Open Access
Add to BibSonomy
Abstract
In this work, we introduce a novel multimode fiber (MMF) – seven core fiber (SCF) – MMF (MCM) optical fiber biosensor, also known as the WaveFlex biosensor (plasma wave assisted fiber biosensor), based on localized surface plasmon resonance (LSPR) for qualitative detection of xanthine. Xanthine is a purine base widely distributed in human blood and tissues, and commonly used as an indicator for various disease detections. The MCM sensor incorporates a tapered optical fiber structure, fabricated using the combiner manufacturing system (CMS), and is designed with SCF and MMF. By effectively harnessing LSPR, the sensor boosts the attachment points of biomolecules on the probe surface through immobilized tungsten disulfide (WS2)-thin layers, gold nanoparticles (AuNPs), and carbon nitride quantum dots (C3N-QDs). The functionalization of xanthine oxidase (XO) on the sensing probe further enhances the sensor's specificity. The proposed WaveFlex biosensor exhibits a remarkable sensitivity of 3.2 nm/mM and a low detection limit of 96.75 µM within the linear detection range of 100 - 900 µM. Moreover, the sensor probe demonstrates excellent reusability, reproducibility, stability, and selectivity. With its sensitivity, biocompatibility, and immense potential for detecting human serum and fish products, this WaveFlex biosensor presents a promising platform for future applications.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Xanthine is a purine base widely distributed in the organs and body fluids of humans and other organisms. A variety of stimulants are derived from xanthine, including caffeine and theobromine [1,2]. It is produced by the breakdown of adenosine triphosphate (ATP), an important intermediate in purine degradation, and will ultimately be converted to uric acid (UA) by xanthine oxidase (XO). It is mainly produced by guanine deaminase from guanine and/or XO from hypoxanthine [3,4]. During the metabolic process, XO plays a catalytic role in oxidizing hypoxanthine to xanthine, and then XO further catalyzes the conversion of xanthine to uric acid, which is broken down in the bloodstream and enters the kidneys, where it is excreted in the form of urine [5]. Xanthine belongs to a class of purines that also participate in many metabolic processes as cofactors and play a key role in fundamental biological processes [6]. When the purine content in the body is in the normal range, uric acid is excreted smoothly through the kidneys; however, when the xanthine content in the body is too high and exceeds the body's own metabolic capacity, it can cause a variety of disorders (often referred to as hyperuricosuric acidosis or Lesch-Nyhan syndrome), such as gout, diabetes mellitus, high blood pressure, obesity, high cholesterol, nephropathy, heart disease, etc [7]. At present, the prevalence of gout in China is uneven due to differences in regions and dietary habits, with an overall prevalence of 1.0% - 15.3% [8], and the prevalence of hyperuricosuria is gradually increasing. The normal level of xanthine in human blood is around 10 µM, while patients with diseases such as gout, xanthinuria, and renal failure [9] have a more pronounced increase in xanthine values. It has been mentioned that choosing foods with low levels of purines in your rich daily dietary choices can reduce the possibility of the above-mentioned diseases. Most health and nutritional investigations have shown that purine-rich foods such as fish, meat, and animal offal increase serum xanthine levels [10,11]. Therefore, accurate detection and research analysis of xanthine have scientific value in both biomedical analysis and detection in clinical medicine [12,13]. If xanthine can be monitored promptly and accurately, it can play a role in preventing the occurrence of various diseases.
Currently, researchers have investigated the detection of xanthine using electrochemical, fluorescence, and colorimetric analysis. Shan et al. [14] proposed a xanthine biosensor coupling XO with horseradish peroxidase (HRP) using calcium carbonate nanoparticles (Nano-CaCO3-NPs) as an enzyme immobilization carrier. Their sensor is not only stable, but also avoids interference from coexisting substances. Xue et al. [15] reported fluorescence burst based zinc oxide nanomaterials (ZnO-NMs) for the label-free detection of xanthine. The method was successfully applied to the determination of xanthine in fish samples with satisfactory results. Hong et al. [16] established a colorimetric method to detect xanthine by binding WSe2 nanosheets to XO. This sensor can detect xanthine in human serum and has great potential for development. Similarly, Li et al. [17] developed a convenient colorimetric method for xanthine detection in urine using WO3 nanosheets. Hsu et al. [18] proposed a combination of the colorimetric method with nanoparticles (NPs) to measure xanthine that is innovative and effective. In addition to the above-mentioned detection methods, researchers have also mentioned the great importance of fiber based biochemical sensors. Sukanya et al. [19] used unclad fiber to make a lossy mode resonance sensor to detect the refractive index (RI) of liquid. Esposito et al. [20] investigated the effect of gamma radiation on long-period fiber Bragg grating, and also used a miniaturized metal package design for fiber Bragg grating sensors, that significantly improved their sensitivity in the field of temperature detection [21].
Nowadays, SPR-based sensor are quite popular because they utilize the plasmonic properties of metals. In 1982, Nylander et al. [22] introduced the principle of SPR into sensors and proposed a device for detecting gases, which is the prototype based on the SPR sensor. Shi et al. [23] developed an SPR biosensor based on a sensitive and stable polydopamine-modified gold membrane. The SPR biosensor is used for the detection of biomolecules and does not require labeling the object to be measured. The turbidity or color of the solution to be measured does not affect the detection, and it can be detected online in real time. However, SPR biosensors are large and expensive and require strict experimental conditions, such as temperature. Thus, localized surface plasmon resonance (LSPR) technology has been gradually developed on the basis of SPR technology. Compared with SPR, LSPR sensors have a small sample requirement and a lower limit of detection (LoD) range, which allows them to detect biological samples at low concentrations [24]. In addition, it is also relatively easy and cheaper to immobilize metal nanoparticles (MNPs) on optical fibers. Fiber optic LSPR technology has important research significance and application value in the fields of biomedicine, drug delivery, and environmental monitoring. Therefore, combining LSPR technology with fiber optic sensing technology is an inevitable trend in the development of biosensing. Luo et al. [25] developed a cellular detection method based on U-shaped fiber optic LSPR, which is used for the label-free detection of cancer cells as well as the expression of N-glycans on the cell surface. The application of the method was validated by testing six different cell lines. The results showed that the method provides a powerful tool for the study of biological processes related to N-glycans while also providing a promising platform for clinical diagnosis. Recently, researchers [26,27] have developed LSPR-based fiber-optic biosensors for the detection of histamine in seafood and aflatoxin B1 in agricultural by-products.
Due to the control of the various properties of NMs, which has led to a systematic understanding of the surface plasma of NMs. Today, one of the most widely exploited features of MNPs is the LSPR, which refers to the collective oscillation of electrons on MNPs excited by incident photons at the resonance frequency. Incident light is absorbed by the NPs only when its wavelength is larger than the particle size of the conducting NPs. The incident light interacts with the free electrons on the surface of the MNPs to produce a continuous localized surface plasma oscillating at a resonant frequency, thereby producing the unique LSPR optical effect. Resonance is reached when the frequency of the incident light is equal to the collective oscillation frequency of the free electrons and shows characteristic extinction peaks in the UV-visible extinction spectra. The changes in the peak wavelengths of the LSPR absorption spectra can be used as sensing mechanisms. The very strong and highly localized evanescent fields induced by the LSPR enable the NPs to act as energy transducers that are highly sensitive to small changes in the local refractive indices. These changes manifest as alterations in the extinction spectra. The interaction of the target analyte with the ligand on the nanoparticle at the sensor surface induces a change in the RI of the surroundings, leading to a shift in the peak absorption wavelength of the MNPs [28].
The most commonly used NPs that are relevant to the sensing of LSPR are gold and silver. The absorption of incident light by gold nanoparticles (AuNPs) and silver nanoparticles (AgNPs) occurs in the visible range. Of all the metals, Ag exhibits the sharpest and strongest plasma vibrational bands, whereas the chemical inertness and biocompatibility of Au, as well as the chemical bond of the sulfhydryl-gold association that can be used to immobilize biomolecules, make it more suitable for biological applications. Transition metal disulfides (TMDs) are two-dimensional (2D) materials that are similar to graphene. TMDs have made significant advances in photodetection in recent years, due to their stronger light-matter interactions than graphene. Tungsten disulfide (WS2), a typical member of the TMDs group, exhibits excellent thermal stability in a wide range of applications. The combination of WS2 thin layers with AuNPs excites the LSPR effect more efficiently. Carbon nitride quantum dots (C3N-QDs), with similar properties to the WS2 thin layers, have attracted a great deal of attention due to their ultra-high rigidity, thermal conductivity, and excellent electronic and magnetic properties. Previous studies have shown that C3N has good performance in different applications, such as nanoelectronics and catalysis [29,30].
In this work, a WaveFlex biosensor based on LSPR is proposed for xanthine molecular detection. Due to the unique seven-core structure of seven-core fiber (SCF), it has enhanced transmission characteristics. We selected SCF and MMF to fabricate a tapered fiber structure based on MCM (Multimode fiber – Seven Core Fiber – Multimode fiber). The NMs, WS2 thin layer, AuNPs, and C3N-QDs were sequentially immobilized on the surface of the optical fiber, which can effectively stimulate the LSPR effect and improve the performance of the sensor. Moreover, additional attachment points can be provided for the functionalization of the XO enzyme. According to the experimental results, the sensitivity and the LoD of the developed sensor were 3.2 nm/mM and 96.75 µM, respectively. This sensor is of great significance for the design of highly sensitive xanthine sensors.
2. Experiment
2.1 Materials and reagents
The seven-core fiber (SCF, 6.1 µm/125 µm) and photosensitive fiber (PSF, 4.2 µm/125 µm) used in the experiments were purchased from FiberCore Ltd., UK, and the single-mode fiber (SMF, 8.2 µm/125 µm) and multimode fiber (MMF, 62.5 µm/125 µm) were purchased from Shenzhen EB-link Technology Co., China. AuNPs solution was synthesized from tetrachloroauric acid (HAuCl4), trisodium citrate, and deionized (DI) water. Acetone, concentrated sulfuric acid (H2SO4, 98%), hydrogen peroxide solution (H2O2, 30%), (3-mercaptopropyl) trimethoxysilane (MPTMS), and ethanol were used for the cleaning of probe surfaces and the immobilization of NMs (C3N-QDs, AuNPs, and WS2-thin layer). The C3N-QDs and WS2 thin-layer nanomaterials were purchased from Xianfeng Nano, Jiangsu. 11-Mercaptodecanoic acid (MUA), N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS), and phosphate-buffered saline (PBS) were procured from Sigma-Aldrich, Shanghai, for the functionalization of the XO enzyme on the nanomaterial-immobilized probes. The XO enzyme was purchased from Sigma-Aldrich, Shanghai. Different concentrations of xanthine solutions were prepared with PBS and purchased from Macklin, Shanghai.
2.2 Measurement and instrument
The sensing probe was developed using a special fiber fusion splicer (FSM-100P+, Fujikura, Japan) and the 3SAE combiner manufacturing system (CMS, USA). An ultra-large core diameter fiber cutter (CT106, Japan) was used to cut the fiber to a quantitative length. We measured the absorption spectrum of AuNPs using a UV-visible spectrophotometer (Hitachi-U-3310) to characterize the NPs. High-resolution transmission electron microscopy (HR-TEM, Talos L120C, Thermo Fisher Scientific, USA) was used to observe the morphology of the NMs used in the experiment. In addition, a tungsten-halogen light source (HL-2000, Ocean Optics, USA) and a spectrometer (USB2000+, Ocean Optics, USA) were used to study the optical transmission properties of the developed sensing probes. Scanning electron microscopy (SEM, Gemini, Carl Zeiss Microscopy) scans the sample by emitting a high-energy electron beam to observe the coating of NMs on the surface of the sensing probe. In addition, a tungsten-halogen light source (HL-2000, Ocean Optics, USA) and a spectrometer (USB2000+, Ocean Optics, USA) were used to study the optical transmission characteristics of the sensing probes developed.
2.3 Sensing principle
To realize the measurement of WaveFlex biosensors, the formation of an effective evanescent wave (EW) is very important. On the one hand, the effective EW can excite the LSPR phenomenon and generate specific absorption peaks related to the RI. On the other hand, the effective evanescent field can also spatially overlap with the analyte to be measured sufficiently, to enable the perception of biomass or biological behavior. Typically, light is transmitted mainly through the core of the optical fiber. The EW generated at the core-cladding interface has a very low penetration depth. The EW attenuates exponentially along the diameter of the cladding, and the penetration depth is much less than the thickness of the cladding. The EW is difficult to reach at the cladding or the dielectric interface, which limits the interaction of light with the surrounding environment. In fiber optic-based WaveFlex biosensors, sensing mainly depends on the change in RI near the sensor probe. The change in RI depends on the strength of the evanescent field, which directly affects the sensitivity of the sensor. The penetration depth of the EW can be enhanced by changing the geometry of the optical fiber to achieve a correlation between the EW and the surrounding environment and improve the sensitivity. The modified structures of conventional optical fibers include U-shaped fibers, D-shaped fibers, spherical fibers, single mode-multimode-single mode (SMS) fibers, and tapered fibers [31–35]. All these fiber structures can effectively excite the enhanced evanescent field and promote the interaction between light and sensing materials.
In this work, a MCM sensor structure is developed by fusing a multimode fiber (MMF), a seven-core fiber (SCF), and a multimode fiber (MMF), and the SCF is fused and pulled to taper. The schematic diagram of the sensor structure is shown in Fig. 1. Because of the mode field mismatch, when light passes through the MMF for the first time into the tapered region, it excites a new mode in the 6-core (in hexagonal pattern) of SCF and the cladding surrounding the MMF. With the tapered fiber structure, the incident light can be coupled to each core to excite the higher-order cladding mode. In the tapered region, the higher-order cladding mode is transmitted through the cladding region and then coupled back to the core mode at the end of the tapered region. EWs are produced because of the interference between the optical paths of core mode and cladding mode. The LSPR effect is motivated by EWs. When the effective refractive index of the external environment changes, the transmission spectrum also changes and finally reflects the shift of the peak wavelength. This can be analyzed according to the principle of the Mach-Zehnder interferometer (MZI) structure. The interference strength between the core mode and the cladding mode can be expressed as [36]:
Here, ${I_{C1}}$ and ${I_{C2}}$ are the transmitted light intensities of the fiber core and cladding modes, respectively, and $\varphi $ is the phase difference due to the interference between the two modes, which can be expressed as:
where $\lambda $ is the central wavelength of the transmission spectrum, and ${n_{\textrm{co}}}$ and ${n_{cl}}$ are the effective RIs of the fiber core and cladding, respectively. The ratio of core to cladding is constant for an optical fiber after taper pulling. The small size of the fiber produces a weak enough waveguide effect. At this point, the original core will not be able to bind light waves. The entire core and cladding portion of the tapered region will be defined as the new core, with the external medium acting as the cladding. The distance that the EW passes through the cladding, which is the penetration depth, can be expressed as [37]:Here, $\alpha $ is the angle of incidence at the interface between the core and the cladding. In the tapered-waist region, the tail of the evanescent field can penetrate into the surrounding medium, whereby an effective evanescent field can be formed for achieving the sensing purpose.
The LSPR effect of precious MNPs makes them well suited for use in biosensors. Biosensors can enable the detection of small molecules at low concentrations by detecting changes in the RI of the medium surrounding the target analyte. The sensing performance can be enhanced by the interaction of EWs with MNPs, and the MNP layer covering the surface of the optical fiber structure can produce the LSPR effect. The change in peak wavelength or peak absorbance in the absorption spectrum of the MNPs is used as the sensing mechanism of the LSPR sensor. It will detect the change in RI around the NPs due to the interaction of biomolecules. The biomolecules are similar in size to the precious MNPs. If XO is immobilized on the surface of NPs, the xanthine in the analyte to be measured binds to the XO, thus changing the RI around the NPs.
This results in a change in the position of the LSPR resonance peak in the extinction spectrum. When the RI of the surrounding medium changes, the wavelength changes accordingly, which can be expressed as [38]:
Here, m is the nanoparticle sensitivity factor, Δn is the change in effective RI, ${d_p}$ is the penetration depth of the EWs, and l is the thickness of the effective adsorption layer.
2.4 Simulation of sensor probe
In order to analyze the transmission characteristics of light in optical fibers, the Beam PROP function of RSoft software is used to simulate the transmission behavior of light within the MCM fiber structure. The core and cladding diameters of the multimode fiber are set to 62.5 µm and 125 µm, respectively, and the cladding diameter of the SCF is the same as that of the multimode fiber, which is 125 µm, and the diameters of the seven cores are the same, which is 6.1 µm. The RIs of the core and the cladding are 1.4592 and 1.4440, respectively, and the energy distributions of the light passing through the sensors are obtained through the simulations as shown in Fig. 2.
Fig. 2. Simulation of the MCM-based tapered optical fiber structure: light passing through, (a) multimode fiber, (b) seven-core fiber, (c) cross-sectional field distribution in the tapered region, (d) light coupling back to the multimode fiber.
Light is stably bound to be transmitted within the multimode fiber core before it enters the SCF, as shown in Fig. 2(a). Figure 2(b) shows that during the process of light entering SCF propagation, one part of the light is coupled to the SCF mode, while another part of the light continues to propagate in the MMF mode. As shown in Fig. 2(c), when light enters a tapered region with a uniform diameter, a small portion of the light is coupled to the fiber core on the side, and most of the light is transmitted in the cladding. At this point, the cladding mode is excited. The evanescent field in this region is also significantly enhanced, making the sensing region more sensitive. Finally, the light begins to couple from the cladding mode of the SCF back to the core mode of the MMF, as shown in Fig. 2(d). In this process, there is still an interference effect, which affects the performance of the sensing region.
2.5 Fabrication of sensor probe
The MCM sensor structure used in this work consists of two sections of MMF spliced through a fusion splicer to both ends of a SCF. The distribution of the cores of the SCF is in the form of a positive hexagon. One core is distributed in the center of the hexagon, and the other six cores are distributed on the periphery of the central core region [39]. The cladding of the SCF has a diameter of 125 µm and each core has the same diameter of 6.1 µm. The distance between two neighboring cores is the same. Since the seven-core fiber has more cores, it has better stability and higher sensitivity. First, the coatings of MMF and SCF were removed using a fiber stripper, and the cladding was cleaned with dust-free paper dipped in ethanol. Then the end faces of MMF and SCF were cut flat with a fiber cutter and placed on both ends of the electrodes of the fusion splicer. At this time, we adjusted the fiber fusion splicer to automatic mode and performed discharge fusion splicing. The MMF-SCF fiber structure can be obtained. Next, the same section of MMF is taken, and the steps of stripping the coating and cutting the end face flat are repeated. Then the MMF-SCF and MMF are placed on the FSM for fusion splicing. In this way, we can obtain the MCM fiber structure. Thereafter, we tapered this structure using the CMS machine. CMS operates in a near-vacuum environment in the three-electrode mode, providing a heating mode at temperatures lower than the air discharge. It provides higher stability for the discharge arc and enables easy control of the taper profile. The fiber is stretched using a stretching stage, with the stretching length and speed controlled by a computer program. The surface roughness of the fiber is low, and the processing speed is fast. The key to the CMS tapering technique lies in the adjustment of the tapering parameters (electrode power, electrode scanning speed, motor moving speed, etc.). The best parameters were obtained through a large number of trials for optimized tapered fiber. The ability to pull a tapered fiber with high precision and good repeatability is well suited to the needs of sensing [40]. Firstly, it is necessary to tune the appropriate taper program and calibrate the left and right platforms. Then the optical fiber is placed into the clamp for fixation. Move the heating electrode to the position where the fiber needs to be tapered and click ‘run’. At this point, the CMS will automatically complete the tapering process. In this way, we can successfully prepare the final MCM sensor structure. The fusion splicer schematic and the CMS three-electrode heating process are shown in Figs. 3(a) and (b). Figure 3(c) shows the manufacturing process diagram of the MCM structure.
