Listed below are the publications related to the MUPIA project
Sordo et al. "Through Silicon Vias in MEMS packaging, a review", NordPac 2019, June 11-13, Lyngby, Denmark
Wright et al. "Development of mechanically compliant flip chip interconnect using single metal coated polymer spheres", EMPC 2019, September 16-19, Pisa, Italy
Sordo et al. "Wafer bonding process for zero level vacuum packaging of MEMS", ESTC2020, September 15-18, Online
Public Summary of MUPIA project
The MUPIA project addressed the topic Manufacturing process for ultimate performance inertial MEMS Gyrometer (JTI-CS2-2017-CfP06-SYS-01-04) within the Programme Area SYS of the Clean Sky 2 program. (MEMS = Micro Electro-Mechanical Systems). MUPIA answered the request for "Silicon Very High Performances MEMS Gyrometers Technology" by merging SINTEF, a research institute and experienced manufacturer of high-performance MEMS and MICROSS, a company with expertise in assembling bare dies into standard and custom ceramic packages for demanding applications. The consortium set to answer the challenge of bringing the manufacturing process for the gyrometer up to a manufacturing readiness level of 5 by exploiting systematic process development and optimization procedures according to their ISO certified Management Systems.
An extensive improvement of technology is foreseen for future aircrafts, primarily to reduce weight, reduce fuel consumption, related CO2 and NOx emissions and increase reliability. Improved sensors and actuators are foreseen to be used more extensively, using newer technology. Innovative wing systems and more extensive condition monitoring are foreseen examples, reducing also cost of maintenance and improving safety in addition to contributing to reduced fuel consumption.
In the end, the results from the project will have a strong positive impact on the accessibility of cost-efficient high-end MEMS gyrometers in the European market in addition to strengthening the position of the involved partners among the key European providers of high-performance components.
The main objective of MUPIA was to demonstrate a cost-efficient manufacturing process for this sensor element and its package within a sustainable value chain. To fulfil this objective, the MUPIA project set and met many sub-objectives relating to design, silicon processing, assembly and packaging, as detailed below.
The project lasted 39 months and was completed May 2021.
The initial gyro design was integrated into SINTEF's design rules and optimised to meet the Topic Manager's (TM) specifications with respect to electrical characteristics. A suitable ceramic package was identified for packaging purposes.
The next tasks concentrated on setting up the industrially viable processes that would be used for manufacturing the MEMS gyrometer:
- accomplishing high resolution photolithography for the patterning of the active MEMS element
- direct wafer bonding to enable wafer level vacuum sealed MEMS units
- Through silicon vias to enable electrical connection to the device
High resolution patterning was realised by optimalisation of the photolithography process and the demanding etch process that defines the working element of the MEMS.
Typically, direct bonding is done as early in the processing as possible since it requires a very flat and clean surface to succeed. In MUPIA, however, the direct bonding would be done after substantial processing. For this reason it was necessary to have a contingency design that was not dependent on successful direct bonding. The project decided, therefore, to manufacture two variants of the MUPIA MEMS,
- MUPIA Open Design (MOD)
- MUPIA Closed Design (COD)
The MOD variant was processed such that the active MEMS element was exposed to the ambient and therefore needed to be vacuum sealed at package level. The vacuum sealing for the COD variant would be done at wafer level and therefore did not demand vacuum sealing at package level.
The first deliverable of gyrometers consisted of the MOD variant. Due to Covid19, the full characterisation of the MEMS the TM's facilities was delayed significantly. However, simplified testing showed promising results. The main issues were due to insufficient vacuum in the package which in turn lowers the Quality factor and decreases the performance potential of the MEMS gyrometer.
The final delivery was intended to consist of the COD variant due to its inherent advantages. However, the vacuum sealing at wafer level proved more challenging than initially anticipated resulting in significant delay in the processing of the COD variant. In addition, a processing error led to the TSVs not having electrical contact with the device. As a result, the MOD variant was invoked as a contingency for the final delivery of the MEMS gyrometer.
Detailed characterisation was carried by the TM out on a subset of the final delivery. The results showed at least a 70% functionality yield with resonance frequencies and x-axis quality factor within the expected ranges. Other measured parameters suggested that the MEMS processing and packaging could benefit from further optimisation in etching uniformity and vacuum sealing, respectively.
• Through Silicon Vias in MEMS packaging, a review, Guido Sordo, NordPac (IMAPS Nordic / IEEE EPS), June 11-13, 2019, Copenhagen, Denmark
• Development of mechanically compliant flip chip interconnect using single metal coated polymer spheres, Daniel Nilsen Wright, 22nd Microelectronics and Packaging Conference & Exhibition (EMPC, IMAPs Europe), September 16-19, 2019 ,Pisa, Italy
• Wafer bonding process for zero level vacuum packaging of MEMS, Guido Sordo, European Systems-Integration Technology conference (ESTC, IEEE EPS), September 15-18, 2022, Online
• Based on the experience from MUPIA, SINTEF is currently developing a MEMS Strain Gauge that will implement vacuum sealing at wafer level.
• Through MUPIA, Micross expects to increase its UK based capability and technical knowledge of this branch of semiconductor packaging and thus be in a position to introduce the increased capacity necessary to service an increasing market demand for such specialist packaging.
Progress beyond the state of the art
The MUPIA project exploited a newly available state-of-the-art wafers service from a well known silicon wafer supplier to buy silicon wafers with prefabricated cavities for gyrometer fabrication. This, in combination with improved in-house fabrication processes at SINTEF is progress beyond state-of-the-art in the MEMS fabrication industry. A process was developed where all dies across the wafer were successfully etched without the typical damages, as notching and wall etch damages.
Direct bonding of silicon wafers in combination with the gyro and TSV structures was successfully implemented, allowing for vacuum sealing of each individual component on wafer level. The wafer level bonding process was already established at SINTEF at the start of the project, but the complexity of the process used for the gyro component has gained new insight.
The combined new process knowledge can be widely used for development and production of other silicon-based devices for a large range of applications, like optics and microfluidics. SINTEF offers its processing services to National and European industry, with a large potential impact.
Micross has established a value chain with SAES (www.saesgetters.com) to offer high vacuum level sealed ceramic packages. This has expanded Micross' technology offer for the European and international microelectronics industry.
In conclusion, the results from MUPIA will contribute to a greener environment indirectly through contributing to the possible realization of the SYSTEM ITDs.
Relevant results were disseminated at conferences and the project web site.