Aidan Arthur Taylor

Microstructure and adhesion of as-deposited and annealed Cu/Ti films on polyimide

Abstract The ability to measure the adhesion energy of metal thin films on polymer substrates is important for the design of reliable flexible electronic devices. One technique is to create well-defined areas of delamination (buckles) as a consequence of lateral compressive stresses induced by tensile straining of the film–substrate system. The adhesion energy is calculated from the buckle dimensions. In order to improve the adhesion between the metal film and polymer substrate, thin adhesion layers can be incorporated. However, interdiffusion and reactions can occur between the adhesion layer and the metal film when subjected to elevated temperatures. This is detrimental for the interfacial adhesion, as will be discussed for Cu films on polyimide with a Ti interlayer subjected to annealing at 350°C.

Robust mechanical performance of chromium-coated polyethylene terephthalate over a broad range of conditions

Mechanical properties of metal films on polymer substrates are normally studied in terms of the fracture and adhesion of the film, while the properties of the polymer substrate and testing conditions are overlooked. Substrate orientation and thickness, as well as strain rate and temperature effects, are examined using Cr films deposited onto polyethylene terephthalate substrates. A faster strain rate affects only the initial fracture strain of the Cr film and not the crack and buckle spacings in the high strain condition. The substrate orientation slightly changes the average crack spacing while the substrate thickness has little effect on the cracking and buckling behaviour. Straining experiments at high temperature increased the average crack spacing and led to a change in buckling mode. The lack of sizeable changes in the mechanical behaviour over the large range of testing procedures leads to a resilient material system for flexible applications.

A Mechanical Method for Preparing TEM Samples from Brittle Films on Compliant Substrates

Abstract Preparing transmission electron microscopy (TEM) samples from thin films is technically challenging and traditional preparation routes can sometimes introduce unacceptable artefacts or even prove impossible. A novel method of preparing plan view TEM samples from thin films by a purely mechanical method is assessed. Two examples of films prepared by this route are briefly presented, a Cr film on PET and an amorphous AlxOy film on Cu. The application of this method allows for TEM analysis of the Cr film without the problems associated with a polymer such as PET disintegrating under the electron beam. For the AlxOy films it is demonstrated that this purely mechanical preparation prevents crystallisation of the film resulting from conventional ion milling preparation routes. The technique also allows for an upper bound of thickness approximation for these films.

On the limits of the interfacial yield model for fragmentation testing of brittle films on polymer substrates

Fragmentation testing is frequently used to probe film fracture strain and the interfacial properties of thin brittle films on compliant substrates. A model based upon complete yield of the film/substrate interface is frequently used to analyse data after cracking has saturated. Additionally, the film is either assumed to have a single-valued failure stress or a distribution of strengths described by Weibull statistics. Recent work by the authors showed that consideration of film thickness variations and the application of neighbour ratio analysis brought 96% of the data for an Al x O y /Cu film/substrate system into compliance with the predictions for a film with a single-valued failure stress. In the present work Cr/PI (polyimide) and Cr/PET (polyethylene teraphthalate) systems are analysed according to the same methodology. The Cr films on polymer substrates crack such that the neighbour ratios considerably exceed the predicted limit of 2. The influence of the relative thickness of the film and substrate and the strain rate of the test is investigated. A deviation from the idealised mechanical model due to the large difference in elastic moduli of film and substrate is put forward as a possible cause of the observed behaviour. The importance of these results to the application of the interfacial yield model is discussed.