Home Journals Progress in Chemistry
Progress in Chemistry

Abbreviation (ISO4): Prog Chem      Editor in chief: Jincai ZHAO

About  /  Aim & scope  /  Editorial board  /  Indexed  /  Contact  / 
Review

Construction and Application in Food Contaminants Detection of Novel Optical Fiber Biosensors

  • Jialin Huang 1 ,
  • Yaohua Qin 2 ,
  • Sheng Tang , 1, * ,
  • Dezhao Kong 2 ,
  • Chang Liu , 2, *
Expand
  • 1 School of Environmental and Chemical Engineering, Jiangsu University of Science and Technology, Zhenjiang 212003, China
  • 2 School of Grain Science and Technology, Jiangsu University of Science and Technology, Zhenjiang 212003, China
* e-mail: (Sheng Tang);

Received date: 2023-04-24

  Revised date: 2023-08-30

  Online published: 2023-12-20

Supported by

National Natural Science Foundation of China(22276080)

Shuangchuang Ph.D Award of Jiangsu Province(1184902001)

Abstract

Food safety is closely related to people’s quality of life. The establishment of simple, sensitive and intelligent detection methods for food contaminants is an important guarantee for food safety and health management. Nevertheless, traditional analysis methods have certain limitations such as time-consuming detection process, high cost, and complicated operation. Optical fiber biosensors, which rely on the interaction between light and fluids, have the characteristics of good signal sensitivity, rapid detection and real-time response. They have recently emerged as advanced optical sensing methods with diverse functions and high sensitivity, and can realize rapid and accurate detection of various pollutants in food. In this review, we summarized the basic principles, classification and research status of various novel optical fiber biosensors. The applications in the detection of various pollutants such as mycotoxins, heavy metal ions, antibiotics, and pesticide residues in food samples were introduced. Furthermore, the development trend of this novel sensing strategy was also briefly discussed.

Contents

1 Introduction

2 Optical fiber biosensor

2.1 Composition of optical fiber biosensor

2.2 Basic principle of optical fiber biosensor

2.3 Classification of optical fiber biosensor

3 Application of optical fiber biosensors in detection of food contaminants

3.1 Mycotoxin

3.2 Heavy metal ion

3.3 Antibiotic

3.4 Pesticide residue

3.5 Pathogen

4 Conclusion and prospect

Cite this article

Jialin Huang , Yaohua Qin , Sheng Tang , Dezhao Kong , Chang Liu . Construction and Application in Food Contaminants Detection of Novel Optical Fiber Biosensors[J]. Progress in Chemistry, 2024 , 36(1) : 120 -131 . DOI: 10.7536/PC230423

1 Introduction

Food safety is closely related to human health and people's quality of life, and has become one of the major issues facing the world. Mycotoxins, antibiotics, heavy metal ions, pesticide residues and pathogens are the main sources of food contamination. These pollutants accumulate excessively in the human body through food intake, which may lead to serious toxic reactions[1]. In the past few decades, a variety of technologies have been developed for food safety quality testing, such as enzyme-linked immunosorbent assay (ELISA), high performance liquid chromatography-mass spectrometry (HPLC-MS), polymerase chain reaction (PCR), gas chromatography (GC) and so on[2][3][4][5]. These traditional methods have high sensitivity and specificity, but they also have some limitations, such as high cost of instruments, fixed test environment, complex sample processing steps, and long test cycle. Therefore, it is still necessary and urgent to establish and optimize the methods for accurate and rapid detection of pollutant residues in complex food matrices, which has become an important research direction in the field of food detection.
In recent years, with the rapid development of nanotechnology, optical sensors combined with spectroscopy have shown great potential in many fields. Among them, fiber-optic biosensors based on the interaction between light and fluid have the advantages of high sensitivity, fast detection and real-time response, and have been widely used in biomedicine, food processing, environmental safety detection and other fields[6]. The optical fiber biosensor uses the change of local dielectric environment caused by the combination of the sensing surface and the sample to be detected as an output signal, can realize the high-sensitivity quantitative detection of a trace target substance only by a small amount of sample, has the advantages of rapid response and low detection cost, and is suitable for monitoring various food pollutants[7]. In addition, as the transmission carrier of optical signal, optical fiber also has many advantages, such as small size, anti-interference, low cost, strong stability and so on. In this paper, the composition and classification of new optical fiber biosensors and their applications in food safety detection in recent years will be summarized, and the applicability and development prospects of current optical fiber biosensing methods will be briefly described.

2 Optical fiber biosensor

2.1 Composition of optical fiber biosensor

Optical fiber biosensor is a kind of optical sensor, which has the characteristics of small size, high sensitivity, real-time online monitoring and so on. At the same time, as a sensing platform, the optical fiber itself has the properties of anti-electromagnetic interference, corrosion resistance, water resistance, high temperature resistance and so on. The optical fiber sensor system is mainly composed of a light source, an optical fiber, and a spectrometer, as shown in Figure 1A. Common light sources include tungsten halogen lamps, light-emitting diodes and pulsed xenon light sources, which cover a wide range of wavelengths, generally between hundreds and more than 2000 nanometers. There are many kinds of optical fibers, and their composition materials and structures are different according to their different uses. For example, it can be divided into quartz optical fiber, plastic optical fiber and composite optical fiber according to the different raw materials. According to the transmission mode, it can be divided into single-mode fiber and multimode fiber[8]. Optical fibers are generally divided into three layers: a core with a high refractive index at the center, a cladding with a low refractive index in the middle, and a protective coating at the outermost layer. Since light is transmitted in the fiber by total internal reflection (TIR), this requires that the refractive index of the core must be greater than that of the cladding[9]. Therefore, in terms of material selection, the core material is generally high-purity silicon dioxide, and the cladding material is mostly fluorine-doped silicon dioxide with a refractive index lower than that of the core. According to the different amount of fluorine doping, the refractive index distribution of the core and cladding can be slightly different. The outermost coating layer mainly plays a protective role to ensure the overall firmness and applicability of the optical fiber.
图1 (a) 光纤传感器的基本结构;(b) 不同结构的光纤[20]

Fig. 1 (a) Basic structure of optical fiber sensor; (b) different structures of optical fiber[20]. Copyright 2021, Multidisciplinary Digital Publishing Institute

