An accurate treatment of electron correlation in real materials presents a formidable challenge to condensed matter physicists and quantum chemists. There is a real need for a calculational method which is more accurate than current density functional theory (DFT) methods, but which is still viable for applications to real materials. A powerful and accurate tool for attacking such problems is provided by the diffusion quantum Monte Carlo (DMC) technique. This incorporates both the one-electron physics of approaches such as Hartree-Fock and DFT, and a proper description of electron correlation. DMC can be used to calculate zero-temperature ground and excited state energies of materials using a realistic continuum description of their atomic properties.
Since the ground breaking DMC calculations of Ceperley and Alder on the electron gas there has been considerable progress in developing and applying DMC techniques to real materials. Examples of today's state-of-the-art applications of DMC are calculating excited state energies of molecules and solids, formation energies of defects in semiconductors, energies of semiconductor surfaces, binding energies of clusters, and trying to understand DFT and to improve on current approximations to the exchange-correlation energy.
On the technical side, important developments have been the introduction of non-local pseudopotentials, the use of variance minimisation techniques to optimise trial wave functions, the integration of quantum chemistry and condensed matter electronic structure software packages with DMC methods, and the very recent development of so-called `Order-N' DMC algorithms. The proposed meeting will concentrate on fermion problems, particularly electrons in molecules and solids, but it is hoped that those involved in using these methods for other problems such as vibrational modes will also find it of interest.
The DMC method is now arousing widespread interest to the wider electronic structure community. With modern parallel computers they are becoming tractable for all sorts of problems of fundamental and technological importance. Activity in DMC calculations has been hampered by the lack of reliable and general computer codes to perform the calculations, but a number of codes are now available. Technical innovations are now coming at a rapid rate and a workshop of this kind is needed to discuss and disseminate the new ideas. This meeting is intended to allow discussions and exchange of information between DMC experts and will allow less experienced users to obtain access to the developers of QMC codes and the expertise to use them. It is hoped that this meeting will help to build a stronger community of researchers using DMC methods.