Tiger Team: tt-vasp-wannier90
Model development for nanoconfined water in quasi 1D crystals
This project will deal with the investigation of quantum electric dipoles realized by isolated water molecules confined in nanocages of crystals. Specifically, we investigate quantum electric dipoles realized by H2O molecules confined to nanocages of crystals. Since the isolated water molecules do not form hydrogen bonds, they constitute an almost ideal model of a quantum electric dipole lattice. It has been shown that, at low temperatures, tunneling within the nanocage smears out the positions of the protons into a pair of corrugated rings. Here, we aim to study the energy levels of nano-confined water molecules in several dielectric frameworks. The investigation will begin with the prototype beryl, in which the length of dipolar chains in the c-axis channels can be modified by varying the water filling. Beryl contains channels of 5.1 Å in diameter with ≈2.8 Å constriction which can be filled with a single water molecule. By applying pressure, the dipolar coupling can be increased. We will then move beyond beryl with its particular triangular ab-lattice and investigate other nanoporous crystals and altered couplings between the confined H2O molecules. This will modify the strength of the dipolar interaction and cause anisotropy within the ab-plane.
In order to model these systems, quantum-mechanical simulations within the Den- sity Functional Theory (DFT) approach, as well as classical Molecular Dynamics (MD) will be utilized. The simulations will be supplemented by low-frequency optical and dielectric investigations at low temperatures, as well as mean field theoretical studies (from collaborative groups). The overall main objective is to develop a deep understanding of how quantum tunneling and fluctuations affect the properties and behavior of the confined electric dipoles. We also aim to address the dipolar interaction between the H2O molecules that eventually leads to a long-range order and possibly the occurrence of an incipient ferroelectric state of water in beryl. From a numerical point of view, we begin with DFT and the calculation of the ground and excited states, as well as the tunneling potential of the water within the crystal cage. We aim to assess the single particle excitations and collective dynamics of confined H2O for different filling factors in the crystal and compare/support the experimental observations. These will be followed by the study of the influence of doping and impurity atoms, and the coupling between the water molecules along and perpendicular to the crystal channels, as well as the exploration of the dipole dynamics through the influence of pressure, temperature, and applied electric fields on the ordering of the water molecules and the ferroelectric state in the crystal. In order to achieve this, MD simulations will be used.
In order to bridge the scales from DFT to MD and stretch our simulations for obtaining also thermodynamic properties, we need to assess the efficiency of the existing classical potential for beryl. As this, has not been used with any of the existing water models, we need to adjust the potential to a water model or vice versa. Accordingly, we need to extract the partial charges from the DFT simulations in order to bridge the scale to MD and match the models. The partial charges will be obtained through the calculation of localized Wannier functions. Having developed the water model for the classical simulations, we can then enrich our research by including the influence of temperature, pressure, and electric fields on the thermodynamic properties and incipient ferroelectricity of nano-confined water in beryl. In the end, unraveling the interplay of quantum tunneling, fluctuation, and frustration among the coupled electric dipoles could finally provide the possibility to realize a quantum electric dipole liquid and glasses.
One problem when integrating Wanner90  into VASP is the fact that only a rather old implementation of Wannier90 is compatible with the current version of VASP. Furthermore modifications introduced by other patches result in even more complexity when integrating Wannier90 into one VASP binary. The combination of all patches including VTST , VASP sol , beef , and an atom-selective variant of Grimme's DFT-D3  has been implemented in VASP. Due to the fact that some of the patches do not play well together, several binaries with different combinations of patches are available. In addition one unified binary has been created by carefully removing conflicts and ambiguities from different patches. In a final step highly optimized binaries have been created for bwForCluster JUSTUS.
 A. A. Mostofi, J. R. Yates, G. Pizzi, Y. S. Lee, I. Souza, D. Vanderbilt, N. Marzari, Comput. Phys. Commun. 185, 2309 (2014).
 G. Henkelman and H. Jonsson, J. Chem. Phys. 113, 9978 (2000).
 K. Mathew, R. Sundararaman, K. Letchworth-Weaver, T. A. Arias, and R. G. Hennig, J. Chem. Phys. 140, 084106 (2014).
 J. Wellendorff, K. T. Lundgaard, A. Møgelhøj, V. Petzold, D. D. Landis, J. K. Nørskov, T. Bligaard, and K. W. Jacobsen, Phys. Rev. B 85, 235149 (2012).
 S. Grimme, J. Antony, S. Ehrlich, and H. Krieg, J. Chem. Phys. 132, 154104 (2010).
Members of the Tiger-Team:
Institut für computergestützte Physik, Universität Stuttgart;
HPC-Kompetenzzentrum für computergestützte Chemie und Quantenwissenschaften, Universität Ulm