Catalytic versus stoichiometric dehydrocoupling using main group metals

Abstract

A primary factor influencing catalyticversus stoichiometric behaviour of molecular main group species in homogeneous dehydrocoupling reactions is the redox stability of the metal centre. Thus, only in the case of redox-stable metals has catalytic behaviour so far been observed, through genuinely hydrogenic coupling (E–H + E′–H → E–E′ + H₂), whereas for redox-unstable metals oxidative dehydrocoupling is seen (E–H + E′–H → E–E′ + 2H⁺ + 2e). The mechanisms of catalytic P–H/P–H and B–H/N–H dehydrocoupling involving main group systems are closely related to d⁰ transition metal counterparts and produce a similar range of products, although the main group systems reported so far are not as active as the most active transition metal catalysts.

---

Robert J. Less

Robert J. Less was born in London in 1975. He undertook his first degree in Cambridge (1993–1996), moving on to a Ph.D. at Cambridge under Dr J. M. Rawson (1996–1999). He then went on to do postdoctoral research with Prof. P. R. Raithby (Bath) and Dr R. P. Davies (Imperial College, London) before returning to Cambridge in 2007. He is currently a Leverhulme postdoctoral fellow in the Wright group.

Rebecca L. Melen

Rebecca L. Melen was born in Nottingham in 1985. She undertook her first degree in Cambridge (2004–2008) and is currently a Ph.D. student in the Wright group working on phosphane chemistry and main group metal mediated B–H/N–H dehydrocoupling.

Dominic S. Wright

Dominic S. Wright obtained his first degree at Strathclyde University (1982–1986) before moving to Cambridge University where he did his Ph.D. under the late Dr Ron Snaith (1986–1989). After a research fellowship at Gonville and Caius College Cambridge (1989–1991), he was appointed to a lectureship at Cambridge and promoted to Reader in Inorganic Chemistry in 2002 and promoted to a personal chair in Inorganic Chemistry in 2010. He is the author of around 230 academic papers.

---

Introduction

Transition metal catalysts have enjoyed a pre-eminent place in the homogeneous and heterogeneous catalysis of a large number of technologically important reactions. According to conventional wisdom this activity is fundamentally associated with the use of valence d-orbitals in bonding, which results in the special ability of these metals to activate precursor molecules (via synergic bonding). The presence of accessible d-orbitals gives transition metals the ability to change oxidation state readily and results in generally greater covalency within metal–ligand bonds than occurs in main group counterparts.¹ This feature, combined with labile ligand coordination chemistry, underpins the use of transition metal compounds in an extremely broad range of catalytic and non-catalytic bond-forming reactions. But is the behaviour of transition metal complexes always so distinctly different from main group compounds? This important question has been highlighted by Power in regard to recent observations that the chemistry of the heavier main group elements ‘more closely resembles transition metals in their coordination and reactions with small molecules (like H₂, CO, NH₃ and C₂H₄) than that of the lighter main group congeners’.²

Studies in the past decade have uncovered general patterns of reactivity which span the perceived divide between main group and transition metal chemistry, by showing that d⁰ main group metals can also be highly active in catalytic bond-forming reactions.^3–7 There is therefore the potential to replace a number of key transition metal mediated reactions with cheaper common main group metals. There is currently world-wide interest in the synthetic applications of a variety of transition metal complexes in catalytic and non-catalytic main group element bond-forming reactions, via homo- or hetero-dehydrocoupling of E–H bonds (eqn (1)).⁸ This new field has been termed inorganometallic because of its relationship to the now ubiquitous use of transition metal species in organic transformations (organometallic),⁹ and has major future applications in molecular and polymer synthesis as well as hydrogen storage.¹⁰

Pertaining to the field of future energy resource, an emerging area of interest is the use of amine–boranes, R₂NHBH₃, as light-weight materials for the chemical storage of H₂ (eqn (2)).^10,11 One of the prerequisites for the applications of these species (e.g., in fuel cells) is the ability to release H₂ rapidly at low temperatures. A particularly important case is ammonia borane (NH₃BH₃) which is a promising hydrogen-storage material (containing 19.6% by weight hydrogen) but which decomposes at temperatures in excess of 120 °C.¹⁰ Hence, a great deal of effort has gone into the development of transition metal catalysts for this dehydrogenation step. However, the majority of these species are based on precious metals, such as Rh, Ru and Pd.¹⁰ There would be significant advantages in using far cheaper and more plentiful main group metals in this setting.¹²

In a previous review,^3a we focused on the stoichiometric dehydrocoupling of P–H bonds into P–P bonds (eqn (1), E = E′ = P) involving main group metal reagents. Here we explore the latest developments in the area of dehydrocoupling reactions involving main group metal catalysis, with the emphasis on the heteroatomic dehydrocoupling of B–H and N–H bonds (eqn (1), E = B, E′ = N). To the best of our knowledge this area has not previously been reviewed. The key issues addressed are (i) the mechanism and kinetics of dehydrocoupling involving main group metals, (ii) its comparison with transition metal systems, and (iii) the synthetic potential of main group dehydrocoupling reactions.

Stoichiometric vs.catalytic dehydrocoupling in P–H/P–H dehydrocoupling

Transition metal catalysed P–H/P–H or B–H/N–H dehydrocoupling can be seen to involve two overall types of reaction pathway. For organometallic catalysts involving partial dⁿ (n ≠ 0) configurations, the course of the catalytic pathway takes place at least in part through oxidative addition and reductive elimination cycles, although the actual mechanisms are in many cases more complicated, with different products resulting from the use of different catalysts and phosphine and amine–borane precursors.^8,10 The general features of dⁿcatalysts of this class can be illustrated by a number of studies in this area, but a simple example investigated by Tilley and coworkers is the catalytic dehydrocoupling of primary phosphines (RPH₂) to diphosphanes [RP(H)P(H)R] (2) by the Rh^I catalyst [(dppe)Rh(η³–CH₂C₆H₅)] (1) [dppe = Ph₂P(CH₂)PPh₂], in involving a Rh^I/Rh^III cycle (Scheme 1).¹³ A change in the oxidation state of the transition metal centre during the course of the catalytic cycle also occurs for the dehydrocoupling of amine–boranes (R₂NHBH₃) with a range of low-oxidation state and/or later transition metal catalysts. The mechanism of activation, although still uncertain in respect to the individual steps in the sequence, is thought to be dominated by the cycle shown in Scheme 1. An example of this that has been particularly thoroughly studied is the B–H/N–H dehydrocoupling using the 12eRh^I cation [(PPh₃)₂Rh]⁺ by Weller and coworkers (Scheme 2).¹⁴

Scheme 1

Scheme 2

The second class of transition metal species contain d⁰ metal centres. These are distinguished from the dⁿcatalytic systems in that the catalytic cycles do not involve a change in the oxidation state of the metal.⁸ Good examples of this can be seen in studies of the dehydrocoupling of phoshines by the groups of Waterman¹⁵ and Stephan involving Ti^IV and Zr^IV.¹⁶ For example, the catalytic formation of diphosphanes RP(H)P(H)R (2) with (N₃N)ZrMe (3) [N₃N=N(CH₂CH₂NSiMe₃)₃²⁻] from primary phosphines (RPH₂) is illustrated in Scheme 3.¹⁵ The dependence of the product on the catalytic system used is emphasised by the fact that, in contrast to the previous reaction, primary phosphines are converted into cyclic phosphanes ([RP]_n) 4 using the anion [Cp*₂ZrH₃]⁻ (5) as the catalyst (Scheme 4).¹⁶ Recent studies have also extended the applications of Group 4 d⁰ precatalysts like Ti(NMe₂)₄ in the dehydrogenation of amine–boranes.¹⁷

Scheme 3

Scheme 4

The significance of the activity of d⁰ transition metals in catalytic dehydrocoupling reactions, in respect to the focus of this review, is the strong relationship of active species like [(N₃N)ZrMe] (3) and [Cp*₂ZrH₃]⁻ (5) to many main group (s- and p-block) metal compounds. Thus, these catalysts have a high degree of ionic character, are highly basic and there is minimal involvement of valence d-orbitals in bonding (which are contracted in such high oxidation states). It is perhaps no surprise then that simple main group bases like metal dialkyl- and silyl-amides, and organometallics (see later, Fig. 3) have been found to be active in the dehydrocoupling of phosphines and amine–boranes. But what are the conditions for catalytic activity for main group bases of this type?

