Post on 09-Aug-2020
Core shell structure for solid gas synthesis of LiBD4
O. Friedrichs,aJ. W. Kim,
bcA. Remhof,
aD. Wallacher,
dA. Hoser,
dY. W. Cho,
c
K. H. Ohband A. Zuttel
a
Received 22nd December 2009, Accepted 12th February 2010
First published as an Advance Article on the web 17th March 2010
DOI: 10.1039/b927068j
The formation of LiBD4 by the reaction of LiD in a diborane/hydrogen atmosphere was
analysed by in situ neutron diffraction and subsequent microstructural and chemical analysis
of the final product. The neutron diffraction shows that nucleation of LiBD4 already starts at
temperatures of 100 1C, i.e. in its low temperature phase (orthorhombic structure). However,
even at higher temperatures the reaction is incomplete. We observe a yield of approximately
50% at a temperature of 185 1C. A core shell structure of the grains, in which LiBD4 forms
a passivation layer on the surface of the LiD grains, was found in the subsequent
microstructural (electron microscopy) and chemical (electron energy loss spectrometry) analysis.
Introduction
In 2003, borohydrides (M[BH4]x) were proposed as new
hydrogen storage materials.1 Among borohydrides LiBH4
has one of the highest gravimetric and volumetric hydrogen
densities, which exceeds even the ones of gasoline. However,
the hydrogen absorption and desorption mechanism is not yet
understood in detail, and high pressures and temperatures are
required for its formation.
Already in 1953, Hermann I. Schlesinger and Nobel laureate
Herbert C. Brown discovered the important role of diborane
in the formation of borohydrides,2 when they published their
study of the synthesis routes for borohydrides. The synthesis
was based on chemical reactions of diborane with metal
hydrides in a solvent (e.g., diethyl ether). The role of the
solvent was to bring the diborane into contact with the hydride
and more importantly, to dissolve the product, borohydride
formed at the surface of the binary hydride. Schlesinger et al.
claimed that the solvent was mandatory for the preparation.
In our recent investigations we showed, in contradiction to
what was claimed by Schlesinger et al., the synthesis of LiBH4
by a solvent free method.3 By heating LiH in a diborane/
hydrogen atmosphere we were able to synthesize LiBH4 at
150 1C. However, the yield was limited to about 50% and similar
experiments to synthesize other borohydrides as Mg(BH4)2 and
Ca(BH4)2 by this method were not successful. Only by milling
the corresponding metal hydrides in a diborane/hydrogen
atmosphere,4 we succeeded to synthesize LiBH4, Ca(BH4)2 and
Mg(BH4)2 in an almost pure and solvent-free method.
In the present work the results of the investigation of the
reaction of LiD with diborane by in-situ neutron diffraction
are presented. The origin of the incomplete reaction was
analysed by microstructural and chemical characterization of
the resulting product.
Experimental
The synthesis of LiBD4 from LiD and B2D6 was carried out in
a custom made, cylindrical Inconel container (id 10 mm,
length 50 mm), developed and constructed by the DEGAS
laboratory of the Helmholtz Centre for Materials and Energy
(HZB) in Berlin, Germany. A schematic sketch is presented in
Fig. 1.
The container consists of two identical compartments, the
lower one filled with LiD (Sigma-Aldrich), the upper one with
a ball milled mixture of Li11BD4 (Katchem) and ZnCl2(Sigma-Aldrich) in a stoichiometric ratio of 5:2 as diborane
source.4 Milling results in the formation of LiZn2(BD4)5,5
which is known to emit diborane and hydrogen when heated
above 85 1C according to the following reaction:
LiZn2(BD4)5 - 2Zn + 2LiD + 5B2D6 + 4D2 (1)
Fig. 1 Schematic sketch of the custom made Inconel sample con-
tainer. The lower compartment is filled with LiD, the upper one with
a milled mixture of Li11BD4 and ZnCl2 forming LiZn2(BD4)5 as borane
source.
