CHAPTER 2

Optical Biosensing
David R. Walt

INTRODUCTION
Optical sensing offers a number of advantages relative to other
transduction mechanisms. Optical methods are sensitive. For example,
fluorescence is an intrinsically amplified method in that one fluorescent
molecule can generate up to a million emitted photons. In addition,
fluorescence is a black background technique in that the emission signal is
at a wavelength separated from the excitation wavelength, thereby
improving the detection sensitivity because one can measure a signal from
a low background rather than try to detect a small signal difference from a
high background (e.g., change in resistance). Attesting to these advantages,
most of the single molecule detection research employs fluorescence,
primarily due to its sensitivity. While methods such as fluorescence and
absorbance have a long history, newer methods such as surface enhanced
Raman spectroscopy (SERS) and surface plasmon resonance (SPR) have
developed rapidly over the last two decades and are playing an increasing
role in optical biosensing.
Optical methods are readily multiplexed—one can interrogate with
many wavelengths simultaneously without the signals interfering with one
another. Another advantage of optical methods is the ability to employ free
path or remote interrogation without the need for wire connections.
Finally, optical methods benefit from a developing infrastructure. Light
sources, detectors, optical interconnects, and other optical technologies are
being developed for the entertainment and telecommunications industries.
The age of photonics is approaching, and optical methods will likely
displace many of the electronic systems.
32 2. Optical Biosensing

During the WTEC investigation of international research and

development in biosensing, panelists identified a number of technology
themes in the optical biosensing area:
• Surfaces
• Arrays
• Inexpensive sensors
• Distributed/networked systems
• Nanomaterials
• Molecular biology
Tremendous advances have been made over the last two decades in
designing and preparing functional surfaces that can serve as attachment
substrates for biosensing materials (Crooks 2003). In addition, surfaces, in
conjunction with these new surface-binding methods, are being
implemented in a variety of optical methods. The multiple methods
employed for optical sensing correspond to the various optical transduction
mechanisms:

• Luminescence. This category of methods encompasses fluorescence

as well as chemiluminescence and bioluminescence. Luminescence
methods are highly sensitive and probably the most prominent optical
method employed today.
− Fluorescence: intensity, lifetime, polarization. Fluorescence is the
most commonly used optical technique. It involves the excitation
of a fluorescent molecule at one wavelength followed by emission
at a longer wavelength. The typical time between excitation and
emission is the lifetime and is typically in the nanosecond
timeframe. Fluorescence can be measured by its intensity, the
lifetime (duration of the excited state), or its polarization (related
to how rapidly the molecule is rotating during its excited state).
− Phosphorescence. Phosphorescence is a phenomenon that results
in a longer lived excited state leading to longer lived emission,
typically in the micro- to millisecond timeframe.
− Fluorescence resonance energy transfer (FRET). FRET is a
process whereby a donor molecule transfers its energy to an
acceptor molecule. The process depends on distance and is a
sensitive method for measuring interactions between two
molecules.
• Chemiluminescence and bioluminescence. These processes arise
when either chemical or biochemical energy is released in the form of
 David R. Walt 33

light. For example, fireflies and jellyfish glow from bioluminescent

bacteria present in these organisms. These chemistries and
biochemistries can be harnessed to prepare sensing materials.
• Absorbance. Absorbance is the simple phenomenon of a substance
absorbing light at specific wavelengths and is proportional to the
amount of absorbing material present. Absorbance can be used to
measure the amount of a substance present; alternatively, indicators
that bind to a substance and change their absorbance upon binding
can be used to indirectly measure the substance of interest.
• Scattering. Scattering is similar to absorbance but measures the
amount of light reflected. In a true scattering method, the scattering
depends on particle size.
• Surface methods. Methods based on an optical phenomenon occurring
at a surface include the following:
− Surface plasmon resonance (SPR). SPR phenomena are those in
which binding to a metal surface causes an optical change (due to
refractive index change) at a metal-substrate (usually glass)
interface.
− Surface-enhanced Raman scattering (SERS). In SERS methods,
an enhanced Raman effect occurs at certain metal surfaces.
− Interference. Interference methods are those in which two optical
signals are recombined to give an interference pattern due to a
delay in one signal relative to the other caused by binding of an
analyte.

SURFACE-BASED OPTICAL BIOSENSING

While much research and development related to fundamental surface
chemistries is ongoing in both the United States and Europe, most of the
present successes in surface-based optical biosensing are centered in
Europe. The most prominent method is one commercialized by Biacore
(Sweden, www.biacore.com) in 1990 and since refined that employs
surface plasmon resonance (SPR). The Biacore instrument is a highly
successful research apparatus that is able to measure both association and
dissociation rate constants and therefore can determine binding constants.
The success of the instrument is due to its full integration. The
manufacturer has addressed the chemistry, fluidics, optical detection, and
data processing and integrated them into a complete system. A variety of
chemistries are available for immobilizing virtually any molecular entity to
the sensing surface, making it a simple “plug and play” instrument for the
end user. The Biacore SPR instrument is the industry standard and is used
34 2. Optical Biosensing

widely in both academic research laboratories and for performing binding

studies in pharmaceutical laboratories.
Other surface-based methods include the promising work of Professors
Günter Gauglitz (Institute for Physical Theoretical Chemistry, Eberhard
Karls University, Tübingen, see site report Appendix B), and Michael
Sailor (University of California, San Diego), both of whom are developing
interferometric optical sensors in which surface binding shifts the
absorption maximum.

BIOSENSING ARRAYS
Scientific advances in the last ten years have made it possible to display
many different binding materials onto a single substrate and to
simultaneously assay for binding to these materials. These abilities have
revolutionized the fields of sensing/biosensing in particular and analysis in
general. Optical biosensor array types include planar waveguide arrays,
CMOS arrays, fiberoptic bundles, SPR arrays, and interferometry arrays.
Such array types provide a comprehensive or “global” picture of the
components in a complex mixture and enable subtle changes in
composition to be monitored even in the presence of a constant
background. Besides the ability to perform multianalyte sensing, other
advantages of optical arrays include on-chip positive and negative controls,
smaller size, lower cost, and higher speed.
Driving the development of optical biosensing arrays are the fields of
genomics, integrated optics, microfluidics, and bioinformatics. Presently,
the major research focus on arrays is in the area of proteomics. Optical
methods, such as fluorescence, for observing binding to such arrays are the
favored approach. The ability to capture global protein expression data by
employing arrays will revolutionize our understanding of living systems.
As the protein composition of cells changes rapidly, the ability to perform
dynamic measurements using multiple arrays will be crucial; it is therefore
important to address and solve the challenges associated with nonspecific
binding to protein arrays, preparing arrays with a high degree of
reproducibility, and attaching active materials to the array. While these
challenges are all areas of active investigation, the problems are manifold.
By solving these problems, however, there should be significant flow-
through discoveries made that will be applicable to many other fields.
Commercially, the array field has been dominated by DNA arrays, with
fluorescence detection dominating as the detection method. There is still a
tremendous research effort concentrated on DNA arrays, and virtually any
new transduction mechanism or biosensing system is applied to DNA
detection. The WTEC panel’s assessment is that while there is a major
emphasis on DNA detection, it is a mature technology. Although work is
 David R. Walt 35

this area is fashionable, any innovation is incremental, and additional

developments will have low impact due to the established base of DNA
array technology and users’ desire to employ standard methods.
Other work on optical biosensing arrays is focusing on simplifying the
supporting instrumental systems that enable optical sensors to be
interrogated and read, and on wider applications, including detection of
hitherto unknown hazards (e.g., toxins and biological agents), display of
biological information, and monitoring of environmental changes.

INEXPENSIVE AND DISTRIBUTED SENSORS

A distinguishing feature of many European biosensor efforts is a focus
on developing extremely inexpensive sensors for everyday applications.
These sensors are primarily directed toward food and environmental
applications and are intended to be widely incorporated in consumer
products. For example, at Cambridge University in the UK (see site report
in Appendix B), the laboratory of Professor Chris Lowe is developing
holographic sensors that can measure a variety of parameters in food or
can be emblazoned into consumer packaging. A visible hologram image
functions as both the analyte-specific responsive media and the optical
detection mechanism; further, it serves as the test result and therefore
requires no additional electronic processing. Holograms can have
presence/absence readout or can be designed with a built-in dial in which
the dial moves as the concentration of the analyte increases. The
holograms can even be written in the product material (e.g., food),
providing a zero materials cost. Sample holographic sensor test results
from Prof. Lowe’s lab are shown in Figure 2.1 below.

Fig. 2.1. Examples of holographic biosensing before and after a

test. (University of Cambridge Institute of Biotechnology,
www.biot.cam.ac.uk/ ~crl/crl6.html)
Another noteworthy example of development of inexpensive sensors for
everyday applications is at Ireland’s National Centre for Sensor Research
at Dublin City University (site report in Appendix B; www.dcu.ie/~ncsr/
index_home.html), where researchers are printing CO2 optical sensing
films directly onto food packaging material (see Figure 2.2). Perishable
foods, such as meats, are packaged under a CO2 atmosphere. If there is a
breach in the packaging material, the CO2 sensing film changes color and
signals to the consumer that the food is not fresh.
36 2. Optical Biosensing

Fig. 2.2. Inexpensive optical sensor for testing integrity of

meat packaging. (Dublin City University,
www.dcu.ie/~ncsr/commercial/technologies.html)
Both of these techniques employ optical methods for readout, with the
human eye as the detector. Such sensing systems should become
increasingly popular and accepted as consumers become familiar with the
capabilities of the technology and begin to demand this kind of quality
assurance in their food as well as in some household products.
A related area of significant R&D effort, also centered in Europe, is that
devoted to distributed sensors. These sensors are generally coupled to
communications systems so that they can send data back to a central data
repository for processing and possible action. For example, the group at
Cranfield University in the UK has deployed sensors for monitoring
environmental parameters (e.g., lead ion, pH, temperature) at hundreds of
sites throughout the country. Some of the sensors are continuous while
others require a discrete measurement by a technician at the site. The
developing database will be a significant resource for the environmental
community for both remediation and regulatory decisions.
Separate work at Dublin City University is directed at distributed
temperature sensors for monitoring fish from catch to market. By using
widely distributed sensors in fishing fleets, the people involved in the
product chain have an incentive to maintain cold conditions, as the value
of the catch will be reduced if the fish are exposed to temperatures outside
the specified range. Neither the United States nor Japan has prominent
efforts in either distributed or inexpensive biosensing.
Both of these areas—inexpensive and distributed sensors—underscore
the potential for integration of academic research into the fabric of societal
needs. While the United States is regarded as a bastion of
entrepreneurship, both Europe and Japan have strong and transparent
connections to industry and commercial applications. These direct links to
industry help facilitate the commercial introduction of sensors developed
in the research community. With pervasive and inexpensive wireless
communications systems on the horizon as well as the technology to make
 David R. Walt 37

inexpensive biosensors and sensors with additional capabilities, there

needs to be an investment in developing such mundane and ubiquitous
biosensing technologies.

NANOSTRUCTURED MATERIALS
At the WTEC Biosensing Study’s U.S. R&D Overview Workshop held
in Bethesda, MD, on 3–4 December 2002 (wtec.org/biosensing/
proceedings/), a theme in U.S. biosensing R&D became apparent:
nanostructured materials with built-in functionality and binding affinity are
increasingly being used to perform optical sensing. Research in
nanomaterials has led to the discovery of new optical (and other)
transduction mechanisms. This area is particularly promising, as it
leverages existing research investments. The Van Duyne group at
Northwestern University (www.chem.northwestern.edu/~vanduyne/) is
developing structured nanomaterials for surface-enhanced Raman-based
biosensing, as illustrated in Figure 2.3.

Fig. 2.3. Top: Nanoparticle array localized surface plasmon

resonance spectroscopy (LSPR)—local refractive
index change. Bottom: Nanostructured gold
materials on a substrate provide local enhancement
in the plasmon resonance. (Haes and Van Duyne
2002)
The Mirkin group at Northwestern (www.chem.nwu.edu/~mkngrp/) has
been focusing on new optical methods for detecting DNA binding based
on aggregation of gold nanoparticles. The Sailor group at the University of
California, San Diego (UCSD: chem-faculty.ucsd.edu/sailor/), has been
employing the unique material properties of porous silicon (called “smart
38 2. Optical Biosensing

dust”) to detect binding to the silicon surface, resulting in an

interferometric response manifested as a color change (see Figure 2.4).

Fig. 2.4. Porous Si particles can be fabricated and used to

sense analytes. Left: Spectral properties of different
Si particles are due to different interference
patterns. Middle: A single particle has nanometer
dimensions. Right: A suspension of the Si material
dispersed in solvent in a test tube. The background
shows an enlargement of the suspension. (Cunin
et al. 2002)
Multiple groups, such as those of Alivisatos at the University of
California, Berkeley; Nie at Emory University; and Bawendi at
Massachusetts Institute of Technology (MIT), are investigating the novel
properties of quantum dots for potential use in optical sensing and
biosensing applications. Complex optical sensing chemistries and
biochemistries are being attached to nanoparticles called PEBBLES
(Probes Encapsulated By Biologically Localized Embedding) by
Kopelman at the University of Michigan (www.umich.edu/~koplab/
research2/analytical/NanoScaleAnalysis.html) that can be introduced into
living cells to report on intracellular concentrations of key metabolites.
Some of these methods are able to detect extremely low absolute
numbers of molecules. The WTEC team found that many research efforts
on optical biosensing were beginning to push toward single molecule
detection limits. All of these methods are still at the research stage, and
significant additional work will be required to bring them to the stage
where they can be used for routine measurements.
Many of these investigators did not initially start out to develop
biosensing materials. Serendipity has frequently played a role in the
discovery of new transduction mechanisms and biosensing phenomena.
With these new discoveries, the biosensing community is attracting new
researchers to the field who are drawn to biosensing because they can see a
 David R. Walt 39

direct application of their fundamental work to an application. Other

materials researchers such as Swager at MIT (web.mit.edu/tswager/www/)
are not working at the nanoscale but also are contributing new optical
biosensing materials such as polymers that have amplified responses to
analyte binding. This latter approach seems to be the model being pursued
by investigators in Japan and Europe. They are either employing more
traditional techniques of organic synthesis and polymer chemistry to
generate new materials, or they are exploiting nanomaterials developed in
the United States.
An often-overlooked advantage of moving to the nanoscale is the
improved sensitivity exhibited by nanomaterial-based sensing. This
sensitivity is a consequence of binding to very small structures, which
localizes a small number of analyte molecules to an extremely small
volume, causing a locally high concentration and/or significant
perturbation in the vicinity of or on the surface of the nanomaterials. The
United States leads efforts in this area relative to the rest of the world.

APPLICATION OF MOLECULAR BIOLOGY TO OPTICAL

BIOSENSING
A major worldwide effort in optical sensing is aimed at using the tools
of molecular biology to design new sensing schemes. There is exciting
work being performed toward creating new protein constructs for optical
sensing. Particularly exciting work is being conducted in Professor
Umezawa’s laboratory at the University of Tokyo (see the site report in
Appendix C). In one approach, his laboratory is creating protein constructs
of two types. The first construct contains cyan fluorescent protein (CFP)
and yellow fluorescent protein (YFP) linked by a both a peptide substrate
and recognition site (Figure 2.5).
In this construct, when the peptide substrate is modified, the recognition
site binds to the modified peptide and changes the conformation of the
construct. The conformational change causes a change in the efficiency of
fluorescence resonance energy transfer from the CFP to the YFP. A second
construct employs two proteins, each carrying a piece of a variant green
fluorescent protein (EGFP). When the proteins are brought into proximity
due to the presence of an analyte, the two proteins are spliced, and EGFP is
reconstituted and green fluorescence begins to appear.
Molecular biological designs for optical sensing can be employed both
for generating purified proteins that can be immobilized to various
substrates and for whole cell biosensing in which the protein is expressed
and functional in a living cell (see Chapter 4). Many optical transduction
mechanisms are being exploited for cell-based biosensing. These methods
open the door to performing functional assays, in contrast to biosensors
40 2. Optical Biosensing

based on affinity binding. The ability to construct and express functional

molecules with optical tags in vivo presents a major opportunity and
capability for observing localization and dynamics within single cells. This
area presents a rich avenue for performing biosensing at the single cell
level and should lead to both fundamental and practical outcomes.

Fig. 2.5. A fluorescent indicator for protein phosphorylation in living

cells, named “phocus.” CFP and YFP are different colored
mutants of green fluorescent protein (GFP) derived from
Aequorea victoria. Upon phosphorylation of the substrate
sequence within phocus by a protein kinase, the adjacent
phosphorylation recognition domain binds with the
phosphorylated substrate sequence, which increases the
efficiency of fluorescent resonant energy transfer (FRET)
between the GFP mutants within phocus. (Courtesy, Prof.
Yoshio Umezawa, University of Tokyo Department of
Chemistry)

GENERAL OBSERVATIONS
A salient feature of European research in biosensing is the presence of
large integrated research groups and programs. Many groups in Europe
have centers or laboratories with more than 100 researchers. Such large
programs exist at the Universities of Dublin City, Potsdam, Twente,
Cranfield, and Linköping; smaller but still substantial teams exist at
Regensburg, Tübingen, and Cambridge. These groups combine expertise
 David R. Walt 41

in all the areas necessary to research and develop a sensing system:

molecular recognition, sensor fabrication, device assembly, data collection,
and data processing. The WTEC group noted the presence of integrated
biosensing programs in Japan, though generally not on the same order of
complexity as those in Europe. There are no comparable programs for
biosensing in the United States. Most U.S. biosensing efforts are single-
investigator or several-investigator projects rather than biosensing team
programs. Another observation—striking to the WTEC team—was the
lack of interest (and funding) in Europe and Japan for chemical and
biological threat agent detection. Every time the WTEC team broached this
subject, the response was met with a complete lack of enthusiasm.
Overview of Scientific Findings Related to Optical Biosensing
• There is much work on DNA and DNA arrays with low commercial
prospects.
• The United States leads in materials research resulting in new optical
sensing phenomena. Materials researchers are being drawn into the
optical biosensing field.
• Optical methods are pushing toward single-molecule detection levels.
• Molecular biological methods are being developed using optical
(fluorescence) biosensing, particularly whole cells for functional
assays in which the overall quality of the sample is analyzed (e.g.,
toxicity) rather than a specific analyte.
Challenges
Probably the most significant challenge for moving optical methods into
the marketplace is integration. Unlike lithographic techniques employed
for electronic devices, optical systems are most often fabricated by
assembling many individual components into a functional device. These
devices are generally hybrids in that they comprise both optical and
electrical components. Full integration of optical components using
lithography or other assembly methods remains a major challenge.
Research in optical materials for biosensing and for optical signal
transduction, microfabrication, and systems integration are all necessary to
advance the field of optical sensing.
Comparison of Optical Sensing Expertise by Region
Table 2.1 summarizes the WTEC panel’s findings with regard to optical
sensing achievements in Europe, Japan, and the United States in the main
areas of this field.
42 2. Optical Biosensing

Table 2.1
Optical Based Sensing
Topic Knowledge Base Work to Date Leading
Region
Interferometric, • Surface plasmon Europe
Label-free resonance
• Interference
Arrays • Patterning • DNA U.S.

• Surface chemistry arrays

• Protein
arrays
Cheap, • Screen printing Europe
Distributed
Sensors • Optical transduction
Nanotechnology • New signaling • Metal U.S.
mechanisms particles
• New materials
Molecular • Genetic engineering U.S.
Biology Europe
• Cell biology Japan
Integration • Engineering, Europe
Chemistry, Computer
Science

REFERENCES
Boisde, G., and A. Harmerand. 1996. Chemical and biochemical sensing with optical fibers
and waveguides. Norwood, MA: Artech House.
Cooper, M.A. 2002. Optical biosensors in drug discovery. Nature Reviews Drug Discovery
1:515–528.
Crooks, R.M. 2003. Bio/chemical sensing using thin film recognition elements. In WTEC
Biosensing Study U.S. R&D Overview Workshop Proceedings, Ch. 3, pp. 57–63.
Baltimore, MD: World Technology Evaluation Center. Available online:
wtec.org/biosensing/proceedings/03_session02.pdf.
Cunin, F., T.A. Schmedake, J.R. Link, Y.Y. Li, J. Koh, S.N. Bhatia, and M.J. Sailor. 2002.
Biomolecular screening with encoded porous-silicon photonic crystals. Nature Materials
1:39–41.
Epstein, J.R., and D.R. Walt. 2003. Fluorescence-based fibre optic arrays: A universal
platform for sensing. Chemical Society Reviews 32:203–214.
Gardner, J.W., and P.N. Bartlett. 1999. Electronic noses: Principles and applications. New
York: Oxford Univ. Press.
 David R. Walt 43

Haes, A.J., and R.P. Van Duyne. 2002. A nanoscale optical biosensor: Sensitivity and
selectivity of an approach based on the localized surface plasmon resonance
spectroscopy of triangular silver nanoparticles. J. Am. Chem. Soc. 124:10596–10604.
Hurst, W.J. 1999. Electronic noses and sensor array based systems: Design and
applications. Lancaster, PA: Technomic Publishing Company, Inc.
Kordal, R., ed. 2002. Microfabricated sensors: Application of optical technology for DNA
analysis. Journal of the American Chemical Society 124:11224.
Ligler, F.S., and C.A.R. Taitt, eds. 2002. Optical biosensors: Present and future.
Amsterdam: Elsevier Science.
Lopez-Higuera, J. 2002. Handbook of optical fiber sensing technology. West Sussex, UK:
John Wiley and Sons, Ltd.
Murphy, C.J. 2002. Optical sensing with quantum dots. Analytical Chemistry 74:520A-
526A.
Vercoutere, W., and M. Akeson. 2002. Biosensors for DNA sequence detection. Current
Opinion in Chemical Biology 6: 816–822.
Wolfbeis, O.S. 2002. Fiber-optic chemical sensors and biosensors. Analytical Chemistry
74:2663–2677.