Sodium hydrazinidoborane: A chemical hydrogen-storage material

Article abstract of DOI:10.1002/cssc.201200800

Herein, we present the successful synthesis and full characterization (by ¹¹B magic-angle-spinning nuclear magnetic resonance spectroscopy, infrared spectroscopy, powder X-ray diffraction) of sodium hydrazinidoborane (NaN₂H₃BH₃, with a hydrogen content of 8.85 wt%), a new material for chemical hydrogen storage. Using lab-prepared pure hydrazine borane (N₂H₄BH₃) and commercial sodium hydride as precursors, sodium hydrazinidoborane was synthesized by ball-milling at low temperature (-30 °C) under an argon atmosphere. Its thermal stability was assessed by thermogravimetric analysis and differential scanning calorimetry. It was found that under heating sodium hydrazinidoborane starts to liberate hydrogen below 608C. Within the range of 60-150 °C, the overall mass loss is as high as 7.6 wt%. Relative to the parent N ₂H₄BH₃, sodium hydrazinidoborane shows improved dehydrogenation properties, further confirmed by dehydrogenation experiments under prolonged heating at constant temperatures of 80, 90, 95, 100, and 110°C. Hence, sodium hydrazinidoborane appears to be more suitable for chemical hydrogen storage than N₂H₄BH₃.

Full text of DOI:10.1002/cssc.201200800

CHEMSUSCHEM
FULL PAPERS
DOI: 10.1002/cssc.201200800
Sodium Hydrazinidoborane: A Chemical Hydrogen-Storage
Material
Romain Moury,[a] Umit B. Demirci,*[a] Takayuki Ichikawa,[b] Yaroslav Filinchuk,[c]
Rodica Chiriac,[d] Arie van der Lee,[a] and Philippe Miele[a]
Herein, we present the successful synthesis and full characteri-
zation (by 11B magic-angle-spinning nuclear magnetic reso-
nance spectroscopy, infrared spectroscopy, powder X-ray dif-
fraction) of sodium hydrazinidoborane (NaN2H3BH3, with a hy-
drogen content of 8.85 wt%), a new material for chemical hy-
drogen storage. Using lab-prepared pure hydrazine borane
(N2H4BH3) and commercial sodium hydride as precursors,
sodium hydrazinidoborane was synthesized by ball-milling at
low temperature (À308C) under an argon atmosphere. Its ther-
mal stability was assessed by thermogravimetric analysis and
differential scanning calorimetry. It was found that under heat-
ing sodium hydrazinidoborane starts to liberate hydrogen
below 608C. Within the range of 60–1508C, the overall mass
loss is as high as 7.6 wt%. Relative to the parent N2H4BH3,
sodium hydrazinidoborane shows improved dehydrogenation
properties, further confirmed by dehydrogenation experiments
under prolonged heating at constant temperatures of 80, 90,
95, 100, and 1108C. Hence, sodium hydrazinidoborane appears
to be more suitable for chemical hydrogen storage than
N2H4BH3.
Introduction
Among the boranes considered for chemical hydrogen stor-
age,[1] hydrazine borane (N2H4BH3) is one of the most recent
materials.[2] Remarkably, hydrazine borane (HB) presents four
protic hydrogen atoms (Hd+), in tandem with three hydridic
hydrogen atoms (HdÀ). Therefore, it presents material-based[3]
gravimetric and volumetric hydrogen-storage capacities as
high as 15.4 wt% and 146 gLÀ1, respectively. Another benefi-
cial feature is that pure HB is stable at ambient conditions,
when stored under an argon atmosphere.[4] It is also stable in
degassed aqueous solutions over a few weeks.[5] Therefore, the
challenge has been to dehydrogenate HB to a large degree
under ambient/mild conditions (i.e., at ꢀ858C and atmospher-
ic pressure, which are criteria for on-board hydrogen-storage
systems for light-duty vehicles).[6]
HB can be considered as a derivative of ammonia borane
(NH3BH3), in which the NH3 moiety has been replaced by N2H4.
NH3BH3 has attracted a lot of attention in thermolytic dehydro-
genation contexts within the past decade, but the first at-
tempts at its thermolysis date back to the 1970–80s.[7] NH3BH3
dehydrogenates at temperatures higher than 1008C, releasing
H2 and byproducts, such as the unwanted borazine and dibor-
ane, and leading to a solid residue (BNHx).[8] The dehydrogena-
tion of HB, similarly to NH3BH3,[9] can be initiated by adding
a metal-based catalyst (e.g., nickel-based bimetallic alloys) into
the aqueous solution at low temperatures,[5,10] or by heating
the solid under an inert atmosphere.[4,11] Through the first ap-
proach, the catalytic dehydrogenation of HB takes place ac-
cording to a two-step reaction that consists of both the hy-
drolysis of the BH3 group and the decomposition of the N2H4
moiety. This approach presents the advantage of occurring at
relatively low temperatures (508C).[10] However, thermolytic de-
hydrogenation may be more attractive due to the fact that the
reaction is easily performed by heating the borane only.
