Table of Contents

Copyright

Half Title

Title Page

Preface

Contributors

Chapter 1: Surface Plasmons for Biodetection

1.1 Introduction

1.2 Principles of SPR Biosensors

1.3 Optical Platforms for SPR Sensors

1.4 Functionalization Methods for SPR Biosensors

1.5 Applications of SPR Biosensors

1.6 Summary

References

Chapter 2: Microchip-Based Flow Cytometry in Photonic Sensing: Principles and Applications for Safety and Security Monitoring

2.1 Introduction

2.2 Microchip-Based Flow Cytometry

2.3 Microchip-Based Flow Cytometry with Integrated Optics

2.4 Applications

2.5 Conclusion

References

Chapter 3: Optofluidic Techniques for the Manipulation of Micro Particles: Principles and Applications to Bioanalyses

3.1 Introduction

3.2 Optofluidic Techniques for the Manipulation of Particles

3.3 Enhancing Optical Manipulation with a Monolithically Integrated on-Chip Structure

3.4 Applications

3.5 Conclusion

Acknowledgments

References

Chapter 4: Optical Fiber Sensors and Their Applications for Explosive Detection

4.1 Introduction

4.2 A Brief Review of Existing Fiber-Optic-Based Explosive Detectors

4.3 High Performance Fiber-Optic Explosive Detector Based on the AFP Thin Film

4.4 Generating High Quality Polymer Film—Pretreatment with Adhesion Promoter

4.5 Effect of Photodegradation on AFP Polymer

4.6 Optimizing Polymer Concentration for Optimized AFP-Film Thickness

4.7 Explosive Vapor Preconcentration and Delivery

4.8 Future Directions and Conclusions

References

Chapter 5: Photonic Liquid Crystal Fiber Sensors for Safety and Security Monitoring

5.1 Introduction

5.2 Materials and Experimental Setups

5.3 Principle of Operation

5.4 Tuning Possibility

5.5 Photonic Devices

5.6 Photonic Liquid Crystal Fiber Sensors for Sensing and Security

5.7 Conclusion

Acknowledgments

Chapter 6: Miniaturized Fiber Bragg Grating Sensor Systems for Potential Air Vehicle Structural Health Monitoring Applications

6.1 Introduction

6.2 Spectrum Fixed AWG-Based FBG Sensor System

6.3 Spectrum Tuning AWG-/EDG-Based FBG Sensor Systems

6.4 Dual Function EDG-Based Interrogation Unit

6.5 Conclusion

Acknowledgments

References

Chapter 7: Optical Coherence Tomography for Document Security and Biometrics

7.1 Introduction

7.2 Principle of OCT

7.3 OCT Systems: Hardware and Software

7.4 Sensing Through Volume: Applications

7.5 Summary and Conclusion

References

Chapter 8: Photonics-Assisted Instantaneous Frequency Measurement

8.1 Introduction

8.2 Frequency Measurement Using an Optical Channelizer

8.3 Frequency Measurement Based on Power Monitoring

8.4 Other Methods for Frequency Measurement

8.5 Challenges and Future Prospects

8.6 Conclusion

References

Index

Bseries

cover_image

cover_image

Title Page

Preface

Twenty-first century society places high priority on health, environment, and security; the threats of terrorism, climate change, and pollution are of increasing concern for both governments and citizens. While many of these threats can never be completely eliminated, their impact can be significantly mitigated by the presence of effective early-detection and early-warning systems. To provide such protection requires flexible, cost-effective sensing, and monitoring systems in areas such as structural integrity, environmental health, human security and health, and industrial process control. Photonics (the use of light) has the potential to provide highly effective solutions tailored to meet a broad range of specific sensing requirements, particularly, as it can leverage many of the technology platforms that were successfully developed for the communications industry. However, optimal sensing solutions require the development of specialized materials, novel optical devices, and new networking algorithms and platforms. As a sensor technology, photonics offers low power requirements, high sensitivity and selectivity, and immunity from electromagnetic interference. A single optical fiber can be used both to detect disturbances at multiple locations and to transit the information to a central point for data processing. Over the last two decades, we have seen the rapid development of photonic sensing technologies and their application in fields including food bacteria detection, oil/gas pipe structure health monitoring, bio chips, explosives detection, defense platform health monitoring, etc.

This book comprises a series of chapters contributed by leading experts in the field of photonic sensing, with target applications to safety and security. The objective is to provide a most comprehensive, though by no means complete, review of this exciting field. This book aims for multidisciplinary readership. The editors intend that the book serve as an invaluable reference that aids research and development of those areas that concern safety and security. Another aim of the book is to stimulate the interest of researchers from physics, chemistry, biology, medicine, mechanics, electronics, defense and others, and foster collaboration through multidisciplinary programs.

Each chapter of the book deals with a specific area of safety and security monitoring using the photonic technique. It provides discussions on background, operation principles, and applications. Chapter 1 is on surface plasmons and their applications to biodetection, in particular to food pathogen detection. Chapter 2 is on microchip-based flow cytometry and its application in bacteria detection and analysis. Chapter 3 is on optofluidic techniques and their application in bioanalysis. Chapter 4 is on optical fiber sensors and their application in explosives detection. Chapter 5 is on photonic liquid crystal fiber sensors and their application in safety and security monitoring. Chapter 6 is on optic fiber sensor systems targeting air vehicle structural health monitoring applications. Chapter 7 is on optical coherence tomography and its application in document security and biometrics. Chapter 8 is on photonics-assisted instantaneous frequency measurement and its potential for electronic warfare applications.

