B–N compounds for chemical hydrogen storage

Abstract

Hydrogen storage for transportation applications requires high volumetric and gravimetric storage capacity. B–N compounds are well suited as storage materials due to their light weight and propensity for bearing multiple protic (N–H) and hydridic (B–H) hydrogens. This critical review briefly covers the various methods of hydrogen storage, and then concentrates on chemical hydrogen storage using B–N compounds. The simplest B–N compound, ammonia borane (H₃NBH₃), which has a potential 19.6 wt% hydrogen storage capacity, will be emphasised (127 references).

---

Charles W. Hamilton

Charles Wayne Hamilton was born in Houston, Texas, USA. He received his BS in Chemistry from Texas A&M University in May of 2001. During his undergrad, he studied nuclear chemistry at the Cyclotron Institute under the direction of Prof. Joseph Natowitz. He then joined the laboratory of Prof. Joseph Sadighi at the Massachusetts Institute of Technology where he studied oxidation catalysis. He received his PhD in 2007, and subsequently accepted a post-doctoral position at Los Alamos National Laboratory where he currently resides. His current research interests include the formation of molecular catalysts for amine borane dehydrogenation.

R. Tom Baker

R. Tom Baker received his BSc degree from UBC in 1975 and his PhD from UCLA in 1980 with M. Frederick Hawthorne. After a postdoctoral stint with Philip S. Skell at Penn State, he joined DuPont Central Research in Wilmington, DE, where he applied inorganic and organometallic chemistry and homogeneous catalysis to the nylon, fluoroproducts, and titanium dioxide businesses. He joined the Chemistry Division at Los Alamos National Laboratory in 1996 and worked on multifunctional catalysis approaches to low-temperature hydrocarbon functionalization and chemical hydrogen storage. In 2008, he became a chemistry professor at University of Ottawa and the director of the Centre for Catalysis Research and Innovation.

Anne Staubitz

Anne Staubitz obtained a diploma in biochemistry from the University of Tübingen, Germany. For her diploma thesis she worked in the group of Prof. Paul Knochel at the University of Munich with novel Grignard reagents applied to the synthesis of heterocycles. She then did her PhD with Prof. Varinder Aggarwal at the University of Bristol, working on a natural product synthesis. In Ian's group she is working as a postdoctoral fellow on the catalytic dehydrogenation of group 13–group 15 adducts.

Ian Manners

Ian Manners received his PhD from the University of Bristol in 1985 in the area of transition metal chemistry. After completing postdoctoral work in Germany in main group chemistry and the USA on polymeric materials, he joined the University of Toronto, Canada in 1990. After 15 years he returned to his Alma Mater to take up a Chair in Inorganic, Macromolecular and Materials Chemistry. His research interests focus on the development of new synthetic reactions in inorganic chemistry and their applications in molecular synthesis, polymer and materials science, supramolecular chemistry, and nanoscience.

---

Introduction

An abundant, inexpensive energy supply is an essential component to boost developing economies as well as maintain status quo for established economies. Current energy consumption is based on combustion of carbon-based fuels to water and carbon dioxide (as well as environmentally harmful carbon particulates and sulfur and nitrogen oxides). For many years, climate modelling has shown that increased carbon dioxide concentration in the atmosphere will lead to detrimental environmental effects (recently reiterated in the fourth IPCC report).¹ Thus it is vitally important to institute a shift away from carbon-based fuels and toward environmentally-friendly replacements.

The transportation sector represented 31(28)% of the EU(US) energy use in 2005(2006) and accounted for 21(34)% of CO₂ emissions.^2,3 For stationary energy applications, CO₂ can be potentially sequestered.⁴ For portable energy, however, this is impractical and an alternative energy carrier must be used.

Hydrogen has the potential to be a clean, source-independent, energy carrier.⁵ It has a high energy content per mass compared to petroleum (120 MJ kg⁻¹ for hydrogen versus 44 MJ kg⁻¹ for petroleum). Also, hydrogen can be readily used to run a fuel cell, which greatly increases efficiency over internal combustion engines (∼32% efficiency for diesel-electric; 90% potential efficiency for fuel cell with heat capture, 85% for electric motor, 77% overall efficiency) while eliminating formation of carbon particulates and sulfur and nitrogen oxides.⁶ Unfortunately, hydrogen has a poor energy content per volume (0.01 kJ L⁻¹ at STP and 8.4 MJ L⁻¹ for liquid hydrogen vs. 32 MJ L⁻¹ for petroleum).

For transportation applications, an energy carrier should have a high energy content in as small a volume as possible to not intrude on passenger space, and as small a mass as possible to maintain fuel efficiency. The US Department of Energy has established a series of targets for hydrogen storage materials so that a vehicle can travel >500 km on a single hydrogen fill.⁵ Included with this are stringent system volumetric (5.4 MJ L⁻¹ 2010, 9.72 MJ L⁻¹ 2015) and gravimetric targets (6 wt% 2010, 9 wt% 2015). Evaluation of a hydrogen storage system includes all associated components (tank, valves, regulators, piping, mounting brackets, insulation, added coolants, etc.).

There are currently four leading methods to store hydrogen: physical means, sorbents, metal hydrides, and so-called chemical hydrides. All four will be briefly summarised, and then chemical hydrides with B–N bonds will be discussed in greater detail. Details of the potential of ammonia borane as a hydrogen storage material have been presented recently.⁷

Physical hydrogen storage

Safely containing hydrogen at high enough pressures requires strong tanks. The tanks must also be lightweight to maintain a high gravimetric capacity. Carbon-fibre reinforced composite tanks are light and capable of storing hydrogen at pressures up to 700 bar (4.7 MJ L⁻¹). These tanks are nonconformable (normally cylinder shaped), which makes them difficult to adapt to a car design. There is also an energy penalty for gas compression (15–20% of the lower heating value for hydrogen).⁵ General Motors recently demonstrated the Chevy Sequel, which is able to achieve 482 km (300 miles) on a single hydrogen fill using compressed tanks to store the hydrogen fuel.⁸ Honda recently released the FCX Clarity, which is capable of 435 km (270 miles) on a single fill, and has recently become available for lease in California, USA.⁹ Both cars use efficient polymer electrolyte membrane fuel cells. While these fuel cells allow for the large range, they currently require expensive platinum catalysts and suffer from limited durability. Although these specific designs may not yet be practical for world-wide use, their demonstration is an important step to show the viability of hydrogen-fuelled cars.

While liquid hydrogen has a higher density than compressed gas, maintaining the low temperatures requires extra components that adversely affect system volumetric and gravimetric storage capacity. Boil-off control is also a significant problem, with tanks losing approximately 1% per day. Finally, significant energy needs to be input to liquefy hydrogen (30% of the lower heating value for hydrogen).⁵

Sorbents

A wide range of nanoporous materials have been studied as potential hydrogen storage materials. The advantage of these sorbent materials lies in their ready reversibility. This is also their largest disadvantage, however, as they contain hydrogen by weak physisorption forces (van der Waals), which are normally less than 1.4 kcal mol⁻¹. Hence, low temperatures (normally liquid nitrogen temperature, 77 K) are necessary to obtain reasonable hydrogen uptake.¹⁰

Carbon-based materials such as nanotubes, nanofibres, solid foams, and activated carbon have been extensively studied. There was initially some confusion about the hydrogen uptake potential of these materials due to the various methods of measuring hydrogen uptake. Recently, the results have been more consistent, with several methods being employed in the same study.¹¹ The maximum adsorption is ∼5 material wt% hydrogen at 77 K; with addition of tanks and cooling systems, the system storage capacity will be much lower.

Zeolite structures have also been examined.¹² Zeolites are microporous structures of hydrated aluminate and silicate that are more easily prepared than nanotubes and offer greater control over pore size. However, they consist of atoms that are heavier than carbon, which limits gravimetric capacity. An emerging field is using zeolite structures as a template to form carbon networks.¹³ Templating carbon onto zeolites is difficult because zeolite structures can be disordered. Nonetheless, recent results demonstrated hydrogen uptake for zeolite-like carbon materials of 6.9 wt% at 77 K and 20 bar.

Organic polymers with intrinsic microporosity (PIMs) are another alternative.¹⁴ PIMs are made of rigid monomers to maintain microporosity (Fig. 1a), yielding gravimetric storage capacity of up to 3.0 wt% at 77 K and 15 bar.¹⁵ Increasing pore size (and thus surface area) should result in a higher storage capacity.

Fig. 1 (a) Some examples of PIMs made up of rigid monomers. (b) Different linkers result in different pore sizes for MOFs.¹⁷

Metal–organic frameworks (MOFs) have also recently come into focus as potential hydrogen storage materials.¹⁶ These are three-dimensional polymers of metal atoms linked by bridging ligands (Fig. 1b).¹⁷ MOFs are readily synthesised by solution methods and then activated by heating to remove solvent molecules. Pore sizes can be rationally designed by choosing the metal and the linker. However, with large linkers, these tend to form interpenetrated structures which can decrease pore size. Similar to carbon nanotubes, MOFs rely primarily on physisorption to bind hydrogen and low temperatures (typically 77 K) are required.

To make these materials more practical, the temperature of effective adsorption must be raised (ideally to ambient temperatures). By adding in metal atoms or other materials capable of chemisorption, the energy of binding can be adjusted. This combination of chemisorption and physisorption could potentially increase the temperature of effective uptake. There are several recent examples of increasing the enthalpy of hydrogen adsorption for zeolites,¹⁸ MOFs,¹⁹ and carbon materials.²⁰ However, the inclusion of extra elements decreases material gravimetric capacity.

Metal hydrides

Metal hydrides rely on stronger chemical interactions than sorbents, and this results in materials that store hydrogen at higher temperatures. Heating is required to release the hydrogen after uptake. Since most metal hydrides are dense powders, achieving volumetric storage targets is feasible. Unfortunately, the relatively heavy metals make high gravimetric capacity difficult.

One way to achieve higher gravimetric capacity is by including lighter main-group elements. These complex metal hydrides, such as alanates, amide, and borohydride compounds, have been evaluated as reversible hydrogen storage materials.²¹ One of the early examples is titanium-doped sodium alanate (NaAlH₄), which can reversibly release hydrogen to give a maximum material gravimetric capacity of 5.5 wt% (although 3–4 wt% is typically obtained). Chen et al. found that Li₃N can add roughly two equivalents of hydrogen to form lithium amide (LiNH₂) and two equivalents of lithium hydride at elevated temperatures (200–250 °C) to give a 9.3 wt% uptake of hydrogen (Scheme 1).²² Under vacuum and at temperatures below 200 °C, 6.3 wt% of hydrogen desorbs. The remaining 3 wt% could be removed by heating above 320 °C.

Scheme 1 Reactions of some complex hydrides.

Calcium borohydride can be heated to 400 °C to release 9.6 wt% hydrogen (Scheme 1).²³ Upon addition of catalytic amounts of the dopants, TiCl₃ and Pd, the ‘spent fuel’ can be rehydrogenated at 700 bar and 400–440 °C in 60% yield.

