Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

[Licence| Download | New Version Template] ackh_v1_0.gz(16 Kbytes)
Manuscript Title: Exact computation of the Zeeman effect on nuclear quadrupole resonance profiles for powders (spin I = 3/2). Determination of the asymmetry parameter.
Authors: J. Darville, A. Gerard
Catalogue identifier: ACKH_v1_0
Distribution format: gz
Journal reference: Comput. Phys. Commun. 9(1975)173
Programming language: Fortran.
Computer: IBM 370/158.
Operating system: ASP/VS2 -R01.6.
RAM: 28K words
Word size: 32
Keywords: Solid state physics, Asymmetry parameter, Nuclear quadrupole Resonance, Powder simulation, Morino-toyama method, Convolution, Line shape simulation, Zeeman effect, Experiment.
Classification: 7.4.

Nature of problem:
The asymmetry parameter (eta) of the electricfield gradient tensor is to be determined precisely, using powder samples. The shape of the Zeeman effect envelope reflects a typical structure due to eta.

Solution method:
The complete quadrupolar and Zeeman hamiltonian is treated exactly. Gaussian, lorentzian or "expreimental line" convolution of the Dirac profile, followed by derivation, gives the first derivative envelope shape, to be compared to experimental results.

The present version of the program is dimensioned so that 322 crystal- lites are considered to represent the powder. The total number of frequencies involved is then 1932 and 1288 are left in the interesting zone. The variables CO and DER are overdimensioned (1000). In fact their dimension would be in any case AKI plus the number of sampling points on the convolution function, where (AKI+1) is the number of defining points for the interesting frequency zone. If the convolution function is the integrated experimental line, 2*NEXP+1 must be the number of sampling points.

Unusual features:
For the theta and angles, increments of 2 and 15 degrees respectively are chosen, in 1/8 of the sphere. More extended range of variations (0 to 180 degrees, for instance) has been proved unnecessary giving the same envelope shape as in the previous case. But the computation time is mulitplied by a factor of about 4.

Running time:
On the IBM 370/158 of the University of Liege, the loading and assembly of the program and the collection of subroutines takes about 21 s. The execution time is highly dependent on various factors, such as the number of values for the asymmetry parameter, the value of AKI and the number of crystallites (NCOUPL). To give a precise example (only one vaule of eta), for a gaussian type convolution, AKI was chosen as 200. As deduced from the half-height width of the gaussian line, the number of sampling points on this one was equal to 25 and the computation time was then 70 s.