Functionalized carbon nanotubes and device applications

This article has been downloaded from IOPscience. Please scroll down to see the full text article.
2004 J. Phys.: Condens. Matter 16 R901
(http://iopscience.iop.org/0953-8984/16/29/R01)
Download details:
IP Address: 220.225.244.114
The article was downloaded on 08/01/2011 at 06:41
Please note that terms and conditions apply.
View the table of contents for this issue, or go to the journal homepage for more
Home Search Collections Journals About Contact us My IOPscience
INSTITUTE OF PHYSICS PUBLISHING JOURNAL OF PHYSICS: CONDENSED MATTER
J. Phys.: Condens. Matter 16 (2004) R901–R960 PII: S0953-8984(04)77483-4
TOPICAL REVIEW
Functionalized carbon nanotubes and device
applications
S Ciraci
1
, S Dag
1
, T Yildirim
2
, O G¨ ulseren
1
and R T Senger
1.3
1
Department of Physics, Bilkent University, 06800 Ankara, Turkey
2
NIST Center for Neutron Research, National Institute of Standards and Technology,
Gaithersburg, MD 20899, USA
3
T
¨
UB
˙
ITAK-UEKAE, P.K. 74, 41470 Gebze, Kocaeli, Turkey
E-mail: ciraci@fen.bilkent.edu.tr
Received 8 March 2004
Published 9 July 2004
Online at stacks.iop.org/JPhysCM/16/R901
doi:10.1088/0953-8984/16/29/R01
Abstract
Carbon nanotubes, in which the two-dimensional hexagonal lattice of graphene
is transformed into a quasi-one-dimensional lattice by conserving the local
bond arrangement, provide several structural parameters for engineering novel
physical properties suitable for ultimate miniaturization. Recent interest in
nanoscience and nanotechnology has driven a tremendous research activity in
carbon nanotubes, which has dealt with a variety of problems and produced
a number of new results. Most of the effort has gone into revealing various
physical properties of nanotubes and functionalizing them in different ways.
This paper covers a narrow region in this enormous research field and reviews
only a limited number of recent studies which fit within its scope. First,
we examine selected physical properties of bare carbon nanotubes, and then
study how the mechanical and electronic properties of different tubes can be
modified by radial strain, structural defects and adsorption of foreign atoms and
molecules. Magnetization of carbon nanotubes by foreign atom adsorption has
been of particular interest. Finally, we discuss specific device models as well
as fabricated devices which exploit various properties of carbon nanotubes.
(Some figures in this article are in colour only in the electronic version)
Contents
1. Introduction 902
2. Atomic and electronic structure 904
2.1. Geometric structure and energetics of bare SWNTs 904
2.2. Energy band structure 906
3. Hydrogenation of carbon nanotubes 908
0953-8984/04/290901+60$30.00 © 2004 IOP Publishing Ltd Printed in the UK R901
R902 Topical Review
4. Oxygenation of carbon nanotubes 912
4.1. Physisorption of O
2
molecules 913
4.2. Chemisorption of oxygen atoms 916
5. Adsorption of individual atoms on SWNTs 917
5.1. Binding geometry and binding energies 919
5.2. Electronic structure 923
5.3. Transition element covered or filled SWNTs 926
6. Radial deformation of carbon nanotubes 927
6.1. Elasticity 928
6.2. Effect on the electronic structure 930
6.3. Effect on the chemical reactivity 932
6.4. Effect of pressure on nanoropes 934
7. Devices based on carbon nanotubes 936
7.1. Ab initio methods in transport calculations 936
7.2. Device models 942
8. Devices fabricated using carbon nanotubes 952
8.1. Transistors based on carbon nanotubes 952
8.2. Chemical sensors 955
9. Conclusions 955
Acknowledgments 956
References 956
1. Introduction
Research on carbon nanotubes is ever intensifying in diverse fields of science and engineering
in spite of the twelve years that have passed since its first discovery by Iijima [1]. There are
several reasons that so much interest has been focused on these materials. First of all, carbon
nanotubes have been a natural curb for several research programmes, which were tuned to
C
60
but all of a sudden came to an end without any great technological applications having
been found. Secondly, researchers, who can touch and relocate individual atoms have been
challenged to discover the novel properties of these strange materials in order to transform
them into new devices or use them in other technological applications. As a result of the rapid
rise in the speed, as well as the rapid reduction in the size of electronic devices, newparadigms
will be needed to overcome the barriers set by the traditional technologies to produce ever
smaller and faster devices. Extensive research dealing with the modification of electronic
structure for desired device operations has indicated that carbon nanotubes can be considered
as a new frontier in the search for the ultimate miniaturization of electronic circuits with
ultrahigh density components and new functionalities. Several devices fabricated so far with
different functionalities appear to meet the great expectations for carbon nanotubes.
Carbon nanotubes are unique materials, which offer a variety of structural parameters for
engineering their physical and chemical properties [2, 3]. They can be synthesized as single-
wall (SWNT) or multiple-wall (MWNT) nanotubes; they can form ropes or even crystals.
Even an ultimate one-dimensional carbon chain at the centre of a MWNT (and stabilized by
the innermost nanotube) has been discovered in cathode deposits [4]. SWNTs are basically
rolled graphite sheets, which are characterized by two integers (n. m) defining the rolling (or
chiral vector) C = na
1
+ ma
2
, in terms of the two-dimensional (2D) hexagonal Bravais lattice
vectors of graphene, a
1
and a
2
. Then the radius of the tube is given in terms of (n. m) by the
relation R = a
0
(n
2
+ m
2
+ nm)
1¡2
¡2π, where a
0
= [a
1
[ = [a
2
[. SWNTs exhibit different
electronic structures depending on n and m (i.e. on their chirality and radius).
Topical Review R903
The mechanical properties of carbon nanotubes are striking. They are flexible and can
sustain large elastic deformations radially; at the same time they are very strong axially with
high yield strength [5, 6]. Their strength far exceeds that of any other fibre. Even more
striking is the response of the electronic structure to the radial deformation leading to dramatic
changes. As has been predicted theoretically and confirmed experimentally, a semiconducting
zigzag tube becomes metallic with finite state density at the Fermi level as a result of radial
deformation transforming the circular cross section into an ellipse. At the same time the
chemical activity of the surface of the tube undergoes a change; the interaction of adatoms
with the SWNT occurs differently at high and low curvature sites. The metal–semiconductor
transition induced by elastic deformation has important implications.
Physical and chemical properties of a SWNT can also be modified by the adsorption
of foreign atoms or molecules. This process is usually named functionalization, and
carries great potential in tailoring new nanostructures for engineering them according to a
desired application. For example, depending on the pattern of hydrogen atom coverage,
while a metallic armchair SWNT can be transformed to a wide band gap semiconductor, a
semiconducting zigzag tube may become a metal with very high state density. A free SWNT,
which is normally nonmagnetic, becomes magnetic with unpaired spins upon the adsorption
of oxygen molecules or specific transition metal atoms. A recent study demonstrates that a
semiconducting zigzag tube becomes both a magnetic and a high conductance wire as a result of
Ti coating [7]. Aselectable functionalization of the (5, 5) SWNTresulting fromCH
n
(n = 1–3)
adsorption and decoration gives rise to a substantial change in the density of states [8].
Suitably doped carbon nanotubes can be functionalized by selectively forming chemical
bonds with ligands at the chemically active impurity site [9]. Apparently, functionalization
of carbon nanotubes, in particular biological and chemical functionalization, is an extensive
field incorporating several recent studies in diverse fields. This reviewis confined rather to the
functionalized SWNTs which have their electronic and magnetic properties modified.
Clearly, carbon nanotubes offer many parameters to deal with and many options for
generating properties suitable for a desired functionality. One of the great challenges of
research on carbon nanotubes has been the realization of nanometre-sized optoelectronic
devices and nanomagnets. In an effort to discover newfeatures of technological interest, several
theoretical and experimental studies actively explored SWNTs, MWNTs, ropes and their
functionalized forms, which resulted in many papers. However, due to its focus compounded
with the space limitations, the scope of this review article has been kept necessarily limited.
The subject matter that we have left out is in no way less significant than what we have included.
We followed a logical order that starts from fundamental aspects and ends with
technological applications. We first established a background concerning the atomic and
electronic structure of various SWNTs. We then examined various methods which are used
to modify the properties of SWNTs to generate new nanostructures. Finally, we discussed
how these properties have been used to model devices. The organization of the article is as
follows. In section 2, we have presented a discussion of electronic structure obtained fromfirst-
principles calculations together with a comparison made to empirical studies. Hydrogenation
and oxygenation of SWNTs have been dealt separately in sections 3 and 4, respectively, owing
to the several recent papers in this area, as well as due to the relevance to hydrogen storage.
Section 5 has been be devoted to the individual adsorption of 24 different atoms (ranging from
alkali and simple metal atoms to group IV atoms and most transition metal atoms), where their
binding structures and binding energies, and the effect of their adsorption on the electronic
structure, have been investigated. Since the ground state for most of the transition metal atoms
adsorbed on the surface of SWNTs is magnetic, and hence has net spin, this section is important
for the magnetic properties of functionalized nanotubes. In section 6, we have investigated the
R904 Topical Review
effects of radial deformation, and have examined how the electronic and chemical properties
are modified. Here, the atomic structure, elasticity and electronic structure and binding of
adatoms under radial deformation have been discussed. In section 7, we have reviewed recent
developments in transport theory and have presented some device models developed using
the features and physical properties discussed in previous sections. Section 8 reviews briefly
the recent progress made in the fabrication of electronic devices based on carbon nanotubes
and has described a few such devices produced. The paper is concluded with outlooks and
prospective developments.
2. Atomic and electronic structure
The electronic band structure of SWNTs can be deduced by mapping the band structure of
graphene in a 2D hexagonal lattice onto a cylinder [3, 10–14]. In this respect, the SWNT
presents an interesting example, in which dimensionality is reduced from two to one. The
analysis based on the band folding indicates that the (n. n) armchair nanotubes are always
metal with π

conduction, and π valence bands crossing at the Fermi level, and exhibit 1D
quantum conduction [14, 15]. The (n. 0) zigzag SWNTs are generally semiconductor and
are only metal if n is an integer multiple of 3. Although the overall electronic structure of
SWNTs has been described by this simple picture, recent studies [16, 18] have shown much
more complicated structural dependence. For example, the (9. 0) tube is, in fact, a small band
gap semiconductor. The semiconducting behaviour of SWNTs has been of particular interest,
since the electronic properties can be controlled by doping or implementing defects in nanotube
based optoelectronic devices [19–26].
Band calculations of SWNTs were initially performed by using a one-band π orbital
tight binding model [11]. Subsequently, experimental data [27–30] on the band gaps were
extrapolated to confirm the inverse proportionality with the radius of the nanotube [13].
Later, first-principles calculation [31] within the local density approximation (LDA) showed
that the σ

–π

hybridization becomes significant at small R (or at high curvature). Recent
analytical studies [32–34] showed the importance of curvature effects in carbon nanotubes.
Nonetheless, band calculations performed by using different methods have been at variance
on the values of the band gap. Extensive theoretical analysis of the band structure of SWNTs
together with the curvature effects on geometric and electronic structure has been carried out
recently [18] by using first-principles pseudopotential plane wave method [35] calculations
within the generalized gradient approximation (GGA) [36].
2.1. Geometric structure and energetics of bare SWNTs
Because of cylindrical symmetry, the structural parameters of SWNTs deviate from those of
graphene. The inset to figure 1 shows a schematic side viewof a zigzag SWNT which indicates
two types of C–C bonds (d
1
and d
2
) and C–C–C bond angles (O
1
and O
2
). The variations of
the normalized bond lengths (i.e. d
1
¡d
0
and d
2
¡d
0
where d
0
is the optimized C–C bond length
in graphene) and the bond angles with tube radius R (or n) are shown in figures 1(a) and (b),
respectively. Both the bond lengths and the bond angles display a monotonic variation and
approach the graphene values as the radius increases. As pointed out earlier for the armchair
SWNTs [37], the curvature effects, however, become significant at small radii. The zigzag
bond angle (θ
1
) decreases with decreasing radius. It is about 12

less than 120

for the (4. 0)
SWNT, while the length of the corresponding zigzag bonds (d
2
) increases with decreasing R
and the length of the parallel bond (d
1
) decreases to a lesser extent with decreasing R. The
angle involving this latter bond (θ
2
) is almost constant.
Topical Review R905
Figure 1. Inset: a schematic side view of a zigzag SWNT, indicating two types of C–C bonds and
C–C–C bond angles. These are labelled as d
1
, d
2
, θ
1
and θ
2
. (a) Normalized bond lengths (d
1
¡d
0
and d
2
¡d
0
) versus the tube radius R (d
0
= 1.41 Å). (b) The bond angles (θ
1
and θ
2
) versus R.
(c) The curvature energy, E
cur
per carbon atom with respect to graphene as a function of the tube
radius. The solid curves are the fit to the data as α¡R
2
. (Reproduced from [18].)
An internal strain is implemented upon the formation of tubular structure fromthe graphene
sheet. The associated strain energy, which is specified as the curvature energy, E
cur
, is
calculated as the difference of total energy per carbon atom between the bare SWNT and
the graphene (i.e. E
cur
= E
T.SWNT
− E
T.graphene
) for 4 n 15. The calculated curvature
energies are shown in figure 1(c). As expected, E
cur
is positive and increases with increasing
curvature. The cohesive energies of SWNTs are also curvature dependent, and are calculated
fromthe expression E
coh
= E
T
[C] −E
T
[SWNT] in terms of the total energy of the free carbon
atom, and the total energy of a SWNT per carbon atom. For a zigzag tube, it is small for small
n and increases with n, and eventually saturates to the cohesive energy of graphene. Similar
trends also exist for the armchair tubes. In classical theory of elasticity the curvature energy
is given by the following expression [38–40]: E
cur
= α¡R
2
, where α = Yt
3
O¡24. Here Y is
the Young’s modulus, t is the thickness of the tube, and O is the atomic volume. Interestingly,
curvature energies obtained from first-principles calculations yield a perfect fit to the relation
α¡R
2
as seen in figure 1(c). In this fit α is found to be 2.14 eV Å
2
¡atom [18].
R906 Topical Review
Figure 2. A map of chiral vectors that determine the chirality of SWNTs. Each vector is specified
by the (n. m) indices. M, S and C denote metals, semiconductors and curvature induced small gap
semiconducting SWNTs, respectively. (Reproduced from [17].)
Table 1. Band gap, E
g
, as a function of radius R for (n. 0) zigzag nanotubes. M denotes the
metallic state. The first-row values were obtained within the GGA in [18]. The second and third
rows from [31] give LDA results, while all the rest are tight binding (TB) results. The two rows
from [42] are for two different TB parametrizations.
n 4 5 6 7 8 9 10 11 12 13 14 15
R (Å) 1.66 2.02 2.39 2.76 3.14 3.52 3.91 4.30 4.69 5.07 5.45 5.84
[18] M M M 0.243 0.643 0.093 0.764 0.939 0.078 0.625 0.736 0.028
[31] M 0.09 0.62 0.17
[31] 0.05 1.04 1.19 0.07
[11] 0.21 1.0 1.22 0.045 0.86 0.89 0.008 0.697 0.7 0.0
[42] 0.79 1.12 0.65 0.80
[42] 1.11 1.33 0.87 0.96
2.2. Energy band structure
According to the zone folding scheme a (n. m) SWNT has been predicted to be metallic when
n −m = 3integer, since the doubly degenerate π and π

states, which overlap at the Kpoint
of the hexagonal Brillouin zone (BZ) of graphene, fold to the I point of the tube [11, 12].
Thus, to first order all (n. n) armchair tubes are metallic, but (n. 0) zigzag tubes become
metallic when n is a multiple of 3. In figure 2, a map of the chiral vector specified by (n. m)
indices indicates whether a SWNT is a semiconductor or metal. This simple picture provides
a qualitative understanding, but fails to describe some important features, in particular for
small radius or metallic nanotubes. SWNTs specified by ‘C’ in figure 2 form a subgroup of
semiconducting tubes, which have a small band gap induced by the curvature. This situation
is clearly shown in table 1, where the band gaps E
g
of (n. 0) SWNTs calculated by different
methods are compared.
First-principles GGA calculations [18] resulted in small, but nonzero energy band gaps
of 93, 78 and 28 meV for (9. 0), (12. 0) and (15. 0) SWNTs, respectively. These gaps
Topical Review R907
Figure 3. (a) Energies of the doubly degenerate π states (VB), the doubly degenerate π

states
(CB) and the singlet π

state as a function of nanotube radius. Each data point corresponds to
n ranging from 4 to 15 consecutively. (b) The calculated band gaps of [18] are shown by filled
symbols. Solid (dashed) curves are the plots of equation (2) (equation (1)). The experimental data
taken from [16, 29, 30] are shown by open diamonds. (Reproduced from [18].)
are measured by scanning tunnelling spectroscopy (STS) experiments [16] as 80, 42 and
29 meV, in the same order. Recently, Kim et al [17] synthesized ultralong and high percentage
semiconducting SWNTs where further experimental evidence for the small band gap tubes
was provided. The biggest discrepancy noted in table 1 is between the tight binding and the
first-principles values of the gaps for small radius tubes such as (7. 0). These results indicate
that curvature effects are important and the simple zone folding picture has to be improved.
Moreover, the analysis of the LDA bands of the (6. 0) SWNT calculated by Blase et al [31]
brought in another important effect of the curvature. The antibonding singlet π

and σ

states
mix and repel each other in curved graphene. As a result, the purely π

state of planar graphene
is lowered with increasing curvature. For zigzag SWNTs, the energy of this singlet π

state is
shifted downward with increasing curvature.
In figure 3(a), the doubly degenerate π states (which are the valence band edge at the I
point), the doubly degenerate π

states (which become the conduction band edge at I for large
R) and the singlet π

state (which is in the conduction band for large R) are shown. As seen,
the shift of the singlet π

state is curvature dependent, and below a certain radius determines
the band gap. For tubes with radius greater than 3.3 Å (i.e. n > 8), the energy of the singlet π

state at the I point of the BZ is above the doubly degenerate π

states (i.e. the bottom of the
R908 Topical Review
conduction band), while it falls between the valence and conduction band edges for n = 7. 8,
and eventually dips even below the doubly degenerate valence band π states for the zigzag
SWNT with radius less than 2.7 Å (i.e. n - 7). Therefore, all the zigzag tubes with radius less
than 2.7 Å are metallic. For n = 7. 8, the edge of the conduction band is set by the singlet π

state, but not by the doubly degenerate π

state. The band gap derived from the zone folding
scheme is reduced by the shift of this singlet π

state as a result of curvature induced σ

–π

mixing. This explains why the tight binding calculations predict band gaps around 1 eV for
n = 7. 8 tubes, while the self-consistent calculations predict much smaller values. Based on
π orbital tight binding model, it was proposed [13] that E
g
behaves as
E
g
= γ
0
d
0
R
. (1)
which is independent of the helicity. Within the simple π orbital tight binding model, γ
0
is taken
to be equal to the hopping matrix element V
ppπ
. d
0
is the bond length in graphene. However,
as seen in figure 3(b), the band gap displays a rather oscillatory behaviour. The relation given
in equation (1) was obtained by a second-order Taylor expansion of one-electron eigenvalues
of the π orbital tight binding model [13] around the K point of the BZ, and hence it fails to
represent the effect of the helicity. By extending the Taylor expansion to the next higher order,
Yorikawa and Muramatsu [41, 42] included another term in the empirical expression of the
band gap variation,
E
g
= V
ppπ
d
0
R

1 + (−1)
p
γ cos(3θ)
d
0
R

. (2)
which depends on the chiral angle, θ, as well as an index p. Here γ is a constant and the
index p is defined as the integer from k = n − 2m = 3q + p. The factor (−1)
p
comes from
the fact that the allowed k is nearest to either the K or K
/
point of the hexagonal Brillouin
zone. For zigzag nanotubes θ = 0, and hence the solid curves in figure 3(b) are fits to the
empirical expression E
g
= V
ppπ
d
0
¡R ± V
ppπ
γ d
2
0
¡R
2
, obtained from equation (2) by using
the parameters V
ppπ
= 2.53 eV and γ = 0.43. The agreement between the first-principles
calculations [18] and the STS data [29, 30] is very good considering the fact that there might
be some uncertainties in identifying the nanotube in the experiment.
The situation displayed in figure 3 indicates that the variation of the band gap with the
radius is not simply 1¡R, but additional terms incorporating the chirality dependence are
required. Most importantly, the mixing of the singlet π

state with the singlet σ

state due to
the curvature and its shift towards the valence band with increasing curvature are not included
in either the π orbital tight binding model or the empirical relations expressed by equations (1)
and (2). This behaviour of the singlet π

states is of particular importance for the applied
radial deformation that modifies the curvature and in turn induces metallization [24, 43, 44]
as discussed in sections 6.2 and 7.2.
3. Hydrogenation of carbon nanotubes
Owing to their large effective surface area, carbon nanotubes can be suitable for hydrogen
storage. Several studies have explored this capacity of carbon nanotubes [45–51], but the
results reported to date have been conflicting. Diverse hydrogen storage densities ranging from
0.4 to 10 wt% have been reported. Theories based on physisorption have failed to predict high
hydrogen uptake [52]. Recently, hydrogen adsorption has been the subject of various studies
predicting interesting features [53–55, 57, 58]. Tada et al [54] examined the dissociative
adsorption of a hydrogen molecule on various SWNTs. They found that the potential barrier
height for the dissociative adsorption of a hydrogen molecule onto an outer wall decreases as
Topical Review R909
the tube diameter decreases. Andriotis et al [58] calculated the I –V characteristics of finite
(5. 5) and (10. 0) SWNTs by using the Landauer expression [59] within the tight binding
Green function method. They found extreme sensitivity to hydrogen adsorption. On the basis
of first-principles calculations, Chan et al [60] proposed a mechanism for the dissociative
chemisorption of hydrogen molecules on SWNTs in the solid (rope) phase under external
pressure. They argued that the process is reversible upon the release of external pressure.
Extensive analysis of hydrogenchemisorptionon SWNTs has been carried out by Yildirim
et al [55] using the first-principles pseudopotential plane wave method. They calculated the
average binding energy E
b
= (E
T
[SWNT] + 4nE
T
[H] − E
T
[H + SWNT])¡4n, in terms of
the total energies of a fully relaxed bare SWNT, the single hydrogen atom H and a fully
exohydrogenated SWNT, respectively. They found that E
b
increases with decreasing radius
of the exohydrogenated SWNT, R
HC
, monotonically. The binding energies can be very
well described by a one-parameter fit, E
b
= E
0
+ C(n. m)¡R
HC
. Here E
0
= 1.73 eV;
C(n. n) = 4.45 eV and C(n. 0) = 4.62 eV. Interestingly, a dodecahedron and C
60
H
60
have
higher average binding energies for adsorbed hydrogen atom. Also the binding energies
of zigzag tubes are always lower than those of armchair tubes with equal radius by about
30 meV¡atom.
Although the exohydrogenated SWNT (C
4n
H
4n
) is always energetically favourable with
respect to a bare SWNTand 4n Hatoms for all values of the radius, it is of interest to see whether
they are also stable against breaking a single C–H bond (or desorption of a single H atom).
Yildirim et al [55] revealed that for R
HC
> 6.25 Å the required energy becomes negative and
hence the system becomes unstable, in spite of the fact that there might be an energy barrier
preventing the desorption of H atoms. On the other hand, the adsorption of a single H atom to
a bare SWNT is always exothermic no matter what the value of R is. However, as discussed
extensively in section 6.3 the amount of energy gained by the adsorption of a single H atom
decreases as R increases.
Important effects of hydrogenation have been further examined for different isomers at
different hydrogen coverage [57]. At full coverage, these are full exohydrogenationand endo–
exohydrogenationwhere each carbon atomis bonded to a hydrogen alternately frominside and
outside. For half-coverage, i.e. O = 0.5, denoted by C
4n
H
2n
, there are three isomers, namely
(i) a uniform pattern, where every other carbon atom is bonded to a hydrogen atom from
outside, (ii) a chain pattern, where every other carbon zigzag chain is saturated by hydrogen,
and (iii) a dimer pattern where every other carbon dimer rowperpendicular to the zigzag carbon
chains is saturated by hydrogen as described in figure 4(a).
It is found that geometric and electronic structures and binding energies of hydrogenated
SWNTs strongly depend on the pattern of hydrogenation (i.e. decoration). The most
remarkable effect is obtained when zigzag nanotubes are uniformly exohydrogenated at
O = 0.5. Upon hydrogenation the structure undergoes a massive reconstruction, whereby
the circular cross section of the (7. 0) SWNT changes to a rectangular one, and those of (8. 0),
(9. 0), (10. 0) and (12. 0) change to square ones as shown in figure 4(b). These new structures
are stabilized by the formation of newdiamond-like C–Cbonds with d
C−C
∼ 1.51–1.63 Ånear
the corners of rectangular or square C
4n
H
2n
. Hence, triangular and pentagonal C rings are
formed instead of hexagonal ones. Depending on 2n mod 4, either one bond is formed just at
the corners or two bonds are formed at either side of the corners. Most interestingly, all these
structures are predicted to be metallic with a high density of states at the Fermi level [57]. The
uniform adsorption at O = 0.5 is metastable for zigzag nanotubes. Such a local minimum
does not exist for armchair nanotubes, since uniformly adsorbed H atoms are rearranged upon
relaxation by concerted exchange of C–H bonds to form zigzag chains along the tube axis.
The cross sections of the chain isomer at O = 0.5 for armchair tubes are polygonal where
R910 Topical Review
zigzag
a
r
m
c
h
a
i
r
uniform chain dimer
Figure 4. (a) A view of three different isomers of hydrogenated (n. 0) SWNT at half-coverage
(with formula unit C
4n
H
2n
). The left and up arrows indicate the tube axis for armchair and zigzag
nanotubes, respectively. Carbon atoms which are bonded to hydrogens are indicated by dark
colour. (b) A side and top view of a (12. 0)C
48
H
24
showing the square cross section of a uniformly
exohydrogenated nanotube at half-coverage. (Reproduced from [57].)
the corners are pinned by the zigzag H chains along the tube axis. For other isomers at half-
coverage as well as exohydrogenationsand endo–exohydrogenationsat full coverage, the cross
sections remain quasi-circular (see figure 5).
The average binding energies are plotted as a function of R in figure 5, and fitted
to a general expression, E
b
= E
0
(O) +
C
p
(O)
R
p
. A special case of this fit was employed
above for exohydrogenated tubes using R
HC
(i.e. expanded radius upon hydrogenation).
Note that the 1¡R form (i.e. p = 1) is quite common for SWNTs and scales various
properties [16, 18, 26, 27, 55, 61]. Here C
p
(O) is a constant that depends on coverage
O, and represents the curvature effect. The values of the exponent p and C
p
are determined
for different isomers [57]. We note that while E
b
increases with decreasing R in the case of
exohydrogenation, this trend is reversed for endo–exohydrogenation due to increased H–H
repulsion inside the tube at small R. Nevertheless, the endo–exohydrogenation of SWNTs,
which transforms the sp
2
to sp
3
-like bonding, gives rise to the highest binding energy saturating
at 3.51 eV as R → ∞. At this limit, the exo–endohydrogenated graphene (from above and
below) is buckled by 0.46 Å as if two diamond (111) planes had an interplanar distance of
0.50 Å.
The effect of exohydrogenation on the electronic structure is remarkable. The band gaps
of (n. 0) zigzag tubes have increased approximatelyby 2 eVupon full exohydrogenation. Even
more interesting is that the exohydrogenated metallic armchair tube becomes semiconducting
with a band gap larger than the zigzag one for comparable radius. As for the band gap
variation with R, E
g
(R) for exohydrogenated tubes at O = 1 displays a behaviour similar to
that of E
b
(R) for both types of SWNTs and hence decreases with increasing R (see figure 6).
Relatively larger band gaps (in the range of 3.5–4 eV) of endo–exohydrogenated SWNTs
(O = 1) can be explained by the fact that the adsorption of H alternately inside and outside
Topical Review R911
Figure 5. Average binding energies, E
b
, of hydrogen atoms adsorbed on various zigzag and
armchair SWNTs versus bare tube radius R. Filled and open symbols are for zigzag and
armchair nanotubes, respectively. Circles and diamonds are for exohydrogenation and endo–
exohydrogenation at full coverage, respectively. The filled squares show the zigzag nanotubes
uniformly exohydrogenated at half coverage. The chain and dimer patterns of adsorbed hydrogen
atoms at half coverage are shown by down and up triangles, respectively. Curves are analytical fits
explained in the text. Insets show top views of several C
4n
H
2n
isomers. (Reproduced from [57].)
leads to an atomic configuration closer to the diamond structure having a rather large band gap
(E
g
= 5.4 eV).
The effect of hydrogenation on the electronic structure is even more interesting at O= 0.5.
Depending on the pattern of hydrogen adsorption, an isomer can be either a metal or insulator.
For example, all uniformC
4n
H
2n
are metallic. On the other hand, the chain pattern realized on
the (n. 0) SWNTs results in two doubly degenerate, almost dispersionless states at the valence
and conduction band edges. The band gap E
g
between these states decreases with increasing
R. When n is odd, E
g
is large (e.g. E
g
= 2.1 eV for (7. 0)). When n is even, the doubly
degenerate band at the conduction band edge moves towards the valence band edge and splits
into bonding and antibonding states. As a result E
g
is reduced significantly, becoming only
a pseudogap for large and even n values. Finally, the dimer row isomers are insulators and
E
g
increases with increasing radius. Surprisingly, there are two dispersive bands with ∼1 eV
bandwidth at both band edges and the extremum moves from the centre of the Brillouin zone
(I point) to the zone edge (Z point).
Hydrogen adsorption induced dramatic changes of the electronic structure are
demonstrated by the total density of states (DOS), D(E), of (9. 0) nanotubes in figure 7. First
of all, the small band gap of the bare (9. 0) SWNT is opened by ∼2 eVupon exohydrogenation
at O = 1. The band gap is still significant for O = 0.5, having the chain pattern, and increases
R912 Topical Review
Figure 6. The band gaps, E
g
, versus the bare nanotube radius R. Filled and empty symbols
indicate zigzag and armchair SWNTs, respectively. Squares show nonmonotonic variation of the
band gap of the bare zigzag nanotubes. Exohydrogenated and endo–exohydrogenated nanotubes
(O = 1) and chain and row patterns of adsorbed hydrogen atoms at half-coverage are shown by
circles, diamonds, down and up triangles, respectively. Curves are guides to the eye. (Reproduced
from [57].)
to 4 eV for the dimer pattern. However, a similar chain pattern of C
32
H
16
in figure 7 has
a much smaller band gap. Surprisingly, all zigzag nanotubes uniformly exohydrogenated at
O = 0.5 are metals. As displayed in the sixth panel of figure 7 for C
36
H
18
(uniform), their
total densities of states are characterized by a peak yielding a high state density at E
F
. While
carbon states are pushed apart, yielding a ∼4–5 eV gap, a new dispersive metallic band with
∼1–2 eV bandwidth crosses the Fermi level. Apart from this being an ideal 1D conductor,
the very high density of states at E
F
might lead to superconductivity. Note that these uniform
C
4n
H
2n
undergo a massive reconstruction and their circular cross sections change into square
ones with the formation of new C–C bonds at the corners. All C atoms without attached H
(except those at corners) as well as the H atoms at the centre of four planar sides contribute
to the high D(E
F
). In this way four individual conduction paths are formed on each side of a
square tube. It is emphasized that the transformation fromthe sp
2
to the sp
3
bonding underlies
the above effects. For C
4n
H
2n
, especially endo–exohydrogenated SWNTs, C
4n
H
2n
H
2n
, the
structures can be conceived as if they are more diamond-like than graphitic. This argument
is justified by the comparison of D(E) for the endo–exohydrogenated (9. 0), i.e. C
36
H
18
H
18
,
with that of bulk diamond in figure 7. Apart from opening a large band gap, the quasi-metallic
D(E) of the bare (9. 0) is modified to become similar to that of bulk diamond. The latter has
a relatively larger valence bandwidth due to the coupling of distant neighbours.
4. Oxygenation of carbon nanotubes
Remarkable effects on the electrical resistance of a semiconducting single-wall carbon
nanotube (s-SWNT) upon exposure to gaseous molecules such as NO
2
and NH
3
, have been
reported [62]. Collins et al [63] found similar effects for oxygen. Exposure to air or oxygen
influences electrical resistance and the thermoelectric power of a s-SWNT, which can be
converted to a good metal and hence its electronic properties can be reversibly modified by a
Topical Review R913
Figure 7. Comparison of the electronic density of states (DOS) of a bare (9. 0) nanotube (C
36
)
and its various hydrogenated isomers. The DOS of hydrogenated (8. 0) at O = 0.5 (i.e. C
32
H
16
in a chain pattern) and bulk diamond (bottom panel) are also shown for comparison. The zero of
energy is taken at the Fermi energy shown by a vertical dashed line. (Reproduced from [57].)
surprisingly small concentration of adsorbedoxygen. Experimental studies on carbonnanotube
field emitters have shown that the adsorption of ambient gases, in particular O
2
, instantaneously
induces a significant increase in the emission current [64]. In addition to functionalization,
oxygenation is involved in other applications. For example, carbon nanotubes synthesized
by using arc discharge are purified from other undesired, carbon based nanoparticles through
oxidation. At elevated temperatures, oxygen undergoes chemical reaction preferably with
the strained C–C bonds and eliminates carbonaceous nanoparticles as well as the caps of
nanotubes [65–67].
4.1. Physisorption of O
2
molecules
Observed effects on the electronic structure of SWNTs due to O
2
physisorption have been
subject to recent theoretical investigations based on first-principles calculations [68–77]. Spin
unpolarized band structure calculations based on the local density approximation (LDA)
predicted that the semiconducting (8. 0) tube becomes metallic, since the valence band
is hole doped by the Fermi level touching the top of the valence band as a result of O
2
physisorption [68]. The analysis based on the local spin density approximation (LSDA) has
indicated that the physisorbed O
2
favours the triplet state. While the spin up states are fully
occupied, the spin down states are nearly empty and hence give rise to a finite density of states
at the Fermi level [68].
R914 Topical Review
Figure 8. A schematic illustration of the physisorption sites of O
2
molecules on the (8. 0) SWNT.
The GGAoptimized distance from one O atom of the molecule to the nearest C atom of the SWNT
is denoted by d
C−O
. E
s
is the GGA chemical bonding energy for the spin polarized triplet state.
E
b
is the binding energy including the van der Waals interaction. (Reproduced from [75].)
The energetics of oxygen adsorption on the surface of graphite [78] and on an (8. 0) SWNT
were calculated for selected sites [69]. Adsorption and desorption of an oxygen molecule and
various precursor states at the edges of finite size armchair (5. 5) and zigzag (9. 0) SWNTs
were studied to provide an understanding of the oxidative etching process [70]. Similarly, the
mechanism of the oxidative etching of the caps and walls of the small radius (5. 5) armchair
SWNT was investigated [71]. The effects of oxygen adsorption on the field emission from
carbon nanotubes were treated by using an ab initio approach [72]. The dynamics of the thermal
collision of the O atom with a (6. 0) SWNT was simulated by ab initio calculations [73].
A detailed analysis of the physisorption of oxygen molecules based on first principles and
fully relaxed, spin polarized GGA calculations including the long range interactions and the
chemisorption of oxygen atom on the (8. 0) and (6. 6) SWNTs has been carried out by Dag
et al [75, 77] with the aim of revealing how the adsorptions of O
2
and O modify the electronic
properties. The zigzag (8. 0) and armchair (6. 6) tubes are taken as prototypes for the s-SWNT
and metallic m-SWNT, respectively. Similar trends are expected to occur in other tubes with
curvature effects being emphasized at small R [18]. The bonding of O
2
has been studied by
placing the molecule at different sites on the SWNT as shown in figure 8.
The binding energy involves short range chemical interaction and long range van der
Waals interaction, i.e. E
b
= E
s
+ E
vdW
. Dag et al [75] have calculated the contribution of the
short range interaction (i.e. chemical bonding energy) by using the expression
E
s
= E
T
[SWNT] + E
T
[O
2
] − E
T
[O
2
+ SWNT]. (3)
in terms of the GGA spin polarized total energies of the fully optimized bare SWNT
(E
T
[SWNT]), the molecule (E
T
[O
2
]) and O
2
physisorbed on the SWNT (E
T
[O
2
+ SWNT]).
By definition, E
s
> 0 in equation (3) corresponds to a stable and exothermic chemical bonding.
In some cases, E
s
- 0 is endothermic, but it corresponds to a local minimum on the Born–
Oppenheimer surface, where the desorption of O
2
fromthe SWNTis prevented by a barrier. The
spin polarized calculations yield relatively lower (stronger) total energies, E
T
[O
2
+ SWNT],
and hence set the triplet state as the ground state with a net magnetic moment of ∼2 j
B
(j
B
= Bohr magneton) per unit cell. The corresponding values for E
s
are −5, 4, −27 and
37 meV, for A, H, Z and T sites (see figure 8).
In the case of the physisorption, the contributions of short range and long range interactions
are small and comparable. Therefore, in treating the binding energy of the physisorbed O
2
the
attractive vdW energy is calculated by using the asymptotic form of the Lifshitz formula [79–
81], i.e. E
vdW
=

i j
C
6i j
¡r
6
i j
. A positive sign has been assigned to E
vdW
, since stable binding
Topical Review R915
is specified by positive energies. Here r
i j
is the distance between the i th O atom and the j th
C atom, and the constant C
6i j
is calculated within the Slater–Kirkwood approximation [82]
to be 10.604 eV Å
6
. We note, however, that this standard calculation of vdW energy may
involve ambiguities due to the relatively small distance between the physisorbed molecule
and the SWNT. Nevertheless, the vdW interaction is attractive and strengthens the chemical
bond. Using the binding geometries determined fromthe minimization of GGAtotal energies,
the E
vdW
values are calculated to be 125, 155, 185 and 155 meV for A, H, Z and T sites,
respectively [75]. By adding these E
vdW
values to the corresponding GGA chemical bonding
energies E
s
, the binding energies E
b
of physisorbed O
2
are found to be 120, 159, 158 and
191 meV for A, H, Z and T sites, respectively (see figure 8). The binding energy of the T site
is in agreement with the recent measurement by Ulbricht et al [83]. We note that these binding
energies are small, and become exothermic mainly owing to the long range vdW interaction.
The small binding energies are characteristic of physisorption.
Using a similar first-principles method within the local density approximation (LDA), Jhi
et al [68] found E
s
= 250 meV for the A site physisorption. The relatively large value of E
s
is due to the fact that the LDA yields overbinding as compared to the GGA. The physisorption
of O
2
on the (6. 6) armchair SWNT has been studied for two possible physisorption sites;
i.e. above the centre of the hexagon (H site), and on top of the C–C bond and perpendicular to
the axis of the tube (B site) [75]. The B site of a (6. 6) tube is similar to the A site of a (8. 0)
tube, but in the former case the C–C bond under the adsorbed O
2
is highly strained, since it
lies on the circumference. The binding energies are calculated to be 132 and 106 meV for H
and B sites, respectively.
While the spin polarized LDA (LSDA) energy band structure [68] has been interpreted
as hole doping due to the adsorbed oxygen molecule, Derycke et al [84] have argued that
the main effect of oxygen physisorption is not to dope the bulk of the tube, but to modify the
barriers of the metal–(8. 0) SWNT contact. The hole doping picture has been refuted by recent
first-principles calculations [75–77]. The band structure calculated using the spin polarized
GGA for O
2
physisorbed at the A site has yielded a band gap of 0.2 eV between the top of the
valence band of the (8. 0) tube and the empty O
ppπ
∗ (↓) band (see figure 9). Band structures
corresponding to H, Z and T sites also have band gaps, which are large enough to prevent
thermal excitation of electrons from the valence band at room temperature [77].
The situation for the m-SWNT is quite different. Upon O
2
physisorption the metallicity
of the armchair tube is lifted for the spin down bands, while spin up bands continue to cross at
the Fermi level, and make the system metallic only for one type of spin. The splitting of the
bands can be explained by the broken symmetry [85].
The interaction of O
2
with the SWNT as a function of O
2
–SWNT distance d has been
examined by calculating the total energy E
T
, bond distance of O
2
d
O−O
, magnetic moment j
and energy gap E
g
of the O
2
physisorbed (8. 0) SWNT [77]. In these calculations, d has been
constrained, but d
O−O
has been relaxed. Figure 10(a) shows the variation of the ratios of E
T
,
d
O−O
, magnetic moment j and E
g
to their corresponding equilibrium values at d
0
= 2.89 Å.
We see that [E
T
[ decreases in the range 1.6 Å - d - 2.9 Å. For a wide range of O
2
–SWNT
separation, E
g
continues to exist, and the total energy difference between the spin polarized
and spin unpolarized states (i.e. LE
T
= 0.86 eV) induces the gap E
g
and prevents it from
closing. For ∼1.6 Å - d - d
0
a strong perpendicular force F

is generated on the O
2
molecule to push it away from the SWNT. The magnetic moment of O
2
+ SWNT diminishes
at a distance d - 2 Å. Moreover, the singlet state of adsorbed O
2
leads to a bound state at
the A site at d = 1.47 Å, with an energy 0.45 eV above the corresponding physisorption
state (i.e. E
T
− E
0
T
= 0.45 eV). A similar singlet bound state occurs also at the Z site at
R916 Topical Review
Figure 9. The electronic structure of O
2
physisorbed on an (8. 0) SWNT calculated using the spin
polarized GGA with the atomic structure fully optimized in the double cell. The spin up and spin
down bands corresponding to the triplet ground state are shown by broken and continuous lines,
respectively. The zero of energy is set at the Fermi level. E
V
indicates the top of the valence band.
The adsorption site (A site) is shown in the inset. (Reproduced from [75].)
d = 1.47 Å and 0.80 eV above the physisorption state of the Z site at d = 2.70 Å. These
singlet bound states in figure 10(b) correspond to local minima on the Born–Oppenheimer
surface and are separated from the more energetic physisorption states by an energy barrier.
It is noted that these states [77, 86] are neither easily accessible from the physisorption state,
nor support the hole doping picture because the band gap is ∼0.5 eV. However, O
2
adsorbed at
the H site of the (6. 6) SWNT behave differently. Figures 10(c) and (d) show the variation of
E
T
(d), F

(d), j(d) and d
O−O
(d) where at d ∼ 1.25 Å, d
O−O
is increased to 2.5 Å and hence
the bond is broken, i.e. the O
2
molecule dissociates into two O atoms. In contrast to the case
at this site, the O
2
molecule has a bound singlet state at the B site with d
O−O
= 1.51 Å.
4.2. Chemisorption of oxygen atoms
The effect of the adsorbed oxygen on the structural and electronic properties of a SWNT is
important. Breaking of the O–O bond of a physisorbed O
2
molecule is unlikely owing to
the weak interaction with the SWNT. However, it was shown that near the defect sites of the
graphite surface O
2
molecules can dissociate [78]. A carbon nanotube, that can be described
as graphene rolled into a cylinder, is normally more reactive than the surface of graphite. As
a result, O
2
physisorbed near the defect sites of a SWNT is expected to dissociate into atomic
oxygens. In fact, it was shown that there is no activation barrier for the dissociation of O
2
when
it is adsorbed at the zigzag edge of a SWNT [70]. Owing to the concerted motion of the atoms
at the proximity of the molecule and energy gained by the individual oxygen atoms engaging
in the bonding with the SWNT concomitant with the dissociation, the activation energy for
dissociation is expected to be low. In this section, we discuss the interaction between atomic
O and SWNT, and reveal the nature of the chemical bonding. Only the short range interaction
is considered.
Spin polarized calculations yield the singlet state with a net zero magnetic moment as
the ground state. Various adsorption sites together with the geometry and chemisorption
energies E
s
obtained from spin unpolarized calculations are shown in figure 11. Among all
sites considered in [75], the single O adsorbed on top of the zigzag C–C bond (i.e. the z site)
Topical Review R917
Figure 10. (a)Variation of the percentage values of C¡C
0
(i.e. E
T
¡E
0
T
, d
O−O
¡d
0
O−O
, j¡j
0
and E
g
¡E
0
g
) with the O
2
–SWNT separation d. E
0
T
, d
0
O−O
, j
0
and E
0
g
correspond to the stable
physisorption state with d
0
= 2.89 Åat the Asite. (b) The total energies of the singlet bound states
found at small d at the Z site (square) and at the A site (diamond). The total energies of the triplet
ground state corresponding to the physisorption state for Z and A sites at d ∼ 2.9 Å are shown by
continuous and broken lines. (c) Variation of the total energy, E
T
, and the force acting on the O
2
molecule, F

; (d) variation of the bond distance of O
2
, d
O−O
, and the magnetic moment, j, with
the O
2
–(6, 6) SWNT distance d for O
2
adsorbed on the H site of the (6, 6) SWNT. (Reproduced
from [77].)
is energetically most favourable with E
s
= 5.07 eV. The binding energies as large as ∼5 eV
suggest that atomic O is actually chemisorbed with a significant charge transfer from C to O.
Moreover, the energy gained from the chemisorption of two atomic oxygens is more than the
bond energy of O
2
in either magnetic or nonmagnetic states. This implies that the dissociation
of O
2
followed by the chemisorption of individual O atoms is an exothermic process similar
to other oxidation processes. As a manifestation of the curvature effect, the strained C–C
bond at the z site is broken, but the C–O bond is strengthened. Experimentally, it was shown
that oxygen exposure first oxidizes and eventually etches away the nanotubes with smaller
radius [87].
The electronic state of an oxygen atomchemisorbed on a SWNT depends on the coverage
and pattern of adsorption. For example, since a single oxygen (per unit cell) chemisorbed at an
a site gives rise to flat bands below E
V
, the band gap of the (8. 0) SWNT remains unaffected.
However, as a well known and usual effect of oxygen, the band gap of the (8. 0) tube opens
up to 3.64 eV when all a sites are filled by chemisorbed oxygen atoms [75].
5. Adsorption of individual atoms on SWNTs
This sections deals with the adsorption of individual atoms on SWNTs and the resulting
properties. Because of miniaturization aiming at higher and higher device densities, the
R918 Topical Review
Figure 11. Aschematic illustration of the various adsorption sites of atomic Oon the (8. 0) SWNT.
Some relevant geometrical data and GGA chemical bond (chemisorption) energies corresponding
to these sites are also given. E
s
: chemisorption energy; d
C−O
: length of the C–O bond; d
C−C
:
length of the C–C bond under an adsorbed O atom. E
s
is obtained from the spin unpolarized
calculations of the total energies in equation (3). (Reproduced from [75].)
realization of interconnects with high conductance and low energy dissipation appears to
be a real technological challenge. Very thin metal wires and atomic chains were produced by
retracting an STMtip froman indentationandthenthinning the neck of the material that wets the
tip [88–90]. While the nanowires produced so far have played a crucial role in understandingthe
quantumeffects in electronic and thermal conductance [14, 91–94], they were not reproducible
enough to offer any relevant technological application. Nowadays, the most practical and
realizable method for fabricating nanowires relies on carbon nanotubes. Moreover, electron
transport measurements showed that SWNTs can resist high Joule heating [95].
Earlier experimental studies have indicated that SWNTs can serve as templates for
producing reproducible, very thin metallic wires with controllable sizes [96]. These metallic
nanowires can be used as conducting connects and hence are important in nanodevices based
on molecular electronics. Recently, Zhang et al [97] have shown that continuous Ti coating of
Topical Review R919
Figure 12. A schematic illustration of different binding sites of individual atoms adsorbed on
a zigzag (8. 0) tube. H: hollow; A: axial; Z: zigzag; D: top; S: substitution sites. (Reproduced
from [103].)
varying thickness and quasi-continuous coating of Ni and Pd can be obtained by using electron
beam evaporation techniques. However, metal atoms such as Au, Al, Fe, Pb were able to form
only isolated discrete patterns or clusters rather than a continuous coating of the SWNT.
Not only metallic connects, but also the contacts of metal electrodes themselves are crucial
for the operation of devices based on nanotubes. Lowresistance ohmic contacts to metallic and
semiconducting SWNTs were achieved by Ti and Ni [98]. The formation of Schottky barrier
at the contact has been found to be responsible for the operation of field emission transistors
made from SWNTs [99–101].
5.1. Binding geometry and binding energies
Recently, an extensive study on the binding geometry, binding energy and resulting electronic
structure of various atoms (ranging from alkali and simple metals to group IV elements, and
including most of the transition metal atoms) adsorbed on SWNTs has been carried out by
Durgun et al [102, 103]. They considered four possible sites shown in figure 12, and determined
the binding geometry and binding energies by optimizing all atomic positions.
The binding energy is calculated from the short range (chemical) interaction using the
expression E
u (p)
b
. E
u (p)
s
= E
u (p)
T
[SWNT] + E
u (p)
T
[A] − E
u (p)
T
[A + SWNT], similar to that
given in equation (3). Here the total energy of the adsorbed atom is E
u (p)
T
[A]. The superscript
u (p) indicates spin unpolarized (spin polarized) energies. The binding energies E
u (p)
s
are
obtained from the total energies corresponding to either a nonmagnetic (spin unpolarized)
state with zero net spin or a spin relaxed state. The long range interaction, E
vdW
, is expected to
be much smaller than the chemisorption binding energy and is omitted. However, for specific
elements, such as Mg and Zn, the binding energy is small and the character of the bond is
between chemisorption and physisorption. Tables 2 and 3 summarize the results relating to
the adsorption of individual atoms.
The interaction between the SWNT and most of the adatoms in tables 2 and 3 is
significant and results in a chemisorption bond. Thus, the binding energies corresponding
to a nonmagnetic state range from ∼1 to ∼4.5 eV. While alkali and simple metals have
binding energy in the range of 1.5 eV, the chemisorption energy of transition metals is
relatively high. On the other hand, metals such as Cu, Au, Ag and Zn have relatively weak
binding. The centre of the hexagons (H site) appears to be favoured by most of the adatoms.
The average C–adatom bond distance occurs in the range of 2.0–2.3 Å. However,
¨
d
C−A
is
relatively small for H, C, O atoms having small atomic radii. It is well known that the
R920 Topical Review
Table 2. The strongest binding site (shown in figure 12); the adsorbate–carbon distance
¨
d
C−A
; the
difference between spin unpolarized and spin polarized total energies LE
T
; the binding energy E
u
b
obtained from spin unpolarized calculations; the binding energy E
p
b
obtained from spin polarized
calculations; the magnetic moment (j
B
¡supercell) of the magnetic ground state corresponding to
the adsorption of various individual atoms on the (8. 0) SWNT. (Reproduced from [103].)
Atom Site
¨
d
C−A
(Å) LE
T
(eV) E
u
b
(eV) E
p
b
(eV) j (j
B
)
Na H 2.3 — — 1.3 —
Sc H 2.2 0.15 2.1 1.9 0.64
Ti H 2.2 0.58 2.9 2.2 2.21
V H 2.2 1.20 3.2 1.4 3.67
Cr H 2.3 2.25 3.7 0.4 5.17
Mn H 2.4 2.42 3.4 0.4 5.49
Fe H 2.3 1.14 3.1 0.8 2.27
Co H 2.0 0.41 2.8 1.7 1.05
Ni A 1.9 0.02 2.4 1.7 0.04
Cu A 2.1 0.03 0.8 0.7 0.53
Zn H 3.7 0 0.05 0.04 0
Nb H 2.2 0.40 3.9 1.8 2.86
Mo H 2.2 0.32 4.6 0.4 4
Pd A 2.1 0 1.7 1.7 0
Ag A 2.5 0.03 0.3 0.2 0.6
Ta H 2.2 0.73 2.9 2.4 3.01
W H–A 2.1 0.59 3.4 0.9 2.01
Pt A 2.1 0 2.7 2.4 0
Au A–T 2.2 0.04 0.6 0.5 1.02
Al H 2.3 — 1.6 — —
C Z 1.5 — 4.2 — —
Si H 2.1 — 2.5 — —
Pb H 2.6 0.01 1.3 0.8 0
H T 1.1 — 2.5 — —
O Z 1.5 — 5.1 — —
S A 1.9 — 2.8 — —
Table 3. The strongest binding site (as shown in figure 12); the adsorbate–carbon distance
¨
d
C−A
; the difference between spin unpolarized and spin polarized total energies LE
T
; the binding
energy E
u
b
obtained from spin unpolarized calculations; the binding energy E
p
b
obtained from
spin polarized calculations; the magnetic moment j per supercell corresponding to the magnetic
ground state corresponding to the adsorption of individual Ti, Mn, Mo, Au atoms on a (6, 6) SWNT.
(Reproduced from [103].)
Atom Site
¨
d
C−A
(Å) LE
T
(eV) E
u
b
(eV) E
p
b
(eV) j (j
B
)
Ti H 2.2 0.48 2.62 1.81 1.68
Mn H 2.5 2.23 3.25 0.1 5.60
Mo H 2.3 0.2 4.34 0.1 3.61
Au T 2.3 0.02 0.41 0.28 0.79
interaction between the graphite surface and most of the atoms included is actually weak.
The curvature effect is the primary factor that strengthens the binding [18, 26, 43]. Specific
adsorbate–SWNT (A + SWNT) systems are found to be in a magnetic ground state, since
LE
T
= E
u
T
[A + SWNT] −E
p
T
[A + SWNT] > 0. Most of the transition metal atoms adsorbed
on the (8. 0) and (6. 6) tubes are in the magnetic ground state with LE
T
> 0, and hence they
have net magnetic moments ranging from 5.49 j
B
(for Mn) to zero magnetic moment (for Pd
Topical Review R921
and Pt). While Ni adsorbed SWNT has very low magnetic moment (0.04 j
B
), the adsorbates
such as Au, Ag and Cu have magnetic moments in the range of 0.4–0.6 j
B
. Atoms, such as
Na, Al, C, Si, Pb, O, S, H favour a nonmagnetic ground state when they are adsorbed on the
(8. 0) SWNT. The magnetic moment generated upon the adsorption of individual transition
atoms has important implications, and points to an issue: whether molecular magnets (or
nanomagnets) can be produced from carbon nanotubes. Addressing this issue may open an
active field of study on SWNTs, which are coveredor substitutionally doped by transition metal
atoms according to a well defined pattern. Inclusion of transition metal elements inside the tube
is another way to obtain nanomagnetic structures. In this way, these atoms are prevented from
oxidizing. Whether a permanent magnetic moment can be generated by exchange interaction
on these transition metal coated SWNTs would be an interesting question to answer. Recently,
the magnetization and hysteresis loops of iron nanoparticles partially encapsulated at the tips
and inside of aligned carbon nanotubes have been demonstrated by experimental works [104].
Binding energies listed in table 2 are of particular interest for coating of SWNTs by metal
atoms, and hence for the fabrication of nanowires. The atoms which were observed to form
continuous and quasi-continuous coating on the SWNT (Ti, Ni and Pd) have relatively high
binding energies as compared to those atoms (Au, Fe, Pb) which remain as discrete particles
on the surface of the tube [96]. We also note that in forming a good coverage, not only
adatom–SWNT interaction, but also other factors, possibly adatom–adatom interaction, play
a crucial role. Good conductors such as Au, Ag, and Cu have very weak binding. On the
other hand, Na with one 3s electron on the outer shell is bound with a significant binding
energy (E
b
= 1.3 eV). The binding energy of Mg is very weak and is only 0.03 eV at the
H site due to its outer shell (3s)
2
. For the same reason Zn by itself exhibits a similar trend
(E
u
b
= 0.05 eV) with its (4s)
2
valence structure. Owing to the weak binding the type of
the bond between Mg (Zn) and the SWNT is between chemisorption and physisorption with
d
C−A
= 3.8 Å (d
C−A
= 3.7 Å). While an individual Al atom (with (3s)
2
3p valence structure)
is not bound to the graphite surface, its binding on the (8. 0) SWNT is relatively strong. This
can be explained by the curvature effect, since the binding was found to be even stronger at the
high curvature site of the SWNT under uniaxial radial deformation [26, 43]. Note also that the
binding energies E
u
b
, as well as E
p
b
and the magnetic moments of the adatom adsorbed on the
(8. 0) SWNT, came out to be consistently lower for the adatom adsorbed on the (6. 6) tube.
Perhaps this trend can also be explained by the curvature effect, since the radius of (8. 0) is
smaller than that of (6. 6) [18, 31].
The transition metal atoms with a few d electrons, such as Sc, Co, Ti, Nb, Ta, form strong
bonds with a binding energy ranging from 2.4 to 1.8 eV, and hence can be suitable for metal
coating of a SWNT. These metals can also be used as a buffer layer to formuniformcoating of
good conductors such as Au, Ag, Cu. Most of the adatoms yield the strongest binding at the H
site. Ni, Pd, Pt (column VIII elements) and Cu, Ag, Au (column I-B elements) seem to prefer
the A site. The average adsorbate–C distance,
¨
d
C−A
, ranges between 1.9 Å (minimum) and
3.7 Å (maximum); most of them occur at ∼2.1 Å. The interaction of group IV elements with
the SWNT is also crucial, since they have the same valence configuration. C and Si adatoms
form rather strong bonds with the SWNT. The calculated binding energies are rather high,
i.e. E
b
= 4.2 and 2.5 eV, respectively. The Z site is energetically favourable for both C and Si
adatoms.
The character of the bond between the adatom and SWNT can be examined by charge
density analysis. For the fully relaxed, minimum total energy configuration, the SCF total
charge density ρ(r) =

occup
n.k
+

n.k
(r)+
n.k
(r) and the difference charge density Lρ(r) =
ρ[A + SWNT] − ρ[SWNT] − ρ[A] can be calculated. In Lρ(r), the total charge densities
of adatom adsorbed SWNT ρ[A + SWNT], of bare SWNT ρ[SWNT] and of free atom ρ[A]
R922 Topical Review
Figure 13. Contour plots of total ρ(r) and difference Lρ(r) charge densities. For C the charge
densities are calculated on a plane passing through the adatom and zigzag C–C bond. For others
(Na, Al, Ti) the charge density plane passes through the centre of the hexagon and adatom. In
contour plots of Lρ(r), charge is depleted from black regions and is accumulated at white regions.
The counter-plots of ρ(r) for [Na + SWNT] are not presented, since they do not convey relevant
information owing to the low charge density around adsorbed Na atoms. (Reproduced from [103].)
are all calculated in the same supercell with the same atomic positions. The difference charge
density conveys information about the charge rearrangements upon adsorption. Figure 13
shows ρ(r) and Lρ(r) which were calculated for four adatoms, namely Na, Al, C, Ti [103].
Dramatic differences in the valence electron configurations of these adatoms are reflected in
the characters of the bonds they form with the SWNT.
According to the above analysis the charge is generally transferred from the adatom to
the C–C bond of SWNT. The 3s valence electron of Na is weakly bound and is donated to the
Topical Review R923
conduction band of the SWNT, which is derived from the modified conduction band of the
bare SWNT. In the case of Al, which is adsorbed at the H site, electrons are transferred from
Al and nearest C atoms to the region between the Al atom and the centre of the hexagon on
the surface of the SWNT and to the C–C bond. The character of the bond formed between
the C adatom and C atoms of the SWNT is reminiscent of the sp
3
covalent bonds in diamond
crystals. ρ(r) has a maximum value at the centre of the C–C bonds. Because of new bond
formation, charge is transferred to the back-bonds. The charge density of adsorbed Si exhibits
a character similar to that of adsorbed C. Directional bonds form between Si and C atoms of
the SWNT. The Ti atom adsorbed at the H site forms a bond. The d orbitals are responsible for
this bond. According to the Mulliken analysis the charge transferred from a Ti atom is 1.45
electrons for the spin unpolarized case. The interaction between the Au atom and SWNT is
weak, which results in a small binding energy. Accordingly, the charge rearrangement due to
Au adsorption is minute.
5.2. Electronic structure
An individual atom adsorbed on an SWNT may give rise to resonance states in the valence
and conduction bands, and also localized states in the band gaps. The energy states associated
with a single adsorbate form energy bands if the electronic structure is calculated within the
supercell method. Actually these bands correspond to a linear chain of adsorbates with a
1D lattice constant of the supercell, c
s
. The localized states are relevant for the doping of a
s-SWNT. Depending on their position relative to the band edges, they are specified as donor
states (if they are close to the edge of the conduction band, E
C
) or as acceptor states (if they
can occur close to the edge of the valence band, E
V
). The latter case is also known as hole
doping.
Na, Al, C and Si, which are adsorbed individually on the (8. 0) SWNT with a repeat
period of 2c, give rise to a spin paired, nonmagnetic ground state [103]. Figure 14 presents
their calculated band structures and the LDOS at the adatoms. The calculated band structure
of the SWNT with Na adsorbed at the H site forming a regular chain with a lattice parameter
(or interatomic distance) a = c
s
∼ 2c yields a half-filled band. This band is normally
the first empty conduction band of the tube. Since the 3s valence electron of Na has low
ionization energy, it is donated to the SWNT to occupy the empty conduction band. As a
result, this empty band is gradually populated and also modified upon adsorption of Na; hence
the SWNT becomes metallic. Another alkali metal, Li, has been the focus of attention because
of SWNT ropes which were expected to be a candidate for anode material use in Li battery
applications [105]. In fact, recent experiments showed higher Li capacity in SWNTs than that
of graphite [106]. Zhao et al [107] investigated Li intercalation in (10. 0), (12. 0), (8. 0) and
(10. 10) SWNT ropes modelled by a uniform, 2D hexagonal lattice. They found that charge
is transferred from adsorbed Li to the SWNT. Furthermore, the SWNT is slightly deformed
upon Li adsorption. The band structure of the Li intercalated (8. 0) rope has indicated that the
bands of the bare SWNTs are not affected in any essential manner; but their conduction bands
get occupied above E
g
and the system becomes metallic. Their findings are consistent with
the adsorption of Na discussed above.
Similarly, the calculated band structure of the SWNTwith Al adsorbed at the Hsite forming
a regular chain structure gives rise to a half-filled band derived from the empty conduction
band of the SWNT. Therefore, the localized state due to an adsorbed individual Al atom is a
donor state. In the case of C, a small gap occurs between the bands derived from adsorbate
states. The empty and filled bands just above and below E
F
are derived from C adatoms with
a significant contribution from the SWNT. Si yields an almost fully occupied band in the band
R924 Topical Review
Figure 14. Energy band structures, TDOSs and LDOSs of single Na, Al, C, Si adsorbed on a
zigzag (8. 0) tube. LDOSs calculated at the adsorbate. The zero of energy is set at the Fermi level.
Na, Al and Si are adsorbed at the H site; C is adsorbed at the Z site. (Reproduced from [103].)
gap and an almost empty band at the bottom of the conduction band. The dispersion of bands
for Si adsorbed on the H site displays differences from that of the C adatom adsorbed at the Z
site. However, the overall behaviours of state densities are similar.
Topical Review R925
Figure 15. Geometric structures of adsorbed Ti chains on (14. 0) (top) and (8. 8) (bottom) SWNTs.
Along the tube axis, each unit cell contains two Ti atoms for the (14. 0) tube and one Ti atom for the
(8. 8) tube. Several unit cells are shown for the purpose of visualization. The distances between
the two neighbouring Ti atoms in the chains are 2.6 Å for (14. 0) and 2.7 Å for (8. 8) tubes.
(Reproduced from [108].)
The spin polarized band structure and TDOSs of adsorbed transition metal atoms (i.e. Au,
Mn, Mo and Ti) display a different situation due to the magnetic ground state. Au yields two
bands in the band gap, namely filled spin up and empty spin down bands. This is in compliance
with the calculated magnetic moment of ∼1j
B
per adsorbed Au atom. There is a small band
gap of ∼0.2 eV between these Au bands. It is seen that the adsorption of Au does not induce
significant modification in the bands of (8. 0), except the Au 6s(↑) and Au 6s(↓) bands in the
gap; it gives rise to two sharp peaks below and above E
F
in the TDOS. That the contribution
of the SWNT states to these peaks is minute and that the band gap between the conduction and
valence bands of the SWNT is practically unchanged confirm the weak interaction between
Au and the SWNT. Filled Au 5d(↑) and 5d(↓) bands occur in the valence band of the SWNT
2 eV below E
F
. Three bands formed fromTi 3d(↑) are fully occupied and accommodate three
electrons of adsorbed Ti atoms. Other Ti 3d(↑) bands occur above E
F
, but they overlap with
the conduction band of the SWNT. The dispersive and almost fully occupied spin down band
is formed from the states of carbon and hence derived from the conduction band of the bare
SWNT. The SWNT is metallized upon Ti adsorption, since this band crosses the Fermi level
and also overlaps with the other conduction bands. This situation is in accordance with the
Mulliken analysis, which predicts the electrons to be transferred from Ti to the SWNT.
Yang et al [108] have studied the binding energies and electronic structures of Ti, Al and
Au chains adsorbed on (10. 0), (14. 0), (6. 6) and (8. 8) SWNTs using first-principles methods.
The geometric configuration of the adsorbed metal chain is shown in figure 15. In spite of the
fact that the calculated binding energies include significant coupling between adjacent metal
atoms, they show the same trends as the binding energies given by Durgun et al [102, 103]. Ti
is much more likely than Al and Au to form a continuous chain on the surface of SWNTs due
to much higher binding energies. Al adsorbed SWNTs in turn have higher binding energies
than the Au adsorbed ones. On the basis of the works by Bagci et al [109] and Prasen et al
[110], we believe that the stability of the Al and Au chains originates from the stability of
zigzag-free Al and Au chains. In calculating the binding energies the energy references are
crucial. It is well known that adsorbed Ti and Au have magnetic ground states leading to a
R926 Topical Review
Figure 16. Optimized atomic structure of (8. 0), (9. 0) and (6. 6) armchair SWNTs which are
uniformly covered with Ti. Corresponding densities of states for spin up and spin down electrons
are shown. (Reproduced from [7].)
net spin. Therefore, the binding energies reported by Yang et al [108] might change if one
took the magnetic states into account. The important effect of the Ti chain is to metallize the
zigzag SWNTs due to the delocalized d states crossing the Fermi level. However, the situation
is different for the Ti chain adsorbed armchair tube, since a small gap is opened owing to the
symmetry breaking.
5.3. Transition element covered or filled SWNTs
It is clear from the above discussion that individual atoms, such as Ti, Ni, Si, can form rather
strong bonds to the surface of the SWNT. Motivated by the experimental work of Zhang
et al [97] showing the continuous coating of the SWNT by Ti, Dag et al [7] investigated the
coverage of an (8. 0) tube by specific transition element atoms, such as Ti, Fe, Co, Cr, using
a first-principles plane wave method. They found that only Ti can yield a continuous, regular
coverage of the SWNT, as well as giving rise to a dramatic modification of the atomic structure.
Upon optimization of the geometry the circular cross section of the SWNT has changed to
a square. Even more interesting is that the s-SWNT has become a highly conducting metal
due to several Ti bands crossing the Fermi level. The conductance of the Ti covered tube has
been estimated to be 16e
2
¡h. Furthermore, they found that the Ti covered (8. 0), (9. 0) and
(6. 6) tubes have magnetic ground states with significant magnetic moments of 13.3, 13.7 and
9.5 j
B
¡cell, respectively. Figure 16 shows the structure and density of states of Ti covered
(8. 0), (9. 0) and (6. 6) tubes.
Yang et al [111] have examined the magnetism of transition metal covered or filled (9. 0)
SWNTs. They also found through ab initio calculations that these structures exhibit substantial
Topical Review R927
Figure 17. The spin polarized band structure (left: minority spin; right: majority spin) and DOS
(bottom: minority spin; top: majority spin) for Co filled SWNT as shown in the right panel. Both
plots suggest that the magnetic moment resides mostly in the narrow transition metal d band below
the Fermi level. (Reproduced from [111].)
magnetism. In particular, Co atoms packed inside a variety of carbon nanotubes offer strong
spin polarization at the Fermi level, as well as considerable magnetic moments. Figure 17
shows the spin polarized band structure and DOS for the Co filled SWNT.
6. Radial deformation of carbon nanotubes
Modification of electronic properties of condensed systems by an applied external pressure
or strain in the elastic range has been a subject of active study. However, in most cases, the
changes that one can induce by the elastic deformation are minute or even negligible due to the
rigidity of the crystals. On the other hand, the situation is rather different for SWNTs owing to
their tubular geometry [33]. Carbon nanotubes display very interesting mechanical properties.
While the MWNTs showthe greatest values for the Young’s modulus, as high as 2 TPa, SWNTs
appear to be very soft in the radial direction [112]. SWNTs are highly flexible: they can sustain
elastic deformations remarkably. The softness of carbon nanotubes in the radial direction has
been observed experimentally: the contact area is flattened when two nanotubes are brought
close to each other [113]. If R is large, a nanotube takes on a flattened structure [114]. It has
been argued that SWNTs show linear elasticity under hydrostatic pressure up to 1.5 GPa at
room temperature.
R928 Topical Review
It has been shown that the structure and the electronic properties of SWNTs undergo
dramatic changes as a result of these deformations [23, 43, 115–125]. Significant radial
deformation of SWNTs can be realized in the elastic range, whereby the curvature is locally
changed. In this way, zones with higher and lower curvatures relative to the undeformed SWNT
can be attained for the same circumference. Hence, one expects radial deformation to induce
important modifications in the electronic and conduction properties of nanotubes [24, 25, 126–
129]. Tight binding calculations have indicated that a SWNT may undergo an insulator–
metal transition under a uniaxial or torsional strain [24, 124, 125]. Multiprobe transport
experiments [23] on individual SWNTs showed that the electronic structure can be modified
by bending the tube, or by applying a circumferential deformation. Empirical extended H¨ uckel
calculations [115] predicted that the conductance of an armchair SWNT can be affected by the
circumferential deformations and a band gap can develop in a metallic armchair SWNT upon
twisting.
A systematic analysis of the effect of the radial deformation on the electrical properties
has been carried out by G¨ ulseren et al [43] using first-principles total energy and electronic
structure calculations. In their study the radial deformation is generated by applying uniaxial
compressive stress σ
yy
to a narrow strip on the surface of a SWNT. In practice such a
deformation can be realized by pressing the tube between two rigid and flat surfaces. As
a result, the radius is squeezed in the y direction while it is elongated along the x direction and
hence the circular cross section is distorted to an elliptical form with major and minor axes, a
and b, respectively. A natural variable for describing the radial deformation is the magnitude
of the applied strain along the two axes,
c
yy
=
R
0
−b
R
0
(4)
and
c
xx
=
R
0
−a
R
0
(5)
where R
0
is the radius of the undeformed (zero-strain) nanotube. The point group of the
undeformed nanotubes is D
nh
or D
nd
for n even or odd, respectively. Under the radial
deformation described above, the point group becomes C
2h
or D
2h
. However, depending on the
nanotube orientation around the tube axes, the in-plane mirror symmetry can be broken. For
the (6. 6) tube, there are several different orientations to investigate when looking at the effect
of mirror symmetry on the band crossing at the Fermi level. The first-principles calculations
predict that elastic radial deformation does not have a noticeable effect on the first- and second-
nearest neighbour C–C distances but it induces significant changes in the bond angles [43].
This observation is therefore important and has to be taken into account in tight binding studies
of SWNTs with radial deformation.
6.1. Elasticity
For the radial deformation, the relation between stress and strain is given by the generalized
Hooke’s law,
σ
xx
= 0 = C
11
c
xx
+ C
12
c
yy
(6)
and
σ
yy
=
F
y
A
= C
12
c
xx
+ C
11
c
yy
. (7)
Here F
y
is the restoring force applied on the surface area A, C
11
and C
12
are the in-plane elastic
stiffness constants. Assuming the validity of Hooke’s law, the elastic deformation energy E
D
Topical Review R929
Figure 18. (a) The strain component c
xx
= (R
0
−a)¡R
0
along the major axis as a function of the
applied strain c
yy
= (R
0
−b)¡R
0
along the minor axis. The slope is the in-plane Poisson ratio, ν
|
.
(b) Variation of the elastic deformation energy per carbon atom. (c) The restoring force on fixed
carbon atoms. For the (8. 0) SWNT, the force is scaled by 0.5 since it is only on one carbon atom,
while for the other tubes it is on two carbon atoms. (Reproduced from [43].)
(i.e. the difference between the total energies of radially deformed and undeformed SWNTs)
is
E
D
= O

1
2
(1 + ν
2
|
)C
11
−ν
|
C
12

c
2
yy
(8)
in terms of the in-plane Poisson ratio, ν
|
[43]:
ν
|
= −
c
xx
c
yy
=
C
12
C
11
. (9)
Deformation energies obtained fromclassical elasticity in equation (8) are compared with
those calculated from first principles in figure 18. Interestingly, the quadratic form obtained
from classical theory fits very well to the elastic deformation energy calculated from first
principles. Hooke’s relation, and hence the elastic character of the deformations, persists up
to c
yy
= 0.25. It is also noted that the SWNT becomes stiffer as R decreases. The variation
of the restoring forces is expected to be linear in the elastic range. The restoring forces in
figure 18(c) are in overall agreement with this argument, except the deviations at certain data
R930 Topical Review
Figure 19. (a) The variation of the band gap, E
g
. (b) The density of states at the Fermi level D(E
F
)
as a function of the applied strain c
yy
. (Reproduced from [43].)
Table 4. In-plane elastic constants of SWNTs. All elastic constants are in GPa except for ν
|
which is unitless. C
eff
= C
11
(1 −ν
2
|
). (Reproduced from [43].)
Radius (Å) ν
|
C
eff
C
11
C
12
(7. 0) 2.76 0.904 129.88 713.36 645.15
(8. 0) 3.14 0.874 98.70 416.88 364.20
(9. 0) 3.52 0.864 91.02 319.67 270.36
(6. 6) 4.06 0.828 86.12 273.46 226.34
points due to uncertainties in the first-principles calculations, which are amplified because the
force is a derivative quantity. Elastic constants calculated from first principles are listed in
table 4.
6.2. Effect on the electronic structure
First-principles calculations confirm that the band gaps of zigzag tubes reduce with applied
strain and eventually vanish leading to metallization. Figure 19 summarizes the variation of
the band gap and the density of states at the Fermi level, D(E
F
), as a function of the applied
strain. For (7. 0) and (8. 0) SWNTs the band gaps decrease monotonically and the onset of
an insulator–metal transition follows with the band closures occurring at different values of
strain. Upon metallization, D(E
F
) increases with increasing strain. However, the behaviour
of the (9. 0) tube is different. Initially, the band gap increases with increasing strain, but then
Topical Review R931
Figure 20. The variation of the energy eigenvalues of states near the band gap at the I point of
the BZ as a function of the applied strain. The shaded region is the valence band. The singlet state
originating in the conduction band is indicated by squares. (Reproduced from [43].)
decreases with strain exceeding a certain threshold value and eventually diminishes. For all
these zigzag SWNTs the band gap strongly depends on the magnitude of the deformation and
E
g
is closed at 13%, 22% and 17% strain for (7. 0), (8. 0) and (9. 0) nanotubes, respectively,
whereas the armchair (6. 6) SWNT, which is normally metallic, remainedmetallic with a slowly
decreasing D(E
F
) even for significant radial deformation. Earlier, Delaney et al [85] showed
that the π

conduction and π valence bands of a (10. 10) tube, which normally cross at the
Fermi level with quasi-linear dispersion, open a pseudogap in the range of ∼0.1 eV at certain
directions of the BZ perpendicular to the axis of the tube owing to tube–tube interactions
in a rope. The opening of the gap is caused by the broken mirror symmetry. Lammert
et al [120] showed the gapping by squashing (20. 20) and (36. 0) metallic tubes, since then
circumferential regions are brought into close proximity. Uniaxial stress of a few kilobars
can reversibly collapse a small radius tube inducing a 0.1 eV gap, while the collapsed large
radius tubes are stable. In the study of Park et al [44], the band gap of the (5. 5) tube was
monotonically increasing, probably due to bilayer interactions, since the separation of the two
nearest walls of the tube became comparable to the interlayer distance of graphite.
In order to explain the band gap variation of (n. 0) tubes, the energies of a few bands
near the band gap are plotted as a function of strain in figure 20. The singlet π

state in
the conduction band shifts downwards in energy much faster than the other states do with
increasing strain. This is due to the curvature increasing with increasing radial deformation.
Since the singlet π

state lies below the doubly degenerate π

states for both (7. 0) and (8. 0)
SWNTs, their band gaps are closed monotonically with increasing c
yy
. On the other hand, for
the (9. 0) SWNT this singlet π

state is above the doubly degenerate π

states. The increase
of the band gap at the initial stages of radial deformation is connected with the relatively high
rate of downward shift of the doubly degenerate π valence band relative to the π

conduction
band under low strains. Once the singlet π

band, which shows faster decrease with strain,
crosses the doublet conduction band and enters the gap, the band gap begins to decrease with
increasing strain. Applying a radial strain through the tip of an atomic force microscope and
simultaneously measuring the conductance, Minot et al [121] confirmed that the band structure
of carbon nanotubes can be altered as predicted in [24, 43, 44] earlier.
R932 Topical Review
Figure 21. Binding energies E
b
of single hydrogen and aluminium atoms adsorbed on the zigzag
SWNTs versus the radius of the tube R. The solid curve is the fit to E
b.A
(R) = E
0.A
+ C
A
¡R
explained in the text. (Reproduced from [26].)
6.3. Effect on the chemical reactivity
Srivastava et al [58] examined the effect of conformational strain on the chemical reactivity
of a SWNT. A (10. 10) SWNT, which was either bent or subjected to torsional strain, had
kinks or ribbon structure on its surface. Actually the deformed places have sites with higher
curvature as compared to the curvature of the bare (10. 10) SWNT. Using a many-body, bond
order potential, they performed classical trajectory simulations on the interaction between the
hydrogen atomand the strained nanotube. These predicted that the chemisorption of hydrogen
atoms is enhanced by as much as 1.6 eV in regions of high conformational strain. Analysis
of atomic cohesive energies and the local density of states obtained from the empirical tight
binding method showed that the sites with enhanced binding energies of hydrogen atoms
correspond to destabilized cohesive energies and electronic states near the Fermi level. The
heightened chemical reactivity of deformed sites was attributed to the introduction of radical
p orbital character.
First-principles pseudopotential calculations have shown that the adsorption of foreign
atoms on SWNTs can indeed be modified continuously and reversibly [26]. The effect of
radial deformation becomes noticeable through the binding properties of foreign atoms on
undeformed SWNTs. This situation is shown in figure 21, where the calculated binding
energies E
b
of H and Al adsorbed on the zigzag tubes are plotted as a function of the radius
R for n = 7. 8. 9. 10. 12. As described in the previous sections, H is adsorbed at the top site,
directly above the C atom of the tube; Al favours the H site. It is seen that E
b
decreases with
increasing radius (or decreasing curvature) and eventually saturates at a value corresponding
to that for on graphene. The variation of the binding energies of H and Al with the radius of
the zigzag tube fits to the curve given by the expression
E
b.A
(R) = E
0.A
+
C
A
R
(10)
where E
0.A
is the binding energy of the adatom A (here H or Al) on the graphene plane and
calculated to be 1.49 and −0.02 eV for H and Al, respectively [26]. The fitting parameter, C
A
,
is calculated to be ∼3.14 eV Å
−1
for both H and Al. Calculated data for n - 7 slightly deviate
from the above simple E
b
(R) relation, perhaps due to the fact that the singlet π

band, which
is normally in the conduction band, falls into the band gap as a result of increased curvature.
Note that, while the band gap shows significant change with n (for example by going from
Topical Review R933
Figure 22. (a) Variation of the binding energies E
b
of single hydrogen atoms adsorbed on a (8. 0)
zigzag SWNT as a function of the radial deformation c
yy
defined in the text. The upper curve
corresponds to H adsorbed on the high curvature site of the deformed tube. The lower curve is
for the adsorption on the low curvature site. (b) As (a) but for a single Al atom. (Reproduced
from [26].)
(8. 0) to (9. 0), E
g
changes from 0.65 to 0.09 eV), the binding energies vary smoothly with
R
−1
. Increase of E
b
with decreasing R (or with decreasing n) shows that for small R the
character of the surface deviates from that of the graphene. This result has been exploited
to create different zones with different curvatures on the surface of a single SWNT by radial
deformation and hence to attain different chemical reactivities.
Radial deformation can be created by applying a uniaxial compressive stress on a narrow
strip on the surface of the SWNT [26, 43] where the circular cross section of the bare tube is
distorted to an elliptical one with major and minor axes a and b, respectively. In figure 22(a)
the variation of the binding energies of a single hydrogen atom adsorbed on the (8. 0) surface
shows that E
b
at the sharp site, i.e. near one of the ends of the major axis, increases with
increasing c
yy
(defined in equation (4)), and traces out the upper curve. The lower curve is
associated with the adsorption on the flat site, i.e. near one of the ends of the minor axis.
According to the predictions of this study, at the high curvature site, E
b
has increased by
0.85 eV for c
yy
= 0.3. On the other hand, E
b
for the adsorption on the low curvature (flat)
site first decreases with increasing strain and then saturates at an energy 0.25 eV less than
that corresponding to c
yy
= 0. The difference of binding energies of the sharp and flat sites
R934 Topical Review
is substantial and is equal to ∼1.1 eV for c
yy
= 0. This is 44% of the binding energy of H
on the undeformed SWNT. As a result of H adsorption, the sp
2
character of the bonding of
the tube has changed locally and become more like sp
3
. In figure 22(b) the binding energy
of Al exhibits a behaviour similar to that of H, despite H and Al favouring different sites on
the (8. 0) tube; E
b
at the sharp site of the deformed SWNT increases with increasing c
yy
. For
example E
b
increases by ∼0.80 eV for c
yy
= 0.3 which is 80% of the binding energy on
the undeformed tube. For Al adsorbed on the flat site, E
b
first decreases with increasing c
yy
,
then gradually increases. Adsorption of Al induces local changes in the atomic and electronic
structures. For example, the surface of the tube where Al is adsorbed expands.
We see that according to the first-principles calculations the chemical reactivity of a
zigzag SWNT can be modified reversibly and variably by radial deformation. The effects of
deformation are significantly different for the zigzag and armchair SWNTs. It is remarkable
that Al, which does not bind to the graphite surface, can be adsorbed at the high curvature site
of a zigzag tube under radial deformation with a significant binding energy. Similarly, it has
been shown that a H
2
molecule which is bound to the surface of (8. 0) becomes physisorbed
at the high curvature site with a significant binding energy [130]. If the physisorbed H
2
can
approach the surface as closely as 1.5 Å, it dissociates into two Hatoms, each of which becomes
bound to adjacent C atoms of the SWNT at the surface. We believe that the tunable adsorption
can have important implications for metal coverage and selective adsorption of foreign atoms
and molecules on the carbon nanotubes and can lead to a wide variety of applications, ranging
from hydrogen storage to chemical sensing, and new nanomaterials.
6.4. Effect of pressure on nanoropes
It is anticipated that physical properties can also be altered by intertube interactions between
nanotubes packed in a hexagonal lattice, as so-called nanoropes. Intertube interactions in
nanoropes can be probed by applying external pressure to vary the intertube distance. For
fullerenes, such high pressure studies have yielded many interesting results [131] including
new compounds such as the pressure induced polymeric phases of C
60
. Similar covalent
bonding can occur between the nanotubes in a rope. Such a property could have important
consequences for nanoscale device applications and composite materials that require strong
mechanical properties, since nanoropes consisting of interlinked SWNTs will be significantly
stronger than nanoropes composed of vdW packed nanoropes [122].
Raman studies on SWNT ropes have been carried out up to 25.9 GPa, and have
indicated that the mode intensities and energies are not completely reversible upon pressure
cycling [132]. These results have been interpreted as irreversible pressure induced changes
in the atomic structure. Furthermore, observation of very large volume reduction and high
compressibility [119] has suggested that a microscopic volume reducing deformation other
than vdW compression can occur under high pressure. Some of these effects are tentatively
attributed to crushing or flattening of nanotube surfaces in the rope, deforming the circular
cross sections to elliptical or hexagonal ones [119]. Peters et al [133] reported that a structural
phase transition occurred at 1.7 GPa. Furthermore they performed empirical uniform force
field calculations and obtained equilibrium lattice constants of the (10. 10) nanotube rope as
a function of applied pressure P. These calculations have indicated a sudden transition of
structure at 1.75 GPa [133]. Their interpretation of the structural transition has been objected
to by Tang et al [134] on the grounds that their experiments were reversible up to P = 4 GPa.
Yildirim et al [123] were the first to investigate the effects of pressure on the ropes based
on the first-principles calculations. The zigzag tubes (5. 0), (6. 0), (7. 0), (9. 0) and the
armchair tube (6. 6) were included in that study. The pressure dependence was determined
Topical Review R935
Figure 23. Optimized structures of SWNT ropes. (a) (6. 0) and (b) (7. 0) nanotubes are packed by
vdW interaction under zero pressure and (c) (7. 0) one-dimensionally interlinked under pressure.
The interlinked structure in (c) has lower total energy than the vdW packed structure in (b). Lattice
parameters of the ropes, a, b and c and the γ angle, are shown. (Reproduced from [123].)
Figure 24. Total energy versus 2D lattice constant for (7. 0) nanotube ropes in different phases.
The top view along the axis of the rope of the different phases is shown in the inset. The zero of
energy is set to the total energy calculated for the optimized structure of the vdW packed rope at
zero pressure. (Reproduced from [123].)
by minimizing the total energy as a function of nanotube separation, i.e. E
T
(a. b), while the
other parameters, including the atom position c and γ = 2π¡n, were optimized. At critical
pressure, a structural phase transformation from a vdW lattice to a new orthorhombic (space
group C
m
C
m
) lattice was observed in which the nanotubes were interlinked along the [110]
direction. The covalent bonding between nanotubes occurred at the high curvature sites as a
result of rehybridization of the carbon orbitals. Figure 23 compares the structure of a vdW
packed (7. 0) nanorope with that of a 1D interlinked one. The same structural transformation
was observed for the (5. 0), (6. 0), (9. 0) zigzag tubes.
The bonds interlinking two adjacent tubes have bond lengths comparable to those in
diamond. This indicates that the rehybridization occurs fromthe sp
2
to the sp
3
type. However,
the bond angles differ from the exact value of the tetrahedral angle and vary between 100

and 120

indicating some strain. For some (n. 0) tubes the energy difference between the
1D interlinked structure and the vdW packed structure is small. Regardless of that, the 1D
interlinked structure is stable once it is formed, since one has to overcome an energy barrier
of 0.7 eV in order to break the links, as illustrated in figure 24.
The total energies of the vdW and 1D interlinked phases cross each other at a ∼ 9 Å with
a small energy barrier of 46 meV¡unit cell (552 K) from the vdW side. The pressure
R936 Topical Review
for attaining this lattice constant is only about 0.3 GPa for the vdW phase, indicating that
polymerization of vdW (7. 0) nanoropes could occur at modest pressures and temperatures.
Conversely, 0.7 eV¡unit cell (or 25 meV¡atom) is required to go from the 1D interlinked
phase to the vdW packed structure. This energy is comparable to the 1D polymerization
energy of C
60
. Interestingly, at about a ∼ 8 Å there occurs another interlinked phase of the
(7. 0) nanorope, where nanotubes are interlinked along both a and b axes. This 2D interlinked
phase is not favoured energetically. Its minimum lies ∼1 eV¡unit cell above the minimum
of the 1D linked phase. However, once it is formed, it is stable and sixteen times stiffer
(d
2
E¡da
2
= 13.7 eV Å
−2
) than the vdW packed rope. The structures transform to even more
complex structures at higher pressure. The ropes of the (6. 6) armchair tubes do not form
interlinking up to 60 GPa. They are rather hexagonally distorted such that the local structure
of nanotube faces is reminiscent of that of a graphite sheet. Upon releasing the pressure
the hexagonally distorted (or polygonized) structure returns to the original one, indicating
that the deformation of the (6. 6) nanorope is elastic (see figure 25). The above structural
transformations induced by the applied pressure lead to significant changes in the electronic
structure [123]. In spite of the fact that an individual (7. 0) tube is a semiconductor with
E
g
= 0.65 eV, the dispersion of the bands near Fermi level in the direction perpendicular to
the tube axis makes the vdW rope metallic. At high pressure (30 GPa), the (2D interlinked)
high density phase becomes a wide band gap insulator with E
g
∼ 2 eV.
7. Devices based on carbon nanotubes
Much of the research on carbon nanotubes has been carried out with the motivation of
generating devices, such as sensors, transistors and nanomagnets, and hence finding a feasible
way to contribute to the objectives of miniaturization. Attempts to design and fabricate
nanodevices from carbon nanotubes have exploited several properties discovered recently,
some of them reviewed in the previous sections. In this section, we will first review recent
progress made in calculating I –V characteristics of nanostructures and present some of the
device models or circuit elements, which are designed theoretically using SWNTs. Finally we
will discuss the devices fabricated.
7.1. Ab initio methods in transport calculations
Accurate prediction of the transport properties of nanoscale atomic or molecular systems
(nanotubes, in particular) including their current–voltage (I –V) characteristics is essential
for realization of a broad spectrum of device applications. Theoretical investigations of the
transport properties of these nanoscale systems are based on calculations with various levels of
sophistication, including phenomenological, semiclassical and quantum mechanical ab initio
methods.
Semiempirical methods [20, 135–145] based on parametrized Hamiltonians, such as
tight binding or extended H¨ uckel models, have been widely used. Since these parametrized
Hamiltonians are generally derived fromthe bulk or isolatedmolecular systems, the effects such
as external bias and gate potential are usually not accounted for. Moreover, these models do
not allow for structural relaxations to be performed. Semiempirical methods are not typically
self-consistent; however, they are relatively simple and easy to implement. These methods
provide some quantitative insight especially in cases where the charge transfer effects are
relatively small. Some applications to carbon nanotubes include designation of a quantum
dot by introducing pentagon–heptagon defects in a carbon nanotube [20], investigation of
spin coherent quantum transport in carbon nanotube magnetic tunnel junctions [140], study
Topical Review R937
Figure 25. The two-dimensional crystal structure and relevant physical parameters for various
high density phases of carbon nanotubes. (a) A 2D interlinked structure of (5. 0) nanotubes. (b)
A hexagonal network of (9. 0) nanotubes, in which (9. 0) tubes are interlinked along a, b and
[110] directions. (c) A very dense structure of (7. 0) nanotubes obtained under 30 GPa. (d) The
optimizedstructure of (6. 6) armchair tubes under 53 GPa. Nanotubes are distortedin such a fashion
that the local nearest neighbour structure is reminiscent of graphite structure. d
C−C
indicates the
smallest distance between two carbon atoms of nearest neighbour tubes in the rope. (Reproduced
from [123].)
R938 Topical Review
Figure 26. Upper panel: geometry of the bent (4. 4) nanotubes with bending angles (a) θ = 0

,
(b) 3

, (c) 6

. Lower panel: the calculated transmission function and LDOS for the corresponding
geometries. (Reproduced from [144].)
of the dynamic conductance properties of carbon nanotubes of different helicities, both with
and without defects [141], and calculation of transmission properties of extended [144] and
finite size [145] carbon nanotubes. Electronic transport properties of extended nanotubes
in the presence of distortions and mechanical deformations were studied by Nardelli [144]
using the surface Green function matching formalism and a nearest neighbour π orbital tight
binding Hamiltonian [146]. The π orbital bands were described via a single parameter,
V
ppπ
= −2.75 eV. Calculations made for small radius (4. 4) nanotubes show that bending of
the tube by small angles (3

and 6

) does not drastically alter the basic electronic characteristics
of the metallic (4. 4) tubes (see figure 26).
A first-principles approach to the transport properties of nanoscale devices is based on
supercell methods [147–154] where Kohn–Sham(KS) equations for the device are solved using
periodic boundary conditions. The scattering states are then determined through a recursive
technique considering the effective device potential between the perfect electrodes. The
applicability of these methods is limited to the calculation of transport coefficients in the linear
response regime, because the effective device potential is derived using periodic boundary
Topical Review R939
conditions which cannot describe systems under an external bias or having nonidentical
electrodes at the two ends. The method has been used to provide a possible explanation for the
experimentally observed conductance of one quantum unit instead of two for nanotubes with
one end dipped into a liquid metal such as mercury [153].
For nanoscale transport investigations, an alternative method, which was pioneered by
Lang [155], is the open jellium approach [155–160] based on Lippman–Schwinger theory.
Treating the leads in terms of a jellium model, the charge density of the open device is
constructed from the scattering states. The KS equations are solved self-consistently with
the appropriate boundary condition of the device under bias. The method, which allows I –V
curves to be calculated, is however limited by two major approximations. Constructing the
charge density and the device potential using only scattering states is not enough for achieving
a truly self-consistent solution. The omitted bound states, which exist within the device, must
also be considered. Second, the real atomic structure of the electrodes is not represented in
the jellium model.
In recent years, several first-principles methods have been developed to capture the
basic properties of the challenging problem of nanoscale transport in the nonequilibrium
regime [161–163]. These methods, essentially differing only in their numerical
implementation, share most of the main ingredients of the treatment as they are imposed by the
intrinsic properties of the problem at hand. Density functional theory (DFT) is employed to
describe the atomistic nature of both the device and the electrodes self-consistently, including
exchange–correlation and charging effects. Unlike conventional DFT simulation schemes, in
order to account for the infinite and open nature of the system due to the electrodes, DFT is
used in combination with the nonequilibrium Green function (NEGF) formalism. The NEGF
formalism provides a means for representing the effects of infinite electrodes in the form of
self-energy terms. Also the nonequilibrium charge density of the device under the biased
conditions to be used in DFT can be constructed through the Keldysh nonequilibrium Green
function. It should be noted that this formalism also allows for a computationally efficient
evaluation of the contribution of the bound states to the charge density.
More specifically, a nanodevice coupled to two semi-infinite metallic electrodes can be
considered to be divided into three parts. These are the left (l), right (r) and central (c)
regions. The central (device) region must be chosen to extend far enough into the electrodes
that the effective device potential can be matched at the boundaries with the bulk potential
of the electrodes (see figure 27). The effective potential of the device is generated using the
information on the charge distribution inside the central region ρ
c
(¯ r). The charge distribution
is conveniently calculated from the density matrix ˆ ρ, which can be constructed using the
nonequilibrium Green function as
ˆ ρ = −
i

dε G
-
(ε) (11)
where
G
-
(ε) = G
R
Σ
-
G
A
(12)
with G
R
and G
A
= (G
R
)

denoting the retarded and advanced Green functions of the device
(central region). Σ
-
corresponds to the injected charge fromthe electrodes. It can be expressed
in terms of the Fermi distribution functions f
l
(ε − j
l
) and f
r
(ε − j
r
) of the left and right
electrodes and their retarded self-energies Σ
R
l
and Σ
R
r
as
Σ
-
= −2i Im[ f
l
Σ
R
l
+ f
r
Σ
R
r
] . (13)
Here j
l
and j
r
are the Fermi levels deep in the left and right electrodes such that V
b
=
(j
l
− j
r
)¡e is the external applied bias. The Green functions are evaluated by direct matrix
R940 Topical Review
Figure 27. (a) The metallic electrode–nanotube interface. A semi-infinite (4. 4) armchair SWNT
is in contact with a semi-infinite Al(100) metallic electrode. The device region is coupled to perfect
atomic scale electrodes which extend to reservoirs at z = ±∞. (b) Charge density contours of the
interface at equilibrium. The contours of the bulk lead match that of the device at the edges of the
device region. (Reproduced from [161].)
inversions:
G
R
(ε) = [εI −H −Σ
R
l
−Σ
R
r
]
−1
(14)
where H[ρ
c
(¯ r)] is the Hamiltonian matrix of the central region within DFT and evaluated
using the charge density obtained from equation (11). The cyclic procedure summarized
above should be iterated until NEGF–DFT self-consistency is achieved.
The I –V characteristics of the nanoscale device are calculated using the Landauer formula
I (V
b
) =
2e
h

dε T(ε) [ f
l
(ε −j
l
) − f
r
(ε −j
r
)] . (15)
It should be noted that the transmission function T(ε) depends both on temperature and on the
applied bias and is expressed as
T(ε) = Tr[Γ
l
(ε)G
R
(ε)Γ
r
(ε)G
A
(ε)] (16)
where Γ
p
is the function describing the linewidth due to the coupling of the central device
region to the electrode p. It is related to the corresponding self-energies of the electrodes in
the form
Γ
p
= i(Σ
R
p
−Σ
A
p
). ( p = l. r). (17)
Topical Review R941
z-position (nm)
-200 0 200 400 600
0
1
2
3
G

(
G

)
0
c
Fibre
a
b
Figure 28. (a) A transmission electron micrograph (TEM) image of carbon nanotubes at the end
of a nanotube fibre. (b) A schematic diagram of the experimental set-up. Nanotubes are lowered
under SPM control into a liquid metal surface. (c) Variation of conductance with nanotube fibre
position. Plateaus are observed corresponding to additional nanotubes coming into contact with
the liquid metal. (Reproduced from [15].)
With the development of self-consistent ab initio methods based on DFT and the NEGF
formalism, the transport properties of a broad set of nanoscale systems [164–168], including
C
60
, linear chains of carbon, fullerene and silicon, and gold nanowires, are studied under
nonequilibrium conditions. Some interesting features such as negative differential resistance
in I –V curves and the importance of the detailed atomic structure of the electrodes are reported.
Carbon nanotubes having different electronic structures depending on their chirality and
radius are candidates for being main constituents of the future nanoelectronics technology. A
considerable amount of experimental and theoretical effort has been devoted to investigating
quantum transport properties of nanotubes. Calculations show that infinitely long, perfect,
metallic single-wall carbon nanotubes have an equilibrium conductance of G = 2G
0
, where
G
0
= 2e
2
¡h, due to two subbands crossing at the Fermi level [11, 12]. The existence of
pentagon–heptagon pair defects in infinite nanotubes is shown to alter the conductance pattern
slightly [152]. The discrepancy between the theoretical calculations and the much smaller
values of the nanotube conductance measured in experiments is attributed to poor nanotube–
electrode contacts.
In an interesting transport experiment performed by Frank et al [15] the conductance of
MWNTs, dipped into a liquid metal to minimize the contact resistance effects, were found to
be close to G
0
, half the expected value. Figures 28(a) and (b) show the transmission electron
micrograph (TEM) image of the nanotubes used and the experimental set-up. The conductance
is seen to increase as the nanotube fibre is dipped into the liquid metal in steps of G
0
as shown
in figure 28(c). Each step corresponds to an additional nanotube coming into contact with the
metal. Although narrower, two more steps at fractional G
0
values are also observed.
An explanation of the measured conductance pattern is provided by Choi and co-
workers [153] in their ab initio calculation of the conductance of the (10, 10) SWNT with one
end immersed in a jellium metal. They showed that of the two available channels, it is mostly
π band incident electrons that suffer strong back-scattering at the nanotube–metal contact,
while the π
-
band is responsible for the observed conductance of the system, i.e. G = G
0
. In
another account, using a semiempirical scattering technique Sanvito et al [169] calculated the
conductance of a MWNT with a special geometry and provided a possible explanation of the
measured [15] fractional conductance values in terms of interwall interactions in a MWNT.
Later, Taylor et al [161] simulated transport properties of a SWNT–metal interface using the
R942 Topical Review
NEGF–DFT formalism and a semi-infinite Al electrode of finite cross section (see figure 27).
They showed that there are two transmission eigenvectors which connect the two sides of
the electrode–nanotube interface. The two transmission eigenvalues T
1
and T
2
were smaller
than unity; however, their sum at the Fermi energy gave an equilibrium conductance almost
equal to G
0
for a range of distances between the nanotube and the electrode. This result is
consistent with both the nanotube–jellium calculation [153] and the experiment [15]. In a
recent study Palacios et al [170] addressed the problemof conductance of finite length carbon
nanotubes with realistic metal electrodes using a first-principles method called the Gaussian
embedded cluster method [163]. They observed resonance patterns around the Fermi level
in the conductance profiles. As electrodes, elemental Ti was found to make better high
transparency contacts to carbon nanotubes compared to more traditional metals such as Al
and Au.
The electronic and transport properties of carbon nanotubes can be modified by doping
with foreign atoms. The effects of substitutional nitrogen doping in carbon nanotubes under
nonequilibriumconditions were investigated by Kaunet al [171] using NEGF–DFTformalism.
Nitrogen doping of nanotubes does not cause substitutional structural relaxation; however,
the step-like equilibrium conductance pattern of the nanotube is altered due to the action of
a substantial number of nitrogen atoms as elastic scattering centres. It was found that for
doped zigzag nanotubes even a single nitrogen atom substitution increases the current flow
and for small radii tubes narrows the current gap. Periodic doping of nitrogen atoms in a
semiconducting nanotube makes the structure metallic with a nonlinear I –V characteristic and
a negative differential resistance at higher biases. It is interesting that breaking of symmetry
by the impurities can raise the current flow in semiconducting tubes and reduce the current
slightly in metallic ones. All these nonequilibriumtransport properties can be understood from
the simple physical picture of alignment and misalignment of current carrying bands of the
device due to applied bias.
7.2. Device models
Several novel properties of carbon nanotubes, which have been revealed through the active
experimental and theoretical research during the last decade, have been combined in
nanodevice operation or circuit elements. These are semiconductor heterostructures or metal–
semiconductor junctions; resonant tunnelling double-barrier (RTDB) and multiple-quantum-
well structures (MQWS); metallic nanorings for persistent currents and solenoids; metal–
nanotube contacts and Schottky barriers; interconnects; etc.
SWNT–SWNT junctions. Structural defects can affect the atomic and electronic structure
of nanotubes. The role of topological defects, such as 5¡7 pentagon–heptagon defects, has
been investigated by different groups. Charlier et al [19] carried out an extensive study on
the topological 5¡7 defects in different directions on the surface of the (12. 0) SWNT and
revealed their effects on the atomic and electronic structure by using a recursion method
within the tight binding description of the carbon π bands. They found that the 5¡7 defect
can induce changes in diameter and chirality of the nanotube. For example, if the 5¡7 defect
can be aligned exactly on the axis of the (n. 0) zigzag tube, the chirality remains the same,
but the diameter becomes smaller. If the 5¡7 defect is perpendicular to the axis of the tube,
the diameter is conserved, but the zigzag SWNT becomes chiral. Chirality increases with
increasing number of 5¡7 defects perpendicular to the axis of the tube. The 5¡7 defects
in any orientation on the surface of SWNTs change both the diameter and the chirality. Not
only atomic or geometric structure, but also the electronic structure around the Fermi level gets
Topical Review R943
Figure 29. Energies of the discrete states of the (6. 4)¡m(5. 5)¡(6. 4) quantum dots as a function
of the number of (5. 5) unit cells m. (Reproduced from [20, 173].)
modified by the presence of 5¡7 defects. Most importantly, nanotubes with different chiralities
and different electronic properties and diameters can be connected using such a pair defect to
formdifferent kinds of heterostructures. These heterostructures (such as metal–semiconductor
and semiconductor–semiconductor) can form rectifiers or diodes and hence they can play an
important role in nanoelectronics. In fact, the calculation of electronic transport through such
junctions yielded tunnelling conductance and hence indicated that they can be used as building
blocks of nanoscale electronic devices [172]. Chico et al [173] studied the electronic structure
of (8. 0)–(7. 1) and (8. 0)–(5. 3) junctions, which were formed from a single (8. 0) SWNT
by introducing a single and triple 5¡7 defect pair, respectively. For the case of the (8. 0)–
(7. 1) junction, it corresponds to a heterostructure between a semiconductor having a band
gap of 0.62 eV and a semimetal. The second junction, (8. 0)–(5. 3), is a semiconductor–
semiconductor heterostructure with two different band gaps.
It is clear that by the controlled generation of 5¡7 defects on a chiral SWNT one can form
a variety of SWNT–SWNT junctions with different heterostructure characteristics. Chico
et al [20] went further and proposed an interesting heterostructure, namely a new type of
quantum dot based solely on single SWNTs. It is made from a series of (6. 4)¡m(5. 5)¡(6. 4)
SWNT junctions. Here the (6. 4) tube is a semiconductor; m(5. 5) indicates m unit cells of a
metallic armchair tube. The (6. 4) tube was connected to the (5. 5) tube smoothly, introducing
heptagon–pentagon defects. In this way formation of undesirable interface states has been
eliminated. Normally, a metallic nanostructure between two semiconducting ones forms a
quantumdot and this has quantized electronic states. Using the empirical tight binding method
and Green function matching approach along with the transfer matrix technique, Chico et al
[20] showed that electronic states in the metallic m(5. 5) part are indeed confined and exhibit
a sharp, δ function-like density of states. Moreover, their level spacings vary depending on the
value of m, showing a clear quantum size effect (see figure 29).
Quantum structures generated by radial deformation. The metallization of semiconducting
SWNTs under radial deformation (discussed in section 6) has been proposed to generate
quantum structures [24, 44, 118]. The idea is simple and originates from the numerous works
done on semiconductor heterostructures (A
n
B
m
) and quantum dots [174]. It is known that
R944 Topical Review
owing to the band offsets of the semiconductor heterostructures, the energies of the band states
of one semiconductor Bmay fall into the band gap of the adjacent semiconductor A. According
to the effective mass approximation, the height (depth) of the conduction (valence) band edge
of Ameasured fromthat of B, LE
C
(LE
V
), behaves as a potential barrier for electrons (holes).
For example, m layers of B between n layers of two As forma quantumwell yielding confined
electronic states. The depth of the well and the width of the barrier and well (in terms of
number of layers n and m, respectively) are crucial parameters for monitoring the resulting
electronic properties. By repeated formation of (A
n
B
m
) one generates MQWSs or RTDBs
by combination (A
n
B
m
A
n
). Similar to what one achieves in semiconductor heterostructures,
periodic MWQSs or finite RTDBstructures can be realized on an individual s-SWNT and their
electronic properties can be variably and reversibly monitored.
Performing empirical tight binding and also first-principles calculations, Kili¸ c et al [24]
have shown that a MQWS is generated by applying different radial deformations at adjacent
parts of a (7. 0) s-SWNT and repeating them periodically. They considered a supercell
consisting of 16 unit cells of the (7. 0) SWNT; n of the unit cells were kept undeformed
and the remaining m = 16 −n were radially deformed. Tight binding charge densities of the
band edge states for superlattices corresponding to n = 4, 8, 12 have clearly demonstrated
the confinement of states and quantum well structure formation. Later, the strain induced
quantum well structures were treated by means of SCF first-principles calculations on a
superlattice consisting of 12 (8. 0) SWNT unit cells [18]; half of them were radially strained
with c
yy
=0.16; the remaining half were left undeformed—i.e. c
yy
= 0. Within the superlattice
geometry, the adjacent unstrained and strained parts were repeated yielding a periodic band gap
modulation. Applying the standard method for band line-up explained in detail below [175],
using a SCF crystal potential and referring to the energies at the top of the valence band, the
misfit of the valence band edges between adjacent (i.e. undeformed and deformed) zones has
been found to be ∼180 meV. This indicates a staggered band line-up. As in the empirical tight
binding calculations, it has been demonstrated that quantum structures can be formed on a
single SWNT. Tombler et al [118] have shown that a quantumdot can be produced by bending
(and hence inducing radial deformation of) the SWNT at a given point.
Quantumstructure formation by modulating hydrogenation. Amodel of an electronic device
based on a similar band gap modulation and hence MQWS formation has been proposed
recently [176]. In this case the band gap modulation is provided by the modulating hydrogen
adsorption (see figure 30). The localization of states in one of the zones (either a (C
32
)
l
(i.e. specified by l) zone or a (C
32
H
32
)
q
(i.e. specified by q) zone) is based on the physical
grounds discussed above.
Formation of different zones leading to different electronic structures on the same
(8. 0) SWNT upon modulating the H adsorption, and also the effect of the superlattice
size (i.e. l, q and l + q), is further analysed by examining the electronic potential [176].
The electronic potential which consists of Hartree, exchange and ionic pseudopotentials,
V
e
(r) = V
H
(r) + V
x
(r) + V
ion
(r), is calculated self-consistently. Figure 31 shows the planar
averaged electronic potential
¨
V
e
(z) =
1
a
s
b
s

V
e
(x. y. z) dx dy (18)
and the running average of the planar averaged potential,
¨
V
r.e
(z), along the axis of the SWNT.
Here a
s
and b
s
are superlattice parameters. As seen in figure 31(a), the short periodicity,
l = q = 1, is not sufficient to induce two adjacent zones with markedly different electronic
structures. In figure 31(b) the differences between two zones become pronounced and LV
Topical Review R945
Figure 30. Superlattice structures, [(C
32
)
l
(C
32
H
32
)
q
] formed on a single (8. 0) SWNT by
modulating adsorption of hydrogen atoms. (a) l = q = 1; (b) l = q = 2. The positions of
64 H and 128 C atoms, as well as the superlattice parameter c
s
, are calculated from full structure
optimization by using conjugate gradient method. (Reproduced from [176].)
Figure 31. Averages of calculated self-consistent electronic potentials of the superlattices. (a)
[(C
32
)
1
(C
32
H
32
)
1
]; (b) [(C
32
)
2
(C
32
H
32
)
2
]; (c) [(C
32
)
3
(C
32
H
32
)
2
]; (d) [(C
32
)
2
(C
32
H
32
)
3
]. Dashed
and thick curves are planar averaged,
¨
V
e
(z), and running averaged,
¨
V
r.e
(z), respectively. Relatively
high cell averaged potential zones correspond to (C
32
)
l
. See the text. (Reproduced from [176].)
increases for the superlattice; l = q = 2. Figures 31(c) and (d) indicate that the potential,
¨
V
r.e
(z), becomes flatter as l and q increase from2 to 3 and two adjacent zones display markedly
different electronic structures.
Band line-up of zones, (C
32
)
l
and (C
32
H
32
)
q
, is a complex process involving the relaxation
of the atomic structure and charge transfer at the interfaces and also resulting in modification
R946 Topical Review
(C
32
H
32
)
q
(C
32
H
32
)
q
∆E
V
∆E
C
E
g
l
E
g
q
(C
32
)
l
Figure 32. Aschematic illustration of the energy band diagram and band offsets estimated by using
averaged self-consistent electronic potentials,
¨
V
r.e
(z), calculated for the l and q zones, LE
B.l(q)
and band gaps, E
g1
and E
g2
, calculated for l = ∞ and q = ∞. The band diagram indicates a
type-I band line-up appropriate for a resonant tunnelling double-barrier structure. (Reproduced
from [176].)
of the electronic potential. Experimental and theoretical methods have been proposed in
the past to determine the band offsets and hence to obtain the band diagram perpetuating
along the semiconductor heterostructures. A solution of this problem [175] involves the
following steps. First, the average electronic potentials corresponding to
¨
V
r.e
(z) for finite
l and q, as well as for infinite l and q, (C
32
)
l=∞
and (C
32
H
32
)
q=∞
, are calculated. The
actual atomic positions, especially those at the interface, and the charge transfer between
regions are of crucial importance for the band offsets and are incorporated self-consistently
in the calculated average electronic potential. Secondly, for the case of l = q = ∞
the energy difference between the edge of the valence band E
V
and the average potential,
i.e. LE
B
[
l(q)=∞
= E
V.l(q)=∞

¨
V
r.e
[
l(q)=∞
, are calculated. LE
B
for (C
32
)
l=∞
and (C
32
H
32
)
q=∞
is found to be 2.62 and 3.91 eV, respectively. In the third step, one assumes that LE
B
is
unchanged for finite l and q and determines the valence band edges of a particular superlattice
with given l and q by adding LE
B
[
l(q)=∞
to the corresponding average potential values
¨
V
r.e
of the l and q zones. At the end, the energies of the valence band edges of two adjacent
zones (i.e. hydrogen-free and exohydrogenated) are fixed. Then, the valence band offset
of the superlattice is LE
V
= E
V.l
− E
V.q
. The energies of the conduction band edges
of the superlattice are determined by adding band gaps of different zones (calculated for
l = ∞ and q = ∞) to the energies of the valence band edges, i.e. E
C.l
= E
V.l
+ E
g1
and
E
C.q
= E
V.q
+ E
g2
. Here, E
g1
and E
g2
are the band gaps of hydrogenated and hydrogen-free
zones, respectively. The conduction band offset is then determined by subtracting these latter
energies, i.e. LE
C
= E
C.q
− E
C.l
. Figure 32 presents a schematic illustration of the band
diagram obtained in [176].
The present discussion for the periodically repeating quantum wells can be extended to
finite systems, for example to the resonant tunnelling double barriers, made by a zone (C
32
)
l
placed between two zones of (C
32
H
32
)
q
, i.e. (C
32
H
32
)
q
(C
32
)
l
(C
32
H
32
)
q
. Metallic reservoirs
adjacent to both barriers, i.e. (C
32
H
32
)
q
, and metallic connects of the device can be achieved
by metal coating [102, 103, 109]. Also a metal–semiconductor heterostructure can be formed
by modulating the hydrogenation of an (n. n) armchair SWNT. In this way metallic connects
are provided by the hydrogen-free ends of the tube.
SWNT–metal contact. The switching of the current in a s-SWNT at room temperature by an
external electric field has been utilized to fabricate a newfield effect transistor (FET) [99–101].
The interaction between the s-SWNT and metal electrode and the Schottky barrier, +
B
, formed
Topical Review R947
0
5
– 5
3 6 9 12 15 18 21 24
10
V
e
(
z
)
[
e
V
]
z(Å)
Figure 33. (a) Local densities of states (LDOSs) at different Au and C atoms of the (8. 0) zigzag
SWNT side-bonded to the Au(100) surface. The locations of atoms are shown in the inset. The
zero of energy is taken at the Fermi energy E
F
. State densities shown by empty circles and filled
diamonds correspond to six neighbouring carbon atoms. Other densities are for single atoms.
(b) Contour plots of the SCF electronic potential, V
e
(r), on a vertical plane. (c) As (b) but on a
horizontal plane bisecting s. In the dark grey regions, V
e
(r) > E
F
. (d) The variation of V
e
(z) on
a perpendicular line passing through the centre of the SWNT. (Reproduced from [179].)
thereof have been proposed as the origin of the FET operation. Model calculations have been
carried out to provide further understanding of experimental I –V characteristics [177, 178].
The origin of the potential barrier formed between an s-SWNT and a specific metal
electrode has been investigated recently by using first-principles calculations [179]. The
interaction and energy of binding between an individual Au atomand (8. 0) tube is essential for
the contact. The contact on the gold electrode is formedby placing the s-SWNTon the Au(100)
surface, so a surface Au atom faces the centre of the hexagon. The Au–SWNT equilibrium
distance, s, occurred at 3.18 Å according to full structure optimization. This implies a weak
Au–SWNTinteraction which is characterized by physisorption. The LDOSs calculated for two
Au atoms at different sites (one is at the contact just belowthe SWNThexagon; the second being
furthest from the SWNT mimics the clean Au surface) in figure 33(a) have finite state density
at the Fermi level. These two LDOSs are similar; there are only minute changes, whereas the
LDOS of the carbon atoms at the contact still has a band gap. The Fermi level lies near the
top of the valence band of the SWNT and is consistent with a small +
B
in the hole doping
picture. The LDOS at the carbon atom which is furthest from the metal electrode displays a
state distribution similar to that of the carbon atom in the contact region. The calculated total
charge density of the SWNT bonded to the metal electrode, E
M
, i.e. ρ[SWNT + E
M
], and the
difference charge density, Lρ = ρ[SWNT + E
M
] − ρ[E] − ρ[SWNT], show minute charge
rearrangements. These results indicate that the weak Au electrode–SWNT interaction does
not induce any significant changes in the electronic structure. The SCF electronic potential
between the SWNT and the Au electrode, V
e
(r), is presented on a vertical plane and also on a
horizontal plane bisecting s in figures 33(b) and (c). The shaded area shows that the electronic
potential energy at the contact yields a potential barrier +
c
= V
e
− E
F
> 0. At the contact
midway between the SWNT and Au(100) surface, +
c
is calculated to be ∼3.9 eV—that is,
comparable with the calculated work function (+ ∼ 5 eV) of the Au slab. Figure 33(d) shows
the variation of V
e
(z) on a line passing through the centre of the SWNT and perpendicular
R948 Topical Review
Figure 34. (a) Local densities of states (LDOSs) at different Mo and C atoms of the (8. 0) zigzag
SWNT side-bonded to the Mo(110) surface. The locations of atoms are shown in the inset. The
zero of energy is taken at the Fermi energy E
F
. State densities shown by empty circles and filled
diamonds correspond to eight neighbouring carbon atoms, The others are for single atoms. (b)
Contour plots of the total SCF charge density on a vertical plane. State densities shown by empty
circles correspond to ten neighbouring carbon atoms. (Reproduced from [179].)
to the Au(100) surface. The effective potential barrier, +
c.eff
, can be even higher owing to
the increased confinement of electrons in the contact region. Therefore, electrons which are
transferred from the metal to the s-SWNT have to tunnel through a potential barrier +
c.eff
(r).
The above situation is, however, different in the case of an s-SWNT side-bonded to the
Mo(110) surface, as shown in figure 34. Upon relaxation, the tube has rotated slightly so C
atoms tend to approach Mo atoms. The LDOS at the Mo atom interacting with the C atoms
of the SWNT is different from that at the clean Mo(110) surface. Moreover, the LDOS at the
carbon atomclosest to the Mo surface has finite state density at E
F
. In particular, the peak near
E
F
is associated with the C–Mo bond states and hence may be attributed to the metal induced
gap states (MIGS). The LDOS of the C atoms furthest from the contact has a band gap near
E
F
. This situation indicates that the site of the SWNT forming contact is conducting, while the
opposite site furthest from the contact remains semiconducting. Owing to the strong Mo–C
bond, the spacing between the SWNT and the Mo electrode is smaller (s = 1.96 Å) than that
with the Au electrode. The strong C–Mo bond formation is clarified by the calculated total
SCF charge density in figure 34(b). The analysis of the difference charge density Lρ shows
Topical Review R949
the strong charge transfer to the C–Mo bonds. The electronic potentials calculated on the
horizontal plane bisecting s and also on a line from the SWNT to the Mo(110) surface through
the contact show that V
e
(r) - E
F
in the contact region. Accordingly, a potential barrier, +
c
,
does not form at the extended contact between the SWNT and the Mo electrode.
It appears that the above results distinguish two different types of electronic structure for
two different electrodes. A large spacing and sizable potential barrier between the tube and
metal electrode, +
c
∼ 3.9 eV, are characteristics of the contact made with the Au surface.
This explains why the devices made from Au electrodes have high contact resistance. Here,
weak coupling of electronic states may not lead to MIGS and the (8. 0) SWNT is identified
as being semiconducting even after the contact has been set with the Au surface. Because of
weak couplingand hence finite +
c
, the s-SWNT–Au contact is reminiscent of the metal–oxide–
semiconductor (MOS) junction. A small +
B
is estimated for the p-type character. Upon radial
deformation, s decreases and eventually +
c
collapses. Similar features have been observed
recently in STM studies using multiwall carbon nanotube tips and Au(111) samples [180].
However, +
c
has disappeared due to strong coupling between the states of the s-SWNT and
Mo(110) [91]. The height of the Schottky barrier +
B
that forms at the SWNT–Mo(110) junction
having a finite contact region followed by a free s-SWNT depends on the position where the
Fermi level is pinned in the gap of the s-SWNT. A crude estimate based on the LDOS suggests
that +
B
∼ 0.4 eV for the p-type character. The small cross section of the tube prevents us
from identifying a band diagram across the diameter. But the bands are normally bent along
the axis of the s-SWNT and the height of the barrier is monitored via the applied gate voltage.
Finally, we note that the electronic properties of present metal–SWNT junctions, in particular
the Fermi level pinning, exhibit marked differences from those of metal–Si heterostructures,
which are known to be insensitive to the type of metal.
At this point, we also mention an interesting work by Buldum and Lu [181] on the
intermolecular junctions of SWNTs. They found atomic scale characteristics in the quantum
transport properties of junctions between two SWNTs. In particular, the contact resistance was
strongly dependent on the local atomic configuration of the contact. The optimal electronic
transport between nanotubes occurred when the surfaces of SWNTs were in atomic registry.
However, as illustrated in figure 35, the contact resistance can vary over several orders of
magnitude when the relative atomic configuration at the contact has been changed. Their
predictions were confirmed experimentally [182]. It has been suggested that the monitoring
of contact resistance may lead to new atomic scale switches, resistors and amplifiers.
A similar effect was predicted earlier for an Al chain, which is connected to two flat Al
substrates to a weaker extent. It was found that the quantumconductance of the chain depends
on the position of the Al atom(whether in a top site or a hollowsite) where the chain is attached
to the electrodes [183].
Metal nanorings. Metal nanorings are of particular interest, because they may lead to the
realization of nanosolenoids on SWNTs. The small radius of a metallic nanoring wrapping
the carbon nanotube may lead to interesting electromagnetic properties. The magnetic field B
at the centre of the ring can be expressed in terms of the quantized angular momentum L
z
of
the electrons in the direction parallel to the tube axis
B =
j
0
eL
z
4πmr
3
. (19)
where r is the radius of the nanoring. Taking the lowest possible value for L
z
and r = 5.9 Å,
we estimate B to be of the order of 100 G. The current in the metal ring that can induce
such a high magnetic field is comparable to the current attained in suspended, monatomic
R950 Topical Review
Figure 35. (a) Amodel of a four-terminal junction formed by crossing two nanotubes. The terminal
labels, rotation angle, O, and the translation directions are shown. The tubes are considered to be
rigid. The current is passing between 1 and 4 and the voltage is measured between 2 and 3. (b) The
contact resistance of the (18. 0)–(10. 10) junction as a function of the rotation angle O. The tubes
are in registry at O = 30

, 90

, 150

. (c) The resistance of the (10. 10)–(10. 10) junction. The
tubes are in registry at O = 0

, 60

, 120

, 180

. (d) The resistance of the (18. 0)–(10. 10) junction
as a function of one tube relative to the other in the x and y directions. (Reproduced from [181].)
gold chains [88–90]. Relatively higher magnetic fields of the order of teslas can be induced
by a higher current passing through a thick Ti based metal coating around the SWNT, or by
increasing the number of turns and hence by forming a nanocoil. Miyamoto et al [184] have
examined the chiral conductivity in bare BC
2
N nanotubes. They estimated that a magnetic
field of a few tenths of a tesla can be induced at the centre of the tube by assuming a relaxation
time for carriers ∼50 times larger than that in Cu and a homogeneous chiral current density
confined to the tubule wall.
Metal nanorings winding the SWNT are also important, because they may allow the
generation of persistent current at the nanoscale. Persistent currents in the nanoring can
be started by sudden application of an external magnetic field. In this way it is possible
to use a nanotube, with a ring at its end, as a local magnetic probe at the nanoscale. The
persistent currents induced around a SWNT have been proposed as possible qubits in quantum
computation by Kulik et al [185]. A superconducting ring may also be used for Schr¨ odinger’s
cat experiments where one deals with the superposition of macroscopic quantum states [186].
The two supercurrent quantum states (clockwise and anticlockwise flow) sit in two separate
quantumwells. It has been observed that a weak microwave, which does not breakCooper pairs,
can cause quantumtunnelling between these two macroscopic states. In this kind of experiment
the main problemis isolating the superconducting quantuminterference device (SQUID) from
the outside (nonquantum) environment and that is why isolated carbon nanotubes can be very
useful.
Motivated by these exciting future applications, Bagci et al [109] studied the formation
of Al coverage and Al nanorings around a (8. 0) SWNT. Since aluminium atoms have strong
Al–Al interaction, but not so strong Al–SWNT interaction, they tend to form clusters or
nanoparticles (i.e. 3D islands) on the surface of the SWNT. However, Bagci et al showed that
at special configurations both a stable nanoring and a metal tube can be formed. Moreover,
these structures become stable even if the underlying SWNT is taken out. We first discuss the
zigzag Al nanoring coverage which is obtained by placing Al atoms on top of carbon atoms
Topical Review R951
Figure 36. A view of the optimized structure of the Al zigzag nanoring formed on a (8. 0) SWNT.
(Reproduced from [109].)
(T site), forming a zigzag ring (see figure 36). This structure includes 64 C and 16 Al atoms in
the double unit cell. In this initial configuration the Al–Al distance is 2.33 Å and the angle of
the Al–Al–Al bond is ∼137

. After structure optimization, the Al–Al bond length is increased
to 2.56 Åand the Al–Al–Al bond angle is decreased to 124

, yielding the radius of the nanoring
as 5.9 Å. A side view of the optimized structure of the Al nanoring wrapping the (8. 0) SWNT
is illustrated in figure 36. The binding energyof the Al nanoringis calculated to be 0.85 eV. The
stability of the Al nanoring around the nanotube can be understood from the stable structures
of planar Al monatomic chains. Recent studies on low dimensional structures of metals have
revealed several stable atomic structures in one dimension (1D) [37, 94, 110, 187–191]. The
final optimized structure of the zigzag Al nanoring around the (8. 0) nanotube has structural
parameters similar to those of the planar zigzag structure [110], except that it is rolled on a
cylinder. The SWNT initially serves as a template in the formation of the ring structure and
also increases the stability of the ring by preventing transitions to other relatively more stable
structures. Therefore, the Al nanoring around the SWNT is expected to be stable at room
temperature. Interestingly, the nanoring is also stable by itself, since the positions of Al atoms
do not change significantly upon discarding the underlying carbon nanotube.
The electronic properties of the zigzag Al nanoring system are also quite interesting. The
electronic energy bands of the Al metal ring (without a SWNT) are derived fromthe dispersive
bands of the flat zigzag Al chain [110]. When the flat zigzag Al chain is rolled into a ring, its
bands are zone folded at the I point and they appear as a number of discrete energy levels.
For the case of the Al nanoring wrapping the nanotube, these states are mixed with the states
of the nanotube and give rise to the bands and a density of states, which contributes a metallic
behaviour to the combined Al nanoring and (8. 0) nanotube system. According to Mulliken
analysis, 0.15 electrons are transferred from each Al atom to the SWNT. Most importantly,
the Al ring is a conductor that incorporates two channels with an ideal ballistic quantum
conductance of 4e
2
¡h.
Nanomagnets. In section 5.2, we showed that nanomagnets can formonce the SWNT is fully
covered or filled by the specific transition metal atoms. Therefore, first-principles calculations
by Durgun et al [102, 103] present evidence that one can generate nanomagnets from SWNTs
decorated by transition metal atoms (such as Ti, Ta). From the spintronics and molecular
magnetism point of view, the subject is extremely important and requires further study of
the magnetization of transition metal atom covered SWNTs. Yang et al [111] performed
R952 Topical Review
Figure 37. (a) A tapping mode AFM image of an individual carbon nanotube on top of three Pt
electrodes. (b) A schematic side view of the FET device. A single s-SWNT is contacted by two
electrodes. The Si substrate, which is covered by a layer of SiO
2
3000 Å thick, acts as a back-gate.
(c) The suggested band diagram of the device. The nanotube with a gap of ∼0.6 eV is connected
to the leads with Fermi energy E
F
by tunnelling contacts, indicated by the black vertical bars. At
A and C, the valence band edge is pinned to the Fermi energy of the leads. Owing to a difference
in work function between the tube and the electrodes, the bands bend towards lower energy in
between the electrodes. For V
gate
> 0 the bands bend more strongly, leading to an insulating state;
for V
gate
- 0 the bands are flattened, resulting in a metal-like conductance. (d) Application of a
bias voltage results in suppression of the barrier. (Reproduced from [99].)
first-principles calculations of Co filled and Co coated (9. 0) tubes and found that these
systems exhibit substantial magnetic moments (1.4 j
B
per Co atom) and have a very high
spin polarization, P = (D

(E
F
) − D

(E
F
))¡(D

(E
F
) + D

(E
F
)) (P = 86%) for the spin
valve effect. Dag et al [7] have obtained a significant net magnetic moment for Ti covered
(8. 0), (9. 0) and (6,6) SWNTs.
8. Devices fabricated using carbon nanotubes
The objective of most of the effort made so far in the research on carbon nanotubes has been to
discover a technological application. Nanoelectronic or molecular electronics considered as
ultimate miniaturization [192] has attracted much of the interest. The realization of a single-
molecule device has been challenging due to difficulties in manipulating the molecule and
achieving the electrical contact. Intensive laboratory work combined with theoretical studies
during last decade has made remarkable advances in methods and techniques which were able
to produce novel electronic devices based on carbon nanotubes. These devices have kept the
great expectations for these materials alive and have continued to motivate further research.
Some of these devices will be reviewed in this section.
8.1. Transistors based on carbon nanotubes
In 1998 Tans et al [99] achieved the fabrication of a field effect transistor (FET), which consists
of a semiconducting SWNT connected to two metal electrodes. Asemiconducting Si substrate
covered with a 300 nm layer of thermally grown SiO
2
was used as a back-gate (see figure 37).
The nanotube was able to switch froma conducting to an insulating state at roomtemperature as
Topical Review R953
Figure 38. Two-probe I –V
bias
curves for various values of the gate voltage (V
gate
). Data were taken
at room temperature and in vacuum (∼10
−4
mbar) with the voltage applied to contacts 1 (drain)
and 2 (source). A negative V
gate
leads to ohmic behaviour while a positive V
gate
results in a strong
suppression of the current at low bias voltage and nonlinear I –V
bias
curves at higher bias. Inset:
conductance at V
bias
= 0 as a function of V
gate
. The conductance through this single-molecule
switch can be varied over at least six orders of magnitude and saturates at 10
−6
O
−1
for negative
V
gate
. (Reproduced from [99].)
a result of the voltage applied to the gate. As shown in figure 38, at V
gate
= 0, I –V
bias
exhibited
a small nonlinearity. However, a pronouncedgap-like nonlinearity developed around V
bias
= 0
with increasing V
gate
. At negative V
gate
, the I –V
bias
curve becomes linear with a resistance that
saturates around 1 MO. Achange of 10 Vin V
gate
resulted in modulation of the conductance by
about six orders of magnitude at V
bias
= 0. According to the electronic structure and energy
diagram of the FET modelled by Tans et al in figure 37(c) the difference in work function
will result in bending of the bands of the semiconducting SWNT. A local polarization layer
develops at the interface until the valence band edge of the SWNT aligns with the Fermi
level of the metal electrode. The band bending length was estimated to be 400 nm. It was
assumed that the Fermi level is pinned at the valence band edge of the SWNT at the contact,
since the FET has the same room temperature resistance of ∼1 MO and a similar temperature
dependence. This behaviour of the device, namely a mild temperature dependence, has been
attributed to the absence of a Schottky barrier. Since the proposed band diagram of the device
was reminiscent of that of the barrier injection transit time diode (BARITT), its operation is
qualitatively described in terms of the BARITT model. It has been argued that using special
geometry, reducing the contact resistance and the thickness of SiO
2
, the device would allow
a maximum frequency of the order of 10 THz. Also its gain can be raised by one order if the
thickness of SiO
2
is decreased from3000 to 50 Å. It has also been foreseen that the integration
of multiple devices into a circuit would be possible, by molecular self-assembly techniques.
Later, Zhou et al [98] measured I –V characteristics of devices obtained with s-SWNTs
contacted by 200 Åthick Ni electrodes with 600 Åthick Au on top. The lengths of the SWNTs
between electrodes were less than 30 jm. Degenerately doped Si wafers with 5000 Å thick
thermally grown oxide on the surface were used as the substrate. The heavily doped substrate
is conducting at low temperature and was used as a back-gate. Electrical transport through an
s-SWNT involved thermal activation at high temperature and tunnelling behaviour through a
R954 Topical Review
reverse biased metal–SWNTjunction at lowtemperature. Under high V
bias
, I –V characteristics
of s-SWNTs exhibit pronounced asymmetry with respect to the bias polarity, as a result of
local gating. Different samples exhibited I –V characteristics resembling that of the Si based
p metal–oxide–semiconductor field effect transistor (MOSFET). The positive voltage gain of
∼3 recorded in this device was almost one order of magnitude higher than the maximum gain
of ∼0.35 obtained by Tans et al [99]. The transconductance was measured to be ∼200 nA V
−1
which is two orders of magnitude higher than the similar results obtained by Martel et al [101].
These values indicate great progress.
Both transistors discussed above have been realized by the side bonding of SWNTs. These
devices generally have high contact resistance and unipolar (p-type) transport characteristics.
High contact resistance between the side-bonded s-SWNT and high work-function metal
electrode is in accord with the first-principles calculations [179]. Martel et al [101] realized
an end-bonded SWNT–metal contact. A FET was made of a semiconducting SWNT of
∼8000 Å length and 14 Å diameter contacting TiC and passivated with a uniform SiO
2
layer.
In contrast to the above planar devices, in the latter device the apparent barrier height for
carrier injection is modulated by the gate field. This allowed the FET device to be ambipolar
with low contact resistance for both n- and p-type conduction. Devices showing ohmic I –V
curves in both strong accumulation and inversion regimes have been attributed to the very
small Schottky barrier at the contact. The low barriers for both electrons and holes have
been attributed to the effective barrier heights being strongly modulated by the electrostatic
field of the gate. Because of their quasi-one-dimensional structure the electronic behaviour
of nanotube contacts is found to be sensitive to the electrostatics [193]. Fermi level pinning
appears to be relatively unimportant for nanotubes end-bondedto TiC. The needle-like shape of
the metallic carbide contacts focuses the electric field and hence the gate can induce a stronger
electric field at the junction. As a result, whereas the true Schottky barrier height is as high as
one expects (i.e. +
B
= E
g
¡2), it is so thin that carriers can easily tunnel through the barrier.
The symmetry and narrowness of the Schottky barrier for both holes and electrons in a
FET based on an s-SWNT have led to an interesting application. Injection of electrons and
holes simultaneously through the narrow barriers at the source and drain contacts has been
possible at room temperature. These injected carriers have recombined and emitted photons
with energy hν = E
g
and wavelength λ ∼ 0.8–1.5 jm. Because λ depends on the E
g
of
the s-SWNT, the wavelength of the emitted light can be adjusted easily by using an s-SWNT
with a different radius, or one can tune the emission to any wavelength by radial deformation.
This new result holds promise that SWNTs can enable nanoscale photon sources for future
optoelectronics and fibre optics [194].
The charge carrier mobility is crucial for the applications such as in few- and single-
electron memories and chemical/biochemical sensors. Instead of a Schottky barrier as in
previous studies, the work by Durkop et al [195] exhibited an ohmic contact behaviour.
They presented measurements on long, ohmically contacted nanotube devices, showing field
effect mobility in semiconducting SWNT transistors as high as 79 000 cm
2
V
−1
s
−1
at room
temperature, and the intrinsic mobility in s-SWNTs is even higher. These measured values
exceed those of any other semiconductor at room temperature.
Beyond the single FET, research on SWNTbased electronics has focused on the integration
of FETs on a chip, since the advance requires assembly strategies that allowprecise localization
and interconnection. Soh et al [197] achieved the synthesis of SWNTs by chemical vapour
deposition (CVD) of methane at controlled locations on a substrate using patterned catalytic
islands. The CVDsynthesis and microfabrication techniques were combined to produce a large
number of ohmically contacted nanotube devices with controllable length, placed on a Si wafer
with ∼1 jm thick thermally grown SiO
2
. It has been argued that this fabrication technique
Topical Review R955
allows large numbers of integrated nanotube circuits to be built reliably with a contact resistance
of ∼10 kO (or 20 kO two-terminal resistance in the ballistic regime). Later Franklin et al
[198] devised a method for reliable integration of long suspended SWNTs into electrically
addressable devices, which involves patterned growth of SWNTs to bridge predefined Mo
electrodes. This approach affords SWNT based devices without any postgrowth processing.
In the course of active research towards nanoelectronics, Bachtold et al [199] demonstrated
logic circuits with FETs based on SWNTs. Their transistors show device characteristics such
as high gain (>10), a large on–off ratio (>10
5
) and room temperature operation.
Much recently, Keren et al [196] reported the realization of a self-assembled carbon
nanotube FET operating at room temperature where a DNA scaffold molecule provides the
address for precise localization of an s-SWNT as well as a template for the extended metallic
wires contacting it. In order to get the transistor to self-assemble, they attached a SWNT
onto a specific site on a DNA strand, by binding together with proteins. They then made gold
nanowire out of DNA molecules at each end of the nanotube. The device could be switched
on and off by applying voltage to it.
8.2. Chemical sensors
As discussed in section 3, the adsorption of atoms or molecules such as NO, NH
3
and O
2
on
s-SWNTs has been reported to change the electrical resistance and similar properties. It has
been suggested that this effect of carbon nanotubes can be used in chemical sensors to detect
specific molecules in the environment [62, 63]. Owing to their bulky size and high power
consumption, the miniaturization of chemical sensors is, in fact, very important for various
applications. Modi et al [200] achieved production of a miniaturized gas ionization sensor a
few microns in length using MWNTs. The sharp tips of MWNTs generate very high electric
field at relatively low voltages, lowering breakdown voltages several times in comparison
to those of traditional electrodes. This feature enables compact, battery powered and safe
operation of such sensors. In this sensor, controlled dc voltage is applied between an anode
consisting of vertically aligned MWNT film and a cathode made from Al sheet. A MWNT
film (consisting of MWNTs 250–300 Å in diameter, ∼30 jm long, with ∼500 Å separation)
was grown by CVD on a SiO
2
substrate. An individual MWNT, attached to the substrate in
the head-on position, has a tip radius of 150 Å, which, in turn generates a very high nonlinear
electric field. In this way the formation of a corona is promoted at relatively lowvoltage, which
can be provided by special batteries. The breakdown voltage was found to become lower as
the interelectrode spacing was reduced. The breakdown voltage went down from 280 V (at
150 jm) to 130 V (at 25 jm separation). Such voltages can be easily obtained by series
connection of carbon–zinc batteries. That makes the sensor portable. In the operation of the
sensor, it is found that while the breakdown voltage is independent of the gas concentration, the
current discharge is linearly dependent on the concentration. This sensor and the underlying
technique that it uses are very convenient, because: (i) the precise breakdown voltage provides
the fingerprint for the gas to be identified; (ii) by monitoring the self-sustaining discharge
current, the gas concentration can be determined; (iii) the response time is short [200].
9. Conclusions
Because of growing interest in nanotechnology and the quest for novel nanodevices, study of
carbon nanotubes has seen a tremendous explosion of work. Despite the large number of very
important research works, the scope of this reviewhas been focused on a limited field covering
atomic and electronic structure of single-wall carbon nanotubes.
R956 Topical Review
In this review article, various physical properties of carbon nanotubes and specific
applications using these properties have been investigated. Our discussions have been guided
by first-principles calculations within density functional theory, which were proven to be
appropriate and to yield accurate predictions on the atomic and electronic structure of carbon
nanotubes.
It has been shown that bare nanotubes have many important properties which are suitable
for use in potential technological applications. That a (n. m) SWNT can be metallic or
semiconducting depending on n and m offers a wide range of options. Moreover, these options
are multiplied several times by the functionalization of tubes. Here we dealt with specific
cases wherein the properties of the nanotube undergo a dramatic change upon physisorption of
foreign atoms. Not only functionalization but also radial deformation has been seen to modify
the nanotube properties. A semiconducting nanotube can be metallic under a radial strain
which changes the circular cross section into an elliptical one. Even more interesting is that
these changes can be tuned reversibly. Finally, we discussed howproperties of bare nanotubes
as well as those induced by functionalization or radial deformation can be exploited for the
design of new molecular devices. Transistors and chemical sensors fabricated by using carbon
nanotubes, as explained at the end, show that carbon nanotubes are indeed nanostructures
which may potentially allow several technological applications.
Acknowledgments
This work was partially supported by the National Science Foundation under Grant No INT01-
15021 and T
¨
UB
˙
ITAK under Grant No TBAG-U/13(101T010).
References
[1] Iijima S 1991 Nature 354 56
Iijima S, Ichihashi T and Ando Y 1992 Nature 356 776
[2] Dresselhaus M S, Dresselhaus G and Eklund P C 1996 Science of Fullerenes and Carbon Nanotubes (San
Diego, CA: Academic)
[3] Saito R, Dresselhaus Gand Dresselhaus MS1998 Physical Properties of Carbon Nanotubes (London: Imperial
College)
[4] Zhao X, Ando Y, Liu Y, Jinno M and Suzuki T 2003 Phys. Rev. Lett. 90 187401
[5] Yu M F et al 1998 Science 287 637
[6] Salvetat J P et al 1999 Appl. Phys. A 69 255
[7] Dag S, Durgun E and Ciraci S 2004 Phys. Rev. B 69 121407(R)
[8] Li F, Xia X, Zhao M, Liu X, Huang B, Tan Z and Ji Y 2004 Phys. Rev. B 69 165415
[9] Nevidomskyy A H, Csanyi G and Payne M C 2003 Phys. Rev. Lett. 91 105502
[10] Dresselhaus M S, Dresselhaus G and Saito R 1992 Phys. Rev. B 45 6234
[11] Hamada N, Sawada S and Oshiyama A 1992 Phys. Rev. Lett. 68 1579
[12] Mintmire J W, Dunlap B I and White C T 1992 Phys. Rev. Lett. 68 631
[13] White C T, Robertson D H and Mintmire J W 1993 Phys. Rev. B 47 5485
[14] Ciraci S, Buldum A and Batra I 2001 J. Phys.: Condens. Matter 13 537
[15] Frank S, Poncharal P, Wang Z L and Heer W A 1998 Science 280 1744
[16] Ouyang M, Huang J, Cheung C L and Lieber C M 2001 Science 292 702
[17] Kim W, Choi H C, Shim M, Li Y, Wang D and Dai H 2002 Nano Lett. 2 703
[18] G¨ ulseren O, Yildirim T and Ciraci S 2002 Phys. Rev. 65 153405
[19] Charlier J C, Ebbesen T W and Lambin Ph 1996 Phys. Rev. B 53 11108
[20] Chico L, Lopez Sancho M P and Munoz M C 1998 Phys. Rev. Lett. 81 1278
[21] Collins P G, Zettl A, Bando H, Thess A and Smalley R E 1997 Science 278 100
[22] Bockrath M, Cobden D H, McEuen P L, Chopra N G, Zettl A, Thess A and Smalley R E 1997 Science 275
1922
[23] Bezryadin A, Verschueren A R M, Tans S J and Dekker C 1998 Phys. Rev. Lett. 80 4036
Topical Review R957
[24] Kili¸ c ¸ C, Ciraci S, G¨ ulseren O and Yildirim T 2000 Phys. Rev. B 62 16345
[25] Sim H S, Park C J and Chang K J 2001 Phys. Rev. B 63 073402
[26] G¨ ulseren O, Yildirim T and Ciraci S 2001 Phys. Rev. Lett. 87 116802
[27] Wild¨ oer J W G, Venema L C, Rinzler A G, Smalley R E and Dekker C 1998 Nature 391 59
[28] Venema L C, Janssen J W, Buitelaar M R, Wild¨ oer J W G, Lemay S G, Kouwenhoven L P and Dekker C 2000
Phys. Rev. B 62 5238
[29] Odom T W, Huang J, Kim P and Lieber C M 1998 Nature 391 62
[30] Odom T W, Huang J, Kim P and Lieber C M 2000 J. Phys. Chem. B 104 2794
[31] Blase X, Benedict L X, Shirley E L and Louie S G 1994 Phys. Rev. Lett. 72 1878
[32] Kane C L and Mele E J 1997 Phys. Rev. Lett. 78 1932
[33] Kleiner A and Eggert S 2001 Phys. Rev. B 63 073408
[34] Kleiner A and Eggert S 2001 Phys. Rev. B 64 113402
[35] Payne M C, Teter M P, Allen D C, Arias T A and Joannopoulos J D 1992 Rev. Mod. Phys. 64 1045
[36] Perdew J P and Wang Y 1992 Phys. Rev. B 46 6671
Perdew J P, Chevary J A, Vosko S H, Jackson K A, Pederson M R, Singh D J and Fiolhais C 1992 Phys. Rev.
B 46 6671
[37] Sanchez-Portal D, Artacho E, Soler J M, Rubio A and Ordejon P 1999 Phys. Rev. B 59 12678
[38] Robertson D H, Brenner D W and Mintmire J W 1992 Phys. Rev. B 45 12592
[39] Tibbets G G 1984 J. Cryst. Growth 66 632
[40] Kudin K N, Scuseria G E and Yakobson B I 2001 Phys. Rev. B 64 235406
[41] Yorikawa H and Muramatsu S 1995 Solid State Commun. 94 435
[42] Yorikawa H and Muramatsu S 1995 Phys. Rev. B 52 2723
[43] G¨ ulseren O, Yildirim T, Ciraci S and Kili¸ c ¸ C 2002 Phys. Rev. B 65 155410
[44] Park C J, Kim Y H and Chang K J 1999 Phys. Rev. B 60 10656
[45] Dillon A C, Jones K M, Bekkadahl T A, Kiang C H, Bethune D S and Heben M J 1997 Nature 386 377
[46] Liu C, Fan Y Y, Liu M, Cong H T, Cheng H M and Dresselhaus M S 1999 Science 286 1127
[47] Brown C M, Yildirim T, Neumann D A, Heben M H, Gennett T, Dillon A C, Alleman J L and Fisher J E 1999
Chem. Phys. Lett. 329 956
[48] Ahn C C, Ye Y, Ratnakumar B V, Witham C, Bowman R C and Fultz B 1998 Appl. Phys. Lett. 73 378
[49] Chen C C, Wu X, Lin J and Tan K L 1999 Science 285 91
[50] Wang Q, Challa S R, Sholl D S and Johnson J K 1999 Phys. Rev. Lett. 82 956
[51] Gordillo R C, Boronat J and Casulleras J 2000 Phys. Rev. Lett. 85 2348
[52] Simonyan V V, Diep P and Johnson J K 1999 J. Chem. Phys. 111 9778
[53] Ma Y, Xia Y, Zhao M, Wang R and Mei L 2001 Phys. Rev. B 63 115422
[54] Tada K, Furuya S and Watanabe K 2001 Phys. Rev. B 63 155405
[55] Yildirim T, G¨ ulseren O and Ciraci S 2001 Phys. Rev. B 64 075404
[56] Srivastava D, Brenner D W, Schall J D, Ausman K D, Yu M and Ruoff R S 1999 J. Phys. Chem. B 103 4330
[57] G¨ ulseren O, Yildirim T and Ciraci S 2002 Phys. Rev. B 66 121401
[58] Andriotis A N, Menon M, Srivastava D and Froudakis G 2001 Phys. Rev. B 64 193401
[59] Landauer R 1987 Z. Phys. B 68 217
[60] Chan S P, Chen G, Gong X G and Liu Z F 2001 Phys. Rev. Lett. 87 205502
[61] White C T, Robertson D H and Mintmire J W 1993 Phys. Rev. B 47 5485
[62] Kong J, Franklin N R, Zhou C, Chapline M G, Peng S, Cho K and Dai H 2000 Science 287 622
[63] Collins P G, Bradley K, Ishigami M and Zettl A 2000 Science 287 1802
[64] Dean K A and Chalamala B R 1999 J. Appl. Phys. 85 3832
Dean K A and Chalamala B R 2000 Appl. Phys. Lett. 76 375
[65] Ajayan P M, Ebbesen T W, Ichihashi T, IIjima S, Tanigaki K and Hiura H 1993 Nature 362 522
[66] Ebbesen T W, Ajayan P M, Hiura H and Tanigaki K 1994 Nature 367 519
[67] Morishita K and Takarada T 1997 Carbon 35 977
[68] Jhi S H, Louie S G and Cohen M L 2000 Phys. Rev. Lett. 85 1710
[69] Sorescu D C, Jordan K D and Avouris P 2001 J. Phys. Chem. B 105 11227
[70] Zhu X Y, Lee S M, Lee Y H and Frauenheim T 2000 Phys. Rev. Lett. 85 2757
[71] Moon C-Y, Kim Y-S, Lee E-C, Jin Y-G and Chang K J 2002 Phys. Rev. B 65 155401
[72] Park N, Han S and Ihm J 1999 Phys. Rev. B 64 125401
[73] Mann D J and Halls M D 2002 J. Chem. Phys. 116 9014
[74] Ricca A and Drosco J A 2002 Chem. Phys. Lett. 362 217
[75] Dag S, G¨ ulseren O, Yildirim T and Ciraci S 2003 Phys. Rev. B 67 165424
[76] Giannozzi P, Car R and Scoles G 2003 J. Chem. Phys. 118 1003
R958 Topical Review
[77] Dag S, G¨ ulseren O and Ciraci S 2003 Chem. Phys. Lett. 380 1
[78] Lee S M, Lee Y H, Hwang Y G, Hahn J R and Kang H 1999 Phys. Rev. Lett. 82 217
[79] Lifshitz E M 1956 Zh. Eksp. Teor. Fiz. 29 94
Lifshitz E M 1956 Sov. Phys.—JETP 2 73 (Engl. Transl.)
[80] Israelachvili J N 1985 Intermolecular and Surface Forces (London: Academic)
[81] Ciraci S, Tekman E, Baratoff A and Batra I P 1992 Phys. Rev. B 46 10411
[82] Halgren T A 1992 J. Am. Chem. Soc. 114 7827
[83] Ulbricht H, Moos G and Hertel T 2002 Phys. Rev. B 66 075404
[84] Derycke V, Martel R, Appenzeller J and Avouris Ph 2002 Appl. Phys. Lett. 80 2273
[85] Delaney P, Choi H J, Ihm J, Louie S G and Cohen M L 1998 Nature 391 466
Delaney P, Choi H J, Ihm J, Louie S G and Cohen M L 1999 Phys. Rev. B 60 7899
[86] Chan S P, Chen G, Gong X G and Liu Z F 2003 Phys. Rev. Lett. 90 086403
[87] Chiang I V, Brinson B E, Smalley R E, Magrave J L and Hauge H 2001 J. Phys. Chem. B 105 1157
[88] Agra¨ıt N, Rodrigo J G and Vieira S 1993 Phys. Rev. B 47 12345
[89] Ohnishi H, Kondo Y and Takayanagi K 1998 Nature 395 783
[90] Yanson A I, Bollinger G R, Brom H E, Agr¨ ait N and Ruitenbeek J M 1998 Nature 395 783
[91] Ciraci S and Tekman E 1989 Phys. Rev. B 40 R11969
Tekman E and Ciraci S 1991 Phys. Rev. B 43 7145
[92] Buldum A, Ciraci S and Fong C Y 2000 J. Phys.: Condens. Matter 12 3349
Buldum A, Leitner D M and Ciraci S 1999 Europhys. Lett. 47 208
[93] Ozpineci A and Ciraci S 2001 Phys. Rev. B 63 125415
[94] G¨ ulseren O, Ercolesi F and Tosatti E 1998 Phys. Rev. Lett. 80 3775
Tosatti E and Prestipino S 2000 Science 289 561
[95] Yao Z, Kane C L and Dekker C 2000 Phys. Rev. Lett. 84 4613
[96] Dai H, Wong E W, Lu Y Z, Fan S and Lieber C M 1995 Nature 375 769
Han W Q, Fan S S and Li Q Q 1997 Science 277 1287
[97] Zhang Y and Dai H 2000 Appl. Phys. Lett. 77 3015
Zhang Y, Franklin N W, Chen R J and Dai H 2000 Chem. Phys. Lett. 331 35
[98] Zhou C, Kong J and Dai H 2000 Phys. Rev. Lett. 84 5604
[99] Tans S J, Verscheuren A R M and Dekker C 1998 Nature 393 49
Bachtold A, Hadley P, Nakanishi T and Dekker C 2001 Science 294 1317
Nakanishi T, Bachtold A and Dekker C 2002 Phys. Rev. B 66 073307
[100] Soh H T, Quate C, Morpurgo A F, Marcus C M, Kong J and Dai H 1999 Appl. Phys. Lett. 75 627
Zhou C, Kong J and Dai H 2000 Appl. Phys. Lett. 76 1597
[101] Martel R, Derycke V, Lavoie C, Appenzeller J, Chan K K, Tersoff J and Avouris Ph 2001 Phys. Rev. Lett. 87
256805
Avouris Ph, Martel R, Derycke V and Appenzeller J 2002 Physica B 323 6
Martel R, Schmidt T, Shea H R, Hertel T and Avouris Ph 1998 Appl. Phys. Lett. 73 2447
[102] Durgun E, Dag S, Bagci V M K, G¨ ulseren O, Yildirim T and Ciraci S 2003 Phys. Rev. B 67 201401(R)
[103] Durgun E, Dag S, Ciraci S and G¨ ulseren O 2004 J. Phys. Chem. B 108 575
[104] Zhang X X, Wen G H, Huang S, Dai L, Gao R and Wang Z L 2001 J. Magn. Magn. Mater. 231 L9
Satishkumar B C, Govindaraj A, Vanitha P V, Raychaudhuri A K and Rao C N R 2002 Chem. Phys. Lett. 362
301
[105] Winter M, Besenhard J O, Spahr M E and Novac M P 1998 Adv. Mater. 10 7259
[106] Gao B, Kleinhammer A, Tang X P, Bower C, Fleming L, Wu Y and Zhou O 1999 Chem. Phys. Lett. 307 1536
[107] Zhao J, Buldum A, Han J and Lu J P 2000 Phys. Rev. Lett. 85 1706
[108] Yang C-K, Zhao J and Lu J P 2002 Phys. Rev. B 66 041403(R)
[109] Bagci V M K, G¨ ulseren O, Yildirim T, Gedik Z and Ciraci S 2002 Phys. Rev. B 66 045409
[110] Sen P, Ciraci S, Buldum A and Batra I P 2001 Phys. Rev. B 64 195420
[111] Yang C-K, Zhao J and Lu J P 2003 Phys. Rev. Lett. 90 257203
[112] Treacy M M J, Ebbesen T W and Gibson J M 1996 Nature 381 678
[113] Ruoff R S, Tersoff J, Lorents D C, Subramoney S and Chan B 1993 Nature 364 514
[114] Chopra N G, Benedict L X, Crespi V H, Cohen M L, Louie S G and Zettl A 1995 Nature 377 135
[115] Rochefort A, Avouris P, Lesage F and Salahub D R 1998 Chem. Phys. Lett. 297 45
[116] Rochefort A, Avouris P, Lesage F and Salahub D R 1999 Phys. Rev. B 60 13824
[117] Liu L, Jayanthi C S, Tang M, Wu S Y, Tombler T W, Zhou C, Alexseyev L, Kong J and Dai H 2000 Phys. Rev.
Lett. 84 4950
[118] Tombler T W, Zhou C, Alexseyev L, Kong J, Dai H, Liu L, Jayanthi C S, Tang M and Wu S 2000 Nature 405
769
Topical Review R959
[119] Chesnokov S A, Nalimova V A, Rinzler A G, Smalley R E and Fischer J E 1999 Phys. Rev. Lett. 82 343
[120] Lambert P E, Zhang P and Crespi V 2000 Phys. Rev. Lett. 84 2453
[121] Minot E D, Yaish Y, Sazonova V, Park J-Y, Brink M and McEuen P L 2003 Phys. Rev. Lett. 90 156401
[122] Salvetat J P, Briggs G A D, Bonard J, Bacsa R R, Kulik A J, St¨ ockli T, Burnham N A and Forr´ o L 1999 Phys.
Rev. Lett. 82 944
[123] Yildirim T, G¨ ulseren O, Kili¸ c ¸ C and Ciraci S 2000 Phys. Rev. B 62 12648
[124] Yang L, Anantram M P, Han J and Lu J P 1999 Phys. Rev. B 60 13874
[125] Yang L and Han J 2000 Phys. Rev. Lett. 85 154
[126] Shen W, Jiang B, Han B S and Xie S 2000 Phys. Rev. Lett. 84 3634
[127] Lordi V and Yao N 1998 J. Chem. Phys. 109 2509
[128] Yu M F, Kowalewski T and Ruoff R S 2000 Phys. Rev. Lett. 85 1456
[129] Mazzoni M S C and Chacham H 2000 Appl. Phys. Lett. 76 1561
[130]
¨
Ozt¨ urk Y and Ciraci S 2004 at press
[131] Rao A M et al 1997 Appl. Phys. A 64 231
[132] Venkateswaren U D, Rao A M, Richter E, Menon M, Rinzler A, Smalley R E and Eklund P C 1999 Phys. Rev.
B 59 10928
Sood A K, Teredesai P V, Muthu D V S, Sen R, Gavindaraj A and Rao C N 2000 Chem. Phys. Lett. 319 296
Teredesai P V, Sood A K, Muthu D V S, Sen R, Govindaraj A and Rao C N R 2000 Chem. Phys. Lett. 310 296
[133] Peters M J, McNeil L E, Lu J P and Kuhn D 2000 Phys. Rev. B 61 5939
[134] Tang J, Qin L C, Sasaki T, Yudasaka M, Matsushita A and Iijima S 2000 Phys. Rev. Lett. 85 1887
[135] Pernas P L, Martin-Rodero A and Flores F 1990 Phys. Rev. B 41 8553
[136] Tian W and Datta S 1994 Phys. Rev. B 49 5097
[137] de Parga A L V, Hernan O S, Miranda R, Yeyati A L, Mingo N, Martin-Rodero A and Flores F 1998 Phys.
Rev. Lett. 80 357
[138] Emberley E G and Kirczenow G 1999 Phys. Rev. B 60 6028
[139] Brandbyge M, Kobayashi N and Tsukada M 1999 Phys. Rev. B 60 17064
[140] Mehrez H, Taylor J, Guo H, Wang J and Roland C 2000 Phys. Rev. Lett. 84 2682
[141] Roland C, Nardelli M B, Wang J and Guo H 2000 Phys. Rev. Lett. 84 2921
[142] Wei Y, Wang J, Guo H, Mehrez H and Roland C 2001 Phys. Rev. B 63 195412
[143] Mehrez H, Guo H, Wang J and Roland C 2001 Phys. Rev. B 63 245410
[144] Nardelli M B 1999 Phys. Rev. B 60 7828
[145] Orlikowski D, Mehrez H, Taylor J, Guo H, Wang J and Roland C 2001 Phys. Rev. B 63 155412
[146] Chico L, Benedict L X, Louie S G and Cohen M L 1996 Phys. Rev. B 54 2600
[147] Wan C C, Mozos J L, Taraschi G, Wang J and Guo H 1997 Appl. Phys. Lett. 71 419
[148] Wan C C, Mozos J L, Wang J and Guo H 1997 Phys. Rev. B 55 13393
[149] Wang J, Guo H, Mozos J L, Wan C C, Taraschi G and Zheng Q 1998 Phys. Rev. Lett. 80 4277
[150] Taraschi G, Mozos J L, Wan C C, Guo H and Wang J 1998 Phys. Rev. B 58 13138
[151] Mozos J L, Wan C C, Taraschi G, Wang J and Guo H 1997 Phys. Rev. B 56 4351
[152] Choi H J and Ihm J 1999 Phys. Rev. B 59 2267
[153] Choi H J, Ihm J, Yoon Y G and Louie S G 1999 Phys. Rev. B 60 14009
[154] Choi H J, Ihm J, Louie S G and Cohen M L 2000 Phys. Rev. Lett. 84 2917
[155] Lang N D 1995 Phys. Rev. B 52 5335
[156] Hirose K and Tsukada M 1995 Phys. Rev. B 51 5278
[157] Lang N D and Avouris P 1998 Phys. Rev. Lett. 81 3515
[158] Yaliraki S N, Kemp M and Ratner M A 1999 J. Am. Chem. Soc. 121 3428
[159] Di Ventra M, Pantelides S T and Lang N D 2000 Phys. Rev. Lett. 84 979
[160] Landman U, Barnett R N, Scherbakov A G and Avouris P 2000 Phys. Rev. Lett. 85 1958
[161] Taylor J, Guo H and Wang J 2001 Phys. Rev. B 63 245407
[162] Brandbyge M, Mozos J L, Ordejon P, Taylor J and Stokbro K 2002 Phys. Rev. B 65 165401
[163] Palacios J J, Perez-Jimenez A J, Louis E, SanFabian E and Verges J A 2002 Phys. Rev. B 66 035322
[164] Taylor J, Guo H and Wang J 2001 Phys. Rev. B 63 121104
[165] Larade B, Taylor J, Mehrez H and Guo H 2001 Phys. Rev. B 64 075420
[166] Roland C, Meunier V, Larade B and Guo H 2002 Phys. Rev. B 66 035332
[167] Taylor J, Brandbyge M and Stokbro K 2002 Phys. Rev. Lett. 89 138301
[168] Mehrez H, Wlasenko A, Larade B, Taylor J, Grutter P and Guo H 2002 Phys. Rev. B 65 195419
[169] Sanvito S, Kwon Y-K, Tom´ anek D and Lambert C J 2000 Phys. Rev. Lett. 84 1974
[170] Palacios J J, P´ erez-Jim´ enez A J, Louis E, SanFabi´ an E and Verg´ es J A 2003 Phys. Rev. Lett. 90 106801
[171] Kaun C-C, Larade B, Mehrez H, Taylor J and Guo H 2002 Phys. Rev. B 65 205416
R960 Topical Review
[172] Saito R, Dresselhaus G and Dresselhaus M S 1996 Phys. Rev. B 53 2044
[173] Chico L, Crespi V H, Benedict L X, Louie S G and Cohen M L 1996 Phys. Rev. Lett. 76 971
[174] Esaki L (ed) 1991 Highlights in Condensed Matter Physics and Future Prospects (NATO Advanced Science
Institute vol 285) (New York: Plenum)
[175] Colombo L, Resta R and Baroni S 1991 Phys. Rev. B 44 5572
[176] G¨ ulseren O, Yildirim T and Ciraci S 2003 Phys. Rev. B 68 115419
[177] Rochefort A, Di Vientra M and Avouris P 2001 Appl. Phys. Lett. 78 2521
[178] Heinze S, Tersoff J, Martel R, Derycke V, Appenzeller J and Avouris Ph 2002 Phys. Rev. Lett 89 106801
[179] Dag S, G¨ ulseren O, Yildirim T and Ciraci S 2003 Appl. Phys. Lett. 83 3180
[180] Luna M et al 2003 Ultramicroscopy 96 83
[181] Buldum A and Lu J P 2001 Phys. Rev. 63 161403
[182] Paulson S, Helser A, Nardelli M B, Taylor R M, Falvo M R, Superfine R and Washburn S 2000 Science 290
1742
[183] Mehrez H, Ciraci S and Batra I P 1997 Phys. Rev. B 55 R1981
[184] Miyamoto Y, Louie S G and Cohen M 1996 Phys. Rev. Lett. 76 2121
Miyamoto Y 1996 Phys. Rev. B 54 11149
Miyamoto Y, Rubio A, Louie S G and Cohen M L 1999 Phys. Rev. B 60 13885
[185] Kulik I O, Hakioglu T and Barone A 2002 Eur. Phys. J. B 30 219
[186] Friedman J R, Patel V, Chen W, Tolpygo S K and Lukens J E 2000 Nature 406 43
[187] Mehrez H and Ciraci S 1997 Phys. Rev. B 56 12632
[188] Tosatti E, Prestipino S, Kostlmeier S, Dal Corso A and Di Tolla F D 2001 Science 291 288
[189] Okamoto M and Takayanagi K 1999 Phys. Rev. B 60 7808
[190] Tolla F D, Corsa A D, Torres J A and Tosatti E 2000 Surf. Sci. 454–456 947
[191] H¨ akkinen H, Barnett R N, Scherbakov A G and Landman U 2000 J. Phys. Chem. B 104 9063
[192] Aviran A and Ratner M A 1974 Chem. Phys. Lett. 29 277
[193] Leonard F and Tersoff J 2000 Phys. Rev. Lett. 84 4693
[194] Avouris Ph 2003 Science 300 783
[195] Durkop T, Getty S A, Cobas E and Fuhrer S 2004 Nano Lett. 4 35
[196] Keren K, Berman R S, Buchstab E, Sivan U and Braun E 2003 Science 302 1380
[197] Soh H T, Quate C F, Morpugo A F, Marcus C M, Kong J and Dai H 1999 Appl. Phys. Lett. 75 627
[198] Fraklin N R, Wang Q, Tombler T W, Javey A, Shim M and Dai H 2002 Appl. Phys. Lett. 81 913
[199] Bactold A, Hadley P, Nakanishi T and Dekker C 2001 Science 297 1317
[200] Modi A, Koratkar N, Lass E, Wel B and Ajayan P M 2003 Nature 424 171

INSTITUTE OF PHYSICS PUBLISHING J. Phys.: Condens. Matter 16 (2004) R901–R960

JOURNAL OF PHYSICS: CONDENSED MATTER PII: S0953-8984(04)77483-4

TOPICAL REVIEW

Functionalized carbon nanotubes and device applications
¨ S Ciraci1 , S Dag1 , T Yildirim2 , O Gulseren1 and R T Senger1,3
1 2

Department of Physics, Bilkent University, 06800 Ankara, Turkey NIST Center for Neutron Research, National Institute of Standards and Technology, Gaithersburg, MD 20899, USA 3 ¨ ITAK-UEKAE, P.K. 74, 41470 Gebze, Kocaeli, Turkey TUB˙ E-mail: ciraci@fen.bilkent.edu.tr

Received 8 March 2004 Published 9 July 2004 Online at stacks.iop.org/JPhysCM/16/R901
doi:10.1088/0953-8984/16/29/R01

Abstract Carbon nanotubes, in which the two-dimensional hexagonal lattice of graphene is transformed into a quasi-one-dimensional lattice by conserving the local bond arrangement, provide several structural parameters for engineering novel physical properties suitable for ultimate miniaturization. Recent interest in nanoscience and nanotechnology has driven a tremendous research activity in carbon nanotubes, which has dealt with a variety of problems and produced a number of new results. Most of the effort has gone into revealing various physical properties of nanotubes and functionalizing them in different ways. This paper covers a narrow region in this enormous research field and reviews only a limited number of recent studies which fit within its scope. First, we examine selected physical properties of bare carbon nanotubes, and then study how the mechanical and electronic properties of different tubes can be modified by radial strain, structural defects and adsorption of foreign atoms and molecules. Magnetization of carbon nanotubes by foreign atom adsorption has been of particular interest. Finally, we discuss specific device models as well as fabricated devices which exploit various properties of carbon nanotubes. (Some figures in this article are in colour only in the electronic version)

Contents 1. Introduction 2. Atomic and electronic structure 2.1. Geometric structure and energetics of bare SWNTs 2.2. Energy band structure 3. Hydrogenation of carbon nanotubes
0953-8984/04/290901+60$30.00 © 2004 IOP Publishing Ltd Printed in the UK

902 904 904 906 908
R901

R902

Topical Review

4. Oxygenation of carbon nanotubes 4.1. Physisorption of O2 molecules 4.2. Chemisorption of oxygen atoms 5. Adsorption of individual atoms on SWNTs 5.1. Binding geometry and binding energies 5.2. Electronic structure 5.3. Transition element covered or filled SWNTs 6. Radial deformation of carbon nanotubes 6.1. Elasticity 6.2. Effect on the electronic structure 6.3. Effect on the chemical reactivity 6.4. Effect of pressure on nanoropes 7. Devices based on carbon nanotubes 7.1. Ab initio methods in transport calculations 7.2. Device models 8. Devices fabricated using carbon nanotubes 8.1. Transistors based on carbon nanotubes 8.2. Chemical sensors 9. Conclusions Acknowledgments References 1. Introduction

912 913 916 917 919 923 926 927 928 930 932 934 936 936 942 952 952 955 955 956 956

Research on carbon nanotubes is ever intensifying in diverse fields of science and engineering in spite of the twelve years that have passed since its first discovery by Iijima [1]. There are several reasons that so much interest has been focused on these materials. First of all, carbon nanotubes have been a natural curb for several research programmes, which were tuned to C60 but all of a sudden came to an end without any great technological applications having been found. Secondly, researchers, who can touch and relocate individual atoms have been challenged to discover the novel properties of these strange materials in order to transform them into new devices or use them in other technological applications. As a result of the rapid rise in the speed, as well as the rapid reduction in the size of electronic devices, new paradigms will be needed to overcome the barriers set by the traditional technologies to produce ever smaller and faster devices. Extensive research dealing with the modification of electronic structure for desired device operations has indicated that carbon nanotubes can be considered as a new frontier in the search for the ultimate miniaturization of electronic circuits with ultrahigh density components and new functionalities. Several devices fabricated so far with different functionalities appear to meet the great expectations for carbon nanotubes. Carbon nanotubes are unique materials, which offer a variety of structural parameters for engineering their physical and chemical properties [2, 3]. They can be synthesized as singlewall (SWNT) or multiple-wall (MWNT) nanotubes; they can form ropes or even crystals. Even an ultimate one-dimensional carbon chain at the centre of a MWNT (and stabilized by the innermost nanotube) has been discovered in cathode deposits [4]. SWNTs are basically rolled graphite sheets, which are characterized by two integers (n, m) defining the rolling (or chiral vector) C = na1 + ma2 , in terms of the two-dimensional (2D) hexagonal Bravais lattice vectors of graphene, a1 and a2 . Then the radius of the tube is given in terms of (n, m) by the relation R = a0 (n 2 + m 2 + nm)1/2 /2π, where a0 = |a1 | = |a2 |. SWNTs exhibit different electronic structures depending on n and m (i.e. on their chirality and radius).

and the effect of their adsorption on the electronic structure. we have investigated the . the scope of this review article has been kept necessarily limited. In an effort to discover new features of technological interest. The organization of the article is as follows. At the same time the chemical activity of the surface of the tube undergoes a change. the interaction of adatoms with the SWNT occurs differently at high and low curvature sites. this section is important for the magnetic properties of functionalized nanotubes. The metal–semiconductor transition induced by elastic deformation has important implications. MWNTs. at the same time they are very strong axially with high yield strength [5. a semiconducting zigzag tube may become a metal with very high state density. and hence has net spin. For example. and carries great potential in tailoring new nanostructures for engineering them according to a desired application. A free SWNT. A selectable functionalization of the (5. carbon nanotubes offer many parameters to deal with and many options for generating properties suitable for a desired functionality. A recent study demonstrates that a semiconducting zigzag tube becomes both a magnetic and a high conductance wire as a result of Ti coating [7]. Physical and chemical properties of a SWNT can also be modified by the adsorption of foreign atoms or molecules. Hydrogenation and oxygenation of SWNTs have been dealt separately in sections 3 and 4. The subject matter that we have left out is in no way less significant than what we have included. functionalization of carbon nanotubes. However. We followed a logical order that starts from fundamental aspects and ends with technological applications. Since the ground state for most of the transition metal atoms adsorbed on the surface of SWNTs is magnetic. We then examined various methods which are used to modify the properties of SWNTs to generate new nanostructures. Suitably doped carbon nanotubes can be functionalized by selectively forming chemical bonds with ligands at the chemically active impurity site [9]. This review is confined rather to the functionalized SWNTs which have their electronic and magnetic properties modified. This process is usually named functionalization.Topical Review R903 The mechanical properties of carbon nanotubes are striking. respectively. we have presented a discussion of electronic structure obtained from firstprinciples calculations together with a comparison made to empirical studies. which resulted in many papers. We first established a background concerning the atomic and electronic structure of various SWNTs. where their binding structures and binding energies. In section 6.5) SWNT resulting from CHn (n = 1–3) adsorption and decoration gives rise to a substantial change in the density of states [8]. One of the great challenges of research on carbon nanotubes has been the realization of nanometre-sized optoelectronic devices and nanomagnets. we discussed how these properties have been used to model devices. as well as due to the relevance to hydrogen storage.several theoretical and experimental studies actively explored SWNTs. In section 2. which is normally nonmagnetic. Finally. Apparently. Clearly. ropes and their functionalized forms. is an extensive field incorporating several recent studies in diverse fields. have been investigated. Even more striking is the response of the electronic structure to the radial deformation leading to dramatic changes. Their strength far exceeds that of any other fibre. due to its focus compounded with the space limitations. a semiconducting zigzag tube becomes metallic with finite state density at the Fermi level as a result of radial deformation transforming the circular cross section into an ellipse. in particular biological and chemical functionalization. Section 5 has been be devoted to the individual adsorption of 24 different atoms (ranging from alkali and simple metal atoms to group IV atoms and most transition metal atoms). becomes magnetic with unpaired spins upon the adsorption of oxygen molecules or specific transition metal atoms. 6]. As has been predicted theoretically and confirmed experimentally. owing to the several recent papers in this area. depending on the pattern of hydrogen atom coverage. while a metallic armchair SWNT can be transformed to a wide band gap semiconductor. They are flexible and can sustain large elastic deformations radially.

since the electronic properties can be controlled by doping or implementing defects in nanotube based optoelectronic devices [19–26].1. The variations of the normalized bond lengths (i. 0) zigzag SWNTs are generally semiconductor and are only metal if n is an integer multiple of 3. In section 7. Later. and have examined how the electronic and chemical properties are modified. in which dimensionality is reduced from two to one. Subsequently. It is about 12◦ less than 120◦ for the (4. the structural parameters of SWNTs deviate from those of graphene. become significant at small radii. The inset to figure 1 shows a schematic side view of a zigzag SWNT which indicates two types of C–C bonds (d1 and d2 ) and C–C–C bond angles ( 1 and 2 ). The zigzag bond angle (θ1 ) decreases with decreasing radius. For example. Extensive theoretical analysis of the band structure of SWNTs together with the curvature effects on geometric and electronic structure has been carried out recently [18] by using first-principles pseudopotential plane wave method [35] calculations within the generalized gradient approximation (GGA) [36]. 10–14]. Recent analytical studies [32–34] showed the importance of curvature effects in carbon nanotubes. Atomic and electronic structure The electronic band structure of SWNTs can be deduced by mapping the band structure of graphene in a 2D hexagonal lattice onto a cylinder [3. 0) tube is. 2. The semiconducting behaviour of SWNTs has been of particular interest. The angle involving this latter bond (θ2 ) is almost constant. band calculations performed by using different methods have been at variance on the values of the band gap. and exhibit 1D quantum conduction [14. . As pointed out earlier for the armchair SWNTs [37]. recent studies [16. 18] have shown much more complicated structural dependence. Nonetheless. first-principles calculation [31] within the local density approximation (LDA) showed that the σ ∗ –π ∗ hybridization becomes significant at small R (or at high curvature). The (n. experimental data [27–30] on the band gaps were extrapolated to confirm the inverse proportionality with the radius of the nanotube [13]. Here. while the length of the corresponding zigzag bonds (d2 ) increases with decreasing R and the length of the parallel bond (d1 ) decreases to a lesser extent with decreasing R. The paper is concluded with outlooks and prospective developments. in fact. Geometric structure and energetics of bare SWNTs Because of cylindrical symmetry.e. and π valence bands crossing at the Fermi level. the (9. respectively. Section 8 reviews briefly the recent progress made in the fabrication of electronic devices based on carbon nanotubes and has described a few such devices produced. the curvature effects. the SWNT presents an interesting example. Band calculations of SWNTs were initially performed by using a one-band π orbital tight binding model [11].R904 Topical Review effects of radial deformation. Both the bond lengths and the bond angles display a monotonic variation and approach the graphene values as the radius increases. 15]. a small band gap semiconductor. In this respect. d1 /d0 and d2 /d0 where d0 is the optimized C–C bond length in graphene) and the bond angles with tube radius R (or n) are shown in figures 1(a) and (b). we have reviewed recent developments in transport theory and have presented some device models developed using the features and physical properties discussed in previous sections. however. the atomic structure. elasticity and electronic structure and binding of adatoms under radial deformation have been discussed. 2. Although the overall electronic structure of SWNTs has been described by this simple picture. The analysis based on the band folding indicates that the (n. 0) SWNT. n) armchair nanotubes are always metal with π ∗ conduction.

d2 . indicating two types of C–C bonds and C–C–C bond angles. The associated strain energy. As expected. (b) The bond angles (θ1 and θ2 ) versus R. curvature energies obtained from first-principles calculations yield a perfect fit to the relation α/R 2 as seen in figure 1(c). Similar trends also exist for the armchair tubes. and eventually saturates to the cohesive energy of graphene. is calculated as the difference of total energy per carbon atom between the bare SWNT and n 15.e. E cur = E T. (Reproduced from [18]. and are calculated from the expression E coh = E T [C] − E T [SWNT] in terms of the total energy of the free carbon atom. (c) The curvature energy. These are labelled as d1 . E cur per carbon atom with respect to graphene as a function of the tube radius. which is specified as the curvature energy.Topical Review R905 Figure 1. where α = Y t 3 /24. and is the atomic volume.14 eV Å2 /atom [18]. (a) Normalized bond lengths (d1 /d0 and d2 /d0 ) versus the tube radius R (d0 = 1.) An internal strain is implemented upon the formation of tubular structure from the graphene sheet. The cohesive energies of SWNTs are also curvature dependent.41 Å). Inset: a schematic side view of a zigzag SWNT. . In this fit α is found to be 2. θ1 and θ2 . Here Y is the Young’s modulus. The calculated curvature the graphene (i. Interestingly. In classical theory of elasticity the curvature energy is given by the following expression [38–40]: E cur = α/R 2 .SWNT − E T. it is small for small n and increases with n. The solid curves are the fit to the data as α/R 2 . t is the thickness of the tube. E cur is positive and increases with increasing curvature. For a zigzag tube. E cur . and the total energy of a SWNT per carbon atom.graphene ) for 4 energies are shown in figure 1(c).

80 0. The second and third rows from [31] give LDA results. Each vector is specified by the (n. (12. In figure 2.79 1. m) indices indicates whether a SWNT is a semiconductor or metal.12 1. m) SWNT has been predicted to be metallic when n − m = 3 × integer. where the band gaps E g of (n. M denotes the metallic state.643 0. respectively. as a function of radius R for (n.11 8 3. fold to the point of the tube [11.05 0. 0) zigzag nanotubes.21 7 2.14 0.764 11 4.625 14 5.33 9 3.045 10 3.093 0. 0) zigzag tubes become metallic when n is a multiple of 3.66 M 5 2.96 0.736 15 5.02 M 6 2. which overlap at the K point of the hexagonal Brillouin zone (BZ) of graphene. E g .84 0.008 0.939 12 4.243 0. (Reproduced from [17]. This simple picture provides a qualitative understanding. since the doubly degenerate π and π ∗ states.697 0. 12]. but (n.30 0.028 0. m) indices.76 0. Band gap. These gaps .22 1. in particular for small radius or metallic nanotubes.19 1. a map of the chiral vector specified by (n.R906 Topical Review Figure 2.07 0.86 0.52 0. The first-row values were obtained within the GGA in [18]. First-principles GGA calculations [18] resulted in small. S and C denote metals.39 M M 0. but nonzero energy band gaps of 93.09 1.) Table 1. to first order all (n.91 0. while all the rest are tight binding (TB) results.65 0.2. SWNTs specified by ‘C’ in figure 2 form a subgroup of semiconducting tubes.87 0.0 2. M. n R (Å) [18] [31] [31] [11] [42] [42] 4 1.04 1. 78 and 28 meV for (9.7 0. 0) SWNTs. Energy band structure According to the zone folding scheme a (n.45 0. semiconductors and curvature induced small gap semiconducting SWNTs.0 0.17 0. 0). A map of chiral vectors that determine the chirality of SWNTs.89 0. but fails to describe some important features.078 13 5. respectively.62 1. which have a small band gap induced by the curvature. 0) SWNTs calculated by different methods are compared. n) armchair tubes are metallic.69 0. This situation is clearly shown in table 1.07 0. 0) and (15. Thus. The two rows from [42] are for two different TB parametrizations.

0). Kim et al [17] synthesized ultralong and high percentage semiconducting SWNTs where further experimental evidence for the small band gap tubes was provided. (Reproduced from [18]. the doubly degenerate π ∗ states (which become the conduction band edge at for large R) and the singlet π ∗ state (which is in the conduction band for large R) are shown. n > 8). For tubes with radius greater than 3. As seen.e.Topical Review R907 Figure 3. In figure 3(a). the doubly degenerate π ∗ states (CB) and the singlet π ∗ state as a function of nanotube radius. 29.) are measured by scanning tunnelling spectroscopy (STS) experiments [16] as 80. (a) Energies of the doubly degenerate π states (VB). Recently. Moreover. 0) SWNT calculated by Blase et al [31] brought in another important effect of the curvature. the bottom of the . 30] are shown by open diamonds. These results indicate that curvature effects are important and the simple zone folding picture has to be improved. the doubly degenerate π states (which are the valence band edge at the point). The experimental data taken from [16. 42 and 29 meV. and below a certain radius determines the band gap. For zigzag SWNTs. the energy of this singlet π ∗ state is shifted downward with increasing curvature. the shift of the singlet π ∗ state is curvature dependent. Solid (dashed) curves are the plots of equation (2) (equation (1)).e.3 Å (i. As a result. The antibonding singlet π ∗ and σ ∗ states mix and repel each other in curved graphene. Each data point corresponds to n ranging from 4 to 15 consecutively. the analysis of the LDA bands of the (6. the energy of the singlet π ∗ state at the point of the BZ is above the doubly degenerate π ∗ states (i. in the same order. the purely π ∗ state of planar graphene is lowered with increasing curvature. (b) The calculated band gaps of [18] are shown by filled symbols. The biggest discrepancy noted in table 1 is between the tight binding and the first-principles values of the gaps for small radius tubes such as (7.

By extending the Taylor expansion to the next higher order. The band gap derived from the zone folding scheme is reduced by the shift of this singlet π ∗ state as a result of curvature induced σ ∗ –π ∗ mixing. but the results reported to date have been conflicting. obtained from equation (2) by using the parameters Vppπ = 2. (2) R R which depends on the chiral angle. Within the simple π orbital tight binding model. The agreement between the first-principles calculations [18] and the STS data [29. the band gap displays a rather oscillatory behaviour. θ . and hence the solid curves in figure 3(b) are fits to the 2 empirical expression E g = Vppπ d0 /R ± Vppπ γ d0 /R 2 . γ0 is taken to be equal to the hopping matrix element Vppπ . 58]. n < 7). d0 is the bond length in graphene. 8.2 and 7. Most importantly. d0 d0 E g = Vppπ 1 + (−1) p γ cos(3θ ) . The situation displayed in figure 3 indicates that the variation of the band gap with the radius is not simply 1/R. Theories based on physisorption have failed to predict high hydrogen uptake [52]. but not by the doubly degenerate π ∗ state. it was proposed [13] that E g behaves as d0 E g = γ0 . but additional terms incorporating the chirality dependence are required. Tada et al [54] examined the dissociative adsorption of a hydrogen molecule on various SWNTs. carbon nanotubes can be suitable for hydrogen storage. Based on π orbital tight binding model. 30] is very good considering the fact that there might be some uncertainties in identifying the nanotube in the experiment.7 Å are metallic. 8. For zigzag nanotubes θ = 0. 57. Recently. The relation given in equation (1) was obtained by a second-order Taylor expansion of one-electron eigenvalues of the π orbital tight binding model [13] around the K point of the BZ. the mixing of the singlet π ∗ state with the singlet σ ∗ state due to the curvature and its shift towards the valence band with increasing curvature are not included in either the π orbital tight binding model or the empirical relations expressed by equations (1) and (2).43. However. hydrogen adsorption has been the subject of various studies predicting interesting features [53–55. while the self-consistent calculations predict much smaller values.53 eV and γ = 0. This behaviour of the singlet π ∗ states is of particular importance for the applied radial deformation that modifies the curvature and in turn induces metallization [24. 42] included another term in the empirical expression of the band gap variation. Yorikawa and Muramatsu [41. Diverse hydrogen storage densities ranging from 0.7 Å (i. Therefore. 3. and hence it fails to represent the effect of the helicity. while it falls between the valence and conduction band edges for n = 7. 43.2. Hydrogenation of carbon nanotubes Owing to their large effective surface area. This explains why the tight binding calculations predict band gaps around 1 eV for n = 7. They found that the potential barrier height for the dissociative adsorption of a hydrogen molecule onto an outer wall decreases as . The factor (−1) p comes from the fact that the allowed k is nearest to either the K or K point of the hexagonal Brillouin zone. 8 tubes. Here γ is a constant and the index p is defined as the integer from k = n − 2m = 3q + p. all the zigzag tubes with radius less than 2. Several studies have explored this capacity of carbon nanotubes [45–51].e. 44] as discussed in sections 6. as seen in figure 3(b).R908 Topical Review conduction band). For n = 7. (1) R which is independent of the helicity. as well as an index p.4 to 10 wt% have been reported. and eventually dips even below the doubly degenerate valence band π states for the zigzag SWNT with radius less than 2. the edge of the conduction band is set by the singlet π ∗ state.

triangular and pentagonal C rings are formed instead of hexagonal ones. namely (i) a uniform pattern. Interestingly. in terms of the total energies of a fully relaxed bare SWNT. Hence. (ii) a chain pattern. whereby the circular cross section of the (7.it is of interest to see whether they are also stable against breaking a single C–H bond (or desorption of a single H atom). These new structures are stabilized by the formation of new diamond-like C–C bonds with dC−C ∼ 1. They found that E b increases with decreasing radius of the exohydrogenated SWNT. 0) SWNTs by using the Landauer expression [59] within the tight binding Green function method. 0) SWNT changes to a rectangular one.45 eV and C(n. Here E 0 = 1. Most interestingly. decoration). either one bond is formed just at the corners or two bonds are formed at either side of the corners. where every other carbon zigzag chain is saturated by hydrogen. 0) and (12. the single hydrogen atom H and a fully exohydrogenated SWNT. 0) = 4. (9.e. For half-coverage. Chan et al [60] proposed a mechanism for the dissociative chemisorption of hydrogen molecules on SWNTs in the solid (rope) phase under external pressure. as discussed extensively in section 6. RHC . Extensive analysis of hydrogen chemisorption on SWNTs has been carried out by Yildirim et al [55] using the first-principles pseudopotential plane wave method. On the basis of first-principles calculations. there are three isomers.3 the amount of energy gained by the adsorption of a single H atom decreases as R increases. On the other hand. monotonically. these are full exohydrogenation and endo– exohydrogenation where each carbon atom is bonded to a hydrogen alternately from inside and outside. n) = 4. At full coverage. the adsorption of a single H atom to a bare SWNT is always exothermic no matter what the value of R is. 0).5 for armchair tubes are polygonal where . They calculated the average binding energy E b = (E T [SWNT] + 4n E T [H] − E T [H + SWNT])/4n. = 0.Topical Review R909 the tube diameter decreases. The most remarkable effect is obtained when zigzag nanotubes are uniformly exohydrogenated at = 0. The uniform adsorption at = 0. Important effects of hydrogenation have been further examined for different isomers at different hydrogen coverage [57]. all these structures are predicted to be metallic with a high density of states at the Fermi level [57]. E b = E 0 + C(n.5 is metastable for zigzag nanotubes.73 eV. and those of (8. respectively.25 Å the required energy becomes negative and hence the system becomes unstable. Andriotis et al [58] calculated the I –V characteristics of finite (5. 5) and (10. Also the binding energies of zigzag tubes are always lower than those of armchair tubes with equal radius by about 30 meV/atom. Depending on 2n mod 4.5.62 eV.51–1. since uniformly adsorbed H atoms are rearranged upon relaxation by concerted exchange of C–H bonds to form zigzag chains along the tube axis. Yildirim et al [55] revealed that for RHC > 6. denoted by C4n H2n .63 Å near the corners of rectangular or square C4n H2n . The cross sections of the chain isomer at = 0. 0) change to square ones as shown in figure 4(b). C(n. Such a local minimum does not exist for armchair nanotubes.e. and (iii) a dimer pattern where every other carbon dimer row perpendicular to the zigzag carbon chains is saturated by hydrogen as described in figure 4(a). where every other carbon atom is bonded to a hydrogen atom from outside. Upon hydrogenation the structure undergoes a massive reconstruction.5. The binding energies can be very well described by a one-parameter fit. They argued that the process is reversible upon the release of external pressure. 0). However. They found extreme sensitivity to hydrogen adsorption. in spite of the fact that there might be an energy barrier preventing the desorption of H atoms. Although the exohydrogenated SWNT (C4n H4n ) is always energetically favourable with respect to a bare SWNT and 4n H atoms for all values of the radius. a dodecahedron and C60 H60 have higher average binding energies for adsorbed hydrogen atom. m)/RHC . i. (10. It is found that geometric and electronic structures and binding energies of hydrogenated SWNTs strongly depend on the pattern of hydrogenation (i.

At this limit.e. Relatively larger band gaps (in the range of 3. and fitted C ( ) to a general expression. 27.51 eV as R → ∞. expanded radius upon hydrogenation). The left and up arrows indicate the tube axis for armchair and zigzag nanotubes. E b = E 0 ( ) + pR p . 55. The average binding energies are plotted as a function of R in figure 5.) the corners are pinned by the zigzag H chains along the tube axis. (b) A side and top view of a (12. The effect of exohydrogenation on the electronic structure is remarkable.50 Å. and represents the curvature effect. E g (R) for exohydrogenated tubes at = 1 displays a behaviour similar to that of E b (R) for both types of SWNTs and hence decreases with increasing R (see figure 6). which transforms the sp2 to sp3 -like bonding. the endo–exohydrogenation of SWNTs. Note that the 1/R form (i. For other isomers at halfcoverage as well as exohydrogenations and endo–exohydrogenations at full coverage. 61].46 Å as if two diamond (111) planes had an interplanar distance of 0. the exo–endohydrogenated graphene (from above and below) is buckled by 0. 0)C48 H24 showing the square cross section of a uniformly exohydrogenated nanotube at half-coverage. 18. (Reproduced from [57]. Nevertheless.R910 Topical Review zigzag armchair uniform chain dimer Figure 4. Here C p ( ) is a constant that depends on coverage . gives rise to the highest binding energy saturating at 3. A special case of this fit was employed above for exohydrogenated tubes using RHC (i. respectively. 0) zigzag tubes have increased approximately by 2 eV upon full exohydrogenation. As for the band gap variation with R. Carbon atoms which are bonded to hydrogens are indicated by dark colour. this trend is reversed for endo–exohydrogenation due to increased H–H repulsion inside the tube at small R.e. Even more interesting is that the exohydrogenated metallic armchair tube becomes semiconducting with a band gap larger than the zigzag one for comparable radius. (a) A view of three different isomers of hydrogenated (n. 26. The band gaps of (n.5–4 eV) of endo–exohydrogenated SWNTs ( = 1) can be explained by the fact that the adsorption of H alternately inside and outside . The values of the exponent p and C p are determined for different isomers [57]. 0) SWNT at half-coverage (with formula unit C4n H2n ). We note that while E b increases with decreasing R in the case of exohydrogenation. p = 1) is quite common for SWNTs and scales various properties [16. the cross sections remain quasi-circular (see figure 5).

respectively. there are two dispersive bands with ∼1 eV bandwidth at both band edges and the extremum moves from the centre of the Brillouin zone ( point) to the zone edge (Z point). and increases .g.5.4 eV). On the other hand. the chain pattern realized on the (n. E g is large (e. the small band gap of the bare (9. As a result E g is reduced significantly.1 eV for (7. Depending on the pattern of hydrogen adsorption. The band gap E g between these states decreases with increasing R. (Reproduced from [57]. E b . Average binding energies. The filled squares show the zigzag nanotubes uniformly exohydrogenated at half coverage. 0)).5. Insets show top views of several C4n H2n isomers. Finally. The band gap is still significant for = 0. Circles and diamonds are for exohydrogenation and endo– exohydrogenation at full coverage. of (9. respectively. the doubly degenerate band at the conduction band edge moves towards the valence band edge and splits into bonding and antibonding states. an isomer can be either a metal or insulator. of hydrogen atoms adsorbed on various zigzag and armchair SWNTs versus bare tube radius R. D(E). the dimer row isomers are insulators and E g increases with increasing radius. Curves are analytical fits explained in the text. The effect of hydrogenation on the electronic structure is even more interesting at = 0.Topical Review R911 Figure 5. 0) nanotubes in figure 7. Hydrogen adsorption induced dramatic changes of the electronic structure are demonstrated by the total density of states (DOS). When n is even. For example. having the chain pattern. 0) SWNTs results in two doubly degenerate. almost dispersionless states at the valence and conduction band edges. E g = 2. The chain and dimer patterns of adsorbed hydrogen atoms at half coverage are shown by down and up triangles.) leads to an atomic configuration closer to the diamond structure having a rather large band gap (E g = 5. respectively. Filled and open symbols are for zigzag and armchair nanotubes. Surprisingly. becoming only a pseudogap for large and even n values. First of all. all uniform C4n H2n are metallic. When n is odd. 0) SWNT is opened by ∼2 eV upon exohydrogenation at = 1.

0). (Reproduced from [57].e. Apart from this being an ideal 1D conductor. Collins et al [63] found similar effects for oxygen. The band gaps. In this way four individual conduction paths are formed on each side of a square tube. the structures can be conceived as if they are more diamond-like than graphitic. the very high density of states at E F might lead to superconductivity. As displayed in the sixth panel of figure 7 for C36 H18 (uniform). Surprisingly. down and up triangles. Apart from opening a large band gap. i. Note that these uniform C4n H2n undergo a massive reconstruction and their circular cross sections change into square ones with the formation of new C–C bonds at the corners. Squares show nonmonotonic variation of the band gap of the bare zigzag nanotubes. However. 0) is modified to become similar to that of bulk diamond.5 are metals. yielding a ∼4–5 eV gap. versus the bare nanotube radius R. 4. a new dispersive metallic band with ∼1–2 eV bandwidth crosses the Fermi level. respectively. all zigzag nanotubes uniformly exohydrogenated at = 0. C36 H18 H18 . Filled and empty symbols indicate zigzag and armchair SWNTs. While carbon states are pushed apart. All C atoms without attached H (except those at corners) as well as the H atoms at the centre of four planar sides contribute to the high D(E F ). This argument is justified by the comparison of D(E) for the endo–exohydrogenated (9. The latter has a relatively larger valence bandwidth due to the coupling of distant neighbours. a similar chain pattern of C32 H16 in figure 7 has a much smaller band gap. E g .) to 4 eV for the dimer pattern. Curves are guides to the eye. For C4n H2n .R912 Topical Review Figure 6. respectively. Exohydrogenated and endo–exohydrogenated nanotubes ( = 1) and chain and row patterns of adsorbed hydrogen atoms at half-coverage are shown by circles. C4n H2n H2n . diamonds. their total densities of states are characterized by a peak yielding a high state density at E F . Oxygenation of carbon nanotubes Remarkable effects on the electrical resistance of a semiconducting single-wall carbon nanotube (s-SWNT) upon exposure to gaseous molecules such as NO2 and NH3 . especially endo–exohydrogenated SWNTs. the quasi-metallic D(E) of the bare (9. have been reported [62]. It is emphasized that the transformation from the sp2 to the sp3 bonding underlies the above effects. which can be converted to a good metal and hence its electronic properties can be reversibly modified by a . Exposure to air or oxygen influences electrical resistance and the thermoelectric power of a s-SWNT. with that of bulk diamond in figure 7.

in particular O2 . oxygenation is involved in other applications. the spin down states are nearly empty and hence give rise to a finite density of states at the Fermi level [68]. (Reproduced from [57]. At elevated temperatures. The zero of energy is taken at the Fermi energy shown by a vertical dashed line. instantaneously induces a significant increase in the emission current [64]. For example.5 (i. 0) at = 0. since the valence band is hole doped by the Fermi level touching the top of the valence band as a result of O2 physisorption [68]. C32 H16 in a chain pattern) and bulk diamond (bottom panel) are also shown for comparison. Experimental studies on carbon nanotube field emitters have shown that the adsorption of ambient gases.1. 0) tube becomes metallic. Spin unpolarized band structure calculations based on the local density approximation (LDA) predicted that the semiconducting (8. 0) nanotube (C36 ) and its various hydrogenated isomers.Topical Review R913 Figure 7.e. carbon based nanoparticles through oxidation. In addition to functionalization.) surprisingly small concentration of adsorbed oxygen. Comparison of the electronic density of states (DOS) of a bare (9. . While the spin up states are fully occupied. carbon nanotubes synthesized by using arc discharge are purified from other undesired. The analysis based on the local spin density approximation (LSDA) has indicated that the physisorbed O2 favours the triplet state. oxygen undergoes chemical reaction preferably with the strained C–C bonds and eliminates carbonaceous nanoparticles as well as the caps of nanotubes [65–67]. Physisorption of O2 molecules Observed effects on the electronic structure of SWNTs due to O2 physisorption have been subject to recent theoretical investigations based on first-principles calculations [68–77]. The DOS of hydrogenated (8. 4.

E b is the binding energy including the van der Waals interaction. i. 0) and (6. The bonding of O2 has been studied by placing the molecule at different sites on the SWNT as shown in figure 8. The zigzag (8. The effects of oxygen adsorption on the field emission from carbon nanotubes were treated by using an ab initio approach [72]. The GGA optimized distance from one O atom of the molecule to the nearest C atom of the SWNT is denoted by dC−O . the contributions of short range and long range interactions are small and comparable. 0) SWNT were calculated for selected sites [69]. 0) SWNT. the molecule (E T [O2 ]) and O2 physisorbed on the SWNT (E T [O2 + SWNT]). Therefore. 0) and armchair (6. spin polarized GGA calculations including the long range interactions and the chemisorption of oxygen atom on the (8. A detailed analysis of the physisorption of oxygen molecules based on first principles and fully relaxed. the mechanism of the oxidative etching of the caps and walls of the small radius (5.R914 Topical Review Figure 8. respectively. i. The spin polarized calculations yield relatively lower (stronger) total energies. where the desorption of O2 from the SWNT is prevented by a barrier. Z and T sites (see figure 8). 5) armchair SWNT was investigated [71]. and hence set the triplet state as the ground state with a net magnetic moment of ∼2 µB (µB = Bohr magneton) per unit cell. 77] with the aim of revealing how the adsorptions of O2 and O modify the electronic properties. 0) SWNTs were studied to provide an understanding of the oxidative etching process [70]. (Reproduced from [75]. Dag et al [75] have calculated the contribution of the short range interaction (i. in treating the binding energy of the physisorbed O2 the attractive vdW energy is calculated by using the asymptotic form of the Lifshitz formula [79– 81]. but it corresponds to a local minimum on the Born– Oppenheimer surface.e. E s < 0 is endothermic. 6) tubes are taken as prototypes for the s-SWNT and metallic m-SWNT. The binding energy involves short range chemical interaction and long range van der Waals interaction. The dynamics of the thermal collision of the O atom with a (6. 6) SWNTs has been carried out by Dag et al [75. The corresponding values for E s are −5. 5) and zigzag (9. By definition.e. E vdW = i j C6i j /ri6j . E s is the GGA chemical bonding energy for the spin polarized triplet state. for A. E s > 0 in equation (3) corresponds to a stable and exothermic chemical bonding.e. Adsorption and desorption of an oxygen molecule and various precursor states at the edges of finite size armchair (5. 0) SWNT was simulated by ab initio calculations [73]. Similarly. E T [O2 + SWNT]. In the case of the physisorption. (3) in terms of the GGA spin polarized total energies of the fully optimized bare SWNT (E T [SWNT]). H. A positive sign has been assigned to E vdW . 4. since stable binding . Similar trends are expected to occur in other tubes with curvature effects being emphasized at small R [18]. chemical bonding energy) by using the expression E s = E T [SWNT] + E T [O2 ] − E T [O2 + SWNT]. A schematic illustration of the physisorption sites of O2 molecules on the (8. In some cases. −27 and 37 meV. E b = E s + E vdW .) The energetics of oxygen adsorption on the surface of graphite [78] and on an (8.

The magnetic moment of O2 + SWNT diminishes at a distance d < 2 Å. The B site of a (6. 0) tube. magnetic moment µ and E g to their corresponding equilibrium values at d 0 = 2. The relatively large value of E s is due to the fact that the LDA yields overbinding as compared to the GGA. The hole doping picture has been refuted by recent first-principles calculations [75–77]. For ∼1. and on top of the C–C bond and perpendicular to the axis of the tube (B site) [75].45 eV). A similar singlet bound state occurs also at the Z site at .2 eV between the top of the valence band of the (8. Using a similar first-principles method within the local density approximation (LDA). Nevertheless. the binding energies E b of physisorbed O2 are found to be 120. The interaction of O2 with the SWNT as a function of O2 –SWNT distance d has been examined by calculating the total energy E T . H. 6) tube is similar to the A site of a (8.89 Å. We note. 155. respectively [75]. but dO−O has been relaxed. For a wide range of O2 –SWNT separation. Derycke et al [84] have argued that the main effect of oxygen physisorption is not to dope the bulk of the tube. and the total energy difference between the spin polarized and spin unpolarized states (i. The physisorption of O2 on the (6.604 eV Å6 . By adding these E vdW values to the corresponding GGA chemical bonding energies E s . Z and T sites also have band gaps. above the centre of the hexagon (H site). with an energy 0.86 eV) induces the gap E g and prevents it from closing. magnetic moment µ and energy gap E g of the O2 physisorbed (8. and the constant C6i j is calculated within the Slater–Kirkwood approximation [82] to be 10. that this standard calculation of vdW energy may involve ambiguities due to the relatively small distance between the physisorbed molecule and the SWNT. Using the binding geometries determined from the minimization of GGA total energies.6 Å < d < 2. and make the system metallic only for one type of spin. 0) SWNT [77]. Z and T sites. The band structure calculated using the spin polarized GGA for O2 physisorbed at the A site has yielded a band gap of 0. 158 and 191 meV for A. In these calculations. Here ri j is the distance between the i th O atom and the j th C atom.e.6 Å < d < d 0 a strong perpendicular force F⊥ is generated on the O2 molecule to push it away from the SWNT. which are large enough to prevent thermal excitation of electrons from the valence band at room temperature [77]. E T − E 0 T = 0. The binding energies are calculated to be 132 and 106 meV for H and B sites. respectively.9 Å. since it lies on the circumference. E T = 0.45 eV above the corresponding physisorption state (i. 0) SWNT contact. 6) armchair SWNT has been studied for two possible physisorption sites. Band structures corresponding to H.e. the singlet state of adsorbed O2 leads to a bound state at the A site at d = 1. but in the former case the C–C bond under the adsorbed O2 is highly strained. Upon O2 physisorption the metallicity of the armchair tube is lifted for the spin down bands. We see that |E T | decreases in the range 1.e. 0) tube and the empty Oppπ ∗ (↓) band (see figure 9). The small binding energies are characteristic of physisorption. bond distance of O2 dO−O . E g continues to exist.Topical Review R915 is specified by positive energies. respectively (see figure 8). 185 and 155 meV for A. i. We note that these binding energies are small. The situation for the m-SWNT is quite different. d has been constrained. dO−O . the vdW interaction is attractive and strengthens the chemical bond. While the spin polarized LDA (LSDA) energy band structure [68] has been interpreted as hole doping due to the adsorbed oxygen molecule. H. The binding energy of the T site is in agreement with the recent measurement by Ulbricht et al [83]. Jhi et al [68] found E s = 250 meV for the A site physisorption.47 Å. but to modify the barriers of the metal–(8. and become exothermic mainly owing to the long range vdW interaction. The splitting of the bands can be explained by the broken symmetry [85]. the E vdW values are calculated to be 125. however. Z and T sites. while spin up bands continue to cross at the Fermi level. Moreover. Figure 10(a) shows the variation of the ratios of E T . 159.

The adsorption site (A site) is shown in the inset. It is noted that these states [77. F⊥ (d). is normally more reactive than the surface of graphite. it was shown that there is no activation barrier for the dissociation of O2 when it is adsorbed at the zigzag edge of a SWNT [70]. the z site) .51 Å. Only the short range interaction is considered. 86] are neither easily accessible from the physisorption state. Spin polarized calculations yield the singlet state with a net zero magnetic moment as the ground state.47 Å and 0. However. nor support the hole doping picture because the band gap is ∼0. the O2 molecule dissociates into two O atoms.25 Å. In contrast to the case at this site. As a result. 4. µ(d) and dO−O (d) where at d ∼ 1.e. A carbon nanotube.2. i. we discuss the interaction between atomic O and SWNT. respectively. These singlet bound states in figure 10(b) correspond to local minima on the Born–Oppenheimer surface and are separated from the more energetic physisorption states by an energy barrier. Various adsorption sites together with the geometry and chemisorption energies E s obtained from spin unpolarized calculations are shown in figure 11. Owing to the concerted motion of the atoms at the proximity of the molecule and energy gained by the individual oxygen atoms engaging in the bonding with the SWNT concomitant with the dissociation.5 Å and hence the bond is broken. However.e. dO−O is increased to 2. and reveal the nature of the chemical bonding.R916 Topical Review Figure 9. 6) SWNT behave differently. The spin up and spin down bands corresponding to the triplet ground state are shown by broken and continuous lines. The zero of energy is set at the Fermi level. In fact. the single O adsorbed on top of the zigzag C–C bond (i. the O2 molecule has a bound singlet state at the B site with dO−O = 1. Figures 10(c) and (d) show the variation of E T (d).) d = 1.5 eV. 0) SWNT calculated using the spin polarized GGA with the atomic structure fully optimized in the double cell. E V indicates the top of the valence band. In this section. The electronic structure of O2 physisorbed on an (8. Chemisorption of oxygen atoms The effect of the adsorbed oxygen on the structural and electronic properties of a SWNT is important. Breaking of the O–O bond of a physisorbed O2 molecule is unlikely owing to the weak interaction with the SWNT.70 Å. the activation energy for dissociation is expected to be low.80 eV above the physisorption state of the Z site at d = 2. it was shown that near the defect sites of the graphite surface O2 molecules can dissociate [78]. that can be described as graphene rolled into a cylinder. Among all sites considered in [75]. O2 adsorbed at the H site of the (6. (Reproduced from [75]. O2 physisorbed near the defect sites of a SWNT is expected to dissociate into atomic oxygens.

as a well known and usual effect of oxygen. The total energies of the triplet ground state corresponding to the physisorption state for Z and A sites at d ∼ 2. However. and the magnetic moment. As a manifestation of the curvature effect.64 eV when all a sites are filled by chemisorbed oxygen atoms [75]. µ0 and E g correspond to the stable physisorption state with d 0 = 2. since a single oxygen (per unit cell) chemisorbed at an a site gives rise to flat bands below E V . The electronic state of an oxygen atom chemisorbed on a SWNT depends on the coverage and pattern of adsorption. 0) tube opens up to 3. 6) SWNT distance d for O2 adsorbed on the H site of the (6. For example. µ. (d) variation of the bond distance of O2 . Experimentally. (b) The total energies of the singlet bound states found at small d at the Z site (square) and at the A site (diamond). (a)Variation of the percentage values of C/C0 (i. This implies that the dissociation of O2 followed by the chemisorption of individual O atoms is an exothermic process similar to other oxidation processes. the energy gained from the chemisorption of two atomic oxygens is more than the bond energy of O2 in either magnetic or nonmagnetic states.07 eV.e. Moreover. the strained C–C bond at the z site is broken. (Reproduced from [77].89 Å at the A site. E T . with the O2 –(6.9 Å are shown by continuous and broken lines. the . Adsorption of individual atoms on SWNTs This sections deals with the adsorption of individual atoms on SWNTs and the resulting properties. F⊥ . Because of miniaturization aiming at higher and higher device densities. E T . 0) SWNT remains unaffected. it was shown that oxygen exposure first oxidizes and eventually etches away the nanotubes with smaller radius [87]. dO−O . 5. the band gap of the (8. dO−O .) is energetically most favourable with E s = 5. µ/µ0 0 0 0 0 and E g /E g ) with the O2 –SWNT separation d. 6) SWNT. but the C–O bond is strengthened. dO−O /dO−O . (c) Variation of the total energy. the band gap of the (8. The binding energies as large as ∼5 eV suggest that atomic O is actually chemisorbed with a significant charge transfer from C to O.Topical Review R917 0 0 Figure 10. and the force acting on the O2 molecule. E T /E T .

they were not reproducible enough to offer any relevant technological application. Recently. Zhang et al [97] have shown that continuous Ti coating of .R918 Topical Review Figure 11. dC−C : length of the C–C bond under an adsorbed O atom. very thin metallic wires with controllable sizes [96]. 0) SWNT.) realization of interconnects with high conductance and low energy dissipation appears to be a real technological challenge. Very thin metal wires and atomic chains were produced by retracting an STM tip from an indentation and then thinning the neck of the material that wets the tip [88–90]. 91–94]. These metallic nanowires can be used as conducting connects and hence are important in nanodevices based on molecular electronics. E s is obtained from the spin unpolarized calculations of the total energies in equation (3). Earlier experimental studies have indicated that SWNTs can serve as templates for producing reproducible. While the nanowires produced so far have played a crucial role in understanding the quantum effects in electronic and thermal conductance [14. Nowadays. electron transport measurements showed that SWNTs can resist high Joule heating [95]. Moreover. dC−O : length of the C–O bond. E s : chemisorption energy. A schematic illustration of the various adsorption sites of atomic O on the (8. Some relevant geometrical data and GGA chemical bond (chemisorption) energies corresponding to these sites are also given. the most practical and realizable method for fabricating nanowires relies on carbon nanotubes. (Reproduced from [75].

such as Mg and Zn. similar to that expression E b u (p) given in equation (3). D: top. The interaction between the SWNT and most of the adatoms in tables 2 and 3 is significant and results in a chemisorption bond. and including most of the transition metal atoms) adsorbed on SWNTs has been carried out by Durgun et al [102. They considered four possible sites shown in figure 12. Thus. While alkali and simple metals have binding energy in the range of 1. Pb were able to form only isolated discrete patterns or clusters rather than a continuous coating of the SWNT. Binding geometry and binding energies Recently. metal atoms such as Au.5 eV. The superscript u (p) are u (p) indicates spin unpolarized (spin polarized) energies.Topical Review R919 Figure 12. metals such as Cu. Ag and Zn have relatively weak binding. the chemisorption energy of transition metals is relatively high. However. Not only metallic connects. E vdW .1. The formation of Schottky barrier at the contact has been found to be responsible for the operation of field emission transistors made from SWNTs [99–101]. However. is expected to be much smaller than the chemisorption binding energy and is omitted. but also the contacts of metal electrodes themselves are crucial for the operation of devices based on nanotubes. C.and determined the binding geometry and binding energies by optimizing all atomic positions. Fe. Here the total energy of the adsorbed atom is E T [A].3 Å.0–2. The binding energies E s obtained from the total energies corresponding to either a nonmagnetic (spin unpolarized) state with zero net spin or a spin relaxed state. 103]. for specific elements. A schematic illustration of different binding sites of individual atoms adsorbed on a zigzag (8. binding energy and resulting electronic structure of various atoms (ranging from alkali and simple metals to group IV elements. (Reproduced from [103]. A: axial. The long range interaction. The centre of the hexagons (H site) appears to be favoured by most of the adatoms. Low resistance ohmic contacts to metallic and semiconducting SWNTs were achieved by Ti and Ni [98]. The binding energy is calculated from the short range (chemical) interaction using the u (p) u (p) u (p) u (p) u (p) Es = E T [SWNT] + E T [A] − E T [A + SWNT]. an extensive study on the binding geometry. H: hollow. It is well known that the . 0) tube. Al. However. ¯ The average C–adatom bond distance occurs in the range of 2. On the other hand. the binding energies corresponding to a nonmagnetic state range from ∼1 to ∼4.5 eV. S: substitution sites. Au. 5. the binding energy is small and the character of the bond is between chemisorption and physisorption. O atoms having small atomic radii. Tables 2 and 3 summarize the results relating to the adsorption of individual atoms. dC−A is relatively small for H.) varying thickness and quasi-continuous coating of Ni and Pd can be obtained by using electron beam evaporation techniques. Z: zigzag.

41 interaction between the graphite surface and most of the atoms included is actually weak.53 0 2.02 — — — 0 — — — — 2.9 3. and hence they have net magnetic moments ranging from 5.04 1.01 0 1.0 1. the magnetic moment (µB /supercell) of the magnetic ground state corresponding to the adsorption of various individual atoms on the (8.01 — — — u E b (eV) E b (eV) 1.8 Table 3.6 4.79 2. Most of the transition metal atoms adsorbed on the (8.8 0.21 3.25 2. Mo.7 0.7 1.25 4.3 2.4 0. Mn.1 2. the binding energy E b obtained from spin polarized calculations.6 1.61 0.64 2. 6) SWNT.7 3.7 0.5 5.2 2.20 2.7 0.6 3. the u difference between spin unpolarized and spin polarized total energies E T . since p u E T = E T [A + SWNT] − E T [A + SWNT] > 0.04 0.5 — — — 0.2 2. the adsorbate–carbon distance dC−A .4 0.67 5.60 3.05 0.59 0 0.2 2.4 0.1 2.9 2. The strongest binding site (shown in figure 12). (Reproduced from [103].15 0.5 2.3 2.7 0. (Reproduced from [103].4 0. the magnetic moment µ per supercell corresponding to the magnetic ground state corresponding to the adsorption of individual Ti. the binding energy E b obtained from spin polarized calculations.49 µB (for Mn) to zero magnetic moment (for Pd .49 2.1 0.03 0.R920 Topical Review ¯ Table 2.8 2.81 0.2 2.2 2. 0) and (6.40 0.5 1.2 2.6 1.2 2.9 2. 6) tubes are in the magnetic ground state with E T > 0.3 2.4 2.28 p µ (µB ) 1.7 2.5 1.4 1.86 4 0 0.3 1.1 2.03 0 0.32 0 0.3 1. 43].27 1.14 0.34 0.68 5.2 2.7 0.) Atom Na Sc Ti V Cr Mn Fe Co Ni Cu Zn Nb Mo Pd Ag Ta W Pt Au Al C Si Pb H O S Site H H H H H H H H A A H H H A A H H–A A A–T H Z H H T Z A ¯ dC−A (Å) 2.62 3.4 0.41 0.1 0. The curvature effect is the primary factor that strengthens the binding [18.73 0. The strongest binding site (as shown in figure 12).2 2.4 2.2 1.8 0.2 2.1 2.6 1.3 E T (eV) 0.9 2.42 1.04 — — — 0.48 2.1 1.8 1.2 0. Specific adsorbate–SWNT (A + SWNT) systems are found to be in a magnetic ground state.23 0.3 2.02 0. 26.4 0. Au atoms on a (6.05 3.9 3.9 E T (eV) — 0.4 3. 0) SWNT.01 2. the binding energy E b p obtained from spin unpolarized calculations.5 2.) Atom Ti Mn Mo Au Site H H H T ¯ dC−A (Å) 2.8 — — — p µ (µB ) — 0.17 5. the binding p u energy E b obtained from spin unpolarized calculations.2 3.02 u E b (eV) E b (eV) 1. the adsorbate–carbon distance ¯ dC−A . the difference between spin unpolarized and spin polarized total energies E T .1 3.3 2.1 2.5 2.1 2.9 4.1 2.3 2.58 1.

Ta. The transition metal atoms with a few d electrons. Pt (column VIII elements) and Cu. 0) SWNT. The atoms which were observed to form continuous and quasi-continuous coating on the SWNT (Ti.k (r) and the difference charge density n. 6) tube. Nb. possibly adatom–adatom interaction.8 Å (dC−A = 3.04 µB ). Pb) which remain as discrete particles on the surface of the tube [96]. Cu. since they have the same valence configuration. 43]. Perhaps this trend can also be explained by the curvature effect.4 to 1. came out to be consistently lower for the adatom adsorbed on the (6. Pd. H favour a nonmagnetic ground state when they are adsorbed on the (8. i. 0) is smaller than that of (6. Fe.6 µB . respectively. The character of the bond between the adatom and SWNT can be examined by charge density analysis.5 eV. S. and points to an issue: whether molecular magnets (or nanomagnets) can be produced from carbon nanotubes. 31]. not only adatom–SWNT interaction. The calculated binding energies are rather high. form strong bonds with a binding energy ranging from 2. For the same reason Zn by itself exhibits a similar trend u (E b = 0. While Ni adsorbed SWNT has very low magnetic moment (0. as well as E b and the magnetic moments of the adatom adsorbed on the (8. Most of the adatoms yield the strongest binding at the H site. For the fully relaxed. Ni and Pd) have relatively high binding energies as compared to those atoms (Au.k (r) n.7 Å). Pb. 0) SWNT. Al. Whether a permanent magnetic moment can be generated by exchange interaction on these transition metal coated SWNTs would be an interesting question to answer. The Z site is energetically favourable for both C and Si adatoms. In ρ(r).Topical Review R921 and Pt). Au (column I-B elements) seem to prefer ¯ the A site. but also other factors. The average adsorbate–C distance. C. Ag and Cu have magnetic moments in the range of 0. of bare SWNT ρ[SWNT] and of free atom ρ[A] . Ag. Recently. Note also that the p u binding energies E b . We also note that in forming a good coverage. and hence for the fabrication of nanowires. since the radius of (8. Owing to the weak binding the type of the bond between Mg (Zn) and the SWNT is between chemisorption and physisorption with dC−A = 3. While an individual Al atom (with (3s)2 3p valence structure) is not bound to the graphite surface. Na with one 3s electron on the outer shell is bound with a significant binding energy (E b = 1. the magnetization and hysteresis loops of iron nanoparticles partially encapsulated at the tips and inside of aligned carbon nanotubes have been demonstrated by experimental works [104]. These metals can also be used as a buffer layer to form uniform coating of good conductors such as Au. E b = 4.4–0. Inclusion of transition metal elements inside the tube is another way to obtain nanomagnetic structures. Ag.k ρ[A + SWNT] − ρ[SWNT] − ρ[A] can be calculated. 6) [18. 0) SWNT is relatively strong. and Cu have very weak binding. Addressing this issue may open an active field of study on SWNTs. which are covered or substitutionally doped by transition metal atoms according to a well defined pattern. minimum total energy configuration. dC−A . its binding on the (8. since the binding was found to be even stronger at the high curvature site of the SWNT under uniaxial radial deformation [26.e. Ni. O. Binding energies listed in table 2 are of particular interest for coating of SWNTs by metal atoms. Atoms. Ag. The interaction of group IV elements with the SWNT is also crucial. In this way. the adsorbates such as Au. On the other hand.05 eV) with its (4s)2 valence structure. and hence can be suitable for metal coating of a SWNT. play a crucial role. Co. such as Na. C and Si adatoms form rather strong bonds with the SWNT.8 eV. Good conductors such as Au. the total charge densities of adatom adsorbed SWNT ρ[A + SWNT]. The magnetic moment generated upon the adsorption of individual transition atoms has important implications. Ti. most of them occur at ∼2.9 Å (minimum) and 3. such as Sc. Si. the SCF total occup ∗ ρ(r) = charge density ρ(r) = n. The binding energy of Mg is very weak and is only 0. This can be explained by the curvature effect.1 Å.3 eV).03 eV at the H site due to its outer shell (3s)2 .2 and 2. these atoms are prevented from oxidizing.7 Å (maximum). ranges between 1.

Figure 13 shows ρ(r) and ρ(r) which were calculated for four adatoms. Contour plots of total ρ(r) and difference ρ(r) charge densities. The difference charge density conveys information about the charge rearrangements upon adsorption. (Reproduced from [103]. For others (Na. namely Na. since they do not convey relevant information owing to the low charge density around adsorbed Na atoms. Al. C. Ti [103].R922 Topical Review Figure 13. Dramatic differences in the valence electron configurations of these adatoms are reflected in the characters of the bonds they form with the SWNT. For C the charge densities are calculated on a plane passing through the adatom and zigzag C–C bond. charge is depleted from black regions and is accumulated at white regions. In contour plots of ρ(r).) are all calculated in the same supercell with the same atomic positions. The counter-plots of ρ(r) for [Na + SWNT] are not presented. Al. Ti) the charge density plane passes through the centre of the hexagon and adatom. The 3s valence electron of Na is weakly bound and is donated to the . According to the above analysis the charge is generally transferred from the adatom to the C–C bond of SWNT.

In fact. electrons are transferred from Al and nearest C atoms to the region between the Al atom and the centre of the hexagon on the surface of the SWNT and to the C–C bond. E V ). which is adsorbed at the H site.45 electrons for the spin unpolarized case. Zhao et al [107] investigated Li intercalation in (10. This band is normally the first empty conduction band of the tube. they are specified as donor states (if they are close to the edge of the conduction band. They found that charge is transferred from adsorbed Li to the SWNT. Since the 3s valence electron of Na has low ionization energy. The empty and filled bands just above and below E F are derived from C adatoms with a significant contribution from the SWNT. 10) SWNT ropes modelled by a uniform. the SWNT is slightly deformed upon Li adsorption.Topical Review R923 conduction band of the SWNT. Therefore. Their findings are consistent with the adsorption of Na discussed above. the calculated band structure of the SWNT with Al adsorbed at the H site forming a regular chain structure gives rise to a half-filled band derived from the empty conduction band of the SWNT. The band structure of the Li intercalated (8. The localized states are relevant for the doping of a s-SWNT. cs . (8. give rise to a spin paired. which is derived from the modified conduction band of the bare SWNT. the localized state due to an adsorbed individual Al atom is a donor state. a small gap occurs between the bands derived from adsorbate states. Si yields an almost fully occupied band in the band . Because of new bond formation. Depending on their position relative to the band edges. As a result. The Ti atom adsorbed at the H site forms a bond. has been the focus of attention because of SWNT ropes which were expected to be a candidate for anode material use in Li battery applications [105]. 2D hexagonal lattice. In the case of Al. charge is transferred to the back-bonds. the charge rearrangement due to Au adsorption is minute. which are adsorbed individually on the (8. Another alkali metal. Al. Similarly. hence the SWNT becomes metallic.2. recent experiments showed higher Li capacity in SWNTs than that of graphite [106]. 0) and (10. The character of the bond formed between the C adatom and C atoms of the SWNT is reminiscent of the sp3 covalent bonds in diamond crystals. 0) SWNT with a repeat period of 2c. Directional bonds form between Si and C atoms of the SWNT. The energy states associated with a single adsorbate form energy bands if the electronic structure is calculated within the supercell method. Actually these bands correspond to a linear chain of adsorbates with a 1D lattice constant of the supercell. Na. Li. which results in a small binding energy. but their conduction bands get occupied above E g and the system becomes metallic. ρ(r) has a maximum value at the centre of the C–C bonds. C and Si. and also localized states in the band gaps. In the case of C. (12. The interaction between the Au atom and SWNT is weak. 0) rope has indicated that the bands of the bare SWNTs are not affected in any essential manner. The latter case is also known as hole doping. Electronic structure An individual atom adsorbed on an SWNT may give rise to resonance states in the valence and conduction bands. 0). The charge density of adsorbed Si exhibits a character similar to that of adsorbed C. this empty band is gradually populated and also modified upon adsorption of Na. According to the Mulliken analysis the charge transferred from a Ti atom is 1. 0). Furthermore. nonmagnetic ground state [103]. The calculated band structure of the SWNT with Na adsorbed at the H site forming a regular chain with a lattice parameter (or interatomic distance) a = cs ∼ 2c yields a half-filled band. it is donated to the SWNT to occupy the empty conduction band. The d orbitals are responsible for this bond. Figure 14 presents their calculated band structures and the LDOS at the adatoms. Accordingly. E C ) or as acceptor states (if they can occur close to the edge of the valence band. 5.

. (Reproduced from [103]. C. the overall behaviours of state densities are similar. The zero of energy is set at the Fermi level.) gap and an almost empty band at the bottom of the conduction band. TDOSs and LDOSs of single Na. Energy band structures. However. Si adsorbed on a zigzag (8. Al and Si are adsorbed at the H site. Al.R924 Topical Review Figure 14. Na. C is adsorbed at the Z site. The dispersion of bands for Si adsorbed on the H site displays differences from that of the C adatom adsorbed at the Z site. 0) tube. LDOSs calculated at the adsorbate.

7 Å for (8. Al adsorbed SWNTs in turn have higher binding energies than the Au adsorbed ones. 8) SWNTs using first-principles methods. (Reproduced from [108]. There is a small band gap of ∼0. Several unit cells are shown for the purpose of visualization. This situation is in accordance with the Mulliken analysis. It is well known that adsorbed Ti and Au have magnetic ground states leading to a . Ti is much more likely than Al and Au to form a continuous chain on the surface of SWNTs due to much higher binding energies. 6) and (8. In calculating the binding energies the energy references are crucial. The geometric configuration of the adsorbed metal chain is shown in figure 15. 0). Along the tube axis. 0). 0) and 2.2 eV between these Au bands.6 Å for (14. each unit cell contains two Ti atoms for the (14.e. (6. Al and Au chains adsorbed on (10. since this band crosses the Fermi level and also overlaps with the other conduction bands. That the contribution of the SWNT states to these peaks is minute and that the band gap between the conduction and valence bands of the SWNT is practically unchanged confirm the weak interaction between Au and the SWNT. The dispersive and almost fully occupied spin down band is formed from the states of carbon and hence derived from the conduction band of the bare SWNT. 0). 0) tube and one Ti atom for the (8. Geometric structures of adsorbed Ti chains on (14. In spite of the fact that the calculated binding energies include significant coupling between adjacent metal atoms. Mo and Ti) display a different situation due to the magnetic ground state. it gives rise to two sharp peaks below and above E F in the TDOS. namely filled spin up and empty spin down bands. they show the same trends as the binding energies given by Durgun et al [102. Three bands formed from Ti 3d(↑) are fully occupied and accommodate three electrons of adsorbed Ti atoms. 8) tubes.) The spin polarized band structure and TDOSs of adsorbed transition metal atoms (i. The distances between the two neighbouring Ti atoms in the chains are 2. Filled Au 5d(↑) and 5d(↓) bands occur in the valence band of the SWNT 2 eV below E F . The SWNT is metallized upon Ti adsorption. (14. 8) tube. Yang et al [108] have studied the binding energies and electronic structures of Ti. 103]. It is seen that the adsorption of Au does not induce significant modification in the bands of (8.Topical Review R925 Figure 15. but they overlap with the conduction band of the SWNT. Au. Other Ti 3d(↑) bands occur above E F . Au yields two bands in the band gap. we believe that the stability of the Al and Au chains originates from the stability of zigzag-free Al and Au chains. 0) (top) and (8. 8) (bottom) SWNTs. which predicts the electrons to be transferred from Ti to the SWNT. except the Au 6s(↑) and Au 6s(↓) bands in the gap. On the basis of the works by Bagci et al [109] and Prasen et al [110]. This is in compliance with the calculated magnetic moment of ∼1µB per adsorbed Au atom. Mn.

can form rather strong bonds to the surface of the SWNT. 0) and (6. Therefore. such as Ti.3. regular coverage of the SWNT. The important effect of the Ti chain is to metallize the zigzag SWNTs due to the delocalized d states crossing the Fermi level. 6) tubes have magnetic ground states with significant magnetic moments of 13. However. 0) tube by specific transition element atoms. Corresponding densities of states for spin up and spin down electrons are shown. 13. Transition element covered or filled SWNTs It is clear from the above discussion that individual atoms. 0) and (6. 0). Cr. Optimized atomic structure of (8. Dag et al [7] investigated the coverage of an (8.R926 Topical Review Figure 16.) net spin. Even more interesting is that the s-SWNT has become a highly conducting metal due to several Ti bands crossing the Fermi level. 0). (9. (9. since a small gap is opened owing to the symmetry breaking. Yang et al [111] have examined the magnetism of transition metal covered or filled (9. The conductance of the Ti covered tube has been estimated to be 16e2 / h. Ni. They also found through ab initio calculations that these structures exhibit substantial . Si.5 µB /cell. Furthermore. They found that only Ti can yield a continuous. Co. they found that the Ti covered (8. as well as giving rise to a dramatic modification of the atomic structure.7 and 9. Motivated by the experimental work of Zhang et al [97] showing the continuous coating of the SWNT by Ti. Upon optimization of the geometry the circular cross section of the SWNT has changed to a square. 6) tubes. such as Ti. 0) SWNTs. Fe. 0) and (6.3. the binding energies reported by Yang et al [108] might change if one took the magnetic states into account. Figure 16 shows the structure and density of states of Ti covered (8. 0). 5. (9. the situation is different for the Ti chain adsorbed armchair tube. 6) armchair SWNTs which are uniformly covered with Ti. (Reproduced from [7]. using a first-principles plane wave method. respectively.

as high as 2 TPa. It has been argued that SWNTs show linear elasticity under hydrostatic pressure up to 1. However. Co atoms packed inside a variety of carbon nanotubes offer strong spin polarization at the Fermi level. right: majority spin) and DOS (bottom: minority spin. a nanotube takes on a flattened structure [114]. . The spin polarized band structure (left: minority spin. While the MWNTs show the greatest values for the Young’s modulus. On the other hand. the situation is rather different for SWNTs owing to their tubular geometry [33]. In particular. (Reproduced from [111]. Figure 17 shows the spin polarized band structure and DOS for the Co filled SWNT. SWNTs are highly flexible: they can sustain elastic deformations remarkably. 6.5 GPa at room temperature. If R is large. Radial deformation of carbon nanotubes Modification of electronic properties of condensed systems by an applied external pressure or strain in the elastic range has been a subject of active study.Topical Review R927 Figure 17. Both plots suggest that the magnetic moment resides mostly in the narrow transition metal d band below the Fermi level. The softness of carbon nanotubes in the radial direction has been observed experimentally: the contact area is flattened when two nanotubes are brought close to each other [113].) magnetism. as well as considerable magnetic moments. Carbon nanotubes display very interesting mechanical properties. the changes that one can induce by the elastic deformation are minute or even negligible due to the rigidity of the crystals. in most cases. top: majority spin) for Co filled SWNT as shown in the right panel. SWNTs appear to be very soft in the radial direction [112].

Elasticity For the radial deformation. (7) A Here Fy is the restoring force applied on the surface area A. The first-principles calculations predict that elastic radial deformation does not have a noticeable effect on the first. In this way. Under the radial deformation described above. whereby the curvature is locally changed. Tight binding calculations have indicated that a SWNT may undergo an insulator– metal transition under a uniaxial or torsional strain [24. In their study the radial deformation is generated by applying uniaxial compressive stress σ yy to a narrow strip on the surface of a SWNT. Multiprobe transport experiments [23] on individual SWNTs showed that the electronic structure can be modified by bending the tube. the point group becomes C2h or D2h . For the (6. 6. the elastic deformation energy E D σ yy = xx + C12 yy (6) . However. Hence.1. depending on the nanotube orientation around the tube axes. σxx = 0 = C11 and Fy = C12 xx + C11 yy . C11 and C12 are the in-plane elastic stiffness constants. 25. or by applying a circumferential deformation. respectively. 43.and secondnearest neighbour C–C distances but it induces significant changes in the bond angles [43]. Assuming the validity of Hooke’s law. This observation is therefore important and has to be taken into account in tight binding studies of SWNTs with radial deformation. the in-plane mirror symmetry can be broken. 6) tube. As a result. A systematic analysis of the effect of the radial deformation on the electrical properties has been carried out by G¨ lseren et al [43] using first-principles total energy and electronic u structure calculations. The point group of the undeformed nanotubes is Dnh or Dnd for n even or odd. Empirical extended H¨ ckel u calculations [115] predicted that the conductance of an armchair SWNT can be affected by the circumferential deformations and a band gap can develop in a metallic armchair SWNT upon twisting. Significant radial deformation of SWNTs can be realized in the elastic range. In practice such a deformation can be realized by pressing the tube between two rigid and flat surfaces. the radius is squeezed in the y direction while it is elongated along the x direction and hence the circular cross section is distorted to an elliptical form with major and minor axes. R0 − b (4) yy = R0 and R0 − a (5) xx = R0 where R0 is the radius of the undeformed (zero-strain) nanotube. 115–125]. respectively. there are several different orientations to investigate when looking at the effect of mirror symmetry on the band crossing at the Fermi level. the relation between stress and strain is given by the generalized Hooke’s law. 124. 126– 129]. A natural variable for describing the radial deformation is the magnitude of the applied strain along the two axes. one expects radial deformation to induce important modifications in the electronic and conduction properties of nanotubes [24.R928 Topical Review It has been shown that the structure and the electronic properties of SWNTs undergo dramatic changes as a result of these deformations [23. a and b. zones with higher and lower curvatures relative to the undeformed SWNT can be attained for the same circumference. 125].

Hooke’s relation. (Reproduced from [43].5 since it is only on one carbon atom. For the (8. (c) The restoring force on fixed carbon atoms. (b) Variation of the elastic deformation energy per carbon atom. It is also noted that the SWNT becomes stiffer as R decreases. The slope is the in-plane Poisson ratio. Interestingly.Topical Review R929 Figure 18. ν [43]: C12 xx ν =− = . (a) The strain component x x = (R0 − a)/R0 along the major axis as a function of the applied strain yy = (R0 − b)/R0 along the minor axis. The variation of the restoring forces is expected to be linear in the elastic range. ν .25. while for the other tubes it is on two carbon atoms. the quadratic form obtained from classical theory fits very well to the elastic deformation energy calculated from first principles. the force is scaled by 0. except the deviations at certain data . 0) SWNT.e. and hence the elastic character of the deformations. The restoring forces in figure 18(c) are in overall agreement with this argument. the difference between the total energies of radially deformed and undeformed SWNTs) is ED = 1 (1 2 + ν 2 )C11 − ν C12 2 yy (8) in terms of the in-plane Poisson ratio.) (i. persists up to yy = 0. (9) C11 yy Deformation energies obtained from classical elasticity in equation (8) are compared with those calculated from first principles in figure 18.

Initially.R930 Topical Review Figure 19.904 0.36 416. Ceff = C11 (1 − ν 2 ).67 273. Figure 19 summarizes the variation of the band gap and the density of states at the Fermi level.06 ν 0. which are amplified because the force is a derivative quantity.88 319. (Reproduced from [43]. Upon metallization. 6.88 98. 0) and (8. the behaviour of the (9.) Radius (Å) (7.34 points due to uncertainties in the first-principles calculations.) Table 4. (b) The density of states at the Fermi level D(E F ) as a function of the applied strain yy . However.2. In-plane elastic constants of SWNTs. (Reproduced from [43]. E g . 0) (8. D(E F ) increases with increasing strain. All elastic constants are in GPa except for ν which is unitless. 0) (9.52 4.14 3. Effect on the electronic structure First-principles calculations confirm that the band gaps of zigzag tubes reduce with applied strain and eventually vanish leading to metallization. but then .02 86. 0) (6. 6) 2. 0) tube is different.20 270.12 C11 713.864 0. For (7.15 364. (a) The variation of the band gap. Elastic constants calculated from first principles are listed in table 4.46 C12 645. as a function of the applied strain. the band gap increases with increasing strain.36 226.828 Ceff 129.874 0. 0) SWNTs the band gaps decrease monotonically and the onset of an insulator–metal transition follows with the band closures occurring at different values of strain.70 91.76 3. D(E F ).

Minot et al [121] confirmed that the band structure of carbon nanotubes can be altered as predicted in [24. 0) metallic tubes. 10) tube. 0) SWNT this singlet π ∗ state is above the doubly degenerate π ∗ states. 0). which normally cross at the Fermi level with quasi-linear dispersion. probably due to bilayer interactions. Delaney et al [85] showed that the π ∗ conduction and π valence bands of a (10. Applying a radial strain through the tip of an atomic force microscope and simultaneously measuring the conductance. The singlet π ∗ state in the conduction band shifts downwards in energy much faster than the other states do with increasing strain. 0) and (8. the energies of a few bands near the band gap are plotted as a function of strain in figure 20. 0) SWNTs. In order to explain the band gap variation of (n. Once the singlet π ∗ band. The variation of the energy eigenvalues of states near the band gap at the point of the BZ as a function of the applied strain. The singlet state originating in the conduction band is indicated by squares.) decreases with strain exceeding a certain threshold value and eventually diminishes. In the study of Park et al [44]. (8. The shaded region is the valence band. since the separation of the two nearest walls of the tube became comparable to the interlayer distance of graphite. The increase of the band gap at the initial stages of radial deformation is connected with the relatively high rate of downward shift of the doubly degenerate π valence band relative to the π ∗ conduction band under low strains. 43. Lammert et al [120] showed the gapping by squashing (20. Uniaxial stress of a few kilobars can reversibly collapse a small radius tube inducing a 0. while the collapsed large radius tubes are stable. 0) nanotubes. open a pseudogap in the range of ∼0. which shows faster decrease with strain. the band gap begins to decrease with increasing strain.1 eV gap. Since the singlet π ∗ state lies below the doubly degenerate π ∗ states for both (7. For all these zigzag SWNTs the band gap strongly depends on the magnitude of the deformation and E g is closed at 13%. . 44] earlier. 6) SWNT. which is normally metallic. the band gap of the (5. (Reproduced from [43].Topical Review R931 Figure 20. remained metallic with a slowly decreasing D(E F ) even for significant radial deformation. for the (9. whereas the armchair (6. This is due to the curvature increasing with increasing radial deformation. crosses the doublet conduction band and enters the gap. 0) and (9. Earlier. 22% and 17% strain for (7. On the other hand.1 eV at certain directions of the BZ perpendicular to the axis of the tube owing to tube–tube interactions in a rope. 5) tube was monotonically increasing. their band gaps are closed monotonically with increasing yy . 20) and (36. since then circumferential regions are brought into close proximity. The opening of the gap is caused by the broken mirror symmetry. 0) tubes. respectively.

The heightened chemical reactivity of deformed sites was attributed to the introduction of radical p orbital character. they performed classical trajectory simulations on the interaction between the hydrogen atom and the strained nanotube. perhaps due to the fact that the singlet π ∗ band. The variation of the binding energies of H and Al with the radius of the zigzag tube fits to the curve given by the expression CA (10) E b. As described in the previous sections. 10) SWNT. Effect on the chemical reactivity Srivastava et al [58] examined the effect of conformational strain on the chemical reactivity of a SWNT.6 eV in regions of high conformational strain.49 and −0. First-principles pseudopotential calculations have shown that the adsorption of foreign atoms on SWNTs can indeed be modified continuously and reversibly [26]. Note that. This situation is shown in figure 21. Binding energies E b of single hydrogen and aluminium atoms adsorbed on the zigzag SWNTs versus the radius of the tube R. These predicted that the chemisorption of hydrogen atoms is enhanced by as much as 1. falls into the band gap as a result of increased curvature. Actually the deformed places have sites with higher curvature as compared to the curvature of the bare (10. while the band gap shows significant change with n (for example by going from . The effect of radial deformation becomes noticeable through the binding properties of foreign atoms on undeformed SWNTs.A + R where E 0. The fitting parameter. 10) SWNT. 9.A is the binding energy of the adatom A (here H or Al) on the graphene plane and calculated to be 1. which is normally in the conduction band. A (10.A (R) = E 0. 8.R932 Topical Review Figure 21.) 6. Calculated data for n < 7 slightly deviate from the above simple E b (R) relation. had kinks or ribbon structure on its surface. Analysis of atomic cohesive energies and the local density of states obtained from the empirical tight binding method showed that the sites with enhanced binding energies of hydrogen atoms correspond to destabilized cohesive energies and electronic states near the Fermi level.3. It is seen that E b decreases with increasing radius (or decreasing curvature) and eventually saturates at a value corresponding to that for on graphene.02 eV for H and Al.14 eV Å−1 for both H and Al. bond order potential.A + CA /R explained in the text. where the calculated binding energies E b of H and Al adsorbed on the zigzag tubes are plotted as a function of the radius R for n = 7. 12. Using a many-body. 10. H is adsorbed at the top site. (Reproduced from [26]. is calculated to be ∼3. Al favours the H site.A (R) = E 0. CA . The solid curve is the fit to E b. which was either bent or subjected to torsional strain. respectively [26]. directly above the C atom of the tube.

