Bi 2 S 3 /rGO Composite Based Electrochemical Sensor for Ascorbic Acid Detection

: The engineering of an efﬁcient electrochemical sensor based on a bismuth sulﬁde/reduced graphene oxide (Bi 2 S 3 /rGO) composite to detect ascorbic acid (AA) is reported. The Bi 2 S 3 nanorods/ rGO composite was synthesized using a facile hydrothermal method. By varying the amount of graphene oxide (GO) added to the synthesis, the morphology and size of Bi 2 S 3 nanorods anchored on the surface of rGO can be tuned. Compared to a bare glassy carbon electrode (GCE), the GCE modiﬁed with Bi 2 S 3 /rGO composite presented enhanced electrochemical performance, which was attributed to the optimal electron transport between the rGO support and the loaded Bi 2 S 3 as well as to an increase in the number of active catalytic sites. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) analysis of Bi 2 S 3 /rGO/GCE demonstrate that the active Bi 2 S 3 /rGO layer on GCE plays an important role in the electrochemical behavior of the sensor. In particu-lar, the Bi 2 S 3 /rGO/GCE sensor shows a wide detecting range (5.0–1200 µ M), low detection limit (2.9 µ M), good sensitivity (268.8 µ A mM − 1 cm − 2 ), and sufﬁcient recovery values (97.1–101.6%) for the detection of ascorbic acid. writing—original H.L. validation, supervision, and visualization, and conceptualization,


Introduction
Ascorbic acid (AA), as an essential biological molecule in the human blood, is indispensable for various physiological and metabolic activities. Therefore, it is important to develop facile and sensitive techniques to accurately determine amounts of AA for application in the pharmaceutical, chemical, cosmetic, and food sectors. Compared with traditional analysis methods, such as fluorimetry [1], colorimetry [2], voltammetry [3,4], flow injection analyses [5], spectrophotometry [6], and liquid chromatography [7], the electrochemical detection of AA is considered to be a simple low-cost method owing to its easy operation, high sensitivity and low detection limit [8,9]. However, the direct oxidation of AA on bare electrodes results in electrode surface fouling, high overvoltage, low sensitivity, poor reproducibility, and high interference with other biomolecules [10]. Therefore, it is important to choose an appropriate active material to modify glassy carbon electrodes (GCE) improving their selectivity and sensitivity towards an accurate AA determination.
Following the development of nanomaterials in electrochemical sensors, various materials such as organic redox mediators [11], nanoparticles [12], polymers [13], carbon nanotubes [14], and graphene [15] have been used as redox-active sites to modify the standard electrode surface, reducing the activation energy barrier of the electrochemical reaction and speeding up the electron transfer process [16]. Recently, bismuth sulfide (Bi 2 S 3 ) has attracted much attention as the electrochemically active material in gas sensors, photoresponsive devices, and solar cells due to its unique electronic and optical properties [17][18][19][20][21]. However, the poor intrinsic conductivity and easy agglomeration of Bi 2 S 3 nanoparticles result in low sensitivity and narrow linear range in electrochemical detection limiting its practical applications [22]. Up to now, various strategies have been proposed to improve the electrochemical properties of Bi 2 S 3 . Graphene (GO) has been reported to be an ideal substrate material for the development of optimal nanocomposite sensing materials due to its surface properties and electronic transport properties [23,24]. Luo et al. reported on Cu/graphene nanocomposites as active layer for non-enzymatic glucose sensors, with good stability and selectivity [25]. Guo et al. [26] used MoS 2 -rGO/ITO composites as active materials in electrochemical sensors to simultaneously detect uric acid (UA) and dopamine (DA). Yan et al. [27] synthesized rGO/Bi 2 S 3 nanocomposites using thioacetamide as the sulfur source, and used these nanocomposites as the active layers with remarkable increases in the selectivity and sensitivity of the DA sensor.
In this work we describe a facile-yet-efficient strategy to engineer the structure of Bi 2 S 3 nanostructures. The correlation among graphene content, Bi 2 S 3 structure and electrochemical characteristics is evaluated. In addition, the Bi 2 S 3 /rGO modified glassy carbon electrode (GCE) is validated as active layer in electrochemical sensors for ascorbic acid (AA) detection. The fabricated Bi 2 S 3 /rGO/GCE as AA sensor presented a wide linear detection range, high sensitivity and good selectivity in the presence of various interfering compounds in aqueous phosphate buffer solutions. This work provided an additional step for improving the electrocatalytic activity of transition metal chalcogenides and expanding their applications in the field of electrochemical sensing.

Instruments
The morphology and crystal structure of the samples were determined by Environmental scanning electron microscope with a field emission gun (ESEM, Quanta FEG 250, Hillsboro, OR, USA) and transmission electron microscopy (TEM, Talos F 200 X, Waltham, MA, USA). The crystal phase structure and composition of Bi 2 S 3 in the composite were analyzed by X-ray diffractometer (XRD, D/Max2500pc, Akishima-shi, Tokyo, Japan), Raman spectroscopy (Horiba Evolution, Minami-ku, Kyoto, Japan), and X-ray photoelectron spectra (XPS, ESCALAB250Xi, Waltham, MA, USA). The molecular structure and chemical composition of graphene materials were determined by Fourier transform infrared spectrometer (FT-IR, Perkin Elmer, Waltham, MA, USA). Electrochemical studies were measured by CHI 660E electrochemical analyser (Shanghai Chenhua Instrument Co., Ltd., Shanghai, China) using a conventional three-electrode system.

Synthesis of Bi 2 S 3 /rGO Nanocomposites
Bi 2 S 3 samples were synthesized with different sulfur sources to determine the the sulfur source leading to nanostructures with best sensor activity. As shown in Figure  S1 Bi 2 S 3 the sample prepared with thiourea has a higher current response. This may be attributed to the higher specific surface area and presence of more active sites on Bi 2 S 3 nanostructures prepared with thiourea which can improve the electrochemical performance of the working electrode by accelerating electron transfer, compared with nanostructures synthesized with other sulfur sources.
Synthesis procedure: the GO was sonicated for 2 h in deionized water to form a solution of 1 mg mL −1 . Then, different amounts of Bi (NO 3 ) 3 ·5H 2 O were added to the GO solution and stirred for 30 min (solution A). Based on the weight ratio of Bi (NO 3 ) 3 ·5H 2 O:GO, the nanocomposites were named Bi 2 S 3 /rGO-1 (4:1), Bi 2 S 3 /rGO-2 (2:1), Bi 2 S 3 /rGO-3 (1:1) and Bi 2 S 3 /rGO-4 (1:2), respectively. The thiourea was dissolved in 10 mL deionized water (solution B), Bi (NO 3 ) 3 ·5H 2 O and thiourea in a molar ratio of 1:10. Next, solution B was slowly added into solution A with continuous stirring for 20 min. The resulting mixture was transferred into 100 mL Teflon lined stainless steel autoclave, sealed, and heated at 180 • C for 12 h in an electric oven, and the brownish-black products were centrifuged (at 7000 rpm for 5 min) and washed five times with deionized water and alcohol. Finally, the obtained Bi 2 S 3 /rGO products were dried for 5 h in a vacuum oven at 60 • C.

Sensor Fabrication
Firstly, the glassy carbon electrodes (GCEs) were carefully polished using Al 2 O 3 powders (0.5 mm, and 0.05 µm) on a polishing cloth, successively washed with ethanol and deionized water and dried at room temperature. Next, 10 mg of Bi 2 S 3 /rGO sample was dispersed in the mixture solution containing 5.0 mL of N, N-Dimethylformamide (DMF) and 100 µL of Nafion for ultrasonic treatment for 20 min, obtaining homogeneous Bi 2 S 3 /rGO composite suspension. Then, 5.0 µL of the above Bi 2 S 3 /rGO suspension was coated on surface of the cleaned GCE, after drying in hot air for 10 min, the sensor fabricated with Bi 2 S 3 /rGO/GCE was obtained. Likewise, for comparison, the similar sensor based on rGO/GCE and Bi 2 S 3 /GCE were prepared.

Characterization of Bi 2 S 3 /rGO
The morphology of the Bi 2 S 3 /rGO samples was characterized by SEM and TEM as shown in Figure 1. From SEM images (Figure 1a-d), it is clearly seen that well dispersed rod-like structures of Bi 2 S 3 are formed on the surface of rGO in all Bi 2 S 3 /rGO samples. Moreover, we observed that when the weight ratio of Bi (NO 3 ) 3 ·5H 2 O: GO decreased from 4:1 to 1:2, the average size of Bi 2 S 3 nanorods on the rGO sheets increased by three to five times. In contrast, pure Bi 2 S 3 samples prepared using different sulfur sources in the absence of GO ( Figure S2), are highly aggregated, indicating that GO plays an important role in dispersing nanoparticles during the hydrothermal synthesis. The interaction of Bi + with oxygen-containing groups of GO was reported to be responsible for the dispersibility of Bi 2 S 3 nanorods [29]. It is worth noting that using an appropriate amount of GO in the solution resulted in higher interfacial contact between rGO sheets and Bi 2 S 3 nanorods. As shown in Figure 1c, the Bi 2 S 3 rods are wrapped around the thin rGO sheets in Bi 2 S 3 /rGO-3, this sample also presented the best electrochemical performance. From TEM images of Bi 2 S 3 /rGO-3 shown in Figure 1e,f, the diameter of Bi 2 S 3 rods is about 50-120 nm and the length ranges from 0.4-2.3 µm. The Bi 2 S 3 rods are mostly randomly distributed on the surface of rGO. In addition, the elemental mapping ( Figure S3) further illustrates the good dispersion in the composite material, and the EDAX analysis indicates that the synthesized composite has high purity.  Figure 2a shows XRD patterns of the GO, rGO, Bi2S3, and Bi2S3/rGO composite materials. After the hydrothermal treatment in the presence of thiourea, the typical diffraction peak of GO sample at 2θ = 10.2° disappeared; and the presence of the broad peak at 23.9° corresponding to the (002) plane of rGO indicates the conversion of GO to rGO [30,31]. On the other hand, the diffraction peaks of all Bi2S3/rGO samples match with the typical peaks of Bi2S3 in the CPDS card (No. 17-0320), indicating that GO used in the synthesis promotes the crystal growth of Bi2S3. However, the intensity of the peaks of Bi2S3/rGO gradually decreased with increasing amounts of GO, which is probably due to the shielding effect of rGO on the Bi2S3 surface [32]. Moreover, no pronounced diffraction peak of rGO has been found in XRD patterns of Bi2S3/rGO composites, which can be attributed to the relatively low intensity of the rGO peak or the face-to-face stacking of Bi2S3 on both sides [33]. Figure 2b shows the typical Raman spectra of GO, Bi2S3, and Bi2S3/rGO-3 in the range of 150-2000 cm −1 . The low-intensity feature at 239 cm −1 and the peak at 974 cm −1 observed for pure Bi2S3 correspond to the Ag optical phonon modes and surface phonon modes [34]. Similar peaks of pure Bi2S3 and GO can be observed in the spectra recorded on the Bi2S3/rGO-3 sample, indicating that rGO and Bi2S3 are present in Bi2S3/rGO. The typical G  Figure 2a shows XRD patterns of the GO, rGO, Bi 2 S 3 , and Bi 2 S 3 /rGO composite materials. After the hydrothermal treatment in the presence of thiourea, the typical diffraction peak of GO sample at 2θ = 10.2 • disappeared; and the presence of the broad peak at 23.9 • corresponding to the (002) plane of rGO indicates the conversion of GO to rGO [30,31]. On the other hand, the diffraction peaks of all Bi 2 S 3 /rGO samples match with the typical peaks of Bi 2 S 3 in the CPDS card (No. 17-0320), indicating that GO used in the synthesis promotes the crystal growth of Bi 2 S 3 . However, the intensity of the peaks of Bi 2 S 3 /rGO gradually decreased with increasing amounts of GO, which is probably due to the shielding effect of rGO on the Bi 2 S 3 surface [32]. Moreover, no pronounced diffraction peak of rGO has been found in XRD patterns of Bi 2 S 3 /rGO composites, which can be attributed to the relatively low intensity of the rGO peak or the face-to-face stacking of Bi 2 S 3 on both sides [33]. Figure 2b shows the typical Raman spectra of GO, Bi 2 S 3 , and Bi 2 S 3 /rGO-3 in the range of 150-2000 cm −1 . The low-intensity feature at 239 cm −1 and the peak at 974 cm −1 observed for pure Bi 2 S 3 correspond to the A g optical phonon modes and surface phonon modes [34]. Similar peaks of pure Bi 2 S 3 and GO can be observed in the spectra recorded on the Bi 2 S 3 /rGO-3 sample, indicating that rGO and Bi 2 S 3 are present in Bi 2 S 3 /rGO. The typical G band and D band of Bi 2 S 3 /rGO and GO are attributed to the E 2g phonon of sp 2 carbon atoms and the A 1g (breathing mode) of sp 3 vibrations resulting from surface defects and the structural disorder of GO, respectively [35]. The intensity ratio (I D /I G ) of Bi 2 S 3 /rGO is higher than that of GO because of the increasing sp 2 domains from the reduction of GO during the hydrothermal process [36]. The Raman spectra proved the structural integrity of Bi 2 S 3 /rGO composites. band and D band of Bi2S3/rGO and GO are attributed to the E2g phonon of sp 2 carbon atoms and the A1g (breathing mode) of sp 3 vibrations resulting from surface defects and the structural disorder of GO, respectively [35]. The intensity ratio (ID/IG) of Bi2S3/rGO is higher than that of GO because of the increasing sp 2 domains from the reduction of GO during the hydrothermal process [36]. The Raman spectra proved the structural integrity of Bi2S3/rGO composites. The chemical composition of the samples was evaluated by X-ray photoelectron spectroscopy. Figure 3a shows the XPS survey spectra recorded on the surface of Bi2S3 and Bi2S3/rGO-3 samples. The presence of C, Bi, S, and O elements is confirmed. Compared with the pristine Bi2S3 sample, the intensities of C and O peaks are relatively higher in Bi2S3/rGO-3, suggesting the presence of rGO [37]. Figure 3b shows the XPS spectra of Bi 4f photoelectrons, in pure Bi2S3, the two intense peaks can be associated with Bi 4f7/2 and Bi 4f5/2 doublet with a spin-orbit splitting of 5.3 eV. The Bi 4f doublet with components at 158.2 and 163.5 eV can be observed in the Bi2S3/rGO-3 sample, as well as two low intensity peaks centered at 159.1 and 164.3 eV belonging to Bi-O bond, suggesting the Bi 3+ ions interaction with rGO [38,39]. The Bi-O associated doublet peak is also observed in the pure Bi2S3 sample indicating that a low amount of oxygen reacted with surface Bi ions. Moreover, the S 2p3/2 and S 2p1/2 doublet peaks centered at 161.1 and 162.2 eV indicate that S 2− as the primary valence state occurs in Bi2S3/rGO-3 [40]. Figure 3c shows three components of C1s centered at 284.8, 285.9, and 288.9eV corresponding to the C=C, C-O-C, and O-C=O carbon chemical environments, respectively, indicating the residual oxygen functional groups on the surface of rGO [41]. Besides, the corresponding O 1s core-level spectrum of the Bi2S3/rGO-3 sample (Figure 3d) is well reproduced with three components: the peak at the lowest binding energy can be assigned to Bi-O species, and the peaks centered at 531.8 and 533.6 eV are attributed to C=O and O-C=O bonding environments, respectively [42]. Therefore, from the XPS analysis, we can suggest that Bi2S3 nanorods interact with reduced graphene via the formation of Bi-O bonds, although surface oxidation forming Bi-O bonds cannot be discarded. The chemical composition of the samples was evaluated by X-ray photoelectron spectroscopy. Figure 3a shows the XPS survey spectra recorded on the surface of Bi 2 S 3 and Bi 2 S 3 /rGO-3 samples. The presence of C, Bi, S, and O elements is confirmed. Compared with the pristine Bi 2 S 3 sample, the intensities of C and O peaks are relatively higher in Bi 2 S 3 /rGO-3, suggesting the presence of rGO [37]. Figure 3b shows the XPS spectra of Bi 4f photoelectrons, in pure Bi 2 S 3 , the two intense peaks can be associated with Bi 4f 7/2 and Bi 4f 5/2 doublet with a spin-orbit splitting of 5.3 eV. The Bi 4f doublet with components at 158.2 and 163.5 eV can be observed in the Bi 2 S 3 /rGO-3 sample, as well as two low intensity peaks centered at 159.1 and 164.3 eV belonging to Bi-O bond, suggesting the Bi 3+ ions interaction with rGO [38,39]. The Bi-O associated doublet peak is also observed in the pure Bi 2 S 3 sample indicating that a low amount of oxygen reacted with surface Bi ions. Moreover, the S 2p 3/2 and S 2p 1/2 doublet peaks centered at 161.1 and 162.2 eV indicate that S 2− as the primary valence state occurs in Bi 2 S 3 /rGO-3 [40]. Figure 3c shows three components of C1s centered at 284.8, 285.9, and 288.9eV corresponding to the C=C, C-O-C, and O-C=O carbon chemical environments, respectively, indicating the residual oxygen functional groups on the surface of rGO [41]. Besides, the corresponding O 1s core-level spectrum of the Bi 2 S 3 /rGO-3 sample (Figure 3d) is well reproduced with three components: the peak at the lowest binding energy can be assigned to Bi-O species, and the peaks centered at 531.8 and 533.6 eV are attributed to C=O and O-C=O bonding environments, respectively [42]. Therefore, from the XPS analysis, we can suggest that Bi 2 S 3 nanorods interact with reduced graphene via the formation of Bi-O bonds, although surface oxidation forming Bi-O bonds cannot be discarded. FTIR spectroscopy was used to study graphene oxide and Bi2S3/rGO functional groups ( Figure 4). The broad absorption band centered at 3425 cm −1 is assigned to the stretching vibration of -OH groups from adsorbed water molecules [43]. The characteristic peaks ascribed to C=O, C=C, C-O-C, and C-O functional groups were located at 1711 cm −1 , 1564 cm −1 , 1165 cm −1 , and 1051cm −1 , respectively [44,45]. Compared with GO, it is clearly seen that the intensities of the corresponding absorption peaks of Bi2S3/rGO sample decreased, indicating the conversion of GO to rGO during the hydrothermal process, in agreement with the Raman results.

Formation Mechanism of Bi2S3/rGO
Based on the above analyses, the overall synthetic procedure of Bi2S3/rGO and the electrochemical sensing process for Ascorbic Acid (AA) can be seen in Scheme 1. First, the . XPS survey spectra of (a) Bi 2 S 3 , and Bi 2 S 3 /rGO-3 and core-level XPS spectra of (b) Bi 4f and S 2p, (c) C1s, (d) O1s in Bi 2 S 3 /rGO-3.
FTIR spectroscopy was used to study graphene oxide and Bi 2 S 3 /rGO functional groups ( Figure 4). The broad absorption band centered at 3425 cm −1 is assigned to the stretching vibration of -OH groups from adsorbed water molecules [43]. The characteristic peaks ascribed to C=O, C=C, C-O-C, and C-O functional groups were located at 1711 cm −1 , 1564 cm −1 , 1165 cm −1 , and 1051cm −1 , respectively [44,45]. Compared with GO, it is clearly seen that the intensities of the corresponding absorption peaks of Bi 2 S 3 /rGO sample decreased, indicating the conversion of GO to rGO during the hydrothermal process, in agreement with the Raman results.  FTIR spectroscopy was used to study graphene oxide and Bi2S3/rGO functional groups ( Figure 4). The broad absorption band centered at 3425 cm −1 is assigned to the stretching vibration of -OH groups from adsorbed water molecules [43]. The characteristic peaks ascribed to C=O, C=C, C-O-C, and C-O functional groups were located at 1711 cm −1 , 1564 cm −1 , 1165 cm −1 , and 1051cm −1 , respectively [44,45]. Compared with GO, it is clearly seen that the intensities of the corresponding absorption peaks of Bi2S3/rGO sample decreased, indicating the conversion of GO to rGO during the hydrothermal process, in agreement with the Raman results.

Formation Mechanism of Bi2S3/rGO
Based on the above analyses, the overall synthetic procedure of Bi2S3/rGO and the electrochemical sensing process for Ascorbic Acid (AA) can be seen in Scheme 1. First, the

Formation Mechanism of Bi 2 S 3 /rGO
Based on the above analyses, the overall synthetic procedure of Bi 2 S 3 /rGO and the electrochemical sensing process for Ascorbic Acid (AA) can be seen in Scheme 1. First, the positively charged Bi 3+ ions absorb the negatively charged oxygen-containing functional groups of GO sheets via electrostatic interaction [46], while the thiourea added as the sulfur source and reducing agent hydrolyzes to S 2− ions at high temperature. The interaction between Bi 3+ and S 2− ions generates Bi 2 S 3 nucleii which grow into Bi 2 S 3 nanorods on the surface of the rGO to form Bi 2 S 3 /rGO composites [47]. The electrocatalytic response of AA at Bi 2 S 3 /rGO/GCE is performed in 0.1 M PBS (pH = 7.0). The oxidation process of AA with a two-electron and two-proton electro-oxidation process leads to dehydroascorbic acid formation followed by its electrochemical sensing [48].
Chemosensors 2021, 9, x FOR PEER REVIEW 7 of 15 positively charged Bi 3+ ions absorb the negatively charged oxygen-containing functional groups of GO sheets via electrostatic interaction [46], while the thiourea added as the sulfur source and reducing agent hydrolyzes to S 2− ions at high temperature. The interaction between Bi 3+ and S 2− ions generates Bi2S3 nucleii which grow into Bi2S3 nanorods on the surface of the rGO to form Bi2S3/rGO composites [47]. The electrocatalytic response of AA at Bi2S3/rGO/GCE is performed in 0.1 M PBS (pH = 7.0). The oxidation process of AA with a two-electron and two-proton electro-oxidation process leads to dehydroascorbic acid formation followed by its electrochemical sensing [48].

Scheme 1.
Proposed growth mechanism of Bi2S3/rGO and the electrochemical oxidation of AA.

Bi2S3/rGO Electrochemical Evaluation
The electrochemical behavior of the different modified electrodes was evaluated via cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) using ferro/ferricyanide as the redox probe. From Figure 5a, it can be seen that compared with bare GCE, Bi2S3/GCE, and rGO/GCE, the Bi2S3/rGO/GCE shows a dramatic enhancement of the peak current, indicating that Bi2S3/rGO as the active layer plays a vital role for electron transfer behavior. Moreover, Bi2S3/rGO-3/GCE presents the optimal current response in 0.1 M KCl using 5 mM [Fe (CN)6] 3−/4− within the potential range of −0.2 to 0.8 V. The electrochemical impedance spectroscopy (EIS) Nyquist plots obtained for different modified electrodes are presented in Figure 5b,c. The straight line of Bi2S3/rGO has the steepest slope in the lower frequency area, suggesting that the synthesized Bi2S3/rGO composites have the minimum diffusion resistance. According to the Nyquist plots' semicircle evaluating the interfacial charge transfer resistance (Rct), the Bi2S3/rGO-3/GCE has a lower electron transfer resistance than that of the bare GCE and the Bi2S3/GCE [49,50]. This phenomenon can be associated with the introduction of Bi2S3 nanorods improving electron transfer capability. Compared to the Bi2S3/rGO-1/GCE, the Bi2S3/rGO-2/GCE and the Bi2S3/rGO-4/GCE, Bi2S3/rGO-3/GCE have better conductivity (Figure 5c). According to the obtained results, the interfacial charge transfer resistance of the Bi2S3/rGO composites decreases in sequence along with the increase in the amount of GO added to the synthesis. This can be attributed to the graphene that facilitates the transport of electrons and ions. However, adding high amounts of GO increased the composite resistance, which is likely related to the morphology of the Bi2S3 nanorods. The longer length of the nanorods increases the distance between rGO nanosheets retarding facile electron transport [45]. Hence, the Bi2S3/rGO-3 was chosen in the following experiments. Scheme 1. Proposed growth mechanism of Bi 2 S 3 /rGO and the electrochemical oxidation of AA.

Bi 2 S 3 /rGO Electrochemical Evaluation
The electrochemical behavior of the different modified electrodes was evaluated via cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) using ferro/ferricyanide as the redox probe. From Figure 5a, it can be seen that compared with bare GCE, Bi 2 S 3 /GCE, and rGO/GCE, the Bi 2 S 3 /rGO/GCE shows a dramatic enhancement of the peak current, indicating that Bi 2 S 3 /rGO as the active layer plays a vital role for electron transfer behavior. Moreover, Bi 2 S 3 /rGO-3/GCE presents the optimal current response in 0.1 M KCl using 5 mM [Fe (CN) 6 ] 3−/4− within the potential range of −0.2 to 0.8 V. The electrochemical impedance spectroscopy (EIS) Nyquist plots obtained for different modified electrodes are presented in Figure 5b,c. The straight line of Bi 2 S 3 /rGO has the steepest slope in the lower frequency area, suggesting that the synthesized Bi 2 S 3 /rGO composites have the minimum diffusion resistance. According to the Nyquist plots' semicircle evaluating the interfacial charge transfer resistance (Rct), the Bi 2 S 3 /rGO-3/GCE has a lower electron transfer resistance than that of the bare GCE and the Bi 2 S 3 /GCE [49,50]. This phenomenon can be associated with the introduction of Bi 2 S 3 nanorods improving electron transfer capability. Compared to the Bi 2 S 3 /rGO-1/GCE, the Bi 2 S 3 /rGO-2/GCE and the Bi 2 S 3 /rGO-4/GCE, Bi 2 S 3 /rGO-3/GCE have better conductivity (Figure 5c). According to the obtained results, the interfacial charge transfer resistance of the Bi 2 S 3 /rGO composites decreases in sequence along with the increase in the amount of GO added to the synthesis. This can be attributed to the graphene that facilitates the transport of electrons and ions. However, adding high amounts of GO increased the composite resistance, which is likely related to the morphology of the Bi 2 S 3 nanorods. The longer length of the nanorods increases the distance between rGO nanosheets retarding facile electron transport [45]. Hence, the Bi 2 S 3 /rGO-3 was chosen in the following experiments. In order to investigate the electrocatalytic response of the different modified electrodes in the detection of AA, differential pulse voltammetry (DPV) was carried out in 0.1 M PBS (pH = 7.0) containing 5 mM AA at a scan rate of 100 mV s −1 (Figure 6). The rate of electro-oxidation on the AA was varied at different modified electrodes, the peak potential of the AA oxidation was 0.492, 0.556, 0.188, and 0.224 V for the bare GCE, Bi2S3/GCE, rGO/GCE, and Bi2S3/rGO-3/GCE, respectively. The Bi2S3/GCE showed a weak oxidation peak due to the electric blocking effect produced by the semiconductor's presence over the electrode surface. The Bi2S3/rGO-3/GCE showed an exceptional current response at 0.225 V, and the superior electrochemical performance of Bi2S3/rGO-3/GCE can be attributed to the synergetic combination of the nanocrystalline Bi2S3 with the GO [51]. Therefore, it is suggested that Bi2S3 and rGO improve the materials' electrical conductivity and catalytic activity synergistically.

Effect of Solution pH
To evaluate the pH effect on the response to the electrochemical behavior of AA on the Bi2S3/rGO-3/GCE, analyses were carried out in the pH range from 3.0 to 8.0. As shown in Figure 7a, the oxidation peak potential (EP) shifts towards greater negative potential for increasing pH values, indicating that the electro-oxidation process is pH-dependent and In order to investigate the electrocatalytic response of the different modified electrodes in the detection of AA, differential pulse voltammetry (DPV) was carried out in 0.1 M PBS (pH = 7.0) containing 5 mM AA at a scan rate of 100 mV s −1 (Figure 6). The rate of electro-oxidation on the AA was varied at different modified electrodes, the peak potential of the AA oxidation was 0.492, 0.556, 0.188, and 0.224 V for the bare GCE, Bi 2 S 3 /GCE, rGO/GCE, and Bi 2 S 3 /rGO-3/GCE, respectively. The Bi 2 S 3 /GCE showed a weak oxidation peak due to the electric blocking effect produced by the semiconductor's presence over the electrode surface. The Bi 2 S 3 /rGO-3/GCE showed an exceptional current response at 0.225 V, and the superior electrochemical performance of Bi 2 S 3 /rGO-3/GCE can be attributed to the synergetic combination of the nanocrystalline Bi 2 S 3 with the GO [51]. Therefore, it is suggested that Bi 2 S 3 and rGO improve the materials' electrical conductivity and catalytic activity synergistically. In order to investigate the electrocatalytic response of the different modified electrodes in the detection of AA, differential pulse voltammetry (DPV) was carried out in 0.1 M PBS (pH = 7.0) containing 5 mM AA at a scan rate of 100 mV s −1 (Figure 6). The rate of electro-oxidation on the AA was varied at different modified electrodes, the peak potential of the AA oxidation was 0.492, 0.556, 0.188, and 0.224 V for the bare GCE, Bi2S3/GCE, rGO/GCE, and Bi2S3/rGO-3/GCE, respectively. The Bi2S3/GCE showed a weak oxidation peak due to the electric blocking effect produced by the semiconductor's presence over the electrode surface. The Bi2S3/rGO-3/GCE showed an exceptional current response at 0.225 V, and the superior electrochemical performance of Bi2S3/rGO-3/GCE can be attributed to the synergetic combination of the nanocrystalline Bi2S3 with the GO [51]. Therefore, it is suggested that Bi2S3 and rGO improve the materials' electrical conductivity and catalytic activity synergistically.

Effect of Solution pH
To evaluate the pH effect on the response to the electrochemical behavior of AA on the Bi2S3/rGO-3/GCE, analyses were carried out in the pH range from 3.0 to 8.0. As shown in Figure 7a, the oxidation peak potential (EP) shifts towards greater negative potential for increasing pH values, indicating that the electro-oxidation process is pH-dependent and

Effect of Solution pH
To evaluate the pH effect on the response to the electrochemical behavior of AA on the Bi 2 S 3 /rGO-3/GCE, analyses were carried out in the pH range from 3.0 to 8.0. As shown in Figure 7a, the oxidation peak potential (E P ) shifts towards greater negative potential for increasing pH values, indicating that the electro-oxidation process is pH-dependent and protons take part in the electrode reaction [52,53]. The oxidation peak current of AA increases steadily as the pH changes from 3.0 to 4.0 ( Figure 7b); however, further increasing the pH value from 4.0 to 8.0, the anode peak current tends to decrease. The high analytical signal for AA is observed at pH 4.0. The dependence of the AA peak potential on pH can be expressed by Equation (1): In Figure 7c, the slope is a Nernstian value of −0.0484 V/pH, which indicates the participation of electron and proton involving in the process of AA on the Bi 2 S 3 /rGO-3/GCE surface [54].
Chemosensors 2021, 9, x FOR PEER REVIEW 9 of 15 protons take part in the electrode reaction [52,53]. The oxidation peak current of AA increases steadily as the pH changes from 3.0 to 4.0 ( Figure 7b); however, further increasing the pH value from 4.0 to 8.0, the anode peak current tends to decrease. The high analytical signal for AA is observed at pH 4.0. The dependence of the AA peak potential on pH can be expressed by Equation (1): In Figure 7c, the slope is a Nernstian value of −0.0484 V/pH, which indicates the participation of electron and proton involving in the process of AA on the Bi2S3/rGO-3/GCE surface [54].

Effect of Scan Rate
The AA oxidation reaction kinetics of Bi2S3/rGO-3/GCE was studied by CVs at a scan rate range of 20-100 mV s −1 . As shown in Figure 8a, the peak current response increases with the increase in scan rate and the peak potential shifts to the positive side. Furthermore, the linear relationship between the oxidation peak current (IP) and the square roots of scan rates can be seen in Figure 8b, and the linear regression equation was described as follows: IP (A) = 9.85 1/2 + 25.64 ( in mV s −1 ), R 2 = 0.994 (2) The oxidation peak currents are well proportional to the square roots of the scan rates, which indicates that the oxidation reaction of AA is a typical diffusion-controlled process [55].

Effect of Scan Rate
The AA oxidation reaction kinetics of Bi 2 S 3 /rGO-3/GCE was studied by CVs at a scan rate range of 20-100 mV s −1 . As shown in Figure 8a, the peak current response increases with the increase in scan rate and the peak potential shifts to the positive side. Furthermore, the linear relationship between the oxidation peak current (I P ) and the square roots of scan rates can be seen in Figure 8b, and the linear regression equation was described as follows: I P (A) = 9.85 v 1/2 + 25.64 (v in mV s −1 ), R 2 = 0.994 (2) The oxidation peak currents are well proportional to the square roots of the scan rates, which indicates that the oxidation reaction of AA is a typical diffusion-controlled process [55].
Chemosensors 2021, 9, x FOR PEER REVIEW 9 of 15 protons take part in the electrode reaction [52,53]. The oxidation peak current of AA increases steadily as the pH changes from 3.0 to 4.0 ( Figure 7b); however, further increasing the pH value from 4.0 to 8.0, the anode peak current tends to decrease. The high analytical signal for AA is observed at pH 4.0. The dependence of the AA peak potential on pH can be expressed by Equation (1): In Figure 7c, the slope is a Nernstian value of −0.0484 V/pH, which indicates the participation of electron and proton involving in the process of AA on the Bi2S3/rGO-3/GCE surface [54].

Effect of Scan Rate
The AA oxidation reaction kinetics of Bi2S3/rGO-3/GCE was studied by CVs at a scan rate range of 20-100 mV s −1 . As shown in Figure 8a, the peak current response increases with the increase in scan rate and the peak potential shifts to the positive side. Furthermore, the linear relationship between the oxidation peak current (IP) and the square roots of scan rates can be seen in Figure 8b, and the linear regression equation was described as follows: IP (A) = 9.85 1/2 + 25.64 ( in mV s −1 ), R 2 = 0.994 (2) The oxidation peak currents are well proportional to the square roots of the scan rates, which indicates that the oxidation reaction of AA is a typical diffusion-controlled process [55].

Analytical Sensing Performance
Differential pulse voltammograms (DPVs) of the Bi 2 S 3 /rGO-3/GCE for AA in the PBS solution was studied, and the obtained DPV profiles are presented in Figure 9a. The peak current increases linearly with the concentration ranging from 5 µM to 1200 µM (Figure 9b) and can be described by the corresponding Equation (3): From this data, the detection limit of the Bi 2 S 3 /rGO-3/GCE is 2.94 µM (S/N = 3) for the electrochemical detection of AA, and a sensitivity of 268.84 µA mM −1 cm −2 was determined. Comparison of the analytical results of the Bi 2 S 3 /rGO-3/GCE and reported results for ascorbic acid sensors are summarized in Table 1. From this table, the electrocatalytic activity of the Bi 2 S 3 /rGO-3/GCE sensor is comparable or better than for recently reported studies using other materials.

Analytical Sensing Performance
Differential pulse voltammograms (DPVs) of the Bi2S3/rGO-3/GCE for AA in the PBS solution was studied, and the obtained DPV profiles are presented in Figure 9a. The peak current increases linearly with the concentration ranging from 5 µ M to 1200 µ M ( Figure  9b) and can be described by the corresponding Equation (3): From this data, the detection limit of the Bi2S3/rGO-3/GCE is 2.94 µ M (S/N = 3) for the electrochemical detection of AA, and a sensitivity of 268.84 µ A mM −1 cm −2 was determined. Comparison of the analytical results of the Bi2S3/rGO-3/GCE and reported results for ascorbic acid sensors are summarized in Table 1. From this table, the electrocatalytic activity of the Bi2S3/rGO-3/GCE sensor is comparable or better than for recently reported studies using other materials.

Reproducibility and Stability
The reproducibility and stability of the Bi2S3/rGO-3/GCE were examined in 0.1 M PBS (pH = 7), adding a fixed amount of AA (1 mM) ( Figure 10). In five independently prepared electrode tests, the relative standard deviation (RSD) obtained of the peak current was 6.25%, which is within acceptable levels. Moreover, the current response of Bi2S3/rGO-3/GCE was 96.4% of its initial current value after twenty DPV tests, which implies adequate operational stability.

Reproducibility and Stability
The reproducibility and stability of the Bi 2 S 3 /rGO-3/GCE were examined in 0.1 M PBS (pH = 7), adding a fixed amount of AA (1 mM) ( Figure 10). In five independently prepared electrode tests, the relative standard deviation (RSD) obtained of the peak current was 6.25%, which is within acceptable levels. Moreover, the current response of Bi 2 S 3 /rGO-3/GCE was 96.4% of its initial current value after twenty DPV tests, which implies adequate operational stability.

Interference Effect
The selectivity of the target analyte is a significant property for a biosensor. The interference experiments were carried out by successive addition of 200 µ M AA and 200 µ M interfering species (glucose, carbamide, clycine, K + , Na + , Mg 2+ , Zn 2+ , dopamine (DA), and uric acid (UA)) in 0.1 mM PBS (pH = 7.0). The measured effects of different interferents along with AA are shown in Figure 11. As shown in this figure, all these organic and inorganic species did not show any significant fluctuation in the determination of AA. The current response was obviously enhanced when 200 µ M AA was added again. These results indicate that Bi2S3/rGO-3/GCE has high sensitivity and selectivity to AA in the presence of other organic molecules and ions normally found in food and biological samples.

AA Analytical Application
The applicability of Bi2S3/rGO-3/GCE was further examined using vitamin C tablets. Experiments were performed with the standard addition method [62] and recoveries in the range of 97.14-101.55% were obtained ( Table 2). The good recovery values for AA indicate the feasible application of the Bi2S3/rGO-3/GCE electrode for the AA determination in real samples.

Interference Effect
The selectivity of the target analyte is a significant property for a biosensor. The interference experiments were carried out by successive addition of 200 µM AA and 200 µM interfering species (glucose, carbamide, clycine, K + , Na + , Mg 2+ , Zn 2+ , dopamine (DA), and uric acid (UA)) in 0.1 mM PBS (pH = 7.0). The measured effects of different interferents along with AA are shown in Figure 11. As shown in this figure, all these organic and inorganic species did not show any significant fluctuation in the determination of AA. The current response was obviously enhanced when 200 µM AA was added again. These results indicate that Bi 2 S 3 /rGO-3/GCE has high sensitivity and selectivity to AA in the presence of other organic molecules and ions normally found in food and biological samples.

Interference Effect
The selectivity of the target analyte is a significant property for a biosensor. The interference experiments were carried out by successive addition of 200 µ M AA and 200 µ M interfering species (glucose, carbamide, clycine, K + , Na + , Mg 2+ , Zn 2+ , dopamine (DA), and uric acid (UA)) in 0.1 mM PBS (pH = 7.0). The measured effects of different interferents along with AA are shown in Figure 11. As shown in this figure, all these organic and inorganic species did not show any significant fluctuation in the determination of AA. The current response was obviously enhanced when 200 µ M AA was added again. These results indicate that Bi2S3/rGO-3/GCE has high sensitivity and selectivity to AA in the presence of other organic molecules and ions normally found in food and biological samples.

AA Analytical Application
The applicability of Bi2S3/rGO-3/GCE was further examined using vitamin C tablets. Experiments were performed with the standard addition method [62] and recoveries in the range of 97.14-101.55% were obtained ( Table 2). The good recovery values for AA indicate the feasible application of the Bi2S3/rGO-3/GCE electrode for the AA determination in real samples.

AA Analytical Application
The applicability of Bi 2 S 3 /rGO-3/GCE was further examined using vitamin C tablets. Experiments were performed with the standard addition method [62] and recoveries in the range of 97.14-101.55% were obtained ( Table 2). The good recovery values for AA indicate the feasible application of the Bi 2 S 3 /rGO-3/GCE electrode for the AA determination in real samples.

Conclusions
In summary, using a one-pot synthesis method Bi 2 S 3 /rGO composites were successfully synthesized. The obtained Bi 2 S 3 /rGO composites were used as the sensor active layer significantly improving the GCE performance in detecting AA. The size of the Bi 2 S 3 rods can be controlled by adjusting the amount of GO in the synthesis, and the synergistic effect between rGO sheets and Bi 2 S 3 nanorods has an essential effect on the electrochemical performance. The suitably sized nanorods of Bi 2 S 3 anchored on rGO as an electrochemical sensor to detect AA show excellent sensitivity, good selectivity and stability, as well as good reproducibility and a detection limit of 2.94 µM. Moreover, using Bi 2 S 3 /rGO/GCE as a sensor to detect an actual sample containing AA, a good recovery (97.1% to 101.55%) was observed. The research method described in this work opens new avenues for optimizing novel materials to be used as the active layer in electrochemical detection.