In situ-grown carbon nanotube array with excellent field emission characteristics

Article abstract of DOI:10.1063/1.126790

In situ catalytic thermal decomposition method was used for producing aligned multiwalled carbon nanotubes (MWNTs) in bulk quantities on stable and electrically conducting substrates. Very low turn-on electric fields of 0.75 V/μm and low threshold fields of ～1.6 V/μm (for current density of 10 mA/cm²) were obtained from the MWNT arrays grown on TiN substrate. Furthermore, large emission current densities of 1-3 A/cm² were obtained at reasonably low fields of less than ～8 V/μm. These enhanced emission properties are tentatively attributed to the oriented and high-density nature of the emitting carbon nanotube structure and the high-conductivity, stable nature of the TiN substrate onto which the nanotubes are attached.

Full text of DOI:10.1063/1.126790

In situ-grown carbon nanotube array with excellent field emission characteristics
A. M. Rao, D. Jacques, R. C. Haddon, W. Zhu, C. Bower, and S. Jin
Citation: Applied Physics Letters 76, 3813 (2000); doi: 10.1063/1.126790
View online: http://dx.doi.org/10.1063/1.126790
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/76/25?ver=pdfcov
Published by the AIP Publishing
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.113.111.210 On: Mon, 09 Dec 2013 19:11:03

APPLIED PHYSICS LETTERS
VOLUME 76, NUMBER 25
19 JUNE 2000
In situ-grown carbon nanotube array with excellent field emission
characteristics
A. M. Rao,a) D. Jacques, and R. C. Haddon
Center for Applied Energy Research and Departments of Physics and Astronomy and Chemistry,
University of Kentucky, Lexington, Kentucky 40511
W. Zhu, C. Bower, and S. Jin
Bell Laboratories, Lucent Technologies, Murray Hill, New Jersey 07974
͑Received 11 February 2000; accepted for publication 2 May 2000͒
In situ catalytic thermal decomposition method was used for producing aligned multiwalled carbon
nanotubes ͑MWNTs͒ in bulk quantities on stable and electrically conducting substrates. Very low
turn-on electric fields of 0.75 V/␮m and low threshold fields of ϳ1.6 V/␮m ͑for current density of
10 mA/cm2͒ were obtained from the MWNT arrays grown on TiN substrate. Furthermore, large
emission current densities of 1–3 A/cm2 were obtained at reasonably low fields of less than ϳ8
V/␮m. These enhanced emission properties are tentatively attributed to the oriented and
high-density nature of the emitting carbon nanotube structure and the high-conductivity, stable
nature of the TiN substrate onto which the nanotubes are attached. © 2000 American Institute of
Physics. ͓S0003-6951͑00͒05325-0͔
The recent discovery of electron emission from carbon
nanotubes1–9 has generated excitement since carbon nano-
tubes can serve as ideal, tiny electron emitters for developing
efficient and inexpensive field emission devices such as
vacuum microwave power amplifiers and flat panel displays.
Several characteristics of carbon nanotubes suggest that they
are suitable and promising for cold cathode applications
since ͑i͒ they possess an inherently high aspect ratio ͑length/
diameter ratio of Ͼ1000͒ with the resultant, large field am-
plification factor, ͑ii͒ they possess high chemical and me-
chanical stability, and ͑iii͒ they are capable of producing
large current densities at modest, practical electrical field
levels. Nanotube-based emission devices are also capable of
exhibiting bright light emission for over 10 000 h.5
For field emission applications, the carbon nanotubes
must be grown on electrically conductive substrates in order
to be able to serve as a cathode and continuously supply
electrons to the field emitter. The nucleation and growth of
carbon nanotubes is conveniently achieved by a thin film
deposition of a catalyst layer such as Ni, Co, and Fe on
insulating substrates such as silica, alumina, or semiconduct-
ing substrates such as silicon. Typically, the initially continu-
ous catalyst films are broken up into isolated islands whose
average size is comparable to the nanotube diameter for
nucleation of carbon nanotube growth. Thus, on insulating
substrates, the electrical conduction ͑the continuous supply
of electrons to the emitting tips͒ is mostly absent, either
through the substrate thickness or along the substrate surface.
In spite of the encouraging progress in the carbon nano-
tube research, there is still a need for developing a synthesis
method that is inexpensive and readily scalable to produce
high-quality nanotubes for device applications. In particular,
simple synthesis methods for preparing well-oriented nano-
tubes on highly conductive substrates suitable for field emis-
sion applications are lacking. For example, de Heer et al.10
demonstrated field emission from thin films of aligned mul-
tiwalled carbon nanotubes ͑MWNTs͒ that were produced by
drawing nanotube suspension through a ceramic filter ͑0.2
␮m pore size͒. Li et al.11 synthesized oriented MWNTs from
the decomposition of acetylene at 700 °C over iron-
nanoparticle-embedded mesoporous silica which is electri-
cally insulating. High-purity aligned MWNTs were also pro-
duced between laser-etched tracks on cobalt films on
insulating silica substrates from a catalytic decomposition of
triazine at 1000 °C.12 Large arrays of aligned MWNTs were
grown normal to nickel-coated glass substrates by a plasma-
enhanced hot filament chemical vapor deposition of acety-
lene and ammonia gases.13 More recently, from prepatterned
Fe-catalyst islands on porous silicon, Fan and co-workers14
reported the growth of ‘‘MWNT towers’’ that showed good
emission properties.
A common drawback with these methods11–14 is that
they require special processing of substrates or separate, pre-
deposition of a catalyst layer on substrates in order to pro-
duce aligned MWNTs. We have recently reported a simple,
inexpensive in situ method for preparing aligned MWNTs on
electrically insulating substrates such as silica.15 Our synthe-
sis method is based on the catalytic decomposition of
ferrocene–xylene mixture. In this letter, we report an in situ
growth of aligned nanotubes on stable and highly conductive
substrates of TiN, and discuss their excellent field emission
properties.
In the two-stage reactor ͑described previously in Ref.
15͒ the xylene–ferrocene mixture ͑Fe ϳ0.75 at. %͒ is con-
tinuously injected into the preheater ͑maintained at ϳ200 °C͒
using a syringe pump. As the xylene–ferrocene vapors
emerge at the outlet of the preheater a stream of Ar gas ͑with
10% hydrogen͒ sweeps the vapors into the furnace ͑main-
tained at ϳ700 °C͒ where it is decomposed to produce high
purity, aligned MWNT arrays on the inner walls of the fur-
nace and the substrates placed inside the furnace. In a typical
a͒
Corresponding author; electronic mail: rao@pop.uky.edu
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
0003-6951/2000/76(25)/3813/3/$17.00 3813 © 2000 American Institute of Physics
130.113.111.210 On: Mon, 09 Dec 2013 19:11:03

3814
Appl. Phys. Lett., Vol. 76, No. 25, 19 June 2000
Rao et al.
run, the silica tube reactor ͑inner diameter ␾ϳ34 mm͒ with a
foot long preheater and two foot long furnace, are operated
for two hours to produce ϳ0.4 g of high purity MWNTs for
a xylene–ferrocene feed rate ͑R͒ of 1 ml/h. Under identical
operating conditions, the combined wall and substrate nano-
tube material averaged ϳ1.85 g in a similar reactor with
␾ϳ95 mm and Rϳ4.4 ml/h without noticeable degradation
in the quality of the MWNTs. High resolution scanning elec-
tron microscopy ͑SEM͒ images and energy dispersive x-ray
analysis of MWNT arrays produced by the method described
in Ref. 15 suggest that the growth of the MWNT arrays
begin after an initial deposition of an Fe layer on the silica
substrates. Moreover, high resolution transmission electron
microscopy ͑TEM͒ studies indicated good structural integrity
of individual MWNTs with the most dominant outer diam-
eter of ϳ25 nm. Further evidence for structural integrity was
obtained from polarized micro-Raman spectroscopy.16
The field emission measurements were carried out at
room temperature in a vacuum chamber with a 10Ϫ8 Torr
base pressure. The experimental procedures for current den-
sity measurements were described in detail elsewhere.17,18
Briefly, a voltage up to 2 kV was applied to a hemispherical
molybdenum anode probe ͑radius of curvature RϷ250 ␮m͒
which is 10–300 ␮m above the emitter surface. We used a
translation stage in the vacuum chamber to control the dis-
tance between the anode and the cathode. A hemispherical
probe approach is a better measurement technique than a
parallel plate geometry because it avoids uncertainties
caused by field enhancements at the edge of a planar anode
or emitter, and it also allows easy measurements at many
points on the sample. The emission current–voltage (I–V)
characteristics were measured as a function of the anode–
cathode distance (Z). At each distance, the anode voltage
was raised from zero until the current density reached ϳ0.6
A/cm2 and then decreased back to zero. The effective emis-
sion area (A) used for determining the current density was
estimated as described in Ref. 8, and corresponds to an area
within which JуJmax/2, where Jmax is the emission current
density directly beneath the anode where the electric field is
the highest. For ZӶ2R, Aϭ2␲RZ(21/nϪ1), where n
ϭ(V/I)(dI/dV). Typically, nϳ18 for JϽ50 mA/cm2, de-
creasing gradually as the current density increases. The an-
ode was then moved one step ͑3.3 ␮m͒ closer to the emitter
surface, and the cycle was repeated.
FIG. 1. SEM image of aligned MWNTs grown on ͑a͒ TiN substrates, ͑b͒
TaN substrates.
and Y–Ba–Cu–O superconductor material͒, conducting car-
bides ͑such as WC, TiC͒, and conducting nitrides ͑such as
TiN, ZrN, TaN͒. The oxide substrates were seen to be par-
tially reduced and decomposed under our operating condi-
tions, and no noticeable nanotube nucleation occurred. The
carbide substrates produced much smaller number of nano-
tubes per unit substrate area than the nitride substrates. The
best results for aligned nanotube growth were obtained for
the TiN substrates ͑Fig. 1͒. TiN is a relatively stable material
with excellent electrical conductivity ͑resistivity ϳ25
m⍀ cm͒ and high melting point ͑ϳ3200 °C͒, both of which
are desirable for the stability of field emission cathode struc-
ture, especially for high current-density, high-power applica-
tions such as microwave amplifiers.
Shown in Fig. 1͑a͒ is the SEM image ͑taken at an ob-
lique angle, tilted by 30°͒ of the MWNT array grown on a
TiN substrate from a thermal decomposition of xylene–
ferrocene mixture at 700 °C for 15 min. A longer run time
produces a similar microstructure except that the lengths of
the MWNTs are increased. It is seen that a relatively
uniform-sized MWNTs with ϳ40 nm average diameter are
grown in an aligned fashion perpendicular to the substrate
surface. The presence of Fe catalyst particle at the tips of the
nanotubes suggest that the predominant growth occurs via
the tip growth method.19 The density of nanotube tips is
relatively high, and is estimated to be about 30 per (␮m)2
area, equivalent to ϳ3ϫ105 per typical display pixel area of
(100 ␮m)2, or ϳ3ϫ109/cm2. The mechanism of nanotube
alignment in the present work is not clearly understood. It
may be due to the crowding mechanism15 or due to the
nucleation and out-growth of nanotubes from the pores11
present in the TiN substrates. Further study is needed to un-
derstand the underlying mechanism͑s͒ of the observed
aligned growth. Figure 1͑b͒ shows the MWNTs grown on
the TaN substrate using the identical synthesis conditions
noted above. The presence of well-aligned structure is evi-
dent, however, with a coarser structure with an average di-
ameter of ϳ100 nm as compared to the ϳ40 nm diameter in
While our thermal decomposition method15 routinely
produces well aligned MWNTs on silica, interestingly we
found that the nanotube growth ͑aligned or unaligned͒ rarely
occurred on polished metallic substrates, such as Mo, Ni, or
stainless steel. This absence of nanotube growth may be re-
lated to the high mobility and lack of localization of carbon
on the metal surface, or the difficulty of catalyst island for-
mation due to the diffusional reaction and interfacial bonding
between the Fe layer and the metal substrate at the furnace
temperatures. We therefore embarked on a search for alter-
native conducting substrate materials that would allow nucle-
ation and growth of aligned nanotubes.
The substrates utilized in the present investigation were
of bulk ceramic type with the thickness of 1–3 mm. Various
substrates studied here included conducting oxides ͑such as
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
Mn–Zn ferrites, perovskite type La–Sr–Mn–O manganite
the case of the TiN substrates.
130.113.111.210 On: Mon, 09 Dec 2013 19:11:03

Appl. Phys. Lett., Vol. 76, No. 25, 19 June 2000
Rao et al.
3815
for the MWNTs on TiN. The data was obtained under severe
operating conditions, i.e., at high current density levels in
order to provide a measure of stability/lifetime behavior for
the field emission. As can be seen from the figure, the high
emission current density of 1–3 A/cm2 is maintained at the
applied field for ϳ2 h, and in fact improved with time, which
is interesting and deserves further investigation. The current
density is then gradually reduced with time and stabilized
after ϳ24 h at a level of ϳ110 mA/cm2, which is still rela-
tively high compared to 10 mA/cm2. The eventual decrease
of the emission current density seen in Fig. 3 after a few
hours of continuous, high-current field emission may be
caused by a number of different reasons, one possibility be-
ing the damage ͑by excessive resistive heating in the high
current environment͒ in the nanotube structure near the emit-
ting tips or the damage/change of the contact region near the
bond interface of the nanotubes and the TiN substrate. Fur-
ther studies are needed to fully understand the lifetime be-
havior of the carbon nanotube field emitters.
FIG. 2. Electron emission I-V curves for the MWNT array grown on TiN
substrates.
The aligned MWNTs grown on the TiN substrate exhibit
excellent field emission properties. Shown in Fig. 2 is the
measured emission current as a function of applied voltage
for a sample that was prepared at 700 °C in a 2 h run. Rela-
tively smooth and consistent current–voltage (I–V) curves
were obtained. Replotting of the data as log(I/V2) vs 1/V
indicates the Fowler–Nordheim type field emission
behavior.20 The turn-on field, as defined for an emission cur-
rent of 1 nA, was less than 0.75 V/␮m, which is one of the
lowest turn-on field values that we have observed for several
different types of SWNTs and MWNTs evaluated using the
same experimental method. The threshold field ͑defined here
as the field required to produce technologically useful cur-
rent density of 10 mA/cm2͒ was also low, being ϳ1.6 V/␮m.
The aligned nanotubes on TiN also exhibit very high field
emission currents at relatively low applied fields; the field
required to generate an order of magnitude higher emission
current density of 100 mA/cm2 was ϳ3.1 V/␮m, and that for
producing 1 A/cm2 was ϳ7.8 V/␮m. These high emission
current densities and low threshold fields for the present
MWNTs are comparable to or better than those for the
SWNTs reported recently,8 indicating that the MWNTs can
be as efficient as the SWNTs in field emission.
This research was supported in part by NSF MRSEC
Grant No. DMR-9809686 and by the University of Kentucky
Center for Applied Energy Research.
1 A. G. Rinzler, J. H. Hafner, P. Nikolaev, L. Lou, S. G. Kim, D. Tomanek,
P. Nordlander, D. T. Colbert, and R. E. Smalley, Science 269, 1550
͑1995͒.
2 W. A. de Heer, A. Chatelain, and D. Ugarte, Science 270, 1179 ͑1995͒.
3 Q. H. Wang, T. D. Corrigan, J. Y. Dai, R. P. H. Chang, and A. R. Krauss,
Appl. Phys. Lett. 70, 3308 ͑1997͒.
4 P. G. Collins and A. Zettl, Phys. Rev. B 55, 9391 ͑1997͒.
5 Y. Saito, S. Uemura, and K. Hamaguchi, Jpn. J. Appl. Phys., Part 2 37,
L346 ͑1998͒.
6 Q. H. Wang, A. A. Setlur, J. M. Lauerhaas, J. Y. Dai, E. W. Seelig, and R.
P. H. Chang, Appl. Phys. Lett. 72, 2912 ͑1998͒.
7 J. M. Bonard, J. P. Salvetat, T. Stockli, W. A. de Heer, L. Forro, and A.
Chatelain, Appl. Phys. Lett. 73, 918 ͑1998͒.
8 W. Zhu, C. Bower, O. Zhou, G. P. Kochanski, and S. Jin, Appl. Phys.
Lett. 75, 873 ͑1999͒; G. Kochanski, Lucent Technologies ͑private com-
munication͒.
9 W. B. Choi, D. S. Chung, J. H. Kang, H. Y. Kim, Y. W. Jin, I. T. Han, Y.
H. Lee, J. E. Jung, N. S. Lee, G. S. Park, and J. M. Kim, Appl. Phys. Lett.
75, 3129 ͑1999͒.
10 W. A. de Heer, W. S. Bacsa, A. Chatelain, T. Gerfin, R. Humphrey-Baker,
L. Forro, and D. Ugarte, Science 268, 845 ͑1995͒.
11 W. Z. Li, S. S. Xie, L. X. Qian, B. H. Chang, B. S. Zou, W. Y. Zhou, R.
A. Zhao, and G. Wang, Science 274, 1701 ͑1996͒.
Shown in Fig. 3 is the emission current density vs time
at constant applied field of 6.3 V/␮m ͑the applied voltage
was 1000 V and the cathode–anode distance was 160 ␮m͒
12 M. Terrones, N. Grobert, J. Olivares, J. P. Zhang, H. Terrones, K. Korda-
tos, W. K. Hsu, J. P. Hare, P. D. Townsend, K. Prassides, A. K.
Cheetham, H. W. Kroto, and D. R. M. Walton, Nature ͑London͒ 388, 52
͑1997͒.
13 Z. F. Ren, Z. P. Huang, J. W. Xu, J. H. Wang, P. Bush, M. P. Siegal, and
P. N. Provencio, Science 282, 1105 ͑1998͒.
14 S. Fan, M. G. Chapline, N. R. Franklin, T. W. Tombler, A. M. Cassell,
and H. Dai, Science 283, 512 ͑1999͒.
15 R. Andrews, D. Jacques, A. M. Rao, F. Derbyshire, D. Qian, X. Fan, E. C.
Dickey, and J. Chen, Chem. Phys. Lett. 303, 467 ͑1999͒.
16 A. M. Rao, S. Jorio, M. Pimenta, D. D. S. Dantas, R. Saito, G. Dressel-
haus, and M. S. Dresselhaus, Phys. Rev. Lett. 84, 1820 ͑2000͒.
17 W. Zhu, G. P. Kochanski, and S. Jin, J. Vac. Sci. Technol. B 14, 2011
͑1996͒.
18 W. Zhu, G. P. Kochanski, and S. Jin, Science 282, 1471 ͑1998͒.
19 S. Amelinckx, X. B. Zhang, D. Bernaerts, X. F. Zhang, V. Ivanov, and J.
B. Nagy, Science 265, 635 ͑1994͒.
20 R. H. Fowler and L. W. Nordheim, Proc. R. Soc. London, Ser. A 119, 173
͑1928͒.
FIG. 3. Emission current density vs time at a constant applied field.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.113.111.210 On: Mon, 09 Dec 2013 19:11:03