We are grateful to all the authors for their contributions. Without their assistance and cooperation, this book would not have been possible. We also like to thank Ms. Aleksandra Czapla for the cover design, Ms. Kari Capone and Ms. Lucy Hitz of Wiley for their continual support during the course of this book.

Gaozhi Xiao and Wojtek Bock

Contributors

Pavel Adam, Institute of Photonics and Electronics, Academy of Sciences of the Czech Republic, Prague, Czech Republic
Wojtek J. Bock, Centre de recherche en photonique, Département d'informatique et d'ingénierie, Université du Québec en Outaouais, Gatineau, Québec, Canada
Shoude Chang, Institute for Microstructural Sciences, National Research Council Canada, Ottawa, ON, Canada
Costel Flueraru, Institute for Microstructural Sciences, National Research Council Canada, Ottawa, ON, Canada
Honglei Guo, Microwave Photonics Research Laboratory, School of Information Technology and Engineering, University of Ottawa, Ottawa, ON, Canada
Jiimagesí Homola, Institute of Photonics and Electronics, Academy of Sciences of the Czech Republic, Prague, Czech Republic
Jianjun Ma, Centre de recherche en photonique, Département d'informatique et d'ingénierie, Université du Québec en Outaouais, Gatineau, Québec, Canada
Youxin Mao, Institute for Microstructural Sciences, National Research Council Canada, Ottawa, ON, Canada
Nezih Mrad, Air Vehicles Research Section, Defense R&D Canada, Department of National Defense, National Defense Headquarters, Ottawa, ON, Canada
Shilong Pan, Microwave Photonics Research Laboratory, School of Information Technology and Engineering, University of Ottawa, Ottawa, ON, Canada
Marek Piliarik, Institute of Photonics and Electronics, Academy of Sciences of the Czech Republic, Prague, Czech Republic
Hana Šípová, Institute of Photonics and Electronics, Academy of Sciences of the Czech Republic, Prague, Czech Republic
Tomáš Špringer, Institute of Photonics and Electronics, Academy of Sciences of the Czech Republic, Prague, Czech Republic
Milan Vala, Institute of Photonics and Electronics, Academy of Sciences of the Czech Republic, Prague, Czech Republic
Benjamin R. Watts, Department of Engineering Physics, McMaster University, Hamilton, ON, Canada
Tomasz Wolinski, Faculty of Physics, Warsaw University of Technology, Warszawa, Poland
Gaozhi Xiao, Institute for Microstructural Sciences, National Research Council Canada, Ottawa, ON, Canada
Chang-Qing Xu, Department of Engineering Physics, McMaster University, Hamilton, ON, Canada
Jianping Yao, Microwave Photonics Research Laboratory, School of Information Technology and Engineering, University of Ottawa, Ottawa, ON, Canada
Zhiyi Zhang, Institute for Microstructural Sciences, National Research Council of Canada, Ottawa, ON, Canada

Chapter 1

Surface Plasmons for Biodetection

Pavel Adam, Marek Piliarik, Hana Šípová, Tomáš Špringer, Milan Vala and Jiimagesí Homola

Institute of Photonics and Electronics, Academy of Sciences of the Czech Republic, Prague, Czech Republic

1.1 Introduction

The diffusion of inorganic and biological worlds represents an important paradigm of modern science and technology [1]. Biophotonics stands as an emerging field of research at the crossroads of physical, chemical, and life sciences. The integration of photonics, biology, and nanotechnology is leading to a new generation of devices that makes it possible to characterize chemical and other molecular properties and to discover novel phenomena and biological processes occurring at the molecular level. Biophotonics is widely regarded as the key science on which the next generation of clinical tools and biomedical research instruments will be based.

The last two decades have witnessed an increasing effort devoted to the research and development of optical biosensors and biochips worldwide. Recent scientific and technological advances have demonstrated that such devices hold tremendous potential for applications in areas such as genomics, proteomics, medical diagnostics, environmental monitoring, food analysis, agriculture, and security [2–4]. Label-free optical biosensors present a unique technology that enables the direct observation of molecular interaction in real-time and allows for the study of molecular systems, which cannot be labeled and studied by fluorescence spectroscopy [2]. Optical label-free biosensors measure binding-induced refractive index changes and are typically based on interferometric transducers, such as the integrated optical Mach–Zehnder interferometer [5], the integrated Young interferometer [6], and the white light interferometer [7], and transducers based on spectroscopy of guided modes of dielectric waveguides, such as the resonant mirror sensor [8] and the grating coupler sensor [9], or metal-dielectric waveguides, such as the surface plasmon resonance (SPR) sensor.

Since the first demonstration of the SPR method for the study of processes at the surfaces of metals [10] and sensing [11] in the early 1980s, SPR sensors have received a great deal of attention and allowed for great advances both in terms of technology and applications [12]. Thousands of research papers on SPR biosensors have been published and SPR biosensors have been extensively featured in books [1, 2, 4, 13] and reviews [3, 12, 14–18]. SPR biosensors have become a crucial tool for characterizing and quantifying biomolecular interactions. SPR biosensors have also been increasingly developed for the detection of chemical and biological species and numerous SPR biosensors for the detection of analytes related to medical diagnostics, environmental monitoring, food safety, and security have been reported as well.

This chapter describes the principles of SPR biosensors and discusses the advances that SPR biosensors have made both in terms of technology and applications over the last decade. The first part (Section 1.2) describes the fundamentals of SPR biosensors. Sections 1.3 and 1.4 are concerned with the optical configurations and immobilization methods used in current SPR sensors. The last part (Section 1.5) presents examples of applications of SPR biosensors for the detection of chemical and biological species with an emphasis on food safety and security applications.

1.2 Principles of SPR Biosensors

1.2.1 Surface Plasmons

Surface plasmons (SPs) are electromagnetic modes guided by metallic waveguides. The simplest geometry supporting SPs comprises a planar boundary between a semi-infinite metal and a semi-infinite dielectric. The optical properties of the metal are characterized by a complex permittivity images, where images and images are the real and imaginary parts of ε m) and the dielectric is characterized by the refractive index nd. Analysis of Maxwell's equations with appropriate boundary conditions suggests that this structure can only support a single guided mode of electromagnetic field—an SP [19]. The vector of intensity of the magnetic field of SP lies in the plane of the metal–dielectric interface and is perpendicular to the direction of propagation. Such a mode of the electromagnetic field is referred to as the transversally magnetic (TM) mode. A typical profile of the magnetic field of an SP is shown in Figure 1.1(a). The intensity of the magnetic field reaches its maximum at the metal–dielectric interface and decays into both the metal and the dielectric. The field decay in the direction perpendicular to the metal–dielectric interface is characterized by the penetration depth. The penetration depth depends on the wavelength and permittivities of the materials involved. The penetration depth into the dielectric for an SP propagating along the interface of gold and a dielectric with nd = 1.32 increases with a wavelength and ranges from 100 to 600 nm in the wavelength region 600–1000 nm [19].

Figure 1.1 (a) Spatial distribution of the magnetic intensity for SP at the interface between gold and a dielectric (nd = 1.32) in the direction perpendicular to the interface, λ = 850 nm. (b) Effective index of SP propagating along the interface between a dielectric (refractive index—1.32) and metal (gold) as a function of the wavelength.

1.1

Propagation constant of SP βSP at the metal–dielectric interface can be expressed as

1.1 1.1

where c is the speed of light in a vacuum, ω is the angular frequency, and nef is the effective index of the SP [20, 21]. If the structure is lossless images, Equation 1.1 represents a guided mode only if the metal permittivity images is negative and images. Metals such as gold, silver, and aluminum exhibit a negative real part of permittivity in the visible and near-infrared region of the spectrum. Figure 1.1b depicts the wavelength dependence of the effective index of SP nef for the gold waveguide. The imaginary part of the propagation constant is associated with the imaginary part of the metal permittivity images and determines attenuation of the SP in the direction of propagation [20].

A special example of the metallic waveguide is a symmetric dielectric–metal– dielectric planar structure. When the metal film thickness is much larger than the SP penetration depth into the metal, an independent SP may propagate at each metal–dielectric boundary. If the thickness of the metal film is decreased, coupling between the SPs at opposite sides of the metal film can occur, giving rise to mixed modes of electromagnetic field—symmetric and antisymmetric SPs [22, 23]. The profiles of magnetic intensity of symmetric and antisymmetric SPs are symmetric or antisymmetric with respect to the plane of symmetry of the structure. The field of the symmetric SP penetrates much deeper into the dielectric medium than the field of the antisymmetric SP or the field of a conventional SP at a single metal–dielectric interface. Moreover, the symmetric SP exhibits a lower attenuation than its antisymmetric counterpart and therefore it is referred to as a long-range surface plasmon (LRSP) while the antisymmetric mode is referred to as a short-range surface plasmon [22].

1.2.2 Excitation of Surface Plasmons

1.2.2.1 Prism Coupling

The most common approach to the excitation of SPs is by means of a prism coupler and the attenuated total reflection method (ATR). In the Kretschmann geometry of the ATR method [24], a high refractive index prism with refractive index np is interfaced with a metal–dielectric waveguide consisting of a metal film with permittivity ε m and a semi-infinite dielectric with a refractive index nd (nd < np), Figure 1.2.

Figure 1.2 Excitation of surface plasmons in the Kretschmann geometry of the attenuated total reflection (ATR) method.

1.2

When a light wave propagating in the prism totally reflects on the prism base, an evanescent electromagnetic wave decays exponentially in the direction perpendicular to the prism–metal interface [25]. If the metal film is sufficiently thin (less than 100 nm for light in the visible and near-infrared part of spectrum), the evanescent wave penetrates through the metal film and couples with an SP at the outer boundary of the metal film. In terms of the effective index, this coupling condition can be written as follows:

1.2 1.2

where nef is the effective index of the SP, and the perturbation in effective index images, and the respective propagation constant of SP images describe the effect of the presence of the prism.

Figure 1.3 shows the angular and wavelength spectra calculated using a rigorous Fresnel model of light reflection on a multilayer structure calculated at two different wavelengths and for two angles of incidence, respectively. The reflectivity spectra exhibit distinct dips in TM polarization, which are associated with the transfer of energy from the incident light wave into an SP and its subsequent dissipation in the metal film.

Figure 1.3 TM reflectivity as a function of the angle of incidence (a) and the wavelength (b) calculated for two different angles of incidence using the rigorous Fresnel reflection theory. Configuration: BK7 glass, gold film (thickness—48 nm for the wavelength of 650 nm and 50 nm for the wavelength of 850 nm), water.

1.3

The reflectivity spectra can be rigorously calculated using Maxwell equations and the boundary condition of the planar multilayer structure. Assuming that the permittivity of metal ε m obeys images and images, a Lorentzian (with respect to nef) approximation of the reflectivity can be used as follows [20]:

1.3 1.3

where γi = Im{βSP}λ/2π and γrad = Im{β(1)}λ/2π denote the attenuation coefficients of SPs owing to absorption and radiation, respectively. As follows from Equation 1.3, the minimum of the dip in the reflectivity spectrum occurs when the coupling condition (Eq. 1.2) is matched and the shape of the reflectivity dip depends strongly on the strength of the coupling between the excitation wave and SP represented by γrad. This approximation has been shown to provide a good estimate of the position of the reflectivity dip and to predict the shape of the reflectivity curve in the neighborhood of the minimum [19]. In addition, the Lorentzian curve exhibits the same width as the dips calculated using the rigorous approach [26].

1.2.2.2 Grating Coupling

Another approach to optical excitation of SPs is based on the diffraction of light on a diffraction grating. In this method, a light wave is incident at an angle of incidence θ from a dielectric medium with the refractive index nd on a metal grating with the dielectric constant ε m, the grating period Λ, and the grating depth q (Fig. 1.4). The diffracted wave in m—the diffraction order can couple with SPs when their propagation constants are closely phase-matched. In terms of the effective index, the coupling condition can be written as

1.4 1.4

where images accounts for the presence of the grating.

Figure 1.4 Excitation of surface plasmons by the diffraction of light on a diffraction grating.

1.4

The grating-moderated interaction between a light wave and an SP can be modeled by solving Maxwell's equations in differential form with a grating profile approximated by a stack of layers [27, 28], or in an integral form by solving the Helmholtz–Kirchhoff integral [29].

Figure 1.5 shows the reflectivity spectra (angular and wavelength) for light incident from water onto a gold grating (calculated using the integral method). The angular as well as wavelength reflectivity spectra exhibit a characteristic dip caused by the transfer of energy of the incident light into an SP. Figure 1.5 shows resonant spectra narrower by a factor of 10 compared to resonant dips obtained with a prism coupler in Figure 1.3. On shallow diffraction gratings, SPs are excited at the angles of incidence close to the coupling angles predicted from the matching condition, neglecting the effect of the grating. The depth of the reflectivity dips depends on the depth of the grating. For the structure considered in Figure 1.5 (gold–water interface; grating period 672 nm), the strongest excitation of an SP (R = 0) occurs with a grating depth of about 30 nm.

Figure 1.5 Reflectivity as a function of the angle of incidence (a) and the wavelength (b) calculated for two wavelengths and angles of incidence, respectively (denoted in the label). Configuration: gold–water interface; grating period 672 nm, grating depth 30 nm, angle of incidence taken in the air.

1.5

1.2.2.3 Waveguide Coupling

SP can also be excited by modes of dielectric waveguides, such as planar or channel-integrated optical waveguides and optical fibers. Typically, coupling between a dielectric waveguide mode and an SP propagating along a metal layer in the proximity of the dielectric waveguide is achieved by coupling the evanescent tails of the two waves. The evanescent wave of the dielectric waveguide mode can couple with SP when the propagation constant of the mode βM is equal to the real part of the propagation constant of the SP βSP:

1.5 1.5

As the coupling condition is fulfilled for only a narrow range of wavelengths, the excitation of SPs can be observed as a narrow dip in the spectrum of light transmitted through the waveguide structure.

1.2.3 Sensors Based on Surface Plasmons

A change in the refractive index of the dielectric medium produces a change in the propagation constant of SP at the interface of the metal and the dielectric. Thischange in propagation constant alters the coupling condition between the light wave and the SP, which can be observed as a change in the characteristics of the optical wave interacting with SP. The change in SP propagation is associated with the change in resonant coupling conditions, for example the resonant wavelength, the angle of incidence, or the strength of the SP coupling. Figure 1.6 illustrates the shift in the angular of wavelength SPR spectra because of the change in the refractive index of the dielectric medium nd. Depending on which characteristics of the reflected light wave are measured in the SPR sensor, SPR sensors are classified as (i) SPR sensors with wavelength modulation (the angle of incidence is fixed and the coupling wavelength serves as a sensor output); (ii) SPR sensors with angular modulation (the coupling wavelength is fixed and the coupling angle of incidence serves as a sensor output); (iii) SPR sensors with intensity modulation (both the angle of incidence and the wavelength of incident light are fixed at nearly resonant values and the light intensity serves as a sensor output); and (iv) SPR sensors with phase modulation (both the angle of incidence and the wavelength of incident light are fixed at nearly resonant values and the phase of the reflected light serves as a sensor output).

