexible optical
components. Also excellent beam quality can be obtained from a single mode fiber.
2–5 μm mid-IR nonlinear lasers require a nonlinear glass with high transparence
in 2–5 μm to be the fiber host material. The position of the IR absorption edge, i.e.,
the infrared longwave transmission limit, of an optical glass is intrinsically limited
by the multiphonon absorption edge of the glass. This can be simply explained by
Hooke’s law using the two-mass spring model [13]. In general, non-silica glasses,
such as tellurite (TeO2 based), fluoride (typically ZrF4 or AlF3 based), and chalco-
genide (chalcogen S, Se, Te based) glasses [14–16], possess excellent optical
transparence in the wavelength range of 0.5–5 μm, 0.4–6 μm, and 1–16 μm,
respectively, and thus are attractive candidates as fiber materials for mid-infrared
applications over the conventional silica. The latter shows inferior transparence
beyond 2 μm, due to (1) the strong fundamental vibration hydroxyl absorption at
2.7 μm and (2) high loss (>50 dB/m) starting from 3 μm due to the tail of the
multiphonon absorption of Si-O network.
For realizing such a fiber-based compact 2–5 μm mid-IR nonlinear laser, the
conventional high-power or high-energy continuous wave (CW) or pulsed erbium,
thulium, or holmium-doped fiber lasers with the lasing wavelength at 1.5 μm or
2 μm are ideal as the pump source. This requires the nonlinear fiber having a zero-
dispersion wavelength (ZDW) at 1.5 or 2 μm in order to maximize the efficiency of
nonlinear frequency conversion.
Tellurite glasses have been considered as one of the promising candidates of the
fiber materials for mid-infrared applications since 1980 [14]. In the very first work
using tellurite fibers for mid-IR applications, it was found that in the range of
2–5 μm, the selected tellurite glass unclad fiber on the composition of 70TeO2-
20ZnO-10BaO (mol.%) showed reasonably low losses up to 40 dB/m, which was
not useful for long-haul applications at that time but still attractive enough for short-
haul applications [14].
Figure 9.1 illustrates the statistical relation between the linear refractive index
n and nonlinear refractive index n2 of various optical glasses.
Table 9.1 compares the key physical and chemical properties of various optical
glasses, including silica, tellurite, fluoride, and chalcogenide glasses, as the candi-
dates for mid-IR nonlinear applications.
Technically, tellurite glasses can be regarded as a high-index, highly nonlinear
version of fluoride glasses. Both tellurite and fluoride glasses possess (1) steep
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 215

Fig. 9.1 Relation between the linear refractive index n and nonlinear refractive index n2 in
various optical glasses

Table 9.1 Comparison of various optical glasses as host materials for mid-infrared nonlinear
fibers [14–24]
Silica Tellurite Fluoride Chalcogenide
Refractive index n at 1.55 μm 1.46 2–2.2 ~1.5 2.3–3
Nonlinear refractive index n2 2.5 20–50 2–3 100–1000
( 10
20 m2/W)
Raman gain coefficient gR at 0.93 32 1.1  0.3 280–720
1.064 μm ( 10
11 cm/W)
Raman shift (cm
1) 440 650–750 550 250–350
λ0, zero-dispersion wavelength of ~1.3 ~2 ~1.7 >5
material (μm)
IR longwave transmission limit Up to 3 μm 6–7 μm 7–8 μm 12–16 μm
Reported lowest loss dB/m 0.15  10
3 0.0204 0.45  10
3 0.023
(wavelength) (1.55 μm) (1.56 μm) (2.35 μm) (2.3 μm)
Thermal stability for fiber Excellent Good Poor Good
drawing
Viscosity around fiber-drawing Flat Steep Steep Flat
temperature
Durability in environment Excellent Good Poor, Good
hygroscopic
Toxicity Safe Safe Relatively Relatively
high high
216 X. Feng et al.

viscosity curve around the fiber-drawing temperature [17], (2) the zero-dispersion
wavelength of the bulk material of ~2 μm [18], and (3) longwave transmission limit
(6–7 μm for the bulk). As shown in Fig. 9.1, tellurite glasses have high refractive
index n (2.0–2.2) and nonlinear refractive index n2 (20–50  10
20 m2/W), while
fluoride glasses typically have n of ~1.5 and n2 of ~2  10
20 m2/W, as low as silica
glass.
Another rival for 2–5 μm mid-IR nonlinear applications is chalcogenide glasses,
which is a glass family based on chalcogen elements (S, Se, and Te). Chalcogenide
glasses possess high n2 of 100–1000  10
20 m2/W, which is two to three orders
higher than that of tellurite glasses, and show excellent IR transmission up to
16 μm, superior to tellurite and fluoride glasses. These outstanding performances
make chalcogenide glasses a very promising host material as χ(3)-based highly
nonlinear fiber media. However, the material zero-dispersion wavelength of chal-
cogenide glasses is beyond 5 μm.
In order to make the zero-dispersion wavelength of a chalcogenide glass fiber
close to 1.5 or 2 μm, very large waveguide dispersions need to be introduced,
requiring the final fiber core diameter to be submicron. This is disadvantageous for
power scaling because such a small-core fiber suffers from low damage threshold
power. Hence, tellurite glasses are an ideal host material as a fiber nonlinear
medium for 2–5 μm wavelength range.
In addition, as shown in Table 9.1, tellurite glass is chemically durable in the
environment for usage and less toxic due to its oxide nature, compared with other
two families of nonoxide glasses.
In principle, broadband mid-IR supercontinuum or other nonlinear processes
such as four-wave-mixing (FWM)-based nonlinear parametric generation can be
realized using a single-mode large-mode-area (LMA) tellurite fiber pumped with a
high-power CW or pulsed 2 μm thulium (Tm3+)- or holmium (Ho3+)-doped fiber
laser [19]. But, due to the strong water absorption (~1400 dB/m or ~4 cm
1)
peaking at 3.3 μm, this makes even an 8-mm-long “wet” tellurite glass fiber suffer
a 15–20 dB loss in light intensity around 3–4 μm and leads into very poor nonlinear
conversion efficiency in 3–5 μm region [20]. Since the tail of multiphonon absorp-
tion of tellurite glasses starts from ~5 μm, dehydration is the key approach to extend
the high transmission window (i.e., absorption coefficient < 0.1 cm
1) of a tellurite
glass fiber into the range of 3–5 μm.
In this chapter, we first introduce various fabrication approaches for making
preforms for solid core/cladding fiber and microstructured photonic crystal fibers,
the selection of the glass compositions for the host materials of the fibers, and the
dehydration of the tellurite glass for having low loss in the mid-IR region. From the
early works on using small-core, dispersion-tailored tellurite fibers for generating
mid-IR supercontinuum, the power scaling and the huge water-induced impurity
attenuation at 3–4 μm band are found to be the two major technical difficulties. To
overcome these two problems, further works on using (1) dehydrated small-core
and (2) large-mode-area (LMA) tellurite photonic crystal fibers have then been
carried out.
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 217

9.2 Fabrication

Making fiber preform is the most important process to realize the targeted fiber with
decent attenuation.
Casting method has been previously developed for making all-solid fiber with
core/cladding structure [25–29]. On the other hand, for making non-silica glass
microstructured preform for photonic crystal fiber, extrusion, ultrasonic drilling,
and capillary stacking are the typical approaches [30]. Since the ultrasonic drilling
and capillary stacking tend to introduce surface defects and/or water impurity onto
the holes of the fabricated preforms during the processing, here we mainly focus on
the casting method and the extrusion method.

9.2.1 Preform Fabrication

Built-in casting method is a traditional method for making solid core/cladding

preform, which was developed initially for making low-loss fluoride glass fiber
preform in the early 1980s [25]. First, a cladding glass tube is obtained by casting
the cladding glass in a preheated metal mold. The mold is then overturned imme-
diately and the unsolidified melt in the center of the mold flows out. Casting the
core glass into the cladding tube, the core/cladding preform is then obtained. The
great advantage of this method is the good optical quality of the core/clad interface.
Fluoride fibers with low losses of 1 dB/km have been drawn from preforms
prepared by this method [26]. However, long length fibers with a uniform core
diameter along the fiber are difficult to be achieved, because the preform typically
shows tapered core dimensions along the length.
A modified built-in casting technique has then been proposed [27]. A hole is
made in the bottom plate of the mold. The cladding glass melt is first cast into a
preheated metal mold. The core glass melt is then cast onto the cladding glass tube,
before the central part of the cladding glass gets solidified. The mold is then moved
to make the central part of molten glass flowing down through the hole in the
bottom plate. The core glass melt takes place at the central part of the cladding
glass, to form the core/cladding structure (see Fig. 9.2). Generally, the core/clad
diameter ratio is in a narrow range between 0.5 and 0.6 [27].
Rotational casting method is another approach to make core/cladding structured
preform [28]. The cladding tube is prepared by pouring the cladding melt into a
cylindrical mold. The mold is then immediately rotated at a speed ranging above
1000 rpm for a few seconds. Using this technique, the inner diameter of the
cladding tube can be easily controlled by adjusting the volume of glass melt for a
given mold. Fiber drawn from preform prepared by this method gives transmission
losses of about 1 dB/km in fluoride glass fibers [29].
Figure 9.3a shows the photograph of one of our fabricated tellurite glass
multimode fibers from a core/cladding preform made by modified built-in casting
218 X. Feng et al.

Fig. 9.2 Schematic of modified built-in casting method

Fig. 9.3 (a) Optical photograph of a fabricated tellurite glass multimode fiber from a core/
cladding preform made by modified built-in casting method; (b) measured loss spectrum using
cutback method with cutback length of 0.70 m

method. The fiber has a core diameter of 90 μm and an OD of 190 μm. The
numerical aperture (NA) of the fiber is ~0.25. The fiber is based on the composition
of 75TeO2-20ZnO-5Na2O (mol.%). As one can see in Fig. 9.3b, the loss of the
multimode fiber drawn from the preform is measured to be between 0.4 and 1.5 dB/
m in the range of 1.0–2.6 μm. The loss of the fiber increases dramatically beyond
2.7 μm due to the absorption of water impurity inside the glass.
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 219

Fig. 9.4 Schematic plot of making structured preform using glass extrusion method

9.2.2 Fabrication of Structured Preforms for Tellurite

Photonic Crystal Fibers

The extrusion technique is a traditional method for making glass workpiece with
simple or complex geometry [31]. As shown schematically in Fig. 9.4, the solid
glass billet is first loaded inside a metal sleeve. The glass is then heated above the
glass softening temperature, and high pressure is applied onto the glass billet
through a metal ram, forcing the viscous glass flow passing through the channels
inside extrusion die. The glass flow forms a preform with the structure opposite to
the structure on the die. This technique is especially suitable for making glass
workpieces based on those glasses with a short operating range (known as short
glasses) or glasses with high tendency for crystallization [31].
Figure 9.5a–c illustrates optical photographs of some extruded tellurite glass
preforms. The tellurite glass is with the composition of 75TeO2-20ZnO-5Na2O
(mol.%). It is seen that the extrusion method is indeed a powerful tool for making
preforms with simple geometry such as rods and tubes, as well as preforms with
complex structure such as multiple holey structures or air-suspended core (ASC). In
particular, from Fig. 9.5c, one can see that the supporting glass spokes have the
thickness of 30 μm and the length of 1.7 mm, which are obviously very difficult
features to be made by any other known approaches.

9.2.3 Fiber Fabrication

Fiber fabrication is the procedure to reduce the size of preforms from tens of
millimeters down to hundreds of microns. The reduction ratio is the key parameter
that a fabricator must consider in the fiber drawing. If the reduction ratio from the
outer diameter (OD) of the preform to the OD the fiber is much larger than the ratio
between the core diameter on the preform and the core diameter on the fiber, a
second step of the fiber drawing will be necessary. For example, for a preform with
an OD of 12.5 mm and a core diameter of 1.0 mm, one step of fiber drawing is
220 X. Feng et al.

Fig. 9.5 Optical photographs of some extruded tellurite glass preforms. The tellurite glass has
composition 75TeO2-20ZnO-5Na2O (mol.%). (a) Side view of an extruded tube with OD of
10.0 mm and inner diameter (ID) of 2.0 mm. (b) Cross-sectional view of an extruded structured
preform with multiple holes around a solid core. (c) Cross-sectional view of an extruded preform
with a suspended core. The preform has a diameter of 9.7 mm. A suspended core is supported by
three thin and long glass spokes. The spokes have the length of 1.7 mm. The core diameter and the
thickness of the supporting spokes on the preform are 284 μm and 28 μm

enough to obtain a fiber with a core diameter of 10 μm. But if the targeted core
diameter is 1.0 μm, another step of fiber drawing is then required.
Figure 9.6a shows the flow chart of making a small-core tellurite PCF, using the
extrusion method. First, the cylindrical glass billets with a diameter of 29 mm and a
height of ~35 mm are made by conventional melting-casting method. The purity of
the raw chemicals was at least 4 N (99.99%). The preform with a suspended core is
then extruded. The structured preform is then elongated to a small cane. The cane
is then inserted inside an extruded jacket tube with an inner diameter slightly
larger than the outer diameter of the cane. The set of cane and tube is finally
drawn into the fiber. Given the reduction ratio at the last step of fiber drawing is in
the range of 100–200, the final core diameter will be between 1 and 2 μm. The yield
of a fiber draw is typically more than a hundred of meters in length.
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 221

Fig. 9.6 Flow chart of fabricating (a) small-core and (b) large-core tellurite PCF, starting from
extrusion

Figure 9.6b shows the flow chart of making large-core tellurite PCF, using the
extrusion method. After extrusion, the structured preform can be directly drawn
into fiber, since the reduction ratio from the preform to the fiber is less than 100.
222 X. Feng et al.

9.3 Selection of Glass Compositions as Fiber Host

After the pioneering systematical works of J. E. Stanworth [32] on tellurite glasses,

it has been noticed that TeO2-ZnO and TeO2-WO3 are the two glass candidates with
the large glass-forming area among all tellurite glass compositions [33]. Large
glass-forming area indicates that (1) it is easy to find a few glass compositions with
good thermal stability in the forming area and (2) it is relatively easy to find one pair
of glass compositions with acceptable thermal mismatch and good thermal stability
as the core and cladding. This is the reason why, in the later works on tellurite glass
fibers, TeO2-ZnO-Na2O [23], TeO2-ZnO-BaO [14], and TeO2-WO3-La2O3 [34, 35]
are the most preferable compositions.
The thermal stability of a fiber host glass is important for the fabrication, in
particular when multiple steps of thermal processing (i.e., extrusion, caning, and
fiber drawing) are required. Figure 9.7 illustrates the measured differential thermal
analysis (DTA) curve and the measured glass viscosity curve of selected glass
composition of 75TeO2-20ZnO-5Na2O glass (mol.%), in which Tg and Tx are the
temperatures of glass transition and the onset of crystallization, respectively. Since
the usable viscosity ranges for preform extruding and fiber drawing are 109–107.5
poise and 106.5–104 poise, respectively, crystallization will not occur even when the
glass is processed at the upper limit of working temperature, corresponding to the
lower viscosity limit for fiber drawing. This indicates the excellent thermal stability
of this glass for fiber fabrication.
In this chapter, we mainly focus on two tellurite glass compositions, 75TeO2-
20ZnO-5Na2O glass (mol.%) and 70TeO2-20ZnO-10BaO (mol.%) as the host
materials of fibers. Table 9.2 summarizes the major thermal properties of these
two glass compositions. It is seen that both compositions have large difference
between Tg and Tx, indicating excellent thermal stability for glass extrusion and
fiber drawing.

Fig. 9.7 DTA curve and viscosity curve of selected 75TeO2-20ZnO-5Na2O glass (mol.%)
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 223

Table 9.2 Thermal properties of 75TeO2-20ZnO-5Na2O glass (mol.%) and 70TeO2-20ZnO-

10BaO (mol.%)
Glass composition (mol.%) Tg ( C) Tx ( C) ΔT( ¼ Tx
Tg) ( C)
75TeO2-20ZnO-5Na2O 305 465 160
70TeO2-20ZnO-10BaO 320 / 1

Fig. 9.8 Comparison of absorption coefficients of some selected impurities in a conventional

tellurite glass (composition: 75TeO2-20ZnO-5Na2O mol.%) (summarized from results in ref. [36])

The typical impurities in an optical glass are transition metal ions, rare-earth
ions, and hydroxyl groups (the OH groups). Because the purity of the raw chemicals
for melting tellurite glasses here is 4–5 N (99.99–99.999%), the total impurities in
the final glass ought to be at 10–100 ppm.

9.4 Impurities in Tellurite Glasses and Glass Dehydration

Transition metal ions, rare-earth ions, and hydroxyl groups are the three major
categories of impurities having big impact on increasing attenuation of optical
glasses in the wide range from visible wavelengths to mid-infrared region.
Based on the data given in ref. [36], Fig. 9.8 shows the measured transmission
spectra of bulk tellurite glasses (base: 75TeO2-20ZnO-5Na2O mol.%) doped with
the impurities including the representative transition metal ion Ni2+, the represen-
tative rare-earth ion Nd3+, and the hydroxyl group OH
. It is seen that the funda-
mental vibration of the hydroxyl group OH
is the most harmful impurity
for blocking the transmission of a tellurite glass in 2–5 μm region, in terms of
224 X. Feng et al.

the wide absorption range (3–4 μm) and the high peak absorption coefficient
(10 dB/m/ppm).
Water is commonly existing in many optical materials as the hydroxyl group OH
.
Although water is the most important substance on the earth for all the living
creatures, actually it is a troublemaker in optical area. It causes high attenuation
in optical materials from visible to mid-infrared regions. As one of the most
common impurities in optical materials, hydroxyl groups OH
are very hard to
be eliminated.
The absorption spectrum of the OH fundamental vibration is located around
3 μm and varies with the glass composition. For pure silica glass, the fundamental
vibration of OH band is located at 2.70 ( 0.02) μm and with an extinction
coefficient of 10 dB/m/ppm (in weight) [37], while in a ZrF4-based fluoride glass,
it is located at 2.87 μm with an extinction coefficient of 5 dB/m/ppm (in weight)
[38]. For a tellurite glass, the fundamental vibration of OH band has an extinction
coefficient of ~1 dB/m/ppm (in mole) or ~10 dB/m/ppm (in weight) [39] and ranges
from 3 to 4 μm, much wider than the above two glasses. This is because the
wavelength of the fundamental OH vibration reflects the strength of OH bonding
with the surrounding molecules. The large variety of deformed [TeO4] and [TeO3]
units in tellurite glass network causes very different sites of OH groups, and
consequently the bond strength of OH groups shows much larger variation
[39]. This is very different from the situation of pure silica glass, in which the
[SiO4] tetrahedra units are very much identical in the short range leading to the
fundamental OH absorption as narrow as ~150 nm (~200 cm
1) [37].
Because the nonlinearity n2 of tellurite glasses is one order of magnitude higher
than those of silica and fluoride glasses, less than a meter length of tellurite
nonlinear fiber is acceptable for many nonlinear optical applications, indicating
that an acceptable OH peak absorption loss is no more than the level of 10 dB/m. In
other words, the OH impurity in a usable tellurite glass fiber should be no more than
the level of 1 ppm in weight. It must be pointed out that, without any dehydration
process, the tellurite glass, which was prepared in the open atmosphere, has a high
water absorption coefficient of ~1400 dB/m peaking at 3.4 μm [39]. Therefore, even
an 8-mm-long non-dehydrated tellurite glass fiber will suffer a 15–20 dB loss in the
range of 3–4 μm [40]. Since the tail of multiphonon absorption of tellurite glasses
starts from ~5 μm, a proper dehydration process is the key to extend the high
transmission window (i.e., absorption coefficient < 0.1 cm
1) of a tellurite glass
fiber to 5 μm.
Reactive atmosphere processing (RAP) has been proved to be an effective
approach for dehydrating fluoride and silica glasses [41, 42]. In the case of
dehydrating fluoride glasses, during the melting of fluoride glass, dry reactive
halogen gas (such as NF3, HF, SF6, and CCl4) is purged into the melt to react
with the OH groups inside the glass [41]. The hydroxyl groups bonded with the
glass network via hydrogen bonding are converted into the volatile compounds HF
or HCl, which can be removed from the melt naturally at high temperature.
Chlorine drying is also a common method in the fabrication of silica glass preforms
to reduce the OH peaks at 1.38 μm (the first overtone) [41]. For dehydrating tellurite
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 225

glasses, in order to avoid using highly toxic and reactive halogen gases, halogen-
containing solid compounds are preferable. Previously, fluorotellurite glass,
obtained by introducing fluorine into the oxide tellurite glass matrix, has been
proposed to remove the OH groups with the assistance of fluorine, and also to
extend the mid-IR transmission [43]. However, the oxyfluoride glass has strong
tendency to be crystallized easily during the heating process and hence is not a
thermally stable host material for fiber drawing. Also the introduction of fluorides
causes the glass hygroscopic. In addition, introducing fluorides into tellurite glasses
leads into significant decrease of both the refractive index and the nonlinear
refractive index [43].
Solid-state chloride compounds, such as NaCl and BaCl2, were then added for
dehydrating tellurite glasses [44]. Pyrohydrolysis reaction, OH
+ Cl
¼ HCl" + O2
,
occurs during the glass melting for reducing the OH content in the glass. The
compositions of the studied halo-tellurite glasses include 75TeO2-20ZnO-5Na2O
(TZN), 75TeO2-20ZnO-4Na2O-2NaCl (TZNX1), 70TeO2-20ZnO-10BaO (TZB),
70TeO2-20ZnO-9BaO-1BaCl2 (TZBX1), and 70TeO2-20ZnO-10BaCl2 (TZBX10)
(mol.%). Commercial chemicals with purity better than 99.999% were used as raw
materials for the glass melting. For each composition, a batch of 70 g was mixed
well and then melted in a gold crucible inside a furnace in the glove box at
700–900  C for 2 h. Increasing the melting time in dry atmosphere was found
helpful to reduce the OH content in the glass. But it also caused the deviation of the
final glass composition from the targeted one when the melting time is longer than
3 h, since significant evaporation could be observed from tellurite glass melt. The
melt was then cast in a preheated metal mold, and a bubble-free glass rod with an
outer diameter (OD) of 16 mm and length between 45 and 65 mm was obtained.
Each rod was drawn into an unclad fiber with an OD of 180 μm, with the yield
>50 m. A bulk sample was then obtained from the preform remainder with two
parallel ends polished.
The transmission spectrum of each bulk was measured by a Varian 670-IR
FT-IR spectrometer. The absorption coefficient (in the unit of cm
1) of the residual
OH impurity in the bulk glass was calculated according to the Beer-Lambert law
and converted to the bulk attenuation in dB/m. The losses of the unclad fibers were
measured by the cutback method, using a tunable 1064 nm nanosecond laser
pumped periodically with lithium niobate (MgO:PPLN) (Covesion Ltd) as the
laser source [44, 45]. Essentially, the unclad fiber drawn from a dehydrated
TZNX1 glass (i.e., using 2 mol.% of NaCl to replace 1 mol.% of Na2O in TZN
glass) rod showed a largely reduced loss of 45 dB/m at 3.3 μm [44, 45], compared
with the loss of ~1400 dB/m at 3.3 μm in a TZN glass melted in the open
atmosphere [39] and the loss of 100 dB/m of a TZN glass melted in a dry
atmosphere-filled glove box [46].
As shown in Table 9.2 in this chapter, the TZB glass shows better thermal
stability than the TZN glass, indicating it also a good fiber host material. Figure 9.9
shows our latest improvement on measured mid-IR losses from dehydrated TZBX
glass fibers. It is seen that the losses of the dehydrated TZBX10 unclad fiber at
3.3–3.4 μm are reduced down to 5 dB/m, about one order of magnitude lower than
226 X. Feng et al.

Fig. 9.9 Losses of dehydrated TZBX glass and fibers in mid-IR region

those of the dehydrated TZBX1 unclad fiber at the same wavelengths. The OH
impurity level here is estimated to be 0.5 wt ppm. In a TZBX10-based multimode
fiber with a core diameter 60 of μm and an NA of 0.14, the losses in the same
wavelength region are still below 10 dB/m. The interface imperfections between
the core and the cladding in the multimode fiber are responsible for the loss increase
above the unclad fiber. Further improvement on the preform fabrication should be
able to solve this problem. It is seen that the loss curves of the unclad TZBX10 fiber
and TZBX10 bulk show the good agreement beyond 4.2 μm, where the
multiphonon absorption edge starts playing the dominant role on the loss enhance-
ment. Between 4.35 and 4.65 μm, the loss of a TZBX10 glass fiber is between
50 and 100 dB/m. Therefore, this region should be a longwave limit for generating
10 dB bandwidth supercontinuum from a TZBX10 nonlinear fiber with a length
between 10 and 20 cm.

9.5 Experimental Results on Mid-infrared Generation

in Tellurite Glass Fibers

9.5.1 Early Work on Small-Core Tellurite Glass Fiber

for Mid-infrared Nonlinear Optics

With the rapid development of silica photonic crystal fibers (PCF), also called
microstructured optical fibers, broadband χ(3)-based Kerr nonlinear optical pro-
cesses, such as supercontinuum generation [10, 47] and optical parametric
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 227

Fig. 9.10 (a) SEM photos of fiber cross-sectional structure of the ASC fiber. The arrow shows the
direction of the fast axis of the fiber. (b) Dispersion curves of the ASC fiber and the bulk material.
(c) Output supercontinuum spectra with 30 mW output power generated from a 22-cm length of
this fiber, pumped at 1.56 μm wavelength on the fast and slow polarization axes, respectively. The
vertical dashed line at 1.46 μm indicates the ZDW of the fiber

generation [48], have been experimentally realized using highly nonlinear,

dispersion-tailored silica PCFs. In order to extend these nonlinear optical processes
into wavelength ranges that silica fiber cannot cover effectively, that is, the ultra-
violet (UV, <400 nm) and mid-infrared regimes (>2 μm), non-silica glass highly
nonlinear, dispersion-tailored PCFs have been considered [49, 50].
Although a tellurite PCF with a mode area of 21.2 μm2 has been reported as the
world’s first tellurite PCF in 2003 [50], the first work on using tellurite PCFs for
mid-infrared nonlinear optics appeared in 2006 [51], using a single-mode, highly
nonlinear tellurite PCF with an effective mode area of 2.6 μm2 [52]. Figure 9.10a
shows scanning electron microscope (SEM) images of the cross section of this
small-core tellurite PCF. A triangular core with a diameter of ~2.6 μm is isolated
from the outer jacket by three 7.4  0.2-μm-long and 180  20-nm-thin struts.
228 X. Feng et al.

Since the three thin supporting struts have a width that is much smaller and a length
that is much larger than the wavelength of light, the core can be considered
essentially as an air-suspended high-index glass rod.
The calculated group velocity dispersion curve of such an air-suspended-core
(ASC) fiber and the material dispersion of the base glass (75TeO2-20ZnO-5Na2O
mol.%) are illustrated in Fig. 9.10b. Due to the strong waveguide dispersion
introduced by the small-core and the high-index contrast between the glass and
the air surrounding the glass core, the zero-dispersion wavelength (ZDW) of the
ASC fiber is located at 1.46 μm and thus is shifted considerably toward shorter
wavelengths compared to the ZDW of 2.15 μm in the bulk glass.
Experimental observations as well as corresponding numerical simulations indi-
cated that the fiber was effectively single-moded beyond 1 μm wavelength [52]. A
small effective mode area Aeff of 2.6 μm2 was calculated from the simulation result.
Using the Boskovic method [53, 54], the fiber nonlinearity γ ¼ 2πn2/(λAeff), where
n2 is the nonlinear refractive index of the fiber core material, λ the operation
wavelength, and Aeff the effective mode area of the fiber core, respectively, was
measured to be 280  30 W
1 km
1 at 1.55 μm, which is ~280 times higher than
that of standard single mode silica optical fiber (Corning SMF28). Using the
cutback method, the fiber attenuation was measured to be 2.0 dB/m at 1.55 μm.
Using a 1.56 μm femtosecond laser generated from a tunable optical parametric
oscillator (OPO) as the pump source, broadband supercontinuum was observed
from the tellurite ASC fiber [51]. The pulse duration of the OPO output was
measured to be 200 fs. The average output power from the OPO was ~160 mW,
corresponding to a peak power of ~11.7 kW at 1.56 μm or equivalently to 2.1 nJ
energy per pulse, with a spectral width (full width at half maximum (FWHM)) of
~15 nm. The pump wavelength was thus chosen in the anomalous dispersion region
of the fiber but close to the ZDW, which is the standard arrangement for broadband
supercontinuum generation [55]. As the laser pulses propagate along the fiber, they
undergo sudden spectral broadening by soliton fission, followed by further gener-
ation of long wavelength components via Raman-induced soliton self-frequency
shifting. At the same time, these solitons also create phase-matched dispersive
waves at short wavelengths below the ZDW. Figure 9.10c illustrates the observed
broadband supercontinuum with an output power of 30 mW generated in this
experiment with a tellurite ASC fiber of 22-cm length. Slightly different spectra
were obtained when launching the pump light with different polarizations into fiber,
in particular with polarizations parallel and perpendicular to the fast axis of the fiber
(as shown in Fig. 9.10a), suggesting a weak birefringence arising from the trian-
gular fiber core. All in all, broadband supercontinuum from 0.9 to 2.5 μm was
observed, and it was the first report of generating mid-IR supercontinuum from a
non-silica glass fiber.
In 2008, Domachuk et al. [40] used a similar ASC tellurite fiber and obtained a
record of 4000 nm bandwidth supercontinuum. This fiber was made of 75TeO2-
12ZnO-5PbO-3PbF2-5Nb2O5 and had a hexagonal air-suspended core with a diam-
eter of 2.5 μm and a ZDW of 1380 nm. An OPO generating pulses at 1550 nm
wavelength with 110 fs width at a repetition rate of 80 MHz was used to launch
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 229

150 mW of power at the fiber input. The supercontinuum measured at
45 dBm
extended from 789 nm to 4870 nm, demonstrating the significant potential of
tellurite glass fibers for applications all across the atmospheric transmission win-
dow up to 5 μm wavelength. However, because of the large water-induced absorp-
tion in the tellurite material of this fiber, a very short length of 8 mm of fiber was
employed in the experiment, making the experimental alignment difficult while still
leading to a significant slope toward lower power spectral density at the long
wavelength edge.

9.5.2 Dehydrated Small-Core Tellurite Glass Fiber

In order to improve on these early results on mid-infrared supercontinuum gener-

ation in tellurite fibers, two modifications to the fiber should be considered.
(1) Using dehydrated glass, as described in Sect. 9.4 of this chapter, will signifi-
cantly reduce absorption beyond 3 μm wavelength and will thus allow for the use of
longer lengths of fiber and the generation of supercontinuum spectra with flatter
power spectral densities. (2) The near-symmetric cores of the fibers employed in the
experiments described above limit the range of dispersion profiles that can be
achieved. In particular, while the ZDW can be shifted continuously toward shorter
wavelengths, it is generally not possible to also achieve a flat dispersion curve over
a broad wavelength range, that is, a low dispersion slope, which would further
benefit efficient supercontinuum generation.
To address these issues, a dehydrated TZBX1 glass-based PCF (70TeO2-20ZnO-
9BaO-BaCl2 mol.%) with an elliptical air-suspended core was fabricated [56]. A
preform with one air-suspended core (see Fig. 9.11a) was made by extruding a

Fig. 9.11 (a) Photograph of extruded dehydrated TZBX1 preform and (b) SEM image of
elongated cane for fiber drawing
230 X. Feng et al.

Fig. 9.12 (a) and (b) SEM images of the cross section of fabricated TZBX1 elliptical ASC fiber.
(c) Corresponding calculated dispersion profiles

dehydrated TZBX1 billet. A cane with a long axis diameter of 1.62 mm (see
Fig. 9.11b) was then elongated from this structured preform. Next, the cane was
inserted into an extruded jacket tube, and this ensemble was then drawn into an
ASC fiber with an elliptical core (see Fig. 9.12a, b). The elliptical core has a height
(h) of 1.8 μm and a width (w) of 7.8 μm. The core exhibits strong birefringence, as
shown by the calculated dispersion curves of the fundamental TE and TM polarized
modes in Fig. 9.12c. The ZDW of the fundamental TE mode is located at 1.48 μm.
Comparing the dispersion profile of this mode with that of Fig. 9.10b for a
symmetric-core ASC fiber, one notices that lower dispersion values are achieved
over a broad bandwidth with this elliptical core fiber.
Supercontinuum generation with 90 fs, 1500 nm wavelength pump pulses of up
to 1.12 nJ energy, similar to above, was investigated experimentally [56]. With
15 cm of fiber spectral broadening to ~3 μm was observed, but further broadening
was limited by the available pump power.
With the same fiber, nonlinear broadening with a mid-infrared pump at 2.4 μm
was also demonstrated. A linearly polarized OPO with a pulse duration of ~90 fs, a
repetition rate of 80 MHz, and a maximum average power of 290 mW at 2.4 μm was
used as the pump source. The pump pulses were launched into a 3-cm-long TZBX1
ASC fiber. For this fiber, the pump is therefore located deep inside the anomalous
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 231

Fig. 9.13 Measured mid-infrared soliton dynamics in a 3-cm length of the ASC fiber of Fig. 9.12
pumped at 2.4 μm

dispersion region, resulting in the generation of low-order solitons by the pump

pulses, which are then moving to longer wavelengths by Raman-induced self-
frequency shifting, as can be seen clearly in Fig. 9.13. The maximum launched
average power was 48.6 mW, which corresponds to a peak power of 6.8 kW and a
pulse energy of 0.61 nJ. Even at these low, sub-nJ pulse energies, spectral broad-
ening beyond 3.0 μm is readily observed with this dehydrated tellurite fiber.
Further broadening of the mid-IR supercontinuum can be expected in fibers
fabricated from the latest low-loss dehydrated TZBX10 glass, which will reduce
losses in the 3–4 μm wavelength region by another factor of 10; see Sect. 9.4 above.
Moreover, the dispersion profile can be further optimized using the design degrees
of freedom offered by the elliptical core ASC geometry [56]. Work in this area is
still ongoing at the time of writing.

9.5.3 Power Limitations in Small-Core Tellurite Glass Fibers

Power scaling is a highly desirable performance characteristic of a fiber laser, and

for many applications, the achievable output power of the generated
supercontinuum is more important than, for example, the spectral bandwidth.
However, in all the small-core dispersion-tailored tellurite glass fiber experiments
discussed above, the demonstrated mid-IR supercontinuum output power remained
below 100 mW. This was partially limited by the attenuation located at 3–4 μm due
to the water impurity of the material, but more importantly by the small size of the
232 X. Feng et al.

Fig. 9.14 Optical microscope images of the input facet of the tellurite ASC fiber before (left) and
after (right) the damage caused by exceeding the laser power damage threshold

fiber core which leads to a low damage threshold for the fiber facet and thus limits
the maximum pulse peak power and energy that can be launched into the fiber
without destroying it.
For the tellurite ASC fiber discussed above in Sect. 9.5.1, the damage threshold
was established using a 1.06 μm Yb-doped all-fiber master oscillator power ampli-
fier (MOPA) with pulse duration of 20 ps and repetition rate of 14.4 MHz. Damage
of the ASC fiber facet was observed under a launched average power of 140 mW,
corresponding to a peak power of 390 W [57]. Figure 9.14 shows optical micro-
scope images of the tellurite ASC fiber before and after the facet was damaged. It is
seen that the fiber core disappeared after the launched power exceeded the thresh-
old, while the three supporting glass spokes still survived. The damage intensity is
estimated to be 18.7 GW/cm2. This is consistent with the reported surface damage
threshold of tellurite bulk glasses, 15–20 GW/cm2 using a Nd:YAG laser at
1.064 μm with a similar pulse duration of 25 ps [58]. As a comparison, under
similar conditions, the surface damage threshold of bulk silica glass is above
300 GW/cm2 [59]. In order to significantly increase the achievable mid-IR
supercontinuum power, tellurite fibers with much larger core sizes must therefore
be employed.

