scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Performance Comparison of Graphene Nanoribbon FETs With Schottky Contacts and Doped Reservoirs

19 Aug 2008-IEEE Transactions on Electron Devices (IEEE)-Vol. 55, Iss: 9, pp 2314-2323
TL;DR: In this article, an atomistic 3D simulation study of the performance of graphene-nanoribbon (GNR) Schottky-barrier field effect transistors and transistors with doped reservoirs (MOSFETs) was presented.
Abstract: We present an atomistic 3-D simulation study of the performance of graphene-nanoribbon (GNR) Schottky-barrier field-effect transistors (SBFETs) and transistors with doped reservoirs (MOSFETs) by means of the self-consistent solution of the Poisson and Schrodinger equations within the nonequilibrium Green's function (NEGF) formalism Ideal MOSFETs show slightly better electrical performance for both digital and terahertz applications The impact of nonidealities on device performance has been investigated, taking into account the presence of single vacancy, edge roughness, and ionized impurities along the channel In general, MOSFETs show more robust characteristics than SBFETs Edge roughness and single-vacancy defect largely affect the performance of both device types

Summary (2 min read)

Introduction

  • Graphene is a zero-gap Manuscript received March 24, 2008.
  • GNRFETs that are demonstrated experimentally to date are realized by connecting the channel to metals with Schottky contacts [8], [14], therefore obtaining a Schottky-barrier FET .
  • Different types of nonidealities have been investigated.
  • Vacancies and edge roughness can greatly affect device electrical performance more than ionized impurities actually do.

II. APPROACH

  • Device characteristics of GNRFETs are calculated by solving the Schrödinger equation using the NEGF formalism [25] selfconsistently with the 3-D Poisson equation [18]–[21].
  • Lattice vacancies or edge roughness are considered as atomistic defects of the channel GNR, where the existence of carriers is essentially prohibited.
  • An ionized impurity is treated as an external fixed charge, which can play an important role for the electrostatic potential of the device.
  • In other words, in the self-consistent iterative loop between the transport equation and the Poisson equation, the input charge into the Poisson equation always includes a fixed external charge as well as the output charge from the Schrödinger equation.

III. RESULTS

  • The authors first present results for an SBFET and a MOSFET under ideal conditions.
  • Fig. 2(a) and (b) shows the transfer characteristics for each device.
  • The MOSFET has 50% larger Ion (i.e., current for VG = VDD and VD = VDD) and larger transconductance gm than the SBFET.
  • In Fig. 4(a), the cutoff frequency fT as a function of the applied gate voltage is shown and computed by using the quasi-static approximation [27] as fT = gm 2πCG ∣∣∣∣ VD=VDD (1) where gm is the transconductance and CG is the gate capacitance computed as the derivative of the charge in the channel with respect to the gate voltage.
  • Fig. 4(b) shows the intrinsic delay as a function of on–off ratio:.

B. Atomistic Vacancy

  • Fig. 5(a) and (b) shows the transfer characteristics for SBFET and MOSFET, both in the linear and the logarithmic scale, for different positions of a defect.
  • As shown in Fig. 5, the defect near the source has the largest effect in both devices.
  • For an SBFET with a defect near the source, thicker SB is induced [Fig. 6(a)] due to the electron accumulation, and quantum transmission is reduced [Fig. 6(b)] at the ON state, which result in a smaller Ion.
  • Instead, the reduced number of propagating states due to the lattice vacancy reduces the transmission probability [Fig. 6(d)], which results in a smaller on current.
  • Transport is, indeed, mostly determined by the top-of-the-barrier potential, which, as shown in Fig. 6(c), is only partially influenced by the presence of the defect in correspondence of the drain (and in the middle of the channel).

IV. CONCLUSION

  • GNR SBFETs and MOSFETs are compared by solving the Schrödinger equation self-consistently with the 3-D Poisson equation.
  • His research interests include the physics, modeling, and simulation of nanodevices.
  • Dr. Guo is a member of the technical program committees of the International Electron Devices Meeting and the Device Research Conference.

