Re: RASSI complaining with LU partitioning


[ Molcas user's WWWBoard ]

Posted by Vérot on March 20, 2015 at 12:18:15:

In Reply to: RASSI complaining with LU partitioning posted by Vérot on March 20, 2015 at 09:29:33:

: I'm doing some calculations for nickelocene and I am computing the zero field splitting with the 45 states generated by the d8 in the 3d orbitals.

: With a CAS(8,10) and at the CASSCF or CASPT2 level, everything works fine but when I enlarge the CAS space to a CAS (16,9) I have teh following message from RASSI :
: Nr of states: 25

: State: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
: JobIph: 1 2 2 2 3 3 3 4 4 4 5 5 5 5 5 5 6 6 6 7
: Root nr: 1 1 2 3 1 2 3 1 2 3 1 2 3 4 5 6 1 2 3 1

: State: 21 22 23 24 25
: JobIph: 7 7 8 8 8
: Root nr: 2 3 1 2 3


: RASSI CANNOT CONTINUE. THE PROBLEM AT HAND
: IS PROBABLY NOT SOLUBLE. THE TWO ORBITAL
: SPACES ARE TOO DISSIMILAR.
: LU PARTITIONING IS TROUBLESOME. DIAGONAL
: ELEMENT NR. 13 IS TOO SMALL:
: IN MATRIX CXA IT IS 1.00000000000000
: IN MATRIX CYB IT IS 1.596537582592096E-004
: EVEN AFTER OPTIMAL PIVOT-TRANSFORMATION.

:
: I checked my orbitals, it seems ok (the d orbitals are in the CAS) but I don't know how to solve this problem. It points to the element 13, is it because it is my thirteenth state (among the 25) which is causing trouble ?

: If anyone has any hint on how to solve this problem, I would be truly grateful.


Here is the input :
&SEWARD &END
title
Nicp.xyz
Symmetry
Z X
basis set
C.ano-rcc.Roos.14s9p4d3f2g.3s2p1d.
C2 -0.000000 1.217907 1.810490 Angstrom
C5 0.715868 -0.985307 1.810490 Angstrom
C6 1.158298 0.376354 1.810490 Angstrom
end of basis
basis set
Ni.ano-rcc.Roos.21s15p10d6f4g2h.5s4p2d1f.
Ni1 -0.000000 -0.000000 0.000000 Angstrom
end of basis
basis set
H.ano-rcc.Widmark.8s4p3d1f.2s1p.
H7 -0.000000 2.305577 1.794018 Angstrom
H9 1.355185 -1.865251 1.794018 Angstrom
H11 -2.192735 0.712462 1.794018 Angstrom
end of basis
end of input

&RASSI &END
NROF jobiphs
8 1 3 3 3 6 3 3 3
1
1 2 3
1 2 3
1 2 3
1 2 3 4 5 6
1 2 3
1 2 3
1 2 3
SPIN

I did not use cholesky decomposition in any calculation.


And here is an example of the eight calculations I did previously (the only differences are the spin and the symmetry)
&SEWARD &END
title
Nicp.xyz
Symmetry
Z X
basis set
C.ano-rcc.Roos.14s9p4d3f2g.3s2p1d.
C2 -0.000000 1.217907 1.810490 Angstrom
C5 0.715868 -0.985307 1.810490 Angstrom
C6 1.158298 0.376354 1.810490 Angstrom
end of basis
basis set
Ni.ano-rcc.Roos.21s15p10d6f4g2h.5s4p2d1f.
Ni1 -0.000000 -0.000000 0.000000 Angstrom
end of basis
basis set
H.ano-rcc.Widmark.8s4p3d1f.2s1p.
H7 -0.000000 2.305577 1.794018 Angstrom
H9 1.355185 -1.865251 1.794018 Angstrom
H11 -2.192735 0.712462 1.794018 Angstrom
end of basis
end of input

&RASSCF &END
Title
Nicp
Symmetry
1
Spin
1
nactEl
16 0 0
Inactive
15 12 8 6
Ras2
5 3 3 3
CIRoot=6 6;1 2 3 4 5 6; 10 1 1 1 1 1
ORBListing
ALL
ORBAppear
COMPACT
Lumorb
End of input

&CASPT2 &END
IPEAshift
0
IMAGINARY
0.15
Multistate
6 1 2 3 4 5 6
End of Input



Follow Ups:



Post a Followup

Name:
E-Mail:

Subject:

if B is 1s22s22p1, what is Li?

Passfield:

Comments:


[ Follow Ups ] [ Post Followup ] [ Molcas user's WWWBoard ]