Full text of DOI:10.1557/mrs2001.65

www.mrs.org/publications/bulletin
process is already much studied, its intro-
duction into the MEMS fabrication sequence
has been problematic, due to the require-
ments of controlling trace amounts of water
on the surface and in the solution, stability
of the precursor solutions (the organic
compounds tend to agglomerate in the
bulk solvent), and stability of the resulting
films to temperature and humidity. Hy-
drolysis and condensation reactions in the
solvent compete with surface reactions,
leading to micelle deposition rather than
monolayer deposition.8 The morphology
of the coating strongly affects release yield.9
According to contact-angle measurements,
films of ODTS thermally degrade in air or
oxygen below 200ꢀC, but survive to 400ꢀC
in vacuum or nitrogen, while FDTS sur-
vives in air to 400ꢀC.10 However, FDTS
films have also been reported to degrade
in high-humidity environments (ꢁ80% RH
for extended periods), perhaps due to film
restructuring to form surface micelles or
water adsorption at defects.11
Alternative CVD processes using volatile
fluoroalkylsilane precursors alleviate some
of the problems encountered in solution-
based processes.12 Precursor chemistry is
easily controlled in the gas phase, efficient
transport at low pressure ensures coating
of high-aspect-ratio structures, and self-
limiting surface reactions lead to conformal
monolayer coverage. Plasma-enhanced
deposition has also been used to form thin
fluorocarbon films with low surface en-
ergy,13 but issues of conformal coating,
coverage of inaccessible surfaces, and con-
trol of film properties remain unresolved.
Organic films directly bonded to silicon
(without a surface oxide present) are also
being explored with alkenes forming Si–C
linkages to the surface14 and amines form-
ing Si–N linkages15 emerging as candidates.
The structure, performance, and stability
of such organic films have yet to be thor-
oughly investigated.
The most sophisticated MEMS devices
employ sliding contacts in rotary- or linear-
motion devices. None have yet been com-
mercialized, but they have great appeal
because of the high functionality at low
cost that can be achieved with gears, linear
racks, rotating platforms, and pop-up mir-
rors.1 In these devices, issues of friction
and wear become important in addition to
adhesion. Organic monolayer films are
again the focus of efforts to develop lubri-
cants for MEMS. These films have low
friction coefficients, and their performance
naturally depends on the chemical nature
and/or structure of the films. For example,
ODTS films show a lower initial friction
coefficient than shorter-chain fluorinated
FDTS films, presumably due to higher
packing density and ordering of the hy-
Tribology of MEMS
M.P. de Boer and T.M. Mayer
Because of large surface-to-volume ratios
and low restoring forces, unwanted adhe-
sion and friction can dominate the perform-
ance of microelectromechanical systems
(MEMS) devices. To guarantee the func-
tion and reliability of MEMS devices, tri-
bologists must understand the origins of
adhesion, friction, and wear over a broad
range of length scales from the macro-
scopic to the molecular. In this article, we
present an overview of challenges, suc-
cesses, and initial steps toward a funda-
mental understanding.
be implemented to guarantee that free-
standing films exhibit low curvature, and
sufficient design tolerance must be built
into the device to avoid contact and adhe-
sion due to handling after release of the
freestanding structures.
Contacts are often inherent to the func-
tion of a device. Optical switching devices
require rapid contact and disengagement,
creating a dynamic adhesive interface.
Microrelays have conflicting demands of
low contact resistance and low adhesion.
In some applications, switches may remain
in contact for extended periods. In all of
these, long-term changes in adhesive energy
due to the stability of the interface and en-
vironmental factors are of great concern.
Low-surface-energy, hydrophobic coat-
ings applied to oxide surfaces are promising
for minimizing adhesion and static-charge
accumulation. Typically, these are very thin
or monolayer organic coatings, either
physisorbed or covalently bound to the
surface. Nonpolar organic groups on the
surface exhibit low adhesive energy and
low static-charge accumulation; in some
cases, they are self-healing. In addition to
exhibiting low surface energy, coatings
must be compatible with subsequent de-
vice processing, including packaging proc-
esses involving thermal treatments of
400–500ꢀC. Examples of coatings that have
been successfully introduced to commer-
cial MEMS products include Texas Instru-
ments’ report of a perfluorodecanoic acid
coating on structural aluminum that self-
heals damaged areas5 for their Digital
Micromirror Device™. Analog Devices
has patented nonpolar phenylsiloxane
coatings for accelerometers that resist
charge buildup and also survive packag-
ing temperatures as high as 500ꢀC.6
Polycrystalline silicon (polysilicon) is the
material of choice in surface-micromachined
MEMS because it is compatible with
integrated-circuit technology, can be stress-
relieved to less than 10 MPa, and deposits
conformally by low-pressure chemical
vapor deposition (CVD), allowing the fab-
rication of sophisticated hub geometries.1
(For more on the technique of surface mi-
cromachining, see the article by Mehregany
and Zorman in this issue.) However, sili-
con oxidizes readily to form a hydrophilic
surface, making it susceptible to adhesion.
Most metals also oxidize, so the problem
is relevant to other micromachining tech-
nologies. Oxide surfaces are also prone to
accumulating static charge, which can
interfere with capacitive-sensing circuitry.
Given these surface properties, we must
consider aspects of device design, mate-
rials selection, and processing, and spe-
cific surface modifications to minimize the
effects of adhesion, friction, and wear.
Even in applications where contact is
never intended, adhesion arises as a signifi-
cant problem. For example, in accelerome-
ters and gyroscopes, compliant mechanisms
are freely suspended above the substrate.
During fabrication, a sacrificial material
surrounds the structural films and must
be removed to render the films freestanding.
Sacrificial films are etched or dissolved in
liquid in a so-called release step. To avoid
capillary collapse,2 a supercritical carbon
dioxide drying process has been devel-
oped to circumvent the associated surface
tension.3 However, strain gradients through
the thickness of the films can cause struc-
tural members to curl and contact the
substrate, resulting in adhesion. Proper
deposition and annealing sequences4 must
Coatings must not introduce stress gra-
dients and must coat even the most inac-
cessible surfaces, so line-of-sight deposition
techniques of hard materials are not likely
to work. Therefore, much effort has been
focused on organic monolayer films, such
as the silane coupling agents octadecyl-
trichlorosilane (C18H37SiCl3, ODTS) and per-
fluorodecyltrichlorosilane (C8F17C2H4SiCl3,
FDTS).7 These films are typically deposited
from nonaqueous solution after the re-
lease step in MEMS fabrication. While this
302
MRS BULLETIN/APRIL 2001

Tribology of MEMS
drogenated alkyl groups.16,17 The friction
coefficient of organic film-modified sur-
faces increases after moderate periods of
applied shear force, and the surfaces
exhibit significant wear.18,19 These observa-
tions suggest that organic monolayer films
may not be robust enough to serve as
lubricants in devices requiring extended
sliding contact. Other hard coatings may
be necessary to prevent extensive wear in
sliding contacts. A self-limiting conformal
CVD of tungsten on polysilicon to form
a hard coating shows superb resistance
to wear.20
To develop robust materials and proc-
esses for MEMS devices, in situ micro-
tribology tools are needed to quantify
properties of adhesion, friction, and wear
on as-fabricated MEMS parts. Figure 1
shows a microtribological laboratory on
a chip that is being developed at Sandia
National Laboratories.21 Test structures,
metrology, and mechanics models are inte-
grated to enable high-resolution mechani-
cal and interfacial property measurement.
Deflection of electrostatically loaded can-
tilevers, measured by interferometry (Fig-
ure 1a), can be compared to finite-difference
models to determine Young’s modulus to
5% accuracy.22 Residual strain can be
ꢂ6
ꢂ5
measured to ꢀ10 resolution and ꢀ10
accuracy with electrostatically loaded
fixed–fixed beams (Figure 1b).23 Canti-
levers (Figure 1c) are simple but powerful
structures to measure interfacial adhe-
sion24 and adhesion hysteresis,11 lending
themselves to studies of the interplay of
adhesion with interfacial roughness, hu-
midity, and coatings. For example, adhe-
sion of hydrophilic surfaces increases
exponentially with relative humidity. This
observation can be explained by a single-
asperity model.25 Friction is being studied
as a function of pressure and velocity
by use of a cantilevered hinged-pad test
structure (Figure 1d).26,27 Although the prop-
erties of interest can be determined to high
resolution, the test fixtures occupy only a
small percentage of the wafer area. This is
important because most of the available
wafer area is dedicated to MEMS devices.
