TORQUE, a Software Package to Predict Water Orientations in Ionic Crystals.

It is well known that H2O can affect phase stability and materials performance. For example, water content in air can affect efficiencies of organic-inorganic perovskite solar cell. Thus, the experimental identification of water orientations is highly desirable but remains challenging due to the small scattering cross section of hydrogen. Similarly, first-principles computations depend strongly on unit cell size and available computing resources. We have developed a Linux based software package based on rotational equilibrium and point charge electrostatics for predicting the orientation of crystallization water molecules in ionic crystals. This method is at least \textasciitilde 300 times faster than first-principles density-functional-theory (DFT) computations and provides optimized orientations that are consistent with experiment and theory. Interestingly, it provides a new H2O orientation in Kernite crystal that has not been reported previously. Our DFT computations show that the two conformations are energetically in-equivalent. Thus, the torque method provides a new, simple, robust, and fast method to complete initial structures for ab-initio computations of ionic materials that contain crystallization water and provide initial water orientations for experimental structure refinements.