Microelectromechanical systems (MEMS) devices are manufactured utilizing micromachining techniques that are semiconductor-based. They use piezoelectric, electrostatic, magnetic, or thermal methods to manage a micron-scale movable component.
Billions of MEMS transducers and sensors have been manufactured each year since the early 1980s for every aspect of daily life. They can be found, for example, in consumer devices (microphones and accelerometers), vehicles (inertial and pressure sensors), and digital projection systems (optical micromirrors).
The Tyndall National Institute is a leading research center in Europe, specializing in Information and Communications Technology (ICT) research and development, and is the largest facility of its kind in Ireland. It was established with a mission to help academia and industry in driving research to market.
Tyndall has extensive R&D activity concentrated on the development of MEMS devices, mainly for industries covering medical devices, electronics, communication, and energy. As a part of this pursuit, non-harmful optical characterization methods are frequently employed for static, dynamic, and high-resolution characterization of these structures.
This article presents the Polytec laser vibrometry facility at Tyndall and conveys its use in two applications, specifically piezoelectric MEMS cantilevers for energy harvesting, and radio frequency (RF) MEMS for telecommunications.
The laboratory is furnished with a 30 MHz Polytec MSA-400 (new version MSA-600 up to 2.5 GHz was introduced in 2019) with lenses varying from 1X to 50X, mounted on a TS-140 (Table Stable Ltd.) active anti-vibration table for highly accurate analysis, as shown in Figure 1.
A custom-built, 150-mm diameter vacuum chamber with a glass lid is also fitted in the system, which is interfaced to an Edwards TY1A12311 turbo pump. Along with containing multiple feedthroughs for electrical interfaces, the chamber has an Olympus V101-RM ultrasonic transducer for excitation and an AXL345 accelerometer for chamber motion sensing.
The high voltages commonly required for MEMS characterization is supplied by a Falco Systems WMA-300 voltage amplifier. The system enables complete in-plane, out-of-plane, and topographical analysis of MEMS structures at pressures ranging from less than 0.01 mbar to atmospheric.
Figure 1. Left: Polytec MSA-400 at Tyndall, fitted with a custom vacuum chamber, active isolation table, and a high-voltage amplifier. Right: closer view of the vacuum chamber. A micromachined energy harvesting module is under test; a nearby ADXL345 inertial sensor is utilized to correlate excitation levels with transducer output.
RF MEMS components, for example resonators, varactors, and switches, employ electrostatic actuation to adjust the position of a micron-scale, a mechanical element suspended over a transmission line, therefore changing the features of that line and the circuit where it is embedded.
RF MEMS devices provide extraordinary RF performance. They are lightweight, small, and have a high capability for integration. They are enticing options for use in applications like radio front-ends, phase-shifting circuits, reconfigurable antennas, and impedance matching units.
A key feature of a microswitch is the useful mechanical stiffness (k) of the movable electrode, which establishes the voltage needed to actuate the switch. Generally, this effective stiffness is a result of the material properties (material intrinsic stress and Young’s modulus) and the geometry of the movable electrode.
It may be experimentally determined by analyzing the mechanical resonance frequency (f) of an electrode of mass m, and utilizing the expression k = (2πf)2 m.
Figure 2 demonstrates mechanical resonance frequency data from 100 µm square aluminum electrodes suspended employing three types of spring: meander, spiral, and straight springs. The resonance frequency (and stiffness) is greatly dependent on the geometry of the suspension tethers.
The ambient environment (gas pressure and concentration) around a MEMS device may significantly impact peak reliability and performance. The effects of air damping on a straightforward MEMS cantilever-based resonator is shown in Figure 3 utilizing a plot of vibration amplitude at different pressure levels.
From the results, it is clear that to enable a high Q-factor of this resonator, the ambient pressure level in the package must be controlled below a particular critical level, which in the case of this specific device is around 1mbar.
Figure 2. 100 µm square RF-MEMS capacitive switches exhibiting three alternate spring designs, and the mechanical resonance frequencies measured.
Figure 3. Measured resonance frequency of MEMS resonating cantilever compared to pressure (left) and corresponding features of the mechanical quality factor and resonance frequency.
Piezoelectric Energy Harvesting
Converting freely available kinetic energy sourced from the environment into electrical energy, energy harvesters can be utilized to fuel autonomous low-power electrical systems, for example, wireless sensor nodes (WSNs) in structural, machine health monitoring, or healthcare applications.
The harvester alone is a mechanical system with a resonance frequency aligned to a spectral peak given by the environment housing the application. Therefore, when the harvester is excited into resonance by its environment, it performs as a signal amplifier for the external vibration at that specific frequency.
The harvester must then be able to convert the vibration energy into pragmatic electrical energy. Piezoelectric harvesters do this by making use of piezoelectric materials’ ability to gather charges on their crystal faces when mechanically stressed.
Tyndall has designed and engineered MEMS harvesters with a target resonance frequency of less than 150 Hz. The devices were manufactured employing both surface and bulk silicon micromachining technologies.
In cross-section, Figure 4 outlines the profile of a silicon-based, cantilever-type harvester. The beams constituting the system spring must be long (~ 8 mm - 10 mm) and thin (<50 µm) to achieve resonances of less than 150 Hz. The complete wafer thickness is utilized to form the mass.
The low power output and intrinsically narrow bandwidth are two of the key challenges faced by energy harvesters. One way to correct these challenges is to electrically connect several harvesters, thereby combining the output of devices.
This can be performed by using parallel or series configurations (or a combination of the two). The outcomes for three cantilever harvesters with identical masses but varying beam designs are demonstrated in Figure 5.
These devices were engineered to resonate at around the same frequency. The resonance frequency of the cantilever beams was optically measured utilizing laser vibrometry. The devices resonate from 115 Hz to 118 Hz with displacements of multiple micrometers. The measurements were recorded utilizing the setup outlined in Figure 1.
Figure 4. Schematic demonstrating the cross-section of a silicon-based piezoelectric energy harvester.
Figure 5. Resonance peaks analyzed by the vibrometer for three cantilever structures with slightly different beam designs.
This information has been sourced, reviewed and adapted from materials provided by Polytec.
For more information on this source, please visit Polytec.