StrainBased Resistance of SingleWalled Carbon Nanotubes

1
Strain-Based Resistance of Single-Walled Carbon
Nanotubes
Jonathon A. Brame1, Johnathan Goodsell1, Dr.
Stephanie A. Getty2 1 2006 ESMD Faculty/Student
Research Team Participant, Department of Physics,
Provo, UT, jon.brame_at_gmail.com 2 NASA Goddard
Space Flight Center, Materials Engineering
Branch, Code 541, Greenbelt, MD,
Stephanie.A.Getty_at_nasa.gov
Abstract The goal of this project is to fabricate
devices to test the strain-based change in
resistance of Single-Walled Carbon Nanotubes
(SWCNT) for use in micro-scale, high resolution
magnetometry. To do this, we must first
fabricate a device with electrically contacted
SWCNTs, then release the device onto a flexible
substrate for strain testing. We report progress
in growth techniques, testing techniques, and
comparisons between Chemical Vapor Deposition
(CVD) grown tubes and commercially available
SWCNTs.
Stretching Results The initial results of the
stretch-testing show evidence of reversible,
strain-based change in resistance in SWCNT
devices. Figure 10 shows both the characteristic
resistance changes with stretching/releasing, as
well as several distinct resistance levels
possibly activated by individual nanotube
contacts changing in the stretching process.
Applications A device of this scale capable of
measuring strain would be very useful in the
nano-technology industry. Specifically this
nano-sensor is being developed to create a
micro-scale vector magnetometer for use in
magnetospheric science and planetary magnetic
study and mapping (see Figures 6 7).
-Stretch
-Release
Level 2450 k?
slack in device
Background The remarkable effect of strain on the
conductivity of SWCNTs has been demonstrated
through local deformation (see Figure 1), and
through stretching of the tube structure (see
Figure 2). We seek to extend those results to an
array of SWCNTs in a strain sensor.
Level 1300 k?
Fig7 Possible planetary magnetic exploration
Fig 10 A step pattern was used for testing
resistance while stretching, stretching twice by
4 µm, then releasing back 4 µm
Fig 6 Earths magnetosphere
Fig. 1 Conductivity versus strain in a SWCNT
depressed by an AFM tip (Tombler et al. Nature,
2000).
Table 1
Table 1 Due to the vastly different stretching
characteristics of parylene and SWCNTs, uniform
stress is not distributed evenly throughout the
sample during a stretching. Most of the strain
is absorbed in the parylene, while the nanotubes
may slip within the substrate rather than stretch
as desired. As the parylene stretches, however,
it should cause enough displacement of the tubes
to change the resistance, since nanotubes
resistance are subject to change through bending
as well as stretching (see Tombler et al.)
Growth Results Using thin film iron catalyst has
shown marked improvement over the initial iron
nitrate catalyzed CVD growths (see Figure 8),
yielding initial SWCNT resistance decrease of
several orders of magnitude. Additionally,
magnetic tests were performed on tubes grown
using this new method to ensure that there is no
inherent magnetic effect from the iron catalyst
(Figure 9).
Fig 2 Drastic change in conductivity due to
stretching was measured for a semi-conducting
SWCNT (Cao et al. Physical Review Letters, 2003).
Fabrication After depositing a thin film layer of
Iron, SWCNTs were grown on the SiO2 substrate
through a CVD process using methane and ethylene
as the feed gases. Next we used a shadow-masked
gold evaporation to establish electrical contact
with the nanotubes and coated the whole device
with Parylene. Once the SiO2 is etched away from
behind, we are left with an electrically
contacted SWCNT device on a flexible substrate.
(see Figure 3)

• Conclusions and Future Work
• Increased Growth Density of SWCNT
• Transfer of CVD grown SWCNTs onto flexible
  substrate
• Fabrication of device to measure stretch-based
  resistance changes
• Development of methodology for stretching SWCNTs
• Preliminary results from stretch-testing
• Preliminary comparisons of CVD and commercial
  SWCNTs
• Future Work
• Joint-effort testing between NASA and BYU
• Continuation of stretch-testing
• Further testing of commercially available tubes
• Fabrication of micro-magnetometer

SiO2
Fe Thin Film
SWCNTs
Gold Contacts
Parylene
Etching (KOH)
Fig 8 The SEM image on the left shows SWCNTs
grown with Iron Nitrate catalyst, while the image
on the right shows SWCNTs grown using the thin
film iron catalyst technique (Note that the image
on the right is at twice the magnification as the
image on the left).
Fig 3
Once the SWCNTs were released onto the flexible
substrate, the device was mounted onto a frame
made of ceramic chips with gold contact pads
around the outside. The contact pads on the
frame were then wire-bonded to the gold contacts
on the nanotubes (see Figure 4). Once the sample
was attached and contacted to the frame, the
whole device was mounted onto a probe station for
resistance testing (see Figure 5).
Top view
Acknowledgements ESMD Program, Rocky Mountain
Space Grant Consortium BYU- Dr. David Allred,
Johnathan Goodsell Fisk University- Melissa
Harrison C. Taylor, D. Dove, C. Hoffman, L. Wang
Figure 9 The resistance of the nanotube device
remained constant through a changing magnetic
field
Fig 4
Fig 5