Abstract

Introduction State of the art MEMS devices, such as inertial sensors, consist of very fragile, sensitive elements which should be protected by a cap at the wafer level, to allow standard chip dicing and assembly technologies. Typically, wafer bonding is used for this capping. Since, in the bonding process, the type and pressure of the gas can be controlled, the sealed gas in the MEMS device can be tuned and used for device functionality. This is, for example, very important for resonant MEMS devices such as gyroscopes. These essentially require a high Q-factor, which can only be achieved at low cavity pressures. The actual sealed pressure depends on many factors, such as the chamber pressure during bonding, out-gassing from the bonding material (e.g. glass frit), as well as from the sealing behaviour of the bonding process used. Concepts for MEMS Inner Pressure Measurement At the last ECS Semiconductor Wafer Bonding Symposium in 2012, an approach was introduced for measuring the MEMS cavity inner pressure using the thermal conductivity of the sealed gas, which is strongly dependent on the absolute pressure [1]. While this type of measurement principle had already been previously reported in the literature [2], a special test structure, adapted to a MEMS foundry surface micromachining process [3] for inertial sensors, was described, which allows high-resolution measurement. The vehicle itself is very small and can be easily integrated into mask sets as special test elements and even into functional chips for lifetime monitoring. The measurement principle is based on a heater and temperature measurement resistors which can detect the amount of heat transferred from the heater and, by this, measure the inner cavity pressure. The measurement principle was the subject of the original publication [1], but this paper reports various new measurement results. Results of Inner Cavity Pressure Measurement In the MEMS Foundry Open Platform technology [3] the cavity pressure measurement device is fabricated and used for glass frit bonding which is used to seal the devices (gyroscopes or acceleration sensors). This bonding process is still very attractive regarding cost and process integration, since almost any surface layer and surface profile arising from metal wiring can be bonded and hermetically sealed. However, the glass frit bonding shows some specific behaviour regarding the sealed pressure, since a small amount of outgassing of remaining organics results a slightly higher pressure in the sealed MEMS cavity than set by the pressure in the wafer bonding equipment chamber. Hence, the essential question is: what pressure is actually sealed in the MEMS device? This is also very interesting in relation to the bonding temperature, which is at ~450°C, so that the sealed gas is cooled down after bonding which reduces the sealed pressure at room temperature by a factor of ~2.7, related to the classical thermodynamic ideal gas conditions. These questions were investigated by applying different pressures in the bonding chamber, and measuring the MEMS inner cavity pressure after finishing the glass frit wafer bonding and cooling down the wafer stack.These investigations have confirmed the theory and helped to set up the correct bond chamber pressure for achieving the targeted MEMS cavity pressure. The glass frit material is known to be hygroscopic based on its glass composition, so that it can absorb humidity from air if it is stored after its full glazing cycle. The relevance of this phenomenon was investigated using the inner cavity pressure measurement structure. Cap wafers with completely-processed glass frit were stored in air for different times, and, indeed, different MEMS cavity pressures were measured after bonding. With longer storage time, humidity is absorbed into the glass which outgases into the cavity during bonding, and increases the cavity pressure. Conclusions The heat-transfer-based inner cavity measurement structure is well-suited for monitoring the sealed pressure of MEMS structures in production processes, reliability testing and over the lifetime, as well as for in-depth investigation of wafer bonding processes – further examples will be reported in the full paper.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.