MEMS Platforms for Automated and High-Throughput Micromechanical Testing of Silicon Nanowires

Abstract

Silicon nanowires (SiNWs) are key to nanoelectronics, optoelectronics, and sensors due to their unique electrical, mechanical, and optical properties. The broad potential applications of NWs have driven the need for advanced mechanical testing methods. Traditional testing techniques, including MEMS-based platforms, face challenges in (i) manipulating and aligning nanoscale samples and (ii) imposing a specific stress state, both contributing to inconsistencies, i.e., a large scatter in reported mechanical properties. Co-fabricating the sample with the MEMS platform addresses the first challenge by ensuring precise alignment. The second challenge, however, requires a more versatile method that can impose the desired stress state while accounting for its impact on failure mechanisms. i) Uniaxial testing, though straightforward for mechanical property extraction, applies uniform stress across the cross-section, increasing the risk of brittle failure from pre-existing or fabrication-induced defects. ii) Transverse loading, while altering stress distribution, still introduces a uniform axial stress component that can contribute to failure. iii) In contrast, pure bending eliminates axial stress, localizing the highest stresses within a confined region, offering a more reliable approach for exploring deformation behavior. Each method has its strengths and limitations. Hence, a combination of all loading configurations ensuring consistent initial conditions and surface states is essential to provide a better understanding of SiNW mechanics. This study introduces a set of MEMS-based testing platforms, which are monolithically integrated with SiNWs. Twelve MEMS stages (four for each loading configuration) were fabricated on the same chip, with each stage containing one SiNW sample. A commercial micromechanical testing equipment (FemtoTools) with a micromanipulator tip is used to apply load in an automated fashion. The proposed fabrication and automated characterization approach reduces potential errors during the process and minimizes testing time. This technique is also applicable to other 1D nanomaterials, such as metal oxide NWs, for high-throughput mechanical characterization.