Open access peer-reviewed chapter
Abstract
The label-free, all-optical biosensors, have been gaining more and more popularity in recent years. They can provide almost as high sensitivity as conventional techniques (like ELISA) in the detection of pathogens, although in a much faster and cost-effective way from some liquid samples. They are often combined with microchannels as sample holders and handlers. The application of microchannels implies that only small sample volumes are used (no additional time is needed for cultivation in order to have enough samples). On the other hand, the devices are small sized and portable in contrast to the traditional instruments that are only used in special laboratories by properly trained operators. These biosensors sometimes utilize microelectrodes for mechanical manipulation of the cells in the sample
Keywords
- all-optical
- label-free
- biosensor
- whispering gallery mode (WGM)
- optical waveguide lightmode spectroscopy (OWLS)
- optical waveguide
- photopolymer
1. Introduction
Thinking about biosensors is connected to the detection of pathogens that it makes people recall images from the news where experts in white lab coats are working in serious laboratories with big (and possibly very expensive) machines. They are culturing the microbes from the samples in Petry dishes.
Although this kind of technique still has its place in biosensing, nowadays, it is far from the only way used for decades. We have now the technology of integrated optical sensing that is generally a label-free technique, manifested in portable devices that are working fast, accurately, and cost-effectively.
The technology of integrated optics (IO) is analogous to its name predecessor, the integrated circuits (IC).
As the solid-state switching element, the bipolar transistor was introduced as a small-size alternative to the electronic tubes in the electronic circuits; that was a big leap that led to ICs where all the basic electronic parts (transistor, capacitor, resistor, coil, and wire) a whole circuit of them can be created on a small platform called electronic chip.
The history of IO devices had begun similarly when it became possible to place waveguides, optical filters, and other optical parts (even interferometers) together on the same small-size platform or substrate. This has resulted in the birth of integrated optics. Generally, if there are more than three optical elements on the same platform, it is called an IO device.
Next to their name (integrated), what is similar in the two worlds are the techniques used for the production of the devices. IO devices are fabricated also by applying several types of photolithography, using photoresists and photopolymers. Although not all of the techniques are the same, for example, injection molding has not been applied – yet – in electronics. Thanks to this similarity, the IO devices also can be fabricated in a mass-production manner, resulting in cheap and fast manufacturing.
Very soon, the IO devices have taken place in the sensing technology and in biosensing too. Introducing the label-free technique was a great leap on the path to cheap and fast biosensors.
The traditional way of labeling involves a serious laboratory infrastructure, a lot of time, and cost. Enough sample is needed for detection; therefore, at first, the analyte cells (wanted to be detected) of the original sample have to be selected and multiplied. Next, this ‘big enough’ amount of analyte has to be made ‘visible’ to the instruments in the process called labeling. The analyte (in most cases pathogens) gets ‘decorated’ with some fluorescing or light-absorbing molecules. Both steps are time- and cost-consuming. In the label-free techniques, no such steps are required because the presence of the analytes alone makes optical changes (signal) that can be detected by the optical sensor [1, 2]. When this kind of sensing technology popped up in IO devices, the integrated optical biosensors, a pivotal technology able to provide fast and cheap detection of biomolecular interactions or biological cells, generally from liquid samples.
This chapter introduces principles, mechanisms, and key technologies associated with integrated optical biosensors, the most interesting trends and types of IO biosensors and sensing techniques that we meet nowadays.
2. Fundamentals of integrated optical biosensors
An optical biosensor is a device that uses (generally visible) light to detect biological molecules or substances. These devices typically consist of a sensing element integrated with an optical system that measures changes in light behavior (direction, intensity, phase, polarization, direction, or wavelength) induced by biomolecular- or cell-substrate interactions. Because, in most cases, the sample is liquid, microfluidic channels are parts of the IO biosensors, at least for sample handling. Sometimes, microelectrodes are also present. They build up a highly inhomogeneous electric field in the sample volume. If it is an alternating electric field, then the electrode system is able to manipulate (repel or attract) the analyte particles (e.g., biological cells), applying the phenomena called dielectrophoresis (DEP) [3].
The backbones of the IO devices are the optical waveguides. To be very transparent in the applied wavelength region is a basic and important requirement for the materials of the waveguides. The other is that they should have the highest index of refraction among the surrounding materials. Taking advantage of total internal reflection (that can happen in this case), the light once gets totally reflected inside of them, reminds, and propagates inside. As a side effect of total internal reflection, the so-called evanescent waves appear on the outer surface of these waveguides [4, 5]. These light waves decay exponentially by the distance and the local refractive index at their position can be monitored by them. This way, the presence of some substances (biological cells) deposited or anchored to the surface of the optical waveguide can be detected.
Most frequently, the IO devices are produced by some kind of photolithography that involves the use of photopolymers as building materials (utilizing photolithographic masks) or, more recently, maskless lithography (direct laser writing) solutions. Sometimes soft lithography [6] is used not just for creating the microfluidic channel but also for building optical waveguides or optical elements out of a silicone elastomer, the poly(dimethylsiloxane), PDMS [7, 8].
It is quite a challenge to insert light (coupling in) into the IO waveguides because of the very small sizes (~ microns) of them. For that, in some cases, direct coupling is used when the light comes directly from a very proximate, small-diameter core of a single-mode optical fiber [9] or is focused in (by a microscope objective, most cases). The other commonly used way, mainly for planar waveguides, is the application of optical gratings, embedded in the waveguide.
3. Optical waveguide lightmode spectroscopy
In the planar waveguides, the length and the width of the planar layer are bigger by many orders of magnitude than the wavelength of the light that propagates in, although their thickness is comparable or even lesser of that. Optical waveguide lightmode spectroscopy (OWLS) uses a planar waveguide structure, where the light travels along its length as evanescent waves in the substrate [10]. The incoupling (inserting the light in) is generally managed by an embedded optical grating (in practice, it is a periodic thickness variation) in such a way that the first diffracted order is in the direction of the waveguide. The coupling-out of the light happens in the same way – in many cases – by a different, embedded optical grating. The coupling efficiency is strongly depends on the wavelength and the (variation) of the refractive index of the surrounding media of the integrated optical grating of the planar waveguide [11]. An improved version of OWLS is the resonant wave grating (RWG) technique. It takes advantage of the fact that the coupling efficiency for the incoming and outgoing light strongly depends on the angle of incidence. What is more, this efficiency is also the function of the refractive index and the amount of the (deposited) material on the surface of the optical grating [12]. This adlayer (deposition) is presented as the darker gray layer on the upper part of Figure 1(a).
Figure 1.
The (a) part shows the experimental setup where the optical waveguide with the grating is referred to as a ‘Photonic crystal’ [13]. Light of a halogen lamp arrives from an optical fiber and is first collimated and then directed to the grating. That only inserts a narrow wavelength range into the waveguide, according to the angle of incidence. The same grating couples out the light from the waveguide, and which part of that arrives under α angle is detected. The (b) for pure water shows on a heat map (red to blue represents high to low intensity) the output wavelength as a function of α. The (c) and (d) graphs show the TE1 and TM0 spectra of four different liquids (with different refractive indices) at α = 26°. The latter (e) and (f) graphs show the shift in the output intensity maximum due to the analyte present on the waveguide.
The sensor monitors refractive index changes near its surface (in the evanescent region) by tracking very precisely, on the fly, the shifts in the resonant wavelength of the optical grating since the effective refractive index (
Miniaturized readout systems, capable of converting wavelength shifts into intensity changes, are already available, making it feasible to integrate resonant waveguide grating (RWG) sensors operating near the cut-off point with these systems. This opens the door to developing commercially viable clinical sensors due to their sensitive and adaptable measurement capabilities across a broad sensing range.
4. Whispering gallery mode (WGM) sensors
One of the most advanced and sensitive techniques is the whispering gallery mode (WGM) sensor. This kind of IO sensor exploits the phenomenon where light waves are confined within a circular resonator such as a microsphere, microdisk, or microring [16] by continuous total internal reflection. As light travels along the circumference of the resonator, it creates a standing wave that is highly sensitive to changes in the surrounding environment. Especially near the sensor surface where the evanescent waves of the guided light penetrate. The variation in the outer index of refraction modifies the
Quantum dots are colloidal or epitaxial semiconductor nanocrystals. Being 3D objects inside of them, the electron-hole pairs are confined in all three spatial dimensions. Q-dots are famous for their high quantum yields and resistance to photobleaching (their optical properties only depend on their physical size; there are no dye molecules to be degraded). Mainly the latter property makes them a very common lasing material in WGM microlasers for sensing. When biomolecules bind to the surface of a WGM sensor, they alter the refractive index, and this tiny change is enough to detune the standing wave inside and, by that, the lasing or coupling efficiency of the ring resonator. This makes WGM-based biosensors ideal for low concentrations of analytes. That can be vital for cases when early disease detection and monitoring it is crucial (such a low concentrations for the traditional methods are not sensitive enough) [18]. These sensors can be considered as special case of slab waveguides; the events for sensing happen on their (strongly bended) plain surfaces.
5. Rib waveguide-based IO sensors
We have started the ‘quasy 1D’ devices, the slab (planar) waveguides (most of the optical changes related to sensing happen in one direction, perpendicular to the waveguide plane). The next family of IO biosensors can be considered as ‘quasy 2D’ devices. These optical waveguides generally have rectangular cross section, and they are built on a solid, planar substrate, often referred to as rib-type waveguides [1]. As in the integrated optical devices, in their case, the light waves propagate inside of the waveguide by total internal reflection (the index of refraction for these waveguides is chosen to be the highest in the system). The waveguide has a smooth surface to avoid losses due to light scattering, and its material provides minimal optical loss. Mostly, these waveguides are fabricated by traditional photolithography [1] generally in the following way.
A photolithographic mask defines the waveguide structure in a flat and even layer of a photopolymer. The waveguide is formed in this layer by the exposition through the photomask, and its height is defined by the thickness of the layer. As for exposition, generally, a collimated and spatially homogenized beam of a mercury arch lamp is used (a narrow wavelength band of that). Then the layer gets developed (by soaking it in a special liquid, called developer, that only takes off or dissolves the unexposed photopolymer). This happens at the application of the
Next some microfluidic channel (a sample holder and handler) is added, for the analyte that is (suspended in) a liquid.
For biosensor applications a Mach-Zehnder Interferometer (MZI), made up of these photopolymer waveguides, is a classic setup [20]. The application of this kind of IO interferometer goes back to the beginning of electro-optical telecommunication, where they were (and still are) used as a light-switching part. The IO MZI has an input and an output interface. Further, there are two Y-junctions that splits in to two separated (identical) branches (‘arms’) and rejoin them. The light waves from the input arrive at the first junction, which splits them into two parts with the same intensity and phase. These partial light waves, after traveling over the arms meet as the two arms join in the second Y-junction, which leads to the output. Coherent, monochromatic light (in practice, a laser beam) is coupled in and splits up into two at the first junction. Along the arms, both parts of the light travel the same length, and at the second Y-junction, where they meet, interfere, and the result of that goes to the output. The light intensity at the output – in this ideal case – is the same as it was at the input. For waveguides, there
Figure 2.
Experimental setup of an integrated optical MZI for detecting the Spike Protein of the SARS-CoV-2 virus [22]. The MZI structure (the top view of that is shown) is made of a photopolymer. It is on a glass substrate (1), and it was produced by mask less lithography (direct laser writing). On top of that, there are two microfluidic channels (2) made of PDMS. The upper one is for the sample, and the lower one is for the reference (idled antibody coating). The light intensity at output is detected by a photomultiplier (3,4). A microheater (5) sets the local temperature and is capable of adding a controlled phase shift between the two arms in order to compensate for the temperature drift or the minuscule geometrical difference between the arms. The deposited proteins (anchored by the antibody coat) in the reference arm cause a phase shift in the light waves (6) propagate in that arm, resulting in a detectable variation in the output intensity.
The interferometer structure is made of a photopolymer by direct laser writing (maskless photolithography). The microchannels were created by a technique called soft lithography [6]. The mold used in that process is also made of a photopolymer (SU-8, in this case). None of the integrated optical interferometers has perfect, identical arms, so biasing is needed. The other reason why biasing is beneficial is that the best performance of MZI is at the so-called quadratic points when the output intensity is half of the maximum. For that a microheater was added, which can tune the phase (of the guided light in the reference arm). The right set for reaching the quadratic point can be determined by sweeping the value of the heating current (and power). Such a biasing was performed at the very beginning of the measurement. The microheater is made of a gold layer by lift-off technique [23], utilizing maskless photolithography. Before the measurement, the surface of the waveguides in the arms had been covered with antibodies capable of anchoring only the
6. One-waveguide integrated optical biosensor
If high sensitivity is not required, a much simpler and more robust method can be applied to detect biological cells based on just a single, straight rib waveguide. The optical waveguides generally confine and hold the propagating light waves inside by total internal reflection because they have a higher index of refraction than the surrounding materials at the wavelength to guide. The evanescent waves yet pass the border of the waveguide (they should not exist there according to the classic ray optics) and decay exponentially, and generally disappear in less than a wavelength distance. They interact with anything they meet as normal light waves do. Therefore, when if a biological cell touches the surface of the waveguide it absorbs and scatters the evanescent waves, the latter one can be detected as glowing spots on the waveguide. This is the theoretical basis of a method for sensing bacteria, dispersed in a suspension (Figure 3) [24].
Figure 3.
On the left side (a), the schematic drawing a side view of the waveguide is shown. The gray zone represents the waveguide where the lightwave (in red) is propagated. At the bottom, there is the substrate (blue), and at the top, the liquid (stronger blue with dots) with the bacteria cells is shown. There are two evanescent zones. One of them is in the substrate (pink), and the other is in the suspension (purple). The bacterium cell that immerses in the evanescent waves scatters them and makes them visible even though the bacteria itself are invisible (marked as red flairs). On right (b), the building blocks of the device are presented. The straight waveguide (1) is located between the two golden electrode arrays (2). The liquid sample was fed to the microchannel (3) of a PDMS block that was placed over the waveguide. To increase the sensitivity, the bacteria out of the evanescent zone were collected (pulled down) in a controlled way by means of dielectrophoresis. Each small white spot between the dark electrode tips is the evanescent light that was scattered by a bacterium attached to the surface of the optical waveguide.
For specific sensing, the surface of the waveguide is covered (functionalized) by the antibody that captures only the target bacterium cells. After a washing process, only those cells remind that touch the surface. On the other hand, in this case, the thickness of the evanescent zone was not more than a half wavelength and a very little number of cells was close enough to be detected. To make these ‘visible’ to the system two sets of spike-array, gold film electrodes were added for utilizing the dielectrophoresis (DEP). A very strong, inhomogenous, alternating electric field induces electric dipoles in the particles (polarizing them). Since the full scale electric polarization of a particle (biological cell) needs time, the right frequency (range) of the alternating electric field can be selected, which, with the induced electric dipole is about in phase with that. In this case, the biological cells are attracted by and move toward the strongest electric field (highest gradient). This phenomenon is called positive DEP (pDEP). Consequently, in the opposite case with opposite phase of polarization, the same electrode system repels, and it is called negative DEP (nDEP) [3]. The array of pointed electrodes, next to the optical waveguide, the pDEP collects and holds down the cells in the evanescent zone. In dark background, the evanescent light, scattered by them, makes the cells visible. This way, they can be detected, even by such a microscope, which has a (much) lower resolution than is needed to see these individual cells. Counting the bright spots or measuring the cumulative, scattered light intensity gives information about the number of cells attached that is proportional to their concentration in the whole sample. Therefore, after some calibration, the cell concentration can be determined this way.
