From plin()at()chemvx.tamu.edu Mon Jul 28 16:31:22 1997 Received: from chemvx.chem.tamu.edu for plin $#at#$ chemvx.tamu.edu by www.ccl.net (8.8.3/950822.1) id QAA15427; Mon, 28 Jul 1997 16:11:04 -0400 (EDT) Received: from warbird.chem.tamu.edu ([165.91.176.59]) by chemvx.chem.tamu.edu (Netscape Mail Server v2.02) with SMTP id AAA159 for ; Mon, 28 Jul 1997 15:10:33 -0500 Sender: plin #*at*# chemvx.tamu.edu (Lin Ping) Message-ID: <33DCFCC0.41C6 "-at-" chemvx.tamu.edu> Date: Mon, 28 Jul 1997 15:10:40 -0500 From: Lin Ping Organization: Department of Chemistry X-Mailer: Mozilla 3.01Gold (X11; I; IRIX64 6.1 IP26) MIME-Version: 1.0 To: CHEMISTRY <-at-> www.ccl.net Subject: CCL:summary:basis set of DFT calculation Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear CCLers, two days ago, I post a queastion asking about the basis set for DFT calculation. >Original question > I am trying to run DFT calculation resently. Since the ordinary basis > sets are optimized for ab initio calculation, I am wondering if there > are basis sets spectially designed for DFT calculation, which will > surely improve the result. > Any hints or references would be greatly appreciated. > Replies can send to me directly, I will summarize to the list. Thanks to those who replied to my question: Bill Davis John Waite Dale Andrew Braden WU Nelson H Morgon *********************************************************************** Dr. William M. Davis As part of my Ph.D. thesis, I reoptimized the Stevens-Basch-Krauss ECP basis sets for the main group elements boron to iodine. The results indicated that, while it is more "theoretically sound" to use a basis set optimized for a particular method (DFT, HF etc.), the results really do not change all that significantly. It is much more important in DFT to chose a good combination of exchange and correlation functionals to improve your predictions. ************************************************************************ Dr. John Waite Dear CCL, I want to thank Dr. George Fitzgerald and Oxford Molecular Group (http://www.oxmol.com/) for a very needed contribution to our community, namely, the DGauss basis sets. These contracted gaussian sets were optimized for DFT calculations and include elements up to Xe. They also include the auxilliary fitting uncontracted gaussian sets for these programs which use them. The latest version of the basis sets for DGauss version 4.0 is available from Oxford Molecular ftp servers: in US: ftp://ftp.oxmol.com/pub/UniChem/UniChem_basis in UK ftp://ftp.oxmol.co.uk/pub/UniChem/UniChem_basis I was also given permission to mirror them in CCL archives: http://www.ccl.net/cca/data/basis-sets/DGauss/dgauss-basis.html or you can get there via ftp://www.ccl.net/pub/chemistry/data/basis-sets/DGauss/ There is also an older version (for DGauss 3.0 and older) in the CCL archives (which contains XC auxilliary fitting sets) in the same directory. The perl scripts are provided to convert these basis sets to other formats: For version 3.0: http://www.ccl.net/cca/software/PERL/basis/DG3_to_deMon/dg2dm.html and for version 4.0: http://www.ccl.net/cca/software/PERL/basis/DG4_to_deMon/dgn2dm.html Again, I put a plug for CCL. Please contribute chemistry related material (software, data, overviews, course materials, etc., etc.) to CCL -- I can mirror your files on our Web server, and also have files on the ftp site. Please contact me for more information (and do not be impatient if you need to wait for a few days for an answer -- but I am catching up usually {;-)}. We are still running old site: http://www.ccl.net/chemistry.html but at some point (when the damned things start to work as we want them!!!) we will open our new site: http://www.ccl.net/ Thank you for you attention... Jan Labanowski Ohio Supercomputer Center jkl[ AT ]ccl.net ******************************************************************* Dale Braden Have a look at: http://www.emsl.pnl.gov:2080/forms/basisform.html and look through the list for the DFT-optimized basis sets by Godbout, et al. The site gives references for each basis set. ******************************************************************* JH Wu The DZVP (double-zeta split-valence plus polarisation function basis) are optimised for DFT calculation. It is used by Dgauss program (unichem package). I have included a copy of DZVP basis set for C,H,N,O ... in the attachment. From the comment lines, you can search for related reference. I think you can get a whole copy of it from the homepage of unichem3.0 package. (if can't, give me a line) ******************************************************************* NELSON HENRIQUE MORGON Maybe, I could help you, and you can test my basis set. Tell me that basis set you would like, including kind of functional. I can model different basis set. ******************************************************************* LIN Ping -- Department of Chemistry Phone No. : 409-862-9225 Texas A&M University E-mail : plin()at()chemvx.tamu.edu College Station TX 77843-3255 plin -AatT- warbird.chem.tamu.edu URL: http://http.tamu.edu/~p0l1112/