9.5.4 Supercontinuum Generation in Large-Mode-Area

Tellurite Glass Fiber

Power scaling of common fiber lasers, such as rare-earth-doped fiber lasers or

Raman fiber lasers, is routinely realized by using single-mode large-mode-area
(LMA) fibers. The same concept also provides a feasible solution for achieving
power scaling of broadband supercontinuum laser sources. Because the laser
damage thresholds of non-silica glasses, such as tellurite glasses, fluoride glasses,
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 233

Fig. 9.15 (a) SEM image of the cross section of the tellurite LMA PCF; (b) simulated mode
profile of the fundamental mode (LP01) at 1.55 μm; (c) the observed near-field mode profile at
1.55 μm from the fabricated tellurite LMA PCF. All three plots are on the same scale

and chalcogenide glasses, are much lower than that of silica, it is especially
important to develop non-silica glass LMA fibers for achieving medium- or high-
level output power [60].
Figure 9.5b already showed a tellurite preform with three rings of holes sur-
rounding a solid core that was made by extrusion. The preform has an outer
diameter (OD) of 15.9 mm, a hole diameter of 1.0 mm, and hole separation of
2.2 mm. The preform was directly drawn into fiber with 410  10 μm
OD. Figure 9.15a shows an SEM image of the cross section of this tellurite fiber
with ~80 μm core diameter. The fiber has a uniform hole spacing Λ of 53 μm. Due to
effects such as temperature gradients in the radial direction, surface tension, and
residual air pressure inside the holes during the fiber-drawing process, the average
hole diameter di (where i is the ring number counted from the core outward, i ¼ 1 to
3) was reduced from d1 ¼ 28.1 μm for the inner ring to d2 ¼ 23.3 μm for the middle
ring and d3 ¼ 13.8 μm for the outer ring of holes, corresponding to di/Λ ratios of
0.53, 0.44, and 0.26, respectively. Note that for all three rings the di/Λ ratios
deviated from the initial value of 0.454 within the preform.
It is well known from theoretical investigations that a PCF supports only the
fundamental fiber mode at all wavelengths if it has an infinite cladding containing
an infinite number of rings of triangularly arranged holes with a ratio of hole
diameter to hole spacing, d/Λ, of less than 0.40–0.45. This is often referred to as
endlessly single-mode behavior [61], one of the unique optical properties of PCFs.
For a practical PCF with a finite holey cladding and a finite number of air-filled
holes, higher-order modes are more likely to be observed, especially in short fiber
lengths [62]. Increasing the discrimination between the confinement losses (CLs) of
the fundamental mode LP01 and the first higher-order mode LP11 becomes the
practical criterion for realizing single-mode operation in a PCF.
That the above tellurite LMA PCF was effectively single-moded was first
confirmed by numerical simulations. The simulations predicted that the fundamen-
tal mode (LP01) (as shown in Fig. 9.15b) and the first higher-order mode (LP11)
234 X. Feng et al.

have a CL of ~10
4 dB/m and 3 dB/m at 1.55 μm, respectively. The ratio of the
confinement losses of the fundamental mode to the first higher-order mode at
1.55 μm is ~10
5, indicating that only the fundamental mode LP01 can be observed
at 1.55 μm if the fiber is sufficiently long. Further numerical simulations of this
LMA PCF show that the ratio of the CL of the LP01 mode to that of the LP11 mode is
~10
4 in the whole range of 1–3 μm, while the CL of the LP11 mode is more than
1 dB/m. This indicates that this LMA PCF has a broad bandwidth of more than one
octave for single-mode operation, while conventional single-mode optical fibers
only operate over one octave of bandwidth [60], namely, at normalized frequencies
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
V ranging between 1.2 and 2.4 where V ¼ 2πa n2core
n2clad =λ, where a is the fiber
core radius, λ the wavelength of light, and ncore and nclad the core and cladding
refractive index, respectively.
Robust guidance with a hexagonally symmetric mode profile was observed at
1.55 μm (see Fig. 9.15c) in a 1.50-m-long tellurite PCF. During this measurement,
the fiber was kept effectively straight, i.e., with an estimated bending radius of
~2 m, to minimize the influence of bending loss on the fundamental mode. The
observed mode profile was in good agreement with the simulated fundamental
(LP01) mode (see Fig. 9.15b). The beam quality factor M2 was measured as
1.3  0.1 at 1.55 μm. Both experimental and simulation results indicated the
effective single-mode operation of the fiber [60].
From the simulations, an effective mode area Aeff of 3000  200 μm2 was
calculated for the fundamental mode [60]. This is the largest mode area reported
so far in a single-mode non-silica glass fiber and is comparable to the effective
mode area in silica glass LMA PCFs, for example, the 3160 μm2 reported in ref.
[63]. The propagation attenuation of this tellurite LMA PCF was measured as
2.9  0.1 dB/m at 1.55 μm using the cutback method. The effective fiber
nonlinearity γ of the LMA PCF was calculated to be 0.23 W
1 km
1 at 1.55 μm.
Because of the large-core size, the dispersion profile is effectively the same as its
material dispersion, as illustrated in Fig. 9.10b, with a zero-dispersion wavelength
and a relatively flat dispersion slope at 2.15 μm.
Experimental work on generating broadband mid-IR supercontinuum from the
tellurite LMA PCF was carried out using a tunable femtosecond pulsed optical
parametric amplifier (OPA) with a repetition rate of 1 kHz and a pulse duration of
~120 fs [60]. The idler beam from the OPA at 2.15 μm was launched into a straight
piece of the fiber of 9-cm length using a microscope objective with an NA of 0.1.
This pump wavelength was chosen because it coincides with the fiber ZDW.
Supercontinuum generation was then investigated at average pump powers of
2.4–15.2 mW, measured before the launch into the fiber. The measured output
spectra showed significant broadening toward shorter wavelengths down to about
1 μm. Unfortunately this experiment did not record the spectrum above 2.5 μm
because of the long wavelength limit of the used detector, a cooled extended
InGaAs photodiode, and thus the full extent of the supercontinuum into the
mid-IR region is unknown. However, the supercontinuum output power was
measured at the end of the LMA fiber of 6.0  0.2 mW for an incident pump
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 235

power of 15.2 mW, showing high conversion efficiency from the pump to the
supercontinuum [60].
Most importantly, while the average power used in this experiment was rela-
tively modest, the low laser repetition rate compared to the experiments discussed
above in the small-core fibers (1 kHz here compared to 80 MHz) means that the
individual pulses launched into the LMA fiber had peak powers and pulse energies
several orders of magnitude higher than what could be used in the small-core ASC
fibers because of their low damage threshold. In particular, at the maximum
incident pump average power of 15.2 mW, the pulse energy reached 15.2 μJ, and
the peak power was 127 MW. No facet damage of the tellurite LMA fiber was
observed during these experiments, demonstrating the potential for power scaling
of mid-IR supercontinuum generation in tellurite fibers by moving toward large-
core low-NA fiber structures.

9.6 Outlook

Table 9.3 summarizes the results that we have discussed in this chapter on using
tellurite glass nonlinear PCFs for generating mid-IR supercontinuum.
Supercontinuum generation in a small-core tellurite fiber was first demonstrated
in 2006 [51]. Spectral broadening up to nearly 5 μm has been observed in a similar
fiber structure [40], but only impractically short fiber lengths could be used because

Table 9.3 Summary of using tellurite PCFs for generating mid-IR supercontinuum. The long
wavelength cutoff in the experiments marked with + was not properly resolved because of detector
limitations
Water-
induced Spectral
Effective attenuation span
mode at water Fiber Launched Launched (μm)
area peak length pulse peak (20 dB
Fiber (μm2) (at 3.3 μm) (cm) energy (J) power (W) width) References
ASC 2.6 ~1400 22 2.1  10
9 11.7  103 0.9–2.5+ [51]
PCF with
triangular
core
ASC PCF 1.7 ~1400 0.8 1  10
9 8.5  103 0.79–4.87 [40]
with
hexagonal
core
ASC PCF 3.9 ~70 3 0.61  10
9 6.8  103 2.2–3.1 [56]
with
elliptical
core
LMA PCF 3000 ~1400 9 15.2  10
6 127  106 1.0–2.5+ [60]
236 X. Feng et al.

of water absorption. A small-core fiber made from dehydrated tellurite glass has
also been demonstrated [56] with a novel elliptical core geometry allowing for
better dispersion control. In all these small-core fibers, the maximum pulse energy
and peak power were limited by the fiber damage threshold to about 1 nJ and
10 kW, respectively. A large-mode-area tellurite fiber [60] was demonstrated that
could support three to four orders of magnitude higher powers.
Based on these results, it is expected that using 0.1–1 m lengths of dehydrated
LMA tellurite fibers with a water-induced loss of less than 10 dB/m at 3.3 μm will
allow for the generation of flat 2–5 μm supercontinuum spectra with peak output
powers of 100 MW.

Acknowledgments The authors thank Prof. P. Petropoulos, Prof. D. J. Richardson, Mr. K. E.

Frampton, Mr. N. M. White, Dr. M. Segura, Dr. J. Shi, and Dr. W. Loh of the Optoelectronics
Research Centre, University of Southampton, for their contributions.
Xian Feng also thanks Prof. Setsuhisa Tanabe, Dr. Jau-Sheng (Jimmy) Wang, and Prof. Elias
Snitzer for fruitful discussions in the early stages of developing tellurite glass fibers.

References

1. A. Schliesser, N. Picqué, T.W. Hänsch, Mid-infrared frequency combs. Nat Photonics 6,

440–449 (2012)
2. E. Sorokin, I.T. Sorokina, M.S. Mirov, V.V. Fedorov, I.S. Moskalev, S.B. Mirov Ultrabroad
continuous-wave tuning of ceramic Cr:ZnSe and Cr:ZnS lasers. Paper presented at the Adv.
Solid-State Photon., San Diego, CA, (2010)
3. IPG Photonics, TM:ZnSe/S SERIES. http://www.ipgphotonics.com/Collateral/Documents/
English-US/TM_ZnSE_S_Series_IPG_datasheet.pdf. Accessed 29 June 2015
4. M. Razeghi, N. Bandyopadhyay, Y. Bai, Q. Lu, S. Slivken, Recent advances in mid infrared
(3-5 μm) Quantum Cascade Lasers. Opt Mater Express 3, 1872–1884 (2013)
5. N. Bandyopadhyay, Y. Bai, S. Tsao, S. Nida, S. Slivken, M. Razeghi, Room temperature
continuous wave operation of λ ~ 3-3.2 μm quantum cascade lasers. Appl. Phys. Lett. 101,
241110 (2012)
6. X. Zhu, N. Peyghambarian, High-power ZBLAN glass fiber lasers: Review and prospect. Adv
Optoelectron 2010, 501956 (2010)
7. S.D. Jackson, Towards high-power mid-infrared emission from a fibre laser. Nat Photonics 6,
423–431 (2012)
8. M. Ebrahimzadeh, in Topics in applied physics, volume 89, solid-state mid-infrared laser
sources, ed by I. T. Sorokina, K. L. Vodopyanov. Mid-infrared ultrafast and continuous-wave
optical parametric oscillators, solid-state mid-infrared laser sources (Springer-Verlag, Berlin
Heidelberg, 2003), pp. 184–224
9. P. Russell, J. St, Photonic crystal fibers. Science 299, 358–362 (2003)
10. J.K. Ranka, R.S. Windeler, A.J. Stentz, Visible continuum generation in air-silica microstruc-
ture optical fibers with anomalous dispersion at 800 nm. Opt. Lett. 25, 25–27 (2000)
11. D.K. Serkland, P. Kumar, Tunable fiber-optic parametric oscillator. Opt. Lett. 24, 92–94
(1999)
12. S.T. Cundiff, J. Ye, Colloquium: Femtosecond optical frequency combs. Rev. Mod. Phys. 75,
325–342 (2003)
13. B. Szigeti, Polarisability and dielectric constant of ionic crystals. Trans. Faraday Soc 45,
155–166 (1949)
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 237

14. J.Y. Boniort, C. Brehm, P.H. DuPont, D. Guignot, C. Le Sergent, Infrared glass optical fibers
for 4 and 10 micron bands. Paper presented at 6th European Conference on Optical Commu-
nication, University of York, UK, 16–19 September 1980 (1980)
15. M. Poulain, M. Poulain, J. Lucas, P. Brun, Verres fluores au tetrafluorure de zirconium
proprietes optiques d’un verre dope au Nd3+. Mater. Res. Bull. 10, 243–246 (1975)
16. N.S. Kapany, R.J. Simms, Recent developments of infrared fiber optics. Infrared Phys 5, 69–75
(1965)
17. A. Mori, H. Masuda, K. Shikano, M. Shimizu, Ultra-wide-band tellurite-based fiber Raman
amplifier. IEEE J Lightwave Technol 21, 1330–1306 (2003)
18. T. Mizunami, H. Iwashita, K. Takagi, Gain saturation characteristics of Raman amplification
in silica and fluoride glass optical fibers. Opt Commun 97, 74–78 (1993)
19. D. Szebesta, S.T. Davey, J.R. Williams, M.W. Moore, OH absorption in the low loss window
of ZBLAN(P) glass fibre. J Non-Cryst Solids 161, 18–22 (1993)
20. J.S. Sanghera, L. Brandon Shaw, I.D. Aggarwal, Chalcogenide glass-fiber-based mid-IR
sources and applications. IEEE J Sel Top Quantum Electron 15, 114–119 (2009)
21. R.E. Slusher, J. Hodelin, J.S. Sanghera, L.B. Shaw, I.D. Aggarwal, Large Raman gain and
nonlinear phase shifts in high-purity As2Se3 chalcogenide fibers. J Opt Soc Am B 21,
1146–1155 (2004)
22. A.V. Vasilev, G.G. Devyatykh, E.M. Dianov, A.N. Guryanov, A.Y. Laptev, A.Y. Laptev,
V.G. Plotnichenko, Two-layer chalcogenide-glass optical fibers with optical losses below
30 dB/km. Quantum Electron 23, 89–90 (1993)
23. J.S. Wang, E.M. Vogel, E. Snitzer, Tellurite glass: A new candidate for fiber devices. Opt.
Mater. 3, 187–203 (1994)
24. G. Ghosh, Sellmeier coefficients and chromatic dispersions for some tellurite glasses. J. Am.
Ceram. Soc. 78, 2828–2830 (1995)
25. S. Mitachi, T. Miyashita, T. Kanamori, Fluoride-glass-cladded optical fibres for mid-infra-red
ray transmission. Electron Lett 17, 591–592 (1981)
26. T. Kanamori, S. Sakaguchi, Preparation of elevated NA fluoride optical fibers. Japan J Appl
Phys 25, L468–L470 (1986)
27. S. Sakaguchi, S. Takahashi, Low-loss fluoride optical fibers for midinfrared optical commu-
nication. J Lightwave Technol 5, 1219–1228 (1987)
28. D.C. Tran, C.F. Fisher, G.H. Sigel, Fluoride glass preforms prepared by a rotational casting
process. Electron Lett 18, 657–658 (1982)
29. G. Lu, I. Aggarwal, Recent advances in fluoride glass fiber optics in the USA. Mater. Sci.
Forum 19-20, 375–380 (1987)
30. X. Feng, A.K. Mairaj, D.W. Hewak, T.M. Monro, Nonsilica glasses for holey fibers. J
Lightwave Technol 23, 2046–2054 (2005)
31. E. Roeder, Extrusion of glass. J Non-Crystall Solids 5, 377–388 (1971)
32. J.E. Stanworth, Tellurite glasses. Nature 169, 581–582 (1952)
33. H. Rawson, Inorganic glass forming systems (Academic Press, London, New York, 1967)
34. A.K. Yakhkind, Tellurite glasses. J. Am. Ceram. Soc. 49, 670–675 (1966)
35. V.V. Dorofeev, A.N. Moiseev, M.F. Churbanov, G.E. Snopatin, A.V. Chilyasov, I.A. Kraev,
A.S. Lobanov, T.V. Kotereva, L.A. Ketkova, A.A. Pushkin, V.V. Gerasimenko,
V.G. Plotnichenko, A.F. Kosolapov, E.M. Dianov, High-purity TeO2–WO3–(La2O3, Bi2O3)
glasses for fiber-optics. Opt. Mater. 33, 1911–1915 (2011)
36. X. Feng, J.C. Flanagan, K.E. Frampton, P. Petropoulos, N.M. White, J.H.V. Price, W.H. Loh,
H.N. Rutt, D.J. Richardson, Developing single-mode tellurite glass holey fiber for infrared
nonlinear applications. Adv. Sci. Tech. 55, 108–117 (2008)
37. O. Humbach, H. Fabian, U. Grzesik, U. Haken, W. Heitmann, Analysis of OH absorption
bands in synthetic silica. J Non-Crystall Solids 203, 19–26 (1996)
38. P.W. France, S.F. Carter, J.R. Williams, K.J. Beales, J.M. Parker, OH-absorption in fluoride
glass infra-red fibres. Electron Lett 20, 607–608 (1984)
238 X. Feng et al.

39. X. Feng, S. Tanabe, T. Hanada, Hydroxyl groups in erbium-doped germanotellurite glasses. J

Non-Crystall Solids 281, 48–54 (2001)
40. P. Domachuk, N.A. Wolchover, M. Cronin-Golomb, A. Wang, A.K. George, C.M.B. Cordeiro,
J.C. Knight, F.G. Omenetto, Over 4000 nm bandwidth of mid-IR supercontinuum generation
in sub-centimeter segments of highly nonlinear tellurite PCFs. Opt. Express 16, 7161–7168
(2008)
41. A.E. Comyns, Fluoride glasses (Wiley, Hoboken, NJ, 1989)
42. S. Nagel, J.B. MacChesney, K. Walker, An overview of the modified chemical vapor depo-
sition (MCVD) process and performance. IEEE Trans Microwave Theory Tech 18, 459–476
(1982)
43. M.D. O’Donnell, C.A. Miller, D. Furniss, V.K. Tikhomirov, A.B. Seddon, Fluorotellurite
glasses with improved mid-infrared transmission. J. Non-Crystall Solids 331, 48–57 (2003)
44. X. Feng, J. Shi, M. Segura, N.M. White, P. Kannan, W.H. Loh, L. Calvez, X. Zhang,
L. Brilland, Halo-tellurite glass fiber with low OH content for 2-5 μm mid-infrared nonlinear
applications. Opt. Express 21, 18949–18954 (2013)
45. X. Feng, J. Shi, M. Segura, N.M. White, P. Kannan, L. Calvez, X. Zhang, L. Brilland,
W.H. Loh, Towards water-free tellurite glass fiber for 2–5 μm nonlinear applications. Fibers
1, 70–81 (2013)
46. H. Ebendorff-Heidepriem, K. Kuan, M.R. Oermann, K. Knight, T.M. Monro, Extruded
tellurite glass and fibers with low OH content for mid-infrared applications. Opt Materials
Express 2, 432–442 (2012)
47. J.M. Dudley, L. Provino, N. Grossard, H. Maillotte, R.S. Windeler, B.J. Eggleton, S. Coen,
Supercontinuum generation in air–silica microstructured fibers with nanosecond and femto-
second pulse pumping. J Opt Soc Am B 19, 765–771 (2002)
48. A.Y.H. Chen, G.K.L. Wong, S.G. Murdoch, R. Leonhardt, J.D. Harvey, J.C. Knight,
W.J. Wadsworth, P.S.J. Russell, Widely tunable optical parametric generation in a photonic
crystal fiber. Opt. Lett. 30, 762–764 (2005)
49. T.M. Monro, Y.D. West, D.W. Hewak, N.G.R. Broderick, D.J. Richardson, Chalcogenide
holey fibres. Electron Lett 36, 1998–2000 (2000)
50. V.V.R.K. Kumar, A.K. George, J.C. Knight, P.S.J. Russell, Tellurite photonic crystal fiber.
Opt. Express 11, 2641–2645 (2003)
51. T. Delmonte, M.A. Watson, E.J. O’Driscoll, X. Feng, T.M. Monro, V. Finazzi, P. Petropoulos,
J.H.V. Price, J.C. Baggett, W. Loh, D.J. Richardson, D.P. Hand, Generation of mid-IR
continuum using tellurite microstructured fiber. Paper presented at Conference of Lasers and
Electro-Optics (CLEO) and Quantum Electronics and Laser Science Conference (QELS),
Long Beach, California, USA 21–25 May 2006 (2006)
52. X. Feng, T.M. Monro, V. Finazzi, R.C. Moore, K. Frampton, P. Petropoulos, D.J. Richardson,
An extruded single-mode high-nonlinearity tellurite glass holey fibre. Electron Lett 41,
835–837 (2005)
53. A. Boskovic, S.V. Chernikov, J.R. Taylor, L. Gruner-Nielsen, O.A. Levring, Direct
continuous-wave measurement of n2 in various types of telecommunication fiber at 1.55 μm.
Opt. Lett. 21, 1966–1968 (1996)
54. N.G.R. Broderick, T.M. Monro, P.J. Bennett, D.J. Richardson, Nonlinearity in holey optical
fibers: Measurement and future opportunities. Opt. Lett. 24, 1395–1397 (1999)
55. J.M. Dudley, G. Genty, S. Coen, Supercontinuum generation in photonic crystal fiber. Rev.
Mod. Phys. 78, 1135–1184 (2006)
56. M. Belal, L. Xu, P. Horak, L. Shen, X. Feng, M. Ettabib, D.J. Richardson, P. Petropoulos,
J.H.V. Price, Mid-infrared supercontinuum generation in suspended core tellurite
microstructured optical fibers. Opt. Lett. 40, 2237–2240 (2015)
57. J. Shi, X. Feng, P. Horak, K. Chen, P.S. Teh, S. Alam, W.H. Loh, D.J. Richardson, M. Ibsen,
1.06 μm picosecond pulsed, normal dispersion pumping for generating efficient broadband
infrared supercontinuum in meter-length single-mode tellurite holey fiber with high Raman
gain coefficient. J Lightwave Technol 29, 3461–3469 (2011)
9 Tellurite Glass Fibers for Mid-infrared Nonlinear Applications 239

58. R. Stegeman, L. Jankovic, H. Kim, C. Rivero, G. Stegeman, K. Richardson, P. Delfyett,

Y. Guo, A. Schulte, T. Cardinal, Tellurite glasses with peak absolute Raman gain coefficients
up to 30 times that of fused silica. Opt. Lett. 28, 1126–1128 (2003)
59. B.C. Stuart, M.D. Feit, A.M. Rubenchik, B.W. Shore, M.D. Perry, Laser-induced damage in
dielectrics with nanosecond to subpicosecond pulses. Phys. Rev. Lett. 74, 2248–2251 (1995)
60. X. Feng, W.H. Loh, J.C. Flanagan, A. Camerlingo, S. Dasgupta, P. Petropoulos, P. Horak,
K.E. Frampton, N.M. White, J.H.V. Price, H.N. Rutt, D.J. Richardson, Single-mode tellurite
glass holey fiber with extremely large mode area for infrared nonlinear applications. Opt.
Express 16, 13651–13656 (2008)
61. T.A. Birks, J.C. Knight, P.S.J. Russell, Endlessly single-mode photonic crystal fiber. Opt. Lett.
22, 961–963 (1997)
62. A.M. Grassi, F. Casagrandea, M. D’Alessandroa, S. Marinonia, Single-modeness of short large
mode area fibers: An experimental study. Opt Commun 273, 127–132 (2007)
63. L. Dong, X. Peng, J. Li, Leakage channel optical fibers with large effective area. J Opt Soc Am
B 24, 1689–1697 (2007)
Chapter 10
Tellurite Thin Films Produced by RF
Sputtering for Optical Waveguides
and Memory Device Applications

Luciana R.P. Kassab, Davinson M. da Silva, Vanessa D. del Cacho,

Leonardo Bontempo, Sebasti~ao G. dos Santos Filho,
and Marco Isaias Alayo Chavez

10.1 Introduction

Integrated optics, which allows the integration of multiple electronic devices and
optical components to fulfill some complex functions on a single substrate, has been
widely adopted for optical communications and signal processing systems due to
the innumerous advantages that can be attained, such as higher processing speed,
wider bandwidth for data processing, and reduced power consumption, among
others [1, 2]. The utilization of integrated optical devices in the sensing field has
also attracted more and more attention because of their advantages, when compared
with conventional electronic sensors, such as insensitivity to electromagnetic noise,
free short circuit and ground loops, wide range of temperature operation [3], safety
in environmental fuels and explosives, and low cost [4, 5].
The technology for the fabrication of passive devices such as planar optical
waveguides splitters and multiplexers is well developed; devices based on this
technology are now commercially available. To improve this technology, it has
been necessary to develop optical amplifiers that can be integrated with these
devices compensating the losses in splitters or other components and serving as
preamplifiers for active devices such as detectors. Over the last decade, there has
been a great effort to exploit efficient rare-earth optical amplifiers [6]. Erbium-
doped materials are of great interest in optical communications technology as they

L.R.P. Kassab (*) • D.M. da Silva • V.D. del Cacho

Faculdade de Tecnologia de S~ao Paulo/CEETEPS, S~ao Paulo, SP, Brazil
e-mail: [email protected]
L. Bontempo
Faculdade de Tecnologia de S~ao Paulo/CEETEPS, S~ao Paulo, SP, Brazil
Escola Politécnica da Universidade de S~ao Paulo, S~ao Paulo, SP, Brazil
S.G. dos Santos Filho • M.I.A. Chavez
Escola Politécnica da Universidade de S~ao Paulo, S~ao Paulo, SP, Brazil

© Springer International Publishing AG 2017 241

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_10
242 L.R.P. Kassab et al.

can serve as the gain medium in optical amplifiers operating at the standard
telecommunication wavelength (1.5 μm) [7]. In this regard, the search of new
materials for the fabrication of optical devices is crucial. In this aspect, tellurite
glasses are very promising materials for waveguide amplifiers due to some of their
important characteristic features such as high linear refractive index (up to 2.2), low
phonon energy (around 600–800 cm
1), wide transmission window (0.4–6 μm),
and high solubility for rare-earth ions. For example, pedestal-type waveguides
based in Bi2O3–WO3–TeO2 thin films as core layers are reviewed in this chapter,
for applications in optical devices such as the Mach–Zehnder interferometers (MZI)
[8]. Also, the possibility of using pedestal-type waveguides with Yb3þ/Er3þ
codoped Bi2O3–WO3–TeO2 thin films as optical amplifiers is reviewed for photonic
applications [9]. Optical gain has been demonstrated in tellurite glass and fiber
amplifiers [10–13]. The higher refractive index in the core enhances the confine-
ment factor of the optical field, which is a favorable characteristic for the develop-
ment of narrower waveguides. These characteristics are also important for
applications with optical devices such as MZI as they can provide lower values of
propagation losses that are important for wavelength guiding and for the perfor-
mance of the device. In both cases pedestal-type waveguides were used as they
present favorable characteristics for confining the optical field [14] and are largely
employed in the construction of microdisks for integrated laser microcavities
[15]. This structure is obtained by etching the lower cladding before the deposition
of the core layer. Thus, the geometrical definition of the optical waveguide is
simplified because it is no longer necessary to perform an etching of the core
layer [16]. It is very important to emphasize that avoiding the etching of the core
layer with this technique is very advantageous because when newly developed
materials are used as a core layer, the etching parameters are not known yet and
additional time consuming studies have to be performed before the utilization of
these materials in the fabrication of optical devices.
Also, TeO2–ZnO thin film glasses exhibit physical properties that make them
attractive candidates for matrix materials in memory devices: long-term chemical
stability when compared to other types of glasses and control of the semi-insulating
properties due to the binary phase when compared to silicate glasses [17]. In the last
years, much attention has been paid to memory devices that utilize nanomaterials
with bistable properties [18–23]. In particular, gold nanoparticles have been embed-
ded in a variety of matrices to obtain abrupt conductivity changes from a
low-conductivity state to a high-conductivity state, activated by an applied voltage,
which exceeds a well-established value [18, 24]. In this case, the bistable memory
effect has been taken into account considering the modeling of the current transport
mechanism, including the charge transfer between donor–acceptor pairs [25], the
hoping process associated to charged or non-charged metallic nanomaterials [25],
and the carrier–transport–path model [24].
In this chapter we review some of our previous results obtained with samples
based on different tellurite compositions. Firstly, we present results of pedestal-type
waveguides with undoped Bi2O3–WO3–TeO2 thin films as core layer for applica-
tions in MZI optical devices viewing the fabrication of temperature and pressure
10 Tellurite Thin Films Produced by RF Sputtering for Optical Waveguides and. . . 243

sensors in the future; secondly, we show recent results of pedestal-type waveguides

with Yb3þ/Er3þ codoped Bi2O3–WO3–TeO2 thin films for optical amplifier appli-
cation. Finally, the results of TeO2–ZnO thin films with Au nanoparticles, used in
bistable memory devices, are also shown.

10.2 Experimental Details

10.2.1 Thin Film Fabrication

For the production of the thin films with the RF sputtering technique, ceramic
targets were produced by mixing the high purity starting powders of TeO2 and ZnO
(for the TeO2–ZnO thin film) and of TeO2, Bi2O3, and WO3 (for the Bi2O3–WO3–
TeO2 thin film), submitting them to 8 t uniaxial press and then sintering at 515  C/
760  C, respectively, for 10 h. Targets with 5 cm diameter and a 0.4 cm thickness
were obtained with the following final composition (in wt%): 85.5TeO2–14.5ZnO
and 10Bi2O3–20WO3–70TeO2. The sputtering was carried out with argon and
oxygen gases (18 and 6 sccm flow, respectively), at 5 mTorr pressure and 40 W
RF power, for Bi2O3–WO3–TeO2 target, and 20 W for Er2O3 and Yb2O3 targets.
For the case of the TeO2–ZnO thin films with gold nanoparticles produced for
memory device application, two targets were sputtered simultaneously: a commer-
cial gold target and a ceramic one prepared as explained before. It was used RF
power of 6 W for the metallic gold target. The films were annealed at 325  C in air
for 10 or 20 h to thermally nucleate the crystalline nanoparticles [17]; thin films
without gold nanoparticles were also prepared to be used as reference. Finally,
aluminum was evaporated (1 μm thick) through a mechanical mask to define the
area of metal–insulator–semiconductor (MIS) structures (A ¼ 7.8  10
3 cm 2).

10.2.2 Production of the Pedestal Waveguide

The pedestal waveguides based on Yb3þ/Er3þ codoped Bi2O3–WO3–TeO2 thin

films and undoped Bi2O3–WO3–TeO2 thin films with the MZI structure were
produced in a silicon substrate using conventional microelectronics procedures
[8, 9]: chemical cleaning, thermal oxidation, optical lithography, plasma etching,
and sputtering deposition. One advantage of pedestal waveguides is that the etching
to define the sidewalls of the optical waveguides is always done on the same
material used as low cladding layer, in this case, the SiO2 film. Thus the material
used as core layer does not need to be etched. This simplifies the process, as new
materials need sometimes the development of different procedure for efficient
plasma etching.
244 L.R.P. Kassab et al.

10.2.3 Characterization Techniques

Passive characterization was performed for the pedestal optical waveguides, in

order to determine propagation losses in the visible (VIS) and infrared
(IR) regions. Propagation losses at 630 nm (He–Ne laser) and 1050 nm (Yb fiber
laser) were determined using the “top view” technique [14]. In this technique a
microscope and a charge-coupled device (CCD) camera are positioned above the
waveguide, in order to capture the scattered light intensity profile along the length
of the waveguides. The optical losses are calculated from the slope of the measured
curve of light intensity as a function of distance in centimeters.
Optical gain was measured in the IR region for the pedestal waveguides based on
Yb3þ/Er3þ codoped Bi2O3–WO3–TeO2 thin films, in order to evaluate the potential
of these thin films for applications as optical amplifiers at 1530 nm [9]. The optical
gain measurements were performed using 980 and 1530 nm diode lasers for pump
and signal wavelengths, respectively. Pump and signal beams were combined by a
suitable WDM coupler into the waveguides using a single-mode fiber. The signal
power coupled to the waveguides was kept constant at l μW, to prevent gain
saturation. The input fiber delivering both pump and signal was terminated by a
fiber lens with a mode field diameter of 3 μm. After propagation through the
waveguide amplifier, both pump and signal beams were coupled to a spectrometer
through a second lensed tip fiber focused on the output of the waveguide.
Scanning electron microscopy (SEM) was performed to analyze the pedestal
waveguide profile of Yb3þ/Er3 codoped and undoped Bi2O3–WO3–TeO2 wave-
guides. A profilometer was employed to determine film thicknesses. The concen-
tration of the rare-earth ions in Yb3þ/Er3þ codoped Bi2O3–WO3–TeO2 thin films
was determined by Rutherford backscattering spectrometry (RBS) and particle-
induced X-ray emission (PIXE) analysis, following the studies of Miritello et al.
[26]. The measured concentrations of Er3þ and Yb3þ were 0.05  1020 ions/cm3
and 4.18  1020 ions/cm3.
In the case of the TeO2–ZnO thin films fabricated for memory device applica-
tions, the transmission electron microscope (TEM) was used to determine the size
distribution of the embedded nanoparticles and the crystalline nature of the
obtained structures by means of electron diffraction analysis. Also, after the
fabrication of the MIS capacitors, they were electrically characterized with the
aid of an HP4140 picoammeter to obtain parameters from the current–voltage (I–V)
characteristics. Leakage current was measured submitting the capacitors to an
external voltage ramp and then recording the I–V characteristics. C–V curves
were also measured using an impedance analyzer (Agilent 4294A). All the electri-
cal measurements were performed at room temperature for all thin films.
10 Tellurite Thin Films Produced by RF Sputtering for Optical Waveguides and. . . 245

10.3 Results and Discussion

10.3.1 Production of Mach–Zehnder Interferometer

Structures on Bi2O3–WO3–TeO2 Pedestal-Type
Waveguides for the Production of Optical Devices

The utilization of integrated optical devices in the sensing field have attracted
attention because of the advantages, when compared with conventional electronic
sensors. In this work we review recent results of Bi2O3–WO3–TeO2 (TWB) thin
films for the fabrication of pedestal-type waveguides with Mach–Zehnder struc-
tures aiming application with optical sensing devices [8]. The important character-
istics of these materials such as high refractive index (n  2.0), low phonon energy
between 700 and 900 cm
1, low chemical reactivity, and wide transmission win-
dow from the visible to infrared region (400–5000 nm) [27] allow the utilization of
these thin films as core layers in optical waveguides. A great interest in pedestal-
type waveguides has emerged in integrated optics as they present favorable char-
acteristics of optical field confining as mentioned before. This procedure represents
an alternative method for achieving the lateral confinement in optical waveguides
fabricated with silicon technology that consists in doing the reactive ion etching
(RIE) step before the core definition. Because of these advantages, this process was
used in the present study. Figure 10.1 presents the propagation loss measurements

Fig. 10.1 Propagation losses as a function of pedestal waveguide width at 633 and 1050 nm [8]
246 L.R.P. Kassab et al.

Fig. 10.2 SEM image of Bi2O3–WO3–TeO2 (TWB) pedestal waveguides [8]

for the pedestal optical waveguides with core width in the 1–100 μm range.
Propagation losses around 2.0 and 2.5 dB/cm at 633 and 1050 nm, respectively,
were obtained for waveguides in the 20–100 μm width range. Lower values of
propagation losses were obtained in the visible region and can be attributed to the
fact that lower wavelengths have better light confinement. We observed that the
larger width waveguides presented smaller propagation losses. For waveguide
width smaller than 20 μm, an increase of the propagation losses was observed,
due to the influence of the roughness on sidewall surfaces that are normally higher
in these waveguides, as already reported [28–30].
Figure 10.2 shows the results of the scanning electron microscopy (SEM)
measurements showing the pedestal structure; we observe that the pedestal height
and the core layer height are of about 1.0 μm and 600 nm, respectively. Also the top
of the pedestal structure has almost no surface roughness. However high roughness
was observed, in the lower regions of the waveguide; this may be related to the
micromasking effect [31] caused by the migration of some chromium particles to
the SiO2 exposed region, during the plasma etching. These particles act as
micromasks that protect small regions of the SiO2 surface, causing the observed
roughness. Despite the presence of this roughness, low propagation losses were
obtained, as shown in Fig. 10.1.
Figure 10.3 shows the near-field profile of the waveguides, at 633 nm of the
optical waveguides with different widths. Elliptical shapes are observed for the
larger waveguides, as expected. Besides, for waveguide widths larger than 9 μm,
multimode behavior is observed. We also accomplish preliminary characterizations
in MZI structures that were fabricated using the pedestal-type technique and Bi2O3–
WO3–TeO2 thin films as core layer. Figure 10.4 presents the images obtained at the
output of the MZI consisting in 30 μm width waveguides. This result demonstrates
that the laser beam coupled into the input waveguide of the MZI device was guided
to the output waveguide. MZI structures consisting in optical waveguide width
smaller than 30 μm also exhibit good light guiding.
10 Tellurite Thin Films Produced by RF Sputtering for Optical Waveguides and. . . 247

Fig. 10.3 Scan near-field Bi2O3–WO3–TeO2 waveguides for waveguides with different widths at
633 nm [8]

Fig. 10.4 Scan near-field Bi2O3–WO3–TeO2 waveguides in the output of the MZI for 30 μm at
633 nm [8]

10.3.2 Optical Amplifier Based on Pedestal Waveguides

Produced with Yb3þ/Er3þ Codoped Bi2O3–WO3–TeO2
Thin Films

Experimental results on pedestal waveguides produced with Yb3þ/Er3þ codoped

Bi2O3–WO3–TeO2 thin films deposited by RF magnetron sputtering technique [9]
are reviewed. As stated before, erbium-doped materials are of great interest in
optical communications technology, as they can serve as the gain medium in optical
amplifiers operating at the standard telecommunication wavelength (1.5 μm)
[7]. When only Er3þ ions are present in waveguides, the pumping at 980 nm is
248 L.R.P. Kassab et al.

