No abstract

Published by the AIP PublishingArticles you may be interested in Electrodeposition and characterization of magnetostrictive galfenol (FeGa) thin films for use in microelectromechanical systemsMicropatterning of metal films coated on polymer surfaces with epoxy mold and its application to organic field effect transistor fabrication

No abstract

The adhesion strength between a thin film and substrate is often the critical parameter that controls the initiation as well as the mode of film failure. In this work, a laser-based spallation method is used to determine the adhesion strength of "as deposited" lead zirconate titanate (PZT) sol-gel thin films on the two functionally different substrates. For the first case, PZT sol-gel film is deposited onto bare Si/SiO 2 substrates via spin-casting. The extremely high adhesion strength between the film and the substrate necessitated an additional platinum mass superlayer to be deposited on top of the PZT film in order to induce interfacial failure. For the superlayer film system, a hybrid experimental/numerical method is employed for determining the substrate/film interfacial strength, quantified to be in the range of 460-480 MPa. A second substrate variation with lower adhesion strength is also prepared by applying a self-assembled octadecyltrichlorosilane (ODS) monolayer to the Si/SiO 2 substrate prior to the film deposition. For the monolayer-coated substrate case the adhesion strength is observed to be significantly lower (54.7 MPa) when compared to the earlier case.

Adhesive failure and the attendant delamination of a thin film on a substrate is controlled by the fracture energy required to propagate a crack along the interface. Numerous testing protocols have been introduced to characterize this critical property, but are limited by difficulties associated with applying precise loads, introducing well-defined pre-cracks, tedious sample preparation and complex analysis of plastic deformation in the films. The quasi-static four-point bend test is widely accepted in the microelectronics industry as the standard for measuring adhesion properties for a range of multilayer thin film systems. Dynamic delamination methods, which use laser-induced stress waves to rapidly load the thin film interface, have recently been offered as an alternative method for extracting interfacial fracture energy. In this work, the interfacial fracture energy of an aluminium (Al) thin film on a silicon (Si) substrate is determined for a range of dynamic loading conditions and compared with values measured under quasi-static conditions in a four-point bend test. Controlled dynamic delamination of the Al/Si interface is achieved by efficient conversion of the kinetic energy associated with a laser-induced stress wave into fracture energy. By varying the laser fluence, the fracture energy is investigated over a range of stress pulse amplitudes and velocities. For lower amplitudes of the stress wave, the fracture energy is nearly constant and compares favourably with the critical fracture energy obtained using the four-point bend technique, about 2.5 J m−2. As the pulse amplitude increases, however, a rate dependence of the dynamic fracture energy is observed. The fracture energy increases almost linearly with pulse amplitude until reaching a plateau value of about 6.0 J m−2.

A spatially varying cohesive failure model is used to simulate quasi-static fracture in functionally graded polymers. A key aspect of this paper is that all mechanical properties and cohesive parameters entering the analysis are derived experimentally from full-scale fracture tests allowing for a fit of only the shape of the cohesive law to experimental data. The paper also summarizes the semi-implicit implementation of the cohesive model into a cohesive-volumetric finite element framework used to predict the quasi-static crack initiation and subsequent propagation in the presence of material gradients.

The dynamic fracture of functionally graded materials (FGMs) is modeled using an explicit cohesive volumetric finite element scheme that incorporates spatially varying constitutive and failure properties. The cohesive element response is described by a rate-independent bilinear cohesive failure model between the cohesive traction acting along the cohesive zone and the associated crack opening displacement. A detailed convergence analysis is conducted to quantify the effect of the material gradient on the ability of the numerical scheme to capture elastodynamic wave propagation. To validate the numerical scheme, we simulate dynamic fracture experiments performed on model FGM compact tension specimens made of a polyester resin with varying amounts of plasticizer. The cohesive finite element scheme is then used in a parametric study of mode I dynamic failure of a Ti/TiB FGM, with special emphasis on the effect of the material gradient on the initiation, propagation and arrest of the crack.

No abstract