Final Report Summary - DIAMOND (Discovery and Insight with Advanced Models Of Nanoscale Dimensions)
This project advanced the state of atomistic simulation at the nanoscale with new methods that were either more efficient or more accurate than alternatives. Reducing the computational cost for systems containing many atoms is key, either by exploiting novel computer architectures, or by inventing new theoretical methods and algorithms that solve the problem more effectively. With these methods, simulations have been performed that give important insight in common systems. For example, for the first time, a non-empirical method could predict correctly that ice floats on water, an anomalous material property that results from a delicate balance of weak interactions between water molecules. Similarly, efficient methods allow us to build precise models containing millions of atoms and compute their properties from the fundamental equations, we computed (a very short part) of the dynamics of a viral particle, composed of a million atoms, without resort to experimental data. The focus of much of the work was on understanding an important oxide (TiO2), which is used in a wide range of products, ranging from toothpaste and sunscreen to solar cells and chemical catalysts. Charge transport in this oxide and on its surface was studied in detail, to understand solar cell materials, and hydrogen spillover in heterogeneous catalysis. Our results have been published in peer-reviewed journals, and made available for free and as open source in the form of computer simulation package (see http://www.cp2k.org/).