According to the predictions of this study. near one of the ends of the minor axis. at the high curvature site. The difference of binding energies of the sharp and flat sites . i. near one of the ends of the major axis. 0) zigzag SWNT as a function of the radial deformation yy defined in the text. (a) Variation of the binding energies E b of single hydrogen atoms adsorbed on a (8. Increase of E b with decreasing R (or with decreasing n) shows that for small R the character of the surface deviates from that of the graphene.25 eV less than that corresponding to yy = 0.65 to 0. 0) to (9. the binding energies vary smoothly with R −1 . and traces out the upper curve. E b for the adsorption on the low curvature (flat) site first decreases with increasing strain and then saturates at an energy 0. In figure 22(a) the variation of the binding energies of a single hydrogen atom adsorbed on the (8. (Reproduced from [26]. This result has been exploited to create different zones with different curvatures on the surface of a single SWNT by radial deformation and hence to attain different chemical reactivities. The lower curve is for the adsorption on the low curvature site. E g changes from 0.09 eV). The upper curve corresponds to H adsorbed on the high curvature site of the deformed tube. i. The lower curve is associated with the adsorption on the flat site. respectively.e. 0) surface shows that E b at the sharp site. E b has increased by 0. (b) As (a) but for a single Al atom. 43] where the circular cross section of the bare tube is distorted to an elliptical one with major and minor axes a and b.3.e.Topical Review R933 Figure 22.) (8. increases with increasing yy (defined in equation (4)). On the other hand. Radial deformation can be created by applying a uniaxial compressive stress on a narrow strip on the surface of the SWNT [26. 0).85 eV for yy = 0.

then gradually increases. 0). since nanoropes consisting of interlinked SWNTs will be significantly stronger than nanoropes composed of vdW packed nanoropes [122].1 eV for yy = 0. it has been shown that a H2 molecule which is bound to the surface of (8. Intertube interactions in nanoropes can be probed by applying external pressure to vary the intertube distance. For example. which does not bind to the graphite surface. Such a property could have important consequences for nanoscale device applications and composite materials that require strong mechanical properties. deforming the circular cross sections to elliptical or hexagonal ones [119]. For example E b increases by ∼0. In figure 22(b) the binding energy of Al exhibits a behaviour similar to that of H. 6. These results have been interpreted as irreversible pressure induced changes in the atomic structure. it dissociates into two H atoms. (6. E b first decreases with increasing yy . Similar covalent bonding can occur between the nanotubes in a rope. the sp2 character of the bonding of the tube has changed locally and become more like sp3 . 0). the surface of the tube where Al is adsorbed expands. For fullerenes.75 GPa [133]. Some of these effects are tentatively attributed to crushing or flattening of nanotube surfaces in the rope.3 which is 80% of the binding energy on the undeformed tube. and have indicated that the mode intensities and energies are not completely reversible upon pressure cycling [132].R934 Topical Review is substantial and is equal to ∼1.7 GPa. 10) nanotube rope as a function of applied pressure P. 0).5 Å. 6) were included in that study. As a result of H adsorption. despite H and Al favouring different sites on the (8. Yildirim et al [123] were the first to investigate the effects of pressure on the ropes based on the first-principles calculations. such high pressure studies have yielded many interesting results [131] including new compounds such as the pressure induced polymeric phases of C60 . 0) tube. If the physisorbed H2 can approach the surface as closely as 1. observation of very large volume reduction and high compressibility [119] has suggested that a microscopic volume reducing deformation other than vdW compression can occur under high pressure. and new nanomaterials. Effect of pressure on nanoropes It is anticipated that physical properties can also be altered by intertube interactions between nanotubes packed in a hexagonal lattice. This is 44% of the binding energy of H on the undeformed SWNT. can be adsorbed at the high curvature site of a zigzag tube under radial deformation with a significant binding energy. The effects of deformation are significantly different for the zigzag and armchair SWNTs. Raman studies on SWNT ropes have been carried out up to 25. Their interpretation of the structural transition has been objected to by Tang et al [134] on the grounds that their experiments were reversible up to P = 4 GPa.80 eV for yy = 0. The pressure dependence was determined . as so-called nanoropes.4. 0) becomes physisorbed at the high curvature site with a significant binding energy [130]. Furthermore. Adsorption of Al induces local changes in the atomic and electronic structures. E b at the sharp site of the deformed SWNT increases with increasing yy . It is remarkable that Al. The zigzag tubes (5. ranging from hydrogen storage to chemical sensing. These calculations have indicated a sudden transition of structure at 1. 0) and the armchair tube (6. Peters et al [133] reported that a structural phase transition occurred at 1. each of which becomes bound to adjacent C atoms of the SWNT at the surface. We believe that the tunable adsorption can have important implications for metal coverage and selective adsorption of foreign atoms and molecules on the carbon nanotubes and can lead to a wide variety of applications. (7. Furthermore they performed empirical uniform force field calculations and obtained equilibrium lattice constants of the (10. (9. We see that according to the first-principles calculations the chemical reactivity of a zigzag SWNT can be modified reversibly and variably by radial deformation. For Al adsorbed on the flat site.9 GPa. Similarly.

Optimized structures of SWNT ropes. while the other parameters.) by minimizing the total energy as a function of nanotube separation.Topical Review R935 Figure 23. Lattice parameters of the ropes. The pressure . i. 0) zigzag tubes. are shown. E T (a. b). 0) one-dimensionally interlinked under pressure. as illustrated in figure 24. 0) tubes the energy difference between the 1D interlinked structure and the vdW packed structure is small. This indicates that the rehybridization occurs from the sp2 to the sp3 type. b and c and the γ angle. including the atom position c and γ = 2π/n.e. (a) (6. The zero of energy is set to the total energy calculated for the optimized structure of the vdW packed rope at zero pressure. Total energy versus 2D lattice constant for (7. For some (n. 0) nanotubes are packed by vdW interaction under zero pressure and (c) (7. The bonds interlinking two adjacent tubes have bond lengths comparable to those in diamond. The top view along the axis of the rope of the different phases is shown in the inset. 0) nanorope with that of a 1D interlinked one. 0). were optimized. (6. The same structural transformation was observed for the (5.) Figure 24. a structural phase transformation from a vdW lattice to a new orthorhombic (space group Cm Cm ) lattice was observed in which the nanotubes were interlinked along the [110] direction. 0). the 1D interlinked structure is stable once it is formed. The total energies of the vdW and 1D interlinked phases cross each other at a ∼ 9 Å with a small energy barrier of 46 meV/unit cell (552 K) from the vdW side. 0) nanotube ropes in different phases. The covalent bonding between nanotubes occurred at the high curvature sites as a result of rehybridization of the carbon orbitals. a. 0) and (b) (7. since one has to overcome an energy barrier of 0.7 eV in order to break the links. However. the bond angles differ from the exact value of the tetrahedral angle and vary between 100◦ and 120◦ indicating some strain. (Reproduced from [123]. The interlinked structure in (c) has lower total energy than the vdW packed structure in (b). Regardless of that. (Reproduced from [123]. (9. At critical pressure. Figure 23 compares the structure of a vdW packed (7.

Interestingly. 0) tube is a semiconductor with E g = 0. at about a ∼ 8 Å there occurs another interlinked phase of the (7.7 eV Å−2 ) than the vdW packed rope. these models do not allow for structural relaxations to be performed. Devices based on carbon nanotubes Much of the research on carbon nanotubes has been carried out with the motivation of generating devices. They are rather hexagonally distorted such that the local structure of nanotube faces is reminiscent of that of a graphite sheet. 0. Attempts to design and fabricate nanodevices from carbon nanotubes have exploited several properties discovered recently. The above structural transformations induced by the applied pressure lead to significant changes in the electronic structure [123]. Semiempirical methods [20. investigation of spin coherent quantum transport in carbon nanotube magnetic tunnel junctions [140]. where nanotubes are interlinked along both a and b axes. In spite of the fact that an individual (7. Finally we will discuss the devices fabricated. however. The structures transform to even more complex structures at higher pressure. semiclassical and quantum mechanical ab initio methods.the effects such as external bias and gate potential are usually not accounted for. Ab initio methods in transport calculations Accurate prediction of the transport properties of nanoscale atomic or molecular systems (nanotubes. In this section. it is stable and sixteen times stiffer (d 2 E/da 2 = 13. we will first review recent progress made in calculating I –V characteristics of nanostructures and present some of the device models or circuit elements. have been widely used. Its minimum lies ∼1 eV/unit cell above the minimum of the 1D linked phase. However. These methods provide some quantitative insight especially in cases where the charge transfer effects are relatively small. Some applications to carbon nanotubes include designation of a quantum dot by introducing pentagon–heptagon defects in a carbon nanotube [20]. once it is formed. 0) nanorope. the dispersion of the bands near Fermi level in the direction perpendicular to the tube axis makes the vdW rope metallic.R936 Topical Review for attaining this lattice constant is only about 0. 6) armchair tubes do not form interlinking up to 60 GPa. indicating that polymerization of vdW (7. such as sensors. This 2D interlinked phase is not favoured energetically. 7. 0) nanoropes could occur at modest pressures and temperatures. transistors and nanomagnets. 6) nanorope is elastic (see figure 25). 7. indicating that the deformation of the (6. This energy is comparable to the 1D polymerization energy of C60 . including phenomenological. in particular) including their current–voltage (I –V ) characteristics is essential for realization of a broad spectrum of device applications. The ropes of the (6.3 GPa for the vdW phase. 135–145] based on parametrized Hamiltonians. some of them reviewed in the previous sections. Since these parametrized u Hamiltonians are generally derived from the bulk or isolated molecular systems. Theoretical investigations of the transport properties of these nanoscale systems are based on calculations with various levels of sophistication. and hence finding a feasible way to contribute to the objectives of miniaturization.1. Conversely. At high pressure (30 GPa). such as tight binding or extended H¨ ckel models.65 eV. the (2D interlinked) high density phase becomes a wide band gap insulator with E g ∼ 2 eV. Semiempirical methods are not typically self-consistent. Upon releasing the pressure the hexagonally distorted (or polygonized) structure returns to the original one.7 eV/unit cell (or 25 meV/atom) is required to go from the 1D interlinked phase to the vdW packed structure. which are designed theoretically using SWNTs. Moreover. study . they are relatively simple and easy to implement.

Nanotubes are distorted in such a fashion that the local nearest neighbour structure is reminiscent of graphite structure. (b) A hexagonal network of (9. 0) nanotubes. in which (9. 0) tubes are interlinked along a. dC−C indicates the smallest distance between two carbon atoms of nearest neighbour tubes in the rope. 0) nanotubes. The two-dimensional crystal structure and relevant physical parameters for various high density phases of carbon nanotubes. (a) A 2D interlinked structure of (5.Topical Review R937 Figure 25. (c) A very dense structure of (7. (d) The optimized structure of (6. b and [110] directions. 6) armchair tubes under 53 GPa.) . (Reproduced from [123]. 0) nanotubes obtained under 30 GPa.

R938

Topical Review

Figure 26. Upper panel: geometry of the bent (4, 4) nanotubes with bending angles (a) θ = 0◦ , (b) 3◦ , (c) 6◦ . Lower panel: the calculated transmission function and LDOS for the corresponding geometries. (Reproduced from [144].)

of the dynamic conductance properties of carbon nanotubes of different helicities, both with and without defects [141], and calculation of transmission properties of extended [144] and finite size [145] carbon nanotubes. Electronic transport properties of extended nanotubes in the presence of distortions and mechanical deformations were studied by Nardelli [144] using the surface Green function matching formalism and a nearest neighbour π orbital tight binding Hamiltonian [146]. The π orbital bands were described via a single parameter, Vppπ = −2.75 eV. Calculations made for small radius (4, 4) nanotubes show that bending of the tube by small angles (3◦ and 6◦ ) does not drastically alter the basic electronic characteristics of the metallic (4, 4) tubes (see figure 26). A first-principles approach to the transport properties of nanoscale devices is based on supercell methods [147–154] where Kohn–Sham (KS) equations for the device are solved using periodic boundary conditions. The scattering states are then determined through a recursive technique considering the effective device potential between the perfect electrodes. The applicability of these methods is limited to the calculation of transport coefficients in the linear response regime, because the effective device potential is derived using periodic boundary

Topical Review

R939

conditions which cannot describe systems under an external bias or having nonidentical electrodes at the two ends. The method has been used to provide a possible explanation for the experimentally observed conductance of one quantum unit instead of two for nanotubes with one end dipped into a liquid metal such as mercury [153]. For nanoscale transport investigations, an alternative method, which was pioneered by Lang [155], is the open jellium approach [155–160] based on Lippman–Schwinger theory. Treating the leads in terms of a jellium model, the charge density of the open device is constructed from the scattering states. The KS equations are solved self-consistently with the appropriate boundary condition of the device under bias. The method, which allows I –V curves to be calculated, is however limited by two major approximations. Constructing the charge density and the device potential using only scattering states is not enough for achieving a truly self-consistent solution. The omitted bound states, which exist within the device, must also be considered. Second, the real atomic structure of the electrodes is not represented in the jellium model. In recent years, several first-principles methods have been developed to capture the basic properties of the challenging problem of nanoscale transport in the nonequilibrium regime [161–163]. These methods, essentially differing only in their numerical implementation, share most of the main ingredients of the treatment as they are imposed by the intrinsic properties of the problem at hand. Density functional theory (DFT) is employed to describe the atomistic nature of both the device and the electrodes self-consistently, including exchange–correlation and charging effects. Unlike conventional DFT simulation schemes, in order to account for the infinite and open nature of the system due to the electrodes, DFT is used in combination with the nonequilibrium Green function (NEGF) formalism. The NEGF formalism provides a means for representing the effects of infinite electrodes in the form of self-energy terms. Also the nonequilibrium charge density of the device under the biased conditions to be used in DFT can be constructed through the Keldysh nonequilibrium Green function. It should be noted that this formalism also allows for a computationally efficient evaluation of the contribution of the bound states to the charge density. More specifically, a nanodevice coupled to two semi-infinite metallic electrodes can be considered to be divided into three parts. These are the left (l), right (r) and central (c) regions. The central (device) region must be chosen to extend far enough into the electrodes that the effective device potential can be matched at the boundaries with the bulk potential of the electrodes (see figure 27). The effective potential of the device is generated using the information on the charge distribution inside the central region ρc (r ). The charge distribution is conveniently calculated from the density matrix ρ, which can be constructed using the ˆ nonequilibrium Green function as i ˆ ρ=− (11) dε G< (ε) 2π where G< (ε) = GR Σ< GA (12) with GR and GA = (GR )† denoting the retarded and advanced Green functions of the device (central region). Σ< corresponds to the injected charge from the electrodes. It can be expressed in terms of the Fermi distribution functions fl (ε − µl ) and f r (ε − µr ) of the left and right electrodes and their retarded self-energies ΣR and ΣR as r l
Σ< = −2i Im[ f l ΣR + f r ΣR ] . l r

(13)

Here µl and µr are the Fermi levels deep in the left and right electrodes such that Vb = (µl − µr )/e is the external applied bias. The Green functions are evaluated by direct matrix

R940

Topical Review

Figure 27. (a) The metallic electrode–nanotube interface. A semi-infinite (4, 4) armchair SWNT is in contact with a semi-infinite Al(100) metallic electrode. The device region is coupled to perfect atomic scale electrodes which extend to reservoirs at z = ±∞. (b) Charge density contours of the interface at equilibrium. The contours of the bulk lead match that of the device at the edges of the device region. (Reproduced from [161].)

inversions: GR (ε) = [εI − H − ΣR − ΣR ]−1 l r (14) where H[ρc (r )] is the Hamiltonian matrix of the central region within DFT and evaluated using the charge density obtained from equation (11). The cyclic procedure summarized above should be iterated until NEGF–DFT self-consistency is achieved. The I –V characteristics of the nanoscale device are calculated using the Landauer formula 2e I (Vb ) = (15) dε T (ε) [ f l (ε − µl ) − f r (ε − µr )] . h It should be noted that the transmission function T (ε) depends both on temperature and on the applied bias and is expressed as T (ε) = Tr[Γl (ε)GR (ε)Γr (ε)GA (ε)] (16) where Γ p is the function describing the linewidth due to the coupling of the central device region to the electrode p. It is related to the corresponding self-energies of the electrodes in the form
Γ p = i(ΣR − ΣA ), p p

( p = l, r).

(17)

They showed that of the two available channels. Each step corresponds to an additional nanotube coming into contact with the metal. Some interesting features such as negative differential resistance in I –V curves and the importance of the detailed atomic structure of the electrodes are reported. half the expected value. Figures 28(a) and (b) show the transmission electron micrograph (TEM) image of the nanotubes used and the experimental set-up. the transport properties of a broad set of nanoscale systems [164–168]. The existence of pentagon–heptagon pair defects in infinite nanotubes is shown to alter the conductance pattern slightly [152]. dipped into a liquid metal to minimize the contact resistance effects. Carbon nanotubes having different electronic structures depending on their chirality and radius are candidates for being main constituents of the future nanoelectronics technology.Topical Review a b Fibre R941 3 c 2 G (G0 ) 1 0 -200 0 200 400 600 z-position (nm) Figure 28. metallic single-wall carbon nanotubes have an equilibrium conductance of G = 2G 0 . In another account. (b) A schematic diagram of the experimental set-up. Plateaus are observed corresponding to additional nanotubes coming into contact with the liquid metal. Calculations show that infinitely long. The discrepancy between the theoretical calculations and the much smaller values of the nanotube conductance measured in experiments is attributed to poor nanotube– electrode contacts. while the π band is responsible for the observed conductance of the system.e. 10) SWNT with one end immersed in a jellium metal. due to two subbands crossing at the Fermi level [11. using a semiempirical scattering technique Sanvito et al [169] calculated the conductance of a MWNT with a special geometry and provided a possible explanation of the measured [15] fractional conductance values in terms of interwall interactions in a MWNT. and gold nanowires. are studied under nonequilibrium conditions. Nanotubes are lowered under SPM control into a liquid metal surface. including C60 . 12]. fullerene and silicon. In an interesting transport experiment performed by Frank et al [15] the conductance of MWNTs. Taylor et al [161] simulated transport properties of a SWNT–metal interface using the . An explanation of the measured conductance pattern is provided by Choi and coworkers [153] in their ab initio calculation of the conductance of the (10. where G 0 = 2e2 / h. linear chains of carbon. G = G 0 . perfect. were found to be close to G 0 . Later. i. (a) A transmission electron micrograph (TEM) image of carbon nanotubes at the end of a nanotube fibre. it is mostly π band incident electrons that suffer strong back-scattering at the nanotube–metal contact. Although narrower. A considerable amount of experimental and theoretical effort has been devoted to investigating quantum transport properties of nanotubes. (Reproduced from [15]. (c) Variation of conductance with nanotube fibre position. two more steps at fractional G 0 values are also observed.) With the development of self-consistent ab initio methods based on DFT and the NEGF formalism. The conductance is seen to increase as the nanotube fibre is dipped into the liquid metal in steps of G 0 as shown in figure 28(c).

They showed that there are two transmission eigenvectors which connect the two sides of the electrode–nanotube interface. their sum at the Fermi energy gave an equilibrium conductance almost equal to G 0 for a range of distances between the nanotube and the electrode. In a recent study Palacios et al [170] addressed the problem of conductance of finite length carbon nanotubes with realistic metal electrodes using a first-principles method called the Gaussian embedded cluster method [163]. 0) zigzag tube. which have been revealed through the active experimental and theoretical research during the last decade. The 5/7 defects in any orientation on the surface of SWNTs change both the diameter and the chirality. but also the electronic structure around the Fermi level gets . Chirality increases with increasing number of 5/7 defects perpendicular to the axis of the tube. if the 5/7 defect can be aligned exactly on the axis of the (n. Periodic doping of nitrogen atoms in a semiconducting nanotube makes the structure metallic with a nonlinear I –V characteristic and a negative differential resistance at higher biases. the step-like equilibrium conductance pattern of the nanotube is altered due to the action of a substantial number of nitrogen atoms as elastic scattering centres. If the 5/7 defect is perpendicular to the axis of the tube. such as 5/7 pentagon–heptagon defects. metal– nanotube contacts and Schottky barriers. The effects of substitutional nitrogen doping in carbon nanotubes under nonequilibrium conditions were investigated by Kaun et al [171] using NEGF–DFT formalism. however. Structural defects can affect the atomic and electronic structure of nanotubes. 0) SWNT and revealed their effects on the atomic and electronic structure by using a recursion method within the tight binding description of the carbon π bands. resonant tunnelling double-barrier (RTDB) and multiple-quantumwell structures (MQWS). 7. have been combined in nanodevice operation or circuit elements. the chirality remains the same. The role of topological defects. They observed resonance patterns around the Fermi level in the conductance profiles. the diameter is conserved. has been investigated by different groups. Device models Several novel properties of carbon nanotubes. interconnects. metallic nanorings for persistent currents and solenoids. The electronic and transport properties of carbon nanotubes can be modified by doping with foreign atoms. The two transmission eigenvalues T1 and T2 were smaller than unity. These are semiconductor heterostructures or metal– semiconductor junctions.R942 Topical Review NEGF–DFT formalism and a semi-infinite Al electrode of finite cross section (see figure 27). It was found that for doped zigzag nanotubes even a single nitrogen atom substitution increases the current flow and for small radii tubes narrows the current gap.2. however. SWNT–SWNT junctions. Charlier et al [19] carried out an extensive study on the topological 5/7 defects in different directions on the surface of the (12. This result is consistent with both the nanotube–jellium calculation [153] and the experiment [15]. All these nonequilibrium transport properties can be understood from the simple physical picture of alignment and misalignment of current carrying bands of the device due to applied bias. but the diameter becomes smaller. As electrodes. etc. Nitrogen doping of nanotubes does not cause substitutional structural relaxation. It is interesting that breaking of symmetry by the impurities can raise the current flow in semiconducting tubes and reduce the current slightly in metallic ones. Not only atomic or geometric structure. For example. but the zigzag SWNT becomes chiral. They found that the 5/7 defect can induce changes in diameter and chirality of the nanotube. elemental Ti was found to make better high transparency contacts to carbon nanotubes compared to more traditional metals such as Al and Au.

introducing heptagon–pentagon defects. The (6. 0) SWNT by introducing a single and triple 5/7 defect pair. 4)/m(5. The second junction. 3). the calculation of electronic transport through such junctions yielded tunnelling conductance and hence indicated that they can be used as building blocks of nanoscale electronic devices [172]. 0)–(7. m(5. 4) SWNT junctions. nanotubes with different chiralities and different electronic properties and diameters can be connected using such a pair defect to form different kinds of heterostructures.) modified by the presence of 5/7 defects. 5) part are indeed confined and exhibit a sharp. The metallization of semiconducting SWNTs under radial deformation (discussed in section 6) has been proposed to generate quantum structures [24. 1) junction.Topical Review R943 Figure 29. Using the empirical tight binding method and Green function matching approach along with the transfer matrix technique. a metallic nanostructure between two semiconducting ones forms a quantum dot and this has quantized electronic states. 173]. The idea is simple and originates from the numerous works done on semiconductor heterostructures (An Bm ) and quantum dots [174]. 4) tube was connected to the (5. 4)/m(5. (Reproduced from [20. 4) tube is a semiconductor. 0)–(5. 3) junctions. which were formed from a single (8. 5) unit cells m. respectively. Normally. their level spacings vary depending on the value of m. it corresponds to a heterostructure between a semiconductor having a band gap of 0. 0)– (7. In fact. Chico et al [20] went further and proposed an interesting heterostructure. It is known that . Here the (6. 5) tube smoothly. These heterostructures (such as metal–semiconductor and semiconductor–semiconductor) can form rectifiers or diodes and hence they can play an important role in nanoelectronics.62 eV and a semimetal. For the case of the (8. 1) and (8. 4) quantum dots as a function of the number of (5. namely a new type of quantum dot based solely on single SWNTs. Most importantly. showing a clear quantum size effect (see figure 29). 5)/(6. δ function-like density of states. 0)–(5. It is made from a series of (6. 5)/(6. (8. Chico et al [173] studied the electronic structure of (8. Quantum structures generated by radial deformation. Energies of the discrete states of the (6. Moreover. 118]. In this way formation of undesirable interface states has been eliminated. It is clear that by the controlled generation of 5/7 defects on a chiral SWNT one can form a variety of SWNT–SWNT junctions with different heterostructure characteristics. 44. is a semiconductor– semiconductor heterostructure with two different band gaps. 5) indicates m unit cells of a metallic armchair tube. Chico et al [20] showed that electronic states in the metallic m(5.

the strain induced quantum well structures were treated by means of SCF first-principles calculations on a superlattice consisting of 12 (8. exchange and ionic pseudopotentials. Similar to what one achieves in semiconductor heterostructures. The electronic potential which consists of Hartree. the short periodicity. In this case the band gap modulation is provided by the modulating hydrogen adsorption (see figure 30). Ve (r) = VH (r) + Vx (r) + Vion (r). The localization of states in one of the zones (either a (C32 )l (i. Within the superlattice geometry. Applying the standard method for band line-up explained in detail below [175].e. respectively) are crucial parameters for monitoring the resulting electronic properties. is further analysed by examining the electronic potential [176]. y. it has been demonstrated that quantum structures can be formed on a single SWNT. A model of an electronic device based on a similar band gap modulation and hence MQWS formation has been proposed recently [176]. l = q = 1. z) dx dy (18) ¯ and the running average of the planar averaged potential. Performing empirical tight binding and also first-principles calculations. is calculated self-consistently.16. This indicates a staggered band line-up. and also the effect of the superlattice size (i. the misfit of the valence band edges between adjacent (i. periodic MWQSs or finite RTDB structures can be realized on an individual s-SWNT and their electronic properties can be variably and reversibly monitored.R944 Topical Review owing to the band offsets of the semiconductor heterostructures.e (z). They considered a supercell consisting of 16 unit cells of the (7. In figure 31(b) the differences between two zones become pronounced and V . 0) SWNT. 0) SWNT unit cells [18].e. specified by l) zone or a (C32 H32 )q (i. along the axis of the SWNT. l. the energies of the band states of one semiconductor B may fall into the band gap of the adjacent semiconductor A. Formation of different zones leading to different electronic structures on the same (8. yy = 0. Tight binding charge densities of the band edge states for superlattices corresponding to n = 4. Vr. is not sufficient to induce two adjacent zones with markedly different electronic structures. Quantum structure formation by modulating hydrogenation. For example. As seen in figure 31(a).e. the adjacent unstrained and strained parts were repeated yielding a periodic band gap modulation. q and l + q). the height (depth) of the conduction (valence) band edge of A measured from that of B. Tombler et al [118] have shown that a quantum dot can be produced by bending (and hence inducing radial deformation of) the SWNT at a given point. 0) s-SWNT and repeating them periodically. m layers of B between n layers of two As form a quantum well yielding confined electronic states. behaves as a potential barrier for electrons (holes).e. Later. E C ( E V ). half of them were radially strained with yy = 0. As in the empirical tight binding calculations. 0) SWNT upon modulating the H adsorption. 8. using a SCF crystal potential and referring to the energies at the top of the valence band. n of the unit cells were kept undeformed and the remaining m = 16 − n were radially deformed. According to the effective mass approximation. By repeated formation of (An Bm ) one generates MQWSs or RTDBs by combination (An Bm An ). undeformed and deformed) zones has been found to be ∼180 meV.e. the remaining half were left undeformed—i. Here as and bs are superlattice parameters. The depth of the well and the width of the barrier and well (in terms of number of layers n and m. Kili¸ et al [24] c have shown that a MQWS is generated by applying different radial deformations at adjacent parts of a (7. Figure 31 shows the planar averaged electronic potential 1 ¯ Ve (z) = as bs Ve (x. 12 have clearly demonstrated the confinement of states and quantum well structure formation. specified by q) zone) is based on the physical grounds discussed above.

Figures 31(c) and (d) indicate that the potential. Vr.) increases for the superlattice. (b) l = q = 2. (a) [(C32 )1 (C32 H32 )1 ]. are calculated from full structure optimization by using conjugate gradient method. (Reproduced from [176]. (C32 )l and (C32 H32 )q . Relatively high cell averaged potential zones correspond to (C32 )l . Superlattice structures. Averages of calculated self-consistent electronic potentials of the superlattices. respectively.Topical Review R945 Figure 30. [(C32 )l (C32 H32 )q ] formed on a single (8. 0) SWNT by modulating adsorption of hydrogen atoms. (c) [(C32 )3 (C32 H32 )2 ].) Figure 31. Dashed ¯ ¯ and thick curves are planar averaged. as well as the superlattice parameter cs . l = q = 2. (b) [(C32 )2 (C32 H32 )2 ]. (Reproduced from [176]. (a) l = q = 1. Ve (z). Band line-up of zones.e (z).e (z). ¯ Vr. becomes flatter as l and q increase from 2 to 3 and two adjacent zones display markedly different electronic structures. is a complex process involving the relaxation of the atomic structure and charge transfer at the interfaces and also resulting in modification . See the text. The positions of 64 H and 128 C atoms. (d) [(C32 )2 (C32 H32 )3 ]. and running averaged.

e |l(q)=∞ .e. n) armchair SWNT. Secondly.q = E V. for the case of l = q = ∞ the energy difference between the edge of the valence band E V and the average potential. the valence band offset of the superlattice is E V = E V. 109]. The band diagram indicates a type-I band line-up appropriate for a resonant tunnelling double-barrier structure. E B |l(q)=∞ = E V.e. especially those at the interface. E C. The actual atomic positions. (C32 )l=∞ and (C32 H32 )q=∞ . B . E g1 and E g2 . Then. (C32 H32 )q . SWNT–metal contact. In the third step.l .R946 Topical Review (C32H32)q (C32)l Egl (C32H32)q ∆EC Egq ∆EV Figure 32. 103. for example to the resonant tunnelling double barriers.) of the electronic potential. The conduction band offset is then determined by subtracting these latter energies. First.e (z).e. Here. i. E B. (C32 H32 )q (C32 )l (C32 H32 )q . Metallic reservoirs adjacent to both barriers.l − E V. A solution of this problem [175] involves the ¯ following steps. respectively. respectively. E C = E C. formed . In this way metallic connects are provided by the hydrogen-free ends of the tube. as well as for infinite l and q. The present discussion for the periodically repeating quantum wells can be extended to finite systems. and the charge transfer between regions are of crucial importance for the band offsets and are incorporated self-consistently in the calculated average electronic potential. calculated for l = ∞ and q = ∞. are calculated. one assumes that E B is unchanged for finite l and q and determines the valence band edges of a particular superlattice ¯ with given l and q by adding E B |l(q)=∞ to the corresponding average potential values Vr. i. and metallic connects of the device can be achieved by metal coating [102. Figure 32 presents a schematic illustration of the band diagram obtained in [176].l + E g1 and E C.q . The energies of the conduction band edges of the superlattice are determined by adding band gaps of different zones (calculated for l = ∞ and q = ∞) to the energies of the valence band edges.e. are calculated. calculated for the l and q zones.l(q) and band gaps. i.q − E C.q + E g2 . A schematic illustration of the energy band diagram and band offsets estimated by using ¯ averaged self-consistent electronic potentials.91 eV.l(q)=∞ − Vr. E g1 and E g2 are the band gaps of hydrogenated and hydrogen-free zones.e. i. the energies of the valence band edges of two adjacent zones (i. hydrogen-free and exohydrogenated) are fixed. At the end. The switching of the current in a s-SWNT at room temperature by an external electric field has been utilized to fabricate a new field effect transistor (FET) [99–101].l = E V. (Reproduced from [176]. The interaction between the s-SWNT and metal electrode and the Schottky barrier. E B for (C32 )l=∞ and (C32 H32 )q=∞ is found to be 2. Also a metal–semiconductor heterostructure can be formed by modulating the hydrogenation of an (n.62 and 3. made by a zone (C32 )l placed between two zones of (C32 H32 )q . ¯ i.e of the l and q zones. Experimental and theoretical methods have been proposed in the past to determine the band offsets and hence to obtain the band diagram perpetuating along the semiconductor heterostructures. Vr.e.e (z) for finite l and q. the average electronic potentials corresponding to Vr.

The SCF electronic potential between the SWNT and the Au electrode. there are only minute changes. 0) tube is essential for the contact. (d) The variation of Ve (z) on a perpendicular line passing through the centre of the SWNT. (a) Local densities of states (LDOSs) at different Au and C atoms of the (8. (b) Contour plots of the SCF electronic potential. Other densities are for single atoms. ρ = ρ[SWNT + E M ] − ρ[E] − ρ[SWNT]. (c) As (b) but on a horizontal plane bisecting s. The calculated total charge density of the SWNT bonded to the metal electrode.Topical Review R947 Ve(z)[eV] 10 5 0 –5 3 6 9 12 15 18 z(Å) 21 24 Figure 33. Ve (r) > E F . whereas the LDOS of the carbon atoms at the contact still has a band gap. Ve (r). In the dark grey regions. c is calculated to be ∼3. so a surface Au atom faces the centre of the hexagon. 178]. This implies a weak Au–SWNT interaction which is characterized by physisorption. Model calculations have been carried out to provide further understanding of experimental I –V characteristics [177. i. show minute charge rearrangements. Figure 33(d) shows the variation of Ve (z) on a line passing through the centre of the SWNT and perpendicular .9 eV—that is. on a vertical plane. The LDOS at the carbon atom which is furthest from the metal electrode displays a state distribution similar to that of the carbon atom in the contact region.18 Å according to full structure optimization. The Fermi level lies near the top of the valence band of the SWNT and is consistent with a small B in the hole doping picture. occurred at 3. is presented on a vertical plane and also on a horizontal plane bisecting s in figures 33(b) and (c). ρ[SWNT + E M ]. 0) zigzag SWNT side-bonded to the Au(100) surface. The locations of atoms are shown in the inset.) thereof have been proposed as the origin of the FET operation. The zero of energy is taken at the Fermi energy E F . At the contact midway between the SWNT and Au(100) surface. s. These two LDOSs are similar. comparable with the calculated work function ( ∼ 5 eV) of the Au slab. State densities shown by empty circles and filled diamonds correspond to six neighbouring carbon atoms. The interaction and energy of binding between an individual Au atom and (8. Ve (r). The shaded area shows that the electronic potential energy at the contact yields a potential barrier c = Ve − E F > 0. The LDOSs calculated for two Au atoms at different sites (one is at the contact just below the SWNT hexagon. The origin of the potential barrier formed between an s-SWNT and a specific metal electrode has been investigated recently by using first-principles calculations [179]. The contact on the gold electrode is formed by placing the s-SWNT on the Au(100) surface. (Reproduced from [179]. These results indicate that the weak Au electrode–SWNT interaction does not induce any significant changes in the electronic structure. E M . The Au–SWNT equilibrium distance. and the difference charge density.e. the second being furthest from the SWNT mimics the clean Au surface) in figure 33(a) have finite state density at the Fermi level.

The effective potential barrier. (a) Local densities of states (LDOSs) at different Mo and C atoms of the (8. The locations of atoms are shown in the inset.R948 Topical Review Figure 34. as shown in figure 34. The above situation is. the tube has rotated slightly so C atoms tend to approach Mo atoms. The strong C–Mo bond formation is clarified by the calculated total SCF charge density in figure 34(b). the peak near E F is associated with the C–Mo bond states and hence may be attributed to the metal induced gap states (MIGS). 0) zigzag SWNT side-bonded to the Mo(110) surface.96 Å) than that with the Au electrode. while the opposite site furthest from the contact remains semiconducting. different in the case of an s-SWNT side-bonded to the Mo(110) surface. c. The analysis of the difference charge density ρ shows . The LDOS of the C atoms furthest from the contact has a band gap near E F . the spacing between the SWNT and the Mo electrode is smaller (s = 1. Therefore. State densities shown by empty circles and filled diamonds correspond to eight neighbouring carbon atoms. (b) Contour plots of the total SCF charge density on a vertical plane. the LDOS at the carbon atom closest to the Mo surface has finite state density at E F . however. State densities shown by empty circles correspond to ten neighbouring carbon atoms. In particular.) to the Au(100) surface. This situation indicates that the site of the SWNT forming contact is conducting. Owing to the strong Mo–C bond. Upon relaxation. Moreover.eff (r).eff . (Reproduced from [179]. can be even higher owing to the increased confinement of electrons in the contact region. The LDOS at the Mo atom interacting with the C atoms of the SWNT is different from that at the clean Mo(110) surface. The others are for single atoms. The zero of energy is taken at the Fermi energy E F . electrons which are transferred from the metal to the s-SWNT have to tunnel through a potential barrier c.

A large spacing and sizable potential barrier between the tube and metal electrode. does not form at the extended contact between the SWNT and the Mo electrode. The optimal electronic transport between nanotubes occurred when the surfaces of SWNTs were in atomic registry. In particular. Here. c . The current in the metal ring that can induce such a high magnetic field is comparable to the current attained in suspended. we also mention an interesting work by Buldum and Lu [181] on the intermolecular junctions of SWNTs. Upon radial deformation. because they may lead to the realization of nanosolenoids on SWNTs. They found atomic scale characteristics in the quantum transport properties of junctions between two SWNTs. The electronic potentials calculated on the horizontal plane bisecting s and also on a line from the SWNT to the Mo(110) surface through the contact show that Ve (r) < E F in the contact region. Metal nanorings are of particular interest. c ∼ 3. exhibit marked differences from those of metal–Si heterostructures. monatomic . we note that the electronic properties of present metal–SWNT junctions. which is connected to two flat Al substrates to a weaker extent. Accordingly. Taking the lowest possible value for L z and r = 5.9 Å.Topical Review R949 the strong charge transfer to the C–Mo bonds. we estimate B to be of the order of 100 G. Similar features have been observed recently in STM studies using multiwall carbon nanotube tips and Au(111) samples [180].9 eV. A small B is estimated for the p-type character. Metal nanorings. 4πmr 3 (19) where r is the radius of the nanoring. The small radius of a metallic nanoring wrapping the carbon nanotube may lead to interesting electromagnetic properties. resistors and amplifiers. The small cross section of the tube prevents us from identifying a band diagram across the diameter. which are known to be insensitive to the type of metal. the s-SWNT–Au contact is reminiscent of the metal–oxide– semiconductor (MOS) junction. However. Because of weak coupling and hence finite c . It has been suggested that the monitoring of contact resistance may lead to new atomic scale switches. A similar effect was predicted earlier for an Al chain. the contact resistance was strongly dependent on the local atomic configuration of the contact. a potential barrier. Finally. The magnetic field B at the centre of the ring can be expressed in terms of the quantized angular momentum L z of the electrons in the direction parallel to the tube axis B= µ0 eL z . It was found that the quantum conductance of the chain depends on the position of the Al atom (whether in a top site or a hollow site) where the chain is attached to the electrodes [183]. 0) SWNT is identified as being semiconducting even after the contact has been set with the Au surface. However. are characteristics of the contact made with the Au surface. Their predictions were confirmed experimentally [182]. c has disappeared due to strong coupling between the states of the s-SWNT and Mo(110) [91]. It appears that the above results distinguish two different types of electronic structure for two different electrodes. This explains why the devices made from Au electrodes have high contact resistance. But the bands are normally bent along the axis of the s-SWNT and the height of the barrier is monitored via the applied gate voltage. as illustrated in figure 35. in particular the Fermi level pinning. the contact resistance can vary over several orders of magnitude when the relative atomic configuration at the contact has been changed. weak coupling of electronic states may not lead to MIGS and the (8.4 eV for the p-type character. s decreases and eventually c collapses. A crude estimate based on the LDOS suggests that B ∼ 0. At this point. The height of the Schottky barrier B that forms at the SWNT–Mo(110) junction having a finite contact region followed by a free s-SWNT depends on the position where the Fermi level is pinned in the gap of the s-SWNT.

(b) The contact resistance of the (18. 0) SWNT. 10) junction. The tubes are in registry at = 30◦ . 10) junction as a function of the rotation angle . In this kind of experiment the main problem is isolating the superconducting quantum interference device (SQUID) from the outside (nonquantum) environment and that is why isolated carbon nanotubes can be very useful. (Reproduced from [181]. they tend to form clusters or nanoparticles (i. because they may allow the generation of persistent current at the nanoscale. We first discuss the zigzag Al nanoring coverage which is obtained by placing Al atoms on top of carbon atoms . It has been observed that a weak microwave. Metal nanorings winding the SWNT are also important. The tubes are in registry at = 0◦ .) gold chains [88–90]. Bagci et al showed that at special configurations both a stable nanoring and a metal tube can be formed. 60◦ . 0)–(10. Moreover. 150◦ . (c) The resistance of the (10. 10)–(10. Persistent currents in the nanoring can be started by sudden application of an external magnetic field. as a local magnetic probe at the nanoscale. with a ring at its end. Miyamoto et al [184] have examined the chiral conductivity in bare BC2 N nanotubes. 90◦ . 120◦ .R950 Topical Review Figure 35. Motivated by these exciting future applications. A superconducting ring may also be used for Schr¨ dinger’s o cat experiments where one deals with the superposition of macroscopic quantum states [186]. rotation angle. these structures become stable even if the underlying SWNT is taken out. can cause quantum tunnelling between these two macroscopic states. which does not break Cooper pairs. 10) junction as a function of one tube relative to the other in the x and y directions. 180◦ . Bagci et al [109] studied the formation of Al coverage and Al nanorings around a (8. They estimated that a magnetic field of a few tenths of a tesla can be induced at the centre of the tube by assuming a relaxation time for carriers ∼50 times larger than that in Cu and a homogeneous chiral current density confined to the tubule wall. (a) A model of a four-terminal junction formed by crossing two nanotubes. and the translation directions are shown. The terminal labels.e. . 3D islands) on the surface of the SWNT. but not so strong Al–SWNT interaction. 0)–(10. Relatively higher magnetic fields of the order of teslas can be induced by a higher current passing through a thick Ti based metal coating around the SWNT. The persistent currents induced around a SWNT have been proposed as possible qubits in quantum computation by Kulik et al [185]. (d) The resistance of the (18. Since aluminium atoms have strong Al–Al interaction. The two supercurrent quantum states (clockwise and anticlockwise flow) sit in two separate quantum wells. The tubes are considered to be rigid. However. or by increasing the number of turns and hence by forming a nanocoil. The current is passing between 1 and 4 and the voltage is measured between 2 and 3. In this way it is possible to use a nanotube.

forming a zigzag ring (see figure 36). 0) SWNT is illustrated in figure 36. except that it is rolled on a cylinder. Interestingly. The SWNT initially serves as a template in the formation of the ring structure and also increases the stability of the ring by preventing transitions to other relatively more stable structures. A side view of the optimized structure of the Al nanoring wrapping the (8. 187–191]. This structure includes 64 C and 16 Al atoms in the double unit cell.15 electrons are transferred from each Al atom to the SWNT.85 eV. Yang et al [111] performed . After structure optimization. 0) SWNT. A view of the optimized structure of the Al zigzag nanoring formed on a (8.9 Å. For the case of the Al nanoring wrapping the nanotube. Ta). 94. The final optimized structure of the zigzag Al nanoring around the (8. the Al ring is a conductor that incorporates two channels with an ideal ballistic quantum conductance of 4e2 / h. first-principles calculations by Durgun et al [102. In section 5. Recent studies on low dimensional structures of metals have revealed several stable atomic structures in one dimension (1D) [37. Therefore. 0) nanotube has structural parameters similar to those of the planar zigzag structure [110]. Nanomagnets. The electronic properties of the zigzag Al nanoring system are also quite interesting. 110. From the spintronics and molecular magnetism point of view. The electronic energy bands of the Al metal ring (without a SWNT) are derived from the dispersive bands of the flat zigzag Al chain [110]. we showed that nanomagnets can form once the SWNT is fully covered or filled by the specific transition metal atoms.2. these states are mixed with the states of the nanotube and give rise to the bands and a density of states. which contributes a metallic behaviour to the combined Al nanoring and (8. the Al–Al bond length is increased to 2. The binding energy of the Al nanoring is calculated to be 0. Most importantly. the Al nanoring around the SWNT is expected to be stable at room temperature. Therefore. its bands are zone folded at the point and they appear as a number of discrete energy levels.) (T site). yielding the radius of the nanoring as 5. the nanoring is also stable by itself. According to Mulliken analysis. the subject is extremely important and requires further study of the magnetization of transition metal atom covered SWNTs. 103] present evidence that one can generate nanomagnets from SWNTs decorated by transition metal atoms (such as Ti. In this initial configuration the Al–Al distance is 2. since the positions of Al atoms do not change significantly upon discarding the underlying carbon nanotube.33 Å and the angle of the Al–Al–Al bond is ∼137◦. When the flat zigzag Al chain is rolled into a ring. (Reproduced from [109].56 Å and the Al–Al–Al bond angle is decreased to 124◦ . The stability of the Al nanoring around the nanotube can be understood from the stable structures of planar Al monatomic chains. 0) nanotube system.Topical Review R951 Figure 36. 0.

1. Devices fabricated using carbon nanotubes The objective of most of the effort made so far in the research on carbon nanotubes has been to discover a technological application. The realization of a singlemolecule device has been challenging due to difficulties in manipulating the molecule and achieving the electrical contact.6) SWNTs. For Vgate > 0 the bands bend more strongly. Intensive laboratory work combined with theoretical studies during last decade has made remarkable advances in methods and techniques which were able to produce novel electronic devices based on carbon nanotubes. for Vgate < 0 the bands are flattened. the bands bend towards lower energy in between the electrodes. A single s-SWNT is contacted by two electrodes. the valence band edge is pinned to the Fermi energy of the leads. (Reproduced from [99]. leading to an insulating state. At A and C. Nanoelectronic or molecular electronics considered as ultimate miniaturization [192] has attracted much of the interest. indicated by the black vertical bars. 8. (9.6 eV is connected to the leads with Fermi energy E F by tunnelling contacts. A semiconducting Si substrate covered with a 300 nm layer of thermally grown SiO2 was used as a back-gate (see figure 37). Transistors based on carbon nanotubes In 1998 Tans et al [99] achieved the fabrication of a field effect transistor (FET). (d) Application of a bias voltage results in suppression of the barrier. (c) The suggested band diagram of the device.R952 Topical Review Figure 37. P = (D↓ (E F ) − D↑ (E F ))/(D↓ (E F ) + D↑ (E F )) (P = 86%) for the spin valve effect. These devices have kept the great expectations for these materials alive and have continued to motivate further research. which consists of a semiconducting SWNT connected to two metal electrodes. 0) tubes and found that these systems exhibit substantial magnetic moments (1. (a) A tapping mode AFM image of an individual carbon nanotube on top of three Pt electrodes. Dag et al [7] have obtained a significant net magnetic moment for Ti covered (8. acts as a back-gate. Owing to a difference in work function between the tube and the electrodes. 0) and (6. (b) A schematic side view of the FET device. resulting in a metal-like conductance.4 µB per Co atom) and have a very high spin polarization. The Si substrate. The nanotube with a gap of ∼0. Some of these devices will be reviewed in this section. The nanotube was able to switch from a conducting to an insulating state at room temperature as . which is covered by a layer of SiO2 3000 Å thick. 8. 0).) first-principles calculations of Co filled and Co coated (9.

A change of 10 V in Vgate resulted in modulation of the conductance by about six orders of magnitude at Vbias = 0. The conductance through this single-molecule switch can be varied over at least six orders of magnitude and saturates at 10−6 −1 for negative Vgate . A local polarization layer develops at the interface until the valence band edge of the SWNT aligns with the Fermi level of the metal electrode. Degenerately doped Si wafers with 5000 Å thick thermally grown oxide on the surface were used as the substrate.) a result of the voltage applied to the gate. It has been argued that using special geometry. the device would allow a maximum frequency of the order of 10 THz. A negative Vgate leads to ohmic behaviour while a positive Vgate results in a strong suppression of the current at low bias voltage and nonlinear I –Vbias curves at higher bias. As shown in figure 38. by molecular self-assembly techniques. reducing the contact resistance and the thickness of SiO2 . Also its gain can be raised by one order if the thickness of SiO2 is decreased from 3000 to 50 Å. The lengths of the SWNTs between electrodes were less than 30 µm. It has also been foreseen that the integration of multiple devices into a circuit would be possible. at Vgate = 0.Topical Review R953 Figure 38. Data were taken at room temperature and in vacuum (∼10−4 mbar) with the voltage applied to contacts 1 (drain) and 2 (source). The heavily doped substrate is conducting at low temperature and was used as a back-gate. has been attributed to the absence of a Schottky barrier. namely a mild temperature dependence. Zhou et al [98] measured I –V characteristics of devices obtained with s-SWNTs contacted by 200 Å thick Ni electrodes with 600 Å thick Au on top. I –Vbias exhibited a small nonlinearity. the I –Vbias curve becomes linear with a resistance that saturates around 1 M . At negative Vgate . (Reproduced from [99]. It was assumed that the Fermi level is pinned at the valence band edge of the SWNT at the contact. The band bending length was estimated to be 400 nm. This behaviour of the device. Electrical transport through an s-SWNT involved thermal activation at high temperature and tunnelling behaviour through a . since the FET has the same room temperature resistance of ∼1 M and a similar temperature dependence. Since the proposed band diagram of the device was reminiscent of that of the barrier injection transit time diode (BARITT). Inset: conductance at Vbias = 0 as a function of Vgate . a pronounced gap-like nonlinearity developed around Vbias = 0 with increasing Vgate . According to the electronic structure and energy diagram of the FET modelled by Tans et al in figure 37(c) the difference in work function will result in bending of the bands of the semiconducting SWNT. Later. its operation is qualitatively described in terms of the BARITT model. Two-probe I –Vbias curves for various values of the gate voltage (Vgate ). However.

Because λ depends on the E g of the s-SWNT. Different samples exhibited I –V characteristics resembling that of the Si based p metal–oxide–semiconductor field effect transistor (MOSFET).e. The positive voltage gain of ∼3 recorded in this device was almost one order of magnitude higher than the maximum gain of ∼0. whereas the true Schottky barrier height is as high as one expects (i. as a result of local gating. This allowed the FET device to be ambipolar with low contact resistance for both n. They presented measurements on long. the wavelength of the emitted light can be adjusted easily by using an s-SWNT with a different radius. showing field effect mobility in semiconducting SWNT transistors as high as 79 000 cm2 V−1 s−1 at room temperature. These values indicate great progress. As a result. Because of their quasi-one-dimensional structure the electronic behaviour of nanotube contacts is found to be sensitive to the electrostatics [193]. Under high Vbias. The low barriers for both electrons and holes have been attributed to the effective barrier heights being strongly modulated by the electrostatic field of the gate. Martel et al [101] realized an end-bonded SWNT–metal contact. placed on a Si wafer with ∼1 µm thick thermally grown SiO2 . The charge carrier mobility is crucial for the applications such as in few. Instead of a Schottky barrier as in previous studies. The needle-like shape of the metallic carbide contacts focuses the electric field and hence the gate can induce a stronger electric field at the junction. High contact resistance between the side-bonded s-SWNT and high work-function metal electrode is in accord with the first-principles calculations [179]. B = E g /2). in the latter device the apparent barrier height for carrier injection is modulated by the gate field. Both transistors discussed above have been realized by the side bonding of SWNTs. since the advance requires assembly strategies that allow precise localization and interconnection. The symmetry and narrowness of the Schottky barrier for both holes and electrons in a FET based on an s-SWNT have led to an interesting application.8–1.35 obtained by Tans et al [99].5 µm. ohmically contacted nanotube devices. it is so thin that carriers can easily tunnel through the barrier.and singleelectron memories and chemical/biochemical sensors. Devices showing ohmic I –V curves in both strong accumulation and inversion regimes have been attributed to the very small Schottky barrier at the contact. or one can tune the emission to any wavelength by radial deformation. the work by Durkop et al [195] exhibited an ohmic contact behaviour.and p-type conduction.R954 Topical Review reverse biased metal–SWNT junction at low temperature. Injection of electrons and holes simultaneously through the narrow barriers at the source and drain contacts has been possible at room temperature. It has been argued that this fabrication technique . and the intrinsic mobility in s-SWNTs is even higher. Fermi level pinning appears to be relatively unimportant for nanotubes end-bonded to TiC. In contrast to the above planar devices. Soh et al [197] achieved the synthesis of SWNTs by chemical vapour deposition (CVD) of methane at controlled locations on a substrate using patterned catalytic islands. research on SWNT based electronics has focused on the integration of FETs on a chip. A FET was made of a semiconducting SWNT of ∼8000 Å length and 14 Å diameter contacting TiC and passivated with a uniform SiO2 layer. The transconductance was measured to be ∼200 nA V−1 which is two orders of magnitude higher than the similar results obtained by Martel et al [101]. These measured values exceed those of any other semiconductor at room temperature. This new result holds promise that SWNTs can enable nanoscale photon sources for future optoelectronics and fibre optics [194]. The CVD synthesis and microfabrication techniques were combined to produce a large number of ohmically contacted nanotube devices with controllable length. These devices generally have high contact resistance and unipolar (p-type) transport characteristics. These injected carriers have recombined and emitted photons with energy hν = E g and wavelength λ ∼ 0. Beyond the single FET. I –V characteristics of s-SWNTs exhibit pronounced asymmetry with respect to the bias polarity.

which. Owing to their bulky size and high power consumption. the adsorption of atoms or molecules such as NO. Their transistors show device characteristics such as high gain (>10). Keren et al [196] reported the realization of a self-assembled carbon nanotube FET operating at room temperature where a DNA scaffold molecule provides the address for precise localization of an s-SWNT as well as a template for the extended metallic wires contacting it. the scope of this review has been focused on a limited field covering atomic and electronic structure of single-wall carbon nanotubes. That makes the sensor portable. Chemical sensors As discussed in section 3. 63]. The device could be switched on and off by applying voltage to it. The sharp tips of MWNTs generate very high electric field at relatively low voltages. In the operation of the sensor. a large on–off ratio (>105 ) and room temperature operation. battery powered and safe operation of such sensors. This feature enables compact.Topical Review R955 allows large numbers of integrated nanotube circuits to be built reliably with a contact resistance of ∼10 k (or 20 k two-terminal resistance in the ballistic regime). In the course of active research towards nanoelectronics. which can be provided by special batteries. which involves patterned growth of SWNTs to bridge predefined Mo electrodes. Such voltages can be easily obtained by series connection of carbon–zinc batteries. NH3 and O2 on s-SWNTs has been reported to change the electrical resistance and similar properties. They then made gold nanowire out of DNA molecules at each end of the nanotube. This sensor and the underlying technique that it uses are very convenient. Bachtold et al [199] demonstrated logic circuits with FETs based on SWNTs. because: (i) the precise breakdown voltage provides the fingerprint for the gas to be identified. In this way the formation of a corona is promoted at relatively low voltage. study of carbon nanotubes has seen a tremendous explosion of work. attached to the substrate in the head-on position. it is found that while the breakdown voltage is independent of the gas concentration. lowering breakdown voltages several times in comparison to those of traditional electrodes. by binding together with proteins. Despite the large number of very important research works. 9. the current discharge is linearly dependent on the concentration. has a tip radius of 150 Å. the gas concentration can be determined. in turn generates a very high nonlinear electric field. The breakdown voltage was found to become lower as the interelectrode spacing was reduced. with ∼500 Å separation) was grown by CVD on a SiO2 substrate.2. . This approach affords SWNT based devices without any postgrowth processing. they attached a SWNT onto a specific site on a DNA strand. very important for various applications. Much recently. in fact. The breakdown voltage went down from 280 V (at 150 µm) to 130 V (at 25 µm separation). An individual MWNT. It has been suggested that this effect of carbon nanotubes can be used in chemical sensors to detect specific molecules in the environment [62. controlled dc voltage is applied between an anode consisting of vertically aligned MWNT film and a cathode made from Al sheet. Conclusions Because of growing interest in nanotechnology and the quest for novel nanodevices. In order to get the transistor to self-assemble. Modi et al [200] achieved production of a miniaturized gas ionization sensor a few microns in length using MWNTs. Later Franklin et al [198] devised a method for reliable integration of long suspended SWNTs into electrically addressable devices. ∼30 µm long. A MWNT film (consisting of MWNTs 250–300 Å in diameter. (ii) by monitoring the self-sustaining discharge current. (iii) the response time is short [200]. In this sensor. 8. the miniaturization of chemical sensors is.

Dresselhaus G and Dresselhaus M S 1998 Physical Properties of Carbon Nanotubes (London: Imperial College) [4] Zhao X. Robertson D H and Mintmire J W 1993 Phys. Lett. Ichihashi T and Ando Y 1992 Nature 356 776 [2] Dresselhaus M S. A 69 255 [7] Dag S. Here we dealt with specific cases wherein the properties of the nanotube undergo a dramatic change upon physisorption of foreign atoms. Durgun E and Ciraci S 2004 Phys. Xia X. 81 1278 [21] Collins P G. show that carbon nanotubes are indeed nanostructures which may potentially allow several technological applications. It has been shown that bare nanotubes have many important properties which are suitable for use in potential technological applications. Lett. 65 153405 u [19] Charlier J C. Lett. Rev. as explained at the end. Rev. McEuen P L. Ando Y. Thess A and Smalley R E 1997 Science 275 1922 [23] Bezryadin A. Tans S J and Dekker C 1998 Phys. Finally. Even more interesting is that these changes can be tuned reversibly. Not only functionalization but also radial deformation has been seen to modify the nanotube properties. CA: Academic) [3] Saito R. Lett. Rev. B 69 165415 [9] Nevidomskyy A H. Li Y. various physical properties of carbon nanotubes and specific applications using these properties have been investigated. B 69 121407(R) [8] Li F. these options are multiplied several times by the functionalization of tubes. Dresselhaus G and Saito R 1992 Phys. Cheung C L and Lieber C M 2001 Science 292 702 [17] Kim W. Zettl A. Liu X. Shim M. Chopra N G. 68 1579 [12] Mintmire J W. Lopez Sancho M P and Munoz M C 1998 Phys. which were proven to be appropriate and to yield accurate predictions on the atomic and electronic structure of carbon nanotubes. Jinno M and Suzuki T 2003 Phys. Buldum A and Batra I 2001 J. That a (n. Yildirim T and Ciraci S 2002 Phys. Rev. Zettl A. Huang B. B 45 6234 [11] Hamada N. Rev. 15021 and TUB˙ References [1] Iijima S 1991 Nature 354 56 Iijima S. Phys. Moreover. Our discussions have been guided by first-principles calculations within density functional theory. Rev. Phys. Dresselhaus G and Eklund P C 1996 Science of Fullerenes and Carbon Nanotubes (San Diego. Acknowledgments This work was partially supported by the National Science Foundation under Grant No INT01¨ ITAK under Grant No TBAG-U/13(101T010). m) SWNT can be metallic or semiconducting depending on n and m offers a wide range of options. Bando H.R956 Topical Review In this review article. 2 703 [18] G¨ lseren O. Rev. Rev. Ebbesen T W and Lambin Ph 1996 Phys. Zhao M. 80 4036 . Sawada S and Oshiyama A 1992 Phys. Wang D and Dai H 2002 Nano Lett. Dunlap B I and White C T 1992 Phys. Lett. we discussed how properties of bare nanotubes as well as those induced by functionalization or radial deformation can be exploited for the design of new molecular devices. A semiconducting nanotube can be metallic under a radial strain which changes the circular cross section into an elliptical one. 90 187401 [5] Yu M F et al 1998 Science 287 637 [6] Salvetat J P et al 1999 Appl. Csanyi G and Payne M C 2003 Phys. Rev. Liu Y.: Condens. Rev. B 53 11108 [20] Chico L. Cobden D H. Lett. Matter 13 537 [15] Frank S. Thess A and Smalley R E 1997 Science 278 100 [22] Bockrath M. Rev. Choi H C. Poncharal P. 68 631 [13] White C T. 91 105502 [10] Dresselhaus M S. Wang Z L and Heer W A 1998 Science 280 1744 [16] Ouyang M. Verschueren A R M. Rev. Huang J. Transistors and chemical sensors fabricated by using carbon nanotubes. B 47 5485 [14] Ciraci S. Tan Z and Ji Y 2004 Phys.

Rev. Yildirim T and Ciraci S 2002 Phys. 85 2757 Moon C-Y. Rev. Challa S R. B 105 11227 Zhu X Y. IIjima S. Brenner D W and Mintmire J W 1992 Phys. Brenner D W. Ciraci S. Rev. 82 956 Gordillo R C. Rev. Arias T A and Joannopoulos J D 1992 Rev. B 62 16345 c ¸ u Sim H S. Chapline M G. Han S and Ihm J 1999 Phys. Gong X G and Liu Z F 2001 Phys. Huang J. 116 9014 Ricca A and Drosco J A 2002 Chem. Growth 66 632 Kudin K N. Ciraci S and Kili¸ C 2002 Phys.Topical Review [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] R957 [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] Kili¸ C. Rev. Wu X. B 64 235406 Yorikawa H and Muramatsu S 1995 Solid State Commun. Bekkadahl T A. Rev. B 64 075404 u Srivastava D. 85 2348 Simonyan V V. Car R and Scoles G 2003 J. 329 956 Ahn C C. Phys. Lett. Cheng H M and Dresselhaus M S 1999 Science 286 1127 Brown C M. Artacho E. Yu M and Ruoff R S 1999 J. Lett. G¨ lseren O. Rev. B 67 165424 u Giannozzi P. Jackson K A. Wang R and Mei L 2001 Phys. B 65 155410 u c ¸ Park C J. Rev. Jin Y-G and Chang K J 2002 Phys. Rev. B 65 155401 Park N. Neumann D A. Rev. Ajayan P M. B 63 115422 Tada K. Phys. Ratnakumar B V. B 62 5238 Odom T W. B 64 113402 Payne M C. B 63 073402 G¨ lseren O. Rev. Lett. Peng S. Rev. Soler J M. Rubio A and Ordejon P 1999 Phys. Phys. Rev. Rev. Vosko S H. Lett. Cong H T. Huang J. Phys. Yildirim T and Ciraci S 2003 Phys. 87 205502 White C T. Schall J D. B 63 155405 Yildirim T. Fan Y Y. Rev. Rev. B 45 12592 Tibbets G G 1984 J. B 66 121401 u Andriotis A N. Phys. Teter M P. Pederson M R. Chem. Janssen J W. Kim Y-S. 85 3832 Dean K A and Chalamala B R 2000 Appl. Alleman J L and Fisher J E 1999 Chem. Louie S G and Cohen M L 2000 Phys. Lett. Lett. Kim P and Lieber C M 2000 J. Lemay S G. B 60 10656 Dillon A C. Tanigaki K and Hiura H 1993 Nature 362 522 Ebbesen T W. Park C J and Chang K J 2001 Phys. Lett. Zhao M. Kouwenhoven L P and Dekker C 2000 o Phys. Menon M. Hiura H and Tanigaki K 1994 Nature 367 519 Morishita K and Takarada T 1997 Carbon 35 977 Jhi S H. Chevary J A. 85 1710 Sorescu D C. Appl. B 64 125401 Mann D J and Halls M D 2002 J. Gennett T. B 59 12678 Robertson D H. Cryst. Rev. Smalley R E and Dekker C 1998 Nature 391 59 o Venema L C. Shirley E L and Louie S G 1994 Phys. Phys. Allen D C. Phys. Chem. Kiang C H. Ichihashi T. Ausman K D. Xia Y. Phys. B 46 6671 Perdew J P. Rinzler A G. Wild¨ er J W G. Kim Y H and Chang K J 1999 Phys. B 103 4330 G¨ lseren O. Chem. B 63 073408 Kleiner A and Eggert S 2001 Phys. Rev. Boronat J and Casulleras J 2000 Phys. Lin J and Tan K L 1999 Science 285 91 Wang Q. Dillon A C. Rev. Phys. Scuseria G E and Yakobson B I 2001 Phys. 111 9778 Ma Y. Rev. Diep P and Johnson J K 1999 J. Franklin N R. Jones K M. B 52 2723 G¨ lseren O. 73 378 Chen C C. Witham C. Rev. Yildirim T. Rev. 87 116802 u Wild¨ er J W G. Phys. Kim P and Lieber C M 1998 Nature 391 62 Odom T W. Srivastava D and Froudakis G 2001 Phys. G¨ lseren O and Yildirim T 2000 Phys. Ye Y. B 47 5485 Kong J. Venema L C. 78 1932 Kleiner A and Eggert S 2001 Phys. Bradley K. Ishigami M and Zettl A 2000 Science 287 1802 Dean K A and Chalamala B R 1999 J. B 64 193401 Landauer R 1987 Z. Cho K and Dai H 2000 Science 287 622 Collins P G. Rev. Bowman R C and Fultz B 1998 Appl. Mod. Phys. Lett. Yildirim T and Ciraci S 2001 Phys. Rev. Lett. Rev. Lee E-C. Chem. B 46 6671 Sanchez-Portal D. Chem. Liu M. Chen G. Singh D J and Fiolhais C 1992 Phys. Chem. Benedict L X. Jordan K D and Avouris P 2001 J. B 68 217 Chan S P. Lett. 118 1003 . Lett. Lett. Rev. Rev. Rev. Heben M H. Lee S M. 72 1878 Kane C L and Mele E J 1997 Phys. Buitelaar M R. Rev. Robertson D H and Mintmire J W 1993 Phys. 64 1045 Perdew J P and Wang Y 1992 Phys. 76 375 Ajayan P M. Lee Y H and Frauenheim T 2000 Phys. Zhou C. 362 217 Dag S. Sholl D S and Johnson J K 1999 Phys. 94 435 Yorikawa H and Muramatsu S 1995 Phys. Ebbesen T W. Phys. Yildirim T. Phys. B 104 2794 Blase X. Bethune D S and Heben M J 1997 Nature 386 377 Liu C. Furuya S and Watanabe K 2001 Phys. Rev. G¨ lseren O and Ciraci S 2001 Phys.

Nakanishi T and Dekker C 2001 Science 294 1317 Nakanishi T. Phys. Moos G and Hertel T 2002 Phys. Hahn J R and Kang H 1999 Phys. Lett. Chem. 80 2273 [85] Delaney P. Hertel T and Avouris Ph 1998 Appl. Lett. Louie S G and Zettl A 1995 Nature 377 135 [115] Rochefort A. Rev. Lett. Eksp. Magn. B 66 041403(R) [109] Bagci V M K. Mater. Jayanthi C S. Ebbesen T W and Gibson J M 1996 Nature 381 678 [113] Ruoff R S. Crespi V H. Louie S G and Cohen M L 1999 Phys. Rev. 90 257203 [112] Treacy M M J. Tang M and Wu S 2000 Nature 405 769 . Wong E W. Lett. 85 1706 [108] Yang C-K. B 63 125415 [94] G¨ lseren O. Chen R J and Dai H 2000 Chem. Vanitha P V. Phys. 114 7827 [83] Ulbricht H. Lett. Dag S. Phys.) [80] Israelachvili J N 1985 Intermolecular and Surface Forces (London: Academic) [81] Ciraci S. Kong J and Dai H 2000 Appl. Kane C L and Dekker C 2000 Phys. Rev. 87 256805 Avouris Ph. Rev. Rev. Lett. Schmidt T. B 66 045409 u [110] Sen P. Teor. Lett. B 64 195420 [111] Yang C-K. Dai H. Alexseyev L. 10 7259 [106] Gao B. Magn. Liu L. Avouris P. Rev. Soc. B 60 7899 [86] Chan S P. B 47 12345 ı [89] Ohnishi H. Phys. Wen G H. Tombler T W. B 105 1157 [88] Agra¨t N. B 40 R11969 Tekman E and Ciraci S 1991 Phys. Franklin N W. B 66 075404 [84] Derycke V. Govindaraj A. Lett. Kong J and Dai H 2000 Phys. Rev. Zhao J and Lu J P 2002 Phys. Kleinhammer A. Brinson B E. Phys. Bower C. Kondo Y and Takayanagi K 1998 Nature 395 783 [90] Yanson A I. Phys. Chen G.—JETP 2 73 (Engl. Jayanthi C S. Magrave J L and Hauge H 2001 J. Mater. Fan S and Lieber C M 1995 Nature 375 769 Han W Q. Appenzeller J and Avouris Ph 2002 Appl. Rev. Buldum A and Batra I P 2001 Phys. Lesage F and Salahub D R 1999 Phys. Marcus C M. Lett. 77 3015 Zhang Y. Rodrigo J G and Vieira S 1993 Phys. G¨ lseren O. Rev. Fan S S and Li Q Q 1997 Science 277 1287 [97] Zhang Y and Dai H 2000 Appl. Phys. Quate C. Ihm J.: Condens. Rev. Lett. Martel R. Kong J and Dai H 1999 Appl. Rev. 84 4950 [118] Tombler T W. B 108 575 u [104] Zhang X X. 75 627 Zhou C. Rev. 297 45 [116] Rochefort A. Rev. Chem. Morpurgo A F. Ihm J. Appenzeller J. Huang S. Lett. Lett. Gao R and Wang Z L 2001 J. G¨ lseren O. Dai L. 90 086403 [87] Chiang I V. 331 35 [98] Zhou C. Bagci V M K. Tang X P. Rev. Lett. 380 1 u [78] Lee S M. Lesage F and Salahub D R 1998 Chem. Subramoney S and Chan B 1993 Nature 364 514 [114] Chopra N G. Alexseyev L. Chem. Besenhard J O. Smalley R E. Lett.R958 Topical Review [77] Dag S. Rev. Buldum A. Gong X G and Liu Z F 2003 Phys. Fleming L. Zhao J and Lu J P 2003 Phys. Avouris P. Wu S Y. Bollinger G R. Gedik Z and Ciraci S 2002 Phys. Tekman E. Zhou C. Rev. Lee Y H. Phys. Ciraci S and G¨ lseren O 2004 J. 84 4613 [96] Dai H. Matter 12 3349 Buldum A. Rev. Fiz. 73 2447 [102] Durgun E. Rev. Bachtold A and Dekker C 2002 Phys. Lett. Yildirim T. Yildirim T and Ciraci S 2003 Phys. Brom H E. Lorents D C. 29 94 Lifshitz E M 1956 Sov. Phys. Phys. 362 301 [105] Winter M. Choi H J. B 66 073307 [100] Soh H T. Raychaudhuri A K and Rao C N R 2002 Chem. Derycke V and Appenzeller J 2002 Physica B 323 6 Martel R. Tang M. Hwang Y G. Ercolesi F and Tosatti E 1998 Phys. Hadley P. Lavoie C. Lett. Rev. Phys. Am. Kong J. Han J and Lu J P 2000 Phys. Ciraci S and Fong C Y 2000 J. Baratoff A and Batra I P 1992 Phys. Lett. Phys. Agr¨ it N and Ruitenbeek J M 1998 Nature 395 783 a [91] Ciraci S and Tekman E 1989 Phys. Verscheuren A R M and Dekker C 1998 Nature 393 49 Bachtold A. Benedict L X. Shea H R. 307 1536 [107] Zhao J. 82 217 [79] Lifshitz E M 1956 Zh. Ciraci S. B 43 7145 [92] Buldum A. Chan K K. 80 3775 u Tosatti E and Prestipino S 2000 Science 289 561 [95] Yao Z. B 46 10411 [82] Halgren T A 1992 J. Rev. Tersoff J. Phys. Rev. Wu Y and Zhou O 1999 Chem. Derycke V. Spahr M E and Novac M P 1998 Adv. B 67 201401(R) u [103] Durgun E. G¨ lseren O and Ciraci S 2003 Chem. Choi H J. Lu Y Z. Transl. Zhou C. Kong J and Dai H 2000 Phys. Louie S G and Cohen M L 1998 Nature 391 466 Delaney P. Lett. Lett. B 60 13824 [117] Liu L. 47 208 [93] Ozpineci A and Ciraci S 2001 Phys. 231 L9 Satishkumar B C. Phys. Tersoff J and Avouris Ph 2001 Phys. Lett. Rev. Dag S. Leitner D M and Ciraci S 1999 Europhys. 84 5604 [99] Tans S J. Martel R. 76 1597 [101] Martel R. Cohen M L.

Rev. Kobayashi N and Tsukada M 1999 Phys. B 61 5939 Tang J. Louis E. Nardelli M B. Anantram M P. Rev. Wang J and Roland C 2001 Phys. Kili¸ C and Ciraci S 2000 Phys. Taylor J and Stokbro K 2002 Phys. Soc. Lett. Meunier V. Lett. 76 1561 ¨ u Ozt¨ rk Y and Ciraci S 2004 at press Rao A M et al 1997 Appl. Lett. 319 296 Teredesai P V. Rev. Rev. Rev. Guo H. Benedict L X. B 63 245410 Nardelli M B 1999 Phys. B 63 155412 Chico L. St¨ ckli T. Kulik A J. Lett. Barnett R N. 84 2917 Lang N D 1995 Phys. Taraschi G. Pantelides S T and Lang N D 2000 Phys. Larade B and Guo H 2002 Phys. Menon M. Bacsa R R. Lett. Jiang B. Taylor J. 82 343 Lambert P E. Rev. 84 979 Landman U. Taylor J and Guo H 2002 Phys. Lett. Rev. Lett. Rev. A 64 231 Venkateswaren U D. Lett. Rev. Teredesai P V. Louie S G and Cohen M L 1996 Phys. B 56 4351 Choi H J and Ihm J 1999 Phys. Ordejon P. Qin L C. Lu J P and Kuhn D 2000 Phys. Wang J and Guo H 2000 Phys. Lett. Wan C C. B 59 10928 Sood A K. Sasaki T. Phys. Brandbyge M and Stokbro K 2002 Phys. Lett. Rev. Ihm J. B 60 17064 Mehrez H. Yaish Y. 81 3515 Yaliraki S N. Wan C C. Louis E. Mehrez H and Roland C 2001 Phys. Lett. 85 1456 Mazzoni M S C and Chacham H 2000 Appl. Brink M and McEuen P L 2003 Phys. Phys. B 66 035332 Taylor J.Topical Review [119] [120] [121] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] R959 [133] [134] [135] [136] [137] [138] [139] [140] [141] [142] [143] [144] [145] [146] [147] [148] [149] [150] [151] [152] [153] [154] [155] [156] [157] [158] [159] [160] [161] [162] [163] [164] [165] [166] [167] [168] [169] [170] [171] Chesnokov S A. SanFabian E and Verges J A 2002 Phys. o o Rev. Rev. Rev. B 60 13874 Yang L and Han J 2000 Phys. 71 419 Wan C C. Lett. Rev. Sazonova V. Rinzler A. Mingo N. B 58 13138 Mozos J L. Taylor J. Nalimova V A. Wang J and Guo H 1997 Phys. B 65 205416 . Guo H. 84 3634 Lordi V and Yao N 1998 J. Chem. Yoon Y G and Louie S G 1999 Phys. Lett. 84 2453 Minot E D. Ihm J. Rev. Taylor J. Mozos J L. Rev. Guo H and Wang J 2001 Phys. B 66 035322 Taylor J. Rev. Rev. Rev. Burnham N A and Forr´ L 1999 Phys. Rev. Rev. Phys. 80 4277 Taraschi G. Rev. McNeil L E. Chem. Lett. Lett. Louie S G and Cohen M L 2000 Phys. Lett. Rev. Rev. Guo H. 80 357 Emberley E G and Kirczenow G 1999 Phys. Lett. Park J-Y. 84 1974 a Palacios J J. Wlasenko A. Miranda R. Larade B. B 55 13393 Wang J. Sood A K. B 65 165401 Palacios J J. Rev. Yeyati A L. Guo H and Wang J 1998 Phys. Rev. Kwon Y-K. Wang J and Roland C 2001 Phys. Kowalewski T and Ruoff R S 2000 Phys. B 59 2267 Choi H J. Perez-Jimenez A J. Rev. Rev. Rev. Rev. B 63 121104 Larade B. B 54 2600 Wan C C. Tom´ nek D and Lambert C J 2000 Phys. Rev. Mozos J L. Mehrez H. Rev. B 60 7828 Orlikowski D. Mozos J L. 90 106801 e e a e Kaun C-C. Taylor J. P´ rez-Jim´ nez A J. Yudasaka M. B 63 245407 Brandbyge M. Wang J and Guo H 1997 Phys. 85 1887 Pernas P L. 109 2509 Yu M F. Rev. Larade B. Rev. B 62 12648 u c ¸ Yang L. Am. B 41 8553 Tian W and Datta S 1994 Phys. B 60 14009 Choi H J. Zhang P and Crespi V 2000 Phys. Govindaraj A and Rao C N R 2000 Chem. 85 1958 Taylor J. Guo H. Rev. Rev. Rao A M. 84 2682 Roland C. Smalley R E and Eklund P C 1999 Phys. Rev. Wang J and Guo H 1997 Appl. Rev. Briggs G A D. Smalley R E and Fischer J E 1999 Phys. Taraschi G. Richter E. Rinzler A G. Taraschi G and Zheng Q 1998 Phys. B 52 5335 Hirose K and Tsukada M 1995 Phys. Muthu D V S. 310 296 Peters M J. Phys. Lett. Wang J. Rev. 90 156401 Salvetat J P. Rev. 84 2921 Wei Y. B 60 6028 Brandbyge M. Martin-Rodero A and Flores F 1998 Phys. Rev. Hernan O S. Sen R. Grutter P and Guo H 2002 Phys. Rev. Lett. SanFabi´ n E and Verg´ s J A 2003 Phys. Muthu D V S. Han J and Lu J P 1999 Phys. Guo H and Wang J 2001 Phys. 85 154 Shen W. Lett. Scherbakov A G and Avouris P 2000 Phys. B 63 195412 Mehrez H. Wang J and Roland C 2000 Phys. G¨ lseren O. Lett. B 64 075420 Roland C. B 49 5097 de Parga A L V. Mozos J L. Lett. Rev. Rev. 82 944 Yildirim T. 121 3428 Di Ventra M. Kemp M and Ratner M A 1999 J. Mehrez H and Guo H 2001 Phys. Gavindaraj A and Rao C N 2000 Chem. Matsushita A and Iijima S 2000 Phys. Bonard J. Martin-Rodero A and Flores F 1990 Phys. Sen R. Mozos J L. Guo H. Mehrez H. Wan C C. Phys. Han B S and Xie S 2000 Phys. B 65 195419 Sanvito S. Rev. Rev. B 51 5278 Lang N D and Avouris P 1998 Phys. Rev. Lett. Phys. 89 138301 Mehrez H.

Barnett R N. Corsa A D. 83 3180 u [180] Luna M et al 2003 Ultramicroscopy 96 83 [181] Buldum A and Lu J P 2001 Phys. Wel B and Ajayan P M 2003 Nature 424 171 . Phys. Patel V. Yildirim T and Ciraci S 2003 Appl. Lett. 76 2121 Miyamoto Y 1996 Phys. Morpugo A F. Cobas E and Fuhrer S 2004 Nano Lett. Rev. Berman R S. Lass E. Rev. Rev. Louie S G and Cohen M L 1999 Phys. 76 971 [174] Esaki L (ed) 1991 Highlights in Condensed Matter Physics and Future Prospects (NATO Advanced Science Institute vol 285) (New York: Plenum) [175] Colombo L. Yildirim T and Ciraci S 2003 Phys. Appenzeller J and Avouris Ph 2002 Phys. Ciraci S and Batra I P 1997 Phys. Benedict L X. Phys. Rev. Rev. Phys. Getty S A. 63 161403 [182] Paulson S. Javey A. Lett. Sivan U and Braun E 2003 Science 302 1380 [197] Soh H T. Phys.R960 Topical Review [172] Saito R. Koratkar N. Tombler T W. B 56 12632 [188] Tosatti E. Lett. Phys. B 60 13885 [185] Kulik I O. Shim M and Dai H 2002 Appl. Hakioglu T and Barone A 2002 Eur. B 104 9063 a [192] Aviran A and Ratner M A 1974 Chem. Hadley P. 29 277 [193] Leonard F and Tersoff J 2000 Phys. Rubio A. Lett. Torres J A and Tosatti E 2000 Surf. Quate C F. Rev. Nakanishi T and Dekker C 2001 Science 297 1317 [200] Modi A. J. Falvo M R. B 60 7808 [190] Tolla F D. 75 627 [198] Fraklin N R. Lett. Phys. Rev. Rev. Chem. B 54 11149 Miyamoto Y. Scherbakov A G and Landman U 2000 J. 81 913 [199] Bactold A. Louie S G and Cohen M L 1996 Phys. Crespi V H. Marcus C M. Phys. Wang Q. Kong J and Dai H 1999 Appl. Lett. B 55 R1981 [184] Miyamoto Y. Resta R and Baroni S 1991 Phys. Rev. Helser A. Tersoff J. Rev. Louie S G and Cohen M 1996 Phys. Lett. Di Vientra M and Avouris P 2001 Appl. Derycke V. Nardelli M B. 84 4693 [194] Avouris Ph 2003 Science 300 783 [195] Durkop T. B 30 219 [186] Friedman J R. Martel R. 4 35 [196] Keren K. Lett. Rev. Dal Corso A and Di Tolla F D 2001 Science 291 288 [189] Okamoto M and Takayanagi K 1999 Phys. Dresselhaus G and Dresselhaus M S 1996 Phys. Buchstab E. Lett 89 106801 [179] Dag S. B 68 115419 u [177] Rochefort A. B 53 2044 [173] Chico L. Superfine R and Washburn S 2000 Science 290 1742 [183] Mehrez H. Chen W. B 44 5572 [176] G¨ lseren O. 454–456 947 [191] H¨ kkinen H. Sci. Rev. G¨ lseren O. Kostlmeier S. Rev. Tolpygo S K and Lukens J E 2000 Nature 406 43 [187] Mehrez H and Ciraci S 1997 Phys. Prestipino S. Taylor R M. 78 2521 [178] Heinze S.

Sign up to vote on this title
UsefulNot useful