From owner-nwchem-users@emsl.pnl.gov Tue Dec 7 17:45:15 2004 Received: from odyssey.emsl.pnl.gov (localhost [127.0.0.1]) by odyssey.emsl.pnl.gov (8.12.10/8.12.10) with ESMTP id iB81jF0h006892 for ; Tue, 7 Dec 2004 17:45:15 -0800 (PST) Received: (from majordom@localhost) by odyssey.emsl.pnl.gov (8.12.10/8.12.10/Submit) id iB81jFX5006891 for nwchem-users-outgoing; Tue, 7 Dec 2004 17:45:15 -0800 (PST) Date: Tue, 07 Dec 2004 17:45:12 -0800 From: "Bylaska, Eric J" Subject: RE: Default XC functional in PSPW module To: Lev Gelb , nwchem-users@emsl.pnl.gov Message-id: MIME-version: 1.0 X-MIMEOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-type: text/plain; charset=US-ASCII Thread-Topic: Default XC functional in PSPW module Thread-Index: AcTcpPYzv3PrqgUtSyG0CpMAUNbBpgAIkyZA Content-class: urn:content-classes:message X-MS-Has-Attach: X-MS-TNEF-Correlator: X-OriginalArrivalTime: 08 Dec 2004 01:45:13.0201 (UTC) FILETIME=[8F2E8610:01C4DCC7] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by odyssey.emsl.pnl.gov id iB81jE0h006888 Sender: owner-nwchem-users@emsl.pnl.gov Precedence: bulk Lev, Also, please make sure that you delelte all *.psp and *.vpp files from your permanent_dir between lda and pbe96 runs. By default the program won't regenerate psp files. So If you run an lda calculation followed by a pbe96 calculation, you'll be using a lda generated psp instead of a pbe96 psp. Eric Bylaska -----Original Message----- From: owner-nwchem-developers@emsl.pnl.gov [mailto:owner-nwchem-developers@emsl.pnl.gov] On Behalf Of Lev Gelb Sent: Tuesday, December 07, 2004 1:31 PM To: nwchem-users@emsl.pnl.gov Subject: Default XC functional in PSPW module The appended job script (below) for a plane-wave calculation on a lithium dimer is behaving strangely with regards choice of XC functional (and default Li pseudopotential) With "XC VOSKO", it gives a total energy: -0.4402776830E With no XC specified anywhere, it gives: -0.4359656451E With XC PBE96, it gives -0.4402776830E Likewise, if I do the single-atom energies, I get: VOSKO/UNR: -0.2054801969E NO XC: -0.1992073370E PBE96: -0.2054801969E And thus the bond energies are: VOSKO: 0.0293173 NO XC: 0.03755097 PBE96: 0.02957483 According to the NWChem documentation, the "VOSKO" LDA parameterization is the default choice - but then why do I get different values for explicitly specifying VOSKO or not specifying anything? This was run on an SGI Altix, but I am seeing very similar behavior for on a linux/x86 machine as well. Any suggestions would be appreciated. Job-script follows. #---------------------------------------------- scratch_dir ./working permanent_dir ./working Start Lithium_test echo memory 750 mb charge 0 geometry units angstrom noautosym noautoz print system crystal lat_a 15.0000 lat_b 15.0000 lat_c 15.0000 alpha 90.0 beta 90.0 gamma 90.0 end Li 0.5000 0.5000 0.30 Li 0.5000 0.5000 0.483 end nwpw pseudopotentials Li library library1 end simulation_cell ngrid 34 34 34 end UNRESTRICTED XC PBE96 tolerances 1.0e-7 1.0e-8 ewald_ncut 3 end task pspw energy