FY24 Q4 quarterly report: LLNL's node activity within HydroGEN STCH

Tae Wook Heo, Brandon M. Wood, Sang-Wook Kang, Y. Song2024 · DOI

Abstract

This quarterly report from Lawrence Livermore National Laboratory (LLNL) details progress in modeling Solar Thermochemical Hydrogen (STCH) production. We computed the driving force for the water splitting reaction as a function of reaction progress and continued development of a numerical method for water splitting reactions of $H_2$ production on particle surfaces within 3D porous microstructures. By combining thermodynamic analysis with 3D mesoscale simulations, the research identifies why water-splitting reactions slow down over time and how material microstructure dictates hydrogen yield.

Introduction

Solar Thermochemical Hydrogen (STCH) production utilizes concentrated solar energy and metal oxides, such as perovskites, to split water into hydrogen and oxygen. Despite its potential, the process is often hindered by "sluggish" kinetics—a significant slowdown in the reaction rate as it proceeds. This challenge arises from a complex interplay between atomic-level thermodynamics and mesoscale physical structures. This report aims to bridge this gap by quantifying the thermodynamic driving forces and developing numerical models to simulate hydrogen production within complex, 3D porous materials.

Methods

The research team employed a multi-scale approach to optimize hydrogen production in perovskite oxides. The primary methods included:

  1. Thermodynamic Mapping: Experimental data on partial molar enthalpy and entropy for $SrTi_{0.5}Mn_{0.5}O_{3-\delta}$ (sourced from Qian et al., 2020) were converted into the Gibbs free energy ($\Delta G_{rxn}$) for the water-splitting reaction. This mapping allows for the quantification of the chemical "push" as the material's reduction state changes.
Thermodynamic mapping of partial molar enthalpy (a) and entropy (b) for the perovskite oxide $SrTi_{0.5}Mn_{0.5}O_{3-\delta}$.
  1. Mesoscale Simulation: A Langmuir adsorption model was used to simulate surface reaction kinetics at the solid-pore interface, where $H_2O$ is converted to $H_2$.
  2. Digital Microstructure Synthesis: To accurately represent real-world materials, the team utilized the Dream3D program to generate digital 3D microstructures matching experimental parameters such as porosity, particle size, and surface area. This approach replaces less specific synthetic databases.

Key Insights and Results

A critical finding of this study is the identification of a Thermodynamic "Cliff." The analysis revealed that the driving force for the water-splitting reaction is initially high (approximately 300 kJ/mol O) but drops significantly to around 100 kJ/mol O as the reaction nears completion (i.e., as $\delta$ decreases). This quantitative decline provides a clear thermodynamic explanation for the observed kinetic sluggishness in the later stages of the process.

Gibbs free energy ($\Delta G_{rxn}$) as a function of reaction progress ($\delta$), showing the decrease in driving force for water splitting as the material re-oxidizes.

Furthermore, simulations confirmed that $H_2$ production is strictly localized at solid surfaces. This underscores the importance of physical architecture; porosity and the surface-to-volume ratio are primary bottlenecks for production rates. The team successfully validated their numerical framework, demonstrating its ability to track $H_2$ concentration evolution within a 3D volume.

Conclusion and Future Work

While the current numerical framework is structurally sound, it currently utilizes some arbitrary parameters for validation. Future efforts will focus on:

  • Parameter Integration: Incorporating specific DFT-calculated energy barriers and experimental rate constants to enhance the model's predictive capabilities.
  • Fluid Dynamics: Introducing $H_2O$ and $H_2$ flow rates into the porous regions to account for gas flow limitations observed in experiments.
  • Time-Dependent Analysis: Computing $H_2$ production over time across various 3D digital microstructures to identify the optimal geometry for maximum hydrogen yield.

Summary updated

Abstract

In this quarterly report, we Computed the driving force for the water splitting reaction as a function of reaction progress and continued Development of the numerical method for water splitting reactions of H<sub>2</sub> production on the particle surface within 3D porous microstructures.
BibTeX
@misc{heo2024fy,
  author = {Tae Wook Heo and Brandon M. Wood and Sang-Wook Kang and Y. Song},
  title = {{FY24 Q4 quarterly report: LLNL's node activity within HydroGEN STCH}},
  year = {2024},
  doi = {10.2172/2476128},
}
Figure