This document summarizes work done to develop a photolithography process for fabricating microscale structures on quartz crystals. The structures will be used to study friction at the microscale using a nanoindenter and quartz crystal microbalance. The process involves cleaning the crystals, spinning on photoresist, exposing the photoresist to UV light through a photomask, developing the structures, and hard baking them. The authors optimized the exposure time, development time, and photoresist coating parameters to produce structures with straight edges for future friction experiments exploring the effects of contact size, speed, material, and thin films.

1. Abstract
Apparatus
Photolithography Process
Summary
What determines the friction and wear between two objects in
sliding contact? Although Da Vinci, Amontons and Coulomb
discovered the classical laws of friction centuries ago, this
question remains a puzzle for researchers seeking to
understand friction at the atomic level. Rapid progress has been
made over the last thirty years with the arrival of experimental
techniques capable of probing surface forces on ever smaller
length scales. These advances promise to help solve the
technical challenges currently plaguing the development of
microscopic machines. The devices, known as
microelectromechanical systems (MEMS), may serve as the
mechanical counterpart to integrated circuits, but progress
remains slow as structural materials and lubricant strategies
continue to be developed.
We discuss experiments in which a microscale probe and a
resonating disk of single-crystal quartz are put into contact to
replicate the sliding conditions found inside micromachines,
with rough contacts, high shear speeds, and lubricant films only
one molecule thick. This semester, we learned and implemented
a method for fabricating microscale structures with ultraviolet
photolithography. Characterization of the structures using
optical and electron microscopies played a key role in
optimizing our procedures. The fabricated features will serve
both as test sites for friction studies and as location markers to
enable post-test, ex-situ wear analysis..
Photolithography of microstructures
for investigations of high-speed micro-scale friction
Kieran Berton, Kali Gustafson, Halle Hund, Prof. Brian Borovsky
Physics Department, St. Olaf College, Northfield, MN
Acknowledgements
We gratefully acknowledge Jay Demas and his DUR group, Eric Cole, Elliot Schmidt, and Greg Cibuzar
and Jim Marti from the Minnesota Nano Center for helpful discussions, assistance, and technical
expertise.
References1. Brian Borovsky, Adam Booth, and Erin Manlove, Applied Physics Letters 91, 114101 (2007).
2. B. Borovsky, J. Krim, S. A. Syed Asif, and K. J. Wahl, Journal of Applied Physics 90, 6391 (2001).
3. B. Borovsky, B. Mason, and J. Krim, Journal of Applied Physics 88, 4017 (2000).
4. Top Image Courtesy of Nanorian Technologies: http://www.nanoriantech.com/photolitography/.
Bottom Image Courtesy of MNX: MEMS and Nanotechnology Exchange: https://www.mems-
exchange.org/MEMS/processes/lithography.html.
5 µm
Phase Contrast and Dissecting Microscope Images
Scanning Electron Microscope (SEM) Images
Future Work
Nanoindenter and Quartz Crystal Microbalance1,2
(NI-QCM)
Replicating contacts in
microdevices:
• Small contact radii: ~1 µm
• Fast sliding speeds: ~1 m/s
• Randomly rough surfaces
• Relevant materials and
lubricants
0 Tim
e
~30 min
Amplitude
0
50 nm
Amplitude at resonance is
ramped while normal load is
held fixed
Quartz resonance
curves
Far Right:
NI-QCM apparatus
and scanner base.
Near Right:
Close-up of
microsphere tip
with reflection on
electrode surface.
~0.02”
Measurement technique: Changes in the resonant frequency and
quality factor of the quartz disk (upon contact with the tip) are
determined as functions of oscillation amplitude, at a fixed normal
load.
fres

Change in resonant frequency upon contact with
the tip
Change in half bandwidth upon contact with the tip
1
2
resQ f
  
  
 
Change in inverse quality factor (dissipation)
0 1.4( / ) peakU pm V Q V  
Transverse oscillation amplitude of the
quartz surface is determined from the peak
drive voltage Vpeak.3
50 µm diameter
Gold electrode, with
organic monolayer
(octadecanethiol)
Aluminum oxide probe
30 second UV-light
exposure
(phase contrast)
33 second UV-light exposure
(dissecting)
34 second UV-light exposure
(phase contrast)
Topographic Mode
30 second UV-light
exposure
32 second UV-light
exposure
33 second UV-light
exposure
Full Mode
30 second UV-light
exposure
32 second UV-light
exposure
33 second UV-light
exposure
Scanning electron microscope (SEM) images were taken using the two modes shown above. An electron beam
interacts with atoms in the sample, producing various signals that contain information about the sample’s
surface topography and composition. In topographic mode, one electron beam is directed at the sample to
produce information about the sample’s surface topography. In full mode, two electron beams allow us to
characterize the sample’s composition.
We prepared the crystal by
cleaning it with isopropyl in a
10 minute ultrasonic bath. The
quartz crystal was then dried
with nitrogen and spin coated
with the photoresist for 30
seconds at 3000 RPM. The
prebake consists of a 1
minute, 105°C hardening of
the substrate. We then placed
our photomask over the
sample and exposed both to
UV-light at various time
increments. To develop, we
agitated the exposed crystal in
the microposit for 50 seconds
and then quenched the
developing process by setting
the crystal in DI water for 2
minutes. To set the
microstructures the we “hard-
baked” the crystal for 20
minutes at 120°C.
We used S1805 Photoresist and Microposit
351, which were both obtained from the
Minnesota Nano Center.
Now that we have refined the
fabrication process to
produce micro structures with
straight edges, we look ahead
to make structures which will
be more suited to our friction
testing goals. 2x2 micron
squares with a 20 micron
pitch will allow us to reliably
use the nanoindenter to
scrape both a structure and
the bare crystal with every
friction test. We hope to use
these structures in order to
know where we have
performed a friction test. In
future research, we also plan
to coat the crystal in a various
thin films to characterize how
the films affect friction
testing.
Preparation process4
Our goal this semester was to optimize the photolithographic
fabrication process to successfully create micro structures with
straight edges atop our gold-film quartz crystals. We varied
parameters such as UV-light exposure, microposit development,
and photoresist coating. Through experimentation, we
determined that a UV-light exposure time of 33 seconds, a
microposit development of 50 seconds, and a photoresist
coating of 30 seconds at 3000 RPM optimized the structures.
Informed by what we learned, we hope to perform friction tests
on our microstructures and continue to refine our
photolithographic development process.