a Empa, Materials Science & Technology, Department ofEnvironment, Energy and Mobility, Div. Hydrogen and Energy,CH-8600 Dubendorf, Switzerland.E-mail: oliver.friedrichs@empa.ch; Fax: +41-44 823 4153;Tel: +41-44 823 4022
bDepartement of Materials Science and Engineering,Seoul National University, Seoul 151-742, Republic of Korea
cMaterials Science and Technology Research Division,Korea Institute of Science and Technology, Seoul 136-791,Republic of Korea
dHelmholtz Centre Berlin for Materials and Energy GmbH,Glienicker Strasse 100, 14109 Berlin, Germany
4600 | Phys. Chem. Chem. Phys., 2010, 12, 4600–4603 This journal is �c the Owner Societies 2010
PAPER www.rsc.org/pccp | Physical Chemistry Chemical Physics
Dow
nloa
ded
by S
eoul
Nat
iona
l Uni
vers
ity o
n 10
Jan
uary
201
1Pu
blis
hed
on 1
7 M
arch
201
0 on
http
://pu
bs.r
sc.o
rg |
doi:1
0.10
39/B
9270
68J
View Online
The two parts of the container were separated by a sinter filter
(mesh size: 0.5 mm), suppressing intermixing of the powders
but enabling gas exchange. The quantities of LiD and milled
Li11BD4 and ZnCl2 were selected to ensure a large excess of
emitted diborane during the reaction. The sample container
was filled and sealed in a glove box under inert helium atmos-
phere. To control the temperature, the sample container
was placed into a high-temperature furnace (HTF) which
was finally mounted on the sample stage of the focusing
diffractometer E6 of the HZB. The diffractometer is equipped
with a horizontally and vertically bent monochromator con-
sisting of 105 pyrolytic graphite crystals (20 � 20 � 2 mm)
mounted on a 15 � 7 matrix. The wavelength of 2.445 A was
chosen to maximize the neutron flux at the sample position
(5� 106 neutrons cm–2 s–1). During the experiment, the sample
was heated from rt to 185 1C and the diffraction pattern of
the LiD containing part of the container was recorded
consecutively.
The microstructure of the sample was investigated by secondary
electron imaging, induced by ion or electron irradiation using
a double (ion and electron) beam focussed ion beam (FIB,
FEI, Nova 200). For the chemical composition analysis of the
sample, transmission electron microscopy-electron energy loss
spectroscopy (TEM-EELS) is introduced to detect the light
elements (i.e. Li and B). A cross-sectional TEM sample
was obtained from the specific interest region of the final
product by FIB equipped with a manipulating probe
(100.7t, Omniprobe). A TEM sample preparation process
using FIB and air-lock loading chamber without air-exposure
are described in the previous report.7 Using this technique,
TEM sample which has a final thickness of E50 nm and a
large observation area (10 � 5 mm2) was obtained. The pre-
pared TEM sample was loaded into a 200 keV TEM (FEI,
Tecnai F20) using a portable glove-bag under Ar atmosphere
(99.999%).
Results and discussion
In an in-situ neutron powder diffraction experiment LiD is
heated stepwise from room temperature to 185 1C in diborane/
hydrogen atmosphere, while monitoring the diffraction pattern.
Fig. 2 shows the neutron diffraction pattern measured at
different temperatures.
Heating the cell to 85 1C leads to the decomposition of
LiZn2(BD4)5 and to the formation of the diborane/hydrogen
atmosphere. At this stage of the experiment, the diffraction
pattern of the lower part of the sample container remains
unchanged. The intensities of the LiD reflections6,7 as well
as the background stay constant, no new reflections appear.
Then we heated the sample container to 100 1C. Already
at this temperature, the diborane released from the source
reacts with LiD to form LiBD4 according to the following
reaction:
LiD + 1/2B2D6 - LiBD4 (2)
The corresponding Bragg reflections of the low temperature
phase of LiBH48 can be clearly identified in the diffraction
pattern. Heating to higher temperatures leads to a phase
transition, i.e. to the formation of the high temperature phase
of LiBD4.9,10 The reaction proceeds until about 50% of the
LiD has reacted. Then the reaction stops and further heating
for several hours at the final temperatures of 185 1C has no
more influence on the sample. This is in agreement to our
former studies, where we observed that only about 50% of
LiH is converted to LiBH4 while exposing to diborane at a
temperature of 150 1C.3
In order to analyse the origin of the incomplete transfor-
mation a microstructural study of the final product is carried
out using dual beam FIB. Fig. 3 shows secondary electron
images of the final product which are induced by ion beam
irradiation with an acceleration voltage of 30 keV at a working
distance of 19.5 mm. The corresponding cross-sectional view
of Fig. 3a is prepared by FIB milling treatment using Ga+ ion
beam11 and displayed in Fig. 3b.
