From owner-nwchem-users@emsl.pnl.gov Thu Mar 6 11:00:05 2008 Received: from odyssey.emsl.pnl.gov (localhost [127.0.0.1]) by odyssey.emsl.pnl.gov (8.14.1/8.14.1) with ESMTP id m26J04u3007099 for ; Thu, 6 Mar 2008 11:00:05 -0800 (PST) Received: (from majordom@localhost) by odyssey.emsl.pnl.gov (8.14.1/8.14.1/Submit) id m26J04Wu007098 for nwchem-users-outgoing-0915; Thu, 6 Mar 2008 11:00:04 -0800 (PST) X-Authentication-Warning: odyssey.emsl.pnl.gov: majordom set sender to owner-nwchem-users@emsl.pnl.gov using -f X-IronPort-AV: E=Sophos;i="4.25,457,1199692800"; d="scan'208";a="69274015" X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Subject: RE: [NWCHEM] imaginary freq using auxiliary functions-DGauss A2 Date: Thu, 6 Mar 2008 11:00:01 -0800 Message-ID: <37275E1A43F1EC4C938AEFA1CFFAA0E302CD00D5@EMAIL02.pnl.gov> In-Reply-To: <47D00448.1000201@zsr.uni-hannover.de> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: [NWCHEM] imaginary freq using auxiliary functions-DGauss A2 Thread-Index: Ach/mtZrgzvnHbYSTzaNuTcWSILZnAAHbBNA References: <47AC785F.1090101@ornl.gov> <37275E1A43F1EC4C938AEFA1CFFAA0E302CCFFAD@EMAIL02.pnl.gov> <47CEAC8C.4060905@zsr.uni-hannover.de> <47D00448.1000201@zsr.uni-hannover.de> From: "Vorpagel, Erich R" To: "Henryk Wicke" , "Alvaro Vazquez" Cc: "users nwchem" X-OriginalArrivalTime: 06 Mar 2008 19:00:02.0285 (UTC) FILETIME=[4840E1D0:01C87FBC] Sender: owner-nwchem-users@emsl.pnl.gov Precedence: bulk Greetings Folks, =20 Yes, I did use a different approach. The smallest two exponents are used to generate a ratio which is used to calculate the diffuse exponent for a given shell. ratio =3D (smallest exponent)/(next to smallest exponent). So from = the oxygen atom TZVP basis set, the two smallest outer shell s exponents are 0.276623 and 0.9398526 to give a ratio of 0.294324131... The diffuse exponent is then calculated by multiplying the ratio and the smallest exponent. Thus, (Diffuse oxygen s) =3D 0.294324313 * 0.276623 =3D 0.081416874434999=20 which I rounded off to 0.08142 for the TZVP+ Oxygen 2s diffuse exponent. This was the same method used in ECCE for adding Diffuse functions. Erich R. Vorpagel, Ph.D. Chief Scientist Manager Computational Grand-Challenge and Pilot Projects Capability Steward MSCF Visualization and User Services Group=20 Environmental Molecular Sciences Laboratory Pacific Northwest National Laboratory =20 P. O. Box 999, MS: K8-83 3335 Q Ave. Phone: 509-371-6448 Richland, WA 99352 FAX: 509-371-6445 Email: erich.vorpagel@pnl.gov=20 Web: http://mscf.emsl.pnl.gov/ -----Original Message----- From: owner-nwchem-users@emsl.pnl.gov [mailto:owner-nwchem-users@emsl.pnl.gov] On Behalf Of Henryk Wicke Sent: Thursday, March 06, 2008 6:49 AM To: Alvaro Vazquez Cc: users nwchem Subject: Re: [NWCHEM] imaginary freq using auxiliary functions-DGauss A2 Dear Alvaro, thanks a lot for your message and the link to the interesting article I=20 wasn't aware of before. In the paper they actually apply that equation 22 to the Ahlrichs DZ=20 basis ("Ahlrichs VTZ" at EMSL) for Selenium, so it's quite interesting=20 for me. Anyway, I think Erich Vorpagel must have used a different technique=20 because I tried the equation on the TZVP basis by Godbout et al. which=20 you've used and got different exponents: 0.02859 and 0.02989 for C instead of 0.04441 and 0.02922 0.05294 and 0.05 for O instead of 0.08142 and 0.04812 0.04538 for H instead of 0.04573 Thanks again and best regards, Henryk Alvaro Vazquez wrote: > Dear Henryk, > > Is true, there isnot benefits when you are using fitting with hybrid > funtionals. > I was calibrating functionals and basis sets for a specific system. > > About to how to get exponents for diffuse functions, I found a review > where is commented the necessity to use diffuse funcions to charged species and > how to determine the exponents (Chem. Rev.; 2002; 102(1); 231-282. > http://dx.doi.org/10.1021/cr990044u see eq. 22). > > Perhaps Erich Vorpagel could comment us other way to get exponents for diffuse > functions. > > Thanks > Alvaro