MEMS Community in Canada
The MEMS community in Canada held it's first workshop this February. These are
the records of that workshop. The second workshop will be held this August
13-15 at the University of Alberta in Edmonton, Alberta. The mailing list in
Canada for MEMS activity is mems@cmc.ca.
Records of the First Canadian Workshop on Micromachining
Compiled by Christopher Raum - craum@robinhood.engg.uregina.ca
The first Canadian Workshop on Micromachining took place on the 20th and 21st
of February in Bromont, Quebec where the IC fabrication plant, Mitel, is
located. The Workshop opened up with tours of the facilities at University of
Montreal's Ecole Polytechnique and Concordia University.
BROMONT SESSION: DAY 1
The actual workshop began at the Auberge Bromont Hotel with a round table
discussion attended by about 35 people >from both the academic and industrial
components. The intent of the discussion, chaired by Ash Parameswaran of SFU,
was to both introduce everyone, and to establish a dialogue for the direction
that micromachining should take in Canada.
Dan Gale of CMC (Canadian Microelectronics Corporation), one of the chief
organizers of the workshop started out the program by setting the goals for the
session:
1. A chance for people to get to know more about each other and their work.
2. Things that are going on at CMC and the role that Mitel will play.
Ash's plan was to have a free flowing informal dialogue on where the future of
MEMS research and industry is. It started with a round the table introduction
of the attendees, including a one line description of what their research area
or interest is. The following is a list of the people in attendance.
Abdellah Azelmad abdellah@goal.waterloo.on.ca
Albert Leung aleung@SFU.CA
Behrouz Nikpour behrouz@vlsi.concordia.ca
Chris Raum craum@robinhood.engg.uregina.ca
John Currie currie@vlsi.polymtl.ca
Dan Gale dan@cmc.ca
Robert Antaki fm79@polytec1.polymtl.ca
Genevieve Beique genevieve_beique@mitel.com
Gegi George gg@vax2.concordia.ca
Michel Meunier gn00@polytec1.polymtl.ca
Gordon Harling gord@goal.waterloo.on.ca
Graham McKinnon graham@amc.ualberta.ca
Ahmed Haider haider@vlsi.concordia.ca
Baher Haroun haroun@vlsi.concordia.ca
Phil Haswell haswell@ee.ualberta.ca
Hsu Ho ho@cmc.ca
Ion Stiharu istih@vax2.concordia.ca
Alain Jean jean@ino.qc.ca
Ricardo Izquierdo JG01@polytec1.polymtl.ca
Jagdish Patel jpatel@vax2.concordia.ca
Les Landsberger leslie@ece.concordia.ca
Mojtaba Kahrizi mojtaba@ece.concordia.ca
Garry Tarr ngt@doe.carleton.ca
Nicholas Swart nrswart@venus.uwaterloo.ca
Pak Ko pak.ko@nrc.ca
Ash Parameswaran param@cs.sfu.ca
Makarand Paranjape paran@ece.concordia.ca
Rama Bhat rbhat@vax2.concordia.ca
Remi Meingan remi_meingan@mitel.com
Sandy Robinson sandy@ee.ualberta.ca
Jim Seary seary@cmc.ca
Bing Shen shen@eigen.ee.ualberta.ca
Sundar Cheltar sundar@synergie.polymtl.ca
Next the following issues were discussed:
Academic issues discussed
1. The preferred direction of each MEMS group. This was essentially covered in
the round table discussion.
2. Does the group feel the need for an organized collaboration or is the
individual research we are doing enough? Is there more we can do? What is
each person's vision for the future of MEMS. This too was partially covered
by the round table introduction.
3. If there is a collaborative effort, besides doing individual research. What
can Mitel offer and what else will this allow us to do. I.E. applying for a
grant together or tackling a design problem together.