The sensitive response of the optical path inside the fiber core to the external environment is the basis for rapid detection of the target. In order to improve the interaction between light and fluid, a variety of optical fibers with different geometric modifications have been developed, as shown in Figure 1b. The common ones are U-bend optical fiber, D-type optical fiber, tapered optical fiber and end reflection optical fiber[10]. U-structure fiber can be obtained by bending the fiber with a fixed bending diameter and burning it with a flame, which has been proved to significantly increase the penetration depth of the evanescent field in the bending region and outside the direction of light propagation, thus improving the sensitivity of the fiber[11,12][13]. D-type fiber is generally treated by single-sided mechanical polishing, which only removes part of the cladding on one side of the fiber, and then further modifies the exposed flat side[14,15]. The diameter of a geometrically modified fiber with a tapered structure is usually only a few microns, and the diameter of the fiber can be reduced by locally heating the fiber with a heat source and stretching the glass in both directions along the axis in a molten state[16,17]. By adjusting the taper ratio, the taper length and the incident light angle, the evanescent wave can be fully exposed to the external medium, the penetration depth can be increased, and the sensitivity of the sensor can be improved[18]. Light is transmitted unidirectionally in the above optical fibers with various structures, while the optical fiber based on end reflection relies on the reflection of light at the cut end of the optical fiber to obtain the signal. The evanescent wave sensor model is formed by modifying the polished end face to realize the detection of the target analyte[19]. For different practical application scenarios, the choice of optical fiber is a key point to be concerned.
Optical fiber spectrometer is an important scientific instrument which uses optical fiber as a signal coupling device to analyze the spectrum of the substance to be measured. Because the incident light is in the whole band, the signal measurement is usually carried out by intensity detection or wavelength detection. The core configuration of a fiber spectrometer typically includes a grating, a slit, and a detector. The choice of grating depends on the spectral range as well as the resolution requirements. At present, most of the fiber spectrometers on the market can be configured with gratings of various specifications to achieve a wide spectral range. The height design of the slit has an important impact on the sampling sensitivity. A narrower slit can improve the resolution, but will lead to a reduction in the luminous flux. A wider slit can increase the sensitivity, but will sacrifice the resolution. With the development of technology, the slit height can be customized to meet different application scenarios. At present, the types of detectors can be roughly divided into CCD detectors, CMOS detectors and InGaAs detectors. For different types of optical fibers and the spectral range to be measured, the selection of the detector is also very important. The CCD detector is suitable for both the ultraviolet range and the near-infrared range; The CMOS detector has the advantages of low delay, fast response, high cost performance and high sensitivity in the ultraviolet band; InGaAs detector has very high sensitivity in the near infrared region.

2.2 Basic Principle of Fiber Optic Biosensor

In an optical fiber, when the incident light strikes the interface between the core and the cladding, it will produce an electromagnetic wave that is exponentially attenuated by about one wavelength depth into the cladding, which is called the evanescent wave. The distance of the evanescent wave beyond the core-cladding interface is called the penetration depth. In the presence of a target in actual detection, the evanescent wave can be absorbed by molecules or atoms constituting the target within the effective penetration distance of the evanescent wave, and the concentration of the target is quantified according to the degree of absorption, so as to realize the accurate monitoring of the target[20]. At present, the common evanescent wave optical fiber sensors are usually divided into surface plasmon resonance (SPR)/localized surface plasmon resonance (LSPR) sensors based on noble metal materials and optical fiber sensors based on fluorescence resonance energy transfer.
The principle of SPR optical fiber sensor is similar to that of LSPR optical fiber sensor, but it is different because of the different structure of the surface modified metal material. SPR optical fiber sensor usually deposits a certain thickness of gold, silver and other metal films on the optical fiber core, the evanescent wave generated at the interface between the fiber core and the metal interacts with the free electrons on the surface of the metal dielectric to excite the surface plasmon resonance, and the energy transfer occurs due to the mismatch between the evanescent wave and the surface plasmon, resulting in the resonance absorption peak[21,22][23]. LSPR optical fiber sensor is usually modified with metal nanomaterials such as gold and silver on the core of the optical fiber. When the frequency of the evanescent wave matches the vibration frequency of the noble metal nanoparticles, the nanoparticles will have a strong absorption of photon energy, resulting in an enhanced local electromagnetic field and a more sensitive response to the target molecules[24,25]. When the analyte interacts with the recognition site on the surface of the fiber core, the refractive index and the morphology of the nanoparticles in the local area change, which changes the position of the SPR or LSPR absorption peak, so that quantitative detection can be achieved according to the relationship between the deviation degree and the concentration of the target.
The principle of the fluorescence resonance energy transfer sensor is that a molecule modified with a fluorophore as a sensing element is fixed on a fiber core in advance, and a corresponding complementary sequence labeled with a quencher is combined with the sensing element,The fluorescence intensity is maintained at a low level, and when the target molecule is close to the complementary sequence, the fluorescence intensity is restored through the competitive binding relationship, and the quantitative detection is carried out according to the relationship between the concentration of the target substance and the fluorescence intensity[27].

2.3 Classification of Fiber Optic Biosensor

In recent years, with the innovation and development of detection methodology, the combination of biosensing strategy and optical fiber sensing technology to construct a new type of optical fiber biosensor can significantly improve the detection sensitivity and practicability of optical fiber sensing method. The interaction between the fiber core and the external modification is the basis of quantitative detection. The construction of different types of fiber optic biosensors mainly includes two processes: fiber pretreatment and core surface sensing strategy design. The pretreatment of the optical fiber is mainly to remove the cladding by physical or chemical methods, expose part of the fiber core as the sensing area, and activate it. And then modifying various specific recognition elements on the surface of the fiber core according to the type and application of the optical fiber biosensor to construct a sensing area. According to the different sensing elements and application scenarios, the types of optical fiber biosensors can be simply divided.

2.3.1 Enzyme-based optical fiber biosensor

Enzymes are highly selective and sensitive, and they act quite rapidly compared to other biological receptors and can be used in combination with different transduction mechanisms[28]. The coupling of the analyte enzyme and the target analyte is used to trigger the change of the optical fiber sensing signal.
In the process of constructing enzyme-based optical fiber biosensors, the enzyme is usually immobilized on the optical fiber or the nanomaterial loaded on the surface of the optical fiber by covalent coupling. Semwal et al. Developed an enzyme biosensor for detecting hydrogen peroxide, as shown in Figure 2[29]. First, a layer of gold film was coated on the core without cladding, and then a layer of graphene oxide (GO) was coated, and then catalase was immobilized on GO. The surface of GO is usually modified with carboxyl, epoxy, hydroxyl and other functional groups, which can achieve covalent binding with biomolecules. GO interacts with dissociated water molecules after the enzymatic hydrolysis of hydrogen peroxide by catalase, which changes the effective refractive index of the core surface, thus achieving quantitative analysis. At the same time, GO can also provide a larger surface area and increase the loading capacity of the enzyme, thereby enhancing the sensitivity and sensing mechanism. Nag et al. Immobilized lactamase on the surface of optical fiber modified with polyaniline nanofiber coating by covalent crosslinking of glutaraldehyde for the detection of ceftazidime[11]. Different from the common covalent coupling method, Kant et al. Constructed an enzyme biosensor to detect sorbitol by modifying tantalum pentoxide nanoflowers embedded with sorbitol dehydrogenase on a silver-coated fiber core[23]. The flower-like nanomaterials have high specific surface area and good stability in enzyme entrapment biosensing applications. When the probe is placed in different concentrations of D-sorbitol solution, the resonance wavelength will be blue-shifted with the increase of D-sorbitol concentration around the sensing area. Wang et al. Modified tyrosinase-functionalized nanoparticles on a tapered optical fiber to achieve specific detection of p-cresol[30].
图2 光纤传感设备检测过氧化氢原理图[29]

