Inductive Detection of Magnetic Beads in a Microfuidic Channel

Report
Electrical Characterization of Graphite Oxide
by Ira Jewell
Alternative, chemical, methods for isolating graphene have been pursued and this
work intends to follow one of the most popular. In this method, graphite oxide (GO)
is first synthesized. For this we used the technique of Hummers as modified by
Kovtyukhova [2]. This graphite oxide is hydrophilic and readily disperses in water
with ultrasonic agitation. Single-layer platelets can then be isolated by
centrifugation. The resulting product is sprayed onto a heated substrate so that it
flash evaporates, leaving a distribution of platelets. These GO platelets are nonconductive and not of much use, but under certain conditions they can be made to
conduct.
Unfortunately, results from this current design have not yet been obtained (see Future Work
below). We have, however, shown with two-point measurements of rather thick samples of
graphite oxide that the resistance of the sample can be significantly reduced by simple exposure
to a forming gas ambient at an elevated temperature (300°C)
To a magnetician, graphene is interesting because spins can remain coherent over
long distances because of the long times between scattering events. This makes
graphene a good candidate material for spintronics applications. One group of
researchers have demonstrated the spin valve effect with a sample of several-layer
graphene forming the “spacer layer” between a pair of Cobalt contacts. The
magnetic electrodes were separated by 500nm and the experiment can be taken as
evidence that the “spinformation” contained in the injected electron survives at least
that distance.
sample #2
1.80E-12
1.60E-12
1.40E-12
1.20E-12
1.00E-12
Amps
Since the discovery in 2004 that single atomic planes of graphite could exist
individually [1], there has been much interest in this two-dimensional material known
as graphene. The original technique for isolating graphene was simply micromechanical cleavage from bulk HOPG (highly-oriented pyrolitic graphite) with the
“sticky-tape” method being among the most successful innovations. With this
technique, the experimenter repeatedly divides a piece of HOPG using regular
“Scotch” tape and presses it against a substrate (usually Si oxidized to a specific
thickness to ensure maximal visibility of graphenes) and transfers it with gentle
rubbing. While this method does transfer single and few-layer graphenes, it also
transfers many larger graphitic particles (as seen in this slide’s background) and
finding the desired material is much like finding the needle in the haystack.
Results
8.00E-13
6.00E-13
y = 4E-13x - 1E-13
4.00E-13
2.00E-13
0.00E+00
-2.00E-13 0
0.5
1
1.5
2
2.5
3
3.5
4
1
1.5
2
Volts
sample #2 (reduced)
1.50E-05
1.00E-05
5.00E-06
The resistance (inverse slope of plots
shown) is seen to decrease dramatically
after reduction.
-2.5
-2
-1.5
-1
0.00E+00
-0.5
0
-5.00E-06
0.5
y = 5E-06x + 3E-09
-1.00E-05
-1.50E-05
Volts
We propose to use direct-write laser lithography to define four-point resistivity
structures atop individual single-layer GO platelets. The number of atomic layers in
the samples will be characterized using both atomic force microscopy (AFM) and
Raman spectroscopy. Resistivity measurements will be taken in situ as the GO
sample is exposed to a variety of reducing gasses over a range of temperatures.
We have already shown the possibility of increasing the conductivity of a sample by
several orders of magnitude (using two-point contact measurements on a multilayered sample) and hope to refine our technique to produce device-quality
graphenes by this method in the near future.
Why is Graphene Interesting?
Spin valve effect showing lower resistance state when the direction
of magnetization of the two electrodes are aligned. [6]
Experimental Design
A solution of graphite oxide was first prepared
according to the method of Hummers as modified
by Kovtyukhova [2]. The solution was filtered and
allowed to dry, forming GO paper. The graphite
oxide was then resuspended in solution and
exfoliated by stirring for several days.
Single-layer graphene is interesting for a large number of reasons, both
theoretical and practical. Perhaps the most unique feature of graphene lies in its
bandstructure. At certain points in reciprocal space the conduction band and the
valence band actually touch, making graphene a “zero-bandgap” semiconductor.
Even stranger, in the region of k-space near these points the dispersion is linear.
To a physicist, this is interesting because it forces the (conduction) electrons in
graphene to behave as “massless Dirac fermions” and not like an ordinary
electron at all. These electrons possess zero rest mass and travel with a constant
velocity ≈ 106 m/s [4]. In addition, graphene displays such exotic effects as the
anomalous (half-integer) quantum Hall effect and a nonzero Berry’s phase [5].
Gold (~100nm) and a thin chrome adhesion layer
(~5nm) was then deposited using thermal
evaporation. Liftoff was performed using acetone
and ultrasonic agitation to remove the
undeveloped photoresist, thus defining the metal
contact structures.
The next step in the project is to
characterize each of the potential samples
of GO, keeping only those with one or two
layers. We will use atomic force
microscopy, as step height profiles can
reveal the number of layers in the sample.
Once characterized by AFM, the next step is
to connect each of the 90μm pads (of the
structure with the correctly situated GO
sample) to larger contact pads via
wirebonding. We will use a Mech-El 827
manual bonder.
Finally, the appropriate electrical connections
will be made, and the entire substrate placed
in a “ProboStat” measurement cell so that
resistivity can be monitored in situ as
reducing gasses are flowed and temperature
is varied.
Acknowledgements
The substrate was then heated on a hotplate and
the solution of GO spray deposited as a fine mist
so that the droplets flash-evaporate. This is
necessary because the GO is hydrophilic and
would agglomerate if larger drops were used [7].
The substrates were then analyzed by optical
microscope to find samples that have by chance
landed in the correct place (spanning all four
probes of a measurement structure)
Hall coefficient (which is inversely proportional to carrier
concentration) as function of applied field [1]
Future Work
AFM image of single folded sheet of GO [8]
A Heidelberg DWL66fs laser lithography system
was used to pattern photoresist atop a Si wafer
substrate (with 300nm thermally grown SiO2 to
enhance visibility of graphenes). The pattern is a
repeating grid of four-point measurement
structures consisting of 1μm wide lines connecting
to larger (90μm x 90μm) pads for probing or
wirebonding.
Bandstructure of single-layer graphene [3]
4.5
-4.00E-13
Amps
Abstract
To an engineer, this is interesting because the number and type of carriers (electrons
or holes) can be tuned by application of an electric field. Moreover, both carrier types
have quite high mobilities (μ ≈ 10,000 cm2/V-s) even with carriers approaching
concentrations as high as the theoretical limit ≈ 7*1012 cm-2. Graphene’s high mobililty,
coupled with the fact that its conductivity can be smoothly modulated by application of
an electric field through a “gate” electrode, make it an attractive material for FET
applications.
I would like to thank Dr. Albrecht Jander, Chris Tasker, Sean Smith, and ChienChih Huang. This project is funded by the Army Research Labs
contract #/280710A/
References
[1] Novoselov. Science. 306, 666 (2004)
[2] Kovtyukhova. Chemistry of Materials. 11, 771 (1999)
[3] Geim. Physics Today. 60 (Issue 8), 35 (2007)
[4] Novoselov. Nature. 438, 197 (2005)
[5] Zhang. Nature. 438, 201 (2005)
[6] Nishioka. Applied Physics Letters. 90, 252505 (2007)
[7] Gilje. Nano Letters. 7, 3394 (2007)
[8] Gomez-Navarro. Nano Letters. 7, 3499 (2007)
2.5

similar documents