4. Does the group feel that enough members and organizations are participating
right now. Are there any groups that should have participated in this
workshop but didn't this time? How can they be convinced to attend the next
workshop? The idea of each researcher inviting an industrial partner to the
next workshop was discussed. Also the idea of a collective news letter that
everyone would contribute to be distributed to all the provincial
industries was brought up.
This would also have the benefit of helping the micromachining community of
Canada stick together. The need for a more structured form of communications
between researchers, and between researchers and the industry was definitely
felt.
One group talked about their experiences in trying to drum up industrial
support. They have done a couple of sensor related projects. Sensor products
often take a large number of years to develop. Getting a commercial sponsor who
is willing to invest in your research for that length of time is difficult.
Industry sponsors are often not aware of the nature of micromachining in that
they determine progress with metrics such as: What is your product development?
They want to know what the part numbers of devices being developed are. This
isn't the case with micromachines. In the future, educating the industry will
be necessary in order to break down the barriers to getting funding.
Another group talked about their interest in medical devices. At present there
is not a large degree of utilization of micromachining technology. One
promising area to develop, however are analysis and diagnostics devices for a
low volume samples. These areas that are open to analysis are very wide,
meaning there is lots of room for everyone to develop.
To get an idea of the attendees areas of interest, the participants of the
round table were asked to come up with a one paragraph summary of a project for
applying to NSERC for a Centre of Excellence in micromachining. Some of the
ideas were:
* Biocompatability issues, with new strategies for intervention.
* Special purpose sensors for biotechnology: intelligent micro stimulators,
prosthetics.
* Micromanipulators for optical fiber alignment and biological applications.
* The integration of micromachining applications.
* Micromachining technology to reduce substrate losses.
* Laser processing technology for research into three dimensional structures.
* To come up with sensors and actuators that can be fabricated by standard
processes so that structures can be developed without sophisticated
equipment; thereby increasing the number of participants in the field.
* Developing reliable and effective ways of packaging sensors to expose the
sensor yet protect the circuitry.
In education, simple designs involving basic post-processing are an easy way to
give students an appreciation for micromachining and the microelectronics
industry at the same time.
It was also recognized that micromachining is not only the domain of
microelectronics. Faculties such as Mechanical and Chemical Engineering have a
natural interest in this field. It was also agreed that students even at the
undergraduate level should leave with an understanding of micromachining.
CMC issues discussed
How and why should CMC support this effort? This issue was discussed over the
course of the entire workshop. A collective four point conclusion was reached
by the end of the workshop.
CMC's experience with silicon and gallium arsenide based technology provides a
strong industrial connection in Canada. CMC has historically drawn a strong
benefit from this connection. In the direction that CMC is headed, greater
emphasis will be placed on the systems and applications aspects of using
silicon and gallium arsenide and less on physical design opportunities. The
question with MEMS in general is if there is no developing industry in the
country, why should CMC be involved?
There must be some anticipation of economic and/or social benefits derived from
the use of micromachining in Canada for CMC to be directly involved. For
example, even for very exploratory work is there some capacity for Canadian
industry to absorp results (ie., is there "receptor capacity")? It is extremely
important for CMC to be shown that there is more than just potential in MEMS
research in Canada.
The general consensus at the end of the session was that by reaching a
collective agreement on a blueprint for the industrial and academic development
MEMS science, CMC can be encouraged to support the community's efforts and
goals. The nature of the support will have to be determined in the future, but
at least there will be a positive message of support.
After the first session finished the group adjourned for its soon to be
traditional Pizza Supper. After supper, many people went for drinks and more
informal discussion about the future of MEMS research and support from the
industry.
At first there was limited discussion among the people, not so much because
people were afraid to share ideas, but because no one had a feel yet for what
level everyone was operating on.
After a while, John Currie made a suggestion that opened things up a bit. He
went around asking everyone present if they could come up with a research idea
worth $50,000. The response was limited indicating a lack of direction and
solid ideas. At this point, a grand project was suggested which would involve
the participation of all universities on a common micromachine project. Some
ideas were in the biomedical, automotive, and industrial sensor industry.
