Nano-Newton sensitive MEMS Fx sensors made using CMOS-derived piezoresistors and HAR 3D-printed metal and polymer beams_IshaLodhi et al.
Description
Force characterization data of MEMS pillar-based in-plane force sensors. Active stress-sensing elements are CMOS-derived diffused N-type silicon piezoresistors of widths ranging 300 nm to 1 micron, (length =1 micron for all). The silicon-based, substrate-diffused, piezoresistors have a fan-out "butterfly" design which allows connection of larger metal contacts (1x1 um2 to 2x2um2) to sub-micron width devices. The Fx sensor pillars are made by two different approaches, and both using additive technologies. The first MEMS Fx sensor type is made with 3D-metal-printed high-aspect-ratio (HAR) pillars. We use local metal electrodeposition for 3D-metal-printing (Exaddon Ceres AG, Switzerland). The active piezoresistor microdevices and chips were modified for compatibility with the acidic electrochemical metal deposition process and connection of print surfaces to the electrodeposition chamber work electrode. (WE). The second MEMS Fx sensor presented have pillars exposed in a two-photon polymerization process (Nanoscribe GmbH), and open, unfilled, trenches around the piezoresistor for mechanical stress-concentration in the piezoresistor body. The pillar material for the second type of sensor pillars is the photo-sensitive acrylic-based polymer, IP-Dip (Nanoscribe). This dataset includes experiment data from electromechanical characterization of both sensor types, with (1) HAR copper (Cu) pillars and, (2) polymer TPP pillars. Sinusoidal force (Fx) waveforms are applied against pillar tips, and sensor resistance is recorded simultaneously (sample interval =0.2s). Sensitivity is measured as the fractional change of resistance (delta-R/R0) of the piezoresistive Fx sensors for approaching, and retreating force directions, as well as for varying force application speeds from 0.01 micro-Newton/s to 1.7 micro-Newton/s. A separate sheet includes all COMSOL FEA simulation study results for the 3D piezoresistive multiphysics models of both sensors. Sampling rate of the applied pillar deflection (displacement of the piezoactuated stage, delta-X) depends on wave-table-rate applied. Force (Fx) is extracted by scaling deflection applied on pillar tip with COMSOL simulated pillar spring constants in K(N/m) using SEM-measured pillar geometries and Young's modulus of 130 GPa from literature. Fx sensitivity responses are repeatable with low hysteresis, and have maximum Fx range determined by pillar delamination force (> 3 micro-Newton). Highest sensitivity measured for the HAR Cu pillar sensors was 0.26% per micro-Newton. Best sensitivity measured with the open trenches and TPP pillars was 0.44% per micro-Newton. At the highest Fx force application rate, sensitivities only drops up to 30%.
Files
Institutions
- Georgia Institute of Technology
- Georgia Institute of Technology School of Electrical and Computer Engineering