1. Introduction
Small molecules, typically defined as molecules under 1000 Da, have been of great interest due to their ubiquitous presence in biological processes and their intrinsic properties in human anatomy
[1,2]. Notably, small molecules are capable of crossing the blood-brain barrier and have relative ease in cell membrane permeation compared to larger molecules, thus small molecules can both enter the body readily and traverse to targets with high specificity
[1,2]. These properties have great appeal in the pharmaceutical industry, with 15 of the 37 drugs approved by the FDA in 2022 being small molecules
[3]. On the other hand, the properties that make small molecules appealing for drug development also make them effective toxins, carcinogens, mutagens, and endocrine disruptors both natural, such as several secondary metabolites of fungi, cyanobacteria, plants, and other organisms; and artificial, such as chemical warfare agents
[1,2,4,5,6,7]. Furthermore, biological systems themselves utilize small molecules such as amino acids, steroids, sugars, or metabolites for many of the processes essential for life
[1,2,7].
Considering the biological importance of small molecules, the development of biosensors capable of real-time detection has been in great demand with widespread applications such as for: ensuring correct dosage of drugs, early diagnosis of disease, or continuous detection of toxins. Biosensors are generally composed of a receptor that captures an analyte and a transducer that changes in property upon the binding between the analyte and the receptor to produce a recordable signal
[2]. Unfortunately, due to the low abundance and mass of small molecules, many techniques, especially those with the transducer’s signal strength scales with analyte mass, are challenging to achieve the sensitivity necessary for the effective detection of small molecules under biologically relevant conditions
[7,8]. In addition, label-free techniques are greatly preferred for small molecules, even more so compared to larger analytes like proteins. The low mass and small size of small molecules only exacerbate the concerns of labeled techniques in that a label such as fluorescent protein, fluorescent dyes, and quantum dots can alter the binding behavior and physical properties of an analyte
[9,10]. In many cases, the label is much larger than the analyte itself, for example, a common fluorescent label: green fluorescent proteins, have a molecular weight of 27 kDa to 69 kDa, and even small molecule organic fluorescent probes such as fluorescein has a molecular weight of approximately 332 Da; a very significant addition to a small molecule with mass less than 1000 Da
[9,10,11,12]. Currently, one of the common techniques for molecular detection is enzyme-linked immunosorbent assay (ELISA) and its variations, which offers high accuracy and sensitivity and can be used for small molecule detection, however ELISA is an endpoint detection method that does not provide real time kinetic information
[13] . Small molecule interaction kinetics and affinity with their target or receptor molecules are essential information for studying small molecule functions [
14]. Thus, real-time measurement is preferred and correspondingly transduction methods that can be measured continuously such as changes in refractive index, surface conductance, and resonance frequency have been developed into widespread label-free techniques such as surface plasmon resonance (SPR), field effect transistor (FET), and quartz crystal microbalance (QCM) for real-time small molecule detection using the corresponding receptor or target molecules as recognition elements
[2,15,16,17,18,19,20].
In this review, we explore recent advances in label-free real-time biosensing techniques utilizing optical, electrochemical, and piezoelectric transducers and the innovations or adaptations to meet the growing demands of small molecule detection. Specifically, we will cover works that provide innovative solutions for the challenges of detecting small molecules that can reasonably be applied in biosensors utilized in point-of-care diagnostic, and real-time screening.
2. Optical Transduction
Optical transducers have great potential in the real-time detection of small molecules, but also face intrinsic challenges before being able to fully utilize their advantages. Optical transductions can measure a variety of signals related to the presence of small molecules, such as absorption, scattering, luminescence, or refractive index
[2]. Optical biosensors typically offer fast results with high temporal resolution for real-time monitoring of binding, making them ideal for diagnostics and point-of-care devices
[7]. However, the signal of optical transducers generally scales with the size and/or mass of the analyte of interest, especially when avoiding the use of fluorescent labels
[2,7,21]. For example, one of the commonly used label-free optical detection technique, surface plasmon resonance, has been lauded as the golden standard for molecular binding kinetic measurement, being used for analytes such as proteins, DNA, RNA, peptides, and other biological macromolecules, but attempts to utilize SPR for the detection of small molecules required advanced instrumentation or significant enhancement of receptor surface
[15,22]. Thus, the development of optical techniques for the measurement of small molecules has primarily focused on either enhancing the weak signal of small molecule binding or by measuring binding indirectly via measuring phenomena induced by analyte binding events.
Surface plasmon resonance utilizes surface plasmons, electron oscillations at a metal-dielectric interface, typically gold-coated glass, which respond via oscillation at resonance with a light wave
[23]. The evanescent waves of this oscillation are sensitive to changes close to the metal surface, notably as a change in refractive index due to the binding of an analyte, which shifts the SPR signal and produces a signal proportional to the analyte’s mass
[24]. Unfortunately, this signal dependency on the mass of an analyte makes it challenging for SPR to detect small molecules, often requiring enhancement of either binding site density or signal strength such as through the usage of dextran chips or utilizing localized surface plasmon resonance with nanostructures
[15]. Thus, there has been great interest in developing sensing platforms with high sensitivity to small molecules, requiring simpler sample preparation and instrument operation, and for diagnostic use, having high portability and low cost.
Since the emergence of SPR in the 1970s, surface plasmon based techniques have been developed with great interest and different strategies have emerged to expand techniques into the range of small molecules and overcome their inherent mass dependency [
25]. The use of gold coated optical fibers instead of traditional gold chips for SPR has been in development since the 1990s, offering reduced cost and size compared to traditional SPR facilitating it’s application in point-of-care or diagnostic fields and the ability to optimize sensor attributes to suit the sensor’s purpose by changing the fabrication of the fiber probe [
26,
27]. For example, Liu et al. developed a sensor capable of detecting estradiol, a small molecule drug and hormone, by utilizing fiber grafting to produce high-sensitivity narrow cladding mode spectral combs shown in
Figure 1A, which increased the resolution of the refractive index from 10
-6 to 10
-8 RIU (refractive index unit) and displayed minimal temperature sensitivity by calibrating the light within the fiber core, and reached a limit of detection (LOD) of 1.5 × 10
12 g/mL [
28]. A different approach to the utilization of surface plasmons for detection is exploiting the rapid decay of evanescent waves as distance increases from the sensor surface. Nano-oscillators have been developed in which a receptor is tethered to the surface with a polymer chain and oscillated using an alternating electric field [
29,
30]. Upon binding to the receptor, the charge of the receptor and thus the oscillation amplitude is changed resulting in a change in the oscillation amplitude and the scattering of plasmonic waves that is reflected in the intensity of the plasmonic imaging [
29,
30]. This surface plasmon technique thus circumvents the mass dependency of traditional SPR by instead relying on the charge dependency of the oscillation amplitude. HSV-1 virions tethered to polyethylene glycol linkers and nanodisc encapsulated membrane protein KcsA-Kv1.3 tethered to DNA linkers were able to detect tocrifluor at a detection limit of 16 molecules per virion and 4-(2-ethylpiperidin-1-yl)-2-methyl-6-phenyl-5H-pyrrolo[3,2-d]pyrimidine (EMPPP) at a detection limit of 4.0 × 10
-15 g/mm
2 respectively [
29,
30].
Although SPR is an effective surface sensitive tool with growing in capability for small molecule detection, other evanescent optical sensing principles have been utilized as alternatives. One class of them is grating based optical biosensors, including commercial products, such as Corning Epic or Malvern Creoptix. In general, they have similar performance and cost compared to SPR based systems [
33,
34,
35]. An example of recent development in this class of sensor is guided-mode resonance (GMR), which is also a kind of grating structure-based refractive index sensor [
36]. This reported system us an LED light source and photodetector for reduced cost and improved portability. The small changes of guided-mode resonance wavelength shifts is converted into changes in light intensity, allowing real-time quantification of dinitro-phenyl (DNP) with a LOD of 7.5 ×10
-8 g/mL [
36]. Both SPR and grating based system need special nanofabricated sensor chips that represent considerable cost for consumables. Critical Angle Reflection (CAR) is another recently developed alternative method that measures molecular interaction induced refractive index change via change in reflectivity when the incident light is close to the critical angle, the angle at which the incident light produces an angle of refraction of 90° [
31]. CAR instrumentation is virtually the same as conventional SPR as shown in
Figure 1B and utilizes bare cover glass rather than gold chips, allowing easy implementation where conventional SPR is practiced and at reduced cost
[31]. At high incident angles, CAR surpasses SPR in sensitivity by up to five times, but at the cost of reducing dynamic range and was able to measure small drugs such as furosemide, sulpride, and methyl sulfonamide with a LOD as low as 1.5 × 10
-12 g/mm
2 [
31].
Evanescent techniques have been a great asset in molecular detection with their high sensitivity, but these techniques typically struggle with the detection of small molecules due to their signal strength being mass dependent. Thus, some recent works have been exploring biosensing platforms that detect phenomena induced by small molecule binding to indirectly measure its binding kinetics. One approach is to measure molecular binding induced nanoscale cell membrane deformation on a whole cell assay. Single cells are immobilized on a substrate for the detection of small molecules binding to cell outer membrane receptors, e.g. acetylcholine binding to nicotinic acetylcholine receptors [
37]. Upon binding of the small molecules, the cell membrane is deformed proportionally to the number of ligands bound to receptors. The nanometer scale deformation is detected precisely via a simple differential detection algorithm that calculates the change in relative image intensity within and outside of the cell edge region. Note that cells can be live or fixed, as the deformation is caused directly by the binding events, not downstream events. Another approach utilized an optical fiber suspended over a camera or position-sensitive diode, and is oscillated by an applied alternating current with amplitude proportional to the surface charge of the optical fiber [
38,
39,
40]. Upon binding of analyte onto the surface, the surface charge density of the optical fiber is changed, producing a quantifiable signal that is dependent on the charge of the molecule rather than the mass of the molecule. The method has a LOD of 0.14 e
-/μm
2 for imatinib, a small molecule drug. This charge sensitive optical detection method is an appealing alternative for label-free detection of small molecule binding kinetics [
38,
39,
40].
Although its classification of being label-free is debated, it is worth considering fluorescence-based techniques in which the analyte is not directly labeled. Fluorescence techniques have been highly popular due to their high sensitivity and rapid data analysis. Though many fluorescence techniques rely on labeling analytes with dye molecules, fluorescent techniques that exclude the labeling process can utilize excited auto-fluorescence or analyte-induced quenching to avoid directly modifying analyte behavior
[1,41]. However, these techniques struggle with analytes having high electron affinity when using photo-induced electron transfer for excitation
[42]. A graphitic carbon nitride semiconductor nanosheet layer was used to overcome this issue for the detection of picric acid, a nitroaromatic explosive that can accumulate in water, and shows great promise in utilizing the high sensitivity that fluorescent techniques offer. By exploiting the strong inner filter effect, which is typically considered a hinderance that reduces fluorescence signal, between picric acid and the nanosheets to quench nanosheet fluorescence upon binding of picric acid, an LOD of approximately 1.9 × 10
-9 g/mL was obtained [
42]. Another approach is the use of binding induced cleavage to measure small molecule binding events. A CRISPR-Cas12a derived biosensing platform was developed in which upon small molecule binding onto aTFs proteins bound to double-stranded DNA (dsDNA), the dsDNA is dissociated due to the resulting conformation change of aTF. The free dsDNA can then bind to a Cas12a-crRNA complex resulting in activation of Cas12a. The cleaving of a fluorescence quencher labeled single-stranded DNA (ssDNA) by the activated Cas12a induced a change in fluorescent intensity proportional to the small molecule concentration and binding process [
43]. Although this technique uses a molecular label, the label is applied to an indirectly related ssDNA. This approach allowing use of highly sensitive fluorescence techniques with minimal impact on the binding and properties of the analyte, obtaining a LOD of 1.68 × 10
-9 g/mL for uric acid detection
[43]. Another example of indirect fluorescence signal is the use of fluorescent semiconductor polymer dots that were developed as nanoparticle sensing platforms
[32]. A polymer dot transducer was developed in which an enzyme-catalyzed reaction with glucose would deplete an internal oxygen reservoir resulting in a fluorescent signal visible through skin, allowing in vivo detection of glucose in cells and tissue with an LOD of ~2-8 × 10
-3 g/mL portrayed in
Figure 1C [
32]. This technology can apply to many oxygens consuming enzyme-related small molecules as a means for in vivo detection in patients as an alternative to electrochemical methods that have similar speeds, size, and convenience, but require extracted bodily fluid samples for each measurement [
44].
One unique approach in the development of an optical biosensor for small molecules is the liquid crystal biosensor, which has emerged for small molecule detection recently
[45]. Liquid crystals have long-range orientational order that is sensitive to change at the binding interface allowing label-free high sensitivity detection. However, the optical signal of liquid crystal sensors is generally proportional to the size and number of biomolecules producing topographical changes on glass slides, and thus were not used for small molecules
[45]. By modifying a glass slide using DNAzyme, upon cleaving by L-histidine, a partial substrate sequence is realized and hybridizes with the capture probe producing a DNA duplex on the surface, thus inducing an amplified optical signal compared to the small molecule directly binding to a capture probe with an LOD of 7.8 × 10
-3 g/mL [
45]
.
3. Electrochemical Transduction
Electrochemical transducers share many of the advantages of optical transducers in that they also offer fast, real-time measurements of binding and have the added benefit of generally requiring low-cost instrumentations that can be designed for high portability and accessibility for use with little to no training. Furthermore, the signal of electrochemical transducers is generally not significantly dependent on the size or mass of analytes, but rather on their electronic properties, such as electrochemical reactivity or charge
[16,46,47]. Traditionally, electrochemical biosensors utilized potentiometry or amperometry to detect redox reactions at an electrode surface, offering a mass-independent signal, but this greatly limits the range of molecules that can be detected via electrochemical biosensors to electroactive molecules only. Thus, although real-time electrochemical detection of oxygen and glucose have already reached the commercialization stage for diagnostics and point-of-care monitoring, the detection of other small molecules through electrochemical detection is still developing
[44]. For this reason, innovations in the field have explored other means for analytes to modify the electrical properties of transducers. In particular, impedance-based transducers are promising alternatives that are primarily dependent on the charge of small molecules. Thus, impedance biosensors such as FET sensors have grown in popularity for the detection of small molecules due to their detection mechanism, in which the signal is produced by a change in conductance upon binding of charged molecules, allowing high sensitivity mass independent measurement
[17,18]. Unfortunately, charge-sensitive techniques face challenges from nonspecific binding and Debye screening from the biologically relevant high ionic strength that hinder their sensitivity in serum, plasma, or even phosphate-buffered saline (PBS)
[6,48]. Sensing platforms for small molecules have developed methods to use electrochemical transduction that can detect a larger range of different analytes by enhancing the signal strength of impedance-based biosensors to compensate for charge screening in biologically relevant ionic concentrations. Furthermore, enhancement of signal strength increases the sensitivity of electrochemical instrumentation, and helps to detect the relatively low abundance small molecules typically have in nature [
49,
50].
One of the weaknesses of electrochemical detection, particularly in amperometry or potentiometry is that the analyte of interest must be electrochemically active in order to produce a signal. One innovative approach utilized a customized DNA nanostructure attached at a fixed distance from an electrochemical transducer surface with an attached redox active molecule shown in
Figure 2A [
51]. Upon analyte binding, the change in mass shifts the tethered diffusion between the redox active molecule and the electrochemically active surface. An increased current is resulted from the faster diffusion and reduced distance between the redox active molecule and the transducer surface. An LOD of 9.0 × 10
-7 g/mL and 6.9 ×10
-5 g/mL for biotin and digoxigenin were obtained, respectively [
51]. Similarly, an aptamer with a DNA triplex assembled on nanotetrahedron on screen-printed electrodes was utilized for the detection of saxitoxin and obtained an LOD of 2.8 × 10
-10 g/mL [
52]. Upon small molecule binding, the switch of aptamer triplex lead to dissociation of the pyrimidine arms from the electrode and induce an increase in current that measured with square wave voltammetry and CV [
52].
Field Effect Transistors have been popular candidates for small molecule detection because they are charge-sensitive rather than mass-sensitive. However, they face challenges in biologically relevant conditions due to reduced sensitivity in high ionic concentration, which shield the charge of analyte molecules [
17]. Strategies have been developed to enhance the detection of small molecules, particularly those with weak surface charge. An aptameric graphene field effect transistor sensor was developed by Wang et al. for the detection of small molecules [
53]. Analyte specific aptamers were hybridized onto the sensor surface. Analyte binding release the aptamer anchor and induce a large charge change due to the highly charged nature of aptamers as shown in
Figure 2B. A strong signal can be obtained in this way with an LOD of 1.65 × 10
-8 g/mL [
53]. Another approach sought to overcome the charge screening limitation of FET by utilizing oligonucleotide receptors with adaptive loop binding on nanometer-thin In
2O
3 to detect dopamine, serotonin, glucose, and sphingosine-1-phosphate [
55]. An electromechanical detector was developed using aptamer probes bound to a ssDNA cantilever attached to a double-stranded DNA tetrahedral structure [
56]. An alternating electric field induces the cantilever to raise and lower, which can be monitored after attachment of a fluorescent dye cyanine3 at the tip of the cantilever and further measured in parallel using a graphene layer and a field effect transistor to detect ATP [
56]. Although a fluorescent dye is utilized, it labels the cantilever rather than the analyte of interest [
56].
Another impedimetric technique, electrochemical impedance spectroscopy (EIS), measures the charge transfer resistance of an electrode surface upon binding. EIS has recently gained the capability to detect small molecules using modified sensing layers with enhanced electronic properties with high selectivity and sensitivity for change in impedance such as with aptamers or molecularly imprinted polymers (MIP) [
48]. Hui Lee et al. developed nanoscale molecularly imprinted composite polymer with a thickness of less than 5 nm as an alternative to traditional thicker receptors, offering synthesized thinner layers with higher target to receptor ratios, and resulting in increased sensitivity with an LOD of 5.76 × 10
-10 g/mL for cortisol [
57]. A quantum electrochemical impedance spectroscopy technique was used to measure resonant quantum conductance changed induced by cortisol binding, because the change in charge-transfer resistance was insufficient for traditional electrochemical impedance spectroscopy to detect, due to the low resistance of the thin MIP film [
57]. Another approach, metal oxide nanofibers have grown in popularity as biosensing material due to their high sensitivity and electronic properties. A Mn
2O
3 nanofiber was developed by Supraja et al. as a low bandgap working electrode for the EIS detection of atrazine with full setup in
Figure 2C and a proposed LOD of 2.2 × 10
-22 g/mL [
54]. A high throughput impedimetric sensing platform was developed by using peptide aptamers screened via molecular docking in silica to determine the potential interaction between L-arginine and the extracted peptide aptamer for specificity and binding energy and obtained an LOD of 1.92 × 10
-16 g/mL [
58]. This represents a potential model for future electrochemical impedance spectroscopy-based sensor design workflow as different molecular docking groups can be quickly screened in silica before validation by a well-established experimental method such as isothermal titration calorimetry and then be easily incorporated into an assay for small molecules [
58].
One of the most common targets for small molecule electrochemical detectors is glucose, an important molecule both as a source of energy and a biomarker of diabetes. The detection of glucose has advanced greatly over the years due to the high demand of patients that regularly need to check blood sugar levels [
44]. Thus, miniaturization and accessibility are particularly high priorities for glucose detection, both features that electrochemical transducers excel at offering. Radiofrequency-based biosensors have been an interesting alternative impedimetric technique that measures glucose concentrations by measuring the level of electromagnetic coupling dependent on glucose permittivity [
59]. A miniaturized RF resonator biosensor was recently developed using an interdigital capacitor embedded between two divisions of a spiral-inductor allowing high sensitivity of glucose detection both in water and in serum with LOD of 1.1 × 10
-8 g/mL and 5.9 ×10
-9 g/mL respectively [
59].
As previously mentioned, small molecules have low mass and size, but another common feature of small molecules is their relatively low abundance in their native environments [
49,
50,
60]. Most studies that hope to compensate or overcome this challenge by developing surfaces that can bind to target analyte more readily and selectively than previously discovered materials. However, Cui et al. developed an interesting approach to improve sensitivity to low concentrations of small molecules by increase the transport rate of the analyte to the sensor surface [
61]. Most biosensors rely on passive diffusion for analytes to reach receptors, but this group used directed particle motion toward sensor electrodes driven by AC dielectrophoresis [
61]. An AC capacitive affinity sensor was developed in which low voltage AC dielectrophoresis was used to carry analyte to the sensor via microfluidic movements that are independent of range and size of the analyte [
61].
4. Piezoelectric Transduction
In comparison to optical and electrochemical transducers, piezoelectric transducers are a relatively recent addition to the repertoire of techniques for detecting small molecules, most popularly utilizing quartz crystal microbalances (QCM)
[19]. Piezoelectricity, the phenomenon in which a material produces voltage under mechanical stress or vice versa, allows for the fabrication of sensors that utilize anisotropic crystals that oscillate upon the application of voltage
[19,62]. Piezoelectric biosensors typically measure change in oscillation due to analyte binding for the measurement of analyte properties and kinetic information
[62]. For example, in a conventional quartz crystal microbalance, the added mass upon binding increases the damping of the oscillation and a change in dissipation rate upon ceasing of voltage application that is related to the mass of the bound analyte following the Sauerbrey Equation
[20,63]. Unfortunately, Piezoelectric transducers find difficulty in the measurement of small molecule binding due to the mass dependency of the oscillation’s frequency change and though piezoelectric transducers are resistant to interference from non-transparent mediums compared to optical transducers, they are responsive to changes in viscosity
[20]. Nevertheless, piezoelectric biosensors can be versatile and robust methods for small molecule detection and much progress has been made in enhancing the sensitivity of piezoelectric-based techniques. Furthermore, most relevant biosensing conditions require the sensor to be in liquid, which produces an additive damping to the measured frequency. Thus, the development of piezoelectric biosensors for small molecule detection has been in amplifying the change of frequency upon binding or utilizing alternative means to collect data from the piezoelectric transducer.
