The thermoelasticity of swollen cellulose filaments

Publisher Website
Google Scholar
Abstract
Although dry cellulose exhibits only the normal short range elasticity of most crystalline or glassy solids, swollen filaments can be elastically stretched to about 5% elongation. It is to be expected that the mechanism of this longer range deformation is different from that of rubbery polymer networks because of strong inter‐ and intra‐ molecular forces and because of the uncertain role of crystalline regions as bona fide crosslinks. Information regarding the elastic mechanism can in general be obtained from a study of the retractive force (f) as a function of temperature. From thermodynamics, it follows that (∂f/∂T)LVN (V = volume, N = solvent content, L = length of filament) is related to the desired entropy of deformation (∂S/∂L)TVN of the swollen filament. The experimentally obtained force‐temperature data should therefore be corrected for thermal expansion and changes in the degree of swelling. A useful form of these correction terms is as follows: (∂f/∂T)LVN  (∂f/∂T)LPe − (∂f/∂N)LPT[(∂N/∂T)LPe + λL(∂N/∂L)fTP] where “e” denotes that equilibrium should be maintained with the surrounding pure swelling agent and λ0 is the linear expansion coefficient of the unstretched filament a t constant solvent content. For the dry filament, correction for thermal expansion only is necessary. All quantities in the correction terms are measured for the dry, as well as the swollen, model filaments. Measurements of force and length are performed with a strain gauge and micrometer. Evaluation of the correction term requires the measurement of dimensional changes of the filament induced by a temperature change as well as by a change in water content a t constant temperature. The latter is achieved by dissolving a polymer in the surrounding water and thus lowering the solvent activity. Cross‐sectional changes are measured by projecting an image of the filament onto the ceiling and photographing the edges. The dry filament exhibits a constant negative slope in the force‐temperature diagram After correction for thermal expansion, the ratio of energy force fE to total force at 90°C., fE/f, is = +3.4. The swollen filaments exhibit a maximum. Here, however, the corrections for thermal expansion and swelling are of such a magnitude that (∂f/∂T)LVN has the opposite sign of (∂f/∂T)Lpe. Under a 2% elongation, fE/f = 4.0 at 30°C. and fEM/f = −4.5 a t 90°C. Energies and entropies of deformation calculated from the data for dry as well as swollen cellulose lead to the following interpretation. In dry cellulose, energy as well as entropy will increase with stretching, over the whole temperature range. In swollen cellulose similar increases are found at temperatures below 60–65°C., But the elastic range is much larger. This type of elasticity can be explained on the basis of a model proposed by Huggins, and may bear some resemblance to the α ⇋ β elasticity in certain proteins. At higher temperature, we find an energy as well as entropy decrease with stretching. It is tentatively proposed that this transition is due to a reversible penetration of water into the better ordered (crystalline) regions of cellulose, without the elastic mechanism which stems from the amorphous deformable regions being affected. It appears that, in general, great care is required before conclusions regarding the elasticity may be drawn from measurements of swollen filaments. In particular, minute changes in swelling may have a very large influence on the thermoelastic behavior.

This publication has 12 references indexed in Scilit: