N-Heterocyclic carbenes/imidazolium salts as substrates in catalysis: the catalytic 2-substitution and annulation of heterocyclic compounds

Abstract

N-Heterocyclic carbenes (NHCs) are commonly regarded as strong donor ligands that are valuable in coordination chemistry and catalysis. Many reports describing this aspect of their chemistry have been published. The alternative view of NHCs as reaction intermediates has been little explored, and yet excellent examples exist and will be reviewed in this perspective. Group 10 hydrocarbyl complexes of NHCs [(R)(NHC)ML₂ where R = H, alkyl, aryl, acyl; M = Ni, Pd, Pt] undergo a facile reductive elimination to generate M(0) and R-substituted-azolium salt as products. On the other hand, 2-Hazolium salts will oxidatively add to M(0) complexes to afford reactive NHC–M–H compounds, suitable as catalysts for selected reactions. Combining the oxidative addition and reductive elimination steps into a single cycle, in the presence of an alkene, provides of a novel and potentially exciting, atom efficient catalytic C–C bond forming process for the substitution, and annulation of heterocyclic rings.

---

Kingsley Cavell

Kingsley Cavell was appointed Professor of Inorganic Chemistry, Cardiff University in 2000, where he is currently the Head of School and Group leader of the Metals in Synthesis Research Group. He received his PhD from La Trobe University (Melbourne, Australia) and then undertook post-doctoral studies at the University of Manchester (with Professor Hank Skinner and Dr Geoff Pilcher) and at the University of Amsterdam (with Professor Dr Kees Vrieze). This was followed by a period with the CSIRO (Div. of Materials Sci. and Tech., Melbourne) (1980–1986), and the University of Tasmania (1986–2000). He has been Guest Professor at the Institut für Chemie und Petrochemie der RWTH Aachen, Germany (1989/90 and 1994; with Professor Dr W. Keim), and in the Department of Chemistry, Huazhong University of Science and Technology, Wuhan, PR China (1997; with Professor Li Guangxing). He was awarded the 2001 RSC Industrially-Sponsored Award in Homogeneous Catalysis, sponsored by Synetix. His research is focussed on organometallic chemistry, homogeneous catalysis and on elucidation of reaction mechanisms relating to catalytic processes; a key component of the research has been directed at the study of heterocyclic carbenes as ligands and as reaction intermediates.

---

Introduction

The isolation of free N-heterocyclic carbenes (NHCs) and their use as ligands in organometallic chemistry and catalysis has generated considerable activity in the chemistry community. Developments have been rapid, and an exponential increase in publications has occurred over the past ten years. Numerous, excellent reports and reviews have appeared providing an experimental and theoretical description of these unique species, and very recently a whole volume of Coordination Chemistry Reviews was dedicated to various aspects of NHC chemistry.^1,2 The two-electron donor NHCs have been widely compared to the ubiquitous phosphines as ligands. However, features such as the high basicity of the NHC and hence strong carbene–metal bonding, particularly with later transition metals, and the structural features (directional, wing-tip distribution of steric bulk) of the NHC set these ligands apart. Some debate exists as to the ability of NHCs to undergo π-back-bonding; acknowledging that the electron distribution within the NHC and the C_carbeneπ-orbital is dynamic, a degree of backbonding, in some circumstances, can be expected. However, what is clear is that it is not necessary to invoke back-bonding to provide a strong bonding interaction and hence a stable metal complex.

In the current literature NHCs have been viewed almost exclusively as strong donor ligands for the synthesis of metal complexes. Therefore, not surprisingly, NHCs have been widely applied in catalysis, and with some success, particularly in selected reactions such as olefin metathesis.³ Such powerful ligands should in principle provide stable catalyst systems, and in some cases examples of carbene complexes have been used at very high temperatures (120–180 °C or higher) offering some promise in this respect.

