Two methods are traditionally used to characterize gas adsorption properties in porous solids: volumetric and gravimetric. They have a number of limitations, but most importantly, they yield a macroscopic picture of interactions (properties), without access to a microscopic picture (mechanisms on an atomic level). Diffraction is commonly used as a complementary technique to explain these properties, giving insight into structure and thus revealing the underlying guest-host and guest-guest interactions. Various anomalies (deviations from a typical behaviour) detected by the macroscopic methods require an in situ diffraction experiment, aiming to identify the responsible phenomena like a guest rearrangement / repacking, framework deformation etc. Thus, a separate diffraction experiment is usually providing a microscopic picture for the properties found by other physico-chemical methods.

In this presentation we will show examples of using in situ powder diffraction to simultaneously access the structure and adsorption properties of a small pore crystalline solid. (Quasi)-equilibrium isotherms and isobars can be built directly from sequential Rietveld refinements, both on adsorption and desorption, thus addressing the hysteresis and kinetics of gas adsorption/desorption. Detailed picture of guest reorganization with an increasing uptake can be obtained. Note that the reorganization of the individual guest sites is not accessible to volumetric and gravimetric methods, as they give only total amounts of gas uptake.

Interestingly, the adsorption isobars and isotherms obtained directly from diffraction data can be fitted by known equations, such as a logistic function (isobars) or a Langmuir equation (isotherms). Thermodynamic properties, such as enthalpy and entropy of gas adsorption can be extracted from these curves. The limitations of this technique are very different from traditional methods, thus making it highly complementary.

Lastly, the adsorption kinetics can be followed by in situ powder diffraction at given P,T conditions versus time. The guest uptake extracted by a sequential Rietveld refinement can be fitted and analysed in terms of Arrhenius theory giving access to the activation energies for gas diffusion. Thanks to the microscopic picture these barriers can be tentatively attributed to various diffusion paths inside the solid.

This talk will be illustrated by examples of noble gas adsorption in a porous hydride, γ-Mg(BH₄)₂ [1], featuring 1D channels suitable to distinguish and likely separate some of these gases. Besides published results [2,3], a lot of unpublished data will be shown.

[1] Filinchuk, Y., Richter, B., Jensen, T. R., Dmitriev, V., Chernyshov, D. & Hagemann, H. (2011). Angew. Chem. Int. Ed. 50, 11162. [2] Dovgaliuk, I., Dyadkin, V., Vander Donckt, M., Filinchuk, Y. & Chernyshov, D. (2020). ACS Appl. Mater. Interfaces 12, 7710. [3] Dovgaliuk, I., Senkovska, I., Li, X., Dyadkin, V., Filinchuk, Y. & Chernyshov, D. (2021). Angew. Chem. Int. Ed. 60, 5250.

Complex early transition metal oxides have emerged as leading candidates for fast charging lithium-ion battery anode materials [1,2]. Framework crystal structures with frustrated topologies are good electrode candidates because they may intercalate large quantities of guest ions with minimal structural response. Starting from the empty perovskite (ReO₃) framework, shear planes and filled pentagonal columns are examples of motifs that decrease the structural degrees of freedom. As a consequence, many early transition metal oxide shear and bronze structures do not readily undergo the tilts and distortions that lead to phase transitions and/or the clamping of lithium diffusion pathways that occur in a purely corner-shared polyhedral network[1].

In this work, we explore the relationship between composition, crystal structure, and reduction pathway in a variety of recently synthesized mixed alkali, transition metal, and main group oxides (Fig. 1), moving beyond the archetypal Ti-Nb-O and W-Nb-O phase spaces. Solid-state NMR spectroscopy, X-ray absorption spectroscopy (XANES and EXAFS), synchrotron and neutron diffraction, and DFT are combined with electrochemical experiments to present a comprehensive picture of the charge storage mechanisms. Prospects for tunability and implications for charge rate and structural stability will be discussed.

[1] Griffith, K. J., Wiaderek, K., Cibin, G., Marbella, L. M. Grey, C. P. (2018). Nature 559, 556.

[2] Griffith, K. J., Harada, Y., Egusa, S., Ribas, R. M., Monteiro, R. S., Von Dreele, R. B., Cheetham, A. K., Cava, R. J., Grey, C. P., Goodenough, J. B. (2021). Chem. Mater. 33, 4.

