Freestanding vertically aligned arrays of individual carbon nanotubes on metallic substrates for field emission cathodes

Article abstract of DOI:10.1063/1.1773366

The direct growth of individual and vertically aligned carbon nanotubes (CNT) onto a metallic tip apex using a two-chamber radio-frequency plasma-enhanced chemical vapor deposition was discussed. The arrays of CNTs, having a low-density spatial distribution to avoid mutual electrostatic field screening, were also obtained. It was shown that the individual Ni nanocatalysts, obtained by a sol-gel combustion technique, were dots for the nucleation of individual CNTs that were freestanding, clean and vertically aligned by the presence of a controlled applied field. It was demonstrated that the Ni nanoparticles were mechanically dispersed onto the metallic support surface at desired density.

Full text of DOI:10.1063/1.1773366

Freestanding vertically aligned arrays of individual carbon nanotubes on
metallic substrates for field emission cathodes
M. Mauger, Vu Thien Binh, A. Levesque, and D. Guillot
Citation: Appl. Phys. Lett. 85, 305 (2004); doi: 10.1063/1.1773366
View online: http://dx.doi.org/10.1063/1.1773366
View Table of Contents: http://apl.aip.org/resource/1/APPLAB/v85/i2
Published by the AIP Publishing LLC.
Additional information on Appl. Phys. Lett.
Journal Homepage: http://apl.aip.org/
Journal Information: http://apl.aip.org/about/about_the_journal
Top downloads: http://apl.aip.org/features/most_downloaded
Information for Authors: http://apl.aip.org/authors
Downloaded 17 Aug 2013 to 160.36.192.221. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions

APPLIED PHYSICS LETTERS
VOLUME 85, NUMBER 2
12 JULY 2004
Freestanding vertically aligned arrays of individual carbon nanotubes
on metallic substrates for field emission cathodes
M. Mauger, Vu Thien Binh,a) A. Levesque, and D. Guillot
Equipe Emission Electronique, Laboratoíre de Physique de la Matiere Condensée et Nanostructures–CNRS,
University Claude Bernard Lyon 1, 69622 Villeurbanne, France
(Received 22 December 2003; accepted 19 May 2004)
Direct growth of individual and vertically aligned carbon nanotubes (CNTs) onto a metallic tip apex
using a two-chamber radio-frequency plasma-enhanced chemical vapor deposition is reported.
Individual Ni nanocatalysts, obtained by a sol–gel combustion technique, were dots for the
nucleation of individual CNTs that were freestanding, clean, and vertically aligned by the presence
of a controlled applied field. The arrays of CNTs obtained, having a low-density spatial distribution
to avoid mutual electrostatic field screening, gave uniform stable overall field emission patterns after
a conditioning process. Effective total current densities up to 1 A/cm2 can be extracted. © 2004
American Institute of Physics. [DOI: 10.1063/1.1773366]
Favorable electron emission properties from carbon
compounds have been reported.1–9 Even more interesting,
from a field emission (FE) viewpoint, is the carbon nanotube
(CNT) due to its high aspect ratio and surface stability.10,11
They can be directly grown on a flat electrode/surface, thus
removing the need of a three-dimensional lithographic fabri-
cation of sharp tips for FE. Vertically aligned CNTs tend to
possess better FE properties than nonaligned CNTs,12 and
this was attributed to the potentially large number of exposed
tip ends of the aligned CNTs in arrays, which are sites that
present high electric field enhancement,13 i.e., FE areas. In
the literature today, aligned CNTs have been deposited from
various plasma-enhanced chemical vapor deposition
(PECVD) techniques.14–18 More recently, it has been shown
that the nucleation of single isolated vertical CNTs was ob-
tained when the size of the catalyst is under 100 nm, with