[a] R. Moury, Dr. U. B. Demirci, Dr. A. van der Lee, Prof. P. Miele
IEM (Institut Europeen des Membranes)
UMR 5635 (CNRS-ENSCM-UM2), Universite Montpellier 2
Place E. Bataillon, 34095, Montpellier (France)
Fax: (+33)04-67-14-91-19
The thermolytic dehydrogenation of HB initiates at 608C, si-
multaneously with HB melting, revealing an endothermic peak
with an enthalpy of 15 kJmolÀ1 in the differential scanning cal-
orimetry (DSC) curve. A mass loss of 1.2 wt% at 958C, as deter-
mined by thermogravimetric analysis (TGA) at a heating rate of
18CminÀ1 was reported.[12] A second mass loss of 28.7 wt%
occurs within 105–1608C due to the release of N2H4. A third
mass loss of 4.3 wt% is observed at about 2508C. In this way,
2 equiv of H2 are released and a shock-sensitive solid residue
forms.[12] The high dehydrogenation temperatures and the for-
mation of hazardous byproducts are significant problems,
making HB unsuitable for mobile and portable hydrogen stor-
age. Consequently, strategies have been recently adopted to
E-mail: umit.demirci@um2.fr
[b] Dr. T. Ichikawa
Institute for Advanced Materials Research
Hiroshima University
1-3-1 Kagamiyama, Higashi-Hiroshima, 739-8530 (Japan)
[c] Prof. Y. Filinchuk
Institute of Condensed Matter and Nanosciences
Universitꢀ Catholique de Louvain
Place L. Pasteur 1, 1348 Louvain-la-Neuve (Belgium)
[d] Dr. R. Chiriac
Laboratoire des Multimatꢀriaux et Interfaces
Universitꢀ Lyon 1, CNRS, UMR 5615
43 Boulevard du 11 Novembre 1918, 69622 Villeurbanne (France)
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201200800.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 8
&
1
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
improve the dehydrogenation properties of HB and avoid the
evolution of unwanted gaseous and solid byproducts. The first
strategy involves mixing HB with LiH[11] and the second chemi-
cal modification of HB by replacing one Hd+ atom by an alka-
line cation, such as Li+, which leads to the formation of
LiN2H3BH3.[13]
sharp, which is characteristic of high crystallinity. The difference
in the chemical shifts is consistent with a change of the chemi-
cal environment of B in NaHB.
Analysis by IR spectroscopy (Figure 2) showed slight changes
relative to HB. The changes are mainly visible for the NÀH
stretching (2600–3600 cmÀ1), bending (1800–1300 cmÀ1), and
rocking (1100–900 cmÀ1) regions. The bands shown by NaHB
In view of the issues encountered in HB thermolysis, we de-
cided to follow (right after our work dedicated to this
borane)[12] the chemical modification approach. We then start-
ed to work on the elaboration of metal hydrazinidoboranes
(MN2H3BH3, M=Li or Na), which attractively show parity in
protic and hydridic hydrogen atoms (three Hd+ and three HdÀ
atoms, whereas metal amidoboranes have just two Hd+ for
three HdÀ atoms). Herein, we report for the first time the suc-
cessful synthesis and characterization of NaN2H3BH3 (NaHB).
Results and Discussion
Solubility and stability of NaHB
Sodium hydrazinidoborane (hydrogen content of 8.85 wt%) is
a colorless solid, paste-like at 3–48C, but powdery at 208C. It is
insoluble in common organic solvents, such as THF and tolu-
ene, but reacts with water or methanol (i.e., hydrolysis and
methanolysis of the BH3 moiety; see the Supporting Informa-
tion, Figure S1). For stability reasons, it has to be stored under
an argon atmosphere.
Figure 2. IR spectra of NaHB and HB with: NÀH stretching at 2600–
3600 cmÀ1, BÀH stretching at 2000–2600 cmÀ1, NÀH bending (also BÀN
stretching) at 1300–1800 cmÀ1, BÀH bending at 1100–1300 cmÀ1, NÀH rock-
ing at 900–1100 cmÀ1, and BNÀN stretching at 600–900 cmÀ1
.
Molecular structure of NaHB
are either broadened or weakened, consistent with the change
of the environment of the NÀH bonds by the introduction of
Na+. The BÀH bonds display large and strong bands in the
The molecular structure of NaHB was analyzed by solid-state
11B magic-angle-spinning (MAS) NMR spectroscopy. The spec-
trum was compared to that of HB (Figure 1). Both samples
have a single NBH3 environment. Unlike HB, displaying a split
signal due to a quadrupolar coupling centered at À23.2 ppm,
NaHB shows a single symmetric resonance at À16.5 ppm, sug-
gesting isotropy around the boron atom and a higher molecu-
lar degree of freedom. This indicates a different destabilized
stretching (2000–2600 cmÀ1
) and bending regions (1100–
1300 cmÀ1). With respect to the BÀN bond, several bands in
the BNÀN asymmetric and symmetric stretching regions (600–
900 cmÀ1) are observed.