Fig. 3. (a) Schematic of the fusion splicer machine (FSM), (b) actual image of the three-electrode heating process inside the CMS machine, (c) schematic diagram of the fabrication process of the MCM-based tapered fiber structure.
2.6 Gold nanoparticles, WS2-thin layer and C3N-quantum dots synthesis process
AuNPs with a particle size of 10 nm were synthesized using the conventional Turkevich method [41]. AuNPs solution was synthesized by using tetrachloroauric acid and trisodium citrate. The aqueous solution containing tetrachloroauric acid was heated to boiling, and then trisodium citrate was added to this solution. The final solution was stirred and continued to be heated for 15 min until its color changed to burgundy. 10 mL of C3N-QDs were successfully prepared by placing C3N-QDs (1 mg/mL) into ethanol and then sonicating it for 2 hours with a sonicator. WS2 thin-layer solution is directly used, as purchased. WS2 is a 2D transition metal sulfide material. In biosensors, a thin layer of WS2 can be used as a substrate material or a modification layer, which can enhance the sensitivity of signal detection due to its high specific surface area. The LSPR effect of AuNPs can enhance the intensity of optical signals. By binding specifically to the target biomolecule, the sensitivity and detection limit of the sensor can be improved. C3N-QDs are a novel carbon-based material, and combining them with AuNPs can achieve the optical enhancement effect. This can improve the performance of the biosensor.
2.7 Nanomaterials immobilization and xanthine oxidase enzyme functionalization
Before immobilizing the NPs on the surface of the optical fiber, the fiber needs to be cleaned. First, in order to remove the organic impurities from the surface of the optical fiber, it was kept in an acetone solution for 20 minutes. Then, the optical fiber was immersed in a mixture of concentrated sulfuric acid and 30% hydrogen peroxide solution (Piranha solution), with the volume ratio of the piranha solution being 7:3, for 30 minutes. Due to its strong oxidizing property, the silica hydroxyls on the surface of the optical fiber will be fully exposed after the treatment in order to facilitate the subsequent immobilization of NPs. Finally, the treated fiber was rinsed with deionized water and dried in an oven at 70 °C for 30 min. Next, we performed the immobilization of NPs, which can be divided into three steps. Step 1: Soak the sensing region of the optical fiber in WS2 solution for 10 minutes, and then dry it in an oven for 20 minutes, repeated three times, to get a uniformly thin layer of WS2 on the probe surface. Step 2: The probe was placed in an ethanol solution with a concentration of 1% MPTMS for 12 hours. Then, in order to remove the unbound MPTMS monomer from the fiber surface, the probe was rinsed with anhydrous ethanol and dried with nitrogen gas. The MPTMS-coated fiber was then immersed in a solution of AuNPs for 48 hours. MPTMS is a multifunctional silane that reacts with the silica hydroxyl groups on the fiber surface. The hydrophobic group at one end of the silane is covalently connected to AuNPs, which can immobilize AuNPs very well. Finally, the fiber was rinsed with ethanol and dried with nitrogen gas to remove the unbound AuNPs. Step 3: Place the probe in the prepared C3N-QDs for 10 min, dry it in the oven for 20 min, and repeat this step three times.
Before performing enzymatic functionalization of the optical fibers, the optical fibers with immobilized NPs were rinsed with DI water. Then, it was placed in 5 mL of MUA solution (0.5 mM) for 5 hours. The MUA solution was used to modify the surface of the optical fiber with immobilized NPs with carboxyl groups. Next, the optical fiber was placed in a mixture of 10 ml of EDC (200 mM) and NHS (50 mM) for 30 min to activate the carboxyl groups. EDC can transform the carboxyl groups into amino-active intermediates and undergo a condensation reaction with the amino groups in the oxidase. Due to the instability of the amino-active intermediate, hydrolysis and reconversion of the carboxyl group occur. Mixing EDC and NHS produces amino-active NHS lipids. It will not hydrolyze and can improve the efficiency of functionalizing XO on the surface of the optical fiber. Finally, the optical fiber was immersed in XO solution for 12 h, so that the amino group in the oxidase formed a strong covalent bond with the activated carboxyl group, and the functionalization of the enzyme was achieved. The immobilization of NMs on the fiber structure and the enzyme functionalization process of XO are shown in Fig. 4. In addition, the transmission intensity of the fiber optic probes before and after functionalization were compared. As shown in Fig. 5, the transmission intensity decreased after functionalization with NPs.
Fig. 4. Schematic of the NMs-immobilization and xanthine oxidase enzyme functionalization over the MCM-based taper optical fiber structure.
Fig. 5. Comparison of transmitted intensity between bare fiber probe and fiber with different NMs-functionalization.
2.8 Preparation of xanthine solutions
Xanthine is a group of compounds based on purine-base derivatives. Xanthine is a product of purine metabolism and is converted to uric acid by the XO enzyme. Therefore, the level of xanthine is also an important indicator of some pathological conditions. In this work, xanthine solutions with concentrations of 100 µM, 200 µM, 400 µM, 700 µM, and 900 µM were prepared using PBS buffer. Firstly, a 900 µM stock solution of xanthine was prepared and then diluted with PBS to get the other concentrations. Measuring the sample concentrations from low to high will reduce the error in the experiment.
2.9 Experimental setup
The light source used in this experiment was a tungsten-halogen source (HL-2000) that transmitted the optical signal (in the range of 200 - 1000 nm) through a sensing probe. The other end of the optical fiber was connected to a spectrometer with a spectral range of 200 - 1000 nm. The spectrometer was then connected to a computer that was used to analyze the received signal. When the light signal passed through the probe, the immobilized NPs underwent a resonance reaction. The XO enzyme immobilized on the nanomaterial combined with the xanthine, which led to a change in the surrounding RI. At this point, the spectrometer received a different LSPR peak wavelength, which provided the wavelength drift. The concentration of the prepared xanthine solution was measured from low to high, and corresponding spectral data was recorded. After recording, the probe surface needed to be cleaned with PBS solution and dried to remove the previously bound xanthine biomolecules. Figure 6 shows the principle of xanthine detection and the effective experimental setup used for the detection of different xanthine concentrations.
Fig. 6. Experimental setup for the detection of xanthine solutions using the developed WaveFlex biosensor.
3. Results and discussions
3.1 Optimization of sensor probe
In this work, we first explored the sensing potential of single-mode fiber (SMF), photosensitive fiber (PSF), and seven-core fiber (SCF) for the WaveFlex biosensor. Three fiber structures were fabricated using multimode fibers (MMFs) at both ends and SMF, PSF, and SCF in the middle part, respectively. That formed the MMF-SMF-MMF (MSM) structure, the MMF-PSF-MMF (MPM) structure, and the MMF-SCF-MMF (MCM) structure. Then their transmitted intensity spectra were tested, as shown in Figs. 7(a), (b), and (c). Three fiber structures were prepared for each of the three probes, and their average transmitted intensity was measured and calculated. The measured results are shown in Fig. 7(d). It can be seen that the MCM-based fiber structure has a lower transmitted intensity. The lower the transmission intensity, the higher the transmission depth of the EWs present on the surface of the fiber, and the stronger the evanescent field in the corresponding taper waist region. As a result, the sensing region becomes more sensitive to the surrounding medium. This is consistent with the simulated results shown in Fig. 2. Therefore, we chose the MCM structure to complete the subsequent experiments and develop the WaveFlex biosensor. The diameter of the tapered region can affect the characteristics of the tapered fiber structure; the smaller the waist diameter, the lower the transmission intensity. This indicates that the light in the tapered region penetrates into the cladding and that the evanescent field is stronger. Although the small diameter of the taper region can improve the sensitivity of sensing, the smaller the diameter, the easier it is to break during the functionalization of the probe. Therefore, we selected a probe with a diameter of 40 µm to complete the experiment. We prepared five sensors based on the MCM structure and scanned their diameters as shown in Fig. 8(a). In addition, their transmission intensities were measured simultaneously. As shown in Fig. 8(b), the MCM sensor structure has high reproducibility and can be used well for sensing purposes.
Fig. 7. Transmitted intensity measurement through (a) MMF- SMF - MMF (MSM), (b) MMF- PSF - MMF (MPM), (c) MMF- SCF - MMF (MCM), and (d) comparison of transmitted intensity of MSM, MPM and MCM-based fiber structures.
Fig. 8. Repeatability analysis of sensor structure, (a) diameter measurement, and (b) transmitted intensity measurement using a 40 µm MCM-based WaveFlex sensor structure.
3.2 Characterization of nanomaterials
HR-TEM was used to observe the distribution and shape of AuNPs, as shown in Fig. 9(a). The histogram of synthesized AuNPs is shown in Fig. 9(b), and the results show that the average size of the AuNPs is around 10 nm. It can be seen that the synthesized NPs are spherical and of uniform size, which can better excite the LSPR phenomenon. Since the size and shape of AuNPs have a direct influence on the sensing results, the synthesized AuNPs solution was characterized by a UV-visible spectrophotometer, and the absorption spectra are shown in Fig. 9(c). The resonance wavelength of the absorption peak was 519 nm, which indicated that the synthesized AuNPs possessed a diameter of about 10 nm. Meanwhile, we characterized the images of C3N-QDs and WS2 thin layers, as shown in Figs. 9(d) and (e), respectively. This makes it all the more beneficial for us to carry out the immobilization of NMs.
Fig. 9. Characterization of nanomaterials, (a) HR-TEM image -, (b) histogram of -, (c) absorbance spectrum of AuNPs, (d) TEM image of C3N-QDs, and (e) TEM image of the WS2-thin layer.
3.3 Characterization of nanomaterials-immobilized structure
The structure of the fiber and the immobilization of NMs on the fiber surface were observed using SEM. First, Fig. 10(a) shows the SEM of the MCM fiber structure, and the cylindrical probe after tapering can be clearly seen. We characterized the immobilization of the WS2-thin layer, AuNPs, and C3N-QDs on the probe surface, and the SEM results are shown in Figs. 10(b) and (c). It can be seen that the AuNPs are uniformly distributed on top of the thin layer of WS2, and C3N-QDs are immobilized on the surface of the AuNPs in the form of dots. The sensing region was measured using energy spectroscopy (EDX). Figure 10(d) shows the presence of C, Au, N, W, and S elements on the probe surface. This confirms that the NMs are well immobilized.
Fig. 10. (a) MCM-based WaveFlex biosensor, (b), (c) WS2/AuNPs/C3N-QDs-immobilized optical fiber sensor probe, (d) EDX of a NMs-coated sensor structure.
3.4 Measurement of analytes
We prepared different concentrations of xanthine solutions and measured them using the experimental setup shown in Fig. 6. First, we added the solution of 100 µM concentration to the reaction tank and recorded stable LSPR spectra. Here, the linear detection range of the sensor was 100-900 µM, and we used sequential measurements from low to high concentrations. Then, in order to avoid cross-influence between solutions of different concentrations, the sensing probe was washed several times with PBS solution and dried to remove residual molecules on the probe surface before the measurement of the next concentration. To minimize experimental errors, three fiber-optic probes were selected to measure all concentrations of xanthine solution, and the above steps were repeated. Then the LSPR spectra were averaged and normalized as shown in Fig. 11(a). It can be seen that the peak wavelength of the spectrum shifts to higher wavelengths as the concentration of xanthine increases. This indicates that the XO immobilized on the surface of the probe reacts with the measured xanthine solution, which affects the change in RI and leads to a red-shift in the spectra. The linear fit curve for the peak wavelength is shown in Fig. 11(b), and the linear fit degree of the probe is 0.98. The linear fit curve can be expressed as;
Here, λ is the peak wavelength of the LSPR spectrum, and c is the concentration of the xanthine solution.
The sensitivity of the sensor probe is the slope of the linear fit curve. Also, the LoD is an important parameter of the sensor and is calculated as:
In Eq. (6), in order to calculate the standard deviation (SD) value, the PBS solution was measured ten-times with the same sensor probe. It can be concluded that the LoD of the sensor is 96.75 µM.
