Organic electrochemical transistors, or OECTs, have quietly become one of the most versatile building blocks in organic bioelectronics. They are being explored for biosensing, biomolecular detection, monitoring and stimulation of cellular activity, and even neuromorphic computing architectures, largely because they combine mechanical flexibility, biocompatibility and mixed ionic–electronic conductivity in a single device. Yet despite a decade of rapid progress, the details of what happens at the interfaces inside these devices — where the gate meets the electrolyte and where the channel meets the electrolyte — have remained stubbornly difficult to pin down. A new study published in Advanced Modeling and Simulation in Engineering Sciences tackles that gap head-on, pairing a detailed multiphysics model with careful device fabrication to show exactly how interfacial electrochemistry governs transistor performance.
The work, led by Vijay Kumar Lamba, Sankit Kassa, Deepika Lamba, Aditi Kalsh, Archana Kumari and Pratibha Thakur, centers on transistors built from poly(3,4-ethylenedioxythiophene) doped with poly(styrene sulfonate), the workhorse conducting polymer known as PEDOT:PSS. In an OECT, a thin film of this polymer connects source and drain electrodes, while an electrolyte mediates the coupling to a gate electrode. Applying a gate voltage drives cations from the electrolyte into the polymer, where they compensate the negatively charged sulfonate groups and reduce the oxidized PEDOT⁺ species to its neutral form. This reversible de-doping lowers the hole density in the channel and, with it, the drain current — an electrochemical switching mechanism that gives OECTs their characteristically high transconductance.
What sets the new study apart is the level of quantitative rigor brought to the interface problem. The researchers modeled the potential distribution within the device using the Poisson–Boltzmann formalism with Stern-layer corrections, traced ion-induced doping and de-doping behavior with the Nernst–Planck and Nernst equations, and ran finite-element simulations to track changes in charge mobility and the electrical double layer. Crucially, rather than assuming a fixed interfacial capacitance — an assumption the authors found introduced significant errors at high gate bias — they fed experimentally extracted capacitance values from impedance measurements directly into the simulations. Incorporating these measured parameters reduced the mean absolute error of the model from 8.7 percent to 5.2 percent.
The experimental side of the study was equally deliberate. The devices use carbon paper gate electrodes modified with platinum nanoparticles to enhance electrochemical reactivity, and PEDOT:PSS formulations tuned with ethylene glycol, dodecylbenzenesulfonic acid and a GOPS crosslinker. The electrolyte was 0.01 M potassium chloride in a borate buffer with ferrocene methanol to boost redox-mediated charge transfer. Source and drain gold electrodes were patterned by photolithography with a 5 micrometer line width, and the polymer channel was deposited by spin coating and annealed under nitrogen. Film thicknesses from 50 to 500 nanometers were measured by spectroscopic ellipsometry, with 27 thickness measurements per condition across three independently fabricated samples — and, notably, all simulations were repeated after peer review using the measured rather than nominal thickness values, with uncertainty propagation shown in the figures.
One of the clearest findings concerns how film thickness controls conductivity. The measured conductivity climbed from 330 ± 15 S cm⁻¹ at roughly 47 nanometers to 603 ± 6 S cm⁻¹ at nearly 500 nanometers, with the steepest rise occurring between about 50 and 150 nanometers. The team interprets this as a crossover from percolation-dominated transport, where conductive PEDOT pathways are incomplete, to bulk-like charge transport through a fully interconnected network. Atomic force microscopy backed this up, showing surface roughness falling from 1.8 to 0.7 nanometers as films thickened and densified. Beyond roughly 150 to 200 nanometers, conductivity increased only gradually, signaling the dominance of bulk conduction.
Thickness also proved decisive for transistor performance, but in a more nuanced way. Thinner channels, between 50 and 100 nanometers, delivered reduced drain current because of higher sheet resistance and incomplete percolation. As thickness grew to 150–200 nanometers, the polymer network became fully connected and both transconductance and on/off ratio peaked: at 150 nanometers the peak transconductance was 15.8 ± 0.4 mS with an on/off ratio of 580 ± 30. Pushing past 300 nanometers, however, degraded performance, dropping to 12.1 ± 0.5 mS and 480 ± 25 at 300 nanometers and 7.8 ± 0.4 mS and 350 ± 20 at 500 nanometers. The explanation lies in the finite penetration depth of ions — roughly 150 to 200 nanometers in this system — which leaves the interior of thick films as a weakly modulated “dead layer” that conducts but barely responds to the gate.
Gate geometry mattered as well, though with diminishing returns. Enlarging the gate area from 0.12 to 2.00 square millimeters — a factor of 16.7 — raised the peak drain current by 138 percent, from −26 to −62 microamperes, and boosted transconductance by 156 percent, from 4.8 to 12.3 mS, with the maximum transconductance of 12.3 mS recorded at a gate voltage of −0.52 V. The on/off ratio improved from 260 to 620. But the sub-linear scaling — only a 2.4-fold current increase for a 16.7-fold area increase — reveals a saturation effect rooted in the limited ionic mobility of the bulk electrolyte. Once the gate’s ion-supplying capacity exceeds a critical threshold, ion movement through the electrolyte and into the polymer becomes the rate-limiting step, so geometry alone cannot indefinitely improve performance. Interestingly, the threshold voltage stayed constant at about −0.15 ± 0.05 V across all gate sizes, confirming it is set by electrochemistry rather than layout — an advantage for circuit design, since it can be tuned through material chemistry instead of geometry.
The simulations also mapped the internal electrostatics of the operating device in fine detail. The computed potential profiles show well-defined double layers at the gate interfaces and a distinct potential minimum about 1.8 micrometers from the source, where PEDOT⁺ is reduced to its neutral form — a reduction zone that breaks local charge neutrality and defines the active modulation region of the channel. The depth of this potential well grows with applied gate voltage, and the computed values agree closely with published operando measurements, matching a literature value at 96 percent agreement. The analysis further suggests that gate biases between 0.6 and 0.8 V induce internal field gradients of up to 4.2 × 10⁵ V/m, which may contribute to non-linear behavior and the increased degradation rates observed at the most negative gate voltages.
Stability under continuous operation is often the Achilles heel of organic electrochemical devices, and here the results are striking. Eight independently fabricated devices were run under continuous pulsed bias — the gate voltage pulsed between 0 V and −1.0 V at 0.1 Hz under ambient conditions — for more than 500 hours. Over that period, the ensemble-averaged drain current degraded by only 13.5 percent and transconductance by 16.0 percent, following first-order kinetics with a time constant of 1200 ± 40 hours, equivalent to a degradation rate of about 0.027 percent per hour. The on/off ratio after 500 hours was still 540, and the threshold voltage shifted by only −23 ± 10 mV, a slight negative drift attributed to gradual accumulation of trapped charge. The authors attribute this durability to the GOPS-crosslinked film formulation, which reduces swelling and delamination; the platinum-nanoparticle gate, which stabilizes the gate potential; and the pulsed operation scheme, which lets ions relax between pulses.
The final measure of the framework is how well simulation and experiment agree across a full device population. Comparing simulated and measured transfer characteristics from 35 devices — 805 measurement points in total — the model achieved a Pearson correlation coefficient of R² = 0.987, a root mean square error of 2.8 microamperes, and a mean relative error of 1.8 percent, with 94.2 percent of data points falling within ± 6 percent relative error and 98.5 percent within ± 10 percent. Electrochemical impedance spectroscopy, analyzed with an equivalent circuit capturing both interfaces plus a Warburg diffusion term, passed Kramers–Kronig consistency checks and revealed two distinct relaxation processes corresponding to the gate–electrolyte and channel–electrolyte interfaces, with faster ionic diffusion at higher bias. Device-to-device variability also shrank from roughly 18 percent at low gate bias to about 4 percent at high bias, indicating that gate-controlled doping dominates over fabrication variation in the on-state. Together, the study delivers both mechanistic insight into interface-controlled electrochemical gating and practical design guidelines — optimal thickness near 150–200 nanometers, gate voltages between −0.5 and −1.0 V, and area scaling within electrolyte transport limits — for the next generation of OECT biosensors and organic bioelectronic devices.
Subject of Research: Multiphysics modelling and interface optimization of PEDOT:PSS organic electrochemical transistors for biosensing
Article Title: Multiphysics modelling and interface optimization of PEDOT:PSS OECTs for stable and sensitive biosensing
Article References: Lamba, V. K., Kassa, S., Lamba, D., Kalsh, A., Kumari, A., & Thakur, P. (2026). Multiphysics modelling and interface optimization of PEDOT:PSS OECTs for stable and sensitive biosensing. Advanced Modeling and Simulation in Engineering Sciences, 13(1), Article 14. https://doi.org/10.1186/s40323-026-00336-7
Image Credits: AI Generated
DOI: 10.1186/s40323-026-00336-7
Keywords: organic electrochemical transistors, PEDOT:PSS, biosensing, multiphysics modelling, Poisson-Boltzmann, Nernst-Planck, electrochemical impedance spectroscopy, operational stability, transconductance, thin-film transport, organic bioelectronics, finite-element simulation
News Source: Denise Maddox. (October 29, 2026). Multiphysics Model and Interface Engineering Push PEDOT:PSS Transistors Toward Stable Biosensing. Scienmag.
