Changes in the chemical composition of softwood pulp fibres induced by kraft pulping were readily monitored using analytical pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) combined with principal component analysis and partial least squares analysis. For quantification of lignin and carbohydrates in kraft pulps, a multivariate calibration model was calculated by correlating the Py-GC/MS results with those obtained from the traditional acid hydrolysis/derivatization-GC method. The ability of this calibration model to predict the lignin and carbohydrate content was studied using a number of test samples. In these test samples, lignin and the main carbohydrate constituents (glucose, xylose and mannose) could readily be predicted from Py-GC/MS analysis. In contrast, for the minor sugar constituents, arabinose and galactose, the difference between the values predicted from Py-GC/MS and those determined by acid hydrolysis was larger, owing to the poor reproducibility of these sugars in wet-chemical analysis. However, the pyrolysis method showed very good reproducibility, which indicates that improved quantification of lignin and all the five sugar constituents can be achieved by using a rapid Py-GC/MS technique.
