A self-contained regeneration scheme for spent ammonia borane based on the catalytic hydrodechlorination of BCl3

Article abstract of DOI:10.1002/anie.201201134

Recycling: A self-contained procedure for the recycling of BNH-waste, based on the three major steps: polymer break-up, amine supported catalytic hydrodehalogenation of boron halogens, and the base exchange in borane amine adducts, is developed (see picture). Beyond the original task of recycling spent ammonia borane, the process provides a new means to produce borohydride species efficiently, by the direct use of molecular hydrogen. Copyright

Full text of DOI:10.1002/anie.201201134

Angewandte
Chemie
DOI: 10.1002/anie.201201134
Hydrogen Storage
A Self-Contained Regeneration Scheme for Spent Ammonia Borane
Based on the Catalytic Hydrodechlorination of BCl3**
Christian Reller and Florian O. R. L. Mertens*
More than a decade ago, following the great interest shown by
the automotive industry in hydrogen fuel-cell propelled
vehicles, a quest for chemical-based hydrogen-storage and
hydrogen-source systems exceeding the capacity of classical
metal hydrides and complex metal hydrides began. As
a compound with one of the highest hydrogen contents,
ammonia borane (AB) was quickly identified as a promising
candidate, especially because of its benign hydrogen-release
temperatures starting at 958C.[1,2] The hydrogen release is
exothermic, which explains why direct re-hydrogenation of
spent AB is impossible. This low dehydrogenation temper-
ature and the exothermicity both result from the simulta-
neous presence of hydridic and protic hydrogen in the AB
molecule.
overcome leading to a self-contained recycling procedure,
that is, a procedure that uses molecular hydrogen and spent
AB as input materials and delivers AB utilizing only mass
commodity chemicals as recyclable auxiliary reagents
(Scheme 1). Because the current production of AB requires
Early suggestions of recycling procedures contain three
important steps that pose great challenges for their practical
implementation; namely: 1) the digestion of the polymeric
spent AB by the formation of oxidized, or more specifically,
halogenated boron species, BX3 (X = Cl, Br, I), 2) hydro-
genation, or more precisely, the hydrodehalogenation of these
compounds, and 3) replacement of the auxiliary reagents
(bases) used as thermodynamic drivers in the hydrodehalo-
genation process by ammonia. An alternative recycling
strategy was recently presented by Sutton et al., which is
based on the strong reducing power of hydrazine.[3] The
procedure has the advantage that the regeneration can be
carried out as a one-pot synthesis. However, the use of
energetically demanding, dangerously unstable, and toxic
hydrazine does not make the process appear to be favorable
on an industrial scale.[4] In industry, hydrazine is almost
always, except for the production of rocket fuel, used as
a hydrate rather than a pure compound because of safety
concerns. In the hydrazine-based recycling scheme, however,
it cannot be used as a hydrate because of the sensitivity to
hydrolysis of the hydrazine borane adduct.[5] Consequently,
a recycling scheme using inexpensive and comparably safe
reagents is still highly desired.
Scheme 1. Regeneration of BNH waste based on the catalytic hydro-
dechlorinaton of BCl3. Dashed lines indicate the recycling pathways of
the by-products. Solid boxes indicate the results of the key processes,
BNH break-up, addition of thermodynamic driver, hydrodechlorination,
and base exchange.
the use of expensive hydrides, the demonstrated recycling
scheme also strikes a new path to potentially produce AB and
other boranes more economically. Given the importance of
the three major road blocks for realization of the AB
regeneration scheme outlined above, discussion of the
BNH-waste break-up, hydrodehalogenation of BCl3 species,
and base exchange will each start with a short overview
defining the main obstacles.
Herein we demonstrate that the three major obstacles for
the implementation of the early recycling schemes based on
the break-up of spent AB by boron halogenation can be
BNH-waste break-up: Several groups suggested proce-
dures consisting of a digestion step preceding the separate
hydrogenation of the boron- and nitrogen-carrying moiet-
ies.[6–8] One fundamental problem for breaking up spent AB
lies in the very different waste products that result from the
various dehydrogenation processes. For example, if AB is
decomposed by thermal treatment at 958C, the main compo-
nent of the residue was identified as polyaminoborane
(BH2NH2)x.[9] Thermal treatment at 1508C of AB or
(BH2NH2)x leads to the formation of cross-linked polybor-
azylene (PB).[10–12] Another product obtained from the
dehydrogenation processes is borazine, which is equally
difficult to break-up into smaller borane- and nitrogen-
[*] C. Reller, Prof. Dr. F. O. R. L. Mertens
Department of Physical Chemistry
Technische Universitꢀt Bergakademie Freiberg
Leipziger Strasse 29, 09599 Freiberg (Germany)
E-mail: florian.mertens@chemie.tu-freiberg.de
[**] We thank Mr. Klaus Heinemann from Ullner and Ullner GmbH for
technical support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201201134.