Fig. 10.5 SEM image of Bi2O3–WO3–TeO2 pedestal waveguide (5 μm width) [9]

not efficient; however this problem can be bypassed by the addition of Yb3þ as an
efficient sensitizer of Er3þ ions that provides the possibility of obtaining wave-
guides for VIS and IR optical amplification as well as for other photonic
applications.
In this work, as in the previous review, the fabrication process of pedestal-type
waveguides is proposed to define the sidewalls before deposition of Bi2O3–WO3–
TeO2 core waveguides. This process allowed significant reduction of losses in the
5–100 μm width range.
Comparing scanning electron microscopy (SEM) images of the samples fabri-
cated at different times, it was possible to ensure that the fabrication process has
good reproducibility. Figure 10.5 shows SEM images of the Yb3þ/Er3þ codoped
Bi2O3–WO3–TeO2 pedestal waveguide where we can see that the core surface
presents low roughness.
Low propagation losses were obtained in the 5–100 μm width range. These
results are similar to those measured for the undoped Bi2O3–WO3–TeO2 pedestal
waveguides, shown before. Propagation losses around 2.0 and 2.5 dB/cm were
obtained at 633 and 1050 nm, respectively, for core waveguides in the 20–100 μm
width range. Besides, in the 10–15 μm width range propagation losses, around 2.7
and 3.0 dB/cm were obtained at 633 and 1050 nm, respectively.
The luminescence spectrum of Fig. 10.6 shows the emission band centered at
1.53 μm due to the 4I13/2!4I15/2 transition of Er3þ ions for Bi2O3–WO3–TeO2
waveguides showing the incorporation of rare-earth ions in the trivalent form.
The internal gain of 3.7 dB/cm was achieved at 1530 nm, under 980 nm
excitation, with only 9 mW of pump power in the waveguide and low Er3þ
concentration (0.05  1020 ions/cm3) for core waveguides of 20 μm width
(Fig. 10.7). Results of TeO2 rib waveguide amplifier at 1530 nm, doped only with
Er3þ, were reported recently [32]. In this case, even using a higher Er3þ concen-
tration (2.2  1020 ions/cm3) than the one used in the present study and also a higher
pump power of 200 mW, it was obtained internal gain of 2.8 dB/cm; so, in our case,
10 Tellurite Thin Films Produced by RF Sputtering for Optical Waveguides and. . . 249

Fig. 10.6 Luminescence spectra of Yb3þ/Er3þ codoped Bi2O3–WO3–TeO2 pedestal waveguide in

the infrared region (excitation at 980 nm) [9]

Fig. 10.7 Optical gain at 1530 nm, for Yb3þ/Er3þ codoped Bi2O3–WO3–TeO2 pedestal waveguide
(18 μm width) [9]
250 L.R.P. Kassab et al.

a higher internal gain of 3.7 dB/cm could be obtained, with a lower pump power of
9 mW, probably because of the use of Yb3þ as sensitizer (4.18  1020 ions/cm3). So,
the adequate concentration of Yb3þ (donor) and Er3þ (acceptor) enabled an efficient
energy transfer that provided a considerable luminescence and internal gain at
1530 nm. The present results demonstrate, for the first time, possible applications
of Yb3þ/Er3þ codoped Bi2O3–WO3–TeO2 pedestal waveguides as optical amplifiers
at 1530 nm.

10.3.3 TeO2–ZnO Thin Films with Au Nanoparticles

for Memory Devices

In this section, we review some recent results of the observation of a reversible

memory phenomenon in co-sputtered TeO2–ZnO thin films containing gold
nanoparticles (NPs) [33].
Figure 10.8 shows typical TEM plane view for TeO2–ZnO/Au films with 100 and
500 nm thickness and heat treatment of 10 and 20 h (darker dots represent the gold
NPs).
Gold NPs size distribution after annealing at 325  C for 10 and 20 h are shown in
Fig. 10.9. For samples submitted to heat treatment for 10 h, the average gold NPs
size was about 2.2 and 2.7 nm for film thicknesses of 100 and 500 nm, respectively.
After annealing the films for 20 h, the average size of gold NPs increased more than
100%. The dispersion observed after heat treatments may be attributed to the
occurrence of coalescence concomitantly with the growth of the gold NPs. Since
the gold incorporation in the films was approximately the same for each thickness

Fig. 10.8 TEM plane view

for B1 films containing Au
NPs with different thickness
and heat treatment: (a)
100 nm, 10 h; (b) 100 nm,
20 h; (c) 500 nm, 10 h; (d)
500 nm, 20 h. The
dimensions of the
micrograph are indicated by
the numbers above and on
the left of them [33]
10 Tellurite Thin Films Produced by RF Sputtering for Optical Waveguides and. . . 251

Fig. 10.9 Frequency distribution of NPs diameters for films with thickness and heat treatment:
(a) 100 nm, 10 h; (b) 100 nm, 20 h; (c) 500 nm, 10 h; (d) 500 nm, 20 h [33]

during the co-sputtering, the aerial concentration of NPs should decrease when the
annealing time increases from 10 to 20 h. This is illustrated in Fig. 10.10 and
corroborates the occurrence of coalescence concomitantly with the growth of the
gold NPs.
Typical I–V characteristic curves ([log I]  V) for 100 nm film with gold NPs,
annealed at 325  C during 10 h, are shown in Fig. 10.11. A state of low conductivity
(approximately 10
10–10
8A) was recorded from 0 to 6.5 V for the first voltage
scan, corresponding to a “0” state of a bistable memory; low conductivity occurs
until the applied voltage achieved 6.5 V. An abrupt current transition from 10
8 to
10
3 A, followed by permanence around 10
3 A, was observed for applied voltages
higher than 6.5 V, indicating a state of high conductivity corresponding to a “1”
state of the bistable memory. The high-conductivity state remained for the second
scan, even for voltages below the threshold observed in the first scan [34].
Figure 10.12 shows typical I–V characteristics for 100 nm dielectrics without
incorporation of gold NPs, annealed at 325  C during 10 and 20 h. The observed
252 L.R.P. Kassab et al.

Fig. 10.10 Aerial concentration of NPs as function of the annealing time for 100 and 500 nm

Fig. 10.11 Typical I–V characteristics for 100 nm film with gold NPs, annealed at 325  C during
10 h [34]

current for bias voltages below 6.5 V was at least an order of magnitude higher (10

7
A) for TeO2–ZnO thin films without gold NPs showing that the presence of gold
NPs contributes for the reduction of current leakage in the device.
Figure 10.13 shows the typical I–V characteristic curves for 100 nm and 500 nm
dielectrics with gold nanoparticles, annealed at 325  C during 10 h and 20 h. It was
10 Tellurite Thin Films Produced by RF Sputtering for Optical Waveguides and. . . 253

Fig. 10.12 Typical I–V characteristics for 100 nm dielectrics without gold NPs annealed at 325  C
during 10 and 20 h

Fig. 10.13 Typical I–V characteristics for 100 and 500 nm dielectrics with gold nanoparticles,
annealed at 325  C during 10 and 20 h [34]

observed that transition voltage (Vt) is higher for the thicker films (500 nm). This
behavior may be attributed to the greater difficulty to fill and saturate the higher
amount of traps for the thicker films, which is crucial for current abrupt
transition [34].
254 L.R.P. Kassab et al.

Fig. 10.14 Typical C/Co  VG Characteristics for 100 and 500 nm [33]

Lower transition voltage was observed for films thicknesses of 100 nm heat
treated for 20 h. This is consistent to the Lampert’s model [34], since this film
presented lower aerial concentration of NPs, in respect to the film submitted to a
heat treatment for 10 h. However, the opposite behavior was noticed for the 500 nm
thickness film, indicating that the amount of traps did not reduce significantly after
heat treatment. Observing the NPs size distribution (Fig. 10.9d), a considerable
amount of smaller particles (~ 3 nm) are still present in the sample, despite the
formation of larger particles after the heat treatment. Consequently, the presence of
smaller particles in the sample contributed to inhibit the reducing of traps and,
therefore, the transition voltage.
Figure 10.14 shows typical C–V measurements of TeO2–ZnO with gold NPs
annealed at 325  C during 20 h for films thicknesses of 100 and 500 nm. The
capacitance decrease at the “knee point” is related to the fact that the leakage
current is sufficiently high to start removing the inversion layer, favoring the
appearance of the deep depletion. This occurs approximately at 0.1 nA (1.0  10

8
A/cm2) for 100 nm thickness film and at 1.0 nA (1.0  107 A/cm2) for 500 nm
thickness (see Fig. 10.13).
Drift current is lower for 100 nm thickness film since the charging effect is lower
in the film and at the interface with silicon substrate. The less pronounced stretching
of the C–V curve also corroborates the lower charging effect in the sample
[35]. Similar behavior was reported for PbO–GeO2 thin films with silver NPs
[36]. Thus, the present review showed the use of Au-NPs incorporated TeO2–ZnO
thin films for memory device applications.
10 Tellurite Thin Films Produced by RF Sputtering for Optical Waveguides and. . . 255

10.4 Conclusions

In this chapter, recent studies related to tellurite thin films production by RF

sputtering for optical waveguides and memory device applications were reviewed.
Results of pedestal waveguides based on Yb3þ/Er3þ codoped Bi2O3–WO3–TeO2
thin films fabricated using conventional microelectronics fabrication processes
aiming optical amplifier applications, at 1530 nm, were presented. The purpose of
the study aimed to fulfill the lack of results in the literature related to Yb3þ/Er3þ
codoped tellurite waveguides for optical amplifier applications. 5.6 dB internal gain
(~3.7 dB/cm) was achieved, under 980 nm excitation, demonstrating the possibility
of providing gain in short length waveguides based in tellurite materials.
Also, pedestal waveguides using Bi2O3–WO3–TeO2 thin films as core layer were
reviewed for applications in optical devices, such as Mach–Zehnder interferometry.
Low propagation losses were obtained at 633 and 1050 nm; 2.0 and 2.5 dB/cm at
633 and 1050 nm, respectively, were obtained for waveguides in the 20–100 μm
width range. This was the first report that showed the feasibility of fabricating
Mach–Zehnder interferometers based on tellurite amorphous thin films present in
pedestal-type waveguides for applications with optical sensing devices.
In the cases previously exposed, pedestal geometry was used for the waveguides
production. The possibility of using pedestal-type waveguides as optical devices
was reviewed. Pedestal geometry presents several advantages in relation to other
waveguide geometries, like rib waveguides, for example, since the material used as
core layer does not need to be submitted to etching procedures. Thus, the fabrica-
tion process of the optical waveguides is simplified because the reactive ion etching
occurs before the core definition, alloying the application of different materials in
the waveguides, without the necessity of the development of a fabrication new
route. Pedestal geometry represents, as well, a simple alternative method to obtain
the adequate lateral confinement for light guiding and could be successfully
employed in optical devices fabrication based on tellurite glasses.
A review showing the use of TeO–ZnO composition for memory device appli-
cations was presented. For the MIS structure with TeO–ZnO containing gold
nanoparticles, the current–voltage (I–V) and capacitance–voltage (C–V ) character-
istic curves were shown. A switching behavior, starting from a low-conductivity to
a high-conductivity state was observed, upon application of an external voltage in
3–10 V range. The current increased about five orders of magnitude within a very
narrow applied voltage, proportional to the increase of the film thickness. It was
also observed that the presence of smaller gold NPs (2–3 nm) tends to increase the
transition voltage in the MIS structures.
In addition, capacitance–voltage (C–V ) characteristic curves were correlated to
the current–voltage (I–V) results and to NP size distribution obtained from TEM
analysis. The significant reduction of the capacitance in the inversion region,
starting from different gate voltages (“knee points”), is attributed to the current
leakage, which promotes a deep depletion regime in silicon. This was the first report
256 L.R.P. Kassab et al.

that showed that TeO2–ZnO thin films containing gold NPs are adequate hosts for
the production of memory devices.
Finally, the results herein show that tellurite thin films produced by RF
sputtering technique are suitable hosts for the nucleation of gold nanoparticles
and can be used with success in optical and memory devices.

Acknowledgments We thank the financial support from the Conselho Nacional de

Desenvolvimento Cientı́fico e Tecnológico (CNPq) through the National Institute of Photonics
(INCT Project). The Laboratório de Microscopia Eletrônica (Instituto de Fı́sica—USP) and the
Nanotechnology National Laboratory CNPEM are acknowledged for the TEM images.

References

1. M.P.Y. Desmulliez, VLSI system integration technological challenges. Mater. Sci. Eng. B 74,
269 (2000)
2. A.V. Krishnammoorthy, Optoeletronics — VLSI: photonics integrated with VLSI circuits.
IEEE J. Sel. Top Quant. Electron. 4, 899 (1998)
3. W. Quick, K. James, J. Coker, Fiber optic sensing techniques. Proc. First Int. Conf. Opt. Fiber
Sensors IEEE 6, 6 (1983)
4. K. Okamoto, Fundamentals of Optical Waveguides (Academic Press, Elsevier, San Diego,
CA, 2005)
5. E. Suhir, Microelectronics and photonics—the future. Microchem. J. 31, 839 (2000)
6. R.T. Zhao, L.F. Shen, H. Lin, Optik Int. J. Light Electron. Opt. 122(24), 2237 (2011)
7. Y.C. Yan, A.J. Faber, H. de Waal, P.G. Kik, A. Poman, Appl. Phys. Lett. 71(20), 2922 (1997)
8. M.E. Camilo, L.R.P. Kassab, T.A.A. Assumpç~ao, V.D.D. Cacho, M.I. Alayo, Thin Solid Films
571, 225 (2014)
9. V.D. del Cacho, D.M. da Silva, T.A.A. de Assumpç~ao, L.R.P. Kassab, M.I. Alayo, E.G. Melo,
Opt. Mater. 38, 198 (2014)
10. A. Mori, J. Cerma. Soc. Jpn. 116, 1040 (2008)
11. C. Chryssou, F. Di Pasquale, C.W. Pitt, IEEE J. Sel. Top. Quant. Electron. 6, 114 (2000)
12. S. Dai, C. Yu, G. Zhou, J. Zhang, G. Wang, L. Hu, JOL 117, 39 (2006)
13. S. Shen, A. Jha, X. Liu, M. Naftaly, K. Bindra, H.J. Bookey, A.K. Kar, J. Am. Ceram. Soc. 85,
1391 (2002)
14. T.A.A. de Assumpç~ao, D.M. da Silva, V.D. Del Cacho, L.R.P. Kassab, M.I. Alayo, J. Alloys
Compd. 586, 5368 (2013)
15. J. Cheng, W. Zhang, Q. Zhou, Y. Wang, Y. Huang, J. Peng, Opt. Lett. 36, 1797 (2011)
16. D.O. Carvalho, M.I. Alayo, Oxide-based materials and devices. Proc. Spie 2(7940), 17 (2014)
17. L. Bontempo, S.G. dos Santos Filho, A.L. Siarkowski, L.R.P. Kassab, ECS Trans. 39,
137 (2011)
18. J. Ouyang, C.W. Chu, C.R. Szmanda, L. Ma, Y. Yang, Nat. Mater. 3, 918 (2004)
19. Z. Fan, D. Wang, J.G. Lu, Silver-tetracyanoquinodimethane (Ag-TCNQ) nanostructures and
nanodevice. Proc. IEEE Nanotechnol. 2, 588 (2003)
20. R.J. Tseng, J. Ouyang, C.W. Chu, J. Huang, Y. Yang, Appl. Phys. Lett. 88, 123 (2006)
21. Z. Liu, F. Xue, Y. Su, K. Varahramyan, IEEE Electron. Device Lett. 151, 27 (2006)
22. Y. Song, Q.D. Ling, C. Zhu, E.T. Kang, D.S.H. Chan, Y.H. Wang, IEEE Electron. Device Lett.
154, 27 (2006)
23. L. Shi, Z. Liu, B. Xu, L. Gao, Y. Xia, J. Yin, Nanoscale Res. Lett. 1178, 4 (2009)
24. H.-T. Lin, Z. Pei, Y.-J. Chan, IEEE Electron. Device Lett. 569, 28 (2007)
25. J.G. Simmons, R.R. Verderber, Proc. R. Soc. A 77, 301 (1967)
10 Tellurite Thin Films Produced by RF Sputtering for Optical Waveguides and. . . 257

26. M. Miritello, P. Cardile, R. Lo Savio, F. Priolo, Opt. Express 19, 20761 (2011)
27. J. Wang, E. Vogel, E. Snitzer, Opt. Mater. 3, 187 (1994)
28. F.P. Payne, J.P.R. Lacey, Opt. Quant. Electron. 26, 977 (1994)
29. K.K. Lee, D.R. Lim, H.-C. Luan, A. Agarwal, J. Foresi, L.C. Kimerlin, Appl. Phys. Lett. 77,
1617 (2000)
30. T. Barwicz, H.A. Haus, Three-dimensional analysis of scattering losses due to sidewall.
J. Lightwave Technol. 23, 2719–2732 (2005)
31. S.P. Zimin, E.S. Gorlachev, I.I. Amirov, H. Zogg, J. Phys. D Appl. Phys. 42, 165205 (2009)
32. K. Vu, S. Madden, Opt. Express 18, 19192 (2010)
33. L. Bontempo, S.G. dos Santos Filho, L. R. P. Kassab, TeO2-ZnO thin films with gold
nanoparticles as passivating materials for power devices applications. in 29th International
Symposium on Microelectronics Technology and Devices -SBMicro 2014, 2014, Aracajú.
Proceedings of 29th International Symposium on Microelectronics Technology and Devices
– SBMicro 2014, IEEEXPlore, New Jersey, 2014, pp. 1–3
34. L. Bontempo, S.G. dos Santos Filho, L.R.P. Kassab, Thin Solid Films 611, 21 (2016)
35. E.H. Nicollian, J.R. Brews, MOS (Metal Oxide Semiconductor) Physics and Technology (John
Wiley & Sons, New York, 1982)
36. S.G. dos Santos Filho, V. Sonnenberg, W.G. Hora, D.M. da Silva, L.R.P. Kassab, Thin Solid
Films 520, 2695 (2012)
Chapter 11
Laser Writing in Tellurite Glasses

T. Toney Fernandez, S.M. Eaton, G. Jose, R. Osellame, P. Laporta,

and J. Solis

11.1 Introduction

Direct laser writing of waveguides in tellurite glass was first introduced to surpass
the technical difficulties faced by the existing and well-developed ion-exchange
technique. Since then, it became one of the preferred methods to produce 2D/3D
waveguide structures within the bulk glass and thin films. With the current trend of
emerging mid-infrared market exploiting the fundamental fingerprint region of
molecules and tissues, mid-infrared transparent glasses such as tellurites could
play a pivotal role. They aim at several vital fields like medical, environmental
sensing/monitoring, military surveillance, astronomy, and high-sensitivity gas

T. Toney Fernandez (*)

Dipartimento di Fisica, Politecnico di Milano, Piazza Leonardo da Vinci 32,
Milano 20133, Italy
e-mail: [email protected]
S.M. Eaton • P. Laporta
Dipartimento di Fisica, Politecnico di Milano, Piazza Leonardo da Vinci 32,
Milano 20133, Italy
Istituto di Fotonica e Nanotecnologie-Consiglio Nazionale delle Ricerche (IFN-CNR),
Piazza Leonardo da Vinci 32, Milano, Italy
G. Jose
Faculty of Engineering, Institute for Materials Research, School of Chemical and Process
Engineering, University of Leeds, Leeds LS2 9 JT, UK
R. Osellame
Istituto di Fotonica e Nanotecnologie-Consiglio Nazionale delle Ricerche (IFN-CNR),
Piazza Leonardo da Vinci 32, Milano, Italy
J. Solis
Laser Processing Group, Instituto de Optica, CSIC, Serrano 121, Madrid 28006, Spain
e-mail: [email protected]

© Springer International Publishing AG 2017 259

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_11
260 T. Toney Fernandez et al.

detection utilizing the relatively unexplored mid-infrared wavelength region. The

preference of tellurite glass over other low phonon energy glasses is due to its
comparatively higher mechanical robustness, thermal stability, and proven track
record in the near-infrared (NIR) region, thanks to the telecom industry boom in the
mid-1990s.
The mid-infrared wavelength window is abundant in rotational and vibrational
molecular transitions with stronger line strengths than its visible NIR overtones. A
highly resolved finger printing of the species is possible as the spectrum is less
congested, allowing error-free identification of large number of molecules. The
3 μm region is the center-focus due to the strong vibrational bands of C-H, N-H, and
OH. Currently fluoride- and tellurite-based optical fibers are the main platforms on
which mid-infrared active devices are being realized. The fabrication of active
waveguide functionality in the bulk of such glasses would enable their application
in the integrated optics industry.

11.2 Femtosecond Laser Waveguide Writing

Since femtosecond laser writing is the main technology in forming waveguides in

tellurites, we will provide an introduction to the method here. Femtosecond laser
microprocessing is a direct, maskless technique capable of inducing a permanent
refractive index increase buried beneath the surface of transparent glasses, enabling
photonic circuit fabrication in 3D geometries. Among the many variables in
femtosecond laser waveguide writing, the repetition rate has the most important
role since it influences the heat accumulation between laser pulses, which deter-
mines the regime of modification and the resulting morphological change. In
silicate, phosphate, and tellurite glasses, higher repetition rates are shown to be
beneficial for driving increased heat accumulation, leading to rapid fabrication of
low-loss optical waveguides with tunable size and refractive index contrast [1].

11.2.1 Nonlinear Absorption

Focused femtosecond laser pulses, with wavelengths typically in the visible or near-
infrared, do not have enough photon energy to be linearly absorbed in glasses.
Instead, valence electrons may be promoted to the conduction band through
nonlinear photoionization, which proceeds by multiphoton ionization and/or
tunneling photoionization pathways depending on the laser and glass properties
[2]. In addition to nonlinear photoionization, avalanche photoionization also
occurs, explaining the small variation in threshold intensity for breakdown with
bandgap [3]. Because of this low dependence of the breakdown threshold on the
bandgap energy, femtosecond laser nanofabrication can be applied to a wide range
of glasses and other transparent materials.
11 Laser Writing in Tellurite Glasses 261

Depending on the laser frequency and intensity, Keldysh [4] showed that
nonlinear photoionization can occur by tunneling or multiphoton ionization.
Multiphoton absorption occurs due to the simultaneous absorption of several
photons by an electron in the valence band. The number of photons m required to
bridge the bandgap must satisfy the relation mhν > Egap, where Egap is the bandgap
and ν is the laser frequency. For a combination of higher laser intensities and longer
laser wavelength, the strong laser field distorts the band structure and reduces the
energetic barrier between the valence and conduction bands, allowing for direct
band-to-band transitions by tunneling. For typical waveguide writing conditions in
most glasses, nonlinear photoionization is a combination of both tunneling and
multiphoton ionization.
Electrons already present in the conduction band, provided by thermally excited
impurities or nonlinear photoionization, may absorb laser radiation by free carrier
absorption. After the sequential linear absorption of several photons, a conduction
band electron’s energy exceeds the conduction band minimum, allowing the hot
electron to impact ionize a bound electron in the valence band, giving rise to two
excited electrons at the conduction band minimum. These two electrons can then
undergo free carrier absorption and impact ionization, and the process can repeat
itself as long as the laser field is present and sufficiently strong. This process is
referred to as avalanche ionization.
For subpicosecond laser pulses, absorption is faster than energy coupling to the
lattice, decoupling the absorption and lattice heating processes [5]. Seeded by
nonlinear photoionization, the electron density in the conduction band increases
via avalanche ionization until the plasma frequency approaches the laser frequency,
at which point the plasma becomes strongly absorbing. For a typical femtosecond
laser with 1-μm wavelength, the plasma frequency equals the laser frequency when
the free carrier density approaches ~1021 cm
3, known as the critical density,
commonly used to define optical breakdown. In terms of intensity, this breakdown
threshold is typically ~1013 W/cm2 in glasses.
Since the electron-phonon lattice coupling time is about 10 ps, the absorbed laser
energy is transferred to the lattice well after the laser pulse is gone. As short pulses
require less energy to achieve the intensity for breakdown and because the absorp-
tion is decoupled from the lattice heating, more precise laser nanofabrication is
possible relative to longer pulses. Another advantage of using femtosecond laser
pulses is a deterministic breakdown, since nonlinear photoionization can seed the
electron avalanche. This is in contrast to the stochastic breakdown with longer
pulses which relies on the low concentration of impurities (about one impurity
electron in conduction band per focal volume), randomly distributed in the substrate
to seed an electron avalanche [6].
262 T. Toney Fernandez et al.

11.2.2 Relaxation and Material Modification

Although it is well accepted that nonlinear photoionization and avalanche ioniza-

tion are responsible for the creation of a free electron plasma, the subsequent
mechanisms are less clear when the electrons have transferred their energy to the
lattice and the material is modified. In the 2000 published articles on optical
waveguide writing citing the first work by the Hirao group [7], the reported
morphological changes can be generally classified into three types of modifications:
a smooth refractive index change [8], a form of birefringent refractive index
modification [9–12], and microexplosions resulting in empty voids [13]. The type
of modification depends on many exposure parameters such as pulse energy,
duration, repetition rate, wavelength, polarization, focal length, and scan velocity
but also on glass properties such as bandgap, thermal conductivity, and
composition.

11.2.3 Repetition Rate

The physical interpretations for the structural changes induced by focused femto-
second lasers typically assumed single pulse interactions, but can likely be extended
to the explained modification from multipulse interactions during waveguide writ-
ing, assuming the repetition rate is low enough that thermal diffusion has carried the
heat away from the focus before the next pulse arrives [14]. In this case, the
following pulses may add to the overall modification, but still act independently
of one another.
For high repetition rates (>100 kHz), the time between laser pulses is less than
the time for heat to diffuse away, giving rise to a buildup of temperature in the focal
volume. For sufficiently high pulse energy, the glass near the focus is melted, and as
more laser pulses are absorbed, this melted volume continues to expand until the
laser is removed, when due to rapid cooling results in a region of altered refractive
index. For scanned waveguide exposures, the size of the melted volume is deter-
mined by the effective number of pulses in the laser spot size, N ¼ 2w0R/v, where
2w0 is the spot size, R is the repetition rate, and v is the scan velocity.
Figure 11.1 shows microscope images of borosilicate glass modified by static
laser exposure of 400-nJ pulse energy with different repetition rate and number of
pulses. Spherical laser-modified zones were observed for all static exposures tested
and arise from the three-dimensional symmetry of heat diffusion from a small laser
absorption volume of ~2-μm diameter. These refractive index structures are due to
localized melting within a cumulative heating zone that is built up over many laser
pulses, which then cools rapidly to resolidify after the exposure. Evidence of heat
accumulation is noted at repetition rates above 200 kHz, where the diameter of the
modified volume significantly exceeds the ~2-μm laser spot size. Within each row
(constant repetition rate) in Fig. 11.1, one can see a modest increase in the diameter
11 Laser Writing in Tellurite Glasses 263

Fig. 11.1 Optical microscope images showing modified volumes created in borosilicate glass
with 400-nJ pulses from a femtosecond laser. Total pulse (top) and net fluence (bottom) is shown
for each column, and the repetition rate is indicated for each row. Laser direction is normal to the
page [1]

of the heat-affected zone despite a four order-of-magnitude increase in exposure.

More striking is the tenfold increase in modified zone size when the repetition rate
is increased from 0.1 to 1 MHz in each column. Since the total laser exposure is
identical within any column, 200-kHz repetition rate represents the onset for
cumulative heating effects above which thermal diffusion controls the properties
of the structure formed by the femtosecond laser. One can also appreciate that the
size of the modification zone grows more quickly with the number of pulses when
in the cumulative heating regime.

11.2.4 Linear and Nonlinear Propagation

Linear effects such as dispersion, diffraction, aberration, and nonlinear effects such
as self-focusing, plasma defocusing, and energy depletion influence the propagation
of focused femtosecond laser pulses in glasses, resulting in an altered energy
distribution at the focus, distorting the final refractive index modification.
Neglecting spherical aberration and nonlinear effects, the spatial intensity profile
of a focused femtosecond laser beam, can be well represented by the paraxial wave
equation and Gaussian optics. The diffraction-limited minimum waist radius w0
(1/2 the spot size) for a collimated Gaussian beam focused in glass is given by:

M2 λ
w0 ¼ ð11:1Þ
πNA
264 T. Toney Fernandez et al.

Fig. 11.2 Calculated

intensity distribution for an
800 nm wavelength laser
beam focused 1 mm inside
fused silica, for NA ¼ 0.1,
0.2, and 0.6 [40]. z is the
beam propagation axis

where M2 is the Gaussian beam propagation factor [15] and NA is the numerical
aperture of the focusing objective. The Rayleigh range z0 (1/2 the depth of focus)
inside a glass of refractive index n is given by:

M2 nλ
z0 ¼ ð11:2Þ
πNA2

Chromatic and spherical aberration alters the intensity distribution near the focus
so that Eqs. (11.1) and (11.2) are no longer valid approximations. Chromatic
aberration as the result of dispersion in the lens can be corrected by using chromatic
aberration-corrected microscope objectives for the wavelength spectrum of interest.
For lenses made with easily formed spherical shapes, light rays that are parallel to
the optic axis but at different distances from the optic axis do not converge to the
same point, resulting in spherical aberration. This can be addressed by using
multiple lenses such as those found in microscope objectives or using an aspheric
focusing lens. In waveguide writing where light is focused inside glass, the index
mismatch at the air-glass interface introduces additional spherical aberration. As a
result, there is a strong depth dependence for femtosecond laser-written buried
structures [16, 17], which is even more pronounced for higher NA objectives [5], as
shown in Fig. 11.2 [40]. To avoid such issues, oil immersion lenses [19] or dry
objectives with collars that can correct for spherical aberration at different focusing
depths may be used [16].
Dispersion from mirror reflection and transmission through materials can
broaden the pulse width which can reduce the peak intensity and alter the energy
dissipation at the focus. However, it is only for short pulse <40-fs oscillators with
large bandwidths that dispersion becomes an issue. In this case, precompensation of
the dispersion through the microscope objective is required to obtain the shortest
pulse at the focus [18].
As a consequence of third-order nonlinearities, the spatially varying intensity of
a Gaussian laser beam can create a spatially varying refractive index in glasses. As
the nonlinear refractive index n2 is positive in glasses, the refractive index is higher
at the center of the beam compared to the wings. This variation in refractive index
11 Laser Writing in Tellurite Glasses 265

creates a positive lens which focuses the beam inside a dielectric with a strength
dependent on the peak power. If the peak power of the femtosecond laser pulses
exceeds the critical power for self-focusing [5]:

3:77λ2
Pc ¼ ð11:3Þ
8πn0 n2

a collapse of the pulse to a point is predicted. However, as the beam self-focuses,

the increased intensity drives nonlinear ionization which creates a free electron
plasma, which acts as a diverging lens that counters self-focusing. A balance
between self-focusing and plasma defocusing leads to filamentary propagation,
which results in elongated refractive index structures, which are undesirable for
transversely written waveguide structures (the standard geometry for waveguide
fabrication). Filaments can be exploited to fabricate waveguide devices by scanning
the sample parallel along the optical axis. Self-focusing can be avoided in wave-
guide fabrication by tightly focusing the laser beam with a microscope objective to
reach the intensity for optical breakdown without exceeding the critical power.
In fused silica, n0 ¼ 1.45 and n2 ¼ 3.5  10
20 m2/W [20] so that for
λ ¼ 1030 nm, the critical power is ~3 MW. From Eq. (11.3), the critical power is
proportional to the square of the laser wavelength; therefore, lower critical powers
result when working with the second harmonic wavelength of femtosecond
lasers. Also, the critical power is inversely related to the nonlinear (and linear)
refractive index, making it more difficult to form waveguides in tellurite glasses
with n0 ~ 2.0 and n2 ~ 10
18 m2/W [21] and chalcogenide glasses with n0 ~ 2.5,
n2 ~ 10
17 m2/W [22].
The properties of femtosecond laser-written waveguides in glass depend
strongly on laser exposure conditions such as scan speed, average power, and
numerical aperture. These parameters influence the net fluence, NF:

2w0 RFp
NF ¼ ð11:4Þ
v

where w0 is the waist radius, R is the repetition rate, Fp is the per-pulse fluence
(pulse energy per area), and v is the scan speed. Related to the dwell time, the
effective number of pulses per spot size during scanning is given by N ¼ 2w0(R/v).
Refractive index modification in femtosecond laser waveguide writing occurs
above a bulk modification threshold intensity, typically ~1013 W/cm2 in glasses
[5]. Above this threshold, increasing net fluence via decreased scan speed or
increased fluence results in an increased refractive index, and, at sufficiently high
net fluence, in irregular and overly damaged modification tracks.
266 T. Toney Fernandez et al.

11.3 Laser Writing in Tellurite Glasses

11.3.1 Near-Infrared Applications

The first ever attempt to produce a channel waveguide in a tellurite glass using a
laser was carried out by Chierici et al. [23]. They used a continuous wave 244 nm
doubled frequency Ar ion laser focused on the surface of a tellurite glass. The
composition of the glass was 67.5TeO2-18ZnO-4.5Na2CO3-10GeO2 with a refrac-
tive index of 2.0. Single mode (1550 nm) subsurface waveguides were produced
with positive index change of 1.5  10
3 with high waveguide propagation losses
of 8 dB/cm.
Focusing intense ultrashort pulses inside a high refractive index glass can trigger
nonlinear processes that are detrimental for producing good waveguides. The high
nonlinear third-order susceptibility of tellurite glasses results in filamentary prop-
agation leading to highly asymmetric structures not suitable for photonic applica-
tions [24]. Therefore, the initial investigations were mostly focused on controlling
these harmful filamentation effects. The first study of the interaction between
femtosecond laser and tellurite glass matrix was carried out by Sundaram et al.
[25]. The paper reported on the microexplosions and microstructuring caused by
interaction with focused femtosecond laser pulses and their resulting impact on the
chemistry and structure of tellurite glasses. For the study they used an 800 nm
wavelength laser with 1 kHz repetition rate and 130 fs pulse width, focused inside a
glass mainly composed of a 10Na2O-90TeO2 matrix modified with lanthanum
(2 mol%) or aluminum (1 mol%). The study revealed that the elements migrated
away from the laser-affected zone forming a less dense region without appreciable
structural changes, with the energy dispersive X-ray spectroscopy (EDX) charac-
terization performed along the longitudinal section of the waveguide (parallel to the
laser irradiation direction). A first demonstration of positive index change and light
guiding in a femtosecond laser-written tellurite glass waveguide was reported by
Y. Tokuda et al. [26] in a niobium tellurite glass (xNb2O5 (1-x)TeO2 (x ¼ 5, 15, 23))
using a similar laser. An estimated Δn of 9  10
3 was reported for the waveguide
written with energy >0.16 μJ using a 0.55 NA objective. In this report, the authors
concluded that since in a niobium tellurite glass, a Te–O bond changes from
covalent to ionic nature with the increase in the amount of Nb, the highly intense
femtosecond laser pulses caused dissociation of the weak Te–O bonds followed by
the rearrangement and compaction of the glass structure, inducing the increase of
the refractive index. This was the first time it was demonstrated that modifying the
tellurite-based matrix could in fact help achieving a positive refractive index
change for laser-written waveguides.
Shimotsuma et al. [27, 28] carried out laser irradiation experiments in TeO2
single crystal and zinc-tellurite glass (20ZnO-80TeO2) using a 800 nm, 120 fs Ti:
sapphire laser (Fig. 11.3). The laser was focused using a 100 objective (0.95 NA)
at depths ranging between 20 and 200 μm beneath the glass surface. The experi-
ments were carried out with two different repetition rates, namely, 1 and 200 kHz to
11 Laser Writing in Tellurite Glasses 267

Fig. 11.3 Guide mode at

633 nm through a niobium
tellurite glass waveguide
[26]

Fig. 11.4 Optical microscope images of induced structure by 0.8 μJ laser pulses inside (a, b) zinc-
tellurite glass and (c, d) TeO2 single crystal at 1 and 200 kHz, respectively (e). Pulse number
dependence of the diameter of an induced structure inside zinc-tellurite glass during the 1.0 μJ
laser pulses irradiation at 200 kHz repetition rate [27]

clarify the effect of thermal diffusion on the morphology of induced structures. The
irradiation times were adjusted for two experiments, 1 s for 1 kHz hence delivering
1000 pulses and 1/125 s (i.e., 1600 pulses) for 200 kHz repetition rate. Since the
thermal diffusivity of TeO2 single crystal (1.3  10
6 m2/s) is larger than that of
zinc-tellurite glass (5.2  10
7 m2/s), the size of the structures induced in zinc-
tellurite glass showed a stronger dependence on repetition rate (Fig. 11.4) compared
to the TeO2 crystal, which showed negligible difference between the repetition
rates.
268 T. Toney Fernandez et al.

