From d.w.price@reading.ac.uk Fri Dec 6 05:51:29 1996 Received: from sussdirt.rdg.ac.uk for d.w.price@reading.ac.uk by www.ccl.net (8.8.3/950822.1) id FAA27805; Fri, 6 Dec 1996 05:02:45 -0500 (EST) Received: from indyp (actually host scsgind2) by sussdirt.rdg.ac.uk; Fri, 6 Dec 1996 10:01:37 +0000 Received: from Indyp by indyp via SMTP (940816.SGI.8.6.9/930416.SGI) for id JAA01442; Fri, 6 Dec 1996 09:59:09 GMT Sender: dave@reading.ac.uk Message-ID: <32A7EE6D.41C6@reading.ac.uk> Date: Fri, 06 Dec 1996 09:59:09 +0000 From: Dave Price Organization: none-what-so-ever X-Mailer: Mozilla 3.0Gold (X11; I; IRIX 5.3 IP22) To: Comp Chem Mailing List Subject: Earlier Post on QM treatment of metals MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Dear all, I think that my earlier posting regarding QM treatment of metals may not have been clear enough. I am interested in doing calculations on bulk metal, i.e. a 3D lattice of uncharged metal atoms (not organometallic complexes). I should like to thank the respondees for their suggestions of (mainly) DFT, but I already have experience with DFT and organometallic TM complexes. I am having convergence problems with small clusters of metal atoms and wondered if there were any methods out there (that I may have been too lazy to follow up in the past when I came across them and have now forgotten) that I can use to model (particularly) bulk actinide metal and actinide metal atom clusters. Thanks to earlier respondees and future ones. Dave ------------------------------------------------------------------------ Dr. David W. Price, Tel: +44 (0)118 9875123 extn 7415 Department of Chemistry, Fax: +44 (0)118 9316331 University of Reading, mailto:d.w.price@reading.ac.uk Whiteknights, READING http://www.chem.rdg.ac.uk/g50/mmrg/dave/dave.html RG6 6AD U.K. ------------------------------------------------------------------------ From lutz@ri.ac.uk Fri Dec 6 06:05:37 1996 Received: from mail-d.bcc.ac.uk for lutz@ri.ac.uk by www.ccl.net (8.8.3/950822.1) id FAA27870; Fri, 6 Dec 1996 05:39:33 -0500 (EST) Received: from ri.ac.uk (actually davy.ri.ac.uk) by mail-d.bcc.ac.uk with SMTP (PP); Fri, 6 Dec 1996 10:39:09 +0000 Received: from faraday.ri.ac.uk by ri.ac.uk (5.x/SMI-SVR4) id AA25782; Fri, 6 Dec 1996 10:40:53 GMT From: lutz@ri.ac.uk (Dr Lutz Ackermann) Received: by faraday.ri.ac.uk (950413.SGI.8.6.12) id KAA00776; Fri, 6 Dec 1996 10:40:52 GMT Date: Fri, 6 Dec 1996 10:40:52 GMT Message-Id: <199612061040.KAA00776@faraday.ri.ac.uk> To: CHEMISTRY@www.ccl.net, SCAN@ri.ac.uk, in@ri.ac.uk, UNITS@ri.ac.uk Original-To: CHEMISTRY@www.ccl.net, SCAN@davy.ri.ac.uk, in@davy.ri.ac.uk, UNITS@davy.ri.ac.uk, G94:@davy.ri.ac.uk PP-Warning: Parse error in original version of preceding To line Subject: CCL: Hello CCL'ers Gaussian 94' s SCAN function causes me some headache. The way, units for the stepsize are interpreted seems to be rather unpredictable. For example, I ran two rather similar jobs (see below) in which g94 acts stunningly different: ************************************** job1.log **************************** Variables: zCe -0.162 0 0. R1 2.27 2 0.05 A1 142. 