Implementation and simulation of drift-diffusion models for organic mixed conductor devices

• Ano de defesa: 2022
• Autor(a) principal: Unigarro, Andres David Peña
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Diferenças finitas; Drift-diffusion; Eletrônica orgânica; Finite differences; OECT; Organicelectronics
• Link de acesso: https://www.teses.usp.br/teses/disponiveis/76/76134/tde-19082022-111554/

In the past years, organic electrochemical transistors (OECTs) have emerged as potential transducers inapplications that require the conversion of ion fluxes to electronic current. For the rational optimization and understanding of the fundamentals of OECTs and OECT-based applications, however, it is essential to have theoretical models capable to predict experimental data. Within drift-diffusion models,ion flux from the electrolyte into the organic semiconducting layer is considered to take place due to the action of an electrical field, but also because of diffusion processes generated by the concentration gradients. The governing equation of the drift-diffusion model is the Nernst-Planck equation.Thus, in this project, a numerical approach is followed in order to solve the Nernst-Planck equation in one dimension,and model the ion migration from the electrolyte to the semiconductor. To evaluate the accuracy of the implementation, standard boundary conditions used in the literature to solve analytically the drift-diffusion equations were considered.In doing so,the numerical results were in good agreement with the analytical solutions,achieving maximum errors in the order of 1%. Aiming to a better representation of OECTs, closed boundary conditions are considered. Here, the temporal evolution of the concentration profiles showed a convergence to an exponential steady state distribution, which is in good agreement with the result expected theoretically. A further situation investigated was the consideration of a non-uniform electric field acting on the system, assumed to be finite in the electrolyte regionandzerointhe semiconductor.This Consideration impacts principally the temporal evolution of the concentration in each region. In order to consider the distinct compositions in electrolyte and semiconductors, different values of diffusion coefficients were introduced for each region. This extension has visible impacts in the time that the system needs to achieve the steady state. Moreover, the introduction of the chemical potential gradient as the driving force of diffusion leaded to significant variations in the results obtained with the model. Here,the so-called uphill diffusion reported in the literature was observed. With the numerical approach,it was possible to consider different types of pulsed gate voltages, which allowed to simulate the charge and discharge processes of OECTs. For all cases, oscillatory curves similar to experimental measurements were obtained. Therefore, the numerical approach allowed to go beyond the analytical description, and develop an extensive investigation of the impact that different considerations have in the results.