At the more fundamental level, the
challenge to understanding the tribologi-
cal properties of MEMS devices extends
down the length scale from micrometers
to angstroms. As seen in Figure 2, macro-
Figure 2. Length scales that must be
linked to understand micromachine
device performance include
(a) as-deposited and potentially
degraded coatings (angstrom scale),
(b) local adhesion measurement
(nanometer scale), and (c) asperity
interaction influencing micromachine
deflections (micrometer scale).
scopic contacts in MEMS devices are
composed of a distribution of small-area
contacts at nanometer-scale asperities. The
characteristics of surface morphology and
properties of single asperity contacts can
be studied by techniques such as atomic
force microscopy.28,29 The characteristics of
these nanometer-scale contacts are in turn
strongly affected by the atomic-scale
chemistry and physics of the surfaces and
films coming into contact. Adhesion, fric-
tion, wear, and stability of surface films
are dependent on molecular structure, end-
group chemistry, molecular orientation, de-
fects, and energy-dissipation mechanisms,
which are not yet well understood.30,31
Van der Waals forces between noncontact-
ing portions of the surfaces also contribute
strongly to adhesion.11 Linking these length
scales from the macroscopic to the atomic
will be necessary in order to develop a
fundamental understanding of the ad-
hesion and friction values measured for
MEMS devices.
It is likely that no single technical ap-
proach will address all of the problems of
process compatibility, charging, adhesion,
friction, and wear, and their interactions
with environmental factors. Therefore, a
given MEMS application and packaging
technology must be carefully considered
in the context of which tribological issues
need to be resolved. The fundamental study
of tribological coatings through proc-
essing, surface characterization, and mod-
eling will be necessary to bring a wider
class of MEMS to the market.
Acknowledgment
Sandia National Laboratories is a multi-
program laboratory operated by Sandia
Corp., a Lockheed Martin company, for
Figure 1. Microtribology laboratory on a chip. Integrated interferometry and finite element
modeling in concert with optimized test structure design are used to determine thin-film and
interfacial properties. See text for description of (a)–(d).
MRS BULLETIN/APRIL 2001
303

Tribology of MEMS
18 (2000) p. 2433.
Proc. SPIE Materials and Device Characterization
in Micromachining, Vol. 3875, edited by Y.
Vladimirsky and C.R. Friedrich (SPIE—The
International Society for Optical Engineering,
Bellingham, WA, 1999) p. 61.
24. M.P. de Boer and T.A. Michalske, J. Appl.
Phys. 86 (1999) p. 817.
the U.S. Department of Energy under con-
tract No. DE-AC04-94AL85000.
13. B.K. Smith, J.J. Sniegowski, G. LaVigne, and
C. Brown, Sens. Actuators, A 70 (1998) p. 159.
14. W.R. Ashurst, C. Yau, C. Carraro, R.T. Howe,
and R. Maboudian, in Proc. Solid-State Sensor
and Actuator Workshop (Transducers Research
Foundation, Cleveland, 2000) p. 320.
References
1. J.J. Sniegowski and M.P. de Boer, Annu. Rev.
Mater. Sci. 30 (2000) p. 297.
2. C.H. Mastrangelo and C.H. Hsu, IEEE/ASME
J. MEMS 2 (1993) p. 33.
15. X.-Y. Zhu, V. Boiadjiev, J.A. Mulder, R.P.
Hsung, and R.C. Major, Langmuir 16(2000) p. 6766.
16. V. DePalma and N. Tillman, Langmuir 5
(1989) p. 868.
17. J.D. Kiely, J.E. Houston, J.A. Mulder, R.P.
Hsung, and X.-Y. Zhu, Tribol. Lett. 7 (1999) p. 103.
18. R.R. Rye, G.C. Nelson, and M.T. Dugger,
Langmuir 13(1997) p. 2965.
19. U. Srinivasan, J.D. Foster, U. Habib, R.T.
Howe, R. Maboudian, D.C. Senft, and M.T.
Dugger, Proc. Solid-State Sensor and Actuator
Workshop (Transducers Research Foundation,
Cleveland, 2000) p. 156.
20. S.S. Mani, J.G. Fleming, J.J. Sniegowski,
M.P. de Boer, L.W. Irwin, J.A. Walraven, D.M.
Tanner, and D.A. La Van, in Materials Science of
Microelectromechanical Systems (MEMS) Devices II,
edited by M.P. de Boer, A.H. Heuer, S.J. Jacobs,
and E. Peeters (Mater. Res. Soc. Symp. Proc. 605,
Warrendale, PA, 2000) p. 135.
25. M.P. de Boer, P.J. Clews, B.K. Smith, and
T.A. Michalske, in Microelectromechanical Struc-
tures for Materials Research, edited by S. Brown,
J. Gilbert, H. Guckel, R. Howe, G. Johnson, P.
Krulevitch, and C. Muhlstein (Mater. Res. Soc.
Symp. Proc. 518, Warrendale, PA, 1998) p. 131.
26. M.P. de Boer, J.M. Redmond, and T.A.
Michalske, in Proc. SPIE Materials and Device
Characterization in Micromachining, Vol. 3512
(SPIE—The International Society for Optical
Engineering, Bellingham, WA, 1998) p. 241.
27. B.T. Crozier, M.P. de Boer, J.M. Redmond,
D.F. Bahr, and T.A. Michalske, in Materials Science
of Microelectromechanical Systems (MEMS) De-
vices II, edited by M.P. de Boer, A.H. Heuer, S.J.
Jacobs, and E. Peeters (Mater. Res. Soc. Symp.
Proc. 605, Warrendale, PA, 2000) p. 129 .
28. R.W. Carpick and M. Salmeron, Chem. Rev.
97 (1997) p. 1163.
3. G.T. Mulhern, D.S. Soane, and R.T. Howe, in
Proc. 7th Int. Conf. on Solid-State Sensors and Ac-
tuators (Transducers ‘93) (IEEE Electron Devices
Society, Yokohama, Japan, 1993) p. 296.
4. H. Guckel, D.W. Burns, C.C.G. Visser, H.A.C.
Tilmans, and D. Deroo, IEEE Trans. Electron
Devices 35 (1988) p. 800.
5. S.A. Henck, Tribol. Lett. 3 (1997) p. 239.
6. J.R. Martin and Y. Zhao, U.S. Patent
No. 5,694,740 (1997).
7. R. Maboudian, Surf. Sci. Rep. 3 (1998) p. 209.
8. B.C. Bunker, R.W. Carpick, R. Assink, M.L.
Thomas, M.G. Hankins, J.A. Voigt, D.L. Sipola,
M.P. de Boer, and G.L. Gulley, Langmuir 16
(2001) p. 7742.
9. J.G. Kushmerick, M.G. Hankins, M.P. de Boer,
P.J. Clews, R.W. Carpick, and B.C. Bunker, Tribol.
Lett. 10 (1–2) (2000) p.103.
10. U. Srinivasan, M.R. Houston, R.T. Howe, and
R. Maboudian, IEEE/ASME J. MEMS 7 (1998)
p. 252.
11. M.P. de Boer, J.A. Knapp, T.A. Michalske,
U. Srinivasan, and R. Maboudian, Acta Mater.
48 (2000) p. 4531.
29. M. Salmeron, CHEMTECH (September 1998)
p. 17.
30. E. Barrena, S. Kopta, D.F. Ogletree, D.H.
Charych, and M. Salmeron, Phys. Rev. Lett. 82
(1999) p. 2880.
21. M.P. de Boer, N.F. Smith, N.D. Masters,
M.B. Sinclair, and E.J. Pryputniewicz, in ASTM
STP 1413 (2001) in press.
22. B.D. Jensen, M.P. de Boer, N.D. Masters,
F. Bitsie, and D.A. La Van, IEEE/ASME J. MEMS
(2001) in press.
31. H.I. Kim, M. Graupe, O. Oloba, T. Koini, S.
Imaduddin, T.R. Lee, and S.S. Perry, Langmuir
12. T.M. Mayer, M.P. de Boer, N.D. Shinn, P.J.
Clews, and T.A. Michalske, J. Vac. Sci. Technol., B
23. B.D. Jensen, M.P. de Boer, and F. Bitsie, in
15 (1999) p. 3179.
ꢀ
Abstract Deadlines — In fairness to all potential authors, late abstracts will not be accepted.
June 5, 2001: for abstracts sent via fax or mail
ꢁ
June 19, 2001: for abstracts sent via the MRS Web site
ꢀ
Materials Development
Characterization Methods
Process Technology
2001 MRS FALL MEETING
ꢀ
ꢀ
New
www.mrs.org/meetings/fall2001/
MEETING ACTIVITIES
SYMPOSIA
Inorganic Electronic Materials and Devices
T: Statistical Mechanical Modeling in Materials
Research
U: Advanced Fibers, Plastics, Laminates, and
Composites
Symposium Tutorial Program
A: Materials Issues in Novel Si-Based Technology
B: Materials Science of Microelectromechanical
Systems (MEMS) Devices IV
Available only to meeting registrants, the symposium tutorials
will concentrate on new, rapidly breaking areas of research.
C: Ferroelectric Thin Films X
Nanoscale Materials and Processes
Exhibit and Research Tools Seminars
D: Polarization Dynamics in Ferroic Materials
E: Materials for High-Temperature
Superconductor Technologies
V: Nanophase and Nanocomposite Materials IV
W: Nanoparticulate Materials
Y: Nanopatterning—From Ultralarge-Scale Integration
to Biotechnology
Z: Making Functional Materials with Nanotubes
AA: Self-Assembly Processes in Materials
Over 225 international exhibitors will diplay a full spectrum of
equipment, instrumentation, products, software, publications,
and services. Research Tools Seminars, an educational seminar
series that focuses on the scientific basis and practical applica-
tion of commercially available, state-of-the-art tools, will be held
again this fall.
and
Exhibit
F: Spintronics
G: Thermoelectric Materials 2001—
Research and Applications
Photonic/Optoelectronic Materials and Devices
Organic/Biological Materials and Devices
H: Progress in Semiconductor Materials for
Optoelectronic Applications
I: GaN and Related Alloys
J: Materials Engineering for Solid-State Lighting
K: Microphotonics–Materials, Physics, and Applications
For additional meeting information,
visit the MRS Web site at
BB: Organic Optoelectronic Materials, Processing,
and Devices
CC: Advances in Liquid Crystalline Materials and
Technologies
DD: Polymer Interfaces and Thin Films
EE: Electroactive Polymers and Their Applications as
Actuators, Sensors, and Artificial Muscles
FF: Physical Characterization of Biological Materials
and Systems
GG: Polymeric Biomaterials for Tissue Engineering
HH: Bio-Inspired Materials—Moving Towards
Complexity
Publications Desk
A full display of over 700 books, plus videotapes and electronic
databases, will be available at the MRS Publications Desk.
www.mrs.org/meetings/
or contact:
Thin Films and Surfaces
Symposium Assistant Opportunities
L: Thin Films–Stresses and Mechanical Properties IX
M: Surface Science and Thin-Film Growth in Electrolytes
N: Current Issues in Heteroepitaxial Growth—Stress
Relaxation and Self Assembly
O: Complex Oxide Heteroepitaxy
P: Advances in Surface Engineering—Fundamentals
and Applications
Graduate students planning to attend the 2001 MRS Fall
Meeting are encouraged to apply for a Symposium Assistant
(audio-visual assistant) position.
Member Services
Materials Research Society
Employment Center
Materials and Society
506 Keystone Drive
Warrendale, PA 15086-7573
Tel 724-779-3003
Materials Science, Processing, and Evaluation
An Employment Center for MRS meeting attendees will be
open Tuesday through Thursday.
II: Materials Issues in Art and Archaeology VI
JJ: Scientific Basis for Nuclear Waste Management XXV
KK: Design, Characteristics, and Properties of
Cementitious Materials
Q: Rapid Prototyping Technologies—Tissue Engineering
to Conformal Electronics
R: Electrically Based Microstructural Characterization III
S: Combinatorial and Artificial Intelligence Methods
in Materials Science
Fax 724-779-8313
E-mail: info@mrs.org
X: Frontiers of Materials Research
01-0024
The 2001 MRS Fall Meeting will serve as a key forum for discussion of interdisciplinary leading-edge materials research from around the world.
Various meeting formats—oral, poster, round-table, forum and workshop sessions—are offered to maximize participation.
www.mrs.org/publications/bulletin
304
MRS BULLETIN/APRIL 2001