MITEL SESSIONS: DAY 2
Presentations that various profs and agencies gave (9)
1) Les Landsberger and Baher Haroun (Concordia)
Concordia has a variety of interests centered around MEMS processing and MEMS
post processing, VLSI system architecture and design. The Chemical Engineering
department has an interest in fabrication that extends to the Physics
department. These groups all have collaboration with each other with a lot of
joint activities. The active program involves a reasonable number of graduate
students and even undergraduate projects.
Students typically love this topic. The observation has been made that students
get a charge out of noticing the link between the high level design and the
process. By fostering an early interest in micromachining, the result could be
a great contribution to training highly qualified personnel in Canada.
The overall set of projects span a wide range including integration with
circuitry and design rules, integrated sensors (a topic of great interest to
us) including integrated accelerometers, magnetic pole effect, gas sensor, and
a variety of tactile sensor projects.
Other interests within Concordia include the characterization of mechanical
properties by test structures. John Currie of Polytechnique with Omega have
collaborated to make a simple pressure transducer using anodic bonding. In
general, Concordia is interested in any MEMS topic and are open to
collaborations. With several collaborations going they would like to continue
this trend.
Upcoming work will include CMC submissions involving MEMS design rules,
characterization of mechanical properties, integration of circuitry, and TMAH
etching. In terms of sensor research the approach is to form a systems
perspective with the idea of designing a sensor that you will eventually have
to interface to.
Applications include sensing in remote areas in very rugged environments in
terms of non-destructive testing for motors. Where Mitel's collaboration can
fit in is through their expertise in telecommunications.
2) Gary Tarr (Carleton)
Carleton's lab can do most processing steps required to make the silicon IC's
including photo mask generation right through oxidation infusion, layer
deposition, a few other process features.
Due to the community environment that is beginning to develop in the MEMS
community, it would be useful for the MEMS community if CMC or some agency
where to make an inventory of the processing capabilities that the universities
in Canada are willing to share. Carleton is willing to share most of their
equipment, the only problem being the technical manpower crunch to run the
equipment.
In terms of micromachining and sensors Carleton has been doing some work on
radiation sensors that involve micromachining for a small Ottawa company,
Thompson Electronics, as well as looking at micromachining techniques to reduce
substrate losses in FIC's on silicon.
Another ongoing project is the use of a micromachined spiral inductor for the
design of a micro-transformer or inductor. Another high current device is a
magnetic field sensor. Finally, an interesting sensor idea under development is
radiation detectors. By irradiating a MOSFET, electron pairs are generated in
the oxide and the threshold voltage is shifted. Very low levels, less that a
rad, are measurable.
3) Alan Jean (National Optics Institute)
oAlan gave a report on Micromachining and related technologies at the National
Optics Institute. First there was a list of facilities and technology at the
institute.
Facilities:
Clean room
Plasma and ion beam etching facility
Thin film and coating deposition facility
Laser-assisted processing facility
Crystal growth facility
Technology:
Vacuum deposition of thin films and coatings
Plasma and ion beam etching of materials and films
Photolithography
Laser direct writing by material deposition and
etching
Silicon micromachining
Crystal Growth.
Among NOI's principle goals are the development of expertise in designing and
fabrication of various types of "smart" multifunctional micro sensors and MEMS
devices with the focus on uncooled IR radiation sensors and deformable
micro-optical devices such as micro mirrors.
4) Pat Koh (NRC)
NRC (National Research Council) is beginning to expand into mechatronics
through the IMR (Institute for Machinery Research) in terms of actuators for
machinery components, or linkages. They are interested in creating projects for
this area of research. For example, one of the research programs of IMR is
machinery condition monitoring. In this program, systems would be developed to
detect and diagnose machine faults through vibration analysis and oil/gas flow
debris monitoring. This is seemingly an ideal application for micromachining.
5) Sandy Robinson (University of Alberta)
U of A has a small research group. A lot of work and collaboration with the CMC
occurs. Of particular interest is the characterization of polysilicon layers,
design and testing very simple devices, and also numerical simulation of these
devices in steady state conditions and time variant conditions.
Some of the devices that are being investigated at U of A are suspended glass
platforms that were originally developed by Ash Parameswaran of SFU. Because
the polysilicon is fairly well isolated, it doesn't take much power (min. 40
mW) to get the resistive elements up to high temperature and luminescence. The
stability of the polysilicon is not stable. When the temperature is cycled up
to 500 degrees C, the room temperature resistance and TCR has changed. They are
trying to determine if there is a predictable or systematic variation in this
phenomena.
6) Micheal Meunier (Ecole Polytechnique)
The Laser Processing Lab includes about 10 people. By using various types of
lasers such as Ar+, eximer, diode, and CO2, various types of processing such as
deposition (W, Si, WSi, Cu) and etching (Si, Al, Cu) can be achieved.
Through these processes, a number of three dimensional structures can be
deposited or etched.
1) Curved or round structures in silicon for curved mirrors
2) Etching under a nitride or oxide for tunnels or cavities
3) Growing columns 100's of microns high for pins or axles.
The lab is looking for collaboration, as well as a provider of services.
7) John Currie (Ecole Polytechnique)
The Groupe des couches minces GCM or Thin Film Group has three service
facilities for NSERC funded members with a minor user fee that are available
for fabrication, post- processing, and ion beam analysis.
The LISA (Laboratory for the Integration of Sensors and Actuators) lab is a
smaller group that is interested in integrating sensors and actuators. In terms
of integration, LISA has been working with Mitel for the last 4.5 years from a
process materials point of view. MEMS structures are now under investigation as
well as their integration with control and conditioning circuitry.
In the sensor arena, work is being done principally with environmental
monitoring of trace multicomponent gasses. LISA is working closely with the
resource industry, i.e. the pulp and paper groups on detecting NOx and CO2
emissions >from heavy processing equipment.
The objectives are to:
1. Develop integrated chemical sensors for detecting small concentrations of
nitrogen and sulfur oxides, organics, and dioxines.
2. Analyze the performance and basic physico-chemical working principles of
these sensors.
Problems to be solved: To attain a degree of chemical selectivity.
Solution: develop and array of integrated multiple thin film electrochemical
sensors.
8) Ash Parameswaran (Simon Fraiser University)
The MEMS lab at SFU is intended for research. It offers micro fabrication and
micromachining courses for academic and industrial organizations.
Typically the course contents are based on a prototype development. Course
duration ranges between a few weeks to a few months. At the end of the course
the participants take a set of the fabricated devices with them.
The SFU lab has also developed a low-cost emulsion mask procedure which is
currently available via email for no cost.
9) Nicoles Swart (University of Waterloo)
The sensors group at Waterloo has industry involvement in Japan. Some
activities include using micromachining for thermal isolation, including hot
plate designs with very uniform surface temperatures.
These structures were also used to determine what happens to polysilicon at
high temperatures.
Areas of research within the group include: 1) Test structures for heat
transport analysis in small structures including thermal conductivity studies
and film stresses.
2) Micromachines for mechanical isolation with pressure sensors.
3) Etching kinetics
4) Micromachining for stress reduction in magnetic sensors.
DISCUSSION AFTERWARDS
Informal discussion with MEMS community:
The informal discussion centered around possible strategie for furthering MEMS
and micromaching research in Canada. The reason this is necessary is that CMC
has given the message that they have not included micromachining in their next
five year plan. They would like to include micromachining in subsequent plans,
but they would like to see direction and planning from the academic
micromachining community as a whole.
One common idea from the first informal session (held the evening before at the
Chateau Bromont) was that we need time. Micromachining work started in a hodge
podge fashion with few researchers working on mainly isolated projects. Now
with one processing industry, Mitel, willing to look at micromachining more
carefully as a creditable science and not just as a per project basis, more
time is needed to coordinate planning efforts for the evolution of the science
in the future.