The LiBD4 particle shows a glazed, smooth morphology
on the outside with a size range of 30 to 100 mm. In the cross-
sectional view, a core shell structure with a shell thickness of
about 3 mm is clearly observed.
In order to analyse the chemical compositions of the final
product using TEM-EELS, a uniformly thin TEM sample is
prepared by an FIB milling process. Fig. 4a and Fig. 4b
show an intermediate stage of the sample preparation, and a
Fig. 2 Selected diffraction pattern from the reaction of LiD and B2D6
observed by in situ neutron diffraction. The sample is heated from rt to
185 1C and the assigned phases correspond to LiBD4 high temperature
phase9,10 (�), LiBD4 low temperature phase8 (|), LiD6,7 (J) and Al
(+) from the sample holder.
Fig. 3 Secondary electron images (induced by 30 keV ion beam) of
LiD after reaction with diborane showing a grain (a) with its corres-
ponding cross sectional view (b) of a core shell structure.
This journal is �c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 4600–4603 | 4601
Dow
nloa
ded
by S
eoul
Nat
iona
l Uni
vers
ity o
n 10
Jan
uary
201
1Pu
blis
hed
on 1
7 M
arch
201
0 on
http
://pu
bs.r
sc.o
rg |
doi:1
0.10
39/B
9270
68J
View Online
prepared final TEM sample, respectively. Both secondary
electron images induced by electron beam are obtained with
an acceleration voltage of 5 keV.
In Fig. 4b, two different contrast regions are clearly observed,
which implies that the two regions have different chemical
compositions apart from the platinum (Pt) protective layer on top.
The sample is transferred from FIB to TEM and the chemical
compositions are analysed by TEM-EELS with 200 keV
operating voltage. The spectra from Li (K edge at 51 eV)
and B (K edge at 188 eV) absorption are displayed in Fig. 4c
and Fig. 4d, respectively. The outer region of the particle
showing the darker contrast shows signals originating from Li
and B, while the inner region contains Li but no B. This is
explained by the formation of LiBD4 on the surface of LiD
forming a core shell structure observed by the ion beam
analysis displayed in Fig. 3.
Fig. 5 shows a schematic illustration of the core shell
formation during the synthesis of LiBD4.
LiD reacts on the surface with diborane and forms LiBD4.
A surface layer of LiBD4 is formed and grows to a certain
thickness. Then the reaction stops and leaves a core shell
structure with LiD in the interior and LiBD4 on the outside.
For the formation of LiBD4 either boron in the form of a B–H
species has to diffuse to the interior passing the already formed
layer of LiBD4 or Li has to diffuse to the exterior. Thereby the
overall charge neutrality of the reaction has to be conserved.
We explain the reason for the incomplete formation by a
limited diffusion of either species with the increasing thickness
of the LiBD4 layer. Fig. 5 illustrates one possible reaction
mechanism in which BH4� ions are diffusing into the interior
while D� ions are diffusing to the exterior. In this mechanism
the reaction is limited by the diffusion of BH4� and D� species
through the LiBH4 layer. A mechanism based on Li diffusion is
favored by the high mobility of Li even at low temperatures.12
With less mobile elements such as Mg and Ca no formation of
the corresponding borohydride could be observed under similar
experimental conditions.
Conclusions
LiBD4 forms at the surface of LiD in a diborane/hydrogen
atmosphere. The reaction already starts in the temperature
range of the orthorhombic phase of LiBD4 and stops after
about 50% of LiD is consumed for the formation of LiBD4. A
core-shell structure of lithium hydride surrounded by lithium
borohydride is observed. The reaction stops most probably
due to diffusion problems of either B–H species into the grain
or Li towards the exterior. The results are in agreement with
the passivation layer proposed by Schlesinger et al., who
synthesized different borohydrides in solvents in order to
prevent the formation of the passivation layer. It also explains
the new method to synthesize borohydrides by milling metal
hydrides in diborane atmosphere.4 By the milling procedure
the passivation layer is broken and further reaction is enabled
as we presented on the solvent free synthesis of LiBH4,
Ca(BH4)2 and Mg(BH4)2.
