Research and Development of Advanced Hydrogen Storage Materials: Metal Hydrides

Recipient Sandia National Laboratories/SNL (PI: Mark Allendorf)

Abstract Extensive thermodynamic and kinetic investigations conducted during Phase 1 of the HyMARC program demonstrated that metal hydrides (MH) remain a viable class of solid-state storage materials. However, as a result of the new synthetic methods, advanced characterization tools, and validated models we developed, several key technical issues were identified, including a critical need for MH thermodynamic data that accounts for both enthalpic and entropic effects; expanded assessment and validation of novel nanoscaling strategies for thermodynamic tuning; and development of integrated models to predict MH behavior across a range of length scales. Consequently, our Phase 2 goals are to: 1) Predict and validate comprehensive phase diagrams for complex metal hydride systems to optimize reaction conditions and pathways that could lead to >10 wt% reversible hydrogen capacity with a heat of desorption ≤27 kJ/mol H2; 2) Develop an operational capability for assessing the effects of surfaces and interfaces on metal hydride thermodynamics and kinetics, including “non-ideal factors” such as surface oxides; 3) Identify and evaluate strategies for activating B-H and B-B bonds via incorporation of chemical additives and defects; 4) Innovate synthetic strategies to nanoscaling MHs that will allow control over material properties (e.g. particle size, surface chemistry, confinement stress) governing H2 release and absorption while minimizing the effect of “dead mass and volume” on capacity; 5) Develop new foundational understanding of the effects of reaction- and diffusion-limited kinetic processes on MH microstructure evolution, and use this to manipulate kinetic pathways for improved uptake/release rate; and 6) Establish a machine learning capability focused on materials discovery for metal hydrides. We will continue to employ the synergistic approach applied successfully in Phase 1, which coordinates computational modeling (LLNL, LBNL, and SNL), novel synthetic approaches (SNL, LBNL), and cutting-edge characterization tools (LBNL, SNL, NIST) to develop in-depth foundational understanding of processes controlling MH behavior. Furthermore, we will expand this to include additional diagnostic capabilities now available to us, including solid-state NMR (PNNL), hard X-ray probes (SLAC), and high-pressure vibrational spectroscopy (LBNL).