From owner-nwchem-users@emsl.pnl.gov Sat Apr 26 21:33:54 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 m3R4XrUR009061 for ; Sat, 26 Apr 2008 21:33:54 -0700 (PDT) Received: (from majordom@localhost) by odyssey.emsl.pnl.gov (8.14.1/8.14.1/Submit) id m3R4Xrod009060 for nwchem-users-outgoing-0915; Sat, 26 Apr 2008 21:33:53 -0700 (PDT) X-Authentication-Warning: odyssey.emsl.pnl.gov: majordom set sender to owner-nwchem-users@emsl.pnl.gov using -f X-Ironport-SG: OK_Domains X-Ironport-SBRS: 5.4 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: Am4DANqeE0igW1Ybmmdsb2JhbACIPYh7IQEBAQEBCAUIBxEFRJcmAYE1 X-IronPort-AV: E=Sophos;i="4.25,712,1199692800"; d="scan'208";a="74713018" Date: Sun, 27 Apr 2008 00:33:02 -0400 From: "Harrison, Robert J." Subject: RE: [NWCHEM] Does the MO use 3p or 4p functions? To: Eric Yezdimer Cc: nwchem-users@emsl.pnl.gov Message-id: <003301c8a81f$e3619bf9$16015ba0@ornl.gov> MIME-version: 1.0 X-MIMEOLE: Produced By Microsoft Exchange V6.5 Content-type: text/plain; charset=iso-8859-1 Content-transfer-encoding: quoted-printable Importance: normal X-Priority: 3 Thread-Topic: [NWCHEM] Does the MO use 3p or 4p functions? thread-index: AcinvosiFFs4OhsNRaizGHNsRG+KIAAKcy8QAA3i4Og= Content-class: Sender: owner-nwchem-users@emsl.pnl.gov Precedence: bulk This is correct and the manual describes how to print a detailed list of = the basis functions and their numbering RJ Harrison (from phone 865 274 8544) -----Original Message----- From: Eric Yezdimer Sent: Saturday, April 26, 2008 7:01 PM To: nwchem-users@emsl.pnl.gov Subject: Re: [NWCHEM] Does the MO use 3p or 4p functions? Thank you very much for your answer Kirk. It was very helpful. I would like to ask everyone if there is someone else who can "confirm" Kirk's comment that the coefficients for the linear combination of = multiple GTOs, on the same atomic site, outputted in the "NWChem vector analysis" section appear in the order they are listed in the basis set = description.=20 So for a basis set described by (this table is outputted by NWChem as = well): alpha coefficient C 4 P 9.439*10^0 0.038109 2.002*10^0 0.209480 5.456*10^-1 0.508557 C 5 P 1.517*10^-1 1.00000 with a MO output of: Coeff Atom# AtomType Function 0.302387 2 C pz 0.267493 2 C pz means the MO is given by: phi(MO) =3D 0.302387 * GTO(atom 2, #4 P, pz) + 0.267493 * GTO (atom 2, = #5 P, pz) + .... =20 and not by phi(MO) =3D 0.302387 * GTO(atom 2, #5 P, pz) + 0.267493 * GTO (atom 2, = #4 P, pz) + .... because 0.302387 2 C pz appears first in the basis set list so it is a = GTO(# 4 P pz) and 0.267493 2 C pz is a GTO (#5 P pz) because it appears second (where pz refers to the pz non radial component of a 2 p orbital = geometry) One other issue, as I think Kenneth was also describing, it is possible = the GTO's center might not even be localized to a given atom site. I am = assuming that since the NWChem output is listing "some as yet 100% identified" = GTOs being associated with various atoms that the GTOs are centered on that atom's nuclear position (and not on a bond bisector or some other = place). But I do not know if this is true (or how to check for this possibility = in the NWChem output). David speaks of "print[ing] the basis out". This there a command (beyond = the normal NWChem output) that tells NWChem to print out the "exact and complete" GTOs used to compose each MO it uses? Cheers, Eric PS> I know the difference between GTOs, MOs, and the exact wave function solution for H atom (which is only valid for that and maybe the H2(+) or He(+1) cases), so it would be more efficient if we could restrict the discussion to the technical aspects of obtaining more detailed MO information from the NWChem software.=20 > -----Original Message----- > From: owner-nwchem-users@emsl.pnl.gov=20 > [mailto:owner-nwchem-users@emsl.pnl.gov] On Behalf Of Kirk Peterson > Sent: Saturday, April 26, 2008 12:53 PM > To: Eric Yezdimer > Cc: nwchem-users@emsl.pnl.gov > Subject: [SPAM-MAIL] Re: [NWCHEM] Does the MO use 3p or 4p functions? >=20 > Eric, >=20 > admittedly I'm not much of a NWCHEM user so I can't say for=20 > sure, but usually the basis functions in MO outputs are=20 > ordered consecutively, so I would assume your first choice=20 > below is correct. >=20 > -Kirk >=20 > On Apr 26, 2008, at 9:47 AM, Eric Yezdimer wrote: > > Kirk, > > > > Great. Now we are making progress. Okay now in the basis set=20 > > description outputted by NWChem there are two GTO functions=20 > with p for=20 > > a carbon atom. > > They have the INDEX of 4 and 5. The functions are listed below. > > > >>>> alpha coefficient > >>>> Carbon 4P 9.439*10^0 0.038109 > >>>> 2.002*10^0 0.209480 > >>>> 5.456*10^-1 0.508557 > >>>> > >>>> Carbon 5P 1.517*10^-1 1.00000 > > > > In the MO output I get > > > >>>> Coeff Atom# AtomType Function > >>>> 0.302387 2 C pz > >>>> -0.302387 11 C pz > >>>> 0.267493 2 C pz > >>>> -0.267493 17 C pz > >>>> -0.199437 4 C pz > >>>> 0.302387 7 C pz > >>>> -0.302387 17 C pz > >>>> 0.267493 7 C pz > >>>> -0.267493 11 C pz > >>>> -0.199437 9 C pz > > > > So my question is how should I combine the two p like=20 > orbitals, #4 or=20 > > #5, associated with atom #2. > > > >>>> phi(MO) =3D 0.302387 * GTO(atom 2, #4 P, pz) + 0.267493 *=20 > GTO (atom=20 > >>>> 2, #5 > > P, pz) + .... or > >>>> phi(MO) =3D 0.302387 * GTO(atom 2, #5 P, pz) + 0.267493 *=20 > GTO (atom=20 > >>>> 2, #4 > > P, pz) + .... > > > > Or is my understanding of the output completely wrong. > > > > Eric > > > >> -----Original Message----- > >> From: Kirk Peterson [mailto:kipeters@wsu.edu] > >> Sent: Saturday, April 26, 2008 12:31 PM > >> To: Eric Yezdimer > >> Cc: nwchem-users@emsl.pnl.gov > >> Subject: [SPAM-MAIL] Re: [NWCHEM] Does the MO use 3p or 4p=20 > functions? > >> > >> Dear Eric, > >> > >> All standard Gaussian basis sets only use basis functions of the=20 > >> following type: > >> > >> s: 1s > >> p: 2p > >> d: 3d > >> f: 4f > >> g: 5g > >> etc. > >> > >> Multiple functions of the same angular momentum do not=20 > correspond to=20 > >> higher values of the principle quantum number as you=20 > assume, but only=20 > >> to different values of the Gaussian exponent (for the primitive=20 > >> functions) or different contractions of the primitive Gaussians. > >> > >> Hope that helps some. > >> > >> -Kirk > >> > >> On Apr 26, 2008, at 8:57 AM, Eric Yezdimer wrote: > >>> Does anyone have a more informative and less superficial > >> response that > >>> could help me obtain the exact equation of the MO outputted > >> by NWChem? > >>> Any help > >>> would be greatly appreciated. > >>> > >>> Thanks, > >>> Eric > >>> > >>>> -----Original Message----- > >>>> From: Dixon, David [mailto:dadixon@as.ua.edu] > >>>> Sent: Saturday, April 26, 2008 11:36 AM > >>>> To: Eric Yezdimer > >>>> Subject: [SPAM-MAIL] RE: [NWCHEM] Does the MO use 3p or 4p > >> functions? > >>>> > >>>> You are completely wrong. There is no such thing as a 4p or 5p=20 > >>>> orbital in the correlation consistent basis sets. You > >> really need to > >>>> read about Gaussian wave functions. > >>>> > >>>> -----Original Message----- > >>>> From: owner-nwchem-users@emsl.pnl.gov=20 > >>>> [mailto:owner-nwchem-users@emsl.pnl.gov] On Behalf Of=20 > Eric Yezdimer > >>>> Sent: Saturday, April 26, 2008 10:21 AM > >>>> To: nwchem-users@emsl.pnl.gov > >>>> Subject: Re: [NWCHEM] Does the MO use 3p or 4p functions? > >>>> > >>>> > >>>> Thank you for the response. I will check out the visualization=20 > >>>> software. > >>>> > >>>> I understand that NWChem does not see 1 s and 2 s as in the=20 > >>>> traditional sense. However it does have several GTO to=20 > approximate=20 > >>>> these orbitals. I am using the cc-pVDZ basis set. > >>>> For the carbon atom this basis set has descriptions of the > >> 4P and 5P > >>>> orbitals. > >>>> > >>>> alpha coefficient > >>>> Carbon 4P 9.439*10^0 0.038109 > >>>> 2.002*10^0 0.209480 > >>>> 5.456*10^-1 0.508557 > >>>> > >>>> Carbon 5P 1.517*10^-1 1.00000 > >>>> > >>>> After I run NWChem several MO are reported. One=20 > antibonding pi MO,=20 > >>>> for example, is: > >>>> > >>>> Coeff Atom# AtomType Function > >>>> 0.302387 2 C pz > >>>> -0.302387 11 C pz > >>>> 0.267493 2 C pz > >>>> -0.267493 17 C pz > >>>> -0.199437 4 C pz > >>>> 0.302387 7 C pz > >>>> -0.302387 17 C pz > >>>> 0.267493 7 C pz > >>>> -0.267493 11 C pz > >>>> -0.199437 9 C pz > >>>> > >>>> In this MO the atoms 2, 11, 7 , and 17 appear to have 2=20 > pz atomic=20 > >>>> orbitals on each atomic site. I assume one of the atomic > >> orbitals is > >>>> described by the GTO for 4P and one is described for the > >> GTO for 5P. > >>>> > >>>> In this list however (outputted from NWChem) it does not > >> tell me (as > >>>> far as I can see) which is which. For example should I write > >>>> > >>>> phi(MO) =3D 0.302387 * GTO(atom 2, 4P, pz) + 0.267493 *=20 > GTO(atom 2,=20 > >>>> 5P, > >>>> pz) + > >>>> .... or > >>>> phi(MO) =3D 0.302387 * GTO(atom 2, 5P, pz) + 0.267493 *=20 > GTO(atom 2,=20 > >>>> 4P, > >>>> pz) + > >>>> .... > >>>> > >>>> I am using Mathematica (a program like Matlab) to build > >> out the all > >>>> the MO for the molecules I am interested with so I can > >> perform some > >>>> non-standard calculations of molecular properties. This is > >> why I need > >>>> to know the "exact" > >>>> functional form for the MO found by NWChem. > >>>> > >>>> Any additional insights you could provide would be helpful. > >>>> > >>>> Thanks, > >>>> Eric > >>>> > >>>> > >>>>> -----Original Message----- > >>>>> From: JR Schmidt [mailto:jordan.schmidt@yale.edu] > >>>>> Sent: Saturday, April 26, 2008 9:53 AM > >>>>> To: Eric Yezdimer > >>>>> Subject: [Norton AntiSpam] [SPAM-MAIL] Re: [NWCHEM] Does > >>>> the MO use 3p > >>>>> or 4p functions? > >>>>> > >>>>> > >>>>>> The normal vector output display shows each molecular > >>>>> orbital of course > >>>>>> and each MO is given as a linear combination of basis > >>>>> functions on the > >>>>>> separate atoms. The output also gives the harmonics of > >> each basis > >>>>>> function (s, px, py, pz, etc) however it does not tell me if > >>>>> it is a 2px or a 3px function. > >>>>>> There is only a Bfn. value, that I assume is an internal > >>>>> reference to > >>>>>> the exact basis functions being used. > >>>>> > >>>>> NWChem (nor any other electronic structure program) knows > >> anything > >>>>> about 1s vs. 2s vs 2p 'shells'. These are simply > >>>> approximations based > >>>>> on hydrogen like orbitals. > >>>>> > >>>>> Most common basis sets include MANY sets of s,p,d,f > >>>> functions, even if > >>>>> you do a calculation on an isolated hydrogen atom. For > >>>> example, you > >>>>> would find that the 1s orbital of an H atoms would consist > >>>> of a linear > >>>>> combination of several 's'-like basis functions in the=20 > MO output. > >>>>> This is necessary to increase accuracy (to approximate the > >>>> exponential > >>>>> decay of the 1s orbital with gaussian basis functions), > >> because the > >>>>> orbital exponent of H varies with its chemical environment, etc. > >>>>> > >>>>> In practice, you will find that the calculation (on an > >> atom) still > >>>>> yields a shell-like structure, with identifiable > >>>> 1s,2s,2p,3s orbitals. > >>>>> However, as I mentioned, each of those orbitals will be > >> composed of > >>>>> several s or p type basis functions. The exception would > >> be if you > >>>>> use a minimal basis set (such as STO-3G) which contains > >>>> only 1 basis > >>>>> function per electron. In this case, one basis function > >> represents > >>>>> each orbital, and you will get results like you originally > >>>> expected. > >>>>> Unfortunately, the results will be quantitatively terrible, and=20 > >>>>> possible also qualitatively poor! > >>>>> > >>>>> I hope that helps. Forgive me if I am simply telling=20 > you what I=20 > >>>>> already know. If not, I recommend the book 'Introduction to=20 > >>>>> Computational Chemistry', by Jenson. It is quite good, > >> and an easy > >>>>> read. > >>>>> > >>>>> JR Schmidt > >>>>> > >>>>> > >>>>> > >>>> > >>>> > >>>> > >>> > >> > >> > >> > > >=20 >=20 >=20