示例:NMR自旋旋转耦合常数:C2H2

Download CPL_C2H2.run

#!/bin/sh

# Non-relativistic calculation
# ============================

# A calculation of NMR nuclear spin-spin coupling constants (NSCCs).

# As explained in the ADF manual, the quality of a calculation for spin-spin
# coupling constants, using the program 'CPL', depends largely on the preceding
# ADF calculation, which produces the Kohn-Sham orbitals and orbital energies,
# used as a starting point.

# One of the quality-determining factors is the chosen basis set. It should be
# sufficiently flexible near the nucleus. Although the all-electron basis TZ2P
# is chosen in this example, it is recommendable to add more functions to the
# basis sets near the nucleus in case of heavy elements. One could start from a
# ZORA/QZ4P basis for example.

# The NOSYM symmetry currently needs to be specified in ADF to enable the CPL
# program to work correctly.


$ADFBIN/adf <<eor
  TITLE C2H2 nrel
  BASIS
    Type TZ2P
    Core None
  END
  ATOMS  
    C 0.0  0.0  0.0
    C 0.0  0.0  1.20692
    H 0.0  0.0  2.26672
    H 0.0  0.0 -1.0598
  END
  XC
    GGA Becke Perdew
  END
  BeckeGrid
    Quality good
  end
  SYMMETRY NOSYM
  SAVE TAPE10
eor
rm logfile
cp TAPE10 t10
cp TAPE21 t21


# The CPL program can run in parallel.

# The specification of what needs to be calculated is given in the nmrcoupling
# block key.

# coupling without the SD term
$ADFBIN/cpl <<eor
  nmrcoupling
    dso
    pso
    scf converge=1e-7
    nuclei 1 2 3 4 
    nuclei 3 4
  end
eor
rm logfile
cp t10 TAPE10
cp t21 TAPE21


# In this first example, the SD subkey is left out, as this would lead to a very
# strong increase in the required CPU time. The SD subkey is included in the
# second CPL run. That subkey controls the calculation of the so-called spin-
# dipole term.

# The subkeys dso and pso specify that, respectively, the diamagnetic and
# paramagnetic orbital terms will be calculated. The often dominant Fermi
# contact term (FC) is calculated by default and therefore does not have to be
# specified explicitly.

# The scf convergence subkey, in this context, refers to the convergence for the
# solution of the coupled-perturbed Kohn-sham equations which need to be solved
# to obtain to spin-spin couplings.

# The following lines
#   nuclei 1 2 3 4
#   nuclei 3 4
# specify that one coupled-perturbed Kohn-Sham calculation is performed where
# nucleus number 1 (according to the ordering in the ADF output) is the
# perturbing nucleus, and nuclei 2, 3, and 4 are the perturbed nuclei, and
# another coupled-perturbed Kohn-Sham calculation is performed where nucleus 3
# is the perturbing nucleus and nucleus 4 is the perturbed nucleus.

# The second CPL run also includes the spin-dipole (SD) term, through the SD
# subkey.

# The output of the CPL program first contains a lot of general information, a
# summary of the specified input, and then produces the desired numbers:

# It prints separately the different contributions (FC, DSO, PSO, SD) if
# specified in input and sums them up to a total number. Experimental NSCCs
# between two nuclei A and B are usually reported as J(A,B) in Hertz. From a
# computational point of view, the so-called reduced NSCCs K(A,B) are more
# convenient for comparisons. CPL outputs both. In this example, the Fermi-
# contact term is indeed dominant.

# The first part of the output refers to the line
#   nuclei 1 2 3 4
# then the same thing is done for the second similar line where nucleus 3 is the
# perturbing nucleus.

# The output for the second CPL run looks very similar, but now the SD term is
# added to the Fermi contact term, resulting in much longer execution times.


# == Scalar relativistic and spin-orbit calculations ==

# The CPL program also enables calculations using scalar relativistic effects
# (ZORA) and/or spin-orbit effects.

# Schematically, this requires the following changes to the input file with
# respect to a regular spin-orbit calculation and a non-relativistic CPL
# calculation:

# steep (1s) functions may need to be added to the standard basis sets. the
# full-potential option for ZORA is needed in the create runs and all further
# runs: relativistic zora scalar full the molecular ADF calculation should
# contain the line relativistic zora full spinorbit the CPL input is unmodified
# with respect to the example given here. Please check the ADF manual for
# details on relativistic input options. 


# coupling including the SD term

$ADFBIN/cpl <<eor
  nmrcoupling
    dso
    pso
    sd
    scf converge=1e-7 
    nuclei 1 2 3 4                                              
    nuclei 3 4
  end                                                         
eor
rm logfile

mv TAPE21 C2H2_1.t21

# redo the same set of calculations, but with a different orientation
# of the molecule this time. Only the CPL calculation without the SD
# term is carried out in order to keep the execution time for this example
# within reasonable limits

rm logfile
rm TAPE10

$ADFBIN/adf <<eor
  TITLE C2H2 nrel
  BASIS
    Type TZ2P
    Core None
    CreateOutput None
  END
  ATOMS  
    C 0.0  0.0     0.0
    C 0.0  1.20692 0.0
    H 0.0  2.26672 0.0
    H 0.0 -1.0598  0.0
  END
  XC
    GGA Becke Perdew
  END
  BeckeGrid
    Quality good
  end
  SAVE TAPE10
  SYMMETRY NOSYM
eor
rm logfile

# coupling without the SD term

$ADFBIN/cpl <<eor
  nmrcoupling
    dso
    pso
    scf converge=1e-7
    nuclei 1 2 3 4 
    nuclei 3 4
  end
eor
rm logfile

mv TAPE21 C2H2_2.t21