Electron field emission from composite electrodes of carbon nanotubes-boron-doped diamond and carbon felts

Article abstract of DOI:10.1063/1.2178247

The electron field emission of carbon nanotube (CNT)/boron-doped diamond (BDD)/carbon felt electrodes (CNT/BDD/felt) have been investigated. The composite electrode was initially prepared with the growth of BDD on carbon felt and the subsequent growth of

Full text of DOI:10.1063/1.2178247

Electron field emission from composite electrodes of carbon nanotubes-boron-doped
diamond and carbon felts
J. Mauricio Rosolen, Simone Tronto, Marcel S. Marchesin, Erica C. Almeida, Neidenei G. Ferreira, C. H. Patrick
Poá, and S. Ravi P. Silva
Citation: Applied Physics Letters 88, 083116 (2006); doi: 10.1063/1.2178247
View online: http://dx.doi.org/10.1063/1.2178247
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/88/8?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Enhanced field emission from cerium hexaboride coated multiwalled carbon nanotube composite films: A
potential material for next generation electron sources
J. Appl. Phys. 115, 094302 (2014); 10.1063/1.4866990
Field-emission properties of carbon nanotube composite in side-electron emission configuration
J. Appl. Phys. 109, 074307 (2011); 10.1063/1.3567913
Composite electrode of carbon nanotubes and vitreous carbon for electron field emission
J. Appl. Phys. 104, 054303 (2008); 10.1063/1.2974090
Field emission from multiwall carbon nanotubes on paper substrates
Appl. Phys. Lett. 90, 173124 (2007); 10.1063/1.2734379
Field Emission from Multiwalled Carbon Nanotubes
AIP Conf. Proc. 633, 548 (2002); 10.1063/1.1514180
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:
128.252.67.66 On: Tue, 23 Dec 2014 14:52:50