Crystal structure of NaHB
H
d+···HdÀ network relative to that found in HB. The signal is
NaHB was analyzed by powder XRD. Figure 3 shows an XRD
pattern of single-phase NaN2H3BH3 and its comparison with
patterns of the starting compounds. Diffraction data were in-
dexed in a monoclinic unit cell. The systematic absences sug-
gest the space group P21/n with a=4.97437(11) ꢁ, b=
7.95806(15) ꢁ, c=9.29232(19) ꢁ, and b=93.8137(11)8. The crys-
tal structure has been solved by global optimization in direct
space, using the FOX program, and refined by the Rietveld
method using the FullProf program suite (see the Supporting
Information, Figure S2). Hydrogen atoms were located and re-
fined using eight restraints on bond distances (X-ray average
of 0.9 ꢁ for NÀH and 1.13 ꢁ for BÀH) and fourteen restraints
on bond angles. Localization of hydrogen atoms around nitro-
gen atoms is geometrically unambiguous because at least two
heavier atoms are connected to the sp3-hybridized pivots. The
refined orientation of the ÀBH3 group corresponds to the ex-
pected staggered configuration, further supported by the pres-
ence of two dihydrogen bonds (see the Supporting Informa-
Figure 1. 11B MAS NMR spectra of NaHB and HB.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 8
&
2
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Figure 3. XRD pattern of NaHB and its comparison with patterns of the start-
ing compounds.
tion, Figure S3). In addition, the ÀBH3 group is coordinated to
Na+ through the H···H edge, as in most other metal amidobor-
anes and borohydrides.[14] A common isotropic displacement
factor was refined for each atom type. The final discrepancy in-
dices are: Rp =1.48%, Rwp =2.13% (not corrected for back-
ground), Rp =12.0%, Rwp =10.0% (conventional), RB =5.4%,
c2 =155. The latter value reflects high counting statistics accu-
mulated by the 2D detector. The reliable determination of hy-
drogen atom positions is highlighted by the fact that a removal
of a single hydrogen atom, followed by the Rietveld refine-
ment of all remaining parameters, cannot absorb the differen-
ces, but leads to an increase of RB to about 7.5% and c2 to
more than 210. Our tests show that the missing hydrogen
atom can be located from difference Fourier maps typically as
the highest residual peak.
The crystal structure confirms the formation of NaHB. As in
the case of LiN2H3BH3,[13] Na+ substitutes an hydrogen atom
dissociated from a boron-linked nitrogen atom in the middle
of the molecule. Na+ is surrounded by five symmetry-equiva-
lent anions (Figure 4): four of them form a slightly distorted
square environment created by two nitrogen atoms [NaÀN dis-
tance is 2.396(4) and 2.428(4) ꢁ, NÀNaÀN angle is 168.7(2)8]
and two ÀBH3 groups coordinated symmetrically through the
BH2 edges [Na···B distance is 2.947(5)–3.074(5) ꢁ, B···Na···B angle
is 140.9(2)8]. The fifth nearest neighbor completes the coordi-
nation sphere to a tetragonal pyramid, with a Na···B distance
of 3.425(6) ꢁ and a coordination through the BÀH vertex. The
apical distance links 2D sheets oriented in the (À101) plane in
the 3D structure. The BÀN distance in NaHB is 1.537(6) ꢁ. All
distances are consistent with those in the structure of
NaNH2BH3.[15] The NÀN distance is 1.453(5) ꢁ. To sum up, XRD,
IR, and NMR results confirm the successful substitution of the
Hd+ atom in N2H4 by the electron-acceptor cation Na+[16] and
the formation of NaHB.
Figure 4. Crystal structure of NaN2H3BH3: view (top) along the a axis with
H···H network (dashed lines) for distances below 2.4 ꢁ (here: 2.35 ꢁ). The co-
ordination sphere of Na with the distances of coordinated symmetry-equiva-
lent anions is shown in the structure at the bottom. H, B, N, and Na atoms
are represented by grey, pink, blue, and purple spheres.
Behavior of NaHB under heating
Figure 5. TGA and DSC profiles (top curves) of NaHB (heating rate of
28CminÀ1), and evolution of H2 and gaseous byproducts (bottom curve) fol-
lowed by MS coupled to the TGA apparatus.
The behavior of NaHB under heating was analyzed by TGA
(Figure 5). NaHB starts to liberate H2 below 608C. Concomitant-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 8
&
3
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
ly, the solid melts. For LiN2H3BH3, dehydrogenation starts at
708C.[13] Over the 60–1008C range, NaHB loses 6 wt% of its ini-
tial weight mainly due to the loss of H2, as discussed below.
