Application of Chitosan-Based Molecularly Imprinted Polymer in Development of Electrochemical Sensor for p-Aminophenol Determination

Molecularly Imprinted Polymers (MIPs) have specific recognition capabilities and have been widely used for electrochemical sensors with high selectivity. In this study, an electrochemical sensor was developed for the determination of p-aminophenol (p-AP) by modifying the screen-printed carbon electrode (SPCE) with chitosan-based MIP. The MIP was made from p-AP as a template, chitosan (CH) as a base polymer, and glutaraldehyde and sodium tripolyphosphate as the crosslinkers. MIP characterization was conducted based on membrane surface morphology, FT-IR spectrum, and electrochemical properties of the modified SPCE. The results showed that the MIP was able to selectively accumulate analytes on the electrode surface, in which MIP with glutaraldehyde as a crosslinker was able to increase the signal. Under optimum conditions, the anodic peak current from the sensor increased linearly in the range of 0.5–35 µM p-AP concentration, with sensitivity of (3.6 ± 0.1) µA/µM, detection limit (S/N = 3) of (2.1 ± 0.1) µM, and quantification limit of (7.5 ± 0.1) µM. In addition, the developed sensor exhibited high selectivity with an accuracy of (94.11 ± 0.01)%.


Introduction
Molecular Imprinting Technology (MIT) is currently a synthetic approach to designing molecular recognition materials with the capability of mimicking natural recognition as biological receptors. MIT applications include separation and purification, chemical sensors, catalysis, and receptor systems [1][2][3][4][5][6][7]. MIT is based on the formation of complexes between analytes (templates) and functional monomers. In the presence of an excess crosslinking agent, a three-dimensional polymer network is formed. After the polymerization process, the template is removed from the polymer, leaving specific recognition sites that complement each other in shape, size, and chemical function to the template molecule [8][9][10][11]. The polymerization process is not easy to carry out; thus, several MIPs have been developed from functional polymers by adding controlled crosslinkers. Functional polymers that have been developed as MIP membranes and applied to electrochemical sensors are chitosan, which was used for MSG (monosodium glutamate) sensors [12], poly vinyl alcohol (PVA) for paracetamol sensors [13], and arrowroot starch-PVA for acid sensors [14]. In this study, chitosan-PVA was used as a functional polymer, glutaraldehyde and sodium tripolyphosphate (STPP) as the crosslinkers, and acetic acid as a catalyst.
Para-Aminophenol (p-AP) is the main degradation product of paracetamol and the main impurity of its synthesis process, which, by law, should not exceed the threshold of 0.005% and 0.1% by weight [15]. Acetaminophen or paracetamol (PA) is an analgesic,

Preparation of MIP
A total of 1 g of chitosan was added to 50 mL of 5% (v/v) acetic acid. The mixture was then stirred for 2 h at 50 • C, until a clear and homogeneous solution was obtained. Next, 1 mL of 0.1% (w/v) p-AP solution was added into 7 mL of chitosan solution, and the mixture was stirred and then 1 mL of 2% (v/v) glutaraldehyde or STPP 2% (w/v) was added. The mixture was stirred at room temperature for 1 h, followed by the addition of 1 mL polyvinyl alcohol 1% (w/v). After the mixture was stirred for a few minutes, it was ready to be used as an SPCE modifier.

MIP Characterization
The characterization was carried out on a MIP with a p-AP concentration of 1.0% (M_4), with crosslinkers, both glutaraldehyde (M-4G) and STTP (M-4S). MIP properties was characterized based on the FTIR spectra and MIP surface morphology by scanning electron microscopy (SEM). The MIP preparation procedure was identical to (2.2), but instead, after the mixture was heated at room temperature for 1 h, it was then poured into a Petri dish and heated at 50 • C to form a thin layer. Next, the thin layer was removed from the Petri dish and washed with 0.1 M NaOH followed by distilled water until the washing water was neutral, for the extraction of the template. The MIP was then dried until it was free of water. The FTIR analyses were carried out at the Chemistry Department of Brawijaya University, while the SEM analyses were carried out at the State University of Malang.

Modification of SPCE
A total of 20 µL of MIP (from Section 2.2) was dripped on the WE surface in SPCE, and then smoothed with a small brush. The coating was carried out twice, in which the first coating was heated at 50 • C for 2 min, whereas the second coating was heated at 50 • C for 5 min. A thin layer was formed on the WE surface, and then the surface was dripped with 50 µL of 0.1 M NaOH several times, followed by 50 µL of distilled water, until the water was not alkaline (which was checked by Litmus paper). This was carried out carefully so that the thin film was not damaged or loose.

p-AP Sensor Testing
Sensor evaluation was performed using a 50 µM p-AP solution which was dissolved in phosphate buffer pH 6.2. Evaluation of the effect of p-AP concentration on MIP was carried out by square wave voltammetry (SWV) at an amplitude potential of 50 mV, a frequency Polymers 2023, 15, 1818 4 of 12 of 10 Hz, and a potential step of 10 mV, while the influence of the type of crosslinker was evaluated by cyclic voltammetry (CV) at −1.0 to 0.8 volts with a scan rate of 50 mV/s and potential step 10 mV. In both methods, 200 µL of the p-AP solution was dripped on the SPCE-modified surface.

p-AP Performance
Standard p-AP solutions in several concentrations from 0.5 to 50.0 µM were prepared in 0.1 M phosphate buffer pH 6, then determined by SWV using SPCE-M-4G, SPCE-M-4S, and SPCE. In addition, a sample solution from commercial paracetamol tablets was prepared. The tablets were weighed and then crushed. A total of (500.0 + 0.1) mg of the sample was dissolved in phosphate buffer pH 6, filtered off, and dissolved again in a 10 mL volumetric flask. The SWV was recorded at −0.15 to 0.20 Volts.

Preparation of MIP
In the synthesis of MIP, optimization of concentration of p-AP, as a template, and selection of cross-linking reagents, glutaraldehyde, and STPP, were studied. Glutaraldehyde as a cross-linker was used in the optimization of p-AP concentration. The concentrations of p-AP were 0; 0.1; 0.5; 1.0; 1.5; and 2.0% (w/v) in MIP. To simplify, the six MIPs were coded as M-1 to M-6. Evaluation results of the six MIPs, using square wave voltammetry (SWV), are presented in Figure 1 and Table 1.
alkaline (which was checked by Litmus paper). This was carried out carefully so that the thin film was not damaged or loose.

p-AP Sensor Testing
Sensor evaluation was performed using a 50 µM p-AP solution which was dissolved in phosphate buffer pH 6.2. Evaluation of the effect of p-AP concentration on MIP was carried out by square wave voltammetry (SWV) at an amplitude potential of 50 mV, a frequency of 10 Hz, and a potential step of 10 mV, while the influence of the type of crosslinker was evaluated by cyclic voltammetry (CV) at −1.0 to 0.8 volts with a scan rate of 50 mV/s and potential step 10 mV. In both methods, 200 µL of the p-AP solution was dripped on the SPCE-modified surface.

p-AP Performance
Standard p-AP solutions in several concentrations from 0.5 to 50.0 µM were prepared in 0.1 M phosphate buffer pH 6, then determined by SWV using SPCE-M-4G, SPCE-M-4S, and SPCE. In addition, a sample solution from commercial paracetamol tablets was prepared. The tablets were weighed and then crushed. A total of (500.0 + 0.1) mg of the sample was dissolved in phosphate buffer pH 6, filtered off, and dissolved again in a 10 mL volumetric flask. The SWV was recorded at −0.15 to 0.20 Volts.

Preparation of MIP
In the synthesis of MIP, optimization of concentration of p-AP, as a template, and selection of cross-linking reagents, glutaraldehyde, and STPP, were studied. Glutaraldehyde as a cross-linker was used in the optimization of p-AP concentration. The concentrations of p-AP were 0; 0.1; 0.5; 1.0; 1.5; and 2.0% (w/v) in MIP. To simplify, the six MIPs were coded as M-1 to M-6. Evaluation results of the six MIPs, using square wave voltammetry (SWV), are presented in Figure 1 and Table 1.    From the voltammogram data (Table 1), it can be seen that there are two peaks for p-AP at SPCE without modification: 0.10 and 0.26 volts vs. Ag/AgCl. The peak potential (E p ) of p-Ap, resulting from modified SPCE, is consistent at 0.09 volts. There are two stages of p-AP oxidation ( Figure 2): the first was from p-AP to quinonimine and continued to quinone [41,42]. In the second stage of oxidation, a broad peak was not detected in the modified SPCE. Thus, the MIP membrane on the SPCE surface can cause the second oxidation undetected. This is due to the presence of a thin-film layer on the SPCE surface, which may inhibit the electron transfer from WE to CE at the oxidation of quinonimine to quinone; consequently, the peak current did not appear. From the voltammogram data (Table 1), it can be seen that there are two peaks for p-AP at SPCE without modification: 0.10 and 0.26 volts vs. Ag/AgCl. The peak potential (Ep) of p-Ap, resulting from modified SPCE, is consistent at 0.09 volts. There are two stages of p-AP oxidation ( Figure 2): the first was from p-AP to quinonimine and continued to quinone [41,42]. In the second stage of oxidation, a broad peak was not detected in the modified SPCE. Thus, the MIP membrane on the SPCE surface can cause the second oxidation undetected. This is due to the presence of a thin-film layer on the SPCE surface, which may inhibit the electron transfer from WE to CE at the oxidation of quinonimine to quinone; consequently, the peak current did not appear. The peak current (Ip) on the voltammogram is inversely related to the p-Ap concentration in the MIP, except at 1.0% (M-4). At 1.5 and 2.0% p-AP (M-5 and M-6), the voltammogram peaks were not clearly observed. SPCE-M_4 resulted in the highest peak current and the most symmetrical voltammogram shape. Based on the data in Table 1, the peak current from M-1 is higher than that of M-2 and M_3. In M-1, there is no p-AP, but there is still cross-linking of chitosan by glutaraldehyde, which forms a cavity; hence, p-AP can pass through to diffuse to the WE surface. In M-2 and M-3, it is possible that there is still p-AP remaining, which was not released during the washing, thus blocking the diffusion of p-AP from the bulk solution to WE during measurement. The p-AP in M-4 is 2 times higher than in M-3; thus, more templates are presumably formed, which then cause the peak current to be higher. However, if there is too much p-AP in the MIP, as in M-5 and M-6, the diffusion is more likely to be blocked.
Glutaraldehyde and STPP were compared as chitosan crosslinkers in MIP synthesis at 1% p-AP concentration. The two MIPs were used as SPCE modifiers and tested by cyclic voltammetry for 50 µM p-AP in pH 6.2 buffer solution. M-4G and M-4S are glutaraldehyde and STPP as crosslinkers, respectively, and the p-AP concentration in MIP is 1%. As shown in Figure 3, the oxidation-reduction reaction of p-AP is reversible in the SPCE-unmodified instance, with Ipa/Ipc = 1 and ΔEp = 60 mV. The Ipa/Ipc for SPCE modified by M-4G and M-4S, are 1.6 and 1.2, respectively, whereas ΔEp are 100 and 110 mV, respectively. Modification of SPCE by M-4G and M-4S causes p-AP to be oxidized more slowly; Ep shifted to the positive direction, and so ΔEp increased and the oxidation-reduction reaction of p-AP showed a quasi-reversible property. The peak current (I p ) on the voltammogram is inversely related to the p-Ap concentration in the MIP, except at 1.0% (M-4). At 1.5 and 2.0% p-AP (M-5 and M-6), the voltammogram peaks were not clearly observed. SPCE-M_4 resulted in the highest peak current and the most symmetrical voltammogram shape. Based on the data in Table 1, the peak current from M-1 is higher than that of M-2 and M_3. In M-1, there is no p-AP, but there is still cross-linking of chitosan by glutaraldehyde, which forms a cavity; hence, p-AP can pass through to diffuse to the WE surface. In M-2 and M-3, it is possible that there is still p-AP remaining, which was not released during the washing, thus blocking the diffusion of p-AP from the bulk solution to WE during measurement. The p-AP in M-4 is 2 times higher than in M-3; thus, more templates are presumably formed, which then cause the peak current to be higher. However, if there is too much p-AP in the MIP, as in M-5 and M-6, the diffusion is more likely to be blocked.
Glutaraldehyde and STPP were compared as chitosan crosslinkers in MIP synthesis at 1% p-AP concentration. The two MIPs were used as SPCE modifiers and tested by cyclic voltammetry for 50 µM p-AP in pH 6.2 buffer solution. M-4G and M-4S are glutaraldehyde and STPP as crosslinkers, respectively, and the p-AP concentration in MIP is 1%. As shown in Figure 3, the oxidation-reduction reaction of p-AP is reversible in the SPCE-unmodified instance, with I pa /I pc = 1 and ∆E p = 60 mV. The I pa /I pc for SPCE modified by M-4G and M-4S, are 1.6 and 1.2, respectively, whereas ∆E p are 100 and 110 mV, respectively. Modification of SPCE by M-4G and M-4S causes p-AP to be oxidized more slowly; E p shifted to the positive direction, and so ∆E p increased and the oxidation-reduction reaction of p-AP showed a quasi-reversible property.
The identification of functional groups based on FTIR spectra, compared between chitosan, M-4G, and M-4S, is presented in Figure 4. There is no difference in functional groups between chitosan and M-4S. Meanwhile, in the FTIR spectrum for M-4G, there is a difference in the peaks at wave numbers of 1563 and 1402 cm −1 , which indicates the presence of an -NH secondary amine group and tertiary alcohol [43]. Both M-4G and M-4S lost the peak at 575 cm −1 from chitosan, indicating an -OH out-of-plane bend [43], and that cross-linking possibly occurred in the -OH group of chitosan. Peak wavenumbers and determination of chitosan functional groups, M-4G, and M-4S are presented in Table 2. The cross-linking between chitosan and STPP, which occurs at pH 3, is an electrostatic interaction between the protonated amine group of chitosan (-NH 3 + ) and the phosphate ion in STPP. The washing process by NaOH solution during the SPCE modification was predicted to cause the breakdown of the cross-links. This is indicated by the FTIR spectrum of M-4S which is identical to that of chitosan.  The identification of functional groups based on FTIR spectra, compared between chitosan, M-4G, and M-4S, is presented in Figure 4. There is no difference in functional groups between chitosan and M-4S. Meanwhile, in the FTIR spectrum for M-4G, there is a difference in the peaks at wave numbers of 1563 and 1402 cm −1 , which indicates the presence of an -NH secondary amine group and tertiary alcohol [43]. Both M-4G and M-4S lost the peak at 575 cm −1 from chitosan, indicating an -OH out-of-plane bend [43], and that cross-linking possibly occurred in the -OH group of chitosan. Peak wavenumbers and determination of chitosan functional groups, M-4G, and M-4S are presented in Table 2. The cross-linking between chitosan and STPP, which occurs at pH 3, is an electrostatic interaction between the protonated amine group of chitosan (-NH3 + ) and the phosphate ion in STPP. The washing process by NaOH solution during the SPCE modification was predicted to cause the breakdown of the cross-links. This is indicated by the FTIR spectrum of M-4S which is identical to that of chitosan.   The identification of functional groups based on FTIR spectra, compared between chitosan, M-4G, and M-4S, is presented in Figure 4. There is no difference in functional groups between chitosan and M-4S. Meanwhile, in the FTIR spectrum for M-4G, there is a difference in the peaks at wave numbers of 1563 and 1402 cm −1 , which indicates the presence of an -NH secondary amine group and tertiary alcohol [43]. Both M-4G and M-4S lost the peak at 575 cm −1 from chitosan, indicating an -OH out-of-plane bend [43], and that cross-linking possibly occurred in the -OH group of chitosan. Peak wavenumbers and determination of chitosan functional groups, M-4G, and M-4S are presented in Table 2. The cross-linking between chitosan and STPP, which occurs at pH 3, is an electrostatic interaction between the protonated amine group of chitosan (-NH3 + ) and the phosphate ion in STPP. The washing process by NaOH solution during the SPCE modification was predicted to cause the breakdown of the cross-links. This is indicated by the FTIR spectrum of M-4S which is identical to that of chitosan.   Surface morphology assessment of M-4G and M-4S by SEM, Figure 5) confirms that the p-AP is printed on the surface of M-4G, with no cavity formation observed, which is one of the characteristics of MIP formation. Meanwhile, on the surface of M-4S, the printed p-AP is not visible; only cavities are formed on the surface. This shows that the two crosslinkers have their own weaknesses. Chitosan cross-linking by STPP occurs due to electrostatic interactions and is highly dependent on pH, in which an increase in pH can cause a decrease in the number of cross-links. Chitosan cross-linking by glutaraldehyde occurs in the unprotonated chitosan amine group (-NH 2 ); this reaction occurs more frequently at non-acidic pH, but at pH > 6.2, chitosan would have precipitated out. Therefore, for the synthesis of MIP using glutaraldehyde as a crosslinker, the pH of chitosan in acetic acid was adjusted to 5. Figure 6 shows an illustration of the formation of MIP using glutaraldehyde and STPP as the cross-linker. static interactions and is highly dependent on pH, in which an increase in pH can cause a decrease in the number of cross-links. Chitosan cross-linking by glutaraldehyde occurs in the unprotonated chitosan amine group (-NH2); this reaction occurs more frequently at non-acidic pH, but at pH > 6.2, chitosan would have precipitated out. Therefore, for the synthesis of MIP using glutaraldehyde as a crosslinker, the pH of chitosan in acetic acid was adjusted to 5. Figure 6 shows an illustration of the formation of MIP using glutaraldehyde and STPP as the cross-linker.  In the MIP preparation, chitosan was hydrolyzed in acid (acetic acid) to produce chitosan with shorter chains, but not monomers. Therefore, in this process, chitosan was added with acetic acid and heated for about 2 h to produce a clear mixture. p-AP was added before glutaraldehyde to form a complex, or the interaction between chitosan and p-AP, the crosslinking reaction of chitosan by glutaraldehyde, was expected to be more directed so that the formation of MIP could occur properly. PVA was added last, as a film reinforcement, and was not involved in the reaction. The condition of chitosan cross-linking by glutaraldehyde was different from that of chitosan by STPP. Cross-linking of chitosan by glutaraldehyde occurs at a not too acidic pH, but at a pH ≤ pKa of chitosan (6.2), so that the fraction of the -NH2 group in chitosan can be ≥50%. It was necessary to adjust In the MIP preparation, chitosan was hydrolyzed in acid (acetic acid) to produce chitosan with shorter chains, but not monomers. Therefore, in this process, chitosan was added with acetic acid and heated for about 2 h to produce a clear mixture. p-AP was added before glutaraldehyde to form a complex, or the interaction between chitosan and p-AP, the crosslinking reaction of chitosan by glutaraldehyde, was expected to be more directed so that the formation of MIP could occur properly. PVA was added last, as a film reinforcement, and was not involved in the reaction. The condition of chitosan cross-linking by glutaraldehyde was different from that of chitosan by STPP. Cross-linking of chitosan by glutaraldehyde occurs at a not too acidic pH, but at a pH ≤ pK a of chitosan (6.2), so that the fraction of the -NH 2 group in chitosan can be ≥50%. It was necessary to adjust the pH (≈5) before adding glutaraldehyde. The condition of cross-linking reaction of chitosan by STPP takes place at pH 3; thus, it is unnecessary to adjust the pH. STPP is alkaline; hence, when the STPP is added, it will change the pH of the mixture. The formation of MIP, both by glutaraldehyde and STPP, is illustrated in Figure 6. The illustration is modified from various literature. To release p-AP from MIP, template formation was carried out by washing MIP with 0.1 M NaOH solution, followed by distilled water to rinse off the remaining NaOH. This process was performed at MIP on a modified SPCE surface.

Performance of p-AP Sensor
Based on the results in the section above, SPCE-M-4G was chosen as the best sensor. The sensor was tested in 50 µM p-AP solution. Sensor performance can be maximized if the pH of the p-AP solution is suitable for oxidation, as shown in Figure 2, in which the oxidation of p-AP is affected by pH. The p-AP is a weak acid with pK a1 = 5.48 and pK a2 = 10.46, and can be oxidized under acidic conditions. In this study, optimization of pH in the range of 3-8 was carried out by a mixture of citrate and phosphate buffers. Figure 7 shows the voltammogram of a 50 µM p-AP solution at various pHs. As observed in Figure 6, peak current increases in direct proportion to pH in the range 3-6. The peak current decreased at pH 7 and 8. This can be explained by the illustration in Figure 8; at pH < 5.48, the -NH 2 group on p-AP protonates to -NH 3 + , to form structure A. At pH > 5.48, the -NH 3 + in p-AP decreases to form structure B. Oxidation of p-AP occurs in structure B, as shown in Figure 2, thus the peak current increases from pH 3 to 6. In theory ( Figure 8D), from pH 7 to 9, the B structure of p-AP remains dominant (>95%), but p-AP oxidation occurred under acidic conditions; hence, the peak current of p-AP decreased at pH 7, whereas at pH 8, the peak current was not identified. The highest p-AP peak current was obtained at pH 6. 10.46, and can be oxidized under acidic conditions. In this study, optimization of pH in the range of 3-8 was carried out by a mixture of citrate and phosphate buffers. Figure 7 shows the voltammogram of a 50 µM p-AP solution at various pHs. As observed in Figure  6, peak current increases in direct proportion to pH in the range 3-6. The peak current decreased at pH 7 and 8. This can be explained by the illustration in Figure 8; at pH < 5.48, the -NH2 group on p-AP protonates to -NH3 + , to form structure A. At pH > 5.48, the -NH3 + in p-AP decreases to form structure B. Oxidation of p-AP occurs in structure B, as shown in Figure 2, thus the peak current increases from pH 3 to 6. In theory ( Figure 8D), from pH 7 to 9, the B structure of p-AP remains dominant (>95%), but p-AP oxidation occurred under acidic conditions; hence, the peak current of p-AP decreased at pH 7, whereas at pH 8, the peak current was not identified. The highest p-AP peak current was obtained at pH 6. As presented in Figure 8D, the mole fraction of B at pH 3 is very small, only 0.3%, 10.46, and can be oxidized under acidic conditions. In this study, optimization of pH in the range of 3-8 was carried out by a mixture of citrate and phosphate buffers. Figure 7 shows the voltammogram of a 50 µM p-AP solution at various pHs. As observed in Figure  6, peak current increases in direct proportion to pH in the range 3-6. The peak current decreased at pH 7 and 8. This can be explained by the illustration in Figure 8; at pH < 5.48, the -NH2 group on p-AP protonates to -NH3 + , to form structure A. At pH > 5.48, the -NH3 + in p-AP decreases to form structure B. Oxidation of p-AP occurs in structure B, as shown in Figure 2, thus the peak current increases from pH 3 to 6. In theory ( Figure 8D), from pH 7 to 9, the B structure of p-AP remains dominant (>95%), but p-AP oxidation occurred under acidic conditions; hence, the peak current of p-AP decreased at pH 7, whereas at pH 8, the peak current was not identified. The highest p-AP peak current was obtained at pH 6. As presented in Figure 8D, the mole fraction of B at pH 3 is very small, only 0.3%, causing the slow diffusion rate of B and thus a higher potential is required for oxidation of B. At pH 4 to 5, B species increased, respectively, to 3.2 and 24.9%; thus, the rate of  Figure 8. Structure of p-AP at pH < pK a1 (A); pK a1 < pH < pK a2 (B); pH > pK a2 (C) and mole fractions of A, B, and C (D). As presented in Figure 8D, the mole fraction of B at pH 3 is very small, only 0.3%, causing the slow diffusion rate of B and thus a higher potential is required for oxidation of B. At pH 4 to 5, B species increased, respectively, to 3.2 and 24.9%; thus, the rate of diffusion increases, and the potential required for oxidation decreases. Similarly, this also happened at pH 6 (76.8%) and 7, where species B was 76.8 and 97.0%, respectively. Therefore, E p is inversely related to pH at 3 to 7, in which the linear regression of the relationship is: E p = 0.119-0.011 pH with R 2 is 0.995.
Quantitatively, the relationship between p-AP concentration and peak current (I p ) is shown in Figure 9, in the range of 0 to 50 µM. Figure 9 shows that the linear concentration range of the SPCE-M-4G sensor is 0-35 µM, while for SPCE-M-4S and SPCE, it ranges from 0 to 50 µM. The sensitivity of SPCE-M-4G is the highest compared to that of SPCE-M-4S and SPCE (Table 3). SPCE-M-4G has the highest sensitivity compared to the others because M-4G contains p-AP, as shown in Figure 4a, which can be oxidized during measurement and then triggers the diffusion of p-AP from the bulk to the WE surface. Meanwhile, in SPCE-M-4S, the presence of M-4S can block diffusion of p-AP to the WE surface; hence, the sensitivity of SPCE-M-4S is lower than that of SPCE. Comprehensively, SPCE-M-4G performance was the best, with sensitivity of (3.7 ± 0.1) µA/µM, limit of detection of (2.1 ± 0.1) µM, and limit of quantification of (7.5 ± 0.1) µM, at a linear concentration range of 0 to 35 µM. The short linear concentration range is probably caused by the presence of p-AP in M-4G, which has not been released. hence, the sensitivity of SPCE-M-4S is lower than that of SPCE. Comprehensively, SPCE-M-4G performance was the best, with sensitivity of (3.7 ± 0.1) µA/µM, limit of detection of (2.1 ± 0.1) µM, and limit of quantification of (7.5 ± 0.1) µM, at a linear concentration range of 0 to 35 µM. The short linear concentration range is probably caused by the presence of p-AP in M-4G, which has not been released.  Accuracy was determined by standard addition to real samples. Out of the three samples tested, only one sample gave a positive signal, which was an expired drug sample, sample F. Figure 10 shows that sample F produced a signal, and the signal increased very sharply for sample F plus p-AP standard (F + 7µM), with recovery of p-AP concentration being (94.11 ± 0.01)%. It can be concluded that the p-AP sensor based on chitosan MIP can be applied to detect p-AP in paracetamol samples.  Accuracy was determined by standard addition to real samples. Out of the three samples tested, only one sample gave a positive signal, which was an expired drug sample, sample F. Figure 10 shows that sample F produced a signal, and the signal increased very sharply for sample F plus p-AP standard (F + 7µM), with recovery of p-AP concentration being (94.11 ± 0.01)%. It can be concluded that the p-AP sensor based on chitosan MIP can be applied to detect p-AP in paracetamol samples.
Based on the results of this study, to improve the performance of SPCE-M-4G, it is necessary to optimize PVA because it is likely that PVA can inhibit the release of p-AP from M-4G when washing with NaOH. The presence of p-AP on M-4G which has not been released is advantageous on one hand, but it has an impact on short linear concentration range. With the PVA optimization in the MIP preparation, it is hoped that the best composi-tion will be produced to obtain the best sensor. In addition, in order to guarantee that MIP is not easily separated from the SPCE, it is necessary to develop a sensor manufacturing technique by adding MIP directly to the carbon ink for WE. SPCE y = 0.49x + 1.07 0.9907 0.5 ± 0.1 3.1 ± 0.1 15.3 ± 0.1 SPCE-M-4S y = 0.24x + 0.53 0.9779 0.2 ± 0.1 6.0 ± 0.1 25.0 ± 0.1 SPCE-M-4G y = 3.74x + 0.68 0.9957 3.7 ± 0.1 2.1 ± 0.1 7.5 ± 0.1 Accuracy was determined by standard addition to real samples. Out of the three samples tested, only one sample gave a positive signal, which was an expired drug sample, sample F. Figure 10 shows that sample F produced a signal, and the signal increased very sharply for sample F plus p-AP standard (F + 7µM), with recovery of p-AP concentration being (94.11 ± 0.01)%. It can be concluded that the p-AP sensor based on chitosan MIP can be applied to detect p-AP in paracetamol samples.

Conclusions
Electrochemical sensors can be developed from MIPs based on chitosan polymers, especially for p-aminophenol sensors. As a template of an MIP, the best p-aminophenol concentration was 4% in an MIP. Glutaraldehyde as cross-linker gave better sensitivity to the sensor than sodium tripolyphosphate. When identifying MIPs based on the FTIR spectrum, there was a change in the functional group in MIPs with glutaraldehyde as a crosslinker; in contrast, there was no change in the functional group when using STPP as a crosslinker. Based on identification of surface morphology from SEM images, p-aminophenol binds stronger to MIP with glutaraldehyde as a crosslinker than with STPP. The p-aminophenol sensor produced in this work has a working concentration range of 0.5-35 µM, a sensitivity of (3.7 ± 0.1) µM/µA, LoD of (2.1 ± 0.1) µM, LoQ of (7.5 ± 0.1) µM, and an accuracy of (94.11 ± 0.01)%. The sensors can be applied to paracetamol drug samples after several pre-treatments. From this research, development of a disposable p-AP sensor for the determination of p-AP in paracetamol drug samples remains possible.

Data Availability Statement:
The data presented in this study are available in this same article.