518
JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 10, NO. 4, DECEMBER 2001
Design and Fabrication of Submicrometer, Single
Crystal Si Accelerometer
Jason W. Weigold, Member, IEEE, Khalil Najafi, Fellow, IEEE, and Stella W. Pang, Fellow, IEEE
Abstract—A lateral accelerometer has been designed, simulated,
and fabricated using a 3-mask high-aspect ratio technology. Electron beam lithography and high-density plasma etching in an inductively coupled plasma source enabled aspect ratios 30 to be
achieved. This makes possible beams with very small spring constants. Combining the ability to measure very small displacement
of a proof mass due to narrow capacitive gaps between comb fingers, a highly sensitive accelerometer can be obtained. The fabricated accelerometer with 1 m beams and 0.2 m comb gaps
had a spring constant of 0.127 N/m, which is close to the calculated
values of 0.146 N/m. Based on the capacitance measurements, the
accelerometer sensitivity is calculated to be 6.3 fF/g. Reducing the
beam width to 0.4 m lowered the spring constant to 0.03 N/m,
and an improved equivalent sensitivity of 79.2 fF/g is calculated.
The minimum detectable acceleration is on the order of a few microgravity over a range of hundreds of gravities.
[612]
Index Terms—Accelerometer, deep etching, electron beam
lithography, MEMS, silicon.
I. INTRODUCTION
T
HERE are fairly simple processes available to fabricate
medium performance acceleration sensors capable of detecting milligravity (mg) acceleration signals [1]–[3]. However,
to obtain devices with noise floors in the microgravity ( g)
range, fabrication becomes more complicated in order to add
features and gain performance [4]–[7]. The use of single crystal
Si can increase the quality factor of the sensor which would
allow decreased thermo-mechanical noise due to the force of
gas molecules pushing against the sensor structure [8]. This
noise is probably the performance-limiting factor in many g
accelerometers. Also, the high-aspect ratios obtained using electron beam lithography and high-density plasma etching allow
very small capacitive gaps to be fabricated and very small sensor
motion to be detected [9]–[11]. This increases the sensitivity
of the device and makes possible the measurement of g acceleration signals. There are many transduction methods currently used in acceleration sensors, such as capacitive and tunneling displacement transducers [12]–[16]. In this work, capacitive transduction was used in order to gain performance from
capacitive gaps patterned using submicrometer lithography. Accelerometers have been designed which benefit from these improvements. These devices have been fabricated and tested in
Manuscript received August 2, 2000; revised April 21, 2001. This work was
supported by the Defense Advanced Research Project Agency. Subject Editor
D.-I. Cho.
The authors are with the Department of Electrical Engineering and Computer
Science, University of Michigan, Ann Arbor, MI 48109-2122 USA (e-mail:
pang@eecs.umich.edu; najafi@umich.edu).
Publisher Item Identifier S 1057-7157(01)05870-X.
order to verify the performance gains provided by the new technology.
Surface micromachined accelerometers are available commercially in high volumes with typical noise floors of 500 to
[17]. This noise floor makes possible detec7000 g/Hz
tion of mg signals, but no less. These accelerometers are fabricated using similar technology to the proposed accelerometer
so that comparisons may be made. This device has a 0.1 microgram proof mass with 1.3 m gaps between comb fingers.
It has a rest capacitance of 10 fF and a capacitance change
of 20 aF can be detected which translates to a 0.02-nm movement of the proof mass. New high performance accelerometers
have also been fabricated but are not yet commercially available.
An accelerometer has been fabricated using silicon-on-insulator
(SOI) wafers with a 52 microgram proof mass, a rest capacitance
of 9.7 pF, and a resonant frequency of 3 kHz [18]. The noise
floor of this accelerometer was measured to be 25 g/Hz .
These accelerometers with integrated readout circuitry can be
mass produced at low cost.
