More in

Metrology research aims to improve the commercial application of MEMS

4 mins read

Today's applications for MEMS are hugely varied. They range from accelerometers in automotive airbag deployment systems, detecting the rapid negative acceleration of the vehicle, to inkjet printer heads, reacting to patterns of heat from electric current by dispensing tiny droplets of ink at precise locations to form the image on the paper. They are also seen in smartphones, measuring the rotation of the device to create an intuitive user interface.

Despite these niche applications, the commercial growth of MEMS devices since their first development 50 years ago hasn't met expectations. Research by the Metrology for Energy Harvesting Project, backed by the European Metrology Research Programme (EMRP), is looking to address this by enabling 'joined up' collaborative research across Europe. Through this, a suite of complementary measurement techniques has been produced for the assessment of microscale energy harvesting performance, as well as for broader applications in the electrical and functional characterisation of MEMS devices. Using MEMS for energy harvesting is critical because we need to try to ensure the power source can be integrated with the device we are powering through the harvested energy. This entails either the development of devices that use existing MEMS fabrication techniques or which integrate functional materials into silicon MEMS fabrication. The National Physical Laboratory (NPL) – the UK's National Measurement Institute – has been undertaking cutting edge work into piezoMEMS metrology. PiezoMEMS supports many rapidly growing markets, with applications including sensing, resonators, rf switches and energy harvesting. For example, piezoelectrically transduced piezoMEMS resonators are seen as potential replacements for quartz technology in high frequency, low phase noise reference applications. To date, the only fully commercialised piezoMEMS products are oscillators and inkjet printer heads. One of the reasons behind this is the requirement for high volume production of well characterised, good quality functional films – such as piezoelectric or ferroelectric films. The advance in processing methodologies has launched ferroelectric films into many MEMS based applications, including rf MEMS, ultrasonic infrared sensors, and accelerometers. The production of ferroelectric thin films is expected to rise from 881,000 6in equivalent wafers in 2010 to 1.26million in 2015. Inkjet print head and capacitors are expected to consume more than 90% of this total in 2013. Multinational companies using (and investing in) piezo thin films include NXP, STMicroelectronics, Epson, Panasonic, Fujitsu, Oki and Ramtron. NPL has developed a new MEMS device which could be used for the in situ rapid evaluation of the piezoelectric properties of thin films or micro structured components. While this measures the properties of piezoelectric materials and devices at the microscale, NPL plans to scale the device down further to allow the examination of ferroelectric scaling phenomena and to probe electromechanical coupling across individual ceramic grains. Another barrier to the widescale commercial adoption of MEMS is a lack of understanding of the potential power requirements and outputs of devices. The problem for many MEMS applications is that mechanical components must be embedded in a protective wafer level package. Unless you can access the mechanical system itself without breaking it, developers and customers may find it hard to understand how to best use their product. To address this issue, EMRP scientists in France have recently developed a new way to measure the power requirements and outputs of existing and future MEMS devices accurately. They have achieved this using electrical methods to characterise electromechanical performance of electrostatic MEMS. The cheap and easy to apply technique was developed by a research team from Laboratoire national de métrologie et d'essais (LNE) in France. Lead researcher Dr Alexandre Bounouh and his team developed an experimental set up to gain accurate information on the mechanical values and properties of any MEMS device through electrical measurement. The researchers believe this will help manufacturers to improve product performance, develop new functionalities, reduce energy consumption of mass production, respond to market demands for miniaturisation and increase reliability of MEMS devices. Dr Bounouh's technique works by applying a current with a varying frequency across the device, followed by analysis of the harmonic content of the output voltage of the component parts. With some additional calculations, the technique can determine electrically all the mechanical characteristics of the MEMS device, including the damping factor (a negative impact on the amplitude of oscillations) and the frequency at which the maximum amount of electrical power is generated by the mechanical vibrations of MEMS transducers. Since its development, several MEMS devices have been tested at LNE using the technique and their mechanical resonant frequencies have been measured with only a tiny uncertainty. In future, Dr Bounouh and his colleagues believe the technique may be used to provide feedback on production methods that will allow manufacturers to design MEMS to meet the needs of each system in which they operate. More accurate knowledge of the product output and energy requirements will also affect the choice of device from potential consumers, who will be able to select devices with performance optimised for their particular sector. This accurate and traceable technique could be implemented for online production tests and measurements. It could deliver key competitive edge to EU companies and support large scale manufacturing excellence by introducing metrological principles into industrial processes. Finally, another NPL project is bringing everything together through the development of an energy harvesting workbench. This is a multifrequency vibrational energy harvesting facility with integrated and traceable charge and vibrational amplitude measurement capability. It is designed to measure mechanical performance and electrical power output and provide the link between the electrical and mechanical regimes. It can be applied to electrostatic or piezoelectric energy harvesters. NPL's energy harvesting workbench allows independent three axis vibrational excitation of MEMS based energy harvesting structures. The workbench integrates with the NPL's laser Doppler vibrometer (LDV), allowing measurement of the complex vibration profiles of energy harvesting devices in vacuum. This will provide a unique capability to relate the microscale complex device vibration profile to the energy harvesting output under complex broadband vibrational input. NPL and LNE are two of seven national research centres across Europe that make up the Metrology for Energy Harvesting Project. The project, jointly funded by the EMRP participating countries within EURAMET and the European Union through the EMRP, represents the first coordinated international attempt to apply the principles of metrology (measurement science) to energy harvesting products and materials. As we have shown, this has important implications in how MEMS devices are developed and characterised. The suite of metrology tools developed across the EMRP project can help European industry to commercialise new products through integrated piezoelectric and electrostatic MEMS energy harvesters. This is a hugely important area of growth for Europe in a sector that has already seen the US steal a march through Microgen's announcement of commercial scale production in this area. Cross border work on projects like EMRP offers the best way to rise to this challenge. Paul Weaver and Jenny Wooldridge are researchers with NPL.