Early investigations into the application of NHCs in catalysis were initiated in the 1970s.⁴ However, it was only after the isolation of the first free NHCs,⁵ and the influential catalytic studies by Herrmann that the field really emerged.⁶ Nevertheless, NHCs have not been universally applicable as ligands in catalysis, and it soon became apparent that despite strong metal–NHC bonding interactions, fundamental “problems” exist in the chemistry of these ligands; NHCs are susceptible to loss by reductive elimination, leading to decomposition of the catalyst, in some instances before effective catalysis can take place.^7,8 Unlike their role as spectator ligands in catalysis, an alternative function of NHCs, as intermediates in chemical reactions, which is more in keeping with the role of “traditional” carbenes in synthesis and Schrock type carbenes in catalysis, is virtually unexplored. In the following discussion the view that NHCs act only as simple donor ligands towards transition metals will be clearly dispelled, and a facile coupling reaction between NHCs and olefinic moieties is presented. The application of NHCs as reaction intermediates, or “substrate”, which finish up as products, will be described.

Fundamental aspects of the chemistry of NHCs and related imidazolium salts are reported, and subsequently how this reaction behaviour has been utilised to construct a novel atom efficient C–C coupling reaction via C–H activation. This perspective is not intended as a comprehensive review and hence there has been a selective reporting of the literature most relevant to the topic under discussion.

Reductive elimination: N-heterocyclic carbenes as “participative” ligands

Following the synthesis of the first examples of methyl–palladium NHC complexes an investigation into their thermal decomposition showed an unexpected reaction pathway.^9–11 All compounds investigated readily decomposed by a common process (for example complexes 1, 2 in Scheme 1); the carbene is lost as 2-methylimidazolium salt and the Pd is reduced to Pd(0). The reaction could be followed by NMR, and in several cases the imidazolium salt was isolated and fully characterized.¹⁰Bis-carbene complexes and complexes of donor functionalised carbenes were also found to decompose by a similar route (eqn (1)).^10,11

Scheme 1

A comprehensive experimental and theoretical study was able to demonstrate that the observed reaction was a concerted reductive elimination process; kinetic studies were consistent with reductive elimination, and DFT calculations supported a reductive elimination mechanism.⁷ Calculations on a “real” system indicated an “associative” pathway with a small activation barrier (14.1 kcal mol⁻¹) and overall endothermic reaction (−9.2 kcal mol⁻¹) (Scheme 2). In the transition structure (TS) the change in orbital populations is in accord with reductive elimination; there is mixing of C_carbene p(π), the C_Me and the Pd d orbitals. This mixing of orbitals allows the C_carbene–C_Me bond to form (Fig. 1). The plane of the NHC ligand is at a substantial angle to the coordination plane (commonly 60–80°) and hence the C_carbene p(π), which is perpendicular to the NHC plane is correctly oriented to interact with the Me group (Fig. 1). The d orbital with the appropriate orientation for the three-centre interaction is the d_xy orbital. In the encounter complex the population of the d_x²−y² orbital increases along with the population of the 5s orbital; the C_carbene–C_Me bond is fully occupied with no d involvement and the N p(π) → C_carbene p(π) is purely p in character, as in the starting complex. The p(π) occupancy of the C_carbene has increased due to greater delocalization of the imidazolium cation.

Fig. 1

Scheme 2

Impact of reductive elimination on the application of NHCs in catalysis

Electronic and structural factors associated with NHC ligands generate conditions wherein they are particularly susceptible to reductive elimination. This reaction has major consequences for the application of NHC–M complexes in catalysis. The alkylcarbene reductive coupling can be facile resulting in decomposition of the catalyst. Aryl–M–carbene complexes (M = Pd, Ni) also readily decompose by a reductive elimination process to give 2-arylimidazolium salts. In a fundamental study, investigating oxidative addition of aryl halides to Pd(0), and the Heck reaction, the addition of aryl iodide to Pd(tmiy)₂(MAH) (tmiy = 1,2,3,4-tetramethylimidazol-2-ylidene, MAH = maleic anhydride) at 60 °C yielded a mixture of products, which included the desired oxidative addition product Pd(tmiy)₂(Ph)I and the product 2-phenylimidazolium salt, 4, resulting from reductive coupling of the aryl group and tmiy.¹² Moreover, warming the complex (tmiy)₂Pd(Ar)I (Ar = 4-nitrobenzene), formed from the room-temperature oxidative addition of aryl iodide to Pd(tmiy)₂, gave rise to the reductive elimination product 5 as the major product. Caddick, Cloke et al. observed a similar behaviour when investigating catalytic amination reactions using Pd(NHC)₂ complexes as precatalysts; the oxidative addition of aryl chloride to Pd(NHC)₂ [where NHC = 2,6-di(PrⁱC₆H₃)imidazolidine-2-ylidene] gave only the product 6.¹³

The competitive interplay between decomposition on one hand, and effective catalysis on the other is readily seen from an investigation of a stoichiometric Heck reaction between n-butyl acrylate and the complex (tmiy)₂Pd(Ar)I (Ar = 4-nitrobenzene), at sub-ambient temperatures. The reaction beautifully illustrates this competition between reductive elimination and Heck coupling (Scheme 3).¹² Under these conditions, both Heck chemistry and reductive elimination are occurring, and the relative degree of each is temperature dependent; at lower temperatures reductive elimination predominates, however, as the temperature is raised and as catalytic conditions are applied the Heck reaction proceeds almost exclusively. In the former case the coordinated NHC is behaving as a reaction intermediate, whereas in the latter it is a “non-participative” spectator ligand. Interestingly, the decomposition products isolated from the reaction (products in boxes) provide a clear steer on the mechanism of the Heck reaction. Furthermore, from Scheme 3 it is evident that as well as carbene–aryl reductive coupling carbene–hydride reductive elimination also occurs, to give 2-hydridoimidazolium salt, 7.

Scheme 3

In subsequent studies it was demonstrated that reductive elimination is also extremely facile for carbene–Pd–acyl complexes, yielding 2-acylimidazolium salts and Pd(0).¹⁴ The product distribution in such cases was shown to depend on the structure of the complexes from which reductive coupling took place (Scheme 4); a cis-(NHC)/(Me) arrangement (pathway A) gave 2-methylimidazolium as the major product, some 2-acylimidazolium was also observed; a probable cis-Me/CO structure (pathway B) gave 2-acylimidazolium salt as the major product, and 2-methylimidazolium as the minor product. Significantly, in each of these examples the NHC is again acting as substrate and ends up in the products.

Scheme 4

An investigation of Ni-NHC based olefin dimerization catalysis has provided a further notable example of the impact of reductive coupling on the coordinated NHC, and on the catalytic behaviour of the metal complex.⁸Catalyst systems of the type [NiL₂X₂ + AlEt_nX_3−n] (where L = PR₃ and X = halide) are well known and generally afford highly active catalysts for olefin dimerization and isomerization. In the study of interest here, Ni(NHC)₂I₂ complexes, 8 (Scheme 5), were synthesised, and treated with AlEt₂Cl as cocatalyst, in the presence of 1-butene, in toluene at 20 °C. Analysis of the products from the reaction showed only reductive elimination products, I, II, III (Scheme 5). No butene dimers were observed. Decreasing the temperature to −15 °C, and using the complex with 1,3-diisopropylimidazol-2-ylidene as the NHC ligand generated very small amounts of butene dimers. These results unambiguously demonstrate that Ni–NHC complexes are capable of olefin dimerization, however, decomposition of the catalystvia reductive elimination prevents completion of the desired catalytic cycle. Moreover, the specific decomposition products observed here provide unequivocal evidence that this dimerization reaction follows the Cossee–Arlman mechanism of activation, insertion and elimination, commonly invoked for olefin chain-growth processes.

Scheme 5

When the matching dimerization process, using the same NHC–Ni complexes as catalyst precursors, was conducted in ionic liquid (IL) solvent [1-butyl-3-methylimidazolium chloride–AlCl₃–N-methylpyrrole (0.45 : 0.55 : 0.1 ratio)] rather than toluene, the catalyst was found to be highly active with no evidence of decomposition.⁸ Furthermore, product distributions for each of the catalyst systems studied was surprising similar, indicating a common active species may have been formed in each case. To explain such an interpretation it was suggested that reductive elimination of the NHC–Ni did indeed occur, as outlined in Scheme 5; however, the ionic liquid solvent could now oxidatively add to the Ni(0) thus formed to yield a new Ni–NHC complex, stabilized by the IL solvent, and able to effectively catalyse the dimerization process (Scheme 6). Such a proposal would of course require evidence that oxidative addition of imidazolium salt is indeed feasible (vide infra).

