## Firefly and PC GAMESS-related discussion club

Learn how to ask questions correctly

Re^2: Conical intersection search with multi-states average

Anna Le
tuc59193@temple.edu

Hi Alex,

I increased the ITERMX value to 1100 and the calculation has been running for the past two weeks . It has gone through 252 NSERCH and the calculation has not yet converged. Do you recommend loosening the convergence criteria for this CI search?

Since I am optimizing for a conical intersection between the ground state and the first excited state, does the calculation give me the energy of the ground state for each NSERCH? I was only able to extract the energy of the first excited state for each point.

Anna Le

On Mon Jul 7 '14 4:49pm, Alex Granovsky wrote
---------------------------------------------
>Dear Anne,

>try to set ITERMX to 500 or larger value.

>Kind regards,
>Alex Granovsky

>On Mon Jul 7 '14 2:50pm, Anne Le wrote
>--------------------------------------
>>Dear Firefly Users,

>>This is my first time running MCSCF calculation with Firefly. I am optimizing for a conical intersection between the ground state (S0) and the first excited state (S1) with five states in the average. This is my input,

>> INPUT CARD> \$CONTRL SCFTYP=MCSCF RUNTYP=optimize inttyp=hondo
>> INPUT CARD>    icut=12 d5=.true. nzvar=1 \$END
>> INPUT CARD> \$SYSTEM MWORDS=200 nojac=1 kdiag=0  \$END
>> INPUT CARD> \$moorth nostf=1 nozero=1 tole=0 tolz=0 syms=1 symden=1 symvec=1 \$end
>> INPUT CARD> \$BASIS EXTFIL=.t. gbasis=cc-pvdz \$END
>> INPUT CARD> \$GUESS GUESS=MOREAD NORB=189 \$END
>> INPUT CARD> \$trans altpar=.t. mptran=2 dirtrf=.t. mode=112 \$end
>> INPUT CARD> \$DRT NMCC=32 NDOC=7 NVAL=4 GROUP=C1 FORS=.t. \$END
>> INPUT CARD> \$GUGDIA NSTATE=11 ITERMX=300 cvgtol=1d-7 memmax=999999 \$end
>> INPUT CARD> \$GUGDM2 cutoff=1d-12 WSTATE(1)=1,1,1,1,1 \$end
>> INPUT CARD> \$GUGEM  CUTOFF=1.0D-15 \$END
>> INPUT CARD> \$MCSCF CISTEP=GUGA soscf=.t. fullnr=.f. ISTATE=1 acurcy=3d-8
>> INPUT CARD>    ENGTOL=1.0d-12 maxit=250 \$END
>> INPUT CARD> \$mcaver target=pure jstate=2 conic=2 multiw=.t. \$end
>> INPUT CARD> \$statpt method=conic nstep=1000 \$end
>> INPUT CARD> \$conic shift=1.0E-4 \$end
>> INPUT CARD> \$data
>> INPUT CARD>conical search for pc_sofa
>> INPUT CARD>C1
>> INPUT CARD>C     6.0     3.34565997    -0.14160000    -0.07642000
>> INPUT CARD>H     1.0     1.96000004    -0.28766999     1.53061998
>> INPUT CARD>C     6.0     1.88378000     0.87823999    -1.43701005
>> INPUT CARD>C     6.0     3.23283005     0.37110999    -1.32635999
>> INPUT CARD>C     6.0     0.85118002     0.98475999    -2.47463012
>> INPUT CARD>H     1.0     1.09862006     1.20088005    -3.50523996
>> INPUT CARD>N     7.0    -0.33440000     0.21582000    -2.32835007
>> INPUT CARD>O     8.0    -1.62452996    -0.79584998    -0.71752000
>> INPUT CARD>H     1.0    -0.54900998    -0.49006999    -3.00898004
>> INPUT CARD>C     6.0     4.36014986    -0.72294998     0.65132999
>> INPUT CARD>H     1.0     4.84456015    -1.55017996     0.03438000
>> INPUT CARD>H     1.0     3.93372011    -1.16517997     1.61162996
>> INPUT CARD>H     1.0     5.14190006     0.06531000     0.91036999
>> INPUT CARD>H     1.0     4.03015995     0.44310001    -2.04832006
>> INPUT CARD>N     7.0     2.13144994    -0.03258000     0.58301002
>> INPUT CARD>C     6.0     1.29933000     0.67783999    -0.21716000
>> INPUT CARD>N     7.0    -0.08823000     0.88062000    -0.21032000
>> INPUT CARD>C     6.0    -0.76103002    -0.03120000    -1.03929996
>> INPUT CARD> \$END
>> INPUT CARD> \$VEC
>> INPUT CARD> 1  1-3.06278831E-05-5.48793857E-05 2.59810829E-04 1.41265811E-05-9.66989194E-06

>>The calculation stops after step 8 with the following error message, DAVIDSON DIAGONALIZATION DID NOT CONVERGE IN  300 ITERATIONS.  BYEBYE.

>>Could anyone tell me what I am doing wrong? Is it possible to average five states?

>>Thank you,

>>Anna Le
>>

Thu Jul 24 '14 8:08pm