Date of Award

3-10-2021

Publication Type

Doctoral Thesis

Degree Name

Ph.D.

Department

Civil and Environmental Engineering

First Advisor

Paul Henshaw

Keywords

Black liquor, Deep eutectic solvent, Extraction, Lignin, Protic ionic liquid

Rights

info:eu-repo/semantics/embargoedAccess

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Abstract

This dissertation investigates lignin extraction from black liquor (BL), a waste stream from Kraft pulping, using protic ionic liquids (PILs) and deep eutectic solvents (DESs). Fourteen PILs were synthesized by pairing seven ammonium cations with two anions (lactate [La] and acetate [Ac]), and their viscosity, electrical conductivity, and refractive index were characterized as functions of temperature. Experimental data for the physical properties were curve-fit using the least-squares method. The solubility of alkaline lignin in PILs was measured at 303 K to 368 K. The three best lignin solubilities were 633.5±16.5 mg g-1, 559.2±12.8 mg g-1, and 660.2±19.9 mg g-1 for pyrrolidinium acetate ([Pyrr][Ac]), ethanolammonium acetate ([Eth][Ac]) and propylammonium acetate ([Prop][Ac]), respectively, at 368 K. Solvents with high lignin solubility under similar conditions were selected to extract lignin from black liquor (BL). The PILs and a control IL (1-ethyl-3-methylimidazolium acetate ([Emim][Ac])) were evaluated to extract lignin under different reaction conditions. Several screening studies were employed to select the highest extracting lignin PIL, [Eth][Ac], and determine the range of extraction parameters (reaction time, temperature, and the IL/PIL:BL ratio (w/w)). The extraction process was optimized using the Box-Behnken Design (BBD) of experiments, and a quadratic prediction model was developed for lignin extraction as a function of the parameters affecting the process. It was concluded that reaction time, operating temperature, and PIL:BL ratio produce significant effects on the total lignin extracted. A numerical optimization analysis of the D-optimality index was performed to assign the optimum lignin extraction conditions at 97°C, 4.5 h reaction time and a PIL:BL ratio of 19:1. The predicted and experimental values for total lignin extracted under optimum conditions were70.0% and 75.0±2.9% (283.7±9.1 mg g-1), respectively. The characterization of the extracted lignin samples by Fourier transform infrared (FTIR) spectroscopy confirmed that the observed peaks for the extracted lignin were in agreement with the peaks for a commercially available alkaline lignin. One ionic liquid, [Emim][Ac] (control solvent), and four acetate-protic ionic liquids (PILs), [Eth][Ac], diethylenammonium acetate ([DiEt][Ac]), [Prop][Ac], and [Pyrr][Ac], were employed for kinetic studies of lignin extraction from BL. The kinetics of lignin extraction were examined for a 270-min. reaction time period at three different temperatures. The lowest activation energy (Ea = 6.5 kJ·mol-1) was affiliated with [Prop][Ac]. The PIL with the maximum total lignin extracted, [Eth][Ac], was regenerated using vacuum distillation, and the lignin extraction ability of the regenerated solvent was found to be equivalent to the fresh solvent. FTIR spectroscopy and proton nuclear magnetic resonance (1H-NMR) were employed to characterize the extracted lignin. Gel permeation chromatography revealed molecular weights of 2283±67 g mol-1 and 1892±122 g mol-1 for the lignin extracted using [Eth][Ac] and [Prop][Ac], respectively. The control solvent plus four DESs were synthesized using choline chloride (ChCl) as a salt and four hydrogen bond donors (HBDs): lactic acid (La), oxalic acid, malic acid, and urea. A screening study using the four DESs under different reaction conditions: temperature, DES:BL ratio (w/w), and HBD:salt molar ratio, examined the effect of reaction conditions on total lignin extracted (w%) and determined that lactic acid-ChCl at a 11:1 molar ratio was the best DES for further study. The DES selected was regenerated three times and the total lignin extraction values using regenerated DES were statistically similar to the extraction value using DES before regeneration. FTIR spectroscopy peaks for the extracted lignin samples (using DES before regeneration and after regeneration) agreed with the peaks for a commercially available alkaline lignin. 1H-13C HSQC NMR spectroscopy detected the side chain structure and aromatic linkage peaks in the lignin extracted. Lignin extraction conditions were optimized for La:ChCl using response surface methodology (RSM). A BBD was carried out to develop a quadratic model to predict the total lignin extraction based on operating factors: time, temperature, HBD:salt ratio and DES:BL ratio. The highest lignin extraction conditions using the D-optimality index were: reaction time=4.6 h, operating temperature=99ºC, DES:BL ratio (20:1) (w/w), and La:ChCl (11:1). The maximum predicted value for total lignin extracted under optimum conditions was 83.8% (w%) which was approximately 4.6% (w%) higher than the experimental value, 79.2±1.86% (w%) (299.5±11.0 mg g-1) under similar conditions. The Anderson–Darling (AD) statistic attested to a normal distribution to the residuals, which demonstrated a reasonable correlation between predicted and experimental data over the factor-space studied.

Available for download on Thursday, March 10, 2022

Share

COinS