nano wires

[ranjith shanmugam]

Patterned Growth and Field-Emission
Properties of AlN Nanocones
Ning Liu,† Qiang Wu,*,† Chengyu He,† Haisheng Tao,† Xizhang Wang,† Wei
Lei,‡ and
Zheng Hu*,†
Key Laboratory of Mesoscopic Chemistry of MOE, School of Chemistry and
Chemical Engineering, Nanjing
University, Nanjing 210093, China, and School of Electronic Science
and Engineering, Southeast University,
Nanjing 210096, China
ABSTRACT Patterned growth of AlN nanocones on a Ni-coated Si substrate
is demonstrated through the reaction between AlCl3 and
NH3 at 700 °C with Mo grid as a mask. The AlN nanocones are
selectively deposited in the hollow region of the mask with diameters
of ∼10 nm at the tips and 50-60 nm at the roots. The field-emission
(FE) performance is effectively enhanced by the patterned
growth mainly because of the decreased screening effect, and both turn-
on and threshold fields are dramatically decreased, less than
half of the corresponding ones for the unpatterned product with
similar sizes. The results indicate that patterned growth is an
efficient
and reproducible way to enhance the FE performance of AlN nanocones,
which could be applied to optimize the FE properties of
other nanoscale field emitters.
KEYWORDS: one-dimensional nanostructure • aluminum nitride • field
emission • patterned growth
1. INTRODUCTION Cold cathode materials have promising applications
in military and domestic industries, and now the
related researches are focused on application in a
field-emission (FE) flat-panel display (1). The evolution of
cold cathodes started with Spindt-type field-emitter arrays
such as Mo and Si microtips in the late 1960s. The subuliform
microtips are ideal morphologies for FE because of the
strong local electric field at the tips and the unique direction
of electron emission. Because of the expensive and complicated
fabrication technique as well as the large surface work
function of Mo and Si microtips, they have not been successfully
put into practice. Soon after that, diamond and
diamondlike carbon films received much attention, which
are regarded as good candidates for cold cathodes owing to
the low surface electron affinity, high chemical stability, and
thermal conductivity. However, the field emitters in these
films are randomly oriented and divergence of the emission
electrons is unfavorable for the addressable fabrication,
which limits their applications (1, 2). Currently, one-dimensional
(1D) nanostructures, represented by carbon nanotubes,
have attracted increasing interest for their sharp
apexes and large aspect ratios, which could efficiently
enlarge the enhancement factor for FE. In seeking appropriate
materials, low surface work function or electron affinity,
high stability, and excellent thermal conductivity are the
main guidelines.
AlN is well-known for its small (even negative) electron
affinity, high stability, and superior thermal conductivity
(3, 4). Recently, various 1D AlN nanostructures such as
nanotubes (5-7), nanowires (8), nanobelts (9) and nanocones
(10-13) have been synthesized. Among them, AlN
nanocone arrays exhibit good FE properties because of their
sharp apexes, large aspect ratios, and good alignment on the
substrate (10-13). To further improve the FE properties, two
means are worth attempting. One is doping other elements
such as Si (14) to increase the electron concentration, and
the other is adjusting the nanocone density to decrease the
screening effect. As experimentally and theoretically confirmed
by Nilsson et al., an interemitter distance of about 2
times the height of the 1D nanostructures optimizes the
emitted current per unit area (15). Hence, the density of the
AlN nanocones plays a crucial role for their FE properties.
Screening effects would reduce the field enhancement factor
and thus the emission current when the density of the AlN
nanocones is high, just like most cases reported (16). Patterned
growth is a good method to control the density of the
nanostructures, and there are mainly two routes. One is to
construct the patterned substrate through selective modification
of the surface hydrophilic/hydrophobic nature via,
e.g., electron beam lithography, ultraviolet light irradiation,
or etching, which could influence the deposition location.
The other is to construct the patterned catalyst for the
sequential patterned growth of nanostructures (17-19).
Because AlN nanostructures can deposit on various substrates
without the assistant of a catalyst (10, 20, 21), these
two routes are impracticable for the patterned growth of 1D
AlN nanostrutures. In this study, patterned growth of AlN
nanocones is realized by introducing Mo grid as a mask,
which could result in the locatable deposition of AlN nanocones
in the hollow region of the mask. The patterned AlN
* To whom correspondence should be addressed. E-mail:
wqc...@nju.edu.cn
(Q.W.), zhe...@nju.edu.cn. (Z.H.). Tel.: 0086-25-83686015. Fax:
0086-25-83686251.
Received for review May 15, 2009 and accepted August 3, 2009
† Nanjing University.
‡ Southeast University.
DOI: 10.1021/am9003304
© 2009 American Chemical Society
ARTICLE
www.acsami.org VOL. 1 • NO. 9 • 1927–1930 • 2009 1927
Published on Web 08/17/2009
Downloaded by BHARATHIAR UNIV on September 30, 2009 | http://pubs.acs.org
Publication Date (Web): August 17, 2009 | doi: 10.1021/am9003304
nanocones show a much better FE performance in comparison
with the unpatterned sample, indicating potential applications.
2. EXPERIMENTAL SECTION
Preparation of Patterned AlN Nanocones. Patterned AlN
nanocones were prepared through the reaction between AlCl3
vapor and NH3 gas at 700 °C (10) under the restriction of a Mo
mask, as schematically illustrated in Figure 1A. Briefly, a Mo
grid was tightly attached to a Si(111) substrate coated with a
10 nm Ni film. The substrate was placed in the center of an
alumina tube in a horizontal tubular furnace. When the temperature
was raised to 700 °C, AlCl3 was vaporized upstream
at 135 °C and subsequently transported to the reaction zone
by an Ar flow (300 sccm), where it reacted with NH3 gas (20
sccm). AlN nanocones were deposited onto the interspaces of
the Mo grid, and the places covered by the Mo cross-bars were
bare. The reaction lasted for 4 h, and then the system was
cooled down to room temperature under an Ar flow. After
removal of the Mo grid, patterned AlN nanocones were obtained.
In this paper, two Mo grids with different dimensions
have been attempted. For comparison, a Si wafer without a Mo
grid was also used as the substrate to deposit AlN products
under a similar growth process.
Characterization of the Product. The as-prepared products
were characterized by X-ray diffraction (XRD; Philips X’pert Pro
X-ray diffractometer) and scanning electron microscopy (SEM;
Hitachi S-4800). FE measurements were performed by using a
parallel-plate configuration in a vacuum chamber at a pressure
of about 1 × 10-4 Pa.
3. RESULTS AND DISCUSSION
SEM images of the patterned AlN nanocones synthesized
using two masks with different pore sizes and gaps (patterns
I and II) are shown in Figure 1B-F. For the patterned sample
I, the dimension of each deposited unit is about 185 μmand
the distance between the two neighboring regions is about
35 μm, in good agreement with the pattern of the Mo grid
used. The AlN nanocones inside the block show the quasialigned
morphologies (Figure 1D), while no AlN nanocones
could be observed for the place in between the blocks (Figure
1E), which was covered by the grid cross-bar during the
synthesis. Compared with the patterned sample I, the patterned
sample II has smaller dimension of about 100 μm
for each block and a larger distance between two neighboring
blocks of about 60 μm (Figures 1F and S1 in the
Supporting Information). The patterned products are quite
uniform (Figure 1B,F), which favors application as patterned
pixels for FE. The AlN nanocones vary from submicrometer
to micrometer in length, with diameters of ∼10 nm at the
tips and 50-60 nm at the roots (see Figure S2 in the
Supporting Information). The corresponding energy-dispersive
X-ray analysis (EDX; see Figure S3 in the Supporting
Information) shows that the atomic ratio of Al to N is about
1:0.93, rather near the stoichiometric ratio of AlN.
Figure 2 shows the SEM images of the unpatterned AlN
product without restriction of the Mo grid. The nanocones
are quasi-aligned and uniform with geometrical parameters
similar to those of the patterned AlN nanocones, indicating
the little influence of the Mo mask on the morphology of the
AlN nanocones. The XRD curve is in agreement with that of
hexagonal AlN (see Figure S4 in the Supporting Information).
The growth of the AlN nanocones could be understood as
the vapor-solid epitaxial mechanism, as proposed in our
previous report (10).
In comparison with the unpatterned AlN nanocones,
there exists a larger marginal area for the patterned samples
and a better FE performance could be expected because of
the enhanced field strength at the marginal region (22).
Figure 3A depicts the FE curves of the current density versus
the applied field (J-E) for the two patterned and one
unpatterned AlN nanocones with a cathode-anode spacing
of 100 μm. The turn-on electric field Eto (the electric field to
generate an emission current density of 10 μA/cm2) and the
threshold field Ethr (the electric field to generate an emission
current density of 1 mA/cm2 (23), which could satisfy the
primary requirement for application in conventional flatpanel
displays) are listed in Table 1. It is seen that both Eto
FIGURE 1. Schematic illustration for the patterned growth process
(A) and SEM images of the patterned AlN nanocones synthesized
using two masks with different pore sizes and gaps (B-F). Patterned
sample I: the overview (B), the enlarged region (C), and the part
inside the block (D) and in between the neighboring blocks (E) as
marked by the panes in part C. Patterned sample II: the overview
(F). Compared with pattern I, pattern II has a smaller block size and
a larger gap between two neighboring blocks.
FIGURE 2. SEM images of unpatterned AlN nanocones. The inset
shows the magnified image.
ARTICLE 1928 VOL. 1
•
NO. 9
•
1927–1930 •
2009 Liu et al. www.acsami.org
Downloaded by BHARATHIAR UNIV on September 30, 2009 | http://pubs.acs.org
Publication Date (Web): August 17, 2009 | doi: 10.1021/am9003304
and Ethr of the two patterned AlN nanocones are lower than
half of the corresponding ones of the unpatterned AlN
nanocones. Moreover, for the two patterned samples, Eto
and Ethr for the patterned sample II are smaller than the
corresponding ones for the patterned sample I. Taking into
account the fact that pattern II has a smaller block size and
a larger gap between two neighboring blocks than pattern I
does, it could be deduced that the enhanced FE of the
patterned sample is mainly attributed to the decreased
screening effect at the edge of the blocks. This is also
supported by the FE images of the patterned AlN array (see
Figure S5 in the Supporting Information). Although some
other factors such as the shape, defect, impurity, or surface
property of the nanocones near the edges could also influence
the FE performance of AlN patterns, they could not be
the main reason responsible for such a significant increase
of the FE properties because all of the samples were synthesized
under the same experimental conditions except for
the pattern sizes. Hence, the most obvious difference for the
three samples is the degree of the screening effect, which
should be the dominant factor responsible for their different
FE performances. It is seen that the Ethr values are still higher
than those of carbon nanotubes (24); they are comparable
to those of ZnO (25) and B (26) nanowire arrays. These
results clearly indicate that patterned growth is an efficient
and reproducible method (see Figure S6 in the Supporting
Information) to enhance the FE of AlN nanocones mainly
because of the decreased screening effect.
The Fowler-Nordheim (F-N) plots of the J-E curves, i.e.,
ln(J/E2) vs 1/E, are shown in Figure 3B. According to the F-N
theory, the slope of the F-N plots is equal to -6830φ3/2/ ,
where φ is the work function and the enhancement factor.
Taking φ ) 3.7 eV for AlN (6), is estimated to be 430 for
the unpatterned AlN nanocones. As shown in Table 1, the
enhancement factors of the two patterned AlN nanocones
are increased because of the decreased screening effect. For
the patterned sample I, the F-N plots show a two-sectional
feature, and is calculated to be ∼950 and ∼600 at high
and low electric fields, respectively. The two-sectional feature
of the F-N plot may result from the space charge effect
(27). As is deduced from Figure 3A, at least a 5 times larger
current density was obtained for the patterned sample I than
that for the unpatterned sample at high electric field. The
high current density may generate the space charge by
ionizing the trace gas molecules existing in the gap, which
include the residual gases, the gases desorbed from the
anode due to the electron bombardment, and those from
the cathode due to local heating. With the positive ions
forced to the AlN emitter, the local electric field on the
surface of the emitter would be further enhanced, resulting
in a larger value at high electric field. For the patterned
sample II, the F-N plot is close to linear and the value is
as high as 1561. This may result from the appearance of the
space charge effect even at low electric field because of the
fast increase of the current density, as seen in Figure 3A.
