In situ formation and rapid decomposition of Ti (BH4) 3 by mechanical milling LiBH4 with TiF3

Article abstract of DOI:10.1063/1.3076106

Mechanically milled 3 LiBH4 / TiF3 mixture can rapidly release over 5 wt % hydrogen at moderate temperatures (70-90 °C) without undesired gas impurity. Structure analyses results show that the favorable dehydrogenation performance of the material should b

Full text of DOI:10.1063/1.3076106

In situ formation and rapid decomposition of
with
by mechanical milling
Z. Z. Fang, L. P. Ma, X. D. Kang, P. J. Wang, P. Wang, and H. M. Cheng
Citation: Appl. Phys. Lett. 94, 044104 (2009); doi: 10.1063/1.3076106
View online: http://dx.doi.org/10.1063/1.3076106
View Table of Contents: http://aip.scitation.org/toc/apl/94/4
Published by the American Institute of Physics

APPLIED PHYSICS LETTERS 94, 044104 ͑2009͒
In situ formation and rapid decomposition of Ti„BH … by mechanical
4
3
milling LiBH with TiF
4
3
a͒
Z. Z. Fang, L. P. Ma, X. D. Kang, P. J. Wang, P. Wang, and H. M. Cheng
Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy
of Sciences, Shenyang 110016, China
͑
Received 23 October 2008; accepted 7 January 2009; published online 27 January 2009͒
Mechanically milled 3LiBH /TiF mixture can rapidly release over 5 wt % hydrogen at moderate
4
3
temperatures ͑70–90 °C͒ without undesired gas impurity. Structure analyses results show that the
favorable dehydrogenation performance of the material should be associated with the in situ
formation and rapid decomposition of an intermediate titanium borohydride ͑Ti͑BH ͒ ͒. These
4
3
findings demonstrated a viable chemical activation approach for promoting hydrogen release from
the thermodynamically stable borohydrides. Particularly, it shows the potential of transition metal
borohydrides for hydrogen storage applications. © 2009 American Institute of Physics.
͓
DOI: 10.1063/1.3076106͔
Lithium borohydride ͑LiBH ͒ has received considerable
with eight steel balls ͑10 mm in diameter͒. The ball-to-
powder ratio was around 40:1. All sample operations were
performed in an Ar-filled glovebox.
4
interest as potential hydrogen storage medium due to its ex-
1
–4
tremely high gravimetric hydrogen density ͑18.4 wt %͒.
5
–7
Several approaches ͑reactant destabilization,
nanoporous
The hydrogen release from the postmilled sample was
examined by using both synchronous thermal analyses
͓thermogravimetry/differential scanning calorimetry/mass
spectroscopy ͑TG/DSC/MS͒, Netzsch 449C Jupiter, and
QMS 403C͔ and Sievert-type volumetric apparatus. In the
thermal analyses, the sample was heated to 130 °C at a
ramping rate of 2 °C/min under a flowing Ar atmosphere
͑99.999% purity͒. In the isothermal measurements using
volumetric apparatus, the oil bath was preheated to a desig-
nated temperature ranging from 70 to 90 °C, and then the
measurement was started immediately after immersing the
reactor containing samples ͑with a typical amount of around
100 mg͒ into the oil bath.
8
,9
10–12
scaffolds incorporation, and cation/anion substitution
͒
have been recently developed to address its kinetic and ther-
modynamic limitations that are essentially imposed by the
1
–4
strong and highly directional covalent and/or ionic bonds.
With the aid of these technical advances, several material
systems that possess superior hydrogen storage property to
5
–12
the parent LiBH have been identified.
For example, me-
4
chanically milling LiBH with 1/2MgH produces a Li–Mg–
4
2
5
,6
B–H system that exhibits favorable reversibility. Incorpo-
ration of LiBH4 into nanoporous scaffolds gives rise to
significant improvements on both kinetic and thermody-
8
,9
namic aspects. Quite recently, considerable efforts were
further extended to alkaline earth and transition metal boro-
hydrides, aiming at obtaining a more favorable combination
of hydrogen capacity, kinetics, and reversibility relative to
The samples were characterized by x-ray diffraction
͑XRD, Rigaku D/MAX-2500, Cu K␣ radiation͒, micro-
Raman spectrometer ͑JobinYvon HR800, 632.8 nm-laser͒,
x-ray photoelectron spectroscopy ͑XPS, ESCALAB 250,
Al K␣ radiation͒, and Fourier transform infrared spectros-
1
3–15
LiBH4.
However in a general view, no known metal
borohydrides and related systems could fulfill reversible de-
hydrogenation within the temperature range acceptable for
on-board hydrogen sources ͑−40–85 °C͒.
−
1
copy ͑FTIR, Bruker TENSOR 27, MCT detector, 4 cm
resolution͒. All the sample preparations were performed in
the Ar-filled glovebox. The in situ FTIR measurements were
carried out under Ar atmosphere ͑99.999% purity͒ in tem-
peratures ranging from room temperature to 90 °C at a
ramping rate of 1 °C/min. FTIR spectra were collected in
diffuse reflectance infrared Fourier transform mode.
We herein report our latest experimental finding regard-
ing chemical activation of LiBH for promoted hydrogen re-
4
lease. It was found that the mechanically milled
3
LiBH /TiF mixture can rapidly release 5.7 wt % hydro-
4
3
gen at moderate temperatures without undesired gas impuri-
ties. Furthermore, according to the structure analyses results,
the favorable dehydrogenation performance of the material
should be associated with the in situ formation and rapid
decomposition of the intermediate titanium borohydride
The LiBH /TiF mixture in a 3:1 molar ratio was me-
4
3
chanically milled under Ar atmosphere. The phase/structure
analyses ͑Fig. 1͒ of the postmilled sample show that both
LiBH and TiF well maintain their phase stability during the
4
3
͑
Ti͑BH ͒ ͒.
4 3
milling process. This is consistent with the observation that
no appreciable pressure buildup in the milling vessel was
detected after the intensive milling of 3LiBH /TiF mixture.
The starting materials of LiBH ͑95% purity͒, TiF , and
4
3
TiCl ͑99.999% purity͒ powders were all purchased from
3
4
3
Sigma-Aldrich Corp. and used as received. LiBH powder
Interestingly, a comparative study found that TiCl additive
4
3
was mechanically milled with TiCl or TiF in a 3:1 molar
behaves quite differently from its analog TiF in terms of
3
3
3
ratio for 2 h under argon ͑Ar͒ atmosphere by using a Fritsch
planetary mill at 500 rpm in a stainless steel vial together
reactivity toward LiBH . XRD analysis ͑Fig. 1͒ of the post-
4
7
milled 3LiBH /TiCl sample showed sharp and intense dif-
4
3
fraction peaks of LiCl. XPS examinations ͑the result not
shown here͒ evidenced the formation of elemental B and
a͒Author to whom correspondence should be addressed. Electronic mail:
pingwang@imr.ac.cn. FAX: ϩ86 24 2389 1320.
TiH in the milling process. These results clearly indicate
2
0003-6951/2009/94͑4͒/044104/3/$25.00
94, 044104-1
© 2009 American Institute of Physics

044104-2
Fang et al.
Appl. Phys. Lett. 94, 044104 ͑2009͒
FIG. 1. XRD patterns ͑left panel͒ and Raman spectra ͑right panel͒ of ͑a͒
postmilled 3LiBH /TiF , ͑b͒ postmilled 3LiBH /TiCl , and ͑c͒ postheated
4
3
4
3
3
LiBH /TiF samples. For comparison, the Raman spectrum of pure LiBH
4 3 4
FIG. 3. The isothermal dehydrogenation profiles of the postmilled
was also presented.
3
LiBH /TiF sample at different temperatures. The inset shows the Arrhen-
4 3
ius profile of the dehydrogenation kinetics of the postmilled sample. The
average dehydrogenation rate was determined using the data between 1 and
that LiBH had completely reacted with TiCl following Eq.
4
3
4
wt % hydrogen desorption amount ͑as shown in dashed lines͒.
͑1͒ in the milling process,
3
LiBH + TiCl → 3LiCl + 3B + TiH + 5H .
͑1͒
4
3
2
2
release over 4 wt % hydrogen within about 30 min at 70 °C.
This kinetic performance can meet the practical requirements
for on-board hydrogen sources on the operation temperature
From a thermochemical point of view, similar reaction is
anticipated to occur in the 3LiBH /TiF sample. This is sub-
sequently experimentally confirmed. However unlike TiCl ,
4
3
1
6,17
3
and flow rate.
perature to 90 °C, the released hydrogen amount increases to
.6 wt %. This value agrees well with the results obtained
Upon increasing the dehydrogenation tem-
TiF reacts with LiBH in a more controllable way.
3
4
Figure 2 presents the synchronous TG/DSC/MS results
of the postmilled 3LiBH /TiF sample. Notably, the addition
of TiF results in a substantially lowered dehydrogenation
temperature from around 400 °C for pure LiBH to moder-
ate temperatures. As seen from the TG profile, the weight
loss of the sample starts around 80 °C and ends at approxi-
mately 90 °C. As no diborane ͑B H ͒ and other gas impuri-
ties were detected throughout the heating process, the total
weight loss ͑5.7 wt %͒ can be safely attributed to the hydro-
gen release from the sample. According to the combined
analyses of the DSC/TG results, the dehydrogenation en-
thalpy of the 3LiBH /TiF sample is determined to be exo-
thermic by 13 kJ/mol H . This is in great contrast to the
endothermic enthalpy of ϳ67 kJ/mol H for pure LiBH ,
indicative of the pronounced destabilizing effect of TiF .
5
4
3
from TG measurements, thus further confirming the high-
purity of the released hydrogen. Arrhenius treatment of the
temperature-dependent rate data ͑inset of Fig. 3͒ yields an
apparent activation energy of ϳ19 kJ/mol for hydrogen re-
lease from the 3LiBH /TiF sample. This is substantially
3
4
1
4
3
2
6
lower than the values that were experimentally or theoreti-
1
,18
cally determined for pure LiBH4.
Combined XRD/Raman spectroscopy analyses ͑Fig. 1͒
and XPS examinations ͑the result not shown here͒ of the
postheated sample showed that all the starting materials,
LiBH and TiF , transformed into LiF, elemental B, and
4
3
4
3
2
TiH . These results clearly indicate that TiF reacts with the
2
3
5
2
4
host LiBH in a similar way to TiCl . However compared to
4 3
3
Eq. ͑1͒, the thermodynamic driving force of the reaction be-
The dehydrogenation property of the postmilled
LiBH /TiF sample was further examined by using volu-
metric method. As seen in Fig. 3, the sample could rapidly
tween LiBH and TiF is much smaller, Ϫ123.0 versus
4
3
1
,19
3
Ϫ216.4 kJ/mol.
This may, at least partially, account for
4
3
the observed difference between TiF and TiCl on the reac-
3 3
tivity toward LiBH . Clearly, the moderate reactivity of TiF
4
3
toward LiBH lays an important foundation for controllable
4
hydrogen release from the milled 3LiBH /TiF sample.
