In order to investigate the difference in moisture adsorptivity between the three hydroxyl groups in cellulose-glucopyranose unit, the moisture sorption isotherm and the heat of moisture sorption were observed for a series of cellulose acetates over whole range of relative humidity from dryness up to saturation at 30°C. The series of cellulose acetates, including a regenerated cellulose and a tri-acetate as extremes, were well-defined in terms of the partial degrees of substitution of the three hydroxyl groups, <
fi>, where
i is 2, 3, or 6 indexing the carbon atom to which the hydroxyl group is bonded.
It was revealed that the B. E. T.'s multilayer adsorption model was not applicable to the analysis of the isotherm not only for the cellulose acetates, but even for the regenerated cellulose, explicitly, at low relative humidities of less than 5%. The isotherms in the low humidity range were analyzed successfully by a composite isotherm model of three components; D'Arcy and Watt's model composed of two Langmuir terms and a linear term. One of the two Langmuir terms with relatively large adsorptive energy factor, the other Langmuir term with relatively small adsorptive energy factor, and the linear term, were attributed to the adsorptions due to OH-6 (the hydroxyl group bonded to the C-6 carbon), to OH-2, 3 (the hydroxyl groups bonded to the C-2 and C-3 carbons), and to OAc-2, 3, 6 (the substituted acetyl groups), respectively.
The differential heat of moisture sorption at dryness,
QL(α=0), was found to decrease sequentially with increase in the total degree of substitution; from 341±22 for the regenerated cellulose to 184±10 cal/gram of liquid water for the tri-acetate. Assuming the differential heat of moisture sorption in bulk at dryness to be simple sum of these of the polar groups, the
QL, OH-6(α=0),
QL, OH-2, 3(α=0), and
QL, OAc-2, 3, 6(α=0) were determined as 383, 320, and 180cal/gram of liquid water, respectively, in good accordance with the above assignments of the three kinds of moisture adsorption mechanism, at least qualitatively.
View full abstract