error on single point unrestricted dft calculation


Just Got Here
Hello,
I am running a single point on a rubrene molecule. The geometry was optimized in the neutral species, but I am running a single point on it with charge +1 and -1 and got some errors. I am not sure if it is a problem with SCF convergence in NWchem, or perhaps a problem with the machines I was running on?

This is the error of the -1 species
   convergence    iter        energy       DeltaE   RMS-Dens  Diis-err    time
 ---------------- ----- ----------------- --------- --------- ---------  ------
 d= 0,ls=0.0,diis     1 -12067.7714639427 -2.08D+04  1.71D-03  2.33D+00  2289.4
                                                     1.48D-03  2.33D+00
 WARNING: error on integrated density =  0.65D-07
 greater than required accuracy of  0.10D-07
 d= 0,ls=0.5,diis     2 -12068.0882149425 -3.17D-01  4.04D-04  9.08D-02  2461.7
                                                     3.93D-04  8.63D-02
 WARNING: error on integrated density =  0.66D-07
 greater than required accuracy of  0.10D-07
 d= 0,ls=0.5,diis     3 -12068.1351351350 -4.69D-02  1.31D-04  3.38D-02  2698.9
                                                     1.16D-04  3.48D-02
 WARNING: error on integrated density =  0.66D-07
 greater than required accuracy of  0.10D-07
 d= 0,ls=0.5,diis     4 -12068.1551019650 -2.00D-02  9.47D-05  8.08D-03  2901.0
                                                     8.79D-05  8.40D-03
 WARNING: error on integrated density =  0.66D-07
 greater than required accuracy of  0.10D-07
 d= 0,ls=0.5,diis     5 -12068.1650635412 -9.96D-03  7.32D-05  3.25D-03  3055.9
                                                     6.96D-05  3.35D-03
 WARNING: error on integrated density =  0.67D-07
 greater than required accuracy of  0.10D-07
 d= 0,ls=0.5,diis     6 -12068.1717681063 -6.70D-03  6.88D-05  1.72D-03  3294.5
                                                     6.59D-05  1.68D-03
 WARNING: error on integrated density =  0.67D-07
 greater than required accuracy of  0.10D-07
 d= 0,ls=0.5,diis     7 -12068.1771293039 -5.36D-03  5.42D-05  1.14D-03  3504.1
                                                     4.86D-05  1.03D-03
 WARNING: error on integrated density =  0.67D-07
 greater than required accuracy of  0.10D-07
 d= 0,ls=0.5,diis     8 -12068.1806017421 -3.47D-03  5.82D-05  7.56D-04  3699.2
                                                     5.40D-05  7.03D-04
 WARNING: error on integrated density =  0.67D-07
 greater than required accuracy of  0.10D-07
 d= 0,ls=0.5,diis     9 -12068.1835184527 -2.92D-03  4.37D-05  5.90D-04  3958.5
                                                     3.94D-05  5.63D-04
 WARNING: error on integrated density =  0.67D-07
 greater than required accuracy of  0.10D-07
 d= 0,ls=0.5,diis    10 -12068.1853915968 -1.87D-03  4.04D-05  4.15D-04  4163.4
                                                     3.68D-05  3.94D-04
0:Terminate signal was sent, status=: 15
(rank:0 hostname:node1193 pid:54957):ARMCI DASSERT fail. signaltrap.c:SigTermHandler():463 cond:0

Forum Vet
Yao
I don't see any evince of installation problems in your output, just a slow convergence.
If you diminish the level shift parameter, you might be able to converge faster; on the other hand, reducing the level
shift too much might cause convergence problems.

The following URL describes the usage of the lshift parameter
http://nwchemgit.github.io/index.php/Release62:Density_Functional_Theory_for_Molecules#CONVE...


Forum >> NWChem's corner >> Running NWChem