4
3
An in situ FTIR study was performed to further the
mechanistic understanding of the hydrogen release process
of 3LiBH /TiF sample. As shown in Fig. 4, the character-
4
3
−
1
istic absorption peaks of LiBH at 1050–1300 cm and
4
1
2
7
150–2400 cm− gradually weakened at temperatures over
0 °C and largely disappeared at 84 °C. Meanwhile, two
adsorption bands were observed to emerge at
1
500–1650 cm− and 2450–2650 cm ͑highlighted by
1
−1
rectangles͒ at around 70 °C. With increasing temperature,
these bands initially gained increased intensity, followed by
gradual intensity decrease, and finally disappeared at 90 °C.
A survey of the open literatures found that these IR bands are
close to those observed in group IVB transition metal boro-
FIG. 2. TG/DSC profiles ͑upper panel͒ of the postmilled 3LiBH /TiF
4
3
sample and synchronous MS profiles ͑lower panel͒ of m/e=2͑H ͒ and
2
2
0
m/e=27͑B H ͒. The ramping rate was 2 °C/min.
hydrides ͓e.g., Zr͑BH4͒4 and Hf͑BH4͒4͔. This, together
2
6

044104-3
Fang et al.
Appl. Phys. Lett. 94, 044104 ͑2009͒
ably modified thermodynamics may lay foundation not only
for reversible dehydrogenation at moderate temperatures but
also for the removal of inactive by-product using wet chemi-
cal method.
In summary, the thermodynamically stable LiBH can be
4
effectively destabilized by milling with TiF in a 3:1 molar
3
ratio. Associated with the in situ formation of Ti͑BH ͒ fol-
4
3
lowed by its rapid decomposition, the sample could rapidly
release over 5 wt % hydrogen at 70–90 °C without undes-
ired gas impurity. These findings demonstrated a viable
chemical activation approach for promoting hydrogen re-
lease from borohydride. Additionally, it shows the potential
of transition metal borohydrides for hydrogen storage
applications.
The financial support from the Hundred Talents Project
of the Chinese Academy of Sciences, the National Natural
Science Foundation of China ͑Grant Nos. 50571099,
FIG. 4. In situ FTIR spectra of the 3LiBH /TiF sample.
4
3
with the coincidence between the appearance of the IR bands
5
0671107, 50771094, and 50801059͒, and the special
and the consumption of LiBH , strongly suggests that an-
4
funding for CAS President Prize winner are gratefully
acknowledged.
other borohydride was generated via simple substitution of
BH4− group for halide ion. In view of the reaction stoichio-
metricity ͑LiBH :TiF =3:1͒, we speculated that the in situ
4
3
1
A. Züttel, S. Rentsch, P. Fischer, P. Wenger, P. Sudan, P. Mauron, and C.
formed species is Ti͑BH ͒ , and accordingly, the hydrogen
4
3
Emmenegger, J. Alloys Compd. 356–357, 515 ͑2003͒.
S. Orimo, Y. Nakamori, J. R. Eliseo, A. Züttel, and C. M. Jensen, Chem.
2
release process of 3LiBH /TiF mixture is described as
4
3
Rev. ͑Washington, D.C.͒ 107, 4111 ͑2007͒.
Eq. ͑2͒,
3
A. Züttel, A. Borgschulte, and S. Orimo, Scr. Mater. 56, 823 ͑2007͒.
4
3
LiBH + TiF → 3LiF + Ti͑BH ͒
P. Wang and X. D. Kang, Dalton Trans. 40, 5400 ͑2008͒.
J. J. Vajo, S. L. Skeith, and F. Mertens, J. Phys. Chem. B 109, 3719
4
3
4 3
5
→
3LiF + 3B + TiH + 5H .
͑2͒
2
2
͑2005͒.
6
U. Bösenberg, S. Doppiu, L. Mosegaard, G. Barkhordarian, N. Eigen, A.
This speculation is further supported by the facts or ob-
Borgschulte, T. R. Jensen, Y. Cerenius, O. Gutfleisch, T. Klassen, M.
servations as follows. ͑1͒ The synthesis of Ti͑BH ͒ via re-
Dornheim, and R. Bormann, Acta Mater. 55, 3951 ͑2007͒.
4
3
7
F. E. Pinkerton and M. S. Meyer, J. Alloys Compd. 464, L1 ͑2008͒.
action of LiBH with vaporized TiCl had been experimen-
4
4
8
2
1
A. F. Gross, J. J. Vajo, S. L. Van Atta, and G. L. Olson, J. Phys. Chem. C
tally demonstrated. ͑2͒ Careful examinations of the TG/
MS/FTIR results found that the start/end of hydrogen release
from the sample corresponds well with the in situ formation
and disappearance of the intermediate borohydride, respec-
tively. ͑3͒ The observed near-room-temperature decomposi-
tion behavior of the intermediate borohydride agrees well
112, 5651 ͑2008͒.
9
Z. Z. Fang, P. Wang, T. E. Rufford, X. D. Kang, G. Q. Lu, and H. M.
Cheng, Acta Mater. 56, 6257 ͑2008͒.
S. Orimo, Y. Nakamori, G. Kitahara, K. Miwa, N. Ohba, S. Towata, and
A. Züttel, J. Alloys Compd. 404–406, 427 ͑2005͒.
1
0
11
K. Miwa, N. Ohba, S. Towata, Y. Nakamori, and S. Orimo, J. Alloys
Compd. 404–406, 140 ͑2005͒.
12
2
1
with the established thermal stability of Ti͑BH ͒ .
4
3
L. C. Yin, P. Wang, Z. Z. Fang, and H. M. Cheng, Chem. Phys. Lett. 450,
318 ͑2008͒.
3
The reaction between TiF and thermally stable LiBH
3
4
1
results in the formation of thermally instable Ti͑BH ͒ . Here,
H. W. Li, K. Kikuchi, Y. Nakamori, N. Ohba, K. Miwa, S. Towata, and S.
Orimo, Acta Mater. 56, 1342 ͑2008͒.
4
3
the moderate reactivity of TiF toward LiBH is essentially
3
4
14
15
E. Rönnebro and E. H. Majzoub, J. Phys. Chem. B 111, 12045 ͑2007͒.
Y. Nakamori, H. W. Li, K. Kikuchi, M. Aoki, K. Miwa, S. Towata, and S.
Orimo, J. Alloys Compd. 446–447, 296 ͑2007͒.
important for controllable hydrogen release from the sample,
as it necessitates a thermal activation process for the forma-
tion of Ti͑BH ͒ . This finding provides an approach for uti-
1
1
6
F. Schüth, B. Bogdanović, and M. Felderhoff, Chem. Commun.
4
3
7͑
Cambridge͒ 2004, 2249.
lization of the thermally instable transition metal borohy-
drides for controllable hydrogen release. As demonstrated,
Hydrogen, Fuel Cells and Infrastructure Technologies Program: Multi-
Year Research, Development and Demonstration Plan, Technical
Plan—Storage, 2007 ͑http://www1.eere.energy.gov/hydrogenandfuelcells/
mypp/pdfs/storage.pdf͒.
the LiBH /TiF system possesses favorable dehydrogenation
4
3
kinetics, particularly at moderate temperatures. However, its
vehicular application is restricted by the lack of reversibility
and the relatively low hydrogen capacity that is associated
with the inactive by-product. Our current efforts focus on
screening suitable reactants to carry out single or multiple
cation substitution in the intermediate Ti͑BH ͒ . The favor-
1
1
8
9
Q. F. Ge, J. Phys. Chem. A 108, 8682 ͑2004͒.
I. Barin, Thermochemical Data of Pure Substances ͑Wiley-VCH,
Weinheim, 1988͒.
T. J. Marks and J. R. Kolb, Chem. Rev. ͑Washington, D.C.͒ 77, 263
20
2
1͑
1977͒.
H. R. Hoekstra and J. J. Katz, J. Am. Chem. Soc. 71, 2488 ͑1949͒.
4
3