Figure 1.6 Change in the SPR spectra associated with the change in the refractive index of the dielectric media. Parameters images, images, and images denote the shift in the resonant wavelength, the resonant angle of incidence, and the change in light intensity.

1.6

A change in the refractive index Δnd of the dielectric film with thickness h, produces a change in the effective index of the SP Δnef. If the thickness of the layer is much higher that the penetration depth Lpd of the SP field, the change in the effective refractive index of the SP can be calculated by differentiating the dispersion relation (1.1), which for the conventional SP yields [19]:

1.6 1.6

A change in the effective index of an SP owing to refractive index changes within a thin layer h images Lpd can be estimated using the perturbation theory [19] as

1.7 1.7

1.2.4 SPR Affinity Biosensors

SPR biosensors employ special biomolecules—referred to as biorecognition elements—that can recognize and capture target analytes. Such biorecognition elements are immobilized in the form of a sensitive layer on the surface of the SPR metallic waveguide. When a solution containing analyte molecules is brought into contact with the SPR sensor, analyte molecules in solution bind to the molecular recognition elements, producing an increase in the refractive index of the sensitive layer (Fig. 1.7).

Figure 1.7 Principle of SPR affinity biosensing.

1.7

The change in refractive index Δnd occurring within a layer of thickness h can be expressed as

1.8 1.8

where (dn/dc)vol is the refractive index increment, Δcb is the wt/vol concentration of bound molecules within the sensitive layer with the thickness h, and images is the corresponding surface concentration (mass per surface area). The linear relation between the refractive index change and the surface concentration of bound molecular mass in Equation 1.4 is often referred to as the de Feijter formula [30]. The refractive index increment (dn/dc)vol is a well-characterized property for most biochemical species and ranges typically from 0.1 to 0.3 cm3/g [31]. Proteins and nucleic acids exhibit quite a consistent refractive index increment value, which falls within 8% from the value of (dn/dc)vol = 0.18 cm3/g [31]. As follows from Equations 1.6–1.8, a change in the effective index of the SP owing to the capture of analyte can be expressed as

1.9 1.9

where K is a constant.

1.2.5 Performance Characteristics of SPR Biosensors

The performance of SPR biosensors is usually characterized in terms of the sensitivity, resolution, limit of detection (LOD), linearity, accuracy, reproducibility, and dynamic range [19, 26].

The sensitivity of an SPR sensor is the ratio of the change in sensor output to the change in the quantity to be measured (e.g., the refractive index nd). The sensitivity of an SPR sensor to a refractive index S can be written as

1.10 1.10

where Y denotes the sensor output and depending on the modulation approach usually represents the resonant angle of incidence, or the resonant wavelength, or the reflectivity. Equation 1.10 describes the decomposition of the sensor sensitivity in two parts: (i) the sensitivity of the sensor output to the change in the effective index of an SP and (ii) the sensitivity of the effective index of an SP to the change in the refractive index [19, 26]. Therefore, the second term (ii) is independent of the modulation method and the method of excitation, and the first term (i) represents the instrumental factor which is independent of the measurand.

Analogously, the sensitivity of an SPR biosensor to a concentration of analyte c derives from the change in the refractive index of the sensitive layer nd caused by the analyte binding:

1.11 1.11

As with Equation 1.10, the first term is the instrumental factor and the second term characterizes the properties of SP mode. The third term in Equation 1.11 is derived from the relationship between the analyte concentration and the refractive index change described in Equation 1.8, the binding capabilities of biorecognition elements, and analyte transport to the sensor surface.

The instrumental factor ΔYnef has been analyzed in detail for different combinations of the SPR couplers and modulation approaches in recent publications [26]. Clearly, the instrumental factor depends on which method of excitation of the SPs and modulation approach are used. For the most common sensors based on SPR spectroscopy with a prism coupler, the instrumental factor can be calculated by differentiating coupling conditions (1.2) as follows:

1.12 1.12

1.13 1.13

where λr and θr denote the resonant wavelength and angle of incidence respectively. Instrumental factors are thus determined by the geometry and material constants of the SPR coupler and dispersion of the SP mode. In similar manner, the instrumental factor ΔRnef can be calculated for SPR sensors based on intensity modulation using the Lorentzian approximation Equation 1.3 as

1.14 1.14

Unlike instrumental factors calculated for SPR spectroscopy (Eqs. 1.12 and 1.13), Equation 1.14 suggests a significant dependence of the sensitivity on the strength of SPR coupling represented by the radiation coefficient γrad. This behavior is associated with the effect of γrad term on the shape of the SPR dip as discussed in Section 1.2.2.