In this work, a high-sensitivity accelerometer is designed
which takes advantage of submicrometer lithography and
high-density plasma etching to gain performance over existing accelerometers. The single crystal Si accelerometer is
fabricated from the surface and requires no wafer bonding
or backside processing. It is 3 m thick with 0.2 m comb
gaps and compatible with integrated circuit (IC) processing
techniques. The design methodology for this accelerometer will
be described, followed by simulation of the sensor structure
and sensor fabrication. Finally, test results which highlight the
operation and performance gains of the new design will be
presented.
II. EXPERIMENT
The sensors were fabricated using a simple 3-mask process
which is shown schematically in Fig. 1. This process is completely compatible with previously developed processes which
integrate the sensor with conventional circuitry with the addition
to the circuit process of a single mask [11]. However, in order
to first demonstrate sensor performance gains, the circuitry was
not included.
The process begins with a B diffusion into an n-type Si wafer.
Heavily B-doped layer is formed by a B diffusion using a solid
boron nitride source at 1175 C for 30 min with 3 standard liters
per minute of N and 250 sccm of O to form a 3 m thick
layer with a B concentration of
cm . These p
p
B doped microstructures have some intrinsic stress and could
1057–7157/01$10.00 © 2001 IEEE
WEIGOLD et al.: SUBMICROMETER, SINGLE CRYSTAL Si ACCELEROMETER
Fig. 1.
519
A schematic process flow for fabrication of the acceleration sensor with Al bond pads.
introduce beam tip deflection for cantilevered beams [19]. However, for our accelerometer design and for many other sensor apmicrostructures have been used without stress
plications, p
related problems. The B diffusion determines the thickness of
the final device and in this case is 3 m. Next, a 1- m-thick Al
layer is lifted off to form the Al bond pads as well as to serve
as alignment marks in the electron beam lithography system. A
low temperature oxide (LTO) layer is deposited in a low pressure chemical vapor deposition (LPCVD) furnace and patterned
to cover the Al bond pads. The 900 nm thick LTO is deposited
at 420 C with a pressure of 360 mtorr for 110 min. This will
protect the bond pads from the release etch in ethylenediamine
pyrocatechol (EDP).
Electron beam lithography is then used to pattern polymethylmethacrylate (PMMA) to define the accelerometers with submicrometer beams and comb gaps. The 950 K molecular weight
PMMA resist was baked at 180 C for 30 min, giving a final
resist thickness of 400 nm. The PMMA was then exposed to
a 50 keV electron beam with a typical dose of 274 C/cm .
The exposed resist was then developed in a 1 : 1 solution of
methylisobutylketone and isopropyl alcohol (IPA) for 1 min followed by a 1 min rinse in IPA. This patterned resist is used to
lift off a 230-nm-thick evaporated Ni dry-etch masking layer.
A deep Si etch is performed in an inductively coupled plasma
(ICP) system using 80 W source power and 100 W stage power
at 1 mtorr with 10 sccm of Cl flow and a source to sample
distance of 13 cm. This etch condition is chosen to provide a
vertical etch profile for submicrometer beams and combs. The
dry etch time to completely etch through the 3 m B diffused
layer was about 45 min. Next, the samples are etched in EDP
for 15 min to release the structures, since EDP can selectively
etch the lightly doped n-type Si underneath the 3 m thick p
layer. Finally the Ni mask is removed in HCl and the LTO is
removed in BHF. In this case the BHF etch was carefully timed
so that the Al bond pads would not get attacked. This was found
to work well but must be characterized carefully and test etches
must be performed. In order to ease this process, a wet oxide
etch that does not etch Al could be used, or the oxide could be
removed through a dry etch. Even though the release etch was
performed in liquid, stiction of the movable structure to the fixed
structure or substrate was not observed. This could be due to the
large separation between the movable structure and the substrate
below it.
III. RESULTS AND DISCUSSION
A. Accelerometer Design
Preliminary calculations have been performed to estimate device characteristics which might limit the detectable acceleration. Fig. 2 shows an acceleration sensing structure which can
be used to calculate basic limitations and device characteristics.
