[molpro-user] A question about density-fitting in SAPT

Grant Hill jghill at wsu.edu
Wed May 6 16:41:27 BST 2009


Cong,

Try putting your explicit basis definition within the basis block, e.g.

basis={
default,avdz

set,jkfit
s, He, 66.205029000,13.171713600,3.1622361000,0.81536220000
c,1.1, 0.53033986000E-01
c,2.2, 0.39465220220
c,3.3, 0.91729877120
p, He, 9.6128819820,3.2042939700,1.1643102000
d, He, 7.6739799000,0.98779470000

set,mp2fit
default,avdz/mp2fit
}

HTH,

Grant


On 6 May 2009, at 08:03, cwang at mappi.helsinki.fi wrote:

> Glen,
>
>   Hi,
>
>   Many thanks for your reply, I try to add the comma before the
> angular momentum, still the same problem exists, here is the whole
> output file :(
> Sorry for trouble you
>
> Best regards
> Cong
>
>  CPU=Opteron 2600.401:2600.401:2600.401:2600.401 MHz
>  mxmblk= 64  mxmbln= 64  ncache=  16384  mindgm= 11  mindgv= 33
> mindgc=  4  mindgl=  4  mindgr=  1  noblas=0  nroll=2  minvec=7
>  default implementation of scratch files=df
>
>  ***,
>  r=5.6
>  geometry={nosym; he1; he2,he1,r}
>  basis={
>  default,avdz
>  set,jkfit
>  default,TM/jkfit
>  set,mp2fit
>  default,avdz/mp2fit
>  }
>
>  TM={
>  s, HE, 66.205029000,13.171713600,3.1622361000,0.81536220000
>  c,1.1, 0.53033986000E-01
>  c,2.2, 0.39465220220
>  c,3.3, 0.91729877120
>  c,4.4, 1.0000000000
>  p, HE, 9.6128819820,3.2042939700,1.1643102000
>  c,1.1, 1.0000000000
>  c,2.2, 1.0000000000
>  c,3.3, 1.0000000000
>  d, HE, 7.6739799000,0.98779470000
>  c,1.1, 1.0000000000
>  c,2.2, 1.0000000000
>  }
>
>
>
>            !wf records
>  ca=2101.2
>  cb=2102.2
>
>
>
>            !monomer A
>  dummy,he2
>  {hf; save,$ca}
>  sapt;monomerA
>
>
>            !monomer B
>  dummy,he1
>  {hf; start,atdens; save,$cb}
>  sapt;monomerB
>
>
>            !interaction contributions
>  {sapt;intermol,ca=$ca,cb=$cb,fitlevel=3
>  dfit,basis_coul=jkfit,basis_exch=jkfit,basis_mp2=mp2fit,cfit_scf=3}
>
>
>
>  Variables initialized (517), CPU time= 0.01 sec
>  Commands  initialized (293), CPU time= 0.01 sec, 437 directives.
>  Default parameters read. Elapsed time= 0.06 sec
>  Checking input...
>
>  Unknown command or directive S
>
>  Unknown command or directive C
>
>  Unknown command or directive C
>
>  Unknown command or directive C
>
>  Unknown command or directive C
>
>  Unknown command or directive P
>
>  Unknown command or directive C
>
>  Unknown command or directive C
>
>  Unknown command or directive C
>
>  Unknown command or directive D
>
>  Unknown command or directive C
>
>  Unknown command or directive C
>
>    12 input errors found
>
>  ? Error
>  ? Input errors!
>  ? The problem occurs in check_input
>
>  ERROR EXIT
>  CURRENT STACK:      MAIN
>
>
>
> **********************************************************************************************************************************
>  PROGRAMS   *        TOTAL
>  CPU TIMES  *         0.14
>  REAL TIME  *         0.20 SEC
>  DISK USED  *         0.00 MB
>
> **********************************************************************************************************************************
>
>
>
>
>
> Quoting "Glen Jenness" <grj3+ at pitt.edu>:
>
>> Cong,
>> So there are a couple things for you to do.
>>
>> I tried it myself, and there is no jk fitting basis sets for helium- 
>> there
>> is however mp2 fitting basis sets.  Unfortunately, if you want to use
