Author Topic: Weight Derivatives in Gradient Calculation (For NumForce)  (Read 4703 times)


  • Jr. Member
  • **
  • Posts: 19
  • Karma: +0/-0
Weight Derivatives in Gradient Calculation (For NumForce)
« on: March 02, 2022, 01:41:13 PM »
Dear Users,

After having some trouble with NumForce I realised it would be more accurate to add weight derivatives to $dft in my control file as well as run NumForce with the -central tag. See page 276 of the Turbomole 7.3 manual:

4. Numforce additionally requires the file gradient and will not work, if the calculation
is not done at a stationary point of the molecular total energy. For reliable results,
always use Numforce with the option -central (i.e. central differences) and be aware
of effects due to the step length (option -d real;, default value is 0.02 a.u.). It is
strongly recommended to use Numforce in DFT calculations only with the option
weight derivatives in $dft, since this provides more accurate gradients and thus
frequencies, see Section 22.2.12

This did help my situation but I cannot find any details about how these change the calculation. Is there a paper about Turbomole explaining these options? Could someone give a brief explanation for why adding weight derivatives improves the accuracy of my gradient?