See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/23660335

High-power, continuous-wave, second-harmonic generation at 532 nm in

periodically poled KTiOPO4

Article in Optics Letters · December 2008

DOI: 10.1364/OL.33.002955 · Source: PubMed

CITATIONS READS
55 474

7 authors, including:

Goutam K Samanta Suddapalli Chaitanya Kumar

Physical Research Laboratory ICFO Institute of Photonic Sciences
176 PUBLICATIONS 1,864 CITATIONS 193 PUBLICATIONS 2,298 CITATIONS

SEE PROFILE SEE PROFILE

Manoj Mathew Carlota Canalias

Carl Zeiss AG KTH Royal Institute of Technology
31 PUBLICATIONS 486 CITATIONS 182 PUBLICATIONS 2,163 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Laser-patterned magnetic media for data recording and storage View project

Quantum entanglement View project

All content following this page was uploaded by Suddapalli Chaitanya Kumar on 23 May 2014.

The user has requested enhancement of the downloaded file.

 December 15, 2008 / Vol. 33, No. 24 / OPTICS LETTERS 2955

High-power, continuous-wave, second-harmonic

generation at 532 nm in periodically
poled KTiOPO4
G. K. Samanta,1,* S. Chaitanya Kumar,1 M. Mathew,1 C. Canalias,2 V. Pasiskevicius,2
F. Laurell,2 and M. Ebrahim-Zadeh1,3
1
ICFO–Institut de Ciencies Fotoniques, Mediterranean Technology Park, 08860 Castelldefels, Barcelona, Spain
2
Royal Institute of Technology, Roslagstullsbacken 21 10691 Stockholm, Sweden
3
Institucio Catalana de Recerca i Estudis Avancats, Passeig Lluis ompanys 23, Barcelona 08010, Spain
*Corresponding author: [email protected]
Received October 7, 2008; accepted October 15, 2008;
posted November 3, 2008 (Doc. ID 102494); published December 5, 2008
We report efficient generation of high-power, cw, single-frequency radiation in the green in a simple, compact
configuration based on single-pass, second-harmonic generation of a cw ytterbium fiber laser at 1064 nm in
periodically poled KTiOPO4. Using a crystal containing a 17 mm single grating with period of 9.01 ␮m, we
generate 6.2 W of cw radiation at 532 nm for a fundamental power of 29.75 W at a single-pass conversion
efficiency of 20.8%. Over the entire range of pump powers, the generated green output is single frequency
with a linewidth of 8.5 MHz and has a TEM00 spatial profile with M2 ⬍ 1.34. The demonstrated green power
can be further improved by proper thermal management of crystal heating effects at higher pump powers
and also by optimized design of the grating period to include thermal issues. © 2008 Optical Society of
America
OCIS codes: 140.7300, 140.3510, 190.2620, 190.4400.

Quasi-phase-matched (QPM) nonlinear materials creased powers has not been reported, to our knowl-
have had a major impact on optical frequency- edge. Here, we demonstrate the generation of 6.2 W
conversion technology. In particular, the advent of pe- of cw, single-frequency radiation at 532 nm with a
riodically poled LiNbO3 (PPLN) has enabled the de- single-pass conversion efficiency as high as 20.8%.
velopment of a new generation of devices with We also investigate thermal issues relating to power
unprecedented performance capabilities. At the same scaling of SHG to fundamental powers up to 30 W
time, the onset of photorefractive damage under ex- and discuss strategies for increased output power
posure to visible radiation has confined the utility of and efficiency at higher pump powers.
PPLN mainly to the IR. Doping with MgO can par- A schematic of the experimental setup is shown in
tially alleviate this problem, permitting the use of Fig. 1. The fundamental pump source is a cw Yb fiber
MgO:PPLN at higher powers and lower tempera- laser (IPG Photonics, YLR-30-1064-LP-SF). The laser
tures [1,2]. However, residual effects of photorefrac- delivers up to 30 W of single-frequency radiation at
tive damage continue to restrict operation of devices 1064 nm in a linearly polarized beam with a diam-
based on MgO:PPLN mainly to the near- and mid-IR. eter of 3 mm, M2 factor ⬍1.01, and a measured in-
For frequency conversion in the visible, QPM ma- stantaneous linewidth of 12.5 MHz. To maintain
terials such as MgO-doped periodically poled stoichi- stable output characteristics, we operated the pump
ometric LiTaO3 (MgO:sPPLT) [2,3] and periodically laser at the maximum power and used an attenuator
poled KTiOPO4 (PPKTP) [2,4–7] are attractive alter- comprising a half-wave plate (HWP) and a polarizing
natives, offering increased resistance to photorefrac- beam splitter to vary the input fundamental power. A
tive damage and relatively high effective nonlineari- second HWP was used to yield the correct pump po-
ties 共deff ⬃ 10 pm/ V兲. Advances in material growth larization for phase matching. The fundamental
and fabrication have also led to the availability of beam was focused to a measured beam waist radius
such QPM materials with high optical quality, of 37 ␮m (1 / e2-intensity) at the center of the crystal
shorter grating periods, and long interaction lengths 共␰ = 1.28兲. The bulk PPKTP crystal is fabricated in-
(up to 30 mm), enabling the development of house from a c-cut 19-mm-long, 1-mm-thick, and
frequency-conversion devices for the visible in the cw 6-mm-wide flux-grown KTP sample. The c− face of
[2,4–6] and low-intensity pulsed regimes [7]. the crystal was patterned with a photoresist grating
In this Letter, we report efficient generation of
high-power cw radiation at 532 nm in a compact and
practical design based on simple single-pass, second-
harmonic-generation (SHG) of a cw ytterbium (Yb) fi-
ber laser in PPKTP. Earlier reports of cw SHG in the
visible based on PPKTP include intracavity [4], exter-
nal enhancement [5], and single-pass configurations Fig. 1. Schematic of the experimental design for single-
[6], but the highest cw SH power achieved so far has pass SHG of cw Yb fiber laser in PPKTP. ␭ / 2, half-wave
been 2.3 W in multimode output [4]. As such, power plate; PBS, polarizing beam splitter; L, lens 共f = 20 cm兲; M,
scaling of SHG and performance of the material at in- dichroic mirror.

0146-9592/08/242955-3/$15.00 © 2008 Optical Society of America

2956 OPTICS LETTERS / Vol. 33, No. 24 / December 15, 2008

of 9.01 ␮m period and was covered with an Al film

over 17 mm length at the center, keeping a 1 mm free
zone on each side of the grating. The actual length of
the grating is thus 17 mm. The crystal was poled by
applying 5-ms-long electric field pulses of 2.5 kV/ mm
in magnitude [8]. The PPKTP is housed in an oven
with a temperature stability of ±0.1° C. The crystal
faces are antireflection coated 共R ⬍ 1 % 兲 at 1064 nm
and 532 nm. A dichroic mirror, M, coated for high re-
flectivity 共R ⬎ 99% 兲 at 1064 nm and high transmis-
sion 共T ⬎ 99% 兲 at 532 nm, extracted the generated
Fig. 3. Dependence of the measured SH power and the
green output from the fundamental. corresponding conversion efficiency on the incident funda-
To characterize the PPKTP sample, we first deter- mental power.
mined the dependence of SHG power on crystal tem-
perature. To avoid unwanted contributions from ther- damental and generated green, and linear absorption
mal effects, we performed the measurements at a low at both wavelengths.
fundamental power of 1 W, resulting in 8.3 mW of To verify any contribution from GIIRA to thermal
green output at a phase-matching temperature of dephasing, we focused 6 W of green radiation from a
30.8° C. The result is shown in Fig. 2. The solid curve cw laser at 532 nm (Coherent, Verdi-10) to a beam ra-
is a sinc2 fit to the data, confirming the expected tem- dius of 34 ␮m inside the crystal to obtain similar
perature dependence of SHG. The FWHM of the maximum green power density 共0.33 MW/ cm2兲 as in
curve is ⌬T = 3 ° C, which is slightly wider than the the SHG experiments. The arrangement resulted in
calculated value of 2.62° C from the Sellemier equa- the simultaneous focusing of the fundamental fiber
tions [9]. The effective interaction length can be de- laser IR beam to a waist radius of 28 ␮m inside the
fined as Leff = L ⫻ 2.62/ 3 = 14.84 mm, where L crystal, close to 37 ␮m used in the SHG experiments.
= 17 mm is the physical length of the grating. The We then measured the variation in transmitted IR
discrepancy between the physical length and mea- power at different green power levels over 10 min pe-
sured effective length is attributed to imperfections riods. However, we observed no difference in the
in the grating structure and nonideal focusing. transmitted IR power with and without green radia-
We performed measurements of SHG efficiency tion, indicating the absence of GIIRA up to the maxi-
and power up to the maximum fiber laser power of mum green power of 6 W. We also investigated the
30 W, with the results shown in Fig. 3. The funda- possible role of TPA by recording the crystal trans-
mental power was measured at the input to the crys- mission at both fundamental and green wavelengths
tal, while the generated green power was corrected at power levels up to 30 W and 6 W, respectively. We
for the 6% transmission loss of mirror M. We ob- obtained a linear increase in transmitted power with
tained 6.2 W of green power at the full input power of input power at both wavelengths, from which linear
29.75 W, representing a single-pass conversion effi- absorption losses of 0.3% / cm at 1064 nm and
ciency of 20.8%. As evident in Fig. 3, the quadratic in- 4.5% / cm at 532 nm were deduced, in agreement
crease in SH power and corresponding linear varia- with previous results [10]. These measurements thus
tion in efficiency are maintained up to 15 W of input confirmed that thermal dephasing in the PPKTP
power, after which saturation sets in. Beyond this crystal was also not due to TPA, but a result of intrin-
value, SH power increases linearly with fundamental sic linear absorption of IR and green, leading to in-
power. The saturation effect is attributed to depletion creased heating at higher powers. We determined a
of the fundamental [2] and thermal phase mismatch normalized conversion efficiency of ⬃1.2% / W below
in the PPKTP crystal, owing to a number of factors. 15 W of fundamental power, monotonically decreas-
These include green-induced IR absorption (GIIRA) ing to ⬃0.7% / W with the rise in input power to
of fundamental, two-photon absorption (TPA) of fun- 29.75 W, confirming increased thermal dephasing
due to linear absorption at higher powers. The nor-
malized efficiency values do not account for absorp-
tion losses in the crystal. From the normalized effi-
ciency of 1.2% / W and the focusing parameter of ␰
= 1.28 共h ⬃ 0.85兲, the effective nonlinearity is calcu-
lated as deff ⬃ 7.63 pm/ V for the 17 mm grating, cor-
responding to d33 ⬃ 12 pm/ V. This value is 77.9% of
previous results [11], which we attribute to crystal
absorption of the green and IR, laser linewidth, and
imperfections in the grating structure such as miss-
ing periods and duty factor variations. Considering
the absorption losses of 0.3% / cm at 1064 nm and
Fig. 2. Temperature dependence of SH power (filled 4.5% / cm at 532 nm, we determined a normalized ef-
circles) and the sinc2 fit (solid curve). The resulting tem- ficiency of ⬃1.33% / W below 15 W of fundamental
perature bandwidth is 3 ° C. power, resulting in deff ⬃ 8.6 pm/ V, corresponding to
 December 15, 2008 / Vol. 33, No. 24 / OPTICS LETTERS 2957

d33 ⬃ 13.5 pm/ V for a 14.8 mm PPKTP crystal with

50% duty factor. As the missing periods are included
in the effective length, the deviation in the calculated
d33 from earlier value [11] can be due to the grating
duty factor and the laser linewidth. However, using
the Sellmeier equations [9], the calculated FWHM
spectral acceptance of the 17 mm crystal 共0.157 nm兲
is much wider than the laser linewidth 共12.5 MHz兲,
confirming the discrepancy due to duty factor of the
grating.
To gain further insight into thermal effects in the
PPKTP crystal, we recorded the SHG phase- Fig. 5. (Color online) TEM00 energy distribution, beam
matching temperature at different fundamental pow- profiles, and Gaussian fits (solid curves) of the generated
ers. Although no significant shift in the phase- green beam in the far-field.
matching temperature was observed at the lower The far-field energy distribution of the green out-
input powers, a monotonous decrease in phase- put beam at 5 W together with the intensity profile
matching temperature, at a rate of 0.19° C / W, was and the Gaussian fits, recorded at 25 W of input
measured above 15 W, further confirming crystal power, are shown in Fig. 5. Using a 25 cm focal
heating at higher input powers. At lower powers, lin- length lens and scanning beam profiler, we measured
ear absorption does not significantly affect the phase- M2 values of the beam as M2x ⬃ 1.29 and M2y ⬃ 1.31,
matching temperature, whereas at higher powers it
confirming TEM00 spatial mode. Similar M2 values
heats the crystal, reducing the optimum value for
were measured at different input power levels, show-
phase matching and thus requiring a reduction in
crystal temperature for maximum green generation. ing a small variation in M2x from 1.25 to 1.29 and M2y
Since phase matching for the present PPKTP sample from 1.2 to 1.34.
is near room temperature, further optimization of the We have thus generated 6.2 W of cw, single-
crystal temperature using Peltier cooling or the use frequency radiation at 532 nm at 20.8% efficiency in
of a grating period with a higher phase-matching a highly compact and practical design using simple
temperature could potentially lead to increased green single-pass SHG of a cw Yb fiber laser in PPKTP near
power and efficiency at the higher input powers up to room temperature. With proper thermal manage-
ment and optimized grating design, further improve-
30 W and above, with the possibility of power scaling
ments in green power and efficiency as well as power
of green beyond 6 W. Proper management of the gen-
scaling are feasible.
erated heat or inclusion of the crystal heating effects
in the grating design are other possible strategies to This work was supported by the Ministry of Edu-
achieve the highest SH power and efficiency. cation and Science, Spain, under grant TEC2006-
We analyzed the output spectrum at 532 nm using 12360.
a scanning confocal interferometer (free spectral References
range, 1 GHz; finesse, 400). A typical fringe pattern
1. H. Furuya, A. Morikawa, K. Mizuuchi, and K.
at 25 W of fundamental power (5 W of green), is Yamamoto, Jpn. J. Appl. Phys. 45, 6704 (2006).
shown in Fig. 4, verifying single-frequency operation. 2. F. J. Kontur, I. Dajani, Y. Lu, and R. J. Knize, Opt.
The instantaneous linewidth, measured directly Express 15, 12882 (2007).
without accounting for the interferometer resolution, 3. S. V. Tovstonog, S. Kurimura, I. Suzuki, K. Takeno, S.
is 8.5 MHz. The same behavior was observed Moriwaki, N. Ohmae, N. Mio, and T. Katagai, Opt.
throughout the pumping range with similar line- Express 16, 11294 (2008).
width, confirming robust single-mode operation at all 4. S. Greenstein and M. Rosenbluh, Opt. Commun. 238,
pump powers. 319 (2004).
5. R. Le Targat, J. J. Zondy, and P. Lemonde, Opt.
Commun. 247, 471 (2005).
6. Z. Sun, G. K. Samanta, G. R. Fayaz, M. Ebrahim-
Zadeh, C. Canalias, V. Pasiskevicius, and F. Laurell, in
Proceedings of the Conference on Lasers and Electro-
Optics (Optical Society of America, 2007), paper
CTuK1.
7. G. R. Fayaz, M. Ghotbi, and M. Ebrahim-Zadeh, Appl.
Phys. Lett. 86, 061110 (2005).
8. C. Canalias, S. Wang, V. Pasiskevicius, and F. Laurell,
Appl. Phys. Lett. 88, 032905 (2006).
9. K. Kato and E. Takaoka, Appl. Opt. 41, 5040 (2002).
10. G. Hansson, H. Karlsson, S. Wang, and F. Laurell,
Appl. Opt. 39, 5058 (2000).
Fig. 4. Single-frequency spectrum of the green output re- 11. M. V. Pack, D. J. Armstrong, and A. V. Smith, Appl.
corded by a scanning Fabry–Perot interferometer. Opt. 43, 3319 (2004).

View publication stats