A Novel Ratiometric Photoelectrochemical Biosensor Based on Front and Back Illumination for Sensitive and Accurate Glutathione Sensing

The ratiometric detection method has a strong attraction for photoelectrochemical bioanalysis due to its high reliability and real-time calibration. However, its implementation typically depends on the spatial resolution of equipment and the pairing of wavelength/potential with photoactive materials. In this paper, a novel ratiometric photoelectrochemical biosensor based on front and back illumination was prepared for the detection of glutathione (GSH). Unlike traditional ratio methods, this ratiometric biosensor does not require voltage and wavelength modulation, thereby avoiding potential crosstalk caused by voltage and wavelength modulation. Additionally, the formation of a heterojunction between mTiO2 and Ag2S is conducive to enhancing light absorption and promoting charge separation, thereby boosting the photocurrent signal. Apart from forming a heterojunction with TiO2, Ag2S also shows a specific affinity towards GSH, thus enhancing the selectivity of the mTiO2/Ag2S ratiometric photoelectrochemical biosensor. The results demonstrate that the ratiometric photoelectrochemical biosensor exhibits a good detection range and a low detection limit for GSH, while also possessing significant interference elimination capability. The GSH detection range is 0.01–10 mmol L−1 with a detection limit of 6.39 × 10−3 mmol·L−1. The relative standard deviation of 20 repeated detections is 0.664%. Impressively, the proposed novel ratiometric PEC biosensor demonstrates enviable universality, providing new insights for the design and construction of PEC ratiometric sensing platforms.


Introduction
GSH is a ubiquitous non-protein thiol found in mammalian cells, consisting of glutamic acid, cysteine, and glycine.It plays a crucial role in various physiological processes, including the maintenance of redox homeostasis, the transmission of biological information, and immune response [1,2].Moreover, GSH is used as a biomarker for the early identification of many diseases, such as liver disease, Alzheimer's disease, and cancer [3][4][5].Therefore, it is of great significance for human health to detect changes in GSH levels with a simple, accurate, and sensitive method.In this regard, several analytical techniques have been developed for the detection of GSH, including high-performance liquid chromatography [6], photoelectrochemistry [7], fluorescence analysis [8,9], colorimetry [10,11], the electrochemical method [12,13], and the surface-enhanced Raman effect [14,15].Among these techniques, the photoelectrochemical method has been widely used in the field of bioanalytic chemistry because of its low background signal, high sensitivity, ease of equipment miniaturization, and cost-effectiveness.Various nanomaterials have been utilized to construct glutathione photoelectrochemical sensing platforms.For instance, an Au NPs/ZnO NRs array/FTO electrode demonstrated a detection limit of 2.7 µM within a concentration range of 10-800 µM, with a sensitivity of 0.06336 [16].Similarly, a Pt-IrO 2 /TiO 2 NTs/Ti electrode exhibited a detection limit of 0.8 µM and a sensitivity of 54.174 within the concentration range of 1-10 µM [17].
For typical photoelectrochemical (PEC) bioanalysis, quantitative detection refers to the process whereby a semiconductor is illuminated with light (excitation signal), resulting in the generation of charge carriers (electron-hole pairs).Subsequently, these charge carriers are captured by the electron acceptor or donor, causing the transfer of electrons to the electrode and generating a photocurrent signal (detection signal).However, in the traditional single-signal readout mode, even minor disturbances in the detection instrument and environment may result in unreliable results, particularly for the trace analysis of target substances in complex detection environments.To reduce the interference of external factors and improve the accuracy of trace analysis, ratiometric PEC sensors have been extensively developed [18][19][20].Unlike traditional single-signal sensors, ratio sensors exhibit a dualsignal response mode, where the ratio of two signals is used as the output signal instead of the absolute value of a single signal.The ratio sensor platform is constructed in two modes.In one mode, the target analyte induces a significant change in one signal as the response signal, while the other signal remains relatively unchanged as the reference signal.In another mode, the target analyte induces significant changes in both signals, both of which can serve as either the response or reference signal.Ultimately, a ratio-type sensor is constructed based on the ratio of these two signals.By self-calibrating the two output signals, the interference of the microenvironment and the detection system can be corrected, thereby improving the overall accuracy of the analysis.
Ratiometric photoelectrochemical sensors can be classified into three types: the potential-resolved, the spatial-resolved and the wavelength-resolved ratiometric photoelectrochemical sensors [21][22][23].The spatial-resolved technology relies on adjusting the light source irradiation position to generate different signals, which usually requires the different electrode areas to be modified with different materials and tags [24,25].To avoid interference from adjacent electrode tag diffusion, different regions are divided by insulating and hydrophobic materials or individually connected to multichannel instruments using conductive wires.The fabrication process is complex, and the partition material occupies most of the space of the sensor, severely limiting the electrode density and making high-resolution measurements difficult.Importantly, multi-area modification reduces the repeatability of the sensor, resulting in potential errors.Wavelength-resolved technology relies on light-absorbing active materials under different wavelengths.To avoid signal interference caused by multiple materials absorbing the same wavelength simultaneously, multiple light-absorbing materials with non-overlapping absorption regions need to be selected.Finding corresponding light-active materials to achieve ratiometric photoelectrochemical determination is still a huge challenge [26].On the other hand, different wavelengths of light require the combination of multiple filters, which may cause uneven photon efficiency and induce errors [27].Potential-resolved technology requires the different electrode areas on the same substrate to be modified with different light-active materials.When the applied voltage changes, one material produces a photocurrent while the other material remains "silent".The electrochemical signals of the two materials do not overlap with each other, and this type of method is limited by material screening.Moreover, most light-active materials rely on high positive and negative voltages to make them "silent".High voltage inevitably brings a series of side effects, such as electrode passivation, baseline drift, and biological damage [28].To address the challenges of constructing traditional ratio-based sensors, an electronic transfer tunnel distance modulation strategy has been developed, albeit still relying on wavelength modulation.Recently, the illumination direction control of the photoanode has been demonstrated to influence the electron diffusion distance, which indicates that the transfer ability of interface charge carriers can be controlled by changing the direction of incident light irradiation on the electrode [29,30].
In light of the aforementioned findings, a novel ratiometric photoelectrochemical sensor based on front and back illumination has been developed for the accurate detection of GSH.By leveraging the differential charge carrier transport capabilities between front and back illumination, a ratio-based photoelectrochemical strategy has been implemented, eliminating the need for a light source or potential modulation and overcoming limitations associated with specific photoactive materials, thus demonstrating greater universality.A TiO 2 /Ag 2 S heterojunction structure has been designed to enhance photocurrent response and widen the disparity in charge carrier transport capabilities between front and back illumination configurations.Through the ratio determination of front-to-back illumination photocurrent signals (I F /I B ) for GSH detection, significant improvements in linear curve correlation and reduction in relative standard deviation have been achieved.This study not only offers a facile extension to the determination of other analytes but also expands the diversity in the design of ratiometric sensing strategies.

Apparatus
The surface morphologies of different samples were observed using a field emission scanning electron microscope (SEM, S-4800, Hitachi, Tokyo, Japan) equipped with an energy-dispersive X-ray spectrometer (EDS).X-ray diffractometer (XRD, Bruker AXS D8, Karlsruhe, Germany) was employed to explore the crystal structures of different samples.A fiber optic spectrometer (PG2000 pro, Shanghai Ideaoptics Company, Shanghai, China) was used to record the absorption spectra of the samples.The photoelectrochemical performance analysis was conducted in an electrochemical workstation (CHI660, Shanghai Chenhua Company, Shanghai, China), with a xenon lamp (Meiruichen MC-PF300C, Beijing, China) as the excitation light source.The simulated solar irradiation intensity was adjusted using a radiometer (CEL-FZ-A, China Education Au-light, Beijing, China).The standard three-electrode system was adopted for testing.The modified FTO, Ag/AgCl electrodes, and platinum net were working electrodes, reference electrodes, and pair electrodes, respectively.

Preparation of Initial Sols for F127-TiO 2
Tetrabutyl titanate was dissolved and stabilized in HCl aqueous solution through intense stirring at room temperature.After 15 min, the titanium dioxide precursor solution was slowly added drop by drop to the ethanol solution containing F127.The molar ratio of each component in titanium dioxide sol was as follows: TBOT:F127:HCl:H 2 O:EtOH = 1:0.005:1.7:10:24.Subsequently, the TiO 2 sol was stirred at room temperature and allowed to age for 3 h [31].The aged TiO 2 sol was then stored at a low temperature.

Preparation of mTiO 2 Thin Films
The FTO conductive glass substrate (2 cm × 2 cm) was cleaned with acetone, chloroform, ethanol, and water in sequence by ultrasonication for 10 min to ensure surface cleanliness.Then, F127-TiO 2 sol was spin-coated onto the cleaned FTO surface, with 100 µL of F127-TiO 2 sol applied onto the FTO conductive side, and spun for 30 s at 3000 rpm.After repeating the spin-coating process 4 times, the sample was annealed at 450 • C for 1 h at a heating rate of 5 • C/min in an air atmosphere.The sample prepared using TiO 2 sol containing F127 was named mTiO 2 , while the sample prepared using TiO 2 sol without F127 was named TiO 2 .

Fabrication of mTiO 2 /Ag 2 S
The mTiO 2 sample was first immersed in 0.1 M AgNO 3 solution for 30 s, rinsed with deionized water, and thereafter annealed at 65 • C for 3 min.Next, the sample was immersed in 0.1 M Na 2 S solution for 30 s to allow a displacement reaction with sulfur, followed by rinsing with methanol, and finally annealed at 65 • C for 1 min.The immersion steps were repeated 5 times.Finally, the sample was heated to 400 • C at a heating rate of 10 • C/min in nitrogen and then annealed for 30 min.

Photoelectrochemical Measurements
A three-electrode system was employed for the detection of GSH, with a modified FTO serving as the working electrode, a saturated Ag/AgCl electrode as the reference electrode, and a Pt electrode as the counter electrode.The actual surface area of the working electrode was determined by fixing it with epoxy resin, and an optical photograph containing the electrodes and a ruler was captured using a digital camera.Subsequently, the optical photograph of the electrodes and ruler was imported into ImageJ software (imagej 1.52 h) for processing and analysis to determine the actual area.The PEC measurement was performed using I-T curve testing, with a 0.1 M Tris-HCl buffer solution (pH = 7.4) containing 0.1 M sodium chloride and 0.05 M potassium chloride serving as the electrolyte.Xenon lamps, with a wavelength range spanning from 300 nm to 2500 nm, serve as the excitation light source.The incident light intensity was adjusted to 100 mW/cm 2 using a radiometer.A 0 V (V vs. Ag/Agcl) bias voltage was applied during all photoelectrochemical measurements.Electrochemical impedance spectroscopy (EIS) was performed under illumination conditions in the frequency range of 0.1 Hz to 100 kHz.

Characterization of the Materials
The preparation process of mTiO 2 /Ag 2 S is illustrated in Figure 1a.Firstly, a TiO 2 sol containing F127 is spin-coated onto an FTO substrate, followed by calcination to obtain mesoporous TiO 2 film, which is attributed to the decomposition of F127 during calcination resulting in mesoporous structure formation [31].As shown in Figure 1b,c, compared to FTO, the surface of mTiO 2 /FTO is denser and smoother, indicating the successful loading of mTiO 2 film on the FTO surface.The mesoporous structure of mTiO 2 is not only beneficial for the adsorption of photosensitizer Ag 2 S but also reduces the transfer distance of photo-generated charge carriers, facilitating the suppression of electron-hole recombination.Subsequently, the Ag 2 S film is prepared on the surface of mTiO 2 by the successive ionic layer adsorption and reaction (SILAR) method.As depicted in Figure 1d, Ag 2 S nanoparticles with diameters of 50-200 nm are observed distributed on the mTiO 2 surface.To further investigate the crystal phase structure of mTiO 2 /Ag 2 S, X-ray diffraction (XRD) characterization was performed.Figure 1e shows that the XRD pattern of mTiO 2 /Ag 2 S exhibits sharp and intense diffraction peaks, indicating good crystallinity.According to the JCPDS 00-021-1272 card, the diffraction peaks at 25.8 • and 48.5 • are attributed to the (101) and (200) planes of rutile phase TiO 2 , while according to the JCPDS 00-001-1151 card, the peaks at 28.5 • , 31.1 • , and 34.4 • are ascribed to the monoclinic phase Ag 2 S (110) plane, and the peaks at 36.7 • and 52.9 • are attributed to the monoclinic phase Ag 2 S (200) and (220) planes, respectively [32,33].These results demonstrate the successful preparation of mTiO 2 /Ag 2 S. Energy-dispersive X-ray spectroscopy (EDS) analysis in Figure 1f confirms the uniform distribution of Ti, O, Ag, and S on the substrate surface, providing further evidence for the successful preparation of mTiO 2 /Ag 2 S.
Biosensors 2024, 14, 285 5 of 12 the peaks at 36.7° and 52.9° are attributed to the monoclinic phase Ag2S ( 200) and (220) planes, respectively [32,33].These results demonstrate the successful preparation of mTiO2/Ag2S.Energy-dispersive X-ray spectroscopy (EDS) analysis in Figure 1f confirms the uniform distribution of Ti, O, Ag, and S on the substrate surface, providing further evidence for the successful preparation of mTiO2/Ag2S.

PEC Performance
The photoelectrochemical properties play a crucial role in the sensing performance of sensors.Therefore, the influence of mTiO2/Ag2S heterojunction formation on the photoelectric performance was investigated.The UV-visible absorption spectra in Figure 2a depict the light-capturing abilities of Ag2S, mTiO2 and mTiO2/Ag2S.It can be observed that TiO2 exhibits significant absorption capability in the ultraviolet region, with an absorption edge at approximately 390 nanometers, while Ag2S demonstrates remarkable absorption in both the ultraviolet and visible light regions [34,35].In contrast, upon Ag2S modification on the surface of mTiO2, there is a clear absorption in the visible light region.This is attributed to the strong visible light absorption of Ag2S and the effective improvement of the photogenerated carrier generation and separation by the formation of Type-II heterojunction between Ag2S and mTiO2 [36].To validate the promotion of electron-hole pair separation by the mTiO2/Ag2S heterojunction, Mott-Schottky tests were performed on the mTiO2, Ag2S, and mTiO2/Ag2S samples.In Figure 2b, both mTiO2 and Ag2S display positive-slope N-type semiconductor characteristics, with flat band potentials (Efb) of −0.38 and −0.52 V (V vs. NHE), respectively.Efb serves as an indicator of the internal field strength that drives charge transfer at the semiconductor/electrolyte interface.A higher Efb value indicates a higher chemical potential gradient formed from the semiconductor to the electrolyte, thereby ensuring rapid charge transfer at the semiconductor/electrolyte interface [37].Compared to mTiO2, mTiO2/Ag2S demonstrates a higher Efb value, which confirms that the Type-II heterojunction between mTiO2 and Ag2S promotes rapid charge

PEC Performance
The photoelectrochemical properties play a crucial role in the sensing performance of sensors.Therefore, the influence of mTiO 2 /Ag 2 S heterojunction formation on the photoelectric performance was investigated.The UV-visible absorption spectra in Figure 2a depict the light-capturing abilities of Ag 2 S, mTiO 2 and mTiO 2 /Ag 2 S. It can be observed that TiO 2 exhibits significant absorption capability in the ultraviolet region, with an absorption edge at approximately 390 nanometers, while Ag 2 S demonstrates remarkable absorption in both the ultraviolet and visible light regions [34,35].In contrast, upon Ag 2 S modification on the surface of mTiO 2 , there is a clear absorption in the visible light region.This is attributed to the strong visible light absorption of Ag 2 S and the effective improvement of the photogenerated carrier generation and separation by the formation of Type-II heterojunction between Ag 2 S and mTiO 2 [36].To validate the promotion of electron-hole pair separation by the mTiO 2 /Ag 2 S heterojunction, Mott-Schottky tests were performed on the mTiO 2 , Ag 2 S, and mTiO 2 /Ag 2 S samples.In Figure 2b, both mTiO 2 and Ag 2 S display positive-slope N-type semiconductor characteristics, with flat band potentials (E fb ) of −0.38 and −0.52 V (V vs. NHE), respectively.E fb serves as an indicator of the internal field strength that drives charge transfer at the semiconductor/electrolyte interface.A higher E fb value indicates a higher chemical potential gradient formed from the semiconductor to the electrolyte, thereby ensuring rapid charge transfer at the semiconductor/electrolyte interface [37].Compared to mTiO 2 , mTiO 2 /Ag 2 S demonstrates a higher E fb value, which confirms that the Type-II heterojunction between mTiO 2 and Ag 2 S promotes rapid charge transfer at the interface.This conclusion is also supported by EIS Nyquist plots, where the diameter of the semicircle is directly related to the interface charge transfer resistance (R ct ).A smaller semicircle diameter indicates lower electrode charge transfer resistance.As shown in Figure 2c, the R ct of mTiO 2 is lower than that of TiO 2 , possibly due to the porous structure providing abundant reaction sites or enhanced light absorption.A significantly reduced R ct is observed for mTiO 2 /Ag 2 S compared to mTiO 2 .To further confirm the significantly improved photoelectrochemical performance of the mTiO 2 /Ag 2 S composite material compared to TiO 2 , the PEC activity of TiO 2 , mTiO 2 , and mTiO 2 /Ag 2 S photoelectrodes was tested through I-t experiments.According to Figure 2d, the photocurrent signal of wide bandgap mTiO 2 is relatively low due to its limited light absorption range under illumination.However, after Ag 2 S modification, the photocurrent signal is significantly enhanced.These results validate that the mesoporous TiO 2 and the mTiO 2 /Ag 2 S Type-II heterojunction structure effectively enhance light absorption and suppress electron-hole pair recombination, thereby improving the photoelectrochemical performance.