Fig. 11.5 (a) Optical microscope images (transverse view) of induced structures inside niobium
tellurite glass written with 1.0 μJ, 120 fs laser pulses at a 200 kHz repetition rate for 1 s at depths of
30, 60, 90, and 120 μm. (b) Dependence of the induced structure following the focused irradiation
of 2  105 pulses with 1.0 μJ on the various focal point depths. Symbols of (triangle) and (circle)
show the experimental data of niobium tellurite glass and fused silica glass, respectively. (c)
Dependence of the length of induced structure 60 μm below the surface for different pulse
energies [28]

The higher repetition rate-induced structures in zinc-tellurite glass were

analyzed by backscattering electron imaging and spectrum mapping imaging of
the elements in the glass sample. It was found that zinc diffuses away from the
center of the focal volume, whereas tellurium aggregates into the focal volume
indicating a bright central region (densified zone) in the backscattered electron
image due to Z-contrast (Z—atomic number).
In the same work, the authors also carried out waveguide inscription in niobium
tellurite glass (15Nb2O5-85TeO2) to demonstrate that the length of induced struc-
ture varies with the focal spot depth and pulse energy. The results are compiled in
Fig. 11.5.
Following these initial results, a series of works [29–33] were carried out in
phosphate-modified (20 mol%) tellurite glass matrix doped with erbium and ytter-
bium, eventually reporting a net gain and a total gain bandwidth of 100 nm covering
the entire C and L communication bands. Initial work in the phospho-tellurite glass
(67TeO2-30P2O5-1Al2O3–1.75La2O3–0.25Er2O3 (all in mole %)) was carried out
with a low repetition rate laser of 1 kHz and 45 fs pulse width operating at a
wavelength of 800 nm [8]. Waveguides were written using a 0.45 NA objective,
3 μJ energy and translation speeds between 10 and 30 μm/s. 2 dB/cm propagation
loss was measured at 1310 nm in the best case scenario. Figure 11.6a shows the
optical microscope image of the transverse section of the waveguides, and
Fig. 11.6b shows the 1310 nm propagated laser mode. The explanation for the
positive index modification was supported by the energy dispersive X-ray analysis
in and around the laser-affected zone (along the transverse section of the wave-
guide, parallel (perpendicular) to the waveguide direction (laser irradiation)). The
paper finally reports a decrease in the La and P atomic concentration at the guiding
region; hence the modulation of Te:P ratio resulted in the positive index change.
11 Laser Writing in Tellurite Glasses 269

Fig. 11.6 (a) Optical microscope images of waveguide written with speeds 10 μm/s (top channel),
20 μm/s (middle channel), and 30 μm/s (bottom channel) scan speed. (b) Mode-profile of 1310 nm
laser light from the channel (taken from [29])

Upto this point, there had been no waveguides demonstrated within the cumu-
lative heating regime [1] in tellurite glass, even though heat accumulation in single
spot irradiation had been shown in zinc-tellurite glass with a thermal diffusivity of
5.2  10
7 m2/s (comparable to those of fused silica and borosilicate having values
of 8.4  10
7 and 5.1  10
7 m2/s, respectively [34]). The lack of studies with
MHz repetition rate conditions could be attributed to the commonplace of Ti:
sapphire 1-kHz fs-laser amplifiers in laboratories at the time. In 2008, a high
repetition rate fs-laser was used to produce waveguides in a phospho-tellurite
glass doped with 0.5 wt% erbium [30]. A diode-pumped cavity-dumped Yb:
KYW femtosecond laser oscillator, operated at 600 kHz having a pulse width of
about 350 fs at a wavelength of 1040 nm, was employed. Due to heat accumulation
effects, larger waveguide structures were produced (Fig. 11.7), and the waveguide
writing speeds improved dramatically compared to the previous low repetition rate
reports. A propagation loss of 1.35 dB/cm and internal gain within the C+L
communication band was reported for the first time (Δn ¼ 10
3).
Following this report, improved results were communicated in 2010 using an
optimized base glass with erbium and ytterbium co-doping [31, 32]. This time a
Yb:KYW femtosecond laser with 1-MHz repetition rate (model High Q
femtoREGEN), 1040-nm wavelength, and 400-fs pulse duration was used. The
improvements were quite remarkable as the best waveguide demonstrated a prop-
agation loss of only 0.9 dB/cm, the refractive index change was 2.5  10
3, and the
best waveguides were written at speeds of 2–6 mm/s. A maximum internal gain
value of 3.2 dB was achieved at the erbium peak, yielding for the first time a net
gain (~1 dB) and enabling the waveguide to operate as an optical amplifier device
(Fig. 11.8a). The gain saturation profile, which describes the small-signal internal
gain as a function of the incident pump power (Fig. 11.8b) at 1534, 1550, and
1610 nm wavelength, shows an excellent inversion of active ions over the entire
waveguide length along with a good improvement in the optical transparency at
each wavelength compared to its earlier report. Though the phospho-tellurite
passive glass matrix was already optimized, a further optimization in the active
ion doping concentration was required. This was due to the fact that the presence of
270 T. Toney Fernandez et al.

Fig. 11.7 (a) Differential interference contrast microscope images, of the transverse section of the
waveguides written with a speed of 0.1 mm/s. (b) Comparison between the heat-accumulated
zones when the scan speed is increased from 0.1 to 1 mm/s at a constant pulse energy of 420 nJ.
The fs-laser for inscription was incident along the +y axis [30]

Fig. 11.8 (a) Measured absorption spectrum (black dots) of the 21-mm long active waveguide in a
phospho-tellurite glass matrix co-doped with Er and Yb along with the enhancement (red dashed
line) and the internal gain (blue solid line) at 400 mW incident pump power. (b) Internal gain at
1534 nm (squares), 1550 nm (circles), and 1610 nm (triangles) as a function of incident pump
power [32]
11 Laser Writing in Tellurite Glasses 271

Fig. 11.9 Measured

enhancement and
absorption in two different
active Er:Yb waveguides in
phospho-tellurite glass with
two different Yb3+
co-doping level. The
waveguide length was
22 mm

phosphorous greatly enhances Yb to Er transfer efficiency in the phospho-tellurite

glass matrix as highlighted in Fig. 11.9. Doubling the Ytterbium co-doping con-
centration increased the signal enhancement by 40% maintaining the same absorp-
tion values, indicating an improved and efficient Yb3+ to Er3+ transfer.
The same laser-written waveguides were analyzed to study the correlation of ion
migrations and structural changes, directionality of migrating ions and types of
migrating ions in the tellurite glass waveguides, giving further insight into the
underlying physics during the femtosecond laser pulse-matter interaction [33]. The
responsible factor for the densification was the strong cross migrations of elements
within and outside the guiding region. In contrast to the migration observed in the
low repetition rate laser writing in phospho-tellurite glass, the tellurium was found
to migrate toward the guiding region and therefore densifying it. The rarefied zones
were enriched by the monovalent sodium ions, resulting in a reduced refractive
index. From the best writing condition window, two waveguides having signifi-
cantly different morphologies were chosen for electron probe microanalysis.
Waveguide A, written with a lower writing speed, has a single positive and negative
index change zones, whereas waveguide B has a single positive index change
region sandwiched between two negative index contrast zones (Fig. 11.10).
Raman spectroscopy carried out on the waveguides revealed that the increase in
refractive index was due to the local increase of Te, enabled by the conversion of its
trigonal bipyramids (tbp) to trigonal pyramids (tp), increasing the packing fraction
and its local density. This is summarized in Table 11.1 showing the magnitude of
refractive index change and relative Te ion increase. The large contrast in FWHM
variation of TeO4 and TeO3 favors the model that indicates densification with
conversion from tbp to tp units. Comparatively tp units have shorter average
bond length (including a short double bond), increasing the packing fraction with
large relative increase of atomic mass of Te and Na (Na—22.99, Te—127.6). These
results should help to improve the refractive index change obtained in tellurite glass
272 T. Toney Fernandez et al.

Fig. 11.10 Line scans showing strong ion migration in (a) waveguide A and (b) waveguide B
along with respective (a1) and (b1) secondary electron images [33]. The waveguides have been
inscribed in a phospho-tellurite glass matrix. Waveguide A has been written with a lower speed
than B

Table 11.1 Refractive index change, Te ion increase, and changes to Raman bands for wave-
guides A and B (taken from [33])

Te ion relative TeO4 (cm
1) TeO3 (cm
1)

Waveguide Δn increase (%) Peak shift FWHM Peak shift FWHM
A 3.5  10
3 7.4
15
18
9 +5.5
B 2.5  10
3 6.1
9
14
7 +4.5

waveguides by optimizing the Te-Na ratio within the glass matrix. Additionally this
should also help achieving a faster waveguide writing as the optimization of ion
migration should reduce the fluence needed to achieve the desired refractive index
change.
Other related works were by Ramme et al. [34] in a 5Na2O-20ZnO-75TeO2-
2La2O3 glass, which focused on the fundamental light radiation-glass interaction.
A femtosecond laser operating at a wavelength of 1043 nm was used, and for this
study the following parameters were varied: (1) repetition rate between 0.1 and
1 MHz, (2) pulse width 350–450 fs, (3) writing speed from 0.5 to 2 mm/s, and
(4) three different focusing objectives 40 (0.65 NA), 20 (0.4 NA), and 10
(0.25 NA). The maximum induced index change was approximately 2  10
4 for a
pulse energy of 250 nJ and the 20 objective. For increased pulse energies, the
creation of a negative index change combined with the generation of voids was
observed (Fig. 11.11).
Gross et al. [35] studied laser inscription in a tellurite glass (73TeO2–20ZnO–5
Na2O–2La2O3 mol%) in both athermal and thermal regimes. In the athermal
regime, the range of pulse energies that resulted in a controlled index change in
tellurite glass was highly restricted to 30–40 nJ using a 20 objective. In the
thermal regime, a 5.1 MHz, 800 nm, 50 fs Ti:sapphire laser was used to write
waveguides with a 100 (1.25 NA) oil immersion objective. The produced struc-
tures were continuous and smooth. With translation speed ranging between 100 and
11 Laser Writing in Tellurite Glasses 273

Fig. 11.11 DIC image of a waveguide cross-section, showing the modification with filamentation
in a tellurite glass modified with Na, Zn, and La. Laser inscription was obtained with a 10
focusing objective and 750 nJ pulse energy; the translation direction was normal to the page [34]

Fig. 11.12 DIC images of the waveguides in longitudinal and transverse cross-section of the
laser-written waveguides in tellurite glass [35]

2500 mm/s and energies between 10 and 50 nJ generated uncracked smooth

waveguides, but led to an undesirable negative index change.
Olivero et al. [36] using a low repetition rate laser operating at 1 kHz with 150 fs
pulses demonstrated an internal gain of 0.6 dB/cm at 1535 nm in zinc-tellurite glass
doped with 2 wt% erbium and 0.5 wt% ytterbium.
The major drawback for the progress of laser-written active devices in tellurite
glass was not specifically an intrinsic problem, but ion migration related to laser
writing which is discussed above. It was recently demonstrated that due to ion
migration, the active ion also migrate, affecting the performance for active device
applications. This detrimental migration will have a more complex effect for low
phonon energy glasses like tellurites, tellurides, and chalcogenides, as these glasses
have a poor energy transfer rate coefficient. A detailed discussion with simulations
and experimental data can be found in [37].
It could be clearly understood from the different development stages, tellurite
glass was one of the interesting laser inscribable glasses that matured along with the
development of ultrafast lasers for waveguide writing. The major challenge was the
lack of availability of good optical quality commercial glasses which lagged its
forward advancement. Since 2010 there has been a surge of new low phonon-
energy commercial glasses such as tellurites, tellurides, and chalcogenides from
glass manufacturers aiming for the burgeoning mid-infrared device market.
274 T. Toney Fernandez et al.

11.3.2 Mid-Infrared Applications

Fluoride-based glass matrices are the most popular host matrices for active
mid-infrared applications, but tellurites are becoming strong contenders mainly
because of their wider operational bandwidth and chemical and mechanical stabil-
ity. The only literature which is so far available reports a preliminary attempt to
produce waveguides within a Dy3+-doped zinc-tellurite glass [38]. The work was
carried out with a low repetition rate (1 kHz) Ti:sapphire laser at a wavelength of
800 nm and 100 fs pulse width. Laser inscription was performed using a 0.65 NA
aspheric lens objective with various powers ranging from 300 nJ to 5 μJ and writing
speeds from 0.01 to 6 mm/s. The maximum refractive index change was calculated
to be around 610
3. A fiber pigtailed 808 nm laser diode source was butt coupled
to obtain the amplified spontaneous emission (ASE) from the waveguide. The
mid-infrared ASE spectrum of the Dy3+ tellurite waveguide largely matches the
line shape of the spontaneous fluorescence from bulk Dy3+ tellurite glass, except
with slightly enhanced intensity around 3 and 3.9 μm, suggesting potential enhance-
ments in the bandwidth of this transition in waveguiding structures, crucial for
future waveguide and fiber laser applications. Once again as outlined in the
previous section, the lack of availability of low OH content tellurite glasses
(<10 ppm) is a stumbling block, but currently there are emerging spinoffs and
existing companies that are trying to fill this void.
A target focus region for mid-infrared active channel waveguides would be 3–4
μm where the characteristic fundamental vibration bands (molecular fingerprints)
of most molecules lie, thus making mid-infrared spectroscopy a univocal way to
identify and quantify molecular species, including isotopologues, in a given envi-
ronment. Current NIR detection techniques are based on molecular vibration over-
tones which lacks specificity and sensitivity, which makes mid-IR spectroscopy far
more superior. Additionally, mid-IR spectroscopy provides not only a powerful tool
for understanding the structure of molecular matter and its governing physical laws
but also for performing non-intrusive diagnostics of composite systems of physical,
chemical, or biological interest, in the gas, liquid, or solid phase. The development
of tunable, compact, and high power laser sources for the mid-infrared region still
remains a major challenge simply due to technical reasons, and hence a large
portion of the high potential mid-IR spectrum remains inaccessible.
A recent attempt to introduce tellurite fiber into the mid-IR light source inven-
tory was limited by the presence of high OH content [39]. The broad wavelength
fluorescence line width that could be achieved from the Te matrix (290–500 nm at
3 μm for Dy3+) can be over two times that from the ZBLAN matrix (225 nm at 3 μm
for Dy3+). This is the major feature that could be made use of in the future by the
tellurite active waveguides to produce ultrashort pulses in the mid-infrared wave-
length regime.
11 Laser Writing in Tellurite Glasses 275

References

1. S.M. Eaton et al., Heat accumulation effects in femtosecond laser-written waveguides with
variable repetition rate. Opt. Express 13(12), 4708–4716 (2005)
2. B.C. Stuart, M.D. Feit, et al., Nanosecond-to-femtosecond laser-induced breakdown in dielec-
trics. Phys. Rev. B 53(4), 1749 (1996). doi:10.1103/PhysRevB.53.1749
3. R.R. Gattass, E. Mazur, Femtosecond laser micromachining in transparent materials. Nat.
Photon. 2(4), 219–225 (2008). doi:10.1038/nphoton.2008.47
4. L.V. Keldysh, Ionization in the field of a strong electromagnetic wave. Sov. Phys. JETP 20,
1307–1314 (1965)
5. C.B. Schaffer, A. Brodeur, E. Mazur, Laser-induced breakdown and damage in bulk transpar-
ent materials induced by tightly focused femtosecond laser pulses. Meas. Sci. Technol. 12(11),
1784–1794 (2001)
6. D. Du et al., Laser-induced breakdown by impact ionization in fused silica with pulse widths
from 7 ns to 150 fs. Appl. Phys. Lett. 64(23), 3071 (1994). doi:10.1063/1.111350
7. K.M. Davis et al., Writing waveguides in glass with a femtosecond laser. Opt. Lett. 21(21),
1729–1731 (1996). doi:10.1364/OL.21.001729
8. K. Miura et al., Photowritten optical waveguides in various glasses with ultrashort pulse laser.
Appl. Phys. Lett. 71(23), 3329–3331 (1997). doi:10.1063/1.120327
9. S. Richter et al., On the fundamental structure of femtosecond laser-induced nanogratings.
Laser Photon. Rev. 6(6), 787–792 (2012)
10. L. Sudrie et al., Writing of permanent birefringent microlayers in bulk fused silica with
femtosecond laser pulses. Opt. Commun. 171(4–6), 279–284 (1999)
11. Y. Shimotsuma et al., Self-organized nanogratings in glass irradiated by ultrashort light pulses.
Phys. Rev. Lett. 91(24), 247405 (2003)
12. V.R. Bhardwaj et al., Optically produced arrays of planar nanostructures inside fused silica.
Phys. Rev. Lett. 96(5), 057404 (2006)
13. E.N. Glezer, E. Mazur, Ultrafast-laser driven micro-explosions in transparent materials. Appl.
Phys. Lett. 71(7), 882 (1997)
14. K. Itoh et al., Ultrafast processes for bulk modification of transparent materials. MRS Bull. 31
(8), 620–625 (2006)
15. T.F. Johnston, Beam propagation (M2) measurement made as easy as it gets: the four-cuts
method. Appl. Optics 37(21), 4840–4850 (1998)
16. C. Hnatovsky et al., High-resolution study of photoinduced modification in fused silica
produced by a tightly focused femtosecond laser beam in the presence of aberrations.
J. Appl. Phys. 98(1), 013517 (2005)
17. S.M. Eaton et al., Transition from thermal diffusion to heat accumulation in high repetition rate
femtosecond laser writing of buried optical waveguides. Opt. Express 16(13), 9443–9458
(2008)
18. R. Osellame et al., Optical properties of waveguides written by a 26 MHz stretched cavity Ti:
sapphire femtosecond oscillator. Opt. Express 13(2), 612–620 (2005)
19. R. Osellame et al., Waveguide lasers in the C-band fabricated by laser inscription with a
compact femtosecond oscillator. IEEE J. Sel. Top. Quant. Electron. 12(2), 277–285 (2006)
20. L. Sudrie et al., Femtosecond laser-induced damage and filamentary propagation in fused
silica. Phys. Rev. Lett. 89(18), 186601 (2002)
21. R.A.H. El-Mallawany, Tellurite Glasses Handbook: Physical Properties and Data (CRC
Press, Boca Raton, FL, 2001)
22. Ta’eed V et al. (2007) Ultrafast all-optical chalcogenide glass photonic circuits. Opt. Express,
15(15):9205–9221.
23. E Chierici et al. Direct writing of channel waveguide on a tellurite glass using a focused
ultraviolet laser beam. Proceedings of 2002 IEEE/LEOS Workshop on Fibre and Optical
Passive Components, University of Glasgow, UK, 6–8 June 2002.
276 T. Toney Fernandez et al.

24. M. Liao et al., Ultrabroad supercontinuum generation through filamentation in tellurite glass.
Laser Phys. Lett. 10, 036002 (2013)
25. S.K. Sundaram et al., Microexplosions in tellurite glasses. Appl. Phys. A 76(3), 379–384
(2003)
26. Y. Tokuda et al., Waveguide formation in niobium tellurite glasses by pico- and femtosecond
laser pulses. J. Non Cryst. Solids 326–327, 472–475 (2003)
27. Y. Shimotsuma et al., Nano-modification inside transparent materials by femtosecond laser
single beam. Mod. Phys. Lett. B 19(5), 225–238 (2005)
28. Y. Shimotsuma et al., Nanofabrication in transparent materials with a femtosecond pulse laser.
J. Non Cryst. Solids 352(6–7), 646–656 (2006)
29. P. Nandi et al., Femtosecond laser written channel waveguides in tellurite glass. Opt. Express
14(25), 12145–12150 (2006)
30. T. Toney Fernandez, Active waveguides written by femtosecond laser irradiation in an erbium-
doped phospho-tellurite glass. Opt. Express 16(19), 15198–15205 (2008)
31. S.M. Eaton, et al. 100-nm internal gain bandwidth in Er: Yb-doped phospho-tellurite wave-
guides written by femtosecond laser. Lasers and Electro-Optics (CLEO) and Quantum
Electronics and Laser Science Conference (QELS), San Jose, California, 16–21 May 2010.
32. T. Toney Fernandez, Femtosecond laser written optical waveguide amplifier in phospho-
tellurite glass. Opt. Express 18(19), 20289–20297 (2010)
33. T. Toney Fernandez, Role of ion migrations in ultrafast laser written tellurite glass wave-
guides. Opt. Express 22(12), 15298–15304 (2014)
34. Ramme M et al. Waveguide writing and characterization in tellurite glass. Conference on
Lasers and Electro-Optics, Baltimore, US, 31 May – 5 Jun 2009
35. S. Gross et al., Ultrafast laser inscription in soft glasses: a comparative study of athermal and
thermal processing regimes for guided wave optics. Special issue: glass and photonics – an
overview. Int J Appl Glass Sci 3(4), 332–348 (2012)
36. M. Olivero, Amplification properties of femtosecond laser-written Er3+/Yb3+ doped wave-
guides in a tellurium-zinc glass. Adv. Opt. Technol. 2013, 621018 (2013)
37. T. Toney Fernandez et al., Dual regimes of ion migration in high repetition rate femtosecond
laser inscribed waveguides. IEEE Photon. Technol. Lett. 27(10), 1068–1071 (2015)
38. B.D.O. Richards et al., Mid-IR (3–4 μm) fluorescence and ASE studies in Dy3+ doped tellurite
and germanate glasses and a fs laser inscribed waveguide. Laser Phys. Lett. 10, 085802 (2013)
39. L. Gomes et al., Spectroscopy of mid-infrared (2.9 μm) fluorescence and energy transfer in
Dy3+-doped tellurite glasses. J. Opt. Soc. Am. B 31(3), 429–435 (2014)
40. V. Diez-Blanco et al., Deep subsurface waveguides with circular cross section produced by
femtosecond laser writing. Appl. Phys. Lett. 91, 051104 (2007)
Chapter 12
Supercontinuum Generation in Tellurite
Optical Fibers

Frédéric Désévédavy, Gregory Gadret, Jean-Charles Jules,

Bertrand Kibler, and Frédéric Smektala

12.1 Bibliographic Data

Table 12.1 gathers the data on which this review work is based. For each reference
article, the following results are presented: the supercontinuum bandwidth mea-
sured, the pump conditions (pump wavelength and power), the zero dispersion
wavelength (ZDW) of the waveguide, the fiber geometry, and the glass composition
(when known). We do not pretend to be exhaustive, but, to the best of our
knowledge, the main reference articles in the field are taken into consideration
and are presented in reverse chronological order.

12.2 Brief Considerations on Tellurite Glass Compositions

for Supercontinuum Generation in Optical Fibers

Tellurite glasses present quite high linear refractive indices, around 2, while the
nonlinear refractive indices are varying between 1.7.10
19 m2 W
1 and 5.9.10
19
m2 W
1 [26, 28, 29, 32] at 1.55 μm, which is one order of magnitude above the
nonlinearity of silica glass, highly suitable for nonlinear applications such as
supercontinuum generation.
The first work on tellurite glass was carried out by Stanworth [33], and the first
characterization of optical properties on tellurite fibers was reported by
Wallenberger [34] on TeO2-Al2O3 system and by Wang [35] on the TeO2-ZnO-

F. Désévédavy • G. Gadret • J.-C. Jules • B. Kibler • F. Smektala (*)

Laboratoire Interdisciplinaire Carnot de Bourgogne ICB, UMR 6303 CNRS-Université de
Bourgogne Franche-Comté, 21078 Dijon, France
e-mail: [email protected]; [email protected]; jean-charles.
[email protected]; [email protected]; [email protected]

© Springer International Publishing AG 2017 277

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_12
278 F. Désévédavy et al.

Table 12.1 Experimental supercontinuum generation in tellurite fibers, bibliographic data

Pump
Measured (wavelength/ Fiber Glass composition
SC (μm) power) ZDW (μm) geometry (mol. %) Reference
0.6–2.8 1.5 μm/ 1.48 SCF 70TeO2- 20ZnO- 9BaO- [1]
2–3.4 12.4 kWa 1BaCl2
2.4 μm/
6.8 kWa
0.6–3.3 1.73 μm/ 1.6 T 75TeO2-15ZnO-5ZnF2- [2]
12 kWa Na2O
0,9–3.9 1.958 μm/ 1.74 SCF 76.5TeO2 -6ZnO- [3]
1–2.8 9.2 kWb 11.5Li2O-6Bi2O3
2 μm/
22 kWa
1.92 μm/ 1.8 and 2.1 SI 70TeO2 – 5La2O3- [4]
4.7 kWb 20ZnO-5ZnF2
1.3–2.4 1.56 μm/ 1.689 SCF 78TeO2-5ZnO-12Na2O- [5]
19 kWa 5Bi2O3
0.6–6 1.6 μm/ 2.0 Bulk 72TeO2–18ZnO– [6]
0.63GW (7 mm) 5Na2O–5La2O3
69.8TW/cm
2a

0.84–3.0 1.745 μm/ 1.55 SCF 80TeO2-(10-x)ZnO- [7]

6.5 kWa xZnF2–10Na2O.
x ¼ 0, 5, 10
0.5–5 1.92 μm/ 1.92 SI 70TeO2–5La2O3- [8]
4.7 kWb 20ZnO-5ZnF2
0.5–5 1.92 μm/ 1.92 SI 70TeO2–5La2O3- [9]
4.7 kWb 20ZnO-5ZnF2
0.8–2.5 1.58 μm/ 1.41 and 4.24 MRH 65TeO2-28WO3–7 [10]
230 kWa (Na2O + Nb2O5)
0.65–2.85 1.96 μm/ 1.358 MRH 76.5TeO2-6Bi2O3- [11]
580 Wb 6ZnO-11.5Li2O
1.56 μm/
17 kWa
Generated Pump ZDW (μm) Fiber Glass composition (mol. Reference
SC (μm) (wavelength/ geometry %)
power)
1.4–2.4 1.56 μm/ 0.88 and 1.99 SCF Unspecified [12]
37 kWa
0.8–2 1.56 μm/ 1.545 SCF Unspecified [13]
3.25 W
0.75–2.8 1.745 μm/ 1.66 SCF 80TeO2 -10ZnO- [14]
7 kWa 10Na2O
0.6–1.8 1.064 μm/ 1.51 and 0.94 SCF 76.5TeO2-6Bi2O3- [15]
387 Wb 6ZnO-11.5Li2O
0.73–1.7 1.064 μm/ 1.10 and 1.53 T Unspecified [16]
28 Wb
(continued)
12 Supercontinuum Generation in Tellurite Optical Fibers 279

Table 12.1 (continued)

Pump
Measured (wavelength/ Fiber Glass composition
SC (μm) power) ZDW (μm) geometry (mol. %) Reference
1–1.8 1.56 μm/ 1.45 SCF 80TeO2-10ZnO- [17]
1.3 kWa 10Na2O
0.9–1.6 1.064 μm/ 1.19 and 0.75 SCF 76.5TeO2–6Bi2O3 – [18]
24 Wb 6ZnO–11.5Li2O
1.064–1.6 1.064 μm/ 1 and 1.5 SCF 76.5TeO2–6Bi2O3 – [19]
28 Wb 6ZnO–11.5Li2O
1.2–2.4 1.56 μm/ 1.55 SCF + SI 70TeO2 –8Li2 O– [20]
3.8 kWb 17WO3 –3MoO3 –
2Nb2O5 (core)
75TeO2 –15ZnO–5Na2
O–5La2O3 (clad)
1.15–2.25 1.55 μm/ 1.434 SCF 76.5TeO2–6Bi2O3 – [21]
2.2 kWa 6ZnO–11.5Li2O
0.6–2.4 1.55 μm/ 1.434 T 76.5TeO2–6Bi2O3 – [22]
13 kWa and 1.029 6ZnO-11.5Li2O
0.47–2.4 1.55 μm/ 1.337 SCF 76.5TeO2–6Bi2O3 – [23]
9.4 kWa 6ZnO–11.5Li2O
0.8–2.4 1.557 μm/ 1.608 SCF Unspecified [24]
12.4 kWa
0.8–1.6 1.064 μm/ 0.88 and 1.61 SCF 76.5TeO2–6Bi2O3 – [25]
33 Wb 6ZnO–11.5Li2O
1.557 μm/
1443 Wa
Generated Pump ZDW (μm) Fiber Glass composition (mol. Reference
SC (μm) (wavelength/ geometry %)
power)
0.8–>2.4 1.557 μm/ 1.39 MRH 76.5TeO2–6Bi2O3 – [26]
1.7 kWa 6ZnO–11.5Li2O
1–2.1 1.1–1.6 μm/ 1.275 SCF Unspecified [27]
19 MWa
0.789–4.87 1.55 μm/ 1.380 SCF 75TeO2 -12ZnO-5PbO– [28]
17 kWa 3PbF2–5Nb2O5
0.9–2.5 2.15 μm/ 2.15 MRH 75TeO2 -20ZnO-5Na2O [29]
127 kWa
0.6–3 1.55 μm/ 1.4 SCF 75TeO2 -12ZnO - 5PbO [30]
17 kWa – 3PbF2 – 5Nb2O5
0.9–2.5 1.5 μm/ 1.4 SCF 75TeO2 – 20ZnO- [31]
11.7 kWa 5Na2O
SCF (suspended core fiber), MRH (multiple ring hole), SI (step-index fiber), T (tapered fiber)
a
femtosecond source
b
picosecond source.
280 F. Désévédavy et al.

Na2O system. While many glass compositions exist, only some of them are suitable
for fiber drawing because of their good thermal stability. For fiber drawing pur-
poses, a commonly accepted criterion is the temperature difference ΔT between the
glass transition temperature Tg and the crystallization temperature Tx which must
be greater than 100  C. The thermal stability ΔT of a glass is indeed the first
parameter determining the choice of glass compositions for fiber drawing.
From Table 12.1, typical compositions of tellurite glasses, suitable for fiber
drawing and subsequent supercontinuum generation, contain 70–80 mol. % of
tellurium oxide TeO2 as the glass former and 20–30 mol. % of glass modifier or
intermediate oxides such as ZnO, WO3, Li2O, or Na2O for the common ones. Most
of the used compositions belong then to ternary or quaternary systems, and two
main families of glasses can be identified: TeO2-ZnO-R2O and TeO2-WO3-R2O
where R is Li, Na, or K (see Table 12.1).
Another important parameter is the glass transparency at high wavelengths. One
can note that for infrared (IR) supercontinuum (SC) generation applications, the
modifier or intermediate oxides are preferably chosen so that their phonon energy is
below the phonon energy of TeO2 [4]. For illustration, Fig. 12.1 shows the trans-
mission spectra of pure TeO2 paratellurite mono-crystal and two binary glasses,
80TeO2–20ZnO and 85TeO2–15WO3 (mol. %).
The addition of WO3 clearly shifts the multiphonon absorption edge toward the
shorter wavelengths, and for this reason, despite its positive effect on the glass
nonlinearity, WO3 is actually used in only few compositions (see Table 12.1).
The infrared glass transparency depends on both the glass multiphonon edge
location and the possible presence of extrinsic –OH hydroxyl groups. The latter
comes from the raw materials or the atmosphere and is incorporated into the glass

Fig. 12.1 Transmission

spectra of monocrystalline
TeO2 and two binary
tellurite glasses TeO2-ZnO
and TeO2-WO3
12 Supercontinuum Generation in Tellurite Optical Fibers 281

network during the synthesis. The –OH group vibrations result in broad and intense
absorptions located around 3.3 and 4.4 μm (Fig. 12.1). Besides the multiphonon
edge, these –OH groups, even in limited quantities, are responsible for the infrared
transmission limitation. Atmospheric moisture is an evident source of high water
content in the glass melted in ambient atmosphere. Hence, the glass synthesis is
generally performed using a platinum or a gold crucible in a dry oxidizing-
atmosphere-filled furnace connected to a glove box. However, even in this case,
the residual –OH is sometimes sufficient to prevent fiber transmission in the
mid-infrared (mid-IR) range beyond 3 μm. Sodium oxide is known to decrease
glass durability, and several teams are working on Na-free or Na-poor composi-
tions. For compositions containing BaO, Li2O, and Bi2O3, it appears that the
amplitude of OH bands decreases [36] compared to compositions in which Na2O
and ZnO are used. Liao [15, 25, 26], Qin [22, 23], and Cheng [3] replace Na2O by
Li2O and a part of ZnO by Bi2O3. Belal [1] replaces Na2O by BaO. Another way to
remove contaminating hydroxyl groups is to use halide chemical reagents like
ZnF2, PbF2, or BaCl2 [1, 4, 7, 30, 36]. For illustration, Fig. 12.2a shows the
transmission spectra of 4-mm-thick samples of 80TeO2-10ZnO-10Na2O (mol. %)
glass prepared in ambient atmosphere and in a glove box, as well as the transmis-
sion of an 80TeO2-5ZnO-5ZnF2-10Na2O (mol. %) sample prepared in a glove
box [7].
Figure 12.2b depicts the attenuation spectra measured by the cutback technique
on fibers drawn from the previous glasses [7]. A high absorption band located
around 3350 nm (2985 cm
1) is identified as a combination of weakly H-bonded –
OH and free –OH (Fig. 12.2a). The lower absorption band around 4350 nm
(2300 cm
1) is due to strongly H-bonded –OH. As seen in Fig. 12.2a, the presence
of OH absorption bands significantly decreases the transmission above 3 μm for the
sample cast in air. The glass fabricated in a glove box under a dry oxygen gas flow

Fig. 12.2 (a) Transmission of tellurite bulk samples as a function of the synthesis procedure and
(b) corresponding fibers attenuations
282 F. Désévédavy et al.

exhibits a lower absorption at 3.3 μm when it is strongly reduced for the ZnF2-
containing glass leading to 77% transmission from 1500 nm to 4130 nm. The
spectral transmission window for the corresponding fibers is enlarged toward the
mid-IR (see Fig 12.2b). One can note however that during the synthesis of fluorine-
containing glasses, depending on the amount of the used fluoride reagent, fluoride
ions are incorporated into the glass network and can lead to decrease of the thermal
stability. In the above glasses [7], the ΔT parameter decreases from 165  C for the
fluoride-free glass down to 110  C for the glass composition containing 5 mol. %
ZnF2, and the glass preform drawing then becomes a more tricky process. A
solution is to better control the remaining fluoride amount in the glass, and,
recently, Chavez-Pirson [4] obtained a final glass composition containing only
0.5% of ZnF2.

12.3 Fiber Structures

We report mainly four structures of tellurite fibers used for supercontinuum

(SC) generation: suspended core fibers (SCF), all solid fibers (SF), and multiple
ring hole (MRH) fibers, i.e., more than one ring of low index inclusions and tapered
microstructured fibers (T). These different kinds of fibers are presented in Fig. 12.3.
Fabrication techniques employed in the works reported in Table 12.1 and used to
draw the fibers are mainly based on the preform-to-fiber approach [38] (Fig. 12.3).
One can note however that more generally speaking, for tellurite fibers, the different
fabrication processes are extrusion process [28, 29, 39], rod in tube [8, 18], drilling
[14], and stack and draw [10, 40].

Fig. 12.3 Tellurite fibers used for SC generation: (a) suspended core [31], (b) all solid [8], (c)
multi-rings of holes [10], and (d) taper [37]
12 Supercontinuum Generation in Tellurite Optical Fibers 283

To date, most of the research efforts on SC generation in tellurite fibers have

been performed on suspended core ones. In these fibers the high refractive index
small core is suspended in a low index material, mostly air. The core is surrounded
by longitudinally holes (generally 3), as a consequence it holds in place by thin
struts of a hundred of nanometers wide and several microns long [28, 31]. This
geometry gathers several properties suitable for SC generation: high light confine-
ment, small core inducing small mode effective area and a high nonlinear coeffi-
cient, dispersion management, short nonlinear length, and a relative fabrication
convenience whatever the fabrication process.
The core diameter and hole size can be adjusted to manage the waveguide
chromatic dispersion. In order to optimize the supercontinuum generation, anom-
alous dispersion management is required for soliton generation and phase matching
with dispersive wave [11, 23]. The decreasing of the SCF core (down to 1–3 μm)
not only increases the fiber nonlinearity but also decreases the zero dispersion
wavelength (ZDW), allowing the fiber pumping in its anomalous dispersion regime
with common optical sources (1.55 or 2 μm); see Fig 12.4a. To finely tune the
dispersion, it is necessary to control the hole’s size. Indeed, it is known that a lower
effective index difference between the fiber core and the cladding gives a smaller
dispersion slope and further leads to a smaller value of high-order dispersion, which
is preferable for generating flattened SC [11, 27]. A lower effective index difference
is reached by decreasing the hole’s size, typically to a size comparable with core
diameter; see Fig. 12.4b. A smaller anomalous dispersion slope allows to use a
pump wavelength shifted from the fiber’s ZDW leading to a wider SC expending
more in the mid-IR [11].
The main drawbacks of this geometry are related to the small core size leading to
a more difficult light coupling. In suspended core fibers, the average coupling
efficiency is generally around 10–30% [1, 27, 31]. A fine control of the hole’s
size is also tricky, thus making the dispersion management uncertain, especially for
flat dispersion design. Another drawback is the possible aging of the fiber

a b
Core diameter (µm)
1.6 2.4 3.2 4.0
Nonlinear coefficient (/W/km)

200
1600
Dispersion (ps/(nm×km))

0
1200
Core diameter (µm)
-200 1.5 2.9
1.9 3.2 800
-400 2.4 3.6
4.2

-600 400

800 1000 1200 1400 1600 1800 2000

Wavelength (nm)

Fig. 12.4 Dispersion evolution (a) as a function of the core size with hole diameter >> core
diameter [13] and (b) with hole diameter of the same magnitude than the core one (here 1.2 μm)
[27]
284 F. Désévédavy et al.

depending on the glass composition. This aging phenomenon is due to the atmo-
spheric steam diffusion into the fiber’s holes and the following physicochemical
interaction with the fiber core [41]. Moreover, the light modes propagating through
the small core extend into the air holes and are then subject to water absorption or
other atmospheric absorbing species in the infrared [42].
These issues can be addressed by using other fiber geometries. Tapered fibers
(Fig. 12.3d) allow to launch the pump signal in a large core, then with a suitable
taper profile, the light is adiabatically coupled to the tapered region [43] which
presents opto-geometric properties (small core, ZDW) allowing supercontinuum
generation [2]. In this kind of fiber, deceleration of the solitons due to the changing
group velocities also enhances the soliton trapping of dispersive wave, indepen-
dently of Raman gain [22]. This results in both a blue shift and a red shift of the
supercontinuum. Moreover, the tapered segment not only greatly increases the
nonlinearity but acts as a compensation for the dispersion of the untapered
segment [16].
Multiple ring hole (MRH) fibers (Fig. 12.3c) have been barely investigated for
supercontinuum generation [10, 26, 29]. Hexagonal microstructure, also known as
holey fiber (HF), enables a broad control of both dispersion and waveguide losses of
guided modes. Supercontinuum generation from 0.8 to over 2.5 μm was reported in
a 2.7 μm core diameter HF fiber [10]. However to scale the output SC power up to
medium or high power levels, a nonlinear fiber with a larger mode area may be a
more practical option rather than a small core fiber, as the damage power thresholds
of the former would be considerably higher. Feng et al. [29] demonstrate IR
supercontinuum from 0.9 to 2.5 μm with 6 mW output power in a 9-cm-long
tellurite fiber with a mode area as large as 3000 μm2.
All solid fibers can offer high numerical aperture associated to a small core
leading to a high nonlinearity. Recently, step-index fibers have been used to
generate wide infrared supercontinuum [8, 9]. The authors used a depressed inner
clad fiber consisting of three layers: a core, an inner cladding, and an outer one
presenting refractive indices n0, n1, n2, respectively, with n1 < n2 < n0 (W-type
tellurite fiber). This allows a tight confinement of light and a dispersion manage-
ment in all solid fibers more convenient to manipulate compared to other ones.
Fiber robustness is improved, and there is no more atmospheric steam diffusion into
the glass core through the holes. The fiber can be designed for single-mode
guidance for the pump signal and the mid-IR wavelengths. The coupling is also a
drawback which can be improved by using again a tapered fiber in such a way that
the light is coupled in the large core of the untapered segment. The core/clad
interfaces may be a source of extra losses. These are decreased by using built-in
casting preforms preparation technique [44, 45] and are not necessarily a critical
issue insofar as the fiber length magnitude for SC generation is only of several
centimeters.
It is worth mentioning fibers combining two structures as the one of Duan et al.
[20], which can be designed as a microstructured step-index fiber. It is a composite
microstructured fiber made with two tellurite glasses, one for the solid core and one
for the solid cladding in which six air holes are incorporated. This novel fiber
12 Supercontinuum Generation in Tellurite Optical Fibers 285

structure provides new degrees of freedom for chromatic dispersion engineering.

