CCL:G: Gaussian 09 SA-CASSCF and Forces



 Sent to CCL by: "Jim Kress" [ccl_nospam_+_kressworks.com]
 Did you try:
 http://www.gaussian.com/g_tech/1.htm
 ?
 Jim
 -----Original Message-----
 > From: owner-chemistry+ccl_nospam==kressworks.com+*+ccl.net
 [mailto:owner-chemistry+ccl_nospam==kressworks.com+*+ccl.net] On Behalf Of
 Mireia Segado Centellas mireia.segado[-]sns.it
 Sent: Tuesday, November 12, 2013 8:52 AM
 To: Kress, Jim
 Subject: CCL: Gaussian 09 SA-CASSCF and Forces
 Sent to CCL by: "Mireia Segado Centellas" [mireia.segado.+*+.sns.it]
 Dear all,
 I am launching several calculation at FC zone with different state average
 (a).Also I am interest in forces of some excited state (b). Here it is an
 example of some computation with state average and state average with
 forces.
 _________ non state average
 %chk=nicA_cas108_opt_s0_forces2.chk
 %mem=500MB
 #P CAS(10,8,nroot=3)/cc-pvdz pop=full guess=(read) nosymm
      gfinput  TEST scfcon=6  iop(6/7=3)
      scfcyc=150 geom=checkpoint force
 orbitals
 0 1
 ________ state average
 %mem=500MB
 #P CAS(10,8,nroot=3,stateaverage)/cc-pvdz pop=full guess=(read) nosymm
      gfinput  TEST scfcon=6  iop(6/7=3)
      scfcyc=150 geom=checkpoint
 orbitals
 0 1
  0.3333333 0.3333333 0.3333333
 ______ state average with forces
 %chk=nicA_cas108_opt_s0_nroot2_sa3.chk
 %mem=500MB
 #P CAS(10,8,nroot=3,stateaverage)/cc-pvdz pop=full guess=(read) nosymm
      gfinput  TEST scfcon=6  iop(6/7=3) force
      scfcyc=150 geom=checkpoint
 orbitals
 0 1
  0.3333333 0.3333333 0.3333333
 ___________
 I have found different problems
 1) Non state average calculation give a r3-wavefunction composed by 0.68 GS
 and 0.47 double excitation. Very strange an also not the state expected.
 2) State average computation also do not show the state expected in other
 methods and the sa-casscf computed in molcas.
 3) when force keyword is incorpored the program not read the line where the
 weights of state are incorpored. And put a 0.5 0.5 for the high in energy
 states demanded. In g09 when you compute forces they incopore the computing
 of derivative coupling. Some manner of making read the weights?
 The active orbitals are checked and are correct.
 Thank you very much in advance
 mireia segadohttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp-:-//www.ccl.net/chemistry/sub_unsub.shtmlhttp-:-//www.ccl.net/spammers.txt