(305f) Kinetic Monte Carlo Simulation of the Interfacial Electrochemistry in a Hydrogen-Powered Solid-Oxide Fuel Cell

Turner, C. H., University of Alabama
Wang, X., University of Alabama
Dunlap, B., Naval Research Laboratory
Lau, K. C., George Washington University

A kinetic Monte carlo (KMC) model is developed to simulate the elementary chemical reactions taking place in a Yttria Stabilized Zirconia (YSZ) fuel cell, in order to translate experimental, and ultimately theoretical rates into an atomistic model of the solid-oxide fuel cell (SOFC). While our original model focused exclusively on the cathode-side of a SOFC model, our more recent efforts will be presented, which incorporate both the cathode- and anode-side of the SOFC. The KMC model consists of a set of several electrochemical reaction rates, adopted from experiments and first-principles calculations. The KMC simulations are used to model these simultaneously occurring events, in order to determine potential limitations in the SOFC performance at different operating conditions.

The main focus of this work is the ionic current density (J), studied as a function of various physical parameters: partial pressure of the gases (PO2, PH2, PH2O), external applied bias voltage (Vext), temperature (T), dopant concentration (mol% Y2O3), relative permittivity (εr) of YSZ, and geometrical features of the YSZ electrolyte. This simple model can be used as a baseline to translate elementary chemical reaction rates into atomistic simulations of working SOFCs, pertinent to the experimental operating conditions. As a supplement to conventional macroscopic scale modeling, atomistic modeling of fuel cell operation is important for understanding the underlying physical processes and developing a truly predictive model. Understanding SOFC performance at microscopic length and time scales (i.e., identifying the possible reaction pathways, as well as the effect of local electric fields on chemical reactions, ion and vacancy diffusion rates, and chemical interactions at the three-phase boundary) will be helpful for determining various boundary conditions in macroscopic studies for the general design of fuel cells.

The cathodic and anodic reactions can simultaneously occur via several competing paths, but here we assume that they can all be represented by a minimal set of a few well-defined reaction rates. For each of the allowed reaction pathways, all of the physical parameters influencing the relevant reactions and transport are incorporated within the framework of kinetic Monte Carlo simulations, including polarization resistance and effects from the electrostatic fields. In order to better understand the atomistic behavior of the YSZ model, our simulated predictions of the ionic current density, J (mA/cm2), are divided into three general categories, based upon the simulation parameters being investigated:

1. Materials independent: gas partial pressure (PO2, PH2, PH2O), external applied bias voltage (Vext), and temperature (T)

2. Materials dependent: doping levels (i.e., concentration of Y in YSZ) and relative permittivity (εr)

3. Geometrical parameters: surface area (A) and electrolyte thickness (D)

To identify the influence of each of the parameters in a well-defined manner, we varied each of these parameters independently, while keeping all others fixed during the simulation. Moreover, some parameters (PO2, PH2, PH2O, T, Vext, etc.) are rather easy to control experimentally, whereas other parameters (e.g., impurity segregation, thermally or electrically induced chemical and morphological changes of the electrode/electrolyte interfaces), are experimentally ill-defined and a methodical variation and ultimate prediction of these influences is beyond the scope of this study.

Broadly speaking, all of the results obtained with our model [1,2] are qualitatively consistent with the experimental findings and previous theoretical predictions. Among the physical parameters that we studied, temperature, dopant fraction of Y2O3, and the relative permittivity of YSZ are found to have the most profound influence on the calculated ionic current density of the fuel cell. Our most recent results from the KMC model include frequency-response analysis, generated from simulations of alternating applied voltages, over a range of frequencies. From these simulations, we can capture the properties of geometric and double-layer capacitance and resistance within the SOFC.

[1] K. C. Lau, C. H. Turner, and B. I. Dunlap, "Kinetic Monte Carlo simulation of the Yttria Stabilized Zirconia (YSZ) fuel cell cathode," Solid State Ionics 179, 1912-1920 (2008).

[2] K. C. Lau, C. H. Turner, and B. I. Dunlap, "Kinetic Monte Carlo simulation of O2- incorporation in the Yttria Stabilized Zirconia (YSZ) fuel cell," Chemical Physics Letters 471, 326-330 (2009).