The obtained flattened dispersion profile together with the high optical nonlinearity
enhance the performance of the tellurite fiber for SC generation.
Finally, one can report also the SC generation in Tm3+-doped tellurite micro-
structured fiber [5]. In comparison with SC generation in undoped tellurite
microstructured fiber, the SC spectral bandwidth and the spectral intensity in the
wavelengths region above 1.9 μm are enlarged in Tm3+-doped fibers owing to the
contribution of Tm3+ linear gain and nonlinear optical effect combination to
spectral broadening.

12.4 Pumping Conditions and Coupling Issues

The laser systems used for pumping can be either compact fiber laser devices
(λ ¼ 1.55 or around 2 μm), fiber laser diode (λ ¼ 1.55 μm), diode-pumped solid-
state (DPSS) Nd:YAG lasers (λ ¼ 1.064 μm, eventually fiber output), or Ti:
sapphire-pumped optical parametric oscillators (OPO) or optical parametric ampli-
fiers (OPA). The latter exhibit large and easy tuning possibilities, from visible to
mid-IR depending on the system. This can be a strong advantage when the zero
dispersion wavelength (ZDW) is not known with a sufficiently good precision.
Indeed, in order to generate a supercontinuum in a glass fiber (either specialty IR
glass or more classical silica glass), it appears necessary to meet several optical
requirements. The ZDW, depending on both the intrinsic glass (bulk) dispersion
and the fiber geometry, is a very important characteristic and will mostly define the
wavelength at which the fiber will be pumped. Taking into account the bulk
dispersion of the various compositions used in tellurite glasses, the calculated or
measured ZDW of resulting optical fibers appears mainly located between 1 and
2 μm, with most of them around 1.5 μm (see, for instance, [20]). When tapers are
produced, the ZDW is found shifted to higher energy, typically in the visible and
near IR range [15, 18]. The fibers can be pumped in normal (λ < ZDW) or
anomalous (λ > ZDW) dispersion regime in order to take profit of the related
nonlinear processes (Fig. 12.5).
In fact, most of the works found in the literature have been done using pump
wavelength less than 0.5 μm far from the ZDW, either in normal or anomalous
dispersion regime. Both regimes are tested in a few papers [18, 22].
Since the aim is to take profit of the glass optical nonlinearities, the other relevant
pump parameters are the pulse time duration TFWHM on one hand, which should be
short enough (ps-fs) to benefit from soliton dynamics, and on the other hand the peak
power of pump laser pulses, roughly proportional to Ep/TFWHM ratio, where Ep is the
energy per pulse. Pulse duration lies from 50 to 100 fs (fiber lasers, [11]) up to
600 femtoseconds (OPO/OPA) or 15–20 picoseconds (Tm or Tm-Ho fiber lasers
around 2 μm or DPSS at 1.064 μm). One particular work implements a quasi-CW
sub-microsecond source [13]. For OPO systems, the energy per pulse is strongly
correlated to the selected wavelength and is generally lower in the idler range (higher
286 F. Désévédavy et al.

Fig. 12.5 Pump wavelength versus zero dispersion wavelength (ZDW) for supercontinuum
generation from the works cited in Table 12.1

wavelengths, when technically available) than in the signal range (lower wave-
lengths). Since Ti:sapphire-pumped OPO/OPA systems are also quite expensive, the
choice of a pump laser system in terms of peak power should firstly be determined by
the choice of the pump wavelength. When the source is not homemade, holmium- or
holmium/thulium-doped fiber systems delivering up to 10 kW peak power seem
quite indicated for pumping around 2 μm (a few systems offer some tunability from
now on). Ultrashort pulse lasers at telecommunication wavelengths delivering tenths
of kW are a good alternative for pumping at 1.55 μm as DPSS Nd:YAG systems for
pumping at 1.064 μm (used peak powers under a few hundreds of W). Finally,
OPO/OPA are able to emit tenths of kW (up to MW) covering all these ranges and
adding large tunability, although a usual drawback of such systems is the rather bad
beam quality (in terms of M2, the beam propagation factor characterizing the beam
quality) which can decrease the coupling efficiency since the beam has to be strongly
focused onto an often very small fiber core [3].
Pump light launching into fibers under test is performed depending on the pump
laser technology. OPO/OPA systems supply an atmospheric propagating beam which
necessitates the use of an aspheric lens, a silica, ZnSe lenses and microscope objective
(20 to 40 magnification with potentially adapted numerical aperture), or seldom
silver/gold coating parabolic off-axis concave mirror if pump wavelength needs to be
tuned over a rather large band. When a fiber laser is used, the output can be either
collimated, reducing the problem to the former case, or a transport fiber which allows
direct tip-to-tip injection into the tested fiber core. When evaluated, coupling effi-
ciency ranges from 10 up to 30% [24]. Nevertheless, it is worth noting that in a given
optical configuration, the coupling efficiency decreases with the tested fiber core
12 Supercontinuum Generation in Tellurite Optical Fibers 287

diameter (indeed, the effective mode area). Of course, the pump power density at the
coupling tip should stay under the damage threshold of the glass.
The tested fiber output is often butt coupled into a large mode area (to avoid
supplementary nonlinear effects) IR fiber (fluoride ZBLAN, for instance), whose
other connectorized end goes to the detection device (see, for instance, [3]).
The reader will refer usefully to Table 12.1 to have a deeper insight on technical
characteristics of the pump lasers used in works found in literature.

12.5 Supercontinuum Generation Achievements

The best supercontinuum-spanning range obtained in tellurite optical fibers has

surprisingly long been that of one of the first published results on the subject [28],
where the SC extends from 0.789 to 4.87 μm in a –OH-contaminated suspended
core fiber. This paper has strongly stimulated the community, and several groups
have then explored IR supercontinuum generation in tellurite fibers from that time.
The main challenge they faced up is that of the –OH groups’ contamination of
tellurite glasses or fibers, which, even in small amount, is a real drawback, since the
attenuation associated with –OH vibration is spectrally large and quite intense
(10 dB/m/ppm around 3.4 μm) [7]. Then, for several years, supercontinuum exten-
sion is remained limited around 3 μm (Fig. 12.6), despite the efforts of these groups
(see Table 12.1). However, the comeback to step-index all solid fibers associated
with a drastic synthesis of the initial glass has allowed to get rid of this limitation
and to avoid OH contamination of fibers (Fig. 12.7) [9].

Fig. 12.6 IR supercontinuum spanning in various lengths (12 mm, 27 mm, 15 cm, and 50 cm) of a
suspended core tellurite optical fiber femtosecond pumped with an OPO at 1745 nm in its
anomalous dispersion regime [7]
288 F. Désévédavy et al.

120
100
Spectral Power Density (mW/nm)
100
10

Glass Transmission (% T)
1 80

0.1
60
0.01

1E-3 40

1E-4
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Wavelength (um)

Fig. 12.7 Supercontinuum generation (SCG) in W-type tellurite fiber pumped by 3 W of ~20 ps
pulses from a 32 MHz repetition rate amplified mode-locked laser at 1.92 μm (note: dotted line is
the transmission loss as measured in 10-mm-thick tellurite glass sample which is used to fabricate
the nonlinear tellurite fiber for SCG) [9]

Thanks to a pulsed Tm-doped fiber pumping around 1.9 μm, a 1.2-W output
supercontinuum, spanning from 0.5 to 5 μm, has been obtained in a step-index
tellurite fiber, covering almost the whole available fiber spectral range, between the
intrinsic bandgap and multiphonon absorptions of the glass (Fig. 12.7). This is a
demonstration of the strong potential of tellurite glass optical fibers for the devel-
opment of efficient broadband-tunable all-fibered laser sources spanning from the
visible to the mid-IR and covering the 3–5-μm atmospheric transparency window.
One can precise also that recently ultrabroad supercontinuum generation
(0.6–6 μm) has been obtain by filamentation in a bulk sample of tellurite glass
(Fig. 12.8) [6].
In this case, all the considerations about fibers, their geometry, and their eventual
–OH group contamination are no longer appropriate, since in this case a several
mm-long bulk sample of tellurite glass is sufficient to generate the SC. However,
this particular optical interaction can only be obtained through a very high power
pumping (around 1 GW at 1.6 μm) available with cumbersome laser sources
actually incompatible with compact devices.

12.6 Nonlinear Dynamics and Modeling

The fundamental material properties of tellurite glasses (i.e., large transmission

bandwidth and high nonlinearity) are very promising to develop compact
supercontinuum (SC) sources beyond 2 μm and across the mid-IR. To this purpose,
12 Supercontinuum Generation in Tellurite Optical Fibers 289

Fig. 12.8 Supercontinuum

generation through
filamentation in a bulk
tellurite glass sample
excited under various
pumped conditions
(PC) with a TOPAS in the
femtosecond regime at
1600 nm [6]

the study of SC generation in tellurite fibers benefited from the large refinements
recently developed to describe nonlinear processes in silica fibers. To illustrate the
underlying nonlinear dynamics and potential limitations of SC generation in
tellurite fibers toward the mid-IR region, we provide below numerical simulations
considering a particular range of fiber parameters and pumping conditions that
mimic most of previous experimental demonstrations (see Table 12.1). In particu-
lar, we performed simulations for a small-core microstructured optical fiber (MOF)
pumped by a nJ-level femtosecond erbium-doped fiber laser whose wavelength is
close to the MOF’s zero dispersion wavelength (ZDW), which is the usual config-
uration for efficient SC generation in optical fibers [46]. Similar results can be
obtained with the recent development of thulium-doped fiber lasers to pump larger-
core tellurite fibers (i.e., even based on a simple step-index design) [8]. Pumping
further in the infrared is expected to make easier the covering of mid-IR region, but
it usually requires uncommon laser chains.
The underlying mechanisms of spectral broadening in this pumping regime are
related to the tailoring of soliton dynamics (in anomalous dispersion) and associ-
ated dispersive waves (in normal dispersion) but also their mutual interactions. In
most cases the SC bandwidth is fully driven by interactions such as the soliton
290 F. Désévédavy et al.

trapping of dispersive waves, which modifies the short-wavelength edge (i.e.,

dispersive waves) as a function of the long-wavelength edge (i.e., solitons) in a
way that satisfies group-velocity matching [47]. For this purpose, spectrograms are
calculated to highlight simultaneous frequency and time domain pictures of such
SC dynamics and their correlation.
Moreover the usual pumping with high peak power pulses in the anomalous
dispersion of the MOF may result in SC generation with significant sensitivity to
input noise, which leads to significant shot-to-shot fluctuations in the SC bandwidth
and a low average SC coherence as well. We then performed a numerical investi-
gation with different input noise imposed on the initial pulse to reveal the stability
issue of the SC source with respect to potential applications.
For the MOF design and glass composition, we chose the suspended core
structure with core diameter of 2.7 μm and ZDW at 1.53 μm and typical 80TeO2-
10ZnO-10Na2O (TZN) glass composition. We calculated the spectral dependence
of both dispersion and effective mode area for the two orthogonal components of
the fundamental guided mode. The impact of fiber losses on SC generation, in
particular extra losses due to OH impurities, is also evaluated.

12.6.1 Numerical Model

Our simulations are based on a well-known generalized nonlinear Schr€odinger

equation for single mode and single polarization that has successfully described
the SC generation process over a wide parameter range [46]:
  ð þ1 
∂A b b
α ∂ 0 0 2 0

iDA þ A ¼ iγ 1 þ iτshock Aðz; tÞ Rðt ÞjAðz; t
t Þj dt : ð12:1Þ
∂z 2 ∂t
1

where A (z,t) is the electric field envelope (dimension in W1/2) as a function of the
longitudinal position z in a retarded time frame t moving with the group velocity
vg ¼ 1/β1(ω0) and γ ¼ n2ω0/cAeff is the nonlinear coefficient determined at ω0 the
center angular frequency, with n2 ¼ 3.8 10
19 m2 W
1 as the nonlinear
refractive index of the TZN glass composition [48]. To introduce modal properties
in the modeling, we calculated the wavelength dependence of both effective refrac-
tive index neff(ω) and effective mode area Aeff(ω) for the two orthogonal components
of the fundamental guided mode by means of a commercial software using a fully
vectorial finite-element model [49]. These calculations used Sellmeier coefficients
that give the wavelength dependence of the linear refractive index of the TZN glass
composition under study [50]. Figure 12.6a, b presents the corresponding curves of
the two fiber parameters obtained for one of the polarizations.
Group-velocity dispersions (GVD, β2) and mode areas are very similar for the
two polarization components; the slight differences are a 15-nm shift in the position
of the ZDW and 2–3% offset in the effective mode area. Consequently, we used an
average of both curves for the following simulations based on Eq. (12.1).
12 Supercontinuum Generation in Tellurite Optical Fibers 291

(a) (b) (c)

8 100
0.5
80

Loss (dB/m)
0 7
β2 (ps2/m)

Aeff (μm2)
60
-0.5 6
40
-1 5 20
-1.5
4 0
1 2 3 4 1 2 3 4 1 2 3 4
Wavelength (μm) Wavelength (μm) Wavelength (μm)

Fig. 12.9 Frequency-dependent properties used for SC modeling in our TZN-suspended core
MOF. (a) Calculated group velocity dispersion for our MOF (solid line) and TZN bulk (dashed
line). Inset: suspended-core MOF structure. (b) Calculated effective mode area. (c) Measured
optical losses using a single-material fiber and the cutback technique (solid line). Impact of OH
impurities is also investigated through extra losses added for wavelengths above 2.8 μm, which
corresponds to the beginning of the mid-IR loss bump (dashed line)

We also compare the GVD in our MOF to the intrinsic dispersion of the bulk
material in Fig. 12.9a. It clearly appears that the strong waveguide contribution
allows to shift the initial ZDW of the dispersion located around 2.2 μm toward
lower wavelengths (i.e., for suitable SC generation with 1.55-μm erbium-doped
fiber laser sources).
The fiber dispersion requires particular care because it greatly varies over the SC
bandwidth; our accurate modeling includes the full dispersion profile through
dispersion operator D. We applied this operator directly in the frequency domain
through multiplication of the complex spectral envelope by β(ω)-ωβ1-β0, with βk as
the coefficients related to the Taylor series expansion of the propagation constant
β(ω) ¼ neff ω/c (βk are evaluated at the center frequency). Likewise, the frequency-
dependent fiber losses are introduced through the loss operator α. As a first step, our
model takes into account the measured losses of a single-material fiber made from a
low-OH TZN glass composition including some fluoride ions [7] (i.e., with typical
background losses of ~1 dB/m up to 3 μm and about 10 dB/m in the range 3–4 μm;
see Fig. 12.9). Possible extra losses compared to the bulk or the single-material fiber
are often introduced around 3 μm due to OH impurities [14]; their impact on SC
generation will be evaluated later.
The right-hand side of Eq. (12.1) models nonlinear effects. To this end,
the response function R(t) ¼ (1
fR)δ(t) + fRhR(t) includes both instantaneous
electronic (Kerr) contribution and delayed Raman response with fR ¼ 0.25
representing the Raman fraction. For the Raman response function hR, we
used an intermediate-broadening model using convolutions of Lorentzians and
Gaussians adapted from spontaneous Raman-scattering spectra and calculated
Raman gain spectra of the TZN glass [51, 52]. More details about both origin
and impact of tellurite-based glass structure on the outstanding nonlinear optic
susceptibility and Raman gain are available [53–57]. In general, the Raman
response function of tellurite fibers did not exhibit a simple damped harmonic
292 F. Désévédavy et al.

oscillation, which corresponds to a large Raman spectra bandwidth that can be

tailored by combining multiple components in tellurite glass compositions.
The time-derivative term is usually associated to self-steepening effect, but it
also models the dispersion of the nonlinearity. It is characterized by a time scale
τshock ¼ 1/ω0-{d ln(Aeff)/dω}ω0 that takes into account the frequency dependence of
Aeff. We solved the propagation equation by using a modified version of the split-
step Fourier algorithm. The linear operators are treated in the frequency domain and
the full nonlinear step in the time domain with a second order Runge-Kutta method
(more details are given in Ref. [46]).
To investigate noise properties of SC spectra generated with different pumping
conditions, in particular the presence of significant shot-to-shot fluctuations in the
SC bandwidth, we performed 50 simulations with different input noise imposed on
the initial pulse. The input pulse shot noise was modeled by adding a noise seed of
one photon per mode with random phase on each frequency discretization bin [46].

12.6.2 Simulation Results

At first simulation, we considered a 10-cm length of our TZN MOF. We fixed our
pump wavelength at 1.56 μm and input pulse peak power (P0) at 12 kW. The initial
injected pulse used has a hyperbolic secant field profile with an intensity full width
at half maximum (TFWHM) equal to 400 fs. Fiber parameters calculated at the
pump wavelength are β2 ¼
11.7 ps2 km
1, γ ¼ 296 W
1 km
1, and τshock ¼ 0.95 fs.
The dynamics of the spectral and temporal evolution obtained through the numer-
ical solution of Eq. (12.1) is shown in Fig. 12.10a, b. Our pumping occurs in the

a Spectral evolution b Temporal evolution c Output spectrogram Intensity

10 10 0
DW S 15
8 8
–10
Delay (ps)

10
Distance (cm)

6 6
–20 5
4 4
0
–30
2 2
spectrum

0 0 –40
1 2 3 0 5 10 15 1 2 3
Wavelength (µm) Time (ps) (dB) Wavelength (µm)

Fig. 12.10 Results from a single-shot numerical simulation showing density plots of the
(a) spectral and (b) temporal evolution of SC generated as a function of propagation distance in
our TZN MOF. (c) Calculated output SC spectrogram projected onto the spectrum and temporal
intensity. The gate function used is a 100-fs sech pulse. Group-velocity matching between the most
red-shifted soliton (S) and the trapped dispersive wave (DW) is revealed by the straight white
dashed line. Dotted lines indicate that the SC bandwidth is fully determined by spectral positions
of S and DW
12 Supercontinuum Generation in Tellurite Optical Fibers 293

anomalous dispersion regime with input conditions corresponding to a high soliton

order N ¼ (LD/LNL)1/2 ¼ 125. The corresponding characteristic dispersive and
nonlinear length scales are LD ¼ (TFWHM/1.763)2/β2 ¼ 4.4 m and LNL ¼ 1/γ
P0 ¼ 0.28 mm, respectively. In the first millimeters of propagation, the input
pulse undergoes a first compression stage through self-phase modulation (SPM)
as an ideal higher-order soliton. In the absence of noise perturbations, the soliton
fission occurs generally due to higher-order effects included in Eq. (12.1) after a
propagation distance predicted by Lfiss ~ LD/N ¼ 3.5 cm [46]. Here in the first
centimeter, we clearly observe the emergence of noise-driven modulation instabil-
ity (MI) spectral bands far from the pump due to the intense input power. In this
case, the MI gain competes with usual SPM and then dominates the initial spectral
broadening. The characteristic distance for MI is estimated to be LMI ~ 16
LNL ¼ 4.5 mm (i.e., LMI < Lfiss or N > 16), thus confirming that MI drives the
pulse breakup but also coherence properties of the subsequent SC [58]. After this
pulse breakup into several fundamental and ultrashort soliton structures, the
following spectral expansion is fully driven by soliton dynamics in non-integrable
systems such as inelastic collisions, soliton-self-frequency shift due to the Raman
effect (i.e., red-shifted solitons), and dispersive wave generation (located in the
normal dispersion regime). Finally, the SC bandwidth is usually determined by
the soliton trapping of dispersive waves, which modifies the short-wavelength edge
(i.e., dispersive waves) as a function of the long-wavelength edge (i.e., solitons) in a
way that satisfies group-velocity matching [47]. Figure 12.10c shows the calculated
spectrogram of the output SC field that reveals spectro-temporal signatures of the
dispersive waves and Raman-shifted solitons. One can clearly observe the parabolic
group-delay dependence of spectral components (i.e., related to the group-index
curve of the MOF guided mode), which allows to check the group-velocity
matching dynamics.

12.6.3 Discussion

In general, such nonlinear dynamics rule SC generation for a wide range of

pumping conditions, i.e., from femtosecond to continuous wave (CW) pumps
injected into the anomalous dispersion regime or close to the ZDW of the fiber
under study [46, 58]. Most of previous experimental demonstrations of SC gener-
ation in tellurite fibers used high-power pumps to obtain the largest spectral
bandwidth and overcome possible high losses or nonideal dispersion properties.
However, enhanced SC generation with a moderate pulse energy is possible
through the fine control of nonlinear propagation dynamics in order to fully cover
the glass transmission window. For instance, enhanced blue and red expansion of
the SC bandwidth can be achieved through nonlinear and dispersion engineering
along the fiber axis, i.e., tapered MOFs, the fabrication of micro-wires, or the
cascaded fiber approach [47]. But even in such cases, the emergence of MI may
lead to the recurrent issue of significant shot-to-shot fluctuations of the SC
294 F. Désévédavy et al.

(a) 100fs/12kW (N~31) (b) 400fs/12kW (N~20) (c) 1ps/12kW (N~312)

Spectrum (20dB/div)

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5
Wavelength (μm) Wavelength (μm) Wavelength (μm)

Fig. 12.11 Results from numerical simulation for 50 shots with different input noise (grey lines)
showing the output SC spectra obtained in our 10-cm-long MOF segment when using different
pulse widths (a) 100 fs, (b) 400 fs, and (c) 1 ps, with the same high peak power. The averaging over
50 simulations is superposed to highlight the shot-to-shot fluctuations with respect to an average
measurement (black lines)

properties in tellurite fibers [2]. Figure 12.11 illustrates the significant shot-to-shot
fluctuations of the resulting SC spectrum in our MOF when using different pulse
widths with high peak power. We also superposed the averaging over simulations
with different input noise, thus highlighting typically smooth SC spectra similar to
average spectral measurements (i.e., without the significant fine structure of single
shots). Note that the most significant spectral fluctuations are observed in the fine
structure (i.e., low average SC coherence) and for the mid-IR SC edge, and it
strongly depends on the input pulse width or energy (i.e., on the corresponding N
value). Due to multiple inelastic collisions, some rogue red-shifted solitons clearly
emerge on the mid-IR SC edge for specific input noise conditions (see Fig. 12.11c).
Even if for the shortest pulses the fluctuations decrease, it requires highest pulse
energy if one wants to obtain SC expansion over the full transmission window of
tellurite glasses with near-infrared lasers.
The main approach to avoid this noise-driven regime is when pumping occurs in
the normal dispersion regime and the subsequent spectral broadening develops
entirely in this dispersion regime (i.e., since MI cannot occur), but this solution is
usually associated with reduced spectral bandwidths. However, the generation of
Raman stokes that amplifies noise background is another issue when using long
pulses or CW (for both dispersion regimes), so that one must focus on the use of
ultrashort pulses (i.e., TFWHM < 1 ps). SC generation dynamics for ultrashort pulses
is always driven by self-phase modulation and optical wave breaking in all-normal
dispersion (ANDi) fibers. The resulting spectral profile is extremely flat with
excellent stability and coherence properties, as already demonstrated in silica fibers
[59, 60]. Moreover, this SC regime allows to preserve a single pulse in the time
domain that contains all the spectral components distributed almost linearly with
frequency. Such exceptional properties allows the use of such SC sources for
applications in which coherence properties and temporal profile of the SC pulse
are also of importance. To optimize SC bandwidth in this specific regime, one has to
design fibers that exhibit low and flattened normal dispersion at the pump wave-
length. It is worth mentioning that such properties can be obtained by using
W-index profile fiber without requiring the fabrication of MOFs. This simpler
12 Supercontinuum Generation in Tellurite Optical Fibers 295

fiber architecture combined with novel 2-μm-ultrashort-pulse thulium-doped fiber

lasers should pave the way to stable and compact mid-IR broadband light sources.
Another parameter that significantly affected SC generation in tellurite fibers is
related to the detrimental fiber attenuation in particular close to well-known OH
absorption peaks. Extra losses beyond 3 μm compared to the bulk or the single-
material fiber were already noticed in SC experiments using tellurite MOFs
[2]. Such detrimental loss related to different combinations of bonded OH groups
may be stimulated during MOF fabrication, drawing, or post-processing through
preliminary OH contamination surrounding the small fiber core. The previous SC
simulations (for 400-fs pulses) were again performed by introducing high extra
losses for wavelengths above 2.8 μm, which corresponds to the beginning of the
mid-IR loss bump revealed by measurements in the single-material fiber as shown
in Fig. 12.9c. The results obtained with the different loss profiles are presented in
Fig. 12.12. We note a 300-nm bandwidth reduction, thus confirming the detrimental
effect of extra losses beyond 3 μm on mid-IR SC generation even for a very short
fiber segment. The strong fluctuations in the fine structure of the supercontinuum
are still present. However, the presence of the mid-IR loss edge at 3 μm reduces the
fluctuations of the overall SC bandwidth [61].
Thanks to the recent development of compact mid-IR laser sources, the use of
very small core fibers through MOF technology (i.e., to design low dispersion fibers
at available pump wavelengths) to obtain efficient SC generation is not required
anymore. Indeed, the issue of extra losses can be easily avoided by using all-solid
fibers with simple step-index or more complex W-index profiles.

(a) (b)
Spectrum (20dB/div) Loss (dB/m)

100

50

1 2 3 4 5 1 2 3 4 5
Wavelength (μm) Wavelength (μm)

Fig. 12.12 Similar results than those reported in Fig. 12.11b but considering different fiber loss
profiles: (a) when using the loss profile from the single material fiber and (b) when extra losses are
taking into account above 3 μm
296 F. Désévédavy et al.

12.7 Conclusion

The last decade has been very productive with regard to supercontinuum studies in
tellurite optical fibers. Actually, the potential offered by tellurite glasses in terms of
infrared supercontinuum spectral spanning has been reached, since different groups
can almost cover today the window ranging from 0.5 to 6 μm, either on step-index
fibers or bulk glasses. The case of fibers is of particular interest since this is
a demonstration of the strong potential of tellurite glass optical fibers for the
development of efficient broadband-tunable all-fibered laser sources spanning
from the visible to the mid-IR and covering the 3–5-μm atmospheric transparency
window. These devices are now expected to be commercially available in a near
future. They are suitable for various applications such as, for example, atmospheric
monitoring, environment monitoring, biological or medical diagnostics, as well as
eventually military applications depending on the future available output powers.

References

1. M. Belal, L. Xu, P. Horak, L. Shen, X. Feng, M. Ettabib, D.J. Richardson, P. Petropoulos,

J.H.V. Price, Mid-infrared supercontinuum generation in suspended core tellurite
microstructured optical fibers. Opt. Lett. 40(10), 2237–2240 (2015)
2. J. Picot-Clemente, C. Strutynski, F. Amrani, F. Désévédavy, J. Jules, G. Gadret, D. Deng,
T. Cheng, K. Nagasaka, Y. Ohishi, B. Kibler, F. Smektala, Enhanced supercontinuum gener-
ation in tapered tellurite suspended core fiber. Opt. Commun. 354, 374–379 (2015)
3. T. Cheng, L. Zhang, X. Xue, D. Deng, T. Suzuki, Y. Ohishi, Broadband cascaded four-wave
mixing and supercontinuum generation in a tellurite microstructured optical fiber pumped at
2 μm. Opt. Express 23(4), 4125–4134 (2015)
4. A. Chavez-Pirson, D. L. Rhonehouse, and D. T. Nguyen, U.S. Patent 8,818,160, 26 Aug. 2014
5. Z.-X. Jia, L. Liu, C.-F. Yao, G.-S. Qin, Y. Ohishi, W.-P. Qin, Supercontinuum generation and
lasing in thulium doped tellurite microstructured fibers. J. Appl. Phys. 115(6), 063106 (2014)
6. M. Liao, W. Gao, T. Cheng, Z. Duan, X. Xue, H. Kawashima, T. Suzuki, Y. Ohishi, Ultrabroad
supercontinuum generation through filamentation in tellurite glass. Laser Phys. Lett. 10(3),
036002 (2013)
7. I. Savelii, F. Désévédavy, J.-C. Jules, G. Gadret, J. Fatome, B. Kibler, H. Kawashima,
Y. Ohishi, F. Smektala, Management of oh absorption in tellurite optical fibers and related
supercontinuum generation. Opt. Mater. 35, 1595–1599 (2013)
8. R. Thapa, D. Rhonehouse, D. Nguyen, K. Wiersma, C. Smith, J. Zong, and A. Chavez-Pirson,
Mid-IR supercontinuum generation in ultra-low loss, dispersion-zero shifted tellurite glass
fiber with extended coverage beyond 4.5 μm, in Proc. of SPIE Security+Defense, International
Society for Optics and Photonics, 2013, pp. 889808–889808
9. D. L. Rhonehouse, J. Zong, D. Nguyen, R. Thapa, K. Wiersma, C. Smith, and A. Chavez-
Pirson, Low loss, wide transparency, robust tellurite glass fibers for mid-IR (2–5 μm) appli-
cations, in Proc. Of SPIE Security+Defense, International Society for Optics and Photonics,
2013, pp. 88980D–88980D
10. M. Klimczak, G. Stepniewski, H. Bookey, A. Szolno, R. Stepien, D. Pysz, A. Kar, A. Waddie,
M.R. Taghizadeh, R. Buczynski, Broadband infrared supercontinuum generation in
hexagonal-lattice tellurite photonic crystal fiber with dispersion optimized for pumping near
1560 nm. Opt. Lett. 38(22), 4679–4682 (2013)
12 Supercontinuum Generation in Tellurite Optical Fibers 297

11. D. Deng, W. Gao, M. Liao, Z. Duan, T. Cheng, T. Suzuki, Y. Ohishi, Supercontinuum

generation from a multiple-ring-holes tellurite microstructured optical fiber pumped by a
2 μm mode-locked picosecond fiber laser. Appl. Opt. 52(16), 3818–3823 (2013)
12. L. Liu, Q. Tian, M. Liao, D. Zhao, G. Qin, Y. Ohishi, W. Qin, All-optical control of group
velocity dispersion in tellurite photonic crystal fibers. Opt. Lett. 37(24), 5124–5126 (2012)
13. M. Liao, W. Gao, Z. Duan, X. Yan, T. Suzuki, Y. Ohishi, Supercontinuum generation in short
tellurite microstructured fibers pumped by a quasi-cw laser. Opt. Lett. 37(11), 2127–2129
(2012)
14. I. Savelii, O. Mouawad, J. Fatome, B. Kibler, F. Désévédavy, G. Gadret, J. Jules, P. Bony,
H. Kawashima, W. Gao, T. Kohoutek, T. Suzuki, Y. Ohishi, F. Smektala, Mid-infrared 2000-
nm bandwidth supercontinuum generation in suspended-core microstructured sulfide and
tellurite optical fibers. Opt. Express 20(24), 27083–27093 (2012)
15. M. Liao, W. Gao, Z. Duan, X. Yan, T. Suzuki, Y. Ohishi, Directly draw highly nonlinear
tellurite microstructured fiber with diameter varying sharply in a short fiber length. Opt.
Express 20(2), 1141–1150 (2012)
16. M. Liao, X. Yan, W. Gao, Z. Duan, G. Qin, T. Suzuki, Y. Ohishi, Five-order SRSS and
supercontinuum generation from a tapered tellurite microstructured fiber with longitudinally
varying dispersion. Opt. Express 19(16), 15389–15396 (2011)
17. I. Savelii, J. Jules, G. Gadret, B. Kibler, J. Fatome, M. El-Amraoui, N. Manikandan, X. Zheng,
F. Desevedavy, J. Dudley, J. Troles, L. Brilland, G. Renversez, F. Smektala, Suspended core
tellurite glass optical fibers for infrared supercontinuum generation. Opt. Mater.
33, 1661–1666 (2011)
18. M. Liao, G. Qin, X. Yan, T. Suzuki, Y. Ohishi, A tellurite nanowire with long suspended struts
for low-threshold single-mode supercontinuum generation. J. Lightwave Technol. 29(2),
194–199 (2011)
19. M. Liao, X. Yan, Z. Duan, T. Suzuki, Y. Ohishi, Tellurite photonic nanostructured fiber.
J. Lightwave Technol. 29(7), 1018–1025 (2011)
20. Z. Duan, M. Liao, X. Yan, C. Kito, T. Suzuki, Y. Ohishi, Tellurite composite microstructured
optical fibers with tailored chromatic dispersion for nonlinear applications. Appl. Phys.
Express 4(7), 072502 (2011)
21. G. Qin, X. Yan, C. Kito, M. Liao, T. Suzuki, A. Mori, Y. Ohishi, Widely tunable narrowband
soliton source generation in tellurite microstructured fibers. Laser Phys. 21(8), 1404–1409
(2011)
22. G. Qin, X. Yan, M. Liao, A. Mori, T. Suzuki, Y. Ohishi, Wideband supercontinuum generation
in tapered tellurite microstructured fibers. Laser Phys. 21(6), 1115–1121 (2011)
23. G. Qin, M. Liao, C. Chaudhari, X. Yan, C. Kito, T. Suzuki, Y. Ohishi, Second and third
harmonics and flattened supercontinuum generation in tellurite microstructured fibers. Opt.
Lett. 35(1), 58–60 (2010)
24. X. Yan, G. Qin, M. Liao, T. Suzuki, A. Mori, Y. Ohishi, Soliton source generation in a
few-mode tellurite microstructure fiber. Appl. Phys. Lett. 95(17), 171107 (2009)
25. M. Liao, C. Chaudhari, G. Qin, X. Yan, T. Suzuki, Y. Ohishi, Tellurite microstructured fibers
with small hexagonal core for supercontinuum generation. Opt. Express 17(14), 12174–12182
(2009)
26. M. Liao, X. Yan, G. Qin, C. Chaudhari, T. Suzuki, Y. Ohishi, A highly nonlinear tellurite
microstructured fiber with multi-ring holes for supercontinuum generation. Opt. Express
17(18), 15481–15490 (2009)
27. G. Qin, M. Liao, C. Chaudhari, Y. Arai, T. Suzuki, Y. Ohishi, Spectrum controlled
supercontinuum generation in microstructure tellurite fibers. J. Ceram. Soc. Jpn. 117(1365),
706–708 (2009)
28. P. Domachuk, N.A. Wolchover, M. Cronin-Golomb, A. Wang, A.K. George, C.M.B. Cordeiro,
J.C. Knight, F.G. Omenetto, Over 4000 nm bandwidth of mid-IR supercontinuum generation
in sub-centimeter segments of highly nonlinear tellurite PCFs. Opt. Express 16(10),
7161–7168 (2008)
298 F. Désévédavy et al.