>> density fitting in your DFT-SAPT you'll need to have a jkfit basis  
>> set.
>> The reason for this is in the following paper:
>>
>> @ARTICLE{Hesselmann2005,
>>  author = {A. He{\ss}elmann and G. Jansen and M. Sch\"{u}tz},
>>  title = {Density-functional theory-symmetry-adapted intermolecular
>> perturbation theory with density fitting: A new efficient method to
>> study intermolecular interaction energies},
>>  journal = {The Journal of Chemical Physics},
>>  year = {2005},
>>  volume = {122},
>>  pages = {014103},
>>  number = {1},
>>  doi = {10.1063/1.1824898},
>>  url = {http://link.aip.org/link/?JCP/122/014103/1}
>> }
>>
>> In short, the jk basis sets are used for everything but dispersion,  
>> and
>> mp2 is used for dispersion and exchange dispersion (basically,  
>> jkfit is
>> for anything that's not correlation, and mp2fit is for anything  
>> that is
>> correlation).
>>
>> For your second issue, I believe this is a formating issue.  Try  
>> adding a
>> comma between the angular momentum type and the atom label:
>>
>> TM={
>> s,HE,66.205029000,13.171713600,3.1622361000,0.81536220000
>> ....
>> }
>>
>> Glen
>>
>>
>>
>>>
>>> Thank you, take your time :)
>>>
>>> Cong
>>>
>>> Quoting "Glen Jenness" <grj3+ at pitt.edu>:
>>>
>>>> Cong,
>>>> I'll take a look, but I'm going to be slighty busy all day.   
>>>> Hopefully I
>>>> can respond to you by tonight :)
>>>>
>>>> Glen
>>>>
>>>>
>>>>> Glen,
>>>>>
>>>>>  Hi!
>>>>>
>>>>>  Many thanks for your reply. There still several problem exists.
>>>>> First, I try to use your input, the output file complains
>>>>>
>>>>> *****************
>>>>> cannot find default basis VTZ for atomic number  2  type=FIT
>>>>> context=JKFIT                             attributes
>>>>> ERROR EXIT
>>>>> CURRENT STACK:      MAIN
>>>>> ************
>>>>>
>>>>> It seems there's no JK fit basis for Helium in Molpro. (a) If I  
>>>>> try to
>>>>> skip DF-HF treatment, just do DF-MP2-FIT, modify the input like  
>>>>> that
>>>>>
>>>>>
>>>>> ***************
>>>>> r=5.6
>>>>> geometry={nosym; he1; he2,he1,r}
>>>>> basis={
>>>>> default,avdz
>>>>> set,mp2fit
>>>>> default,avdz/mp2fit  (by the way, is that should be mp2fit or  
>>>>> jkfit)
>>>>> }
>>>>>
>>>>> !wf records
>>>>> ca=2101.2
>>>>> cb=2102.2
>>>>>
>>>>>
>>>>> !monomer A
>>>>> dummy,he2
>>>>> {hf; save,$ca}
>>>>> sapt;monomerA
>>>>>
>>>>> !monomer B
>>>>> dummy,he1
>>>>> {hf; start,atdens; save,$cb}
>>>>> sapt;monomerB
>>>>>
>>>>> !interaction contributions
>>>>> {sapt;intermol,ca=$ca,cb=$cb,fitlevel=3
>>>>> dfit,basis_mp2=mp2fit,cfit_scf=3}
>>>>>
>>>>> ***************
>>>>>
>>>>> The output file complains for
>>>>>
>>>>> ******
>>>>> Memory available:                             7272627 words
>>>>> ? Error
>>>>> ? Basis set  not found
>>>>> ? The problem occurs in load_basis
>>>>>
>>>>> ERROR EXIT
>>>>> CURRENT STACK:      MAIN
>>>>> ******
>>>>> What is the reason for that and how to solve this problem?
>>>>>
>>>>> (b) Try do find a JK fit basis set, (although density fitting in  
>>>>> HF
>>>>> may not very helpful than DF-DFT/MP2) Turbomole claim they have a
>>>>> universal JK basis set for
>>>>>
>>>>> he def2-QZVP
>>>>> he def2-QZVPP
>>>>> he def2-TZVP
>>>>> he def2-TZVPP
>>>>> he small
>>>>> he d-aug-cc-pVDZ
>>>>> he d-aug-cc-pVTZ
>>>>> he d-aug-cc-pVQZ
>>>>> he d-aug-cc-pV5Z
>>>>>
>>>>> all these cases (?!)
>>>>>
>>>>> I try to import this JK basis as
>>>>>
>>>>> input file