This Interest in metal hydrides was initially driven by the potential to develop efficient and safe on-board hydrogen stores working close to ambient pressure and temperature. In search for hydrides with higher gravimetric storage capacity, the researchers concentrated on hydrides based on light atoms, among others on Li and Na salts containing hydroborate anions such as borohydride BH₄⁻ or closo-hydroborate B₁₂H₁₂²⁻ [1]. The hydrogen absorption-desorption cycling in complex hydrides still needs more chemical ideas due to relatively strong covalent bonding. Unexpectedly, the high mobility of alkali metal cations in some complex hydrides has opened the door for their application as battery materials, mainly as solid-state electrolytes (SSE).

Replacing the liquid electrolyte by SSE offers several advantages: i) a solid material is more thermally stable, thus enhancing the overall safety of the battery; ii) being less prone to the dendrite penetration, it enables the use of alkali metals as negative electrodes and iii) acting as physical layer between the two electrodes, it has a beneficial effect on the cell performance [2].

Among the different classes of SSE, the metal hydroborates have received particular interest, being soft, highly stable toward oxidation and exhibiting fast ion conductivity, enabled by an entropically-driven phase transition. Such transitions generally occur above room temperature (rt), and it is therefore necessary to frustrate the anionic lattice, for example by anion mixing to bring the superionic regime down to rt [3-6].

The hydrogen storage and mobility of the cations in light complex hydrides depends on the structural features, pathways available in the anion packing and on the anion thermal motion. While the latter requires important experimental and theoretical effort, the first two parameters can be easily quantified from crystal structures obtained by X-ray powder diffraction.

Examples of crystallography and crystal chemistry analyses of novel solid-state electrolytes as well as proof-of-concept Na-ion all-solid-state batteries will be shown.

[1] Paskevicius M. et al. Chem. Soc. Rev. 2017, 46, 1565

[2] Zeier W. & Janek J. Nat. Energy 2016, 1, 16141

[3] Tang W.-S. et al. ACS Energy Lett. 2016, 1, 659

[4] Duchêne L. et al. Energy Environ. Sci. 2017, 10, 2609

[5] Murgia F. et al. Electrochem. Comm. 2019, 106, 106534

[6] Brighi M. et al. Cell Reports Phys. Sci. 2020, 1, 100217

Diffraction experiments typically provide clear picture of a crystal structure and basis for understanding material properties. However, for materials with high static or dynamic disorder and/or weakly occupied atomic sites, e.g. ionic conductors, the diffraction data reflecting space- and time-averaged state may struggle to distinguish several alternative models yielding similar χ². In that case, atomistic modelling may help not only to identify the more energetically stable configuration but also provide insights into the mechanism of its formation. I will present several recent examples of studies of disordered oxide-ion and proton conductors, where ab initio static and geometry optimisation calculations and molecular dynamics simulations not only helped to validate neutron diffraction analysis but also revealed the mechanism driving the disorder.

Prussian blue analogues (PBAs) with formula A_xM[M’(CN)₆]_1-y.zH₂O, show considerable promise as highly sustainable electrodes in sodium ion batteries. PBAs are formed of metal that are octahedrally coordinated by cyanide groups which act as bridges between the metal centers. This corner linked framework creates a highly porous structure into which either cations such as Na⁺ or molecules such as H₂O can insert into. However, PBAs receive criticism on their thermal stability and moisture sensitivity, which can be detrimental to the electrochemical performance or compromise safety. However, existing pessimism towards the material is based on studies of traditional Prussian blue (Fe₄[Fe(CN)₆]₃), whereas the vacancy-free compounds such as iron hexacyanoferrate (Fe-HCF), Na_xFe[Fe(CN)₆]_1-y.zH₂O (x≈2, y≈0, z≈0), do not show any similarity in terms of structural transitions or performance in a battery. In this contribution, our efforts at understanding the thermal and moisture stability of vacancy-free Fe‑HCF are presented.

We have optimised a method of consistently producing Fe-HCF with <5% vacancies on the Fe(CN)₆ site. Consequently, the effect of sodium content and moisture on structure and stability has been independently quantified. In the absence of vacancies, the moisture sensitivity of the material is determined by the Na⁺ content, with a sodium-rich structure absorbing more water and binding with higher affinity. Interestingly, despite a higher moisture sensitivity, the Na⁺ rich system features higher thermal stability. The interplay between the host framework, sodium and water also appears to influence the phase transitions of the material. The sodium-free material does not undergo any phase transitions, remaining cubic (Fm-3m) from 4K to 300K, whereas the sodium rich (x>1.5) systems exhibit several phase transitions between R-3 and P2₁/n as a function of temperature and water content. These are driven by octahedral tilting (cf. perovskites) and given that such transitions are generally rare in PBAs, their presence within a single system provides a platform for investigating driving factors.

As described, the moisture sensitivity of PBAs is often understood as the tendency to absorb water into the bulk structure. However, water can negatively affect cation rich Fe-HCF via other mechanisms. We identified that contact with airborne moisture during storage can lead to a loss of capacity in Fe-HCF. The capacity fading mechanism proceeds via two steps, first by sodium from the bulk material reacting with moisture at the surface to form sodium hydroxide and partial oxidation of Fe²⁺ to Fe³⁺. The sodium hydroxide creates a basic environment at the surface of the PW particles, leading to decomposition to Na₄[Fe(CN)₆] and iron oxides. Although the first process leads to loss of capacity, which can be reversed, the second stage of degradation is irreversible. The combination of each process ultimately leads to a surface passivating layer which prevents further degradation.

Thus, the interaction of water with cation rich PBAs is complex and should not be overlooked. Gaining an understanding of the degradation mechanisms, including structural and chemical driving forces provides substantial insight into effective design strategies for increasing the performance.

The overarching goal of this work is to understand and overcome the performance limitations of industrially relevant battery materials using powder diffraction studies, both through ex situ studies of materials and operando studies of cycling battery cells. However, the normal modalities for the collection and analysis of powder diffraction data typically lack the sensitivity to resolve the structure of battery materials with sufficiently low uncertainty to effectively resolve structure-properties correlations. We have therefore been actively been developing new approaches to data collection and analysis that overcome these limitations, permitting us to obtain robust structure-properties correlations for industrially relevant cathode materials and for industrially relevant battery cell designs.

We have recently developed a novel perspective for systematically exploring occupancy defects which we have applied to the study of the important family of NMC battery cathode materials [1]. Using these f* diagrams, we have demonstrated sufficient sensitivity to site occupancies to resolve problems with the conventional atomic form factors used for X-ray diffraction – an error of about 3% in the case of oxygen. After correcting for these problems and robustly determining atomic displacement parameters, we have demonstrated the ability to unambiguously resolve the nature of key defects as well as to determine defect concentrations with an unprecedented sensitivity of ~0.1% (absolute), as judged by the agreement between independent refinements of synchrotron and neutron powder diffraction data. From the refined occupancies for a series of NMC compounds, it was possible to determine the energy associated with the formation of anti-site defects, and to conclusively demonstrate that the conventionally accepted mechanism for defect formation was incorrect [2]. Additionally, we have utilized rapid synchrotron powder diffraction methods to carry out multidimensional diffraction studies with fine resolution not just in time but in space as well. In this manner, it has been possible to resolve both vertical [3] and lateral [4] inhomogeneity in battery cells with a sensitivity to the local state of charge (SOC) of ~0.1%. The former has illuminated the performance limitations of exceptionally thick battery cathodes with very high energy densities, while the latter has allowed us to distinguish between different potential failure mechanisms.

[1] L. Yin, G. Mattei, Z. Li, J. Zheng, W. Zhao, F. Omenya, C. Fang, W. Li, J. Li, Q. Xie, J.-G. Zhang, M.S. Whittingham, Y.S. Meng, A. Manthiram and P. Khalifah (2018). Rev. Sci. Instrum. 89, 093002.

[2] L. Yin, Z. Li, G. Mattei, J. Zheng, W. Zhao, F. Omenya, C. Fang, W. Li, J. Li, Q. Xie, E. Erickson, J.-G. Zhang, M.S. Whittingham, Y.S. Meng, A. Manthiram, and P. Khalifah (2019). Chem. Mater., 32, 1002-1010.

[3] Z. Li, L. Yin, G. Mattei, M. Cosby, B.-S. Lee, Z. Wu, S.-M. Bak, K. Chapman, X.-Q. Yang, P. Liu, and P. Khalifah (2020). Chem. Mater., 32, 6358.

[4] G. Mattei, Z. Li, A. Corrao, C. Niu, Y. Zhang, B.-Y. Liaw, C. Dickerson, J. Xiao, E. Dufek, and P. Khalifah (2021). Chem. Mater., 33, 2378.