Scheme 6

Factors effecting reductive elimination

A number of studies have been undertaken in an attempt to understand factors that effect the susceptibility of a complex to reductive elimination—these include: (a) a combined experimental and DFT investigation of Me–Pd–CNC pincer complexes;¹⁵ studies on (b) the influence of geometry in Me–Pd–NHC complexes;¹⁶ (c) the impact of N-substituents on reductive elimination behaviour;¹⁷ and (d) an experimental study on the influence of chelating spectator ligands in complexes of the type [Pd(NHC)Me(P–P)]BF₄.¹⁸ In the first report (a), complementary experimental and computational studies on Pd–CNC pincers, Scheme 7, has shown that the pincers are extremely thermally stable, however, when decomposition does occur it is by a partial reductive elimination mechanism.^15,19 Complexes of type 9 (n = 1, R = Z = Me), and 10, when operated under catalytic conditions, show long-term stability at high temperatures (>180 °C in several cases), with excellent catalytic activities and selectivities.^20–22

Scheme 7

A DFT computational study (b), on the influence of the spectator ligand bite angle and the twist angle of the NHC on reductive elimination from PdL₂ complexes (L = phosphine), demonstrated that bite angle has a significant impact.¹⁶ Increasing the bite angle (across a range 80–130°; 12) lowered E_act and increased the exothermicity of the reaction. Complexes with a large bite angle >110° showed a thermodynamic preference for products, complexes with a bite angle <110° prefer the reverse reaction—oxidative addition. Interestingly, steric bulk on the phosphines appears to be just as important as bite angle in influencing the reaction.^16,23 Rotation of the NHC with respect to the coordination plane of the complex 13 had little influence on E_act but has significant effect on the thermodynamics of the decomposition reaction. The instability of the complexes where the carbene twists towards coplanarity with the coordination plane can be attributed to steric strain rather than electronic influences.

A further computational study (c), using DFT methods, investigated the influence of the NHC N-substituents on both the E_act and the overall thermodynamics of the reductive elimination reaction, 14.¹⁷ Electronic factors have a major influence on the activation barrier. The NHC pπ orbital plays a key role in the reductive elimination process and N-substituents that remove π-density act to promote elimination, whereas increased electron donation from the N-substituents stabilise against reductive elimination.¹⁷

An experimental study, (d), found that chelating spectator ligands impart a degree of stability to the complexes of type 15.¹⁸ When two monodentate phosphine ligands are used decomposition is rapid at 20 °C, however, if the monodentate ligands are replaced by dppp no decomposition is detected after 24 h.⁷ Comparing different chelating phosphine ligands, it was found that the rate of decomposition could ostensibly be linked to the chelate ring size; at 65 °C, with dppp decomposition was complete after 6 hrs, with dppe only a small amount of decomposition had occurred in this time.¹⁸

Having recognised the problem presented by reductive elimination it is possible to consider methods of limiting the impact of the reaction. A number of solutions have been proposed and tried with mixed success. For example the use of chelating ligands incorporating an NHC component has worked relatively well but is limited to catalyst systems in which a high degree of coordinative unsaturation is not essential.^15,20,24

Another approach is the application of imidazolium based ILs as solvents, which has been shown to have clear benefits (vide supra), by providing highly active and very stable catalyst systems.⁸

Oxidative addition of azolium salts (ILs) to low valent metal centres: generation of M–N-heterocyclic carbene complexes

In early studies, Roper et al. and co-workers found that 2-chloro-derivatives of azolium salts oxidatively add to low valent d⁸ and d¹⁰ metal centres to form M–NHC complexes (M = Ir, Pd, Pt, Ni).²⁵ Much later it was shown, using both experiment, and DFT calculations, that oxidative addition of 2-H-azolium salts is a low energy pathway, readily yielding carbene–M–H complexes [Scheme 8, eqn (3)–(5)].^26–28 The main features of this reaction are that it is a useful synthetic method for the preparation of novel carbene complexes^27–31 and, possibly more importantly, oxidative addition of 2-hydridoimidazolium salt generates, in situ, NHC–M–H complexes, which are considered to be active species in many catalytic reactions.