Recent effort has concentrated on mixtures of complex metal hydrides as potential hydrogen storage systems. Yang and Sudik found that ternary mixtures of MgH₂, LiNH₂ and LiBH₄ have increased rates and extent of hydrogen release compared to binary mixtures of these components.²⁴ At elevated temperatures (∼350 °C), 8–11 wt% hydrogen is released depending on the amount of MgH₂ in the mixture. At lower temperatures, hydrogen release is reversible (2.8 wt% at 140 °C). Soloveichik and co-workers reported the decomposition of a compound that contains borohydride and amines, Mg(BH₄)·2NH₃.²⁵ This compound has a maximum storage capacity of 16.0 wt%. Though hydrogen loss is observed at 150 °C, even heating to 400 °C results in a net loss of only 13.1 wt% hydrogen. Rehydrogenation has not yet been realised, but loss of hydrogen is reported to be endothermic, indicating that direct rehydrogenation may be possible.

Chemical hydrides

The so-called chemical hydrides typically use lighter elements than metal hydrides, which result in much higher gravimetric storage capacity. The necessary cleavage of covalent element–hydrogen bonds, however, puts more stringent requirements on reversibility. A reversible system will have a Gibb’s free energy (ΔG) of hydrogen release at or near 0 kcal mol⁻¹. These materials will have a positive entropic term (ΔS) as hydrogen gas is being released. Thus, a slightly endothermic (ΔH > 0) dehydrogenation reaction is required to achieve a system that is reversible under reasonable/practical conditions.

Most potential chemical hydrogen storage materials suffer from reaction enthalpies that are too endothermic or exothermic for reversible hydrogen release. Reaction kinetics can also be a significant problem. Indeed, for many years, the only known examples of hydrogen activation under mild conditions by non-metals involved complexes only isolable in matrices.²⁶ However, there have been several recent examples of isolable non-metal compounds that add hydrogen under mild conditions. Power and co-workers demonstrated addition of two hydrogen molecules to a digermyne compound (Scheme 2).²⁷ This was followed by the discovery of facile hydrogen addition across two diaryltin molecules.²⁷ The reverse reaction, the conversion of two E–H bonds to hydrogen and an E–E bond, has been demonstrated by Himmel and co-workers for a B–H species.²⁸ Bertrand, Schoeller, and co-workers designed N-heterocyclic carbene analogues, with an N-aryl group substituted with a carbon group, which add hydrogen under mild heating.²⁹ Stephan and co-workers reported the reversible heterolysis of hydrogen over a phosphinoborane species [1,4-(R₂P)C₆F₄(BR₂)] to make a phosphonium-borate zwitterionic compound.³⁰ Finally, mixtures of Lewis acids (BR₃) and Lewis bases (PR′₃,³¹N-heterocyclic carbenes,^31,32 and NR′₃³³) that are typically too sterically encumbered to form a dative bond (frustrated Lewis pairs) are capable of adding hydrogen under mild conditions to form borate salts. The ease of hydrogen addition depends on the sterics and electronics of the frustrated Lewis pairs. All of these systems have very low gravimetric capacity, but studying the basic reactions of hydrogen addition to non-metal systems can garner insight into the design of reversible systems with higher storage capacity.

Scheme 2 Non-metal compounds that add hydrogen under mild conditions.

Cyclic organic molecules have also been evaluated as potential hydrogen storage materials.³⁴ Dehydrogenation of organic molecules is typically too endothermic for hydrogen release at moderate temperatures. However, incorporation of fused rings and heteroatoms (O and especially N) can lower the heat requirements for hydrogen release.³⁵ Due to the large kinetic barriers, a catalyst (typically noble metal-based) is also used to aid dehydrogenation and rehydrogenation. For instance, in a recent patent, Pez and co-workers reported that N-ethylcarbazole can be hydrogenated over Ru/LiAlO₂ at 1000 psi of hydrogen and 160 °C. The hydrogen can then be released over Pd/LiAlO₂ at 199 °C to give 5.6 wt% hydrogen (Scheme 3). More importantly, the system can be cycled 5 times with no detectable degradation of the N-ethylcarbazole.

Scheme 3 Reversible dehydrogenation of N-ethylcarbazole.

There are several systems that rely on hydrolysis of active metals³⁶ or chemical hydrides to produce hydrogen. Thorn and co-workers reported that a C–H in the 2-position of benzimidazoles is sufficiently hydridic to be protonated by a variety of acids including water over a Pd catalyst to produce hydrogen.³⁷ Hydrolysis of another chemical hydride, sodium borohydride, has also been investigated. Although it can be stabilised in basic aqueous solutions, the addition of catalysts or acids can be used to initiate a controlled release of hydrogen by hydrolysis.³⁸ Unfortunately, efficiently regenerating the BO₂⁻ to BH₄⁻ is difficult due to the stability of B–O bonds.

Under the current technology, chemical hydrides with high storage capacity will require off-board regeneration, which adds to the complexity of the hydrogen storage system. However, these compounds could offer advantages for the fuel distribution system. Current fuel distribution is based on transporting liquid hydrocarbons; converting this system to transport compressed or liquid hydrogen will be an expensive endeavour (potentially costing trillions of dollars).³⁹ A stable chemical hydride could circumvent this problem, allowing for transportation of hydrogen using the existing infrastructure.

B–N compounds

Many of the previously mentioned hydrogen storage materials have gravimetric material capacities that may be too low for on-board hydrogen storage requirements. However, there are several B–N compounds with the potential to meet these requirements. B–N compounds are well suited for hydrogen storage because both boron and nitrogen are lightweight elements capable of bearing multiple hydrogens. Also, B–H and N–H bonds tend to be hydridic and protic, respectively, resulting in normally facile hydrogen release.⁴⁰ A successful B–N hydrogen storage material must contain multiple hydrogen equivalents per main-group element, have a good match between the number of hydridic B–H bonds and protic N–H bonds, and have the stability requirements necessary for safe storage of hydrogen. Several classes of B–N materials that may be suitable for hydrogen storage applications will be mentioned, and then the simplest B–N compounds, amine boranes, will be covered in more detail.

Hydrazine borane compounds

Much of the early interest in hydrazine borane compounds was centred around their use as propellants.⁴¹ Both hydrazine borane and -bis(borane) have been synthesised by various methods, the most common of which is addition of hydrazine salts to borohydride (Scheme 4). Hydrazine borane [-bis(borane)] has a potential hydrogen storage capacity of 13.1 [13.4] wt% (assuming loss of 3[4] eq. of H₂). Unfortunately, hydrazine bis(borane) is a shock-sensitive explosive,⁴² explodes when heated in air,⁴³ and is therefore poorly suited for hydrogen storage applications unless it can be stabilised.

Scheme 4 Synthesis of hydrazine bis(borane) and hydrazine monoborane.

Amine triborane compounds

The first high yield synthesis of ammonia-triborane (H₃N–B₃H₇) was reported by Kodama and co-workers in 1959.⁴⁴ Tetraborane (B₄H₁₀) is treated with tetrahydropyran to yield tetrahydropyran-triborane and a half equivalent of diborane. Tetrahydropyran can be readily displaced by ammonia to yield the product. Unfortunately, tetraborane and diborane are toxic gasses causing large-scale synthesis by this route to be impractical. Yoon and Sneddon reported an alternate synthesis of ammonia-triborane in which Bu₄NB₃H₈⁴⁵ salt (Scheme 5) is treated with 0.5 eq. I₂ in glyme to yield glyme–B₃H₇, Bu₄NI and 0.5 eq. of H₂.⁴⁶ Similar to the previous method, glyme is displaced by ammonia to yield the product. The structure of ammonia-triborane was established by X-ray methods and has a trigonal arrangement of B atoms with one bonded to ammonia.⁴⁷ Ammonia-triborane has a storage potential of 10.6 wt% (assuming 3 eq. H₂ are released), but direct dehydrogenation methods have yet to be reported. Kodama noted that upon treatment with sodium in ammonia, one equivalent of hydrogen is released giving a mixture of products including NaBH₄. Sneddon found that H₃N–B₃H₇ undergoes rapid hydrolysis after treatment with 1 M HCl (7.85 eq. of H₂ after 120 min) or various metal catalysts, the best being Rh/Al₂O₃ (approx. 7.5 eq. after 25 min at 21 °C with 7 mol% catalysts).⁴⁶ Similar to borohydride hydrolysis, products with B–O bonds are produced which will be energetically costly to regenerate.

Scheme 5 Synthesis and structure of ammonia-triborane.

Ammonium hydrotriborate ([NH₄][B₃H₈]) has also been synthesised by treatment of pentaborane with ammonium hydroxide.⁴⁸ Surprisingly, this is a stable, colourless, crystalline solid in pure form, and no decomposition was evident on heating the solid at 60 °C for 70 hours. It is apparently stable in water and alcohols and slowly decomposes to form H₃N–B₃H₇ and H₂ when treated with benzene or ether. This is in stark contrast to [NH₄][BH₄]⁴⁹ which decomposes at temperatures above −40 °C. Ammonium hydrotriborate has a potential hydrogen storage capacity of 13.9 wt% (assuming loss of 4 eq. H₂). Hydrogen loss from alkylammonium hydrotriborate salts has been realised by addition of metallic Si or Al (Scheme 6).⁵⁰ When [NH₃Me][B₃H₈] is treated with Si, 5 eq.⁵¹ of hydrogen gas are released giving 9.9 wt% H₂.

Scheme 6 Synthesis of ammonium hydrotriborate and dehydrogenation of alkylammonium salts.

Guanidinium hydrotriborate, [C(NH₂)₃][B₃H₈], has been prepared by treating di(guanidinium) sulfate with sodium hydrotriborate.⁵² At ∼100 °C, this compound violently decomposes to form 6.2 eq. H₂ (by MS), which provides 12.3 material wt%. It was noted that some decomposition was evident if the compound was stored at 20 °C for ∼2 months. Also, triaminoguanidinium hydrotriborate, [C(NHNH₂)₃][B₃H₈] was synthesised by treatment of pentaborane with triaminoguanidine.⁵³ A decomposition point was not reported, but a melting point of 72–78 °C was determined. The potential hydrogen storage capacity is 11.0 wt% (assuming loss of 8 eq. H₂).

It is important to note that these polyborane compounds are likely to be explosive, as many related compounds have been used as rocket propellants.

Amine compounds of higher-order polyboranes

Borane has a very rich chemistry, and there are many polyborohydride compounds known. When ammonia is bubbled through a solution of decaborane (B₁₀H₁₄), the tris(ammoniate) of decaborane (TAD) is formed.⁵⁴ This compound likely has a [NH₄][B₁₀H₁₃(NH₃)₂] formulation, and one of the NH₃ groups is only loosely bound. This formulation is thus similar to the compound [NH₂Et₂][B₁₀H₁₃(NHEt₂)].⁵⁵ Although TAD is stable under ambient conditions, upon heating to 75 °C, one equivalent of hydrogen and one equivalent of ammonia are released. Complete conversion to BN compounds was achievable by addition of hydrazine to TAD, and heating to 800 °C under an ammonia stream.⁵⁶ It is noteworthy that decaborane reacts vigorously with hydrazine to induce a fire in air.