APPLIED PHYSICS LETTERS 88, 083116 ͑2006͒
Electron field emission from composite electrodes of carbon
nanotubes-boron-doped diamond and carbon felts
a͒
b͒
J. Mauricio Rosolen, Simone Tronto, Marcel S. Marchesin, Erica C. Almeida,
b͒
c͒
d͒
Neidenei G. Ferreira, C. H. Patrick Poá, and S. Ravi P. Silva
Nano-Electronics Centre, ATI, University of Surrey, Guildford GU2 7XH,
United Kingdom and Dep. de Química, FFCLRP, Universidade de São Paulo, SP-14040-930 Brazil
͑
Received 31 October 2005; accepted 27 January 2006; published online 23 February 2006͒
The electron field emission of carbon nanotube ͑CNT͒/boron-doped diamond ͑BDD͒/carbon felt
electrodes ͑CNT/BDD/felt͒ have been investigated. The composite electrode was initially prepared
with the growth of BDD on carbon felt and the subsequent growth of CNT by chemical
decomposition of methanol. The composite electrodes were characterised using scanning electron
microscopy and transmission electron microscopy. For the CNT/BDD/felt samples, the electron field
−
1
emission was observed at macroscopic fields as low as 1.1 V ␮m . The emission current versus
time plot shows significant potential for future field emission applications. © 2006 American
Institute of Physics. ͓DOI: 10.1063/1.2178247͔
Diamond thin films are a class of carbon materials that
mounts on a carbon felt have previously been explored for
1
0,11
have been used for electron field emission devices since
electrochemical applications.
CNTs grown on BDD
1
1
991. Its electron emission properties depend on several
films, as described in this letter, is a further interesting tech-
nique to increase its potential uses.
2
factors, such as the work function, grain boundaries, and
3
–7
small protuberances. For example, it has been shown that
The felt substrate was obtained from carbonization of
polyacrinitrile at 2000 °C. The obtained carbon fibers have a
diameter of around 10 ␮m and resistivity equal to
2
the boundaries between grains, sites rich in sp hybridiza-
tion, emit electrons more easily than the diamond facets of
7
−1
boron-doped diamond ͑BDD͒ polycrystalline films. Interest-
0.96 ⍀ cm ͑determined by the four-probe method͒. BDD
2
1
−3
ingly, despite boron being an acceptor state in diamond it
appears to enhance the electron emission properties of the
material. We believe this is based on the substrate playing an
important role in the emission process of diamond films. In
fact, it has been observed that better emission results from
films with acceptor concentrations of about 1.5ϫ10 cm
were then grown by the hot filament chemical vapor deposi-
9
,10
tion ͑HFCVD͒ method as described in literature.
The Ra-
man spectra of polycrystalline BDD films studied in the
present letter are identical to those found in Ref. 10, which
3
chemical vapor deposited diamond films grown on SnO in-
shows diamond grains with only sp hybridization. The
2
stead of silicon, suggesting that the electron supply layer or
substrates plays an active role in the emission process.
CNTs were grown on each BDD/carbon fiber using methanol
vapor. The catalysts were a mixture of cobalt and manga-
nese. These samples were dispersed on BDD/carbon fiber by
immersion of the BDD/felt substrate in an ethanol solution
8
However, despite the well-known attractive properties of dia-
mond ͑negative electron affinity, thermal, mechanical, and
chemical inertness͒, there are still several problems in the
production of field emission devices based on this material
associated with obtaining high current densities from emit-
ters, obtaining uniform microstructures, etc., which affect
9
long-term stability. In addition, diamond films, as well as
other carbon materials which have been used for emission,
are subject to thermal effects at the interface between the
substrate and the carbon film. The presence of heating at the
interface of the film and substrate generates different me-
chanical stresses on the film. This will increase the resistance
and/or the voltage barrier needed to maintain electron emis-
sion at low voltages for an extended periods of time.
In order to circumvent the above problem, we present
here an approach based on a combination of carbon nano-
tubes ͑CNTs͒ and BDD films mounted on a carbon felt to
obtain reproducible electron emission at low fields. BDD
a͒Author to whom correspondence should be addressed; electronic mail:
rosolen@ffclrp.usp.br
b͒
Also at: Instituto Nacional de Pesquisas Espaciais, INPE, São Jose dos
Campos-12245-970, Brazil.
Present address: Institute of Materials Research and Engineering, 3 Re-
c͒
search Link, Singapore 117602.
Electronic mail: s.silva@surrey.ac.uk
FIG. 1. Typical SEM micrographs of the composite CNT/BDD/felt elec-
trode and single diamond/fiber and single CNT/BDD/fiber.
d͒
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:
003-6951/2006/88͑8͒/083116/3/$23.00 88, 083116-1 © 2006 American Institute of Physics
28.252.67.66 On: Tue, 23 Dec 2014 14:52:50
0
1

