Research Areas

For the past few years, the research program at BUCAS has focused on the application of chemometrics to problems in analytical spectroscopy. 

Textile Classification

One area of research has involved the classification of textiles on the basis of near-infrared spectroscopy using soft-independent modeling of class analogies (SIMCA), a form of disjoint modeling involving principal component analysis.  This project, which is being conducted in collaboration with the Department of Family and Consumer Sciences at Baylor, has resulted in a fabric database with over 800 different fabric samples.

Forensic Analysis

Another area of interest to BUCAS researchers has been the application of near-infrared spectroscopy combined with multivariate regression modeling to predict the post-mortem time interval from bone samples.

Chiral Analysis

A major research area for BUCAS researchers has been the study of new potential spectroscopic methods of chiral analysis.  The need for improved strategies for the determination of enantiomeric composition arises from increased pressure on the pharmaceutical industry by government agencies for documentation on the pharmacological effects of single-enantiomer drugs combined with the simultaneous demand in drug development for the determination of enantiomeric excess in large combinatorial libraries.  While many techniques for chiral analysis have been developed over the years, gas- and liquid chromagraphy, capillary electrophoresis, and nuclear magnetic resonance are currently the most widely used.  For high-throughput screening, separation methods are relatively slow and rapid spectroscopic techniques are most desirable.  As a result, chiroptical techniques like vibrational circular dichroism and Raman optical activity are gaining popularity in the area of chiral analysis. 

Experimental discrimination of enantiomers with non-chiroptical techniques is carried out conventionally by means of chiral auxiliary agents such as chiral shift reagents (in NMR), chiral complexing agents (in chromatography), and chiral solvents.  This enantiomeric discrimination arises when a given enantiomer of the chiral auxiliary interacts with two enantiomers of a chiral analyte to produce diastereomeric pairs with different physical properties.  Cyclodextrins are barrel-shaped sugar molecules that can form transient, non-covalent diastereomeric guest-host complexes with chiral guest molecules.  Because the complexes that are formed are diastereomeric, they have different physical and spectral properties. 

Recent research in our laboratories has shown that small spectral differences can be observed in the electronic spectra of solutions containing a fixed amount of a cyclodextrin and a fixed amount of a chiral analyte as the enantiomeric composition of the chiral analyte is varied.  This observation has led to the development of a spectroscopic method of chiral analysis known as CARMSD, which stands for Chiral Analysis by Regression Modeling of Spectral Data.  CARMSD combines ordinary spectral data (like UV absorption or fluorescence) with guest-host chemistry and chemometrics to produce a simple procedure to determine the enantiomeric composition of unknown samples.  Research at BUCAS has revealed that good regression models can be prepared that can predict the enantiomeric composition of unknown samples of chiral analytes with a prediction accuracy of ± 3%.  At the present time, BUCAS researchers are actively studying the technique to optimize its potential.