Micro and Nanoelectromechanical Systems: A New Approach for Low Cost 6D Inertial Sensor

Philippe Robert, Patrice Rey, Patrice, Arnaud Walther, Guillaume Jourdan and Mylène Savoye, CEA-LETI
Corresponding author:


We are presenting a novel approach for very low cost 6D inertial sensor. This concept is based on the idea to mix on same device MEMS and NEMS technologies. The MEMS part is used for the mass to keep sufficient inertial force, and the NEMS is used as a very sensitive sub-µm suspended stress gage. This concept allows both in-plane and out-of-plane acceleration or Coriolis force detection on a same device. It is also compatible with differential detection to reduce thermal drifts.

Technological realization and first characterizations of 3-axis accelerometer and 3-axis gyrometer have been achieved and will be detailed in the presentation. On the basis of these results, one provides a footprint smaller than 3.5 mm2 for the integration of 3 axis accelerometer and 3 axis gyrometer on the same chip. To our knowledge, this level of integration and miniaturization has never been presented.


The MEMS market growth comes mainly from consumer market (cell phone, game, …). For this market, a very strong pressure is exerted on MEMS manufacturers. Typically 5 to 15% of cost reduction is expected each year for these components. At the end, a simple optimization of design and process will be insufficient and then a technological breakthrough is clearly expected to drastically miniaturize the MEMS sensors.

Nevertheless, this size reduction has major impacts on inertial sensor, in particular with regard to the performance: Reducing the seismic mass has a direct impact on the sensitivity, and lowers the nominal capacitance, with consequences on signal to noise ratio. To overcome these limitations, a new concept is proposed mixing micro and nanoscale structures, thus named M&NEMS. The basic idea is to combine on a same device a thick MEMS layer for the inertial mass, with a thin and narrow NEMS part to realize a suspended strain gage. A high sensitivity can be obtained due to the very high stress concentration induced by the very small cross-section of the silicon nanowire gage and also by the lever arm effect of the accelerometers and gyrometers designs (see Fig. 1). The two thicknesses of the M&NEMS approach offer also the ability to have on a same chip an in-plane and out-of-plane detection of the inertial mass movement (see Fig. 2). It means that with this concept and technology, inertial sensors can be integrated in less than 1 mm2 for 3D-accelerometer and less than 2.5 mm2 for the 3D-gyrometer.

Figure 1: Concept of the M&NEMS accelerometer: In-plane acceleration causes the mass to rotate around the rotating shaft that applies an axial stress in the NEMS suspended gage. This stress is amplified by a lever arm effect induced by the design (amplification x30), and also by the very small section of the gage increased by thinned suspended gage (magnification x5)
Figure 2: Concept of the M&NEMS out-plane accelerometer. In that configuration, a vertical acceleration causes the mass to rotate around the hinges. This rotation applies an axial stress in the NEMS suspended gage (as the gage is thinner than the mass). As for the in-plane case, this stress is amplified by a lever arm effect.


An example of in-plane accelerometer and X-axis gyrometer are shown in Fig. 3 and Fig. 5.

Figure 3: SEM view of a in-plane accelerometer
Figure 4: Electrical characterization of a 50g accelerometer (relative variation of the gage resistance vs acceleration)
Figure 5: SEM view of a Z-axis gyrometer

A focus on the gage lets clearly appear the MEMS inertial mass of 15µm thick, and the sub-µm gage that has a section of 0.25x0.25µm2. The 6 mask levels of the M&NEMS accelero and gyro technology will be detailed in the presentation (Fig. 7). This process is based on a SOI technology where the NEMS part is manufactured in the thin silicon active layer. The MEMS part is defined within a 15µm silicon epitaxial layer. The electrical characterizations of these two kinds of sensors are still in progress, but so far, all the measured parameters are in perfect agreement with the simulations.

Figure 6: Q-factor measurement on the Z-gyro
Figure 7: M&NEMS accelerometer process flow

It concerns:

  • The sensitivity (Fig. 4), linearity and thermal drift for the accelerometer;
  • The drive and sense resonant frequencies, Q-factors (Fig. 6), thermal and pressure behavior for the gyrometer.

Concerning the gyrometer, the sensitivity of the nano-gages is such that it can work in an open-loop mode. An operation in rough vacuum packaging (without getter) seems also very likely.


New designs and technological runs are in progress to go further in the development of this concept, in particular to integrate in the same flow a 3D magnetometer and a pressure sensor. The goal is to achieve at the end the demonstration of an IMU sensor module with nine-degrees-of-freedom (3-axis accelerometer + 3-axis gyrometer + 3-axis magnetometer) and a pressure sensor integrated on a same chip.


Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback