Preparation and electrochemical hydrogen storage of boron nitride nanotubes

Article abstract of DOI:10.1021/jp050105u

Boron nitride (BN) nanotubes were synthesized through chemical vapor deposition over a wafer made by a LaNi₅/B mixture and nickel powder at 1473 K. Scanning electron microscopy, transmission electron microscopy, energy-dispersive spectroscopy, X-ray diffraction, and X-ray photoelectron spectroscopy were performed to characterize the microstructure and composition of BN nanotubes. It was found that the obtained BN nanotubes were straight with a diameter of 30-50 nm and a length of up to several microns. We first verify that the BN nanotubes can storage hydrogen by means of an electrochemical method, though its capacity is low at present. The hydrogen desorption of nonelectrochemical recombination in cyclic voltammograms, which is considered as the slow reaction at BN nanotubes, suggests the possible existence of strong chemisorption of hydrogen, and it may lead to the lower discharge capacity of BN nanotubes. It is tentatively concluded that the improvement of the electrocatalytic activity by surface modification with metal or alloy would enhance the electrochemical hydrogen storage capacity of BN nanotubes. ? 2005 American Chemical Society.

Full text of DOI:10.1021/jp050105u

J. Phys. Chem. B 2005, 109, 11525-11529
11525
Preparation and Electrochemical Hydrogen Storage of Boron Nitride Nanotubes
X. Chen,† X. P. Gao,*,† H. Zhang,† Z. Zhou,† W. K. Hu,† G. L. Pan,† H. Y. Zhu,*,‡
T. Y. Yan,† and D. Y. Song†
Institute of New Energy Material Chemistry, Department of Materials Chemistry, N & T Joint Academy,
Nankai UniVersity, Tianjin 300071, China, and Australian Key Centre of Microanalysis & Microscopy and
School of Chemistry, The UniVersity of Sydney, New South Wales 2006, Australia
ReceiVed: January 7, 2005; In Final Form: April 12, 2005
Boron nitride (BN) nanotubes were synthesized through chemical vapor deposition over a wafer made by a
LaNi5/B mixture and nickel powder at 1473 K. Scanning electron microscopy, transmission electron microscopy,
energy-dispersive spectroscopy, X-ray diffraction, and X-ray photoelectron spectroscopy were performed to
characterize the microstructure and composition of BN nanotubes. It was found that the obtained BN nanotubes
were straight with a diameter of 30-50 nm and a length of up to several microns. We first verify that the BN
nanotubes can storage hydrogen by means of an electrochemical method, though its capacity is low at present.
The hydrogen desorption of nonelectrochemical recombination in cyclic voltammograms, which is considered
as the slow reaction at BN nanotubes, suggests the possible existence of strong chemisorption of hydrogen,
and it may lead to the lower discharge capacity of BN nanotubes. It is tentatively concluded that the
improvement of the electrocatalytic activity by surface modification with metal or alloy would enhance the
electrochemical hydrogen storage capacity of BN nanotubes.
I. Introduction
recently that BN nanotubes can uptake 1.8-2.6 wt % hydrogen
under ∼10 MPa at room temperature17 and collapsed BN
Boron nitride (BN) nanotubes have been attracted much
attention in recent years due to their unique structures and
semiconductive properties. Several methods have been devel-
oped to synthesize BN nanotubes, including arc discharge,1
chemical vapor deposition (CVD),2 laser ablation,3 and reactive
ball-milling,4 similar to carbon nanotubes. The preparation of
BN nanotubes with a high purity is more important for their
potential application. The growth of BN nanotubes depends
strongly upon the presence of catalysts and precursors. Metallic
particles,5-7 alloy catalysts,8 and boron compound precursors9,10
were found to be very effective for the production of BN
nanotubes. It was demonstrated that rare-earth-based AB5 and
Zr-based AB2 hydrogen storage alloys, which are extensively
used in metal hydride-nickel (MH/Ni) batteries, are good
catalysts for the synthesis of carbon nanotubes.11,12 Therefore,
the utilization of the hydrogen storage alloy (for example, LaNi5
alloy) as the catalyst precursor would be feasible and economical
as compared with the LaB6 alloy and metallic particles in the
synthesis of BN nanotubes.
Hydrogen storage is a vital issue in the application of
hydrogen as a future energy carrier. Many alloys are capable
of reversibly storing large amounts of hydrogen, either through
absorbing gaseous hydrogen at certain pressures and tempera-
tures or by an electrochemical process to form a metal hydride.13
However, traditional metal hydrides usually have the problems
of low gravimetric storage capacities, and thus are not able to
satisfy the requirements for practical hydrogen storage. In recent
years, nonmetallic nanotube materials such as carbon nanotubes
and BN nanotubes have attracted much attention as candidates
for high hydrogen storage capacities.14-18 It has been reported
nanotubes exhibit an even higher hydrogen adsorption capacity
(4.2 wt %) than any multiwalled carbon nanotubes.18 Theoretical
calculations have also been performed to investigate hydrogen
adsorption on BN nanotube and other fullerene materials.19,20
It was suggested that the large uptake of hydrogen in BN
nanotubes was primarily due to chemisorption.6 In addition to
the gas-phase reaction, hydrogen can also be stored in nano-
structured carbon through electrochemical processes in aqueous
solutions (KOH or H2SO4 electrolyte) under moderate condi-
tions.21-27 Therefore, hydrogen storage in BN nanotubes is also
worth exploring through electrochemical processes.
In the present study, an effective method to synthesize large-
scale BN nanotubes and control their morphologies was
described using a LaNi5 alloy as a catalyst precursor through
chemical vapor deposition, and hydrogen storage of the BN
nanotubes was also investigated in an electrochemical process.
It was found for the first time that BN nanotubes can store
hydrogen electrochemically, and such a study can also provide
a further understanding of the mechanism of hydrogen storage
in BN nanotubes through a gas-phase reaction.
II. Experimental Section
A. Sample Preparation and Characterization. The LaNi5
hydrogen storage alloy powder and amorphous boron were
mixed by ball-milling for 5 h with the mass ratio of 1:4. The
mixture of 50 mg of the as-prepared LaNi5/B and 150 mg of
the Ni powder was pressed into a wafer under a pressure of 30
MPa. The wafer located in a ceramic boat was heated to 1473
K in a resistance furnace, while the argon gas passed through
the quartz tubes during the process. Then ammonia gas was
introduced to the above reaction system at a flow rate of 100
sccm for 1 h. After the reaction was completed, the wafer was
dispersed ultrasonically in ethanol, and a drop in the above layer
* Authors to whom correspondence should be addressed. E-mail:
xpgao@nankai.edu.cn; h.zhu@emu.usyd.edu.au.
† Nankai University.
‡ The University of Sydney.
10.1021/jp050105u CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/19/2005

11526 J. Phys. Chem. B, Vol. 109, No. 23, 2005
Chen et al.
Figure 1. SEM images of the as-prepared wafer containing (a) dense BN nanotubes and (b) purified BN nanotubes and (c) TEM and (d) HRTEM
images of an individual BN nanotube.
was placed on a copper grid. Scanning electron microscopy
(SEM, Hitachi S-3500N), transmission electron microscopy
(TEM, Tecnai 20), and X-ray photoelectron spectroscoy (XPS,
PHI-5300ESCA) were employed to characterize the as-prepared
samples.
discharge curve and cycle life measurements. The discharge
cutoff potential was set to be -0.4 V vs a Hg/HgO reference
electrode. The cyclic voltammetric (CV) experiments were taken
at different scan rates after fully charging using an IM6e
electrochemical workstation (ZAHNER). The working elec-
trodes for the CV experiments were prepared using the same
method described above, except that the apparent surface of
the porous nickel substrate was about 1.0 cm2 and the carbonyl
nickel powder was not added.
B. Electrochemical Hydrogen Storage. The powder ob-
tained on the surface of the as-prepared wafer was immersed
in concentrated nitric acid for more than 24 h for purification
and then washed with distilled water until pH ) 7. The purified
BN nanotubes (10 mg) and carbonyl nickel powders (INCO
255, 100 mg) were mixed completely and then pasted into the
substrate of porous nickel. After being pressed, the above BN
nanotube electrodes were applied to charge and discharge
measurements. Electrochemical measurements of BN nanotube
electrodes were performed in a three-electrode compartment
using a sintered nickel electrode as the counter electrode and
an Hg/HgO electrode as a reference electrode in 6 mol/L KOH
electrolyte at room temperature in air (normal atmosphere). For
comparison, the electrodes with commercial BN powders
(hexagonal structure) were also prepared and measured under
the same conditions as the blank. The working electrodes were
fully charged at 500 or 1000 mA g-1 for 1 h and discharged at
the same current density after a rest of 5 min in both charge/
III. Results and Discussion
The as-prepared and purified BN nanotubes are up to several
tens of microns long and 30-50 nm in diameter as shown in
Figure 1. The tubes have a perfect multiwalled cylindrical
structure and do not contain internal caps. The inner diameter
of the BN nanotubes is almost uniform, approximately 6-10
nm. In Figure 2a, the strongest diffraction peaks of the purified
BN nanotubes correspond to the (002) and (100) planes of BN
compounds with a hexagonal structure (Joint Committee on
Powder Diffraction Standards Card No. 34-0421), respectively.
There are no impurities that can be detected from the X-ray
diffraction (XRD) pattern, indicating that all of the catalysts
and precursors can be removed during the purification. The

Electrochemical Hydrogen Storage in BN Nanotubes
J. Phys. Chem. B, Vol. 109, No. 23, 2005 11527
Figure 2. (a) XRD patterns of the purified BN nanotubes and (b) the
calcined products of a LaNi5 alloy and B powders mixed with nickel
powders.
calculated interference fringe spacing is about 0.34 nm from
the high-resolution TEM (HRTEM) image in Figure 1c, which
corresponds to the (002) plane of the purified BN nanotubes
and is in good agreement with the value derived from the XRD
data. It is demonstrated from B1s and N1s core level XPS spectra
of the purified products (Figure 3) that distinct peaks are
observed at binding energies of 190 and 398 eV, respectively,
revealing the existence of the BN compound. The atomic ratio
of B to N was determined to be 0.97, which is close to the
stoichiometric BN composition.2,9 It is also confirmed by
energy-dispersive specrometry (EDS, Figure 4) that there are
not any impurities, indicating that La-Ni, Ni-B, and La-B
alloys and metallic Ni are completely removed during purifica-
tion. The Cu and C signals come from the Cu grid and the
carbon film supporting the specimen in the TEM observation.
It can be seen from the XRD patterns of the calcined products
of the LaNi5 alloy and the B powders mixed with nickel powder
that LaB6, NiB, and a small amount of the residual Ni phases
can be detected (Figure 2b). It is considered that boron reacted
with LaNi5 and Ni to form LaB6 and NiB during the heat
treatment process at the first stage. When the temperature is
controlled to 1473 K and ammonia gas is introduced, BN is
then synthesized by the reaction of superfluous boron and
ammonia over NiB/LaB6 particles. As for the growth mechanism
of BN nanotubes, the chemical reactions in the whole process
can be written as
Figure 3. (a) B1s and (b) N1s core level spectra of the purified BN
nanotubes.
particles from agglomerating and keeps the melting point low
of small-sized NiB particles. LaNi5 alloy used here is more
effective to form NiB intermetallic compounds for the subse-
quent synthesis of BN nanotubes.
To examine hydrogen storage in the electrochemical process,
the first charge and discharge curves are shown in Figure 5a
for the purified BN nanotube electrodes at different current
densities. The initial electrochemical discharge capacities of 68
mA h g-1 (corresponding to ∼0.25 wt % hydrogen) and 54
mA h g-1 were obtained in the BN nanotube electrodes at the
discharge current densities of 500 and 1000 mA g-1, respec-
tively. The discharge capacity of carbonyl nickel powders
(INCO 255) is only 0.4 mA h g-1 as reported previously,29
inferring that carbonyl nickel powders give a little contribution
to the BN nanotube electrodes. In the meantime, for comparison,
another electrode was prepared with commercial BN (hexagonal
structure) powder as the blank electrode, and its discharge
capacity was only 16 mA h g-1, so the nanotubes have much
higher electrochemical activities. It should be noted that the
electrochemical hydrogen storage capacity obtained in this
investigation is much lower than the hydrogen storage capacity
reported in gas-phase reaction under a high pressure of 10 MPa
or above.17,18 The lower discharge capacity of the BN nanotubes
may result from the normal atmosphere of 0.1 MPa (from the
steady-state potential of -0.93 V vs a Hg/HgO reference
electrode according to Nernst equation in 6 mol/L KOH solution
at room temperature) in the electrochemical process. The cycle
LaNi5(s) + Ni(s) + B(s) f LaB6(s) + NiB(s,l)
B(s) + NH3(g) NiB(l)/LaB (s)8 BN(s) + H2(g)
6
The formation of liquid NiB here is a key catalyst for the
second reaction.28 It is observed from a SEM image that few
BN nanotubes can be formed if only LaNi5/B particles are
performed. Once the nickel reagent is added with a suitable
amount of boron powders, the NiB yield increases obviously.
Moreover, the morphology of the BN nanotubes can be
improved greatly. BN nanotubes are formed by the reaction of
boron and ammonia gas in the liquid NiB droplet, which can
be supported by previous literature.28 And there is no evidence
to suggest that LaB6 is used as catalyst by the CVD method to
produce BN nanotubes, but it probably prevents the NiB

11528 J. Phys. Chem. B, Vol. 109, No. 23, 2005
Chen et al.
Figure 4. EDS spectrum of the purified BN nanotubes.
life of the purified BN nanotube electrodes at two current
densities is shown in Figure 5b. A longer activation period was
needed at the higher current density. The discharge capacities
remain over 50 mA h g-1 after 50 cycles at the discharge current
densities of 500 and 1000 mA g-1. This suggests that the BN
nanotubes have a strong resistance against oxidation and
corrosion.
Cyclic voltammograms (CVs) were carried out to further
examine the electrochemical hydrogen adsorption-oxidation
reactions of purified BN nanotubes. As shown in Figure 6, the
anodic oxidation peaks of hydrogen in the CVs are in the
potential range between -0.50 and -0.55 V vs Hg/HgO, and
the cathodic adsorption peaks of hydrogen are observed around
-0.96 V, which is close to that of metal hydride electrodes
and carbon nanotube electrodes.22,30 Interestingly, the desorption
peaks of hydrogen appear in the range between -0.78 and
-0.85 V (depending on the scan rates) in the anodic process,
indicating a slow desorption process of hydrogen at BN
nanotubes. The electrochemical oxidation of hydrogen in the
anodic process is dominant as usually observed in the CVs of
metal hydride electrodes. Clearly, the desorption peak of
hydrogen appears prior to the electrochemical oxidation peak
of hydrogen, suggesting the possible existence of the strong
chemisorption of hydrogen. The chemisorption was also found
in the gas-phase reaction17,18 and verified by theoretical calcula-
tion.31 The mechanism of the electrochemical hydrogen storage
process can be summarized as follows
BN + H2O + e- S BN-Hads + OH-
BN-Hads S BN-Habs
2BN-Hads w 2BN + H2
BN-Hads and BN-Habs are denoted as adsorbed hydrogen
and absorbed hydrogen on BN nanotubes, respectively. Gener-
ally, a pair of adsorption and desorption peaks of hydrogen in
the electrochemical reaction in the acidic media can be observed
in Pd or Pd alloy electrodes due to the strong interaction of
hydrogen with Pd.32,33 Hydrogen desorption from the Pd
Figure 5. (a) First charge/discharge curves and (b) cycle life of purified
BN nanotube electrodes at different current densities.

Electrochemical Hydrogen Storage in BN Nanotubes
J. Phys. Chem. B, Vol. 109, No. 23, 2005 11529
TNSF (033804411), and CEM foundation for the N & T Joint
Academy, China. Financial support from the Australian Re-
search Council (ARC) is also gratefully acknowledged, and H.
