Schematic microstructure of the [LiNH2+2LiH]/[Li2NH+LiH]/Li3N system. Panels (a) and (b) indicate Hydrogenation and dehydrogenation, respectively. The nitrogen-containing phases (blue = LiNH2, pink = Li2NH, gray = Li3N) form a contiguous core–shell structure with phase boundaries propagating along the directions indicated by small arrows. a) During hydrogenation, LiH (green) is assumed to evolve from (i) a dispersed molecular, cluster, or solid-solution state to (ii) nucleated crystallites that eventually (iii) coarsen. b) Dehydrogenation follows the reverse pathway. Smaller particles favor the dispersed LiH state (i) due to interfacial penalties associated with nucleation. Credit: Wood et al. Click to enlarge.

Complex metal hydrides are a promising class of hydrogen storage materials, but their viability is usually limited by slow hydrogen uptake and release. Nanoconfinement—infiltrating the metal hydride within a matrix of another material such as carbon—can, in certain instances, help make this process faster by shortening diffusion pathways for hydrogen or by changing the thermodynamic stability of the material.

However, the researchers showed that nanoconfinement can have another, potentially more important consequence. They found that the presence of internal “nano-interfaces” within nanoconfined hydrides can alter which phases appear when the material is cycled.

The key is to get rid of the undesirable intermediate phases, which slow down the material’s performance as they are formed or consumed. If you can do that, then the storage capacity kinetics dramatically improve and the thermodynamic requirements to achieve full recharge become far more reasonable. In this material, the nano-interfaces do just that, as long as the nanoconfined particles are small enough. It’s really a new paradigm for hydrogen storage, since it means that the reactions can be changed by engineering internal microstructures.

—Brandon Wood, an LLNL materials scientist and lead author

The Livermore researchers used a thermodynamic modeling method that goes beyond conventional descriptions to consider the contributions from the evolving solid phase boundaries as the material is hydrogenated and dehydrogenated. They showed that accounting for these contributions eliminates intermediates in nanoconfined lithium nitride, which was confirmed spectroscopically.

Beyond demonstrating [LiNH₂ + 2LiH]/Li₃N@npC as a reversible, high-performing hydrogen-storage material, our work establishes that proper consideration of solid–solid nanointerfaces and particle microstructure is necessary for understanding hydrogen-induced phase transitions in complex metal hydrides. This highlights the analogy between hydrogen storage reactions and solid-state reactions in battery electrode materials, where internal interfaces have similarly been identified as important factors for altering phase transformation path- ways.

Significantly, this introduces the possibility of tuning solid-state hydrogen-storage materials by tailoring morphology and internal microstructure, representing a new paradigm for engineering materials that could meet established performance targets. Further gravimetric capacity improvements may also be achieved using emerging encapsulants with higher theoretical surface areas and loading ratios. Because nanointerface effects are amplified by the coexistence of multiple phases upon cycling, particularly prominent changes may be expected for nanoparticles with multicomponent mixtures.

—Wood et al.

The idea came from Mahidol University graduate student Natchapol “Golf” Poonyayant, who approached Sandia with the idea of using nanoconfinement to enhance hydrogen storage reactions in nitrogen-containing compounds. Working with the Sandia researchers, Poonyayant, his adviser, Pasit Pakawatpanurut, and fellow Mahidol student Natee “Game” Angboonpong found that liquid ammonia could be used as a gentle and efficient solvent for introducing metals and nitrogen into the pockets of carbon nanoparticles, producing nano-confined lithium nitride particles.

The new material that emerged from Poonyayant’s idea showed some unusual and unexpected properties. First, the amount of lithium nitride in the carbon nanoparticle host was quite high for a nano-confined system, about 40%.

Second, the nano-confined lithium nitride absorbed and released hydrogen more rapidly than the bulk material. Furthermore, once the lithium nitride had been hydrogenated, it also released hydrogen in only one step and much faster than the bulk system that took two steps.

To better understand the mechanism responsible for this improvement, the Sandia scientists reached out to computational scientist Brandon Wood of LLNL, a leading expert in the theory of solid-state reactions. Wood and his LLNL colleagues Tae Wook Heo, Jonathan Lee and Keith Ray discovered that the reason for the unusual behavior was the energy associated with two material interfaces.

Since the lithium nitride nanoparticles are only 3 nanometers wide, even the smallest energetically unfavorable process is avoided in the hydrogen storage properties. For lithium nitride nanoparticles undergoing hydrogenation reactions, the avoidance of unfavorable intermediates—extra steps in the chemical process—increases efficiency.

Taking the path of least resistance, the material undergoes a single-step path to full hydrogenation. Similarly, once hydrogenated, the nanoparticles release hydrogen by the lowest energy pathway available, which in this case is direct hydrogen release back to lithium nitride.

According to the Sandia and LLNL researchers, the next step is to further understand how the dehydrogenated and hydrogenated phases of lithium nitride change at the nanoscale. This is a stiff challenge to the team, as it requires imaging different chemical phases within a particle that is only several nanometers wide.

Hydrogenation forms a mixture of lithium amide and hydride (light blue) as an outer shell around a lithium nitride particle (dark blue) nano-confined in carbon. Nanoconfinement suppresses all other intermediate phases to prevent interface formation, which has the effect of dramatically improving the hydrogen storage performance. Source: LLNL. Click to enlarge.

The team will draw on the capabilities within the DOE’s Hydrogen Storage Materials Advanced Research Consortium (HyMARC), led by Sandia and comprised additionally of scientists from LLNL and Lawrence Berkeley National Laboratory. The team plans to use spatially resolved synchrotron radiation from LBNL’s Advanced Light Source to probe interface chemistry and structure.

In addition, since the nanoporous carbon host is “dead weight” from a hydrogen storage perspective, the team is examining ways to “lighten the load” and find carbon materials with more nano-pockets for a given carbon mass.

The work was funded by the Department of Energy’s (DOE) Fuel Cell Technologies Office and the Boeing Co.

Resources

• Wood, B. C., Stavila, V., Poonyayant, N., Heo, T. W., Ray, K. G., Klebanoff, L. E., Udovic, T. J., Lee, J. R. I., Angboonpong, N., Sugar, J. D. and Pakawatpanurut, P. (2017) “Nanointerface-Driven Reversible Hydrogen Storage in the Nanoconfined Li–N–H System” Adv. Mater. Interfaces doi: 10.1002/admi.201600803