International MEMS/MST Industry Forum - Abstracts and Biographies 2012
Leopold Beer, Director Marketing
Bosch Sensortec is a subsidiary of the Bosch Automotive Division focusing specifically on consumer electronics applications for MEMS inertial, pressure & geomagnetic sensors.
While Bosch is leading the MEMS sensor market globally, Bosch Sensortec achieved a leading position in consumer electronics applications.
Today Bosch Sensortec is the only supplier having accelerometer, gyroscope, geomagnetic and barometric pressure sensor technologies in-house.
This presentation will address the current status of MEMS sensor integration in mobile devices by highlighting system requirements and subsequent sensor requirements. Further the evolution of MEMS sensor integration in mobile devices will be projected based on overall system optimization aspects.
Some of the addressed aspects:
1. 10 DoF Sensor Data Fusion: Challenges & Solutions for mobile device manufacturers.
2. Requirements for employed technologies & derived sensor parameters.
3. Physical versus functional integration – KSF`s and market pull.
MEMS sensors now are a standard for smartphones – but the functional and physical integration is just in the first stage. There are still significant technical hurdles to be overcome for having – MEMS Sensors everywhere.
Leopold Beer joined Bosch Sensortec shortly after its foundation in January 2006 as sales director and currently holds the responsibility of global marketing director.
Prior to joining Bosch Sensortec, Leopold Beer held various engineering and managerial positions in the semiconductor and the automotive industry.
Gerhard Lammel is Senior Manager MEMS at Bosch Sensortec, responsible for advanced development. Since 2005 he managed several sensor development projects and is part of the founding team of Bosch Sensortec. Before, he was responsible for the process development for an integrated pressure sensor in the business unit Automotive Electronics of Robert Bosch GmbH. He obtained his PhD in 2001 from the Swiss Federal Institute of Technology in Lausanne, Switzerland. Gerhard Lammel studied physics and economics at the University of Munich. He founded two small high-tech companies for computer networks and computer graphics in 1991 and 1995. He is inventor and co-inventor of over 40 patents.
Gregory J. Galvin, CEO
MEMS inertial sensors, specifically accelerometers, first found widespread commercial use in automotive airbag systems. Although that was, and is, a high volume application, it is rapidly being dwarfed by the use of MEMS inertial sensors in mobile electronic products. In particular, mobile handsets’ rapid adoption of inertial sensors is driving the consumption of such sensors into the billions of units. Through the use of MEMS inertial sensors, accelerometers and gyroscopes, the mobile device becomes aware of its position, orientation, and motion environment. This in turn leads to a more sophisticated and intuitive user interface between the device and its user. At this point to speculate that every mobile electronic product will one day contain one or more MEMS inertial sensors no longer seems farfetched – the only question is how soon will it happen?
Dr. Galvin is President and CEO of Kionix, a wholly owned subsidiary of ROHM Co., Ltd. and one of the world’s top three suppliers of micro-electromechanical (MEMS) inertial sensors. A leading authority on MEMS product innovation, Dr. Galvin founded Kionix in 1993 to commercialize a novel silicon-micromachining technology pioneered by researchers at Cornell University. Prior to this, he served Cornell for nine years, first as Deputy Director of the Cornell Nanofabrication Facility in which micromechanical research was conducted and later as Director of Corporate Research Relations, focusing on technology transfer. Dr. Galvin was elected to the Cornell University Board of Trustees in 2011; he also serves on the Advisory Council of Cornell’s College of Engineering. He has a B.S. in Electrical Engineering from the California Institute of Technology, a Ph.D. in Materials Science and an M.B.A. from Cornell University.
Silicon MEMS Resonator Die Integration Inside SOC Package: Improve System Performance and Reduce BOM Cost
Markus Lutz, Founder & Executive VP
MEMS resonators, used in time-keeping applications, are the new holy grail of integration. In the past, the electronics industry had to rely on traditional quartz-based timing references and adapt to the inherent limitations of quartz. Engineers have not been able to overcome the technical challenges of integrating quartz into semiconductor devices due to size, reliability and cost issues. MEMS have broken through these barriers and can be integrated into semiconductor packaging with other silicon die.
Starting with sensors, silicon MEMS technology has been used in electronics systems for over two decades and this technology has become omnipresent in a multitude of applications. More recently, MEMS have expanded into timing applications and have begun to rapidly replace quartz-based timing references. The recent proliferation of MEMS timing has come through the adoption of MEMS oscillators and clock generators across all major electronics segments. These MEMS oscillators and clock generators are comprised of a MEMS resonator stacked on top of an analog oscillator IC, and together they are packaged into a standard plastic package. Now, MEMS resonators are being integrated with other types of semiconductor die (e.g., processors, SOCs), taking integration to the next level and creating an exciting emerging growth opportunity in the $2B (12BU) resonator market.
This session will address how ultra low power, zero-footprint MEMS resonators can be cost-effectively integrated into any standard semiconductor package, replacing bulky external quartz crystals and oscillators in the system. MEMS resonators are manufactured using standard CMOS processes leveraging semiconductor foundries. The resonators are delivered as known good die and can be integrated into ICs to provide time keeping (e.g. real time clocks) and power management(e.g. sleep, wake-up) functionality. With sub-uA power consumption, these resonators are ideal for low-power radios, microcontrollers/ microprocessors, and SOCs for portable handheld and mobile consumer applications.
The elimination of external timing components and circuits results in a host of benefits for these applications. For designers and manufacturers of semiconductors, MEMS resonators bring enhance performance and new functionality to the SOC. At the system level, integrated MEMS resonators enable improved system performance and reliability, reduced support costs and reduced BOM. MEMS resonators are highly accurate and offer frequency stability of ± 5 PPM. Additionally, using silicon MEMS enable a standard supply chain with scalability, shorter lead times and reduced costs.
Markus Lutz, CTO, Executive Vice President Initial inventor of InChipMEMS™ technology, which allows vacuum-sealed MEMS structures to be manufactured in ultra-pure wafer cavities with integrated CMOS and shipped in low-cost industry standard packages. SiTime is using this key intellectual property to bring to market the lowest cost, high performance resonators and oscillators, which are 1/8th the size of leading-edge competitive timing devices. Mr. Lutz received his Diplom Ingenieur Elektrotechnik at the Technical University of Munich in 1992. He started his career at Robert Bosch GmbH in Reutlingen Germany, where over 4 years he invented and managed the development of Bosch's first silicon based MEMS gyroscope, now a $200M/year business. In 1999 he joined the newly founded Research and Technology Center of Bosch in Palo Alto as MEMS Program Director. Together with Aaron Partridge (CSO) and Professor Tom Kenny's (Technical Advisory Board, Board Member) team he further developed InChipMEMS and wrote the first business plan for commercialization of InChipMEMS in the timing market. Markus holds 80 patents, and has authored and co-authored 16 publications.
Marc Osajda, Global Automotive Strategy Manager
Mobility enhances quality of life, but can come at a steep cost. The World Health Organization (WHO) counts 1.2 million people killed on world roads every year. WHO forecasts annual road fatalities to rise to 1.9 million by 2020. Road deaths are currently the number one cause of death for young people worldwide. The economic cost to developing countries is estimated at close to $100 billion per year. Making vehicles safer is not only a moral imperative but an economic one as well.
Electronic Systems Save Lives:
The introduction of airbags, followed by active safety systems have already reduced fatalities in developed countries significantly. Airbag regulations in Europe, for example, drove road fatalities down from about 75,000 a year to less than 60,000 in the 1990s. The European Union’s goal is that predictive safety and advanced driver assistance systems (ADAS) will ultimately reduce road fatalities per year to less than 13,000 by 2020.
Contrast this with the safety market in developing countries. A critical safety measure that consumers in developed nations take for granted—the airbag—is still uncommon in nations where car culture is growing the fastest. There is no doubt that airbags improve survivability in accidents. Passive safety systems such as front and side airbags are becoming mandated by governments in developing countries. In Brazil, for example, only 20 percent of vehicles currently have airbags, but government regulation beginning in 2014 will require all vehicles to have driver and passenger airbags. Fewer than 12 percent of vehicles in India have airbags, but front airbags will be mandated beginning in 2013.
Freescale is committed to better health and safety in the automotive segment. As the number one supplier of sensors for airbags and a leader in many other global automotive semiconductor segments, Freescale is well positioned to help automotive manufacturers make a difference.
This presentation will specifically highlights the latest development in airbag sensor and highlight what makes an automotive accelerometer different from a smartphone accelerometer. What are the requirements in terms of performances, packaging, testing, quality and product longevity and how those challenges are addressed by Freescale.
The presentation will also highlight how complex an automotive sensor can be through the example of a Tire Pressure Monitoring Sensors developed by Freescale.
Marc Osajda is Global Automotive Strategy Manager for Freescale Semiconductor in Munich, Germany. He is responsible for market strategy and analysis of the automotive semiconductor market with a specific focus on Sensors.
He joined Motorola's European semiconductor business in 1992 as a design engineer. From 1995 to 2004, Marc was involved with the European Automotive industry, respectively as Sensor application engineer, Sensor Marketing Engineer and Program Manager.
In 2005 Marc Osajda moved to Munich and joined Freescale Semiconductor Automotive Global Sales & Marketing organization. He is now in charge of global automotive strategy.
He holds an engineering degree in mechanics and electronics from the French “Ecole Nationale Superieure d’Arts et Métiers” (ENSAM).
Jeffrey L. Hilbert, COO and Founder
The consumer wireless market is characterized by very large volumes, very short product development cycles, decreasing product life times, and very low and rapidlydeclining product prices. These market characteristics make adoption of a new technology a risky proposition particularly when there is little to no high volume manufacturing and fieldperformance data to demonstrate the technology is proven and truly ready for insertion.
3G and now rapidly growing 4G mobile applications have intensified the requirements for a new, disruptive technology to implement next generation radio hardware particularly in the RF front-end of these devices. RF micro-electro-mechanical systems (RF-MEMS) is one such technology. Deep integration of RF-MEMS with high volume industry mainstream CMOS technologyprovides a viable approach to accessing the numerous benefits enabled by micro-mechanical devices while achieving the cost points and scalability required for success in the consumer mobile market.
