From jaouad[ AT ]ugr.es Wed Mar 1 06:08:38 1995 Received: from ocelote.cica.es for jaouad - at - ugr.es by www.ccl.net (8.6.10/930601.1506) id FAA15433; Wed, 1 Mar 1995 05:17:15 -0500 Received: (from ean ( ( at ) ) localhost) by ocelote.cica.es (8.6.9/8.6.9) id LAA08939 for CHEMISTRY ( ( at ) ) ccl.net; Wed, 1 Mar 1995 11:09:49 +0100 X400-Received: by mta rica in /PRMD=/ADMD=/C=/; Relayed; Wed, 1 Mar 1995 11:09:38 UTC+0100 X400-Received: by /PRMD=iris/ADMD=mensatex/C=es/; Relayed; Wed, 1 Mar 1995 11:09:38 UTC+0100 X400-Received: by /PRMD=es/ADMD=/C=/; Relayed; Wed, 1 Mar 1995 11:10:08 UTC+0200 Date: Wed, 1 Mar 1995 11:10:08 UTC+0200 X400-Originator: jaouad <-at-> ugr.es X400-Recipients: non-disclosure:; X400-Content-Type: P2-1984 (2) X400-MTS-Identifier: [/PRMD=es/ADMD=/C=/;950301111008] Content-Identifier: 42 Conversion: Prohibited From: Jaouad el bahraoui To: CHEMISTRY (+ at +) ccl.net (confirm) Message-ID: <42*/S=jaouad/O=ugr/PRMD=iris/ADMD=mensatex/C=es/ ( ( at ) ) MHS> Subject: SUMMARY DGAUS and G92/DFT Return-Receipt-To: Jaouad el bahraoui MIME-Version: 1.0 (Generated by Ean X.400 to MIME gateway) Dear CCLers First, thanks to those who responded to my query. the original message as well as the responses are at the and of this mail. ******************************************************************************** EL BAHRAOUI JAOUAD Ph.D.STUDENT Instituto de Biotecnologia Grupo de Modelizacion y Diseno Molecular Universidad de GRANADA,Facultad de Ciencias, GRANADA 18071,Spain TEL: 34-58-243186. FAX: 34-58-243186 EMAIL: jaouad : at : ugr.es ******************************************************************************** the original question and the answers: I'm doing some calculations with G92/DFT and I'want to performe it with DGAUS basis set, knowing that gaussian dont accept auxiliary basis set Do you know if it possible. thanks in advance. jaouad ******************************************************************************************************** From: Clifford LeMaster: CLEMAST "at@at" quartz.es I would like to hear the replies. I am also interested in DFT *************************************************************************************** From: steinke (- at -) ZIB-Berlin.DE I can't suggest this in generally since the DGauss basis set is specially developed for the "DFT-LDA method" and it makes no sense for me (at least theoretically) to use such a basis set in a HF-based density approach like the implementation in G92/DFT. I'll expect some curious results using such a "mixed" approach. The auxiliary basis set is used for the representation of the one-electron density in a set of analytically based functions for a slightly better integral computation. Best regards, Thomas ***************************************************************************************************************************************************** From: m10!aefrisch ^at^ uunet.uu.net Yes, it is possible; Gaussian does allow you to read in a basis set for the calculation. Look at the discussion of the Gen keyword in the G92/DFT manual to see the format that is required. **************************************************************************************************************************************** From: dobbskd-0at0-esvax.dnet.dupont.com Yes, it is possible. I have been doing it for well over 6 months now. G92/DFT does not need auxiliary basis sets. DGauss basis sets work fine, but do not expect DGauss energies/structures to be exactly the same as those of G92/DFT since they each implement the DFT code differently. ****************************************************************************************************************************************** From: "Oscar N. Ventura" Dear Jaouad, I can not help you with that topic because I don't have the necessary experience. However, I have performed DFT calculations using G92/DFT and conventional ab initio basis sets. My experience with some hydrogen-bonded complexes, organic radical cations and some organometallics is that you can obtain accurate results in such a manner. If you are interested I can send you some numbers. Best regards, O. *********************************************************************************************************************** From: Nathaniel Malcolm hpf.ch.man.ac.uk> as far as i am aware Gaussian does allow the use of general basis sets with the keyword "gen" instead of a standard basis set (see the users manual) NOJ Malcolm *********************************************************************************************** From: I do not think G92 currently accepts auxillary functions for DFT calculations. However, there is the program DeFT of Alain St.-Amant at Ottawa that does. DGAUSS, DEMON, and DeFT all take auxillary functions....for good reason...St-Amant wrote DEMON and DeFT, and there are many similarities with DGAUSS. Regards, John ***************************************************************************************************** From: Dear Jaouad, The DGauss basis set can be used in G92. But you need to modify it into acceptable format to be read into G92. G92 doesn't use fitting for the charge density. (This is one of the reasons DGauss is so much faster). Therefore, the auxiliary basis set, which is used only for the density fitting, is not needed in G92 DFT calcultions. Please let us know if you have any further questions. Regards, Susan -------------------------------------------------------- Susan M. Gustafson, Ph.D. Phone: (612) 683-3662 Cray Research, Inc. e-mail: sgustaf { *at * } cray.com -------------------------------------------------------- ******************************************************************************************* From: Hola, Si es posible usar las bases de dgauss, yo hecho un par de calculos con G92/DFT usando las bases de dgauss, todo lo que tienes que hacer es ignorar las "auxiliary basis set". Ademas tienes que traducir el formato que usa dgauss con el que usa G92/DFT. Si tienes mas preguntas enviamelas directamente, yo trabajo con Gaussian en Cray. Tuve el placer de conocer tu Universidad hace mas de un anyo. Saludos.