Studies of the Alkaline Degradation of Cellulose and the Isolation of Isosaccharinic Acids

Cellulosic materials are expected to form a significant proportion of the waste proposed for disposal in underground repositories being designed for the storage of radioactive waste. Under the alkaline conditions of these facilities, cellulose degrades by a so called „peeling‟ reaction resulting in the production of a complex mixture of products (CDPs), the major components being α- and β isosaccharinic acid (α and β-ISA). A significant amount of research has been performed on ISA as part of the safety assessment for the development of these underground repositories due to the ability of ISA to complex with, and increase the solubility of radioactive isotopes. Until now, the vast majority of this research has involved the readily-available α-ISA, only a limited number of studies have involved β-ISA because no simple procedure is available for its isolation. Therefore, in this project, a method for the synthesis and isolation of β-ISA was developed.

Cellulose degradation experiments which were performed to maximise solution concentrations of β-ISA are described in chapter 3. Microcrystalline cellulose was degraded under anaerobic conditions at either RT, 50 °C or 90 °C and comparisons were made between the use of NaOH and Ca(OH)2 as the base catalyst. As expected, the major products of all degradation reactions were α- and β-ISA, in addition, small amounts of free metasaccharinic acid (MSA) was detected in the Ca(OH)2 reactions. The largest solution concentrations of β-ISA were produced when cellulose was degraded at 90 °C using NaOH; after 24 hrs of reaction, solution concentrations of 12.7 g L-1 were achieved, whereas, in the equivalent Ca(OH)2 reaction, after 4 days a maximum concentration of only 5.1 g L-1was produced. For this reason, cellulose was degraded at 90 °C using NaOH to produce degradation solutions to be used in procedures to isolate β ISA. An additional finding was that significant amounts of ISA were being removed from degradation solutions due to absorption on to unreacted cellulose fibres; in the NaOH reaction, absorption was occurring rapidly and the percentage of ISA in both the solution and solid phases were very similar. In the Ca(OH)2 reaction, the absorption was a slow process and the percentage of ISA on the solid phase (61 %) was lower than the percentage of ISA in the solution phase (84 %) suggesting that solid Ca(OH)2 was affecting both the rate at which absorption was occurring and the composition of the absorbed species; this was possibly due to solid Ca(OH)2 physically obstructing the access of ISA to the cellulose fibres and also catalysing the oxidation of some of the ISA into smaller fragmentation products.

Methods which were developed to isolate β-ISA are described in chapter 4. Isolation of β-ISA was initially achieved by eluting crude cellulose degradation solutions directly through a column of anion exchange resin. Using an automated system, a large throughput of material was possible resulting in the accumulation of relatively large amounts of β-ISA; after repeating the column 17 times, 1 g of pure β-ISA was isolated. However, using this method, the crude solutions severely fouled the anion exchange resin, concluding that anion exchange was more suited to small scale isolations of β-ISA. A final isolation procedure was developed which involved the elution of mixtures of benzoylated CDPs through normal phase silica columns. It was determined that prior to elution, coloured impurities could be efficiently removed by passing the derivatised mixture through a wide bed of silica. Slow elution of the resulting clean syrup through a large silica column allowed up to 7 g of tribenzoylated β-ISAL to be isolated and following de-benzoylation procedures, 2.6 g of β-ISA was isolated from a single column. The large protecting groups also allowed single crystals of both α- and β-tribenzoate to be produced and the resulting X-ray structures confirmed the absolute configuration of tribenzoylated β-ISAL as being 2R, 4S. Additional NMR analysis of collected fractions allowed several other polyhydroxylated compounds to be identified, also present as their perbenzoylated esters, these being: 3,4-dihydroxybutanoic acid, 2,5-dihydroxypentanoic acid, 2,3-dideoxypentanoic acid and 2,4,5-trihydroxypentanoic acid.

The isolation of large amounts of β-ISA allowed several solution phase physical properties of β ISA to be measured and these are reported in chapter 5, including the aqueous pKa (3.61) which was determined using NMR methods. The rate constants for the inter-conversion between ISAH and ISAL were also studied for both α- and β-ISA. In acidic environments, ISAH undergoes an acid catalysed lactonisation to generate isosaccharino-1,4-lactone (ISAL), conversely in basic environments, ISAL undergoes a base catalysed ring-opening to produce ISAH. Using pH-stat autotitration, the second-order rate constants for the lactone hydrolysis reaction were determined, to which values of 25.3 M-1 s-1 for β-ISAL and 97.0 M-1 s-1 for α-ISAL were observed. The acid catalysed lactonisation of ISAH was studied using 1H NMR spectroscopy; the second-order rate constant for the lactonisation of β-ISAH (3.10 x 10-3 M-1 s-1) was larger than the second order rate constant for the lactonisation of α-ISAH (7.04 x 10-4 M-1 s-1).