Scheme 8

Recent DFT computational studies explored different aspects of the oxidative addition reaction.^32,33 In the first study, oxidative addition of azolium salts to a model Wilkinson's catalyst was investigated.³² Ironically, the calculations show that the reverse reaction (reductive elimination to give the original imidazolium salt and the Rh(I) complex) is thermodynamically preferred, suggesting that reductive elimination could occur in NHC–Rh(I)–H complexes. The second computational study looked at the oxidative addition of 2-substituted azolium salts to group 10 metals (eqn (6)) to give complexes 19.³³ It appeared that all three metals would have little trouble undergoing oxidative addition, particularly when the 2-substituent was H. Interestingly, it was predicted that “under the right conditions” even the C2–Me “blocked” IL [R = Me, eqn (6)], would also undergo oxidative addition. Furthermore, oxidative addition of oxazolium salts [X = O, eqn (6)] to a group 10 metal centre is highly energetically favourable, providing a valuable synthetic route to new NHC complexes.

Redox catalysis involving the NHC–M–H intermediate: imidazolium salts/NHCs as substrate

By combining the two half-reactions described above (oxidative addition of imidazolium salts, and the reductive elimination from hydrocarbyl–M–NHC complexes), in the presence of an unsaturated substrate, a novel atom efficient, C–C coupling reaction was developed (Scheme 9).³⁴ The reaction, which demonstrates the application of the imidazolium salt/NHC couple as substrate, integrates a potentially destructive decomposition reaction (reduction elimination) into a useful process. In the reaction the imidazolium salt 20 (an IL) is converted to the 2-substituted salt 21, i.e. a reactive IL is converted to a far less reactive alternative, and one which offers new design opportunities for IL solvents.

Scheme 9

A range of olefins and of azolium salts have been utilised as substrates in this reaction (Table 1), providing a variety of 2-substituted heterocyclic compounds.^34,35Catalytic performance is modest and high catalyst loadings are required. Of significance, neutral azoles could also be used in the reaction [eqn (8)], further extending its potential.³⁵ However, in the case of the neutral azole it was necessary to activate the 2-H by adding a strong Lewis acid to the system.

Table 1 Substrates and products from the novel redox reaction between olefins and of azolium salts

---

A comprehensive DFT investigation of the mechanism has been carried out, and provides a valuable indication why catalyst performance is only modest.³⁵ The reaction mechanism appears quite complex, and subtle changes to the catalyst lead to changes in the rate-determining step. For small, difficult to dissociate (very basic ligands), Scheme 9, L= PMe₃, the main activation barrier is ligand dissociation prior to alkene coordination. For bulkier, more readily dissociated ligands (L = PPh₃, PBu^t₃) the main energy barrier is the reductive elimination step, although the coordination of alkene can also be high in energy in these systems. In all cases, the oxidative addition step was facile and unlikely to be rate-determining. Concurrent, and extensive, experimental studies assessing ligand (L) sterics and electronic properties, and L : Ni ratios were in very good agreement with the computational studies.³⁵

In subsequent studies the intramolecular 2-substitution, or annulation of a range of azolium salts was investigated.³⁶ The reaction converts N-alkenyl and N-substituted alkenyl substituents into five- and six-membered fused ring azolium species, efficiently and in high yield (Scheme 10). Catalyst loadings of 2–5% gave the best results. Lower catalyst loadings lead to reduced conversion; at lower catalyst loadings, i.e. lower Ni : substrate ratios the equilibrium favours coordination of a second molecule of alkenylimidazolium to the metal centre, shutting down the catalyst. Highly branched rings and larger ring products were also generated by this reaction. A common feature of the fused ring products is the creation of a chiral centre on ring closure, a potentially important feature.

Scheme 10

Structures such as 23 and 27 are oils at room temperature and therefore have potential as IL solvents. Therefore, an asymmetric version of this reaction could provide a catalytic route to chiral ILs. For this reason (S)-QUINAP and (R)-BINAP were tested as ligands, L, in the reaction, eqn (10). However, no annulated products were observed, only alkene isomerisation products were obtained. It appears that the chelating spectator ligands block a coordination site and shut down the final reductive elimination step in the reaction (Scheme 9); the Ni(II) then acts as a standard isomerisation catalyst. Chiral monodentate ligands, which we are currently investigating, will be required for this reaction.

Reaction behaviour of C2-blocked imidazolium salts; the possibility of further ring substitutions

In 2001, Crabtree and co-workers reported the first example of “abnormal” (C4/5) coordination of an NHC, complex 30 [eqn (11), Scheme 11].³⁷ Following that initial study a number of other groups reported the serendipitous C4/5 coordination of NHCs.³⁸ Abnormal coordination and the reaction behaviour of complexes of abnormally bound NHCs have recently been reviewed.³⁹

Scheme 11

The oxidative addition of imidazolium salts to generate “normal” C2-bound carbene complexes had been shown to be a facile process and ensuing studies showed that 2-blocked imidazolium salts will also oxidatively add to Pt(0)⁴⁰ and Ni(0)⁴¹ to form abnormal, C4/5 bound carbene–M–H complexes, Scheme 12. The C4/5-NHC–M–H (M = Ni) thus formed could be expected to also undergo insertion of olefins and subsequent reductive elimination to generate C4/5 substituted imidazolium salts, in much the same way that has been demonstrated for normal C2 bound NHCs (vide supra).

Scheme 12

In preliminary studies styrene and also dimethyl fumarate were reacted with complexes 33/34 to observe whether insertion and possibly reductive elimination occurred.^40,41

Unexpectedly, the reaction that took place was an apparent equilibrium process in which reductive elimination of the hydride and abnormal carbene predominated forming C4/5 hydridoimidazolium salt, 38, and a Pt(0) diolefin complex, 37, eqn (15). There was no evidence of olefin insertion, nor was the normal C2 bound NHC eliminated. Therefore, initial studies indicate that both oxidative addition of 2-blocked imidazolium salt, and reductive elimination of the abnormal NHC does occur, whereas, migratory insertion did not.

These results are entirely consistent with the behaviour of catalysts for 2-substitution of azolium salts (eqn (7), Scheme 9), discussed above. The reaction of azolium salt and olefinic species in the presence of the most active catalyst systems developed to date, results in the formation of 2-substituted azolium salt only (Table 1); further reaction to give C4/5 substituted azolium has not been observed.

Alternative catalytic process for 2-alkylation of N-heterocycles via C–H activation

In an elegant series of studies, Bergman and co-workers have also investigated the catalytic substitution of N-heterocyclic ring systems, via a C–H activation process. The initial reports focused on intramolecular annulation of N-alkenyl-substituted benzimidazoles using [RhCl(COE)₂]₂ + added PCy₃ as the catalyst system (Scheme 13).⁴² These studies were followed by further reports looking at the intermolecular and intramolecular reaction between a wide variety of heterocycles and olefinic substrate to provide a range of 2-functionalised products.^43,44 During the course of these studies, clear evidence was provided to show that Rh–NHC complexes were generated as intermediates in the reaction; similar to observations for the Ni/Pd catalysed reactions, (vide supra). Metal complexes of benzimidazoles are well know and they have been prepared by a variety of methods.⁴⁵

Scheme 13

Despite similarities between the Rh(I) catalysed reaction and our Ni(0)/Pd(0) catalysed reaction, thorough mechanistic studies by the Bergman group (Scheme 14),⁴⁶ in combination with our investigations on the mechanism of the Ni/Pd catalysed process (Scheme 9),^32,33,35 demonstrate the distinct mechanistic differences between the two processes. The former mechanism proposes a 1,2-H shift followed by C–H cleavage and migration of the H to the Rh centre, i.e. a unique C–H activation and intramolecular H-transfer process. The mechanism for the M(0) catalysed reaction involves a primary oxidative addition step to generate an NHC–M–H intermediate into which the olefinic moiety inserts to give NHC–M–alkyl species, which undergoes reductive elimination to generate the product.

Scheme 14

Bergman extended the reaction to include; catalytic arylation of heterocyclic compounds (again highlighting the likely intermediacy of an NHC–Rh complex);⁴⁷Rh(I) catalysed alkylation of quinolines and pyridines;⁴⁸ and enantioselective annulation of an indole, using a catalyst system {[RhCl(COE)₂]₂ plus added phosphoramidite as the chiral ligand}, ultimately generating biologically active dihydropyrroloindoles.⁴⁹ In important, related, studies Bergman and co-workers applied the Rh-catalysed C–H activation and functionalisation of heterocyclic species to the total synthesis of vasicoline⁵⁰ and to the synthesis of kinase inhibitors.⁵¹ In the former reaction 3,4-dihydroquinazolines were both inter- and intra-molecularly coupled with olefinic species (Scheme 15). The studies of Bergman and co-workers beautifully illustrate several key applications for this type of unique chemistry—involving azole substitution and annulation.

Scheme 15

Concluding remarks and outlook

The interplay between N-heterocyclic carbenes and the closely related azolium salts is an interesting one. Their interconversion has been shown to be facile, promoted by a metal-mediated redox process. The consequent application of N-heterocyclic carbenes, and their transition metal complexes, as intermediates in chemical reactions and in catalysis, i.e. more in the manner of “traditional” carbenes, has opened up new opportunities in synthetic chemistry. The use of heterocyclic compounds (and azolium salts) as substrates in catalysis, as demonstrated here, has the possibility of generating new molecules as building blocks for the pharmaceutical industry, and of influencing the study of ionic liquids. The challenge is now to develop new and better catalysts for this reaction, and to investigate the possibility of multiple ring substitutions, thus extending the range of possible products. The chemistry described here has also demonstrated the direct generation of active catalytic species NHC–M–H, formed from the oxidative addition of imidazolium salt to low-valent transition metal centres. When generated in imidazolium-based ionic liquid solvents there is the possibility of producing solvent-stabilised, long-lived active species.

Acknowledgements

Thanks are extended to students and collaborators who have made important contributions to the research described here; their names appear in the references below. In particular I would like to acknowledge Professor Brian Yates whose very important contributions of to the computational studies have led to a long and fruitful association. I also wish to acknowledge funding from the Australian Research Council (ARC), EPSRC and Cardiff University.

References

1. (a) W. A. Herrmann and C. Köcher, Angew. Chem., Int. Ed. Engl., 1997, 36, 2162; (b) D. Bourissou, O. Guerret, F. P. Gabbaï and G. Bertrand, Chem. Rev., 2000, 100, 39; (c) F. K. Zinn, M. S. Viciu and S. P. Nolan, Annu. Rep. Prog. Chem., Sect. B: Org. Chem., 2004, 100, 231; (d) K. J. Cavell and D. S. McGuinness, Coord. Chem. Rev., 2004, 248, 671; (e) E. Peris and R. H. Crabtree, Coord. Chem. Rev., 2004, 248, 2239; (f) C. M. Crudden and D. P. Allen, Coord. Chem. Rev., 2004, 248, 2247; (g) V. César, S. Bellemin-Laponnaz and L. H. Gade, Chem. Soc. Rev., 2004, 33, 619; (h) K. J. Cavell, D. S. McGuinness, Palladium Complexes with Carbonyl, Isocyanide and Carbene Ligands, in Comprehensive Organometallic Chemistry III, ed. R. H. Crabtree, D. M. P. Mingos and A. J. Canty, Elsevier, Oxford, 2007, vol. 8, ch. 4, pp. 197–268; (i) F. E. Hahn and M. C. Jahnke, Angew. Chem., Int. Ed., 2008, 47, 3122; (j) O. Kaufhold and F. E. Hahn, Angew. Chem., Int. Ed., 2008, 47, 4057.
2. R. H. Crabtree (editor), Coord. Chem. Rev., 2007, 251, 595.
3. E. Colacino, J. Martinez and F. Lamaty, Coord. Chem. Rev., 2007, 251, 726.
4. (a) D. J. Cardin, M. J. Doyle and M. F. Lappert, J. Chem. