Low-temperature synthesis of nanosized metal borides through reaction of lithium borohydride with metal hydroxides or oxides

Article abstract of DOI:10.1016/j.jallcom.2015.08.149

In this study, we report a novel and facile synthesis approach of boron-rich transition metal borides such as LaB₆ and TiB₂ through reaction of lithium borohydride (LiBH₄) with corresponding metal hydroxide or oxide at temperatures below 600 °C. A fast endothermic reaction occurred at around 350 °C in the ball milled mixture of 6LiBH₄ + La(OH)₃ or 12LiBH₄ + La₂O₃, efficiently producing crystalline LaB₆ nanoparticles of a size smaller than 100 nm. In comparison, the reaction of LiBH₄ with TiO₂ proceeded within a wide temperature range from 120 °C to 500 °C, resulting in the formation of nanocrystalline TiB₂ of only a few nanometers. This synthesis method proved to be a facile and general route to fabricate nanosized transition metal borides.

Full text of DOI:10.1016/j.jallcom.2015.08.149

Journal of Alloys and Compounds 651 (2015) 666e672
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: http://www.elsevier.com/locate/jalcom
Low-temperature synthesis of nanosized metal borides through
reaction of lithium borohydride with metal hydroxides or oxides
,
Wei Yuan Pan a, Qian Wen Bao a, Yang Jun Mao a, Bin Hong Liu a *, Zhou Peng Li b
a Dept. of Materials Sci. & Eng., Zhejiang University, Hangzhou 310027, PR China
b Dept. of Chemical & Biological Eng., Zhejiang University, Hangzhou 310027, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
In this study, we report a novel and facile synthesis approach of boron-rich transition metal borides such
as LaB6 and TiB2 through reaction of lithium borohydride (LiBH4) with corresponding metal hydroxide or
oxide at temperatures below 600 ꢀC. A fast endothermic reaction occurred at around 350 ꢀC in the ball
milled mixture of 6LiBH4 þ La(OH)3 or 12LiBH4 þ La2O3, efficiently producing crystalline LaB6 nano-
particles of a size smaller than 100 nm. In comparison, the reaction of LiBH4 with TiO2 proceeded within
a wide temperature range from 120 ꢀC to 500 ꢀC, resulting in the formation of nanocrystalline TiB2 of
only a few nanometers. This synthesis method proved to be a facile and general route to fabricate
nanosized transition metal borides.
Received 26 July 2015
Received in revised form
17 August 2015
Accepted 18 August 2015
Available online 20 August 2015
Keywords:
Metal boride
Lithium borohydride
Hydroxide
© 2015 Elsevier B.V. All rights reserved.
Oxide
Nanostructure
1. Introduction
metal borides are being developed as functional materials.
However, synthesis of metal borides is challenging mainly due
Metal borides are a group of chemicals with unique structures
and properties, and thus are used in versatile applications [1,2]. For
example, MB2 (M ¼ Mg, Al, Ti, V, Zr, Nb etc.) has a layered structure
with alternating layers of metal and boron. MB6 (M ¼ Ca, La, Ce etc.)
has a CsCl structure in which the site of Csþ is occupied by a metal
ion, while Clꢁ is replaced by a covalently bonded B6 octahedron [1].
These boron-rich metal borides are known as refractory materials
that commonly have high hardness and melting points, high
resistance to wear and corrosion, high chemical stability, excellent
thermal and electric conductivity [1,2]. They can thus be used in
extreme environments such as for aerospace applications. In recent
years, several metal borides are found to have functional properties
[1e8]. For example, MgB2 is a superconductor with a super-
conducting temperature of 39 K [1,2]. LaB6 and CeB6 are excellent
field emission materials used in scanning and transmission electron
microscopes [1,2]. VB2 or TiB2 is found with extremely high elec-
trochemical capacities around 4500 mAh gꢁ1 in a VB2 (TiB2)-air
battery [3,4]. TiB2 is also used as a catalyst support in proton ex-
change membrane fuel cell (PEMFC) [5]. Therefore, more and more
to the lack of reactive elemental boron sources [1,2] They are usu-
ally synthesized using elemental materials of metals and boron by
hot pressing, spark plasma sintering, fused salt electrolysis, self-
propagating high-temperature synthesis, chemical vapor deposi-
tion, etc. Due to the very high melting point of 2075 ꢀC and
chemical inertness of elemental boron, the fabrication tempera-
tures of metal borides are usually higher than 1100 ꢀC [1]. High
synthesis temperatures of metal borides not only lead to high en-
ergy consumption, but also induce pollution from crucible, raw
materials as well as atmosphere. Moreover, metal borides prepared
at high temperatures suffer from poor size control. On the other
hand, nanosized powders with homogeneous composition and
narrowly distributed particle size are in demand to improve their
sinterability.
Considering more and more important roles played by metal
borides and relatively less knowledge of their synthesis, consider-
able research effort is being made to develop new, facile and gen-
eral synthesis routes for metal borides [9e16]. For example, a
general solution route to synthesize metal boride nanocrystals has
recently been developed using metal chlorides and sodium bor-
ohydrdie (NaBH4) as the reactants [9]. A eutectic LiCl/KCl salt
mixture was used to initiate the reaction in the liquid state. Using
this synthesis process, nanocrystals of CaB6 and NbB2 as examples
* Corresponding author.
E-mail address: liubh@zju.edu.cn (B.H. Liu).
http://dx.doi.org/10.1016/j.jallcom.2015.08.149
0925-8388/© 2015 Elsevier B.V. All rights reserved.

W.Y. Pan et al. / Journal of Alloys and Compounds 651 (2015) 666e672
667
have been obtained under a synthesis temperature of 800e900 ꢀC.
In another research work, nano powders of TiB2 and VB2 have been
mechanochemically synthesized by using LiBH4, LiH and clorides
such as TiCl3 or VCl3 [10]. Nanosized LaB6 crystals have been
fabricated at 400 ꢀC starting from Mg powder, NaBH4 and LaCl3
[14].
In this study, we introduce a facile and general synthesis
method of transition metal borides such as LaB6 and TiB2 at tem-
peratures below 600 ꢀC. LiBH4 was used as the boron source as well
as the reductive agent. It reacted with metal hydroxide such as
La(OH)3 or oxides such as La2O3 or TiO2 to fabricate nanosized
crystalline LaB6 or TiB2. The results shown in this study provide a
new, facile and general synthesis route for nanosized and crystal-
line transition metal borides, thus shedding light on developments
of novel synthesis approaches and performance enhancements of
these metal borides.
Fig. 1. DSC curve and hydrogen release behavior during heating the ball milled
mixture of 6LiBH4 þ La(OH)3.
2. Experimental details
hydrogen release can be observed in Fig. 1: it first started at about
70 ꢀC and then speeded up at around 170 ꢀC, but paused at 260 ꢀC.
Then at around 350 ꢀC, a large amount of hydrogen was released
significantly within a narrow temperature range, implying a fast
reaction. After this reaction, almost no hydrogen gas was further
emitted even though the reactor was subsequently heated to
600 ꢀC.
2.1. Materials
The chemicals and raw materials used in this study were
commercially purchased as follows: LiBH4 (95%, Acros), La(OH)3
(99.9%, Aladdin), La2O3 (99.99%, Aladdin), TiO2 (99.8%, Aladdin).
2.2. Synthesis process of metal borides
The DSC curve of the sample heated from room temperature to
500 ꢀC is also presented in Fig. 1. During heating, two small
exothermic peaks were observed at 82.6 ꢀC and 252.3 ꢀC respec-
tively. Also two endothermic peaks at 117.5 ꢀC and 277.1 ꢀC were
detected, which are close to the structural change and the melting
temperatures of LiBH4. These two endothermic peaks are weak
compared to those of pure LiBH4, implying that only a small amount
of LiBH4 was left at this stage. Then a large and sharp endothermic
peak appeared at 351.5 ꢀC, which is coincident with the significant
hydrogen release at this temperature. Fig. 1 demonstrates that the
reaction was accomplished within a very narrow temperature
range and accompanied with a significant endothermic effect and
considerable hydrogen release.
The mixtures of LiBH4 with La(OH)3, La2O3 or TiO2 in stoichio-
metric molar ratios were first ball milled at 400 rpm for 16 h in a
planetary mill. The stainless steel vessel for ball milling was 100 ml
and the weight ratio of ball to sample was 180:1. The ball milled
mixture was then introduced into a stainless steel reactor. The
handling of ball milling samples was undergone in a glove box filled
with high purity argon gas in which the levels of H2O and O2 were
kept below 1 ppm. The sealed reactor was then set up on a Sievert's
apparatus. The reactor was first evacuated and then heated from
room temperature to 600 ꢀC at 2 ꢀC minꢁ1. The pressure in the
system was recorded during the experiments. The amount of gas
release was then determined according to the pressure rise in the
reactor. The final pressure in the reactor was lower than 0.15 MPa.
The ex situ XRD analysis results of the mixtures heated to
different temperatures are shown in Fig. 2. Several new phases such
as LiBO2, La2O3 and B20H26O were detected in the ball milled
mixture, indicating that some reactions occurred during the ball
milling treatment. Both La2O3 and B20H26O seemed to remain until
300 ꢀC while LiBO2 was transformed. However, all these phases
were not detected anymore in the sample heated to 400 ꢀC, sug-
gesting that they were only intermediates. Perfect diffraction pat-
terns of LaB6 were exhibited by the samples heated to 400, 500 and
600 ꢀC, as shown in Fig. 2. In comparison, only extremely weak
peaks of other substances such as Li2O were detected in these
samples. Based on the results shown in Figs. 1 and 2, it can be
deduced that the significant reaction at 351.5 ꢀC produced crys-
talline LaB6. The further heating to 600 ꢀC only induced slight
growth of LaB6 grains because the XRD peaks became somewhat
stronger and narrower at higher temperatures. According to the
Scherrer equation, the crystallite sizes of LaB6 were calculated to be
23.3, 23.5, 26.0 nm at 400, 500 and 600 ꢀC respectively. The cell
parameters of LaB6 obtained at 400, 500 and 600 ꢀC were
a ¼ 4.1525, 4.1531 and 4.1443 Å, slightly smaller than the value of
a ¼ 4.157 Å given in the JCPDS card (PDF #65-1831).
2.3. Instrumental characterizations and analyses
X-ray diffraction (XRD) analyses were performed on a PAN-
alytical X'Pert PRO using the CuKa radiation. A special sample stand
was designed to protect the samples from air exposure during XRD
measurements. Differential Scanning Calorimetry (DSC) analyses
were conducted on a Netzsch STA449F3 at a heating rate of
5
ꢀC minꢁ1 under argon atmosphere with
a flow rate of
50 ml minꢁ1. Field emission scanning electron microscopy (FE-
SEM) observations were carried out on a Hitachi SU-70 microscope.
The samples were protected from air exposure by Ar gas blowing
during transferring to the sample chamber. High resolution trans-
mission electron microscopy (HRTEM) observations were per-
formed on Tecnai G2 F30 S-Twin, Philips-FEI.
3. Results
3.1. LaB6 synthesis through reaction of LiBH4 with La(OH)3 or La2O3
The SEM photos in Fig. 3 demonstrate the morphologies of the
products at three temperatures. Nearly spherical particles with a
homogeneous size of less than 100 nm were achieved in all the
samples. As shown in Fig. 3, only slight particle growth occurred in
the samples with the increase of the temperature, which is in
After ball milling of 6LiBH4 þ La(OH)3 at 400 rpm for 16 h, the
mixture was transferred into a reactor under Ar atmosphere in a
glovebox. Fig. 1 shows hydrogen release of the mixture during
heating from room temperature to 600 ꢀC, which is obtained by
assuming that the emitted gas was only hydrogen. Two stages of

668
W.Y. Pan et al. / Journal of Alloys and Compounds 651 (2015) 666e672
Fig. 2. XRD patterns of the 6LiBH4 þ La(OH)3 mixture at different reaction stages.
Fig. 3. SEM photos of the reaction products for 6LiBH4 þ La(OH)3 heated to 400 ꢀC, 500 ꢀC and 600 ꢀC respectively. (a) and (b): 400 ꢀC; (c) and (d): 500 ꢀC; (e) and (f): 600 ꢀC.

W.Y. Pan et al. / Journal of Alloys and Compounds 651 (2015) 666e672
669
agreement with the XRD results.