Using a 3- m-thick proof mass, the mass of the sensor can be
calculated. 5 5 m holes in the proof mass are included in
order to allow the undercutting and release of the structure using
the EDP etch. The holes are spaced at a period of 8.5 m and
rotated 45 relative to the manhattanized layout. These dimensions will allow the large proof mass to be completely released
in just a 15-min EDP etch. The volume of the proof mass is calm and the mass is calculated to be
culated to be
kg, or about 9.0 micrograms.
If we assume that a comb array is used to sense motion and
that the comb fingers are 2 m wide with 0.1 m gaps between them, we can calculate the nominal capacitance and the
minimum resolvable movement of the mass. Using these dimensions, we have enough room for 807 comb fingers. Note that not
only do the small comb gaps increase sensitivity by increasing
the capacitance change, but also allow more fingers per area to
be included. We will assume a nominal finger overlap of 5 m,
although this could be increased or decreased in order to change
the nominal capacitance. If we assume that we can detect 0.1 fF
520
JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 10, NO. 4, DECEMBER 2001
difference in thermal expansion coefficient between the sensor
structure and the material to which it is anchored. However,
since this device is made of Si and the substrate is also Si, both
will expand and contract at the same rate with varying temperature, and the output should remain stable over temperature.
The comb drive on this sensor utilizes a varying overlap area
for capacitance change. This is often not used because the capacitance change is inversely proportional to the gap between
combs as shown in (3) [21]:
(3)
Fig. 2. A schematic of a 1 mm by 3 mm die containing the acceleration sensor.
The proof mass is 970 m wide, 1795 m long, and the thickness can be varied.
changes in capacitance, then this translates to a minimum detectable sensor movement of 0.12 nm.
Assuming dual 0.2- m-wide, 3- m-thick, and 2000- m-long
suspension beams, we can calculate a spring constant ( ) of
0.00093 N/m. A Young’s modulus of 150 GPa is assumed.
Using Hook’s law, the minimum detectable force ( ) of
N is obtained. This provides a minimum dem/s or about 1.3 g
tectable acceleration of
resolution. The resonant frequency of this device is about 51.2
Hz with a sensitivity of 75.99 pF/g. With such a high-capacitance sensitivity, the noise floor of the device may be limited
by thermo-mechanical noise caused by gas molecules pushing
against the structure. The noise equivalent acceleration ( )
due to Brownian motion of gas molecules is given by [8]
(1)
is the Boltzmann constant and is equal to
J/K and is the temperature and is assumed to be 300 K.
The damping term for one finger, , is given by the following
equation [20]:
where
(2)
where
viscosity of air and is equal to
kg/m-s;
area of the comb finger overlap and is equal to 3 m
5 m;
gap between comb fingers.
This damping term assumes a Couette-type damper model
which is appropriate in this type of comb drive. However, this is
only an approximation and this model assumes a plate with the
velocity of the fluid on one side of the plate decaying linearly
to zero at the boundary. The two sides of the comb were taken
into account by multiplying the damping term by 2 for each
side of the comb and then multiplying by the number of combs.
This should give a good approximation for the damping term,
. Using these methods, the Brownian noise limit is calculated
to be 4.4 g/Hz .
This structure has some advantages that make it particularly
strong in certain applications. Since the substrate and the sensor
are both made of single crystal Si, it is expected that the sensor
performance will be less sensitive to temperature variations.
Typically, sensor output will vary with temperature due to the
In a comb utilizing a varying gap instead of varying overlap area,
the capacitance change is inversely proportional to the square of
the gap which will provide a higher sensitivity:
(4)
Also, since the gap varies, the combs can be made long to increase the capacitance change. However, since the gaps utilized
here are so small, adequate capacitance change can be achieved
with the varying area comb drive and its advantages can be exploited. One advantage of this type of comb drive is that the
pull-in voltage can be made very large. As the length of the comb
fingers is increased, the pull-in voltage can be increased. The
only limit to this occurs when the comb fingers are too long,
they become less stiff in the off axis direction and can pull-in
with off axis acceleration. However, if they are kept stiff, large
accelerometer displacement can occur without any change in
sensitivity. This large displacement was simulated and accelerations as high as 1000 g could be tolerated before the maximum
stress in the device reached the yield strength of 7 GPa for Si.
