35.6 Options

There are many options available, but these are hardly needed. Normally, when standard orbital basis sets such as aug-cc-PVTZ or VTZ-F12 are used, appropriate defaults are used and no further options are needed. A typical input simply reads


We recommend to specify options only when necessary and when it is well understood what they mean!

Options for canonical and local versions:

Select the basis for density fitting (see section 15 for details). basis can either refer to a set name defined in the basis block, or to a default MP2 fitting basis (e.g., DF_BASIS=VTZ generates the VTZ/MP2FIT basis). By default, the MP2FIT basis that corresponds to the orbital basis is used.
Select the density fitting basis for computing the exchange and Fock operators. By default, the JKFIT basis sets which correspond to the orbital basis are used.
Select the basis for the resolution of the identity (RI). This can refer to a default basis set or a set name defined in a basis block. For F12 methods the Hartree-Fock JKFIT basis sets perform well for the RI, despite having been optimized for other purposes. These sets are used by default for the AVnZ orbital basis sets. The basis type can be appended to the basis name after a slash, e.g.
RI_BASIS=AVQZ/JKFIT would use the specified JKFIT set, or RI_BASIS=AVQZ/OPTRI would use the optimized CABS basis sets of Peterson et al. These are recommended for the AVnZ and VnZ-F12 basis sets and used by default for the latter ones. Note that each OPTRI basis is associated to a specific orbital basis. Therefore, the name of the OPTRI basis must either be the same as that of the orbital basis, or be omitted, e.g., RI_BASIS=OPTRI selects automatically the correct set for the current orbital basis.

In case of R12-methods (which are not recommended to be used), the RI basis should be chosen to be a large uncontracted AO basis (at least AVQZ). Contraction/uncontraction can be forced appending (CONTRACT) or (UNCONTRACT) to the basis name, e.g.,
RI_BASIS=AVQZ(UNCONTRACT)/ORBITAL. If other options are given in parenthesis, these can be separated by commas, e.g.,
Alternative forms, which should work as well, are
Note that the CONTRACT/UNCONTRACT option cannot be used with basis set names previously defined in a basis block.

Can be used to change the default type for the RI basis, e.g. CONTEXT=OPTRI will use the OPTRI basis sets that correspond to the VnZ-F12 or AVnZ basis sets.
Select the explicitly correlated ansatz ansatz methods. See section 35.7 for the possibilities and further details.
Basis set name for geminal expansion; atom labels are ignored. This can either be OPTFULL (full nonlinear fit of the geminal expansion), EVEN (even tempered fit), or refer to a set name defined in a previous BASIS block. Default is OPTFULL.
Frozen geminal type: LINEAR or SLATER, default is SLATER.
Number of Gaussian geminal functions (default 6).
Centre of even tempered geminal exponents, if GEM_BASIS=EVEN (default 1.0).
Ratio of even tempered geminal exponents, if GEM_BASIS=EVEN (default 3.0).
Exponent for Slater-type frozen geminal, or parameter for weight function in other frozen geminal models (default 1.0 $a_0^{-1}$). It is possible to specify extra exponents for core-core and core-valence correlation. If two values are given (in square brackets), the first is used for valence pairs, the second for core-core (cc) and core-valence (cv) pairs. If three values are given, the first is used for vv, the second for cv, and the third for cc correlation.
Exponent for weighting function (default -1, which means a value derived from GEM_BETA).
Exponent for r in omega fitting (default 0).
Exponent for r in weighting function (default 0).
Max. number of iterations in geminal optimization (default 200).
Print parameter for geminal optimization (default 0).
Debug option for geminal optimization (default 0).
Convergence threshold for geminal line search (default 0.001).
Scaling factor for exponents in geminal optimization (default 1.0).
Geminal optimization method (augmented Hessian (AH) or Newton-Raphson (NR), default AH).
Trust ratio in AH geminal optimization (default 0.4).
Hessian shift in AH geminal optimization (default 0).
Flags numerical integration in geminal optimization (default 0).
Geminal plot file (default blank).
If nonzero (default), fit each geminal independently to Gaussians (if several exponents are used). If zero, the first exponent is fitted, unless GEM_BETA_OPT is specified.
Exponent used to fit the Gaussian expansion.
Only for calculation with multiple exponents: if nonzero, loop externally over integral program for different exponents. This is implied automatically if each geminal is fitted independently. If the same Gaussian exponents are used for each Slater exponent, SIM_MULTGEM=0 can be used (default). In this case the integral program handles the general contractions (slightly faster).
Select output level:
Standard output
Standard output plus more detailed information about integral evaluations.
Debugging output
Threshold below which non-physical eigenvalues are projected from approximate B matrices
Threshold for integral screening contribution.

molpro@molpro.net 2019-06-16