Gradient and Hessian printing on S0

Questions
#1

Hi, I would like to know whether it is possible or not to run a frequency calculation, or force calculation, and printing both the gradient and hessian for a ground state of singlet simmetry. It turns out that I can print the gradient with “force” as jobtype, and the hessian if I do a “frequency” job, but I wonder if it is possible to have both.

Thank you again for your support and help

#2

I don’t think so, because for frequency jobs Q-Chem does not test that the point is a minimum (which allows one to compute the Hessian at arbitrary points on the potential surface), but this means there’s no need to evaluate the gradient in order to compute the Hessian. That said, cost of JOBTYPE=FORCE is much smaller than JOBTYPE=FREQ (certainly for large systems). I would make a compound input, FORCE calculation first then use SCF_GUESS=READ for the 2nd job so there’s very little additional overhead.

1 Like
#3

Here’s a simple example that demonstrates @jherbert 's idea:

$molecule
0 1
  O   -3.0324244    1.8346984   -0.0000000
  H   -2.0624244    1.8346984   -0.0000000
  H   -3.3557542    2.4883633   -0.6395937
$end

$rem
JOBTYPE FORCE
METHOD B3LYP
BASIS 6-31G*
$end

@@@

$molecule
READ
$end

$rem
JOBTYPE FREQ
METHOD B3LYP
BASIS 6-31G*
SCF_GUESS READ
SCF_MAX_CYCLES 0
$end
1 Like