Did you find this useful? Give us your feedback

Figures (14)

Content maybe subject to copyright    Report

Università di Pisa
!"#!$$%&#'"#()$*)&#+"#,$%-&#'"#./%%/00$%1&#2"#'3$&#!"#$%#&'()"*+%&,'#-.%(*%$*/#',0"("*1'(%#-22%(*345.*
6-70*8)0%779:*+%(7')7.*'(;*<%,";*="."#>%-#.ƞ&#.444#5*/%6/07)$%6#$%#48107*$%#91:)016&#!!&#;&#<<"=>?@A
=>=>#B=CCDE#
#
Performance Comparison of
Graphene Nanoribbon FETs With
Schottky Contacts and Doped
Reservoir
"#$%&'()"##%)
91</*7F1%7#$G#48107*)0/8#/%H#I$F<371*#4%-)%11*)%-&#J%):1*6)7K#$G#(8$*)H/#
*(+%,$-+).(#/()
9)</*7)F1%7$#H)#.%-1-%1*)/#H188L.%G$*F/M)$%1N#48177*$%)0/&#.%G$*F/7)0/&#51810$F3%)0/M)$%)&#
J%):1*6)7O#H)#P)6/#
01#'2(%)3#%&)
91</*7F1%7#$G#48107*)0/8#/%H#I$F<371*#4%-)%11*)%-&#J%):1*6)7K#$G#(8$*)H/#
*($41551)6+%%+--#%1)
9)</*7)F1%7$#H)#.%-1-%1*)/#H188L.%G$*F/M)$%1N#48177*$%)0/&#.%G$*F/7)0/&#51810$F3%)0/M)$%)&#
J%):1*6)7O#H)#P)6/#
7(%&)*$#)
91</*7F1%7#$G#48107*)0/8#/%H#I$F<371*#4%-)%11*)%-&#J%):1*6)7K#$G#(8$*)H/#

2314 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 55, NO. 9, SEPTEMBER 2008
Performance Comparison of Graphene Nanoribbon
FETs With Schottky Contacts and Doped Reservoirs
Youngki Yoon, Student Member, IEEE, Gianluca Fiori, Seokmin Hong,
Giuseppe Iannaccone, Member, IEEE,andJingGuo,Member, IEEE
Abstract—We present an atomistic 3-D simulation study
of the performance of graphene-nanoribbon (GNR) Schottky-
barrier field-effect transistors (SBFETs) and transistors with
doped reservoirs (MOSFETs) by means of the self-consistent so-
lution of the Poisson and Schrödinger equations within the non-
equilibrium Green’s function (NEGF) formalism. Ideal MOSFETs
show slightly better electrical performance for both digital and
terahertz applications. The impact of nonidealities on device per-
formance has been investigated, taking into account the presence
of single vacancy, edge roughness, and ionized impurities along the
channel. In general, MOSFETs show more robust characteristics
than SBFETs. Edge roughness and single-vacancy defect largely
affect the performance of both device types.
Index Terms—Defect, device simulation, graphene field-
effect transistor, graphene nanoribbon, impurity, nonequilibrium
Green’s function (NEGF), quantum transport.
I. INTRODUCTION
I
NTHElastdecade,carbonnanostructureshaveattracted
much attention from the device research community because
its electrical properties make it very appealing for electronic
applications. Carbon nanotubes were demonstrated, for the first
time, by Iijima [1], and from there, huge effort has been directed
to understand the physical properties of the new material and
to exploit its potentials in electronic applications to come after
Moore’s law and ITRS requirements [2]–[4]. Carbon atoms
can not only be combined in the form of tubes of nanoscale
dimensions but can also be arranged in a stable 2-D graphene
sheet [5]–[7]. Electrons in graphene behave as massless fermi-
ons and travel through the lattice with long mean free path, as
shown by the high mobility [5]–[7]. Graphene is a zero-gap
Manuscript received March 24, 2008. The works of Y. Yoon, S. Hong and
J. Guo were supported in part by the Semiconductor Research Corporation
and the Office of Naval Research N000140810861. The works of G. Fiori
and G. Iannaccone were supported in part by the EC Seventh Framework
Program under project GRAND (Contract 215752), by the Network of Excel-
lence NANOSIL (Contract 216171), and by the European Science Foundation
EUROCORES Program Fundamentals of Nanoelectronics, through funds from
CNR and the EC Sixth Framework Program, under project DEWINT (Contract
ERAS-CT-2003-980409). The review of this paper was arranged by Editor M.
Reed.
Y. Yoon and J. Guo are with the Department of Electrical and Computer En-
gineering, University of Florida, Gainesville, FL 32611 USA (e-mail: ykyoon@
ufl.edu).
G. Fiori and G. Iannaccone are with the Dipartimento di Ingegneria
dell’Informazione: Elettronica, Informatica, Telecomunicazioni, Università di
Pisa, 56122 Pisa, Italy (e-mail: g.fiori@iet.unipi.it).
S. Hong was with University of Florida. He is now with Purdue University,
West Lafayette, IN 47907 USA.
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TED.2008.928021
material, with linear dispersion in correspondence of the Fermi
energy, which makes it particularly unsuitable for transistor
applications. However, energy gap can be induced by means
of lateral confinement [8], which is realized, for example, by
etching the graphene sheet in narrow stripes, so-called graphene
nanoribbons (GNRs).
Theoretical works have shown that GNRs have energy gap
which is inversely proportional to their width [9], [10], and
due to their reduced dimensions, edge states play an impor-
tant role, defining non-null energy gap, for all ribbon widths
[11], [12]. GNR field-effect transistors (GNRFETs) have been
fabricated very recently [13]–[15]. GNRFETs that are demon-
strated experimentally to date are realized by connecting the
channel to metals with Schottky contacts [8], [14], therefore
obtaining a Schottky-barrier FET (SBFET). In addition, ohmic
contacts can, in principle, be obtained by heavily doping the
GNR source and drain extensions, which makes device opera-
tion MOSFET-like (therefore, it is referred to as a MOSFET in
the s ubsequent discussion). Because fabrication techniques are
at the very first steps, simulations can represent an important
tool to evaluate device performance. Semiclassical top-of-the-
barrier simulations have been performed [16], [17], whereas
quantum simulations based on a tight-binding approach have
followed [18]–[21] in order to assess device potential. However,
due to the embryonic stage of this new field of research, many
issues still remain unsolved. It is, for example, not clear how
much performance improvement can be obtained by using a
MOSFET device structure, as compared to the Schottky-contact
counterpart, as well as the extent to which nonidealities can
affect device characteristics. State-of-the-art etching techniques
are, for instance, far from atomistic resolution, so that edge
roughness can play an important role on device performance
[22]–[24]. In addition, defects or ionized impurities can repre-
sent elastic scattering centers, which can greatly degrade the
expected fully ballistic behavior.
In this paper, GNR SBFET and MOSFET are numerically
studied in order to establish their potential and the performance
that can be expected if technological challenges are met. The
approach is based on the self-consistent solution of the 3-D
Poisson and Schrödinger equations within the nonequilibrium
Green’s function (NEGF) formalism [25], by means of a real-
space p
z
tight-binding Hamiltonian, in which energy relaxation
at the GNR edges is considered. Different types of nonideali-
ties have been investigated. In particular, we have studied the
effect of a single-vacancy defect, an ionized impurity in the
channel, and edge roughness on the device performance. Doped
source and drain reservoir devices show better performance as
0018-9383/$25.00 © 2008 IEEE

YOON et al.:PERFORMANCECOMPARISONOFGNRFETsWITHSCHOTTKYCONTACTSANDDOPEDRESERVOIRS 2315
Fig. 1. Simulated device structure. (a) SBFET with metal contacts.
(b) MOSFET with doped source and drain extensions. The SiO
2
gate insulator
is 1.5 nm thick with a relative dielectric constant κ = 3.9. N = 12 A-GNR is
used as a channel material, which is 15 nm long and 1.35 nm wide, and the
bandgap is E
g
0.6 eV. The SB height in (a) is a half band gap.
compared to Schottky GNRFETs. Vacancies and edge rough-
ness can greatly affect device electrical performance more than
ionized impurities actually do.
II. A
PPROACH
Device characteristics of GNRFETs are calculated by solving
the Schrödinger equation using the NEGF formalism [25] self-
consistently with the 3-D Poisson equation [18]–[21]. A tight-
binding Hamiltonian with an atomistic p
z
orbital basis set is
used to describe atomistic details of the GNR channel. Coherent
transport is assumed. Simulated device structures are shown
in Fig. 1. The source and the drain are doped extensions of
GNRs in MOSFETs, and metals in SBFETs with SB height
of Φ
Bn
=Φ
Bp
= E
g
/2. Double-gate geometry is used through
1.5-nm SiO
2
gate oxide (κ = 3.9).Foranidealdevicesimula-
tion, perfectly patterned 15-nm-long N = 12 [9] armchair-edge
GNR (A-GNR) is used as a channel material, which has a width
of 1.35 nm and a bandgap of 0.6 eV. Edge bond relaxation
is treated according to ab initio calculation, and a tight-binding
parameter of t
0
= 2.7 eV is used [11]. Power supply voltage
is V
DD
= 0.5 V. Room-temperature (T = 300 K) operation is
assumed.
Nonidealities are treated as follows. Lattice vacancies or
edge roughness are considered as atomistic defects of the
channel GNR, where the existence of carriers is essentially
prohibited. These atomistic vacancies or edge roughness can
be implemented by breaking the nearest bonds (t
0
= 0) in the
device channel Hamiltonian matrix of the perfect lattice accord-
ing to the geometry of the defective lattice. For simplicity, it is
assumed that the topological structure of GNR is not affected by
the defect, which may provide a perturbation to the quantitative
results, but the qualitative conclusions of this paper will not be
changed. An ionized impurity is treated as an external fixed
charge, which can play an important role for the electrostatic
potential of the device. In other words, in the self-consistent
iterative loop between the transport equation and the Poisson
equation, the input charge into the Poisson equation always
includes a fixed external charge as well as the output charge
from the Schrödinger equation.
III. R
ESULTS
A. Ideal Structures
We rst present results for an SBFET and a MOSFET under
ideal conditions. Fig. 2(a) and (b) shows the transfer charac-
teristics for each device. SBFET shows the typical ambipolar
behavior [Fig. 2(a)], so that, for a fair comparison, a common
off current I
o
= 10
7
AisselectedandthattheON state
is defined at V
on
= V
o
+ V
DD
.Then,theoperatingvoltage
ranges are shown by the gray windows in Fig. 2(a) and (b) for
each device. Through the gate work-function tuning, V
o
can be
shifted to V
G
= 0V(V
D
= V
DD
),andthetransfercharacteris-
tics after the work-function engineering are shown in Fig. 2(c):
The MOSFET has 50% larger I
on
(i.e., current for V
G
=
V
DD
and V
D
= V
DD
)andlargertransconductanceg
m
than the
SBFET. This observation agrees to a conclusion in a previous
literature that the on current of a ballistic SBFET with positive
SB height is smaller than that of a ballistic MOSFET due to
the tunneling barrier at the source end of the channel [26].
MOSFETs can have, in addition, a significantly larger maxi-
mum on–off ratio than SBFETs due to the absence of ambipolar
transport, as shown in Fig. 2(d).
Fig. 3(a) shows the output characteristics for V
G
= 0.5 V:
MOSFET shows a better saturation behavior. This is confirmed
by the output conductance g
d
defined as the derivative of the
output characteristic with respect to V
D
.Ascanbeseenin
Fig. 3(b), g
d
in MOSFET is almost half the value found for
SBFET.
We now focus on switching and high-frequency perfor-
mances of GNR devices. In Fig. 4(a), the cutoff frequency f
T
as a function of the applied gate voltage is shown and computed
by using the quasi-static approximation [27] as
f
T
=
g
m
2πC
G
!
!
!
!
V
D
=V
DD
(1)
where g
m
is the transconductance and C
G
is the gate capaci-
tance computed as the derivative of the charge in the channel
with respect to the gate voltage. As can be seen, MOSFET
has 30% higher f
T
as compared to the SBFET counterpart.
For what concerns the intrinsic switching time τ instead, which
represents the typical figure of merit for digital applications, we
have used a previously developed comparison method that takes
into account the power supply,
ON,andOFF states [28]. This
quantity is typically used to estimate the time it takes an inverter
to switch, when its output drives another inverter. Fig. 4(b)
shows the intrinsic delay as a function of on–off ratio: In this
case, MOSFET exhibits 20% faster switching speed than a
middle-bandgap SBFET. The very high cutoff frequency and
the very small delay shown in Fig. 4 are due to the extremely
short channel length (15 nm) and the assumption of purely
ballistic transport. In general, f
T
is inversely proportional to the
channel length, and for longer channel SBFETs, for example,
it can be expressed as f
T
73 GHz/(L
ch
in micrometers) at
the
ON state. In addition, additional parasitic capacitance could
largely affect the estimated f
T
and delay.
B. Atomistic Vacancy
We now focus our attention on the effect of a single-vacancy
defect on device performance. Fig. 5(a) and (b) shows the
transfer characteristics for SBFET and MOSFET, both in the
linear and the logarithmic scale, for different positions of a
defect. All defects are placed in the middle of the channel along
the width direction, whereas three different positions along the

