Self-catalyst growth of LaB6 nanowires and nanotubes

Article abstract of DOI:10.1016/j.cplett.2006.03.049

LaB₆ nanowires have been successfully fabricated with self-catalyst method using lanthanum (La) powders and boron trichloride (BCl₃) gas mixed with hydrogen and argon. Our results show that lanthanum hexaboride (LaB₆) nanowires are 20-200 nm thick and several micrometers long. Transmission electron microscopy (TEM) reveals that the nanowires are highly crystalline. In our experiment LaB₆ nanotubes have also been observed for the first time. Our results indicate that LaB₆ nanotubes are polycrystalline with 200-400 nm thick and around several hundred nanometers to one micrometer long. A growth mechanism based on the self-catalyst process is proposed for the formation of the nanowires.

Full text of DOI:10.1016/j.cplett.2006.03.049

Chemical Physics Letters 423 (2006) 138–142
www.elsevier.com/locate/cplett
Self-catalyst growth of LaB6 nanowires and nanotubes
*
Junqi Xu, Yanming Zhao , Chunyun Zou
School of Physics, South China University of Technology, Wushan Road, Tianhe, Guangzhou, Guangdong 510640, PR China
Received 5 January 2006; in final form 24 February 2006
Available online 27 March 2006
Abstract
LaB6 nanowires have been successfully fabricated with self-catalyst method using lanthanum (La) powders and boron trichloride
(BCl3) gas mixed with hydrogen and argon. Our results show that lanthanum hexaboride (LaB6) nanowires are 20–200 nm thick and
several micrometers long. Transmission electron microscopy (TEM) reveals that the nanowires are highly crystalline. In our experiment
LaB6 nanotubes have also been observed for the first time. Our results indicate that LaB6 nanotubes are polycrystalline with 200–400 nm
thick and around several hundred nanometers to one micrometer long. A growth mechanism based on the self-catalyst process is pro-
posed for the formation of the nanowires.
Ó 2006 Elsevier B.V. All rights reserved.
1. Introduction
known materials and its performance is unaffected by the
presence of either nitrogen or oxygen [15,16]. Therefore,
lanthanum hexaboride has attracted considerable contem-
porary interest from both fundamental and practical
viewpoint.
LaB6 whiskers have been synthesized from the following
chemical reaction: 2LaCl3(g) + 12BCl3(g) + 21H2 = 2-
LaB6(s) + 42HCl(g) [17–19]. Recently, using the similar
chemical reaction process, single-crystalline LaB6 nano-
wires have been produced by Zhang et al. [20] with the
aid of catalysts. Here we report the growth of LaB6 nano-
wires and nanotubes which was produced by the direct
reaction on substrates without using any catalysts, i.e., with
the self-catalyst method.
Rare-earth alkali metal borides belong to a group of
refractory, non-oxide-type, metal-like compounds that are
characterized by high melting point, high strength, and
high chemical stability, as well as other special peculiarities,
such as a low electronic work function, stable specific resis-
tance, a low expansion coefficient in some temperature
ranges, and high neutron absorbability [1]. All these out-
standing properties result in a wide range of applications,
such as high-energy optical systems [2–4], sensors for
high-resolution detectors [5], electrical coatings for resis-
tors, and thermionic materials [6,7]. In all these materials,
lanthanum hexaboride (LaB6) occupies a unique place as
a type of electron gun for an electron microscope and elec-
tron beam writing unit due to its high rate of thermoionic
emission, fair chemical stability at high melting point
(above 2500 °C) [8]. A common feature of LaB6 is that it
possess very low work function, which is lower than any
other known materials except YB6 (work functions of
2.74 and 2.22 eV for LaB6 and YB6, respectively) [9–14].
In fact, LaB6 has the highest electronic emissivity of any
2. Experimental
The method employed to synthesize the nanowires was
based on the following chemical reaction: La(s) +
6BCL3(g) + 9H2(g) = LaB6(s) + 18HCl(g). The reaction
was carried out in a quartz tube furnace. Si substrate with
La powders (mass purity: 99.9%) on it was first loaded in a
quartz boat, and the boat was put into a long quartz tube
under N2 atmosphere. Before reaction, the quartz tube was
evacuated and washed with nitrogen for three times. After
that the quartz tube was heated to an expected furnace
*
Corresponding author. Fax: +86 20 87112844.
E-mail address: zhaoym@scut.edu.cn (Y. Zhao).
0009-2614/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.cplett.2006.03.049

J. Xu et al. / Chemical Physics Letters 423 (2006) 138–142
139
temperature (1070 °C) under a mixed gas (50%H2 +
50%Ar) flow of about 30 ml/min, a steady BCl3 flow of
30 sccm was started and maintained 60 min. After the
above procedure the quartz tube was cooled down to the
room temperature under the mixed gas (50%H2 +
50%Ar) atmosphere. The product obtained from the Si
substrate was first washed with distilled water to remove
the HBO3. After drying in the vacuum at 60 °C for 4 h,
the final reddish deep purple or blush deep powder product
was obtained.
The products were characterized and analyzed using
X-ray diffractometer with Cu Ka radiation, LEO 1530
VP field-emission scanning electron microscopy and JEOL
JEM-2010 transmission electron microscopy.
Fig. 1. X-ray diffraction patterns of the sample obtained from Si
substrate.
3. Results and discussion
ground is the silicon substrate which was eroded by the
BCl3 gas. Fig. 2b shows a typical SEM image of the
obtained products. It reveals that the products consist of
wire-like nanostructures with lengths ranging from several
to several tens micrometers and diameters varying from
100 nm to 200 nm. Fig. 2c shows an individual LaB6 nano-
wire at a relatively low magnification, which has the uni-
form diameter of 100 nm. The nanowire appears straight
and has a round tip. In addition, the thinner nanowires
are also found in the products and minimum diameter of
the nanowire is about 20 nm [seen in Fig. 2d].
For electron diffraction and microscopy, the as-prepared
sample was first dispersed in ethanol and then collected
onto a copper grid that was coated with a thin holy carbon
film. As shown in Fig. 3a, under relatively low magnifica-
tion, the products obtained on the clean Si substrate are
composed of nanorods with a sharp tip on its top. From
The phase identification of the product was carried out
by X-ray powder diffractometer. A step scan mode was
adopted with a scanning step of 0.02° and a sampling time
of 3 s. Fig. 1 shows the X-ray diffraction patterns of the as-
prepared LaB6 product. No impurity phase is detected
under the resolution of our X-ray diffractometer. The
X-ray diffraction patterns of the product were successfully
indexed with a cubic lattice using the program Dicvol, and
the lattice parameters were further least-squares refined by
˚
the program PIRUM as a = 4.145 A which agrees well
with the data from reference (JCPDS card: No. 73-1669).
Our X-ray diffraction analysis results suggest that the as-
prepared product is the LaB6 with cubic phase.
Fig. 2a gives an SEM image of the nanowires on the Si
substrate at lower magnification, and the rough back-
Fig. 2. (a) The general view of the products. (b) SEM image of the LaB6 nanowires. (c) A SEM image of an individual nanowire with a round tip. (d) A
typical TEM image of a thinner LaB6 nanowire with diameter about 20 nm.

140
J. Xu et al. / Chemical Physics Letters 423 (2006) 138–142
Fig. 3. (a) Low-magnification TEM image of a LaB6 nanowire. The globule in the graph is a LaB6 nanoparticle. (b) The HRTEM observation suggests
that the growth direction of LaB6 nanowires is parallel to the [111] direction. The inset shows a typical selected-area electron diffraction pattern of the
ꢀ
nanowire from the ½224ꢀ direction.
Fig. 3a it can be seen that globules are also exist in our
product. The diameter of the LaB6 nanowires is around
25 nm and the length extend to a few micrometers. Detailed
structural characterization of the LaB6 nanowires was per-
formed using HRTEM. Fig. 3b shows a typical lattice-
resolved HRTEM image of the LaB6 nanowire shown in
Fig. 3a. It reveals that the LaB6 nanowire is single crystal-
line in nature and free of dislocation and structure defects.
The lattice spacing of 0.240 nm corresponds to the d-spac-
ing of the (111) crystal faces, which indicates that the
growth direction of the nanowires is parallel to the [111]
direction. Also the high-resolution transmission electron
microscopy lattice image indicates that the nanowire was
solid inside. In addition the same structural features are
found in other nanowires by careful HRTEM observation.
In order to identify the crystal lattice of LaB6 nanowire
selected-area electron diffraction patterns were performed,
and the inset of Fig. 3b shows electron diffraction pattern
nearly 30 nm. From the TEM figures we present here it is
hard to conclude that the nanotubes are open in the tips
or not, and this may need to be clarified in the future work.
It is worth noticing that the two nanotubes have the similar
morphology, that is to say the inner wall of the nanotubes
is rough. The selected-area electron diffraction (SAED)
pattern in Fig. 4b shows that the nanotubes are polycrys-
talline structure, which is contrast with that for nanowires
where the single structure is adopted. It is difficult to attri-
bute the polycrystalline structure of the LaB6 nanotubes to
electron beam damage because one of the most excellent
properties of LaB6 is the resistance to electron bombard-
ment [21]. The diffraction pattern of nanotube can be read-
ily indexed as (100), (111), (311) of the primary cubic
structure with lattice constant a agreeing with our X-ray
diffraction measurement result of the products obtained
on the clean Si substrate as well as selected-area electron
diffraction result of nanorods.
One question that need addressing is how the LaB6
nanowires grow. The TEM analysis shows that the nano-
wires may not be dominated by the conventional vapor–
liquid–solid (VLS) mechanism proposed for nanofibers
grown by a catalyst-assisted process, in which a transition
metal particle is capped at the tip of the fiber and serves as
the active catalystic site [22]. In our experiment we did not
observed any metal particles in LaB6 nanowires and nano-
tubes in extensive TEM observations, which give no evi-
dence to support the VLS mechanism. Thus, the growth
procedure may be describe as ‘self-catalytic’ growth [23],
that is to say, in our experiment the melting metal La plays
the role of reactant and catalyst simultaneously. In order to
give a conceptual description of the growth of the LaB6
ꢀ
taken along ½224ꢀ crystal zone axes. The electron diffrac-
tion pattern shows that the LaB6 nanowires adopt single
crystal structure which has a cubic structure and the lattice
constant a agrees well with our X-ray diffraction measure-
ment result.
It is interesting that LaB6 nanotubes were also found at
the same experiment. Fig. 4a shows a representative TEM
image of the LaB6 nanotubes. In Fig. 4a two worm-like
nanotubes were observed and the diameters of them are
not uniform. It can be seen from Fig. 4a that the diameter
of bigger nanotube is ranging 200–400 nm and the length
extending to 1 lm with the wall thickness around 60 nm,
and the smaller one has the diameter of about 200 nm
and the length around 400 nm with the wall thickness

J. Xu et al. / Chemical Physics Letters 423 (2006) 138–142
141
Fig. 4. (a) TEM image of LaB6 nanotubes. The two arrows point at the two nanotubes. (b) Selected-area electron diffraction pattern of the nanotube. The
diffraction rings are indexed according to the LaB6 crystal structure.
nanowires, we draw a model sketched in Fig. 5. When the
furnace temperature reached the melting point of the rare-
earth metal grains the metals start to melt and form a large
quantity of liquid droplets with size close to that of the ori-
ginal solid metal particles (Fig. 5, part A); Subsequently,
the pyrolyzed B3+ from BCl3 is supposed to be absorbed
by the melting metal at some point of the surface, resulting
in there being some LaB6 nanoclusters, which serve as
nuclei for the nanowire growth following (Fig. 5, part B).
The growth of metal boride nanowires begins and contin-
ues as long as appropriate quantities of B3+ and the source
metal reactants are available (Fig. 5, part C). With the
nanowire growing continuously, at some time, the diameter
of metal La droplet decrease to approximate to the diame-
ter of the nanowire (Fig. 5, part D), and at this time the
diameter of the nanowire begins to decrease and generate
a sharp or a round tip with the source being exhausted
(Fig. 5, part E). However, the growth mechanism for
LaB6 nanotubes may be more complicated, and a more
detail study should be carried out.
4. Conclusion
In conclusion, we have successfully fabricated LaB6
nanowires and nanotubes with self-catalysts method using
lanthanum (La) powders and boron trichloride (BCl3) gas
mixed with hydrogen and argon. The nanowires have a
diameter around 20–200 nm and the length extending to
several micrometers. The growth direction of the nanowires
is parallel to the [111] direction. As compared with LaB6
nanowires prepared by Zhang et al. [20], the efficiency of
ours lies in a series of features, such as no catalyst Au or
Pt were used throughout the whole growth process, and
some nanowires are thinner and have a very sharp or a
round tip, which is very important for the field-induced
electron emission. In our experiment LaB6 nanotubes have
also been observed for the first time. The formation mech-
anism of LaB6 nanotubes is not clear. The LaB6 nanowires
and nanotubes can be potentially used as providing ther-
moionic emission, field-induced emission, and thermal
Fig. 5. The sketch of LaB6 nanowire growth process: (A) When the
temperature of the furnace reaches the melting point of metal La, the grain
starts to melt and form a large quantity of liquid droplets with size
identical to those of the original solid metal particles; (B) The pyrolyzed
B3+ from BCl3 is supposed to be absorbed by the melting metal La at some
point of the surface, resulting in some metal boride nanocluster, which
serve as nuclei for the nanowire growth following; (C) The growth of metal
boride nanowires begins after the formation of the nuclei and continues as
long as appropriate quantities of B3+ and the source metal reactants are
available; (D) With the nanowire growing continuously, the diameter of
metal La droplet decrease to equal to the diameter of the nanowire at some
time and (E) after that, the diameter of the nanowire begins to decrease
and generate a sharp or a round tip with the source being exhausted.

