Fast stress relaxation at the surface of silica glass : mechanism and effects

Abstract

It is proposed that the fast relaxation of surface stress takes place as water is able to diffuse into the glass surface. The water in the glass surface causes local fluctuations in the composition which act to relax the glass modulus thereby causing a relaxation in stress. This argument is tested by showing the existence of composition fluctuations through the memory effect, small angle X-ray scattering, and phase separation by the formation of bubbles in the silica glass and water system. The ability of water to cause relaxation is explored through dynamic mechanical and delayed elasticity studies in literature of glasses which contain water or were tested in the presence of water.

The fast relaxation of surface stress in the presence of water at temperatures below the annealing range has been known to occur in oxide glasses. This surface stress relaxation is responsible for the permanent bending of thin glass specimens to which stress was previously applied. Surface stress relaxation has also been leveraged as a method of strengthening glass fibers and has been utilized to explain longstanding mysteries in glass science such as the fatigue limit. Despite the experimental evidence that stress relaxation occurs faster when the glass is exposed to water, the mechanism of this fast relaxation caused by water has remained elusive.

In this thesis, further strengthening experiments revealed that the relaxation of stress takes place nearly instantaneously and bending studies showed that the rate of surface stress relaxation is nearly independent of the applied stress until tensile stresses in the GPa range are applied. While most of the studies focus on silica glass, bending of multicomponent glass thin plates revealed that there may be an initial relaxation which occurs immediately. This initial relaxation is temperature dependent, increasing with increasing temperature. These findings should be viewed in light of the time-dependent elastic properties of glass.

As an alternative to the relaxation of stress, other authors proposed a mechanism of surface swelling in order to explain the strengthening of fibers in which water diffusion accelerated by tensile stress causes the glass surface to swell and leads to a compressive stress at the surface of the glass which causes it to become stronger. This mechanism can be ruled out on the basis of the occurrence of stress relaxation in shear.