Ehrenfried Zschech, Sven Niese1, Kristina Kutukova, Juergen Gluch

Fraunhofer IKTS Dresden, Germany
1 now with AXO Dresden GmbH, Dresden, Germany

The study of the fracture behaviour of multi-component materials is of high interest in materials science and engineering, and the nondestructive 3D imaging of crack propagation in mechanically stressed samples, both bulk samples (e.g. composites) and thin film stacks (e.g. 3D patterned microelectronic structures), with high spatial resolution is essential for the validation of models based on materials physics and chemistry as well as fracture mechanics, and eventually for the design of advanced materials with tailored mechanical properties. In addition, it is needed for gauging the in-service mechanical performance and reliability of advanced products and systems.

Laboratory transmission X-ray microscopy (TXM) and nano X-ray computed tomography (nano-XCT) offer intrinsic advantages for 3D in-situ imaging of cracks in multi-component materials and structures, since it is nondestructive, i.e. the local strain state is not modified by sample preparation. In this talk, we present a novel micro-double cantilever beam (micro-DCB) test in an X-ray microscope for a nondestructive 3D visualization of crack propagation with high spatial resolution during mechanical loading. This approach allows to identify the weakest (cohesive or adhesive) bindings in the material or of an interface and it provides several benefits over existing methods. It is applicable across a range of disciplines, from materials science and microelectronics to life sciences.

The potential of the novel approach is demonstrated for two exaamples: 1) Nafion®, a synthetic polymer, filled with platinum particles: The mechanical stability of Nafion®, which is caused by the polymer backbones, is of particular importance for the use as solid polymer electrolyte (proton conductor) in proton exchange membrane (PEM) fuel cells, e.g. for automotive applications. 2) an on-chip interconnect stack of an integrated circuit: The robustness of the backend-of-line (BEoL) stack against chip-package interaction (CPI) is evaluated, and the weakest layers and interfaces are identified, which is valuable information for reliability engineering and design-for-reliability (DFR) in semiconductor industry.