Y. Zhu is indebted to the ARC for a QE II fellowship.
References and Notes
(1) Chopra, N. G.; Luyren, R. J.; Cherry, K.; Crespi, V. H.; Cohen,
M. L.; Louie, S. G.; Zettl, A. Science 1995, 269, 966.
(2) Lourie, O. R.; Jones, C. R.; Bartlett, B. M.; Gibbons, P. C.; Ruoff,
R. S.; Buhro, W. E. Chem. Mater. 2000, 12, 1808
(3) Laude, T.; Matsui, Y.; Marraud, A.; Jouffrey, B. Appl. Phys. Lett.
2000, 76, 3239.
(4) Chen, Y.; Gerald, J. F.; Willams, J. S.; Bulcock, S. Chem. Phys.
Lett. 1999, 299, 260.
(5) Bai, X. D.; Yu, J.; Liu, S.; Wang, E. G. Chem. Phys. Lett. 2000,
325, 485.
(6) Tang, C. C.; Bando, Y.; Golberg, D.; Ding, X. X.; Qi, S. R. J.
Phys. Chem. B 2003, 107, 6539.
(7) Ma, R. Z.; Bando, Y.; Sato, T. Chem. Phys. Lett. 2001, 350, 1.
(8) Bando, Y.; Ogawa, K.; Golberg, D. Chem. Phys. Lett. 2001, 347,
349.
(9) Ma, R. Z.; Bando, Y.; Sato, T.; Kurashima, K. Chem Mater. 2001,
13, 2965.
(10) Narita, I.; Oku, T. Diamond Relat. Mater. 2003, 12, 1146.
Figure 6. Cyclic voltammograms of the purified BN nanotube
electrode at different scan rates. To avoid the perturbation of hydrogen
adsorption of the metallic Ni powders, the electrodes used here were
constructed only with porous Ni substrate and without the addition of
carbonyl Ni powders.
(11) Gao, X. P.; Qin, X.; Wu, F.; Liu, H.; Lan, Y.; Fan, S. S.; Yuan, H.
T.; Song, D. Y.; Shen, P. W. Chem. Phys. Lett. 2000, 327, 271.
(12) Shaijumon, M. M.; Ramaprabhu, S. Chem. Phys. Lett. 2003, 374,
513.
(13) Schlapbach, L.; Zu¨ttel, A. Nature 2001, 414, 353.
(14) Liu, C.; Fan, Y. Y.; Liu, M.; Cong, H. T.; Cheng, H. M.;
Dresselhaus, M. S. Science 1999, 286, 1127.
(15) Pinkerton, F. E.; Wicke, B. G.; Olk, C. H.; Tibbetts, G. G.; Meisner,
electrode takes place in two mechanisms, electrochemical
oxidation and nonelectrochemical recombination. However, in
alkaline solution hydrogen desorption by nonelectrochemical
recombination is seldom observed in metal hydride electrodes.30
The Heyrovsky electrochemical oxidation process is dominant
because of the high electrocatalytic activity of metal hydride
electrodes. It is well-known that the electrocatalytic activity of
nonmetallic BN compounds in alkaline solution is relatively
poor, compared with that of metal hydride electrodes. The strong
hydrogen desorption of nonelectrochemical recombination
without charge transfer may lead to the low discharge capacity
of BN nanotubes. The improvement of the electrocatalytic
activity by surface modification with metals or alloys is expected
to enhance the electrochemical hydrogen storage capacity of
BN nanotubes, because it has been found that carbon nanotubes
decorated with metallic Ni or Pd as catalysts showed a higher
hydrogen storage capacity in the electrochemical and gas-phase
reactions.34-35
G. P.; Meyer, M. S.; Herbst, J. F. J. Phys. Chem. B 2000, 104, 9460.
(16) Hirscher, M.; Becher, M.; Haluska, M.; Quintel, A.; Skakalova,
V.; Choi, Y. M.; Dettlaff-Weglikowska, U.; Roth, S.; Stepanek, I.; Bernier,
P.; Leonhardt, A.; Fink, J. J. Alloys Compd. 2002, 330, 654.
(17) Ma, R. Z.; Bando, Y.; Zhu, H. W.; Sato, T.; Xu, C. L.; Wu, D. H.
J. Am. Chem. Soc. 2002, 124, 7672.
(18) Tang, C. C.; Bando, Y.; Ding, X. X.; Qi, S. R.; Golberg, D. J. Am.
Chem. Soc. 2002, 124, 14550.
(19) Jhi, S. H.; Kwon, Y. K. Phys. ReV. B 2004, 69, 245407.
(20) Wu, X. J.; Yang J. L.; Hou, J. G.; Zhu, Q. S. J. Chem. Phys. 2004,
121, 8481.
(21) Nu¨tzenadel, C.; Zu¨ttel, A.; Chartouni, D.; Schlapbach, L. Electro-
chem. Solid-State Lett. 1999, 2, 30.
(22) Qin, X.; Gao, X. P.; Liu, H.; Yuan, H. T.; Yan, D. Y.; Gong, W.
L.; Song, D. Y. Electrochem. Solid-State Lett. 2000, 3, 532.
(23) Rajalakshmi, N.; Dhathathreyan, K. S.; Govinraraj, A.; Satishkumar,
B. C. Electrochim. Acta 2000, 45, 4511.
IV. Conclusion
BN nanotubes were synthesized through chemical vapor
deposition using a LaNi5 alloy as a catalyst precursor at 1473
K. It was found that the obtained BN nanotubes are straight
with a diameter of 30-50 nm and a length of up to several
microns and can store hydrogen in an electrochemical process.
An initial electrochemical discharge capacity of 68 mA h g-1
was obtained in the BN nanotube electrode. The anodic
oxidation peak and cathodic adsorption peak of hydrogen were
found in the cyclic voltammograms. The hydrogen desorption
peak of nonelectrochemical recombination appeared at the range
between -0.78 and -0.85 V, which is attributed to the slow
reaction at BN nanotubes. The results reported here also suggest
the possible existence of the strong chemisorption of hydrogen,
and it may lead to the low discharge capacity of BN nanotubes.
It is tentatively concluded that the improvement of the electro-
catalytic activity by surface modification with metals or alloys
would enhance the electrochemical hydrogen storage capacity
of BN nanotubes.
(24) Lombardi, I.; Bestetti, M.; Mazzocchia, C.; Cavallotti, P. L.; Ducati,
U. Electrochem. Solid-State Lett. 2004, 7, A115.
(25) Jurewicz, K.; Frackowiak, E.; Beguin, F. Electrochem. Solid-State
Lett. 2001, 4, A27.
(26) Skowronski, J. M.; Scharff, P.; Pfander, N.; Cui, S. AdV Mater.
2003, 15, 55.
(27) Jurewicz, K.; Frackowiak, E.; Beguin, F. Appl. Phys. A 2004, 78,
981.
(28) Tang, C. C.; Bando, Y.; Sato, T. Chem. Phys. Lett. 2002, 362, 185.
(29) Yan, X. Q.; Gao, X. P.; Li, Y.; Liu, Z. Q.; Wu, F.; Shen, Y. T.;
Song, D. Y. Chem. Phys. Lett. 2003, 372, 336.
(30) Gao, X. P.; Liu, J.; Ye, S. H.; Song, D. Y.; Zhang, Y. S. J. Alloys
Compd. 1997, 253-254, 515.
(31) Wu, X. J.; Yang, J. L.; Hou, J. G.; Zhu Q. S. Phys ReV. B 2004,
69, 153411.
(32) Lukaszewski, M.; Kusmierczyk, K.; Kotowski, J.; Siwek, H.;
Czerwinski, A. J. Solid State Electrochem. 2003, 7, 69.
(33) Czerwinski, A.; Kiersztyn, I.; Grden, M. J. Solid State Electrochem.
2003, 7, 321.
(34) Gao, X. P.; Lan, Y.; Pan, G. L.; Wu, F.; Qu, J. Q.; Song, D. Y.;
Shen, P. W. Electrochem. Solid-State Lett. 2001, 4, A173.
(35) Yoo, E.; Gao, L.; Komatsu, T.; Yagai, N.; Arai, K.; Yamazaki, T.;
Matsuishi, K.; Matsumoto, T.; Nakamura, J. J. Phys. Chem. B 2004, 108,
18903.
Acknowledgment. This work was supported by the 973
Program (2002CB211800), NCET (040219), NSFC (50381037),