MEMS Trends No.9 2012.01

MEMS'
Trends
,668( 1
JANUARY 2011
Ma g a z i n e o n ME MS T e c h n o l o g i e s & Ma r k e t s
FEATURE STORIES - BIOMEMS
Fragmented microHuidics
market means fragmented
materials opportunities
The gene sequencing gold
rush goes to the next level
FINANCIAL BUZZ
DALSA acquisition lets us
look at the MEMS world
differently
F r e e r e g i s t r a t i o n o n www.i-micronews.com
P
r
i
n
t
e
d
o
n
r
e
c
y
c
l
e
d
p
a
p
e
r
+ Semicon Korea
January 26 to 28, 2011 - Seoul, Korea
+ LabAutomation
January 29 to February 2, 2011 - Palm Springs, CA
+ MEPTEC - 9th Annual MEMS Technology
Symposium
May 19, 2011 - San Jose, CA
E V E N T S
E D I T O R I A L
The Yin and Yang of MEMS
f you were gone the last 2 weeks of 2010, as many were, it was upon your New Year return
that you learned of the acquisition of Dalsa by Teledyne while the strongest apparent
affnity may be CMOS it is also clear that "kid brother MEMS played more than a minor
role. Paraphrasing one individual close to the transaction who sent me an email the day of
the news: "there are a lot of synergies and opportunities across all technologies between
these two companies. This deal is an indication of MEMS strength and by all appearances
it is a well-balanced acquisition.
With such big news so early, is that it for the year? Or is there more in store for 2011? With
this issue of MEMSTrends' the focus is back on BioMEMS where there is much that bears
watching in the coming year. n some cases it has been a long time coming, but mostly it is a
simple matter that problems and solutions, costs and benefts, are fnally fnding the winning
equilibrium and BioMEMS will be gaining momentum.
n addition, inside this issue Eric Mounier highlights a number of key areas where the balance
is reaching a point where good things are likely to come: the market pulling and good solutions
fnding the way to the front to deliver benefts. One of the more interesting among them is the
return of the fber optic switch market.
A lot has changed since the telecom crash and this time things may well be different. n
optical switching 1.0, optimized switching solutions were developing at the same time that
everything was in place for a capacity glut and the solution suddenly had no place to go.
A decade later, internet growth seems to have no end, smart phones are more and more
abundant, streaming and other data heavy services are just beginning and cloud computing
is here to stay. And, while the capacity limits of fbre networks have an incredibly long way
to go before they reach their limits, optical switching using MEMS enabled devices provides
solutions that have enough benefts that indeed for certain applications the demand and
solution balance will be met.
And that is of course the key problems with well matched solutions, complementary
opposites fnding equilibrium to yield a greater whole. Across a broad range of applications
and devices, as 2011 unfolds we see MEMS poised for a strong and sustained run of growth,
not because they are MEMS, but because MEMS devices are providing the right solutions
that yield real benefts for everyone in the value chain.
This is the frst anniversary issue of MEMSTrends' which replaced MEMSentry which itself
had a great run, and we would like to take a moment to thank all of our readers. Your on-going
feedback has been tremendous. As our audience grows and we continue to supply you with
information and analysis, we like hearing from you so please keep the comments coming,
and have a great 2011!
Jeff Perkins, President, YoIe Inc.
perkins@yoIe.fr
M E M S ' T r e n d s 3
J A N U A R Y 2 0 1 1 I S S U E N 5
PLATINUM PARTNERS:
...As 2011 unfoIds
we see MEMS poised
for a strong and
sustained run
of growth...
REGISTER NOW:
www.semiconkorea.org
co-located with:
Exhibiting opportunities available!
Its elementalplan now
to participate at SEMICON Korea 2011!
SEMICON
Korea brings together

industry leaders and innovators to identify
new opportunities and explore the future
of microelectronics.
s #ONNECT TO PEOPLE AND PRODUCTS
over 20,000 visitors and 500+ exhibitors
s /PTIMIZE THE LANDSCAPE
co-located with LED Korea
s ,EVERAGE FACETOFACE OPPORTUNITIES
meet customers, connect with suppliers
www.semiconkorea.org
JANUARY 2628 Convention & Exhibition Centre (COEX), Seoul, Korea
CONTACT:
SEMI Korea
Hye-Jin Jang
Tel: +82.2.531.7802
Email: hjang@semi.org
3%-) (EADQUARTERS
Tel: 1.408.943.6900
US toll free: 1.888.500.7364
Email: sales@semi.org

B I O M S
J A N U A R Y 2 0 1 1 I S S U E N 5
Medical device makers turn to
microsystems to gather information
from inside the body
I
magesensors andingestionsensors onpills report
out from the digestive tract. Image sensors and
force sensors on endoscopes let doctors look and
feel around from natural openings and tiny incisions.
Pressure sensors and accelerometers on implants
monitor and adjust heart pacing and pumping.
Suppliers say this is just the beginning, with a second
generation of more sophisticated systems now in the
pipeline.
"The bottom line is the market is demanding more
and more ways to monitor patients - and the
technology is now hnally there to deliver it, says
Enrique Malaret, CEO of Tronics MedTech (Sunnyvale,
CA), a medical product development consulting
company now owned by its MEMS partner Tronics
Microsystems.
Given Imaging targets using PillCam
as delivery platform for sensors
and actuators for diagnosis
and treatment
Given Imagings PillCam pill that takes pictures inside
the digestive track and sends them out wirelessly is
now widely used to diagnose problems in the small
intestine, and is entering the much larger colonoscopy
market. The Yoqneam, Israel-based company next
aims to use the pill as a delivery platform for a wide
range of other sensors, using optics, RF, ultrasound
or biomarkers to identify disease from within the
digestive tract, and then even perhaps for actuators
to take samples and apply treatments.
Recent advances in the sensitivity and size of MEMS and other electronic sensors,
and in ultralow power control and wireless transmission solutions, are enabling a
whole new world of looking at the inside of the human body.
"Going forward youre going to see more MEMSin
whatever we do, says Ori Braun, SVP of business
development, also in charge of acquisitions and
new technologies. "Our objective is to enhance
and automate hndings. The capsule is just
the infrastructure. He notes the company is
particularly interested in adding sensors for 3D
localization to locate the capsule in all dimensions
in the body in real time, in energy harvesting
to supply more energy for the ever increasing
demands on imaging, and in sensors that
identify diseases. Different solutions swallowed
by the patient and different kinds of image
sensors could highlight different abnormalities.
Maneuvering the capsule to a specihc location
could enable it to do biopsies or deliver drugs
using microfuidics. Given Imaging aims to be
the platform for all sorts of such developments
from other suppliers. "We encourage companies
and universities to approach us with interesting
solutions, Braun urges.
The company has recently added a sensor to
measure gastric acid refux in the esophagus
to its toolkit, using a capacitive pH sensor
inserted with an endoscope and attached to the
esophagus wall for a few days with a sort of
suction cup and needle.
The progress of imaging, wireless transmission
and low power technologies for consumer
electronics have enabled Given Imaging to
considerably improve its current generation
product, with the colon tool now provided with
a camera at each end to record images in all
directions as it rotates through the passage,
and with an RF transmitter that sends out
images at close-to-full- video rates of 35 frames
per second. Back when it started developing
the PillCam a dozen years ago, the tiny, low
power imagers, LEDs and wireless transmission
electronics werent widely available. The
company worked with Micron, now Aptina, to
develop a low power CCD, and with Zarlink
to develop a low power RF transmitter. Given
Imaging designed and integrated the system,
including clever optics to solve the problem
of the light causing refections on the lens
because the LED and camera had to be so
close together on the tiny unit, and algorithms
to highlight the areas of interest in the hours of
video produced.
Power consumption remains a major issue.
The company continues to look for lower
power solutions. Off-the shelf image sensors
are still too power-hungry to use with the tiny
conventional batteries that ht in the pill, and
demand for more energy-consuming capabilities
is only accelerating. "Demand for more energy
is never going to end, says Braun. "People
always want more pixels, then theyll want 3D,
high dehnition, spectrum analysis. But the FDA
limits how much power can be transmitted, so
were looking at energy harvesting.
Wide interest in implanted pressure
sensors, now in animal trials
Tronics MedTech reports wide interest across
its medical device development clients for
applications using MEMS pressure sensors with
low power RF output, for implantable systems
to monitor pressure in the eyes, the brain, and
the circulatory system, and for implantable
drug delivery pumps. "Theres most interest in
pressure sensors by far, says Malaret, though
he also notes some interest in accelerometer
systems to monitor heart performance, and
in a next generation of multisensor clusters,
which get more reliable results by combining
input from multiple different sensors.
Among projects in the pipeline are small MEMS
transducers implanted in the eye or on a contact
lens to monitor intraocular pressure in patients
diagnosed with severe glaucoma, so treatment
with medication and physical shunts to release
the abnormal pressure can be monitored and
adjusted to prevent blindness. Work is also
underway on similar implanted pressure sensors
to monitor intracranial pressure in hydrocephaly
or brain injury, allowing doctors to tell for the
hrst time if the shunt thats supposed to release
pressure is actually working.
And theres increasing demand for more
sophisticated monitoring of blood pressure and
fow for improving cardiac care, from pressure
sensors implanted in the arteries and fow
meters using two pressure sensors to monitor
the patterns of fow and pressure through the
heart, to provide information to adjust ventricle
assist pumping devices and pacemakers.
"People are interested in understanding the
I S S U E N 5 J A N U A R Y 2 0 1 1
7 M E M S ' T r e n d s
5000
4500
4000
3500
3000
2500
2000
1500
1000
500
0
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Microsystem devices for healthcare applications market 2005-2015 in $M
(Source: BioMEMS report, October 2010, Yole Dveloppement)
Given Imagings PillCam
dynamic prohle of blood pressure, notes Ariel
Cao, director of business development at Tronics
Microsystems. "The industry is going to more detailed
measurement.
He says these various pressure sensor implant
development programs are now mostly in animal
testing, will move to human testing in 2012-2013,
and products will likely be on the market by 2015.
"And what were seeing is just the tip of the iceberg,
he argues, noting that theres also a lot going on in
house at the big medical device companies.
Tronics worked with MedTech for several years
to develop a small, hermetic pressure sensor in
biocompatible titanium packaging for a client, then
acquired the development company. It makes the
pressure sensors for all these different systems with
a standard design platform and process fow, typically
customized only by different dimensions for different
applications, and by either wireless or wired interfaces
in the ASIC. But the mechanical biocompatible
titanium or polymer packaging has to be adopted
to the interfaces required by the specihc fuids and
locations, whether blood in arteries, cerebral fuid in
the brain, urine in the bladder, or whether the sensor
needs to be integrated into a micropump or other
complete system.
Pressure sensors are also needed to monitor and
control drug delivery pumps, whether implanted or
external. Primarily used in systems to deliver insulin,
and to deliver chronic pain medication directly where
needed, the pumps use MEMS sensors to monitor the
fuid dynamics of delivery volume, and to make sure
the channel is free fowing.
The company says it also has quite a number of
systems under development for remote monitoring.
Though the smart phone now provides a convenient
ready-made consumer interface for tracking and
transmitting data to enable an affordable system,
the long approval process for medical devices, and
the necessity of demonstrating better outcomes for
lower cost from remote monitoring so insurers will
pay for it, likely means remote monitoring networks
remain some years in the future.
Proteus Biomedical enables systems
with alternative packaging
Proteus Biomedical (Redwood City, CA) takes a
different approach to sensing and reporting from
within the human body, by adding a detector chip
to any pill. These silicon chips include a thin hlm of
common mineral-based materials, which reacts with
fuid once in the stomach to send a high frequency,
not RF, electrical current through body tissue.
Company ofhcials are fond of pointing out that this
biogalvanic power system uses minerals similar to
those in common foods or vitamin pills.
The signal is picked up by a smart skin patch, which
records the time, the type of medication, and some
vital signs, using skin sensors that measure electrical
activity to monitor heart rate, some that monitor the
changes in the impedance of the electrodes as the
chest expands and contracts to measure respiratory
rate, an accelerometer to measure activity level and
position, and perhaps a glucose sensor that takes
its sample with a tiny needle through the skin. The
patch then communicates results by Bluetooth to a
mobile phone or computer.
The monitor patch got US FDA clearance last spring,
and the pill and patch system received the European
Union consumer CE Mark approval in August. The
system of pill, patch and smart phone app is in
various states of trials for monitoring compliance
and effectiveness of drug treatment for cardiac
conditions, tuberculosis, psychiatric disorders,
diabetes and organ transplants.
Though the smart pills have gotten all the recent
press attention, the company is also at work on
implantable sensors and stimulators using what it
says is a unique biocompatible chip-scale packaging
hlm, a thin and durable protective layer used with
MEMS-based chip-scale packaging that prevents
disintegration of the active electronics in the harsh
environment inside the body, without the need for
relatively large metal packages. This allows multiple
sub-millimeter MEMS sensors and IC controls to
be included on a single lead for a pacemaker, for
example, to intelligently sense movement of the
chambers of the heart and stimulate their proper
contraction across multiple sites.
The company has raised more than $100 million to
date in venture capital, include a Series E round last
year of $25.4 million from Novartis, Medtronic and
ON Semiconductor.
Paula Doe for Yole Dveloppement
J A N U A R Y 2 0 1 1 I S S U E N 5
Ariel Cao,
Director
of Business
Development
for North
America,
Tronics
Microsystems
During his career, Ariel Cao has
accumulated over +20 years of
industry experience covering marketing,
management, semiconductor and
MEMS technologies, EDA software and
telecommunications IC design as well as
a strong understanding of business and
industry issues.
Now with Tronics Microsystems as
Director of Business Development
for North America, he focused on
developing strategic partnerships
along the technology platforms for
custom component in application
domains such inertial or biomedical
systems.
Enrique Malaret, CEO,
Tronics MedTech
Enrique Malaret has over 28 years of
experience in medical device product
development. Mr. Malaret became a
founding partner MedTech Development,
in a medical product development
consulting company that was acquired
by Tronics Microsystems, N.A. in 2008.
Since the acquisition, Mr. Malaret
has served in the role of CEO for Tronics
MedTech, the medical systems division
of Tronics.
Ori Braun,
SVP, Given
Imaging
Mr Braun has served as Sr Vice President
- Business Development since October
2006. Prior to Given Imaging, Mr. Braun
served as President of Valor Inc. From
2004 to 2005, he served as President
of LifeWatch Inc. From 1985 to 2004
he founded and held various CEO and
executive positions at 3DV Systems
Inc., at Helios Software Engineering Ltd.
and at Lansoft Computing Ltd.
(Courtesy of Tronics Microsystems)
Miniature capacitive pressure sensor chipset
(Courtesy of Tronics Microsystems)

I S S U E N 5 J A N U A R Y 2 0 1 1
The new device, which checks levels of glucose,
cholesterol and 17 other substances in the blood, is
smaller and cheaper than conventional blood-testing
equipment, and targets small- and medium-sized
hospitals. Yole Dveloppement recently had the
opportunity to interview Mr. Hanshin Kim, Director,
HME Business Team at Samsung Electronics:
Yole Dveloppement: Samsung recently
launched a point of care analysis system.
Could you describe this technology?
Hanshin Kim: It is a fully integrated bench top
device that can perform multiple biochemical
analysis within 12 minutes. It features a sliding
tray approximately the size of a CD where blood
samples are placed, and the device automatically
performs the test and provides results at the press
of a button. The whole blood is applied directly to
the disposable "lab-on-a-disc containing different
kinds of freeze dried reagents for the blood test.
It is able to check 19 clinical chemistry analytes
including cholesterol, glucose level for diabetes,
and symptoms of heart disease or liver and kidney
impairments. It employs proprietary technologies
including micro-fuid mechanics and micro-valves
within the design.
YD: Could you describe a bit more the micro-
Huidic disc for our readers?
HK: Centrifugal force is utilized as propulsion force
in lab-on-a-disc. Various fuidic processing steps
such as the metering of liquid sample, mixing of
reagents, fuid separation, and sample splitting
are automated simply by implementing different
spinning prohles. And also the laser irradiated
ferrowax microvalves (LIFM) are embedded in
the discs for valving. Upon excitation with a laser
diode, iron oxide nanoparticles within the parafhn
wax act as integrated nanoheaters, causing the
parafhn wax to melt. Due to LIFM, various liquid
buffers and reagents can be prehlled in the discs,
and multiple analysis steps are integrated on a
single centrifugal microfuidic platform.
YD: The product has been launched
exclusively on the Korean market. What are
the different types of user today, and what are
your future targets?
HK: The main users are small clinics and hospitals.
It is used as a point of care diagnostic device for
blood testing in intensive care and emergency
units. The product is particularly appreciated by the
users because of fast turnaround time, accurate
as the laboratory analyzers, and ease of use. Our
objective is to provide the point of care technology
which will support physicians prompt and accurate
diagnosis. The next version of the product will be
suited to immunoassays as well, which will open
opportunities for bigger markets.
YD: How is the product distributed among the
users? Are you working with partners?
HK: Although Samsung Electronics has a signihcant
reputation and a strong network in South Korea,
we contracted a JW Pharmaceuticals, a Korean
pharmaceutical manufacturer, to promote and
distribute the product locally. Due to the fact
that the diagnostic market is a very conservative,
JW Pharmaceuticals, which already have a good
healthcare network, was chosen since this is a new
market for Samsung Electronics.
Interview Note
with Samsung Electronics
Samsung Electronics giant is looking to invest 1.2 trillion won ($9.85 billion) in the
healthcare business by 2020 with the goal of achieving 10 trillion won in annual sales
that year.
F E A T U R E S T O R I E S - B I O M E M S
Blood testing device (Courtesy of Samsung)
J A N U A R Y 2 0 1 1 I S S U E N 5
10
YD: When do you expect to propose this product on the
international market?
HK: Since it is the hrst product for Samsung Electronics on the
diagnostic market, we only launched the product domestically with
KFDA approval. We are planning to launch the product internationally
in the very near future, so we will prepare for the regulatory approvals
such as 510(k).
About Samsung Electronics
Samsung Electronics Co., Ltd. is a global leader in semiconductor,
telecommunication, digital media and digital convergence
technologies with 2009 consolidated sales of US$116.8 billion.
Employing approximately 188,000 people in 185 ofhces across
65 countries, the company consists of eight independently
operated business units: Visual Display, Mobile Communications,
Telecommunication Systems, Digital Appliances, IT Solutions,
Digital Imaging, Semiconductor and LCD. Recognized as one
of the fastest growing global brands, Samsung Electronics is a
leading producer of digital TVs, memory chips, mobile phones and
TFT-LCDs.
For more information, please visit www.samsung.com.
Entry system disc (Courtesy of Samsung)
Microsystems
devices driving
healthcare
applications
Discover the NEW
report on
i-Micronews.com/reports
BioMEMS
Mr. Hanshin
Kim, Director,
Samsung
Electronics
Mr. Hanshin Kim
is a director of
HME (Health & Medical Equipment)
Business team of Samsung Electronics.
He is in charge of IDS (in-vitro
Diagnostics Solutions) Group. He
received his M.S. in Biochemistry
from Yonsei University, Korea in
1993. From 2000 to 2009 he was
a researcher at the Samsung
Advanced Institute of Technology.
In 2005, he received MBA from
Sogang University, Korea. His
current research interests are
diagnostic system, POCT (Point of
Care Test) devices and connected
healthcare solutions.
M E M S ' T r e n d s
I S S U E N 5 J A N U A R Y 2 0 1 1
11
DNA, Cells, Proteins +
Semiconductor Technology
=
Life science innovation on fast forward
The hrst draft of the human genome was declared complete in the year 2000.
T
he 10 year, $3 Billion effort was funded by
the United States Department of Energy
and carried out collaboratively by large
academic genome centers. The goal of this effort
was to provide life science researchers with a
complete sequence of the human genome to
enable advances in the understanding of molecular
biology which would be used to create the means
to eventually cure basic natural defects found and
improve human life.
The process by which the human genome was
decoded was through the implementation of
genome centers. These centers were equipped
with rows and rows of capillary electrophoresis
sequencers instruments working in conjunction
with liquid handling robots running 24/7 decoding
a single human beings 3 billion base pair sequence
of As, Ts, Cs, and Gs. This $3B effort provided
the world with the first "reference human
genome and as it turns out the human genome is
approximately 99.9% the same but the differences
in that 0.1% give rise to the large variations in the
human population from physical characteristics to
susceptibility or predisposition of diseases.
Looking beyond the human genome there is the
need to decode the genomes of plants, animals,
and other organisms that are commercially vital or
ecologically critical to the life cycles.
Next generation sequencing
Fast forward 10 years to today and life science
companies like Illumina, Life Technologies, Pacihc
Biosciences, Complete Genomics, and Roche
have leveraged semiconductor technology and
processes to develop "Next Generation Sequencing
instrument platforms. The development of these
instrument platforms leverage the ongoing advances
in semiconductor processes to gain the advantages
of highly reproducible microstructures that can
be patterned or arrayed into silicon or glass and
in some cases leveraging sensing electronics
directly in the same "chip eliminating the need
for expensive optics and discrete sensing systems.
The culmination of advances in chemistry, sensors
and microfuidic/MEMS technology has fueled the
continued drive to decreasing sequencing costs
with the current goal of $1000 per human genome
set by the X-Prize for Genomics and the reduction
in sequencing time to approximately several
weeks. The realization of this goal will likely be
achieved in the next year and will continue to
drive the adoption of next generation sequencing
platforms by researchers around the world.
SVTC
Commercialization Gap
Universities
and Labs
Foundries
Captive Operations
Bridging the Gap
Research Development Manufacturing
Figure 1 - Bridging the Lab to Fab Commercialization Gap
Michael Lee - Director
of Marketing, Life Science
Mi chael Lee studi ed El ectri cal
Engineering at Carnegie Mellon
University and completed his SMEE
at the Massaschusettes Institute of
Technology. Michael has held various
engineering and marketing roles at
Polaroid Medical Imaging Systems,
Actel, Mentor Graphics, and most
recently at Fluidigm Corporation
where he was managing all of the
marketing efforts for the BioMark
Genetic Analysis System. In 2010
Michael joined SVTC Technologies
as the Director of Marketing, Life
Science.
J A N U A R Y 2 0 1 1 I S S U E N 5
Innovation on fast forward
The use of semiconductor technologies has had a dramatic effect
on Next Generation Sequencing. It has sparked a race in the life
science tools industry where the decreases in cost and increases in
data output has exceeded that of Moores Law. Innovation continues
to be driven in the sequence space as nanopore sequencing that is
being developed will be the "next Next Generation Sequencing and
this will also leverage and push semiconductor technology from both
the process and detection/sensing perspectives. Nanopores on the
order of 1-3nm will need to be arrayed and instrumented on CMOS
to detect the nucleic acid bases passing through the nanopores.
The resulting "chip that will ultimately be commercialized will
revolutionize the sequencing market.
As the world of sequencing has been placed in fast forward, the
same semiconductor technologies and processes are already being
applied to the protein, high content screening, and drug delivery
markets. These markets will likely see similar gains thus advancing
the research and applications that will beneht the life science
market.
Lab to fab
Innovations that are developing these "next generation platforms
typically come from internal company research groups or university
labs. As development moves forward and a proof concept is achieved
the next steps are to make the move from "Lab to Fab. From the
perspective of developing the microfuidic/MEMS chip the requirements
typically cannot be met with standard fabrication processes of CMOS
fabs. SVTC is a nanotechnology development services company that is
uniquely positioned to help bridge the gap to commercialization (Fig. 1)
and has been helping our customers move from the "Lab to the Fab
since the company was founded in 2004. An advanced technology and
secure IP infrastructure combined with 24/7 state-of-the-art production
capabilities to support companies that are developing novel microfuidic,
MEMS, and semiconductor devices for the Life Science, Aerospace and
Defense, Semiconductor markets. SVTC provides access to over 250
production ready semiconductor tools in over 95,000 square feet of
clean room space and process libraries with over 2500 recipes on
200mmwafers. SVTCs engineering services provides access to a highly
skilled and experienced engineering teamto support development from
prototype to production and work collaboratively with our customers
as an extension of their own teams. The SVTC development model
is designed to support unique materials and advanced lithographic
processes and long with a very fexible engagement model that enables
rapid product commercialization while enabling customer intellectual
property independence.
www.svtc.com
12
Overview
of the technical
& economic
requirements
of POC
applications
Discover the NEW
report on
Point of Care
Testing
Applications
of Microfluidics
Technologies
I S S U E N 5 J A N U A R Y 2 0 1 1
Nano Imprint Technologies,
the most exciting recent
advances, by EV Group
Yole Dveloppement interviewed EV Group about its Nano Imprint technologies.
What are the current technologies challenges? What are the main applications of
such innovative process? How EV Group addresses the market? . This discussion
will give you an overview of Nano Imprint Lithography (NIL), its benehts and the EV
Group approach.
Yole Dveloppement: What are the recent
advances in Nano Imprint technologies?
Ron Miller: I think that some of the most exciting
recent advances in Nano Imprint Lithography (NIL)
have to do with lowering both the cost of ownership
and the process barriers to bringing the technology
to a high volume manufacturing environment.
These advances are really on two fronts. The
hrst is the advances to equipment throughput
of the various imprint processes available (UV-
NIL, Hot Embossing and micro-Contact Printing).
Only a few years ago, the majority of available
equipment for nano-imprint lithography processes
had throughputs that could only facilitate R&D and
prototyping applications. Today there are systems
available with throughputs approaching, and in
some cases exceeding, 40 to 50 wph.
The other main aspect driving the reduction
in cost of ownership is the work that has been
done to develop more cost efhcient processes
and materials for working stamps. Typically, in
the imprint process, a hard master is made via,
for instance, e-beam lithography. This allows for
very hne feature dehnition. However, writing hne
features over a large area with e-beam lithography
can be both time consuming and costly. Thus, a
hard master is a high value part that one would like
to avoid using directly for imprinting. In addition to
this, and particularly with UV-NIL, a hard master
will need to have an anti-stiction layer applied to
allow for facile and defect free release - adding
extra cost and time to the overall process.
To avoid using the high value hard master during
the imprint process, a working stamp can be made
from the hard master and used for imprinting. This
working stamp methodology has been around for
quite some time - and has employed soft masters
(with materials like - polydimethylsiloxane (PDMS)
- for prototyping) and hard masters made of thin,
fexible pieces of etchedglass for volume applications.
However, problems arise with materials like PDMS
since they are difhcult to use in stamp fabrication.
They can absorb solvents (thus causing hdelity
issues) and have a limited lifetime. Hard working
stamps, on the other hand, require several costly
steps during fabrication and thus require additional
and costly processing and infrastructure. But recent
developments of new material sets and processes
now allow for rapid fabrication of polymeric
working stamps. These stamps have the capability
to dehne feature sizes from < 15 nm all the way up
to millimeter sized structures. They can be used
for thousands of imprints and, because of their
inherently low surface energy, do not require the
application of an anti-stiction coating. Additionally,
the materials have thermal stability to around
200C allowing them to be used for hot embossing
of the majority of relevant thermoplastic materials
for microfuidic applications rather than just UV-
NIL process fows
Ron Miller, Imprint
Lithography Business
Development Manager
and External R&D Programs
Manager, EV Group
EVG770 Step-and-Repeat System (Courtesy of EVG)
J A N U A R Y 2 0 1 1 I S S U E N 5
YD: How does Nano imprint differentiate from
other structuring technologies?
RM: NIL provides the lithography community
with several unique advantages over other forms
of lithography. For instance, NIL can be used to
pattern very hne features (again, < 15nm) with
a low cap-ex/low cost-of-ownership. Additionally,
NIL can directly pattern 3D structures, aspherical
patterns and non-repeating patterns which would
not be possible with some types of conventional
lithography methods. The use of the working
stamps mentioned above also allow for cost-
effective, rapid prototyping of devices.
Competing methods for the various imprint
lithography applications include a wide variety of
technologies such as immersion lithography, EUVL,
massively-parallel e-beam lithography, interference
lithography, self-assembling monolayers and
injection molding. Each patterning technique
has drawbacks and benehts in comparison to NIL
processes. For instance, immersion lithography
(and, in the future EUVL) have been developed
for high throughput, high value applications and
will most assuredly dominate that space going
into the future. However, their equipment costs
make them prohibitive to other products (e.g. laser
gratings on a photonic integrated circuit for internet
communications) that are either more cost-sensitive