In a theoretical study, Nguyen, Matus, and Dixon investigated the heats of formation of several ammonium salts of polyboranes.⁵⁷ They found that formation of [B₁₂H₁₂][NH₄]₂ from H₂ (10 eq.), BN (2 eq.), and B (10 eq.) has a heat of formation of only 10 to 12 kcal mol⁻¹ (the reverse reaction would result in loss of 8.8 wt% of H₂ from [B₁₂H₁₂][NH₄]₂). In a recent DOE progress report, Hawthorne and co-workers investigated a series of anionic polyborane compounds for hydrogen storage applications.⁵⁸ Ammonium salts of (B₁₁H₁₄)⁻, (B₁₂H₁₂)²⁻, and (B₁₀H₁₀)²⁻ were all synthesised and hydrolyzed in the presence of a metal catalyst to yield hydrogen, ammonium borate, and boric acid (Scheme 7).

Scheme 7 Hydrolysis of ammonium polyborane salts.

Amine borane compounds

Amine boranes are the most thoroughly studied B–N hydrogen storage materials. By varying the substituents on B and N, a variety of properties can be altered such as melting and decomposition points as well as dehydrogenation enthalpy and nature of the reaction products. Nöth and Beyer investigated the physical properties of a variety of alkylamine boranes obtained by addition of the alkylammonium salt to lithium borohydride (Table 1).⁵⁹ An alternative way to synthesise these compounds is by substitution of H₃B·L (L = Me₂S, THF, Me₃N) with the amine.⁶⁰ The physical properties are difficult to predict. For instance, H₂EtNBH₃ is one of the least stable of the group of monoalkylamine boranes, whereas HEt₂NBH₃ is one of the most stable dialkylamine boranes. Linear H₂ⁿBuNBH₃ decomposes above 10 °C, whereas the branched isomer, H₂^tBuNBH₃, is stable up to 120 °C.

Table 1 Physical properties of some alkylamine boranes^a

Alkylamine borane	Melting point/°C	Decomp. point/°C
a Table adapted from ref. 59.
H3NBH3	104	∼100
H2MeNBH3	56	70
H2EtNBH3	19	30–40
H2^nPrNBH3	45	50–70
H2^iPrNBH3	65	90–100
H2^nBuNBH3	−48	10–15
H2^tBuNBH3	96	120–140
HMe2NBH3	37	150
HEt2NBH3	−18	200
H^nPr2NBH3	30	140
H^iPr2NBH3	23	250
H^nBu2NBH3	15	120
H^iBu2NBH3	19	150

---

Hawthorne reported that B-substituted amine boranes can be prepared by reduction of alkylboroxines [(BOR)₃] using lithium aluminium hydride in the presence of trimethylamine.⁶¹ The majority of the Me₃NBRH₂ (R = ⁿPr, ⁱPr, ⁿBu, 2-Bu, ⁱBu, ^tBu, n-pentyl, and n-hexyl) compounds are liquids at room temperature with the exception of R = cyclohexyl (mp = 40–41 °C) and benzyl (mp = 58–60 °C). Treatment of these compounds with excess ammonia in the presence of catalytic ammonium chloride at 100–150 °C affords B-substituted borazines, [HNBR]₃, in moderate to good yield (65–91%).⁶²

In a theoretical study, Manners, Harvey and co-workers investigated the effect of B- and N-substituents on the ΔG and ΔH of dehydrogenation of HR₂NBR′₂H.⁶³ Loss of hydrogen from amine boranes normally is too exergonic for reversibility (see the section: Thermal dehydrogenation of amine boranes). By altering the substituents (R and R′) on HR₂NBR′₂H, however, the ΔG of dehydrogenation can be made more neutral. A strong dative N–B σ-bond in the reactant (HR₂NBR′₂H) and a weak dative N–B π-bond in the product (R₂NBR′₂) results in a less exergonic dehydrogenation. In general, the σ-bond plays a more important role in determining the overall energetics, so HR₂NBR′₂H compounds with electron donating groups on nitrogen (resulting in a more Lewis-basic amine) and electron withdrawing groups on boron (resulting in a more Lewis-acidic borane) are best suited for reversible dehydrogenation. Evaluation of a series of cyclic amines indicates that four- and five-membered rings exhibit very similar dehydrogenation enthalpies, whereas mechanisms involving six-membered rings are more endothermic. This appears to be an effect of added ring strain going from an sp³- to an sp²-hybridised nitrogen centre.

Hydrogen loss from amine boranes is typically effected by solvolysis (including acid- and metal-catalysed variants) or thermolysis, in which the product distribution depends on the reaction conditions and presence of additives or catalysts. The easiest way to evaluate the extent of hydrogen loss is by measuring the amount of hydrogen gas generated by either Toepler pump or GC/MS methods. Gas burette and thermogravimetric analysis (TGA) may also be used, but other gaseous products may also contribute to the volume measured without a method for direct identification or quantification. Temperature-programmed desorption (TPD) provides a means to analyze the volatile products by MS. If soluble products are formed, NMR spectroscopy (especially ^1,2H, ¹¹B, and ^14,15N)⁶⁴ can also give a good indication of the extent of dehydrogenation by both the chemical shift and multiplicity of peaks formed (Fig. 2). The wide chemical shift range for ¹⁵N NMR is particularly useful, with sp²-hybridised N in substituted borazines from −230 to −280 ppm and sp³-hybridised N in alkylamine boranes from −340 to −375 ppm.⁶⁵ Many of these nuclei are quadrupolar and observed line widths will depend on the electric field gradient at the nucleus and the nuclear correlation time. Large molecules with low symmetry (around the quadrupolar nucleus) in viscous solvents will thus give rise to broad resonances. The increased molecular motion and decreased viscosity at higher temperatures can often be used to reduce linewidths in solution NMR experiments. Conversely, cooling the sample effectively ‘decouples’ the quadrupolar interaction to adjacent nuclei thus reducing line widths of, for example, ¹H and ¹³C NMR signals in alkylboranes. Finally, double rotation experiments or use of high magnetic fields allows for useful information to be obtained for quadrupolar nuclei in the solid state as well.⁶⁶

Thermal solvolysis of amine boranes

One of the conceptually easiest ways to release hydrogen from molecules with B–H bonds is by solvolysis (see borohydride and polyborane hydrolysis). Reactions of amine boranes with alcohol or water are thermodynamically downhill. However, high temperatures (above the boiling point of water) are necessary to induce hydrolysis under neutral or basic conditions.⁶⁷ In a recent article, thermally-induced solvolysis was used in two different ways.⁶⁸ The first capitalised on an exothermic hydrogen release to induce a self-sustained reaction, and the second relied on pressurising water to increase its boiling point. A gelled mixture of AB, Al, and H₂O in a 2 : 3 : 3 ratio can be ignited. The excess heat generated causes the reaction to be self-sustaining, releasing 7.7 wt% of hydrogen in the process. Also, a mixture of water and AB can be heated to 135 °C (under Ar pressure) to release 3 equivalents of hydrogen. The reaction was very clean by ¹¹B-NMR and MS; no borazine formation was detected.

Acid-catalysed solvolysis

Acid-catalysed hydrolysis is the oldest known process for releasing hydrogen from amine boranes (Scheme 8).⁶⁹ It is likely that the acid functions by protonating the amine, which releases BH₃ for subsequent hydrolysis. The nature of the amine has a profound influence on the reaction rate. For instance, AB is hydrolysed 600 times faster than H₂MeNBH₃ and 4.8 × 10⁴ times faster than HMe₂NBH₃. More interesting from an engineering point of view is successful use of immobilised acids such as ion exchange resins or various zeolites to activate ammonia borane.⁷⁰ Even CO₂ (which generates carbonic acid in situ) proved to be a catalyst for the dehydrogenation (albeit with a slow rate).

Scheme 8 Acid catalysed hydrolysis of ammonia borane.

Metal-catalysed solvolysis

Many metals and metal complexes have been found to catalyse amine borane solvolysis (Table 2).^71–80 The recent focus in catalyst development has been on the use of non-precious metals. In two recent heterogeneous systems, catalyst morphology greatly affected catalytic activity. Hollow spheres of nickel metal were shown to exhibit substantial catalytic activity versus nickel powder. By doping the nickel hollow spheres with Pt, the rate could be increased such that three equivalents of hydrogen were released within 30 minutes.⁷⁹ In a recent study, iron nanoparticles were also found to catalyse solvolysis. Borohydride reduction of Fe(SO₄) results in nanoparticles that slowly catalyse hydrolysis (160 min at RT with 12% cat. loading). However, if the nanoparticles are generated in the presence of ammonia borane, a very active catalyst forms (8 min, RT, 12% cat. loading), which can be recycled without loss of activity, and was similarly effective in air.⁸⁰ It was found that Fe(SO₄) reduction forms crystalline material in the absence of AB, but forms amorphous nanoparticles in the presence of AB, which may account for the difference in activity.

Table 2 Different catalytic systems for solvolysis of amine boranes

	Amine borane	Catalyst	Solvent	Eq. of H2 released	Temp/°C	Time	Ref.
1	H2^tBuNBH3	10% Pd/C (50%wet)	MeOH	Approx 3	30	100 min	71, 72
	Me3NBH3					20 h
2	Various	10% Pd/C (50%wet)	H2O, various alcohols	High efficiency	20	5 min (MeOH) to 190 min (^tBuOH)	73
		Raney Ni (5 mol%)
3	H3NBH3	Pt (20% on C) (2 mol%)	H2O	Approx 3	20	2 min	74
		[Rh(1,5-cod)(μ-Cl)]2 (2 mol%)		Approx 2.7		20 min
		Pd (2 mol%)		Approx 2.5		250 min
4	H3NBH3	Dowex (12 wt%)	H2O	Approx 2.8	20	8 min	70
		CO2		No AB left		7 days
5	H3NBH3	Co (10% on C) (2 mol%)	H2O	Approx 2.9	20	60 min	75
		Ni (10% on γ-Al2O3) (2 mol%)		Approx 2.9		60 min
6	H3NBH3	Ni0.88Pt0.12 hollow sphere (2 mol%)	H2O	Approx 3	20	30 min	79
7	H3NBH3	Rh colloids (1 mol%)	H2O	Approx 2.8	20	40 s	76
		Ir colloids (1 mol%)		Approx 3		105 min
		Co colloids (1 mol%)		Approx 3		60 min
8	H3NBH3	RuCl3 (0.5 mol%)	MeOH	Approx 3	20	5 min	77
9	H3NBH3	Amorphous Fe nanoparticles	H2O	Approx 3	20	8 min	80
10	H3NBH3	Various Co, Ni, Cu nanoparticles	H2O	Approx 3	20	20–300 min	78

---

Thermal dehydrogenation of amine boranes

Although amine boranes are isoelectronic to alkanes, the energetics and process of dehydrogenation are much different. Dehydrogenation of ethane to ethylene, for example, is endothermic (ΔH = 32.6 kcal mol⁻¹) as cleavage of two strong C–H bonds is not totally compensated for by formation of H₂ and the C=C π-bond. In contrast, dehydrogenation of ammonia borane to aminoborane (H₂NBH₂) is exothermic (ΔH = −5.09 kcal mol⁻¹)⁸¹ as the dative B–N bond is converted into a stronger covalent one. Upon evaluating the kinetics of intramolecular hydrogen loss from ammonia borane in the gas phase, Dixon and co-workers found that the intramolecular activation barrier (32–33 kcal mol⁻¹)⁸² is actually larger than the B–N bond dissociation energy (25.9 kcal mol⁻¹).⁸³ According to this result, AB should dissociate into NH₃ and BH₃ before H₂ loss. However, the newly formed BH₃ can catalyse AB dehydrogenation through a six-membered transition state (Scheme 9) at a barrier only 6.1 kcal mol⁻¹ higher in energy than separated AB and BH₃.

