**
pyVib **

** -molecular graphics program- **

I.About pyVib

1) What is pyVib ?

pyVib is a molecular graphics program. On the one hand, pyVib allows the visualization of various spectra such as Infra-Red (IR), Vibrational Circular Dichroism (VCD), Raman and Raman Optical Activity (ROA). On the other hand, pyVib is a tool for visualizing molecular structures and the related vibration modes. In addition, pyVib allows the representation of group coupling matrices (GCM's) and atomic contribution patterns (ACP's) as defined by W. Hug (Chemical Physics 264(2001) 53-69).

2) Input file(s)

To date, pyVib is able to read the latest versions of fchk output files of Gaussian03(TM). In the near future, the output files of Dalton (ab-initio program) will be handled.

3) Output file(s)

Any convoluted/line spectrum is written as either ir.dat, vcd.dat, ram.dat or roa.dat in the working directory. These files are in the usual ascii format as X Y columns, X being the frequencies while Y are the corresponding intensities. The idea behind these outputs is to allow the plotting of spectra with other programs.

II. technical information

In this section, the definitions and formulae used in pyVib are summarized.

1) vibrations

a) Representation of motion

The diameters of spheres are always chosen proportional to the excursions. For the excursions there are the following options:

(L^x)^2 summed to 1 like in Gaussian

Advantages: Mostly of a practical nature, the sum of the surfaces of all spheres is 1 and the scale hardly ever needs to be readjusted when going from one vibration to another.

For a given vibration, but not between vibrations, the relative size of the surfaces of the spheres shows the relative importance nuclear motion has for generating a particular cross-section.

Disadvantages: No comparison of the size of nuclear motion is possible for different vibrations.

b) Representation of fractions of vibrational energy

The fraction of vibrational energy of a nucleus in a normal mode is given by the square of its mass-weighted excursion L, where L is normalized by putting the sum of all L^2 equal to 1.In the visualization program, it is the volume of the spheres which is chosen proportional to L^2. This is meaningful as the energy is a scalar quantity, and as a geometrical interpretation, similar to Lx as a normalized nuclear displacement, is not possible for mass-weighted coordinates.

2) Cross-sections

For both ACP's and GCM's, individual invariants occurring in the formulae for scattering cross sections can be drawn in the same way as the scattering cross-sections themselves are drawn.For Raman and ROA, for the separate viewing of e.g. the quadrupole and the magnetic dipole contributions, or of the isotropic and the anisotropic contributions, there are numerous options, depending on whether one includes or not the numerical and the frequency factors with which invariants are multiplied in scattering cross sections. The two basic options are:

Representation of that part of a scattering cross-section which is due to e.g. quadrupole contributions. This means that the quadrupole invariant will be multiplied with all factors with which it is multiplied in the scattering cross-section. (This is what can best be extracted from measurements done with different polarization schemes and scattering geometries.)

Representation of the reduced invariants. This excludes numerical factors, and also the integral over vibrational wave-functions. (This is an option which is also meaningful for vibrational absorption and VCD.)

a) ACP's (atomic contribution patterns)

With cross-sections representing areas, and considering the fact that they represent quantities for an isotropic sample and thus rotational averages, the logical choice is to draw spheres with a surface proportional to the computed cross-sections. Spheres, viewed from different directions, evidently always present the same surface, and so does a molecule rotated in space with spheres drawn on its atoms (unless they cover each other, of course).

At present, the program can only draw spheres on nuclei. It is equally possible to define such spheres for groups of atoms, e.g. a methyl group, or the H-atoms only of a methyl group, but this is not yet implemented. In this case, we no longer deal with ACPs, but with GCPs (Group Contribution Pattern).

b) GCM's (group coupling matrices)

Drawn as circles with surfaces proportional to scattering cross-sections.

III.Next developments

The following features will be added soon:

- Reading of various output files (Dalton, Turbomole)
- Pattern recognition
- Any idea you have is welcome

- Photo gallery: picture 1
picture 2
picture 3
picture 4
picture 5
picture 6

picture 7 picture 8 picture 9 picture 10 picture 11 picture 12 picture 13 picture 14

Document written by Dr. Mohamed Zerara Mohamed.Zerara@unifr.ch