>>>>>
>>>>> *****
>>>>> r=5.6
>>>>> geometry={nosym; he1; he2,he1,r}
>>>>> basis={
>>>>> default,avdz
>>>>> set,jkfit
>>>>> default,TM/jkfit
>>>>> set,mp2fit
>>>>> default,avdz/mp2fit
>>>>> }
>>>>>
>>>>> TM={
>>>>> s HE, 66.205029000,13.171713600,3.1622361000,0.81536220000
>>>>> c,1.1, 0.53033986000E-01
>>>>> c,2.2, 0.39465220220
>>>>> c,3.3, 0.91729877120
>>>>> c,4.4, 1.0000000000
>>>>> p HE, 9.6128819820,3.2042939700,1.1643102000
>>>>> c,1.1, 1.0000000000
>>>>> c,2.2, 1.0000000000
>>>>> c,3.3, 1.0000000000
>>>>> d HE, 7.6739799000,0.98779470000
>>>>> c,1.1, 1.0000000000
>>>>> c,2.2, 1.0000000000
>>>>> }
>>>>>
>>>>>
>>>>> !wf records
>>>>> ca=2101.2
>>>>> cb=2102.2
>>>>>
>>>>>
>>>>> !monomer A
>>>>> dummy,he2
>>>>> {hf; save,$ca}
>>>>> sapt;monomerA
>>>>>
>>>>> !monomer B
>>>>> dummy,he1
>>>>> {hf; start,atdens; save,$cb}
>>>>> sapt;monomerB
>>>>>
>>>>> !interaction contributions
>>>>> {sapt;intermol,ca=$ca,cb=$cb,fitlevel=3
>>>>> dfit 
>>>>> ,basis_coul=jkfit,basis_exch=jkfit,basis_mp2=mp2fit,cfit_scf=3}
>>>>>
>>>>> *********
>>>>>
>>>>> The output file complains
>>>>>
>>>>> ******
>>>>>
>>>>> TM={
>>>>> s HE, 66.205029000,13.171713600,3.1622361000,0.81536220000
>>>>> c,1.1, 0.53033986000E-01
>>>>> c,2.2, 0.39465220220
>>>>> c,3.3, 0.91729877120
>>>>> c,4.4, 1.0000000000
>>>>> p HE, 9.6128819820,3.2042939700,1.1643102000
>>>>> c,1.1, 1.0000000000
>>>>> c,2.2, 1.0000000000
>>>>> c,3.3, 1.0000000000
>>>>> d HE, 7.6739799000,0.98779470000
>>>>> c,1.1, 1.0000000000
>>>>> c,2.2, 1.0000000000
>>>>> }
>>>>> ...skipping...
>>>>> Unknown command or directive C
>>>>>
>>>>> Unknown command or directive P
>>>>>
>>>>> Unknown command or directive C
>>>>>
>>>>> Unknown command or directive C
>>>>>
>>>>> Unknown command or directive C
>>>>>
>>>>> Unknown command or directive D
>>>>>
>>>>> Unknown command or directive C
>>>>>
>>>>> Unknown command or directive C
>>>>>
>>>>>   12 input errors found
>>>>>
>>>>> ? Error
>>>>> ? Input errors!
>>>>> ? The problem occurs in check_input
>>>>>
>>>>> ERROR EXIT
>>>>> *********
>>>>>
>>>>> What is the reason for that? and how to solve this problem?
>>>>>
>>>>> Thank you very much
>>>>> Best regards
>>>>>
>>>>> Cong Wang
>>>>>
>>>>> Department of Chemistry
>>>>> Laboratory for Instruction in Swedish
>>>>> University of Helsinki
>>>>> A.I. Virtanens plats 1
>>>>> P.O. Box 55
>>>>> FI-00014 University of Helsinki
>>>>> FINLAND
>>>>>
>>>>>
>>>>>
>>>>> 2009/5/4 Glen Jenness <grj3+ at pitt.edu>:
>>>>>> Cong,
>>>>>> The problem is that you didn't specify the jkfit basis set in  
>>>>>> your
>>>>>> basis
>>>>>> input area. Â So if you wanted to use aug-cc-pVDZ (just as an  
>>>>>> example)
>>>>>> your
>>>>>> basis input area should look something like:
>>>>>>
>>>>>> basis={
>>>>>> default,avdz
>>>>>> set,jkfit
>>>>>> default,vtz/jkfit
>>>>>> set,mp2fit
>>>>>> default,avdz/jkfit
>>>>>> }
>>>>>>
>>>>>> So if your basis is avxz (where x=d,t,q etc.) then jkfit is v(x 
>>>>>> +1)z
>>>>>> and
>>>>>> mp2fit is avxz.
>>>>>>
>>>>>> Glen
>>>>>>
>>>>>>> Dear everyone,
>>>>>>>
>>>>>>> Â Hello!
>>>>>>>
>>>>>>> Â I have a question about density-fitting in SAPT in molpro  
>>>>>>> 2006,
>>>>>>> according to the manual,
>>>>>>> http://www.molpro.net/info/current/doc/manual/node410.html
>>>>>>>
>>>>>>> "For this one may use the input:
>>>>>>>
>>>>>>> {sapt;intermol,ca=$ca,cb=$cb,fitlevel=3
>>>>>>> dfit 
>>>>>>> ,basis_coul=jkfit,basis_exch=jkfit,basis_mp2=mp2fit,cfit_scf=3}
>>>>>>> "
>>>>>>>
>>>>>>> Suppose "this one" refer to
>>>>>>> http://www.molpro.net/info/current/doc/manual/node407.html
>>>>>>>
>>>>>>> therefore I made such an input
>>>>>>> *********************************************************************
>>>>>>> r=5.6
>>>>>>> geometry={nosym; he1; he2,he1,r}
>>>>>>> basis=avqz
>>>>>>>
>>>>>>> !wf records
>>>>>>> ca=2101.2
>>>>>>> cb=2102.2
>>>>>>>
>>>>>>>
>>>>>>> !monomer A
>>>>>>> dummy,he2
>>>>>>> {hf; save,$ca}
>>>>>>> sapt;monomerA
>>>>>>>
>>>>>>> !monomer B
>>>>>>> dummy,he1
>>>>>>> {hf; start,atdens; save,$cb}
>>>>>>> sapt;monomerB
>>>>>>>
>>>>>>> !interaction contributions
>>>>>>> {sapt;intermol,ca=$ca,cb=$cb,fitlevel=3
>>>>>>> dfit 
>>>>>>> ,basis_coul=jkfit,basis_exch=jkfit,basis_mp2=mp2fit,cfit_scf=3}
>>>>>>> *********************************************************************
>>>>>>>
>>>>>>> The output tells me
>>>>>>>
>>>>>>> ***************************
>>>>>>> Â PROGRAM * SAPT Â  Â  Â Authors: A. Hesselmann, G. Jansen, 2003
>>>>>>>
>>>>>>> Â (Calculation of intermolecular interaction energy  
>>>>>>> contributions
>>>>>>> Â for single-determinant wave-functions)
>>>>>>>
>>>>>>>  Monomer A wave function in record             2101.2
>>>>>>>  Monomer A electrostatic potential in record   1001.3
>>>>>>>  Monomer B wave function in record             2102.2
>>>>>>>  Monomer B electrostatic potential in record   1002.3
>>>>>>>  Overlap matrix in record                       
>>>>>>> 1000.3
>>>>>>>
>>>>>>> Â Factor for nonlocal exchange: Â  Â 1.000000
>>>>>>> Â Print factor: Â  0
>>>>>>> Â Density fitting level: Â  3
>>>>>>> Â Level of calculation: Â  Â 3
>>>>>>>
>>>>>>> Â CPKS solver: Â  Â 1
>>>>>>> Â CPKS convergence threshold: Â  0.10000000E-05
>>>>>>> Â CPKS maximumb number of iterations: Â  Â  50
>>>>>>>
>>>>>>> Â Occupied space for monomer A: Â  Â 1
>>>>>>>  Virtual  space for monomer A:   91
>>>>>>>
>>>>>>> Â Occupied space for monomer B: Â  Â 1
>>>>>>>  Virtual  space for monomer B:   91
>>>>>>>
>>>>>>> Â Memory available: Â  Â  Â  Â  Â  Â  Â  Â  Â  Â  Â  Â  Â  Â   
>>>>>>> 7272627
>>>>>>> words
>>>>>>> Â ? Error
>>>>>>> Â ? Basis set JKFIT not found
>>>>>>> Â ? The problem occurs in load_basis
>>>>>>>
>>>>>>> Â ERROR EXIT
>>>>>>> Â CURRENT STACK: Â  Â  Â MAIN
>>>>>>>
>>>>>>> *********************************
>>>>>>> Seems there's some problem in the JKFIT basis, then, actually  
>>>>>>> what is
>>>>>>> "this one" in the manual, Â what is the reason for that? how  
>>>>>>> to solve
>>>>>>> this problem?
>>>>>>>
>>>>>>> Thank you very much
>>>>>>> Best regards
>>>>>>> Cong Wang
>>>>>>>
>>>>>>> Department of Chemistry
>>>>>>> Laboratory for Instruction in Swedish
>>>>>>> University of Helsinki
>>>>>>> A.I. Virtanens plats 1
>>>>>>> P.O. Box 55
>>>>>>> FI-00014 University of Helsinki
>>>>>>> FINLAND
>>>>>>> _______________________________________________
>>>>>>> Molpro-user mailing list
>>>>>>> Molpro-user at molpro.net
>>>>>>> http://www.molpro.net/mailman/listinfo/molpro-user
>>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Molpro-user mailing list
>>>>>> Molpro-user at molpro.net
>>>>>> http://www.molpro.net/mailman/listinfo/molpro-user
>>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>
>>
>>
>>
>
>
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user




More information about the Molpro-user mailing list