Stable nanoporous alkali halide polymorphs: a first principles bottom-up study

The stability of nanoclusters and nanocluster-based polymorphs of a large range of alkali halides were investigated using state-of-the-art plane wave density functional theory (DFT) calculations. Specifically, the most energetically stable cluster isomers of (MX)(12) (M = Li, Na, K, Rb, Cs, X = F, Cl, Br, I) were considered (i.e. the slab and cage) with respect to two bulk polymorphs: rock-salt (rs-MX) and a nanoporous analogue of the zeolite sodalite (SOD-MX). In both cases, these bulk materials can be regarded as being assembled from their respective cluster building block (slab -> rs, cage -> SOD). For all alkali halides the dense rs-MX phase was found to be more stable than the low-density nanoporous SOD-MX phase. For the (MX) 12 clusters, the dense slab cluster isomers were also generally found to be the most stable cluster type except, however, for the LiX series where the cage isomer was energetically preferred. The energy difference between the rs-MX and the SOD-MX bulk polymorphs (per MX unit) was found to follow the same trend as that between the respective (MX) 12 slab and cage clusters. Correspondingly, the cage-based SOD-LiX phases were all found to be only marginally metastable with respect to the rs-LiX forms (Delta ESOD-rs <= 0.05 eV per LiX). From DFT calculations on the low enthalpy landscape of LiF polymorphs, the energy versus volume equations of state of rs-LiF and SOD-LiF were compared with those of a number of other LiF polymorphs showing SOD-LiF to be stable with respect to compression and expansion and very competitive energetically with a number of denser phases. Classical molecular dynamics calculations were also performed to confirm the thermal stability of the SOD-LiF phase, further strengthening our prediction as to the viability of these novel low density nanoporous ionic materials.