| AVS 66th International Symposium & Exhibition | |
| MEMS and NEMS Group | Monday Sessions |
| Session MN-MoM |
| Session: | MEMS, BioMEMS, and MEMS for Energy: Processes, Materials, and Devices I |
| Presenter: | Bogdan Vysotskyi, CEA/LETI-University Grenoble Alpes, France |
| Authors: | B. Vysotskyi, CEA/LETI-University Grenoble Alpes, France SH. Lai, CEA/IRIG-University Grenoble Alpes, France M. Defoort, CEA/LETI-University Grenoble Alpes, France M. Sansa, CEA/LETI-University Grenoble Alpes, France K. Clement, CEA/IRIG-University Grenoble Alpes, France M. Gely, CEA/LETI-University Grenoble Alpes, France C. Masselon, CEA/IRIG-University Grenoble Alpes, France S. Hentz, CEA/LETI-University Grenoble Alpes, France |
| Correspondent: | Click to Email |
Nanomechanical resonators have recently shown their potential to extend mass spectrometry towards a mass range inaccessible to commercial spectrometers [1]. The frequency shift-to-particle mass conversion requires precise knowledge of the resonator’s effective mass. When using a single resonator, uncertainty on the effective mass translates into a shift into central mass of the measured mass profile. If this resonator can be used for a large amount of time, time and effort can be spent into proper morphological characterization such as scanning electronic microscopy or local stress measurement. While these techniques can be suitable for MEMS-type devices [2], they prove much more complex and less effective in the case of nanomechanical resonators due to limited precision (c.a. 5 to 10nm). Moreover, the issue becomes way more acute when using arrays of resonators [3]: in this case, effective mass uncertainty and variability within the array leads to shifts in central mass, but also changes in mass profile. Lastly, routine particle measurements demand frequent changes in devices and time-effective calibration techniques are required. This crucial issue for mass spectrometry applications is very little discussed in the literature, or is addressed with complex procedures [4]. FEM simulations show that two main parameters impact effective mass assessment in the case of our monocrystalline silicon resonators (160nm thickness, 300nm width and c.a. 10um long): width and residual plane stress. The resonance frequencies of all resonators in the array are measured, thus both deviation from the theoretical frequency spacing and absolute frequency of our 20 resonators in the array are used for calibration of effective mass. A two-step optimization routine is used in conjunction with a physical model and internal stress and beam width are deduced. With this method an extremely low absolute mass error (<1%) is demonstrated to be reached. This non-destructive technique based on electrical measurement is amenable to the future use of very large arrays (>1000 resonators) for very short analysis time. This method can be extended for non-destructive characterization of nanomechanical resonators for different applications.
[1] S. Dominguez-Medina et al., Science 362, 918-922 (2018)
[2] A. Brenes et al., Mechanical Systems and Signal Processing 112, 10-21 (2018)
[3] E. Sage et al., Nature Communications 9 : 3283 (2018)
[4] O. Malvar et al., Nature Communications 7 : 13452 (2016).