Quartz crystal microbalance is accepted as the most popular piezoelectric detection technique, however, though it displays great sensitivity, it struggles to have sufficient sensitivity for the detection of small molecules due to its signal dependency on mass and is prone to nonspecific binding. An indirect competitive strategy utilized gold nanoparticles conjugated with a secondary antibody that can then be bound to the captured primary antibodies on the sensor surface after competition between the antigen on the surface and the analyte shown in
Figure 3A [
64]. The additional mass of the gold nanoparticle amplifies the change in frequency and dissipation that is produced from binding on the surface allowing increased Ochratoxin detection sensitivity withs a LOD of 4 × 10
-11 g/mL in PBS and 1.6 × 10
-10 g/mL in red wine [
64]. Another approach to compensate for the mass dependence of QCM is in the development of an electromagnetic piezoelectric acoustic sensor [
65]. An ultrahigh frequency piezoelectric aptasensor was developed based on QCM that utilized excitation of an acoustic resonance with a piezoelectric quartz substrate through an external magnetic field induced by a spiral coil, which allows operation at high frequencies without metal contacts [
65]. It is also one of the first uses of an aptamer immobilized on an organic adlayer for the detection of small molecules using an acoustic wave sensor, obtaining an LOD of 2.7 × 10
-7 g/mL [
65]. The increased frequency decreases the penetration depth of the sensor allowing it to have higher sensitivity to binding events that occur at the biosensor surface [
59]. Koutsoumpeli et al. utilized affimirs, an antibody mimetic, as a high stability, low cost alternative receptor surface on a self-assembled monolayer of long-chained alkanethiols with an oligoethylene glycol component [
66]. The resulting monolayer provided resistance to nonspecific binding and was able to detect methylene blue with micromolar limit of detection despite not yet being optimized [
66].
Another popular acoustic wave sensor is the surface acoustic wave (SAW) sensor, which utilizes similar technology, but the wave propagates only at the guiding layer of the substrate surface instead of the entire substrate allowing the use of higher frequencies and increasing sensitivity at the surface [
67]. The most sensitive SAW sensors are Love-wave sensors that protect the interdigital transducer by the waveguide layer from harsh liquid environments [
67]. Compared to QCM, thin film materials have not been widely explored, thus Sayago et al. investigated graphene oxide layers as an alternative to traditional gold film for the detection of chemical warfare agent (CWA) small molecules with LOD as low as 2 × 10
-7 g/mL [
8]. Similarly, a graphene oxide layer was produced with carbon vapor deposition, which is able to distinguish endotoxin from aflatoxin with aptamer receptors on a shear horizontal surface acoustic wave sensor portrayed in
Figure 3B, achieving a LOD of 2.53 × 10
-9 g/mL [
68].
Figure 3.
A) Schematic of competitive immunoassay combined with QCM Reprinted from [
64], Copyright (2017), with permission from Elsevier. B) Surface functionalization for shear horizontal SAW with graphene film for detection of endotoxin [
68].
Figure 3.
A) Schematic of competitive immunoassay combined with QCM Reprinted from [
64], Copyright (2017), with permission from Elsevier. B) Surface functionalization for shear horizontal SAW with graphene film for detection of endotoxin [
68].
Another piezoelectric-based transducer being explored is complementary metal oxide semiconductors that utilize a piezo-resistant membrane bridge sensor [
69]. A microcantilever-based biosensor with a 2D array of suspended thin film and bridge structure offered higher sensitivity than traditional microcantilever biosensors, measuring phenytoin with an LOD of 4.06 × 10
-6 g/mL [
69]. Compared to SPR, the membrane bridge has increased reaction area and sensitivity, high stability due to increased stiffness, and its relatively small size makes it more compatible for point-of-care devices [
69].