using radio-frequency plasma-enhanced chemical vapor
deposition (rf PECVD). We also show that, after a condition-
ing process, these arrays of isolated CNTs become stable and
uniform emitters in vacuum of 10−6–10−8 Torr.
Our CVD growth was performed in a homemade two-
chamber reactor [Fig. 1(a)]. The concept of a two-chamber
reactor allowed splitting between the plasma production
from the applied local field on the metallic growth support.
By monitoring the values of the polarization potentials V1
and V2 applied to electrodes E1 and S, we have direct control
of not only the flux and the energy of the incident ions from
the plasma, but also of the local electric field at the cathode
substrate during CNT growth. The metallic supports we used
here were Ta tips with apex radii of about 400 ␮m.
Our catalyst nanodots were Ni nanoparticles synthesized
by a sol–gel combustion process.21 We mixed an aqueous
solution containing nickel acetate tetrahydrate and methyl
hydrazine with 2-propanol solution at room temperature,
with constant stirring for 1 day. This solution of metal ac-
etate was then dried at ϳ50 °C in air atmosphere until com-
plete evaporation of the solvent, and then it was heated at
400 °C for 1 h under nitrogen atmosphere. Finally, by me-
chanical filtration, the powder obtained was calibrated in or-
der to have Ni nanoparticles with diameters in the range of
100 nm.22
practically 100% yield using
a
glow discharge
PECVD.17,19,20 The vertical direction of the CNTs was re-
lated to the presence of the electric field intrinsic in this
process. The most important observation and result, for the
use of CNTs as cold cathodes, were that these isolated CNTs
in the array structures exhibited remarkable uniformity in
terms of diameter and height.20 They were determined, re-
spectively, by the size of the catalyst dots and by the growth
conditions (gas composition, flux and total pressure P, tem-
perature of the substrate T and growth time).
The main characteristics of growth and the results are
the following.
However, in most of these studies electron lithography
microfabrication processes prepared catalyst nanodots hav-
ing diameters less than 100 nm. For possible extension of
these vertically aligned individual CNT array cathodes to
large production applications with high throughput fabrica-
tion process, we have to tackle the two following problems.
(1) Is there a lower-cost, nonlithographic fabrication process
available for direct growth of individual vertically aligned
CNTs on metallic surfaces, and (2) are their FE properties
compatible with the industrial environment, as in field emis-
sion displays, for example?
(1) The Ni nanoparticles were mechanically dispersed onto
the metallic support surface at the desired density. Fig-
ure 1(b) indicates that dispersion of individual nanosize
Ni clusters can be obtained with this lithographic-free
process. This deposit was cleaned in the growth chamber
by sputtering with hydrogen ions for ϳ5 min at T be-
tween 700 and 800 °C, prior to the growth process. The
parameters V1, V2, T and P defined the height of the
CNTs for a given growth time. As an example, we have
obtained CNTs with height between 2 and 10 ␮m in
20 min for V1=50 V and 100 VഛV2ഛ300 V, under
typical conditions of C2H2: NH3 flow of 2:3 sccm and
P=0.1 Torr. Figure 1(c) reveals that the nucleation area
of the CNTs was delineated by the presence of the Ni
We report in this letter the direct growth of arrays of
isolated and vertically aligned CNTs on a metallic surface
a)Author to whom correspondence should be addressed; electronic mail:
vuthien@lpmcn.univ-lyon1.fr
0003-6951/2004/85(2)/305/3/$22.00
305
© 2004 American Institute of Physics
Downloaded 17 Aug 2013 to 160.36.192.221. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions

306
Appl. Phys. Lett., Vol. 85, No. 2, 12 July 2004
Mauger et al.
FIG. 1. (a) Schematic drawing of the two-chamber rf PECVD reactor. A resistive coil heater obtained the temperature T of the sample, measured by a
thermocouple and a micropyrometer. (b) Dispersion on the surface of the Ni nanoclusters (diameter ϳ100 nm) obtained by a sol–gel combustion process. In
this example the distances between the nanoclusters were in the range of 1–2 ␮m. (c) Overall and (d) detailed scanning electron microscopy photos of the
direct growth of freestanding, isolated vertically aligned CNTs on the apex of a metallic support tip. Confinement of the CNTs on the apex of the Ta tip was
the consequence of voluntary localized deposition of the Ni nanocatalysts only onto this area. (e) Characteristic histogram of the CNT height distribution over
the metallic tip apex showing that more than 65% of the CNTs have heights between 3.5 and 4.5 ␮m.
clusters, and Figs. 1(d) and 2 show scanning electron
In the last part of this letter, we report about the FE
microscopy images characteristic of the CNTs obtained.
efficiency of these cathodes. In our FE experimental proce-
dure, the metallic support tip covered with the CNT array
was in front of a fluorescent screen/metallic anode and all the
I–V measurements were done within this diode setup, where
I and V were, respectively, the FE total current and voltage.
Since the emission area was confined to the end of the sup-
(2) The majority of CNTs were freestanding, isolated and
vertically aligned on the metallic surface [Figs. 1(c) and
1(d)]. In most cases, with a catalysts size of ϳ100 nm,
there was only nucleation of a single CNT per dot, in
agreement with former analysis.20 Note that, as observed
in Ref. 20, the ion sputtering during growth resulted in
port tip covered with the CNT arrays, it was estimated to be
the nucleation of “clean” CNTs.
a zone of about 100 ␮m diameter or less [Fig. 1(c)]. The FE
(3) At some catalyst sites, two or more CNTs can nucleate
behavior of the arrayed CNTs, after a conditioning process
[Figs. 2(b) and 2(c)]. This situation happens when the
described later, shown Fowler–Nordheim behavior, i.e., lin-
size of the catalyst is greater than 100 nm,20 or when an
aggregation of several Ni nanoclusters happened. How-
ear variation of ln͑I/V2͒ vs ͑1/V͒. Figure 3 is a characteristic
example of the I–V evolution of these cathodes. The FE
ever, within such situations, only one of the CNTs had a
stability was good even for a working pressure of 10−7 Torr
dominant height and it will be then the field emitting
CNT.
(inset in Fig. 3). We obtained reproducible and stable total
currents up to ϳ20 ␮A from these cathodes, which means
(4) The statistical spacing between the different nanotubes
across the arrayed distribution presented good spatial
distribution, mainly greater than the height of the CNTs
[Fig. 2(a)]. Note that the optimum FE conditions were
reached when the emitters were spaced about twice their
height apart to avoid electrostatic field screening.23
(5) The histogram of the heights of the CNTs within an
array showed tight dispersion around a mean value. Fig-
ure 1(e) is a characteristic distribution for a given
growth. Note that to achieve good emission uniformity
the structural variation between emitters must be as
small as possible.
effective current densities from 0.3 to 1 A/cm2. This behav-
ior was perfectly coherent with the FE behavior of individual
CNTs that has been analyzed in Ref. 10, showing that one
CNT acted as a high aspect ratio metallic tubular tip with a
work function in the range of 4 eV.
However, in the case of an array having the height dis-
tribution given in Fig. 1(e), the following conditioning pro-
cess is mandatory to end up with a uniform stable overall FE
current. The conditioning process we propose is a two-step
procedure. In the first step, the heights of the CNTs of the
emitting array were first standardized. Then, in the second
step, the apexes of these field emitters were cleaned in situ
by using field desorption and Nottingham effects.
FIG. 3. I–V characteristics of the arrayed CNTs followed Fowler–Nordheim
behavior. The inset shows the stability of the FE current for more than 6 h of
continuous emission at 1 ␮A in 10−7 Torr. Note that the likely cause of the
fluctuation of the stability curve shown in the inset was retro-ion bombard-
ment due to the bad vacuum conditions ͑10−7 Torr͒.
FIG. 2. Scanning electron microscopy observations of freestanding CNTs on
a metallic cathode support. (a) One CNT per Ni catalyst dot (majority of the
CNTs obtained). (b) Two CNTs per Ni dot. (c) Multiple CNTs per Ni dot.
Downloaded 17 Aug 2013 to 160.36.192.221. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions

Appl. Phys. Lett., Vol. 85, No. 2, 12 July 2004
Mauger et al.
307
tallic surfaces of freestanding, individual, vertically aligned
CNTs allowed direct use of these arrays as cold cathodes,
after a two-step conditioning process that assumed an opti-
mum setting for a uniform and stable overall FE of the ar-
rays. The results establish a scalable method of field emitter
fabrication, which can address the present uniformity prob-
lem of FE from CNT mats and is well suited for field emis-
sion display production.
One of the authors (M.M.) acknowledges a CIFRE
graduate grant from INANOV and ANRT.
1W. Zhu, G. P. Kochanski, S. Jin, and L. Seibles, J. Appl. Phys. 78, 2707
(1995).
FIG. 4. Numerical simulation showing the contribution of CNTs to the
overall FE current during the conditioning process. The two insets are FE
patterns before (curve 1) and after conditioning (curve 3) coming from a
CNT array area about 100 ␮m in diameter. The ring on the FE pattern in
inset 3 was an artifact due to the reflection of light on the glass support of
the fluorescent screen.
2B. S. Satyanarayana, A. Hart, W. I. Milne, and J. Robertson, Appl. Phys.
Lett. 71, 1430 (1997).
3G. A. J. Amaratunga and S. R. P. Silva, Appl. Phys. Lett. 68, 2529 (1996).
4A. N. Obraztsov, I. Yu. Pavlovsky, and A. P. Volkov, J. Vac. Sci. Technol.
B 17, 654 (1999).
5B. F. Coll, J. E. Jaskie, J. L. Markham, E. P. Menu, A. A. Talin, and P. von
Allmen, Mater. Res. Soc. Symp. Proc. 498, 185 (1998).
6B. S. Satyanarayana, J. Robertson, and W. I. Milne, J. Appl. Phys. 87,
3126 (2000).
From the as-growth height distribution, only the few
highest CNTs of the array will contribute to the FE current
(curve 1 in Fig. 4), the FE patterns were then nonuniform
and spotty (inset 1 in Fig. 4). To standardize the height of the
CNTs we increased the voltage applied to induce selective
field evaporation among the CNTs. The field evaporation10
will act preferentially on the highest CNTs, leading to short-
ening (or eventual destruction) of them. As a consequence,
there is a concomitant increase in the uniformity of the FE
patterns due to the increase in the number of contributing
CNTs for FE. The optimum was reached when the majority
of the CNTs of the array having the same height contribute to
the formation of the FE pattern (inset and curve 3 in Fig. 4).
A numerical simulation taking into account such evolution
due to the conditioning process confirmed that a greater
number of CNTs would then contribute to the FE pattern, as
shown by the evolution from curve 1 to curve 3 in Fig. 4. As
the height of the emitting CNTs decreased by field evapora-
tion during the conditioning process, the FE voltages for the
same overall current must increase in relation with the de-
crease of the field enhancement factor.13 In the example in
Fig. 4, if V0 is the FE voltage for curve 1, the FE voltages for
curves 2 and 3 to obtain the same overall current are, respec-
tively, 1.06 and 1.20V0. These values are comparable to our
experimental data for a 2 ␮A total current a voltage of
2200 V was needed before conditioning and 2650 then
2800 V after mild and severe conditioning processes.
7G. A.J. Amaratunga, M. Baxendale, N. Rupesinghe, I. Alexandrou, M.
Chhowalla, T. Butler, A. Munindradasa, C. I. Kiley, L. Zhang, and T.
Sakai, New Diamond Front. Carbon Technol. 9, 31 (1999).
8I. Musa, D. A. Munindrasdasa, G. A. Amaratunga, and W. Eccleston,
Nature (London) 395, 362 (1998).
9J. M. Bonard, J. P. Salvetat, T. Stockli, L. Forro, and A. Chatelain, Appl.
Phys. A: Mater. Sci. Process. 69, 245 (1999).
10V. Semet, V. Thien Binh, P. Vincent, D. Guillot, K. B. K. Teo, M. Chhow-
alla, G. A. J. Amaratunga, W. I. Milne, P. Legagneux, and D. Pribat, Appl.
Phys. Lett. 81, 343 (2002).
11S. T. Purcell, P. Vincent, C. Journet, and V. Thien Binh, Phys. Rev. Lett.
88, 105502 (2002).
12M. Sveningsson, R.-E. Morjan, O. A. Nerushev, Y. Sato, J. Bäckström, E.
E. B. Campbell, and F. Rohmund, Appl. Phys. A: Mater. Sci. Process. 73,
409 (2001).
13R. Forbes, C. J. Edgecombe, and U. Valdre, Ultramicroscopy 95, 57
(2003).
14Z. F. Ren, Z. P. Huang, J. W. Xu, J. H. Wang, P. Bush, M. P. Siegal, and P.
N. Provencio, Science 282, 1105 (1998).
15C. Bower, W. Zhu, S. Jin, and O. Zhou, Appl. Phys. Lett. 77, 830 (2000).
16V. L. Merkulov, D. H. Lowndes, Y. Y. Wei, G. Eres, and E. Voelkl, Appl.
Phys. Lett. 76, 3555 (2000).
17M. Chhowalla, K. B. K. Teo, C. Ducati, N. L. Rupesinghe, G. A. J.
Amaratunga, A. C. Ferrari, D. Roy, J. Robertson, and W. I. Milne, J. Appl.
Phys. 90, 5308 (2001).
18L. Delzeit, I. McAninch, B. A. Cruden, D. Hash, B. Chen, J. Han, and M.
Meyyappan, J. Appl. Phys. 91, 6027 (2002).
19Z. F. Ren, Z. P. Huang, D. Z. Wang, J. G. Wen, J. W. Xu, J. H. Wang, L.
E. Calvet, J. Chen, J. F. Klemic, and M. A. Reed, Appl. Phys. Lett. 75,
1086 (1999).
20K. B. K. Teo, S.-B. Lee, M. Chhowalla, V. Semet, V. Thien Binh, O.
Groening, M. Castignolles, A. Loiseau, G. Pirio, P. Legagneux, D. Pribat,
D. G. Hasko, H. Ahmed, G. A. J. Amaratung, and W. I. Milne,
Nanotechnology 14, 204 (2003).
In a following second step, the stability of the overall FE
current was obtained by field desorbing the adsorbed species
at the apexes of the CNTs with the help of the Nottingham
effect similar to the cleaning procedure analysis described in
detail in Ref. 10. Note that this last step can be done either at
room temperature or under heat treatment.
In conclusion, we have presented a fabrication process
of arrays of individual CNTs using a cost-effective,
lithographic-free, chemical technology. Direct growth on me-
21Chn Syukri, T. Ban, Y. Ohya, and Y. Takahashi, Mater. Chem. Phys. 78,
645 (2003).
22V. Thien Binh and A. Levesque, French Patent application No. 03,80,973
(filed July 2003).
23L. Nilsson, O. Groening, Ch. Emmenegger, O. Kuettel, E. Schaller, L.
Schlapbach, H. Kind, J. M. Bonard, and K. Kern, Appl. Phys. Lett. 76,
2071 (2000).
Downloaded 17 Aug 2013 to 160.36.192.221. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions