3.4.1 ANALYSIS OF CARBOHYDRATES AND THEIR TRANSFORMATION PRODUCTS

Let us consider first the analysis of a sample of hemicelluloses dissolved in water. The general analytical strategy is given in Figure 10. An ana­lytical procedure using GC based on acid methanolysis consists of the following steps [23]. Freeze drying of a 2 mL solution of hemicellulose in water with the subsequent addition of 2 M HCl in water-free metha­nol, is followed by keeping the sample at 100 °C for three hours of neu­tralization with pyridine, addition of internal standard (sorbitol), evapo­ration, silylation (hexamethyldisilazane and trimethylchlorosilane), and finally GC analysis. The latter could use, for instance, a split injector (260 °C, split ratio 1:15) with a 30 m/0.32 mm i. d. column coated with dimethyl polysiloxane (DB-1, HP-1), hydrogen or helium as a carrier gas and FID with a following temperature programme: 100-280 °C and ramping 4 °C/min.

An advantage of direct methanolysis of wood samples is that essen­tially only hemicelluloses are cleaved and very little cellulose. Moreover, contrary to hydrolysis, it allows less degradation of released monosaccha­rides. Methanolysis can be used also for direct analysis of solid wood and fiber samples. A typical chromatogram is presented in Figure 11, showing several peaks for a particular sugar due to the presence of a & P anomers of pyranoses & furanoses (Figure 12).

Due to the complexity of the product mixture and the analytical pro­cedure correction factors are needed. For instance, cleavage (the meth — anolysis) could be incomplete for certain glycosidic bonds, such as the Xyl-MeGlcA bond. Some degradation of formed sugars, especially uronic acids may happen and the products can have different detector responses. In order to determine correction factors it is recommended to perform methanolysis, silylation and GC analysis on a sample containing equal amounts of Ara, Xyl, Man, Glc, Gal, GlcA, GalA, etc., and pure hemicel­luloses and pectins (if present) and to compare peak areas with the area of the internal standard.

Another example worth considering is the gas-phase catalytic trans­formation of levoglucosan over zeolites [24,25]. The reaction scheme is given in Figure 13. In [24,25] for HPLC analysis an acid Aminex cation H+ column with sulfuric acid (0.005 M) as a mobile phase with a flow of 0.5 ml/min at 338 K was used, along with an Aminex HPX — 87C column and mobile phase-calcium sulfate (1.2 mM) with a flow rate of 0.4 ml/min at 353 K. A refractive index detector was applied. Figure 14 illustrates that the separation is very much dependent on the analytical conditions.

Stability of the samples is another important issue, which should also be carefully considered, as illustrated in Figure 15. Samples stored in a freezer exhibited another peak, which is certainly a result of transforma­tions happening during storage.

An even more prominent difference in analysis was noticed in the aqueous reforming of sorbitol [27-29]. Comparison of the analysis for different columns is given in Figure 16 demonstrating that for the iden­tification of reaction products tedious and time-consuming analytical work is required.