This is an attractive result in comparison to that obtained with
pure HB (mass loss of 1.2 wt% at 958C),[12] or even to that ob-
tained with neat ammonia borane (stable up to ꢁ1008C).[9]
Within 60–1508C, NaHB dehydrogenates and the overall mass
loss is 7.6 wt%. For comparison, LiN2H3BH3 (11.69 wt%) shows
a mass loss of 8.8 wt% within the temperature range of 70–
1508C.[13]
The released gases were analyzed by mass spectroscopy
(Figure 5). Traces of the unwanted byproducts N2H4 and NH3
were detected in the released hydrogen. Ammonia was also
detected during the decomposition of LiN2H3BH3.[13] The
amounts of N2H4 liberated by NaHB are much lower than those
generated by the decomposition of HB (see the Supporting In-
formation, Figure S4). This may be attributed to the presence
of Na in NaHB, with Na inhibiting the evolution of N2H4 and fa-
voring the reactions between the Hd+ and HdÀ atoms. A signal
at m/zꢁ28 could be attributed to N2 (28 gmolÀ1) and/or B2H6
(27.6 gmolÀ1). IR analysis of the released gases was helpful in
this identification. The spectrum (see the Supporting Informa-
tion, Figure S5) shows the absence of any signal in the BÀH
stretching region (2000–2600 cmÀ1), evidencing the absence of
B2H6. The signal at m/zꢁ28 was thus ascribed to N2, which
Figure 6. Dehydrogenation of NaHB under prolonged heating at five differ-
ent temperatures.
Table 1. Time for the release of 1 equiv of H2 from HB and NaHB with
temperature increase.
T
t [min]
[8C]
NaHB
HB
80
90
95
111
28
5
>120
100
96
À
100
110
1
0.42
64
56
suggests that the N2H3 moiety completely dehydrogenates.
From the point of view of H2 purity, the presence of N2 is not
problematic because it is inert towards the metal electrocata-
lysts used in fuel cells. This gas was also found during the de-
composition of LiN2H3BH3.[13] Notably, borazine was not detect-
ed under our experimental conditions.
After heating for 2 h, about 1.1, 1.3, 1.8, 2.8, and 2.8 equiv of
H2 evolve at 80, 90, 95, 100, and 1108C, respectively. Neverthe-
less, it is important to note that most of the H2 released was
generated in less than 10 min at 100 and 1108C. That implies
a material-based net effective gravimetric hydrogen-storage
capacity of 3.1, 3.8, 5.2, 8.1, and 8.4 wt%, respectively. Despite
the fact that different experimental conditions were used, the
dehydrogenation of NaHB may be compared to that of
NaNH2BH3 (9.45 wt%). It was reported that in 1 h the latter re-
leases about 64% of its hydrogen content (6 wt%) under heat-
ing at about 918C,[15] whereas our compound liberates 50%
(4.2 wt%) at 958C. Despite its inferior dehydrogenation perfor-
mance, NaHB is a potential candidate for chemical hydrogen
storage.
The behavior of NaHB under heating was also analyzed by
DSC (Figure 5). Dehydrogenation seems to be a complex reac-
tion, consisting of four exothermic processes. These processes
may be ascribed to three successive dehydrogenation stages
(peaking at ꢁ75, 90, and 958C), followed by denitrogenation
(peaking at ꢁ1078C). The exothermic character of dehydro-
genation lets us suggest that direct rehydrogenation under H2
is not possible.[1,9,15]
The dehydrogenation of NaHB under prolonged heating at
a constant temperature (80, 90, 95, 100, and 1108C; Figure 6)
was compared to that of HB (see the Supporting Information,
Figure S6). The results obtained with HB are consistent with
the results reported by Hꢂgle et al.,[11] even though at 1108C
and within 2 h HB releases more H2 in our experiments (3.7
versus 2.4 wt%). The dehydrogenation process has a two-stage
profile, with the second stage having slower H2 release rates.
At 1108C, the major fraction of H2 (7.9 wt%) is liberated within
30 min. With respect to NaHB, the dehydrogenation profile is
similar, but H2 release is much faster. At 808C, the H2 release
rate is close to that found with HB, but beyond this tempera-
ture it is faster. NaHB is able to liberate 1 equiv of H2 within
111 min, whereas HB releases less than 0.7 equiv of H2 within
the same time frame. At higher temperatures, the kinetics and
extent of dehydrogenation are enhanced. The results are sum-
marized in Table 1. For example, NaHB generates 1 equiv of H2
in 5 min at 958C (versus 96 min for HB).
The gases liberated at 95 and 1008C were analyzed by GC.
