Dear siesta users,I'm now puzzled with the results by siesta in calculating
the interaction energy of benzene and molecule hydrogen.According to a
reference(Fabien Tran et al,J. Phys. Chem. B 2002, 106, 8689-8696),they used
the Gaussian packages,and got the results much agreed with the experiment
datas:when the orientation of the H2 axis is parallel to X axis ,the
interation energy is 0.42 kcal/mol,and when the orientation is parallel to Z
axis,the interaction energy is 0.72 kcal/mol.But I tried with siesta (the
parameters has been kept the same as possible as I can ),the former results
is 3.7997kcal/mol,and the latter is becoming to 4.252kcal/mol.Both of the
calculation has a big difference after I checked all the possible reasons.I
could't understand why siesta can't make an accurate results.So I want to
know is there any possible reason about the precision of siesta?
*the following is an input file of two examples:*
SystemName Ben+h2_z
SystemLabel ben+h2_z
NumberOfAtoms 14
NumberOfSpecies 2
%block ChemicalSpeciesLabel
1 6 C
2 1 H
%endblock ChemicalSpeciesLabel
#%block PAO.BasisSizes
# O DZP
# Zn DZP
# C DZP
# H DZP
#%endblock PAO.BasisSizes
%block PAO.Basis
C 2
n=2 0 3
4.088 3.347 3.002
1.000 1.000 1.000
n=2 1 3 P 1
4.870 3.347 3.002
1.000 1.000 1.000
H 1
n=1 0 3 P 1
4.593 3.713 3.002
1.000 1.000 1.000
%endblock PAO.Basis
#PAO.BasisSize DZP
PAO.EnergyShift 100 meV
PAO.BasisType split
LatticeConstant 15.000 Ang
#%block LatticeParameters
# 2.460 2.460 6.800 90.000 90.000 120.000
#%endblock LatticeParameters
%block LatticeVectors
1.000 0.000 0.000
0.000 1.000 0.000
0.000 0.000 1.000
%endblock LatticeVectors
#Kgrid_cutoff 5.00000 Ang
%block kgrid_Monkhorst_Pack
1 0 0 0.0
0 1 0 0.0
0 0 1 0.0
%endblock kgrid_Monkhorst_Pack
# %block BandLines
# 1 0.000 0.000 -1.000
# 100 0.000 0.000 1.000
# %endblock BandLines
#%block GeometryConstraints
# position from 1 to 424
#%endblock GeometryConstraints
#SPIN options
XC.functional GGA
XC.authors PBE
NetCharge 0
SpinPolarized F
FixSpin F
TotalSpin 0.0
MeshCutoff 300.0 Ry
FilterCutoff 100. Ry
# SCF options
#%block DM.InitSpin
#%endblock DM.InitSpin
MaxSCFIterations 100 # Maximum number of SCF iter
DM.MixingWeight 0.10 # New DM amount for next SCF cycle
DM.NumberPulay 10
DM.Tolerance 1.d-4 # Tolerance in maximum difference between
input and output DM
DM.UseSaveDM T # to use continuation files
DM.MixSCF1 F
DM.PulayOnFile F # Store in memory ('F') or in files ('T')
SolutionMethod diagon # OrderN or Diagon
ElectronicTemperature 100.0 K #Temp. for Fermi smearing
# MD options
MD.TypeOfRun cg
MD.NumCGsteps 150
MD.MaxCGDispl 0.1 Ang
MD.MaxForceTol 0.01 eV/Ang
MD.VariableCell F
MD.MaxStressTol 0.05 GPa
#MD.TypeOfRun fc
#MD.FCDispl 0.04 Ang
#MD.FCfirst 1
#MD.FClast 200
AtomicCoordinatesFormat Ang
%block AtomicCoordinatesAndAtomicSpecies < ben+h2_z.xyz
%endblock AtomicCoordinatesAndAtomicSpecies
# OUTPUT options
WriteCoorInitial T
WriteCoorStep T
WriteCoorXmol T
WriteForces T
WriteEigenvalues T # If .false., it writes them in the
file Systemlabel.EIG
WriteMullikenPop 1 # Write Mulliken Population Analysis
#WriteKpoints T
#WriteKbands T
#WriteBands T
WriteMDCoorXmol F
WriteMDhistory F
# options for saving or reading information
MD.UseSaveZM F # Use stored positions and velocities
MD.UseSaveCG F # Use stored positions and velocities
SaveRho T # Write valence pseudocharge at the mesh
#SaveDeltaRho T # Write RHOscf-RHOatm at the mesh
#SaveElectrostaticPotential T # Write the total elect. pot. at the mesh
# (local pseudopotential + Hartree)
SaveHS T
SaveTotalPotential T # write the valence total effective local
potential
# (local pseudopotential + Hartree + Vxc)
#WriteSiestaDim T # Write minimum dim to siesta.h and stop
#WriteDenchar T # Write information for DENCHAR
