Single‐Component Electroactive Polymer Architectures for Non‐Enzymatic Glucose Sensing

Abstract Organic mixed ionic‐electronic conductors (OMIECs) have emerged as promising materials for biological sensing, owing to their electrochemical activity, stability in an aqueous environment, and biocompatibility. Yet, OMIEC‐based sensors rely predominantly on the use of composite matrices to enable stimuli‐responsive functionality, which can exhibit issues with intercomponent interfacing. In this study, an approach is presented for non‐enzymatic glucose detection by harnessing a newly synthesized functionalized monomer, EDOT‐PBA. This monomer integrates electrically conducting and receptor moieties within a single organic component, obviating the need for complex composite preparation. By engineering the conditions for electrodeposition, two distinct polymer film architectures are developed: pristine PEDOT‐PBA and molecularly imprinted PEDOT‐PBA. Both architectures demonstrated proficient glucose binding and signal transduction capabilities. Notably, the molecularly imprinted polymer (MIP) architecture demonstrated faster stabilization upon glucose uptake while it also enabled a lower limit of detection, lower standard deviation, and a broader linear range in the sensor output signal compared to its non‐imprinted counterpart. This material design not only provides a robust and efficient platform for glucose detection but also offers a blueprint for developing selective sensors for a diverse array of target molecules, by tuning the receptor units correspondingly.


General experimental
1 H NMR spectra were recorded at 400 MHz on a Bruker Avance III spectrometer.Chemical shifts (δ) are quoted to the nearest 0.01 ppm relative to tetramethylsilane, with the residual solvent peak used as the internal standard: CHCl3 (7.26 ppm), DMSO (2.50 ppm) or CH3CN (2.10 ppm).Coupling constants (J) are given to the nearest 0.1 Hz.Peak multiplicities for resonances are noted as: s, singlet; d, doublet; dd, doublet of doublets; t, triplet; q, quartet; m, unresolved multiplet. 13C NMR spectra were recorded at 101 MHz on a Bruker Avance III spectrometer.Chemical shifts (δ) are quoted to the nearest 0.1 ppm, with reference to the given solvent CDCl3 (77.0 ppm) or DMSO-d6 (39.5 ppm) as the internal standard.
All mass spectra were obtained from the analytical laboratory in the Department of Chemistry at Queen Mary University of London as follows: low resolution mass spectra were obtained to 1 d.p. from either an Agilent 1100 Series Liquid Chromatograph with a SL Ion Trap mass selective detector, or an Agilent 6890N Gas Chromatograph with an Agilent 5973N mass selective detector, or a Bruker Autoflex MALDI-TOF spectrometer.High resolution mass spectra were obtained to 4 d.p. from a Waters Synapt G2-Si High-Definition Mass Spectrometer.Melting points were obtained using a Stuart SMP11 melting point apparatus and are uncorrected.

Synthesis
Note: compounds EDOT-Cl and EDOT-N3 are the same intermediates reported in our previous work; 1 for completeness, syntheses and characterization data are reproduced here.

2-(Chloromethyl
Under an atmosphere of nitrogen, 3,4-dimethoxythiophene (5.0 g, 35 mmol, 1.0 eq.), 3-chloro-1,2-propanediol (7.25 mL, 88 mmol, 2.5 eq.) and p-toluenesulfonic acid monohydrate (66 mg, 0.35 mmol, 1 mol%) were dissolved in anhydrous toluene (140 mL) in a 250 mL 2-necked round bottom flask equipped with distillation apparatus.The solution was heated at 95 °C for 24 h, over which time a black oily precipitate was formed.After this time, another equal portion of the diol (7.25 mL, 88 mmol, 2.5 eq.) was added and the mixture was stirred at the same heat for a further 4 days, then cooled to room temperature.The toluene solution was decanted from an insoluble dark blue oil and concentrated in vacuo.The crude material was purified by chromatography over silica gel, with the product EDOT-Cl eluting in 7:3 hexane:DCM as a white solid (2.76 g, 42%).Rf = 0.25 Under an inert atmosphere, EDOT-Cl (100 mg, 0.52 mmol, 1.0 eq.) was dissolved in anhydrous DMF (5.3 mL).To this solution, NaN3 (44 mg, 0.68 mmol, 1.3 eq.) was added and the solution was heated to 120 °C and stirred at this temperature for 24 h.After this time, the solution was cooled to room temperature and H2O (5.0 mL) was added.The mixture was extracted with EtOAc (50 mL) and washed with H2O (1 × 50 mL) and brine (3 × 50 mL).