Fig. 2 Schematic diagram of an optical fiber sensing device for detecting H2O2[29]. Copyright 2021, Elsevier

Enzyme-based sensing detection methods usually have high specificity and sensitivity due to the high specificity and catalytic activity of enzymes, but the activity of enzymes is easily affected by external environmental conditions such as temperature and pH, which has great limitations in the detection of actual samples. By combining the enzyme with optical fiber sensing, the enzyme is immobilized on the optical fiber carrier by common chemical covalent coupling and other methods, which not only simplifies the analysis operation process, but also helps to achieve higher acid-base tolerance, thermal stability, storage stability and long-term reusability of the immobilized biomaterial. Enzyme-based fiber optic biosensor has a wide range of applications in the detection of actual samples.

2.3.2 Immuno-optical fiber biosensor

Immunological fiber-optic biosensor is a very important kind of fiber-optic biosensor, whose principle is mainly based on the specific binding of antibody and antigen. This binding changes the intensity of the reflection and can yield specific information about the analyte concentration based on the extent of the intensity change. Similar to enzymes, antibodies are usually immobilized on the surface of the carrier by condensation reaction using chemical crosslinkers such as 11-mercaptoundecanoic acid (11-MUA) as a representative in the construction of immunofiber biosensors. Chaudhari et al. Developed an evanescent wave fiber optic biosensor for ampicillin detection with a detection limit as low as 7.4×10−10g/mL by immobilizing ampicillin antibody on the surface of an optical fiber[31]. Kim et al. Established a simple fiber-optic sensor for direct and rapid measurement of thyroglobulin (Tg) by immobilizing thyroglobulin antibody on the fiber-optic section[32]. Jing et al. designed a double-antibody sandwich immuno-optical fiber biosensor, as shown in Figure 3, by adding a lossless matching layer with a refractive index lower than that of the fiber core between the fiber core substrate and the metal layer as a buffer layer, so as to reduce the attenuation loss of the electromagnetic field at a long distance[33]. Then gold nanospheres are modified on the surface of the metal layer, and the electronic coupling between the metal layer and the gold nanospheres further enhances the electric field intensity, so that the sensor has higher sensitivity; and then a self-polymerization dopamine layer is formed on the gold nanospheres, and the antibody is fully fixed by covalent binding of quinone groups in the polydopamine and amino groups in the antibody. The detection limit of the sensor for human immunoglobulin G (IgG) was as low as 0.11 μg/mL. Kim et al. Designed a nano-dome array optical fiber sensor and modified the gold surface with an antibody to detect thyroglobulin, forming a dome array with a uniform nano-gap on the end face of the optical fiber. The "hot spot" of the particle spacing in the array enlarges the sensing area and greatly enhances the intensity and detection sensitivity of the electromagnetic field, with a detection limit of 38 fg/mL[34].
图3 双抗体夹心免疫传感器检测人免疫球蛋白G示意图[33]

Fig. 3 Schematic diagram of double antibody sandwich immunosensor for detecting human-immunoglobulin G[33]. Copyright 2021, Multidisciplinary Digital Publishing Institute

The key to the construction of immuno-fiber-optic biosensor is the selection and modification of the antibody on the surface of the fiber. Polyclonal antibodies can bind to a variety of epitope antigens, so their specificity is relatively low, and they are more likely to produce cross-reactions, which limits their application in the detection of a variety of target analytes, and the antibodies produced in different batches also have batch differences, resulting in relatively low repeatability. Monoclonal antibodies can recognize a single specific antigen and have high specificity, but they also have the limitations of high cost and low thermal stability. How to select the appropriate antibody and stably modify it on the optical fiber under the premise of low cost and easy detection is one of the problems that need to be solved in the current immune optical fiber biosensor.

2.3.3 Nucleic acid optical fiber biosensor

Nucleic acid fiber optic biosensor is a new type of optical fiber sensing method based on the reaction between nucleic acid strands for signal recognition and amplification. Thanks to the high specificity and flexible design of nucleic acid strands, signals can be significantly amplified based on different signal recognition technologies and nucleic acid amplification strategies in practical detection scenarios.
There are many methods to construct nucleic acid fiber optic biosensors, such as modifying biotin on the target nucleic acid strand, which can stably combine with the fiber modified with streptavidin after silylation to form a biotin-streptavidin complex, thus realizing the specific recognition of the target molecule. Yang et al. Reported a CRISPR-based fiber-optic evanescent fluorescence biosensor for detecting novel coronavirus (SARS-CoV-2), as shown in Fig. 4[35]. The Cas13a protein interacts with CRISPR RNA (crRNA) after specific binding, and guides Cas13a to recognize and bind the target viral RNA to form an RNA duplex according to the sequence of the crRNA, and the formation of the duplex double helix structure induces the conformational change of the Cas13a protein,It activates the catalytic site and enters the enzymatic "active" state, where the activated Cas13a protein molecule can cleave any exposed RNA molecule, including the free hairpin beacon molecule in solution, and the part released by the cleaved beacon molecule triggers the downstream HCR amplification process. The biotin-labeled HCR product was captured by the streptavidin-functionalized optical fiber, and the labeled Cy5.5 fluorophore on the product was excited by the evanescent wave of the optical fiber, and the detection of viral RNA was realized by collecting the fluorescent signal.
图4 无酶扩增偶联CRISPR-Cas13a检测SARS-CoV-2[35]

Fig. 4 SARS-CoV-2 detection with enzyme-free amplification coupled CRISPR-Cas13a[35]. Copyright 2021, Elsevier

