Search:

# Generation Of QRHF Orbitals

Running QRHF-CC calculations

There are two possible approaches for running QRHF-CC calculations. For these methods, energies can be obtained at any level of CC theory while gradients are limited to the CCSD level. Analytic second derivatives are not available. The first input option involves the use of positive and negative integer arguments to the QRHF_GENERAL keyword, and is described in the keyword input section of this manual. However, this approach can be a nuisance in some instances, particularly when one wishes to calculate vibrational frequencies by finite difference as problems arise in getting the right state at lower symmetries. To alleviate this problem and make the input simpler in general, another input option is available. This newer approach is recommended for all QRHF calculations and is described below. A perusal of the descriptions of the QRHF keywords in the keyword section is useful, although attention should be paid only to the slash delimiters that are used when the closed-shell reference determinant and final state reference determinant differ in the occupation number of two or more orbitals.

The procedure that should be followed is:

a. Set QRHF_G parameters to ZERO. b. Set QRHF_O parameters to the absolute orbital offset. c. Set QRHF_S parameters to the appropriate values.

The code now analyzes the eigenvalue vector and converts the absolute orbital offset (which runs from 1 to the number of basis functions) to the relative offset within the symmetry block (the value of QRHF_O if QRHF_G was set to that particular IRREP) and also figures out if the orbital was occupied or unoccupied in the closed-shell reference. If the QRHF_S keyword is omitted, the value is set to 2 automatically if the orbital was occupied in the reference and to 1 if the orbital was unoccupied.

An example should make the input clear. For DZ water, the closed-shell eigenvalues are as follows:

    1     1         -20.5626573732        -559.5429901106      A1        A1 (1)
2     2          -1.3344673703         -36.3130041514      A1        A1 (1)
3    11           -.7025649081         -19.1179214961      B1        B1 (3)
4     3           -.5503201546         -14.9750968089      A1
5     9           -.5012216770         -13.6390482392      B2        B2 (2)
6     4            .2063804242           5.6159433857      A1        A1 (1)
7    12            .2980294496           8.1098608204      B1        B1 (3)
8    10            .8676455317          23.6100308614      B2        B2 (2)
9     6            .8962711236          24.3889792707      A1        A1 (1)
10    13            .9524138412          25.9167129446      B1        B1 (3)
11     5           1.1454741226          31.1701938136      A1        A1 (1)
12    14           1.1994450316          32.6388290803      B1        B1 (3)
13     7           1.6475850033          44.8334387180      A1        A1 (1)
14     8          43.3199085654        1178.8044088906      A1        A1 (1)


The following QRHF inputs are equivalent:

• CFOUR(CALC=CCSD,BASIS=DZ,QRHF_G=1/1,QRHF_O=1/3)
• CFOUR(CALC=CCSD,BASIS=DZ,QRHF_G=0/0,QRHF_O=6/11)
Page last modified on April 02, 2011, at 08:04 PM
This page has been visited 549 times since April 2020.
CFOUR is partially supported by the U.S. National Science Foundation.