Angew. Chem. Int. Ed. 2012, 51, 11731 –11735
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
11731

.
Angewandte
Communications
containing units because of its tendency to dehydropolymer-
ize. If the dehydrocoupling is performed in solution using
certain solvents and catalysts, the main product is also
PB.[13–15] A possible means to break the stable B N bonds in
ꢀ
these compounds is the hydrolysis of the spent fuel leading to
[6,8]
ꢀ
the formation of even more stable B O bonds.
Although
ꢀ
ꢀ
the conversion of B O bonds into B H bonds is in principle
possible using strong reducing agents, that is, metal hydrides
such as LiAlH4, the problem is then only shifted to the
recycling of other high energy materials. We therefore
focused on break-up procedures aiming at the formation of
ꢀ
ꢀ
B X bonds (X = Cl, Br, I), which are weaker than B O
bonds, assuming that these systems would be more easily
ꢀ
accessible for the catalytic generation of B H bonds. In
contrast to the hydrazine-based recycling procedure, the
successful recycling scheme should also be capable of digest-
ing strongly dehydrogenated AB, that is, heavily cross-linked
polyborazylene which is insoluble in common solvents. To
achieve this goal the Sneddon group suggested the use of
Brønsted acid (HBr and HCl) based superacids.[6] To date, the
systems AlBr3/HBr/CS2 and AlCl3/HCl/toluene were inves-
tigated, but no significant break-up of strongly dehydrogen-
ated AB materials, such as borazine or PB was achieved by
the superacid treatment.[6,16]
Since the difference in the break-up between less-
dehydrogenated material, such as (BH2NH2)x, and a more
strongly dehydrogenated one has proven to be of importance,
we took great care to obtain the more strongly dehydro-
genated type of material in a well-defined and reproducible
way. We therefore first dehydrogenated AB at 958C for 24 h
to obtain predominantly (BH2NH2)x (checked by NMR and
diffuse reflectance infrared Fourier transform (DRIFT)
spectroscopy, see Figure S2 and S4 in the Supporting Infor-
mation) and then continued the dehydrogenation at 1508C
for 20 h under vacuum (10ꢀ3 mbar) to obtained strongly cross-
linked PB that is not soluble in THF (checked by DRIFT;
Figure S4, S1).
Herein, we report an improved procedure for the
regeneration of strongly dehydrogenated species, such as
borazine, PB, and cross-linked PB by using the AlCl3/HCl/CS2
superacid/solvent system running at 40 bar HCl pressure and
808C. The superacid protonates the nitrogen atoms in the
material and the excess of HCl ensures chlorination of all
boron atoms leading to the formation of NH4Cl and BCl3 with
a yield of 90% BCl3 (for details concerning the quantification
procedures please see Section S 6.2 of the Supporting
Information). The formation of NH4Cl was confirmed by
powder X-ray diffraction (PXRD; Figure S6) and that of BCl3
was quantitatively determined from the integration of the
11B NMR signals. The monitoring of the process by 11B NMR
spectroscopy is displayed in Figure 1. During the first 3 h of
the reaction almost no soluble components were detected in
the reaction solution. After another hour, signals at d =
32 ppm, d = 29.8 ppm, and d = 25.4 ppm appeared from
which the formation of B-chlorinated borazanaphthalene
(B5N5H4Cl4; see Figure S5) and of smaller soluble chlorinated
PB fragments can be concluded.[17,18] The chlorination of
borazine to 2,4,6-trichloroborazine (BCl3NH)3 was conducted
as a control experiment, which, beside the possibility to
Figure 1. Temporal development of the digestion of cross-linked poly-
borazylene by the superacid system HCl/AlCl3/CS2, recorded by
11B NMR (C6D6) spectroscopy.
break-up this compound also showed that the only effect of
the chlorination is the conversion of the borazine doublet
11B{1H} NMR at d = 32 ppm (see Figure S3) into a singlet
essentially with no effect on the signal position. This fact
allows us to conclude that a significant amount of chlorinated
BN ring species has formed after 4 h.
The reaction proceeds by the formation of BCl3, detected
by the development of the intense signal at d = 46 ppm and
the fading out of the original resonance signals. The overall
yield of the reaction is greater than 90% BCl3. The solid
residue was determined by PXRD (Figure S6) to be AlCl3 and
ammonium chloride. Its inductively coupled plasma–optical
emission spectroscopy (ICP-OES) analysis gave a boron
content of 0.03 wt%.
All volatile components were collected by a cryo trap
allowing the determination of the overall yield to be more
than 90% (by 11B NMR spectroscopy and ICP-OES). We also
tested if a break-up of less-dehydrogenated material, in this
case (BH2NH2)x, is achievable at milder conditions. For this
material, a yield of BCl3 of at least 80% was accomplished at
ambient temperature and 30 bar HCl pressure within 4 h. For
borazine, a yield of 90% BCl3 at pHCl = 30 bar and 808C was
obtained in 4 h. Thus, the applicability of the superacid
digestion is demonstrated for essentially all types of relevant
spent AB material.
Hydrodehalogenation of BCl3: The second major obstacle
to an efficient recycling route for spent AB based on the
superacid concept is the successful hydrodehalogenation of
the halogenated boron species, in this case BCl3, formed
during the spent-fuel break-up. Since the direct hydrogena-
tion of BCl3 to (di)borane is thermodynamically unfavorable
1
(BCl3 + 3H2! = B2H6 + 3HCl
DRG8(298 K) =+
2
148.0 kJmolꢀ1),[19] the reaction will need an auxiliary reactant
as a thermodynamic driver. It is clear that for a self-contained
regeneration scheme, this auxiliary reactant needs to be easily
recoverable.[20] For the proof of concept, we chose the
ꢀ
relatively strong Lewis base triethylamine that forms a B N
adduct with BCl3. The regeneration scheme was intended to
11732 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 11731 –11735

Angewandte
Chemie
avoid expensive or difficult to handle (solid, toxic etc.)
hydrogenation compounds, such as metal hydrides or hydra-
zine, and to conduct the hydrogenation preferably with
molecular hydrogen. As a comparably inexpensive hydro-
genation catalyst, finely dispersed nickel appeared to be
a possible choice, but elemental nickel forms nickel borides in
the presence of Et3NBH3. Since nickel boride itself was shown
to be a catalyst for the hydrogenation of alkenes,[21] with the
boron serving as a structural promoter to prevent the
coarsening of the finely dispersed material, a phenomenon
observed, for example, in the case of Raney nickel,[22] we
decided to try Ni3B as the catalyst for the hydrodechlorination
step. From the presence of chloride, a certain risk of poisoning
the nickel boride catalyst by the irreversible formation of
nickel chloride surface species must be assumed. The com-
parably fast kinetics (Figure 2) over the course of the reaction
fast for well-resolved curves to be obtained by extracting and
characterizing individual samples from the reaction solution
in the autoclave. Also, at 1808C, the established decomposi-
tion of all triethylamine adducts, that is, BClxH3ꢀxNEt3 (x =
1,2,3), under the formation of molecular HCl, ethene, and
cyclic N-alkyl-B-chloroborazines sets in.[23,24]
In Figure 3 the hydrodechlorination of BCl3 at 1308C and
60 bar hydrogen pressure is shown (for a corresponding
experiment at 1608C see Figure S15 in the Supporting
Information). In experiments leaving out only the catalyst
Figure 2. Temporal development of the stepwise hydrodechlorination
Figure 3. a) Comparison of the 1H NMR (D2O) coupled spectra of
BH3NH3 obtained from the recycling process and produced by the
standard salt metathesis.[25] b) The 11B{1H} NMR (D2O) spectrum of
the product of the NEt3–NH3 exchange reaction.
of Et3NBCl3 using molecular hydrogen and Ni3B as a catalyst.
T=1308C, p(H2)=60 bar.
1
and the fact that X-ray photoelectron spectroscopy (XPS)
investigations of the catalyst after the completion of the
reaction did not show any signs of chloride formation,
illustrate that this type of catalyst poisoning is not relevant
for this system. It is safe to assume that the presence of the
large amounts of triethylamine protects the catalyst. Some
information on the cycle life and catalyst leaching can be
found in the Supporting Information.
Hydrodechlorination experiments with triethylamine in
various solvents showed that an excess of this base also helped
to prevent the formation of Et3NHBCl4 as an ionic by-
product. Consequently, we conducted subsequent experi-
ments solvent free, that is, with triethylamine also assuming
the role of the solvent as well as its role as a thermodynamic
driver.
Figure 2 illustrates that the reaction proceeds stepwise
from Et3NBCl3 via the intermediates Et3NBHCl2 and
Et3NBH2Cl to the product Et3NBH3. Some optimization
with respect to the reaction rate and the formation of by-
products was carried out by investigating the reaction at
60 bar hydrogen pressure and T= 60, 100, 139, 1608C. The
corresponding maxima of the first intermediate, the
Et3NBHCl2 adduct, were reached at 600, 270, 90, 30 min,
respectively. At 1808C the reaction already proceeded too
or the hydrogen gas, no hydrodechlorination was detected,
which demonstrates the decisive role of these two compo-
nents. Long-term experiments show that the equilibrium even
in the case of the 1308C experiment is practically reached
within 48 h. (Equilibrium composition Et3NBHCl2/
Et3NBH2Cl/Et3NBH3: (at 1308C) 12%/70%/18%, (at
1608C) 0%/70%/30%). A single subsequent removal of the
produced Et3NHCl and continued reaction, shifts the equi-
librium composition to 15%/40%/45% at 1308C and 0%/
52%/48% at 1608C.
Base exchange: The last key step to arrive at the final
product AB is the exchange of the auxiliary base in the
borane adduct for ammonia. Although this step is just
a simple exchange reaction, reports in the literature demon-
strate that it is also a major problem for the proposed
recycling procedures in organic solvents. For example, it was
reported the AB cannot be precipitated from a Me2EtNBH3/
toluene solution by bubbling ammonia through it at ambient
conditions.[6] Although hydrazine borane should be strongly
favored in respect to the base exchange with ammonia
because of its comparably weak base strength (pKa = 7.94)
compared to that of ammonia (pKa = 9.26), no exchange was
Angew. Chem. Int. Ed. 2012, 51, 11731 –11735
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11733

.
Angewandte
Communications
observed on exposure to liquid ammonia (at ꢀ788C) whereas
a complete exchange was achieved at 608C within 24 h.[3]
From these results it is clear that for using NEt3 (pKa =
10.6) as auxiliary base, a large excess of ammonia, best put
into practice by working solvent free with pure ammonia, and
elevated temperatures are required. Since the product AB
starts to dehydrogenate at 958C, a temperature of 808C was
chosen as the upper limit. The result of such a base-exchange
experiment conducted by exposing the Et3NBH3 adduct to
liquid ammonia at 808C for 12 h is displayed in Figure 3. At
room temperature, however, the exchange rate is very low
and both types of adducts (Et3NBH3, H3NBH3) are stable
break-up procedure. A comparably simple technical imple-
mentation for each step in the recycling scheme can always be
found.
To our knowledge, the process developed demonstrates
the first hydrodehalogenation of a BX3 species based on the
catalytic activation of molecular hydrogen. It will allow
applications other than the one presented, such as the
ꢀ
generation of valuable B H containing species applicable as
hydrogenation agents in organic synthesis, without the use of
consumable metal hydrides.
even when exposed to pure liquid base of the other type. Experimental Section
All reactants were handled under argon. The digestion processes were
Consequently, after thermal quenching by ammonia release,
the adduct obtained from the base-exchange step can be
separated. The comparison of the 1H NMR (D2O) spectrum:
d = 1.37 ppm (q, 1JBH = 92 Hz) of the compound obtained
with that of pure AB substantiates its formation. Weak signals
of the reactant NEt3, the triplet at d = 1.27 ppm, and a quartet
at d = 3.17 ppm, are still present in the spectrum. Evidence for
the synthesis of the desired product AB is given by the
occurrence of the quartet at d = ꢀ24 ppm in the 11B{1H} NMR
spectrum with a coupling constant of 1JB,H = 92 Hz, a value in
the typical range for BH3 moieties. The absence of the BH3
quartet in the 11B{1H} NMR spectrum at d = ꢀ12.1 ppm for
Et3NBH3 gives evidence that the exchange of NEt3 for
ammonia was complete.
conducted in 15 mL stainless steel pressure vessel (alloy 59) using
magnetic stirring. For the hydrodechlorination experiments the same
autoclave was used with a PTFE insert. Solvents were purchased from
ABCR and distilled from CaH2 and stored under Argon. Aluminum
chloride (ABCR) was sublimed prior to use. Gases: HCl 5.0 (Linde),
Hydrogen 5.0 (Praxair), NH3 3.8 (Linde), and BCl3 5.0 (Linde) were
used without further purification.
NMR spectroscopy: All NMR investigations were carried out at
room temperature. 11B and 1H NMR spectra were recorded with
a Bruker Avance III 500 MHz spectrometer. The 11B shifts were
referenced to BF3OEt2 (d = 0 ppm). For all samples spectra were
measured with and without decoupling. NMR solvents were used
without further purification.
Digestion of polyborazylenes (PB): PB (33 mg) was suspended in
CS2 (5 mL) and mixed with AlCl3 (350 mg). The mixture was then
kept at 808C and under 40 bar of HCl for 12 h.
As a test of the overall procedure, including all separation
steps, we subjected a 40 mg sample of spent AB (PB) to the
full recycling sequence, that is, BNH break-up and formation
of BCl3, hydrodechlorination (without removal of Et3NHCl),
BH2Cl disproportionation (conversion according to
2BH2Cl!BHCl2 + 0.5B2H6), and base exchange, and
obtained 25.7 mg AB (total yield 60%, details and thermo-
dynamic considerations see the Supporting Information).
In summary, it was demonstrated that the superacid/
solvent system HCl/AlCl3/CS2 is capable of completely
digesting borazine, polyaminoborane, and even cross-linked
PB, the cross-linked PB being the form of the most strongly
dehydrogenated AB material. The digestion is accompanied
by the simultaneous conversion of the boron content of these
compounds into volatile BCl3 with high yields. The catalytic
hydrodehalogenation was achieved with the Lewis base NEt3
as a thermodynamic driver. In contrast to expectations from
the respective pKa values, the exchange of the auxiliary base
NEt3 in its BH3 adduct for ammonia can be conducted in
a way that is fast and goes to completion. Taking these points
together allows a closed regeneration scheme to be set-up that
combines the advantages of using inexpensive standard
chemicals (HCl, AlCl3, CS2, NEt3, Ni3B) and the generation
The gas phase was then expanded into a cooling trap (ꢀ788C)
which contained CDCl3 (5 mL). Afterwards, the residue was identi-
fied to consist of AlCl3, NH4Cl. In the cooling trap BCl3
(11B{1H} NMR (C6D6): d = 46 ppm (s)) and small amounts of 2,4,6-
trichloroborazine (11B{1H} NMR (C6D6): d = 32 ppm (s)) and B-
chlorinated borazanaphthalene (11B{1H} NMR (C6D6): d = 32 ppm
(s), d = 29.4 ppm (s), d = 25.4 ppm (s) were identified as the only
compounds present. The quantification by ICP-OES showed that
about 90% of PB boron content was converted into BCl3. The
digestion of (BH2NH2)x and borazine was conducted analogously.
Hydrodechlorination of Et3NBCl3: Et3NBCl3 (300 mg; 1.3 mmol)
of and Ni3B (18 mg; 0.1 mmol) were added to triethylamine (7 mL)
and loaded into a sealed pressure vessel with inner PTFE coating.
Samples for 11B NMR analysis were extracted under inert gas
atmosphere every 30 min. The last sample was taken after 24 h.
Exchange of the base triethylamine for ammonia in the adduct
Et3NBH3: Et3NBH3 (0.5 mL; 3.3 mmol) were dissolved in liquid
ammonia (10 mL) and heated in a sealed pressure vessel using
a temperature program for 12 h. Afterwards, the solution was
thermally quenched by quickly releasing the NH3 atmosphere from
the pressurized vessel. Finally, BH3NH3, 11B{1H} NMR (D2O): d =
ꢀ24 ppm 1JB,H = 92 Hz (q), was isolated by removing NEt3 by vacuum
distillation.
See the Supporting Information for further details.
Received: September 10, 2012
Published online: October 24, 2012
ꢀ
of B H bonds by direct hydrogenation, that is, by using
molecular hydrogen. For the recycling scheme to be self-
contained all the by-products need to be easily convertible
into the main reactants without generating new undesired
products. For example, the auxiliary base, in the presented
case NEt3, can be regained by the reaction of its hydrochlo-
ride with NaOH. The sodium hydroxide used in this process is
obtained during the chlorine production by chloralkali
electrolysis necessary for the generation of HCl used in the
Keywords: ammonia borane · BHN waste · boron compounds ·
hydrodechlorination · nickel boride
.
[1] F. Baitalow, J. Baumann, G. Wolf, K. Jaenicke-Rçßler, G.
Leitner, Thermochim. Acta 2002, 391, 159 – 168.
11734 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 11731 –11735

Angewandte
Chemie
[2] M. Bowden, T. Autrey, I. Brown, M. Ryan, Curr. Appl. Phys.
2008, 8, 498.
[3] A. D. Sutton, A. K. Burrell, D. A. Dixon, E. B. Garner, J. C.
Gordon, T. Nakagawa, K. C. Ott, J. P. Robinson, M. Vasiliu,
Science 2011, 331, 1426 – 1429.
[4] J.-P. Schirmann, P. Bourdauducq in Ullmannꢀs Encyclopedia of
Industrial Chemistry, Wiley-VCH, Weinheim, 2000.
[5] R. Moury, G. Moussa, U. B. Demirci, J. Hannauer, S. Bernard, E.
Petit, A. van der Lee, P. Miele, Chem. Phys. 2012, 14, 1768 –
1777.
[6] L. G. Sneddon, Amineborane Based Chemical Hydrogen Stor-
age—Final Report, University of Pennsylvania, web address:
http://www.osti.gov/bridge/purl.cover.jsp?purl = /
1011765A1gU1x/.
[7] 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.
[8] P. V. Ramachandran, P. D. Gagare, Inorg. Chem. 2007, 46, 7810 –
7817.
[9] N. C. Smythe, J. C. Gordon, Eur. J. Inorg. Chem. 2010, 509 – 521.
[10] J. Baumann, F. Baitalow, G. Wolf, Thermochim. Acta 2005, 430,
9 – 14.
[14] W. J. Shaw, J. C. Linehan, N. K. Szymczak, D. J. Heldebrant, C.
Yonker, D. M. Camaioni, R. T. Baker, T. Autrey, Angew. Chem.
2008, 120, 7603 – 7606; Angew. Chem. Int. Ed. 2008, 47, 7493 –
7496.
[15] a) C. A. Jaska, K. Temple, A. J. Lough, I. Manners, J. Am. Chem.
Soc. 2003, 125, 9424 – 9434; b) C. A. Jaska, K. Temple, A. J.
Lough, I. Manners, Chem. Commun. 2001, 962 – 963.
[16] S. Hausdorf, F. Baitalow, G. Wolf, F. O. R. L. Mertens, Int. J.
Hydrogen Energy 2008, 33, 608 – 614.
[17] A. W. Laubengayer, O. T. Beachley, R. F. Porter, Inorg. Chem.
1965, 4, 578 – 582.
[18] P. J. Fazen, E. E. Remsen, J. S. Beck, P. J. Carroll, A. R. McGhie,
L. G. Sneddon, Chem. Mater. 1995, 7, 1942 – 1956.
[19] M. W. Chase et al., JANAF Thermochemical Tables, 3rd ed.
[20] L. G. Sneddon, Amineborane Based Chemical Hydrogen Stor-
age-2008
http://www.hydrogen.energy.gov/annual_review08_
storage.html#chemical.
[21] C. A. Brown, J. Org. Chem. 1970, 35, 1900 – 1904.
[22] V. Bagockij, V. S. Bagockij, Fundamentals of Electrochemistry,
Wiley, Hoboken, 2006.
[23] I. M. Butcher, W. Gerrard, J. Inorg. Nucl. Chem. 1965, 27, 823 –
829.
[11] J. Nylꢀn, T. Sato, E. Soignard, J. L. Yarger, E. Stoyanov, U.
Hꢁussermann, J. Chem. Phys. 2009, 131, 104506.
[12] F. H. Stephens, V. Pons, R. T. Baker, Dalton Trans. 2007, 2613 –
2626.
[24] H. Nçth, B. Wrackmeyer, Nuclear Magnetic Resonance Spec-
troscopy of Boron Compounds, Springer, Berlin, 1978.
[25] M. E. Bowden, G. J. Gainsford, W. T. Robinson, Aust. J. Chem.
2007, 60, 149 – 153.
[13] J. F. Kostka, R. Schellenberg, F. Baitalow, T. Smolinka, F.
Mertens, Eur. J. Inorg. Chem. 2012, 49 – 54.
Angew. Chem. Int. Ed. 2012, 51, 11731 –11735
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11735