Hence, the space charge effect functions almost in the whole
range of the applied field, different from the case for the
patterned sample I, leading to the linear feature of the F-N
plot. More direct experimental evidence seems to be needed
to clarify this point, e.g., by monitoring the evolution of the
space charge distribution with the applied field.
After FE measurements, no obvious morphological change
could be observed for the AlN nanocones, which is much
different from the case for ZnO nanowire arrays (see Figure
S7 in the Supporting Information) (16). This could be attributed
to the good flexibility and high stability of AlN
nanocones (28), which facilitates the FE application of AlN
nanocones.
4. CONCLUSIONS
In summary, a simple route has been developed to
prepare patterned AlN nanocones on a Ni-coated Si substrate
through the reaction of AlCl3 and NH3 at 700 °C with Mo
grid as a mask. The AlN nanocones have diameters of ∼10
nm at the tips and 50-60 nm at the roots. The patterned
growth could effectively decrease the screening effect because
of the increased marginal region. As a result, both Eto
and Ethr for the patterned AlN nanocones are dramatically
decreased, less than half of the corresponding ones for the
unpatterned AlN nanocones. The results indicate that pat-
FIGURE 3. (A) J-E curves of the patterned and unpatterned AlN
nanocones with an electrode distance of 100 μm. (B) F-N plots
corresponding to the J-E curves in part A.
Table 1. Eto, Ethr, and of the Patterned and
Unpatterned AlN Nanocones
pattern I pattern II unpatterned
Eto (V/μm) 7.7 4.8 15.2
Ethr (V/μm) 13.1 11.2 29.5
977/552 1561 436
ARTICLE
www.acsami.org VOL. 1 • NO. 9 • 1927–1930 • 2009 1929
Downloaded by BHARATHIAR UNIV on September 30, 2009 | http://pubs.acs.org
Publication Date (Web): August 17, 2009 | doi: 10.1021/am9003304
terned growth is an efficient approach to enhancing the FE
properties of AlN nanocones, which should be applicable to
optimization of the FE properties of other nanoscale field
emitters.
Acknowledgment. This work was jointly supported by the
National Basic Research Program of China (2007CB935503),
NSFC(Grants20525312,20601013,20873057,and20833002),
and the program for Changjiang Scholars and Innovative
Research Team in University.
Supporting Information Available: Figures giving additional
SEM images, an XRD pattern, EDX results, and FE
measurements. This material is available free of charge via
the Internet at http://pubs.acs.org.
REFERENCES AND NOTES
(1) Xu, N. S.; Huq, S. E. Mater. Sci. Eng. R 2005, 48, 47–189.
(2) Xu, N. S.; Latham, R. V.; Tzeng, Y. Electron. Lett. 1993, 29,
1596–
1597.
(3) Wu, C. I.; Kahn, A.; Hellman, E. S.; Buchanan, D. N. E. Appl.
Phys.
Lett. 1998, 73, 1346–1348.
(4) Benjamin, M. C.; Wang, C.; Davis, R. F.; Nemanich, R. J. Appl.
Phys. Lett. 1994, 64, 3288–3290.
(5) Wu, Q.; Hu, Z.; Wang, X. Z.; Lu, Y. N.; Chen, X.; Xu, H.; Chen, Y.
J. Am. Chem. Soc. 2003, 125, 10176–10177.
(6) Tondare, V. N.; Balasubramanian, C.; Shende, S. V.; Joag, D. S.;
Godbole, V. P.; Bhoraskar, S. V.; Bhadbhade, M. Appl. Phys. Lett.
2002, 80, 4813–4815.
(7) Yin, L. W.; Bando, Y.; Zhu, Y. C.; Golberg, D.; Li, M. S. Adv.
Mater.
2004, 16, 929–933.
(8) Wu, Q.; Hu, Z.; Wang, X. Z.; Lu, Y. N.; Huo, K. F.; Deng, S. Z.;
Xu,
N. S.; Shen, B.; Zhang, R.; Chen, Y. J. Mater. Chem. 2003, 13,
2024–2027.
(9) Wu, Q.; Hu, Z.; Wang, X. Z.; Chen, Y.; Lu, Y. N. J. Phys. Chem. B
2003, 107, 9726–9729.
(10) Liu, C.; Hu, Z.; Wu, Q.; Wang, X. Z.; Chen, Y.; Sang, H.; Zhu, J.
M.;
Deng, S. Z.; Xu, N. S. J. Am. Chem. Soc. 2005, 127, 1318–1322.
(11) Shi, S. C.; Chen, C. F.; Chattopadhyay, S.; Chen, K. H.; Chen, L.
C.
Appl. Phys. Lett. 2005, 87, 073109.
(12) Zhao, Q.; Xu, J.; Xu, X. Y.; Wang, Z.; Yu, D. P. Appl. Phys.
Lett.
2004, 85, 5331–5333.
(13) Tang, Y. B.; Cong, H. T.; Chen, Z. G.; Cheng, H. M. Appl. Phys.
Lett. 2005, 86, 233104.
(14) Tang, Y. B.; Cong, H. T.; Wang, Z. M.; Cheng, H. M. Appl. Phys.
Lett. 2006, 89, 253112.
(15) Nilsson, L.; Groening, O.; Emmenegger, C.; Kuettel, O.; Schaller,
E.; Schlapbach, L.; Kind, H.; Bonard, J. M.; Kern, K. Appl. Phys.
Lett. 2000, 76, 2071–2073.
(16) Wang, X. D.; Zhou, J.; Lao, C. S.; Song, J. H.; Xu, N. S.; Wang,
Z. L. Adv. Mater. 2007, 19, 1627–1631.
(17) Weintraub, B.; Deng, Y. L.; Wang, Z. L. J. Phys. Chem. C 2007,
111, 10162–10165.
(18) Masuda, Y.; Kinoshita, N.; Sato, F.; Koumoto, K. Cryst. Growth
Des. 2006, 6, 75–78.
(19) He, J. H.; Hsu, J. H.; Wang, C. W.; Lin, H. N.; Chen, L. J.;
Wang,
Z. L. J. Phys. Chem. B 2006, 110, 50–53.
(20) Shi, S. C.; Chen, C. F.; Chattopadhyay, S.; Lan, Z. H.; Chen, K.
H.;
Chen, L. C. Adv. Funct. Mater. 2005, 15, 781–786.
(21) Yazdi, G. R.; Syvajarvi, M.; Yakimova, R. Appl. Phys. Lett. 2007,
90, 123103.
(22) Di, Y. S.; Yang, X. X.; Lei,W.; Zhang, X. B.; Cui, Y. K.; Wang,
Q. L.;
Yang, G. D. Nanotechnology 2007, 18, 505701.
(23) Wang, Q. H.; Corrigan, T. D.; Dai, J. Y.; Chang, R. P. H.;
Krauss,
A. R. Appl. Phys. Lett. 1997, 70, 3308–3310.
(24) Fan, S. S.; Chapline, M. G.; Franklin, N. R.; Tombler, T. W.;
Cassell,
A. M.; Dai, H. J. Science 1999, 283, 512–514.
(25) Lee, C. J.; Lee, T. J.; Lyu, S. C.; Zhang, Y.; Ruh, H.; Lee, H.
J. Appl.
Phys. Lett. 2002, 81, 3648–3650.
(26) Liu, F.; Tian, J. F.; Bao, L. H.; Yang, T. Z.; Shen, C. M.; Lai,
X. Y.;
Xiao, Z. M.; Xie, W. G.; Deng, S. Z.; Chen, J.; She, J. C.; Xu, N. S.;
Gao, H. J. Adv. Mater. 2008, 20, 2609–2615.
(27) Xu, N. S.; Chen, Y.; Deng, S. Z.; Chen, J.; Ma, X. C.; Wang, E.
G.
J. Phys. D: Appl. Phys. 2001, 34, 1597–1601.
(28) Zhang, Y. J.; Liu, J.; He, R. R.; Zhang, Q.; Zhang, X. Z.; Zhu,
J. Chem.
Mater. 2001, 13, 3899–3905.
AM9003304
ARTICLE 1930 VOL. 1
•
NO. 9
•
1927–1930 •
2009 Liu et al. www.acsami.org
Downloaded by BHARATHIAR UNIV on September 30, 2009 | http://pubs.acs.org
Publication Date (Web): August 17, 2009 | doi: 10.1021/am9003304