From extensive investigation of the reactivity of a range of p-block dimethylamido base, M(NMe₂)_n (n = 2, M = Sn; n = 3, M = As, Sb, Bi)^3a it is clear that although these species are highly active in the dehydrocoupling of phosphines, these reactions are nonetheless stoichiometric. This is because this is not the same type of hydrogenic coupling that is found for transition metal catalytic systems (Type 1, eqn (1)) but rather oxidative coupling (Type 2, eqn (3)). This is a consequence of the combination of the basic character of these reagents (i.e., –2H⁺) and their redox instability (i.e., –2e) under the reaction conditions. Thus the net result is the formation either of metals (M⁰) or Zintl ions resulting from metal reduction from the noxidation state.^3a,b,18 A recent example of this is seen in the dehydrocoupling of Mes*PH₂ (Mes* = 2,3,5–^tBu₃C₆H₄) with Sn(NMe₂)₂ (6) (Scheme 5), giving Sn⁰ and the P=P bonded product Mes*P=PMes* (7).¹⁹ The reduction of M in this type of reaction does not appear to be reversible and hence this can be seen as the primary reason for the stoichiometric (rather than catalytic) outcome of the reaction. These observations highlight one potential difference between transition metal catalysed dehydrocoupling systems and those of the main group, that redox stability of the metals (M) is important to the maintenance of catalytic behaviour for main group reagents, and that there is currently no counterpart to the dⁿcatalytic situation that occurs for transition metals.

Scheme 5

Further recent studies reiterate these conclusions. Firstly, it is found that [Cp*₂Sn^II] (8) behaves in much the same way as Sn(NMe₂)₂ (6) (as a redox-active base) and functions as a stoichiometric reagent in the coupling of primary phosphines into diphosphanes (2) (eqn. (4a); cf. the catalytic behaviour in Scheme 1). However, switching to [Cp*₂Sn^IVCl₂] (9) as the reagent produces at least substoichiometric behaviour with Turn Over Numbers (TON) of up to 8 (at 60 °C, 4 days) which are not dissimilar to those found for the Zr^IV catalysts discussed above (with TON = 11–20,^8,15 at 90–120 °C, 1–9 days) (eqn. (4b)). This Sn^IV-mediated reaction is apparently hydrogenic (Type 1) and involves P–H activation by the C–Sn bonds, as suggested by the formation of Cp*H during the initial stage. A tentative mechanism is outlined in Scheme 6. The product is a mixture of the diphosphanes [RP(H)P(H)R] (2), the cyclic tetraphosphane [RP]₄ (10), the pentaphosphane [RP]₅ (11) and higher-chain oligomers [RP(H)(RP)_nP(H)R] (n ≥ 1) (12). This range of products is similar to that obtained in the reactions of using [(N₃N)ZrR], mentioned previously, as is the dependence of product distribution on the nature of the R-group.¹⁵

Scheme 6

In situ ³¹P NMR spectroscopic studies, combined with fractional crystallisation of the products produced over the course of the reaction, support the view that the low TON found for Cp*₂SnCl₂ (9) is due to rapid termination of the catalytic cycle via the irreversible reduction of Sn^IV to Sn^II. Relevant to the lower activity of Sn^IIversusSn^IV is the recent observation that the catalytic activity of Group 2 vs.Group 3 amides in dehydrocoupling of amine–boranes decreases markedly with decreased Lewis acidity of the metal.^5b In the dehydrocoupling of FcPH₂ (Fc = ferrocenyl) with Cp*₂SnCl₂ the first product observed is the diphosphane [FcP(H)P(H)Fc] (13), then the tetraphosphane [FcP]₄ (14) (Fig. 1a), next the pentaphosphane [FcP]₅ (15) (Fig. 1b) and only at the end of the reaction (after 4 days) is the presumed Sn^II termination product [(FcPPFc)SnCl₂]₂ obtained (16) (Fig. 1c). The most likely source of Sn^II is from reductive elimination of ‘[RP=PR]’ or [RP]₄ from intermediates like B and C in Scheme 6.

Fig. 1 a) the tetraphosphane [FcP]₄ (14), the pentaphosphane [FcP]₅(15) and c) the presumed Sn^II termination product (16).

The lower redox stability of Ga^III intermediates compared to Al^III counterparts, especially hydrides,²⁰ also has the general effect of decreasing the catalytic activity of Ga^III amides in B–H/N–H dehydrocoupling compared to Al^III counterparts (see later discussion).⁶ A dramatic example of this has been seen recently in the reaction of the Ga^III base [Ga{N(SiMe₃)₂}₃] (17) with ammonia–borane (NH₃BH₃).^6b This reaction is stoichiometric (Type 2) rather than catalytic and gives the unusual, conjugated imido–borane [B{(NHBH)N(SiMe₃)Si(Me₂)N(SiMe₃)₂}₃] (18) (Fig. 2) in which a combination of B–N bond formation and N–Si bond cleavage of the N(SiMe₃)₂ groups occurs (Scheme 7). This result can be traced to the powerful reducing effect of the finely-divided Ga metal decomposition product.²¹

Fig. 2 Structure of the conjugated imido borane generated from stoichiometric dehydrocoupling of NH₃BH₃ with Ga {N(SiMe₃)₂}₃ (17).

Scheme 7

Main group metal catalysed dehydrocoupling of amine–boranes

Developments in the area of catalytic dehydrocoupling of amine boranes using main group elements have proceeded rapidly in the past three years. It is perhaps no surprise, bearing in mind the previous discussion of catalyticvs. stoichiometric behaviour, that the realms of catalysis of amine boranes has so far been restricted to redox-stable alkaline earth, rare earth metals, and aluminium. The most likely reason for this is the redox stability of the metals involved and the consequent stability of metal hydride intermediates in these systems, which function as the active catalysts.

The first evidence that molecular main group reagents could be active in B–N bond formation was obtained by Hill and coworkers who showed that the σ-bond metathesis reaction of β-diketiminato calcium amide complex 19 with the secondary organoborane 9-BBN (20) (9-borobicyclo[3,3,1]nonane) results in the B–N bonded product 21 and the calcium borohydride hydride complex 22 (Scheme 8).²² Since then a range of reagents have been investigated in stoichiometric and catalytic dehydrocoupling of amine boranes (Fig. 3).

Fig. 3 Main group reagents which have been employed in stoichiometric and catalytic studies with ammonia borane and amine–boranes.

Scheme 8

Later in situNMR spectroscopic and structural studies of the stoichiometric reactions of the β-diketiminato calcium and magnesium complexes (of Type A, Fig. 3), and silylamide and organometallic complexes (of Type B, Fig. 3) with ammonia–borane (NH₃BH₃),^4a and a range of amine–boranes [R¹R²NHBH₃, R¹ = H, R² = Me, ⁱPr, Dipp;[4bc,e]; R¹ = R² = Me or ⁱPr]^4d,f,g by the groups of Harder and Hill were able to identify key intermediates in the catalytic reactions. Some important representatives which were structurally characterised are shown in Fig. 4. For consistency, all of the species shown possess β-diketiminato ligands and result from reactions of type A reagents, however, similar types of intermediates have been uncovered using Type B reagents of Group 2 (Mg, Ca, Sr)⁴ and also Group 3 (Sc, Y).⁵

Fig. 4 Key types of intermediates that have been structurally characterised.

Although there is still some debate about precise mechanistic details,^4c these studies lead Hill to suggest a potentially general mechanism for the catalytic cycle involved in the formation of the range of B–N products (Scheme 9).^4d These products are highlighted in red in Scheme 9 and their formation depends primarily on the amine–borane used, and also to some extent on the kinetics of individual steps (particularly β–BHvs δ–BH elimination, producing the alternative products 27 and 28). For example, using the less sterically bulky Me₂NHBH₃ the products are (Me₂N)₂BH (27a), [Me₂NBH₂]₂ (28a) and traces of Me₂N=BH₂ (26a) (using cycles 1 and 2),^4d,f,g,5 using more bulky ⁱPr₂NHBH₃ the unsaturated imido–boraneⁱPr₂N=BH₂ (26c) is produced (using only cycle 1),^4f and using DippNH₂BH₃ gives (DippNH)₂BH (27d) (cycles 1 and 2) (see Table 1).^4c Similar products are generated using the same amine–boranes in a variety of transition metal catalysed reactions.¹⁰ The important observation that the greater the charge density of the Group 2 or 3 metals involved, the greater the catalytic activity has been ascribed to the effect of polarisation on the rate of σ-metathesis (β–BH and δ–B–H elimination) and on insertion of unsubstituted intermediates (like that in the proposed transition state 29, Scheme 9).^4d,5

Scheme 9

Table 1 Showing the products formed in catalytic reactions involving Type A and B precatalysts using different amine boranes

Amine–borane	B–N Products
a the distribution of products depends on the precatalyst employed and thus the relative rates of β–BH and δ–BH elimination.
Me2NHBH3^a
(CH2)4NHBH3
^i Pr2NHBH3
DippNH2BH3

---

In common with transition metal catalytic systems (e.g., like that shown in Scheme 2) metal hydrides are proposed as key intermediates in these dehydrocoupling reactions. Although definitive evidence has been understandably difficult to obtain from in situNMR spectroscopic studies so far,^4f strong support for this proposition, albeit indirect, was obtained from the isolation of the Zn hydride [CH{[horiz bar, triple dot above]C(Me)[horiz bar, triple dot above]N(Dipp)}₂ZnH] (31) in high yield from the stoichiometric reaction of the chloride precursor 30 with the potassium salt K[ⁱPrNHBH₃] (Scheme 10).²³

