|
Internet Electronic Journal of Molecular Design - IEJMD, ISSN 1538-6414, CODEN IEJMAT
| ABSTRACT - Internet Electron. J. Mol. Des. February 2002, Volume 1, Number 2, 94-107 |
Quantitative Structure-Retention Relationship Study of Gas
Chromatographic Retention Indices for Halogenated Compounds
Teodora Ivanciuc and Ovidiu Ivanciuc
Internet Electron. J. Mol. Des. 2002, 1, 94-107
|
Abstract:
A high interest in predicting physical, chemical, and
biological properties of halogenated compounds is
generated by the urgent need to develop alternatives to
chlorofluorocarbons, new compounds with low ozone
depletion potential and a low global warming potential.
Quantitative structure-retention relationships (QSRR) for
halogenated compounds can be used to predict gas
chromatographic retention indices by using theoretical
descriptors computed from the chemical structure.
QSRR models for the gas chromatographic retention
indices of 207 halogenated compounds are established with
the CODESSA (Comprehensive Descriptors for Structural
and Statistical Analysis) program.
The best results are obtained with a group of QSRR models
with six structural descriptors, with a correlation coefficient
between 0.994 and 0.993 and a leave-one-out cross-validation
correlation coefficient between 0.992 and 0.991.
All QSRR models contain the Kier and Hall connectivity
index 1χv, the number of F atoms NoF, and the gravitation
index for all pairs of atoms G1. A fourth descriptor comes
from the class of charged-partial surface area indices,
while the remaining two descriptors are related to the
number of I or Br atoms, bond order, or Randic
connectivity index 1χ.
QSRR models developed with CODESSA allow accurate
computation of the gas chromatographic retention indexes
of halogenated compounds using simple constitutional,
topological, geometric, electrostatic and quantum
descriptors that can be computed with standard quantum
chemistry packages.
|