The resolution of an SPR sensor is defined as the smallest change in the refractive index that produces a detectable change in the sensor output. The magnitude of the change in sensor output that can be detected depends on the level of uncertainty of the sensor output—the output noise. Resolution of an SPR sensor σn, is typically expressed in terms of the standard deviation of noise of the sensor output σY translated to the refractive index of the bulk medium:

1.15 1.15

The noise in the sensor output originates from the noise of individual light intensities involved in the calculation of the sensor output. The propagation of noise to the sensor output was investigated by Piliarik and Homola [26]. Their study revealed that, independent of the SPR coupling principle, the refractive index resolution of an SPR sensor can be expressed as follows:

1.16 1.16

where K is the noise distribution factor, r is the noise correlation factor, N is the number of intensities involved in the measurement, I0 is the intensity of the incident light which corresponds to one detector (e.g., spectrometer pixel), and σI(max) is the standard deviation of intensity noise (for amplitude sensors this corresponds to the level of measured intensity and for spectroscopic sensors this corresponds to the level at the threshold of the SPR dip). The noise distribution factor is between K = 0.38 for amplitude SPR sensors and K = 0.50 for SPR spectroscopy with noise homogeneously distributed across the SPR dip (K = 0.43 for shot noise limited SPR spectroscopy being the most common case). The noise correlation factor r = 1 for uncorrelated intensities. The factor r is derived from the Pearson correlation coefficient ρ of individual pixels of the spectrum as images for spectroscopic SPR sensors. For SPR sensors with intensity modulation, images, where N is the number of detector pixels.

As follows from the analysis presented above, the resolution of the SPR sensor depends on the noise of the used optoelectronic components, the strength of the coupling between the light wave and the SP, and material parameters of the metallic waveguide [26]. In contrast, although the sensitivity of SPR sensors depends strongly on the coupling principle and modulation, a comparable resolution can be achieved regardless of the SPR coupling principle and modulation.

The LOD of an SPR biosensor represents the ability of a biosensor to detect an analyte. LOD is defined as the concentration of analyte derived from the smallest measure of the sensor output Y, which can be distinguished from the sensor output corresponding to a blank sample. The value of the sensor output corresponding to the LOD, YLOD, can be expressed as

1.17 1.17

where Yblank is the mean of the blank measures, σblank is the standard deviation of the blank measures, and m is a numerical factor chosen for the required confidence level (typically m = 3).

1.3 Optical Platforms for SPR Sensors

Since the first demonstration of SPR sensors in the early 1990s, optical platforms of SPR sensors have made substantial advances in terms of both performance [32–45] and new capabilities. SPR instruments have become compact enough to be used for routine bioanalytical tasks in the field [46–52] and have expanded to enable parallelized detection of hundreds of different analytes at a time [35, 53–56].

In this section, we present a brief overview of the advances in SPR optical platforms, in particular within the last decade. The section is organized into four subsections, based on the method of coupling of light to SPs (prism, grating, and waveguide). The last subsection presents examples of the available commercial SPR systems.

1.3.1 Prism-Based SPR Sensors

The use of prism couplers to couple light to SPs is straightforward, versatile, and does not require complex optical instrumentation. Therefore, SPR sensors based on the attenuated total reflection method and prism couplers have been the most widely used. All major types of modulation have been implemented in prism-based SPR sensors. In SPR sensors based on spectroscopy of SPs, the angular or wavelength spectrum of the optical wave coupled with the SP is measured. Alternatively, changes in the intensity or phase of the reflected wave can be measured at a fixed wavelength and an angle of incidence. While the spectroscopic SPR sensors usually offer higher resolution, they provide a rather limited number of sensing channels. By contrast, intensity or phase modulations can be adopted by SPR imaging configurations, where independent measurements are performed in as many as hundreds of sensing channels simultaneously.

1.3.1.1 Spectroscopic Prism-Based SPR Sensors

The use of angular modulation in SPR sensors has a long history. In 1991, Sjolander et al. reported an angular modulation-based SPR sensor utilizing a light-emitting diode (LED), a glass prism with a gold layer, and a detector array with imaging optics to detect the angular spectrum of light reflected from the gold layer (Fig. 1.8) [32]. A divergent beam emitted by the LED was collimated in the plane parallel to the gold layer and focused by a cylindrical lens in the perpendicular plane to produce a wedge-shaped beam illuminating a thin gold film. The imaging optics shaped the reflected beam in such a way that the angular spectrum of each sensor channel was projected on a separate row (or rows) of the array detector. This design has been further advanced by Biacore AB (Sweden) and has resulted in a variety of commercial SPR sensors (subsection 1.3.4).

Figure 1.8 (a) Side view and (b) top view of a prism-based SPR sensor in angular configuration with three parallel channels. a—light-emitting diode, b—lenses, c—sensor chip, d—microfluidic cartridge, e—coupling prism, f—polarizer, g—photodiode array detector.

Reprinted, by permission, from Reference 32

1.8

An interesting approach in the miniaturization of SPR sensors with angular modulation was reported by Thirstrup et al. [49]. They developed a polymeric exchangeable sensor chip containing two diffractive optical coupling elements (DOCEs). The DOCEs (chirped relief gratings) were used to focus an incident parallel beam to a sensing spot and project the reflected beam onto a detector array (Fig. 1.9). The sensor provided an angular sensitivity of 140°/RIU and a resolution of 5 × 10−7 RIU. The SPR chips with DOCEs were produced in plastic by injection molding, which provides a low-cost method for the mass production of SPR chips. However, the limited quality of the diffraction coupling elements caused background illumination, which had a negative influence on the performance of the sensor.