Results of the First Canadian Workshop on Micromachining
Four items were identified which were presented to CMC at the end of the
meeting.
1) Ask for more time under the case that such an involved technology requires a
high degree of training; similar to the training and evolution of processes
that was needed in the early 80's when microelectronics was first supported.
Now that Mitel is taking the processes being asked for by the MEMS community
seriously, time is needed for these processes to evolve.
If a coordinated plan that can be given to CMC they can turn it into a proposal
which can then in turn be submitted to NSERC for funding. Anywhere from 6 to 18
months would be needed to come up with such a plan.
18 months was preferred by the group for a number of reasons:
a) To facilitate proper learning cycles for using Mitel's process in an
interactive fashion. This will allow engineers and scientists to develop
design capabilities that will be more useful to industry.
b) 6 months is not enough to sell the idea to the industry. To start the
interaction between industry that is needed to get them interested the
extra time is justified.
2) To create a Canadian MEMS Challenge to help unite and coordinate efforts. It
would also help to put individual expertise toward achieving one goal.
There is a need to identify this challenge, and a few topics were initially
proposed.
a) Micromachined components for use on satellites for the Canadian Space
Agency.
b) Micromachines to aid in the human geneome project.
c) Micro sensors for the intelligent house project underway in Quebec.
d) Micro sensors for non-destructive self testing of structures (i.e. a
bridge)
In any event, it is important to identify a grand challenge in which the entire
MEMS community in Canada can contribute; >from the industry side and the
university side. Such a unifying challenge could be presented as a vehicle to
justify CMC's support of our efforts. It was recognized by Mitel that such a
product would have to be justified in that it is marketable.
3) To involve industry as much as possible in these workshops. One suggestion
is for each attendee to bring along an industry friend to take part in the
proceedings. There would be a presentation at the workshop to let the industry
know what the Canadian micromachining community is capable of. This would allow
us to show "unenlightened" industry representative how they can use MEMS
solutions for their problems. The workshops could conceivably take place after
each CMC production run at Mitel.
The main concern of this point is to educate the industry as to the
capabilities of MEMS so that they take an active interest in progress. It's
important that the industry realizes that our field is results driven, not
curiosity driven.
4) Training highly qualified personnel, and enhancing the skills of people with
expertise in processing. This operation will create more knowledgeable people
in micromachining and enhance the application of high tech to various
applications.
CMC has indicated they want to participate in this process, and that they are
very interested in the future of micromachining research in Canada. However,
they currently have no plan for including micromachine research into the
current funding layout they are submitting to NSERC.
They indicated that they would like some sort of communication from the
Canadian MEMS community as a whole as to a blueprint for the future within a
time frame of 6 to 12 months. This will give CMC an idea of the kind of
resources they should allocate to sustain research as well as their degree of
involvement with help and preparation.
CMC anticipates getting some kind of feedback from NSERC in September. It's
entirely possible that the whole nature of their core area activity could
change anyway, therefore there is uncertainty relating to that area.
Recommendations
The immediate outcome of the workshop is an understanding to continue
communication and circulation of information through the use of the cmc MEMS
mailing list. This mailing list would be used as a tool for continued
discussion on projects such as the Canadian MEMS Challenge, and to organize the
next workshop. The mailing list address is: mems@cmc.ca. It is unmoderated.
To create a national repository for project documentation and results to be
reserved at cmcCache and available by ftp. ftp keeper.ic.cmc.ca. Logon as
anonymous.
To create a Usenet group to deal with topics in micromachining on an
international level. Information from each level could naturally flow outwards
from local to national to the international level.
Further Discussion and Future Workshops
Packaging was recognized by Mitel in particular, but by workshop attendees as
well, as an important topic for future workshops. It was suggested that future
projects that are taken on include the packaging as part of the design right
from the very start. In many sensor designs, the CMOS electronics must be in
close proximity with the micromachined membrane or sensor components.It is here
that packaging becomes particularly important.