At <1008C, H2 was pure, but at >1008C trace amounts of NH3
were detected (see the Supporting Information, Figure S7). It is
noteworthy that emission of NH3, but to a higher extent, was
also reported in the case of NaNH2BH3.[17]
The dehydrogenation of NaHB at temperatures higher than
1008C (Figure 6, Table 2) is abrupt and extremely rapid, which
might be due to a different mechanism of dehydrogenation in-
volving different reaction intermediates. At 100 and 1108C, hy-
drogen-release rates (calculated for the time range necessary
to liberate 2.5 equiv of H2) are about 1.5 and 4 LminÀ1. Accord-
ing to Gao and Shreeve,[18] the fast kinetics of H2 generation
might open a new perspective of energy applications (i.e, fuels
in hypergolic propellant systems).
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 8
&
4
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Table 2. H2 release rates (r) from NaHB at various temperatures. Stages 1
and 2 refer to the dehydrogenation profile, in which two different H2 re-
lease kinetics are observed. The conversion of NaHB after heating for 2 h
is also provided.
[a]
[a]
[b]
T
[8C]
r1
r2
a2 h
[mLH minÀ1
]
[mLH minÀ1
]
[%]
2
2
80
90
95
100
110
--[c]
1.3
0.1
33
44
59
91
95
--[d]
3.1
0.08
0.01
0.02
ꢁ1500
ꢁ4000
[a] Rate values at 100 (r1) and 1108C (r2) are calculated on the basis of
time required to generate 2.5 equiv of H2. [b] Conversion at 2 h. [c] The
time of stage 1 is too short to give a relevant rate. [d] The curve is not
linear, making rate difficult to determine.
Figure 7. IR spectra of NaHB and of solid residues recovered at 95 and
110 8C with: NÀH stretching at 2600–3600 cmÀ1, BÀH stretching at 2000–
2600 cmÀ1, NÀH bending (also BÀN stretching) at 1300–1800 cmÀ1, BÀH
bending at 1100–1300 cmÀ1, NÀH rocking at 900–1100 cmÀ1, and BNÀN
Characterization of solid residues
stretching at 600–900 cmÀ1
.
The solid residues recovered after the heating of NaHB at 95
and 1108C for 2 h were analyzed because their identification,
though often complicated,[9,11] is important for approaching
their recyclability. The XRD patterns of the residue (see the
Supporting Information, Figure S8) resemble those reported in
the case of heated ammonia borane, amidoboranes, or
HB,[12,15,19] thus suggesting the formation of an amorphous
polymeric phase. Nevertheless, for the sample heated at 958C,
some diffraction peaks are distinguished. Despite the fact that
a thorough analysis was undertaken, we were unable to unam-
biguously assign XRD peaks, showing once more the difficulty
to characterize the boron- and nitrogen-based solid residues.
Nevertheless, the formation of NaH or Na could be eliminated.
The formation of boron nitride (BN) was also ruled out by the
fact that this ceramic material generally forms from boranes at
temperatures much higher than 5008C.[9] Finally, a survey of
the literature dedicated to boron-based materials was per-
formed, according to which the diffraction peaks might be at-
tributed to crystalline linear polyaminoborane (À(NH2BH2)nÀ).[19]
The IR spectra (Figure 7) are typical of dehydrogenated
boron-based materials.[20] The intensity of stretching bands of
NÀH and BÀH weakened with dehydrogenation temperature.
This is consistent with the increase of the degree of dehydro-
genation with temperature. The intensity of the bands at 400–
1300 cmÀ1 also decays. The band at 1300–1600 cmÀ1, becom-
ing stronger upon dehydrogenation, may be attributed to the
BÀN bonds in polymeric boron- and nitrogen-based com-
pounds (e.g., polyborazylene-like).[21]
1.6 and 5.9 ppm are ascribed to N3ÀxBHx environments (1<x<
3), typical of polycyclic boron- and nitrogen-based com-
pounds.[24] By increasing the temperature to 1108C (see the
Supporting Information, Figure S9), almost all of NaHB dehy-
drogenates, and BN3, N2BH, and N3ÀxBHx species form.
The 23Na MAS NMR spectrum of NaHB (see the Supporting
Information, Figure S10) shows a complex signal. The signal
centered at À75.7 ppm is due to a quadrupolar interaction on
a single Na site in NaHB.[23] The peak at À13.6 ppm may be at-
tributed to an amorphous phase. With the increase of temper-
ature, the quadrupolar lineshape broadens and that gives
a single resonance at À19.9 ppm at 958C and À7.1 ppm at
110 8C. This confirms the amorphization of NaHB.[23]
To sum up, the characterization of the solid residues sug-
gests the formation of an amorphous polyborazylene-like ma-
terial. Goubeau and Ricker suggested the formation of various
linear and cyclic polymeric materials based on the structural
units ÀNHxÀNHyÀBHzÀ (with x+y+z=2.5 at 958C and 0.5 at
110 8C).[4] It is assumed that Na would enter the polymeric
structure because in our conditions the formation of NaH (or
of Na) has not been observed.