29. X. Feng, W.H. Loh, J.C. Flanagan, A. Camerlingo, S. Dasgupta, P. Petropoulos, P. Horak,
K.E. Frampton, N.M. White, J.H. Price, H.N. Rutt, D.J. Richardson, Single-mode tellurite
glass holey fiber with extremely large mode area for infrared nonlinear applications. Opt.
Express 16(18), 13651–13656 (2008)
30. P. Domachuk, N. A. Wolchover, M. Cronin-Golomb, F. Omenetto, K. K. Jang, J. Ahn,
A. Wang, A. C. George, and J. Knight, IR supercontinuum in compact tellurite PCFs, in
Proc. of CLEO Conference on Lasers and Electro-Optics 2007, Optical Society of America,
2007, p. CWO6
31. T. Delmonte, M. Watson, E. O’Driscoll, X. Feng, T. Monro, V. Finazzi, P. Petropoulos,
J. Price, J. Baggett, W. Loh et al., Generation of mid-IR continuum using tellurite
microstructured fiber, in Proc. of CLEO/QELS Conference on Lasers and Electro-Optics
and Quantum Electronics and Laser Science Conference 2006, 2006, pp. 1–2
32. A. Lin, A. Ryasnyanskiy, J. Toulouse, Fabrication and characterization of a water-free
mid-infrared fluorotellurite glass. Opt. Lett. 36(5), 740–742 (2011)
33. J. Stanworth, Tellurite glasses. Nature 169, 581–582 (1952)
34. F.T. Wallenberger, N.E. Weston, S.D. Brown, Infrared optical tellurite glass fibers.
J. Non-Cryst. Solids 144, 107–110 (1992)
35. J.S. Wang, E.M. Vogel, E. Snitzer, Tellurite glass: a new candidate for fiber devices. Opt.
Mater. 3(3), 187–203 (1994)
36. J. Massera, A. Haldeman, J. Jackson, C. Rivero-Baleine, L. Petit, K. Richardson, Processing of
tellurite-based glass with low OH content. J. Am. Ceram. Soc. 94(1), 130–136 (2011)
37. G. Brambilla, Optical fiber nanowires and microwires: a review. J. Opt. 12, 043001 (2010)
38. G. Tao, H. Ebendorff-Heidepriem, A.M. Stolyarov, S. Danto, J.V. Badding, Y. Fink, J. Ballato,
A.F. Abouraddy, Infrared fibers. Adv. Opt. Photon. 7(2), 379–458 (2015)
39. H. Ebendorff-Heidepriem, S.C. Warren-Smith, T.M. Monro, Suspended nanowires: fabrica-
tion, design and characterization of fibers with nanoscale cores. Opt. Express 17(4),
2646–2657 (2009)
40. E.F. Chillcce, C.M.B. Cordeiro, L.C. Barbosa, C.H. Brito Cruz, Tellurite photonic crystal fiber
made by a stack-and-draw technique. J. Non-Cryst. Solids 352(32–35), 3423–3428 (2006)
41. O. Mouawad, P. Vitry, C. Strutynski, J. Picot-Clémente, F. Désévédavy, G. Gadret, J.-C. Jules,
E. Lesniewska, F. Smektala, Atmospheric aging and surface degradation in As2S3 fibers in
relation with suspended-core profile. Opt. Mater. 44, 25–32 (2015)
42. C.R. Petersen, U. Møller, I. Kubat, B. Zhou, S. Dupont, J. Ramsay, T. Benson, S. Sujecki,
N. Abdel-Moneim, Z. Tang, D. Furniss, A. Seddon, O. Bang, Mid-infrared supercontinuum
covering the 1.4–13.3 μm molecular fingerprint region using ultra-high NA chalcogenide step-
index fiber. Nat. Photonics 8(11), 830–834 (2014)
43. J. Love, W. Henry, W. Stewart, R. Black, S. Lacroix, F. Gonthier, Tapered single-mode fibers
and devices. 1. Adiabaticity criteria. J. Optoelectronics, IEE Proceedings 138(5), 343–354
(1991)
44. M. Boivin, M. El-Amraoui, Y. Ledemi, S. Morency, R. Vallée, Y. Messaddeq, Germanate-
tellurite composite fibers with a high-contrast step-index design for nonlinear applications.
Opt. Mater. Express 4(8), 1740–1746 (2014)
45. K. Clarke, Y. Ito, Manufacture of fluoride glass preforms. J. Non-Cryst. Solids 140, 265–268
(1992)
46. J.M. Dudley, G. Genty, S. Coen, Supercontinuum generation in photonic crystal fiber. Rev.
Mod. Phys. 78(4), 1135–1184 (2006)
47. J.M. Dudley, J.R. Taylor, Supercontinuum Generation in Optical Fibers (Cambridge Univer-
sity Press, 2010)
48. A. Lin, A. Zhang, E. Bushong, J. Toulouse, Solid-core tellurite glass fiber for infrared and
nonlinear applications. Opt. Express 17(19), 16716–16721 (2009)
49. www.comsol.com, “Comsol multiphysics”
50. G. Ghosh, Sellmeier coefficients and chromatic dispersions for some tellurite glasses. J. Am.
Ceram. Soc. 78(10), 2828–2830 (1995)
12 Supercontinuum Generation in Tellurite Optical Fibers 299

51. M. O’Donnell, K. Richardson, R. Stolen, C. Rivero, T. Cardinal, M. Couzi, D. Furniss,

A. Seddon, Raman gain of selected tellurite glasses for ir fibre lasers calculated from sponta-
neous scattering spectra. Opt. Mater. 30(6), 946–951 (2008)
52. X. Yan, G. Qin, M. Liao, T. Suzuki, Y. Ohishi, Transient Raman response and soliton self-
frequency shift in tellurite microstructured fiber. J. Appl. Phys. 108(12), 123110 (2010)
53. A. Mirgorodsky, M. Soulis, P. Thomas, T. Merle-Méjean, M. Smirnov, Ab initio study of the
nonlinear optical susceptibility of TeO2-based glasses. Phys. Rev. B 73(13), 134206 (2006)
54. M. O’Donnell, K. Richardson, R. Stolen, A. Seddon, D. Furniss, V. Tikhomirov, C. Rivero,
M. Ramme, R. Stegeman, G. Stegeman, et al., Tellurite and fluorotellurite glasses for
fiberoptic Raman amplifiers: glass characterization, optical properties, Raman gain, prelimi-
nary fiberization, and fiber characterization. J. Am. Ceram. Soc. 90(5), 1448–1457 (2007)
55. R. Jose, G. Qin, Y. Arai, Y. Ohishi, Tailoring of Raman gain bandwidth of tellurite glasses for
designing gain-flattened fiber Raman amplifiers. JOSA B 25(3), 373–382 (2008)
56. G. Guery, A. Fargues, T. Cardinal, M. Dussauze, F. Adamietz, V. Rodriguez, J.D. Musgraves,
K. Richardson, P. Thomas, Impact of tellurite-based glass structure on Raman gain. Chem.
Phys. Lett. 554, 123–127 (2012)
57. S. Manning, H. Ebendorff-Heidepriem, T.M. Monro, Ternary tellurite glasses for the fabrica-
tion of nonlinear optical fibers. Opt. Mater. Express 2(2), 140–152 (2012)
58. G. Genty, S. Coen, J.M. Dudley, Fiber supercontinuum sources. JOSA B 24(8), 1771–1785
(2007)
59. C. Finot, B. Kibler, L. Provost, S. Wabnitz, et al., Beneficial impact of wave-breaking for
coherent continuum formation in normally dispersive nonlinear fibers. JOSA B 25(11),
1938–1948 (2008)
60. A.M. Heidt, A. Hartung, G.W. Bosman, P. Krok, E.G. Rohwer, H. Schwoerer, H. Bartelt,
Coherent octave spanning near-infrared and visible supercontinuum generation in all-normal
dispersion photonic crystal fibers. Opt. Express 19(4), 3775–3787 (2011)
61. D. Buccoliero, H. Steffensen, H. Ebendorff-Heidepriem, T.M. Monro, O. Bang, Midinfrared
optical rogue waves in soft glass photonic crystal fiber. Opt. Express 19(19), 17973–17978
(2011)
Chapter 13
Tellurite Glasses for Plasmonics

V.A.G. Rivera, F.A. Ferri, and E. Marega Jr.

13.1 Introduction

We center our attention in the tellurite glasses doped with rare-earth ions (REIs) and
their optical properties that are widely in optical devices, since the REIs should
ideally be confined as a delta function in the center of the core for maximum gain
per unit pump power. Therefore, there is a trade-off between the confinement and
the high RE ion concentration.
This attribute of tellurite glass, high concentration of REIs, is used in devices
that provide gain and can be exploited for nanostructure plasmonics application.
For instance, many authors suggested increasing the propagation length of the
surface plasmon polariton (SPP) by producing the population inversion in the
dielectric medium adjacent to the metallic nanostructure [1–7] or via energy
transfer between REIs and the metallic nanostructure [8].
Plasmonic nanostructure in interface with a tellurite glasses doped with REIs or
not can produce a field enhancement sustained that resonant excitation can lead to a
reduction in the threshold for achieving inversion in the optically active surround-
ing medium, and the presence of gain can counteract the inherent absorption losses
in the plasmonic nanostructure [9]. This enhancement of plasmon resonances in
gain medium is experimentally confirmed, e.g., amplification of fluorescence due to
field enhancement in gain medium with metal nanostructure has been observed [8].

V.A.G. Rivera (*)

Facultad de Ciencias Fı́sicas, Universidad Nacional Mayor de San Marcos, Lima, Peru
e-mail: [email protected]
F.A. Ferri
Departamento de Fı́sica, Universidade Federal de S~ao Carlos, S~ao Carlos, SP, Brazil
E. Marega Jr.
Instituto de Fı́sica de S~ao Carlos – INOF, USP, Caixa Postal 369, 13560-970 S~ao Carlos,
SP, Brazil

© Springer International Publishing AG 2017 301

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3_13
302 V.A.G. Rivera et al.

It is worth mentioning that the main purpose of this chapter is to show some of
concrete concept of plasmonic nanostructure interaction with REIs exhibiting a
field enhancement in transparent medium with such nanostructures via a localized
surface plasmon resonance (SPR).

13.2 Localized Surface Plasmon Resonance and Plasmonic

Nanostructures

13.2.1 Introduction

Materials that possess a negative real and small positive imaginary dielectric
constant are capable of supporting a surface plasmon resonance (SPR). This
resonance is a coherent oscillation of the surface conduction electrons excited by
electromagnetic (EM) radiation. Plasmonics is the study of these particular light–
matter interactions, which have enabled a vast array of applications, including
surface-enhanced spectroscopies [10–17], biological and chemical sensing [18–
22], and lithographic fabrication [23, 24]. Plasmonic materials can already be
found in commercial instruments, such as the BIAcore®, which monitors the
thermodynamics and kinetics of biological binding processes using SPR spectros-
copy [22, 25, 26]. In these experiments, researchers use thin (
50 nm) metal films
as the sensing platform; these yield propagating plasmons, also known as surface
plasmon polaritons [27, 28]. However, with recent advances that allow the con-
trolled fabrication and manipulation of metallic structures on the nanoscale,
researchers have demonstrated new applications that take advantage of the local-
ized surface plasmon resonance (LSPR) [18, 20, 21, 29–31].
In the case of surface plasmon polaritons, plasmons propagate in the x- and
y-directions along the metal–dielectric interface, for distances on the order of tens
to hundreds of microns, and decay evanescently in the z-direction with 1/e decay
lengths on the order of 200 nm [28, 32]. The interaction between the metal surface-
confined EM wave and a molecular surface layer of interest leads to shifts in the
plasmon resonance condition, which can be observed in three modes: (a) angle
resolved, (b) wavelength shift, and (c) imaging [22]. In the first two modes, one
measures the reflectivity of light from the metal surface as a function of either angle
of incidence (at constant wavelength) or wavelength (at constant angle of inci-
dence). The third method uses light of both constant wavelength and incident angle
to interrogate a two-dimensional region of the sample, mapping the reflectivity of
the surface as a function of position.
For the case of localized surface plasmons, light interacts with particles much
smaller than the incident wavelength [18, 20–83]. This leads to a plasmon that
oscillates locally around the nanoparticle with a frequency known as the LSPR
13 Tellurite Glasses for Plasmonics 303

[30, 33]. Similar to the SPR, the LSPR is sensitive to changes in the local dielectric
environment [33, 34]. Typically researchers sense changes in the local environment
through an LSPR wavelength-shift measurement, although a variant of angle-
resolved sensing for the LSPR is also possible [35].
Significant interest in understanding propagating and localized surface plasmons
developed on the discovery of surface-enhanced Raman scattering (SERS)
[36]. Researchers proposed competing mechanisms of both chemical and
EM-field enhancements to explain the observed Raman enhancement on roughened
metallic substrates [37–39]. Because the EM-field enhancement mechanism
required coupling of the incident radiation to the metal surface, researchers devoted
much theoretical and experimental effort to understanding surface plasmons
[40, 41]. In particular, they studied the role of size, shape, material, and local
dielectric properties—all of which determine the LSPR wavelength
[42, 43]. Whereas these studies provided a fundamental understanding of how
plasmons are influenced by local structure and environment, they also suggested
the usefulness of plasmons as a sensing modality. Today, plasmon spectroscopy
enjoys a reputation as an ultrasensitive method for detecting molecules of both
biological and chemical interest, in addition to its continued role in enabling
surface-enhanced spectroscopic methods, including SERS, surface-enhanced
hyper-Raman scattering, surface-enhanced infrared spectroscopy, second harmonic
generation, and surface-enhanced fluorescence [15, 44, 45].
Although SPR spectroscopy currently dominates commercial instrumentation,
LSPR spectroscopy offers many of the same advantages for sensing and spectros-
copy experiments, along with several additional benefits [20, 29]. Both SPR and
LSPR spectroscopy can provide thermodynamic and real-time kinetic data for
binding processes. Moreover, although SPR spectroscopy provides much higher
sensitivity to changes in the bulk refractive index than LSPR spectroscopy [29], the
response of the two techniques becomes comparable when measuring short-range
changes in the refractive index owing to a molecular adsorption layer [29]. This is a
result of the much smaller sensing volume offered by LSPR sensors, as the
EM-field decay length is 40–50 times shorter than that of the SPR sensors [46]. -
Single-nanoparticle LSPR spectroscopy is another option, offering sensitivity that
not only rivals, but can surpass, that of nanoparticle array-based LSPR spectroscopy
[47, 48]. Moreover, advances in both synthetic and lithographic fabrication tech-
niques allow researchers to tune the LSPR wavelength throughout the visible-,
near-infrared, and the infrared region of the EM spectrum, by varying the shape,
size, and material of the nanoparticles that support the surface plasmons [49–
51]. This offers additional flexibility when designing LSPR sensing experiments.
First, we present several key equations for understanding the theory behind the
LSPR. This is followed by a discussion of experimental techniques for fabricating
metallic nanoparticles and performing spectroscopic measurements. Next, we pre-
sent several fundamental studies of plasmon spectroscopy, illustrating key relation-
ships that govern the LSPR spectral location and its sensitivity to the local dielectric
environment.
304 V.A.G. Rivera et al.

13.2.2 Theory

A full theoretical treatment of localized surface plasmons is quite lengthy and

beyond the scope of this chapter. For our more compact treatment, we consider a
spherical nanoparticle of radius that is irradiated by z-polarized light of wavelength
λ (where a is much smaller than the wavelength of light λ; i.e., a/λ < 0.1). In this
limit, the magnitude of the electric field appears static around the nanoparticle,
allowing us to solve Maxwell’s equations using a quasi-static approximation
[33, 52]. The resulting solution for the EM field outside the particle is given by
   
εin  εout bz 3z
Eout ðx; y; zÞ ¼ E0bz  a3 E0 3  5 ðxb
x þ yb
y þ zbz Þ ð13:1Þ
ðεin þ 2εout Þ r r

Here, εin is the dielectric constant of the metal nanoparticle, and εout is the
dielectric constant of the external environment. Because εin is strongly dependent
on wavelength, the first term in square brackets determines the dielectric resonance
condition for the particle. When the dielectric constant of the metal is roughly equal
to 2εout, the EM field is enhanced relative to the incident field. In the case of silver
and gold, this condition is met in the visible region of the spectrum, which has
important implications for surface-enhanced spectroscopies. The size (a) and exter-
nal dielectric constant (εout) also play key roles in determining the EM field outside
the particle, consistent with experimental results.
We can calculate the extinction spectrum of the metal sphere as follows [53, 54]:
" #
24π 2 Na3 εout 3=2 εi ð λ Þ
Eð λ Þ ¼ : ð13:2Þ
λ ln ð10Þ ðεr ðλÞ þ χεout Þ2 þ εi ðλÞ2

Here, εr and εi are the real and imaginary components of the metal dielectric
function, respectively. Again, we note the wavelength dependence of the metal
dielectric function. Of additional interest is the factor of χ that appears in front of
εout. The value of χ is 2 for the case of a sphere; however, it takes on values as large
as 20 to account for particle geometries with high aspect ratios [55]. Because we
can only solve the value of χ analytically for spheres and spheroids, and we must
approximate it for all other geometries [55], researchers have developed additional
numerical methods; these include the discrete dipole approximation and the finite-
difference time-domain methods [56, 57]. In these calculations, the particle of
interest is taken as N finite polarizable elements, each of which can interact with
the applied electric field. In the case of the discrete dipole approximation method,
this interaction is simulant in the frequency domain. Both methods allow the
evaluation of the extinction of particles of arbitrary shape and size, and the results
typically match well with experimental results.
In addition to modeling the extinction of particles, several equations describe
how the LSPR is used for both sensing and spectroscopic experiments. For exam-
ple, the LSPR extinction (or scattering) wavelength maximum, λmax, is sensitive to
13 Tellurite Glasses for Plasmonics 305

the dielectric constant ε (or refractive index, n; since ε ¼ n2). Thus, changes in the
local environment—such as through the presence of an adsorbed species—should
cause a shift in λmax. This leads to the following relationship [58]:
  
2d
Δλmax ¼ mΔn 1  exp ð13:3Þ
ld

Here m is the bulk refractive-index response of the nanoparticle(s), Δn is the

change in refractive index induced by the adsorbate, d is the effective adsorbate
layer thickness, and ld is the characteristic EM-field decay length (approximated as
an exponential decay). This relationship is the basis of LSPR wavelength-shift
sensing experiments.
We can describe the enhancement factor for SERS as shown [59, 60]:
   2
Eout ðωÞ2 Eout ðω  ωv Þ2  I SERS
ðωv Þ=N 
EFSERS ðωv Þ ¼ ¼ 4h surf
ð13:4Þ
E40 I NRS
ðωv Þ=N 
vol

Equation ((13.4)) describes how the Raman enhancement effect is a result of

enhancing both the incident excitation, Eout(ω), and the resulting Stokes’ shifted
Raman, Eout(ω  ωv), EM fields. The calculation of the enhancement factor from
experimental measurements is given by the right-hand side of Eq. (13.4)—i.e., the
SERS-enhanced Raman intensity, ISERS(ωv)—normalized by the number of mole-
cules bound to the enhancing metallic substrate, Nsurf, divided by the normal Raman
intensity, INRS(ωv), normalized by the number of molecules in the excitation
volume, Nvol. This equation is important for characterizing the enhancing ability
of substrates of various material, geometry, and LSPR wavelength and is applied to
several fundamental spectroscopy studies.

13.2.3 Experimental Details

13.2.3.1 Spectroscopy

Fundamental spectroscopic studies on plasmonic materials provide an understand-

ing of the properties that govern both the LSPR wavelength and the surface
enhancement factor of these substrates, which is critical when designing materials
for sensing applications. In this section, we describe experiments that demonstrate
several of these fundamental relationships.
306 V.A.G. Rivera et al.

13.2.3.2 Size and Shape Dependence

Because the shape and size of a metallic nanoparticle dictate the spectral signature
of its plasmon resonance, the ability to change these two parameters and study the
effect on the LSPR is an important experimental challenge. Researchers first
systematically studied this using NSL-fabricated arrays with variable diameter
nanosphere masks as well as different Ag film thicknesses [61]. A second approach
is to use electrochemistry to oxidize the nanoparticles, thus modifying their surface
structure in a highly controlled manner [62]. For these experiments, NSL-fabricated
silver triangles were prepared on an indium tin oxide substrate and subjected to
multiple chronocoulometric runs. The particular power of this approach is that the
triangles are selectively oxidized, allowing the LSPR response to morphological
changes to be directly correlated. For example, we can study the relation between
the in-plane width and the LSPR because the triangle height does not change during
the initial oxidation events. Thus, electrochemical oxidation is a unique tool for
changing the shape and size of NSL-fabricated nanoparticles in a controlled
manner, allowing the observation of systematic trends.
A third approach for examining the relationship between the shape, the size, and
the LSPR wavelength is through single-nanoparticle spectroscopy [41, 48, 63,
64]. Because there is inherent heterogeneity among individual nanoparticles, each
LSPR spectrum is different, revealing the true distribution of resonance wave-
lengths [65]. For example, silver colloids prepared via a citrate reduction have a
variety of shapes and sizes, which results in a vast array of LSPR spectra [66]. The
different shapes show different refractive-index sensitivities, with rods showing the
highest sensitivity followed by triangles and then spheres [48]. More refined
synthetic approaches produce nanoparticles with less heterogeneity in shape and
size and even higher refractive-index sensitivity [49, 67]. For instance, triangular
nanoprisms functionalized with alkanethiols of increasing chain length show a 4.4-
nm shift in maximum LSPR wavelength per each additional methyl group, the
highest short-range, single-particle refractive-index sensitivity reported thus far
[66]. Moreover, the height, width, and tip sharpness of these nanoprisms can be
controlled, allowing the characterization of the relationships between these struc-
tural properties and both the LSPR wavelength and refractive index sensitivity [66].
Single-particle spectroscopy has also revealed unique spectroscopic properties.
For example, when individual cubic nanoparticles with a 30-nm edge length are
placed on a dielectric substrate, the LSPR spectrum reveals two peaks: a broad red
peak that is also present in solution and a narrow blue peak that emerges owing to its
interaction with the dielectric substrate [41]. A nanoparticle to yield a new plasmon
resonance peak when placed on a dielectric surface, it must satisfy two conditions:
(a) its near fields must be most intense at the polar (rather than the equatorial)
regions of the nanoparticle, and (b) it must be thicker than the skin depth of the
material (
25 nm for silver). This new peak also shows improved refractive-index
sensitivity owing to its narrow line width, defined by a figure of merit in which the
bulk refractive-index sensitivity is divided by the full width half max of the peak
13 Tellurite Glasses for Plasmonics 307

[41]. Thus, this novel cubic architecture may be valuable for future LSPR sensing
experiments.

13.2.3.3 Localized Surface Plasmon Resonance Distance Dependence

Equation (13.3) models the change in the LSPR wavelength as the refractive index
of the nanoparticles’ surroundings is changed. In this equation, the EM-field decay
length was approximated as a single exponential decay with characteristic decay
length ld.
Although Eq. (13.3) provides a reasonable approximation of the behavior of the
field, it provides no insight into the effect of nanoparticle structure on the EM-field
decay length and related properties. To understand this relationship, researchers
have conducted initial experiments by measuring the LSPR wavelength shift of
NSL-fabricated triangles using self-assembled alkanethiol mono- and multilayers
to probe the short-range and long-range distance dependence of the field, respec-
tively [66]. The results of the short-range studies suggest that the LSPR shift
decreases when either (a) the nanoparticle in-plane width increases or (b) the
nanoparticle out-of-plane height increases, whereas the long-range studies suggest
the opposite trend [66]. However, these experiments had less-than-optimal control
over layer thickness and local refractive index owing to the effects of alkanethiol-
chain tilt angle and packing density.
Recently, researchers carried out a second-generation LSPR distance-
dependence study. They employed atomic layer deposition (ALD) [42] to provide
the ultimate in distance resolution and a highly uniform refractive index indepen-
dent of layer thickness [68]. Here, a single atomic layer of alumina (Al2O3) is
deposited onto the nanoparticle surface, which provides 1.1-Å distance resolution
[69]. Whitney et al. [46] shows the LSPR shift from NSL-fabricated nanoparticles
of varying out-of-plane heights as the alumina film thickness increases. The
nanoparticles show LSPR shifts even with a single atomic layer of Al2O3, demon-
strating the impressive sensitivity of nanoparticles to their local environment.
However, this sensitivity diminishes as the nanoparticle out-of-plane height
increases: the thickest nanoparticles show less of an LSPR shift in response to the
alumina layer than the thinnest nanoparticles; see [46]. Moreover, the LSPR shift
begins to flatten out (at a layer thickness of approximately 600 Å), owing to the
decay of the EM field further from the particle surface.
Because the enhanced EM field decreases further from the nanoparticle surface,
ALD also offers a mechanism for probing the distance dependence of SERS [70].
308 V.A.G. Rivera et al.

13.2.3.4 Localized Surface Plasmon Resonance Spectroscopy near

Molecular Resonances

Surface-enhanced resonance Raman scattering (SERRS) has been known for pro-
ducing enhancement factors of the order 1011–1012 dating back to the original
report of SERS [71]. In SERRS, the Raman excitation light is of sufficient energy
to promote an electronic transition in the molecule of interest, from the ground to
some electronically excited state. When the LSPR of the enhancing substrate is also
in the proper energy region, this provides SERRS enhancement factors that are
roughly the product of the enhancement factors for nonresonant SERS of the
substrate and the resonance Raman spectrum of the adsorbate [72].
To study this relationship, recent work focused on the LSPR shift induced by the
adsorption of a molecule whose absorption spectrum is in the same energy region as
the plasmon resonance [73].

13.3 Tellurite Glasses Doped with REIS

13.3.1 A Summary of the Optical, Structural, and Thermal

Properties of Tellurite Glasses

Tellurium oxide (TeO2) is a conditional glass former, meaning that pure tellurium
oxide (unlike SiO2) cannot form glass by itself when quenched from the liquid or
vapor state. The structure of solid (α and β forms) TeO2 has a lone-pair electron
(LPE) [74, 75], which reduces the number of permutations and combinations [76]
possible in a glassy structure. However, when a small concentration of solute
(<2 mol%) is incorporated in pure TeO2 liquid, the tendency for glass formation
increases dramatically [75], which suggests that the multivalent cations from
solutes, e.g., ZnO, WO3, and trivalent oxides, bond with the LPE and provide
rigidity in the random glass network structure by reducing the overall Coulombic
repulsion from LPE sites. A number of oxides, e.g., R2O, RO, R2O3, RO2, R2O5,
and RO3, are known to dissolve in tellurium oxide liquid including the rare-earth
oxides (RE2O3), by forming thermodynamically metastable liquids, which upon
quenching yield stable glass compositions, resisting crystallization. Vogel
discussed that the multiple structural sites in tellurite glass can provide a means
for dissolving larger concentrations of rare-earth oxide than in silicates, phosphates,
and fluorides [75]. Detailed information about the structural characteristics (as well
as the optical, thermal, and spectroscopic properties) of tellurium oxide glasses can
be found in [77, 78].
Unlike most network glass formers, TeO2 glass has three basic structural units as
stated in [79]: a TeO4 (trigonal bipyramid), a TeO3 (pyramid), and an intermediate
with TeO3+δ polyhedron. Each of these structures has a LPE [75–79].
13 Tellurite Glasses for Plasmonics 309

The TeO4 unit has four oxygen atoms which are covalently bonded with the
central tellurium atom to form a trigonal bipyramid (tbp), with one of the equatorial
oxygen positions unoccupied. In the bipyramid structure, the two equatorial and
two apex oxygen sites are called the bridging oxygens (BO), whereas the third
equatorial site is a LPE, available from the valence band of Te. In the trigonal
pyramid (tp) structure, there are two BO sites and one non-bridging oxygen (NBO);
the latter is a double-bond Te ¼ O. The intermediate polyhedron is effectively a
distortion of the tp due to the presence of excess oxygen (δ). In the tellurite glass
structure, a LPE site, in terms of static charge equivalence, is like an oxide (O2)
ion, and therefore it might offer equilibrium positions through interchange via
delocalization for achieving the network continuity in a tellurite glass structure.
The structure of tellurite glass permits incorporation of silicates, borates, ger-
manium oxide, and phosphate [80], in increasing order of solubility [81], which
means that the glass network is capable of forming more than one electric dipole
environment for rare-earth ions. Consequently, the fluorescence line shape and
radiative and nonradiative rates may be planned in such hybrid glass structures [79].
Several methodologies for controlling and reducing the absorption of OH ions
during glass melting were developed [79]. The OH absorption in TeO2-based
glass is strongly dependent on the constituents, especially the alkali, phosphate, and
B2O3, which are inherently hygroscopic. The raw material processing, followed by
melting, requires drying in a chlorine–oxygen mixture at the temperature at which
the alkali components of the raw materials do not convert into chlorides via
Na2CO3 + Cl2 ¼ 2NaCl + CO2 + 0.5O2 and NaCl + H2O ¼ NaOH + HCl. Any resid-
ual H2O in the gas stream will be reabsorbed by the chloride components of the
glass-forming liquid and increase the OH absorption. The tendency for carbonate
to chloride conversion during melting is also observed with BaO which invariably
has a small layer of carbonate on the surface. During chlorine drying, BaCO3 may
also convert into BaCl2. Once the chlorine-drying condition is optimized for a
composition, the melting must be carried out in ultradry oxygen gas. O’Donnell
et al. [82] utilized the treatment of tellurite glass-forming melts with ZnF2 and
demonstrated significant reduction in OH, but not sufficient. For instance, the
incorporation of ZnF2 in the tellurite glass-forming liquid provides a reasonable
reaction pathway as PbF2 in a GeO2 liquid [79].

13.3.2 Spectroscopic Analysis

The spectroscopic properties of rare-earth ions (REIs) in tellurite glasses can be

found in details in [79]. As stated there, the electronic band gap of these materials is
in the region of 300–350 nm, which means that, except for a few transitions in the
UV–visible below 400 nm, a vast majority of visible transitions are likely to be
favorable in these two types of lower phonon energy glass hosts since many of the
visible transitions have been reported also in high-phonon-energy (1000 cm1)
silicate, phosphate, and borate and low-phonon-energy (<700 cm1) fluoride
310 V.A.G. Rivera et al.

glasses. At the other end of the spectrum in the near- and mid-IR, the majority of
optical transitions in rare-earth ions are possible because the relatively low-phonon
energies of TeO2 (700–800 cm1)-based glasses favor such optical transitions.
However, as the energy gap for an optical transition narrows to less than 3500 cm
1
, such optical transitions are strongly dominated by multi-phonon-activated
decay processes. In [79], we can find examples of optical transitions with detailed
discussions for appreciating the challenges and opportunities in engineering laser
and amplifier devices. For each rare-earth ion, [79, 94] analyzes the ion as a single
dopant, followed by key co-dopants which are essential in the exploration of
significant optical transitions and the mechanism of controlling them via energy
exchange processes (energy transfer and/or cross-relaxation).

13.4 Plasmonic Nanostructure Coupled with REIS:

Plasmon–Photon Interaction

Collective charge excitations on the surface of a metallic surface are the origin of
LSPR; such phenomenon is caused by the incidence light excitation (electromag-
netic—EM—field) in the metallic surface. Since EM radiation is formally
explained by quantum mechanics, besides, it is reasonable to consider the plasmon
as a quantum of energy associated with the collective excitations. This statement is
crucial in order to understand the plasmon–photon interaction or conversion pro-
cess, since recent examples in the literature have reported the interaction between
plasmonic systems with REIs [83, 84], for instance. To describe such interactions,
the Green function formalism for the electric potentials is commonly used [85].
The spontaneous emission of a REI can be produced for an external perturbation.
When a REI is placed within a site (amorphous system or cavity), their interaction is
said to be weak. Depending on the characteristics of the cavity, only specific EM
modes will be supported, because the cavity can only allow certain radiative decay.
In this way, if a REI is placed close to a plasmonic nanostructure, then the final
density of states will peak at λLSPR, as it provides a new and or strong decay
radiative channel on the REI. We consider two points: (1) If the lifetime of the
LSPR is very short compared with the emission spontaneous lifetime of an isolated
REI, we are at the limit of weak coupling. Hence, photons are engaged for a short
time and interact very little with the REI. (2) If the lifetime of the LSPR is
compared with the spontaneous emission lifetime of an isolated REI, the plasmon–
photon interaction is reliable for the collective excitation of the electrons present on
the plasmonic nanostructure, with a dynamic ruled by a Hamiltonian operator H.
We write the Hamiltonian that describes the dynamics of this system as [8, 85]:
 
n X 1 2
H¼ Π þ mωw Ψw :
2 2
ð13:5Þ
2 w m w
13 Tellurite Glasses for Plasmonics 311

The main consequence of this H relies on the fact that plasmons are quantized,
similar to the well-known case of photons from quantum theory. Additionally, let us
assumed that photons and plasmons are also bosons, because they are defined by a
real vector function that corresponds to neutral and massless particles. Since
plasmons possess similar features to photons, for instance, creation/destruction
operators, plasmon modes have different degrees of alignment with the polarization
of the external photon.
The Hamiltonian operator of Eq. (5) defines the plasmon–photon interaction.
Thus, it is possible to determine the probabilities of conversion from a plasmon to a
photon (and vice versa) using the perturbation method. The first process we will
consider is the conversion of a plasmon, as follows [85]:

2π 2 Ne2 h
T w, k,α ¼ nw jΘw, k, α j2 δðhωw  hωk Þ: ð13:6Þ
mNP V

Such probability rate was achieved from the assumption that the plasmon decays
spontaneously. Additionally, the second process is the conversion of a photon, and
here the probability rate is similar to the first process [85]:

2π 2 Ne2 h
T w, k, α ¼ nk, α jΘw, k, α j2 δ ðhωw  hωk Þ: ð13:7Þ
mNP V

The increased lifetime of the plasmon is critical to the enhancement of the

electric field near the plasmonic nanostructure. Therefore, a strong coupling
between plasmons and photons must be made.
In order to obtain a strong coupling we can put REIs near the nanostructure
plasmonic, because additional photons in a dielectric medium transparent originat-
ing from the electronic transitions of these ions will improve the plasmon–photon
interaction.

13.4.1 Local Field Enhancement

An SPP is a collective motion of conduction band electrons relative to fixed positive

ions in a metal, driven by the electric field component of incident light [86, 87], and
ca generate an extraordinary confinement that can couple between neighboring
REIs. Additionally, resonance conditions are achieved when light is coupled in
phase with the natural frequency of the SPP (LSPR), i.e., the plasmonic structure
absorbs the maximum volume of incident EM radiation, causing the highest amount
of charge displacement.
An effective coupling between the plasmonic nanostructure and the REI pro-
duces a tremendously high local electric field, which may not be possible from
individual REI. Fields are disposed to concentrate at the tips of the protrusions in an
effort to be nearly perpendicular to a metallic surface. These volumes of intense EM
312 V.A.G. Rivera et al.

radiation are called “hot spots” [88]. They will also produce an intense EM
radiation, “lightning rod” effect, due to the high crowding of electric field lines
[89]. These effects are the dominant mechanisms at plasmonic nanostructures and
are valuable for many surface-enhanced spectroscopic methods [30].
In the experimental demonstration of these effects that we can find in Ref. [90],
in that work, it is shown that gold nanoparticles can be excited by preferential
incident radiation, achieving a direct coupling between Er3+ ions, subsequent in
(1) a local field increase (Frohlich condition), (2) nonradiative decay (Joule effect),
or (3) radiative energy which depends on the albedo of the nanoparticle. In this
manner, the response of the nanoparticle will depend on the physical system and is
not strictly symmetric about the resonance frequency. Therefore, in a local field
correction [91]:
   1 !
ðε0 þ 2ÞE0 ωp 2
Eeff ¼ 1 þ Vωp 2 ð1  V Þ  ω2 þ iγω , ð13:8Þ
3 3ε0

where Eeff is the effective electric field. The effective field is found to be
Eeff ¼ Eloc + E. Hence, the enhancement luminescence intensity is related to the
strong local electric field owing to the metallic nanoparticles, which increases
the quantum yield η on luminescence. In this manner, both the “hot spot” and the
“lightning rod” effects are also consequences of the short distance, between two
equipotential particles (e.g., a REI coupled with a metallic nanoparticle). Hence, for
small separations, it becomes essential to adapt a quantum mechanical description
in order to predict reliable electric field enhancements, as discussed in [84].
In this framework, let us assume that tellurite glass can be labeled as a mixture
(distribution) of N different oscillators, hence, a dielectric function as a continuous
linear superposition of individual dielectric functions [92, 93]:
Z Y
ωi 2
εðωÞ ¼ ε0 þ ω0  ω00 Δω 02
dω0 , ð13:9Þ
0  ω2  iγ 0 ω 0

where ε(ω) (¼εd(ω)) is the dielectric function of the host matrix (glass) and
∏(ω0  ω0’) is a rectangular function of width Δ that represents the relative
numbers of oscillators at each frequency from the transparent material. Therefore,
we can write a complete equation that describes the emission intensity of the NP as
function of ω (INP(ω)) [84]:

4πω4 jEj2 ða1 a2 a3 Þ2

I NP ðωÞ ¼
9c3 ε0
8 RQ 92
< ω 2 0 2
= ð13:10Þ
1 2 p  ε0 þ
ω þiγω
ðω0 ω0 ÞΔ ω 0 2 ωωi2 iγ ωdω00
R Q
0 0
:
: ε0 þ ðω0 ω00 ÞΔ ω 0 2 ωωi2 iγ ωdω00 ð1Li ÞþLi 1ω2ωþiγω
2 p2 ;
0 0
13 Tellurite Glasses for Plasmonics 313

Equation (13.10) considers the geometrical effect of the nanoparticle and the
environment transparent glass. We consider the following two cases. (1) If
εm(ω)  εd(ω) > 0, then INP(ω)>0 is always the case. Therefore, we have an
improvement in the luminescence, as mentioned above. (2) If εm(ω)  εd(ω) < 0,
then we most likely (based on a geometrical dependency, Li  1) have a quenching
in the luminescence. This manifests the asymmetric behavior of the nanoparticle
emissions (INP(ω)) for different frequencies [94].