This presentation will discuss the current status of RF-CMOS MEMS in mobile applications as well as look at future product applications and integration opportunities. By way of example, state-of-the-art results from the implementation and production of software tunable, digital, RF-CMOS MEMS circuits will be discussed.
Jeffrey L. Hilbert is the President, COO and founder of Wispry, Inc. Wispry, based in Irvine, California is a fabless semiconductor company utilizing radio frequency micro-electro-mechanical systems (RF-MEMS) technology integrated with CMOS technology to develop tunable electronic products for the cellular communications and wireless consumer electronics markets. Jeff has over 35 years of executive management and technical experience in a number of leading semiconductor and MEMS companies including LSI Logic, Compass Design Automation, AMCC, Motorola, Harris and Coventor. An experienced entrepreneur, Jeff has raised over $100M in financing to fund two consecutive start-up semiconductor companies over the past fourteen years. Mr. Hilbert holds a BS in Chemical Engineering from the University of Florida and an MS in Computer Science from Florida Institute of Technology.
Keynote: Marco Angelici, Audio and Sound Product Management Director
Consumer world boosts the continuous innovation of MEMS technology, raising the bar of sensing capability and pushing with unbelievable miniaturization and packaging options the frontiers of possible applications far away from what conceivable today.
Sensor Fusion, providing up to 10 degrees of freedom, is today the key enabler of motion MEMS pervasion in Smartphones and Game consoles. Environmental Sensors (P, T, and H) are now part of this fusion process, by enriching the usability of Smart products. Even Audio MEMS fusion with the other sensors can pave the path of Personal Listening experience towards the bionic ear’s concept. High performance sensors, processing and connectivity are the bricks for new augmented applications. Fusion in MEMS will go in parallel to MEMS Industry suppliers’ consolidation: it’s time for leaders in MEMS one-stop-shop.
Marco Angelici, Product Management Director of Audio and Sound Business Unit, STMicroelectronics
He joined STMicroelectronics’ R&D labs in Grenoble, France, in 1995 as designer of low power audio converters (ADC, DAC) and digital interfaces. In 1997 he moved to Audio Division in Agrate, Italy, extending his IC design responsibilities to audio amplifiers and digital signal processing for consumer applications. After four years of successful design activity, with several patents, he became responsible for the worldwide Audio Division R&D team.
In 2008, Marco was appointed Product Management Director, with full responsibility of strategic marketing, business management, application and program management, for consumer audio. In 2011, ST regrouped Audio activities under Analog, Mems and Sensors Group, by the creation of Audio and Sound Business Unit, to enlarge the Audio offering from “legacy consumer products” to innovative Audio Solutions, including MEMS Microphones and Transducers. Marco is Deputy Director of the management board office of the Audio BU.
Eric Lautenschlage, MEMS Engineering Manager
Throughout our history humans have continued to develop technology to serve the needs of those who use it. This trend has continued into modern times with the progression of electronic technology through the fixed computing, network interconnectedness, and personal mobile phases. Human interfacing with mobile devices today is possible using several options, but for some applications, might be best served using the Knowles MEMS Joystick. A technical review of the joystick will be presented, including both a high level summary of the design and architecture, as well as the process flow used for fabrication. Finally, some potential market applications will be presented which are well suited to joystick use.
Dr. Eric Lautenschlager has been working for the past 5 years at Knowles Acoustics with primary R&D responsibility to develop MEMS technology for use in the SiSonic™ microphone product family. Prior to Knowles, he worked with Honeywell Aerospace on their MEMS gyroscope technology, and previous to that, for Cypress Semiconductor doing transistor design and device process integration. Dr. Lautenschlager earned his Ph.D. in Physics from the University of Texas – Austin as his B.A. in Anthropology from the University of Illinois – Champaign.
Uwe Schwarz obtained a degree in Physics at the University in Leipzig in 1988. He joined X-FAB in 1992. He worked first as a development and process engineer on photolithographic processing, and was also involved in some of the CMOS technology development programs of the company.
In 1997 he started the first activities in the field of MEMS process development at X-FAB. He has been deeply involved in the development of a MEMS foundry business. From 2001 he has been head of the MEMS process development team.