2316 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 55, NO. 9, SEPTEMBER 2008
Fig. 2. I
D
V
G
characteristics of (a) an ideal SBFET and (b) an ideal MOSFET. For a fair comparison between two different devices, the minimal leakage
current I
min
of SBFET is chosen as a common off current I
o
= 10
7
A, and ON state is defined at V
on
= V
o
+ V
DD
,whereV
DD
= 0.5 V is the power
supply voltage. The gray windows in (a) and (b) show the operating voltage ranges of each device. (c) Transfer characteristics of the ideal devices after gate
work-function engineering, by which V
o
can be shifted to V
G
= 0V.AnidealMOSFEThas50%largerI
on
than an ideal SBFET. (d) I
on
versus I
on
/I
o
.
MOSFETs can have a significantly larger on–off ratio than SBFETs.
Fig. 3. (a) I
D
V
D
characteristics at V
G
= V
DD
= 0.5 V. (b) Output conductance g
d
versus V
G
for V
D
= V
DD
.MOSFETshowsbettersaturationbehavior,
which can also be pointed out by smaller g
d
.
Fig. 4. (a) Cutoff frequency f
T
versus V
G
. (b) Intrinsic delay τ versus I
on
/I
o
. MOSFETs can have higher cutoff frequency and smaller intrinsic delay than
SBFETs.

YOON et al.:PERFORMANCECOMPARISONOFGNRFETsWITHSCHOTTKYCONTACTSANDDOPEDRESERVOIRS 2317
Fig. 5. Effect of a lattice vacancy along the transport direction. I
D
V
G
of (a) SBFETs and (b) MOSFETs in the presence of a single lattice vacancy, in
(left axis) a log scale and in (right axis) a linear scale. The lattice vacancy is placed in the middle of the channel width direction and at the different positions along
the transport direction: near the source, in the middle of the channel, and near the drain.
Fig. 6. Conduction band profile along the channel position for (a) SBFETs and (c) MOSFETs in the presence of a lattice vacancy at the ON state. Energy-resolved
current spectrum for (b) the SBFET and (d) the MOSFET in the presence of a vacancy near the source.
propagation direction are considered: In particular, the defect
has been placed near the source, in the middle of the channel,
and near the drain.
As shown in Fig. 5, the defect near the source has the largest
effect in both devices. As compared to the ideal device, the
defect results in 46% and 17% smaller I
on
in SBFET and
MOSFET, respectively. This is because the carrier transport in
the device is totally controlled by the SB at the source end for
SBFETs and by the top of the barrier, which is also located near
the source, for MOSFETs.
The details of the I
on
reduction can be explained by the
reduced quantum transmission and self-consistent electrostatic
effect. For an SBFET with a defect near the source, thicker SB
is induced [Fig. 6(a)] due to the electron accumulation, and
quantum transmission is reduced [Fig. 6(b)] at the
ON state,
which result in a smaller I
on
. When a defect is located at
halfway along the channel or near the drain of an SBFET, the
accumulated electrons lift up the potential barrier and reduce
the energy window of electron injection from the source to the
channel, which results in reduced current with a lattice vacancy.
In case of a MOSFET with a lattice vacancy near the source,
the s elf-consistent potential barrier is not increased, as shown
in Fig. 6(c). Instead, the reduced number of propagating states
due to the lattice vacancy reduces the transmission probability
[Fig. 6(d)], which results in a s maller on current. On the other
hand, defects near the drain and in the middle of the channel do

Citations
More filters
Journal ArticleDOI
09 Jul 2009-Langmuir
TL;DR: Graphene sheets produced through chemical exfoliation of natural graphite flake and hydrazine conversion are found to be hydrophilic and hydrophobic, and isolated graphene layers seem more difficult to wet in comparison to graphite, and low adhesion work was found in the graphene-liquid interface.
Abstract: Graphene sheets were produced through chemical exfoliation of natural graphite flake and hydrazine conversion. Subsequently, graphene sheets were assembled into a thin film, and microscale liquid droplets were placed onto the film surface for measurement of wettability and contact angle. It is found that a graphene oxide sheet is hydrophilic and a graphene sheet is hydrophobic. Isolated graphene layers seem more difficult to wet in comparison to graphite, and low adhesion work was found in the graphene-liquid interface. Approximation of solid-liquid interfacial energy with the equation of state theory was applied to determine the graphene surface energy. The results indicate that surface energy of graphene and graphene oxide is 46.7 and 62.1 mJ/m2, respectively, while natural graphite flake shows a surface free energy of 54.8 mJ/m2 at room temperature. These results will provide valuable guidance for the design and manufacturing of graphene-based biomaterials, medical instruments, structural composites, electronics, and renewable energy devices.

927 citations

Journal ArticleDOI
TL;DR: The I-V characteristics measured at 100, 300, and 400 K indicate that temperature strongly influences the ideality factor of graphene-silicon Schottky diodes, and the optical transparency of the thin graphene layer allows the underlying silicon substrate to absorb incident laser light and generate a photocurrent.
Abstract: We have fabricated graphene-silicon Schottky diodes by depositing mechanically exfoliated graphene on top of silicon substrates. The resulting current–voltage characteristics exhibit rectifying diode behavior with a barrier energy of 0.41 eV on n-type silicon and 0.45 eV on p-type silicon at the room temperature. The I–V characteristics measured at 100, 300, and 400 K indicate that temperature strongly influences the ideality factor of graphene–silicon Schottky diodes. The ideality factor, however, does not depend strongly on the number of graphene layers. The optical transparency of the thin graphene layer allows the underlying silicon substrate to absorb incident laser light and generate a photocurrent. Spatially resolved photocurrent measurements reveal the importance of inhomogeneity and series resistance in the devices.

457 citations

Journal Article
TL;DR: In this article, the authors have fabricated graphene-silicon Schottky diodes by depositing mechanically exfoliated graphene on top of silicon substrates, and the resulting current-voltage characteristics exhibit rectifying diode behavior with a barrier energy of 0.41 eV on n-type silicon and 0.45eV on p-type Silicon at the room temperature.
Abstract: We have fabricated graphene-silicon Schottky diodes by depositing mechanically exfoliated graphene on top of silicon substrates. The resulting current–voltage characteristics exhibit rectifying diode behavior with a barrier energy of 0.41 eV on n-type silicon and 0.45 eV on p-type silicon at the room temperature. The I–V characteristics measured at 100, 300, and 400 K indicate that temperature strongly influences the ideality factor of graphene–silicon Schottky diodes. The ideality factor, however, does not depend strongly on the number of graphene layers. The optical transparency of the thin graphene layer allows the underlying silicon substrate to absorb incident laser light and generate a photocurrent. Spatially resolved photocurrent measurements reveal the importance of inhomogeneity and series resistance in the devices.

423 citations

Journal ArticleDOI
TL;DR: In this article, an ultrathin epitaxial graphite graphite (NPEG) was grown by thermal decomposition on the (0001) surface of 6H-SiC and characterized by surface-science techniques.
Abstract: We have produced ultrathin epitaxial graphite films which show remarkable 2D electron gas (2DEG) behavior. The films, composed of typically 3 graphene sheets, were grown by thermal decomposition on the (0001) surface of 6H-SiC, and characterized by surface-science techniques. The low-temperature conductance spans a range of localization regimes according to the structural state (square resistance 1.5 kOhm to 225 kOhm at 4 K, with positive magnetoconductance). Low resistance samples show characteristics of weak-localization in two dimensions, from which we estimate elastic and inelastic mean free paths. At low field, the Hall resistance is linear up to 4.5 T, which is well-explained by n-type carriers of density 10^{12} cm^{-2} per graphene sheet. The most highly-ordered sample exhibits Shubnikov - de Haas oscillations which correspond to nonlinearities observed in the Hall resistance, indicating a potential new quantum Hall system. We show that the high-mobility films can be patterned via conventional lithographic techniques, and we demonstrate modulation of the film conductance using a top-gate electrode. These key elements suggest electronic device applications based on nano-patterned epitaxial graphene (NPEG), with the potential for large-scale integration.

290 citations

Journal ArticleDOI
TL;DR: In this article, the effects of a model disorder potential (Anderson-type) on the elastic mean free path of carbon nanotubes and graphene nanoribbons are investigated.
Abstract: Two-dimensional graphene, carbon nanotubes, and graphene nanoribbons represent a novel class of low dimensional materials that could serve as building blocks for future carbon-based nanoelectronics. Although these systems share a similar underlying electronic structure, whose exact details depend on confinement effects, crucial differences emerge when disorder comes into play. In this review, we consider the transport properties of these materials, with particular emphasis on the case of graphene nanoribbons. After summarizing the electronic and transport properties of defect-free systems, we focus on the effects of a model disorder potential (Anderson-type), and illustrate how transport properties are sensitive to the underlying symmetry. We provide analytical expressions for the elastic mean free path of carbon nanotubes and graphene nanoribbons, and discuss the onset of weak and strong localization regimes, which are genuinely dependent on the transport dimensionality. We also consider the effects of edge disorder and roughness for graphene nanoribbons in relation to their armchair or zigzag orientation.

261 citations

References
More filters
Journal ArticleDOI
22 Oct 2004-Science
TL;DR: Monocrystalline graphitic films are found to be a two-dimensional semimetal with a tiny overlap between valence and conductance bands and they exhibit a strong ambipolar electric field effect.
Abstract: We describe monocrystalline graphitic films, which are a few atoms thick but are nonetheless stable under ambient conditions, metallic, and of remarkably high quality. The films are found to be a two-dimensional semimetal with a tiny overlap between valence and conductance bands, and they exhibit a strong ambipolar electric field effect such that electrons and holes in concentrations up to 10 13 per square centimeter and with room-temperature mobilities of ∼10,000 square centimeters per volt-second can be induced by applying gate voltage.

55,532 citations


"Performance Comparison of Graphene ..." refers background in this paper

  • ...Electrons in graphene behave as massless fermions and travel through the lattice with long mean free path, as shown by the high mobility [5]–[7]....

    [...]

Journal ArticleDOI
Sumio Iijima1
01 Nov 1991-Nature
TL;DR: Iijima et al. as mentioned in this paper reported the preparation of a new type of finite carbon structure consisting of needle-like tubes, which were produced using an arc-discharge evaporation method similar to that used for fullerene synthesis.
Abstract: THE synthesis of molecular carbon structures in the form of C60 and other fullerenes1 has stimulated intense interest in the structures accessible to graphitic carbon sheets. Here I report the preparation of a new type of finite carbon structure consisting of needle-like tubes. Produced using an arc-discharge evaporation method similar to that used for fullerene synthesis, the needles grow at the negative end of the electrode used for the arc discharge. Electron microscopy reveals that each needle comprises coaxial tubes of graphitic sheets, ranging in number from 2 up to about 50. On each tube the carbon-atom hexagons are arranged in a helical fashion about the needle axis. The helical pitch varies from needle to needle and from tube to tube within a single needle. It appears that this helical structure may aid the growth process. The formation of these needles, ranging from a few to a few tens of nanometres in diameter, suggests that engineering of carbon structures should be possible on scales considerably greater than those relevant to the fullerenes. On 7 November 1991, Sumio Iijima announced in Nature the preparation of nanometre-size, needle-like tubes of carbon — now familiar as 'nanotubes'. Used in microelectronic circuitry and microscopy, and as a tool to test quantum mechanics and model biological systems, nanotubes seem to have unlimited potential.

39,086 citations


"Performance Comparison of Graphene ..." refers background in this paper

  • ...Carbon nanotubes were demonstrated, for the first time, by Iijima [1], and from there, huge effort has been directed to understand the physical properties of the new material and to exploit its potentials in electronic applications to come after Moore’s law and ITRS requirements [2]–[4]....

    [...]

  • ...Carbon nanotubes were demonstrated, for the first time, by Iijima [1], and from there, huge effort has been directed...

    [...]

  • ...REFERENCES [1] S. Iijima, “Helical microtubules of graphitic carbon,” Nature, vol. 354, no. 6348, pp. 56–58, Nov. 1991....

    [...]

