Dear Jonas, Henry and Els,

Thank you for the quick replies!

Actually, I confused the matter; the small U radius was one I tried
just as an experiment. However, the problem with the UO2(H2O)4+2
complex also occurs when the U radius is much larger (e.g., 2.5 Å).
For example, using the radii suggested by Jonas Haller, and the
following COSMO input, I still get a a poor geometry (U-O = 1.92 Å,
U-OH2=2.5:

SOLVATION
    Solvent epsilon=78.8 radius=1.4
    SurfaceType esurf
    DivisionLevel ND=3 min=0.5 Ofac=0.8
    ChargeUpdate Method=Gauss-Seidel
    DiscAttributes SCale=0.01 LEGendre=10 TOLerance=1.0d-2
    SCF PERT
    C-Matrix Exact
    Radii
      U=2.05
      O=1.576
      H 1.172
    Subend
END

Just in case I'm doing something stupid, here is the entire input
file. Again, without the COSMO solvation, I get a very good
prediction of U-O bond lengths.

#! /bin/sh
$ADFBIN/dirac -n1 < $ADFRESOURCES/Dirac/U.5d
$ADFBIN/dirac -n1 < $ADFRESOURCES/Dirac/O.1s
$ADFBIN/dirac -n1 < $ADFRESOURCES/Dirac/H
mv TAPE12 t12.rel

$ADFBIN/adf -n1 << eor
create U $ADFRESOURCES/ZORA/TZP/U.5d
relativistic scaler ZORA
corepotentials t12.rel &
U 1
end
end input
eor

mv TAPE21 t21.U

$ADFBIN/adf -n1 << eor
create O $ADFRESOURCES/ZORA/TZP/O.1s
relativistic scaler ZORA
corepotentials t12.rel &
O 2
end
end input
eor

mv TAPE21 t21.O

$ADFBIN/adf -n1 << eor
create H $ADFRESOURCES/ZORA/TZP/H
relativistic scaler ZORA
corepotentials t12.rel &
H 3
end
end input
eor

mv TAPE21 t21.H

$ADFBIN/adf << eor
Title UO2(H2O)4

relativistic scaler ZORA
corepotentials t12.rel &
U 1
O 2
H 3
end

UNITS
LENGTH angstrom
ANGLE degree
END
ATOMS
U 0. 0. 0.
O 0. 0. 1.8
O 0. 0. -1.8
O 2.4 0. 0.
H 3.21 0. 0.577
H 3.21 0. -0.577
O -2.4 0. 0.
H -3.21 0. 0.577
H -3.21 0. -0.577
O 0. 2.4 0.
H 0. 3.21 0.577
H 0. 3.21 -0.577
O 0. -2.4 0.
H 0. -3.21 0.577
H 0. -3.21 -0.577
END
FRAGMENTS
U t21.U
O t21.O
H t21.H
END
SOLVATION
    Solvent epsilon=78.8 radius=1.4
    SurfaceType esurf
    DivisionLevel ND=3 min=0.5 Ofac=0.8
    ChargeUpdate Method=Gauss-Seidel
    DiscAttributes SCale=0.01 LEGendre=10 TOLerance=1.0d-2
    SCF PERT
    C-Matrix Exact
    Radii
      U=2.05
      O=1.576
      H 1.172
    Subend
END
SYMMETRY NOSYM
SYMMETRY NOSYM
Charge 2.0
XC
LDA SCF VWN
GGA SCF PBE
end
Geometry
End
End Input
eor

rm TAPE21 logfile

David M. Sherman
Professor of Geochemistry
Department of Earth Sciences
University of Bristol
Bristol BS8 1RJ UNITED KINGDOM
Phone: 44-(0)117-954-5446
http://mineral.gly.bris.ac.uk
Received on 2006-03-01 14:39:03