Solid-Contact Potentiometric Sensors Based on Main-Tailored Bio-Mimics for Trace Detection of Harmine Hallucinogen in Urine Specimens

All-solid-state potentiometric sensors have attracted great attention over other types of potentiometric sensors due to their outstanding properties such as enhanced portability, simplicity of handling, affordability and flexibility. Herein, a novel solid-contact ion-selective electrode (SC-ISE) based on poly(3,4-ethylenedioxythiophene) (PEDOT) as the ion-to-electron transducer was designed and characterized for rapid detection of harmine. The harmine-sensing membrane was based on the use of synthesized imprinted bio-mimics as a selective material for this recognition. The imprinted receptors were synthesized using acrylamide (AA) and ethylene glycol dimethacrylate (EGDMA) as functional monomer and cross-linker, respectively. The polymerization process was carried out at 70 °C in the presence of dibenzoyl peroxide (DBO) as an initiator. The sensing membrane in addition to the solid-contact layer was applied to a glassy-carbon disc as an electronic conductor. All performance characteristics of the presented electrode in terms of linearity, detection limit, pH range, response time and selectivity were evaluated. The sensor revealed a wide linearity over the range 2.0 × 10−7–1.0 × 10−2 M, with a detection limit of 0.02 µg/mL and a sensitivity slope of 59.2 ± 0.8 mV/hamine concentration decade. A 40 mM Britton–Robinson (BR) buffer solution at pH of 6 was used for all harmine measurements. The electrode showed good selectivity towards harmine over other common interfering ions, and maintained a stable electrochemical response over two weeks. After applying the validation requirements, the proposed method revealed good performance characteristics. Method precision, accuracy, bias, trueness, repeatability, reproducibility, and uncertainty were also evaluated. These analytical capabilities support the fast and direct assessment of harmine in different urine specimens. The analytical results were compared with the standard liquid chromatographic method. The results obtained demonstrated that PEDOT/PSS was a promising solid-contact ion-to-electron transducer material in the development of harmine-ISE. The electrodes manifested enhanced stability and low cost, which provides a wide number of potential applications for pharmaceutical and forensic analysis.


Polymers' Characterizations
The imprinting approach that was used for MIPs preparation of harmine was the noncovalent molecular imprinting method ( Figure 1). The amino, methoxy and N-pyridine groups of the template harmine can form strong hydrogen bonding with the functional monomer acrylamide (AA). In addition, there is a charge-transfer complex interaction that can take place between the electron-deficient aromatic ring in harmine and the electron-rich amino group of the AA monomer [40,41].
Reproducibility, conditioning, and potential-stability were investigated. The performance characteristics of the presented sensors such as linearity response, selectivity, detection limit and selectivity were also evaluated. The sensors were applied for monitoring harmine hallucinogen in urine specimens, and the results were compared with the standard gas-liquid chromatography method [39].

Polymers' Characterizations
The imprinting approach that was used for MIPs preparation of harmine was the non-covalent molecular imprinting method (Figure 1). The amino, methoxy and N-pyridine groups of the template harmine can form strong hydrogen bonding with the functional monomer acrylamide (AA). In addition, there is a charge-transfer complex interaction that can take place between the electron-deficient aromatic ring in harmine and the electron-rich amino group of the AA monomer [40,41]. As shown in Figure 2, the scanning electron microscope of either MIPs or non-imprinted polymers (NIPs) nano-beads can present a picture for the surface morphologies of the obtained polymers. The MIPs beads showed a semi-uniformed spherical shape with a size distribution of 255.8-486.2 nm, whereas the NIPs particles were of irregular and much smaller (a size distribution of 125.5-218.9 nm). This can indicate that the presence of harmine as a templated molecule has a great influence on the shape of the polymer formed, in which the template-monomer complex can change the solubility of the growing polymer. This has a great influence on altering the polymer morphology [42,43].

Membrane Optimization
These artificial receptors when dispersed in plasticized polyvinyl chloride (PVC)membranes provided highly sensitive and selective sensors for harmine monitoring. Solid-contact sensors (i.e., without internal filling solution), made of glassy carbon (GC)disks coated with the membrane cocktail were prepared, characterized and examined for determining harmine. Four membrane sensors for sensor type were designed and characterized during 8 weeks according to IUPAC recommendations [44].
For membrane optimization, the potentiometric response of the sensor is greatly influenced by the polarity of the membrane medium. Harmine-based sensor incorporating MIP nano-beads with tetrabutyl phosphate (TBP), dioctyl phosphate (DOP) and o-nitrophenyloctyl ether (o-NPOE) plasticizers were designed and checked. The calibration slope and detection limit were found to be 58.0 ± 0.7, 50.1 ± 0.3 and 45.2 ± 0.6 mV/decade and As shown in Figure 2, the scanning electron microscope of either MIPs or nonimprinted polymers (NIPs) nano-beads can present a picture for the surface morphologies of the obtained polymers. The MIPs beads showed a semi-uniformed spherical shape with a size distribution of 255.8-486.2 nm, whereas the NIPs particles were of irregular and much smaller (a size distribution of 125.5-218.9 nm). This can indicate that the presence of harmine as a templated molecule has a great influence on the shape of the polymer formed, in which the template-monomer complex can change the solubility of the growing polymer. This has a great influence on altering the polymer morphology [42,43].

Membrane Optimization
These artificial receptors when dispersed in plasticized polyvinyl chloride (PVC)membranes provided highly sensitive and selective sensors for harmine monitoring. Solidcontact sensors (i.e., without internal filling solution), made of glassy carbon (GC)-disks coated with the membrane cocktail were prepared, characterized and examined for determining harmine. Four membrane sensors for sensor type were designed and characterized during 8 weeks according to IUPAC recommendations [44].
For membrane optimization, the potentiometric response of the sensor is greatly influenced by the polarity of the membrane medium. Harmine-based sensor incorporating MIP nano-beads with tetrabutyl phosphate (TBP), dioctyl phosphate (DOP) and o-nitrophenyloctyl ether (o-NPOE) plasticizers were designed and checked. The calibration slope and detection limit were found to be 58.0 ± 0.7, 50.1 ± 0.3 and 45.2 ± 0.6 mV/decade and 0.03, 0.23 and 0.68 µg/mL upon using o-NPOE (ε = 24) instead of DOP (ε = 7) and TBP (ε = 4), respectively. As shown in Figure 3, the membranes incorporating o-NPOE revealed a more favorable calibration slope and lower limit of detection than those containing either DOP or TBP plasticizer. As a control, harmine-ISEs based on NIP nano-beads were also characterized. The electrodes based on NIP revealed a sub-Nernstian slope of 19.6 ± 0.9 mV/decade (R 2 = 0.998) with a linearity range 5.0 × 10 −5 -1.0 × 10 −2 M of harmine solution and a detection limit of 8.0 × 10 −5 M. All potentiometric features for the presented sensors were shown in Table 1.
In the presence of PEDOT/PSS as an ion to electron transducer and o-NPOE plasticizer, the sensor revealed a Nernstian slope of 59.2 ± 0.8 mV/decade (n = 5, r 2 = 0.9996) and detection limit of 0.02 µ g/mL. This confirmed that PEDOT/PSS layer has no influence on the potentiometric response of the sensor.     Table 2. Effect of solvent polarity on the selectivity behavior of harmine based sensor.

Interfering Ion
Log K pot Harmine,J + SD * o-NPOE DOP TBP * ±Standard deviation is calculated from three measurements.
In the presence of PEDOT/PSS as an ion to electron transducer and o-NPOE plasticizer, the sensor revealed a Nernstian slope of 59.2 ± 0.8 mV/decade (n = 5, r 2 = 0.9996) and detection limit of 0.02 µg/mL. This confirmed that PEDOT/PSS layer has no influence on the potentiometric response of the sensor.

Time Response and pH Effect
The time response of C/PEDOT:PSS/harmine-ISE was shown in Figure 4. The steady time attaining equilibrium is recorded in 1.0 × 10 −8 -1.0 × 10 −2 M harmine solutions. The time was found to be <10 s after a 10-fold rapid increase in concentration. The results confirmed that the presented electrode has high stability and it can be used as a rapid and automated analytical tool. Repeatability was evaluated by using 40 mM Britton-Robinson (BP) buffer, pH 6 over 2.5 h. No remarkable change in the response was observed during this interval. After re-calibration for 10 times/day, it was noticed that no significant change in the recorded slope 59.2 ± 0.8 mV/decade (n = 6) and limit of detection 0.02 µg/mL (n = 10) which reflects the high reproducibility of the presented sensor.
*±Standard deviation is calculated from three measurements.

Time Response and pH Effect
The time response of C/PEDOT:PSS/harmine-ISE was shown in Figure 4. The steady time attaining equilibrium is recorded in 1.0 × 10 −8 -1.0 × 10 −2 M harmine solutions. The time was found to be <10 s after a 10-fold rapid increase in concentration. The results confirmed that the presented electrode has high stability and it can be used as a rapid and automated analytical tool. Repeatability was evaluated by using 40 mM Britton-Robinson (BP) buffer, pH 6 over 2.5 h. No remarkable change in the response was observed during this interval. After re-calibration for 10 times/day, it was noticed that no significant change in the recorded slope 59.2 ± 0.8 mV/decade (n = 6) and limit of detection 0.02 µ g/mL (n = 10) which reflects the high reproducibility of the presented sensor. The effect of pH on the potentiometric response of C/PEDOT:PSS/harmine-ISE was checked over the pH range 2-10. The sensor revealed a constant potential response over the pH range 3-7.5, which can be taken as the working pH range for harmine assessment. At pH values > 8, the potential response of the sensor declined due to the formation of the un-protonated harmine base, which cannot be detected by the sensor. An increase in the potential response at pH values < 3 due to positive interference by H + ions. Therefore, 40 mM Britton-Robinson (BR) buffer solution, pH 6 was chosen for all subsequent measurements. The effect of pH on the potentiometric response of C/PEDOT:PSS/harmine-ISE was checked over the pH range 2-10. The sensor revealed a constant potential response over the pH range 3-7.5, which can be taken as the working pH range for harmine assessment. At pH values > 8, the potential response of the sensor declined due to the formation of the un-protonated harmine base, which cannot be detected by the sensor. An increase in the potential response at pH values < 3 due to positive interference by H + ions. Therefore, 40 mM Britton-Robinson (BR) buffer solution, pH 6 was chosen for all subsequent measurements.

Selectivity Behavior
The selectivity behavior of harmine sensors was tested towards different alkaloids, amines and inorganic cations. The selectivity coefficient values expressed as (Log K pot Harmine, J ) were calculated and are presented in Table 3. The results revealed that harmine sensors exhibited a good selectivity in the presence of many basic organic compounds. Addition of PEDOT/PSS layer has no effect on the potentiometric selectivity pattern. This is clear from the selectivity coefficient values obtained by C/PEDOT:PSS/harmine-ISE and C/harmine-ISE. The presented electrodes showed significant enhanced selectivity other than the selectivity presented by Hassan et al. [30]. This reflects the successful use of MIPs as ionophores for harmine.
a MSSM: Modified separate solution method. b SSM: Separate solution method.* ±Standard deviation of three measurements.

Impedance Spectroscopy and Chronpotentiometry Measurements
The chronopotentiograms (E versus t plots) were shown in Figure 5. The chronopotentiograms showed the potential-jump (∆E) after changing the current-direction and a potential drift (∆E/∆t) at longer times. In addition, this potential jump was used to calculate the total resistance (R b ) of the presented sensor (∆E = I · R b ), which is controlled by the bulk resistance of the ISM. The results of R b were found to be (0.23 ± 0.02 MΩ) and (0.21 ± 0.05 MΩ) for both C/PEDOT:PSS/harmine-ISE and C/harmine-ISE, respectively. The potential drift (∆E/∆t) of C/PEDOT:PSS/harmine-ISE (1.37 ± 0.5 µV/s) was found to be lower than C/harmine-ISE (63.3 ± 1.3 µV/s). The redox capacitance (C = I/(∆E/∆t)) was calculated and found to be 729.9 ± 4.5 and 15.8 ± 1.3 µF for C/PEDOT:PSS/harmine-ISE and C/harmine-ISE, respectively. The results indicated an enhanced potential stability and high capacitance for the presented sensor upon the addition of PEDOT/PSS as a solid contact material. Impedance measurements were carried out for both C/PEDOT:PSS/harmine-ISE and C/harmine-ISE. The impedance spectra were recorded at the open-circuit potential in 10 mM HMR solution and are shown in Figure 6. From the high-frequency semicircle part, the total resistance (R) was found to be 0.25 ± 0.03 and 0.18 ± 0.04 MΩ for both C/PEDOT:PSS/harmine-ISE and C/harmine-ISE, respectively. The capacitance (C) for either C/PEDOT:PSS/harmine-ISE or C/harmine-ISE was estimated from the low-frequency semicircle part. The capacitance (C) was found to be 722.6 ± 3.5 and 13.8 ± 0.7 µF for either C/PEDOT:PSS/harmine-ISE or C/harmine-ISE, respectively. The obtained results confirmed that the lipophilic character of PEDOT/PSS layer can generate a large redox capacitance, which is responsible for the enhanced potential stability of the presented ISEs.

Water-Layer Test
A severe potential drift can arise in the ISE because of the existence of the water-layer between the ISM and the electronic substrate. This potential drift can be eliminated after the insertion of a hydrophobic solid-contact layer at the interface between the ISM and the

Water-Layer Test
A severe potential drift can arise in the ISE because of the existence of the water-layer between the ISM and the electronic substrate. This potential drift can be eliminated after the insertion of a hydrophobic solid-contact layer at the interface between the ISM and the electronic conductor. The water-layer test is performed to distinguish the existence of this layer. The electrodes were inserted at first in 10 −2 M NaCl for 7 h and then inserted in 10 −5 M HMR solution for another 6 h, then changed back to 10 −2 M NaCl solution. The potential was recorded over all these intervals. As shown in Figure 7, C/PEDOT:PSS/harmine-ISE revealed higher potential-stability than C/harmine-ISE, especially when going back to the harmine solution. This confirms the non-existence of the water-layer in the C/PEDOT:PSS/harmine-ISE after the insertion of the lipophilic PE-DOT:PSS layer. The long-term stability of both C/PEDOT:PSS/harmine-ISE and C/harmine-ISE was calculated from the potential-response at the final part of the water-layer test (Figure 7). The potential drift obtained was calculated and found to be 0.1 and 4.2 mV/h for both C/PEDOT:PSS/harmine-ISE and C/harmine-ISE, respectively. This confirmed that the insertion of the PEDOT:PSS layer enhanced the potential-stability and reflected the high lipophilicity of this solid-contact material.

Analytical Assessment of Harmine
The presented C/PEDOT:PSS/harmine-ISE was introduced to determine harmine in urine specimens spiked with different amounts of harmine. Each sample was analyzed in triplicates and the results were in comparison with the standard liquid chromatographic method (HPLC) as a comparative technique. All results were shown in Table 4. The average recoveries varied between 94-102% and 98-105% for the presented potentiometric method and HPLC method, respectively. The t-student and F-tests data emphasized that there is no observed difference between the set of results obtained by the presented potentiometric method and the standard HPLC method. The presented potentiometric method revealed an enhanced applicability as a new methodology for harmine assessment.
DOT:PSS/harmine-ISE after the insertion of the lipophilic PEDOT:PSS layer. The longterm stability of both C/PEDOT:PSS/harmine-ISE and C/harmine-ISE was calculated from the potential-response at the final part of the water-layer test (Figure 7). The potential drift obtained was calculated and found to be 0.1 and 4.2 mV/h for both C/PE-DOT:PSS/harmine-ISE and C/harmine-ISE, respectively. This confirmed that the insertion of the PEDOT:PSS layer enhanced the potential-stability and reflected the high lipophilicity of this solid-contact material.

Analytical Assessment of Harmine
The presented C/PEDOT:PSS/harmine-ISE was introduced to determine harmine in urine specimens spiked with different amounts of harmine. Each sample was analyzed in triplicates and the results were in comparison with the standard liquid chromatographic method (HPLC) as a comparative technique. All results were shown in Table 4. The average recoveries varied between 94-102% and 98-105% for the presented potentiometric method and HPLC method, respectively. The t-student and F-tests data emphasized that there is no observed difference between the set of results obtained by the presented potentiometric method and the standard HPLC method. The presented potentiometric method revealed an enhanced applicability as a new methodology for harmine assessment. Table 4. Harmine assessment in spiked urine specimens using C/PEDOT:PSS/harmine-ISE.

Biomimics Synthesis
"The main-tailored beads were prepared via a thermal polymerization process. A 1.0 mmol of HME (templated molecule) and 3.0 mmol of AA as a functional monomer were pre-complexed together for 30 min and then, they were mixed together with 3.0 mmol of EGDMA as a cross-linking reagent and 80 mg of BPO as an initiator. The mixture was placed in a capped-glass bottle and 25-mL acetonitrile as a porgenic solvent was added. All dissolved oxygen was removed by passing a flow stream of N 2 gas for 15 min. Sonication of the solution for 15 min is enough for complete solution homogenization. The polymer was obtained after heating the reaction mixture for 18 h at 80 • C in an oil. Using Soxhlet extraction, harmine molecules were completely removed using a solution mixture of CH 3 COOH/CH 3 OH (1:9, v/v). The obtained nano-beads were washed several times with CH 3 OH and then, they were left to dry. Non-imprinted polymers (NIPs) were also prepared via the same steps but in absence of an HME molecule and under the same conditions". PVC; all in 2-mL THF. Afterward, the membrane was left to dry until a uniform shape was obtained with good adhesion to the GC substrate. All coated-wire electrodes (CWEs) were also fabricated according the above-mentioned steps but without insertion of the PEDOT/PSS transducing-layer. The prepared SC-ISEs were conditioned in 1 mM harmine solution for 4 hrs and then conditioned in 10 −8 M harmine for 24 h. When not in use, the sensors were kept in the same solution".

Electrochemical Measurements
"All the pre-conditioned sensors were calibrated by spiking 1.0 mL of 10 −1 -10 −5 M HME solution in 9-mL BR buffer. The sensors in conjunction with an Ag/AgCl double junction reference electrode were inserted in the test solution and the potential arisen from each harmine concentration was measured. The potential readings were recorded after stabilization to ±0.2 mV and plotted versus log (harmine) concentration. The constructed calibration plots were used for all subsequent measurements". Potentiometric selectivity coefficients (K Pot harmine,B ) were evaluated using the modified separate solution method (MSSM) [45] and calculated from the equation: where E o 1 and E o 2 represent the potential readings measured by harmonium ion and the interfering ion at the extrapolated calibration curve (a = 1 M), respectively, and S is the slope observed for the primary cation.
Time recovery was evaluated after continuous reading of the potential during the concentration range from 10 −7 to 10 −2 M harmine hydrochloride solutions. Effect of pH on the potentiometric response was also investigated after measuring the potential of the proposed sensor in 10 −3 M harmine solution and changing the pH of this solution from 2 to 10.
Impedance measurements were performed in 10 mM harmine hydrochloride solution within the frequency range 0.01-10 5 Hz using 0.01 V amplitude at open circuit potential 0.2 V. The working electrodes were the studied electrodes, the reference electrode was Ag/AgCl (saturated KCl), and Pt wire was used as the auxiliary electrode. Chronopotentiomtric measurements were also performed through the three-electrode cell and a reversed current of the value ±1 nA was applied to the working electrode for 60 s as previously described by Bobacka' s protocol [46].

Assessment of Harmine in Urinary Specimens
Different urine samples were collected from two different volunteers. After collection, samples were stored untreated at −20 • C until further analysis. The samples were spiked with stock solution of harmine. Each 0.1 mL of fresh urine sample was taken and diluted to 10-mL of Britton-Robinson buffer at pH 6 and then directly analyzed.

Conclusions
In this work, we designed reliable, robust, cost-effective solid-contact potentiometric electrodes for harmine detection. The trace-level assessment of harmine was achieved by integrating a man-tailored artificial receptor for harmine as an ionophores and PEDOT/PSS as a lipophilic solid-contact material. The observed recommended features of PEDOT/PSS as an ion-to electron transducer were reinforced by these studies for designing solid-contact ISEs. The electrodes exhibited a fast-response towards harmine with a Nernstian slope of 59.2 ± 0.8 mV/decade (n = 5, r 2 = 0.9996) over the linear range of 1.0 × 10 −7 -1.0 × 10 −2 M and a detection limit of 0.02 µg/mL. Reasonable selectivity over different common organic and inorganic ions, good accuracy, and possible interfacing with automated systems were presented. These results demonstrated that PEDOT was a promising solid-contact ion-toelectron transducer material in the development of harmine-ISE. The electrodes manifested enhanced stability and low cost that provides a wide number of potential applications for pharmaceutical and forensic analysis. The sensor was successfully applied in monitoring harmine in different urine specimens. The analytical results were compared and agree fairly well with those obtained by gas-liquid chromatography.