In addition, different kinds of groups can be modified on the nucleic acid chain, and then combined with the fiber core or the noble metal nanomaterials on the surface of the fiber core. Janik et al. Designed an optical fiber sensor to monitor the isothermal amplification of DNA on the surface of an optical fiber in real time[36]. The DNA primer modified with DCBO group was combined with the fiber treated by 3- (azidopropyl) triethoxysilane, and the circular DNA designed based on H5N1 influenza virus was used as the target circular DNA to initiate the amplification reaction under the action of the primer DNA and Phi29 DNA polymerase, and finally the double-stranded DNA (dsDNA) product was formed. The biosensor has an ultra-high refractive index sensitivity and requires only a very small amount of sample to detect the signal. Tseng et al. Developed a nucleic acid fiber optic biosensor based on gold nanoparticles. The ssDNA modified with -SH formed an Au-S covalent bond with the gold nanoparticles, and then the ssDNA was specifically bound to the target DNA and RNA oligonucleotides. They successfully explored the effects of ssDNA with different oligonucleotide lengths and directions on the signal, and achieved a sub-nanometer detection limit[37].
The versatility of nucleic acid fiber optic biosensors also benefits from the flexible design of nucleic acid strands. Ngo et al. Proposed a sensing system for detecting β-thalassemia gene mutation by binding of gold nanoparticles to bacterial MutS protein on optical fiber[38]. In this method, MutS binds to gold nanoparticles on the surface of the fiber core to serve as a sensing region for the detection of monobasic mismatched DNA. MutS protein can capture the target double-stranded DNA containing A and C mismatched base pairs in the sample. At this time, the refractive index of the surface of gold nanoparticles increases and the intensity of transmitted light through the optical fiber decreases, thus achieving the detection of AC heteroduplex DNA solutions with different concentrations, with a detection limit as low as 0.49 nM.
Nucleic acid fiber-optic biosensors, represented by aptamer single-stranded nucleic acid, have developed rapidly in recent years. Aptamers can undergo structural changes in a variety of ways to bind to specific analytes with high affinity and selectivity. Compared with the immunoassay sensor, it has the advantages of low cost, simple synthesis and modification process, strong stability and easy long-term storage.

2.3.4 Whole-cell fiber-optic biosensor

Using cells as receptors to construct fiber-optic biosensors is a hot research topic in recent years. Generally speaking, cells immobilized on gold films or gold nanomaterials, when combined with external analytes, cause changes in the refractive index of the optical fiber surface, cell morphology or intracellular changes, thus changing the signal.
Because bacterial cells are mostly negatively charged, it is a common method to construct whole-cell fiber-optic biosensors by depositing a polyelectrolyte coating with positive charges on the surface of the sensing carrier and binding bacteria through electrostatic adsorption. Halkare et al. Developed an optical fiber biosensor using Escherichia coli B40 as a receptor to detect mercury and chromium ions in water by immobilizing bacteria on an optical fiber probe coated with gold-coated nanoparticles, as shown in Figure 5[39]. Two layers of poly (sodium 4-styrenesulfonate) and poly (allylamine hydrochloride) were deposited on the sensor probe modified with gold nanoparticles, respectively, and the probe was positively charged and could electrostatically bind to E. coli B40. When the probe was exposed to different concentrations of mercury and chromium ions, the heavy metal ions interacted with thiols and other groups on the cell surface to form coordination covalent bonds, which changed the local refractive index and changed the LSPR signal, with a detection limit of 0. 5 ppb. Futra et al. Immobilized recombinant green fluorescent protein Escherichia coli on an optical fiber and covered it with a layer of calcium alginate film, and developed a fluorescence-based optical fiber biosensor for the detection of heavy metal ions[40]. The reaction of metal ions with the sulfhydryl functional group of fluorescent protein can inhibit the metabolism of E. coli, and the fluorescence intensity of E. coli will decrease significantly. The method has high detection sensitivity for copper ions, cadmium ions, lead ions and the like, and has the disadvantage that the detection limit for silver ions and iron ions is relatively low. Zajic et al. Immobilized Pseudomonas putida TVA8 on a positively charged tapered fiber treated with polyethyleneimine for the detection of the aromatic contaminant toluene[41]. Pseudomonas putida TVA8 can bioluminesce in the presence of benzene series, and the change of luminous intensity is monitored in real time through a light guide cable connected with an optical fiber.
图5 (a)基于细菌的光纤传感器检测水中重金属离子示意图;(b)实验装置示意图[39]

Fig. 5 (a) Schematic of the bacteria based fiber-optic sensor for monitoring heavy metal in water; (b) schematic of the experimental setup[39]. Copyright 2019, Elsevier

As the basic unit of an organism, cells are more resistant to changes in pH or temperature than other biological components such as antibodies, enzymes, and nucleic acids[28]. On the other hand, whether extracted from nature or cultured in the laboratory, the source of cells is very wide, so the cost of cell fiber optic biosensors is relatively lower. The continuous culture regeneration of cells also allows the regenerative use of biosensors. However, the sensors based on cells inevitably have some disadvantages, such as long cell culture period, relatively strict and complex culture conditions.

3 Application of Fiber Optic Biosensor in Food Contaminant Detection

Mycotoxins, heavy metal ions, antibiotics, pesticides and pathogens are common contaminants in food, which seriously threaten human health and have been the focus of attention in the field of food safety. There are many kinds of food and complex matrices, so how to effectively reduce the interference of sample matrix effect in the process of detecting various pollutants is still a major challenge. Compared with traditional methods, fiber-optic biosensors have their unique advantages in reducing the influence of sample matrix effects: (1) a small region on the surface of the fiber core is sensitive to the change of dielectric environment.And that optical fib signal can be regulated and control or sensitized through various sensing modes such as surface plasma resonance, heat flux, refractive index and the like, and the diversified sen modes can be applied to the detection of different types of food samples[42]; (2) The optical fiber sensor mainly triggers the signal change through the specific binding between the analyte to be detected in the solution and the sensing element on the surface of the optical fiber, and the detection method has strong selectivity[43]; (3) The optical fiber signal can be further sensitized by combining with various new biosensing strategies, and the actual sample used is less. The highly sensitive signal output can effectively improve the responsiveness to specific analytes in food, and can further reduce the interference of sample matrix background signals[44]. Therefore, fiber-optic biosensors have excellent application potential and prospects in the field of sensitive analysis of food pollutants. Some of the applications of fiber optic biosensors are outlined below by pollutant category.

3.1 Mycotoxin

Mycotoxins are secondary metabolites produced by fungi that can cause poisoning in humans and animals, and can lead to serious public safety problems[45]. Because mycotoxins are widely distributed in a variety of cereals and cereal products, food intake has become a routine route of human exposure to mycotoxins, which can lead to infection-related diseases or death[46]. Common toxins include zearalenone (ZEN), deoxynivalenol, ochratoxin A (OTA), and aflatoxin. Therefore, it is very important to construct specific analytical methods for the rapid and accurate detection of mycotoxins in food[42]. In recent years, thanks to the advantages of optical fiber sensors, there are more and more applications in the detection of mycotoxins. Xu et al. Constructed a portable and reusable LSPR-based fiber-optic biosensor for the detection of ZEN in beer, as shown in Fig. 6[19]. First, the thiol group was modified on the end of the fiber by conventional silanization method, and then the gold nanoparticles were connected to the thiol group through Au-S bond, and the surface of the gold nanoparticles was modified by ZEN aptamer. The functionalized optical fiber was used as a sensing probe, and when the probe was immersed in different concentrations of ZEN solution, the ZEN would be selectively trapped, and the LSPR absorption peak would be red-shifted to different degrees. The linear range was 1 ~ 480 ng/mL, and the detection limit was 0. 102 ng/mL. By cutting and polishing the used fiber tip and repeating the modification process mentioned above, it can be simply and conveniently used for the next test. In addition, when beer is used as an actual sample to detect ZEN, the sample processing process is relatively simple, only the beer is degassed, and the ZEN standard solution is added after being diluted by the buffer solution, and the recovery rate is calculated to be 85% -102%, so the method has good detection reliability.
图6 LSPR光纤生物传感器检测玉米赤霉烯酮示意图[19]

Fig. 6 Schematic diagram of the optical fiber-based LSPR biosensors for detecting zearalenone[19]. Copyright 2021, Elsevier

Lee et al. Designed a fiber-based LSPR sensing system for OTA detection[47]. In the sensor, a gold nanorod is modified on a sulfhydryl functionalized optical fiber core, the surface of the gold nanorod is modified by an OTA aptamer through Au-S bond interaction, and when an optical fiber probe is immersed in an OTA sample solution,The combination of OTA and OTA aptamer makes the aptamer change from single-chain structure to G-quadruplex structure and leads to the increase of the refractive index of the gold nanorod surface, resulting in the red shift of the LSPR absorption peak. The developed sensor can directly determine the concentration level of OTA between 10 pM and 100 nM. Then grape was selected as the actual sample to be detected, and the developed sensor was immersed in the mixed solution of grape juice and buffer solution with the volume ratio of 1:1, and different concentrations of OTA were added to the test, and the results showed that the recovery rate was 85. 5% ~ 116.9%.
Song et al. Developed a two-color evanescent wave all-fiber detection system based on fluorescence resonance energy transfer to simultaneously detect aflatoxin M1 (AFM1) and OTA in milk, as shown in Fig. 7[48]. The system is equipped with two lasers (635, 405 nm) for simultaneous acquisition of fluorescence signals at two wavelengths. The aptamers of AFM1 and OTA were labeled with two fluorophores (Cy5.5 and Alexa 405) with different excitation wavelengths, respectively, and their complementary sequences were labeled with the respective quenchers (BHQ3 and Dabcyl). In the absence of a target, the aptamer hybridizes to a complementary sequence, resulting in fluorescence quenching due to the proximity of the fluorophore and the quencher. After the addition of the target, the target binds to the corresponding aptamer to form a stable complex structure, resulting in the recovery of fluorescence intensity. Under the optimal conditions, the system can simultaneously and selectively determine AFM1 and OTA in the range of 1 ng/L ~ 1 mg/L, and the detection limits of AFM1 and OTA are 21 and 330 ng/L, respectively. In the milk sample test, the standard solution of toxin was added to the mixed solution of pure milk sample and acetonitrile, fully shaken and refrigerated, and the supernatant was diluted with buffer by centrifugation, thus reducing the matrix effect and directly used for sensor testing. The results showed that the recoveries of AFM1 and OTA ranged from 71.28% to 116.7% and from 74.13% to 124.8%, respectively.
图7 基于FRET同时检测AFM1和OTA的双色适配体传感器示意图[48]

Fig. 7 Schematic illustration of FRET-based dual-color aptasensor for simultaneous detection of AFM1 and OTA[48]. Copyright 2018, Springer Vienna

The development of biocoupling technology has led to the development of optical fiber biosensors, in which the receptor sensing element can be easily combined with other nanomaterials or signal molecules. When detecting various small molecular substances represented by biotoxins, the toxins can be easily captured by the aptamer modified on the optical fiber. Here, the optical fiber not only serves as a detection device, but also serves as a circulation device of the optical path, so that the equipment is simpler and more practical. However, it still has some problems, such as the binding time of the analyte to be tested is not fixed, and the pretreatment step of the optical fiber takes a long time. In the future, these shortcomings will be gradually overcome with the development of technology.

3.2 Heavy metal ion

Due to the non-degradability, bioaccumulation and strong toxicity of heavy metal ions, the presence of heavy metals in food and environmental samples can lead to serious pollution and health problems[46,49,50]. Some heavy metal ions, such as Pb2+, Hg2+ and Cd2+, can cause serious damage to the human body and even lead to a variety of diseases even at very low concentrations. Fiber optic biosensors have also become a convenient and effective method to detect heavy metal ions[51]. Sadani et al. Designed a U-shaped optical fiber sensor based on LSPR for sensitive detection of heavy metal ions[52]. Firstly, the optical fiber was silanized, and then bovine serum albumin (BSA) was immobilized on the surface of the optical fiber. The detection of mercury ions was realized by immobilizing the synthesized chitosan-coated gold nanoparticles on the BSA. The linear range was from 0.1 to 540 ppb. Fish and spinach samples were tested, and 1 G of each sample was neutralized with sodium hydroxide after being crushed, and then four different concentrations of mercury ions were added to the test, with an overall error of less than 15%.
Verma et al. Prepared a sensor based on surface plasmon resonance to detect cadmium ions, lead ions and mercury ions[53]. In this sensing model, the surface of the fiber core is coated with a 40 nm silver film, a 10 nm antimony tin oxide (ITO) film, and a pyrrole/chitosan coating in turn. ITO can improve the sensitivity of the sensor and protect the silver film from oxidation, and the composite matrix formed by pyrrole and chitosan can be highly sensitive to heavy metal ions. The detection limits of cadmium ion, lead ion and mercury ion were 0. 256,0.440 and 0. 796 ppb, respectively. Based on the ability of mercury ion to specifically bind to thymine to form a T-Hg2+-T complex structure, Zhou et al. Developed an optical fiber biosensor based on fluorescence resonance energy transfer, as shown in Fig. 8[49]. Quenching molecule labeled poly T base DNA strand (BQ-T14) and fluorescent molecule labeled poly A base DNA strand (CY-A14) were used as biological recognition element and signal reporter, respectively. When the sample does not contain the Hg2+, the two DNA molecular chains are hybridized, and the fluorescence signal intensity is low; When there is Hg2+ in the sample, due to the competitive binding between Hg2+ and BQ-T14, a stable T-Hg2+-T mismatch structure is formed, and then CY-A14 is excited by the evanescent wave on the fiber surface, and the fluorescence intensity increases, and the higher the concentration of Hg2+, the higher the fluorescence intensity. Under the optimal conditions, the detection period of a single sample did not exceed 10 min, and the detection limit of Hg2+ was 8. 5 nM. The recovery rate of commercial bottled water is between 89. 2% and 104. 7%, which proves that the sensor has sufficient precision and accuracy.
图8 便携式倏逝波光流控生物传感器检测Hg2+示意图[49]

Fig. 8 Schematic diagram of portable evanescent wave optofluidic biosensor for detecting Hg2+[49]. Copyright 2021, Elsevier

