CCL: FCI/cc-pvQZ optimization



 Sent to CCL by: Grant Hill [hilljg*cardiff.ac.uk]
 Hi Ramachandran,
 
I can't really see the problem you describe in your output files. As I see it you have something like (apologies for the inevitable destruction of the formatting):
 Input_dist    Energy               Output_dist    Energy
 1.46              -2.97748298     1.45443278   -2.98182783
 1.69              -2.97164111     1.40123645   -17.20198531
 Perhaps I have misunderstood your question?
 
The results do raise some other questions though, such as why the energy is so low in the second case, but I'm no expert in FCI so I won't attempt to answer that.
 Regards,
 Grant
 
On 13 Aug 2008, at 14:54, Ramachandran Chelat rcchelat*| *rediffmail.com wrote:
 
 Sent to CCL by: "Ramachandran  Chelat" [rcchelat[A]rediffmail.com]
 Dear friends,
 Thanks to Prof. Peterson for his reply for my earlier posting.
 
About H2+ I agree that FCI calculation for 1 electron H2+ is meaningless.
 But the same problem happens for all other optimizations.
 
For example, below is the scaning results for HeH+ system at FCI/cc- pvQZ level using molpro.
 As you see, the scan results shows a minimum at 1.4634 au.
 
But when I optimize the geometry from two different distances (1.69 & 1.46) , it gives the energy of those two distances as the optimized geometry.
From the out file it is clear that optimization steps initiated properly.
 However, it prints the initial energy as the optimized results.
 
It seems that the convergence is not achieved although it prints the energy in the last. I am not getting the optimized structure either in the .log file or in .out file
 I am using the Molpro Version 2006.1
 I have attached the input and output files for your kind information
 ! $Revision: 2006.0 $
 ***,HeH+intermediate-example-FCI
 memory,50,m
 geometry={He;
 H1,He,r1(i)
 }
 
distances = [0.6,0.7,0.8,0.9,1.0,1.1,1.2,1.3,1.3885,1.4,1.4632,1.4634,1.5,1.6,1.7,1.8 , 1.9,1.931059,2.0,2.1,2.2,2.3,2.4,2.5,2.6,2.7,2.8,2.9,3.0,3.2,3.4,3.6,3.8,4.0,4.5,5
 .0,6.0,7.0,8.0,9.0,10.0,20.0]
 i=0                                   !initialize a counter
 do ir1=1,#distances                   !loop over distances for O-H1
 i=i+1                                 !increment counter
 r1(i)=distances(ir1)                  !save r1 for this geometry
 basis=cc-pvQz             !define basis
 {HF
 wf,charge=1,spin=0
 }
 fci;
 
efci(i)=energy !save scf energy for this geometr
 enddo
 {table,r1,efci   !produce a table with results
 head, r1,FCI    !modify column headers for table
 
save,HeHplus.tab !save the table in file HeHplus.tab
 title,Results for HeHplus, basis $basis   !title for table
 }
   .600000   -2.35001369
      .700000   -2.59653051
      .800000   -2.74877774
      .900000   -2.84392314
     1.000000   -2.90323595
     1.100000   -2.93950114
     1.200000   -2.96070893
     1.300000   -2.97200449
     1.388500   -2.97647841
     1.400000   -2.97677518
     1.463200   -2.97748524
     1.463400   -2.97748528
     1.500000   -2.97728912
     1.600000   -2.97508201
     1.700000   -2.97119995
     1.800000   -2.96635764
     1.900000   -2.96104361
     1.931059   -2.95935255
     2.000000   -2.95559048
     2.100000   -2.95022178
     2.200000   -2.94508373
     2.300000   -2.94026719
     2.400000   -2.93582336
     2.500000   -2.93177518
     2.600000   -2.92812571
     2.700000   -2.92486424
     2.800000   -2.92197097
     2.900000   -2.91942031
     3.000000   -2.91718352
     3.200000   -2.91353120
     3.400000   -2.91077953
     3.600000   -2.90871995
     3.800000   -2.90718198
     4.000000   -2.90603421
     4.500000   -2.90428686
     5.000000   -2.90345146
     6.000000   -2.90282074
     7.000000   -2.90261298
     8.000000   -2.90252412
     9.000000   -2.90248010
    10.000000   -2.90245590
    20.000000   -2.90241368
 
But when I optimize the geometry from two different distances, it gives the energy of those two distances as the optimized geometry.
From the out file it is clear that optimization steps initiated properly.
 However, it prints the initial energy as the optimized results.
 
It also seems that the convergence is not achieved although it prints the energy in the last. I am not getting the optimized structure either in the .log file or in .out file
 I am using the Molpro Version 2006.1
 input file:
 ***,HeH+ example-FCI
 memory,150,m
 geometry={He;
  H1,He,r1
 }
   r1=1.46
 basis=cc-pvQz
 !define basis
 {hf
 wf,charge=1,spin=0
 }
 fci
 optg
 
= = = ======================================================================
 part of the out put file for the distance 1.46 atomic unit
 
= = ======================================================================
 Optimization point 1
 
Variable Last Current Next Gradient Hessian
 E(FCI) / Hartree             .00000000    -2.97748298      .00000000
 
R1 / BOHR .00000000 1.46000000 1.45610632 .00128627 .33034831 Convergence: .00000000 (line search) . 00389368 .00128627 (total)
 Optimization point 2
 
Variable Last Current Next Gradient Hessian
 E(FCI) / Hartree           -2.97748298    -2.98028234      .00000000
 
R1 / BOHR 1.46000000 1.45610632 1.45440416 .00039126 .22986190 Convergence: .00000000 (line search) . 00170216 .00039126 (total)
 Optimization point 3
 
Variable Last Current Next Gradient Hessian
 E(FCI) / Hartree           -2.98028234    -2.98185605      .00000000
 
