[molpro-user] Error in SCF (dsyevd failure)

Ouyang Fengcong John john.ouyang at u.nus.edu
Tue Jan 13 15:27:58 GMT 2015


Dear Molpro Users,


I am running HF calculations on water clusters with the presence of dummy atoms and the SCF program just crashes with the following error:


 ITERATION    DDIFF          GRAD             ENERGY        2-EL.EN.            DIPOLE MOMENTS         DIIS   ORB.
    1      0.000D+00      0.000D+00       -76.00559895     77.730090   -1.04839    0.12985   -0.28672    0    start
    2      0.000D+00      0.733D-02       -76.03190302     74.280063   -0.64024    0.07957   -0.17535    1    diag
 dsyevd failed, info=       206

 If this happens in HF/KS, set HF option dsyevd=0.

 The default value for HF can be changed in hfopt.registry.

 If this happens elsewhere, set gparam,dsyevd=0.

 The global default value can be changed in common.F.
 ? Error
 ? dsyevd failure
 ? The problem occurs in diag2


I have tried setting dsyevd=0 in both SCF and gparam but it does not resolve the problem.
I have also attached the input file for your reference.

***, water
MEMORY,100,M

GEOMETRY={
12
water
 O       2.98672748     -1.74999213     -1.53386319
 H       2.42224336     -2.40248156     -1.91660845
 H       2.47200847     -0.96369874     -1.44619536
 O      -1.89167953      3.26841187      2.98635650
 H      -1.88890386      2.34728956      3.19227934
 H      -2.41837478      3.37121892      2.20988917
 O       0.51658607      3.96277165      2.33908081
 H      -0.35734880      3.83061814      2.67021775
 H       0.95181942      4.54078579      2.94517493
 O       3.40350318      1.89189637     -3.36418319
 H       4.01028299      1.17353439     -3.28261614
 H       3.31268454      2.26931143     -2.50384140
}

BASIS,AVDZ


DUMMY, 4, 5, 6, 7, 8, 9,10,11,12
{HF,ENERGY=1.d-10,ACCU=12};
MP2;

Many thanks in advance.

Regards,
John Ouyang
PhD Student, Chemistry
National University of Singapore
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20150113/a62e24ab/attachment.html>


More information about the Molpro-user mailing list