3.3 Antibiotics

Antibiotics play an important role in the treatment of diseases caused by bacterial infections in animals and are widely used in animal husbandry[54]. The use of high doses of antibiotics can lead to antibiotic residues in animal-derived foods, which will eventually enter the human body along the food chain and eventually affect human health, so it is very important to develop sensitive and effective methods to detect antibiotics in food quickly and accurately[55].
Shrivastav et al. Proposed a method to detect erythromycin in milk and honey by combining molecular imprinting technology with optical fiber sensing method[56]. Firstly, the sensing area of the fiber core was coated with a silver layer by thermal evaporation, and then the polymer nanoparticles with erythromycin as the template were coated on the silver-coated area by dip-coating method to prepare the binding sites for specific recognition of erythromycin. When the erythromycin molecule enters the imprinted binding site and through non-covalent binding, it causes the refractive index of the nanoparticle layer to change, thereby shifting the absorption wavelength to an extent proportional to the analyte concentration. Honey samples and milk samples were treated in the same way. They were diluted in water according to the ratio of weight to volume of 1:5. The diluted samples were filtered with a filter membrane and spiked for later use. The recovery rate was 98.2% ~ 102.0%, and the detection limit was 1.62 nM. The chemiluminescent optical fiber sensor developed by Nie et al. Realizes the multiple analysis of chloramphenicol, sulfadiazine and neomycin with different concentrations in milk samples by controlling the length of the optical fiber sensing area and the number of optical fibers, as shown in Figure 9[57]. Tang et al. Designed a fiber-optic biosensor based on target binding to promote fluorescence quenching for on-line continuous specific detection of kanamycin[58]. In this sensor system, the fluorophore-labeled kanamycin aptamer is prone to form an intermolecular multimeric structure (M-Apt) in the absence of antibiotics, while the binding of the aptamer to kanamycin leads to the destruction of M-Apt and the formation of kanamycin-aptamer-hairpin structure (Kana-Apt). The photoinduced electron transfer between the fluorophore and kanamycin partially quenches the fluorescence of the hairpin structure, and thanks to the ability of GO to efficiently bind to single-stranded DNA, the introduced GO quenches the fluorescence of free single-stranded DNA, reduces the background of the sensor, and further quenchs the fluorescence of Kana-Apt. Unbound aptamer in solution binds to unlabeled aptamer immobilized on the fiber, and the fluorescence signal increases, and the intensity is inversely proportional to the concentration of kanamycin, with a linear range of 200 nM to 200 μM and a detection limit of 26 nM.
图9 可调谐检测范围的光纤介导免疫传感器用于多重检测兽药残留[57]

Fig. 9 Optical fiber-mediated immunosensor with a tunable detection range for multiplexed analysis of veterinary drug residues[57]. Copyright 2019, American Chemical Society

In order to achieve real-time automatic monitoring of streptomycin (STR), Zhu et al. Established a sandwich-type fluorescent biosensor based on separated aptamer (SPA)[59]. First, one fragment of the STR-specific separation aptamer, SPA1STR, was immobilized on the optical fiber by covalent binding, and the other fragment, SPA2STR, modified with Cy5.5 fluorophore, was mixed with the sample. Due to the presence of STR, SPA2STR forms a stable composite structure with the STR captured by SPA1STR. The evanescent wave excites Cy5.5 on SPA2STR to produce a fluorescence signal, and the fluorescence intensity reflects the concentration level of STR. The sensor is sensitive in the range of 60 – 526 nM with a detection limit as low as 33 nM.

3.4 Pesticide residue

The abuse of pesticides can also lead to the residues of environmental pollutants and metabolites, which are likely to flow into the human body along with the food chain, so it is necessary to detect pesticide residues conveniently and quickly[42,60]. Miliutina et al. Developed an optical fiber sensor fixed with a metal-organic framework (MOF-5) layer to detect organophosphorus pesticides, which was tested with fenitrothion in water, as shown in Figure 10[61]. First, a gold film of about 40 nm is deposited on the fiber core by vacuum sputtering, and then a MOF-5 layer with high affinity for pesticides is deposited on the gold surface. When the sensor is immersed in an organophosphorus pesticide solution, the target will fill the gaps in the MOF-5 layer, resulting in an increase in the refractive index of the MOF-5 surface and a change in the wavelength of the plasma absorption band.The detection limit of pesticide can be as low as 10 pM, which is much lower than the detection limit 14552 of the traditional gas chromatography in the national standard (standard number: GB/T 14552-2003), and the sensing response time is affected by the concentration of pesticide, the higher the concentration, the faster the response, which can save time and cost to a greater extent. Atrazine (ATZ), as one of the most common herbicides in the world, has a high risk of carcinogenesis and poses a serious threat to human health. Long et al. Developed an optofluidic biosensing platform to detect ATZ[62]. First, the hapten-carrier conjugate ATZ-bovine serum albumin (ATZ-BSA), which was used as a recognition element, was covalently coupled to the surface of the fiber, and then the Anti-ATZ monoclonal antibody (anti-ATZ-MAb), which was highly specific for ATZ, was bound to the ATZ-BSA. Because Anti-ATZ-Mab was modified with Cy5.5 fluorophore in advance, the evanescent wave excited the fluorophore to produce a fluorescence signal, and the intensity of the fluorescence signal was positively correlated with the concentration of ATZ. The final detection limit was 0. 06 μg/L, which was lower than the detection limit of 0. 25 μg/L in the national standard of liquid chromatography/mass spectrometry (standard number: GB/T 21925-2008), and had high application significance.
图10 用于农药固相提取和检测的功能性等离子体光纤传感器的制备示意图[61]

Fig. 10 Schematic of the preparation of functional plasmonic optical-fiber sensor for pesticides solid-phase extraction and detection[61]. Copyright 2020, Elsevier

3.5 Pathogen

Foodborne pathogens pose a serious threat to human health. After eating food containing foodborne pathogenic microorganisms, human beings are prone to nausea, vomiting, diarrhea and other symptoms, and even cause serious organ damage or cancer[63]. Common pathogens include Campylobacter, non-typhoid Salmonella, Shiga toxin-producing Escherichia coli, Listeria monocytogenes and so on[64,65]. Due to the disadvantages of traditional detection methods, such as long detection time, low sensitivity, complex and expensive instruments, it is of great significance to construct new biosensors for qualitative and quantitative detection of these pathogens[66].
Cui et al. Developed a quantum dot immunofluorescence biosensor for the detection of Staphylococcus aureus, as shown in Figure 11. The sensor has both high specificity of antigen-antibody interaction and high sensitivity and stability of quantum dot fluorescence[67]. After silanization of the optical fiber, the anti-Staphylococcus aureus polyclonal antibody was first immobilized on the surface of the probe to specifically capture Staphylococcus aureus, and when Staphylococcus aureus was captured, the monoclonal anti-Staphylococcus aureus antibody was combined with the captured Staphylococcus aureus again to form a double antibody sandwich structure. Biotin-labeled goat anti-mouse IgG antibody was then combined with mouse monoclonal antibody. Finally, streptavidin-labeled QDs were introduced for fluorescence labeling by specific binding between biotin and streptavidin. The signal cascade amplification of the biotin-streptavidin-quantum dot system can significantly enhance the fluorescence signal to improve the sensitivity, and the fluorescence intensity is significantly linearly and positively correlated with the logarithm of the concentration of Staphylococcus aureus in the range of 103~107CFU/mL, and the detection limit reaches 1×103CFU/mL. Fang et al. Proposed a novel two-color fluorescent aptamer sensor for simultaneous detection of Escherichia coli O157: H7 and Salmonella typhimurium[27]. Firstly, hydrofluoric acid was used to etch a conical optical fiber probe with a nanoporous layer of 50 ~ 100 nm, and Cy3-labeled Escherichia coli O157: H7 aptamer sensor (Cy3-apt-E) and Cy5.5-labeled Salmonella typhimurium aptamer sensor (Cy5.5-apt-S) were used as biological recognition elements and signal reporters, respectively. When the mixture of the two sensors and the two bacteria is introduced to the surface of the probe, the free Cy3-apt-E and Cy5.5-apt-S can enter the nanoporous layer of the optical fiber and be excited by the evanescent wave to generate fluorescence,However, Cy3-apt-E and Cy5.5-apt-S, which bind to two kinds of bacteria respectively, can not be excited because the size of bacteria is too large to enter the nanopore layer, so the relationship between the number of bacteria and the fluorescence intensity can be used to quantitatively detect pathogenic bacteria. Ultimately, the detection limits for E. Coli O157: H7 and S. Typhimurium were 340 and 180 CFU/mL, respectively. The recoveries of E. coli O157: H7 and Salmonella typhimurium were 82. 1% -97. 4% and 73% -89. 6%, respectively, when the samples were sterilized and directly added with different amounts of bacteria for fluorescence signal detection.
图11 基于光纤探针的量子点免疫荧光生物传感器检测金黄色葡萄球菌[67]

Fig. 11 Optical fiber probe-based quantum dots immunofluorescence biosensors in the detection of staphylococcus aureus[67]. Copyright 2021, Frontiers Media S.A.

To sum up, when the optical fiber biosensor detects the above harmful substances, different sensing processes can be designed for different targets. By modifying aptamers, antibodies, enzymes and other molecules on the surface of optical fibers, combined with fluorescence, LSPR effect, nucleic acid amplification and other strategies, a diversified sensing platform suitable for precise detection of various food pests can be constructed, and its detection sensitivity and specificity can also be guaranteed. With the development of materials with better performance and more innovative detection methodologies, fiber optic biosensors are expected to become a detection technology with great practical application value, which is widely used in food safety, environmental monitoring and other fields.

4 Conclusion and prospect

By using the optical fiber as the carrier of the biological recognition element and modifying the surface of the optical fiber, a diversified and functional optical fiber biosensor can be constructed, which can realize the sensitive, convenient and highly specific detection of various trace harmful substances in food. However, at present, there are still some shortcomings, such as the complicated pretreatment steps of optical fiber, the signal stability is greatly affected by the environment, and the reusability of optical fiber is still facing challenges on the premise of ensuring sensitivity. With the continuous development of miniaturization of instruments, continuing to seek progress in technology and operation methods, and trying to combine with other fields or devices such as smart phones, artificial intelligence, mechanical automation, etc., it can be predicted that optical fiber biosensors still have great prospects for development. This new biosensing method has potential application value and practical significance in the fields of rapid detection of food pollutants, environmental monitoring, biomedicine and so on.
[1]
Yao Y, Hu T, Song C, Liu C, Kong D Z, Huang C, Zhu J, Shen W, Shi H W, Tang S. Anal. Chim. Acta, 2021, 1187: 339169.

[2]
Yan C, Teng J, Liu F Y, Yao B B, Xu Z L, Yao L, Chen W. Microchem. J., 2020, 159: 105414.

[3]
Feng Y, Zhang W J, Liu Y W, Xue J M, Zhang S Q, Li Z J. Molecules, 2018, 23(8):1953.

[4]
Wang W J, Fu M, Zhang Q D, Zhen Y R, Liu J J, Xiang S N, Michal J J, Jiang Z H, Zhou X, Liu B. Food Chem., 2021, 341: 128170.

[5]
Al-Dalali S, Li C, Xu B C. Food Chem., 2022, 376: 131881.

[6]
Jia Y X, Zhao S Q, Li D S, Yang J L, Yang L. Food Contr., 2023, 144: 109361.

[7]
Allsop T, Neal R. Sensors, 2019, 19(22): 4874.

[8]
Kadhum Hisham H. Am. J. Remote. Sens., 2018, 6(1): 1.

[9]
Al Mahmud R, Sagor R H, Khan M Z M. Opt. Laser Technol., 2023, 159: 108939.

[10]
Caucheteur C, Guo T, Albert J. Anal. Bioanal. Chem., 2015, 407(14): 3883.

[11]
Nag P, Sadani K, Mohapatra S, Mukherji S, Mukherji S. Anal. Chem., 2021, 93(4): 2299.

[12]
Xu Y, Luo Z W, Chen J M, Huang Z J, Wang X, An H F, Duan Y X. Anal. Chem., 2018, 90(22): 13640.

[13]
Sai V V R, Kundu T, Mukherji S. Biosens. Bioelectron., 2009, 24(9): 2804.

[14]
Zakaria R, Kam W, Ong Y S, Yusoff S F A Z, Ahmad H, Mohammed W S. J. Mod. Opt., 2017, 64(14): 1443.

[15]
Gasior K, Martynkien T, Wojcik G, Mergo P, Urbanczyk W. Opto Electron. Rev., 2017, 25(1): R1.

[16]
Li H T, Huang Y Y, Hou G H, Xiao A X, Chen P W, Liang H, Huang Y G, Zhao X T, Liang L L, Feng X H, Guan B O. Sci. Adv., 2019, 5(12): eaax4659.

[17]
Jia H, Zhang A, Yang Y Q, Cui Y Q, Xu J R, Jiang H W, Tao S C, Zhang D W, Zeng H P, Hou Z Y, Feng J J. Lab Chip, 2021, 21(12): 2398.

[18]
Wang M, Yang F, Dai S X, Cao Z F, Su J X, Ding S J, Zhang P Q. J. Light. Technol., 2021, 39(14): 4828.

[19]
Xu Y C, Xiong M, Yan H. Sens. Actuat. B Chem., 2021, 336: 129752.

[20]
Soares M S, Vidal M, Santos N F, Costa F M, Marques C, Pereira S O, Leitão C. Biosensors, 2021, 11(9): 305.

[21]
Li L K, Zhang Y N, Zheng W L, Lv R Q, Zhao Y. Opt. Lett., 2023, 48(4): 952.

[22]
Wang Q, Jiang X, Niu L Y, Fan X C. Opt. Lasers Eng., 2020, 128: 105997.

[23]
Kant R, Tabassum R, Gupta B D. Sens. Actuat. B Chem., 2017, 242: 810.

[24]
Huang Q, Zhu W J, Wang Y, Deng Z, Li Z, Peng J K, Lyu D J, Lewis E, Yang M H. Sens. Actuators B, 2020, 321: 128480.

[25]
Singh R, Kumar S, Liu F Z, Shuang C, Zhang B Y, Jha R, Kaushik B K. Biosens. Bioelectron., 2020, 168: 112557.

[26]
Zhou J R, Qi Q Q, Wang C, Qian Y F, Liu G M, Wang Y B, Fu L L. Biosens. Bioelectron., 2019, 142: 111449.

[27]
Fang S Y, Song D, Zhuo Y X, Chen Y, Zhu A N, Long F. Biosens. Bioelectron., 2021, 185: 113288.

[28]
Liang G L, Luo Z W, Liu K P, Wang Y M, Dai J X, Duan Y X. Crit. Rev. Anal. Chem., 2016, 46(3): 213.

[29]
Semwal V, Gupta B D. Sens. Actuat. B Chem., 2021, 329: 129062.

[30]
Wang Y, Zhu G, Li M Y, Singh R, Marques C, Min R, Kaushik B K, Zhang B Y, Jha R, Kumar S. IEEE T Nanobioscie, 2021, 20(3): 377.

[31]
Chaudhari P P, Chau L K, Tseng Y T, Huang C J, Chen Y L. Mikrochim. Acta, 2020, 187(7): 396.

[32]
Kim H M, Jeong D H, Lee H Y, Park J H, Lee S K. Sci. Rep., 2021, 11(1): 15985.

[33]
Jing J Y, Liu K, Jiang J F, Xu T H, Wang S, Ma J Y, Zhang Z, Zhang W L, Liu T G. Nanomaterials, 2021, 11(8):2137.

[34]
Kim H M, Lee H Y, Park J H, Lee S K. ACS Sens., 2022, 7(5): 1451.

[35]
Yang Y H, Liu J C, Zhou X H. Biosens. Bioelectron., 2021, 190: 113418.

[36]
Janik M, Hamidi S V, Koba M, Perreault J, Walsh R, Bock W J, Smietana M. Lab Chip, 2021, 21(2): 397.

[37]
Tseng Y T, Li W Y, Yu Y W, Chiang C Y, Liu S Q, Chau L K, Lai N S, Chou C C. Sensors, 2020, 20(11):3137.

[38]
Ngo L T, Wang W K, Tseng Y T, Chang T C, Kuo P L, Chau L K, Huang T T. Anal. Bioanal. Chem., 2021, 413(12): 3329.

[39]
Halkare P, Punjabi N, Wangchuk J, Nair A, Kondabagil K, Mukherji S. Sens. Actuat. B Chem., 2019, 281: 643.

[40]
Futra D, Heng L E, Ahmad A, Surif S, Ling T. Sensors, 2015, 15(6): 12668.

[41]
Zajic J, Ripp S, Trogl J, Kuncova G, Pospisilova M. Sensors, 2020, 20(11):3237.

[42]
Loyez M, DeRosa M C, Caucheteur C, Wattiez R. Biosens. Bioelectron., 2022, 196: 113694.

[43]
Xu B, Xiang X Y, Ding L Y, Luo Z H, Zhao J, Huang J, Li H J, Jiang X D. IEEE Sens. J., 2023, 23(7): 6832.

[44]
Narsaiah K, Jha S N, Bhardwaj R, Sharma R, Kumar R. J. Food Sci. Technol., 2012, 49(4): 383.

[45]
Ganesan A R, Mohan K, Karthick Rajan D, Pillay A A, Palanisami T, Sathishkumar P, Conterno L. Food Chem., 2022, 378: 131978.

[46]
Chen H Y, Zhang L, Hu Y, Zhou C S, Lan W, Fu H Y, She Y B. Sens. Actuators B, 2021, 329: 129135.

[47]
Lee B B, Park J H, Byun J Y, Kim J H, Kim M G. Biosens. Bioelectron., 2018, 102: 504.

[48]
Song D, Yang R, Fang S Y, Liu Y P, Long F. Microchim. Acta, 2018, 185(11): 508.

[49]
Zhou Y, Wang H L, Song D, Li Z G, Han S T, Long F, Zhu A N. Anal. Chim. Acta, 2021, 1155: 338351.

[50]
Han S T, Zhou X H, Tang Y F, He M, Zhang X Y, Shi H C, Xiang Y. Biosens. Bioelectron., 2016, 80: 265.

[51]
Zhang J J, Cheng F F, Li J J, Zhu J J, Lu Y. Nano Today, 2016, 11(3): 309.

[52]
Sadani K, Nag P, Mukherji S. Biosens. Bioelectron., 2019, 134: 90.

[53]
Verma R, Gupta B D. Food Chem., 2015, 166: 568.

[54]
Zhou C, Zou H M, Sun C J, Li Y X. Food Chem., 2021, 361: 130109.

[55]
Khan M Z H. Crit. Rev. Anal. Chem., 2022, 52(4): 780.

[56]
Shrivastav A M, Usha S P, Gupta B D. Biosens. Bioelectron., 2017, 90: 516.

[57]
Nie R B, Xu X X, Chen Y P, Yang L. ACS Sens., 2019, 4(7): 1864.

[58]
Tang Y F, Gu C M, Wang C, Song B D, Zhou X H, Lou X H, He M. Biosens. Bioelectron., 2018, 102: 646.

[59]
Zhu Q, Liu L H, Wang R Y, Zhou X H. J. Hazard. Mater., 2021, 403: 123941.

[60]
Carvalho F P. Food Energy Secur., 2017, 6(2): 48.

[61]
Miliutina E, Guselnikova O, Burtsev V, Elashnikov R, Postnikov P, Svorcik V, Lyutakov O. Talanta, 2020, 208: 120480.

[62]
Long F, Zhu A N, Shi H C, Sheng J W, Zhao Z. Chemosphere, 2015, 120: 615.

[63]
Qu L L, Ying Y L, Yu R J, Long Y T. Mikrochim. Acta, 2021, 188(6): 201.

[64]
Heredia N, García S. Anim. Nutr., 2018, 4(3): 250.

[65]
Franz C M A P, den Besten H M W, Böhnlein C, Gareis M, Zwietering M H, Fusco V. Trends Food Sci. Technol., 2018, 81: 155.

[66]
Zhang R Y, Belwal T, Li L, Lin X Y, Xu Y Q, Luo Z S. Compr. Rev. Food Sci. Food Saf., 2020, 19(4): 1465.

[67]
Cui J W, Zhou M J, Li Y, Liang Z X, Li Y Q, Yu L, Liu Y, Liang Y, Chen L G, Yang C X. Front. Cell. Infect. Microbiol., 2021, 11: 665241.

Outlines

/