Non-Covalent Interactions in Large Molecules and Extended Materials
CECAM-HQ-EPFL, Lausanne, Switzerland
Atomistic simulations are at the forefront for elucidating the intricate interactions that give rise to well-performing devices and drugs, and explain numerous phenomena in nature. With modern computing facilities and more sophisticated codes and theoretical methods being developed, wavefunction based methods for studying non-covalent interactions have expanded considerably. Methods with explicit electron correlation are now being applied to systems with hundreds of atoms [1-3]. This has been considered impracticable for the past decades for the sub-chemical accuracy that we desire from high-level methods. With this progress, we can finally tackle systems that are more relevant on the experimental scale. New approaches have been introduced which address the scaling of expensive "gold-standard" methods based on coupled cluster theory, reducing it from N^7 to near-linearly scaling with N, N being the measure of the system size [4,5]. In quantum Monte Carlo methods, improvements in the algorithm and underlying theory have boosted the efficiency of the method by orders of magnitude [6,7]. These and other recent advancements enable the application of benchmark-accuracy methods to physically more challenging materials [7,8]. As such, we can build our understanding of non-covalent interactions, which has been stuck in the regime of small molecular systems to date. Indeed, most work-horse methods are developed and validated on data sets of small molecular dimers, inevitably losing the non-additive and anisotropic interactions that only manifest in larger systems . High-level wavefunction methods can allow us to unravel non-covalent interactions for different kinds of materials and in doing so, we can also improve work-horse computational methods that are needed for addressing macroscopic properties. However, it is imperative to understand how recent approximations in high-level methods affect the prediction of non-covalent interactions. The focus of this workshop is to bring together leading researchers, many of whom are junior researchers themselves, to consolidate the state of the field and understand the current challenges and directions that need to be taken in order to realize the goal of accurately predicting non-covalent interactions in large and complex systems.
J. Gerit Brandenburg ( Interdisciplinary Center for Scientific Computing, Heidelberg University, 69120 Heidelberg ) - Organiser
Peter Nagy ( Budapest University of Technology and Economics ) - Organiser
Alexandre Tkachenko ( University of Luxembourg ) - Organiser
Yasmine Al-Hamdani ( University of Zurich ) - Organiser