Repetitive ultramicrotome trimming and SEM imaging for characterizing printed multilayer structures

Ultramicrotomy is a well-established technique that has been applied in biology and medical research to produce thin sections or a blockface of an embedded sample for microscopy. Recently, this technique has also been applied in materials science or micro- and nanotechnology as a sample preparation method for subsequent characterization. In this work, an application of ultramicrotomy for the cross-section preparation of an inkjet-printed multilayer structure is demonstrated. The investigated device is a capacitor consisting of three layers. The top and bottom electrodes are printed with silver nanoparticle ink and the dielectric layer with a ceramic nanoparticle/polymer ink. A 3D profilometer is initially used to study the surface morphology of the printed multilayer. The measurements show that both electrodes exhibit a coffee-ring effect, which results in an inhomogeneous layer structure of the device. To obtain precise 3D information on the multilayer, cross-sections must be prepared. Argon ion beam milling is the current gold standard to produce a single cross-section in good quality, however, the cross-section position within the multilayer volume is poorly defined. Moreover, the milling process requires a significant investment of time and resources. Herein, we develop an efficient method to realize repetitive cross-section preparation at well-defined positions in the multilayer volume. Repetitive cross-sections are exposed by trimming with an ultramicrotome (UM) and this blockface is subsequently transferred into a scanning electron microscope (SEM) for imaging. A combination of custom-modified UM and SEM specimen holders allows repeated transfer of the clamped multilayer sample between instruments without damage and with high positioning accuracy. This novel approach enhances the combination of an established ultramicrotome and a SEM for multilayer sample volume investigation. Thus, a comprehensive understanding of printed multilayer structures can be gained, to derive insights for optimization of device architecture and printing process.