From Mega to Micro – How Megatrends Influences the Future of Microsystems in Automotive Applications
Bernhard Schmid, Manager Sensor System & Technology
society underlies an ongoing change. The world population is growing
steadily. The standard of living is developing heavily in the
high-growth markets of the future as Asia, but also Brazil and Eastern
Europe. All these people have the increasing demand for safety, energy,
healthiness, affordable and beneficial goods and last but not least
self-actualization. From this point of view the automobile industry
derivate the four megatrends: Safety: Safe Mobility, Environment: Clean
Power, Information: Intelligent Driving and Standard of Living:
These megatrends impinge on every activity related to the life cycle of a single component up to the whole vehicle, starting from innovation, development, production, usage until the disposal. Besides others, Microsystems are the winner to cope these challenges (more products, more functionality) but also treated by the constraints of the trends (less costs, higher efforts). One example is to reduce further the numbers of fatalities in an increasing traffic density. In this scenario the systems need to know the precise position and speed vector of itself and each possible object to collide with. To cope this we need more precise inertial, wheel speed, steering angle sensors and powerful fusion algorithms to combine the information with GNSS. One other very popular scenario is the transition from a fossil based energy source to the electrical propulsion, feed from regenerative energy sources. This is more revolutionary since it changes not only the complete architecture of the vehicle but also the infrastructure and therefore, more or less the whole knowledge base related to the vehicles lifecycle. This article will highlight these and other scenarios and their impact on Microsystems, respectively MEMS. Further, it will give some examples related to these scenarios on which Continental is working.
the beginning of 2010 Bernhard Schmid is heading the department "Sensor
Systems & Technology" of the Business Unit Sensorics, transformed
into “Generic System Development” after the fusion of the BUs Passive
Safety and Sensorics at Continental in Frankfurt / Main. In this
function he is responsible for the pre-development of all kind of
automotive transducer technologies, sensors and sensor systems and was
assigned as Continental Principal Technical Expert for “Automotive Speed
and Inertial Sensors”.
His professional career started Bernhard Schmid as a radio electronics trainee. Until 1996 he studied physics at the Technical University of Munich and conducted his diploma thesis at the Université de Montreal, Canada in the field of microsystems technology and gas sensors. From 1996 to 2002 he worked as a process and development engineer and later as the head of process and technology engineering in the wafer manufacturing of MEMS airbag sensor elements at Temic microelectronics in Munich / Ottobrunn. By 2009, Mr. Schmid was in charge of the design and technology development of diverse MEMS inertial sensor principles for active and passive safety systems in the R & D segment of the business unit Sensorics at Continental in Frankfurt / Main.
Matthew Crowley, Founder/ VP Business Development
MEMS timing devices have made significant inroads into replacing discrete quartz crystal devices in many applications. As the performance of new MEMS based oscillators improves and new applications are accessible to MEMS, this displacement will only accelerate. One unique capability of MEMS timing devices is their ability to be sold as WLCSP packaged devices that can be integrated in the same package as other ICs. In addition to performance, quality and supply issues, quartz device performance can vary significantly based on their placement on PCBs, rapid thermal changes due to other components, and PCB board noise/harmonics. All of these issues can significantly increase system design complexity and substantially delay time to market for complex systems such as smartphones. CSP MEMS unique ability to be co-packaged with other ICs will prove disruptive to the timing industry and will fundamentally change how timing is provided to electronic systems. By co-packaging a MEMS oscillator with other ICs in a system-in-package implementation, end customers can expect a significant reduction in total solution size, PCB board noise and design complexity. It is expected that the reduction in design complexity will enable faster time to market for system vendors who can choose ICs with integrated timing to implement simple plug and play timing architectures. This will also enable new business models as integrated reference designs eliminate the need for system vendors to directly source or design-in quartz oscillators. This integration ability is one of the most compelling reasons for the displacement of traditional quartz oscillators.
Matt Crowley is a founder of Sand 9, where he manages business relationships with external partners. Before founding Sand 9, Matt was the Director of the Technology Development Fund at Boston University’s Office of Technology Development. At BU, Matt focused on creating and financing new companies formed to commercialize disruptive technologies developed at the University. In addition to forming new companies, Matt also directed venture capital investments for the University. Prior to joining BU, Matt worked at Mars & Co strategy consulting, where he advised Fortune 500 companies on operational and strategic issues. Matt received an interdisciplinary degree in Physics and the Philosophy of Science from Princeton University, where he also studied Japanese.
Tomas Bauer, VP, Sales & Business Development
history of product development is a history of product diversity
through innovation. It is said that over 100,000 patents went into the
creation of the modern car. A quick search on the US Patent Office shows
over 330,000 patents on electric motors; over 50,000 on hinges. Nowhere
is this innovation so prevalent today as in the field of MEMS: applying
micro manufacturing techniques to mechanical structures as diverse as
atomic clocks to micro syringes. And because MEMS as a discipline is
being applied to such a wide diversity of products, it remains an
innovation game. Each product type is unique, each construction needs a
manufacturing process matched to its design. Just like a machine shop
needs flexibility to make everything from hinges to engine blocks, a
MEMS fab needs to support everything from atomic clocks to micro
In the early days of MEMS, only the companies who could build and manage their own fabs could pursue MEMS development. These are the IDMs which provide much of the MEMS today. However this restricts the innovation to a few very large houses which can support this infrastructure. For MEMS to diversify and grow, a foundry model is needed.
As an innovation game, a majority of MEMS products either don’t make it to market or service a defined yet small market space. 49 out of every 50 MEMS products never need more than 1000 wafers per month; within the MEMS world even the very large volume
applications are measured in single thousands of wafers per month – orders of magnitude fewer than the major IC industry drivers like DRAM or CPUs. A foundry suited to servicing this size of customer is needed to enable the MEMS market to flourish.