083116-2
Rosolen et al.
Appl. Phys. Lett. 88, 083116 ͑2006͒
FIG. 3. Emission current vs the applied voltage of the CNT/BDD/felt for
several gap distances at 100 ␮m, 200 ␮m, 300 ␮m, 400 ␮m, 500 ␮m, and
emission curve of BDD/felt at 500 ␮m gap ͑dashed line͒. The curves were
collected beginning at 100 ␮m and the distance was increased.
dimension of the catalytic particles at the boundary of dia-
mond grains and face were different. Thus, the CNT on BDD
films may be a mixture of bamboolike, cup stacked, or sim-
ply disordered multiwalled CNTs.
Figure 2 shows the variation of current versus voltage
͑
I-V͒ curves for the compound fibers of electrodes in contact
with a tungsten probe inside an SEM microscope. In this
setup, the felt was grounded using a silver epoxy and the
current measured was not associated with a single fiber. Be-
sides this, the curves have the same profile and all electrodes
show high electronic conductivity. This allows one to infer
that adherence of the BDD to the felt is very good. It also
points to a well-adhered CNT deposit on BDD films, i.e., the
resistance at the several interfaces of our compound is very
low. As a consequence, the Joule heating at these interfaces
is expected to be low.
FIG. 2. Compound fibers of electrode in contact with a tungsten probe
inside a SEM microscope ͑A͒ and its variation of current versus voltage ͑B͒.
containing 5 wt. % of cobalt and manganese acetate and then
were left at 80 °C under ambient conditions. The tempera-
ture and time of CNT growth was 650 °C and 20 min, re-
Typical field emission I-V curves for the BBD/felt and
CNT/BDD/felt electrodes are shown in Fig. 3. All felt elec-
trode I-V curves were collected using a stainless-steel sphere
anode ͑5 mm diameter͒, and the composite carbon and felt
spectively. N was used as the carrier gas for methanol vapor
2
−
1
2
at a constant flow rate of 0.2 ml min .
electrode samples were of dimensions 0.8ϫ0.8 cm with a
Figure 1 shows scanning electron microscopy ͑SEM͒
images from the CNT/BDD/felt electrodes, the polycrystal-
line BDD film, and the CNT grown on BDD film on a single
fiber. The BDD film covered the whole fiber in the felt and
the CNT was then grown over the surface of the BDD film.
The micrographs show that the CNT growth occurs at the
boundary between the grains as well as on the face of each
BDD crystal. This suggests that the BDD films ͑or substrate͒
should not have a major influence on the type of
CNT growth resulting from the catalytic decomposition of
methanol.
thickness of 0.5 cm. The vacuum chamber was evacuated to
about 2ϫ10− Torr prior to electron emission testing, and
the electrical contact of the CNT-felt was conducted using a
thin layer of silver epoxy. The gap between the sphere and
CNT-felt was control by an automated translator with a step-
per motor. The BDD/felt is capable of emitting 1 nA at a
6
−
1
macroscopic field of 2.2 V ␮m , while the CNT/BDD/felt
emits electrons at 1.1 V ␮m− at the same separation dis-
tance from the anode. The turn-on fields for current at 10 ␮A
1
−
1
are typically around 1.5–3.4 V ␮m . However, it is worth
noting that the effective geometric area used in the emission
of the felt is very complex and difficult to determine for each
fiber taking part in the electron emission process.
Indeed, the catalytic particles were observed at the tip of
the tubes. If we consider previous studies on the growth of
CNT on zeolite and NaCl, with different catalyst using simi-
lar CNT growth conditions as those employed in this study,
the CNTs on BDD films should produce a structure similar to
bamboo or cup stacked in nature. In our studies, we found
multiwalled nanotubes which have a correlation between the
Figure 3 also shows a small hysteresis effect on the
I-V curves for the composite electrodes, i.e., the difference in
the applied fields between the upward and downward voltage
cycles appear mainly at an I-V measurement distance of
500 ␮m. The results suggest that this hysteresis is associated
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:
structure of the CNT and catalyst particle dimension. The
with geometric factors since it was not observed at lower gap
128.252.67.66 On: Tue, 23 Dec 2014 14:52:50

