| Ever increasing global demands for energy and concerns on our environment have permitted increased demand for utilization of lignocellulosic biomass to chemicals and fuels. However, the natural recalcitrance of plant cell walls resulting from complex arrangement and distribution of heterogeneous components impedes its conversion. To overcome the obstacle, low-cost and effective pretreatment technologies are required by opening the rigid cell wall structure to be accessible. In the present work, a series of microscopies combined with wet chemical analysis were carried out to solve the three scientific issues:1) interpreting the natural recalcitrance of lignocellulosic biomass; 2) digging out the dynamic changes of plant cell walls during dilute acid and alkali pretreatments; 3) evaluating the influence of the anatomical and compositional changes on enzymatic digestibility. The results are showed as follows:Compared to Pinus yunnanensis opposite wood (OW), the mild compression wood (MCW) tracheid displayed an additional lignified layer (S2L) adjacent to the compound middle lamellar (CCML). Addtionally, the severe compression wood (SCW) with a helical cavities-rich secondary wall (SW) had intercellular spaces at corners. In OW tracheid the highest lignin concentration was observed in the CCML, whereas the S2L layer has the highest level of lignification in SCM tracheid. As a transition the MCW tracheid showed comparative level of lignification in the CCML and S2L. Moreover, the cellulose distribution displayed and opposite trend with lignin and the largest cellulose microfibril angle was in MCW S1 but the lowest was in SCW S1.By performing sodium hydroxide pretreatment (2% w/v,121℃) on poplar(Triploid populus tomentosa) cell walls, swelling occurred primarily in the secondary wall (S) but alkali had little effect on the cell corner middle lamella (CCML). Correspondingly, there was a preferential delignification in the S at the beginning of pretreatment, while the level of delignification in CCML (-88%) was higher than that in the S (~83%) for the whole process revealed by Raman spectra. It also suggested that prolonging residence time to 180 min would not remove lignin completely but cause rapid loss of carbohydrates, which was further visualized by Raman spectroscopic images. Furthermore, pretreatment with alkali exposed the embedded microfibrils from non-cellulosic polymers clearly, enlarged the diameter of microfibrils, and decreased the surface porosity. These results suggested that there was a synergistic mechanism of lignocellulose deconstruction regarding cell wall swelling, main components dissolution and microfibrils morphological changes that occurred during alkali pretreatment.Compared to Triploid populus tomentosa opposite wood (OW), the tension wood (TW) fiber displayed an additional gelatinous layer (GL) with high cellulose concentration. During dilute acid pretreatment,29%-71% hemicelluloses were dissolved into pretreatment liquid, leaving behind a cellulose and lignin-rich residual. The cellulose microfibrils were disintegrated due to matrix removal. Meanwhile, lignin droplets were dispersed on the inner cell wall with a dense distribution nearby pits. For topochemical changes, hemicelluloses in OW fiber were easier to be acid-catalyzed, while delignification in TW fiber was more severe. It can be concluded that the differential structure features of OW and TW contributed to the diverse response to pretreatment.The internodes of Miscanthus x giganteus were subjected to dilute acid pretreatment to yield a range of samples with altered cell wall structure and chemistry. The consequent morphological and compositional changes and their possible impact on saccharification efficiency were comprehensively investigated. Dilute acid pretreatment of M. x giganteus resulted in solubilization of arabinoxylan and cross-linking hydroxycinnamic acids in a temperature dependent manner. The optimized pretreatment (1% H2SO4,170℃ for 30 min) resulted in significant enhancement in the saccharification efficiency (51.20%) of treated samples in 72 h, which amounted to 4.4-fold increase in sugar yield over untreated samples (11.80%). The remarkable improvement could be correlated to a sequence of changes occurring in plant cell walls due to their pretreatment-induced deconstruction, namely, loss in the matrix between neighboring cell walls, selective removal of hemicelluloses, redistribution of phenolic polymers, and increased exposure of cellulose. The consequently occurred changes in inner cell wall structure including damaging, increase of porosity and loss of mechanical resistance were also found to enhance enzyme access to cellulose and further sugar yield.During dilute acid pretreatment (1% H2SO4,170℃), in addition to being removed lignin can migrate from the cell corner middle lamella (Ccml) into compound middle lamella (Cml) and secondary wall (Sw) regions, as well as from inner to the outer surface. Upon cooling after treatment coalesced lignin could solidify and either become trapped within the wall layers or settle out of the bulk liquid, potentially depositing back onto the biomass surface. The density of lignin droplets was sensitive to pretreatment severity. After 30 min residence, the sizes of lignin droplets appeared to differ between structures of which diameters ranging from 20 nm to 265 nm, with an average of 103 nm. Moreover, a mechanism of lignin migration within cell walls was proposed, showing preferential diffusion of lignin to less lignified regions.The recalcitrance in grasses varies according to cell type and tissue. Dilute acid pretreatment (1% H2SO4 170℃ for 30 min) on M. x giganteus pith effectively hydrolyzed 73.33% hemicelluloses and separated cohesive cell walls from the compound middle lamella due to lignin migration. Lignin droplets with an average diameter of 49.5±29.3 nm were concurrently coalesced on wall surface, that in turn exposed more microfibrils deep in walls to be enzymatically hydrolyzed reaching 82.55%. By contrast, the rind with a relatively intergrated cell structure was covered by larger lignin droplets (101.2±44.1 nm) and filled with inaccessible microfibrils limiting enzymatic sacchrification (31.50%). Taken together, the cellulose digestibility of biomass was not majorly influenced by cellulose crystallinity, while it was strongly correlated with the positive effects of hemicellulose degradation, lignin redistribution, cellulose exposure and loosening cell wall structure. |