Early evolution of the optical MEMS market segment show abundant investment capital allowing market entrants to build their own wafer fabs for MEMS. The absence of Pure Play MEMS foundries and an understanding of the importance to keep MEMS processing in-house to maintain confidentiality was deemed critical to success. Most of these ventures did however realize that maintaining a full wafer fab to make up to 1000 wafers per month was a very expensive adventure.
Most new companies cannot afford, and their volumes cannot justify, construction of a MEMS fab line to service their needs. The only option for these companies is a pure-play MEMS foundry, one dedicated to process integration and manufacturing excellence, but which is product agnostic and has no competing profit motive for quelling competition.
The fabless MEMS market will flourish when companies can innovate freely, without fear of compromising their innovation by giving early notice to established product companies of their new innovations.
This is the model that Silex brings to the market: as a pure MEMS foundry, we focus on the critical areas that companies need:
• Innovation and flexibility to adapt processing to new MEMS designs
• Product agnostic focus on process integration and high volume manufacturing
• A manufacturing and engineering organization that can take MEMS to Market, Faster, servicing the 95% of customers who need <1000 wafers per month, and develop robust processes which can take the 5% of high volume products into large volume IC foundry partners.
• A process integration approach focused on faster time to market and standardization through SmartBlocks™ process step standardization which retains maximum flexibility and agility.
As the battleground of MEMS innovation, Pure Play foundries like Silex can also lead the R&D of novel process capabilities and integration/packaging techniques. Where the market is looking for innovative 3D packaging technologies, Silex has been in high volume production with all silicon TSVs, WLCSP, and other novel heterogenous packaging solutions for over half a decade.
Every new product company envisions themselves as the next revolutionary product on the market. If we could predict with certainty which ones were going to hit, the path to production would be much simpler. Reality, however, says that nobody knows for sure. What is needed is a foundry who excels in supporting the ‘innovation game’ in taking new products and processes to volume production, supporting also the <1000 wafer per month needs, but has the ecosystem and alliances to bridge the gap to true high volume manufacturing. Silex is the leader in establishing this model, one which can truly help fabless innovators to challenge the dominance of the IDMs in the MEMS landscape.
Tomas Bauer was born on 1974, MS.
Vice President, Sales & Business Development since 2006; employed with Silex since 2004.
Tomas Bauer has an outstanding record of deploying successful strategies in the sales of complex technical solutions and manufacturing services. He has played an important role in shaping the global foundry business strategy of Silex and has made a significant contribution to the overall growth of the company. Prior to joining Silex, Tomas held positions with Temex and Ericsson Microelectronics.
Anu Kärkkäinen, Principal Scientist and Sensors Product Line Manager
MEMS technology provides a cost efficient way to produce sensors in high volumes. However, development costs and shortage of ramp-up production facilities makes the development of new products challenging. The ‘MEMS law: one process, one product’ is still mostly valid, and some process development is almost exclusively needed when a new product type is being launched. Existing MEMS manufacturers are mostly interested in high volume production only due to the good productivity of the technology, typically low margins, and profitability requirements. On the other hand, there is a market entry time for new products resulting also in a need for moderate-volume production sites.
Within VTT Group a contract manufacturing company VTT Memsfab Ltd has been established in 2011 with a business model aiming at initial production, moderate-volume final production, and ramp-up services for products striving for high-volume markets. Hence, VTT Technical Research Centre of Finland, together with VTT Memsfab Ltd, can provide a seamless chain from research to production without a need to transfer the technology into another factory with a different toolset in the early stages of production. This is a ‘MEMS highway’ for companies to test their sensor ideas in the open market. Customer benefits from reduced development time and lowered R&D cost. Operating within one site is efficient and enables long term collaboration.
VTT helps companies to fast-forward development processes by exploiting the intellectual property that we have generated over the years in the fields of consumer electronics, automotive, energy technology, healthcare and other areas. VTT offers development services on many technology platforms, and VTT Memsfab a fluent flow to commercial production.
Mrs Anu Kärkkäinen is the Manager of the MEMS sensors product line of VTT Technical Research Centre of Finland. She has a long experience at different measurement related R&D work in Finnish industry. Since 2001 Mrs Kärkkäinen joined VTT where she has, for example, coordinated the EC funded project EMMA, which developed capacitive MEMS components for precision applications.
Paul Lindner is executive technology director at EV Group headquarters in St. Florian, Austria. With 20 years experience in the semiconductor industry, Mr. Lindner's past work includes involvement in many aspects of Semiconductor and MEMS equipment manufacturing.
Based on mechanical engineering background he started working at EV Group in 1988 as mechanical design engineer. His responsibilities included the design of various different semiconductor processing systems and tooling for custom applications. Innovative system designs pioneered in first commercially available wafer bonders, SOI bonding systems or precision alignment systems for 3D Interconnect applications. Prior to his appointment as executive technology director, Lindner has established a product management department at EV Group. During that time he was involved in marketing, sales, manufacturing and on-site process support. Customer orientation throughout all steps of product development, innovation and implementation in a production environment are among the main goals of EV Group's technology groups headed by Mr. Lindner. His current responsibilities as executive technology director include new technology development by heading R&D, product, and quality management, business development and process technology departments.
