Sorption of water vapor by cellulose and polymers at high humidities

Sorption of water by cellulose and other polymers at high humidities is treated as a process of solution of water into the disordered regions of the polymer with deviation from the simple solution theory because of restrications on the swelling of the system by the insoluble ordered regions. The viewpoint taken in this paper is based on the principles presented by Rowen and Simha, but some improvement has been made in the following point. (1) In view of oversimplification made by them in evaluating the partial molal free energy change ΔF̄ 1 (el) due to the elastic deformation of the polymer in the course of sorption, the equation obtained for the swelling of crosslinked polymers is employed: ΔF̄ 1 (el) = RT (1/ z ) v 2 1/3 , where z is the ratio of the molal volume of the polymer between crosslinks to that to solvent, and v 2 is the volume fraction of the polymer. (2) Since the solution of water is considered to take place only into the disordered regions, the volume fraction of the polymer in the amorphous regions must be taken for v 2 in the expression of the sorption isotherm, whereas the existence of crystalline regions has been virtually ignored in the treatment of Rowen and Simha. (3) In examining the sorption data by the solution theory, one must be careful of possible deviations due to the variation of the interaction parameter with polymer concentration, although the present analysis of the sorption isotherm remains unsatisfactory in this point. Along this line of argument, the sorption data for several cellulosic materials and polyvinyl alcohol were examined. The most important feature brought out in this work is the fact that the elastic contribution to the isotherm is considerably great, although recent treatments of sorption by the solution theory have neglected it. Unexpectedly, very small values were obtained for z. This would be an indication that junction points between molecules are present in the solution of amorphous cellulose and that these restrict the swelling of the sample to a great extent. The simple solution theory does not explain the experimental evidence examined here and leads to unreasonably high values for the interaction parameter μ. This is also true for polyvinyl alcohol fibers, but in this case μ diminishes as the vapor pressure increases in the highhumidity region, contrary to the usual case for a number of polymer‐water systems. This anomalous tendency may be accounted for by a hypothesis that the progressive swelling of the fiber would break the hydrogen bonds between the polymer molecules in the sorpation process.