BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Talks.cam//talks.cam.ac.uk// X-WR-CALNAME:Talks.cam BEGIN:VEVENT SUMMARY:What’s the point of linear-scaling electronic structure methods? - Prof. Peter Haynes\, Imperial College London DTSTART:20140213T141500Z DTEND:20140213T151500Z UID:TALK49847@talks.cam.ac.uk CONTACT:Dr G Moller DESCRIPTION:It is over 20 years since Yang proposed the first "divide-and- conquer" linear-scaling method [1]. Since then there has been a lot of fus s about the development of O(N) methods but relatively little to show for it in terms of practical applications. In this talk I will suggest why thi s is the case and how it may be addressed. \n\nFirst\, I will briefly outl ine the scheme implemented in the ONETEP code [2]\, focussing on the in si tu optimisation of local orbitals that enables plane-wave accuracy to be a chieved. I will suggest that the resulting "non-orthogonal generalised Wan nier functions" may be used in a manner similar to maximally localised Wan nier functions e.g. to interpolate band structure. \n\nSecond\, I will sug gest three strategies for exploiting the capability of linear-scaling meth ods to perform large-scale electronic structure calculations\, each illust rated with an example. 1. low-dimensional systems where the configuration space to be explored is relatively simple\, with reference to simulations of entire polar semiconductor nanorods that are amenable to O(N) methods. 2. the role of linear-scaling methods in a multiscale approach\, such as t he fitting of classical force fields to explore configuration space by mol ecular dynamics\, as shown by the prediction of amyloid fibril structure u sing computational NMR spectroscopy. 3. theoretical spectroscopy to establ ish a direct link between simulation and experiment involving recent devel opment of local orbital methods for performing time-dependent density-func tional theory calculations to obtain optical absorption spectra.\n\n[1] Ph ys. Rev. Lett. 66\, 1438 (1991)
\n[2] www.onetep.org LOCATION:TCM Seminar Room\, Cavendish Laboratory END:VEVENT END:VCALENDAR