Journal ArticleDOI
10 Nov 2005-Nature
TL;DR: In this paper, an experimental investigation of magneto-transport in a high-mobility single layer of Graphene is presented, where an unusual half-integer quantum Hall effect for both electron and hole carriers in graphene is observed.
Abstract: When electrons are confined in two-dimensional materials, quantum-mechanically enhanced transport phenomena such as the quantum Hall effect can be observed. Graphene, consisting of an isolated single atomic layer of graphite, is an ideal realization of such a two-dimensional system. However, its behaviour is expected to differ markedly from the well-studied case of quantum wells in conventional semiconductor interfaces. This difference arises from the unique electronic properties of graphene, which exhibits electron–hole degeneracy and vanishing carrier mass near the point of charge neutrality1,2. Indeed, a distinctive half-integer quantum Hall effect has been predicted3,4,5 theoretically, as has the existence of a non-zero Berry's phase (a geometric quantum phase) of the electron wavefunction—a consequence of the exceptional topology of the graphene band structure6,7. Recent advances in micromechanical extraction and fabrication techniques for graphite structures8,9,10,11,12 now permit such exotic two-dimensional electron systems to be probed experimentally. Here we report an experimental investigation of magneto-transport in a high-mobility single layer of graphene. Adjusting the chemical potential with the use of the electric field effect, we observe an unusual half-integer quantum Hall effect for both electron and hole carriers in graphene. The relevance of Berry's phase to these experiments is confirmed by magneto-oscillations. In addition to their purely scientific interest, these unusual quantum transport phenomena may lead to new applications in carbon-based electronic and magneto-electronic devices.

11,122 citations

Journal Article
TL;DR: An experimental investigation of magneto-transport in a high-mobility single layer of graphene observes an unusual half-integer quantum Hall effect for both electron and hole carriers in graphene.
Abstract: When electrons are confined in two-dimensional materials, quantum-mechanically enhanced transport phenomena such as the quantum Hall effect can be observed. Graphene, consisting of an isolated single atomic layer of graphite, is an ideal realization of such a two-dimensional system. However, its behaviour is expected to differ markedly from the well-studied case of quantum wells in conventional semiconductor interfaces. This difference arises from the unique electronic properties of graphene, which exhibits electron–hole degeneracy and vanishing carrier mass near the point of charge neutrality. Indeed, a distinctive half-integer quantum Hall effect has been predicted theoretically, as has the existence of a non-zero Berry's phase (a geometric quantum phase) of the electron wavefunction—a consequence of the exceptional topology of the graphene band structure. Recent advances in micromechanical extraction and fabrication techniques for graphite structures now permit such exotic two-dimensional electron systems to be probed experimentally. Here we report an experimental investigation of magneto-transport in a high-mobility single layer of graphene. Adjusting the chemical potential with the use of the electric field effect, we observe an unusual half-integer quantum Hall effect for both electron and hole carriers in graphene. The relevance of Berry's phase to these experiments is confirmed by magneto-oscillations. In addition to their purely scientific interest, these unusual quantum transport phenomena may lead to new applications in carbon-based electronic and magneto-electronic devices.

10,112 citations

Journal ArticleDOI
TL;DR: It is found that the energy gap scales inversely with the ribbon width, thus demonstrating the ability to engineer the band gap of graphene nanostructures by lithographic processes.
Abstract: We investigate electronic transport in lithographically patterned graphene ribbon structures where the lateral confinement of charge carriers creates an energy gap near the charge neutrality point. Individual graphene layers are contacted with metal electrodes and patterned into ribbons of varying widths and different crystallographic orientations. The temperature dependent conductance measurements show larger energy gaps opening for narrower ribbons. The sizes of these energy gaps are investigated by measuring the conductance in the nonlinear response regime at low temperatures. We find that the energy gap scales inversely with the ribbon width, thus demonstrating the ability to engineer the band gap of graphene nanostructures by lithographic processes.

4,969 citations


"Performance Comparison of Graphene ..." refers background in this paper

  • ...GNRFETs that are demonstrated experimentally to date are realized by connecting the channel to metals with Schottky contacts [8], [14], therefore...

    [...]

  • ...of lateral confinement [8], which is realized, for example, by etching the graphene sheet in narrow stripes, so-called graphene...

    [...]

Frequently Asked Questions (1)
Q1. What are the contributions mentioned in the paper "Performance comparison of graphene nanoribbon fets with schottky contacts and doped reservoir" ?

The authors present an atomistic 3-D simulation study of the performance of graphene-nanoribbon ( GNR ) Schottkybarrier field-effect transistors ( SBFETs ) and transistors with doped reservoirs ( MOSFETs ) by means of the self-consistent solution of the Poisson and Schrödinger equations within the nonequilibrium Green ’ s function ( NEGF ) formalism.