Author Topic: error in definite matrix  (Read 1123 times)

ztehrani

  • Jr. Member
  • **
  • Posts: 14
  • Karma: +0/-0
error in definite matrix
« on: May 12, 2017, 01:00:56 pm »
Dear All,
Hi. I am going to run single point of calculation with triplet spin state on my system with 89 atoms and 2 Cu... I got warning in eht step of define "!!! WARNING ! HOMO/LUMO-SEPARATION LOWER THAN 0.05 AU !!!" and also there is hole in MOs of system before running...
  Nr.   Orbital    Occupation       Energy
  402. b 201 a                     -0.442903 H =       -12.052 eV
  401. a 201 a        1.000        -0.442903 H =       -12.052 eV
  400. b 200 a                     -0.449641 H =       -12.235 eV
  399. a 200 a        1.000        -0.449641 H =       -12.235 eV
  398. b 199 a        1.000        -0.450577 H =       -12.261 eV
  397. a 199 a        1.000        -0.450577 H =       -12.261 eV
  396. b 198 a        1.000        -0.451467 H =       -12.285 eV
  395. a 198 a        1.000        -0.451467 H =       -12.285 eV
  394. b 197 a        1.000        -0.452852 H =       -12.323 eV
  393. a 197 a        1.000        -0.452852 H =       -12.323 eV
  392. b 196 a        1.000        -0.453479 H =       -12.340 eV
  391. a 196 a        1.000        -0.453479 H =       -12.340 eV
  390. b 195 a        1.000        -0.454336 H =       -12.363 eV
  389. a 195 a        1.000        -0.454336 H =       -12.363 eV
  388. b 194 a        1.000        -0.458756 H =       -12.483 eV
  387. a 194 a        1.000        -0.458756 H =       -12.483 eV
  386. b 193 a        1.000        -0.463334 H =       -12.608 eV
I changed orbitals order in alpa beta files and mos... checked again with eiger... it was correct but the job crashed at the beginning with this error:
 <geterg> : there is no data group $energy
  MOs are in ASCII format !
 reading orbital data $uhfmo_alpha  from file alpha .
 orbital characterization : expanded
 reading orbital data $uhfmo_beta  from file beta .
 orbital characterization : expanded
 ERROR: non-positive definite matrix in Cholesky decomposition (routine: usad)
    of MOs in cho_ortho.f

 MODTRACE: no modules on stack

 negative root in usad, see output
 ridft ended abnormally
any suggestion or help will be appreciated...
Cheers
Zahra