Rapid characterization of biomass using fluorescence spectroscopy coupled with multivariate data analysis. I. Yellow poplar (Liriodendron tulipifera L.)

Abstract

This is the first of a two-paper series that reports on the use of fluorescence spectroscopy coupled with multivariate data analysis as a potential process analytical tool to develop calibration and prediction models for some physical and chemical properties of yellow poplar (Liriodendron tulipifera L.). Waste streams emanating from the processing of this wood species may potentially serve as feedstock for biofuels and biochemicals With the exception of holocellulose content, all the properties considered in the study were predicted with moderate to strong coefficient of determination (R2). Fluorescence spectra-based prediction model for each property considered in this study was compared with near infrared (NIR) spectra-based prediction models of similar properties from a previous study using the same population. The NIR-based prediction models exhibited slightly superior model strength over the fluorescence spectra-based prediction models of similar properties.

Recent innovations in analytical methods for the qualitative and quantitative assessment of lignin

“…Fluorescence spectroscopy 2-3 orders of magnitude more sensitive than UV/vis [37] Can probe relative lignin content based on brightness in imaging techniques [103] Capable of localization of lignin in plant cell wall [103,131,176] Lignin reveals strong fluorescence [96] Cheap instrumentation [277] More selective than absorbance [37] Non-destructive [277] Requires fluorophore (intrinsic or through labeling) [277] Has not led to robust predictive modeling [277] GC Better spectral resolution than HPLC [326] Selectively analyze only volatile species [326] Headspace sampling limits clean-up requirements [56,327] Destructive [326] Requires volatile molecules, or must derivatize to promote into gas phase [328] HPLC Diverse separation configurations and molecular polarities [329] Non-destructive; can collect fractions [330] Separation of non-volatile or thermally unstable molecules [330] Long analysis times [328,331] May require sample clean-up [327] MS Can be used for MW determination [23] Rapid analysis [23] Selectivity [23] Sensitivity [23] Variety of different ionization strategies [23] Destructive [23] MIR spectroscopy…”

Determination of concentration of ACQ wood preservative components by UV-Visible spectroscopy coupled with multivariate data analysis

“…ACCEPTED MANUSCRIPT 12 thumb is that a model distance greater than 3 indicates that the two models that are significantly different from each other (Esbensen 2002). A model distance close to 1 suggests that the two models are identical and may belong to the same class.…”

“…Hence there is a need to find a rapid, cheap and environmentally safe method to replace the laborious and time consuming wet process. A spectroscopic tool such as near-infra-red (NIR) spectroscopy coupled with multivariate data analysis has been successfully used for the rapid determination of many wood properties and their treatment processes 8,9,10,] Past studies have also demonstrated the use of other spectroscopic methods such as fluorescence and mass spectroscopy coupled with Chemometrics, to rapidly characterize wood properties (11,12,13].…”

Direct Infusion Electrospray Ionization – Ion Mobility – High Resolution Mass Spectrometry (DIESI-IM-HRMS) for Rapid Characterization of Potential Bioprocess Streams

“…Currently, there are relatively few examples of analyses applied to "whole-sample" biomass. Fluorescence and nearinfrared spectroscopic techniques coupled with multivariate statistics [2,3] and 13 C-NMR spectroscopy [4] have been employed for rapid determination of whole-sample biomass hydrolysate. However, such techniques are best suited for providing scoping and/or macro-component (i.e., bulk) information on biomass samples.…”

“…Accordingly, detailed characterization of potential biofuel process streams is a critical step in optimizing biomass-to-fuel conversions, which are required for viable energy yields [1]. Most current biomass analyses target bulk components [2][3][4][5] or screen only a small fraction [6][7][8][9][10][11][12][13] of the total molecular composition. Moreover, these techniques can involve timeconsuming preparation steps, extended analysis times, and multiple analyses for different compound classes [6][7][8][10][11][12][13][14].…”