The acylation of three cellulose samples by acetic anhydride, Ac 2 O, in the solvent system LiCl/ N,N-dimethylacetamide, DMAc (4 h, 110°C), has been revisited in order to investigate the dependence of the reaction efficiency on the structural characteristics of cellulose, and its aggregation in solution. The cellulose samples employed included microcrystalline, MCC; mercerized cotton linters, M-cotton, and mercerized sisal, M-sisal. The reaction efficiency expresses the relationship between the degree of substitution, DS, of the ester obtained, and the molar ratio Ac 2 O/AGU (anhydroglucose unit of the biopolymer); 100% efficiency means obtaining DS = 3 at Ac 2 O/AGU = 3. For all celluloses, the dependence of DS on Ac 2 O/AGU is described by an exponential decay equation: DS = DS o -Ae -[(Ac2O/AGU)/B] ; (A) and (B) are regression coefficients, and DS o is the calculated maximum degree of substitution, achieved under the conditions of each experiment. Values of (B) are clearly dependent on the cellulose employed: B (M-cotton) [ B (M -sisal) [ B (MCC) ; they correlate qualitatively with the degree of polymerization of cellulose, and linearly with the aggregation number, N agg , of the dissolved biopolymer, as calculated from static light scattering measurements: (B) = 1.709 ? 0.034 N agg . To our knowledge, this is the first report on the latter correlation; it shows the importance of the physical state of dissolved cellulose, and serves to explain, in part, the need to use distinct reaction conditions for MCC and fibrous celluloses, in particular Ac 2 O/AGU, time, temperature.