Scheme 9 Mechanism of BH₃ catalysed AB dehydrogenation.

In the condensed phase, thermolysis of amine boranes such as AB and methylamine borane (H₂MeNBH₃, MeAB) have been shown to proceed by an intermolecular mechanism that involves initial formation of a diaminoboronium borohydride salt (Scheme 10). Further reaction of this salt with additional amine borane molecules builds up aminoborane chains with formation of a new B–N bond for each hydrogen molecule released. Computational investigations of presumed linear polyaminoborane products, H₃N(BH₂NH₂)_nBH₃, showed that low energy coiled and helical conformations are favoured that feature B–H⋯H–N dihydrogen bonding.⁸⁴

Scheme 10 Formation of the diaminoboronium borohydride salt.

Dehydrogenation of amine boranes typically yields a variety of oligomeric products depending on conditions and methods unless they are sterically blocked (Fig. 2, for AB). Dixon and co-workers calculated the thermodynamics of the formation of smaller oligomers [B_xN_xH_y (x = 2, 3)] in both the gas and condensed phase.⁸⁵ Larger oligomeric products were evaluated in the condensed phase by Miranda and Ceder.⁸⁶ These products result from both a polymeric ammonia borane cycle (AB to PAB to PIB; see Fig. 2 for structures) and a cyclic oligomeric pathway (AB to CTB, borazine or 1,4-polyborazylene). While the overall reaction enthalpies depend on the products formed, all reactions in the study are estimated to be mildly exothermic [−1.6 to −20 kcal mol⁻¹ AB]. If one considers that the entropic term contributes ∼8 kcal mol⁻¹ H₂, it is clear that direct rehydrogenation will not be possible under practical conditions and that amine boranes will need to be regenerated in a chemical process. A few products, such as borazine, are volatile. Loss of these products leads to contamination of the hydrogen stream (potentially poisoning the fuel cell), and material loss (limiting regeneration efficiency).

Fig. 2 Some products of dehydrogenation from AB, and ¹¹B NMR shifts.

Solid state thermolysis

Initial studies of alkylamine borane (H₂RNBH₃) thermolysis (at 90–120 °C) afforded mixtures of cyclic amino- and iminoborane oligomers as well as undefined products.⁸⁷ Framery and Vaultier found that heating N- and B-substituted amine boranes (H₂RNBH₃ or H₃NBRH₂) to 200 °C gave the corresponding borazine compounds in good yield.⁶⁵ Detailed studies on the thermolysis of MeAB revealed that hydrogen is released in two stages, one at ∼100 °C, and the second at 190 °C. For the latter, a competing pathway to borazine formation was identified as dehydrogenative cross-linking of (HMeNBH₂)₃ to give an insoluble polymer.⁸⁸

The parent, ammonia borane, decomposes in three distinct steps (Scheme 11).⁸⁹ The first commences at ∼100 °C and peaks between 107 and 117 °C with an initial weight loss of ∼1.1 eq. of dihydrogen (∼7.2 wt%). The second equivalent is lost over a much broader temperature range, with a maximum rate at ∼150 °C, and the rest released at much higher temperatures.⁹⁰ Since all steps are exothermic, the high temperature requirement is a reflection of the significant kinetic barriers. The decomposition temperatures and products of dehydrogenation are dependent on the rate that the temperature is elevated. Lower temperature ramping rates result in a higher decomposition temperature. Analysis (IR and MS) of the volatile thermolysis products for the first dehydrogenation step revealed traces of B₂H₆, H₂N=BH₂⁹¹ and borazine accompanying the evolved hydrogen.⁹² Subsequent detailed ¹¹B solid state NMR studies at high field showed that formation of a new AB mobile phase preceded formation of the diammoniate of diborane (DADB, [BH₂(NH₃)₂][BH₄]). DADB, formed from two AB molecules by a hydride transfer, actually initiates hydrogen loss and concomitant B–N bond formation.⁹³ A similar intermediate was proposed for the thermal decomposition of MeAB.⁸⁸ The second hydrogen-releasing step produces cyclic iminoborane oligomers (including borazine and B–N linked borazines, polyborazylene) whose proposed graphitic structure is reminiscent of hexagonal and rhombohedral phases of boron nitride.⁹⁴

Scheme 11 Thermolysis of ammonia borane.

The dehydrogenation rate can be increased by the inclusion of additives or by intercalation of AB in a solid scaffold. Benedetto and co-workers found that AB samples doped or milled with Pt (ca. 1%) had a greater extent of H₂ release at low temperatures (23% increase in H₂ release at 140 °C).⁹⁵ Autrey and co-workers found that a nanocomposite of mesoporous silica and AB (1 : 1 by weight) releases hydrogen at 50 °C with a half-reaction time of 85 min compared to a half-reaction time of 290 min at 80 °C for neat ammonia borane.⁹⁶ The peak dehydrogenation temperature was lowered from ∼110 °C to ∼98 °C when a heating rate of 1 °C min⁻¹ was used. Encapsulation of AB in a 24 wt% carbon cryogel⁹⁷ decreased the peak dehydrogenation temperature to ∼90 °C, and there was no further decomposition at higher temperatures. Volumetric measurements indicated a 9 wt% loss of hydrogen, and no borazine formation was detected (MS).

The mechanism(s) by which the solid scaffolds increase the rate may be similar for both cases. It is possible that the observed effect may be a function of increasing the surface area of AB, as it is intercalated in nano-scale pores, which is known to lower the phase transition temperature and thereby presumably the dehydrogenation temperature. Another possibility is that dehydrogenation of AB is catalysed by the exposed functional groups (SiOH for silica and carboxylic acid for carbon cryogel) on the nanocomposite. Acid catalysed dehydrogenation of AB is well known (see below). Both mechanisms are consistent with the observation that smaller pore sizes reduce the decomposition temperature.

Solution thermolysis of ammonia borane

Thermal decomposition of AB in a variety of aprotic, polar solvents is slow and results in a mixture of cyclic amino- and iminoborane oligomeric dehydrogenation products.⁹⁸ Sneddon and co-workers found that addition of ionic liquids can greatly increase both the rate and extent of dehydrogenation.⁹⁹ While heating pure ammonia borane at 85 °C gave a negligible amount of dihydrogen after 3 h, heating in certain ionic liquids resulted in immediate release of dihydrogen (0.95 equivalents after 3 h at 85 °C and even 1.5 eq. after 3 h at 95 °C, compared to 0.8 eq. at 95 °C after 3 h for AB). Monitoring these reactions in situ using ¹¹B NMR provided evidence for rapid formation and stabilisation of DADB in the ionic liquid. ¹¹B NMR analysis of pyridine extracts of the colorless non-volatile residue indicated linear and branched acyclic aminoborane chains, such as H₃N(BH₂NH₂)_nBH₃ and H₃NBH(NH₂BH₃)₂, in addition to DADB.

Acid-catalysed dehydrogenation of ammonia borane

Another way to effectively dehydrogenate AB is by the addition of Lewis or Brønsted acids. Treatment of AB with the strong Lewis acid B(C₆F₅)₃ at 25 °C in ether affords the boronium cation salt [BH₂(NH₃)(OEt₂)][BH(C₆F₅)₃] by hydride abstraction. Strong Brønsted acids, such as trifluoromethane sulfonic acid (HOTf), protonate a B–H bond in AB yielding hydrogen and the analogous boronium triflate. These boronium cations are more reactive versions of that found in DADB and can, as a result, initiate hydrogen release from AB even at 25 °C. Computational studies showed that the cation interacts initially with a B–H bond of AB, drawing a protic N–H in proximity to a hydridic B–H, resulting in loss of hydrogen. Further molecules of AB then interact similarly with the resultant cationic complex to build the aminoborane chains stepwise (Scheme 12). The relative concentration of acid needs to be kept low (0.5 mol%) to avoid chain termination to aminodiborane, B₂H₅(μ-NH₂), and concentrated solutions afford high yields of borazine at 60 °C in 4 h.¹⁰⁰

Scheme 12 A Lewis-acidic [H₂BNH₃]⁺ molecule interacts with ammonia borane to lose hydrogen and form a new compound that is capable of attack at two positions.

Anionic dehydropolymerisation of ammonia borane

In a recent DOE progress report, Sneddon and co-workers reported that generation of catalytic amounts of H₂NBH₃⁻ increases the rate of hydrogen release.¹⁰¹ Metal complexes of this anion have also been investigated as potential hydrogen storage materials (see the Metal amidoboranes section). The anion can be generated in situ by addition of LiNH₂, LiH, or proton sponge [1,8-bis(dimethylamino)naphthalene]. The use of proton sponge eliminates the formation of LiBH₄ and NH₃ side products identified when LiNH₂ or LiH was used as the base. Although, the mechanism is currently unknown, it has been suggested that the increased hydricity of the B–H bond in H₂NBH₃⁻ leads to facile H₂ loss from its reaction with AB.¹⁰²

Metal-catalysed dehydrogenation of amine boranes

Metal-catalysed dehydrogenation of amine boranes offers the potential for additional control over both extent and rate of hydrogen production.^103–113 In a 1989 patent, Laine and Blum claimed the dehydrogenation of several amine boranes using Ru₃(CO)₁₂ at 60 °C (Table 3; 8–10).¹⁰³ Using a heterogeneous Pd/C catalyst, Roberts and co-workers reported conversion of HMe^tBuNBH₃ to the corresponding aminoborane at 120 °C.¹⁰⁴ Despite the wide range of catalytic systems developed subsequently (Table 3), there has yet to be a catalyst capable of both a high rate and large extent of hydrogen release.^105–113 Moreover, if these systems are to be practical, base metals need to be used at low catalyst loadings. Finally, engineering controlled hydrogen release has heretofore only been demonstrated for solid catalysts so effective heterogenisation strategies will be required.