3.2. TiB2 synthesis from LiBH4 and TiO2
The HRTEM observation shown in Fig. 4 reveals more clearly the
morphology and structure of the formed LaB6 nanoparticles. In
Fig. 4(a), spherical particles smaller than 100 nm with a low degree
of aggregation are well observed. Moreover, most of these particles
are single crystals as shown by Fig. 4(b and d). The selected area
electron diffraction (SAED) pattern shown in Fig. 4(d) reveals that
the particle shown in Fig. 4(c) is a single crystal of LaB6.
For comparison, La2O3 instead of La(OH)3 was also used as the
precursor to synthesize LaB6. As shown in Fig. 5, the ball milled
mixture of 12LiBH4 þ La2O3 had a significant reaction at around
350 ꢀC, resembling that of 6LiBH4 þ La(OH)3. Their differences in
hydrogen release behavior were that 6LiBH4 þ La(OH)3 released
more hydrogen below 350 ꢀC, while 12LiBH4 þ La2O3 still emitted
hydrogen slightly at temperatures higher than 350 ꢀC. The DSC
curves of 12LiBH4 þ La2O3 and 6LiBH4 þ La(OH)3 are similar to each
other in that both of them show a significant endothermic effect at
356.6 and 351.1 ꢀC respectively. The XRD result of 12LiBH4 þ La2O3
heated to 600 ꢀC displays a predominant pattern of LaB6 with only
minor peaks from Li3BO3 and LiOH. It suggests that both
12LiBH4 þ La2O3 and 6LiBH4 þ La(OH)3 produced crystalline LaB6
through a similar reaction route.
Fig. 6 reveals the hydrogen release behavior of the ball milled
2LiBH4 þ TiO2 mixture. It can be seen that the release of hydrogen
continued from 120 ꢀC to 500 ꢀC, suggesting that the reaction be-
tween LiBH4 and TiO2 proceeded within a wide temperature range.
Probably due to the continuous reaction, the DSC curve shown in
Fig. 6 demonstrates an overall endothermic effect during the whole
heating process. One apparent endothermic peak at 128.5 ꢀC, which
is close to the structural change temperature of LiBH4, was
observed. In addition, two small and wide endothermic peaks at
267.1 ꢀC and 289.4 ꢀC respectively were also detected.
The results of XRD measurements shown in Fig. 7 clearly
demonstrate a gradual transformation from TiO2 to TiB2. After ball
milling, the mixture only exhibited the peaks of Anatase-TiO2
without peaks from LiBH4, most probably because LiBH4 became
amorphous during ball milling. After heating, some new phases
including LiH appeared. Moreover, the peak strengths of these new
phases decreased with the increase of temperature, suggesting that
they are reaction intermediates. Inversely, the peaks from TiB2
became apparent as the temperature was increased up to 500 ꢀC.
The broad peaks of TiB2 also indicate its nanosized structure. The
Fig. 4. HRTEM pictures of the reaction product for 6LiBH4 þ La(OH)3 heated to 600 ꢀC. (a)e(c): morphologies of LaB6 particles; (d) SAED pattern of the particle shown in (c).

670
W.Y. Pan et al. / Journal of Alloys and Compounds 651 (2015) 666e672
Fig. 6. DSC and hydrogen release curves of the ball milled 2LiBH4 þ TiO2 mixture.
occurred around 350 ꢀC with a significant endothermic effect. The
fast reaction resulted in a complete transformation into LaB6
nanoparticles of good crystallinity. Also the low reaction temper-
ature ensured a homogeneous nanosize of smaller than 100 nm.
Though there were compounds such as LiBO2, La2O3, B20H26
O
detected after ball milling or heating, it seems that they were only
reaction intermediates because they were not present in the final
products. As LaB6 and Li2O were detected in the final product, we
thus suppose that the reaction can be written as follows:
6LiBH4 þ La(OH)3 / LaB6 þ 3Li2O þ 13.5H2
(1)
Based on the above equation, about 8.4 wt% hydrogen will be
released with respect to the total weight of 6LiBH4 þ La(OH)3. In our
experiment, about 7.8 wt% hydrogen was obtained, which is close to
the value predicted by the reaction (1).
Both 12LiBH4 þ La2O3 and 6LiBH4 þ La(OH)3 mixtures revealed
the similar endothermic reaction at around 350 ꢀC. It is thus sup-
posed that they had analogous reaction pathway to produce crys-
talline LaB6. However, the Li-containing by-products may be
different for both mixtures in consideration of chemical balance.
Determination of the exact side products other than LaB6 for
12LiBH4 þ La2O3 needs further research work.
As LaB6 is not soluble in water, it is thus possible to separate it
from other reaction by-products such as Li2O or LiOH that are sol-
uble in water. Therefore, pure LaB6 can be achieved simply through
water-washing the solid reaction product.
In contrast to a fast reaction occurring within a narrow tem-
perature range for the mixture of LiBH4 with La(OH)3 or La2O3, the
reaction of TiO2 with LiBH4 proceeded within a much wider tem-
perature range from 120 ꢀC to 500 ꢀC. Based on the XRD results, the
reaction is supposed to be as follows:
Fig. 5. Comparison between 12LiBH4 þ La2O3 and 6LiBH4 þ La(OH)3. (a) hydrogen
release behavior; (b) DSC curves; (C) XRD result of 12LiBH4 þ La2O3 heated to 600 ꢀC.
SEM and HRTEM pictures in Fig. 8 clearly display morphologies and
nanostructures of the products at 400 ꢀC and 600 ꢀC respectively.
Comparison of two SEM images reveals that the particles became
more coalescent at 600 ꢀC. From HRTEM pictures, nanocrystalline
TiB2 domains of only a few nanometers could be observed in the
final product. However, it seems that the present reaction condition
did not ensure a complete transformation from the intermediate
into TiB2. Higher temperature or longer reaction time is thus
required.
2LiBH4 þ TiO2 / TiB2 þ 2LiOH þ 3H2
(2)
In the above equation, LiOH is deduced as a product mainly
based on the chemical balance though it was not directly detected.
According to the above equation, 4.9 wt% hydrogen will be released
during the reaction. About 4.5 wt% hydrogen was actually obtained
in our experiment, thus partially confirming the above reaction
mechanism.
The slow and gradual reaction between LiBH4 and TiO2 is
beneficial for the formation of nanosized TiB2. However, it also
hinders the complete transformation into TiB2, thus decreasing the
conversion efficiency. As LiBH4 or LiOH is soluble in water, it is easy
to separate them from Ti-containing compounds through water
washing. But if the Ti-containing intermediate is not completely
4. Discussion
In this study, nanosized LaB6 was successfully synthesized
through reaction of LiBH4 with La(OH)3 or La2O3. The main reaction

W.Y. Pan et al. / Journal of Alloys and Compounds 651 (2015) 666e672
671
Fig. 7. XRD patterns of the 2LiBH4 þ TiO2 mixture heated to different temperatures.
Fig. 8. SEM and HRTEM photos of the reaction product for 2LiBH4 þ TiO2. (a): SEM, heated to 400 ꢀC; (b): SEM, heated to 600 ꢀC; (c) and (d): HRTEM, heated to 600 ꢀC.
transformed into TiB2, it would be difficult to separate it from TiB2.
Therefore, it is highly required to achieve complete transformation
through optimizing reaction conditions.
also successfully fabricated nanosized ZrB2 through reaction of
LiBH4 with ZrO2 below 600 ꢀC. However, borides of main group
metals such as MgB2 could not be obtained by heating the ball
In our experiment, we found that the reaction of LiBH4 with
oxides or hydroxides can be extended to synthesize other transition
metal borides. For example, this method may also be applied to the
synthesis of other rare earth metal hexaborides such as CeB6. We
milled mixture of 2LiBH4
þ
MgO. In this case, the self-
decomposition of LiBH4 occurred while MgO remained unreacted.
