Transcript An Introduction to Carbon Nanotubes

An Introduction to Carbon
Nanotubes
John Sinclair
Outline


History
Geometry



Electronic Properties



Field Effect Transistors
Quantum Wires
Physical Properties


Rollup Vector
Metallicity
Ropes
Separation
Introduction

High Aspect Ratio
Carbon
nanomaterial



Family inclues Bucky
Balls and Graphene
Single Wall Carbon
Nanotubes
(SWCNT)
Multiwall Carbon
Nanotubes
(MWCNT)
History

1952 L. V. Radushkevich and V. M. Lukyanovich




1991-1992 The Watershed


50 nm MWCNT Published in Soviet Journal of
Physical Chemistry
Cold War hurt impact of discovery
Some work done before 1991 but not a “hot” topic
Iijima discovers MWCNT in arc burned rods
Mintmire, Dunlap, and White‘s predict amazing
electronic and physical properties
1993 Bethune and Iijima independently discover
SWCNT

Add Transition metal to Arc Discharge method
(same method as Bucky Balls)
Geometry

Rollup Vector



Chiral Angle




(n,m)
n-m=3d
tan(θ) =
√3m/(2√(n2+m2+
nm))
Arm Chair (n,n),
θ=30 ○
Zig-zag (n,0),
θ=0 ○
Chiral, 0○< θ<30
○
Field Effect Transistors


FETs work because of applied
voltage on gate changes the
amount of majority carriers
decreasing Source-Drain
Current
SWCNT and MWCNT used



Differences will be discussed
Gold Electrodes
Holes main carriers

Positive applied voltage
should reduce current
SWCNT Transport Properties




Current shape consistent
with FET
Bias VSD = 10 mA
G(S) conductance varies
by ~5 orders of
magnitude
Mobility and Hole
concentration
determined to be large



Q=CVG,T (VG,T voltage to
deplete CNT of holes)
C calculated from
physical parameters of
CNT
p=Q/eL
MWCNT Transport Properties

MWCNT performance
is poor without
defects



See arrow for twists
in collapsed MWCNT
MWCNT has
characteristic shape of
FET
Hole density similar to
SWCNT but Mobility
determined to be
higher

Determined same as
above
FET Conclusions




Higher carrier density than graphite
Mobility similar to heavily p-doped
silicon
Conductance can be modulated by
~5 orders of magnitude in SWCNT
MWCNT FET only possible after
structural deformation
Quantum Wires


SWCNT Armchair
tubes
SWCNT deposited
over two
electrodes

Electrode
resistance
determined with
four point probe
and found to be ~
1 MΩ
Coulomb Charging


Contact Resistance
Lower than
Rquantum=h/e2~26 kΩ
C very low s.t.
EC=e2/2C very large



If EC <<kT, Current
only flows when
Vbias>EC
Various gate V taken
into account
Step-like conductance
Quantum Wire

Strongly Temperature
dependent conduction curve




Occurs when a discrete
electron level tunnels
resonantly though Ef of
electrode
If electron levels of SWCNT
where continuous peak would
be constant
E levels separated by ΔE
The resonant tunneling
implies that the electrons are
being transported phase
coherently in a single
molecular orbital for at least
the distance of the electrodes
(140 nm)
Physical Properties of Ropes


SWCNT rope laid on
ultra-filtration
membrane
AFM tip applies force
to measure Shear
Modulus G and
Reduced Elastic
Modulus Er


Er = Elastic Modulus
when Searing is
negligible
Displacement of
tube/Force was
measured and Er and
G where calculated
Summary of Results

Typical Values




Gdia ~ 478 GPa
Ggla ~ 26.2 GPa
Er-dia ~ 1220 GPa
Er-gla ~ 65-90 GPa
Conclusion On Physical Properties


Shear properties of SWCNT lacking
(Even compared to MWCNT ropes)
Elastic properties very promising
Synthesis and Seperation

One major reason CNT devices have been
so hard to scale up to industry uses is due
to the inability to efficiently separate
different species of CNT


Different types are produced randomly with
1/3 conducting 2/3 semiconducting
It has now been reported that with the
use of structure-discriminating
surfactants one can isolate a batch of CNT
such that >97% CNT within 0.02 nm
diameter
Overview of Technique



Surfactants change buoyancy
properties of CNT
Ultra-centrifugation techniques
(which are scale-able) are used to
separate different CNT
Effective separation is seen


Separation according to metallicity
Separation according to diameter
Conclusion



CNT devices show promise in
molecular electronics both as wires
and FET
Physical properties are very
promising being both strong and
light
Separation techniques continue to
be developed to allow companies to
make CNT devices
Sources









M. S. DRESSELHAUS, G. DRESSELHAUS, and R. SAITO.
Carbon 33, 7 (1995)
R. Martel, T. Schmidt, H. R. Shea, T. Hertel, and Ph.
Avourisa. App. Phys. Lett. 73, 17 (1998)
Sander J. Tans, Michel H. Devoret et al. Nature 386,
474-477 (1997)
Jean-Paul Salvetat et al. Phys. Rev. Lett. 82, 5 (1999)
MICHAEL S. ARNOLD et al. Nature Nanotechnology 1,
60-65 (2006)
www.noritake-elec.com/.../nano/structu.gif
http://en.wikipedia.org/wiki/Carbon_nanotube
academic.pgcc.edu/~ssinex/nanotubes/graphene.gif
nano.gtri.gatech.edu/Images/MISC/figure4.gif