or not as high volume as memory or high-end logic.
Thus immersion lithography and EUVL do not really
directly compete with NIL at this time.
Both interference lithography and self-assembly will
most likely occupy nice niches in the nano-patterning
space. However both are faced with the limitation
that they can only pattern periodic features with
limited pitch variation.
Massively-parallel e-beam lithography is still in
what I consider an early development stage. But,
if the problems associated with this technology
- which basically boil down to controls and data
management and transfer - can be overcome in
a cost-efhcient manner, this will be an interesting
patterning technology.
And hnally, micro (and nano) injection molding
is a high volume alternative to hot embossing
of polymeric substrates for the microfuidics
application space. The initial cost for the molds
for this technique is quite high and requires a
set design and a high volume piece part to make
sense economically. But, at high volume, injection
molding will provide a good value proposition for
certain parts. However, even in high volume, hot
embossing will still be the patterning solution
of choice for certain devices with structures
that present a particular challenge for injection
molding. These typically are structures that have
high aspect ratios and/or high density such as
capture pillars for a Circulating Tumor Cell device.
Also injection molding can encounter issues with
devices that are very thin or require high optical
14
Exploded image of a CMOS Image Sensor stack (Courtesy of EVG)
"The majority
of EVG activities
in NIL are focused
on three application
spaces where
we see the most
end-user activity,
specically optics/
light management,
mastering and
microHuidics."
I S S U E N 5 J A N U A R Y 2 0 1 1
clarity. Hot embossing, particularly in conjunction
with the working stamp processing mentioned
above, presents a very nice value proposition for
rapid and cost-effective microfuidic prototyping
when compared to the high cost and long lead time
for injection molding tooling.
YD: What are the main applications today,
and what are the growing areas?
RM: The majority of EVG activities in NIL are
focused on three application spaces where we
see the most end-user activity (and thus market
potential), specihcally optics/light management,
mastering and microfuidics.
In terms of light management applications, EVG
has found a very receptive market for imprinting
of lenses for Wafer Level Optics (WLO). The WLO
market remains mostly focused on the wafer
level packaging of CMOS Image Sensors (CIS) -
particularly for mobile phone applications. Our
activities in this market have allowed us to rehne
our imprint lithography processes, equipment
and software in an automated, high volume
environment for wafer sizes up to 300mm. Thus,
while the CIS application is not "nano in feature
size - the overall process fows and, to some
extent the material sets, are virtually the same
and have enabled us to run our equipment and
process solutions in high volume manufacturing
environments. This multi-year manufacturing
expertise allows EVG to address the emerging
market for automated nano-scale processing.
This has opened-up a wide held of processing
capability that can be applied to NIL patterning in
light management applications mostly focused on
dehning features such as gratings, waveguides,
resonators and quasi-photonic crystals. The main
application spaces for these applications are
communications, displays, and LEDs.
We believe that NIL will also play an important role
in mastering - or patterning very hne feature sizes
over large areas in a cost effective manner. EVG
addresses this market with our EVG770 NIL Step-
and-Repeat System. With this system we are able
to imprint a repetitive pattern over a large area - up
to 300 mm. Feature sizes can be well below 20nm.
Additionally, we have developed methodologies for
employing working stamp material onto a standard
NIL stepper template in order to facilitate both rapid
prototyping and fabrication. These master stamps
can play a variety of roles but will mostly be focused
on stamp fabrication for roll-to-roll processing,
although we do see a variety of potential standard,
batch processing applications on large areas.
The main application areas here will be light
management for displays, security features and
photovoltaics. We also see potential in life science
applications for large area sensor arrays.
EVG believes that the emerging held of microfuidics
will offer a signihcant market opportunity for
patterning of polymeric substrates at the macro,
micro and nano scale. Along those lines, we have
recently shipped an EVG750 fully automated
embossing system to our North American partner
- Industrial Materials Insitute, NRC, Quebec. With
current conhguration and processes, this system
can achieve over 10 embossed substrates per hour.
Also, our working stamp methodology mentioned
above means that we can offer our customers short
development cycles and reduced cost of ownership
for rapid prototyping hot embossing process
services. This allows them to gain access to molded
microfuidic parts in a very short period of time
(days) versus getting a metal stamp manufactured
for either embossing or injection molding (weeks).
About EV Group
EV Group (EVG) is a world leader in wafer-
processing solutions for semiconductor, MEMS
and nanotechnology applications. Key products
include wafer bonding, lithography/nanoimprint
lithography (NIL) and metrology equipment,
as well as photoresist coaters, cleaners and
inspection systems. With over 35% market
share (installed base of over 150 systems) EVG
holds a leading position in NIL and lithography
for advanced packaging and MEMS.
Other target semiconductor-related markets
include silicon-on-insulator (SOI), 3D interconnects,
compound semiconductor, silicon-based power-
device solutions and microfuidics.
Founded in 1980, EVG is headquartered in
St. Florian, Austria, and operates via a global
customer support network, with subsidiaries
in Tempe, Ariz.; Albany, N.Y.; Yokohama and
Fukuoka, Japan; Seoul, Korea and Chung-Li,
Taiwan.
www.evgroup.com
Super dense, high aspect ratio separation columns
for capturing species such as: proteins, nucleic acids
and whole cells (bacteria, cancer cells, etc) embossed
into a polymer substrate on the EVG750 with working
stamps The columns are 30m tall by 9m wide
with roughly 4m separation Image courtesy of the
Industrial Materials Institute, NRC, Quebec, Canada
Ron Miller has over 13 years
experience in the semiconductor
equipment and materials space in both
large and small company environments.
He has been with EV Group for over six
years starting off in technical sales.
Currently he serves two roles at EVG
with world-wide responsibilities
- Imprint Lithography Business
Development Manager and External
R&D Programs Manager. In the latter
role, Ron directs EV Group activities
relating to universities and start-up
companies, to include R&D and
intellectual property development.
He has co-authored numerous
technical papers and articles focusing
on imprint lithography equipment and
processes and serves as the Chairman
of the NILCOM Consortium. Ron has
a B.S. in Chemical Engineering from
the University of Illinois, Champaign-
Urbana and is a veteran of the United
States Navy.
J A N U A R Y 2 0 1 1 I S S U E N 5
M
ajor segments driving growth require
products with entirely different requirements
for everything down to the substrate
materials. Point-of-care clinical diagnostics
devices demand low cost, disposable cartridges
in automated testing systems to make the tests
affordable. The R&D market, in contrast, looks
for very precise, very complex chips to quickly
test one sample against thousands of targets at
once, to save researchers time and replace large
complex equipment to bring its big savings.
But the choice is not as simple as just low cost
plastic vs. more precisely patterned glass.
Production volume, application, type of patterning,
and optical properties all impact material choice,
for both cost and performance.
Market size is one consideration. Polymer costs
a fraction of a cent per square centimeter, while
glass costs $.02, but glass is actually the lower
cost choice for all but high volume production.
A device on glass can be prototyped and then
directly scaled up to volume production on the
same equipment, so manufacture of low to mid
volumes is cheaper than with polymer. Injection
molding of polymer devices requires hrst making
a costly mold, so costs come down only when that
can be amortized over high volumes. Those volume
requirements also mean that injection molding
is not practical for making prototypes, so that
s usually done using PDMS or some other cheap
and convenient material, so some redesign will
often be required to port the process to injection
molding for production.
The desired features are another issue. Very hne
features can be precisely etched on glass, for
higher resolution than molded plastic. Glass is
also a good insulator, making it easy to provide an
electric potential across the chip if needed. On the
other hand, polymer provides greater freedom of
design for large reservoirs and wells, sometimes
needed in great numbers across plates for testing
thousands of targets, and they can be expensive to
etch or sandblast in glass. Polymer is also good for
integrating valves and pumps, as silicone hlm can
be easily added between polymer layers for fexible
features easily actuated by pressure, for clever
solutions like those from Rheonix and Fluidigm.
Fragmented microHuidics market
means fragmented materials
opportunities
4500
4000
3500
3000
2500
2000
1500
1000
500
O
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Analyt cal devices
Pharmaceutical and life science research
Point of Care diagnostics
Clin cal & veterinary diagnostics
Drug delivery (inhalers, micropumps, microneedles)
Industrial and environmental
"Redesign is often
required to port the
process for large
scale production,"
explains
Frdric Breussin,
Yole Dveloppement
Frdric Breussin,
Project Manager,
Microfuidics,
Yole Dveloppement
Driven by impressive recent progress in automating the identihcation of genes and
proteins, the microfuidics market is approaching ~$1.5 billion and is on a steady
~20% annual growth path going forward. But the highly diverse range of products
in the sector means highly fragmented demand for processes and materials.
MicroHuidic devices market in $M
(Source: Emerging Markets for Microfuidic Applications Report, August 2009, Yole Dveloppement)
Optical properties are still another consideration.
Since most microfuidic devices depend on optical
readouts, they need optical access, for which the
transparency of glass is an advantage. Not only
are many traditional polymers not transparent, but
they may fuoresce at the same wavelengths as
used for detection, so more expensive and hard-
to-process alternatives like COC may have to be
used instead.
Silicon is most used by companies with
semiconductor production experience. And it
makes most sense to use the higher cost material
when silicons active properties are needed to
integrate the processing on the chip.
That means that almost half thecurrent microfuidics
market, by value, opts for polymer, as plastic is
the preferred choice for low cost diagnostics and
point of care devices. Polymer is also used for
the majority of microfuidics for pharmaceutical
research, which often require arrays of wells that
are easier to make in polymer. Many of these
devices were developed for diagnostics markets
needing low cost disposables, but are currently still
used mostly in research.
Silicon and glass each account for nearly equal parts
of the remaining half of the market. Glass is used
for the majority of analytic devices, particularly
for sample preparation for mass spectrometry
and high pressure liquid chromatography. Silicon
dominates the drug delivery applications, which
consist mostly of inhaler nozzles for asthma
medication and micro insulin pumps. Its also
widely used in the small but developing market
for sophisticated and reusable industrial and
environmental sensors.
Others are developing other approaches as well.
Nano imprint technologies may have potential, like
Sonys stamping/embossing replication technique
based on that used to make CDs, though feature
depth is limited. Yet another approach, taken by
Sophion Bioscience, is to combine glass, silicon
and polymer in one device, making the reservoirs
in polymer, the channels in glass, and the controls
in silicon, then assembling the separate parts into
a complex multiplexing device for drug research.
Production technology on all these substrates is
maturing, with more professional, automated,
quality production facilities now becoming readily
available. Typically microfluidics foundries,
such as Dolomite, Micralyne and Microfluidic
ChipShop, have worked mainly with R&D groups
to do relatively small production runs. But now
we see companies leveraging their equipment
from volume production of MEMS to offer mass
production of microfuidics as well, particularly on
glass and silicon.
With the impressive progress coming out
of biomedical research labs in identifying
biomarkers-and in starting to make progress
on the equally challenging issue of hguring out
practical ways to apply this new knowledge to
improve peoples lives-there looks like plenty of
opportunity for both companies that target low
cost, high volume markets, and those that target
the high margin niche markets.
I S S U E N 5 J A N U A R Y 2 0 1 1
"Silicon makes
most sense when
active properties are
needed to integrate
the processing
on the chip," he added
GC, HPLC)
environnment
applications
Drug
Delivery
Total
microfluidic
device market
MicroHuidic devices market in $M
Material breakdown
(Source: Emerging Markets for Microfuidic Applications Report, August 2009, Yole Dveloppement)
Frdric Breussin is expert in
Microfuidics for diagnostics and life
sciences. He has supported many
companies in their innovation and
product development strategy in
making the bridge between micro
systems technologies and their
applications in Life sciences, diagnostics
and medical device industries. He holds
an Engineering diploma from INSA
Rouen and a DEA in fuid mechanics
from University of Rouen.
Others
Si
Glass
Polymer
J A N U A R Y 2 0 1 1 I S S U E N 5
T
he bulk of the chatter last year came
with the acquisition of Ion Torrent by Life
Technologies in August, the IPO of Pacihc
Biosciences in October and the announcement
in June from GnuBio, a new startup out of the
Weitz lab at Harvard, proposing a $30 cost for
sequencing the human genome. The acquisition of
Ion Torrent (which just launched its product line in
late 2010) by Life Technologies was for U.S.$375.0
million in cash and stock, with an option to
increase by a further U.S.$350.0 million if certain
technical and time-based milestones are met by
the end of 2012 and the IPO of Pacihc Biosciences
which sold 12.5 million shares for $16 per share
netted the company U.S.$200.0 million on sales of
U.S.$135,000.00 in 2009. GnuBios sequencing is
based on the microfuidic droplet technology out
of the Weitz lab, which also produced RainDance
Technologies. GnuBio seems to be moving quickly
as it closed a Series A round of U.S.$3.0 million in
December.
What is the scientihc basis for all of this excitement?
The use of molecular diagnostics testing platforms
that use either genomic or proteomic technologies
have increased dramatically. Currently the fastest-
growing business segment in clinical diagnostics,
clinicians hnd the tests highly valuable in helping
them select appropriate treatment strategies.
These new genomic and proteomic technologies
generally include analysis of deoxyribo-nucleic acid
(DNA), ribonucleic acid (RNA), or protein-specihc
targets to predict a disease or predisposition of
an individual to a disease (personalized genetic
medicine). Before development of molecular
diagnostic tests, clinicians were limited to less
sensitive assays that did not capture either host
or infectious pathogen genetic information. Today,
a variety of molecular genetic diagnostic tests are
used for routine patient management. The analysis
of specihc gene sequences provides important
diagnostic information for identifying a pathogenic
agent, predicting disease, monitoring disease,
choosing treatment options, and subsequently,
for determining the effectiveness of the specihed
therapy or treatment. Additionally, novel molecular
technologies with diagnostic applications have
become an important tool in drug discovery and
preclinical drug development.
With little or no revenues to show from these
companies at this stage why are hnancial and
strategic investors participating in or watching
thesedevelopment withsuchahighlevel of interest?
Less than hve years ago, the goal for inexpensive
sequencing was $1000 per genome, the number
believed to be the threshold at which doctors and
insurance companies would begin adoption. Now it
is becoming a reality. In addition to the price issues
is the need is for simple-to-operate, portable, and
affordable small sequencers that can be integrated
into point-of-care diagnostics, as well as used for
research in smaller laboratories. This issue has
been take care of too. The use of small sequencers
is likely to expand considerably, particularly
with the increasing acceptance of personalized
medicine at the clinical level. In addition, there
are over 300,000 scientists working in smaller
peripheral laboratories worldwide, and as a result,
there is a market potential for at least 50,000 small
sequencers over the next hve years. The revenue
model these companies have adopted at this stage
is one where equipment is sold and signihcant cash
fow created by chips and reagents each time a
sequence is run.
What does the amount to in terms of market size
right now? When you look at the numbers the
worldwide molecular diagnostic market, including
peripheral technologies, in vitro and in vivo
diagnostics, and molecular imaging, was worth $9
billion in 2009. It is estimated that the market will
expand to $15.5 billion in 2014 and $42.5 billion in
2019. It is a huge market and can be very prohtable
for the leading players with the right technology
platform and distribution channel reach. New
technologies have been responsible for 15 years of
steady growth in the sector. The value of the DNA
sequencing market in 2009, including reagents and
instruments, was $700 million worldwide, with U.S.
sales contributing over $500 million of that. The
market is estimated to expand to $900 million by
2014 and $1.8 billion by 2019.
The gene sequencing gold rush goes
to the next level
While the hrst wave of Genomic "Gold Rush talk and how billons of dollars was
going to be made from it was all over the place way back in the year 2000 the talk
went to the next level in 2010.
Patrick Keating,
Senior Partner,
Yole Finance
What are the risks? As you can see the total existing and forecasted
market size for DNA sequencing compared to the amounts invested
in Pacihc Biosciences and paid for Ion Torrent would seem to indicate
a lot of money chasing a small market. Yes, it is still early in the
adoption phase for use at the patient and drug-discovery level and
market estimates are hard to formulate with such a new technology.
However, in general, peoples DNA doesnt change, so they only
need it to be sequenced once. In other words, no repeat customers.
Sure, there are a lot of potential customers right now; most of the
people that you know probably havent had their DNA sequenced.
But as the price of sequencing goes down, the barrier to entry for
the industry is likely to fall as well. That could result in a low-margin
business for this application thats counting on a dwindling customer
base. Of course, the drug industry could change the dynamics of
the whole market as it needs different ways to speed the discovery
of new drugs, hnd reuse for old drugs and enhance the delivery of
existing drugs. On the other hand, apart from the market value in
dollars, another risk indicator has been an increase in the number
of major players involved in the sector. The number of companies
actively involved in developing molecular diagnostics increased from
less than 100 in 1995 to over 500 in 2010. Surely, there will be
more new sequencing technologies developed that will threaten the
livelihood of current and emerging players in the market.
In the past the different scenarios for platform technologies have
seen many different transformations of the early players and a wide
variety of results from a prohtability standpoint too. In fact, some
platform players even went down the path of trying to morph into
drug companies as they found their original markets limited and
the lure of hnding the next billion dollar biotech drug with their
propriety knowledge, libraries and technology platforms potentially
more prohtable for them. While it is truly exciting to see the
amazing things these emerging technology platforms can do from
a scientihc standpoint it still remains to be see what will emerge
from a business standpoint in the years ahead. Certainly money has
been and will be made in the gene sequencing market. But, as an
investor, you should do your homework and stay ahead of the curve
in this complex and fast-changing industry.
About Yole Finance
Yole Finance is a boutique hnancial advisory business unit of the
Yole Group of Companies, with an international franchise and
one of the strongest capabilities in advising companies in the
MEMS, compound semiconductors, advanced packaging, Opto
and RF, microfuidics, nanotechnology, photovoltaics and power
electronics helds. As a boutique hnancial advisory hrm Yole
Finance offers a full range of services:
- Capital Raising
- M&A
- Research
- IP Value Creation
For more information, please visit www.yolenance.com
A new tool
offering a real
and
comprehensive
picture
of microfluidics
community
Discover the NEW
report on
World
Microfluidic
Players
Database
I S S U E N 5 J A N U A R Y 2 0 1 1
I N D U S T R Y R E V I E W
MEMS is a very fragmented market
> A limited number of applications have a market
size above $200M (less than 10).
> Simplihcation of manufacturing is still an
objective: The Yole Dveloppement MEMS law
One product, one process, one packaging is still
there.
> MEMS packaging and software development are
more and more adding value: sensor integration
(in silicon or in SiP) and sensor fusion are key
challenges for the industry.
> The development of new MEMS applications is
taking years to be commercialized: In average, 4
years from hrst developments to hrst commercial
product and $45M of investment . and several
CEOs.
But on the other hand, the business has dramatically
changed due to the growth of the consumer
electronic applications of MEMS devices.
> 8 manufacturing infrastructure (either internally
or via an adapted supply chain) is a must have for
companies looking to compete for the long run in
the consumer electronic business.
Process optimisation
It is step by step happening in large companies.
But MEMS foundries are really not pushing such
optimisation, mainly because it is not a request
from their customers. We see more and more
manufacturing platforms in place, which is a good
way to manage the diversity of MEMS front end
and back end processing. Yole Dveloppement
has just released a very new report analysing the
manufacturing process evolution (in technology and
business) and how the companies are trying either
to decrease the complexity of the MEMS front end
process... or embedding it into a plate form.
So what is new for 2011? Well, from my point of
view with 20 years of active analysis of the MEMS
market, I am really surprised by the new business
model which is spreading very rapidly among MEMS
companies:
What can we expect in 2011
in the MEMS business...?
Growth, Growth, Growth... What is funny in the MEMS business is that key structural
aspects of the business have not changed since more than 15 years.
3-axis
gyroscopes
3-axis
accelerometers
3-axis digital
compass
3A + 3M
IMU:
3A + 3G
3A
+ 3G
+ 3M
- Market demand is
validated
- 2009: 10% of compass
are embedded in 3A+3G
modules
- Challenge: compass
location is critical
(interferences.)
=> such integration
is expected
to stay limited to specihc
platforms for the years
to come
- When gyros will become
popular: interest to
integrate them with
accelerometers
- Sensor fusion with
accelerometer data
will provide additional
features (navigation.)
- Similar manufacturing
platforms and
constraints for
integration
=> expected to surge
quickly
- Long term view: demand will start only when 3-axis
gyroscope will become a commodity
Motion sensor combos: strategies of integration
(Source: Motion sensors for mobile and consumer applications report, February 2011, Yole Dveloppement)
3A: 3 axis accelerometers
3M : 3 axis magnetometers
3G : 3 axis gyroscopes
J A N U A R Y 2 0 1 1 I S S U E N 5
22
What will
be the future
of MEMS
manufacturing?
Discover the NEW
report on
Trends
in MEMS
Manufacturing
& Packaging
- Companies are more and more supplying bare MEMS die to
customers... but also competitors. Just take few examples:
> Inhneon is selling silicon microphones mostly as bare die to ECM
manufacturers, its direct competitors. BSE, Hosiden, AAC.. are
among Inhneons main customers... and competitors.
> Omron is selling its microphone die to ST which is using its sales
and marketing power to promote it as its own product.
> Multiple inertial sensor companies have only one device
(accelerometer, gyroscope or electronic compass) but are offering a
complete product portfolio by integration of competitor devices in their
product family: Kionix, AKM, SensorDynamics, Analog Devices... all
these companies are involved in such deals.
> Microbolometer manufacturers are doing the same.
How long can such collaborations stay alive? We will see in 2011...
- Sensor fusion (and linked software knowledge) is more and
more important:
> In inertial sensing, trying to provide a 6 or 9 axis IMU (even for
consumer electronics) without a strong software engineering team
is just nonsense. Companies have been created in order to support
such evolution, like Movea.
> In image sensing, more and more applications are requiring
visible image picture but also infrared imaging. The Microsoft Kinect
is an example of such sensor fusion needed to provide the functions
for the applications targeted.
So what can we wait for 2011? Clearly healthy growth is here and will
diffuse in all part of the industry. Of course consumer electronics will
continue its adoption of MEMS devices but also industrial, medical,
defence, telecommunication... applications are all integrating more
and more MEMS devices for key functions, providing the companies
outside the large volume applications very strong business
perspectives and growth opportunities.
As the companies are able to provide nice forecasts, more M&A will
happen: end of 2010 has seen these signs of increased M&A (see
related article by Yole Finance) and more will happen.
But one thing remains a key problem (or a great innovation provider):
packaging of MEMS devices is more and more important. It is still 30%
to 40% of the cost of a device but it is becoming not only a cost but
a way to innovate, in order to make thinner devices, integrate more
functions, moving to the 3rd dimension...
So, yes, the MEMS business is really moving fast and 2011 will be
another incredible year of innovation, product launch, growth... and
M&A.
Jean-Christophe Eloy, President & CEO,
Yole Dveloppement
B I L L E T
F
irst, as 2010 came to a close, we have identihed
hrst signals of what could be a re-org of the
MEMS business and supply chain. This is just
happening for the microphone business. If we look
are the supply chain, we will see fragmentation of
the activities with some processing wafers, others
packaging the dies and hnally a third player selling
the microphones. This change could also impact other
MEMS devices as well such as inertial. It means that,
at the end, the sensor will not be sold by the one
that make the MEMS, and this change will not concern
only MEMS foundries.
Another point could be a MEMS company going
IPO in 2011. There has been a lot of buzz about
InvenSense IPO at the end of 2010 (which was
postponed). While there are not so many MEMS
companies that are likely to go IPO for 2011, we do
believe another IPO could happen in 2011. On the
M&A side, we believe more M&A are coming: end
of December Dalsa has been acquired by Teledyne.
This is the hrst starting point of a new wave of
change of ownership for MEMS companies.
At the system level, multi-chip integration module is
becoming increasingly used. This is particularly true for
consumer applications for IMUs devices. In 2011, we
will see as common the integration of accelerometers
and gyroscopes and magnetometers as well. One
likely evolution will be the integration of a pressure
sensor, still for positioning applications (altimeter
measurement for GPS). In general, inertial sensors will
be again key growth areas for MEMS devices.
Another important fact is that 2011 could also
be the year when RF MEMS switches will enter
into volume production, specihcally for cell phone
applications. At the end of 2010, we have identihed
two players in qualihcation / pre-production with
big plans for 2011.
Another topic is the growth for the uncooled IR
applications. Uncooled IR players are struggling
to get the cost down to enter the consumer
applications. This is a challenging area for MEMS.
Although microbolometers will be the dominant
uncooled IR technology for the next years, we are far
from a US$10 cost target. We see two approaches
competing together. The microbolometers players,
working hard on the design, materials and packaging
side are trying to lower the cost. Big advantage is
that resolution is good. On the other side, other
approaches (e.g. using pyroelectric sensors) are
being developed to increase the resolution while
keeping very low cost. Through 2011, we believe
some announcements could prove which way will
be the best in efhciency/ cost effectiveness.
The optical MEMS market for fiber optic
telecommunications will prove it is rising again. After a
10-year long cold winter, MEMS-based micro-mirrors
are back on the scene for wavelength switching. For
those of you who have known the 2000 Telecom
Bubble Burst, this will be refreshing news!
On the manufacturing side, an interesting trend
started a few years ago to achieve standardization
of MEMS processing. The different approaches
included both technology platforms and product
platforms. Both aimed at developing custom-
made standard MEMS manufacturing processes.
An example is Dalsa with its technology platform.
The objective will be for the foundries to lower the
development time for new products, to decrease
the time-to-market and of course to lower the cost.
As second-source manufacturing partnership is still
scare in MEMS, such approaches could open the
way to standard MEMS processing.
In 2011, we also believe CMOS-MEMS will stay a
niche technology as room temperature bonding
will allow CMOS & IC interconnection without the
CMOS constraints. Some MEMS will still use CMOS
processes because of array structure (micro-
mirrors) or because of the proximity between IC
and MEMS (e.g. oscillators). But as an example,
Microbolometers are today shifting from CMOS
MEMS where MEMS is processed above the ROIC to
wafer bonding approach.
On the material side, we also see two materials
having more and more importance in the MEMS
manufacturing. Oneis glass. Glass wafers arealready
used for capping and in some cases, with Through
Glass Vias (TGV) for electrical interconnects. But
glass wafers will gain more importance through
microfuidics applications as well. Another material
back on the MEMS scene is PZT. Piezo MEMS seem
to have new interest and this is also largely because
of new promising applications such as Wafer Level
AutoFocus. On the European scene, new European
projects are currently studying the different
processing approaches for PZT thin hlms deposition.
To conclude, there will be many interesting stories
to follow for 2011. Stay tune to our different MEMS
Medias. This is Yoles everyday goal to keep you
aware of latest MEMS business evolution.
Dr Eric Mounier, Project Manager,
Yole Dveloppement
MEMS facts to watch in 2011!
I S S U E N 5 J A N U A R Y 2 0 1 1
For this hrst issue of MEMSTrends 2011, we would like to highlight what we think will
be the most important MEMS areas to watch in 2011.
Dr Eric Mounier,