7. Interferometric device with a single-mode optical fiber
At the end of this chapter, here comes an example of ‘3D sensing’. The most known integrated optical device is a single-mode optical fiber. The laser beam out of that can be utilized for detecting biological cells in suspension by monitoring the interference of its partial beams. A single-mode optical fiber consists of a cylindrical core with a high refractive index and the cladding that surrounds it with a lower index of refraction. The coupled light waves propagate in the core by means of total refraction (and a part of them as evanescent waves in the cladding). The diameter of the core is small enough (less than 5 microns for visible light) therefore he light only can travel in one trajectory; there is no internal interference, and the output is always a single light cone. If a part of a coherent and monochromatic light (a laser beam) is going on a different way (and path length) than the main part, when they meet, an interference occurs according to the phase difference between them. The following new type of all-optical biosensor utilizes that in order to detect low-concentration biological cells from suspensions (Figure 4) [25].
Figure 4.
At the upper part (a) of this schematic, the device is shown from the side. A coherent and monochromatic laser light emerges from the fiber in a cone shape. A lower part of that hits the glass bottom, which reflects it upwards (darker pink). Later, this reflected part meets the one directly arriving (light pink), and they interfere. If cells are present on the glass surface, they disturb the reflection that results in a variation of the interference pattern that appears on a screen. Applying some image processing, from this change the concentration of cells can be determined. The lower part of the figure (b) shows the two partial beams in 3D representation. The typical, undisturbed interference pattern (with no cells in) can be seen on a screen at the position of the thin black line [25].
The output laser beam of the single-mode optical fiber forms a light cone (with a Gaussian intensity distribution). The optical axis is set to be parallel with the flat surface of the glass base plate. At a given distance, the beam is reflected from. So this way it travels on a broken line route, upward. The other part of the laser light goes in a direct, straightway and meets the reflected light. An interference pattern resulted since the reflected light goes on a longer path, and it has some phase shift relative to the directly arriving light. Due to the geometry, this pattern consists of parallel bright lines separated by dark lines. If the reflection is disturbed by biological cells deposited and anchored to the bottom plate, the interference pattern becomes distorted. By means of image processing, the number of cells causing that distortion can be determined. Since the device just ‘counts’ the cells, it is almost irrelevant what volume they were sedimented from; a very small cell concentration can be detected or determined in that way.
8. Conclusions
As one can see from the limited number of examples, IO biosensors, as manifestations of label-free, all-optical sensing provide various ways to detect biological analytes. This is a very promising and innovative way and evolution of the Integrated Optical devices. It can fill the gaps in the practical application where the conventional ‘hightech’ laboratory cannot be applied as their valuable alternative. Among the wide opportunities they represent already in our days, some cases and particular applications, they already perform better than the conventional solutions.
Acknowledgments
Here, I wish to express my deep gratitude to Dr. András Dér, who has introduced me to the world of Integrated Optics.
References
- 1.
Kozma P et al. Integrated planar optical waveguide interferometer biosensors: A comparative review. Biosensors and Bioelectronics. 2014; 58 :287-307. DOI: 10.1016/j.bios.2014.02.049 - 2.
Mohanty SP, Kougianos E. Biosensors: A tutorial review. IEEE Potentials. 2006; 25 :35-40. DOI: 10.1109/MP.2006.1649009 - 3.
Martinez-Duarte R. Microfabrication technologies in dielectrophoresis applications: A review. Electrophoresis. 2012; 33 :3110-3132. DOI: 10.1002/elps.201200242 - 4.
Estevez M-C et al. Integrated optical devices for lab-on-a-chip biosensing applications. Laser Photonics. 2012; 6 (4):463-487. DOI: 10.1002/lpor.201100025 - 5.
Garland PB. Optical evanescent wave methods for the study of biomolecular interactions: A review. Quarterly Reviews of Biophysics. 1996; 29 (1):91-117. DOI: 10.1017/S0033583500005758 - 6.
Whitesides GM et al. Soft lithography in biology and biochemistry. Annual Review of Biomedical Engineering. 2001; 3 :335-373. DOI: 10.1146/annurev.bioeng.3.1.335 - 7.