083116-3
Rosolen et al.
Appl. Phys. Lett. 88, 083116 ͑2006͒
−
1
room atmosphere. For a field of 4 V ␮m , the electrode
emitted about 2 ␮A with good stability. The absence of fluc-
tuations in Fig. 4͑b͒ suggests that this observed variation is
associated with some kind of degradation of the CNTs and/or
formation of amorphous carbon.
In conclusion, this letter shows the CNT/BDD/felt elec-
trode to be an excellent field emission electrode. This com-
posite electrode has good mechanical properties and can be
subject to chemical treatments without significant lost of
CNTs. The CNT/BDD/felt is able to emit electrons at volt-
ages in the range of 100–1500 microns at threshold fields
that appear smaller than the most of flat multiwall or single-
wall CNT electrodes prepared at the present time. This ex-
cellent performance is associated with several factors. The
first is due the capacity of the felt substrate to mount large
amounts of BDD films and CNTs on the emission area. An-
other important factor is the low resistance of the electrode.
Poor electrical contact resulting from low adherence of CNTs
or diamond to substrates generally results in a Joule effect
that will provoke degradation or failure of emission. In this
new approach, the resistance between the CNT-carbon is
smaller than CNT-Si or CNT-metal. Third, is the combina-
tion of materials that are capable of emitting together and/or
to act as a channel for emission of electrons, i.e., BDD film
for carbon fiber and CNT for the BDD film. We believe that
the carbon compound presented in this letter opens the door
for the development of alternative sources for generation of
emission devices.
FIG. 4. Variation of emission as function of time current of CNT/BDD/felt
electrode: ͑a͒ As obtained at 2.1 V/␮m and ͑b͒ at 4 V/␮m after ͑a͒ run. The
voltage was applied without previous voltage scan on electrode.
distances. The presence of amorphous carbon or impurities
are often said to be responsible for this hysteresis. Both I-V
curves increase monotonically with voltage, with the satura-
tion occurring first for the BDD/felt electrode. For the CNT/
BBD/felt, the saturation seems to occur at higher fields than
the BDD/felt. This result suggests that depending on the ap-
plied field, the emission of the CNT/BBD/felt may be selec-
tively chosen. At shorter gap separation distances, the emis-
sion should be dominated by all forms of carbons of the
composite electrode, while at larger separations the emission
of the electrode should be predominantly influenced by the
CNTs due to the higher-field enhancement factors of these
structures. The CNT/BBD/felt appears thus very efficient for
emission. This combination was able to provide a variation
in the I-V curves that is similar to that observed for single-
wall CNTs, where the I-V curves are found to have a sharp
Some authors ͑J.M.R, S.T., and M.M.͒ thank CAPES
BEX 0278/04-1͒, CNPq, and FAPESP. One of the authors
J.M.R.͒ thanks the University of Surrey for hospitality dur-
ing his sabbatical leaving. The University of Surrey ac-
knowledges the support from the EPSRC Portfolio Partner-
ship project.
͑
͑
1
B. C. Djubua and N. N. Chubum, IEEE Trans. Electron Devices 38, 2314
͑1991͒.
2
S. R. P. Silva, G. A. J. Amaratunga, and K. J. Okano, J. Vac. Sci. Technol.
B 557, 17 ͑1999͒.
A. Wisitsora-at, W. P. Kang, J. L. Davidson, Q. Li, J. F. Xu, and D. V.
Kerns, Appl. Surf. Sci. 280, 146 ͑1999͒.
Y. H. Chen, C. Ti, and I. N. Lin, Appl. Surf. Sci. 142, 516 ͑1999͒.
Y. Ando, Y. Nishibayashi, H. Furuta, K. Kobashi, T. Hirao, and K. Oura,
Diamond Relat. Mater. 12, 1681 ͑2003͒.
O. M. Küttel, O. Gröning, Ch. Emmenegge, L. Nilsson, L. Maillard
Diederich, and L. Schlapbach, Carbon 745, 37 ͑1999͒.
S. Y. Chen, M. Y. Lee, C. S. Chen, and J. T. Lue, Phys. Lett. A 436, A313
3
4
5
6
6
,11
rise in emission after the threshold field Eth.
7
Finally, Fig. 4 shows the behavior of emission current
versus time for CNT/BDD/felt at two different fields. The
voltage was applied without any previous voltage scans on
the electrodes. The CNT/BDD/felt shows small fluctuations
during the first 3 h of the emission, and then stabilized at
8͑
2003͒.
P. M. Koinkar, R. J. Mahajan, P. P. Patil, and M. A. More, Thin Solid
Films 275, 474 ͑2005͒.
9
0
E. C. Almeida, N. G. Ferreira, and J. M. Rosolen, Diamond Relat. Mater.
1
E. C. Almeida, A. V. Diniz, J. M. Rosolen, and N. G. Ferreira, Diamond
Relat. Mater. 14, 679 ͑2005͒.
4, 1673 ͑2005͒.
1
−
1
0.2 ␮A at 2.1 V ␮m . After the first run in Fig. 4͑a͒, the
11
sample was removed from the chamber and then subject to
emission tests after long-term exposure to air for a day at
J. M. Bonard, M. Croci, C. Klinke, R. Kurt, O. Noury, and N. Weiss,
Carbon 40, 1715 ͑2002͒.
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:
28.252.67.66 On: Tue, 23 Dec 2014 14:52:50
1