2 3. Scan the potential surface. Variable Value No. Steps Step-Size -------- ----------- --------- --------- 2 2.270000 2 0.0500 3 142.000000 2 3.0000 A total of 9 points will be computed. ... Scan completed. Summary of the potential surface scan: N R1 A1 HF ---- --------- --------- ----------- 1 2.2700 142.0000 -10610.64422 2 2.3200 142.0000 -10610.64868 3 2.3700 142.0000 -10610.64848 4 2.2700 145.0000 -10610.65709 5 2.3200 145.0000 -10610.66256 6 2.3700 145.0000 -10610.66329 7 2.2700 148.0000 -10610.64848 8 2.3200 148.0000 -10610.65383 9 2.3700 148.0000 -10610.65413 ---- --------- --------- ----------- ************************************** job1.log **************************** ************************************** job2.log **************************** Variables: zCe -0.162 0 0. R1 2.353 0 0.05 A1 144. 1 2. Scan the potential surface. Variable Value No. Steps Step-Size -------- ----------- --------- --------- 3 144.000000 1 114.5916 A total of 2 points will be computed. ... Scan completed. Summary of the potential surface scan: N A1 HF ---- --------- ----------- 1 144.0000 -10610.66100 2 258.5916 -10609.94079 ---- --------- ----------- ... ************************************** job2.log **************************** In the first case, both the initial value of the angle A1 as well as the step size are being interpreted in degree (which corresponds to the default, according to the manual). In the second case however, only the initial value of the angle A1 is treated as degrees whereas the stepsize is taken in radians (yielding nonsentic geometries, of course). Has anyone encountered simmilar erratic behaviour? Is ther a way to force G94 to interprete both the initial value and the step size consistently as degrees? Thanks a lot, Cheers, Lutz. <^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^> < Dr. Lutz Ackermann > < The Royal Institution of Great Britain > < Research Assistent > < Davy-Faraday Research Laboratory > < 21 Albemarle Street > < London W1X 4BS > < UK > < > < phone : +44-171-409 2992 ext 424 > < FAX : +44-171-629 3569 > < e-mail: lutz@ri.ac.uk > < http://www.ri.ac.uk/DFRL/L.Ackermann > vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv From raperez@valdivia.uca.uach.cl Fri Dec 6 08:51:31 1996 Received: from valdivia.uca.uach.cl for raperez@valdivia.uca.uach.cl by www.ccl.net (8.8.3/950822.1) id IAA28510; Fri, 6 Dec 1996 08:02:59 -0500 (EST) Received: from [200.2.113.90] by valdivia.uca.uach.cl (AIX 3.2/UCB 5.64/4.03) id AA26686; Fri, 6 Dec 1996 08:59:46 -0500 Message-Id: <1.5.4.32.19961206130451.0069e658@valdivia.uca.uach.cl> X-Sender: raperez@valdivia.uca.uach.cl X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 06 Dec 1996 09:04:51 -0400 To: Comp Chem Mailing List From: Ramiro Arratia Perez Subject: DFT on clusters, >Date: Fri, 06 Dec 1996 09:00:43 -0400 >To: Dave Price >From: Ramiro Arratia Perez >Subject: DFT on clusters, > >Dear Prof. Price, > I understand quite well your dissapointment, but I think I can help you. > 1. Firstly, since you want to calculate cluster of metal atoms or actinides, you must use Relativistic Density Functional methods, i.e., those methods that solves the Dirac equation rather than Pauli type or Schrodinger type equations. >Currently, there are only 2 methods that can do energetics (DS, Ellis et al) and optical and magnetic properties (DSW, Arratia-Perez et al) of large clusters of any type.. See the following references > 1. D. E. Ellis et al, Inorg. Chem. 30, 3986 (1991) > 2. R. Arratia-Perez et al, Phys. Rev. B 39, 3005 (1989) or more recent > Chem. Phys. Lett. 255, 217 (1996). > >Well, all depend of what you want to calculate... > best wishes, > Ramiro > Ramiro Arratia-Perez, Ph.D. Instituto de Quimica Universidad Austral de Chile Casilla 567, Valdivia, CHILE Phone: (5663)-221902 Fax: (5663)-221294 or (5663)-221599