Compared with previously reported synthetic routes of metal
borides, the present synthesis method is superior to the others in

672
W.Y. Pan et al. / Journal of Alloys and Compounds 651 (2015) 666e672
following aspects: First, LiBH4 was employed as the reactive boron
source as well as the reductive agent, thus initiating the reaction at
lower temperatures. Also its low melting temperature of 288 ꢀC
enabled the reaction to proceed in liquidesolid states that would
enhance the reaction kinetics. Second, metal hydroxides or oxides
instead of metal halides were used as the precursors in this study,
resulting in endothermic synthetic reactions of metal borides. This
is the main distinction between the present method and the others
reported previously. Most of the reported synthetic reactions for
metal borides are exothermic, thus suffering from poor control of
reaction temperature and then particle size [1]. In comparison, the
endothermic nature of the synthesis reactions presented in this
study allowed for a precise control of reaction temperature and the
resulting nanostructures including crystallinity, particle shape and
size. Third, the synthesis method used in this study proved to be a
facile and general route to fabricate nanosized transition metal
borides including MB2 and MB6 structures under mild conditions. It
is easy to achieve not only individual metal borides but also their
composites in nanosizes through this method. For example, we
have successfully synthesized a composite of LaB6 and ZrB2 through
simple one-pot reaction, which is an important boride composite
for high temperature applications.
In comparison, the reaction of LiBH4 and TiO2 took place within a
wide temperature range from 120 ꢀC to 500 ꢀC. The slow and
gradual reaction produced TiB2 nanoparticles of only a few nano-
meters. The successful synthesis of these borides suggests that this
approach is a facile and general synthesis route for transition metal
borides.
Acknowledgments
This work is financially supported by the National Natural Sci-
ence Foundation of China under grant Nos. 51271164, 21276229 and
21476200.
References
[1] S. Carenco, D. Portehault, C. Boissiere, N. Mezailles, C. Sanchez, Chem. Rev. 113
(2013) 7981e8065.
[2] B. Albert, H. Hillebrecht, Angew. Chem. Int. Ed. 48 (2009) 8640e8668.
[3] J. Stuart, A. Hohenadel, X. Li, H. Xiao, J. Parkey, C.P. Rhodes, S. Licht,
J. Electrochem. Soc. 162 (2015) A192eA197.
[4] J. Stuart, M. Lefler, C.P. Rhodes, S. Licht, J. Electrochem. Soc. 162 (2015)
A432eA436.
[5] Z. Huang, R. Lin, R. Fan, Q. Fan, J. Ma, Electrochim. Acta. 139 (2014) 48e53.
[6] J. Liu, S. He, C. Li, F. Wang, M. Wei, D.G. Evans, X. D, J. Mater. Chem. A 2 (2014)
7570e7577.
5. Conclusions
[7] B.H. Liu, Q. Li, Int. J. Hydrog. Energy 33 (2008) 7385e7391.
[8] S. Rades, S. Kraemer, R. Seshadri, B. Albert, Chem. Mater. 26 (2014)
1549e1552.
In this work, nanosized transition metal borides such as LaB6
and TiB2 were successfully synthesized through reaction of LiBH4
with the corresponding hydroxides or oxides at temperatures
below 600 ꢀC. In the ball milled mixture of 6LiBH4 þ La(OH)3 or
12LiBH4 þ La2O3, a fast endothermic reaction at around 350 ꢀC
produced crystalline LaB6 nanoparticles of a size smaller than
100 nm. Though some intermediates were formed during the re-
action, LaB6 was found as the only boride in the reaction product
and can be easily separated from other water soluble side products.
[9] D. Portehault, S. Devi, P. Beaunier, C. Gervais, C. Giordano, C. Sanchez,
M. Antonietti, Angew. Chem. Int. Ed. 50 (2011) 3262e3265.
[10] J.W. Kim, J.H. Shim, J.P. Ahn, Y.W. Cho, J.H. Kim, K.H. Oh, Mater. Lett. 62 (2008)
2461e2464.
[11] Y. Yan, Z. Huang, S. Dong, D. Jiang, J. Am. Ceram. Soc. 89 (2006) 3585e3588.
[12] H.E. Camurlu, F. Maglia, J. Eur. Ceram. Soc. 29 (2009) 1501e1506.
[13] D. Lee, G.D. Sim, K. Xiao, J.J. Vlassak, J. Phys. Chem. C 118 (2014) 21192e21198.
[14] M. Zhang, L. Yuan, X. Wang, H. Fan, X. Wang, X. Wu, H. Wang, Y. Qian, J. Solid
State Chem. 181 (2008) 294e297.
[15] L. Wang, L. Xu, Z. Ju, Y. Qian, CrystEngComm 12 (2010) 3923e3928.
[16] J.P. Kelly, R. Kanakala, O.A. Graeve, J. Am. Ceram. Soc. 93 (2010) 3035e3038.