Cai DK et al. Optical absorption in transparent PDMS materials applied for multimode waveguides fabrication. Optical Materials. 2008; 30 (7):1157-1161. DOI: 10.1016/j.optmat.2007.05.041 - 8.
Cairone F et al. Micro-optical waveguides realization by low-cost technologies. Micro. 2022; 2 (1):123-136. DOI: 10.3390/micro2010008 - 9.
Dér A et al. Integrated optical switching based on the protein bacteriorhodopsin. Photochemistry and Photobiology. 2007; 83 (2):393-396. DOI: 10.1562/2006-06-21-RA-944 - 10.
Székács A et al. Optical waveguide light-mode spectroscopy immunosensors for environmental monitoring. Applied Optics. 2009; 48 (4):B151-B158. DOI: 10.1364/AO.48.00B151 - 11.
Ormos P et al. Protein-based integrated optical switching and modulation. Applied Physics Letters. 2002; 80 :4060-4062. DOI: 10.1063/1.1481197 - 12.
Fang Y et al. Resonant waveguide grating biosensor for living cell sensing. Biophysical Journal. 2006; 91 (5):1925-1940. DOI: 10.1529/biophysj.105.077818 - 13.
Kovacs B et al. Near cut-off wavelength operation of resonant waveguide grating biosensors. Scientific Reports. 2021; 11 :13091. DOI: 10.1038/s41598-021-92327-4 - 14.
Paulsen M et al. Intensity-based readout of resonant-waveguide grating biosensors: Systems and nanostructures. Photonics and Nanostructures - Fundamentals and Applications. 2017; 26 :69-79. DOI: 10.1016/j.photonics.2017.07.003 - 15.
Farkas E et al. Label-free optical biosensor for real-time monitoring the cytotoxicity of xenobiotics: A proof of principle study on glyphosate. Journal of Hazardous Materials. 2018; 351 :80-89. DOI: 10.1016/j.jhazmat.2018.02.045 - 16.
Cicek K et al. Single-slot hybrid microring resonator hydrogen sensor. Journal of the Optical Society of America B. 2017; 34 (7):1465-1470. DOI: 10.1364/JOSAB.34.001465 - 17.
Toropov N et al. Review of biosensing with whispering-gallery mode lasers. Light: Science & Applications. 2021; 10 :42. DOI: 10.1038/s41377-021-00471-3 - 18.
Bog U et al. On-chip microlasers for biomolecular detection via highly localized deposition of a multifunctional phospholipid ink. Lab on a Chip. 2013; 13 :2701-2707. DOI: 10.1039/C3LC50149C - 19.
Belokopytov GV et al. Optical maskless lithography. Russian MicroElectronics. 2011; 40 (6):414-427. DOI: 10.1134/S1063739711060023 - 20.
Martens D et al. Study on the limit of detection in MZI-based biosensor systems. Scientific Reports. 2019; 9 :5767. DOI: 10.1038/s41598-019-42305-8 - 21.
Mathesz A et al. Integrated optical biosensor for rapid detection of bacteria. Optofluidics Microfluidics and Nanofluidics. 2015; 2 :15-21. DOI: 10.1515/optof-2015-0002 - 22.
Petrovszki D et al. Penetration of the SARS-CoV-2 spike protein across the blood-brain barrier, as revealed by a combination of a human cell culture model system and optical biosensing. Biomedicine. 2022; 10 (1):188. DOI: 10.3390/biomedicines10010188 - 23.
Park J et al. Micropatterning of poly(dimethylsiloxane) using a photoresist lift-off technique for selective electrical insulation of microelectrode arrays. Journal of Micromechanics and Microengineering. 2009; 19 (6):065016. DOI: 10.1088/0960-1317/19/6/065016 - 24.
Petrovszki D et al. An integrated electro-optical biosensor system for rapid, low-cost detection of bacteria. Microelectronic Engineering. 2021; 239-240 :111523. DOI: 10.1016/j.mee.2021.111523 - 25.
Valkai S et al. Optical interferometric device for rapid and specific detection of biological cells. Biosensors. 2024; 14 (9):421. DOI: 10.3390/bios14090421