Project Manager,
Yole Dveloppement
Wafer level packaged MEMS device
with through silicon vias
(Courtesy of Dalsa)
C O M P A N Y I N S I G H T
J A N U A R Y 2 0 1 1 I S S U E N 5
Introduction
SensorDynamics is a well known manufacturer
of automotive grade MEMS gyroscopes and
inertial combo sensors such as SD721 (single
axis gyroscope), SD755 (gyroscope plus
accelerometer) and SD78x family (gyroscope plus
3D accelerometer) mainly for Rollover and ESC.
Driven by these applications considerable advances
in reduction of manufacturing costs, package size
and power consumption have been made.
With the new family of gyroscopes in QFN package,
namely SD705 (in-plane single axis gyroscope),
SD706 (out-of- plane single axis gyroscope), SD742
(dual axis gyroscope), SD740 (3D gyroscope) and
SD746 (3D gyroscope plus 3D accelerometer)
SensorDynamics now applies this know how to
the non-automotive market where many new
applications are arising. For instance, today, even
small handheld devices have enough computing
power to handle the signals of several inertial
sensors for solving complex tasks like gesture
recognition. Consequently, SensorDynamics today
provides a fully integrated 6DoF-IMU in a tiny
6x6x1.2 mm QFN40 package with a moderate
power consumption of only 16 mW for the consumer
market.
SensorDynamics is addressing the high-end
consumer, industrial, automation, medical and
health-care market and therefore uses the stable
and proven QFN package, a package that allows
sensors to be calibrated with low bias and sensitivity
errors over a wide temperature range up to 125C.
Technology
The SensorDynamics MEMS fabrication makes
use of the reliable PSMX2 process. This silicon
technology based process provides a 10 to
20m thick polycrystalline silicon for the micro-
mechanically active structure, thus making it best
suited for single axis, dual axis, and triple axis
gyroscope sensing elements with high resonant
frequencies, which is highly desired to give the
structures a high resistivity against shock and
vibrations.
Figure 1 shows the mechanically active structure
of a 3D gyroscope. Basically this structure consists
of four coupled masses that are forced to resonant
oscillations by means of electrostatic excitation.
Whenever this oscillating structure is subjected
to a rotation, the so called Coriolis force is acting
perpendicularly to the rotation axis and to the
direction of oscillation and thus forcing the structure
to a secondary movement. The amplitude of this
secondary oscillation is a measure of the rotational
speed.
For a complete 6DoF-IMU a 3D accelerometer must
be added to the 3D gyroscope. SensorDynamics
has a unique dual cavity technology available that
allows integrating the gyroscope together with the
accelerometer on a single die. This technology
provides adjacent, hermetically sealed cavities
with different internal pressure, optimized for the
different needs of two sensor types.
The inertial sensors discussed in this paper are
all packaged in the proven QFN40 package. This
package is extremely stable over temperature and
can be employed up to +150C. SensorDynamics
only uses QFN packages with L-shaped leads
that provide well visible soldering joints at the
package rim. This makes the QFN also suitable for
automotive applications.
Product properties
Inertial Sensors can be optimized for different
performance parameters such as shock resistivity,
vibration robustness, temperature drift, and bias
stability. The QFN package supports all of these
parameters, especially temperature drift and bias
stability.
Next Generation of MEMS
gyroscopes and inertial combo
sensors from SensorDynamics
Figure 1: Active Structure of a 3D gyroscope
Dr. Michael
Kandler
Design-In-
Manager of
inertial sensor
products
and combo sensors at
SensorDynamics
After his studies of electrical
engineering at the RWTH Aachen Dr.
Michael Kandler received his Ph.D.
from the Fraunhofer Institute IMS in
Duisburg where he worked on surface
micromechanical integrated pressure
sensors.
In 1995 he joined the Hoechst AG in
Frankfurt as a production engineer
being responsible for measurement,
control and sensor equipment of
production plants. Beginning of 1999
he changed to Inhneon Technologies
AG in Munich as project leader for
development of pressure sensors for
side-crash detection. Since 2007 he
is the Design-In-Manager of inertial
sensor products and combo sensors at
SensorDynamics GmbH Itzehoe where
he is in charge of worldwide customer
FAE support.
As an example hgure 3 shows the result of
a Root Allan Variance (RAV) measurement
of a single axis gyroscope in QFN package,
SD706. The RAV is used to determine the
intrinsic bias stability of a sensor, i.e. the
stability in the absence of external infuences
like temperature changes and vibration. For
this purpose the sensor signal is sampled over
several days at constant ambient conditions.
Subsequently the measurement series is
divided into certain time intervals and average
values are calculated for these intervals.
The standard deviation (root variance) of
the average values is plotted for different
averaging intervals.
In hgure 3 the interval time is referred to as
cluster time. The minimum of the RAV plot is
considered as the intrinsic bias stability.
Table 1 provides an overview of the family
of SensorDynamics gyroscopes and combo
sensors in QFN40 package. The specihed
errors are maximum errors over the whole
temperature range from -40C to +85C.
The measurement range of the accelerometer
of the SD746 is 8g. All sensors feature
a continuous self diagnosis function that
immediately alarms the user if a MEMS
element has a malfunction.
Applications
SensorDynamics is targeting three main areas
of application for the new inertial sensors.
First area is navigational systems where
the position of a moving object has to be
determined. In this application the sensor
signals must be integrated over time in order
to convert rotational speed into angle and
acceleration into distance traveled. However,
due to the unavoidable bias errors the
position error will increase with integration
time. Hence, sensors such as the SD706 with
outstanding intrinsic bias stability are well-
suited for such applications.
A second area of application is platform
stabilization where unwanted movements
have to be compensated, such as the rocking
movement of a crane or image blur in cameras.
MEMS inertial sensors are well-suited for
such an application, since there is no need
to integrate the sensor output signal. 2D or
3D gyroscopes such as SD742 and SD740 ht
well into these applications, especially if the
application is safety relevant. This is because
the safety level can be improved by making
use of the continuous self test of these devices.
The third main area is motion capturing for
instance gesture recognition for handheld
devices, but also movement analysis in medical
research or for computer animated movies.
The SD746 is an ideal ht for these applications
because it is temperature compensated over
the whole operating range making external
bias compensation dispensable.
Outlook
SensorDynamics is addressing further
optimization of the performance parameters
of its inertial single- and combo-sensors, e.g.
improvement of power consumption and noise
hgures. These parameters are strongly related
to ASIC design and technology.
Another held of continuous improvement is
shock resistivity and vibration hardness. Here
the thickness of the MEMS active structure is a
key parameter because thicker structures can
be made stiffer thus allowing higher resonant
frequencies.
In order to enhance the sensor system
approach, SensorDynamics will add more
functionality to its modules like interfaces for
other sensors (e.g. AMR sensors) and have a
focus on application software for fast and easy
sensor employment.
SensorDynamics will also put efforts into
making a 6DoF-IMU available for the
automotive market. For this purpose the
temperature operating range will be enlarged
from +85C to +125C and the device will be
qualihed according to AEC-Q100.
www.sensordynamics.cc
I S S U E N 5 J A N U A R Y 2 0 1 1
Figure 2: SD746, 6DoF-IMU in QFN40 package
10
4
10
3
10
2
10
1
10
0
10
0
10
-2
10
2
10
4
Cluster Time [s]
R
o
o
t
A
l
l
a
n
V
a
r
i
a
n
c
e
[
/
h
]
Figure 3: Root Allan Variance of a single axis gyroscope in QFN package (SD706)
The bias stability is 1 07 /h
Table 1: SensorDynamics gyroscopes in QFN package For axis dehnition refer to hgure 2 *)
Measurement range up to 4096/s on customer demand
SD705 SD706 SD742 SD740 SD746
Gyroscopes
Measurement range 300/s 1024/s* 2048/s
Sensitive axes X Z X, Y X, Y, Z X, Y, Z
Max. zero rate error 3/s 7,5/s 10/s
Max. sensitivity error 2,5% 5% 7%
Accelerometer - - - - 3D
Interface SPI SPI, I
2
C, Analog SPI, I
2
C
Self test fail-safe continuous self diagnosis
Supply Voltage 3.1V to 3.5V 2.6V to 3.3V 2.55V to 3.6V
J A N U A R Y 2 0 1 1 I S S U E N 5
F
urthermore, there is a higher than ever
market demand for lower prices and ever-
shorter design cycles for these motion enabled
products. The market is pulling hard.
MEMS and silicon manufacturers play a signihcant
role by providing multiple axis sensors, powerful
microcontrollers, and now sensor clusters that offer
the possibility of creating systems of unprecedented
complexity and power. But these technological
solutions introduce new challenges in sensor
selection, management and signal processing.
Product Developers and System Integrators have to
manage the impact of the market requirements but
struggle with the complexities of this new and fast
evolving MEMS machinery.
In this paper, we want to describe the three main
components of this fast evolving MEMS toolbox and
provide a short review of their capabilities, and
limitations. As we will conclude, the trend for high
level motion features is therefore to combine these
three sensors so that these MEMS can be put at the
service of markets needs.
Among the different types MEMS motion sensors
on the market, there are three which are most
useful for building mobile motion applications:
gyroscopes, accelerometers and magnetometers.
These sensors can be used in conjunction with
other position related devices that rely on hxed
instrumentation. For instance, optic, acoustic,
mechanical or electromagnetic systems takeadvantage
of using both mobile and hxed instrumentation, thus
giving access to absolute parameters (position or
orientation). However, for greatest fexibility and ease-
of-use, it is often preferable to use only mobile sensors
in a product development effort. In fact, sometimes
there is no choice because appropriate fixed
instrumentation is not available or such equipment
does not support the product requirements. In these
situations, we often hnd that only mobile sensors hxed
to the body can be used. This last case is the one
considered in this paper.
So what can be accomplished with embedded
gyroscopes, accelerometers and magnetometers?
As individual sensors, they can be quite useful
on their own if the objective is a simple motion
feature. However, individual MEMS sensor types
by themselves are often not sufhcient when ones
goal is a more complex motion feature, like the
computation of attitude or trajectory estimation.
In the text which follows, well present a short
review of these three sensors, their capabilities and
strengths, and how they can be combined.
The Objectives of MEMS-based Motion
Sensing Systems
One of the primary objectives of MEMS-based
motion sensing systems is to dehne position and
orientation of some object or person in a real
Use of MEMS motion sensors
for embedded mobile
applications
Figure 1: 6 Degrees of Freedom (DOF) include 3 translation parameters and 3 orientation parameters
True Acceleration (A) equals the second derivative of position over time
There is a strong trend in consumer electronics which shows a growing interest
for "motion enabled products. This trend for "more motion is characterized by
increasing quantity and complexity that not only relies on scaling sensors but also
on satisfying the increasing demand for more complex features.
world frame of reference. Achieving this
objective enables a wide range of interesting
applications in sports, gaming, health and
other areas of interest.
Determining position and orientation in a real
world frame of reference requires accurate
measurement and tracking in 6 Degrees of
Freedom (DOF) in a frame of reference or,
simply, frame. Those 6 DOF include three
translation and three rotational (see Figure 1).
Lets name the three angles Roll, Pitch and Yaw.
In most applications, the frame in which
we wish to measure and track these six
parameters is hxed to the Earth. Is this
possible?
What can we sense?
If we were in an inertial frame (i.e. a frame in
which no force is applied and we are stationary
or moving at a constant speed), we could only
sense both our angular acceleration and linear
acceleration. For the purposes of our paper,
a frame hxed to the earth is considered as an
inertial frame.
Moreover, on the earth, we are given
two natural and very useful vector helds
that surround us all the time: the earths
gravitational and magnetic helds (see hgure
2). If we can sense these from the mobile
device, we can use them to determine
information about our six DOF. This will give
us some very valuable information.
The gyroscope is the hrst sensor that well
consider which contributes to our complete
six DOF knowledge. It measures the rotation
speed of the mobile device in the earths
frame of reference but expressed in the
mobile devices frame. Rotation speed,
however, is not yet the orientation that were
looking for. It is hrst derivative of orientation.
This is a good start, but to truly solve the
orientation problem over time, we therefore
need to know our initial orientation at time
zero (To), and we need be aware of the drift
in our calculations over time due to inherent
stability characteristics of our gyroscope bias
(or offset).
If we know the initial orientation, then a hrst
order integration computation will give us the
orientation of the device through time. This
is what we can achieve from very accurate
gyroscopes with no bias.
All real word gyroscopes do have bias,
however. If our gyroscope has a 1/sec
bias, for example, then after 60 seconds our
orientation calculations will have an error of
as much as 60.
Next, we consider the accelerometer. At hrst
glance, the accelerometer will give access to
the direction of the local gravitational held and
can thus provide information about the tilt
(Pitch and Roll) of the mobile device relative
to the that held. This would complement the
information from our gyroscope and we would
now be able to compute two of the three
angles we need.
It turns out that this is only true if the
accelerometer is attached to a steady state
device. If the device is not in a steady state,
then the accelerometer is also measuring the
True Acceleration of the device (acceleration
from its trajectory in space) together with
gravitational acceleration (See Figure 4).
With no additional information, there is no way
to separate these two contributing sources of
acceleration from our sensor reading. So, we
either need to know the True Acceleration from
which we can compute the Roll and Pitch, or
we need to know the Roll and Pitch from which
we can compute the True Acceleration.
The third and hnal sensor well consider is
the magnetometer. It has also its pros and
cons, but provides very valuable information.
As seen in Figure 2 and Figure 5, the
magnetometer is able to give orientation
relative to a vector held which is constant in
time. Unfortunately, over large areas, the
magnetic vector held cannot be considered
I S S U E N 5 J A N U A R Y 2 0 1 1
Figure 2: On Earth, we are moving in 2 natural
vector helds that can be considered steady over
time in the earths reference frame
Figure 3: Gyroscopes give access to
the instantaneous rotation speed Through
a hrst order integration, we can compute
the orientation
Figure 4: An accelerometer combines the true acceleration (acceleration from motion)
with the earths gravity held
Figure 5: A magnetometer is reading the magnetic vector held at its position,
therefore providing absolute information about the Yaw angle
G
Gravity
H
Earth
Mag
held
G
Gravity
held
A 3-axis accelerometer measures both acceleration from

Motion and Gravity
2D rotation
var (tilt)
Output (Accelerometer) = R(pitch, roll)
/World Frame
x (A-G)
/World Frame
3D acc var
J A N U A R Y 2 0 1 1 I S S U E N 5
constant in space. So we`d have to know additional
information about variations in the magnetic held
over the area relevant to our application.
For many applications, however, we can make the
assumption that this vector held has a constant
direction in the motion space of our device. The
strength of the magnetometer is that the readings
are not affected by motion (i.e. theres no magnetic
held created by the motion of the device) and it is
the only sensor that can provide information related
to the absolute third angle that we need (Yaw).
Thus we see that any individual sensor cannot
provide enough information to determine the
absolute orientation of a mobile device. Depending
on the application and its context, merging data
from these 3 sensors can allow us to estimate
orientation. Additionally, if we know the angles,
there is even a way to estimate True Acceleration.
In conclusion
- Gyroscopes allow us to measure Rotation Speed.
If the initial orientation at T0 is known, and if
we can estimate the sensor bias, then we can
calculate true orientation.
- Accelerometers allow us to measure the sum of
True Acceleration and the gravitational held (i.e.
Pitch and Roll angle). However, these phenomena
are inseparable unless we know one of these two.
- Magnetometers gives allow us to measure the
hnal angle, Yaw, and is not subject to distortions
from motion. However, the magnetic vector held
can vary in space over large areas.
Fusion techniques are a solution to the drawback of
individual sensors.
Moveas MotionIC offerings provide innovative data
fusion techniques in the form of software, hrmware,
integrated chips and chipsets to put this MEMS
machinery at the service of market needs. Moveas
embedded and application specihc software derives
advanced motion information to enable "motion
responsiveness in any consumer device. System
Integrators and OEMs will have "plug & play
solutions that will allow them to focus on user
experience and reduce their Time-to-Market.
Figure 6 above summarizes how multiple sensors,
initial conditions, and constraints can lead to the
solutions we need for our applications.
Bruno Flament and Yanis Caritu,
Movea SA, France
About Movea
Movea is a leading provider of motion processing
chips, software, hrmware, and IP for the
Consumer Electronics industry. Moveas unique
motion processing capabilities enable customers
and partners to quickly add motion intelligence
to their products, meaning reduced risk, cost,
and Time-to-Market for delivering compelling
new motion-based features that create more
end-user value. Movea has a global reach
with headquarters in Grenoble, France, a US
subsidiary in Silicon Valley, California, as well
as technology partners, manufacturing partners
and distributors around the world.
For further information, please see the
webcast at www.movea.com
Figure 6 : Merging Sensors and A priori information is the best way to get access to True Motion parameters
I S S U E N 5 J A N U A R Y 2 0 1 1
COme to WIN in leveraging
the European competitiveness
in miniaturized smart systems
E U R O P E A N P R O J E C T
I
n converging and integrating technologies,
smart systems make possible to predict,
decide and interact with the environment. The
growth of miniaturized smart systems market is
an attractive opportunity to positively impact the
European economy and industry. In 2008, the
global market was evaluated at about $50B and it
is expected to grow to $200B in 2020 (source: Yole
Dveloppement). Smartphones success story is
just a beginning: more will come! If largest growth
is expected in consumer and diagnostic applications,
miniaturized smart systems hnd applications in
a broader range of key strategic sectors that
are healthcare, energy and environment, safety
and security, logistics, transportation, ICT, and
manufacturing. Smart systems have the ability to
solve key societal issues including reducing energy
consumption, developing solutions for active and
healthy ageing, improving diseases diagnostic and
treatment.
Francisco Ibanez, Deputy Head of Unit, Micro and
Nanosystems ICT programme at the European
Commission, explains "the European industry is a
world leader in miniaturized smart systems. Europe
benehts of state-of-the-art research work supported
by the European Commission in the course of the
6th and 7th R&D Framework Programmes. COWIN
is a key support action to optimize value creation
from European collaborative research results and to
generate business opportunities.
COWIN is supporting innovation stakeholders
- industrial, academic and investors - to
extract the best of the European collaborative
research in:
+ Identifying attractive and advanced
technologies tting with their needs and
expectations
+ Generating added value projects by reusing
advanced technologies already developed
+ Supporting the creation of start-up
companies
+ Supporting investment in companies, key
technologies and IP.
COWIN is already identihed as a key tool for
industrial companies willing to capture innovation
in smart systems. As states Jean Luc Mate, General
Manager at Continental Automotive: COWIN
can be a new radar for industry leaders to detect
innovation from their outside world but he also
adds as President of the EUREKA cluster EURIPIDES
"I was expecting from a long time a new seamless
on demand information process to plug FP R&D
results into our SMART SYSTEM industry based
EURIPIDES cluster. We decide to partner on this
support action to validate an easy to implement
cooperation scheme. Our clear goal with COWIN
is to explore new opportunities for generating new
EURIPIDES projects on innovation from European
collaborative results.
If Europes industry is world leading in smart
systems, most of Venture Capitals backed start-
up companies are still located in North America.
They represent about 70% of global VCs backed
companies in smart systems (source: Yole Finance).
In facilitating interaction between public and private
With the launch of the COWIN initiative, the European Research community has
engaged in a new step towards value creation in miniaturized smart systems.
Cell-cell interactions arrays
(source: Mindseeds Laboratories)
J A N U A R Y 2 0 1 1 I S S U E N 5
investment in Europe, COWIN will also contribute to
the innovation take-up. According to Mr. Geoffroy
Dubus, Partner Venture Capital at Gimv, a major
European Venture Capital provider for high-tech
companies, "Miniaturized smart systems are part of
our investment targets. COWIN will contribute to
our screening process of added-value technologies
and spin-off from world class European research
centres. We are quite enthusiastic about this new
initiative. COWIN has our full support.
Matchmakers between technology providers, users
and investors to turn R&D into commercial offers,
COWIN partners beneht from their experience
and track record in innovation commercialization.
COWINs team is composed of Yole Dveloppement
and Yole Finance (France), coordinator of the
project, VDI/VDE-IT (Germany), Zabala Innovation
Consulting (Spain), Euripides - the Eureka Cluster
in smart systems (France) and Tartu Biotechnology
Park (Estonia).
To make this initiative a success, COWIN partners
are combining their technical and market expertise
to make emerge business opportunities. They
share their unique innovation network composed
of all stakeholders in smart systems including
public and private investors, Eureka community
and network, regulatory bodies, end-users,
industrial and academic communities. COWIN is
also supported by large organizations including
EPoSS, the European Technology Platform on Smart
System Integration.
Graldine Andrieux Gustin, COWINs coordinator,
explains: "COWIN is operational and pragmatic. We
are working hand in hand with research projects
partners. We are providing individual support
to leverage the value of their work and to reach
maturity for further investment. We also give
access to COWIN innovation network to identify
best of best partners.
We concentrate our efforts to reach concrete results
in bridging the gap to market. Our objective is that
with our support new products based on European
R&D research become available on the market.
www.cowin4u.eu
About COWIN
COWIN is a support action launched under the
7th framework Program for 3 years to strengthen
the European competitiveness in miniaturized
smart systems. This initiative is dedicated
to the commercial exploitation of advanced
technologies developed in the framework of
European collaborative research projects.
COWINs mission is to facilitate take-up of the
advanced technologies worthy of investments,
in order to capture innovation, win new markets
and make a proht.
For more information, please contact
Graldine Andrieux Gustin, COWIN
coordinator (andrieux@yole.fr)
Cell-cell interactions arrays (source : Mindseeds Laboratories)
I S S U E N 5 J A N U A R Y 2 0 1 1
Metrology and inspection for MEMS
manufacturing
Micromachining process control
Bulk and surface micromachining processes have
already been used by MEMS manufacturers for a
long time. FOGALE MEMSCAN uses patented optical
device which combines low coherence infrared
interferometric technology and infra red microscopy.
MEMSCAN provides accurate membrane thickness
mapping and uniformity for all kind of materials
transparent to NIR and high aspect ratio cavities
depth measurement. The advanced vision device
(IR transmission) is designed to perform MEMS
device inspection and to detect visual defects.
The automatic focus and auto contrast combined
with the image enhancement software give great
image quality. The visual defects are detected by
comparison with a reference die.
New trends in wafer-level packaging
of MEMS require well suited
metrology and inspection solutions
Roadmaps for MEMS wafer processing and
packaging are following two main trends: cost
reduction and miniaturization.
The roadmaps include:
- Wafer level packaging of MEMS devices by wafer
to wafer bonding.
- Compact hybrid IC-MEMS integration and
corresponding packaging concepts;
- Growing synergies between MEMS manufacturing
equipment and IC/MEMS packaging equipment.
The general idea of wafer-to-wafer bonding is to
cap the wafer containing the MEMS structures with
a separate, micromachined wafer in which a small
cavity is made. This cap can also act as a structural
part for MEMS devices that require operation
in a vacuum, at reduced pressure or in an inert
environment.
After bonding it is very important to detect any
defects at the surface of the MEMS and to measure
the shape of the membrane through the silicon
cap layer. The combination of IR microscopy and
IR metrology is the perfect choice to perform
such process control. Both MEMS inspection and
membrane shape are performed at the same time
through the silicon cap layer.
A curved membrane can be detected by IR
microscopy. Fringes can be observed. Between two
fringes, the gap difference is equal to /2 (: light
wavelength).
To quantify the curvature of the membrane through
the cap, the IR metrology laser beam is scanning
the membrane surface in x and y. From the
measurement of the airgap between the bottom
face of the silicon cap and the MEMS top surface,
the bow of the membrane can be easily calculated.
Conclusion
Cost and miniaturization requirements are driving
the technology toward wafer-level processed
packaging methods that can be applied to a
wide range of MEMS devices. Already used for
micromachining process control, the MEMSCAN
allows the monitoring of bonding processes used for
advance packaging and system integration in multi
chip modules.
www.fogale-semicon.com
The MEMSCAN has been developed by FOGALE Nanotech in partnership with
MEMS manufacturers to ht perfectly with MEMS manufacturing requirements for
surface and bulk micromachining and Wafer Level packaging process control. Both
Metrology and IR inspection are performed in one shot with one tool.
IR visual inspection through Si Cap
Flat membrane Defect: Membrane with bow
3
2
1
0
0 500 -500 1000 1500 -1000 1500
-1
-2
-3
-4
Bow (m)
Membrane bow measurement after
Si cap bonding process
"The advanced
vision device
(IR transmission)
is designed to
perform MEMS
device inspection
and to detect
visual defects,"
said Gilles Fresquet
FOGALE nanotech
Bow measurement through silicon cap:
J A N U A R Y 2 0 1 1 I S S U E N 5
Taking the Risk out of MEMS
Product Development
W H A T S I N S I D E ?
A
ccording to Jean-Christophe Eloy of Yole
Developpement, MEMS startups need
about $45 million and three to four CEOs to
make it to commercialization. Developing a new
MEMS product is a difhcult and risky business.
What makes developing new MEMS devices so hard?
Our hrm, AM Fitzgerald & Associates, located in
Burlingame, California, specializes in MEMS product
development. We work with clients starting from
the concept phase and perform design, modeling
and small volume prototyping, squeezing out
process risk and producing a robust design that can
be quickly ramped to production at a foundry. After
working with more than 70 clients on new MEMS
products as diverse as long wavelength infrared
imagers, tissue scaffolds, microphones, biological
and chemical sensors, pressure sensors and optical
waveguide switches, we have gained insight into
this industry-wide problem.
The MEMS industry, despite its semiconductor
heritage, is still quite young, and as such, is
missing many of the sophisticated capabilities
enjoyed by the semiconductor industry. Thanks to
powerful and verihed electronic design automation
(EDA) tools, foundry-specihc design rules and
kits, and standardized fabrication processes, an
ASIC designer can sit at his computer, design and
completely verify a new chip layout, tape out to
a foundry, and receive high-yielding, functioning
silicon chips in less than 12 weeks.
This fast and high hdelity design process is not yet
possible in MEMS. One of the missing links is EDA;
although several MEMS-specihc EDA tools exist on
the market, they do not yet provide the end-to-end
simulation capability available to the IC industry.
It is hardly the fault of MEMS EDA software suppliers
- the problem is the nature of MEMS themselves.
MEMS are mechanical devices, whose vertical
(Z-axis) architecture dictates process fow, and
for which process details such as sidewall angle,
roughness or etch undercut can have a profound
impact on device performance. Furthermore, MEMS
processes are not standardized and the material
properties of thin hlms vary widely by foundry, tool
and recipe. MEMS process and layout design rules
are difhcult to quantify because so many key MEMS
processes (deep reactive ion etch, in particular) are
extremely sensitive to multiple variables such as
pattern load factor, linewidth and location on the
wafer.
Current MEMS EDA tools can not close the loop
between the idealized model and what the actual
MEMS device will look like, because the process
information is missing. MEMS development
therefore requires repetitive open loops of design-
fab-test to converge a successful design. Because
wafer fabrication cycles take a long time (2-3
months) and are expensive ($100-200K per batch
of 150 mm wafers), development costs add up very
quickly and timelines drag out. Many companies
have run out of money just trying to get their MEMS
to work.
Our Methodology for MEMS
Product Development
Despite these industry shortcomings, minimizing
risk, cost, and time in MEMS product development
can be accomplished by expert engineering and by
hrst performing prototyping in a lower-cost, non-
production environment.
The important feedback loop between design and
process must be closed to produce a successful
MEMS device, and today, that can only be done by
an experienced MEMS team that has deep design
AND process knowledge. In early stage product
development, a MEMS engineer who can mentally
integrate the design requirements, the fab process
limitations and the coupled-physics behavior (e.g.
structural-thermal-electrostatic interactions) of the
device can make up for MEMS EDA dehciencies.
Over the past 20 years, the MEMS industry has seen some very successful products
and companies, but the road is also littered with many failures: failed products,
bankrupt companies, and disgruntled investors.
ANSYS model of an infrared imaging array pixel -
concept design
The MEMS
industry, despite
its semiconductor
heritage, is still
quite young,
and as such, is
missing many of
the sophisticated
capabilities
enjoyed by the
semiconductor
industry,
explains
Alissa M. Fitzgerald
I S S U E N 5 J A N U A R Y 2 0 1 1
Our approach to product development uses expert
engineers to establish a baseline robust design and
to fabricate functional prototypes. Then for late
stage development, we transfer the technology to a
foundry, whose staff and facility are best suited to
the tasks of perfecting the process and improving
yield.
One of AMFitzgeralds strategic advantages in
product development is its ability to quickly
fabricate small quantities of prototypes. Our expert
engineers do the fabrication work themselves, so
they can apply their experience and judgment to
make design and process adjustments during
the course of fabrication. We leverage university
research facilities to enable cost-effective small
batch production and process versatility. While
quality control and tool performance can not
match that of a production MEMS foundry, at the
prototyping phase, it does not need to.
When a prototype design is stable and ready for
foundry transfer, AMFitzgerald helps its customers
to select the right foundry partner and transfers
the new MEMS technology knowledge base to
the selected foundrys engineers. All mask data
hles, runsheets, lessons learned and in-process
metrology requirements are transferred and if
needed, engineering support is provided to the
foundry during pilot batch fabrication.
The transfer to and interaction with the foundrys
engineers reduces risk and provides excellent value
to all parties: the customer enjoys lower costs and
faster transition for bringing a developed design and
process; the foundry benehts from the wealth of
process information provided by AMFitzgerald, which
minimizes their cost and technical risks and enables
faster and more efhcient production ramp-up.
www.amtzgerald.com
Prototype infrared imager pixel array fabricated on top
of CMOS photodiode wafer by AMFitzgerald
Configure your wet fabrication process to increase yields and lower cost with SSECs 3300 Series of Single Wafer Wet Processors.
SSEC provides complete process development services to enable system configuration according to your process and manufacturing requirements.
ssecusa.com
SINGLE WAFER WET PROCESSORS & CLEANERS
F
A
B
R
I
C
A
T
I
O
N

E
Q
U
I
P
M
E
N
T

F
O
R

T
H
E

I
N
T
E
G
R
A
T
E
D

C
I
R
C
U
I
T

I
N
D
U
S
T
R
Y
CLEAN
99% Particle Removal Efficiency at the 88mm, 65mm, and 45mm Nodes
STRI P & LI FT- OFF
Immersion and Single Wafer Processing
High Velocity Spray Rotary PVA Brush Heated Solvent Immersion Heated High Pressure Scrub
WET ETCH
Uniform, Selective Etching on Multiple Process Levels
COAT / DEVELOP
Photolithography Clusters
Stream Etch for Films & Metals Wafer Thinning Low Impact Developing Spin Coating
2010 Solid State Corporation
Solid State Equipment Corporation
S I N G L E WA F E R WE T P R O C E S S I N G A N D C L E A N I N G
F I N A N C I A L B U Z Z
J A N U A R Y 2 0 1 1 I S S U E N 5
I
t had annual sales of $200.0 million with an
EBITDA of $34.3 million and Net Income of
$12.9 million giving it a Price-to-Sales ratio
on the deal of 1.7. In Q3:10 DALSA reported sales
of $55.3 million with digital imaging representing
$38.4 million with a net income of $4.2 million
and its semiconductor business representing $16.9
million with a net income of $0.6 million including
MEMS related revenue of $8.0 million. Teledyne,
established in 1960, based in Thousand Oaks
California, has 8,100 employees. It had TTM sales
of $1.7 billion with an EBITDA of $225.3 million
and a net income of $116.1 million. In the past ten
years Teledyne has made 34 acquisitions with a
combined value of $950.0 million.
There are many options for growing a company.
Integrated companies are generally looking at two
options to develop themselves in new areas. They
can start building new competencies from scratch,
but it generally takes a lot of time and money. Or
they can simply buy it.
In line with the company history, Teledyne has
chosen the second option when acquiring DALSA.
Technology acquisition of uncooled infrared
technologies seems to be the major motivation
for this transaction. Indeed Teledyne has a strong
position in infrared imaging for high-performance
markets. DALSA has complementary competences
with visible CCD and CMOS cameras and extensive
developments in innovative infrared technologies
since several years (in particular uncooled
microbolometers based on MEMS).
However Yole Dveloppement thinks that this is
only a short-term motivation for the deal. Indeed
Teledyne wants to specialize in high-performance
electronics products. If some MEMS developments
can immediately help with the infrared business,
Teledyne will probably see MEMS devices proliferate
in many business units later on. There are many
opportunities of collaboration between both
organizations, their hardest task will be to set the
priorities!
DALSA acquisition lets us look
at the MEMS world differently
DALSA was acquired by Teledyne in an all cash deal for $18.25 a share for a total
of $341.0 million. DALSA, established in 1980, based in Waterloo Ontario Canada
has 1,000 employees, including 400 at its semiconductor foundry in Bromont,
near Montreal.
Laurent Robin,
Market Analyst, MEMS,
Yole Dveloppement
Extract of Teledyne Investor Presentation, October 2010 We can see that in this example Teledyne Defence
Electronics Products Yole thinks that future generations of such products can potentially beneht from MEMS
developments (microbolometers but also RF MEMS phase shifters, inertial sensors for navigation or guidance.)
Teledyne already had a hrst level of involvement
in semiconductors and MEMS (in particular RF-
MEMS switches). Until now most of the work
of Teledyne in semiconductor, MEMS (and
some imaging development as well) was in the
Teledyne Scientihc & Imaging, LLC division.
DALSA capabilities will clearly enhance Teledyne
presence in those areas. With this transaction,
DALSA could be used by Teledyne as an internal
semiconductor lab and fab, which could beneht
to numerous Teledyne divisions.
Although we note that the continuously
growing MEMS foundry activity represents
less than 20% of DALSA revenue, MEMS
becomes a strategic technology in many high-
performance industries. DALSA MEMS foundry
revenue makes them the largest independent
MEMS foundry worldwide (Yole MEMS Company
ranking published in 2010) and DALSA has
great customers like InvenSense. This activity
is quite far fromTeledyne usual business model
which is more focused on products rather than
services. It will thus be very interesting to
see the evolution of the MEMS foundry part
of DALSA, which now has the possibility to
work with its parent company in addition to
the external customers. Probably both will be
needed to move effectively to the next stage of
the company development. DALSA will clearly
continue to work with external customers and
do business as usual. Teledyne projects will
be additional developments and will not be in
competition with the external customers, the
device manufacturers. The high valuation also
means that DALSA will have more hnancial
support to expand its capacities, which is
good. Finally we cou ld also expect major
changes in other MEMS foundry in the coming
months: new market and strategy evolutions
can be expected.
Laurent Robin, Yole Dveloppement
I S S U E N 5 J A N U A R Y 2 0 1 1
Company
Type
of product
Type of
transaction
Price of the
deal (USD)
Acquirer /
Investors
Yole Comment
October
Jyve
(USA)
Inertial MEMS Acquisition Not announced
Unknown
mixed-signal
semiconductor
company
There is a strong motivation for analog mixed
signal companies to get access to innovation
motion sensing technologies
November
Advanced
MicroSensors,
Incorporated
(AMS)
(USA)
MEMS and thin-hlm
magnetic devices
Acquisition Not announced
Plures
Technologies,
Inc. (PTI)
Plures Technologies get access both to a MEMS
facility and to a thin-hlm magnetic technology
(MI and GMR) which is currently very popular
for compass application
Bluechiip Ltd
(Australia)
MEMS RFID chip
+ reader
IPO
From $3M to
$6M
-
Close of the offer period has been extended
to Monday 28 February 2011
Honeywell
AoB Business
unit
(USA)
Cam and crankshaft,
transmission, vane,
vehicle pressure,
position and wheel
speed sensors
Acquisition $140M
Sensata
Technologies
Holding N.
MEMS activity of Honeywell is not concerned:
no impact for inertial sensors, pressure sensors
or magnetic compass
Lilliputian
Systems
(USA)
MEMS -based solid
oxide fuel cell
Equity stake Not announced Intel Capital
Under the agreement, Intel will also provide
silicon wafer manufacturing services
to Lilliputian
December
Sound
Solutions
Business Line
of NXP
(Austria)
Acoustic components:
dynamic speakers and
receivers
Acquisition $855M
Dover
Corporation
(NYSE:DOV)
--> parent
company
of Knowles
Electronics
This will provide to Knowles a leading position
in the mobile phone speaker market, which
is complementary to its microphone devices.
NXP was also working on MEMS microphone
developments and high expertise in packaging.
We expect that part of the corresponding
development/ know-how / engineers / IP will
beneht to Knowles
DALSA
Corporation
(TSX: DSA)
(Canada)
- Digital imaging
products
- MEMS component
manufacturing
- HV ASICs
Acquisition $341M
Teledyne
Technologies
Incorporated
(NYSE: TDY)
See Financial Buzz article
MEMS Finance News - Q4, 2010
J A N U A R Y 2 0 1 1 I S S U E N 5
Micronarc 2011 review
E V E N T R E V I E W
STM targets 1B MEMS units in 2012!
The first day opened with a talk from
STMicroelectronics on MEMS for consumer. STM
showed increasing market demand from consumer
applications for accelerometers, gyroscopes
& magnetometers. STM said main products
integrating inertial MEMS and magnetometers are
smart phones today, while cameras with tablets
being the next targets. STM is also currently
working at increasing performance by integrating
the sensors together in what they call their iNemo
platform. On an integration approach, STM said
they chose to have a stacking of the ASICS above
the MEMS die to have a higher fexibility. By end
2010, STM had produced their billionth MEMS dies.
STM mentioned they could be looking for additional
capacity in the future with fab extension. Their
production objective is 1B units for 2012!
The Next presentation was from Komax Medtech.
They discussed about the challenges of having a
mass production for complex automation process
of devices: mass production is all about automation
they said.
During the development cycle, production
dramatically increases when we move from pilot
production to high speed manufacturing. So the
challenge is to master each of the four stages
from prototyping to mass manufacturing. Komax
masters all the different steps during an automation
process are: dehne product, decide manufacturing
strategy, identify processes analyze risks, cost and
timeline.
During next presentation, Del West (a racing valve
train systems specialist) highlighted lessons we
can learn from F1. They showed the large R&D
effort necessary for F1 development and presented
impressive hgures: a F1 engine has 5,000 parts
of which 1,500 are moving part. In a car race,
winning pressure is huge and innovation is thus
always present. Resources of development at Del
West (100 people with 50 of them being engineers)
can be mobilized in a very short time: from 2 to 24
months max. For manufacturing, Del West outsources
very fewcomponents. On one hand the talk concluded
by saying that F1 is very effective at hunting &
adopting technologies from other industries. And on
the other hand F1 technologies can also be accessible
and transfered to other industries. Del West also
concluded they could be watching other industries as
well (watch industry and medical).
The Next talk was from SUSS MicroTec. It was about
advanced photolithography techniques for mask
aligner lithography systems. To avoid illumination
errors, SUSS MicroTec has developed customized
illumination. One particular feature is that by using
illumination Filter Plates (IFP) there are no optical
element between the illumination source & the
wafer.
Tronic's aims at being the leading
manufacturer of custom MEMS.
The next session was about "Micro-manufacturing
of Si and non-Si MEMS and gathered the talks from
CSEM, EPFL, IMTEK and Tronics.
CSEM (now 400 employees, 66MCHF revenue
in 2008) talked about Microsystems and their
4/6 MEMS foundry service used for research,
prototyping and small scale production. CSEM is
mastering multi-level Si DRIE, SiC processing,
AlN thin hlms as well as electro deposition, metal
implantation in polymers and SAM deposition with
After the success of the hrst edition, the 2nd Micronarc Alpine Meeting 2011 (16-
18 of January) was held at Villars in Switzerland, a charming village and ski resort
located at 1250 meters high. This meeting focused on equipment for manufacturing
microproducts including MEMS of course, and also non Si micro products. This 2-day
conference stimulated networking and discussions in a very casual atmosphere.
I S S U E N 5 J A N U A R Y 2 0 1 1
MVD. CSEM showed nice examples, in particular
for watches. It showed temperature-stable Zerodur
glass spiral hairspring by Cartier, escapement
Pulsomar by Patek Philippe and new Si escapement
by GP. On the more "traditional MEMS side, CSEM
showed piezoelectric resonators & hlters. CSEM has
developed a family of low-frequency Si resonators
by mixing Si DRIE and AlN thin hlms together.
Last example is an "exotic realization for a Braille
display. CSEM called it M-MEMS (micro & macro
system) in which a soft polymer membrane covers
a Si die. CSEM objective is not to compete with STM
as target volumes are rather in the tens of wafers.
EPFL talked about ceramic modules for harsh
environments: aerospace, deep-well drilling,
automotive or chemicals. Deposition is thick-hlm
technology. EPFL described three examples: liquid
level sensor, jet engine AMB (Active Magnetic
Bearing) position sensor and microreactors. Such
technology makes sense when cost is not an issue
and where aggressive environment is present.
The IMTEK / University of Freiburg talked about
Lab on Chip based on fexible foils. Technology is
based on centrifugal effect. So far, positive analyses
for patient have been realized to select the best
antibiotics. The main strategy is to become a
