Enhancing Amperometric Transducer Selectivity with Nanoscale Polyphenylenediamine Films

Background

Biosensors offer a cost‑effective alternative to chromatography, spectroscopy, and colorimetry, yet often lag behind in analytical performance. Current research in biosensorics is rapidly evolving [1]. According to the International Association of Researchers in Fundamental and Applied Chemistry, a biosensor is an integrated device that couples a biological recognition element with a transducer to provide quantitative or semi‑quantitative analysis [2]. Electrochemical biosensors—amperometric, potentiometric, conductometric, and impedimetric—constitute the largest subgroup [3].

Selective detection is critical when analysing complex biological matrices such as blood serum, urine, or cerebrospinal fluid, where endogenous molecules (ascorbic acid, cysteine, homocysteine, uric acid, dopamine, glutathione) can produce false signals. The selectivity of a biosensor stems from both the biological receptor and the transducer; enzymes and antibodies are inherently selective, whereas electrodes are typically non‑selective. Therefore, strategies to shield the electrode from interferences are essential.

Two primary strategies exist: lowering the operating potential or adding a semipermeable membrane that allows the target analyte while rejecting larger interferents [4]. Deposition of nanostructured polymer films is straightforward and minimally affects sensor performance. Among polymeric membranes, phenylenediamine (PD) derivatives have attracted attention for their nanoporous architecture that permits hydrogen peroxide diffusion while blocking larger species [5]. Polyphenylenediamine (PPD) films exhibit pores that accommodate small molecules like H₂O₂ but exclude larger molecules such as ascorbic acid or dopamine, thereby enhancing sensor accuracy. Previous work has explored different PD isomers and polymerization methods; meta‑PD (m‑PD) generally yields the most permeable yet selective membranes, especially when deposited by cyclic voltammetry [6–9]. Recent studies also demonstrate PD‑based membranes in non‑enzyme sensors, e.g., for bovine serum albumin detection via fluorescence quenching [10].

Our objective was to compare two common deposition techniques—cyclic voltammetry and constant‑potential electropolymerization—and identify the optimal protocol for creating PPD membranes on platinum disk electrodes.

Methods

Materials

Ascorbic acid, cysteine, uric acid, dopamine, hydrogen peroxide, m‑phenylenediamine, and HEPES were purchased from Sigma‑Aldrich (USA). Human serum samples were sourced from the Kyiv Municipal Scientific and Practical Center of Nephrology and Hemodialysis (Ukraine).

Design of Amperometric Transducers

Custom platinum disk electrodes were fabricated by sealing a 0.4 mm diameter, 3 mm long platinum wire into a 3.5 mm outer diameter glass capillary. The exposed wire end served as the working surface. The inner wire end was soldered to a silver wire using Wood’s alloy and connected to the potentiostat. Prior to use, the electrodes were cleaned with 30 s HCl, rinsed with ethanol, and polished with P1500 abrasive paper.

Measurement Methods

UV‑vis diffuse‑reflectance spectra were recorded on a Thermo Evolution 600 spectrometer (200–900 nm) using a Spectralon standard. Electrochemical measurements employed a standard three‑electrode cell (working, platinum wire auxiliary, Ag/AgCl reference) connected to a PalmSens potentiostat with an 8‑channel multiplexer. Chronoamperometry was performed at 0.6 V vs. Ag/AgCl in 10 mM HEPES (pH 7.4) with a 3 mL glass cell under continuous stirring. Substrate concentrations were prepared by adding aliquots of stock solutions (50 mM H₂O₂, 20 mM ascorbic acid, 3 mM cysteine, 4.5 mM uric acid, 2.1 mM dopamine). Phenylenediamine was dissolved in 40 mM phosphate buffer (pH 7.4). Cyclic voltammetry used 0–0.9 V scan range at 20 mV s⁻¹, 5 mV steps. All experiments were triplicated; data represent mean ± SD (OriginPro 8.5).

Results and Discussion

To assess the benefit of a PPD layer, we first evaluated the baseline sensitivity and selectivity of bare platinum electrodes toward hydrogen peroxide and common interferents. Using three concentration levels—(1) physiological (serum) concentrations, (2) 20‑fold dilution, and (3) 100‑fold dilution—we observed that dopamine and ascorbic acid elicited the strongest signals, whereas cysteine produced the weakest. Even after 100‑fold dilution, interferent responses remained about 20% of the H₂O₂ signal, indicating that a bare electrode alone is insufficient for complex biological samples.

We then compared two deposition strategies. In the CV method, the electrode was subjected to successive potential sweeps (CVA) in 30 mM PD solution; each CVA took ~2 min. In the constant‑potential method, the electrode was held at +0.7 V for 40 min in the same solution. Table 3 shows that membranes formed by CV significantly suppressed interferents (especially cysteine) and increased H₂O₂ sensitivity by 2.6×, likely due to platinum surface activation. The constant‑potential approach did not enhance sensitivity and required longer deposition time. Therefore, CV is preferable for its speed, selectivity, and sensitivity gains, despite its limitation to a single electrode per sweep.

Spectroscopic analysis confirmed the formation of PPD: the diffuse‑reflectance spectrum of the film displays intense bands at 222 nm and 315 nm (monomer‑like π→π* transitions) and a broad tail from 400–800 nm indicative of extended conjugation in the polymer.

We optimized the number of CVAs and PD concentration. A single CVA produced an insufficient membrane; increasing to three CVAs eliminated dopamine and uric acid responses while maintaining H₂O₂ sensitivity. Reducing PD to 5 mM preserved selectivity for diluted samples but compromised performance with undiluted matrices. A concentration of 30 mM PD with three CVAs yielded complete interference suppression and acceptable H₂O₂ response; higher concentrations (>50 mM) thickened the membrane and reduced sensitivity.

Stability tests revealed that the membrane retained its selectivity during 2 h of continuous operation, with slight improvements in H₂O₂ response and further suppression of interferents—likely due to pore clogging by larger species. Storage at –18 °C for 8 days increased H₂O₂ sensitivity by 2.5× without affecting interferent rejection, suggesting slow membrane swelling enhances permeability.

Finally, application to real biological samples (serum and neuronal lysate) demonstrated that unmodified electrodes produced noisy signals, whereas PPD‑modified electrodes showed negligible responses, confirming their suitability for complex matrices.

Conclusions

Electropolymerization of polyphenylenediamine via cyclic voltammetry offers a rapid, effective way to create semipermeable membranes that suppress interferences on platinum disk amperometric transducers. Three CV scans in 30 mM PD completely eliminate the impact of common interferents, while 5 mM PD is adequate for diluted samples to preserve higher H₂O₂ sensitivity. The resulting membranes remain selective for at least 2 h of operation and are stable after 8 days of frozen storage. These findings support the use of PPD‑modified electrodes in the design of reliable biosensors for complex biological fluids.