This allows an extremely sensitive accelerometer with a very
large dynamic range to be fabricated with minimum detectable
accelerations on the order of a few g while being able to measure over a range of hundreds of g. This will probably be limited
by the detection circuit capability to resolve a small capacitance
change over a large background capacitance.
B. Finite Element Simulation of Accelerometer
Simulation using finite element methods has been performed
to verify mechanical behavior of the structure as well as to optimize accelerometer design. The commercially available finite
element simulation program, ANSYS 5.5, has been used for the
mechanical simulation [22]. During operation of the device, the
plate will remain fairly rigid due to its large stiffness. The etch
holes distributed throughout the plate will make it slightly less
rigid. In order to simplify the simulation, the etch holes were
lumped together into one large hole in the middle of the plate.
The plate would remain rigid, but the mass would be equivalent
to the mass of the actual device. Combs were omitted for simplicity from this model. The combs will add some mass to the
structure, however, this mass is small compared to the mass of
the plate, so it can be neglected for initial simulations. The simulated structure with lumped etch holes is shown in Fig. 3(a).
A close-up of the beam attachment to the plate is shown in
Fig. 3(b) with meshing completed. The simulations are modeled
using a 3 m thick, 1015 835 m plate with 4 beams that are
WEIGOLD et al.: SUBMICROMETER, SINGLE CRYSTAL Si ACCELEROMETER
521
C. Fabrication Results
Fig. 3. Simulated structure without comb fingers showing (a) the etch holes
lumped together in the plate center for simplification of simulation and (b) a
close-up of the attachment of beam to plate after meshing.
each 1228 m long and 0.5 m wide. The hole in the middle is
305 305 m to correctly model the mass of the plate.
Simulation can be performed to determine the positions of
the resonant modes as well as the force required to fracture the
device and the device’s reaction to gravity. From these, spring
constant and sensitivity can also be extracted. When simulated,
this device was broken into 58 739 elements and the mass was
kg. A modal analysis took about 5 min to extract
the first 5 resonant modes. The modes were at 360, 1091, 3029,
4571, and 39 944 Hz. The first mode was the desired mode of vibration in the direction. The second mode was vibration in the
-axis and the third and fourth modes were torsional. The fifth
mode which is way above the others, is a vibration of the springs
alone. The spring constant was calculated to be 0.027 N/m from
the simulation.
Problems were encountered when trying to simulate high
aspect ratio beams with a narrow beam width. Since the
accelerometer structure is quite large, the number of elements
increases rapidly. For beams with a very high aspect ratio, a
large number of elements must be used to satisfy aspect ratio
specifications of the software. Elements with high aspect ratios
cause larger errors in simulation, therefore, multiple elements
with a lower aspect ratio must be used. A large number of
elements has a large memory requirement and the software
package used also has a maximum element limit.
Fig. 4 shows a die photo of the completely released accelerometer. The surface around the structure shows texturing
from the EDP release etch. This is visible under the optical
microscope, however, when viewed under the scanning electron
microscope, the surface is quite smooth as shown in Fig. 5. The
accelerometer was completely released after 25 min in EDP
and was verified to be functional through electrical testing.
The suspended structure is shown in Fig. 5(a). The proof mass
can be seen suspended by one of the four beams and the Al
bond pad is in the top of the figure. The comb fingers are
also visible which provide drive capability and also serve to
electrically sense the motion of the structure. In Fig. 5(b), a
higher magnification of the comb fingers shows the vertical
profile and submicrometer gaps generated by etching the high
aspect ratio resonators in Cl . These comb drives are 3 m
thick with 0.2 m gaps in between.
If the structure is left in EDP for a long time, the whole structure will undercut and the anchored areas will release from the
substrate. The structure has been designed so that the whole device will be totally released in only 15 min in EDP. However, undercut of the bond pads is still undesirable because an undercut
bond pad will not have the mechanical strength that a completely
anchored one would. Fig. 6(a) shows a bond pad without corner
compensation. The pad is undercut even after 25 min in EDP.
However, corner compensation structures on the corners of the
bondpad as shown in Fig. 6(b) will prevent the EDP from undercutting at the corners. This will provide a strong anchor for
the structure, and will also make wire bonding to the pad easier
and resistant to fracture.
D. Accelerometer Testing and Performance Analysis
Testing of the released accelerometers was performed. Since
the structure is designed to be very sensitive to small forces, it
can be pulled in quite easily. The pull-in voltage for the structure
with 1- m-wide beams was found to be 5 V. For the pull-in
voltage of 5 V, the structure was displaced by 6 m. From this
pull-in voltage and the geometry of the structure, the force ( )
on one of the plates of the capacitor per finger is given by
(5)
where
dielectric constant of the material between the comb
fingers;
thickness of the structure;
gap between comb fingers;
voltage placed across the comb.
N and this is multiplied
The force is calculated to be
by 2 for the two plates on each side of the movable finger and
again by 230 for one set of combs. This gives a total force apN. Using Hooke’s Law and a deflection of
plied of
6 m for pull in, this gives a spring constant of 0.127 N/m which
is very close to the simulated spring constant of 0.146 N/m for
the 1- m-wide beams of this particular fabricated device.
A dc voltage was applied to the device across one set of
combs to provide a known force and the capacitance change
522
JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 10, NO. 4, DECEMBER 2001
Fig. 4. Die photo showing the whole accelerometer structure.
2
Fig. 5. (a) Scanning electron micrograph of a completely released accelerometer structure. The structure is completely released after a 25-min EDP etch.
The combs are visible and connected to the Al bond pads. (b) Higher magnification of the comb drive showing the 3-m-thick high-aspect ratio comb fingers
with 0.2 m gaps in between.
is measured. An accelerometer with 0.4- m-wide beams and
0.2- m-wide comb gaps was tested. The voltage across the
combs was set to 2.5 V and swept to 0 V in steps of 0.1 V.
The capacitance at each point was measured and a plot of this
capacitance–voltage ( – ) curve is shown in Fig. 7.
Negative voltages must be applied so that the p /n diodes are
kept reverse biased. The reverse leakage current was measured
to be in the A range and it does not affect the measurement of
[11]. The junction capacitance varies
junction capacitance,
decreased from 6.7 pF
with bias voltage and it was found that
at 0 V to 2.7 pF at 2.5 V. Therefore, this junction capacitance
will serve to mask the sensitivity during – measurements.
Fig. 6. Micrographs of (a) a 50 50 m bond pad without corner compensation after 25 min in EDP and (b) the same pad with corner compensation. The
pad with corner compensation is not undercut and provides good mechanical
support for the wire bond as well as the accelerometer anchor.
However, under normal device operation, a constant bias is applied, so that the junction capacitance will not vary.
From these two measurements, the junction capacitance variation can be subtracted out from the total measured capacitance
change, yielding the capacitance change due to movement of
the structure alone. The capacitance variation due to structure
motion increased from 513 fF at 1.3 V to 706 fF at 2.5 V.
This sensitivity corresponds to 269 fF/V due to movement of
the sensing mass. The variation of capacitance with force can
be calculated using [21]
(6)
WEIGOLD et al.: SUBMICROMETER, SINGLE CRYSTAL Si ACCELEROMETER
523
IV. CONCLUSION
Fig. 7. Measured capacitance change with applied voltage for fabricated
accelerometer with 0.4-m-wide beams. The voltage is only applied to one of
the two sets of comb fingers. This measurement is masked by the variation of
junction capacitance with voltage.
An acceleration sensor with a minimum detectable acceleration in the g range was designed in order to demonstrate and
prove the performance gains provided by the dry etch conditions and fabrication processes. This accelerometer will benefit
by the high aspect ratio gaps, providing increased capacitance
as well as a larger number of sense fingers in the comb sense
structure compared to sensors fabricated using other technologies. With submicrometer beam width and comb gaps, these devices should be able to achieve g sensing with the added benefit that the simple process can result in high yields with very
few processing steps. In addition, the process used is compatible
with circuit integration so that circuitry can be included with the
sensor.
kg and had
The fabricated sensor has a mass of
m . The measured spring
a proof mass that was
constant for 1- m-wide beams was 0.127 N/m, which was close
to the simulated spring constant of 0.146 N/m. From the measurements of capacitance changes, the equivalent sensitivity is
calculated to be 6.3 fF/g, which is similar to the sensitivity of
7.0 fF/g for the 1- m-wide beams obtained from the calculations based on device design. For the device with 0.4- m-wide
beams, an equivalent sensitivity is calculated to be 39.62 fF/g for
one set of combs or 79.2 fF/g for both sides. The accelerometer
sensitivity can be further improved by fabricating even narrower
beams. As shown in the calculations, 0.2- m-wide beams with
0.1- m-wide gaps can provide a sensitivity of 76 pF/g with a
noise equivalent acceleration of 4.4 g/Hz .
REFERENCES
Fig. 8. Capacitance change with acceleration determined by subtracting
out junction capacitance variation. In normal device operation, the sensor is
biased with a constant voltage and the junction capacitance will not vary with
acceleration.
where is the dielectric constant and a is the applied acceleration to the sensor. From the measured – curve, the capacitance change can be related to equivalent acceleration as
shown in Fig. 8. Therefore, the equivalent sensitivity for one
set of combs obtained from the changes in capacitance was
found to be 39.62 fF/g. Thus, operating using both combs will
produce an equivalent sensitivity of 79.2 fF/g, which is higher
than the calculated sensitivity of 44.2 fF/g. We have also fabricated accelerometers with 1- m-wide beam. Since the wider
beams have a large spring constant, the equivalent sensitivity
from the measured capacitance changes was lower to 48.6 fF/V
or 6.3 fF/g, which was very close to the calculated value of
7.0 fF/g for the device with 1 m beams. As the beam width
continues to be scaled down, the spring constant will decrease
further and the predicted resolution of a few g can be realized.
[1] S. J. Sherman, W. K. Tsang, T. A. Core, R. S. Payne, D. E. Quinn, K. H.
Chau, J. A. Farash, and S. K. Baum, “A low-cost monolithic accelerometer: Product/technology update,” in Proc. IEEE Electron. Dev. Mtg.,
San Francisco, CA, 1992, pp. 160–161.
[2] W. Yun, R. T. Howe, and P. R. Gray, “Surface micromachined digitally force-balanced accelerometer with integrated CMOS detection circuitry,” in Proc. IEEE Solid-State Sensor and Actuator Workshop, Hilton
Head Island, SC, June 1992, pp. 126–131.
[3] K. Chau, S. R. Lewis, Y. Zhao, R. T. Howe, S. F. Bart, and R. G. Marcheselli, “An integrated force-balanced capacitive accelerometer for low-g
applications,” in Proc. Eighth International Conf. on Solid State Sensors
and Actuators (Transducers ’95), Stockholm, Sweden, June 1995, pp.
593–596.
[4] A. Selvakumar, F. Ayazi, and K. Najafi, “A high sensitivity Z -axis torsional silicon accelerometer,” in Proc. IEEE Electron. Dev. Mtg., San
Francisco, CA, 1996, pp. 765–768.
[5] C. Yeh and K. Najafi, “Micromachined tunneling accelerometer with a
low-voltage CMOS interface circuit,” in Proc. Ninth International Conf.
on Solid State Sensors and Actuators (Transducers ’97), Chicago, IL,
June 1997, pp. 1213–1216.
[6] C. Liu, A. M. Brazilai, J. K. Reynolds, A. Partridge, T. W. Kenny, J. D.
Grade, and K. Rockstad, “Characterization of a high-sensitivity micromachined tunneling accelerometer with micro-g resolution,” J. Microelectromechan. Syst., vol. 7, pp. 235–244, 1998.
[7] N. Yazdi, A. Ayazi, and K. Najafi, “Micromachined inertial sensors,”
Proc. IEEE, vol. 86, pp. 1640–1659, 1998.
[8] T. B. Gabrielson, “Mechanical–thermal noise in micromachined
acoustic and vibration sensors,” IEEE Trans. Electron Devices, vol. 40,
pp. 903–909, 1993.
[9] J. W. Weigold, W. H. Juan, and S. W. Pang, “Dry etching of deep Si
trenches for released resonators in a Cl plasma,” J. Electrochem. Soc.,
vol. 145, pp. 1767–1771, 1998.
[10] J. W. Weigold and S. W. Pang, “Fabrication of thick Si resonators with a
frontside-release etch-diffusion process,” J. Microelectromechan.Syst.,
vol. 7, pp. 201–206, 1998.
524
JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 10, NO. 4, DECEMBER 2001
[11] J. W. Weigold, A.-C. Wong, C. T.-C. Nguyen, and S. W. Pang, “A merged
process for thick single crystal Si resonators and BiCMOS circuitry,” J.
Microelectromechan. Syst., vol. 8, pp. 221–228, 1999.
[12] F. Rudolf, A. Jornod, J. Berqovist, and H. Leuthold, “Precision
accelerometers with g resolution,” Sensors Actuators A, vol. 21–23,
pp. 297–302, 1990.
[13] K. Ma, N. Yazdi, and K. Najafi, “A bulk-silicon capacitive microaccelerometer with built-in overrange and force feedback electrodes,” in
Proc. IEEE Solid-State Sensor and Actuator Workshop, Hilton Head Island, SC, June 1994, pp. 160–163.
[14] B. Boser and R. T. Howe, “Surface micromachined accelerometers,”
IEEE J. Solid-State Circuits, vol. 31, pp. 366–375, 1996.
[15] B. P. Drieenhuizen, N. Maluf, I. E. Opris, and G. Kovacs, “Force-balanced accelerometer with mg resolution fabricated using silicon fusion
bonding and deep reactive ion etching,” in Proc. Ninth International
Conf. on Solid State Sensors and Actuators (Transducers’97), Chicago,
IL, June 1997, pp. 1229–1230.
[16] H. K. Rockstad, T. W. Kenny, J. K. Reynolds, W. J. Kaiser, and T. B.
Gabrielson, “A miniature high-sensitivity broad-band accelerometer
based on electron-tunneling transducers,” Sensors and Actuators A, vol.
43, pp. 107–114, 1994.
[17] [Online]. Available: http://www.analog.com/imems/library/sensor971.
pdf.
[18] M. A. Lemkin, T. N. Juneau, W. A. Clark, T. A. Roessig, and T.
J. Brosnihan, “A low-noise digital accelerometer using integrated
SOI-MEMS technology,” in Proc. Tenth International Conf. on Solid
State Sensors and Actuators (Transducers’99), Sendai, Japan, June
1999, pp. 1294–1297.
[19] J. W. Weigold, W. H. Juan, S. W. Pang, and J. T. Borenstein, “Characterization of bending in single crystal Si beams and resonators,” J. Vac.
Sci. Technol. B, vol. 17, pp. 1336–1340, 1999.
[20] M. A. Schmidtt, R. T. Howe, S. D. Senturia, and J. H. Haritonidis, “Design and calibration of a microfabricated floating-element shear stress
sensor,” IEEE Trans. Electron Dev., vol. 35, pp. 750–757, 1988.
[21] A. Selvakumar, Ph.D. dissertation, University of Michigan, Ann Arbor,
MI, 1997.