microfuidic platform and also to offer a design
and foundry service.
Finally, Tronics presented MEMS technologies and
product platforms. Tronics (now 60 people, with
two fabs) aims at being the leading manufacturer of
custom MEMS. They just announced the launch of
a new JV in China. Tronics dehned MEMS, not as a
single product or market, but as a tool-kit. In most
cases, new MEMS devices necessitate new processes.
To overcome that, Tronics core process foundation
is based on SOI MEMS to be used for a wide range
of components. For inertial MEMS, Tronics has
developed AGELLAN core technology platform.
On the hrst days afternoon, two sessions were
dedicated to non Si devices manufacturing. The
First talk was given by KIT who presented new
developments for ceramic injection moulding in
small dimensions. The Second talk realized by
the University of Bradford gave a very complete
overview of the facilities and competencies of
the Centres work on Polymer Micro & Nano
Technology. The Centre is currently working on
improved mechanical & optical properties in order
to get better form factors for polymer devices. The
Centre is experimentally determining relationship
between materials, processes and hnal product
properties. Session ends with ARBURG GmbH +
Co KG talking on micro-injection moulding for
complex thermoplastic, MIM and CIM parts. In
the session dedicated on metal micromachining,
Georgia Institute of Technology gave a keynote talk
on Fundamental Mechanics of Micro Machining. The
reason why we need alternatives to Si DRIE and/
or LIGA processes is mainly cost and materials.
Georgia tech insisted on the fact that a new 300
mm fab is about US$4B today! Moreover bio
medical, optics, electronics, robotics and defense
are drivers for micro manufacturing. Georgia Tech
proved dramatic productivity (by a factor of 2)
enhancement for cutting tools.
Georg Bischopink, Bosch, Congress keynote
J A N U A R Y 2 0 1 1 I S S U E N 5
A 0.25 cm resolution pressure sensor
from Bosch!
The second day was dedicated to assembly & test.
The First presentation was from NPL and was about
structured surfaces for traceability. Then Nottingham
University discussed specihc challenges of micro
assemby because of the incoming of new products
on the market (photonics, energy sources, medical
devices). The Industry needs smart reconhgurable
micro-assembly systems. Nottingham University
has developed a micro assembly concept based
on self-adaptive, self-learning & self-awareness
abilities for a system to be fully adaptive. 3D
minifuidics devices have been assembled as
examples. The Next manufacturing session opened
with a talk by Santa Cruz University. This University
aims at mixing different areas together into a single
platform: biology, photonics & fuidics thematic.
They concluded that hybrid devices are a challenge
for the future of manufacturing.
The conference ended with a presentation given
by BOSCH on MST from automotive to consumer
& future applications. Bosch presented impressive
hgures related to MEMS: 78% of all new cars have
pressure sensors for motor control, 28% of all
new cell phones have accelerometers, 35% of all
new DSC have gyros and 90% of new cars have
accelerometers for airbag. Bosch has produced far
more than 1.5B MEMS (more than 400 mio in 2010
only). Bosch believes there will be no standard
MEMS process in the future (the "one MEMS, one
process rule will still be present for a long time!).
Bosch presented an impressive pressure senor with
0.25 cm resolution. Then they have presented the
SMI540, hrst ESP combi-inertial sensor (yaw rate
& accelerometer) in mold package with gold wire
bonding. They also presented a new technology
for consumer pressure sensor membrane: APSM
based in porous Si-surface micromachining (a
completely CMOS compatible technology). Also
using CMOS MEMS technology, Bosch is today
producing microphones (Akustica technology). It
was interesting to note the involvement of Bosch in
the magnetic held sensor as well.
Regarding future applications, Bosch mentioned SiC
pressure sensors for exhaust gas for automotive in
a very demanding environment (high temperature
& corrosive atmosphere). Bosch is completely
separating the electronics circuit from exhaust gas.
A gold coil is used for inductive coupling between
the sensor and the ASIC. In the FIR held, Bosch
demonstrated a resolution > 100x50 pixels. As
STM, Bosch is also working on fusion of different
MEMS. Bosch presented a global vision for sensors
for the Internet of Things (IOT): key elements
such as wireless, low cost & energy autonomous
networked sensors. Examples of applications are
home automation, medical applications, logistics,
smart grid. Bosch also mentioned they are looking
at pico projection using single micro mirror with
piezo detection and novel wafer level packaging.
As a last example Bosch presented a Lab On Chip
development.
Dr Eric Mounier, Yole Dveloppement
MAM 2011 in review
The second edition of the Micronarc Alpine
Meeting (MAM2011) was held in Villars-sur-
Ollon (CH) from January 16-18, assembling 65
participants, including 10 exhibitors.
The 2-day congress represented a healthy mix
of research and private sector interests and
featured a high-powered list of speakers from
Germany, Switzerland, Italy, France, the UK and
the United States.
A dedicated session on MEMS featured talks by
representatives from CSEM (Marc-Alexandre
Dubois), the EPFL-LPM (Thomas Maeder), IMTEK
(Roland Zengerle) and Tronics Microsystems
(Stphane Renard).
As well, two main keynotes were given by
major MEMS players. Edoardo Bortolato from
ST Microelectronics presented a conference-
opening talk entitled, "Sensors: Quo Vadis?
Closing the conference was the Invited Keynote
Georg Bischopink of Robert Bosch, who gave an
excellent overview of Boschs past, present and
future activities in a talk entitled "MST at Bosch
- from automotive to consumer electronics and
future applications.
Edward Byrne, Project Manager, Micronarc
Georg Bischopink, Bosch, Congress keynote
I S S U E N 5 J A N U A R Y 2 0 1 1
ISS Europe 2011 - Connect to
the European Semiconductor
Community through SEMI
E V E N T R E V I E W
T
he past 12 months has seen a resurgence
in the global economy and especially in the
semiconductor industry that is forecasted
to continue into 2011. The European micro/nano-
electronics industry has taken advantage of these
conditions with companies across the EU and
beyond reporting strong business activity.
Europe has strong capabilities in many key areas
of the semiconductor and related industries.
Under the title "How to capitalize on European
strengths along the semiconductor value chain the
conference aims to stimulate discussion that will
allow the exploitation of these strengths to their
fullest extent. To this end, Europe needs to:
- Leverage its strong position in advanced research
- Commercialise R&D achievements
- Build on its diverse manufacturing base
- Maintain advanced manufacturing capability
- Exploit niche and developing markets
There are many examples of such activity across
Europe and SEMIs ISS executive symposium will
once again gather industry leaders to discuss their
strategies and capabilities, stimulate debate on how
to ensure Europe exploits its many strengths and
takes full advantage of developing markets.
The conference will address these key challenges
and encourage discussion and debate. Day 1
focuses on the market and related business
implications with industry sector leaders and
forecasters providing a comprehensive view of
industry developments. This is complemented by
a panel in which manufacturers representing the
diverse nature of Europes capabilities discuss the
issues around successful device manufacturing in
Europe.
Day 2 looks at Europes technical strengths and its
participation in all aspects of the industry - from
450mm through More Moore to More than Moore
- showing capability across all key development
areas.
Europe has world class capabilities in all areas critical
to the developing micro/nano-electronics industry.
ISS Europe 2011 will discuss those strengths and
create an environment for all participants to share
experiences and stimulate the development of a
strong European business.
Top Industry Speakers:
- Carlo Bozotti, STMicroelectronics
- Frdric Barrand, PSA Peugeot Citron
- Diego Olego Philips, Healthcare
- Douglas Gilstrap, Ericsson
- Bill McClean, IC Insights
- Malcolm Penn, Future Horizons
- Robert E. Bruck, Intel
- Richard Koo, Nomura Research Institute
- Paolo Gargini, Intel
- Frank Melzer, Bosch Sensortec
- Elgar Fleisch, University of St. Gallen
- Robert Hagen, Inhneon
- Laurent Malier, CEA-LETI
- Elke Eckstein, Osram
- Michael Lehnert, LFoundry
- Rutger Wijburg, NXP
- Hans-Juergen Straub, X-Fab
- Jens Drews, GLOBALFOUNDRIES
- Jean-Marc Chery, STMicroelectronics
Visit www.semi.org/isseurope for more details
and take a detailed look into the conference
agenda.
Join us at ISS Europe 2011, register on-line
and be a part of this important executive
conference!
See you in Grenoble!
For more information, please contact Carlos
Lee (clee@semi.org)
The annual SEMI ISS Europe (Industry Strategy Symposium) is the recognized
industry event in Europe and a must-attend for every executive leader of the
industry. Keep yourself informed about the latest developments and strategies,
and network with your peers in a convenient atmosphere.
"ISS Europe is a
unique opportunity
for decision-makers
across the entire
microelectronics
supply chain to meet
and collaborate,"
according to
Heinz Kundert,
President, SEMI Europe
J A N U A R Y 2 0 1 1 I S S U E N 5
About Yole Dveloppement
Editorial Staff
Managing Editor: Jean-Christophe Eloy - Editor in chief: Dr Eric Mounier - Editors :
Jean-Christophe Eloy, Dr Eric Mounier, Frdric Breussin, Laurent Robin, Patrick Keating,
Jrme Mouly, Paula Doe - Manager: Sandrine Leroy - Assistant: Camille Favre -
Production: atelier JBBOX
CONTACTS
Beginning in 1998 with Yole Dveloppement, we have grown to become a group of companies providing market research, technology analysis,
strategy consulting, media in addition to hnance services. With a solid focus on emerging applications using silicon and/or micro manufacturing
Yole Dveloppement group has expanded to include more than 40 associates worldwide covering MEMS and Microfuidics, Advanced Packaging,
Compound Semiconductors, Power Electronics, LED, and Photovoltaic. The group supports companies, investors and R&D organizations worldwide
to help them understand markets and follow technology trends to develop their business.
SERVICES
- Market data, market research and marketing analysis
- Technology analysis
- Reverse engineering and reverse costing
- Strategy consulting
- Corporate Finance Advisory (M&A and fund raising)
For more information about :
- Services : Jean-Christophe Eloy (eloy@yole.fr)
- Publications: David Jourdan (jourdan@yole.fr)
- Media : Sandrine Leroy (leroy@yole.fr)
MEDIA
- Critical news, Bi-weekly: Micronews, the magazine
- In-depth analysis & Quarterly Technology Magazines: MEMS Trends - 3D Packaging - PV Manufacturing - EfhcienSi
- Online disruptive technologies website: www.i-micronews.com
- Exclusive Webcasts
- Live event with Market Briehngs
REPORTS
- Collection of market & technology reports
- Players & market databases
- Manufacturing cost simulation tools
- Component reverse engineering & costing analysis
More information on www.yole.fr
Translating concepts and requirements into real-world products
requires great ideas and reliable execution. Tronics guides you to
the rewards and around the risks by offering global solutions with
local support and fullfilling your expectations.
FROM QUALIFIED WAFERS TO INTEGRATED COMPONENTS
THATS WHAT WE PROVIDE!
www.tronicsgroup.com

MEMS Trends No.9 2012.01

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

MEMS Trends No.9 2012.01

Hochgeladen von

Copyright:

Verfügbare Formate

MEMS'

Korea brings together

A 3-axis accelerometer measures both acceleration from

Das könnte Ihnen auch gefallen