Institute of Chemical Engineering page 1Achema 2012

Thermal Process Engineering

Bettina Liebmann, Joana Rodrigues,

Anton Friedl, Kurt Varmuza

bettina.liebmann@tuwien.ac.at

NIR spectroscopy and chemometrics for ionic liquids: lignocellulosic biomass dissolution

Institute of Chemical Engineering page 2Achema 2012

Thermal Process Engineering

The dissolution in ionic liquids (IL) is a new, alternative technology to disrupt the complex fibre network of lignocellulosic biomass at comparatively mild conditions.

Its three main compounds – cellulose, hemicelluloses, and lignin – are interesting renewable sources for biofuels, chemicals, and biomaterials.

Once dissolved in the IL, they can be separated by simple addition of an anti-solvent; e.g., the addition of water immediately precipitates amorphous cellulose from the IL.

In this preliminary study we investigate the applicability of NIR spectroscopy and chemometrics for quantifying the three main compounds of lignocelluloses in IL.

Introduction

Straw composition, schematically

Institute of Chemical Engineering page 3Achema 2012

Thermal Process Engineering

consist entirely of ions (salts) melting point < 100°C (liquid at room T!) negligible vapour pressure excellent thermal stability up to 400°C non flammable electrically conducting, magnetic many interesting applications! “design” properties by ion combination

Ionic Liquids

Molecular structure of the ionic liquid EMIM-Oac (1-ethyl-methyl-imidazolium

acetate)

After mixing the pure cellulose with EMIM-Oac, particles and

agglomerations are dissolved by ultrasound and temperatures 40-

60 °C for 7-24 hours to obtain clear solutions.

Institute of Chemical Engineering page 4Achema 2012

Thermal Process Engineering

NIR spectrometer (Brimrose Luminar 5030, AOTF)

Near-infrared spectroscopy (NIR)

NIR spectrum of the used IL; NIR spectra of pure solid

cellulose, lignin, and xylan.

NIR transflectance probe with variable pathlength for liquid samples (left); NIR diffuse reflectance probe for solid samples (right).

Institute of Chemical Engineering page 5Achema 2012

Thermal Process Engineering

NIR spectra and predicted concentration by PLS regression model

Cellulose + IL Lignin + IL Xylan + IL

SEPCV, standard error of prediction from leave-one-out cross validation; aCV, number of PLS componentsPLS-models with leave-one-out CV in software Unscrambler.

Institute of Chemical Engineering page 6Achema 2012

Thermal Process Engineering

Empirical models for the concentration of cellulose, lignin, and xylan in ionic liquid EMIM-OAc result in good prediction performance.

In the next step, experiments will focus on multi-component mixtures of cellulose, lignin, and xylan in the ionic liquid.

Good reference values for actually dissolved straw components in the IL is a limiting factor in PLS model building for industrial application.

Summary and Outlook