Figure 1.9 Scheme of the SPR sensor based on the diffractive optical coupling element (DOCE).

Reprinted, by permission, from Reference 49

1.9

Another compact SPR sensor based on angular modulation was proposed by the researchers at Texas Instruments and the University of Washington in the 1990s. In recent years, this design (referred to as Spreeta), has been further refined. Spreeta SPR sensor had specific dimensions of only 3 × 1.5 × 4 cm3 and consisted of a plastic prism molded to a printed circuit board (PCB). The PCB contained a microelectronic circuit containing an LED and a linear diode array detector. The divergent light beam emitted by the LED passed through a polarizer and hit the sensor surface at a range of angles above the critical angle. Therefore, different areas on the sensor chip were illuminated under different angles (Fig. 1.10). The light reflected from the mirror placed on the top of the sensor was captured by the detector. The sensor was showed to have a resolution of 1.8 × 10−7 RIU within a time interval of 0.8 s. However, the sensor exhibited deviations from a smooth response of 0.2% during the measurement of a change in the refractive index of 0.04 RIU, which corresponds to a resolution of 8 × 10−5 RIU [47]. Subsequently, Chinowsky et al. demonstrated a portable 24-channel SPR sensor integrating eight 3-channel Spreeta sensors into a single suitcase [48]. The sensor contained supporting electronics and a microfluidic system with a temperature-controlled flow cell and was demonstrated to provide a resolution of 1–3 × 10−6 RIU [48]. Currently, the Spreeta sensor is also commercially available from Sensata Technologies (Section 1.3.4).

Figure 1.10 Scheme of the Spreeta sensor from Texas Instruments.

Reprinted, by permission, from Reference 47

1.10

Kim et al. reported a portable SPR platform based on angular modulation of SPR in eight parallel sensing channels [50]. A special gold-coated replaceable prism served as a coupling prism and a sensing chip at the same time. This made it possible to avoid the use of refractive index matching oil. The system was reported to exhibit a baseline noise of 0.7 mdeg [50], which for a typical angular sensitivity of about 95°/RIU [19] corresponds to a resolution of 7 × 10−6 RIU.

In 2008, Feltis et al. reported a fully integrated handheld SPR device with a size of only 15 × 8 cm2 and a weight of 600 g [51] (Fig. 1.11). The sensor was based on angular modulation of SPR using a laser diode and a photodiode array as a light source and detector, respectively. The sensor only measured in a single channel and was demonstrated to achieve a resolution of about 3 × 10−6 RIU [51].

Figure 1.11 Photograph of a handheld SPR device based on a prism coupler and angular interrogation.

Reprinted, by permission, from Reference 51

1.11

In 2002, Song et al. reported an SPR sensor based on angular interrogation of SPR, which utilized a bi-cell detector instead of a photodiode array [57]. The intensity of the reflected light captured by the detector's two cells, A and B, was analyzed and the position of the SPR minimum was determined by dividing the differential and sum signals (A − B)/(A + B). The presented baseline noise of about images corresponds to a resolution of 4 × 10−6 RIU (assuming an operating wavelength of 635 nm and a sensitivity of about 140°/RIU [19]). A similar approach was reported by Zhang et al. [36], who used a quadrant detector and thus were able to measure in two channels simultaneously. This sensor was reported to provide an angular resolution of 10−5, which corresponded to a refractive index resolution of about 10−7 RIU (assuming a sensitivity of 130°/RIU [19]).

In 2007, researchers from Agilent Technologies (USA) reported an SPR system, which offers both angular modulation and SPR imaging [35]. The sensor configuration with angular is depicted in Figure 1.12a. The light from a laser diode was collimated and passed through an acousto-optic deflector (AOD) where the direction of the beam was modulated within a defined angular span. The deflected light was then imaged by a two-lens telescope onto a sensing chip so that the illuminated region remained at the same place during the angular scanning. The surface of the sensing chip was then imaged onto a high-speed CMOS-based camera. In the performed experiment, the camera frame-rate was 1.1 kHz and scanning through the full angular range was done every 100 ms, which allowed for the acquisition of more than 100 spectra each second. The combination of a powerful light source and a high-speed camera allowed for massive averaging of angular spectra and therefore offered spectra with extremely low noise. The resulting performance of the sensor as a function of the number of imaged spots is shown in Figure 1.12b for two different values of dynamic range. For example, the sensor imaging 30 sensing spots and offering a dynamic range of 0.015 RIU provided a resolution of about 7 × 10−9 RIU [35].

Figure 1.12 (a) Configuration of the high-resolution and high-throughput SPR imaging sensor based on fast angular scanning. (b) System performance as a function of the number of cells and dynamic range. Black data (experimental datapoints and theoretical extrapolations) corresponds to a dynamic range of 0.06 RIU, while the gray data were obtained for a 0.015 RIU dynamic range.

Reprinted, by permission, from Reference 35

1.12

Wavelength modulation has also found its way to spectroscopic SPR sensors. For instance, a high-resolution prism-based SPR sensor with wavelength modulation was reported by Homola et al. (Fig. 1.13) [33, 58]. Light from a halogen lamp was transmitted through a multimode optical fiber and after collimation and polarization it was made incident on an SPR chip with a gold film interfaced with a coupling prism. The reflected polychromatic beam was then coupled to output optical fibers and transmitted to a multichannel spectrometer. With the use of an advanced data processing algorithm [34], a baseline noise as low as 1.5 × 10−3 nm was established. This low noise level makes it possible to achieve a refractive index resolution of 2 × 10−7 RIU (assuming an operating wavelength of 750 nm and sensor sensitivity of 7500 nm/RIU [19]).

