From owner-chemistry -x- at -x- ccl.net Wed Feb 21 08:41:00 2007 From: "Abrash, Samuel sabrash++richmond.edu" To: CCL Subject: CCL:G: G03-Excessive mixing of frozen core and valence orbitals Message-Id: <-33670-070221061427-20516-z0lVAhkLMUy769J+MWtl3g^_^server.ccl.net> X-Original-From: "Abrash, Samuel" Content-class: urn:content-classes:message Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="iso-8859-1" Date: Tue, 20 Feb 2007 10:55:03 -0500 MIME-Version: 1.0 Sent to CCL by: "Abrash, Samuel" [sabrash]~[richmond.edu] Hi Ding, You'd said earlier that when you use MP2=full the job works but the Mulliken distributions don't make sense. Its well known that the Mulliken distributions tend to fail when diffuse functions are included in the basis set. Best, Sam -----Original Message----- > From: owner-chemistry*o*ccl.net [mailto:owner-chemistry*o*ccl.net] Sent: Tuesday, February 20, 2007 2:33 AM To: Abrash, Samuel Subject: CCL:G: G03-Excessive mixing of frozen core and valence orbitals Sent to CCL by: Xunlei Ding [dingxunlei[-]gmail.com] Dear David, Thank you very much for your clear explain and suggestions. If this work is done with 6-311++G**, the error is the same: " Largest valence mixing into a core orbital is 9.94D-01 Excessive mixing of frozen core and valence orbitals." And it succeeds with mp2=full. Best regards! Ding Shobe, David David.Shobe---sud-chemie.com wrote: >Sent to CCL by: "Shobe, David" [David.Shobe=sud-chemie.com] >Because you are using ghost atoms, there are empty core orbitals. The HF routine does not distinguish core orbitals from valence orbitals (at least I think this is true), but the MP2 defaults to MP2=FC (frozen core) which of coures does make the distinction. Note that 9.92D-01 = 0.992 = almost full occupancy! > >Perhaps using mp2=full would help, but then you'd have to redo any other mp2 calculations (you have to be consistent: either mp2=full everywhere or mp2=fc everywhere). > >A larger basis set like 6-311+G* or aug-cc-pVTZ is likely to improve matters. With more valence orbitals to choose from, there is less need to use the core orbital. The 6-31G* basis set is too small for accurate calculations of hydrogen bonding energies anyway, if that's what you're after. Again, you'd have to redo any calculations you've already performed using the smaller basis set. > >Regards, >--David Shobe > Süd-Chemie, Inc. > >Attention to detial is essentail. > > >-----Original Message----- > > >>From: owner-chemistry=-=ccl.net [mailto:owner-chemistry=-=ccl.net] >> >> >Sent: Sunday, February 18, 2007 8:39 PM >To: Shobe, David >Subject: CCL:G: G03-Excessive mixing of frozen core and valence orbitals > > >Sent to CCL by: Ding Xunlei [dingxunlei^-^gmail.com] Dear All, > >I did a calculation using mp2 and I find an error below. >Could you give me some suggestion? Thank you! > >This input file: >------------------------------------------------------------ >#mp2/6-31G* > >test > >0 1 > H 0.000000 -0.917747 0.000000 > F 0.000000 0.000000 0.000000 > H 0.000000 1.956639 0.000000 > F 0.000000 2.875828 0.000000 > H 0.000000 -0.917747 -2.250000 > F 0.000000 0.000000 -2.250000 > H-bq 0.000000 1.956639 -2.250000 > F-bq 0.000000 2.875828 -2.250000 > >------------------------------------------------------------ > >The output file: >------------------------------------------------------------------- > SCF Done: E(RHF) = -199.305447100 A.U. after 21 cycles > Convg = 0.2272D-08 -V/T = 1.9654 > S**2 = 0.0000 > ExpMin= 1.61D-01 ExpMax= 7.00D+03 ExpMxC= 1.05D+03 IAcc=1 IRadAn= > 1 AccDes= 1.00D-06 > HarFok: IExCor= 205 AccDes= 1.00D-06 IRadAn= 1 IDoV=1 > ScaDFX= 1.000000 1.000000 1.000000 1.000000 Largest valence mixing into a core orbital is 9.92D-01 Excessive mixing of frozen core and valence orbitals. > Error termination via Lnk1e in /gaussian/03d02-xlf10/g03/l801.exe at Sat Feb 17 16:37:03 2007. > > >Yours sincerely, >Ding Xunlei >2007-02-19 >______________________________________________http://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txtThis e-mail message may contain confidential and / or privileged information. If you are not an addressee or otherwise authorized to receive this message, you should not use, copy, disclose or take any action based on this e-mail or any information contained in the message. If you have received this material in error, please advise the sender immediately by reply e-mail and delete this message. >Thank you.http://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txt