Testing issues (MEMS design for testability, standardization of testing)
As the quality of MEMS designs increases in the future, and the standards are
raised, there will be a series of gates (or design check points) that will
arise. This will allow MEMS production to move more economically and
efficiently, improving the overall evolution of the science. I.E. "Have you
thought through your packaging?"
The Second Canadian Workshop on Micromachining is set to take place at the
Univerisity of Alberta, August 13-15, in Edmonton Alberta. The contact for this
workshop is Sandy Robinson - sandy@ee.ualberta.ca
APPENDIX ONE
DESCRIPTION OF UNIVERSITY FACILITIES
SFU
Available MEMS Process
1. Standard Silicon isotropic and anisotropic etching HNA, EDP, TMAH, KOH etc
etc.
2. Anodic and Fusion Bonding.
3. Laser direct write lithography and laser-based micromachining.
Fabrication Facilities
The Micromachining Laboratory at the SFU School of Engineering Science includes
a 300 sq.ft. class 100 clean room and more than 1000 sq. ft. class 1,000 clean
area.
The following processing technologies and testing facilities are available for
research.
(a) QuickChip - the direct laser write facility for photolithography and
one-day gate array prototyping,
(b) Quintel Q-4000 double-sided mask aligner with IR light system for
double side mask alignment,
(c) A low cost mask making process with which we can produce 5", 25
micrometer resolution emulsion masks for less than $35. For most of the
micromachining research and prototype development these masks prove to be
quite sufficient.
(d) five target, magnetron sputtering machine,
(e) Photoresist spinners,
(f) Wafer spin rinser/dryer,
(g) Wet benches with fume hoods,
(h) Etching station with programmable control system,
(i) Two 4" diffusion/oxidation Tempress furnaces (4 tubes each),
(j) BHMJL Olympus metallurgical microscope with 35mm and Polaroid
cameras, a high-resolution TV camera and Sony TV monitor, and 3 stereo zoom
Bausch & Lomb microscopes, (k) Scanning electron microscope,
(l) Thermosonic wire bonder, ultrasonic wedge bonder,
(m) Karl Suss wafer scriber, Wafer fracturer, Die expander, Die
bonder, wentworth probe stations, four-point probe, (n) Variable speed
grinder/polisher for angle lapping and parallel lapping of semiconductor
samples,
(o) HP 4145A and HP 4145B programmable semiconductor parametric analyzers
The VLSI design facility contains a Sun server, 3 Sun Sparc10 workstations, 3
Sun Sparc2 workstations and 5 Sun Sparcstation1's. Numerous CAD tools for
design, simulation and analysis are available.
Devices
A. CMOS Micromachined Integrated Dynamic Thermal Scene Simulator
B. CMOS Micromachined electrothermal actuators
C. CMOS Micromachined Visual-to-Thermal Converter
D. CMOS Micromachined Integrated Alcohol Sensor
E. Miniature thermal peristaltic fluid pump
F. Micromachined pressure-time recorder for surgical applications
G. Micromachined accelerometer for consumer electronics
H. Temperature sensor with frequency output
I. Micromachined non-reverse valve
J. Signal conditioned circuitry for integrated transducers
University of Montreal
The Thin Film research group at U of M, known as the GMC (groupe des couches
minces or the Thin Film Group), has three service laboratories. These are the:
1. Surface lab: Studies solid surfaces and interfaces.
2. MODFAB (Microelectronic and Optoelectronic Device Fabrication Facility):
Works with silicon and compound semiconductor technologies
3. Ion Beam Facility: Can modify materials through deep high energy
implantation and characterize thin films by nondestructive ion beam
analysis.
Recent developments have been in areas including microelectronics materials and
processes, photonics materials and processing, and protective coatings.
