Computational Materials Scientist - Electrochemical Systems
Posted on Aug 14, 2019 by Leidos
An opportunity exists for a computational materials scientist to develop and utilize computational multiphysics/electrochemical models and data analysis tools to support NETL's solid oxide fuel cell (SOFC) research.
Research efforts focus primarily on:
(1) The development, refinement, and calibration of models that accurately represent relevant physical and chemical processes (including multiphase flow, electrochemistry, and mass/charge transport) within fuel cell electrodes, (2) the development of computational tools incorporating Bayesian analysis to calibrate the electrochemical models and quantify the uncertainty in the calibrated parameters based on supplied experimental and theoretical data, and (3) the utilization of the models and statistical analysis tools to infer the model parameters that change during long term fuel cell degradation. The models and tools developed and utilized by this researcher must be adapted to couple with multiple platforms (eg, FORTRAN, C++, COMSOL, Matlab) running on different client's workstations, as well as on NETL's Joule 2.0 supercomputer. This position will also provide support to integrate the developed models within other SOFC stack/system models and to communicate the lessons learned from the simulations with external industrial, academic, and national laboratory partners.
- Bachelor's degree and 4 years prior relevant experience; or Master's degree and 2 years prior relevant experience; or PhD and 0-2 years prior relevant experience.
- Modeling electrochemical systems such as fuel cells, including interpreting and coding electrochemical and chemical reactions, modelling fluid dynamics and mass/charge transport in porous materials, and scale bridging to pass simulation data effectively across multiple length scales. Experience is required with the calibration, validation, and utilization of electrochemical performance models using experimental data.
- Bayesian statistical analysis for calibration of models, including using Markov Chain Monte Carlo and Sequential Monte Carlo sampling methods.
- Numerical modelling including ordinary differential equations and partial differential equations. Knowledge and experience with stiff equations solvers using finite volume and finite element discretization methods.
- Submitting jobs and managing simulations and data within a supercomputing facility using Slurm and Grid Engine, including parallelization of tasks when possible.
- Collaborating with experimentalists to request experimental data most useful for gaining a better understanding of the underlying physics and chemistry determining overall cell performance.
- Matlab (including the Parallel, Optimization, and Statistics and Machine Learning Toolboxes and the Foreign Function Interface to COMSOL)
- C/C++ (including the Boost and GSL libraries, parallelization using OpenMP and MPI, and the Foreign Function Interface to FORTRAN). Familiarity preferred in C++, Python, LUA and Visual Basic.
- Python (including the Scipy, Joblib, and NumPy packages, the Matplotlib library, and the Foreign Function Interface to C)
- Experience using Git and SVN for version control of custom codes
- The preferred candidate should have a deep knowledge of conventional and state-of-the-art fuel cell materials and experience in analysing electrochemical performance data such as polarization curves and electrochemical impedance Spectra.
- The preferred candidate will have more than two years postdoctoral experience and have experience in modelling of solid oxide fuel cells in a collaborative work environment such as the National Energy Technology Laboratory.