[22] “ANSYS 5.5.2,” ANSYS Inc., Southpointe, 275 Technology Drive,
Canonsburg, PA, 15 317, http://www.ansys.com.
Jason W. Weigold (S’95–M’00) received the B.S.E.E degree from the University of California, Los Angeles (UCLA) in 1995 and the M.S.E.E. and Ph.D.
degrees from the University of Michigan, Ann Arbor, in 1997 and 2000, respectively.
He has worked at Analog Devices, Inc., on surface micromachined accelerometers. His research interests include sensors, micromachining and
MEMS, dry etching, and other microfabrication technologies.
Dr. Weigold is a student member of the SPIE, the Electrochemical Society,
as well as the American Vacuum Society.
Khalil Najafi (S’84–M’86–SM’97–F’00) was born
in 1958. He received the B.S., M.S., and Ph.D. degrees in 1980, 1981, and 1986, respectively, all in
electrical engineering from the Department of Electrical Engineering and Computer Science, University
of Michigan, Ann Arbor.
From 1986 to 1988, he was employed as a Research Fellow, from 1988 to 1990, as an Assistant Research Scientist, from 1990 to 1993, as an Assistant
Professor, from 1993 to 1998, as an Associate Professor, and since September 1998, as a Professor, and
the Director of the Solid-State Electronics Laboratory, Department of Electrical
Engineering and Computer Science, University of Michigan. His research interests include: micromachining technologies, solid-state micromachined sensors, actuators, and MEMS; analog integrated circuits; implantable biomedical
microsystems; hermetic micropackaging; and low-power wireless sensing/actuating systems.
Dr. Najafi was awarded a National Science Foundation Young Investigator
Award from 1992 to 1997, was the recipient of the Beatrice Winner Award for
Editorial Excellence at the 1986 International Solid-State Circuits Conference,
of the Paul Rappaport Award for coauthoring the Best Paper published in the
IEEE TRANSACTIONS ON ELECTRON DEVICES, and of the Best Paper Award
at ISSCC 1999. In 1994, he received the University of Michigan’s “Henry
Russel Award” for outstanding achievement and scholarship, and was selected
as the “Professor of the Year” in 1993. In 1998, he was named the Arthur F.
Thurnau Professor for outstanding contributions to teaching and research, and
received the College of Engineering’s Research Excellence Award. He has
been active in the field of solid-state sensors and actuators for more than 18
years, and has been involved in several conferences and workshops dealing
with solid-state sensors and actuators, including the International Conference
on Solid-State Sensors and Actuators, the Hilton-Head Solid-State Sensors
and Actuators Workshop, and the IEEE/ASME Microelectromechanical
Systems (MEMS) Workshop. He is the Editor for Solid-State Sensors for
IEEE TRANSACTIONS ON ELECTRON DEVICES, Associate Editor for IEEE
TRANSACTIONS ON BIOMEDICAL ENGINEERING, Associate Editor for IEEE
JOURNAL OF SOLID-STATE CIRCUITS, and an Associate Editor for the Journal
of Micromechanics and Microengineering, Institute of Physics Publishing.
Stella W. Pang (S’81–M’82–SM’91–F’99) received
the Ph.D. degree in electrical engineering and computer science from Princeton University, Princeton,
NJ, in 1981.
She is currently Professor of the Electrical
Engineering and Computer Science Department at
the University of Michigan. From 1981 to 1989,
she was with Lincoln Laboratory, Massachusetts
Institute of Technology, Lexington, working on
submicrometer technology for microelectronic
applications. Her research interests include nanofabrication technology, dry etching, dry deposition, and surface modifications for
microelectromechanical, microelectronic, and optical devices. She has over
250 technical papers, book chapters, and presentations.
Dr. Pang has served as conference organizer for the American Vacuum Society, the Electrochemical Society, IEEE, the Material Research Society, and
SPIE. She is a Fellow of ECS and AVS.