Deletions via Orca

Deletions via Orca

Postby Top_Cat » Mon Mar 19, 2018 10:32 am

Howdy guys,

I've been running various analyses via Orca and am wondering about timings required to do either LEWIS or NEDA deletions. I've added entries into the Orca input file for deletions like so:

Code: Select all
%nbo
 delkeylist="$del lewis $end"
end


And all seems to run normally until the energy evaluation steps after printing the occupancy table.

Code: Select all
Orbital occupancies:

Orbital                         No deletions  This deletion    Change
------------------------------------------------------------------------------
1. CR ( 1) N  1                      2.00000        2.00000       0.00000
2. CR ( 1) N  5                      2.00000        2.00000       0.00000
3. LP ( 1) N  1                      1.99711        2.00000       0.00289

etc.

NEXT STEP:  Perform one SCF cycle to evaluate the energy of the new density
             matrix constructed from the deleted NBO Fock matrix.

------------------------------------------------------------------------------

Copying NBO density...
Calculating new Fock-Matrix...
Calculating Fock-Matrix...


The job is seemingly stopping there.

I'm running it on our SGE cluster and my script is set up to run the job from our /scratch directory. All indications are that the job appears to be running but, as printing the standard NBO output is so fast, I guess it struck me as unusual that this has now been sitting there for over an hour for a relatively small test case (Ammonia dimer). Is the job stuck because, for example, an option to copy the file to /scratch is missing perhaps and it's just hanging? Or does this calculation just take quite a long time and I should be patient? Appreciate any guidance, thanks very much.
Top_Cat
 
Posts: 5
Joined: Mon Mar 19, 2018 10:04 am

Re: Deletions via Orca

Postby ericg » Tue Mar 20, 2018 7:07 am

Deletions only take about as much time as one SCF iteration, so you shouldn't have to wait long for a result. Can you post a copy of the input for a calculation that fails for you?

Eric
ericg
 
Posts: 296
Joined: Sat Dec 29, 2012 9:31 am

Re: Deletions via Orca

Postby Top_Cat » Tue Mar 20, 2018 7:45 am

Yeah I thought as much. Input for Orca is as follows:

Code: Select all
! RI-SCS-MP2 def2-TZVP NormalPrint Grid4 RijCosX GridX4 NormalSCF nbo opt
%nbo
 delkeylist="$del lewis $end"
end
%scf
        MaxIter 200
end
%maxcore 4000
%output
  PrintLevel=Normal
  Print[ P_Basis       ] 2
  Print[ P_GuessOrb    ] 1
  Print[ P_MOs         ] 1
  Print[ P_Density     ] 1
  Print[ P_SpinDensity ] 1
end
%pal
        nprocs 16
end
* xyz 0 1
 N   -1.578718000000   -0.046611000000    0.000000000000
 H   -2.158621000000    0.136396000000   -0.809565000000
 H   -2.158621000000    0.136396000000    0.809565000000
 H   -0.849471000000    0.658193000000    0.000000000000
 N    1.578718000000    0.046611000000    0.000000000000
 H    2.158621000000   -0.136396000000   -0.809565000000
 H    0.849471000000   -0.658193000000    0.000000000000
 H    2.158621000000   -0.136396000000    0.809565000000
*


The example I used in the original post was from another molecule. I tried to attach the output for this one but the 256Kb limit stopped me (roughly double that). Probably not needed I guess? Anyway, thanks for your help.
Top_Cat
 
Posts: 5
Joined: Mon Mar 19, 2018 10:04 am

Re: Deletions via Orca

Postby ericg » Thu Mar 22, 2018 7:32 am

The calculation doesn't fail for me. I'm using Orca 4.0.1 and NBO 6.0.18 (11-Mar-2018).

The calculation is an MP2 optimization. NBO analysis is performed three times at each geometry, first on the SCF density, second on the unrelaxed MP2 density, and finally on the relaxed MP2 density. Deletions can only be performed on the SCF density, and I get the following for the first geometry:

Code: Select all
 LEWIS: Delete all non-Lewis NBOs
 Deletion of the following orbitals from the NBO Fock matrix:
   11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30
   31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50
   51  52  53  54  55  56  57  58  59  60  61  62  63  64  65  66  67  68  69  70
   71  72  73  74  75  76  77  78  79  80  81  82  83  84  85  86  87  88  89  90
   91  92  93  94  95  96  97  98

 Orbital occupancies:

       Orbital                         No deletions  This deletion    Change
 ------------------------------------------------------------------------------
    1. CR ( 1) N  1                      2.00000        2.00000       0.00000
    2. CR ( 1) N  5                      2.00000        2.00000       0.00000
    3. LP ( 1) N  1                      1.99711        2.00000       0.00289
    4. LP ( 1) N  5                      1.99711        2.00000       0.00289
.
.
.
   96. RY ( 3) H  8                      0.00004        0.00000      -0.00004
   97. RY ( 4) H  8                      0.00001        0.00000      -0.00001
   98. RY ( 5) H  8                      0.00000        0.00000       0.00000

 NEXT STEP:  Perform one SCF cycle to evaluate the energy of the new density
             matrix constructed from the deleted NBO Fock matrix.

 ------------------------------------------------------------------------------
Copying NBO density...
Calculating new Fock-Matrix...
Calculating Fock-Matrix...done!

New NBO energy via Fock-Matrix: -112.427666


Starting NBO again for $del/return energy instructions...

 ------------------------------------------------------------------------------
   Energy of deletion :       -112.427665763
     Total SCF energy :       -112.439606604
                          -------------------
        Energy change :          0.011941 a.u.,           7.493 kcal/mol
 ------------------------------------------------------------------------------
Starting NBO again for $del instructions...

Storing NBOs: topcat.nbo

                                *** returned from  NBO  program ***
ericg
 
Posts: 296
Joined: Sat Dec 29, 2012 9:31 am

Re: Deletions via Orca

Postby Top_Cat » Thu Mar 22, 2018 10:39 am

Hi Eric,

According to Ute from the Orca team, jobs hang on parallel runs but the problem is fixed for the next update. When I ran the same job serially it certainly worked and I got a result similar to yours. Appreciate you having a look into this.
Top_Cat
 
Posts: 5
Joined: Mon Mar 19, 2018 10:04 am


Return to General NBO Discussion

Who is online

Users browsing this forum: No registered users and 1 guest

cron