From chemistry-request@ccl.net Fri Apr 9 11:48:32 2004 Received: from pat.cryst.bbk.ac.uk (pat-ext.cryst.bbk.ac.uk [193.61.32.113]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id i39GmWjR017235 for ; Fri, 9 Apr 2004 11:48:32 -0500 Received: from mail.cryst.bbk.ac.uk (chem239.chem.ucl.ac.uk [128.40.76.239]) by pat.cryst.bbk.ac.uk (8.12.8/8.12.8) with ESMTP id i39Gmcog029229 for ; Fri, 9 Apr 2004 17:48:42 +0100 Message-ID: <4076D2D8.5000805~at~mail.cryst.bbk.ac.uk> Date: Fri, 09 Apr 2004 17:44:08 +0100 From: Ricardo Grau-Crespo User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) X-Accept-Language: en-us, en MIME-Version: 1.0 To: chemistry~at~ccl.net Subject: high-temp energy surfaces in magnetic systems Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-cryst-bbk-ac-uk-MailScanner-Information: Please contact CCSG more information (http://www.cryst.bbk.ac.uk/CCSG/) X-cryst-bbk-ac-uk-MailScanner: Found to be clean X-cryst-bbk-ac-uk-MailScanner-SpamCheck: not spam, SpamAssassin (score=-4.9, required 9, autolearn=not spam, BAYES_00 -4.90) Dear all, The energy hypersurface (energy as a function of the position of the nuclei) is normally considered in molecular dynamics as a function that is independent of temperature. This is obviously just an approximation. In particular, for atoms with non-zero magnetic moments (say, for example, a ferromagnetic material) it is clear that the energy surface is not the same at low temperatures (in the ferromagnetic state) as at high temperatures (paramagnetic state). In these magnetic systems, if one wants to use 0 K ab initio calculations to obtain approximates 'high-temperature' energy surfaces (for example, in order to fit interatomic potentials to be used in high temperature MD), one normally has to choose between: 1. spin polarized calculations - which introduce the error associated to the ordering of the frozen magnetic moments (such order is not present at high temperatures), or 2. non spin polarized calculations - which introduce an error because of the wrong electronic configurations of the magnetic atoms. I would like to know which approximation is better, that is, which of these errors is less significant for the HT energy surface. Is there a general answer for this question? I would appreciate any comment or suggestion. Thank you very much, Ricardo. -- Ricardo Grau-Crespo PhD Student - University of London Christopher Ingold Laboratories, room 205 20 Gordon Street, London WC1H 0AJ United Kingdom Tel : +44(0) 20 7679 7465 Fax : +44(0) 20 7631 6803 Web : http://img.cryst.bbk.ac.uk/www/graucrespo/homepage.htm -------------------------------------------------------------