Scheme 10

The mechanistic details found for the catalytic dehydrocoupling of Al^III amide bases Al(NR₂)₃ (Type C) with amine–boranes are similar to those shown in Scheme 9 for early main group (Type A and B) precatalysts.⁶ This situation is, however, complicated by the non-innocent involvement of the ligands. Firstly, the Me₂N⁻groups of the less sterically encumbered Al(NMe₂)₃ (i.e., Me₂N = L, in Scheme 9) can exchange with the R¹R²NH⁻groups of the amine–borane (eqn (5)), leading to a complicated mixture of intermediates present in the catalytic reactions. The tendency for this reaction to occur depends on the amine–borane, R₁R₂NHBH₃, so that (for example) comparatively little exchange of Me₂N⁻ and ^tBuNH⁻ occurs for ^tBuNH₂BH₃ whereas a significant amount of exchange is observed for ⁱPr₂NHBH₃.^6b Recently, it has been suggested that in at least some cases this type of ligand exchange reaction may play a part in the overall catalytic cycle.²⁴ Secondly, the R₂N⁻groups of Al(NR₂)₃ precatalysts (Type C) can be intimately involved in generating the active catalyst species. This is seen in studies of the catalytic reaction of Al(NMe₂)₃ with Me₂NHBH₃ (Scheme 11) in which the active species is the dihydride 33, generated either by base-assisted β-hydride transfer to Al^III (as depicted) or by β–BH elimination followed by addition of Me₂N⁻ to a metal-bonded, unsaturated Me₂N=BH₂. Recent evidence has suggested that reactions involving early main group Type B precatalysts may not be as simple as initially proposed in Scheme 9, and that here also the same type of addition of previously-assumed innocent ligands to unsaturated B=N intermediates can occur.^4g

R¹R²NHBH₃ + Al(NMe₂)₃ → Me₂NHBH₃ + (R¹R²N)Al(NMe₂)₂ (5)

Scheme 11

Once the catalyst 33 is established, the final B–N product [Me₂NBH₂]₂ (28a) (Table 1) appears to be generated via a similar set of reactions to Cycle 2 in Scheme 9, involving δ–BH elimination. The big difference, however, is that the minor product in this reaction (Me₂N)₂BH (27a) (Table 1) is not formed by a β-hydride elimination within this cycle but only in the initiation stage during the formation of the catalyst 33via dissociation of 32 (Scheme 11). The absence of free Me₂N=BH₂ (26a) (Table 1) as even a transient product in this reaction supports the view that the catalytic mechanism involved in this case is entirely metal-based.

An extremely long induction period of up to 30 min occurs in the case of the catalytic reaction of (ⁱPr₂N)₃Al with ⁱPr₂NHBH₃ as a result of the slow formation of the apparently active dihydride species [H₂Al(μ-NⁱPr₂)]₂ (34) (Scheme 12), which is structurally similar to the catalytic species 33 involved in the reaction between Al(NMe₂)₃ and Me₂NHBH₃.^6b In this case, termination of the catalytic cycle (which is similar to Cycle 1 in Scheme 9) apparently occurs by formation of an inactive BH₄⁻ complex (35) which only appears at the end of the reaction. This is closely related to the deactivation of the dihydride pincer catalyst [(POCOP)IrH₂] [POCOP = η¹-1,3-(OP^tBu₂)₂C₆H₃] with BH₃ in the dehydrogenation reaction with ammonia borane, NH₃BH₃.²⁵ However, an alternative explanation for the final formation of a BH₄⁻ complex is that this is an additional catalytic species or a catalyst rest state.^4c

Scheme 12

The reaction of Al(NMe₂)₃ with ^tBuNH₂BH₃ is of particular interest in respect to comparison with transition metal catalysis. This reaction gives a mixture of the trimeric borazane [BH₂N(H)^tBu]₃ (36), as well as the borazine [BHN^tBu]₃ (37) and unidentified polymeric B–N compounds (Scheme 13).^6b The apparent intermediacy of the borazane 36 in this reaction is similar to the dehydrocoupling reactions of a range of amine–boranes [RNH₂BH₃] (R = H, Me, Ph) with the transition metal catalyst [Rh(1,5-cod)(μ-Cl)]₂.²⁶ However, in the case of the Rh^I-catalysed reactions the rate of formation of the borazanes is much faster than that of the subsequent dehydrogenation to the borazines, resulting in slow build up in the borazane with time before it is slowly consumed to give the borazine. This is the opposite to the situation found in the Al^lII system in which the borazane and borazine are formed together as the reaction proceeds.

Scheme 13

Comparison of rate data, transition metal vs. main group metal

A major obstacle to accurate comparison of the activity of main group and transition metal catalysed reactions in the area of dehydrocoupling is the lack of reported quantitative data like TON₅₀ (Turnover Number at 50% conversion) and TOF (Turnover Frequency), and the broad range of solvents and conditions that have been employed, which may differ significantly from one reaction to another. More often than not the activity quoted in the literature is based purely on a semi-quantitative expression of a certain percentage conversion in a particular time. Table 2 lists all of the data currently available using main group precatalysts and converts this into approximate TON₅₀ and TOF values so that some broad comparison can be made. The main conclusion which can be drawn is that, although not anywhere near as active as the most active transition metal dehydrocoupling catalysts recently introduced (with TON₅₀ca. 5000 and TOF 6000 h⁻¹),^14,27 main group catalysts are comparable in activity with a broad range of other transition metal catalysts used to effect the same transformations. Thus the range of activities previously reported for the majority of dehydrocoupling reactions of Me₂NHBH₃, ⁱPr₂NHBH₃ and ^tBuNH₂BH₃ involving transition metals (where data is available, from Table 2 reference 10a) correspond to TON₅₀ = 4–100 and TOF = 0.1–2.8 h⁻¹, which compares to TON₅₀ = 5–100 and TOF = 0.1–30 h⁻¹ for main group catalysis.

Table 2 Table showing the TON₅₀ and TOF data currently available on main group catalysed reactions of amine–boranes. TOF = (mol substrate converted)/(mol. cat × time) = (% conversion/%loading)/time; TON₅₀ = number of molecules generated half way through reaction, taking into account the proportion of converted at the end of the reaction: TON₅₀ = TOF × (t/2) × (100% conversion)

Substrate	Catalyst	% loading	Product	Approx. % conversion	Conditions	TON50	TOF	Ref.
Me2NHBH3	Mg[CH(SiMe3)2]2(thf)2	5 mol (%)	[Me2NBH2]2	100%	60 °C, 72 h, C6D6	10	0.28	4d
Me2NHBH3	Y[N(SiMe3)2]3	3 mol (%)	[Me2NBH2]2	86%	60 °C, 12 h, C6D6	17	2.39	5
Me2NHBH3	Y[N(SiMe3)2]3	3 mol (%)	[Me2NBH2]2	4%	25 °C, 0.25 h, C6D6	17	5.33	5
Me2NHBH3	Sc[N(SiMe3)2]3(thf)2	3 mol (%)	[Me2NBH2]2	10%	25 °C, 1 h, C6D6	17	3.33	5
Me2NHBH3	Sc[N(SiMe3)2]3(thf)2	3 mol (%)	[Me2NBH2]2	90%	60 °C, 1 h, C6D6	17	30.00	5
Me2NHBH3	Al(NMe2)3	8 mol (%)	[Me2NBH2]2	80%	25 °C, 120 h, C7D8	6	0.08	6b
Me2NHBH3	Al(NMe2)3	5 mol (%)	[Me2NBH2]2	100%	50 °C, 48 h, C7D8	10	0.42	6b
^i Pr2NHBH3	Al(N^iPr2)3	2 mol (%)	^i Pr2NBH2	100%	20 °C, 2 h, C7D8	25	25.00	6b
^i Pr2NHBH3	Al(N^iPr2)3	10 mol (%)	^i Pr2NBH2	100%	60 °C, 2 h, C6D6	5	5.00	6b
^i Pr2NHBH3	[H2Al(μ-N^iPr2)]2	0.5 mol (%)	^i Pr2NBH2	50%	20 °C, 96 h, C7D8	100	1.04	6b
^i Pr2NHBH3	Mg[N(SiMe3)2]2	5 mol (%)	^i Pr2NBH2	100%	25 °C, 1 h, C6D6	10	20.00	4f
^t BuNH2BH3	Al(NMe2)3	3 mol (%)	[^tBuNBH]3	13%	20 °C, 96 h, C6D6	17	0.05	6b

---

Conclusions and future prospects

From the survey of the reactivity of main group metals in catalysis in this area it can be concluded that (i) the catalytic activity of homogeneous molecular main group metal catalysts is very dependent on the redox stability of the metal and that the course of dehydrocoupling reactions [whether dehydrogenic (catalytic) or stoichiometric (oxidative)] is determined by this redox stability, (ii) that the mechanisms of dehydrocoupling of phosphines (RPH₂) and amine boranes (R¹R²NHBH₃) with d⁰ transition metals and main group metals are surprisingly similar and involve related intermediates, and (iii) that although the activity of main group catalysts is far lower than that of the most active transition metal systems, the activity is similar to a large number of previously reported transition metal systems.