PEC Ratio Sensing Mechanism of GSH
To investigate the sensing mechanism of the TiO 2 /Ag 2 S ratiometric photoelectrochemical sensor, the band gaps, conduction band positions, and valence band positions of mTiO 2 and Ag 2 S need to be determined.The bandgaps (E g ) of mTiO 2 and Ag 2 S semiconductors were calculated using the Tauc equation: where α is the absorption coefficient, h is the Planck constant, ν is the light frequency, A is the proportionality constant, E g is the band gap, and n is decided by the transition type of the semiconductor [38].As shown in Figure 3a,b, the bandgaps of mTiO 2 and Ag 2 S are 3.24 eV and 1.21 eV, respectively.In the N-type semiconductor, the gap between the conduction band edge and the flat band edge is approximately 0.1 V [39].The valence band positions were estimated according to the following formula: portion of the light is absorbed by FTO, leading to a reduction the photoactive material.Additionally, the FTO and TiO2 lay nesses, which impede the transmission of light to Ag2S.Thes result in a decreased light absorption efficiency of the photoelec ing a relatively weaker output of the photoelectric current sign ometric PEC sensor is developed for the quantitative detection ferent photocurrent responses generated by the front and back  The E VB represents the conduction band potential, the E CB is the valence band potential, and the E fb is the flat band potential.The E fb values of mTiO 2 and Ag 2 S are determined to be −0.38 V and −0.52 V (vs.NHE), respectively, based on the Mott-Schottky (M-S) curves shown in Figure 2b.The valence band positions of mTiO 2 and Ag 2 S are calculated to be 2.76 V and 0.59 V (vs.NHE), respectively, according to Equation (2).Building upon these results, a potential detection mechanism of GSH is established using the mTiO 2 /Ag 2 S PEC sensor, as illustrated in Figure 3c.The mTiO 2 /Ag 2 S system forms a type-II heterojunction, which broadens the light absorption range and significantly enhances the efficiency of photo-induced charge carrier separation.Under illumination, the two photoactive materials (mTiO 2 and Ag 2 S) are excited, with electrons in the valence band of mTiO 2 and Ag 2 S being excited to the conduction band, therefore generating holes in the valence band.Due to the formation of the heterojunction, photogenerated electrons rapidly transfer from the conduction band of Ag 2 S to the conduction band of TiO 2 through the heterojunction structure, subsequently transferring to the external circuit.Meanwhile, photogenerated holes transfer from the valence band of TiO 2 to the valence band of Ag 2 S. The matching of band structures results in the formation of an internal electric field, leading to enhanced separation of electrons and holes.The PEC response mechanism for GSH is based on the acceleration of charge transfer by the electron donor.On one hand, Ag 2 S exhibits a specific affinity for GSH-containing thiol groups.On the other hand, GSH acts as a scavenger for photogenerated holes, effectively reducing the electron-hole recombination rate and thereby amplifying the photocurrent response.Figure 3d illustrates the principle behind the construction of the ratio-type PEC sensing platform.When incident light illuminates the front side of the electrode (the left part of Figure 3d), both TiO 2 and Ag 2 S can be simultaneously excited.The formation of a heterojunction between TiO 2 and Ag 2 S greatly enhances the efficiency of charge separation, resulting in a relatively strong output of the photoelectric current signal from the photoelectrode.However, when incident light irradiates the back side of the electrode (the right part of Figure 3d), several factors come into play.Firstly, due to the strong absorption capability of FTO in the ultraviolet region, a portion of the light is absorbed by FTO, leading to a reduction in the light absorption of the photoactive material.Additionally, the FTO and TiO 2 layers possess certain thicknesses, which impede the transmission of light to Ag 2 S.These dual factors collectively result in a decreased light absorption efficiency of the photoelectrode, consequently yielding a relatively weaker output of the photoelectric current signal.Therefore, a novel ratiometric PEC sensor is developed for the quantitative detection of GSH based on the different photocurrent responses generated by the front and back illumination of GSH.

Optimization of Experimental Conditions
To enhance the performance of the PEC sensor for detecting GSH, experimental conditions were optimized for the concentrations of Ag 2 S, the molar mass of F127, and the spin coating speeds of TiO 2 sol.All PEC tests of the optimization process were performed under a light intensity of 100 mW/cm 2 and a 0 V bias voltage.The optimal conditions were determined based on the photocurrent intensity value of mTiO 2 /Ag 2 S. In Figure 4a, the effect of varying Ag 2 S concentration was studied.The photocurrent intensity initially increases and then decreases with increasing Ag 2 S concentration.Notably, the photocurrent reaches its peak when the Ag 2 S concentration is 0.1 M. The reduction in photocurrent intensity observed at high concentrations of Ag 2 S may potentially be attributed to the agglomeration phenomenon of Ag 2 S. As shown in Figure 4b, the photocurrent response of the prepared photoelectrode is maximized when the amount of F127 added to the TiO 2 sol is 5 mmol.The introduction of the F127 pore-forming agent increases the specific surface area of TiO 2 and provides more binding sites for Ag 2 S loading.This favors the formation of more heterojunctions, which in turn accelerate charge transfer and enhance light absorption.However, excessive F127 may result in too many surface defects acting as composite centers, which increase the electron-hole recombination efficiency and possibly cause a decrease in photocurrent signal [40].The photocurrent intensity is impacted by the spin-coating speeds, as it affects the uniformity and loading amount of TiO 2 on the FTO surface.At low spin-coating speeds, an excessive loading of TiO 2 causes TiO 2 accumulation, producing an increasing number of surface composite centers.As the surface composite centers proliferate, the diffusion resistance of electron motion increases, resulting in a decrease in photocurrent intensity.However, higher spin-coating speeds also contribute to a decrease in photocurrent, as the lower loading amount of TiO 2 leads to overall insufficient light absorption.As depicted in Figure 4c, the photocurrent initially increases and then decreases with the spin-coating speeds varying, with the maximum photocurrent observed at a spin-coating speed of 3500 rpm.

Optimization of Experimental Conditions
To enhance the performance of the PEC sensor for detecting GSH, experimental co ditions were optimized for the concentrations of Ag2S, the molar mass of F127, and t spin coating speeds of TiO2 sol.All PEC tests of the optimization process were perform under a light intensity of 100 mW/cm 2 and a 0 V bias voltage.The optimal conditions w determined based on the photocurrent intensity value of mTiO2/Ag2S.In Figure 4a, t effect of varying Ag2S concentration was studied.The photocurrent intensity initially creases and then decreases with increasing Ag2S concentration.Notably, the photocurre reaches its peak when the Ag2S concentration is 0.1 M. The reduction in photocurrent tensity observed at high concentrations of Ag2S may potentially be attributed to the a glomeration phenomenon of Ag2S.As shown in Figure 4b, the photocurrent response the prepared photoelectrode is maximized when the amount of F127 added to the Ti sol is 5 mmol.The introduction of the F127 pore-forming agent increases the specific s face area of TiO2 and provides more binding sites for Ag2S loading.This favors the f mation of more heterojunctions, which in turn accelerate charge transfer and enhance lig absorption.However, excessive F127 may result in too many surface defects acting composite centers, which increase the electron-hole recombination efficiency and possib cause a decrease in photocurrent signal [40].The photocurrent intensity is impacted the spin-coating speeds, as it affects the uniformity and loading amount of TiO2 on t FTO surface.At low spin-coating speeds, an excessive loading of TiO2 causes TiO2 acc mulation, producing an increasing number of surface composite centers.As the surfa composite centers proliferate, the diffusion resistance of electron motion increases, resu ing in a decrease in photocurrent intensity.However, higher spin-coating speeds also co tribute to a decrease in photocurrent, as the lower loading amount of TiO2 leads to over insufficient light absorption.As depicted in Figure 4c, the photocurrent initially increa and then decreases with the spin-coating speeds varying, with the maximum photoc rent observed at a spin-coating speed of 3500 rpm.

Assay Performance of the PEC Sensor
Under the aforementioned optimized conditions, the analytical performance of t mTiO2/Ag2S photoelectrochemical sensor for GSH was evaluated by measuring the curves of the photoelectrode with different concentrations of GSH in 0.1 mM Tris (pH 7.4) solution.Figure 5a,d

Assay Performance of the PEC Sensor
Under the aforementioned optimized conditions, the analytical performance of the mTiO 2 /Ag 2 S photoelectrochemical sensor for GSH was evaluated by measuring the I-T curves of the photoelectrode with different concentrations of GSH in 0.1 mM Tris (pH = 7.4) solution.Figure 5a,d show the variation curves of front illumination photocurrent (I F ) and back illumination photocurrent (I B ) when 0.01-10 mM GSH was added to 0.1 M Tris (pH = 7.4) solution.Observations reveal that both I F and I B witness a marked increase with the rise in GSH concentration, with I F being greater than I B at any given concentration of GSH.Such findings provide supporting evidence for the validity of the mTiO 2 /Ag 2 S ratiometric photoelectrochemical sensor's proposed design.Figure 5b,e present the linear fitting curves for both I F and I B concerning GSH concentration.A positive correlation between photocurrent density and GSH concentration can be observed.Furthermore, within the GSH concentration range of 0.01-10 mM, two-segment linear relationships are exhibited between GSH concentration and both I F and I B .When the concentration of glutathione is at a low water level, the local concentration on the electrode surface is rapidly depleted as glutathione is converted into a product, resulting in a high sensitivity of the photocurrent response.When the concentration of glutathione is high, diffusion of glutathione is hindered, resulting in decreased sensitivity [40].Under front illumination conditions, two linear regression equations are derived as follows: I F1 = 1.29C + 2.21 (R 2 = 0.939) and I F2 = 0.012C + 2.45 (R 2 = 0.900), with a detection limit (LOD) of 9.33 × 10 −3 mmol•L −1 (S/N = 3).Under the back illumination condition, the two linear regression equations are I B1 = 1.41C + 1.60 (R 2 = 0.923) and I B2 = 1.74C + 0.02 (R 2 = 0.940), with a LOD of 8.01 × 10 −3 mmol•L −1 .The above results indicate that both front and back illumination photocurrent responses exhibit low detection limits for GSH sensing, with the latter having lower detection limits due to its lower background signal compared to the former [41].It is worth noting that the linear relationships between GSH concentration and both I F and I B are poor and can not satisfy the accurate measurement of GSH.The dual photocurrent signals obtained from front-/back-illuminated electrodes can be ratioed, which provides self-calibration and improves the reliability of the results.A strong linear relationship between I F /I B and GSH concentration is observed, as shown in Figure 6c.Two linear regression equations were derived, namely I F1 /I B1 = 0.41C + 1.32 (R 2 = 0.992) and I F2 /I B2 = 0.0035C + 1.36 (R 2 = 0.9931).The detection limit is determined to be 6.39 × 10 −3 mmol•L −1 .These results indicate that the ratio-based sensor displays excellent linear correlation with GSH concentration, thereby effectively reducing errors and achieving satisfactory accuracy in biosensing applications.To evaluate the performance of the sensing system, the anti-interference capability of GSH detection was investigated by the biosensor.0.1 mM GSH solution was prepared, and common interfering substances including glycine, glutamic acid, and arginine were added at a concentration of 0.5 mM for detection.No significant change in photocurrent was observed, as depicted in Figure 6f, indicating the excellent anti-interference ability of the mTiO 2 /Ag 2 S-based photoelectrochemical sensing system.The stability of the photoelectrode is also an important parameter for the sensor.To assess the stability of the mTiO 2 /Ag 2 S ratio-based photoelectrochemical sensor, the mTiO 2 /Ag 2 S electrode was subject to chopped light irradiation for 400 s in a Tris solution (pH = 7.4) containing 0.1 mM GSH.It can be seen from Figure 6a,b that both I F and I B exhibit stable and reproducible photocurrents.Additionally, Figure 6c displays the photocurrent values of I F (a) and I B (b) from 20 chopped light cycles, with the corresponding relative standard deviations (RSD) calculated as 1.683% and 1.377%, respectively.Importantly, the RSD of I F /I B calculated from 20 chopped light cycles is 0.664%, significantly lower than that of the individual signals I F and I B .The repeatability of the sensor is also a crucial parameter.As illustrated in Figure 6d,e, the comparison of photocurrent signals from five parallel electrodes reveals that the signals remain relatively consistent, indicating good reproducibility of the sensor.Figure 6f presents the I F /I B values of the five parallel electrodes, with a calculated RSD of 1.135%.This RSD value is lower than the RSD of the front illumination photocurrent signal (2.192%) and the back illumination photocurrent signal (3.316%).These results collectively highlight the superiority of the ratio-based sensor.

Conclusions
In summary, a novel ratiometric PEC sensor has been successfully established for GSH detection based on the strategy of modulating the charge carrier transport capabilities through front and back illumination.By modifying the surface of mTiO2 with Ag2S to construct a type-II heterojunction, the photocurrent signal has been amplified to enhance the sensitivity of the sensor.Additionally, Ag2S provides specific recognition sites for detecting GSH, significantly improving the sensor's selectivity.Experimental results demonstrate that the ratiometric mTiO2/Ag2S sensor exhibits precise and sensitive detection ca-

Conclusions
In summary, a novel ratiometric PEC sensor has been successfully established for GSH detection based on the strategy of modulating the charge carrier transport capabilities through front and back illumination.By modifying the surface of mTiO2 with Ag2S to construct a type-II heterojunction, the photocurrent signal has been amplified to enhance the sensitivity of the sensor.Additionally, Ag2S provides specific recognition sites for detecting GSH, significantly improving the sensor's selectivity.Experimental results demonstrate that the ratiometric mTiO2/Ag2S sensor exhibits precise and sensitive detection ca-

Conclusions
In summary, a novel ratiometric PEC sensor has been successfully established for GSH detection based on the strategy of modulating the charge carrier transport capabilities through front and back illumination.By modifying the surface of mTiO 2 with Ag 2 S to construct a type-II heterojunction, the photocurrent signal has been amplified to enhance the sensitivity of the sensor.Additionally, Ag 2 S provides specific recognition sites for detecting

Figure 3 .
Figure 3. Band gap values of (a) mTiO 2 and (b) Ag 2 S. (c) The mechanism illustration of the mTiO 2 /Ag 2 S PEC sensor for GSH detection.(d) The ratiometric PEC sensing mechanism.
show the variation curves of front illumination photocurrent and back illumination photocurrent (IB) when 0.01-10 mM GSH was added to 0.1 M T (pH = 7.4) solution.Observations reveal that both IF and IB witness a marked increase w the rise in GSH concentration, with IF being greater than IB at any given concentration GSH.Such findings provide supporting evidence for the validity of the mTiO2/Ag2S ra

Figure 6 .
Figure 6.(a) Front illumination and (b) back illumination photocurrent stability of the mTiO2/Ag2Sbased PEC biosensor in the electrolyte containing 0.1 mM of GSH under intermittent visible light irradiation for 400 s (c) 20 times photocurrent ratio (IF/IB).(d) Front illumination and (e) back illumination photocurrent signal of five parallel photoelectrodes in the electrolyte containing 0.1 mM of GSH.(f) Five parallel photoelectrode signal ratio (IF/IB).

Figure 6 .
Figure 6.(a) Front illumination and (b) back illumination photocurrent stability of the mTiO2/Ag2Sbased PEC biosensor in the electrolyte containing 0.1 mM of GSH under intermittent visible light irradiation for 400 s (c) 20 times photocurrent ratio (IF/IB).(d) Front illumination and (e) back illumination photocurrent signal of five parallel photoelectrodes in the electrolyte containing 0.1 mM of GSH.(f) Five parallel photoelectrode signal ratio (IF/IB).

Figure 6 .
Figure 6.(a) Front illumination and (b) back illumination photocurrent stability of the mTiO 2 /Ag 2 Sbased PEC biosensor in the electrolyte containing 0.1 mM of GSH under intermittent visible light irradiation for 400 s (c) 20 times photocurrent ratio (I F /I B ).(d) Front illumination and (e) back illumination photocurrent signal of five parallel photoelectrodes in the electrolyte containing 0.1 mM of GSH.(f) Five parallel photoelectrode signal ratio (I F /I B ).