Calcium borohydride for hydrogen storage: Catalysis and reversibility

Article abstract of DOI:10.1021/jp0764541

We demonstrate a new solid-state synthesis route to prepare calcium borohydride, Ca(BH₄)₂ by reacting a ball-milled mixture of CaB₆ and CaH₂ in a molar ratio of 1:2 at 700 bar of H₂ pressure and 400-440°C. Moreover, doping with catalysts was found to be crucial to enhance reaction kinetics. Thermogravimetric analysis and differential scanning calorimetry revealed a reversible low-temperature to high-temperature endothermic phase transition at 140°C and another endothermic phase transition at 350-390°C associated with hydrogen release upon formation of CaB₆ and CaH₂, as was evident from X-ray diffraction analysis. Thus, since Ca(BH₄)₂ here is shown to be prepared from its anticipated decomposition products, the conclusion is that it has potential to be utilized as a reversible hydrogen storage material. The theoretical reversible capacity was 9.6 wt % hydrogen.

Full text of DOI:10.1021/jp0764541

12045
2007, 111, 12045-12047
Published on Web 10/03/2007
Calcium Borohydride for Hydrogen Storage: Catalysis and Reversibility
Ewa Ro1nnebro* and Eric H. Majzoub
Sandia National Laboratories, 7011 East AVenue, LiVermore, California 94551-0969
ReceiVed: August 10, 2007; In Final Form: September 8, 2007
We demonstrate a new solid-state synthesis route to prepare calcium borohydride, Ca(BH4)2, by reacting a
ball-milled mixture of CaB6 and CaH2 in a molar ratio of 1:2 at 700 bar of H2 pressure and 400-440 °C.
Moreover, doping with catalysts was found to be crucial to enhance reaction kinetics. Thermogravimetric
analysis and differential scanning calorimetry revealed a reversible low-temperature to high-temperature
endothermic phase transition at 140 °C and another endothermic phase transition at 350-390 °C associated
with hydrogen release upon formation of CaB6 and CaH2, as was evident from X-ray diffraction analysis.
Thus, since Ca(BH4)2 here is shown to be prepared from its anticipated decomposition products, the conclusion
is that it has potential to be utilized as a reversible hydrogen storage material. The theoretical reversible
capacity was 9.6 wt % hydrogen.
Hydrogen storage is accepted to be a key enabling technology
for the advancement of hydrogen and fuel cell power technolo-
gies in transportation, stationary, and portable applications. The
U.S. Department of Energy (DOE) is focusing the efforts
primarily on on-board vehicular hydrogen storage systems that
will allow for a driving range of at least 300 miles. This has
stimulated many research groups to attempt to find new practical
metal hydrides or other hydrogen storage materials that will
meet the U.S. DOE system target of 6 wt % of reversible
hydrogen capacity at around 100 °C and 1-10 atm.1 With
respect to the weight percent of hydrogen, the most promising
class of metal hydrides to meet this target is the complex
hydrides. Recently investigated materials with high gravimetric
densities include alanates,2 amides,3 and metal borohydrides.4
We will here focus on the metal borohydrides, specifically
calcium borohydride. Among the metal borohydrides, the
features of the alkali borohydrides, that is, LiBH4, NaBH4, and
KBH4, are relatively well explored.5 While sodium and potas-
sium borohydrides cannot be rehydrided, lithium borohydride
has been shown by Muller et al. to be hydrogenated at 600-
650 °C at 100-160 bar.6 By destabilizing LiBH4 with MgH2
upon formation of MgB2, LiH, and hydrogen gas, Vajo et al.7
showed a system capable of reversibly storing 8-10 wt % of
hydrogen at 350-450 °C. Also, Mg(BH4)2 has been investigated
for its reversible properties,8,9 but so far, there is no evidence
for reversibility in the literature. Another interesting high-
hydrogen-content compound is Ca(BH4)2, but little information
is available regarding fundamental properties. Calcium boro-
hydride can be prepared by reacting calcium hydride10 or
alkoxides11 with diborane, which is a poisonous gas. Another
preparation method is reaction in THF,12 forming Ca(BH4)2-
(THF)2, which is commercially available. Recently, calcium
borohydride was more conveniently prepared by reacting MgB2
with CaH2 at 350 bar of H2 pressure and 400 °C upon obtaining
MgH2 as a byproduct;13 thus, the maximum hydrogen capacity
would be 8.3 wt %. Nakamori et al.14 prepared calcium
borohydride by ball milling LiBH4 with CaCl2, obtaining LiCl
as a byproduct, thus preventing reversibility. We will here for
the first time demonstrate how to prepare calcium borohydride
from its decomposition products without forming byproducts,
thus obtaining a theoretical reversible capacity of 9.6 wt % of
hydrogen.
Theoretical calculations showed that calcium boro-
hydride may form by reacting CaB6 with CaH2, resulting
in a reaction enthalpy (at T ) 0 K) of ∼53 kJ (mol H2)-1 for
the following reaction:15
CaB6(s) + 2CaH2(s) + 10H2(g) f 3Ca(BH4)2(s)
This encouraged us to prepare this material by solid-state
reactions. We discovered that it was possible to prepare calcium
borohydride by following a specific synthesis route, previously
not reported in the literature. The synthesis route involved three
consecutive steps, which could not be excluded in order to
prepare calcium borohydride. The starting materials were CaB6
(99.5% Alfa Aesar) and CaH2 (99.99%, Aldrich), mixed in a
1:2 molar ratio. All sample handlings were performed in an
argon-filled glove box monitored to have oxygen and water
levels below 1 ppm. First, the mixture of CaB6 and CaH2 was
ball milled in a SPEX mill for 30 min. Second, the powder
was mechanically pressed into a pellet of 10 mm in a hydraulic
Carver press to form better contact between the particles. Third,
the pellet was placed in a steel crucible further enclosed in a
commercial AE Closure (Snap-tite, Inc.), which was attached
to a high-pressure hydrogenation system and placed in a furnace.
The autoclave was initially pressurized to 350 bar of H2 pressure
before being heated up to 400-440 °C, and the obtained
* To whom correspondence should be addressed. E-mail:
ecronne@sandia.gov.
10.1021/jp0764541 CCC: $37.00 © 2007 American Chemical Society

12046 J. Phys. Chem. B, Vol. 111, No. 42, 2007
Letters
Figure 1. Powder X-ray diffraction pattern of a sample made by reacting CaB6 + 2CaH2 + 4 wt % of TiCl3 + 4 wt % of Pd at 70MPa and 440
°C. The sample contains Ca(BH4)2 (|) + CaB6 (O) + CaH2 (b) + CaO (x).
The peak at 2280 cm-1 belongs to the strongest BH4 vibration
in Ca(BH4)2.
Bogdanovic et al.2 showed that by adding TiCl3, sodium
alanate, NaAlH4, can be made reversible, obtaining 4.5 wt %
at 100-200 °C. Among tried dopants to facilitate the reaction
above were TiCl3, Pd, TiCl3 + Pd, and RuCl3. With only TiCl3
added, a small amount of Ca(BH4)2 formed, as was evident from
estimating the amount of the phases by comparing relative
intensities in the XRD pattern. With a mixture of TiCl3 + Pd,
we obtained a much higher yield of ∼60% Ca(BH4)2 in relation
to CaB6 and CaH2, and the pellet changed color from brown to
gray. The XRD pattern is shown in Figure 1. By adding RuCl3,
Figure 2. Raman spectra collected from (a) high-pressure synthesis
a slightly better yield as compared to that of the TiCl3 + Pd
mixture was obtained. Interesting to note is that, depending on
the added catalyst/dopant, other new phases form in addition
or instead of Ca(BH4)2. Only the catalyst mixture of TiCl3 +
Pd resulted in Ca(BH4)2 of the Ba(MnO4)2 structure type, with
no other new XRD peaks present (except for unreacted CaB6
and CaH2). We anticipate that these other phases are polymorphs
of Ca(BH4)2. An extensive investigation has been initiated to
determine the reaction mechanism. All of the samples contained
a significant amount of CaO (10-20 wt %), concluded to
originate from impurities in the lines between the hydrogen
bottle and the autoclave.
of solid-state Ca(BH4)2 and (b) a CaB6 reference sample. Raman
spectroscopy of a synthesized sample of Ca(BH4)2 showing the shifts
from the [BH4-] anion at 2280 cm-1
.
pressure was adjusted to 700 bar and kept for 48 h. Importantly,
a chemical reaction will not occur unless a dopant of 4-8 wt
% is added to the initial mixture of CaB6 and CaH2. As evident
from powder XRD analysis, the appearance of new X-ray peaks
indicated a chemical reaction. X-ray data were collected of
powders contained in 0.7 mm capillaries on a rotating anode
Rigaku diffractometer (RU-300) with a Cu target at 40 kV and
40 mA at 295 K. The new peaks were indexed by utilizing the
trial-and-error program TREOR16 to find the unit cell param-
eters. Thereafter, by refining the unit cell in Pirum,17 the crystal
structure was determined to be face-centered orthorhombic, with
unit cell parameters a ) 13.07(2), b ) 8.700(2), and c ) 7.52(2)
Å (V ) 854.4 Å3), likely of the Ba(MnO4)2-type, space group
Fddd (No. 70), in accordance with the theoretical predictions
by Vajeeston et al.18 and the Rietveld analysis by Miwa et al.19
of an adduct-free sample of calcium borohydride. The XRD-
pattern is shown in Figure 1. The samples were here also
analyzed by Raman spectroscopy, TGA, and DSC. Raman data
of a reference sample of CaB6 and a sample of synthesized
calcium borohydride were collected on a Spex model 1877 0.6
m triple spectrometer, using the 514 nm line of a Coherent
Innova Ar ion laser at a power of about 10 mW at the sample.
Raman spectroscopy showed the presence of the [BH4-] anion
in the spectra, in accordance with the literature.20 The Raman
peaks in Figure 2 at 785, 1150, and 1285 cm-1 belong to CaB6.
In order to investigate the desorption behavior, a sample of
synthesized calcium borohydride was analyzed by thermal
gravimetry analysis (TGA) and differential scanning calorimetry
(DSC) using Mettler Toledo instruments TGA/SDTA851 and
DSC822. The heating rate was 5 K/min-1 in a 20 mL/hour slow
stream of argon at ambient pressures. Figure 3 shows the
experimental curves from analysis of a sample (<5 mg) with a
composition similar to the XRD pattern in Figure 1. The DSC
curve shows two phase transitions at onset temperatures of 140
and 350 °C. The TGA curve shows a weight loss at 350-390
°C, which is most likely related to hydrogen release from
calcium borohydride. The increase in the TGA curve has been
confirmed to be due to background effects. The decomposition
products were predicted to be CaB6 and CaH2 by Konoplev et
al.21 and Miwa et al.19 In order to verify this assumption, we
collected XRD data at two points on the curve; samples were
heated up to 150 and 400 °C, respectively, and XRD data were
collected for cooled down samples. At 150 °C, the XRD pattern

Letters
J. Phys. Chem. B, Vol. 111, No. 42, 2007 12047
Figure 3. DSC curve (- - -) and TGA curve (s) showing two endothermic phase transformations at onset temperatures of 140 and at 350 °C and
release of hydrogen at 350 °C.
is similar to the starting sample; thus, there is a reversible phase
transition from a low-temperature form to a high-temperature
form, similar to that for LiBH4. However, at 400 °C, the XRD
pattern corresponds to CaB6 and CaH2 without Ca(BH4)2
present. Thus, upon release of 9.6 wt % of hydrogen,
we presume that calcium borohydride has completely decom-
posed to form CaB6 and CaH2, which we now know is the
starting mixture to synthesize calcium borohydride at 700 bar
(70MPa) and 440 °C. The above reaction formula is thus
partially reversible at the present described reaction conditions.
In conclusion, this material has potential to be used as a
reversible hydrogen storage material for on-board storage.
High-pressure tanks up to 800 bar have been shown
to be feasible for utilizing high-capacity materials, as pre-
sented by Hardy et al.22 during the 2007 DOE annual review
meeting.
References and Notes
(1) U.S. Department of Energy Hydrogen Program. http://
www.hydrogen.energy.gov.
(2) Bogdanovic, B.; Schwickardi, M. J. Alloys Compd. 2002, 339, 299.
(3) Chen, P.; Xiong, X.; Lou, J.; Lin, J.; Tan, K. L. Nature 2002, 420,
302.
(4) Zu¨ttel, A.; Wenger, P.; Rentsch, S.; Sudan,. P.; Mauron, Ph.;
Emmenegger, Ch. J. Power Sources 2003, 118, 1.
(5) Renaudin, G.; Gomes, S.; Hagemann, H.; Keller, L.; Yvon, K. J.
Alloys Compd. 2004, 375, 98.
(6) Muller, A.; Havre, L.; Mathey, F.; Petit, V. I.; Bensoam. J. U.S.
Patent 4,193,978, 1980.
(7) Vajo, J. J.; Skeith, S. L.; Mertens, F. J. Phys. Chem. B 2005, 109,
3719.
(8) Chlopek, K.; Frommen, C.; Le´on, A.; Zabara, O.; Fichtner, M. J.
Mater. Chem. 2007, 33, 3496.
(9) Zhao, J.-C.; Andrus, M.; Cui, J.; Gao, Y.; Kniajansky, S.; Lemmon,
J.; Raber, T.; Rijssenbeek, J.; Rubinsztajn, G.; Soloveichik, G. DOE Annual
Merit Review Proceedings, Hydrogen Storage Program, Metal Hydride
Center of Excellence. http://www.hydrogen.energy.gov/pdfs/review07/
st_16_zhao.pdf (2007).
(10) Wiberg, E.; Hartwimmer, R. Z. Z. Naturforsch., B: Chem. Sci.
1955, 10, 295.
(11) Wiberg, E.; Noth, H.; Hartwimmer, R. Z. Z. Naturforsch., B: Chem.
Sci. 1955, 10, 292.
(12) Mikheeva, V. I.; Titov, L. V. Zh. Neorg. Khim. 1964, 9, 789.
(13) Barkhordarian, G.; Klassen, T.; Dornheim, M.; Bormann, R. J.
Alloys Compd. 2007, 440, L18.
(14) Nakamori, Y.; Li, H.-W.; Kikuchi, K.; Aoki, M.; Miwa, K.; Towata,
S.; Orimo, S. J. Alloys Compd. 2007, in press.
In summary, we have demonstrated a unique solid-state
synthesis route to prepare calcium borohydride, Ca(BH4)2, for
the first time without forming byproducts. The yield relative to
the starting materials is estimated to 60%. We have also
demonstrated that this chemical reaction can be reversed with
a theoretical capacity of 9.6 wt %; the anticipated decomposition
products, that is, CaB6 and CaH2, can be rehydrided under high-
hydrogen pressures and moderate temperatures when a catalyst
is added, thus forming calcium borohydride. Next, we will
explore structural changes upon decomposition in detail.
Moreover, we will optimize the reaction conditions at lower
pressures and temperatures and continue exploring the effect
of catalysts on kinetics.
(15) Majzoub, E. H. Unpublished result using first-principles DFT
calculations at T ) 0 K.
(16) Werner, P.-E.; Eriksson, L.; Westdahl, M. J. Appl. Crystallogr.
1985, 18, 367.
(17) Werner, P.-E. Ark. Kemi 1969, 31, 513.
(18) Vajeeston, P.; Ravindran, P.; Fjellvåg, H. J. Alloys Compd. 2007,
in press.
(19) Miwa, K.; Aoki, K.; Noritake, T.; Ohba, N.; Nakamori, Y.; Towata,
S.; Zu¨ttel, A.; Orimo, S. Phys. ReV. B 2006, 74, 155122.
(20) Gomes, S.; Hagemann, H.; Yvon, K. J. Alloys Compd. 2002, 346,
206.
Acknowledgment. Funding was provided by the U.S.
Department of Energy, Office of Energy Efficiency and
Renewable Energy under the Hydrogen Storage Grand
Challenge, Center of Excellence within DOE’s National Hy-
drogen Storage Project. We thank Ray Baldonado and Dennis
Morrison for skillful technical assistance on the high-pressure
experiments.
(21) Konoplev, V. N.; Sizareva, A. S. Koord. Khim. 1952, 18, 508.
(22) Hardy, B. J.; Wu, T.; Gorcyzca, J. L.; Anton, D. L. DOE Annual
Merit Review Proceedings, Hydrogen Storage Program, Metal Hydride
Center of Excellence. http://www.hydrogen.energy.gov/pdfs/review07/
stp_37_hardy.pdf (2007).