Future work in the area of main group catalysis in dehydrocoupling is likely to focus on the development of new types of E–E′ bond-forming reactions and the optimisation of catalyst activity. It can be noted in regard to the latter point that there have previously been no attempts to vary systematically the ligand set present in main group catalysts systems in order to optimise activity, as is the norm in the development of transition metal catalytic systems.^1,14d As is already apparent from studies so far,⁴ changing the steric demands of the ligands in the precatalysts (Types A, B and C, Fig. 3) is one obvious way of modifying the kinetics and selectivity. However, a possible additional way of enhancing activity in the future is increasing the electronegativity of the ligands. The effects of this should be to increase the Lewis acidity of the metal centres, with the additional result of increasing redox stability; leading to greater polarisation of metal-bonded intermediates⁵ and potentially longer catalyst life-times.

Acknowledgements

We thank the EPSRC (DSW), The Leverhulme Trust (RJL, DSW), and Cambridge University (RLM).

References

1. J. P. Collman, L. S. Hegedus, J. R. Norton, R. C. Finke, Principles and Applications of Organotransition Metal Chemistry, University Science Books, U.S., 2nd revised edn 1987.
2. P. P. Power, Nature, 2010, 463, 171.
3. (a) P–H/P–H coupling; R. J. Less, R. L. Melen, V. Naseri and D. S. Wright, Chem. Commun., 2009, 4929; (b) V. Naseri, R. J. Less, M. McPartlin, R. E. Mulvey and D. S. Wright, Chem. Commun., 2010, 46, 5000; (c) M. S. Hill, M. F. Mahon and T. P. Robinson, Chem. Commun., 2010, 46, 2498.
4. (a) Group 2 B–H/N–H coupling; J. Spielmann, G. Jansen, H. Bandmann and S. Harder, Angew. Chem., Int. Ed., 2008, 47, 6290; (b) J. Spielman and S. Harder, J. Am. Chem. Soc., 2009, 131, 5064; (c) J. Spielmann, M. Bolte and S. Harder, Chem. Commun., 2009, 6934; (d) D. J. Liptrot, M. S. Hill, M. F. Mahon and D. J. MacDougall, Chem.–Eur. J., 2010, 16, 8508; (e) J. Spielmann, D. F.-J. Piesik and S. Harder, Chem.–Eur. J., 2010, 16, 8307; (f) M. S. Hill, M. Hodgson, D. J. Liprot and M. F. Mahon, Dalton Trans., 2011, 40, 7783; (g) P. Bellham, M. S. Hill, D. J. Liptrot, D. J. MacDougall and M. F. Mahon, Chem. Commun., 2011, 47, 9060.
5. Group 3 B–H/N–H coupling; M. S. Hill, G. Kociok-Köhn and T. P. Robinson, Chem. Commun., 2010, 46, 7587.
6. (a) Group 13 B–H/N–H coupling; H. J. Cowley, M. S. Holt, R. L. Melen, J. M. Rawson and D. S. Wright, Chem. Commun., 2011, 47, 2682; (b) M. M. Hansmann, R. L. Melen and D. S. Wright, Chem. Sci., 2011, 2, 1554.
7. (a) Other catalytic reactions involving Group 2 metal reagents; M. R. Crimmins, M. Arrowsmith, A. G. M. Barrett, I. J. Casey, M. S. Hill and P. A. Procopiou, J. Am. Chem. Soc., 2009, 131, 9670; (b) J. Spielmann and S. Harder, Eur. J. Inorg. Chem., 2008, 1480; (c) J. Spielmann, F. Buch and S. Harder, Angew. Chem., Int. Ed., 2008, 47, 9434.
8. (a) S. Greenberg and D. W. Stephan, Chem. Soc. Rev., 2008, 37, 1482; (b) D. W. Stephan, Angew. Chem., Int. Ed., 2000, 39, 314; (c) T. J. Clark, K. Lee and I. Manners, Chem.–Eur. J., 2006, 12, 8634; (d) R. Waterman, Curr. Org. Chem., 2008, 12, 1322; (e) R. Waterman, Dalton Trans., 2009, 18.