Conclusions
We have synthesized for the first time sodium hydrazinidobor-
ane (hydrogen content of 8.85 wt%). The results presented
show that it has potential as a chemical hydrogen-storage ma-
terial. A material-based gravimetric hydrogen-storage capacity
slightly higher than 8 wt% is achieved with heating at 100 and
110 8C. Furthermore, dehydrogenation rates were found to be
high, with values of 1.5 and 4 LminÀ1, respectively.
The 11B MAS NMR spectrum of the residue obtained at 958C
(see the Supporting Information, Figure S9) shows a weak
À
signal at À42.9 ppm. It corresponds to the BH4 ion, suggest-
ing the formation of an intermediate containing this species.
For example, the formation of [(NH3)2BH2][BH4] was suggested
in the thermolysis of ammonia borane.[22] A similar intermedi-
ate might tentatively be suggested as [Na2(N2H3)2BH2]+[BH4]À
in the title system. At 958C, three 11B resonances at positive
shifts are also observed. The broader one, centered at
17.7 ppm, is assigned to three-fold boron, that is, to the BN3 or
N2BH species, as in polyborazylene.[23] The peaks centered at
As in the case of the parent HB or ammonia borane and de-
rivatives, the technological implementation of NaHB will be
strongly dependent on closing the cycle of hydrogen storage.
In other words, the recyclability of the spent fuel after dehy-
drogenation and, most importantly, the regenerability of the
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 8
&
5
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
borane are the key issues. The recent progress in the regenera-
tion of ammonia borane is useful in this context.[25]
Acknowledgements
The authors would like to acknowledge the support of the Direc-
tion Gꢀnꢀrale des Armꢀes (DGA), the Centre National de la Re-
cherche Scientifique (CNRS) and Fonds de la Recherche Scientifi-
que (FNRS). U.B.D. and T.I. would like to thank the JSPS for par-
tially supporting the present work through the JSPS Summer Pro-
gram given to R.M. U.B.D. and Y.F. would like to thank the COST
“Nanostructured materials for solid-state hydrogen storage” pro-
gram for partially supporting the present work through the
short-term scientific mission given to R.M. We would also like to
thank M. FranÅois Toche (Universitꢀ Lyon 1) for his kind help with
DSC, and SNBL, ESRF for beam-time allocation.
Experimental Section
Hydrazine borane (>99%) was synthesized according to a proce-
dure described elsewhere.[12] Typically, N2H4·0.5H2SO4 (21.42 g,
264.34 mmol, Sigma–Aldrich) and NaBH4 (10 g, 264.34 mmol, Acros
Organics) were suspended under vigorous stirring in anhydrous di-
oxane (150 mL) under argon. The stirring was continued until com-
pletion of the H2 release, which was followed by a bubble counter
filled with paraffin oil. After Schlenk filtration of the mixture, diox-
ane was removed under vacuum. Sodium hydride (Sigma–Aldrich)
was used as received. Both samples were stored in an argon-filled
glove box (MBraun M200B, H2O<0.1 ppm, O2 <0.1 ppm).
Keywords: boron
· dehydrogenation · hydrogen · nmr
spectroscopy · synthetic methods
NaHB was prepared by ball-milling under an argon atmosphere.
In the argon-filled glove box, hydrazine borane (0.1312 g) was
transferred into a stainless steel jar (25 mL). Then, stainless steel
balls (ca. 20 g) were delicately and compactly aligned over the HB
sample. Finally, NaH (0.0724 g) was transferred into the jar and
over the balls, the ball layer acting as a barrier. The molar HB/NaH
[1] a) L. Schlapbach, A. Zꢂttel, Nature 2001, 414, 353–358; b) W. Grochala,
P. P. Edwards, Chem. Rev. 2004, 104, 1283–1316; c) U. Eberle, M. Felder-
hoff, F. Schꢂth, Angew. Chem. 2009, 121, 6732–6757; Angew. Chem. Int.
Ed. 2009, 48, 6608–6630; d) J. Graetz, Chem. Soc. Rev. 2009, 38, 73–82;
e) J. Yang, A. Sudik, C. Wolverton, D. J. Siegel, Chem. Soc. Rev. 2010, 39,
656–675; f) P. E. de Jongh, P. Adelhelm, ChemSusChem 2010, 3, 1332–
1348; g) N. Armaroli, V. Balzani, ChemSusChem 2011, 4, 21–36; h) H.
Reardon, J. M. Hanlon, R. W. Hugues, A. Godula-Jopek, T. K. Mandal,
D. H. Gregory, Energy Environ. Sci. 2012, 5, 5951–5979.
ratio was equal to
1
and the weight ratio balls/(NaH
HB) was equal to 100. The transfer of NaH was performed carefully,
avoiding to put both precursors into contact; this was highly im-
portant because these reactants interacted readily under ambient
conditions, releasing hydrogen. Right after the NaH transfer, the
closed grinding jar was carefully taken out of the glove box and
placed in a vacuum flask filled with dry ice. The temperature of the
jar was monitored and when it decreased to À308C the jar was
placed in the ball mill. The milling process started immediately
(250 rpm for 10 min). NaHB obtained in this way was stable under
an argon atmosphere and under ambient conditions, and could be
handled safely.
[2] N. Vinh-Son, S. Swinnen, M. H. Matus, M. T. Nguyen, D. A. Dixon, Phys.
Chem. Chem. Phys. 2009, 11, 6339–6344.
[3] a) R. von Helmolt, U. Eberle, J. Power Sources 2007, 165, 833–843;
b) U. B. Demirci, P. Miele, Energy Environ. Sci. 2011, 4, 3334–3341.
[4] V. J. Goubeau, E. Ricker, Z. Anorg. Allg. Chem. 1961, 310, 123–142.
[5] S. Karahan, M. Zahmakıran, S. ꢃzkar, Int. J. Hydrogen Energy 2011, 36,
4958–4966.
[6] Targets for Onboard Hydrogen Storage Systems for Light-Duty Vehicles, US
Department of Energy, Office of Energy Efficiency and Renewable
Energy and The FreedomCAR and Fuel Partnership, 2009. Available at
http://www1.eere.energy.gov/hydrogenandfuelcells/storage/pdfs/tar-
gets_onboard_hydro_storage_explanation.pdf.
[7] a) M. G. Hu, R. A. Geanangel, W. W. Wendlandt, Thermochim. Acta 1978,
23, 249–255; b) V. Sit, R. A. Geanangel, W. W. Wendlandt, Thermochim.
Acta 1987, 113, 379–382.
[8] a) G. Wolf, J. Baumann, F. Baitalow, F. P. Hoffmann, Thermochim. Acta
2000, 343, 19–25; b) F. Baitalow, J. Baumann, G. Wolf, K. Jaenicke-
Rçßler, G. Leitner, Thermochim. Acta 2002, 391, 159–168; c) F. Baitalow,
G. Wolf, J. P. E. Grolier, F. Dan, S. L. Randzio, Thermochim. Acta 2006, 445,
121–125; d) J. Baumann, F. Baitalow, Thermochim. Acta 2005, 430, 9–
14.
[9] a) C. W. Hamilton, R. T. Baker, A. Staubitz, I. Manners, Chem. Soc. Rev.
2009, 38, 279–293; b) U. Sanyal, U. B. Demirci, B. R. Jagirdar, P. Miele,
ChemSusChem 2011, 4, 1731–1739; c) H. L. Jiang, S. K. Singh, J. M. Yan,
X. B. Zhang, Q. Xu, ChemSusChem 2010, 3, 541–549.
[10] J. Hannauer, O. Akdim, U. B. Demirci, C. Geantet, J. M. Herrmann, P.
Miele, Q. Xu, Energy Environ. Sci. 2011, 4, 3355–3358.
[11] T. Hꢂgle, M. F. Kꢂhnel, D. Lentz, J. Am. Chem. Soc. 2009, 131, 7444–
7446.
[12] R. Moury, G. Moussa, U. B. Demirci, J. Hannauer, S. Bernard, E. Petit, A.
van der Lee, P. Miele, Phys. Chem. Chem. Phys. 2012, 14, 1768–1777.
[13] H. Wu, W. Zhou, F. E. Pinkerton, T. J. Udovic, T. Yildirim, J. J. Rush, Energy
Environ. Sci. 2012, 5, 7531–7535.
[14] Y. Filinchuk, D. Chernyshov, V. Dmitriev, Z. Kristallogr. 2008, 223, 649–
659.
NaHB and the solid residues were characterized by solid-state 11B
MAS NMR (Varian VNMR400, 1H 400 MHz, 11B 128.37 MHz, 23Na
105.86 MHZ, À108C, 20000 rpm), FTIR (Nicolet 710, 128 scans),
thermogravimetric (Rigaku, TG8120), and DSC (DSC1 Mettler
Toledo) analyses.
XRD data were collected at the Swiss–Norwegian Beam Lines at
the ESRF research institute by means of the Pilatus 2M detector
using a wavelength of 0.69412 ꢁ. Indexing was performed using
the DICVOL04 program.[26] The structure was solved by parallel
tempering in FOX,[27] and refined by the Rietveld method with the
FullProf suite.[28] The structure shown in Figure 4 fitted the diffrac-
tion pattern well (see the Supporting Information, Figure S2) with
an agreement factor (Rwp) of 10.0% and a Bragg R-factor (RB) of
5.4%.
The dehydrogenation of NaHB under prolonged heating at con-
stant temperatures (80, 90, 85, 100, and 1108C) was performed on
our thermolysis bench. A flask was linked to a Schlenk line and
was immersed in an oil bath heated to the desired temperature.
The outlet of the flask was connected to an inverted burette filled
with colored water through a cold trap maintained at 08C. The
whole bench was maintained under a N2 atmosphere. The H2 evo-
lution was video-recorded and the data plotted using the Matlab
software.
The purity of H2 was analyzed by mass spectrometry (Anelva, M-
QA200TS) coupled to the TGA apparatus and by GC analysis (Shi-
madzu, GC-14B). The gases collected in a Tedlar bag were also ana-
lyzed by FTIR (Nicolet Nexus, high-resolution detector MCT/A,
1024 scans, resolution of 0.1 cmÀ1).
[15] a) Z. Xiong, C. K. Yong, G. Wu, P. Chen, W. Shaw, A. Karkamkar, T. Autrey,
M. O. Jones, S. R. Johnson, P. P. Edwards, W. I. F. David, Nat. Mater. 2008,
7, 138–141; b) S. Chua, W. Li, W. J. Shaw, G. Wu, T. Autrey, Z. Xiong,
M. W. Wong, P. Chen, ChemSusChem 2012, 5, 927–931.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 8
&
6
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
[16] Y. S. Chua, G. T. Wu, Z. T. Xiong, T. He, P. Chen, Chem. Mater. 2009, 21,
4899–4904.
[25] a) B. L. Davis, D. A. Dixon, E. B. Garner, J. C. Gordon, M. H. Matus, B.
Scott, F. H. Stephens, Angew. Chem. 2009, 121, 6944–6948; Angew.
Chem. Int. Ed. 2009, 48, 6812–6816; b) A. Sutton, A. K. Burrell, D. A.
Dixon, E. B. Garner, III., J. C. Gordon, T. Nakagawa, K. C. Ott, J. P. Robin-
son, M. Vasiliu, Science 2011, 331, 1426–1429.
[26] a) T. Roisnel, J. Rodrꢄguez-Carvajal, WinPLOTR, Proceedings of the Sev-
enth European Powder Diffraction Conference (EPDIC 7), (Ed.: R. Delhez,
E. J. Mittenmeijer), 2000, 118–123; b) A. Boultif, D. Louer, J. Appl. Crys-
tallogr. 2004, 37, 724–731.
[17] K. J. Fijalłkowski, W. Grochala, J. Mater. Chem. 2009, 19, 2043–2050.
[18] H. Gao, J. M. Shreeve, J. Mater. Chem. 2012, 22, 11022–11027.
[19] T. He, J. Wang, G. Wu, H. Kim, T. Proffen, A. Wu, W. Li, T. Liu, Z. Xiong, C.
Wu, H. Chu, J. Guo, T. Autrey, T. Zhang, P. Chen, Chem. Eur. J. 2010, 16,
12814–12817.
[20] U. B. Demirci, S. Bernard, R. Chiriac, F. Toche, P. Miele, J. Power Sources
2011, 196, 279–286.
[21] P. J. Fazen, E. E. Remsen, J. S. Beck, P. J. Carroll, A. R. McGhie, L. G. Sned-
don, Chem. Mater. 1995, 7, 1942–1956.
[27] V. Favre-Nicolin, R. Cerny, J. Appl. Crystallogr. 2002, 35, 734–743.
[28] J. Rodrꢄguez-Carvajal, Physica B 1993, 192, 55–69.
[22] A. C. Stowe, W. J. Shaw, J. C. Linehan, B. Schmid, T. Autrey, Phys. Chem.
Chem. Phys. 2007, 9, 1831–1836.
[23] K. Shimoda, Y. Zhang, T. Ichikawa, H. Miyaoka, Y. Kojima, J. Mater. Chem.
2011, 21, 2609–2615.
[24] D. Neiner, A. Karkamkar, J. C. Linehan, B. Arey, T. Autrey, S. M. Kauzlarich,
J. Phys. Chem. C 2009, 113, 1098–1103.
Received: October 23, 2012
Revised: December 19, 2012
Published online on && &&, 0000
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 8
&
7
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
R. Moury, U. B. Demirci,* T. Ichikawa,
Y. Filinchuk, R. Chiriac, A. van der Lee,
P. Miele
Borane’s been busted: Sodium hydrazi-
nidoborane (NaN2H3BH3) is a new
boron-based material, derived from hy-
drazine borane (N2H4BH3). It has been
mechano-synthesized and successfully
characterized. Under prolonged heating,
it is able to release almost 3 equiv of H2
at different kinetic rates, depending on
temperature. This makes NaN2H3BH3
a potential candidate for use as a chemi-
cal hydrogen-storage material.
&& – &&
Sodium Hydrazinidoborane: A
Chemical Hydrogen-Storage Material
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 8
&
8
&
ÞÞ
These are not the final page numbers!