Abstract

Catalytic amount of cheap chiral recognition reagent (quinine) combined with chemometric and infrared spectroscopy (IR) was used for quick and accurate determination of the enantiomeric compositions of ibuprofen. First of all, full spectrum was used as the input variable of partial least regression (PLS) to establish the calibration model, after which a good prediction rate of the average error of 4.65% was obtained. Then, wavelet transform (WT) algorithm, with strong compression ability, was employed to obtain a more concise model. The low-frequency coefficients were extracted by the simplest Harr wavelet function, and were used as the input variables to establish the calibration, after which a good prediction rate of the average error of 3.60% was obtained. All the results showed that the combination of partial least squares and infrared spectroscopy can be used for quick and accurate prediction of the enantiomeric excess (ee) value of ibuprofen.

Key words: Ibuprofen, enantiomeric compositions, quinine, infrared, wavelet transform.