R1 / BOHR 1.45610632 1.45440416 1.45443278 -.00000669 .23379242 Convergence: .00000000 (line search) . 00002862 .00000669 (total)
 END OF GEOMETRY OPTIMIZATION.    TOTAL CPU:        11.8 SEC
 
ITER. ENERGY(OLD) ENERGY(NEW) DE GRADMAX GRADNORM GRADRMS STEPMAX STEPLEN STEPRMS 1 -2.97748298 -2.98028234 -.00279936 .00128627 . 00128627 .00128627 .00389368 .00389368 .00389368 2 -2.98028234 -2.98185605 -.00157371 .00039126 . 00039126 .00039126 .00170216 .00170216 .00170216 3 -2.98185605 -2.98182783 .00002822 .00000669 . 00000669 .00000669 .00002862 .00002862 .00002862
 Geometry written to block  1 of record 700
 **********************************************************************************************************************************
 DATASETS  * FILE   NREC   LENGTH (MB)   RECORD NAMES
 
1 18 2.68 500 610 700 900 950 970 1000 1100 1400 1410 VAR BASINP GEOM SYMINP ZMAT AOBASIS BASIS S T V 1200 1210 1080 1600 129 960 1650 1700 H0 H01 AOSYM SMH P2S ABASIS MOLCAS OPER
              2       3         .29       700     1000     2100
                                         GEOM     BASIS     RHF
 PROGRAMS   *        TOTAL       FCI       FCI        HF       INT
 CPU TIMES  *        14.29      2.47      1.88      0.04      0.46
 REAL TIME  *        14.70 SEC
 DISK USED  *        17.01 MB
 SF USED    *          .20 MB
 **********************************************************************************************************************************
        FCI             HF-SCF
     -2.97748298     -2.93287638
 
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = ======================================================================
 part of the out put file for the distance 1.69 atomic unit
 
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = ======================================================================
 Quadratic Steepest Descent - Minimum Search
 Optimization point 1
 
Variable Last Current Next Gradient Hessian
 E(FCI) / Hartree             .00000000    -2.97164111      .00000000
 
R1 / BOHR .00000000 1.69000000 1.41635627 .04380375 .16007582 Convergence: .00000000 (line search) . 27364373 .04380375 (total)
 Optimization point 2
 
Variable Last Current Next Gradient Hessian
 E(FCI) / Hartree           -2.97164111   -15.26931437      .00000000
 
R1 / BOHR 1.69000000 1.41635627 1.38167732 .00492688 .14207111 Convergence: .00000000 (line search) . 03467895 .00492688 (total)
 Optimization point 3
 
Variable Last Current Next Gradient Hessian
 E(FCI) / Hartree          -15.26931437   -19.86401961      .00000000
 
R1 / BOHR 1.41635627 1.38167732 1.40206356 -.00702739 .34471252 Convergence: .00000000 (line search) . 02038624 .00702739 (total)
 Optimization point 4
 
Variable Last Current Next Gradient Hessian
 E(FCI) / Hartree          -19.86401961   -17.09342828      .00000000
 
R1 / BOHR 1.38167732 1.40206356 1.40128096 .00028054 .35847378 Convergence: .00000000 (line search) . 00078260 .00028054 (total)
 Optimization point 5
 
Variable Last Current Next Gradient Hessian
 E(FCI) / Hartree          -17.09342828   -17.19613572      .00000000
 
R1 / BOHR 1.40206356 1.40128096 1.40123645 .00001510 .33918500 Convergence: .00000000 (line search) . 00004450 .00001510 (total)
 END OF GEOMETRY OPTIMIZATION.    TOTAL CPU:        20.8 SEC
 
ITER. ENERGY(OLD) ENERGY(NEW) DE GRADMAX GRADNORM GRADRMS STEPMAX STEPLEN STEPRM
 S
 
1 -2.97164111 -15.26931437 -12.29767326 .04380375 . 04380375 .04380375 .27364373 .27364373 .273643
 73
 
2 -15.26931437 -19.86401961 -4.59470524 .00492688 . 00492688 .00492688 .03467895 .03467895 .03467895 3 -19.86401961 -17.09342828 2.77059132 .00702739 . 00702739 .00702739 .02038624 .02038624 .02038624 4 -17.09342828 -17.19613572 -.10270743 .00028054 . 00028054 .00028054 .00078260 .00078260 .00078260 5 -17.19613572 -17.20198531 -.00584959 .00001510 . 00001510 .00001510 .00004450 .00004450 .00004450
 Geometry written to block  1 of record 700
 **********************************************************************************************************************************
 DATASETS  * FILE   NREC   LENGTH (MB)   RECORD NAMES
 
1 18 2.68 500 610 700 900 950 970 1000 1100 1400 1410 VAR BASINP GEOM SYMINP ZMAT AOBASIS BASIS S T V 1200 1210 1080 1600 129 960 1650 1700 H0 H01 AOSYM SMH P2S ABASIS MOLCAS OPER
              2       3         .29       700     1000     2100
                                         GEOM     BASIS     RHF
 PROGRAMS   *        TOTAL       FCI       FCI        HF       INT
 CPU TIMES  *        23.51      2.65      1.89      0.05      0.60
 REAL TIME  *        25.10 SEC
 DISK USED  *        17.03 MB
 SF USED    *          .54 MB
 **********************************************************************************************************************************
        FCI             HF-SCF
     -2.97164111     -2.92672974
 **********************************************************************************************************************************
 Variable memory released
 
-= This is automatically added to each message by the mailing script =- To recover the email address of the author of the message, please change> Conferences: http://server.ccl.net/chemistry/announcements/ conferences/
 
Search Messages: http://www.ccl.net/htdig (login: ccl, Password: search)>