142
J. Xu et al. / Chemical Physics Letters 423 (2006) 138–142
[5] G.G. Friz, K.S. Wood, D.V. Vechten, A.L. Gyulamiryan, A.S.
Kuzanyan, N.J. Giordano, T.M. Jacobs, H.-D. Wu, J.S. Horwitz,
A.M. Gulian, Proc. SPIE 4140 (2000) 459.
[6] J.M. Lafferty, J. Appl. Phys. 22 (1951) 299.
[7] S.F. Vogel, Rev. Sci. Instrum. 41 (1970) 585.
[8] R.W. Tohnson, A.H. Daane, J. Phys. Chem. 65 (1961) 909.
[9] S.V. Ermakov, B.M. Tsarev, Radiotekhn. Elektron. 10 (1965) 972.
[10] R.W. Decker, D.W. Stebbins, J. Appl. Phys. 26 (1955) 1004.
[11] B.I. Mikhailovs’kii, Ukr. Fiz. Zh. 7 (1962) 75.
field-induced emission of electrons for TEM, SEM, flat
panel displays, as well as other electronic devices that
require high-performance electron source. The experiments
on the electric field-induced electron emission measurement
of LaB6 nanowires and nanotubes are now in progress.
Acknowledgements
[12] G.A. Kundint’seva, B.M. Tsarev, Radiotekh. Elektron. 3 (1958) 428.
[13] G.V. Samsonov, A.E. Grodstein, Zhur. Fiz. Khim. 30 (1956) 379.
[14] R.L. Czechoslov, J. Phys. 6 (1956) 625.
[15] J.H. Westbrook, Proceedings of the International Symposium on
High Temperature Technology, McGraw-Hill, New York, 1960.
[16] A.F. Carter, G.P. Wood, NASA Memo-2-16-59L, 1959..
[17] S. Motojima, Y. Takahashi, K. Sugiyama, J. Cryst. Growth 44 (1978)
106.
The authors thank Dan Jiang at the Sun Yat-sen Uni-
versity for valuable discussion with the TEM results. This
work was funded by NSFC Grant (No. 50472055) sup-
ported through NSFC Committee of China.
References
[18] E.I. Givargizov, L.N. Obolenskaya, J. Cryst. Growth. 51 (1981) 190.
[19] E.I. Givargizov, L.N. Obolenskaya, J. Less-Common Met. 117 (1986)
97.
[1] S.Q. Zheng, G.H. Min, Z.D. Zou, H.S. Yu, J.D. Han, J. Am. Ceram.
Soc. 84 (2001) 2725.
[2] K.R. Peschmann, J.T. Calow, K.G. Knauff, J. Appl. Phys. 44 (1973)
2252.
[20] H. Zhang, Q. Zhang, J. Tang, L.C. Qin, J. Am. Chem. Soc. 127 (2005)
2862.
[3] E.A. Kafadaryan, S.I. Petrosyan, G.R. Badalyan, S.R. Harutyunyan,
A.S. Kuzanyan, Thin Solid Films 416 (2002) 218.
[4] P.E. Oettinger, Appl. Phys. Lett. 56 (1990) 333.
[21] K. Ebihara, S. Hiramatsu, Rev. Sci. Instrum. 67 (1996) 2765.
[22] R.S. Wagner, W.C. Ellis, Appl. Phys. Lett. 4 (1964) 89.
[23] H.Y. Dang, J. Wang, S.S. Fan, Nanotechnology 14 (2003) 738.