Dépolymérisation de la cellulose dans des mélanges fluides binaires et ternaires riches en CO2 à hautes pressions

Abstract

The primary objectives of this thesis have been (1) to investigate the behavior of cellulose in single and muticomponent fluids rich in CO[subscript 2]; (2) to identify the fluids which can depolymerize and dissolve cellulose; (3) to study the effect of operation mode (batch or flow system), extraction time, temperature, pressure and the solvent composition on the dissolution of cellulose by the selected fluids; (4) to characterize the collected material from the extraction process; (5) to develop a theoretical explanation of the dissolution process. The motivation in the study is the potential benefits that could result from successful utilization of supercritical fluids for effective dissolution and separation of complex lignocellulosic materials, such as wood. Based on hydrogen bonding and solubility parameter concepts, different fluids were selected: pure carbon dioxide and water, the binaries: CO[subscript 2] - water; CO[subscript 2] - methanol and CO[subscript 2] - acetone and the ternaries: CO[subscript 2] - water - methanol and CO[subscript 2] - water - ethylene glycol. Cellulose Sigma was used as starting material. The temperature did not exceed 200°C which was considered a limiting value before thermal degradation. An experimental set up was constructed for this study. The system consisted of a high pressure view cell (thermodynamic observations) or reactor and a set of hot and cold separator traps for collection of extracted material after decompression. This system was improved several times in the course of the study. A first set of experiments suggested that the binary CO[subscript 2] (0.6) - water (0.4) was the most promising. As parameters of comparison, % dissolution and degree of polymerization of cellulose were used. Since it is well accepted that the first step in cellulose dissolution is swelling we decided, based on earlier works, to impregnate the cellulose with ethylene glycol before the CO[subscript 2] - water treatment. The best results were obtained when the experiment was conducted in batch mode. We obtained dissolution/extraction of above 90% when the reaction was conducted in the range 160 - 170°C and pressures comprised between 40.0 and 55.0 MPa. With a reaction time of six hours, the dissolution/extraction was 80% and the recovery of cellulose accounted for 70%. With a reaction time of ten hours, the dissolution/extraction reached 90% and the cellulose recovery reached 80%. The infrared spectra of the non-dissolved residues and the depolymerized cellulose samples recovered in the separator traps did not show any significant differences. To obtain larger quantities of products, we have conceived and constructed a new reactor. By running NMR analysis, we have shown that the ethylene glycol penetrates the cellulose structure and/or makes a new hydrogen bond.