Matching SCF energies in QChem and Gaussian

Questions
#1

I am showing the input files of calculation of water SCF energy using wB97XD/cc-pvDZ. I have also referred to the topic-

to check if I have missed anything.

===== Gaussian input ==============
#p wb97xd/cc-pvdz

test against qchem

0 1
H 0.000000 -0.115747 1.133769
H 0.000000 1.109931 -0.113383
O 0.000000 0.005817 -0.020386

====================================================
====== QChem input ===================================
$molecule
0 1
H 0.000000 -0.115747 1.133769
H 0.000000 1.109931 -0.113383
O 0.000000 0.005817 -0.020386
$end

$rem
EXCHANGE wb97x-d
BASIS cc-pvDZ
SCF_CONVERGENCE 8
XC_GRID 3
THRESH 16
MAX_SCF_CYCLES 600
SCF_ALGORITHM diis
MAX_CIS_CYCLES 50
MEM_STATIC 300
MEM_TOTAL 2000
SYMMETRY true
$end

The SCF energy from Gaussian is -76.3609300087 and
from QChem is -76.3604630209.

At present, the two energies differ from the fourth decimal, which is still non-negligible for me to proceed with other calculations. Can anyone suggest if there are other variables to be set in QCHEm for me to match Gaussian results?

Thanks,
Jayashree

#2

SG-3 grid is a pruned version of (99,590) with a different radial quadrature (“DE2”) as compared to Euler-Maclaurin (see: https://doi.org/10.1002/jcc.24761). Try XC_GRID = 000099000590 to get Euler-Maclaurin-Lebedev (99,590),

#3

Thanks very much for this suggestion. The energies now agree.