Change in energy in electric field is not what I would expect?

Questions
#1

I am trying to look into the change in energy for molecules in an electric field. Below I have included my input for hydroxide where I apply a dipole field in the +x and -x directions. I also don’t apply a field. I am confused, however, because the dipole moment of OH- points along +x (in the chosen orientation), so I would expect a positive field to lower the energy of the system. The calculation indicates the opposite though. A negative field lowers the energy of the system and a positive field raises the energy of the system. Can someone provide me some clarity on this? I can’t find anything in the manual about this, so I assume there is some implicit sign convention I am not understanding. Is that true?
Thanks.

$molecule
-1 1
O 0.0 0.0 0.0
H 0.9591126269 0.0 0.0
$end

$rem
mem_total 16000
ideriv 1
incdft 0
incfock 0
jobtype sp
method wB97X-V
unrestricted false
basis def2-qzvppd
scf_algorithm gdm
scf_max_cycles 500
scf_guess sad
scf_convergence 8
thresh 14
symmetry 0
sym_ignore 1
gen_scfman true
gen_scfman_final true
internal_stability false
complex false
chelpg true
$end

$multipole_field
X -0.001
Y 0.0
Z 0.0
$end

@@@

$molecule
-1 1
O 0.0 0.0 0.0
H 0.9591126269 0.0 0.0
$end

$rem
mem_total 16000
ideriv 1
incdft 0
incfock 0
jobtype sp
method wB97X-V
unrestricted false
basis def2-qzvppd
scf_algorithm gdm
scf_max_cycles 500
scf_guess sad
scf_convergence 8
thresh 14
symmetry 0
sym_ignore 1
gen_scfman true
gen_scfman_final true
internal_stability false
complex false
chelpg true
$end

$multipole_field
X 0.001
Y 0.0
Z 0.0
$end

@@@

$molecule
-1 1
O 0.0 0.0 0.0
H 0.9591126269 0.0 0.0
$end

$rem
mem_total 16000
ideriv 1
incdft 0
incfock 0
jobtype sp
method wB97X-V
unrestricted false
basis def2-qzvppd
scf_algorithm gdm
scf_max_cycles 500
scf_guess sad
scf_convergence 8
thresh 14
symmetry 0
sym_ignore 1
gen_scfman true
gen_scfman_final true
internal_stability false
complex false
chelpg true
$end

Meaning of quadrupole field components in $multipole_field
#2

Hello,

This was a known historical issue. When Q-Chem was created, it adopted a convention in which the field direction goes from negative to positive charge, which is opposite to the most common definition on physics textbooks. So you will just need to add a minus sign for your field.

Apparently this issue is not unique to Q-Chem. According to this paper by Shaik and co-workers https://pubs.acs.org/doi/10.1021/jacs.0c05128, the authors found the same in Gaussian 09 and referred to it as the “Gaussian convention”.

Best,
Yuezhi

1 Like
#3

Q-Chem manual has been updated to stipulate this convention. Will appear in v. 6.1.