Romain Fraux, Project Manager
SYSTEM PLUS CONSULTING
Inertial MEMS today are driven by further market trends. One of the main is the price pressure which is very strong (5% drop per quarter for consumer). In order to keep a strong price decrease, manufacturers have to perform technological innovations.
For a few years, inertial MEMS have been subject to market and technological evolution. This has been driven by a large increase of the consumer market: mobile phones and tablets for accelerometers; gaming for gyros; mobile phones for magnetometers. Along with “stand-alone” MEMS devices, inertial combo sensors are also coming.
In order to show the technological evolutions of consumer inertial MEMS, we will make a review of several STMicroelectronics 3-axis accelerometers and gyroscopes. With the largest market share for MEMS inertial sensors for consumer applications, ST is a good representative of technological evolutions. We will highlight the processes used by ST to decrease years after years the footprints and thicknesses of their MEMS devices (technology shrink, new wafer bonding process, TSV) and see how that impact cost.
Romain Fraux is Project Manager for Reverse Costing analyses at System Plus Consulting.
Since 2006, Romain is in charge of costing analyses of MEMS devices, Integrated Circuit and electronics boards. He has significant experience in the modeling of the manufacturing costs of electronics components.
Romain has a BEng from Heriot-Watt University of Edinburgh, Scotland and a master's degree in Microelectronics from the University of Nantes, France.
Stephan Proennecke, Director of Production and Project Manager
The likelihood of someone getting diabetes is growing fast. It is poised to bankrupt the health care system of many advanced countries if nothing is changed. When patients cannot be cured, they need an accurate therapy for preventing the complication linked to diabetes. The need to elaborate a cost effective solution that can offer a safe and efficient therapy for the diabetic patients is of upmost importance.
To address these specific needs, Debiotech has developed in partnership with ST Microelectronics a highly innovative MEMS pump, providing a nanopump capable to deliver highly accurate doses of 200 nL with an accuracy better than ±5%. The purpose is to provide a pump that can mimic the pancreas, thanks to a very small and accurate stroke volume by design, but also to offer additional capabilities that will improves the safety of the pump through the implementation of pressure sensors and detectors. The use of high level integration of MEMS also makes the JewelPump the smallest large-reservoir patch insulin pump, which contributes to the patient comfort and a better therapy acceptance.
Thanks to the large-scale manufacturing experience of ST Microelectronics, the JewelPump also benefits from an exceptional cost effectiveness which shall contribute to its market acceptance and availability.
Clinical use of the JewelPump on diabetic patients already indicates an exceptional acceptance and delivery precision.
Dr Stephan Proennecke is Director of Production and Project Manager in Debiotech, where he leads the JewelPUMP project, based on MEMS technology. Stephan got his PhD in 1992 at the EPF of Lausanne. After 4 years of fundamental research on the chemical properties of water-metal interfaces at CNRS in France, he decided to continue his carrier in the industry, leading research and development projects in the consumer electronic fields, in particular with Logitech. He participated during these years to the industrialization of highly advanced technology products for the retail market, understanding the key steps of product conception, development and large-scale manufacturing. Stephan joined Debiotech in early 2009, taking the lead of the team who developed the JewelPUMP and the responsibility for its large-scale industrialisation.
Martin Schrems, Director Technology R&D,
Dr. Martin Schrems, "Director R&D" for Process Development and Implementation joined ams in 2001. Martin’s responsibilities include managing all technology R&D activities including semiconductor process, device development, circuit modeling, DFM as well as ESD/EMC. Moreover Martin coordinates all national and European collaborative funding project activities for the company. He has 20 years’ experience in the semiconductor industry with previous positions held at TOSHIBA Corporation (Japan), Siemens Semiconductors (US), and Infineon (Germany). He holds a PhD degree in Electrical Engineering (1991) and a Master of Science in Physics (1988) both from the Technical University Vienna as well as an Executive MBA from IMADEC University (2006).
Maik Wiemer, Department Manager System Packaging
Fraunhofer Institute for Electronic Nano Systems
3D integration is of major interest for several applications in the fields of microelectronics and MEMS (Micro-Electro-Mechanical-Systems) technologies providing the opportunity to integrate electronic devices of different functions and technologies, such as MEMS and different electronic devices. Beside important fabrication processes such as wafer thinning, Through-Silicon-Via (TSV) drilling or etching and TSV filling one well established technology for the fabrication of 3D devices is wafer bonding. One critical parameter in wafer level bonding is the process temperature. Reasons therefore are different coefficients of thermal expansion (CTE) of the bonding partners and the need of lower process temperatures during the fabrication process of the whole system.
This paper introduces three low-temperature wafer-bonding processes. The first process uses a specific form of local heat generation, which is based on nano scale reactive material systems. Such systems consist of several layers of minimum two different materials with nano scale thicknesses. These layers generate a self-propagating exothermic reaction during their intermixing. The resulting heat can be used as heat source for bonding processes such as solder bonding. This work focuses on the direct wafer deposition of ultra thin and reactive nano scale multilayer systems. The principle of this method is demonstrated by reactive wafer bonding at room-temperature.