9. T. P. Fehlner, Inorganometallics, Plenum Press, New York 1992.
10. (a) C. W. Hamilton, R. T. Baker, A. Staubitz and I. Manners, Chem. Soc. Rev., 2009, 38, 279; (b) A. Staubitz, A. P. M. Robertson, M. E. Sloan and I. Manners, Chem. Rev., 2010, 110, 4023.
11. A further important issue in H₂storage is regeneration of the unsaturated, dehydrogenated products (‘spent fuel’) back to amine boranes, see A. A. D. Sutton, A. K. Burrell, D. A. Dixon, E. B. Garner III, J. C. Gordon, T. Nakagawa, K. C. Ott, J. P. Robinson and M. Vasiliu, Science, 2011, 331, 1426.
12. V. Pons, R. T. Baker, N. K. Szymczak, D. J. Heldebrant, J. C. Linehan, M. H. Matus, D. J. Grant and D. A. Dixon, Chem. Commun., 2008, 6597.
13. L.-B. Han and T. D. Tilley, J. Am. Chem. Soc., 2006, 128, 13698.
14. (a) See T. M. Douglas, A. B. Chaplin, A. S. Weller and J. Am, J. Am. Chem. Soc., 2008, 130, 14432; (b) R. Dalanegra, A. B. Chaplin and A. S. Weller, Angew. Chem., Int. Ed., 2009, 48, 6875; (c) A. B. Chaplin and A. S. Weller, Angew. Chem., 2010, 122, 591; (d) R. Dallanegra, A. P. M. Robertson, A. B. Chaplin, I. Manners and A. S. Weller, Chem. Commun., 2011, 47, 3763 and references therein.
15. R. Waterman, Organometallics, 2007, 26, 2492.
16. (a) M. C. Fermin and D. W. Stephan, J. Am. Chem. Soc., 1995, 117, 12645; (b) N. Etkin, M. C. Fermin and D. W. Stephan, J. Am. Chem. Soc., 1997, 119, 2954; (c) N. Etkin, M. C. Fermin and D. W. Stephan, Angew. Chem., Int. Ed., 2001, 40, 1865; (d) J. D. Masuda, A. J. Hoskins, T. W. Graham, C. Beddie, M. C. Fermin, N. Etkin and D. W. Stephan, Chem.–Eur. J., 2006, 12, 8696.
17. T. Beweries, S. Hansen, M. Kessler, M. Klahn and U. Rosenthal, Dalton Trans., 2011, 40, 7689.
18. M. A. Beswick, N. Choi, C. N. Harmer, A. D. Hopkins, M. McPartlin and D. S. Wright, Science, 1998, 281, 1500.
19. M. McPartlin, R. L. Melen, V. Naseri and D. S. Wright, Chem.–Eur. J., 2010, 16, 8854.
20. J. L. Atwood, K. Robinson, F. R. Bennett, F. M. Elms, G. A. Koutsantonis and C. L. Raston, Inorg. Chem., 1992, 31, 2673 and references therein.
21. (a) A. G. Avent, P. B. Hitchcock, A. V. Khvostov, M. F. Lappert and A. V. Protchenko, Chem. Commun., 2004, 2272; (b) see also B. Luo, V. G. Young Jr. and W. L. Gladfelter, J. Organomet. Chem., 2002, 649, 268.
22. M. R. Crimmins, A. G. M. Bartlett, M. S. Hill, P. B. Hitchcock and P. A. Procopiou, Organometallics, 2007, 26, 4076.
23. J. Spielmann, D. Piesik, B. Wittkamp, G. Jansen and S. Harder, Chem. Commun., 2009, 45, 3455.
24. S. Harder and J. Spielman, Chem. Commun., 2011, 47, 11945.
25. (a) M. C. Denney, V. Pons, T. J. Hebden, D. M. Heinekey and K. I. Goldberg, J. Am. Chem. Soc., 2006, 128, 12048; (b) T. J. Hebden, M. C. Denny, V. Pons, P. M. B. Piccoli, T. F. Koetzie, A. J. Schultz, W. Kaminsky, K. I. Goldberg and D. M. Heinekey, J. Am. Chem. Soc., 2008, 130, 10812.
26. C. A. Jaska, K. Temple, A. J. Lough and I. Manners, J. Am. Chem. Soc., 2003, 125, 9424.
27. (a) A. Staubitz, A. P. Sato and I. Manners, Angew. Chem. Int. Ed., 2008, 47, 6212; (b) A. Staubitz, M. E. Sloan, A. P. M. Robertson, A. Friedrich, S. Schneider, P. J. Gates, J. Schmedt auf der Günne and I. Manners, J. Am. Chem. Soc., 2010, 132, 13332.

---