Figure 1.13 Four-channel sensor based on a wavelength interrogation of SPR in a Kretschmann geometry.

Reprinted, by permission, from Reference 58

1.13

In order to increase the amount of information in SPR sensors with wavelength modulation, wavelength division multiplexing (WDM) was introduced to SPR sensors with wavelength modulation. Two different approaches to WDM were proposed by Homola's group [39, 40]. In the first approach, a sensing surface consisting of a thin gold film partly coated with a thin dielectric film was used. As the presence of the thin dielectric film red-shifted the resonant wavelength (compared to the bare gold), the reflected light exhibited two dips associated with the excitation of SPs in the area with and without the dielectric overlayer [39]. In the second configuration, a special skewed prism was used in which polychromatic light was sequentially made incident on two sensing channels under different angles [40]. This WDM approach was combined with the previously described 4-channel SPR sensor with four parallel beams yielding an 8-channel SPR sensor. The resolution was 1 × 10−6 RIU and 7 × 10−7 RIU for the short-wavelength (640 nm) channel and the long-wavelength (790 nm) channel, respectively [40].

Bardin et al. [41] reported an SPR sensor combining high resolution of spectroscopy of SPs with a high number of sensing channels. In this spectro-imaging SPR sensor, polychromatic light reflected from a linear array of sensing spots was dispersed by a diffraction grating and projected onto a 2D detector. The spectra of light reflected from individual channels appeared as narrow bands on the detector. The sensor was demonstrated to attain a resolution of 3.5 × 10−7 RIU in up to 20 channels [41].

More complex plasmonic modes, such as LRSPs, have also been exploited in spectroscopic SPR sensors. In 2001, Nenninger et al. [37] reported an SPR sensor utilizing LRSPs. LRSP is a guided mode of an optical structure consisting of a very thin metal film sandwiched between two dielectrics with similar refractive indices (Section 1.2.1). Nenninger et al. investigated LRSP sensing structures employing two dielectrics as a buffer layer—one with refractive indices slightly below (Teflon AF) and above (MgF2) that of water. In the wavelength-modulation-based SPR sensor, the MgF2 chips were found to perform better and to achieve a resolution of 2 × 10−7 RIU [37]. In 2007, Slavík and Homola improved the design and implementation of the LRSP-based sensor (with Teflon AF as a buffer layer) and achieved a resolution of 2.5 × 10−8 RIU [38].

1.3.1.2 Prism-Based Sensors with Intensity or Phase Modulation

Intensity modulation has been increasingly used in SPR sensors to enable spatially resolved measurements [59, 60]. In typical SPR imaging sensors, a parallel TM-polarized beam of monochromatic light is launched into a prism coupler and made incident on a thin metal film at an angle of incidence close to the coupling angle for the excitation of SPs. The intensity of reflected light depends on the strength of the coupling between the incident light and the SP and therefore can be correlated with the distribution of the refractive index along the metal film surface [61–63]. In order to increase the sensor stability and optimize the contrast of SPR images, Fu et al. proposed an SPR imaging employing a white light source and a bandpass interference filter [64]. This SPR sensor was demonstrated to provide a refractive index resolution of 3 × 10−5 RIU [61].

In 2007, Piliarik et al. [53] reported an SPR imaging sensor based on polarization contrast and an array of patterned multilayers. In this configuration, each sensing channel consisted of a pair of sensing spots (type I and II; Fig. 1.14) with different multilayers. The multilayers were designed in such a way that with an increasing refractive index at the sensor surface, the intensity of the reflected light increased for spot type I and decreased for spot type II. When the sensor response is defined as a ratio of the intensities of light reflected from the spots I and II, the sensitivity is higher than the sensitivity for each individual spot. In addition, the defined sensor output is therefore insensitive to light level fluctuations, which are correlated across the sensing area. The sensor was demonstrated to provide an extended dynamic range of 0.012 RIU and a resolution of 2 × 10−6 RIU. Later, the same group reported an SPR imaging sensor combining the polarization contrast with a different concept of internal referencing. This concept is based on two mirrors enabling real-time compensation of fluctuations of the dark current and intensity of incident light [54]. Resolution as low as 2 × 10−7 RIU in 120 sensing spots was achieved with this type of sensor [54].

Figure 1.14 Scheme of an SPR imaging device based on polarization contrast and internal referencing using two types of multilayer structures on each sensing channel.

Reprinted, by permission, from Reference 53

1.14

In 2007, Law et al. reported an SPR sensor with phase modulation and an extended dynamic range [65]. The sensor was based on temporal modulation of the phase and detection of the second and third modulation harmonics. While the detection of the third modulation harmonics resulted in a good resolution (3 × 10−7 RIU) and a limited operating range (0.004 RIU), detection of the second harmonic modulation offered a worse resolution (10−5 RIU) and a larger operating range (0.012 RIU) [65].