The second process is the field-assisted direct bonding of two Si wafers by using a thin glass layer as intermediate material. In order to obtain a flat and uniform glass film, a Si and a glass wafer are first bonded by using standard anodic bonding process. The glass wafer is then thinned down to 10 µm thickness by using grinding tools. After subsequent surface polishing, the composite wafer is then bonded with another Si wafer, with a small negative bias voltage applied on the glass coated Si wafer. Experiment results indicate that full surface wafer bonding can be achieved at 150 °C process temperature, and 120°C bias voltage. The detailed influence of each process parameters will be discussed in the presentation. The process has been applied for MEMS microscopes.
Within the third process we will demonstrate the usage of surface activating procedures prior to bonding, like low-pressure plasma or ion beam treatment. Another opportunity for surface activation is the dielectric barrier discharge (DBD), which offers stable plasma at atmospheric pressure (AP). In the paper the wafer level bonding was examined for silicon to lithium tantalate (LiTaO3, application: SAW filter), Foturan© (application: through glass vias) and Borofloat© by using CMP-processes and ambient pressure plasma activation.
Maik Wiemer received his M.S. degree in electrical engineering from the Dresden University of Technology, Germany, in 1988. In 1999 he received his Ph.D. at the Chemnitz University of Technology, Germany, in the field of MEMS packaging. In 1999 he joints the Fraunhofer IZM as research associate. Since 2009 he is the department leader of department System packaging at the Fraunhofer Institute for Electronic Nanosystems in Chemnitz, Germany. His professional interests lie in the research and development of MEMS packaging technologies, 3D integration technologies and wafer bonding processes. He has authored and co-authored 62 reviewed journal papers and he has 4 patents or patent applications.
Michael Bergman, PhD Researcher
For biomedical applications and biosensors the development of biocompatible materials is an important issue and especially the protein-surface interaction is in the main focus. The function of implanted biosensors is mainly disturbed by the events associated with tissue reactions surrounding the implanted device, such as hemostasis, inflammation, repair, infections, protein deposition and encapsulation of the functional part. In this study a low-pressure magnetron-enhanced 15 kHz discharge polymerization is used to get biocompatible nanofilms on e.g. enzyme based biosensors, which were implanted in animals. The process gases are CH4 and O2 excited between two titanium electrodes with a power of 45 W.
X-ray photoelectron spectroscopy (XPS) and dynamic contact angle measurements gave information about the chemical/physical properties of the plasma coating. Surface Plasmon Resonance (SPR) was used to characterize the performance of protein-surface interaction of the different coatings. By fluorescent microscopy the antimicrobial properties of the nanofilm were investigated. Compatibility of the process with sensor materials e.g. adherence in dry and aqueous environment was tested. In vivo experiments were carried out to monitor glutamate concentration during epilepsy induced by kainate injection. The sensor was inserted vertically into dorsal hippocampus of a rat. First in vivo experiments were carried out, showing a rise in amperometric glutamate measurement, during an epileptic seizure.
With higher amounts of oxygen in the precursor gas mix the contact angle decreases, because more polar groups were embedded into the film. The nanofilm can be tailored to repellent properties in matters of fibrinogen adsorption, suggesting an improvement of biocompatibility for in vivo applications. A reduction in bacteria adhesion up to 99% could be reached.
The wide parameter range of the coating process allows the tailoring of surfaces towards biomedical needs. Especially for biological applications these coatings allocate an opportunity to enhance the biocompatibility but not disturbing the inherent functionality of the material.
Michael Bergmann received his Master degree (German diploma) in Microsystems engineering specialized in Life Sciences at the Albert-Ludwig University in Freiburg in 2010. Currently, he is in the third year of his PhD thesis working on plasma polymers at the Laboratory for Sensors at the Institute of Microsystems engineering in Freiburg.
Josep Montanyà I Silvestre, Co-founder, Chief Technology Officer,
While MEMS devices enable many new functions and applications in todays electronic systems,most, if not all, require dedicated fabrication processes and special packaging techniques, hence manufacturing in high volumes, low cost and multiple sourcing for consumer applications is an industry wide challenge. On top of that, the need for tight integration with the control electronics is a further important consideration. On the other side, most microelectronics benefit from mass market standard CMOS, sourced from multiple large foundries and benefiting from economies of scale. nanoEMS™ technology is a set of design techniques that allow the usage of standard commodity CMOS to build many MEMS devices for the consumer market at a previously unachievable price point.
In this presentation the nanoEMS™processing technology will be shown. The technology is based on applying a vHF etch to the finished CMOS wafers, so that the inter-metal dielectric layers of the back-end are removed as required, releasing the metal interconnections , so as to define the MEMS devices. Although many approaches have been proposed or attempted in the past to build MEMS devices monolithicaly integrated with CMOS, this is by far the simplest post-processing that have ever been proposed and fully demonstrated, that is manufacturable in high volumes and low cost in any mainstream CMOS foundry.
The technology will be further illustrated by a particular case, a 3D digital compass, showing the design and performance achieved, together with prequalification and initial yield data. There will also be a look into the future for what’s next in terms of applications and technology for this exciting development..
Josep received an M.S. degree in telecommunication engineering in 1996 and a Ph.D. in January 2012 from UPC (Barcelona) .
He has 16 years of experience in the electronics and software industries and 9 years in the MEMS field. His Ph.D thesis describing a novel design for an RF MEMS was the inspiration for the founding of Baolab in 2003.
He continues to pioneer the development of MEMS with his work as CTO at Baolab driving the NanoEMS™ technology, building MEMS sensors and devices using standard CMOS processing.
Anna-Riikka Vuorikari-Antikainen is a Senior Vice President, Products at Okmetic, a company making silicon wafers e.g. for MEMS manufactures. Currently she is responsible for Okmetic’s product portfolio and product and process development. She has worked at Okmetic since 1992 in different management positions in production, quality, marketing, R&D and business development areas. Vuorikari-Antikainen is a core group member of MemsCat, Finnish MEMS cluster and chairman of VTT’s advisor committee for ICT and electronics. She has a Master’s degree in Physical Metallurgy from the Helsinki University of Technology, Finland.
Laurent Robin, Activity Leader – Inertial MEMS
MEMS market will continue double digit growth, as demand will drive a doubling of the MEMS market to $21 billion by 2017. As this industry evolves, driven by inertial sensors, we observe that key challenges arise: it becomes key to be able to supply a wide range of sensor elements, to integrate them in combo solutions, and to add the right level of software. The technology issues and cost structure are also essential and are driving quick evolutions of the competitive landscape. The status of this industry will be reviewed in this presentation, as well as the solutions proposed by the key players.The first part will show the global evolution of the MEMS markets and will highlight the importance of inertial sensors. MEMS will see 20% compound average annual growth in units and 13% growth in revenues for the next six years. Along with microfluidics, the inertial sensor sector will increasingly come to dominate the MEMS market totals. Accelerometers, gyros, magnetometers and combos will account for about 25% of the overall market in 2017.The second part of the presentation will details trends on inertial sensors used in mobile devices. There’s plenty of room for the inertial sensor market to grow for at least the next three years, as penetration increases in growing end markets. As accelerometers and magnetometers are now starting to be designed into more feature phones, the hottest consumer products remain 3-axis gyroscopes. The penetration of gyros jumped from 9% of smart phones in 2010 to 36% in 2011. However we believe that the market for discrete sensors will begin to decline, but the growth for combo solutions will be huge. Though currently less than $100 million niche, we expect combos to be a $1.7 billion opportunity by 2017. In 2011, 6-axis accelerometer and magnetometer combo units were shipped in volume, and 6-axis accelerometer and gyro units also started to do so as well, often for only a small additional cost for the accelerometers. Finally the competitive landscape and the complexity of the inertial MEMS supply chain will be detailed. Teardowns of real products will be shown to highlight the different technologies used for front-end as well as for packaging and integration. Information on the cost structure and gross margin of the different components will be given and major evolutions of the value chain will be analyzed.
Laurent Robin is in charge of the MEMS & Sensors market research at Yole Développement. He previously worked at image sensor company e2v Technologies (Grenoble, France). He holds a Physics Engineering degree from the National Institute of Applied Sciences in Toulouse, plus a Master Degree in Technology & Innovation Management from EM Lyon Business School, France.
Jérémie Bouchaud, Senior Principal Analyst, MEMS and Sensors
It might seem that it is a tad too early to predict that MEMS industry will soon need new killer applications given the high rate of expansion the devices have witnessed in the past few years. Yet, in order to continue notable double-digit growth, especially in the key MEMS consumer-mobile and automotive segments, killer apps are going to be essential for suppliers. Since rebounding after the economic recession with a peak expansion rate of 20 percent in 2010, the MEMS industry has added about $1 billion in revenue each year between 2011 and 2014, equivalent to yearly growth of 11 to 14 percent. However, the industry will start running out of steam by 2015 when overall revenue growth falls to 7 percent, due to saturation of some major segments.
Where can the next killer apps and new devices come from which will propel MEMS forward and back into double-digit growth after 2014?
1) On the application side, what will comes next from Apple? Automotive mandates in BRIC countries? Wireless sensor networks?
2) Which emerging MEMS devices will be the next big hit for mobile devices: MEMS speakers? MEMS autofocus? RF MEMS tuners? Environmental and gas sensors?
Jérémie Bouchaud is unique to the MEMS industry – his breadth of MEMS and sensors device and application knowledge is unmatched, particularly in terms of automotive, consumer markets and industrial and medical applications. He was a founder and head of MEMS research for Wicht Technologie Consulting . At IHS, Bouchaud is responsible for the MEMS service area. In the course of his career, he has led more than 100 MEMS-related market research endeavors. Prior to WTC, he oversaw technology transfer for sensors and MEMS at the German office of CEA-LETI.
Bouchaud is a graduate of the Munich University of Applied Sciences and of Ecole Supérieur de Commerce of Grenoble. He speaks German, English and French.