How a blister heals

We use experiments to study the dynamics of the healing of a blister, a localized bump in a thin elastic layer that is adhered to a soft substrate everywhere except at the bump. We create a blister by gently placing a glass cover slip on a PDMS substrate. The pressure jump across the elastic layer drives fluid flow through micro-channels that form at the interface between the layer and the substrate; these channels coalesce at discrete locations as the blister heals and eventually disappear at a lower critical radius. The spacing of the channel follows a simple scaling law that can be theoretically justified, and the kinetics of healing is rate limited by fluid flow, but with a non-trivial dependence on the substrate thickness that likely arises due to channelization. Our study is relevant to a variety of soft adhesion scenarios. Copyright c © EPLA, 2013 Introduction. – Blisters, blebs and boils are protrusions of a skin that separates from a solid substrate. They are ubiquitous in science and technology, and arise in situations ranging from the pesky bubble that is hard to a eliminate from wall-paper to the localized delaminations in thin adhesive layers [1], from the active membrane protrusions in a cell [2] to the fluid-filled blisters in inflamed tissues and their physical counterparts in geophysics. How they nucleate, grow and eventually saturate is now a well-studied problem; following nucleation at a defect or debonded region, the growth of a blister is determined by the balance of the internal pressure, adhesive forces, mode mixity, film elasticity, and the deformation of the film and/or the substrate [3–10]. Indeed, the mechanism by which blisters form due to internal pressure is now a standard method to quantify the work of adhesion between the film and the substrate, a fundamental property of the interface. Once a blister forms, it does not always maintain itself and can slowly heal as the fluid permeates away from the localized zone of high pressure. However, how a blister heals does not seem to have been studied previously. Here, we investigate this process using a simple experiment that follows the disappearance of a blister trapped between a thin glass cover slide and a PDMS film, (a)E-mail: [email protected] relevant for any system where the escape or removal of fluid is essential to form a uniform adhesive bond. Experimental methods. – Our minimal system consists of an elastic plate (glass cover slide) that is gently placed onto a soft sticky substrate (10–240μm thick PDMS film). As the crack between the cover slide and the PDMS film heals (which occurs spontaneously after contact), the edge of a razor blade is used to guide the crack front to trap a pocket of air and form a blister that is initially axisymmetric. The evolution of the size and shape of the blister is recorded using an optical microscope equipped with a monochromatic light source (fig. 1(a)), using interference fringes (fig. 1(b)) to estimate the height profile of the blister, knowing that constructive interference bands occur at heights of (2n−1)λ 4 and destructive interference at nλ 2 , where n = 1, 2, 3 . . . and λ is the wavelength of light. We find that fingering instabilities form on the blister periphery for all film thicknesses. However, as shown in fig. 1(b), once the blister shrinks to a critical size, the distal tips of the fingers become pinned even as the fingers grow inward and the central part of the blister shrinks rapidly. This process forms a set of interconnected channels which display multiple branches. To understand the origin of this branching, we consider the