THE DIFFERENTIATION OF ILLICIT METHAQUALONE TABLET FORMULATIONS USINGPRINCIPAL COMPONENT AND SOFT INDEPENDENT MODELING OF CLASS ANALOGY ANALYSIS OF THEIR NEAR-INFRARED REFLECTANCE SPECTRA

The viability of using principal component analysis (PCA) and soft ind ependent modeling of class analogy analysis (SIMCA) of the near-infrar ed reflectance spectra of illicit methaqualone tablet formulations as an aid in sample differentiation was investigated. Near-infrared spect ra of ten different illicit species occurring on the South African mar ket was obtained through direct diffuse reflectance measurements made on the tablet surfaces using a bidirectional fiber-bundle probe. The l earning set was analyzed using PCA and SIMCA and resolved in well-sepa rated clusters thus making differentiation possible.