3.5 Stability and pH test
The stability test was performed to examine whether the position of the initial wavelength of the sensor remained stable for a certain period of time during multiple measurements. We tested the PBS solution ten times using the same sensor probe. First, the PBS solution was added, and after a stable LSPR spectrum was recorded, the probe was dried and PBS was added again. The above steps were repeated ten times, and the experimental results are shown in Fig. 12(a). It can be seen from the peak wavelength that the fluctuation of the peak line is very small, with a standard deviation (SD) of 0.103. The results show that the sensor has high stability.
The pH test was performed to study the xanthine fraction molecules’ solubility in different pH environments and to determine the most suitable solubilization environment for the probe. In this work, PBS solution was used as a solvent with a pH of 7.4. Different solvent environments with pH values of 5, 8, 11, and 13 were prepared with hydrochloric acid and sodium hydroxide. Xanthine solutions of 100 µM and 900 µM were prepared in the above solvent environments. The pH values were tested in order from low to high. Each test was completed, the corresponding LSPR spectra were recorded, and then the probe surface was cleaned with a blank solution of the same pH value and dried. Measuring the other solutions in the same way, we recorded the peak wavelength shift as shown in Fig. 12(b). There is a maximum drift in the solvent with a pH of 7.4, which indicates that the sensor probe performs best when using PBS solution as the solvent.
3.6 Reproducibility and reusability test
Reproducibility test is the measurement of the same concentration of xanthine solution with different sensor probes, which verifies that the fabrication of the sensor is reasonable. Here, we tested a 200 µM concentration of xanthine solution with three probes and recorded stable LSPR spectra. To minimize experimental error, the probes were rinsed with PBS and dried at the end of each test. The test results are shown in Fig. 13(a), where different probes have the same peak wavelength for the same concentration of solution, indicating that the sensor has good reproducibility.
Reusability is also an important performance indicator of the sensor, which shows the reusability of the sensor. We selected the same probe to measure xanthine solutions at 200 µM and 700 µM concentrations twice. First, 200 µM of xanthine was detected, and the LSPR spectra were saved. Then the probe surface was cleaned with a PBS solution and dried, and the second measurement of 200 µM concentration was performed. Following the above method, we measured the 700 µM concentration of xanthine twice. The experimental results are shown in Fig. 13(b). It can be seen that the peak wavelength of the same concentration is also the same, which indicates that the sensor can obtain high reusability.
3.7 Selectivity test
Selective testing was performed to investigate the specificity of the sensor for the measured molecules in a complex biological environment. Therefore, we selected five interfering biomolecules, sarcosine, creatinine, urea, glucose, and tryptamine, and prepared solutions at the lowest (100 µM) and highest (900 µM) concentrations using PBS buffer, respectively.
The peak wavelengths were measured, and LSPR spectra were recorded at constant room temperature to analyze the difference in peak wavelengths in the linear range (100 - 900 µM). The test results are shown in Fig. 14. It can be seen that the maximum value of peak wavelength shift occurs when the molecule measured is xanthine, while the wavelength shift of other biomolecules is smaller. This indicates a significant change in RI due to the specificity of the immobilized XO. The sensor has high selectivity for xanthine.
Fig. 14. Selectivity test of the XO/C3N-QDs/AuNPs/WS2-immobilized tapered MCM-based WaveFlex biosensor
3.8 Evaluation of sensor performance
In this work, we have developed a WaveFlex biosensor based on LSPR for the detection of xanthine. The performance of the developed sensor is compared with the existing sensors in terms of the use of NMs, sensing mechanism, linear range, sensitivity, and LoD. The developed sensing mechanisms (e.g., electrochemical, fluorescence, and colorimetric) have a few drawbacks, such as high manufacturing costs and a small linear range. Table 1 shows the comparison of sensor performance, and it can be seen that the proposed WaveFlex biosensors have a wide linear range and are highly reproducible, meaning they can provide reliable measurements over a wider operating range. In some areas, where the concentration of the target substance may vary over a wide range, the sensor is able to provide accurate measurements. This gives the sensor greater applicability. We believe that this advantage will help drive the development of sensor technology and provide more reliable and accurate measurement solutions for practical applications.
Table 1. Performance comparison of the proposed sensor with that of existing sensors.
4. Conclusion
In this study, we have successfully developed a novel WaveFlex biosensor based on a tapered-MCM fiber structure for the detection of xanthine. Experimental comparisons showed that the MCM sensor structure exhibited lower transmission intensity but a stronger evanescent field, leading to better excitation of the LSPR effect compared to MSM and MPM probes. We sequentially immobilized nanomaterials (WS2-thin layer, AuNPs, and C3N-QDs) on the probe surface and characterized them using a UV-visible spectrophotometer and HR-TEM. SEM was used to verify the immobilization of the fiber probe. The functionalization of XO confirmed the sensor's specificity for xanthine. The sensor exhibited stable LSPR spectra when testing different concentrations of xanthine solutions. We also conducted tests for reproducibility, reusability, stability, selectivity, and pH sensitivity. The results demonstrated the sensor's linear response within the concentration range of 100-900 µM, with a sensitivity of 3.2 nm/mM and a limit of detection of 96.75 µM. Our sensor shows promise for the detection of xanthine in both human samples and food applications, surpassing conventional biosensors in its potential utility.
Funding
National Natural Science Foundation of China (61905103, 62205136); Double-Hundred Talent Plan of Shandong Province, China; Special Construction Project Fund for Shandong Province Taishan Mountain Scholars; Liaocheng University (318052341); Natural Science Foundation of Shandong Province (ZR2020QC061, ZR2021QF025); Science and Technology Support Plan for Youth Innovation of Colleges and Universities of Shandong Province of China (2022KJ107).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. S. Gunasekaran, G. Sankari, and S. Ponnusamy, “Vibrational spectral investigation on xanthine and its derivatives—theophylline, caffeine and theobromine,” Spectrochim. Acta, Part A 61(1-2), 117–127 (2005). [CrossRef]
2. S. Azam, N. Hadi, N. U. Khan, and S. M. Hadi, “Antioxidant and prooxidant properties of caffeine, theobromine and xanthine,” Medical science monitor: international medical journal of experimental and clinical research 9, BR325–330 (2003).
3. H. M. Schmidt, S. Hahn, G. K. Annarapu, M. Carreo, and A. C. Straub, “Xanthine Oxidase Has a Protective Role during Heme Crisis By Binding and Degrading Heme,” Blood 136(Supplement 1), 12 (2020). [CrossRef]
4. A. S. Hernández-Cázares, M.-C. Aristoy, and F. Toldrá, “Hypoxanthine-based enzymatic sensor for determination of pork meat freshness,” Food Chem. 123(3), 949–954 (2010). [CrossRef]
5. D. A. Kostić, D. S. Dimitrijević, G. S. Stojanović, I. R. Palić, A. S. Đorđević, and J. D. Ickovski, “Xanthine oxidase: isolation, assays of activity, and inhibition,” J. Chem. 2015, 1–8 (2015). [CrossRef]
6. M. Legraverend, “Recent advances in the synthesis of purine derivatives and their precursors,” Tetrahedron 64(37), 8585–8603 (2008). [CrossRef]
7. A. Mehmood, M. Ishaq, L. Zhao, B. Safdar, A. U. Rehman, M. Munir, A. Raza, M. Nadeem, W. Iqbal, and C. Wang, “Natural compounds with xanthine oxidase inhibitory activity: A review,” Chem. Biol. Drug Des. 93(4), 387–418 (2019). [CrossRef]
8. Z. Miao, C. Li, Y. Chen, S. Zhao, Y. Wang, Z. Wang, X. Chen, F. Xu, F. Wang, and R. Sun, “Dietary and lifestyle changes associated with high prevalence of hyperuricemia and gout in the Shandong coastal cities of Eastern China,” J. Rheumatology 35, 1859–1864 (2008).
9. M. Dervisevic, E. Dervisevic, and M. Şenel, “Recent progress in nanomaterial-based electrochemical and optical sensors for hypoxanthine and xanthine. A review,” Microchim. Acta 186(12), 749 (2019). [CrossRef]
10. S. Hayman and W. Marcason, “Gout: is a purine-restricted diet still recommended?” J. Am. Diet. Assoc. 109(9), 1652 (2009). [CrossRef]
11. H. K. Choi, S. Liu, and G. Curhan, “Intake of purine-rich foods, protein, and dairy products and relationship to serum levels of uric acid: the Third National Health and Nutrition Examination Survey,” Arthritis Rheum. 52(1), 283–289 (2005). [CrossRef]
12. J. Ruiz-Jiménez, J. M. Mata-Granados, and M. D. Luque de Castro, “On-line automatic SPE-CE coupling for the determination of biological markers in urine,” Electrophoresis 28(5), 789–798 (2007). [CrossRef]
13. A. A. Ejaz, W. Mu, D.-H. Kang, C. Roncal, Y. Y. Sautin, G. Henderson, I. Tabah-Fisch, B. Keller, T. M. Beaver, and T. Nakagawa, “Could uric acid have a role in acute renal failure?” Clin. J. Am. Soc. Nephrol. 2(1), 16–21 (2007). [CrossRef]
14. D. Shan, Y. Wang, H. Xue, and S. Cosnier, “Sensitive and selective xanthine amperometric sensors based on calcium carbonate nanoparticles,” Sens. Actuators, B 136(2), 510–515 (2009). [CrossRef]
15. G. Xue, W. Yu, L. Yutong, Z. Qiang, L. Xiuying, T. Yiwei, and L. Jianrong, “Construction of a novel xanthine biosensor using zinc oxide (ZnO) and the biotemplate method for detection of fish freshness,” Anal. Methods 11(8), 1021–1026 (2019). [CrossRef]
16. C. Hong, L. Guan, L. Huang, X. Hong, and Z. Huang, “Colorimetric determination of xanthine with xanthine oxidase and WSe2 nanosheets as a peroxidase mimic,” New J. Chem. 45(23), 10459–10465 (2021). [CrossRef]
17. Z. Li, X. Liu, X.-H. Liang, J. Zhong, L. Guo, and F. Fu, “Colorimetric determination of xanthine in urine based on peroxidase-like activity of WO3 nanosheets,” Talanta 204, 278–284 (2019). [CrossRef]
18. H.-C. Hsu, P.-W. Liao, H.-T. Lee, W.-C. Liu, and M.-L. Ho, “Silver Nanoplates for Colorimetric Determination of Xanthine in Human Plasma and in Fish Meat via Etching/Aggregation/Fusion Steps,” Sensors 20(20), 5739 (2020). [CrossRef]
19. S. Choudhary, F. Esposito, L. Sansone, M. Giordano, S. Campopiano, and A. Iadicicco, “Lossy mode resonance sensors in uncoated optical fiber,” IEEE Sens. J. 23(14), 15607–15613 (2023). [CrossRef]
20. F. Esposito, A. Stancalie, A. Srivastava, M. Śmietana, R. Mihalcea, C. Neguț, S. Campopiano, and A. Iadicicco, “The impact of gamma irradiation on optical fibers identified using Long Period Gratings,” J. Lightwave Technol. 41(13), 4389–4396 (2023). [CrossRef]
21. F. Esposito, S. Campopiano, and A. Iadicicco, “Miniaturized strain-free fiber Bragg grating temperature sensors,” IEEE Sens. J. 22(17), 16898–16903 (2022). [CrossRef]
22. C. Nylander, B. Liedberg, and T. Lind, “Gas detection by means of surface plasmon resonance,” Sens. Actuators 3, 79–88 (1982). [CrossRef]
23. S. Shi, L. Wang, R. Su, B. Liu, R. Huang, W. Qi, and Z. He, “A polydopamine-modified optical fiber SPR biosensor using electroless-plated gold films for immunoassays,” Biosens. Bioelectron. 74, 454–460 (2015). [CrossRef]
24. S. Jia, C. Bian, J. Sun, J. Tong, and S. Xia, “A wavelength-modulated localized surface plasmon resonance (LSPR) optical fiber sensor for sensitive detection of mercury (II) ion by gold nanoparticles-DNA conjugates,” Biosens. Bioelectron. 114, 15–21 (2018). [CrossRef]
25. Z. Luo, Y. Wang, Y. Xu, X. Wang, Z. Huang, J. Chen, Y. Li, and Y. Duan, “Ultrasensitive U-shaped fiber optic LSPR cytosensing for label-free and in situ evaluation of cell surface N-glycan expression,” Sens. Actuators, B 284, 582–588 (2019). [CrossRef]
26. W. Zhang, R. Singh, Z. Wang, G. Li, Y. Xie, R. Jha, C. Marques, B. Zhang, and S. Kumar, “Humanoid shaped optical fiber plasmon biosensor functionalized with graphene oxide/multi-walled carbon nanotubes for histamine detection,” Opt. Express 31(7), 11788–11803 (2023). [CrossRef]