13.4.2 Energy Transfer

Energy transfer process occurs between two or more REIs of the same or different
species via an interaction between two induced dipoles. One quantum emitter (the
donor) is in an excited state with more energy than the other quantum emitter and
then transfers part of its energy to other quantum emitter (acceptors) which is in an
excited state of less energy. These can either be between ions of the same or
different rare earth (clusters). Energy transfer efficiency depends on 1/r (r, distance)
between the donor and the acceptor; for more information, see [95, 96]. Owing to
the context of our book, we will focus on talking about the effects of this process in
REIs and plasmonic nanostructures.
The energy transfer process in REIs may be either favorable or detrimental,
which can be confirmed via its luminescence emission spectrum. An engineering of
this energy transfer process consents, for instance, improving the pumping
efficiency of a solid state laser [97], increasing the bandwidth of band emission
[98–101], tuning of the emission color in the visible region [90, 102, 103], obtaining
the white light [104–106], and improving the performance of a solar cell
[107, 171]. The study of luminescence of the REIs is owing to the ever-increasing
demand for optical sources and amplifiers operating at wavelengths compatible
with fiber communication technology, white light, optical displays, diagnostic
images, and solar energy. Additionally, the 4f electrons of the trivalent REI show
a multiplicity of excited states in both emission and absorption (spectra resulting
consist of groups of sharp lines). Furthermore, multipolar interactionprocesses may
be divided into five categories, (a) multipolar resonance, (b) multipolar transfer,
(c) nonresonant transfer interactions, (d) stepwise up-conversion, and
(e) cooperative luminescence, and can be defined using either the short-range
exchange or longer-range electric multipolar mechanisms. The transfer efficiency
may be defined as:

W DA τD
ηD ¼ , ð13:11Þ
1 þ W DA τD

Note that the two ions need not be both of the same REI; WDA is the rate of the
transfer process and τD is the radiative lifetime of the donor ion in the absence of an
314 V.A.G. Rivera et al.

acceptor ion. Nevertheless, such ion–ion interaction reduces the observed lumines-
cence lifetime:

τobs ¼ τDþA ¼ τD ð1  ηD Þ: ð13:12Þ

The 1/rx dependence of the strength of the multipolar interactions yields x ¼ 6,

8, and 10, i.e., dipole–dipole, dipole–quadrupole, and quadrupole–quadrupole
interactions, respectively. Also, nonradiative energy transfer can effectively cover
much longer distances. WDA can be altered due to the presence of metallic nano-
structure, because this produces a change in the local electric field. For instance,
metallic nanostructure within the tellurite glass doped with REI can modify the
bandwidth of the Stark energy levels, which results in a blue or red shift and/or a
broadening or increase in the REI emissions [38–87, 111].
We can conclude that the dynamic coupling mechanism [108–110] shows that
the interaction energy between one metallic nanostructures and the REI can be
written as:
X rij
HDC ¼ e pNPj nj  3 , ð13:13Þ
i, j
 rij 

Here pNPj (¼αjE) is the ED moment of the nanoparticle induced by the incident
light, rij is the separation distance between the plasmonic nanostructure and REI, nj
is the density of the conduction electrons inside a plasma characterized by carriers
with charge N0 e, and αj was defined in [108, 109]. The Hamiltonian HDC is
expressed as a function of the set of eigenmodes associated with the conduction
electron density generated by, for example, (1) different nanoparticle ( j) through its
collective free oscillations at each resonance frequency of the nanoparticle and their
geometric dependence by means of αj and (2) metallic array structures (hole/slit/
pillar).
The quantum behavior of the REI and plasmonic nanostructures (in the LSPR
mode) can be described through HDC, plasmon–photon coupling, which can be
verified as an enhancement/quenching of luminescence emission from the tellurite
glass, a widening of the broadband emission, and a change in its line shape
[91, 111]. Such interaction is given by nj; a resonant or nonresonant interaction
between the metallic nanostructures and REI only occurs if the REI is near enough
to the metallic nanostructure for the wave functions of the two systems to overlap.
Therefore, the total intensity IT can be written as [94]:

I T ¼ I DþA ¼ I D ð1  ηLSPR Þ þ η0 I 2 : ð13:14Þ

Here, ηLSPR ¼ Γrad/(ΓLSPR + Γnonrad + ΓLSPR), ΓLSPR is the energy transfer rate to
the LSPR mode [112, 113], ID is the intensity emission of the REI, and I2 (¼IA) is
the intensity emission of the metallic nanostructure. In the case with multiple
interactions at different distances between metallic nanostructure and REIs, we
can assume that the charge from the REI is the same, because the Einc employed to
13 Tellurite Glasses for Plasmonics 315

excite these systems (REI-doped tellurite glass with metallic nanostrucutre embed-
ded or onto surface) will be absorbed for the REI due to the cross-sectional
absorption that is more than the metallic nanostructure. Case contrary the light
incident (λ0 6¼ λspp—wavelength surface plasmon resonance) is scattered for the
metallic nanostructure, and this scattered light is absorbed for the REI which emits
and this coupled/resonates (nonresonantes) with the metallic nanostructure. Finally,
η0 is the internal quantum efficiency of the energy transfer. Note that here we did
not consider the shape, size, or material of the metallic nanostructures.
Experimentally, we can calculate if we have an improvement in the lumines-
cence and otherwise a quenching via [8]:
 
1 ID
η0 I 2 ¼ I T  I 1 1  þ : ð13:15Þ
τ I DþA τ

If I2 > 0, then we have an improvement in the luminescence intensity, and

otherwise we have quenching. The energy transfer from the REI ! plasmonic
nanostructure is more probable than the energy transfer from the plasmonic nano-
structure ! REI, since the excited state lifetime of metallic nanostructure is
extremely short (on the order of nanoseconds) in comparison with the higher energy
excited states of the REIs (on the order of microseconds). Nonetheless, the lumi-
nescence enhancement is accredited to a local field enhancement [8, 91, 110, 171];
thus, LSPR gives us the ability to significantly alter the “local density of optical
states” (LDOS). The spontaneous emission of REIs is proportional to the LDOS; we
have an enhanced emission resulting from the changes in the LDOS of the REI or
other QEs near to metallic nanostructure.
In addition, the REI–plasmonic nanostructure interaction waves produce an con-
structive and/or destructive interference, since they are located very close to each
other corresponding to a maximum (IT > 0) and a minimum (IT < 0) of the emission
spectra. Thus, these asymmetric line shapes are resulted of a quantum interference
producing rapid variations in intensity of the various diffracted spectral orders of
REIs and metallic nanostructure within/onto the tellurite glass. Similar asymmetric
profiles were observed in various other systems and settings [8, 114, 171].

13.4.3 Resonance Modes and Tuning

Plasmonic nanostructures have been extensively studied revealing amazing effects

and fascinating phenomena [115]. For instance, the ability to tune the resonant
modes in the plane of the nanostructure due to the propagating SPP (λSPP), which,
e.g., can interact with the various nanostructures embedded/onto the REI-doped
tellurite glass, generates an LSPR with the same λSPP. These traveling waves are
continually interacting on the surface (under continuous excitation), i.e., construc-
tive and destructive interference, and depend on the periodic structures or surface
corrugations. In addition, we recall that we can also have hybridization of the SPP
316 V.A.G. Rivera et al.

on both sides of the metal surface and that the transmission of incident light through
the metal film can create induced surface plasmons (ISP), skin effect. Other studies
are periodic slits in metal films, and here three mechanisms were identified:
(1) SPP, (2) waveguide modes, and (3) optical cavities in slits [116, 117].
Resonance modes and tuning of the transmission properties of a periodic
nanohole array (2D) can be explained in terms of the Bloch EM modes determined
for a unit cell of the nanostructure [118, 119]. These SPPs correspond to standing
SPP Bloch waves on a periodic surface:

ω 2π 2π
kspp ¼ uxy δp  i ~
neff sin θ~ ux  j ~ uy : ð13:16Þ
c a0 a0

kspp is the SPP Bloch wave vector on the periodic nanostructure. Neff is the
effective refractive index of the air–metal interface. θ is the angle of incidence. ~ uxy
is the unit vector in the plane of the film in the direction of projection of the incident
light wave vector, δp ¼ 0 or 1 for p- or s-polarized incident light (relative to the
sample surface), respectively, ~ ux and ~ uy are the unit reciprocal lattice vectors of the
periodic structure, and a0 is the periodicity (same in the x- and y-directions). i and
j are integer numbers corresponding to different directions in the SPP Brillouin
zone and determining the direction of the SPP on a metallic surface.
In this manner, two-dimensional finite periodic nanostructures can match the Φ
for LSPR with the Bragg resonance of the grating. For  example,  for normal
incidence in a square lattice, we have θ ¼ 0 and j~ 
ux j ¼ ~ 
uy ¼ 2π=a0 and the
transmission maxima known as Wood’s anomaly [120]:
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
a0 εd εm
λspp ði; jÞ ¼ pffiffiffiffiffiffiffiffiffiffiffiffi2ffi Re : ð13:17Þ
i þj
2 εd þ εm
a0 pffiffiffiffiffi
λWood ði; jÞ ¼ pffiffiffiffiffiffiffiffiffiffiffiffi2ffi εd : ð13:18Þ
i þj
2

However, Eqs. (13. 1) and (13.18) do not consider the presence of holes (their
size or geometry) and the associated scattering losses, and this neglect the interfer-
ence that gives rise to a resonance shift. Consequently, this predicts the peak
positions at wavelengths slightly shorter than those observed experimentally, but
not, e.g., Rayleigh anomaly or Fano resonance.

13.4.3.1 Effects of Nanostructure Arrays on REI

Today, plasmonic nanostructure investigations provide us with an increasing vari-

ety of metallic nanostructures, which can be either resonant or nonresonant, both
with extraordinary capabilities and with the ability to concentrate light into nano-
scale volumes. Resonant nanostructures are based on scaled radiofrequency
antenna designs. In analogy, an optical antenna is the equivalent of a classical
13 Tellurite Glasses for Plasmonics 317

antenna. But, the term “optical antenna” is extended beyond its common definition
in radio wave technology. An optical antenna is not just a resonator or a strong
scattered and also has the functions of transducer between free radiation and
localized energy. Its performance is defined for the degree of localization and the
magnitude of transduced energy (EM waves for electric currents or vice versa)
[121]. In contrast, nonresonant nanostructures can also be enhancing light concen-
tration, e.g., retardation-based resonators called a feedgap [122]. In addition, the
local field intensities can be improved through “lightning rod” effects
[123]. Plasmonic tapers can provide a broadband or nonresonant
enhancement [124].
Under certain circumstances (specific geometrical parameters, wavelengths, and
angles of incidence) from plasmonic nanostructures are possibly obtained linear
and nonlinear effects when SPPs propagate into metal grating, nanoholes, or slits
[125] in the vicinity of the wavelengths where extraordinary optical transmission
(EOT) occurs.
Photonics science and engineering covers the generation, emission, transmis-
sion, modulation, signal processing, switching, coupling, amplification, and detec-
tion/sensing of light. It covers all technological applications of light and covers
almost the entire EM spectrum, from UV to visible and to the near-, mid-, and
far-infrared [126–128]. Nevertheless, photonics, for instance, involves quantum
states of REI in quantum communication, quantum computation, and optical
communications and is a leading approach to quantum information [129–
131]. For new photonic applications are desired for the nanoscale control over
light, here metallic nanostructures play an important part in the development of new
technologies for photonics.
However, there are still several unresolved issues impeding the tremendous
potential of nanophotonic circuits, such as efficient coupling to nanoscale wave-
guides and the limitations imposed by inevitable SPP propagation losses. Engi-
neering on the optical properties of SPPs and LSPRs in nanophotonics will depend
on the efficient photon–plasmon and plasmon–photon interaction and solved the
problems mentioned above. An effective solution to the SPP propagation can be
achieved by a gain-assisted medium [132, 133], which can improve the coupling
and propagation of these SPPs or LSPR with quantum emitter. REI-doped tellurite
glass can couple with the SPP or with LSPR, and this can be divided in two regimes,
the weak-coupling and strong-coupling regimes.
The spontaneous emissions from a REI-doped tellurite glasses depend on the
crystal field, which can be modified by the presence of metallic nanoparticle or by
the vicinity of a nanostructure array. Such “new” EM environment can be called
cavity quantum electrodynamics (CQED), which involves the interaction of a REI
with a tailored EM field and a high-quality factor (Q) and small volumes (Veff)
[134]. Such interactions are divided in two coupling types which depend on the
interaction between the REI and the surrounding field which is defined by a
coupling frequency [8137]:
318 V.A.G. Rivera et al.

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ω
g ¼ pqe : ð13:19Þ
ε0 hV eff

Weak coupling and strong coupling are determined by the CQED and dependent
on the comparison of g and the damping rates of both the QE and the cavity (γ, k): a
weak-coupling regime is given by g  (γ, k), and a strong-coupling regime is given
by g  (γ, k) [135].
In REI-doped tellurite glasses in the presence of metallic nanoparticle or in the
vicinity of a nanostructure array, weak coupling is associated with the Purcell effect
[134] and can have two processes: (1) favorable, where the intense plasmonic field
increases the excitation rate of the REI, and (2) detrimental, where the plasmonic
field improves the decay rate of the REI into SPP or LSPR modes via the Purcell
effect, i.e., emission quenching. These two effects occur close to the metallic
surface; therefore, to perform the plasmon–photon coupling, an optimum distance
is needed.
A strong coupling, known as Rabi oscillation, can occur between REI-doped
tellurite glasses in the presence of metallic nanoparticle or in the vicinity of a
nanostructure array. Rabi oscillations manifest themselves in an energy splitting of
the light–matter energy levels. Hence, emission from the REI or also quantum
emitter and SPP or LSPR cannot be treated separately, because the interaction of
light and matter must be dealt with using a perturbative approach [8, 30, 94, 136–
142].
Plasmonic array can function as a light scattering concentrator, where each
nanohole acts like an antenna or nanocavity; such characteristics go beyond the
well-established frequency-selective technologies and are expanding to new spec-
tral regions, such as the visible and NIR regions [143]. One of the advantages of
plasmonic cavities is their compatibility with a wide variety of REIs and their
broadband cavity spectra, which enable a broad-spectrum enhancement of the REIs
[83, 144]. For instance, Lo et al. [145] showed that single-crystalline Er-doped ZnO
nanorod arrays on Ag island films are a promising device for enhancing 1540-nm
emissions for optical communication.
Single-photon sources operating at telecommunication wavelengths are crucial
building blocks for fiber-based quantum information networks. In this sense, there is
no doubt that Er3+-doped materials have potential for engineering on-chip light
sources and laser structures operating at 1.54 μm [146, 147]. Using plasmon–photon
interaction, they increase the optical properties in optical telecommunications.

13.4.3.2 Transmission Enhancement

SPPs can concentrate light in sub-wavelength volumes, for their EOT and for
guiding light along the surface of a metal. Besides, metallic nanostructures sustain-
ing surface plasmons can change the excitation and emission properties of locally
excited REIs. Of course, depending on the size and geometry of the nanostructure,
13 Tellurite Glasses for Plasmonics 319

quantum effects can be significant on the REI emission via description of the
electrodynamics. Thus, plasmonic nanostructures act as physically reachable clas-
sical oscillator systems at quantum system, with potential applications and benefits
to designing models for a variety of attractive physical processes.
The most common valence state of REIs in solids is the trivalent state; here, the
4f electrons are weakly perturbed by the charges of the surrounding atoms, and thus
the wavelengths of the emission/absorption transitions are relatively insensitive to
the host material. But the line shapes of the emission spectra and the emission
intensity are governed by its surroundings. When a REI is near a metallic nano-
structure, the number of EM modes can profoundly change and decay pathways
available to an emitter, increasing or decreasing the radiative rate of the REI, i.e., it
is possible to manipulate the LDOS; we can found other examples in [148, 149].
In REI-doped tellurite glass, the spontaneous emission rate (γ) of an REI is
proportional to the LDOS, and thus we can have an increased emission owing to the
changes in the LDOS of the REI. A large LDOS can increase not only γ but also the
stimulated emission process within the metallic nanostructure, namely, a lasing
action without requiring stronger pump power [150]. Intense optical fields are
coupled to oscillating charges in the metal nanostructure which can sustain a
large LDOS [151]. In this manner, we can improve the emission of one nanocavity
due the nanocavity array and in consequence increase the stimulated emission from
the REI near or into the nanocavities. Hence, the study of the LDOS in
nanostructures on a gain medium is essential for understanding and developing
novel photonic nanodevices.
We can employed a quantum treatment of SPPs and LSPR to describe the
interaction between light and matter. Then, we use quantization for EM fields in
dispersive and absorptive media introduced by Gruner and Welsch, where the EM
energy associated with surfaces waves is defined as [136, 152]
X
U¼ ε0 ω2 S Ak A∗ ∗
k þ Ak Ak ð13:20Þ
k

where Ak is the amplitude of the vector potential dependent on position and time
and S ¼ s(r, t) is the displacement of a small volume of charge inside a plasma
characterized by carriers with charge e, mass m, and density N [85]. Additionally,
we can suppose that the REIs may occupy different sites in the host matrix; thus, a
direct coupling between the excited states of the REIs and the plasmonic nanocavity
modifies the Stark level energies, i.e., the local field when it is excited, so the
oscillator strength (Pstrength) of a spectral line, corresponding to a transition from the
REI’s ground level, i, to a component, f, of the excited level is given by [8]
 
8π2 mν nX  2 X   o
Pstrength ¼ χ 〈i j Dq nearestneighbor j f 〉 þ 〈i j Dq nanocavity j f 〉2 ,
3hð2J þ 1Þ
ð13:21Þ
320 V.A.G. Rivera et al.

where ν is the frequency of the line and q ¼ 2, 4, and 6. The second term in the
parentheses in this equation has been modified from the theory of Judd [153] and
represents the electric dipole transition due to the LSPR modes of the nanocavity.
The factor χ is an adjustable factor that depends on the refractive index of the
medium in which the REI is embedded and J is the total angular momentum.
An experiment that demonstrates about the improved EOT in both single and
array nanocavities with REIs can be found in [154]. We also find enhanced EOT
with REIs in different plasmonic arrays [143, 144] in the visible and near-infrared
regions.

13.4.3.3 Focusing of Surface Plasmon Polaritons

Lens provide us with a convenient way to manipulate light; however, dielectric

lenses cannot focus light to spots less than about half a wavelength of light (λ/2),
and in addition, dielectric resonators have EM mode volumes, Veff, limited to
Veff  (λ/2)3 On the other hand, plasmonic nanostructures let us move beyond
these limits, since these metallic nanostructures can concentrate and resonate light
at the nanoscale.
Focused light to the nanoscale was a technological challenge; nevertheless,
today’s developments in SPPs allows, here, nanofocusing of SPPs due to optical
energy, which is compressed and concentrated into the nanoscale. Nanofocusing is
one of the key tools for efficient coupling of light and light-carried information into
nano-waveguides, interconnectors, and optical nanodevices [155], resolution of
microscopes [156, 157].
Plasmon nanofocusing can occur in an adiabatic regime or a non-adiabatic
regime [158–161]. For instance, a conical metal tip is capable of focusing light to
a nanoscale spot, permitting the accumulation of energy [162, 163].
Due to the strong confinement by the metallic nanostructures, we can profoundly
alter the light emission properties of nearby quantum emitters [164–166], i.e., we
can manipulate radiative decay by (1) increasing optical excitation rates, (2) mod-
ifying radiative and nonradiative decay rates, and (3) altering emission direction-
ality. Experimental evidence for REI-doped tellurite glass interaction with metallic
nanostructure can be found in [154]. The excitation of the REI ions depends on their
localization to the metallic film and on their proximity to the nanocavity.
Depending on a variety of parameters, the focusing (REI–nanocavity coupling)
can exhibit either increased or decreased (quenching) luminescence [167]. From
these effects, we have the ability to build a nanolaser that combines the advantages
of photonic lasers (e.g., directional beam emission) with those of plasmonic
nanocavity (e.g., nanoscale optical confinement and ultrafast response). In other
words, a promising future of these nanocavities in a REI-doped tellurite glass, we
can anticipate an outstanding role for REIs in the new generation of optoelectronic
nanodevices.
13 Tellurite Glasses for Plasmonics 321

13.4.3.4 Polarization Control

Polarizers are made from birefringent materials or grid plates, which are used for
the conversion of polarization states [168], as half-wave and quarter-wave plates,
for instance. Such materials change the propagation velocities of the medium,
owing to the nature of the refractive index. Metamaterials also offered capacity to
manipulate the polarization state of light [169]. In this manner, the implementation
of nanostructures, combined with REIs into a tellurite glasses, enables an increased
capacity to control the polarization by adjusting the shape and dimensions of the
nanostructure.
Nanostructures of noble metals hosted in REI-doped tellurite glasses [170, 171]
show strong absorption of visible light. However, the size and shape of the metallic
nanostructures, and furthermore the dispersive properties of the glass are critical
factors that characterize the enhancement of the local electric field in the vicinity of
the metallic nanostructures.
Achieving an enhanced electric field from LSPR modes in metallic-
nanostructure arrays is more effective, and such feature occurs because in an
array configuration it is possible to reproduce effects of well-known devices, like
waveguides, lenses, and antennas [172]. Consequently an increase in the excitation
and emission rates of the REI can be achieved, because the radiation pattern from
the localized plasmon provides efficient coupling between the emission (photon
generation) of the radiation array and the emission from the REI-doped tellurite
glasses [173]. In this manner, it is important to choose the best format for the
nanostructure.
Let us assume that a periodic metallic-nanostructure array can be treated as a
nanoantenna set. Also, the mode of each nanoantenna governs the polarization of
the emitted radiation due to the structure (geometry or shape) [174].
The radiation pattern of a REI-doped tellurite glass is generally dipolar, and
coupling between REI and metallic nanostructure (set of nanoantennas) can be
efficient because the radiation pattern of the nanostructure matches with that of the
REI. Quantitatively, such a matching is represented by the collection efficiency η0.
An increased radiation pattern displays a directionality D’ defined by the
nanoantenna array, which defines the shape of the pattern [8]:

Pðθ; φÞ
D0 ¼ 4π R ð13:22Þ
Pðθ; φÞdΩ

where θ and φ are the polar and azimuthal angles, respectively, and P(θ, φ) is the
power of the set of nanoantennas. The relation that describes the coupling between
the RE and this array is represents as:

G ¼ η0 D0 ð13:23Þ

where G defines the gain in the emission of the radiation from the nanostructures,
which is proportional to the collection efficiency η0. In this case, the coupling
322 V.A.G. Rivera et al.

between the REIs and the nanoantennas is coherent, and in this frame, such a system
can function as a polarization selector, because only a specific output may emerge
from the REI-doped tellurite glass with the metallic nanostructure. In other words,
the configuration metallic nanostructure and the REI-doped tellurite glass can also
be used as a polarizer in the nanoscale with a determined wavelength.
A crucial parameter to control the polarizer is the phase, which depends on the
dispersion properties of the materials that constitute the nanostructure (the metallic
thin film and/or the glass). Moreover, the geometry of the nanostructure array also
provides a means of controlling of the amplitudes of the electric field. For instance,
a nanocavity with an elliptical shape, fabricated at the center of the nanostructure,
can be designed to control the polarization state of the output light [175, 176].
Here, the electric component of the incident and reflected waves can be projected
inside the nanocavity in two parallel directions, on the major axis of the nanocavity.
Thus, the nanocavity can be treated as a Lorentz oscillator. In this sense, further-
more to the features of the nanocavity and in order to control the polarization state
of REIs, it is also important to analyze the tellurite glass where the nanostructure
will be fabricated. Specifically, in REI-doped tellurite glasses, it is reasonable to
describe their optical anisotropy resulting from the presence of such ions. The
crystalline potential, owing to the positions of the atoms that form the lattice in
the host matrix, is regarded as a perturbation that splits the energy levels from the
REI. We consider the probability of spontaneous emission of a REI starting from an
excited state J, which is defined by [177]

Σq ΣJ Aq ð φi  φf Þ
Atotal ¼ , ð13:24Þ
3

where index q represents the polarization state by the REI. The numerator is written
as a function of the correction of perturbation levels:
(       )
X 〈φi jV crys φβ 〉〈φβ Pφf 〉 〈φi jPφβ 〉〈φβ V crys φf 〉
Aq ðφi  φf Þ ¼ þ :
β
Ei  E β Ef  Eβ
ð13:25Þ

We notice from the expression above that, in theory, we can control the polar-
ization state of the excitation light P, because it is possible control part of the
polarization state of the light emitted by the REI.

13.5 Potential Applications

The interaction/coupling between a REI with a metallic nanostructure results in a

polarization of the metallic electrons or, in the case of transition between electronic
levels of the REI, a sustained field enhancement resulting from resonance coupling
13 Tellurite Glasses for Plasmonics 323

or an energy transfer owing to nonresonant coupling by the metallic nanostructures

and the optically active surrounding medium. In other words, we have the ability to
manipulate the electronic states and decay pathways available in a REI at the
nanoscale, thus increasing or decreasing the radiative rate of a REI. In addition,
by overcoming the SPP propagation losses within a gain medium, the propagation
can be sustained owing to efficient coupling between a surface plasmon polariton
(SPP) and the REI. Alternatively, the simplest solution is to replace the dielectric
substrate for another dielectric that possesses a higher refractive index.
From this point of view, we have the ability to engineer a way to counteract the
absorption losses in metal, to enhance luminescence, and to control the polarization
and phase of the REIs. This provides us with exciting applications of SPPs and
localized surface plasmon resonance (LSPR) in nanophotonics, with tremendous
commercialization potential in the following areas:
1. Optical telecommunication, owing to their capacity for integrated optics at the
nanoscale.
2. Sensing, since the spectral position of the resonances of a metallic nanostructure
depends on the dielectric environment within the EM near field and can be
improved by the presence of an REI and its unique frequency-converting
capability (both down-conversion and up-conversion).
3. Clean energy, obtained by increasing the response of a p-n junction, by means of
localized SPPs or LSPR in REI-doped materials.
In [8], a roadmap highlighting some exciting and as-yet-unexplored topics
related to these points was presented, with the promise of achieving the next
stage of research in quantum plasmonics, using REI-doped materials (a gain
medium generating hybrid SPP/LSPR and loss compensation) with a very fruitful
and productive future. Such applications are indeed exciting motivation, despite the
fact that we do not know exactly what form future quantum–plasmonic technolo-
gies will take. However, we can say that in the photonic quantum technologies
being developed, it is most likely that quantum states of light will be transmitted
and that information processing will be performed on these states. It is worth saying
that if we are to realize these technologies, then we will need to constantly exploit
the latest developments in the nanophotonics field.
Currently, we have a variety of technological routes open in the photonics field,
since there are a great variety of REI-doped tellurite glasses that are candidates to
be employed in the development of new light sources for nanodevices, such as
lasers, LEDs, and single-photon sources, but also in chemistry and life sciences
where they act as nanoscopic probes and labels.
Nonetheless, REIs are, from our point view, the most promising because of their
unique characteristics, such as their magnetic, luminescent, and electrochemical
properties. They suit newly emerging devices for their greater efficiency, reduced
weight, high speed, greater durability, and miniaturization potential. In this route,
nanophotonics is not only about very small photonic circuits and nanochips but also
about new ways of sculpting the flow of light by means of nanostructures and/or
plasmonic NPs exhibiting fascinating optical properties that we are learning to
324 V.A.G. Rivera et al.

manipulate and control. Therefore, a deep investigation of these effects may

provide us with new perspective on plasmonics and may yield many more future
applications, as listed above. This can be very helpful for developing applications of
such metallic nanostructures.
Outstanding reviews on plasmonic circuitry can be found in [178–180] and in the
renowned paper by Harry A. Atwater about the promise of plasmonics [181]. In his
words, “a technology that squeezes EM waves into minuscule structures may yield
a new generation of superfast computer chips and ultrasensitive molecular
detectors.”
Ultimately, practical photonic circuits use a combination of plasmonic and
dielectric components, taking advantage of the best performance available. In
other words, whether or not the outcome will lead to hybrid SPP/LSPR circuits,
the activity stimulated by plasmonics and REI research will bring the long-held
dream of an integrated optical circuit ever closer to reality. Of course, there are
countless routes to merging the best features of photonics and plasmonics, with the
only limits being our imagination and our determination to realize the promise of
nanophotonic technology.

Acknowledgment This work was financially supported by the Brazilian agencies FAPESP,
CNPq, and CEPOF/INOF. V.A.G. Rivera thanks to FAPESP for financial support (project 2009/
08978-4 and 2011/21293-0) that allowed my post-doctoral.

References

1. D.K. Gramotnev, S.I. Bozhevolnyi, Plasmonics beyond the diffraction limit. Nat. Photon. 4
(2), 83–91 (2010)
2. J. Grandidier et al., Surface plasmon routing in dielectric-loaded surface plasmon polariton
waveguides. Proc. SPIE 7033, 70330S–70330S8 (2008)
3. A.V. Krasavin, A.V. Zayats, Electro-optic switching element for dielectric-loaded surface
plasmon polariton waveguides. Appl. Phys. Lett. 97(4), 041107 (2010)
4. R.M. Briggs, J. Grandidier, S.P. Burgos, E. Feigenbaum, H.A. Atwater, Efficient coupling
between dielectric loaded plasmonic and silicon photonic waveguides. Nano Lett. 10(12),
4851–4857 (2010)
5. V.A.G. Rivera et al., Focusing surface plasmons on Er3+ ions with convex/concave
plasmonic lenses. Proc. SPIE 8269, 82692I-1–82696I-6 (2012)
6. V.A.G. Rivera et al., Luminescence enhancement of Er3+ ions from electric multipole
nanostructure arrays. Proc. SPIE 8269, 826921–882697 (2012)
7. O.B. Silva, V.A.G. Rivera, E. Marega, Engineering of the extraordinary optical transmission
of metallic gratings via Er3 + -doped tellurite glass. SPIE OPTO 93741, 7 (2015)
8. V.A.G. Rivera, O.B. Silva, Y. Ledemi, Y. Messaddeq, E. Arega Jr., Collective Plasmon-
Modes in Gain Media: Quantum Emitters and Plasmonic Nanostructures (Springer,
New York, NY, 2016)
9. S.A. Maier, Plasmonics Fundamentals and Applications (Springer, New York, NY, 2007)
10. D.L. Jeanmarie, R.P. Van Duyne, Surface Raman spectroelectrochemistry, part 1: heterocy-
clic, aromatic, and aliphatic amines adsorbed on the anodized silver electrode. J. Electroanal.
Chem. 84, 1–20 (1977)
13 Tellurite Glasses for Plasmonics 325

11. C.L. Haynes, C.R. Yonzon, X. Zhang, Surface-enhanced Raman sensors: early history and
the development of sensors for quantitative biowarfare agent and glucose detection. J. Raman
Spectrosc. 36, 471–484 (2005)
12. C.R. Yonzon, C.L. Haynes, X. Zhang, J.T. Walsh, R.P. Van Duyne, A glucose biosensor
based on surface-enhanced Raman scattering: improved partition layer, temporal stability,
reversibility, and resistance to serum protein interference. Anal. Chem. 76, 78–85 (2004)
13. S.R. Emory, S. Nie, Probing single molecules and single nanoparticles by surface-enhanced
Raman scattering. Science 275, 1102–1106 (1997)
14. A.C.R. Pipino, G.C. Schatz, R.P. Van Duyne, Surface-enhanced second harmonic diffraction:
selective enhancement by spatial harmonics. Phys. Rev. B 49, 8320–8330 (1994)
15. W. Yang, J. Hulteen, G.C. Schatz, R.P. Van Duyne, A surface-enhanced hyper-Raman and
surface-enhanced Raman scattering study of trans-1,2-bis(4-pyridyl)ethylene adsorbed onto
silver film over nanosphere electrodes. Vibrational assignments: experiment and theorys.
J. Chem. Phys. 104, 4313–4323 (1996)
16. C.D. Geddes, K. Aslan, I. Gryczynski, J. Malicka, J.R. Lakowicz, Noble-metal surfaces for
metal enhanced fluorescence. Rev. Fluoresc. 1, 365–401 (2004)
17. P.J.G. Goulet, R.F. Aroca, Surface-enhancement of fluorescence near noble metal
nanostructures. Top. Fluoresc. Spectrosc. 8, 223–247 (2005)
18. A.J. Haes, L. Chang, W.L. Klein, R.P. Van Duyne, Detection of a biomarker for Alzheimer’s
disease from synthetic and clinical samples using a nanoscale optical biosensor. J. Am.
Chem. Soc. 127, 2264–2271 (2005)
19. A.J. Haes, D.A. Stuart, S.M. Nie, R.P. Van Duyne, Using solution-phase nanoparticles,
surface-confined nanoparticle arrays and single nanoparticles as biological sensing platforms.
J. Fluoresc. 14, 355–367 (2004)
20. C.R. Yonzon, E. Jeoung, S. Zou, G.C. Schatz, M. Mrksich, R.P. Van Duyne, A comparative
analysis of localized and propagating surface plasmon resonance sensors: the binding of
concanavalin A to a monosaccharide functionalized self-assembled monolayer. J. Am. Chem.
Soc. 126, 12669–12676 (2004)
21. C.R. Yonzon, D.A. Stuart, X. Zhang, A.D. McFarland, C.L. Haynes, R.P. Van Duyne,
Towards advanced chemical and biological nanosensors: an overview. Talanta 67, 438–448
(2005)
22. J.M. Brockman, B.P. Nelson, R.M. Corn, Surface plasmon resonance imaging measurements
of ultrathin organic films. Annu. Rev. Phys. Chem. 51, 41–63 (2000)
23. W. Srituravanich, N. Fang, C. Sun, Q. Luo, X. Zhang, Plasmonic nanolithography. Nano Lett.
4, 1085–1088 (2004)
24. A. Sundaramurthy, P.J. Schuck, N.R. Conley, D.P. Fromm, G.S. Kino, W.E. Moerner,
Toward nanometer-scale optical photolithography: utilizing the near-field of bowtie optical
nanoantennas. Nano Lett. 6, 355–360 (2006)
25. C.E.H. Berger, T.A.M. Beumer, R.P.H. Kooyman, J. Greve, Surface plasmon resonance
multisensing. Anal. Chem. 70, 703–706 (1998)
26. M. Malmqvist, Biospecific interaction analysis using biosensor technology. Nature 361,
186–187 (1993)
27. R.P. Van Duyne, Molecular plasmonics. Science 306, 985–986 (2004)
28. W. Knoll, Interfaces and thin films as seen by bound electromagnetic waves. Annu. Rev.
Phys. Chem. 49, 569–638 (1998)
29. A.J. Haes, R.P. Van Duyne, A unified view of propagating and localized surface plasmon
resonance biosensors. Anal. Bioanal. Chem. 379, 920–930 (2004)
30. A.J. Haes, C.L. Haynes, A.D. McFarland, S. Zou, G.C. Schatz, R.P. Van Duyne, Plasmonic
materials for surface-enhanced sensing and spectroscopy. MRS Bull. 30, 368–375 (2005)
31. X. Zhang, M.A. Young, O. Lyandres, R.P. Van Duyne, Rapid detection of an anthrax
biomarker by surface-enhanced Raman spectroscopy. J. Am. Chem. Soc. 127, 4484–4489
(2005)
326 V.A.G. Rivera et al.

