Development Of Methods For Analyzing Raman Imaging Data From Plant Cell Wall And Its Applications

The recalcitrance of plant cell wall is the primary barrier for conversion of such renewable resources to biofules,chemicals and bio-based materials.Key to overcoming this problem is to comprehensively understand the complicated structure of plant cell wall and its chemical nature.As an in-situ technique,Raman spectroscopic imaging is finding ever-increasing applications in the field of cell wall topochemistry for its ability to provide spatial and chemical information of compositions within plant cell wall simutaneously.The acquired Raman imaging data set often contains thousands of spectra measured at hundreds or even thousands of individual frequences.However,most of the available methods in Raman spectral processing are designed for single-spectrum analysis.Until now,there is no anlytical method developed for Raman imaging data set of plant samples.To address this issue,the author established a novel method for denoising and automatic classification of Raman imaging data set of plant cell wall based on the multivariate methods.The above method is utilized to separate pure spectra of cellulose and hemicelluloses,explore the destruction mechanism of plant cell wall pretreated by ionic liquid(EmimAc),and quantitatively monitor the lignin content changes of different cell wall layers during delignification.The main content of this paper are as follows.(1)The author proposed a method named "automatic pre-processing method for Raman imaging data(APRI)" to remove the spectral contaminants in Raman imaging data set of plant cell wall.The method works by introducing the adaptive iteratively reweighted penalized least-squares(airPLS)algorithm and the principal component analysis(PCA).The baseline drifts and cosmic spikes were removed on the basis of the spectral features themselves.It retains the important information of original data and is suitable for denoising Raman spectra from hardwood,softwood and herbage,and eliminates the human influences and makes the denoising results stable and reproducible.The results of APRI can be processed by multivariate analysis such as singular value decomposition(SVD)and multivariate curve resolution.In addition,APRI is computaionally efficient and conceptually simple and potential to be extended to other methods of spectroscopy,such as infrared(IR),nuclear magnetic resonance(NMR),X-Ray Diffraction(XRD).With the help of this method,a typical spectral analysis can be performed by a non-specialist user to obtain useful information from a spectroscopic imaging data set.(2)The author illustrates a novel method for automatically identifying the spectra of different plant cell wall layers in Raman imaging data set based of PCA and cluster analysis.The method falls into five steps:1)collection of the Raman imaging data set,2)baseline correction,3)PCA for specific spectral peaks,4)cluster analysis for PCA scores,and 5)results verification.On the basis of this method,the author calculates the average spectra in various cell wall layers.These spectra were more accurate and representative than those obtained by manual method since they contain the whole spectral data of the corresponding layer.In addition,the relationship between different characteristic peaks within the Raman spectra of wood samples was discussed.The result of correlation coefficients indicated that the peak at 1331 cm-1 in the Raman spectrum of wood samples might be more related to lignin rather than cellulose.(3)As a multivariate curve resolution,self-modeling curve resolution(SMCR)is used to decern the Raman spectral imaging data set(after APRI processing)of model monosaccharides(glucose and xylose)and polysaccharides(cellulose and hemicelluloses).Results indicate that SMCR is a powerful tool for discrimination of Raman spectra of saccharide,while the lignin in plants plays a major role in disturbing the seperation of polysaccharides.To address this issue,the plant cell wall is delignified by NaC102.The semi-quantitative concentrations of polysaccharides are identified based on the corresponding Raman images.Results show that cellulose is mostly concentrated in the secondary wall(SW)of poplar fibres,whilst the distribution of hemicellulose is almost uniform throughout the cell wall of fibres except for a higher concentration found in the S1 and the outer S2 layer.The vessel have relatively high hemicellulose concentrations which is comparable to the outer S2 layer of fibres,but the cellulose concentration is relatively low in these two cell types.(4)The process of ionic liquid pretreatment of poplar cell wall using 1-ethyl-3-methyliidazolium acetate(EmimAc)is monitored on a cellular lavel by employing Raman imaging technique.The results show that the ionic liquid penetrates into poplar cell wall and results in cell wall swelling.There is a significant difference in wall thickness between SW adjacent to cell corner(CC)and compound middle lamella(CML)(the former is thicker),but they present a very similar swelling behaviour:degree of SW swelling increases up to 2.02× and 2.00× with in 2 h,respectively.The poplar cell wall dissolution during the ionic liquid pretreatment can be clearly divided into two stages:1)slow penetration of ionicliquid,and 2)rapid dissolution of lignin and polysaccharides.In this case,the onset of the dissolution of these compositions occurred only after the cell wall of biomass swelled to a certain extent.The swelling and dissolution occur alternatively until all of the components completely dissolve into the ionic liquid.(5)Raman imaging technique is employed to track the delignification of poplar cell wall by NaClO2.Transverse sections of poplar xylem are applied in all the experiments to ensure that the results of various measurements are consistent and corroborate each other.The dynamic images of lignin distribution are achieved by predicting the changes of Raman spectra in untreated samples.The fixed observation area guaranteed that the Raman spectra acquired at different treatment times were comparable.Depending on the available continuous Raman images,the results of component analysis and the method for automatic identification of Raman spectra from different cell wall layers,the quantitative data of delignification of various cell wall layers were finally achieved.The results show that that the lignin content of SW,CML and CC in untreated sample were 72.26%,21.22%and 6.52%(g/g),respectively.The delignification process eventually removed 31.71%(g/g)of lignin:in which 53.83%from SW,33.18%from CML,and 12.99%from CC.Specifically,delignification of the SW can be divided into two stages:1)slow penetration of reaction liquid,2)followed by rapid removal of lignin.