The Mechanism Of Topochemical Dissolving Of Hemicellulose During Hydrothermal Pretreatment Of Poplar

Hemicelluloses, as structure regulators in cell wall formation, are heteropolysaccharides that are comprised of pentose and hexose sugars with different proportion. They are regarded as the key factor contributing to the recalcitrance of lignocellulosic cell walls, because they are tightly associated with lignin by chemical bonding and with cellulose by physical intermixing resulting in a complex, intricate cell wall structure. To interpret the relationship between ultra-structural characteristics of hemicelluloses presented in plant cell wall and biomass recalcitrance, ongoing research is worldwide focusing on the deposition and dissolution of hemicellulose in cell wall formation and deconstruction at ultra-structural level. In this doctoral thesis, poplar was chosen as raw material due to its fast-growth and high yield features. The topochemical characteristics of various types of hemicelluloses in poplar fiber formation and deconstruction were firstly investigated by transmission electron microscopy (TEM) in combination with immuno-gold labeling. Scanning electron microscope (SEM) and atomic force microscope (AFM) detected the alteration in surface morphology of poplar cell walls during hydrothermal pretreatment (HTP). Furthermore, the changes in the micro-distribution of lignin and cellulose in poplar cell walls during HTP were revealed by confocal Raman microscope (CRM) and their effects on the hemicellulose removal during HTP of poplar fibers were explored. On the basis of results obtained, a hypothesis was developed to describe the mechanism of hemicelluloses removal during hydrothermal pretreatment of poplar fibers at the ultra-structural level. The results were shown as following.1. Xylan was initially deposited at the cell corner of the S1 layer during S1 formation. The deposition of xylan in the S1 and S2 layers increased during S2 formation. In mature poplar fibers, the deposition of low substituted xylans and highly substituted xylans showed different patterns. The low substituted xylans showed more deposition in the outer secondary cell wall than inner layer, while highly substituted xylans showed uniform xylan deposition in the whole secondary cell wall. During maturation, the deposition of mannan occurred mainly in middle layer of secondary cell wall.2. HTP led to the removal of hemicellulose from the S2 layer and CML. The amount of removal of hemicellulose gradually increased with increasing pretreatment time. NMR results verified that xylan was removed from the cell wall and that a certain amount of acetyl groups were still bound to the backbone of dissolved xylan. Furthermore, immune-gold labeling results pointed out that HTP caused greater decline in the density of highly substituted xylans labeling in the sub-layers of fiber wall than in the density of low substituted xylans labeling. While the reduction of mannan labeling in poplar fiber during HTP was not obvious.3. In comparison of the relatively smooth cell wall surfaces displayed in the untreated poplar, droplets were observed on hydrothermally pretreated poplars surface, and the amount and size of which varied with the pretreatment time. The lignin droplets were evenly deposited throuthout the cell wall surface at 10 min and started to decrease in the S2 layer (in the cross-section) at longer pretreatment time. For the tangential section, the coalescence of lignin droplets was observed after 30 min. In addition, HTP caused the removal of materials embedding among the cellulose microfibrils. The amout of removal of the materials enhanced with increasing the pretreatment time, and the explosure of cellulose microfibrils arranged in a parallel manner was found. As the hydrothermal pretreatment time extended to 30 min, void among cellulose microfibrils was observed, and the amount of which increased at longer pretreatment time.4. HTP led to more dissolution of lignin in the S2 layer of fiber than in the compound middle lamella (CML). For example, the lignin concentration in the S2 layer decreased by 70.4% for 30 min, while that in the CML decresed by 40.1%. NMR results showed that the removal of S units (mainly existed in the thick S2 layer) was more than that of G units (principally existed in the thin CML) for 30 min. Meanwhile, the dissolution of hemicellulose and lignin from S2 layer and CML was accompanied by the exposure of embedded cellulose.5. There was a greater loss in the density of xylan labeling during HTP in the delignified and enzymatically treated fibers compared to untreated fibers. The D+HTP fibers displayed about 1.5-4.7 and 3.5-63.8 times lower density of xylan labeling in the S1 layer and S2 layers, respectively, compared with untreated fibers. After HTP+E, the S1 layer and S2 layers of the fibers exhibited about 1.1-4.4 and 0.9-1.5 fold lower density of xylan labeling, respectively, when compared with that in the untreated fibers.6. We firstly propose that in the initial stages of HTP lignin-free xylan in the S2 layer was more readily hydrolyzed than in the S1 layer by hydronium ions. With increasing pretreatment time, the xylan covalently bound to lignin was also removed from the S2 layer due to the dissolution of lignin. The xylan tightly bound to cellulose was seldom removed during HTP, but was hydrolyzed in subsequent enzymatic treatment.