32. H. Knobloch, H. Brunner, A. Leitner, F. Aussenegg, W. Knoll, Probing the evanescent field
of propagating plasmon surface polaritons by fluorescence and Raman spectroscopies.
J. Chem. Phys. 98, 10093–10095 (1993)
33. K.L. Kelly, E. Coronado, L. Zhao, G.C. Schatz, The optical properties of metal nanoparticles:
the influence of size, shape, and dielectric environment. J. Phys. Chem. B 107, 668–677
(2003)
34. M.M. Miller, A.A. Lazarides, Sensitivity of metal nanoparticle surface plasmon resonance to
the dielectric environment. J. Phys. Chem. B 109, 21556–21565 (2005)
35. T.R. Jensen, M.L. Duval, L. Kelly, A. Lazarides, G.C. Schatz, R.P. Van Duyne, Nanosphere
lithography: effect of the external dielectric medium on the surface plasmon resonance
spectrum of a periodic array of silver nanoparticles. J. Phys. Chem. B 103, 9846–9853 (1999)
36. K. Hamamoto, R. Micheletto, M. Oyama, A.A. Umar, S. Kawai, Y. Kawakami, An original
planar multireflection system for sensing using the local surface plasmon resonance of gold
nanospheres. J. Opt. A 8, 268–271 (2006)
37. M.G. Albrecht, J.A. Creighton, Anomalously intense Raman spectra of pyridine at a silver
electrode. J. Am. Chem. Soc. 99, 5215–5217 (1977)
38. M. Moskovits, Surface roughness and the enhanced intensity of Raman scattering by mole-
cules adsorbed on metals. J. Chem. Phys. 69, 4159–4161 (1978)
39. A. Wokaun, Surface-enhanced electromagnetic processes. Solid State Phys. 38, 223–294
(1984)
40. C.L. Haynes, R.P. Van Duyne, Nanosphere lithography: a versatile nanofabrication tool for
studies of size-dependent nanoparticle optics. J. Phys. Chem. B105, 5599–5611 (2001)
41. L.J. Sherry, S.H. Chang, G.C. Schatz, R.P. Van Duyne, B.J. Wiley, Y. Xia, Localized surface
plasmon resonance spectroscopy of single silver nanocubes. Nano Lett. 5, 2034–2038 (2005)
42. K.A. Willets, R.P. Van Duyne, Localized surface plasmon resonance spectroscopy and
sensing. Annu. Rev. Phys. Chem. 58, 267–297 (2007)
43. J.J. Mock, M. Barbic, D.R. Smith, D.A. Schultz, S. Schultz, Shape effects in plasmon
resonance of individual colloidal silver nanoparticles. J. Chem. Phys. 116, 6755–6759 (2002)
44. T.R. Jensen, R.P. Van Duyne, S.A. Johnson, V.A. Maroni, Surface-enhanced infrared spec-
troscopy: a comparison of metal island films with discrete and nondiscrete surface plasmons.
Appl. Spectrosc. 54, 371–377 (2000)
45. A.M. Moran, J. Sung, E.M. Hicks, R.P. Van Duyne, K.G. Spears, Second harmonic excitation
spectroscopy of silver nanoparticle arrays. J. Phys. Chem. B 109, 4501–4506 (2005)
46. A.V. Whitney, J.W. Elam, S. Zou, A.V. Zinovev, P.C. Stair, et al., Localized surface plasmon
resonance nanosensor: a high-resolution distance-dependence study using atomic layer depo-
sition. J. Phys. Chem. B 109, 20522–20528 (2005)
47. C.L. Nehl, H. Liao, J.H. Hafner, Optical properties of star-shaped gold nanoparticles. Nano
Lett. 6, 683–688 (2006)
48. A.D. McFarland, R.P. Van Duyne, Single silver nanoparticles as real-time optical sensors
with zeptomole sensitivity. Nano Lett. 3, 1057–1062 (2003)
49. R. Jin, Y. Cao, C.A. Mirkin, K.L. Kelly, G.C. Schatz, J.G. Zheng, Photoinduced conversion
of silver nanospheres to nanoprisms. Science 294, 1901–1903 (2001)
50. C.L. Haynes, A.D. McFarland, L. Zhao, R.P. Van Duyne, G.C. Schatz, et al., Nanoparticle
optics: the importance of radiative dipole coupling in two-dimensional nanoparticle arrays.
J. Phys. Chem. B 107, 7337–7342 (2003)
51. A.A. Athawale, P.P. Katre, M.B. Majumdar, Nonaqueous phase synthesis of copper
nanoparticles. J. Nanosci. Nanotechnol. 5, 991–993 (2005)
52. G.C. Schatz, M.A. Young, R.P. Van Duyne, in Surface-Enhanced Raman Scattering: Physics
and Applications, ed by K. Kneipp, M. Moskovits, H. Kneipp. Electromagnetic mechanism of
SERS (Springer-Verlag, Berlin, 2006), pp. 19–46
53. G. Mie, Contributions to the optics of turbid media, especially colloidal metal solutions. Ann.
Phys. 25, 377–445 (1908)
13 Tellurite Glasses for Plasmonics 327

54. C.F. Bohren, D.R. Huffman, Absorption and Scattering of Light by Small Particles (Wiley,
New York, 1983)
55. S. Link, M.A. El-Sayed, Spectral properties and relaxation dynamics of surface plasmon
electronic oscillations in gold and silver nano-dots and nano-rods. J. Phys. Chem. B 103,
8410–8426 (1999)
56. A. Taflove, Computational Electrodynamics: The Finite-Difference Time Domain Method
(Artech House, Boston, 1995)
57. R.X. Bian, R.C. Dunn, X.S. Xie, P.T. Leung, Single molecule emission characteristics in
near-field microscopy. Phys. Rev. Lett. 75, 4772–4775 (1995)
58. A.J. Haes, R.P. Van Duyne, A nanoscale optical biosensor: sensitivity and selectivity of an
approach based on the localized surface plasmon resonance of triangular silver nanoparticles.
J. Am. Chem. Soc. 124, 10596–10604 (2002)
59. A.D. McFarland, M.A. Young, J.A. Dieringer, R.P. Van Duyne, Wavelength-scanned sur-
face-enhanced Raman excitation spectroscopy. J. Phys. Chem. B 109, 11279–11285 (2005)
60. R.P. Van Duyne, Laser excitation of Raman scattering from adsorbed molecules on electrode
surfaces. Chem. Biochem. Appl. Lasers 4, 101–185 (1979)
61. T.R. Jensen, M.D. Malinsky, C.L. Haynes, R.P. Van Duyne, Nanosphere lithography: tunable
localized surface plasmon resonance spectra of silver nanoparticles. J. Phys. Chem. B 104,
10549–10556 (2000)
62. X. Zhang, E.M. Hicks, J. Zhao, G.C. Schatz, R.P. Van Duyne, Electrochemical tuning of
silver nanoparticles fabricated by nanosphere lithography. Nano Lett. 5, 1503–1507 (2005)
63. J.J. Mock, D.R. Smith, S. Schultz, Local refractive index dependence of plasmon resonance
spectra from individual nanoparticles. Nano Lett. 3, 485–491 (2003)
64. A.M. Michaels, M. Nirmal, L.E. Brus, Surface enhanced Raman spectroscopy of individual
rhodamine 6G molecules on large Ag nanocrystals. J. Am. Chem. Soc. 121, 9932–9939
(1999)
65. L.J. Sherry, R. Jin, C.A. Mirkin, G.C. Schatz, R.P. Van Duyne, Localized surface plasmon
resonance spectroscopy of single silver triangular nanoprisms. Nano Lett. 6, 2060–2065
(2006)
66. A.J. Haes, S. Zou, G.C. Schatz, R.P. Van Duyne, A nanoscale optical biosensor: the short
range distance dependence of the localized surface plasmon resonance of silver and gold
nanoparticles. J. Phys. Chem. B 108, 6961–6968 (2004)
67. R. Jin, Y.C. Cao, E. Hao, G.S. Metraux, G.C. Schatz, C.A. Mirkin, Controlling anisotropic
nanoparticle growth through plasmon excitation. Nature 425, 487–490 (2003)
68. M.D. Groner, J.W. Elam, F.H. Fabreguette, S.M. George, Electrical characterization of thin
Al2O3 films grown by atomic layer deposition on silicon and various metal substrates. Thin
Solid Films 413, 186–197 (2002)
69. M. Ritala, M. Leskela, in Handbook Thin Film Materials, Academic, ed by H. S. Nalwa.
Atomic layer deposition, vol 1 (San Diego, CA, 2002), pp. 103–159
70. A.W. Ott, J.W. Klaus, J.M. Johnson, S.M. George, Al2O3 thin film growth on Si(100) using
binary reaction sequence chemistry. Thin Solid Films 292, 135–144 (1997)
71. J.A. Dieringer, A.D. McFarland, N.C. Shah, D.A. Stuart, A.V. Whitney, et al., Surface
enhanced Raman spectroscopy: new materials, concepts, characterization tools, and applica-
tions. Faraday Discuss. 132, 9–26 (2005)
72. C.L. Haynes, R.P. Van Duyne, Plasmon-sampled surface-enhanced Raman excitation spec-
troscopy. J. Phys. Chem. B 107, 7426–7433 (2003)
73. A.J. Haes, S. Zou, J. Zhao, G.C. Schatz, R.P. Van Duyne, Localized surface plasmon
resonance spectroscopy near molecular resonances. J. Am. Chem. Soc. 128, 10905–10914
(2006)
74. T. Sekiya, N. Mochida, A. Ohtsuka, M. Tonokawa, Raman-spectra of MO1/2–TeO2 (M ¼ Li,
Na, K, Rb, Cs and Tl) glasses. J. Non Cryst. Solids 144, 128–144 (1992)
75. W. Vogel, Chemistry of glass (American Ceramic Society, Westerville, OH, 1985)
76. T. Sekiya, N. Mochida, A. Ohtsuka, Raman-spectra of MO–TeO2 (M ¼ Mg, Sr, Ba and Zn)
glasses. J. Non Cryst. Solids 168, 106–114 (1994)
328 V.A.G. Rivera et al.

77. T. Sekiya, N. Mochida, A. Ohtsuka, A. Soejima, Raman-spectra of BO3/2–TeO2 glasses.

J. Non Cryst. Solids 151, 222–228 (1992)
78. T. Sekiya, N. Mochida, A. Soejima, Raman-spectra of binary tellurite glasses containing
tri-valent or tetra-valent cations. J. Non Cryst. Solids 191, 115–123 (1995)
79. A. Jha, B. Richards, G. Jose, T. Teddy-Fernandez, P. Joshi, X. Jiang, Joris Lousteau. Progr.
Mater. Sci. 57, 1426–1491 (2012)
80. O.V. Mazurin, M.V. Streltsina, T.P. Shavio-Shavaikovskaya, Handbook of glass data
(Elsevier, Amsterdam, 1991)
81. Y. Himei, A. Osaka, T. Nanba, Y. Miura, Coordination change of Te atoms in binary tellurite
glasses. J. Non Cryst. Solids 177, 164–169 (1994)
82. M.D. O’Donnell, C.A. Miller, D. Furniss, V.K. Tikhomirov, A.B. Seddon, Fluorotellurite
glasses with improved mid-infraredtransmission. J. Non Cryst. Solids 331, 48–57 (2003)
83. O.B. Silva, V.A.G. Rivera, F.A. Ferri, Y. Ledemi, A.R. Zanatta, Y. Messaddeq, E. Marega Jr.,
Proc. SPIE 8809, 88092X–880921 (2013)
84. V.A.G. Rivera, Y. Ledemi, S.P.A. Osorio, D. Manzani, Y. Messaddeq, L.A.O. Nunes,
E. Marega Jr., J. Non Cryst. Solids 358(2), 399 (2012)
85. M. Finazzi, F. Ciaccacci, Phys. Rev. B 86, 035428 (2012)
86. G. Schmid (ed.), Nanoparticles (Wiley-VSH, KGaA, Boschstr, 2010)
87. H. Raether, Surface Plasmon on Smooth and Rough Surfaces and on Gratings, Springer
Tracts in Modern Physics, vol 111 (Springer, New York, NY, 1988)
88. V.M. Shalaev, R. Botet, R. Jullien, Phys. Rev. B 44, 12216 (1991)
89. P. Bharadwaj, B. Deutsch, L. Novotny, Adv. Opt. Photon. 1, 438 (2009)
90. V.A.G. Rivera, Y. Ledemi, S.P.A. Osorio, D. Manzani, F.A. Ferri, S.J.L. Ribeiro,
L.A.O. Nunes, E. Marega Jr., J. Non Cryst. Solids 378, 126–134 (2013)
91. O.L. Malta, P.O. Santa-Cruz, G.F. de Sá, F. Auzel, JOL 33, 261 (1985)
92. D.K. Lynch, Astro. J. 467, 894 (1996)
93. C.F. Bohren, D.R. Huffman, Absorption and scattering of light by small particles (John
Wiley & Sons, New York, NY, 1983)
94. V.A.G. Rivera, F.A. Ferri, E. Marega Jr., in ed. By K.Y. Kim, Localized Surface Plasmon
Resonances: Noble Metal Nanoparticle Interaction with Rare-Earth Ions, Chapter 11, Intech,
Rijeka (2012).
95. P. Wu, L. Brand, Anal. Biochem. 218, 1 (1994)
96. R.M. Clegg, Curr. Opin. Biotechnol. 6, 103 (1995)
97. E. Snitzer, R. Woodcock, Appl. Phys. Lett. 6, 45 (1965)
98. D. Milanese, M. Vota, Q. Chen, J. Xing, G. Liao, H. Gebavi, M. Ferraris, N. Coluccelli,
S. Taccheo, J. Non Crys. Solids 354, 1955 (2008)
99. A. Pal, A. Dhar, S. Das, S.Y. Cehen, T. Sun, R. Sen, K.T.V. Grattan, Opt. Express 18(5), 5068
(2010)
100. B. Zhou, L. Tao, Y.H. Tsang, W. Jin, E.Y.-B. Pun, Opt. Express 20(4), 3803 (2012)
101. V.A.G. Rivera, M. El-Amraoui, Y. Ledemi, Y. Messaddeq, E. Marega Jr., JOL 145,
787 (2014)
102. H. Deng, S. Yang, S. Xiao, H.-M. Gong, Q.-Q. Wang, J. Am. Chem. Soc. 130, 2032 (2008)
103. Y. Ledemi, D. Manzani, S.J.L. Ribeiro, Y. Messaddeq, Opt. Mater. 33(12), 1916 (2011)
104. M. Haase, H. Schfer, Angew. Chem. Int. Ed. 50, 5808 (2011)
105. J Zhao, D Jin, EP Schartner, Y Lu, Y Liu, AV Vyagin, L Zhang, JM. Dawes, P Xi, JA Piper,
EM. Goldys, TM Monro, Nat. Nanotechnol. 8, 729 (2013).
106. Y. Ledemi, Andrée-Anne Trudel, V.A.G. Rivera, S. Chenu, E. Véron, L.A.O. Nunes,
M. Allix, Y. Messaddeq, White light and multicolor emission tuning in triply doped Yb3+/
Tm3+/Er3+ novel fluoro-phosphate transparent glass-ceramics (2014). doi:10.1039/
C4TC00455H
107. A. Shalava, B.S. Richards, M.A. Green, Sol. Energ. Mater. Sol. Cells 91, 829 (2007)
108. C.K. Jorgensen, B.R. Judd, Mol. Phys. 8, 281 (1964)
109. F. Michael Reid, F.S. Richardson, J. Chem. Phys. 79(12), 5735 (1983)
13 Tellurite Glasses for Plasmonics 329

110. OL Malta, LD Carlos, Quim. Nova 26(6), 889 (2003).

111. V.A.G. Rivera, Y. Ledemi, S.P.A. Osorio, F.A. Ferri, Y. Messaddeq, L.A.O. Nunes,
E. Marega Jr., Proc. SPIE 8621, 86211J–862111 (2013)
112. A. Gopinath, S.V. Boriskina, S. Yerci, R. Li, L. Dal Negro, Appl. Phys. Lett. 96, 071113
(2010)
113. J.R. Lakowicz, Anal. Biochem. 337, 171 (2005)
114. A.E. Miroshnichenko, S. Flach, Y.S. Kivshar, Rev. Mod. Phys. 82, 2257 (2010)
115. V. Shalaev, Engineering Space for Light with metamaterials (NATO-ASI, Ottawa, ON,
2008)
116. U. Schroter, D. Heitmann, Phys. Rev. B 58, 15419 (1998)
117. E. Popov, M. Neviere, S. Enoch, R. Reinisch, Phys. Rev. B 62, 16100 (2000)
118. W.L. Barnes, T.W. Preist, S.C. Kitson, J.R. Sambles, Phys. Rev. B 54, 6227 (1996)
119. S.A. Darmanyan, A.V. Zayats, Phys. Rev. B 67, 035424 (2003)
120. R.W. Wood, Phys. Rev. 48, 928 (1935)
121. P. Bharadwaj, B Deutsch, L Novotny. Adv. Opt. Phot. 1, 438 (2009)
122. P.J. Schuck, D.P. Fromm, A. Sundaramurthy, G.S. Kino, W.E. Moerner, Phys. Rev. Lett. 94,
017402 (2005)
123. J. Gersten, A. Nitzan, J. Chem. Phys. 73, 3023–3037 (1980)
124. R. Zia, J.A. Schuller, M.L. Brongersma, Near-field characterization of guided polariton
propagation and cutoff in surface plasmon waveguides. Phys. Rev. B 74, 165415 (2006)
125. B. Lee, S. Kim, H. Kim, Y. Lim, Progr Quant. Electron. 34, 47 (2010)
126. Editorial, Nat. Photon. 6, 407 (2012)
127. M.A. Nielsen, I.L. Chuang, Quantum Computation and Quantum Information (Cambridge
Univ. Press, Cambridge, 2000)
128. N. Gisin, R. Thew, Nat. Photon. 1, 165 (2007)
129. D.K. Gramotnev, S.I. Bozhevolnyi, Nat. Photon. 4, 83 (2010)
130. J. Zhang, L. Cai, W. Bai, Y. Xu, G. Song, Opt. Lett. 36(12), 2312 (2011)
131. XR Jin, L Sun, X Yang, J Gao X, Opt. Lett. 38(20), 4078 (2013).
132. M.P. Nezhad, K. Tetz, Y. Fainman, Opt. Express 12, 4072 (2004)
133. S.A. Maier, Opt. Commun. 258, 295 (2006)
134. E.M. Purcell, Phys. Rev. B 69, 681 (1946)
135. D.F. Walls, G.J. Milburn, Quantum Optics (Springer, New York, NY, 2008)
136. M. Finazzi, F. Ciaccacci, Phys Rev. B 86, 035428 (2012)
137. A. Archamabault, F. Marquier, J.-J. Greffet, Phys. Rev. B 82, 035411 (2010)
138. A. Trugler, U. Hohenester, Phys. Rev. B 77, 115403 (2008)
139. C. Van Vlack, P.T. Kristensen, S. Hughes, Sathees. Phys Rev B 85, 075303 (2012)
140. T. Hummer, F.J. Garcı́a-Vidal, L. Martı́n-Moreno, D. Zueco, Phys. Rev. B 87, 115419 (2013)
141. GV Naik, J Kim, A. Boltasseva, Opt. Mater. Exp. 1(6), 1090 (2011).
142. D. Dzsotjan, J. Kaestel, M. Fleischhauer, Phys. Rev. B 84, 075419 (2011)
143. V.A.G. Rivera, Y. Ledemi, M. El-Amraoui, Y. Messaddeq, E. Marega Jr., Proc. SPIE 8632,
863225–863221 (2013)
144. V.A.G. Rivera, Y. Ledemi, Y. Messaddeq, E. Marega Jr., Proc. SPIE 8994, 899421–899421
(2014)
145. J.-W. Lo, W.-C. Lien, C.-A. Lin, J.-H. He, ACS Appl. Mater. Interfaces 3, 1009 (2011)
146. H. Mertens, A. Polman, Appl. Phys. Lett. 89, 211107 (2006)
147. A. Gopinath, S.V. Boriskina, S. Yerci, R. Li, L. Dal Negro, Apl. Phys. Lett. 96, 071113
(2010)
148. C.E. Hofmann, F.J. Garcı́a de Abajo, H.A. Atwater, Nano Lett. 11, 372 (2011)
149. S. Karaveli, R. Zia, Phys. Rev. Lett. 106, 193004 (2011)
150. W. Zhou, M. Dridi, J.Y. Suh, C.H. Kim, D.T. Co, M.R. Wasielewski, G.C. Schatz,
T.W. Odom, Nat. Nanotechnol. 8, 506 (2013)
151. W. Zhou, T.W. Odom, Nat. Nanotechnol. 6, 423 (2011)
152. T. Gruner, D.G. Welsch, Phys. Rev. A 53, 1818 (1996)
330 V.A.G. Rivera et al.

153. B.R. Judd, Phys. Rev. 127(3), 750 (1962)

154. V.A.G. Rivera, FA Ferri, LAO Nunes, AR Zanatta, E Marega Jr E (2012) Focusing surface
plasmons on Er3+ ions through gold planar plasmonic lenses. Appl. Phys. A 109(4):1037.
155. D.K. Gramotnev, J. Appl. Phys. 98, 104302 (2005)
156. F. De Angelis et al., Nano Lett. 8, 2321 (2008)
157. F. De Angelis et al., Nat. Nanotechnol. 5, 67 (2010)
158. S.I. Bozhevolnyi, V.S. Volkov, E. Devaux, J.-Y. Laluet, T.W. Ebbesen, Nature 440,
508 (2006)
159. M. Durach, A. Rusina, M.I. Stockman, K. Nelson, Nano Lett. 7, 3145 (2007)
160. N.A. Issa, R. Guckenberger, Plasmonics 2, 31 (2007)
161. D.K. Gramotnev, D.F.P. Pile, M.W. Vogel, X. Zhang, Phys. Rev. B 75, 035431 (2007)
162. A. Dereux et al., J. Microsc. 202(2), 320 (2001)
163. D. Mehtani et al., J Opt. A 8(4), 183 (2006)
164. V.A.G. Rivera, S.P.A. Osorio, Y. Ledemi, D. Manzani, Y. Messaddeq, L.A.O. Nunes,
E. Marega Jr., Opt. Express 18(24), 25321 (2010)
165. P. Anger, P. Bharadwaj, L. Novotny, Phys. Rev. Lett. 96, 113002 (2006)
166. S. Kuhn, U. Hakanson, L. Rogobete, V. Sandoghdar, Phys. Rev. Lett. 97, 017402 (2006)
167. O.L. Muskens et al., Nano Lett. 7, 2871 (2007)
168. A. Yariv, P. Yeh, Optical Waves in Crystals: Propagation and Control of Laser Radiation
(Wiley Classics Library, New York, NY, 2003)
169. J. Hao et al., Phys. Rev. Lett. 99, 063908 (2007)
170. R.H. Magruder, J.E. Wittig, J. Non Cryst. Solids 163, 162 (1993)
171. V.A.G. Rivera, Y. Ledemi, M.A. Pereira-da-Silva, Y. Messaddeq, E. Marega, Sci. Rep. 6,
18464 (2016)
172. A.G. Curto, Ph.D. thesis, Universitat Politècnica de Catalunya (2013).
173. J. D. Jackson (ed.), Classical electrodynamics (John & Wiley, New York, NY, 1999)
174. C. A. Balanis (ed.), Antenna Theory: Analysis and Design (John & Wiley, New York, NY,
2005)
175. A. Chakrabarty, F. Wang, F. Minkowski, K. Sun, Q.H. Wei, Opt. Express 20, 11615 (2012)
176. J.C. Gutierrez-Vega, R.M. Rodriguez-Dagnino, M.A. Meneses-Nava, S. Chavez-Cerda,
Am. J. Phys. 71, 233 (2003)
177. X. Gong, Y. Chen, Y. Lin, Q. Tan, Z. Luo, Y. Huang, J. Appl. Phys. 103 (2008)
178. E. Ozbay, Plasmonics: merging photonics and electronics at nanoscale dimensions. Science
311, 189 (2006)
179. T.W. Ebbesen, C. Genet, S.I. Bozhevolnyi, Surface-plasmon circuitry. Phys. Today 61,
44–50 (2008)
180. M.S. Tame, K.R. McEnery, S.K. Ozdemir, J. Lee, S.A. Maier, M.S. Kim, Quantum
plasmonics. Nat. Phys. 9, 329 (2013)
181. H.A. Atwater, The promise of plasmonics. Sci. Am. 296, 56–63 (2007)
Index

A Crystal field, 46–51, 157, 317

Absorption coefficient, 8, 34–37, 53, 101, 102, Crystalline phase, 60, 61, 63, 69, 183, 198
120, 185, 186, 188, 216, 223–225 Crystallization temperature, 35, 62, 64, 83, 280
Absorption edge, 3, 214, 226, 280 Crystallographic structure, 60
Acousto-optic modulation, 60
Amorphous system, 310
Angular momentum, 43, 171, 320 D
Anti-stokes photoluminescence, 7, 16–34 2D/3D active and passive waveguides, 241
Attenuation spectrum, 142, 143, 281 Decay time, 2, 151
Distillation-purification process, 74
Double-crucible technique, 133
B Drawbacks of extrusion technique, 140
Barady, G., 94
Bismuth oxide (Bi2O3), 6, 34, 35, 37, 121, 194,
201, 281 E
Bloch electromagnetic modes, 316 Effective phonon mode (EPM), 20
Broadband amplification, 147, 167, 169 Efficient nucleation, 20
Broadband emission in tellurite glasses, 7, 9, Electric dipole (ED), 46, 47, 50, 51, 170, 171,
147, 156–176, 178–199, 201, 202, 314 309, 314, 320
Electronic band gap, 309
Embedded nanoparticles, 244
C Emission cross section, 102, 105, 110, 115,
C and L bands, 51, 169 116, 147, 148, 162, 169, 172, 173, 175,
Cavity quantum electrodynamics (CQED), 317 176, 187, 202
Chalcogen element, 8, 60, 61, 66, 73, 216 Emission intensities, 165, 181–185, 187, 194,
Charge clouds, 48 312, 319
Chromatic aberration, 196, 264 Energy level, 9, 19, 22, 23, 42, 43, 46, 47, 50,
Chromatic dispersion, 2, 10, 122, 283, 285 51, 109, 157–166, 192, 194, 197, 201,
Coalescence, 250 314, 318, 322
Continuous wave (CW), 9, 16, 94, 103, 106, Energy transfer (ET), 5, 7, 10, 20–30, 41, 94,
111, 115, 118, 123, 127, 214, 216, 266, 101, 117, 118, 149, 164, 165, 173, 179,
293, 294 180, 183, 184, 187, 192, 193, 250, 273,
Conventional melt quenching technique, 5, 16 301, 310, 313–315, 323
Coulomb interaction, 43, 44 Erbium-doped fiber amplifier (EDFA), 2,
Cross-relaxation process, 103, 115, 117 51, 158

© Springer International Publishing AG 2017 331

V.A.G. Rivera, D. Manzani (eds.), Technological Advances in Tellurite Glasses,
Springer Series in Materials Science 254, DOI 10.1007/978-3-319-53038-3
332 Index

European Space Agency (ESA), 66, 70 Glass formers and modifiers, 181–182
Excited state absorption (ESA), 26, 53, 54, 102, Glass quantum communication devices, 93
113, 151, 158 Glass transparency, 280
Extrusion, 132, 140, 217, 219–222, 233, 282 Green function, 310
Group theory, 49

F
Femtosecond laser, 10, 159, 228, 260–266, H
269, 271, 272 Heavy-metal oxide (HMO) glasses, 15, 20,
Fiber amplifier, 9, 51, 93, 131, 162, 165, 181, 201
167–170, 193–194, 202, 242 High nonlinear, 7, 15, 94, 266, 283
Fiber drawing, 7, 66, 73, 77, 134, 170, 216, Hooke’s law, 214
219, 222, 225, 229, 233, 280 Hot spots, 312
Fiber evanescent wave spectroscopy (FEWS), Hyperpolarizabilities, 35
66, 74, 76–79
Fiber geometries, 277, 279, 284
Fiber laser, 5, 132, 165, 175, 185, 193, 196, I
213, 214, 216, 231, 232, 244, 285, 286, Induced Grading Autocorrelation (IGA), 94
289, 295 Infrared supercontinuum generation, 229
Fiber on glass waveguide, 8 Infrared-to-visible upconversion, 26
Filamentation effects, 266 Infrared transmission window, 101
Fluoride fibers, 213, 217 Integrated optics, 126, 184, 241, 245, 260, 323
Fluoride glass, 61, 108, 126, 147, 149, 181, Interatomic distances, 60
185, 201, 202, 213, 214, 217, 224, Ion exchange technique, 96, 259
232, 309 Ion–ion interaction, 53, 54, 314
Fluoroindate glass, 20 Ions doping, 15
4f manifold, 42 I–V characteristics, 244, 251–253
4f-4f transitions, 22, 28
Fourier transform infrared spectroscopy
(FTIR), 5, 76 J
Four-level laser, 102, 163 Joule effect, 312
Four-wave-mixing (FWM), 216 Judd-Ofelt parameters, 9, 149
Free electron plasma, 262, 265 Judd-Ofelt theory, 51
Free ion, 6, 42–47, 157
Free OH groups, 149, 184
Frequency upconversion, 7, 18–30, 36 K
Fresnel reflection, 101, 102, 107, 108 Kerr gate technique, 17
Full width half maximum (FWHM), 2, 94, Kerr lens, 105
111–113, 118, 165, 167, 168, 175, Kramers theorem, 49
178–184, 186, 194, 202, 228, 271,
272, 292
Fundamental mode, 233, 234 L
Large-mode-area (LMA), 216, 232–235, 287
Laser generation, 9, 132, 143–147
G Laser microcavities, 242
Gain and laser generation, 9, 132, 143, 145, Laser sources based on tellurium oxide, 8
147 Laser writing, 7, 10, 96, 97, 260–274
Gain saturation profile, 269 Legendre polynomials, 43
Gaussian laser beam, 264 Lifetime, 9, 104, 117, 132, 149–151, 157, 160,
Germanate glass, 28, 30, 34, 116, 163, 180 165, 169, 170, 172, 173, 175, 176,
Glass-ceramics (GCs), 6, 9, 15, 162, 164, 179–185, 187, 196, 197, 202, 310, 311,
182–183, 190, 196, 198 313, 315
Glass durability, 281 Lightning rod, 312, 317
Index 333

Linear effects, 263 Nonlinear refractive index, 94, 132, 214, 215,
Linear refractive index, 16, 35, 214, 215, 225, 228, 264, 290
265, 290 Nonlinear Schr€odinger equation, 290
Ln-doped fiber lasers, 193 Non-oxide glasses, 69, 216
Local density of optical states, 315 Nonradiative decay processes, 149
Local electric field, 184, 311, 312, 314, 321 Nonradiative relaxation, 21, 23
Localized surface plasmon resonance (LSPR), Numerical aperture (NA), 107, 134, 151, 218,
10, 20, 302–308, 323 264, 265, 284, 286
Localized temperature sensing, 98–99
Low dispersion, 214, 229, 295
Luminescence spectrum, 248 O
OH absorption bands, 101, 281
Optical amplifiers, 6, 7, 9, 52, 93, 94, 132–151,
M 156, 157, 160, 162, 165–169, 173, 175,
Mach-Zehnder interferometer (MZI), 9, 242, 182, 187, 193, 195, 241–244, 247–250,
245–246, 255 255, 269
Magnetic dipole, 50, 51, 171 Optical attenuation, 9, 132, 134, 142
Master oscillator power amplifier (MOPA), 232 Optical communication technology, 51,
Materials triangle, 1 53, 167
Medium-frequency phonons, 93 Optical lithography, 9, 243
Melt-quenching method, 16, 20, 133 Optical parametric amplifiers (OPA), 285, 286
Memory device, 7, 54, 83, 84, 241–256 Optical parametric oscillator (OPO), 17, 214,
Metal–insulator–semiconductor (MIS), 243, 228, 285–287
244, 255 Opto-electrophoretic sensing, 8, 67, 78–80
Metastable liquids, 308 Opto-geometric properties, 284
Microdisks, 242 Output power, 9, 103, 106, 108, 109, 111–115,
Microexplosions, 262, 266 118, 120, 127, 196, 227, 228, 231, 234,
Micro-structured fiber, 139, 140, 285 236, 284, 296
Micro-structured optical fiber (MOF), 35, 93, Oxyfluorotellurite glass, 182, 183, 187, 197
132, 138, 140–147, 214, 226, 289–295 Oxyhalide tellurite glasses, 4, 191
Mid infrared (MIR), 5, 7, 9, 10, 63, 66, 67, 101,
126, 156, 236, 259, 260, 273, 274, 288
Mode-locking, 103–105 P
Mode-profile image, 97 Passive devices, 241
Multiphonon absorption, 126, 214, 216, 224, Pedestal waveguide, 9, 243–245, 247–250, 255
226, 280, 288 Phonon absorption, 66
Multiphonon relaxation (MPR), 115, 173, 181 Phonon-assisted upconversion, 7, 18–20
Multiple ring hole (MRH) fibers, 282, 284 Phonon energy, 7, 15, 41, 52, 77, 81, 101, 105,
Multiplet splitting, 50 116, 117, 122, 126, 160, 161, 166, 170,
181, 183, 185, 187, 197–199, 201, 202,
213, 242, 245, 260, 273, 280, 309
N Phospho-tellurite glass, 170, 195, 268–272
Nanoparticle, 6–8, 10, 16, 20–30, 53, 54, 73, Photo induced processes, 6
94, 184, 242, 243, 250–255, 302–307, Photoluminescence quantum yield (PLQY),
312, 314 157, 160, 169, 175, 176, 198, 199, 201
Nanostructure plasmonics, 301, 311 Photonic applications in tellurite glasses, 7, 8,
National Aeronautics and Space 93, 94, 96–99
Administration (NASA), 70, 74 Photonic circuits, 96, 260, 323
Nonlinear absorption, 10, 260–261 Photonic crystal fiber (PCF), 138, 139, 214,
Nonlinear applications, 7, 9, 52, 132, 142, 167, 216, 217, 219, 226
236, 277 Photonic nanowires, 95
Nonlinear laser sources, 214 Photonics science, 317
Nonlinear photoionization, 260–262 Planar waveguides, 96
334 Index

Plasmonic materials, 302, 305 Supercontinuum, 9, 10, 93, 122, 127, 134, 166,
Plasmonics, 7, 10, 94, 301–324 187, 214, 216, 226–229, 231, 232,
Polarizability, 2, 34, 35, 53, 94 234–236, 277, 284, 285, 287, 288,
Propagation losses, 9, 184, 242, 244–246, 248, 295, 296
266, 268, 269, 317, 323 Supercontinuum (SC) generation, 7–10, 94,
Pumping conditions, 285–287, 289, 102, 122, 195–196, 226, 228–230,
292, 293 232–235, 277–296
Surface-enhanced hyper-Raman scattering,
303
Q Surface-enhanced infrared spectroscopy, 303
Q-switched, 9, 17, 103, 104, 118, 127, 196 Surface plasmon resonance (SPR), 10, 20, 94,
Quality factor (Q), 5, 103, 115, 173, 174, 184, 302–308, 315
234, 317 Symmetrical signals, 36
Quantum cutting, 6, 192–193 Synthesis of telluride glasses, 8, 69–70
Quantum–photon interaction, 94 Synthesis process, 59, 60, 62, 70

R T
Raman amplifiers, 9, 11, 121, 167, 194–195 Tapered fiber, 10, 279, 284
Raman fiber lasers, 232 Telecommunications band, 9, 108
Raman spectroscopy, 271 Telluride glasses, 8, 59–86
Rare-earth-doped tellurite glasses, 9, 105–121, Tellurite fibers, 9, 94, 132–134, 142, 147,
124, 126 149–151, 167, 170, 181, 196, 202, 214,
Rare-earths, 54, 93, 132, 149 216, 228, 229, 231–233, 235, 274,
Reabsorption process, 145, 179 277–296
RF sputtering, 7, 9, 241–256 Tellurite glass fiber, 2, 6–9, 122, 168, 201,
Rod-in-tube technique, 7, 72, 74, 131–134, 282 213–236
Rotational casting method, 72, 217 Tellurite microsphere laser, 107
Runge-Kutta method, 292 Tellurite rods, 9, 133, 136–137
Rusell-Saunders coupling, 43 Temperature-dependent absorption cross
sections, 19
Thermal properties, 5, 52, 62, 63, 116, 166,
S 222, 223, 308–311
Second harmonic generation, 53, 303 Thermoelectricity, 8, 59, 81–82
Selection rules, 8, 42, 46, 47, 49–51, 171 Thermoelectric material, 2, 67, 81, 82
Self-focusing, 105, 263, 265 Third order nonlinear, 7, 34–36, 53, 264
Semiconducting behavior, 67 Three-level laser, 102
Shockley–Queisser theory, 188 Thulium-doped fiber amplifiers (TDFA),
Single-mode optical fibers, 234 51, 169
Site symmetry, 49 Total angular momentum, 171, 320
Slope efficiency, 102, 103, 105–107, 109, 111, Trigonal bipyramid, 181, 271, 308
113, 115, 117–120 Trigonal pyramids, 7, 181, 271
Smart materials, 1–2 Trivalent lanthanides, 7, 8, 41–45, 47–54
Solar devices, 9 Tunable laser, 76, 109, 122, 160, 164, 165, 168,
Solid state laser, 6, 11, 53, 104, 126, 159, 172, 182, 187, 193, 213, 225, 228, 274,
214, 313 288, 296
Spherical aberration, 263, 264
Spin–orbit coupling, 43, 45
Sputtering deposition, 243 U
Stable telluride glasses, 60, 94 Ultrashort pulse laser, 105, 111, 286, 294
Stark splitting, 105 Upconversion emission, 98, 121, 197
Stimulated emission cross section, 102, 105, Upconversion spectroscopy, 181
162, 169, 172, 173, 175, 202 Urbach’s tail, 53
Index 335

V Wigner-Eckart theorem, 44
Van der Waals bonds, 60

Z
W Zero dispersion wavelength (ZDW), 94, 122,
Wave functions, 42, 46, 47, 50 160, 187, 214, 216, 228, 234, 277, 283,
Waveguide, 7, 15, 52, 71, 96, 109, 159, 216, 285, 286, 289
241, 259, 277, 316
Wavelength division multiplexing (WDM), 6,
143, 169, 194, 195, 244