Table 3 Selection of reported metal catalysts for amine borane dehydrogenation

	Catalyst (mol%)	Substrate	Conditions	Products	Eq. of H2	Ref.
a Inferred from reported TOF values.b CTB is cyclotriborazane.c NR is “not reported.”d Yield of products not quantified; other products possible.e Isolated yields, actual yield will be higher but other products detected.f κ^3-2,6-[OP(t-Bu)2]2C6H3.g Enders’ carbene: (1,3,4-triphenyl-4,5-dihydro-1H-1,2,4-triazol-5-ylidene).h IDipp is 1,3-bis(2,6-diisopropylphenyl)-1,3-dihydro-2H-imidazol-2-ylidene.
1	Cp2TiMe2 (0.5%)	HMe2NBH3	16 h, 25 °C	No reaction	0	106
2	Cp2Ti (2%)	HMe2NBH3	4 h, 20 °C	(Me2NBH2)2	1	110
3	Cp2Ti (2%)	H(i-Pr)2NBH3	1 h, 20 °C	^iPr2NBH2	1	110
4	{[Cp(SiMe3)2]2Ti}N2 (2%)	HMe2NBH3	7 min, 23 °C^a	(Me2NBH2)2	1	111
5	{[Cp(SiMe3)2]2Ti}N2 (2%)	H3NBH3	161 h, 65 °C^a	CTB^b, borazine	NR^c	111
6	[indenyl-(SiMe3)2]2Zr	HMe2NBH3	147 h, 65 °C^a	(Me2NBH2)2	1	111
7	[P(^iPr)3]2Br2(CH3CN)(NO)Re (1%)	HMe2NBH3	4 h, 85 °C	(Me2NBH2)2	1	105
8	Ru3(CO)12 (0.2%)	H3NBH3	85 h, 60 °C	BN1.13H4.7 (elemental analysis)	NR	103
9	Ru3(CO)12 (0.1%)	Me3NBH3, PrNH2	32 h, 60 °C	[–N(Pr)B(H)–]3 (57%)	NR	103
10	Ru3(CO)12 (0.1%)	Me3NBH3, MeNH2	9.5 h, 60 °C	[–B(NMeH)N(Me)–]3, B(NHMe)3^d	NR	103
11	trans-RuMe2(PMe3)4 (0.5%)	HMe2NBH3	16 h, 25 °C	(Me2NBH2)2	1	106
12	FeH(PMe2CH2)(PMe3)3 (9%)	H3NBH3	96 h, 25 °C	CTB, borazine, polyborazylene^d	NR	108
13	[Rh(1,5-cod)(μ-Cl)]2 (0.5%)	HMe2NBH3	8 h, 25 °C	(Me2NBH2)2	1	106
14	[Rh(1,5-cod)(μ-Cl)]2 (5%)	HMe2NBH3	<2 h, 25 °C	(Me2NBH2)2	1	106
15	RhCl3 (0.5%)	HMe2NBH3	22.5 h, 25 °C	(Me2NBH2)2 (90%)	0.9	106
16	HRh(CO)(PPh3)3 (0.5%)	HMe2NBH3	160 h, 25 °C	(Me2NBH2)2 (5%)	0.05	106
17	[Cp*Rh(μ-Cl)Cl]2 (0.5%)	HMe2NBH3	112 h, 25 °C	(Me2NBH2)2	1	106
18	[Rh(1,5-cod)(μ-Cl)]2 (0.5%)	H(1,4-C4H8)NBH3	24 h, 25 °C	[(1,4-C4H8)NBH2]2 (73%)^e	NR	106
19	[Rh(1,5-cod)(μ-Cl)]2 (0.5%)	HMe(PhCH2)NBH3	24 h, 25 °C	[Me(PhCH2)NBH2]2 (79%)^e	NR	106
20	[Rh(1,5-cod)(μ-Cl)]2 (1%)	H2MeNBH3	∼60 h, 45 °C	(MeNBH)3 (40%)^e	NR	106
21	[Rh(1,5-cod)(μ-Cl)]2 (0.6%)	H2PhNBH3	16 h, 45 °C	(PhNBH)3 (56%)^e	NR	106
22	[Rh(1,5-cod)(μ-Cl)]2 (0.6%)	H3NBH3	∼60 h, 45 °C	Borazine (10%),^e PIB, polyborazylene	NR	106
23	[Rh(1,5-cod)(μ-Cl)]2 (1%)	H^iPr2NBH3	24 h, 25 °C	(^iPr)2NBH2 (49%)^e	NR	106
24	[Ir(1,5-cod)(μ-Cl)]2 (0.5%)	HMe2NBH3	136 h, 25 °C	(Me2NBH2)2 (95%)	0.95	106
25	(POCOP^f)Ir(H)2 (0.5%)	H3NBH3	14 min, 20 °C	Cyclopentaborazane	1	113
26	Ni(1,5-cod)2, 2 NHC^g (9%)	H3NBH3	3 h, 60 °C	Polyborazylene^d	2.8	108
27	Pd/C	HMe^tBuNBH3	1 h, 120 °C	[Me^tBuNBH2]2	1	104
28	Pd/C (0.5%)	HMe2NBH3	68 h, 25 °C	(Me2NBH2)2 (95%)	0.95	106
29	(IDipp)^hCuCl (12.5%)	HMe2NBH3	24 h, 20 °C	(Me2NBH2)2^d	NR	108

---

Under the heavily reducing conditions of amine borane dehydrogenation, the metal complex catalyst precursor will often undergo changes. Frequently, the active species is much different than the precatalyst. Manners and co-workers found that [Rh(1,5-cod)(μ-Cl)]₂ catalyses the dehydrogenation of a variety of amine boranes at room temperature or with mild heating (Table 3; 13, 14, 18–23).¹⁰⁶ The analogous Ir precursor or Rh precursors with different supporting ligands exhibited much lower activity for HMe₂NBH₃ (DMAB) dehydrogenation (Table 3; 15–17, 24). The dehydrogenation of DMAB using [Rh(1,5-cod)(μ-Cl)]₂ exhibits an induction period, during which a black, opaque suspension forms. TEM analysis indicated Rh aggregation; the UV-Vis spectrum was similar to the spectrum of Rh colloids; addition of Hg (either at the onset, or during the reaction progress) resulted in complete loss of activity; and although the dark powder isolated from catalysis still had activity, the solution after filtration had almost no activity. These observations all point to Rh(0) colloids as the catalytically-active species. However, subsequent in situ EXAFS (extended X-ray absorption fine structure) analysis found that the same catalytic precursor, under different conditions, may form soluble Rh clusters that also catalyse this reaction.¹⁰⁷ These Rh_4–6 clusters, observed in solution during the reaction, eventually precipitate, likely due to a ligand exchange process with the formed products. The clusters can be redissolved by treatment with DMAB. The process of forming colloids or clusters can be complicated, and minor variations in conditions can have a profound influence on metal species development. However, it is clear in both cases that the precatalyst is drastically altered during the reaction.

Minor alterations to supporting ligands can also greatly influence the catalytic activity. Baker and co-workers measured the relative catalytic activity of N-heterocyclic carbene (NHC) complexes of Ni.¹⁰⁸ The Enders’ carbene Ni complex was found to be 11.5 times faster than the IDipp complex and 8.8 times faster than the IMes complex.¹⁰⁹ Also, the Enders’ Ni complex was 4.1 times faster than a Rh(NHC) complex and 1.9 times faster than a Ru(NHC) complex. The Enders’ carbene Ni complex has the largest extent of dehydrogenation yet seen (>2.5 eq.) but requires mild heating. Initially, a minor amount of borazine is formed. However, borazine reacts further by crosslinking reactions to form polyborazylene. Using borazine as the substrate under similar catalytic conditions to AB results in almost no activity, indicating that AB either activates the catalyst or is involved in the crosslinking reactions. During the course of the reaction a dark, homogeneous maroon solution forms. Hg addition results in no loss of activity, which is indicative that metal colloids are not the active catalyst. A kinetic isotope effect (KIE) is observed for D₃NBH₃, H₃NBD₃, and D₃NBD₃ (2.3, 1.7, and 3.0, respectively) indicating that both N–H and B–H bond cleavage are involved in the turnover limiting step or that N–H and B–H bond cleavage steps have similar rates.

Solvent and substrate can also play a significant role. An example of this was noted in the titanocene-based dehydrogenation of amine boranes initially reported by Manners and co-workers (Table 3; 2, 3)¹¹⁰ and extended by Chirik and co-workers to include other Cp-based ligands and Zr (Table 3; 4–6).¹¹¹ Chirik and co-workers noted that {[η⁵-C₅H₃(SiMe₃)₂]₂Ti}N₂ has a TOF of >420 h⁻¹ in benzene-d₆ and 0.29 h⁻¹ in THF for the dehydrogenation of DMAB. Also, AB was found to have a much slower rate than DMAB (Table 3; 4 and 5).

Finally, the catalyst identity controls the extent of dehydrogenation. Heinekey, Goldberg and co-workers reported that (POCOP)IrH₂¹¹² is an extremely active catalyst for AB dehydrogenation (Table 3; 25).¹¹³ Unfortunately, only a single equivalent of hydrogen is released per equivalent of AB. During the course of the reaction, an insoluble colorless precipitate is formed. The X-ray powder diffraction and IR data agree closely with that previously reported for cyclopentaborazane, [H₂NBH₂]₅.¹¹⁴ However, Manners et al. found that alkylamine boranes, and mixtures of alkylamine boranes and AB under similar conditions gave soluble aminoborane polymers.¹¹⁵ The measured wide angle X-ray scattering pattern and the IR for the white precipitate from the (POCOP)IrH₂ catalysed dehydrogenation of AB were different from that reported for [H₂NBH₂]₅. They suggested that the white precipitate may be polymeric in nature similar to the products formed from alkylamine borane dehydrogenation. Although the actual nature of the precipitate has not yet been completely established, it is clear from the solid state ¹¹B NMR data (−18 ppm) and the measured amount of hydrogen released that an [H₂NBH₂]_n product is formed.

The mechanisms of metal-catalysed dehydrogenation have been investigated by computational methods for the titanocene system reported by Manners, the (POCOP)IrH₂ system reported by Goldberg and Heinekey, and the Ni(NHC)₂ system reported by Baker. Although the relative energies of intermediates are different, the overall reaction pathways are similar. In the initial step, B–H coordinates to the metal complex. In the titanocene case, the N–H is activated followed by hydride transfer from B–H (Scheme 13).¹¹⁶ Two mechanisms were evaluated in the (POCOP)IrH₂ case. The first follows generation of monovalent (POCOP)Ir with a concerted B–H and N–H bond activation at the Ir^I centre. The second one involves a concerted B–H and N–H bond activation at trivalent (POCOP)IrH₂ (Scheme 14). In the second mechanism, a hydride ligand acts as a proton acceptor yielding pentavalent (POCOP)IrH₄.¹¹⁷ In the Ni(NHC)₂ case, N–H deprotonation is proposed to occur at the Ni-bound NHC carbon, forming a coordinated imidazolium-type ligand (Scheme 15).¹¹⁸ The imidazolium subsequently protonates the Ni centre and hydrogen formation follows. The proposed mechanism is not consistent with the observed KIE, so further investigation is warranted.

Scheme 13 Mechanism of TiCp₂ catalysed dehydrogenation of Me₂HNBH₃ as proposed by Luo and Ohno.¹¹⁶ Gibbs free energies corrected for toluene (CPCM).

Scheme 14 Relative energies of the proposed catalytic cycle in THF (CPCM).

Scheme 15 Simplified partial mechanism for NHC₂Ni catalysed dehydrogenation of AB. (i) Protonation of an NHC ligand. (ii) Transfer of a proton to metal centre. (iii) Formation of hydrogen.

All mechanisms proposed to date are focused on the initial loss of hydrogen from amine boranes. Establishing the role of the metal complex in subsequent oligomerisation and loss of the second equivalent of hydrogen will also be very important.

Regeneration of spent fuel

Spent fuel from solvolysis methods

The advantage of solvolysis methods is their high efficiency, chemical robustness, and the fact that inexpensive base-metal catalysts have been developed. A significant drawback of these systems is that B–H bonds are converted to much stronger B–O bonds, resulting in a more exothermic reaction than dehydrogenation. Regeneration of spent fuel from rehydrogenation methods requires strong reducing agents. Ramachandran and co-workers demonstrated a system based on transition metal catalysed solvolysis of AB to yield [NH₄][B(OMe)₄], which could be converted back to AB by treatment with NH₄Cl and lithium aluminium hydride (Scheme 16).⁷⁷ It will likely be energetically costly to convert the oxidation product, Al(OMe)₃, back to the complex hydride.

Scheme 16 Solvolysis and subsequent regeneration of ammonia borane.

Spent fuel from dehydrogenation methods

The strategies presented to date for regeneration of BNH_x-spent fuel involve two important steps, digestion and reduction. Although a multitude of products can result from AB dehydrogenation, they all contain newly formed B–N linkages that must be broken. Addition of an acid (HX) can protonate these linkages, releasing amine and making B–X bonds (digestion). The B–X bonds can then be reduced by chemical reductants to form B–H bonds. A practical system must be as energetically efficient as possible, so no reaction step can be too exo- or endothermic.

The method mentioned above by Ramachandran and co-workers could potentially be used to regenerate ammonia borane. However, the generation of strong B–O bonds in the reaction pathway will likely limit overall efficiency. Alternatively, Sneddon and co-workers¹⁰² and Mertens and co-workers¹¹⁹ have independently developed reduction schemes that form B–Br and B–Cl bonds, respectively.

Sneddon and co-workers reported that spent BNH_x fuel was successfully digested using HBr/AlBr₃ (super-acid) in CS₂ to form H₂, NH₄Br, BBr₃ as well as [H₂NBBr₂]₃. The relative ratio of BBr₃ to [H₂NBBr₂]₃ depends on the material being digested. Mertens and co-workers treat a THF solution of spent fuel with an ether solution of HCl to generate BCl₃, NH₄Cl and H₂. Both aminoborane [(BH₂NH₂)_x] and iminoborane [(BHNH)_x] materials were successfully digested. Unfortunately, the yields of BCl₃ were low due to decomposition in THF. Switching to a similar super-acid solution that Sneddon and co-workers used increases the yield of BCl₃ to >60%. BCl₃ is difficult to directly hydrodechlorinate and reduction by hydrogen requires high temperatures (600–700 °C) to yield the partially reduced product, BHCl₂ (which must be removed from the side-product HCl and can subsequently disproportionate into BCl₃ and B₂H₆). However, addition of a Lewis-base such as NMe₃ reduces the hydrodechlorination temperature to 200 °C (but still requires high pressures of 2000 atm).¹²⁰ Unfortunately, these conditions are energetically costly, and the yield is poor (25%). However, a similar concept can be used for reduction of BX₃ compounds by chemical hydrides. Another advantage offered by a Lewis-base is that it eliminates formation of B₂H₆, a hazardous material. Thus, Sneddon and co-workers treat BBr₃ with N,N′-diethylaniline, which yields a compound easily reduced by triethylsilane under mild conditions. Mertens and co-workers found that stronger bases such as NEt₃, inhibit the reduction of BCl₃ by triethylsilane and MgH₂. Addition of a weaker base, NPh₃, allows for the complete reduction of Ph₃NBCl₃ by MgH₂ after 20 minutes at 80 °C. The final step is displacement of R₃N with ammonia, to yield H₃NBH₃.

In a DOE progress report regarding Los Alamos National Laboratory’s effort toward BNH_x-spent fuel processing, effective digestion of polyborazylene with 1,2-benzenedithiol was demonstrated to form ammonia adducts of dithioboron compounds.¹²¹ These compounds contain relatively weak B–S bonds that are readily reduced by Sn–H. Optimisation of reaction conditions and energy efficiency are still being developed, but this is an intriguing result.

Horizons

Despite the vast amount of progress made on understanding and controlling the properties of B–N compounds, more work needs to be accomplished for these materials to be practical hydrogen storage materials. From an engineering point of view, fuels and spent fuels that are liquids are preferable, especially for materials that require off-board regeneration. A tailored mixture of B–N compounds, such as substituted amine boranes, could possibly be mixed with AB to afford such a liquid fuel that could still maintain acceptable gravimetric storage capacity. Combinations of this fuel mix with solid catalysts that maintain acceptable extent and rate of hydrogen release would still need to generate a pure hydrogen stream for the fuel cell and a spent fuel that can be regenerated in high yield with minimum energy input. Given the energy required just to transport the spent fuel for chemical processing, research is underway to discover new B–N hybrid compounds that are capable of on-board reversible hydrogen storage. Two promising examples are metal amidoboranes and C–B–N compounds.

Metal amidoboranes

Just as complex hydrides use main-group elements to increase the gravimetric capacity of metal hydrides, metal amidoboranes use metals to potentially decrease the enthalpy of hydrogen loss, eliminate formation of volatile coproducts (increasing purity of hydrogen stream), and help control the rate of H₂ release. AB is readily deprotonated by a variety of metal hydrides to form a new family of metal amidoboranes. M^I(NH₂BH₃) (M^I = Li, Na) were prepared by ball-milling solid AB with sodium- or lithium hydride.¹²² Thermal decomposition of these materials was monitored using TPD, coupled with an MS analyser. Under the conditions of measurement, the Li and Na compounds released hydrogen at considerably lower temperatures (89 and 92 °C, respectively) than AB (108 °C). Formation of borazine, observed at 154 °C for AB, was not observed in the case of the alkali-metal amidoboranes. Both compounds have shorter B–N bond distances than that in AB, which may be an indication of a stronger dative bond.¹²³ The different dative bond may be responsible for the different decomposition temperature, but this is still a matter of speculation.

Burrell and co-workers synthesised Ca(NH₂BH₃)₂ by addition of AB to CaH₂ in THF.¹²⁴ A THF adduct is formed, but the THF can be removed in vacuo. While this compound is reported to lose hydrogen at 90 °C, only <0.3 equivalents of H₂ are released. Slowly ramping the temperature to 170 °C results in 3.6 equivalents of H₂ released (corresponding to 7.2 material wt%) with <0.1% ammonia and borazine released.

In addition to expanding the range of metals used in metal amidoborane complexes, the role of dopants or catalysts in controlling dehydrogenation, and possibly rehydrogenation, needs to be explored.

C–B–N compounds

Another strategy to obtain reversible chemical hydrogen storage materials is through C–B–N compounds. Since dehydrogenation of C–C bonds is significantly endothermic, new compounds containing both C–C and B–N bonds may be tailored to undergo dehydrogenation with ΔG≈ 0. Indeed, computational results from Dixon and co-workers show that replacement of 2 methylene groups of cyclohexane with an H₂NBH₂ group gives a material with 7.1 wt% hydrogen for loss of 3 eq. of H₂ and ΔH = 23.5 kcal mol⁻¹ H₂ (Scheme 17).¹²⁵ Similar compounds have been reported recently and identification of suitable dehydrogenation catalysts is in progress.¹²⁶

Scheme 17 Calculated energy of dehydrogenation for 1,2-azaboracyclohexane.

Conclusions

The high gravimetric capacity of B–N materials makes them particularly appealing for hydrogen storage applications. It is important to note that basic research in this area also impacts several other fields such as transfer hydrogenation,^33,127 ceramic precursor production, and B–N polymer synthesis.¹¹⁵ A wide range of materials have been synthesised, and our understanding of hydrogen release mechanisms is rapidly increasing. As the next generation of hybrid B–N materials are designed to enable energy efficient regeneration or even reversible storage, these compounds continue to show promise for practical hydrogen storage.

References

1. Intergovernmental panel on climate change, Climate change 2007: Synthesis Report, http://www.ipcc.ch/ipccreports/ar4-syr.htm.
2. Annual Energy Outlook 2008 (Revised Early Release) http://www.eia.doe.gov/oiaf/aeo/index.html.
3. Eurostat: Energy, transport and environment indicators http://epp.eurostat.ec.europa.eu/.
4. F. X. Han, J. S. Lindner and C. Wang, Naturwissenschaften, 2007, 94, 170; A. Yamasaki, J. Chem. Eng. Jpn., 2003, 36, 361.
5. S. Satyapal, J. Petrovic, C. Read, G. Thomas and G. Ordaz, Catal. Today, 2007, 120, 246.
6. A. Boudghene Stambouli and E. Traversa, Renewable Sustainable Energy Rev., 2002, 6, 297.
7. F. H. Stephens, V. Pons and R. T. Baker, Dalton Trans., 2007, 2613; T. B. Marder, Angew. Chem., Int. Ed., 2007, 46, 8116; H. W. Langmi and G. S. McGrady, Coord. Chem. Rev., 2007, 251, 925.
8. http://www.gm.com/explore/technology/news/2007/fuel_cells/.
9. http://automobiles.honda.com/fcx-clarity/.
10. A. W. C. van den Berg and C. O. Areán, Chem. Commun., 2008, 668.
11. G. G. Tibbetts, G. P. Meisner and C. H. Olk, Carbon, 2001, 39, 2291.
12. J. Dong, X. Wang, H. Xu, Q. Zhao and J. Li, Int. J. Hydrogen Energy, 2007, 32, 4998.
13. Z. Yang, Y. Xia and R. Mokaya, J. Am. Chem. Soc., 2007, 129, 1673.
14. P. M. Budd, A. Butler, J. Selbie, K. Mahmood, N. B. McKeown, B. Ghanem, K. Msayib, D. Book and A. Walton, Phys. Chem. Chem. Phys., 2007, 9, 1802; N. B. McKeown and P. M. Budd, Chem. Soc. Rev., 2006, 35, 675.
15. J.-Y. Lee, C. D. Wood, D. Bradshaw, M. J. Rosseinsky and A. I. Cooper, Chem. Commun., 2006, 2670.
16. Two recent reviews: J. L. C. Rowsell and O. M. Yaghi, Angew. Chem., Int. Ed., 2005, 44, 4670; D. J. Collins and H.-C. Zhou, J. Mater. Chem., 2007, 17, 3154.
17. A. C. Sudik, A. R. Millward, N. W. Ockwig, A. P. Côté, J. Kim and O. M. Yaghi, J. Am. Chem. Soc., 2005, 127, 7110.
18. Y. Li and R. T. Yang, J. Phys. Chem. B, 2006, 110, 17175.
19. Y. Li and R. T. Yang, J. Am. Chem. Soc., 2006, 128, 726.
20. A. D. Lueking and R. T. Yang, Appl. Catal., A, 2004, 265, 259.
21. S. Orimo, Y. Nakamori, J. R. Eliseo, A. Züttel and C. M. Jensen, Chem. Rev., 2007, 107, 4111.
22. P. Chen, Z. Xiong, J. Luo, J. Lin and K. L. Tan, Nature, 2002, 420, 302.
23. E. Rönnebro and E. H. Majzoub, J. Phys. Chem. B, 2007, 111, 12045.
24. J. Yang, A. Sudik, D. J. Siegel, D. Halliday, A. Drews, R. O. Carter, III, C. Wolverton, G. J. Lewis, J. W. A. Sachtler, J. J. Low, S. A. Faheem, D. A. Lesch and V. Ozolinš, Angew. Chem., Int. Ed., 2008, 47, 882; A. Sudik, J. Yang, D. Halliday and C. Wolverton, J. Phys. Chem. C, 2008, 112, 4384.
25. G. Soloveichik, J.-H. Her, P. W. Stephens, Y. Gao, J. Rijssenbeek, M. Andrus and J.-C. Zhao, Inorg. Chem., 2008, 47, 4290.
26. Some representative publications: Z. L. Xiao, R. H. Hauge and J. L. Margrave, Inorg. Chem., 1993, 32, 642; H.-J. Himmel, L. Manceron, A. J. Downs and P. Pullumbi, Angew. Chem., Int. Ed., 2002, 41, 796; H.-J. Himmel, L. Manceron, A. J. Downs and P. Pullumbi, J. Am. Chem. Soc., 2002, 124, 4448; A. Köhn, H.-J. Himmel and B. Gaertner, Chem.–Eur. J., 2003, 9, 3909.
27. G. H. Spikes, J. C. Fettinger and P. P. Power, J. Am. Chem. Soc., 2005, 127, 12232; Y. Peng, B. D. Ellis, X. Wang and P. P. Power, J. Am. Chem. Soc., 2008, 130, 12268.
28. O. Ciobanu, P. Roquette, S. Leingang, H. Wadepohl, J. Mautz and H.-J. Himmel, Eur. J. Inorg. Chem., 2007, 4530.
29. G. D. Frey, V. Lavallo, B. Donnadieu, W. W. Schoeller and G. Bertrand, Science, 2007, 316, 439.
30. G. C. Welch, R. R. San Juan, J. D. Masuda and D. W. Stephan, Science, 2006, 314, 1124; Y. Guo and S. Li, Inorg. Chem., 2008, 47, 6212.
31. G. C. Welch and D. W. Stephan, J. Am. Chem. Soc., 2007, 129, 1880; P. A. Chase and D. W. Stephan, Angew. Chem., Int. Ed., 2008, 47, 7433; S. J. Geier, T. M. Gilbert and D. W. Stephan, J. Am. Chem. Soc., 2008, 130, 12632; T. A. Rokob, A. Hamza, A. Stirling, T. Soós and I. Pápai, Angew. Chem., Int. Ed., 2008, 47, 2435.
32. D. Holschumacher, T. Bannenberg, C. G. Hrib, P. G. Jones and M. Tamm, Angew. Chem., Int. Ed., 2008, 47, 7428.
33. V. Sumerin, F. Schulz, M. Nieger, M. Leskelä, T. Repo and B. Rieger, Angew. Chem., Int. Ed., 2008, 47, 6001; V. Sumerin, F. Schulz, M. Atsumi, C. Wang, M. Nieger, M. Leskelä, T. Repo, P. Pyykkö and B. Rieger, J. Am. Chem. Soc., 2008, 130, 14117.
34. S. Hodoshima, H. Arai, S. Takaiwa and Y. Saito, Int. J. Hydrogen Energy, 2003, 28, 1255.
35. G. P. Pez, A. R. Scott, A. C. Cooper and H. Cheng, US Pat., 7 101 530, 2006; G. P. Pez, A. R. Scott, A. C. Cooper, H. Cheng, F. C. Wilhelm and A. H. Abdourazak, US Pat., 7 351 395, 2008; A. Moores, M. Poyatos, Y. Luo and R. H. Crabtree, New J. Chem., 2006, 30, 1675; E. Clot, O. Eisenstein and R. H. Crabtree, Chem. Commun., 2007, 2231.
36. M.-H. Grosjean, M. Zidoune, L. Roué and J.-Y. Huot, Int. J. Hydrogen Energy, 2006, 31, 109.
37. D. E. Schwarz, T. M. Cameron, P. J. Hay, B. L. Scott, W. Tumas and D. L. Thorn, Chem. Commun., 2005, 5919.
38. S. C. Amendola, M. Binder, S. L. Sharp-Goldman, M. T. Kelly and P. J. Petillo, US Pat., 6 534 033, 2003.
39. Estimated $380 million for 1.41 million vehicles in California, USA alone: J. M. Ogden, Int. J. Hydrogen Energy, 1999, 24, 709.
40. While one valence bond structure for ammonia borane would place a formal negative charge on B and a formal positive charge on N, in spite of charge transfer from N to B, the nitrogen actually carries a net negative charge due to the electronegativity difference between N and B. R. Hoffmann, J. Chem. Phys., 1964, 40, 2474.
41. L. R. Grant and J. E. Flanagan, US Pat., 4 381 206, 1983.
42. F. C. Gunderloy, Jr, B. Spielvogel and R. W. Parry, Inorg. Synth., 1967, 9, 13.
43. H. J. Emeleus and F. G. A. Stone, J. Chem. Soc., 1951, 840.
44. G. Kodama, R. W. Parry and J. C. Carter, J. Am. Chem. Soc., 1959, 81, 3534.
45. K. C. Nainan and G. E. Ryschkewitsch, Inorg. Nucl. Chem. Lett., 1970, 6, 765; G. E. Ryschkewitsch, K. C. Nainan, S. R. Miller, L. J. Todd, W. J. Dewkett, M. Grace, H. Beall, M. F. Hawthorne and R. Leyden, Inorg. Synth., 1974, 15, 113.
46. C. W. Yoon and L. G. Sneddon, J. Am. Chem. Soc., 2006, 128, 13992.
47. C. E. Nordman and C. Reimann, J. Am. Chem. Soc., 1959, 81, 3538.
48. W. V. Hough and J. M. Makhlouf, US Pat., 3 313 603, 1967.
49. R. W. Parry, D. R. Schultz and P. R. Girardot, J. Am. Chem. Soc., 1958, 80, 1.
50. J. E. Flanagan, US Pat., 4 166 843, 1979.
51. In the original figure of this patent, (6 + n) H₂ should be (6 −n) H₂.
52. L. V. Titov, M. D. Levicheva and G. N. Dubikhina, Izv. Akad. Nauk SSSR, Ser. Khim., 1976, 8, 1856.
53. D. A. L. Carvalho and N. W. Shust, US Pat., 3 564 561, 1971.
54. R. H. Toenikoetter, PhD Thesis, St. Louis Univ., Missouri, 1959; J. Williams, R. L. Williams and J. C. Wright, J. Chem. Soc., 1963, 5816.
55. M. F. Hawthorne, A. R. Pitochelli, R. D. Strahm and J. J. Miller, J. Am. Chem. Soc., 1960, 82, 1825; B. M. Graybill, A. R. Pitochelli and M. F. Hawthorne, Inorg. Chem., 1962, 1, 622.
56. T. Yogo and S. Naka, J. Mater. Sci., 1990, 25, 374.
57. M. T. Nguyen, M. H. Matus and D. A. Dixon, Inorg. Chem., 2007, 46, 7561.
58. M. F. Hawthorne, S. S. Jalisatgi and A. Safronov, University of Missouri-Columbia’s Progress Towards Chemical Hydrogen Storage Using Polyhedral Borane Anion Salts, DoE Hydrogen Annual Progress Report, 2007 (http://www.hydrogen.energy.gov/pdfs/progress07/iv_b_5d_hawthorne.pdf).
59. H. Nöth and H. Beyer, Chem. Ber., 1960, 93, 928.
60. For a general review on amine borane synthesis and properties see: B. Carboni and L. Monnier, Tetrahedron, 1999, 55, 1197.
61. M. F. Hawthorne, J. Am. Chem. Soc., 1961, 83, 831.
62. M. F. Hawthorne, J. Am. Chem. Soc., 1961, 83, 833.
63. A. Staubitz, M. Besora, J. N. Harvey and I. Manners, Inorg. Chem., 2008, 47, 5910.
64. A. R. Siedle, in Annual Reports on NMR Spectroscopy, ed. G. A. Webb, Academic Press, London, 1988, vol. 20, ch. 2, pp. 205–306; H. Nöth, in NMR: Basic Principles and Progress, ed. P. Diehl, E. Fluck and R. Kosfeld, Springer-Verlag, Berlin, 1978, vol. 14.
65. E. Framery and M. Vaultier, Heteroat. Chem., 2000, 11, 218.
66. H. Beall and C. H. Bushweller, Chem. Rev., 1973, 73, 465; A. Jerschow, Prog. Nucl. Magn. Reson. Spectrosc., 2005, 46, 63.
67. P. A. Storozhenko, R. A. Svitsyn, V. A. Ketsko, A. K. Buryak and A. V. Ul’yanov, Russ. J. Inorg. Chem. (Transl. of Zh. Neorg. Khim.), 2005, 50, 980.
68. M. Diwan, V. Diakov, E. Shafirovich and A. Varma, Int. J. Hydrogen Energy, 2008, 33, 1135.
69. H. C. Kelly, F. R. Marchelli and M. B. Giutso, Inorg. Chem., 1964, 3, 431; G. E. Ryschkewitsch, J. Am. Chem. Soc., 1960, 82, 3290; G. E. Ryschkewitsch and E. R. Birnbaum, J. Phys. Chem., 1961, 65, 1087; G. E. Ryschkewitsch and E. R. Birnbaum, Inorg. Chem., 1965, 4, 575; H. C. Kelly and V. B. Marriott, Inorg. Chem., 1979, 18, 2875; A. D’Ulivo, M. Onor and E. Pitzalis, Anal. Chem., 2004, 76, 6342.
70. M. Chandra and Q. Xu, J. Power Sources, 2006, 159, 855.
71. M. Couturier, B. M. Andresen, J. L. Tucker, P. Dubé, S. J. Brenek and J. T. Negri, Tetrahedron Lett., 2001, 42, 2763.
72. M. Couturier, J. L. Tucker, B. M. Andresen, P. Dubé, S. J. Brenek and J. T. Negri, Tetrahedron Lett., 2001, 42, 2285.
73. M. Couturier, J. L. Tucker, B. M. Andresen, P. Dubé and J. T. Negri, Org. Lett., 2001, 3, 465; M. Couturier, B. M. Andresen, J. B. Jorgensen, J. L. Tucker, F. R. Busch, S. J. Brenek, P. Dubé, D. J. am Ende and J. T. Negri, Org. Process Res. Dev., 2002, 6, 42.
74. M. Chandra and Q. Xu, J. Power Sources, 2006, 156, 190.
75. Q. Xu and M. Chandra, J. Power Sources, 2006, 163, 364.
76. T. J. Clark, G. R. Whittell and I. Manners, Inorg. Chem., 2007, 46, 7522.
77. P. V. Ramachandran and P. D. Gagare, Inorg. Chem., 2007, 46, 7810.
78. S. B. Kalidindi, M. Indirani and B. R. Jagirdar, Inorg. Chem., 2008, 47, 7424.
79. F. Cheng, H. Ma, Y. Li and J. Chen, Inorg. Chem., 2007, 46, 788.
80. J.-M. Yan, X.-B. Zhang, S. Han, H. Shioyama and Q. Xu, Angew. Chem., Int. Ed., 2008, 47, 2287.
81. D. A. Dixon and M. Gutowski, J. Phys. Chem. A, 2005, 109, 5129.
82. J. Zhang, S. Zhang and Q. S. Li, J. Mol. Struct. (THEOCHEM), 2005, 717, 33; Q. S. Li, J. Zhang and S. Zhang, Chem. Phys. Lett., 2005, 404, 100.
83. M. T. Nguyen, V. S. Nguyen, M. H. Matus, G. Gopakumar and D. A. Dixon, J. Phys. Chem. A, 2007, 111, 679.
84. J. Li, S. M. Kathmann, G. K. Schenter and M. Gutowski, J. Phys. Chem. C, 2007, 111, 3294; D. Jacquemin, E. A. Perpète, V. Wathelet and J.-M. André, J. Phys. Chem. A, 2004, 108, 9616.
85. M. H. Matus, K. D. Anderson, D. M. Camaioni, S. T. Autrey and D. A. Dixon, J. Phys. Chem. A, 2007, 111, 4411.
86. C. R. Miranda and G. Ceder, J. Chem. Phys., 2007, 126, 184703.
87. M. P. Brown, R. W. Heseltine and L. H. Sutcliffe, J. Chem. Soc. A, 1968, 612.
88. O. T. Beachley, Inorg. Chem., 1967, 6, 870; M. E. Bowden, I. W. M. Brown, G. J. Gainsford and H. Wong, Inorg. Chim. Acta, 2008, 361, 2147.
89. M. G. Hu, R. A. Geanangel and W. W. Wendlandt, Thermochim. Acta, 1978, 23, 249; R. A. Geanangel and W. W. Wendlandt, Thermochim. Acta, 1985, 86, 375; V. Sit, R. A. Geanangel and W. W. Wendlandt, Thermochim. Acta, 1987, 113, 379; G. Wolf, J. Baumann, F. Baitalow and F. P. Hoffmann, Thermochim. Acta, 2000, 343, 19; F. Baitalow, J. Baumann, G. Wolf, K. Jaenicke-Rößler and G. Leitner, Thermochim. Acta, 2002, 391, 159; J. Baumann, F. Baitalow and G. Wolf, Thermochim. Acta, 2005, 430, 9.
90. A good indication of whether pure boron nitride has been obtained is the IR spectrum: often, authors cite WAXS data as proof, where BN can indeed be identified, but some N–H bonds remain, which can be identified by a prominent N–H stretch band around 3600 cm⁻¹.
91. Both compounds were confirmed as volatile products in matrix isolation studies: J. D. Carpenter and B. S. Ault, Chem. Phys. Lett., 1992, 197, 171.
92. P. M. Kuznesof, D. F. Shriver and F. E. Stafford, J. Am. Chem. Soc., 1968, 90, 2557.
93. A. C. Stowe, W. J. Shaw, J. C. Linehan, B. Schmid and T. Autrey, Phys. Chem. Chem. Phys., 2007, 9, 1831; M. Bowden, T. Autrey, I. Brown and M. Ryan, Curr. Appl. Phys., 2008, 8, 498.
94. The thermal dehydrogenation of borazine to form boron nitride has elicited considerable interest due to the material properties of boron nitride. While this expands the chemical knowledge of formally the third dehydrogenation step of ammonia borane, this is of no interest for hydrogen storage purposes, as firstly, borazine is a fuel cell poison and secondly, formation of boron nitride as a thermodynamically extremely stable product makes recycling difficult. Readers who are interested in this aspect of ammonia borane chemistry are referred to, for example: A. W. Laubengayer, P. C. Moews, Jr and R. F. Porter, J. Am. Chem. Soc., 1961, 83, 1337; P. J. Fazen, J. S. Beck, A. T. Lynch, E. E. Remsen and L. G. Sneddon, Chem. Mater., 1990, 2, 96; P. J. Fazen, E. E. Remsen, J. S. Beck, P. J. Carroll, A. R. McGhie and L. G. Sneddon, Chem. Mater., 1995, 7, 1942; D.-P. Kim, K.-T. Moon, J.-G. Kho, J. Economy, C. Gervais and F. Babonneau, Polym. Adv. Technol., 1999, 10, 702.
95. S. De Benedetto, M. Carewska, C. Cento, P. Gislon, M. Pasquali, S. Scaccia and P. P. Prosini, Thermochim. Acta, 2006, 441, 184.
96. A. Gutowska, L. Li, Y. Shin, C. M. Wang, X. S. Li, J. C. Linehan, R. S. Smith, B. D. Kay, B. Schmid, W. Shaw, M. Gutowski and T. Autrey, Angew. Chem., Int. Ed., 2005, 44, 3578.
97. A. M. Feaver, S. Sepehri, P. J. Shamberger, A. C. Stowe, T. Autrey and G. Cao, J. Phys. Chem. B, 2007, 111, 7469.
98. J. S. Wang and R. A. Geanangel, Inorg. Chim. Acta, 1988, 148, 185; W. J. Shaw, J. C. Linehan, N. K. Szymczak, D. J. Heldenbrandt, C. Yonker, D. M. Camaioni, R. T. Baker and T. Autrey, Angew. Chem., Int. Ed., 2008, 47, 7493.
99. M. E. Bluhm, M. G. Bradley, R. Butterick III, U. Kusari and L. G. Sneddon, J. Am. Chem. Soc., 2006, 128, 7748.
100. F. H. Stephens, R. T. Baker, M. Hernandez-Matus, D. J. Grant and D. A. Dixon, Prepr. Pap. - Am. Chem. Soc., Div. Fuel Chem., 2006, 51, 573.
101. L. G. Sneddon, Amineborane Hydrogen Storage - New Methods for Promoting Amineborane Dehydrogenation/Regeneration Reactions, DoE Hydrogen Annual Progress Report, 2007 (http://www.hydrogen.energy.gov/pdfs/progress07/iv_b_5e_sneddon.pdf).
102. L. G. Sneddon, Amineborane Based Chemical Hydrogen Storage, DoE Hydrogen Annual Merit Review, 2007 (http://www.hydrogen.energy.gov/pdfs/review07/st_27_sneddon.pdf).
103. Y. D. Blum and R. M. Laine, US Pat., 4 801 439, 1989.
104. I. G. Green, K. M. Johnson and B. P. Roberts, J. Chem. Soc., Perkin Trans. 2, 1989, 1963.
105. Y. Jiang and H. Berke, Chem. Commun., 2007, 3571.
106. C. A. Jaska, K. Temple, A. J. Lough and I. Manners, Chem. Commun., 2001, 962; C. A. Jaska, K. Temple, A. J. Lough and I. Manners, J. Am. Chem. Soc., 2003, 125, 9424; C. A. Jaska and I. Manners, J. Am. Chem. Soc., 2004, 126, 9776.
107. Y. Chen, J. L. Fulton, J. C. Linehan and T. Autrey, J. Am. Chem. Soc., 2005, 127, 3254; J. L. Fulton, J. C. Linehan, T. Autrey, M. Balasubramanian, Y. Chen and N. K. Szymczak, J. Am. Chem. Soc., 2007, 129, 11936.
108. R. J. Keaton, J. M. Blacquiere and R. T. Baker, J. Am. Chem. Soc., 2007, 129, 1844.
109. Enders’ carbene: 1,3,4-triphenyl-4,5-dihydro-1H-1,2,4-triazol-5-ylidene; IMes: 1,3-bis(2,4,6-trimethylphenyl)-1,3-dihydro-2H-imidazol-2-ylidene; IDipp: 1,3-bis(2,6-diisopropylphenyl)-1,3-dihydro-2H-imidazol-2-ylidene.
110. T. J. Clark, C. A. Russell and I. Manners, J. Am. Chem. Soc., 2006, 128, 9582.
111. D. Pun, E. Lobkovsky and P. J. Chirik, Chem. Commun., 2007, 3297.
112. POCOP is κ³-2,6-[OP(t-Bu)₂]₂C₆H₃.
113. M. C. Denney, V. Pons, T. J. Hebden, D. M. Heinekey and K. I. Goldberg, J. Am. Chem. Soc., 2006, 128, 12048.
114. K. W. Böddeker, S. G. Shore and R. K. Bunting, J. Am. Chem. Soc., 1966, 88, 4396.
115. A. Staubitz, A. P. Soto and I. Manners, Angew. Chem., Int. Ed., 2008, 47, 6212.
116. Y. Luo and K. Ohno, Organometallics, 2007, 26, 3597.
117. A. Paul and C. B. Musgrave, Angew. Chem., Int. Ed., 2007, 46, 8153.
118. X. Yang and M. B. Hall, J. Am. Chem. Soc., 2008, 130, 1798.
119. S. Hausdorf, F. Baitalow, G. Wolf and F. O. R. L. Mertens, Int. J. Hydrogen Energy, 2008, 33, 608.
120. F. M. Taylor and J. Dewing, US Pat., 3 103 417, 1963.
121. K. C. Ott, R. T. Baker, A. K. Burrell, B. L. Davis, H. V. K. Diyabalanage, J. C. Gordon, C. W. Hamilton, M. Inbody, K. K. Jonietz, R. J. Keaton, V. Pons, T. A. Semelsberger, R. Shrestha, F. H. Stephens, D. L. Thorn and W. Tumas, Chemical Hydrogen Storage Research at Los Alamos National Laboratory, DoE Hydrogen Annual Progress Report, 2007 (http://www.hydrogen.energy.gov/pdfs/progress07/iv_b_5g_ott.pdf).
122. Z. Xiong, C. K. Yong, G. Wu, P. Chen, W. Shaw, A. Karkamkar, T. Autrey, M. O. Jones, S. R. Johnson, P. P. Edwards and W. I. F. David, Nat. Mater., 2008, 7, 138.
123. A shorter bond distance on its own cannot per se serve as proof for a stronger bond. For a discussion see for example: F. Bessac and G. Frenking, Inorg. Chem., 2006, 45, 6956.
124. H. V. K. Diyabalanage, R. P. Shrestha, T. A. Semelsberger, B. L. Scott, M. E. Bowden, B. L. Davis and A. K. Burrell, Angew. Chem., Int. Ed., 2007, 46, 8995.
125. A. J. Arduengo and D. A. Dixon, Main Group Element and Organic Chemistry for Hydrogen Storage and Activation, DoE Hydrogen Program Review, 2008 (http://www.hydrogen.energy.gov/pdfs/review08/st_9_dixon.pdf); K. Goldberg and M. Heinekey, Solutions for Chemical Hydrogen Storage: Dehydrogenation of B–N Bonds, DoE Hydrogen Program Review, 2008 (http://www.hydrogen.energy.gov/pdfs/review08/st_10_goldberg.pdf).
126. E. R. Abbey, L. N. Zakharov and S.-Y. Liu, J. Am. Chem. Soc., 2008, 130, 7250; E. R. Abbey, J. T. Jenkins, L. N. Zakharov and S.-Y. Liu, Org. Lett., 2007, 9, 4905; M. Scheideman, G. Wang and E. Vedejs, J. Am. Chem. Soc., 2008, 130, 8669.
127. C. A. Jaska and I. Manners, J. Am. Chem. Soc., 2004, 126, 2698; D. Chen and J. Klankermayer, Chem. Commun., 2008, 2130.

---

Footnotes

† Part of the renewable energy theme issue.

‡ Dedicated to Prof. M. Frederick Hawthorne on the occasion of his 80^th birthday.

---