Fast prediction of crystal morphology Crystals are the building blocks of materials in sectors such as pharmaceuticals, electronics, photovoltaics and catalysis. Scientists developed a model to quickly predict morphology for controlled engineering and optimised design. Industrial Technologies © Thinkstock Controlling crystal morphology is critical to design for function. The appearance of undesirable morphologies can degrade product quality and handling characteristics, even causing blockage of filters and tubes. With increasing knowledge of quantum mechanical behaviours, there is much to be gained from combining classical crystal growth theories with modern quantum mechanical simulations. The EU-funded project 'Crystal surface simulations' (CRYSURFSIM) was launched to develop a unifying model for modern crystal morphology engineering. The amazingly varied shapes of crystals, even those of the same compound, arise from the different growing conditions to which they are subjected. Surface free energy is an important driving parameter. Surface free energy, also called surface tension for liquids, is the difference between the energies of atoms at the surface versus on the inside of the crystal. The bond–valence deficiency (BVD) model can be used to predict surface free energies and crystal morphology. CRYSURFSIM exploited the BVD model to gap the classical and quantum mechanical descriptions of crystal growth. Scientists successfully applied it to describe the surface free energy of common metal crystal structures. Results agreed with those of quantum mechanical lattice simulations and the crystal structures it predicted resembled those found in nature. The most important advantage of the BVD model relative to lattice simulations was its speed. Application to minerals (specifically, with structure types AX and AX2) was more challenging. While the BVD model handles both neutral and charged surfaces, lattice simulations require a 'correction' to convert charged to charge-neutral surfaces, proving a challenge for the comparison of results. However, when halides, sulphides and oxides were considered, the BVD model once again showed a clear agreement with lattice simulations and with morphologies seen in nature. The BVD model has proven to be a fast tool to accurately predict the morphology of crystals with minimal computational load, forming an effective bridge between macroscopic and quantum mechanical crystal formation models. It will support crystal engineers in rapid design of novel morphologies for specific functions in sectors from biomedicine and pharmaceuticals to thin-film semiconductors and photovoltaics. This is turn will have important impact on the competitiveness of the EU economy. Keywords Crystal morphology, quantum mechanical, crystal surface, surface free energy, bond–valence deficiency