Acknowledgements
Financial support from the 6th Framework Program of the
European Commission (NESSHY Contract No.: 518271), the
Swiss National Science Foundation (SNF-Project 200021-
119972/1), the Swiss Federal Office of Energy and the integrated
project of ICC-IMR is gratefully acknowledged. We would also
like to thank for financial support from the European Commis-
sion through the Key Action: Strengthening the European
Research Area, Research Infrastructure (contract number
RII3-CT-2003-505925).
References
1 A. Zuttel, P. Wenger, S. Rentsch, P. Sudan, Ph. Mauron andCh. Emmenegger, J. Power Sources, 2003, 118, 1.
2 H. I. Schlesinger, H. C. Brown, B. Abraham, A. C. Bond,N. Davidson, A. E. Finholt, J. R. Gilbreath, H. Hoekstra,L. Horvitz, E. K. Hyde, J. J. Katz, J. Knight, R. A. Lad,
Fig. 4 SEM secondary electron images and TEM-EELS spectra of
LiD after reaction with diborane. Section (a) shows the sample
preparation using FIB resulting in (b) a uniformly thin sample.
Additionally the figure displays the TEM-EELS spectra of the K
absorption edge of (c) Li and (d) B for two different areas corresponding
to LiBD4 and LiD.
Fig. 5 Schematic presentation of the core shell formation on LiD due
to surface reaction with diborane. The figure illustrates one possible
reaction mechanism, which is based on diffusion of BH4� and D� ions.
4602 | Phys. Chem. Chem. Phys., 2010, 12, 4600–4603 This journal is �c the Owner Societies 2010
Dow
nloa
ded
by S
eoul
Nat
iona
l Uni
vers
ity o
n 10
Jan
uary
201
1Pu
blis
hed
on 1
7 M
arch
201
0 on
http
://pu
bs.r
sc.o
rg |
doi:1
0.10
39/B
9270
68J
View Online
D. L. Mayfield, L. Rapp, D. M. Ritter, A. M. Schwartz, L. Sheft,L. D. Tuck and A. O. Walker, J. Am. Chem. Soc., 1953, 75,186.
3 O. Friedrichs, A. Borgschulte, S. Kato, F. Buchter, R. Gremaud,A. Remhof and A. Zuttel, Chem.–Eur. J., 2009, 15, 5531.
4 O. Friedrichs, A. Remhof, A. Borgschulte, F. Buchter, S. I. Orimoand A. Zuettel, submitted.
5 D. Ravnsbæk, Y. Filinchuk, Y. Cerenius, H. J. Jakobsen,F. Besenbacher, J. Skibsted and T. R. Jensen, Angew. Chem., Int.Ed., 2009, 48, 6659.
6 J. P. Vidal and G. Vidal-Valat, Acta Crystallogr., Sect. B: Struct.Sci., 1986, 42, 131.
7 C. G. Shull, E. O. Wollan, G. A. Morton and W. L. Davidson,Phys. Rev., 1948, 73, 842.
8 M. R. Hartman, J. J. Rush, T. J. Udovic, R. C. Bowman andS. J. Hwang, J. Solid State Chem., 2007, 180, 1298.
9 J.-Ph. Soulie, G. Renaudin, R. Cerny and K. Yvon, J. AlloysCompd., 2002, 346, 200.
10 F. Buchter, Z. Łodziana, Ph. Mauron, A. Remhof, O. Friedrichs,A. Borgschulte, A. Zuettel, D. Sheptyakov, Th. Strassle andA. J. Ramirez-Cuesta, Phys. Rev. B: Condens. Matter Mater.Phys., 2008, 78, 094302.
11 J. W. Kim, J.-P. Ahn, S.-A. Jin, S. H. Lee, H.-S. Chung, J.-H. Shim,Y. W. Cho and K. H. Oh, J. Power Sources, 2008, 178, 373.
12 M. Matsuo, A. Remhof, P. Martelli, R. Caputo, M. Ernst,Y. Miura, T. Sato, H. Oguchi, H. Maekawa, H. Takamura,A. Borgschulte, A. Zuettel and S. I. Orimo, J. Am. Chem. Soc.,2009, 131, 16389.
This journal is �c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 4600–4603 | 4603
Dow
nloa
ded
by S
eoul
Nat
iona
l Uni
vers
ity o
n 10
Jan
uary
201
1Pu
blis
hed
on 1
7 M
arch
201
0 on
http
://pu
bs.r
sc.o
rg |
doi:1
0.10
39/B
9270
68J
View Online