Also associated with the University of Montreal is the LISA lab at Ecole
Polytechnique
LISA Lab (Ecole Polytechnique)
The LISA lab (Laboratory for the Integration of Sensors and Actuators), headed
by John Currie, is an installation for the design and fabrication of advanced
integrated devices using low-cost high-productivity micro-electronics
compatible technologies.
The lab has the following equipment at its disposal.
Fabrication:
sputtering system
ultrasonic/thermocompaction wire bonder
silicon chemical etching system
plasma enhanced CVD
rapid thermal annealing system
Patterning:
mask aligner
laser ablation system
reactive ion etching machine
Characterization:
point probe tester
potentionstat-galvanostat
characterization chamber
humidity analyzer
DEKTAK film thickness profilometer
Stress Tester
SEM
Concordia
Concordia University houses the Microelectronics Device and Fabrication
Laboratory. This facility conducts research into both processing and materials
for conventional semiconductor devices and processing for MEMS devices and
systems (also called micromechatronic systems by this lab).
The Concordia lab has some of the following equipment at its disposal
Infrastructure:
* 40 sq. m. of class 1000-5000 clean room
* 10 sq. m. of class 100 softwall clean room
* wet bench and fume hood
* 4 laminar flow work stations rudimentary mask-making by the rubylith method
Specialized Capabilities:
* KOH anisotropic etching of Si
* TMAH anisotropic etching of Si
* electrochemical discharge drilling
* anodic bonding of Si and Glass
* rapid thermal processor for annealing, oxidation
* corona discharge processing of Si in thermal oxidization furnace
Projects Underway:
* anisotropic etching using KOH, TMAH
* effects of substrate stress on etch anisotropy
* integrated sensors: accelerometer, Hall effect, gas
* tactile sensing
* MEMS device design rules
* characterization of mechanical properties of microsensor materials
* LIGA designs
* pressure micro transducer using anodic bonding and electrochemical
discharge drilling
Upcoming work:
* MEMS device design rules
* characterization of mechanical properties of microsensor materials
* integration of circuitry for "smart sensors"
* TMAH anisotropic etching for post processing
* use of IR aligner for double-sided designs
Carleton
The Department of Electronics at Carleton is exploring the use of
micromachining to produce passive components such as spiral inductors and
transformers suspended on oxide films for high-frequency RF circuits on silicon
substrates. By etching away the substrate, it is possible to considerably
reduce losses and increase Q. Ph.D. student John Long is carrying out most of
this work. So far this technique has been applied on devices formed in the
BATMOS process at Northern Telecom.
University of Alberta
University of Alberta's activities in MEMS include characterization of
polysilicon layers, design and testing of devices for sensor applications,
numerical simulation of MEMS devices, and CMOS micromachining courses.
APPENDIX TWO
CORPORATE AND GOVERNMENT AGENCIES
Canadian Microelectronics Corporation (CMC)
CMC lists micromachining technology as a possible extension to their planned
activities from the year 1995-2000. Currently they are acting as the go between
for Universities participating in micromachining research and Mitel Corp. Mitel
has completed a production run for a number of member Universities that are
currently undergoing micromachine research.
CMC believes there is considerable Canadian research talent in the
micromachining field, with a growing base. They feel, however, that the growth
of the industry in Canada is much slower due to an undeveloped market.
CMC is planing a micromachine design kit that will make use of the three
dimension capable CAD application, Tanner Tools.
Planning milestones for CMC include:
1994:
* Form stakeholder connections; leverage CMC's IC fabrication options for
minimal support.
* Technical task group formed.
1995:
* Initial funding is in place
* First training course offered.
* Service contracts placed for delivery of two flavours of manufacturing
* Initiate preparation of design kits and training workshops
* Investigate laser direct-write options for managing multiple flavours of
processing.
1996:
* Lead-site established for preparing microelectromechanical design kits
(integration of analog, digital, mechanical devices in one CAD tool).
* Library conventions selected or defined.
1998:
* Deliver CAD tools, design kits, and training as the basis of a general
program
* Plan for supporting electronic-mechanical co-design.
Mitel
Mitel is the largest non-captive foundry in Canada. As a Canadian company, they
would like to urge the scientists and engineers in Canada to stay in contact
with the company whenever they have special requirements or concerns.
Mitel's 1.5 micron process is being used by CMC member universities for a
fabrication run of micromachine designs. Features of the 1.5 micron process
include:
Double poly/double metal
3 micron poly and metal 1 pitch
5.5 Volts maximum operating voltage
2.7-3.6 Volts low voltage option
1.2 Volts low voltage option
Direct shrink of Mitel 2 micron design rules
Low TCR resistor module
Design rules for this process reproduces the 2 micron process and betters it by
a factor of 25%. Minimum pitch is available on the 1.5 micron process on all
the main layers: poly1, poly2, contact and metal1. If a double sided process is
desired, most likely the designer will want to get the wafer without bonding
since this would allow for further processing to be done on the wafer.
For some of the post-processing, some designs may require more than a single
die. Partial wafers or multiple die are possible at Mitel.
The issue of compatibility between more traditional IC fabrication and
micromachine design on the same chip was addressed. Due to the space available
for fabrication runs, a chip wafer cannot be devoted solely for micromachine
structures, therefore the compatibility issue has to be dealt with. It must be
determined what a designer can do with his die without affecting his neighbor's
chip.
For example, a micromachine design might need space left around the structure
in order to give enough undercutting capability for etching, and that might be
mistaken for using up space. The issue of affecting the integrity of the
process for the neighboring chip must be delt with.
There may be an implementation of a systematic way to describe design
violations that the designer intends to make, so they can automatically be
excluded from the quality checks that the rest of the designs go through.
One option would be to submit the micromachine designs with known violations
after the other designs had gone through the normal DRC procedures. The DRC
violations that the micromachine designs caused could then be ignored.
Another option would be to come out with a definition for the CIF and GDSII
layers for "open" or DRC exclusions and for other special layers. It was
proposed that a set of general rules of design guidelines be set forth by Mitel
and CMC.
Another very important issue with Mitel is packaging. The company has seen many
interesting designs which are not practical because there was no arrangement
for packaging. As Mitel puts it. The three most important issues they want to
see addressed in MEMS is packaging, packaging and packaging
With questions relating to any processes, Hsu Ho (ho@cmc.ca) of CMC was
designated as the contact. He would then contact Mitel or Goal as needed.
Some questions and answers:
1) Would there be any design rule violations that we may use for a MEMS
structure that would not be allowed because of process or technology
constraints?
Mitel would deal with this on a case by case basis. Every design rule has a
reason, any compromise would be on a case by case basis.
2) What type of material will there be on the backside to etch through. What
films are left on the backside of the wafer at the end of the process?
There's nothing left. There is bare silicon at the end of the process. Mitel
has also done some experimentation in CMOS polished double sided wafers
. 3) When backside etching is part of post-processing, it would be useful to
know which of the processes the backside of the wafer is exposed to and which
are single side processes. This will give a feel for what type of material is
on the backside to etch through.
There were concerns raised about post-processing double sided wafers.
Protecting one side of the wafer with a wax is used at the LISA lab. With a
TMAH etchant, the wax provides protection for about one hour under low
temperature (18 deg C). EDP has not been tried.
Goal Electronics
Goal provides commercial access to a wide range of integrated circuit processes
with supported libraries of analog and digital standard cells. We offer low
cost access, through shared wafer services, to the Mitel's 1.5 micron double
polysilicon, double metal process. Upcoming shared wafer runs are scheduled for
September 94 and January 95.
Goal Electronics also provides design assistance with interfaces to Cadence or
Viewlogic design environments, or can perform full custom design to a customer
specification. They have offices in Montreal, Quebec and Waterloo, Ontario.
Optics Company
See MITEL SESSIONS: DAY 2 Presentation 3
National Research Council
See MITEL SESSIONS: DAY 2 Presentation 4