EDOT-PBA
Under anhydrous conditions in a degassed 5 mL microwave vial, EDOT-PBr (1.79 g, 5.06 mmol, 1.0 eq.), KOAc (1.24 g, 12.7 mmol, 2.5 eq.) and chloro[(tri-tert-butylphosphine)-2-(2aminobiphenyl)] palladium(II) (P(t-Bu)3 Pd G2, 25.9 mg, 50.6 µmol, 1.0 mol%) were suspended in degassed MeOH (33 mL).Separately, a mixture of ethylene glycol (1.79 mL, 31.6 mmol, 6.25 eq.) and tetrahydroxydiboron (567 mg, mmol, 1.2 eq.) in degassed MeOH (17 mL) was prepared, and both mixtures were heated to 35 °C.After 10 min at this temperature, the tetrahydroxydiboron was fully dissolved, and this solution was added dropwise slowly to that of the combined solid reagents.The reaction was stirred at 35 °C for 16 h, after which it was cooled to ambient temperature, and H2O (50 mL) was added.The suspension was filtered, and a 1:1 mixture of H2O:MeOH (2 × 50 mL) followed by H2O (200 mL) were passed through the grey solid.The filtrate was left to stand for 24 h, upon which the MeOH was evaporated and a white precipitate was formed, which was collected and triturated in Et2O (100 mL).The product was filtered and dried under vacuum at 40 °C for 24 h, to give the product EDOT-PBA as a white solid (1.08 g, 67%).While this combined evidence paints a picture of a complex system of mixed 1:1 and 2:1 binding occurring overall, we establish with confidence the novel monomer EDOT-PBA binding to the template methyl ⍺-D-glucopyranoside in acetonitrile.

Spectroelectrochemistry
In this analysis, a stepwise increase in voltage was applied to NIP or MIP films on indium tin oxide (ITO)-coated glass slides in 1X PBS, while concurrently performing UV-Vis-NIR measurements.In many ways, the two systems exhibit similar behavior (Figure S5): discrete polaronic, then bipolaronic species formation phases up to and after 0.0 V are observed, with comparable absorption λmax values of 592 nm (NIP) and 598 nm (MIP) for the neutral absorption band, and 940 nm and 956 nm respectively for the polaron; while both architectures demonstrate efficient bleaching of the neutral absorption feature mostly within a narrow voltage range of -0.3 V to 0.0 V (Figure S6).However, for the NIP film, a well-defined shoulder in the neutral absorption band is present at 640 nm, which is quenched at a lower potential in comparison to the main neutral peak.In the solid state, this feature can be attributed to the 0-0 vibronic transition in a H-or J-like aggregated system, which denotes a high degree of order corresponding to backbone planarity in the polymer. 9,10This feature is far less pronounced for the MIP film, indicating a more amorphous structure for this film architecture.

Figure S6 .
Figure S6.Bleaching of neutral absorption band (absorbance difference at λmax for increasing potentials vs. neutral polymer at -0.6 V) for NIP (recorded at 592 nm) and MIP (recorded at 598 nm).

Figure S7 .
Figure S7.Electrochemical impedance spectra (Bode plots) of (a) NIP and (b) MIP polymer films before and after interactions with glucose at different concentrations.The circuits shown in the insets were used to fit the spectra.RSOL is solution resistance.R, CPE, and C are the resistance, constant phase element, and capacitance of the electrode material, while W is the Warburg element.

Figure S8 .
Figure S8.The OECT biosensor characteristics gated with the MIP electrodes.(a) The output characteristics of the OECT with the MIP electrode as the gate in PBS.Transfer characteristics of the OECT channel gated with the MIP electrodes exposed to (b) glucose and (c) fructose solutions of varying concentrations.

Figure S9 .
Figure S9.The specificity of MIP gated OECT biosensors.(a) Response to common interferents or abundant molecules in human serum samples at their physiological concentrations.(b) Response to PBS and commercial serum with and without the spiked additional glucose molecules.

Table S1 .
Parameters obtained from fitting the EIS spectra

Table S2 .
Statistical data for the calibration plots of glucose detection using NIP and MIP gate sensors The LOD for the NIP is calculated using the linear equation for the lower concentration range.

Table S3 .
Comparison of LOD values for non-enzymatic electrochemical PBA-based glucose sensors