27. X. Liu, R. Singh, M. Li, G. Li, R. Min, C. Marques, B. Zhang, and S. Kumar, “Plasmonic sensor based on offset-splicing and waist-expanded taper using multicore fiber for detection of Aflatoxins B1 in critical sectors,” Opt. Express 31(3), 4783–4802 (2023). [CrossRef]
28. A. J. Haes, S. Zou, J. Zhao, G. C. Schatz, and R. P. Van Duyne, “Localized surface plasmon resonance spectroscopy near molecular resonances,” J. Am. Chem. Soc. 128(33), 10905–10914 (2006). [CrossRef]
29. H. Li, L. Xu, X. Huang, J. Ou-Yang, M. Chen, Y. Zhang, S. Tang, K. Dong, and L.-L. Wang, “Two-dimensional C3N/WS2 vdW heterojunction for direct Z-scheme photocatalytic overall water splitting,” Int. J. Hydrogen Energy 48(6), 2186–2199 (2023). [CrossRef]
30. Y. Zhao, X. Feng, M. Zhao, X. Zheng, Z. Liu, S. Yang, S. Tang, D. Chen, G. Wang, and G. Ding, “High-performance near-infrared photodetectors based on C3N quantum dots integrated with single-crystal graphene,” J. Mater. Chem. C 9(4), 1333–1338 (2021). [CrossRef]
31. P. Mishra, H. Kumar, S. Sahu, and R. Jha, “Flexible and Wearable Optical System Based on U-Shaped Cascaded Microfiber Interferometer,” Adv. Mater. Technol. 8(3), 2200661 (2023). [CrossRef]
32. M. S. Soares, L. C. Silva, M. Vidal, M. Loyez, M. Facão, C. Caucheteur, M. E. Segatto, F. M. Costa, C. Leitão, and S. O. Pereira, “Label-free plasmonic immunosensor for cortisol detection in a D-shaped optical fiber,” Biomed. Opt. Express 13(6), 3259–3274 (2022). [CrossRef]
33. X. Ning, C. L. Zhao, J. Yang, and C. C. Chan, “Zeolite thin film-coated spherical end-face fiber sensors for detection of trace organic vapors,” Opt. Commun. 364, 55–59 (2016). [CrossRef]
34. N. Agrawal, C. Saha, C. Kumar, R. Singh, B. Zhang, R. Jha, and S. Kumar, “Detection of L-cysteine using silver nanoparticles and graphene oxide immobilized tapered SMS optical fiber structure,” IEEE Sens. J. 20(19), 11372–11379 (2020). [CrossRef]
35. G. Zhu, L. Singh, Y. Wang, R. Singh, B. Zhang, F. Liu, B. K. Kaushik, and S. Kumar, “Tapered optical fiber-based LSPR biosensor for ascorbic acid detection,” Photonic Sens. 11(4), 418–434 (2021). [CrossRef]
36. X. Fu, J. Zhou, Z. Fu, M. Xu, S. Huang, W. Jin, G. Fu, and W. Bi, “A Multiparameter Sensor Based on Dumbbell-Shaped Double-Cladding Fiber Structure Cascaded Long Period Fiber Grating,” IEEE Sens. J. 22(14), 14118–14127 (2022). [CrossRef]
37. M. Li, R. Singh, M. S. Soares, C. Marques, B. Zhang, and S. Kumar, “Convex fiber-tapered seven core fiber-convex fiber (CTC) structure-based biosensor for creatinine detection in aquaculture,” Opt. Express 30(8), 13898–13914 (2022). [CrossRef]
38. A. J. Haes and R. P. Van Duyne, “A nanoscale optical biosensor: sensitivity and selectivity of an approach based on the localized surface plasmon resonance spectroscopy of triangular silver nanoparticles,” J. Am. Chem. Soc. 124(35), 10596–10604 (2002). [CrossRef]
39. O. Arrizabalaga, Q. Sun, M. Beresna, T. Lee, J. Zubia, J. Velasco Pascual, I. Sáez de Ocáriz, A. Schülzgen, J. E. Antonio-Lopez, and R. Amezcua-Correa, “High-performance vector bending and orientation distinguishing curvature sensor based on asymmetric coupled multi-core fibre,” Sci. Rep. 10(1), 14058 (2020). [CrossRef]
40. R. Singh, Z. Wang, C. Marques, R. Min, B. Zhang, and S. Kumar, “Alanine aminotransferase detection using TIT assisted four tapered fiber structure-based LSPR sensor: From healthcare to marine life,” Biosens. Bioelectron. 236, 115424 (2023). [CrossRef]
41. J. Turkevich, P. C. Stevenson, and J. Hillier, “A study of the nucleation and growth processes in the synthesis of colloidal gold,” Discuss. Faraday Soc. 11, 55–75 (1951). [CrossRef]
42. B. Zhi-Kun, Y.-T. Zhang, L. She-Hong, and L. Hong-Xia, “A flexible electrochemical sensor based on L-arginine modified chemical vapor deposition graphene platform electrode for selective determination of xanthine,” Chinese Journal of Analytical Chemistry 48(9), 1149–1159 (2020). [CrossRef]
43. B. Y. Sahyar, M. Kaplan, M. Ozsoz, E. Celik, and S. Otles, “Electrochemical xanthine detection by enzymatic method based on Ag doped ZnO nanoparticles by using polypyrrole,” Bioelectrochemistry 130, 107327 (2019). [CrossRef]
44. Y. Wang, H. Zhao, H. Song, J. Dong, and J. Xu, “Monodispersed gold nanoparticles entrapped in ordered mesoporous carbon/silica nanocomposites as xanthine oxidase mimic for electrochemical sensing of xanthine,” Microchim. Acta 187(10), 543 (2020). [CrossRef]
45. M. Wang, J. Zhang, X. Zhou, H. Sun, and X. Su, “Fluorescence sensing strategy for xanthine assay based on gold nanoclusters and nanozyme,” Sens. Actuators, B 358, 131488 (2022). [CrossRef]
46. X. An, Q. Tan, S. Pan, S. Zhen, Y. Hu, and X. Hu, “Determination of xanthine using a ratiometric fluorescence probe based on boron-doped carbon quantum dots and gold nanoclusters,” Microchim. Acta 189(4), 148 (2022). [CrossRef]
47. R. Kant, R. Tabassum, and B. D. Gupta, “Xanthine oxidase functionalized Ta2O5 nanostructures as a novel scaffold for highly sensitive SPR based fiber optic xanthine sensor,” Biosens. Bioelectron. 99, 637–645 (2018). [CrossRef]
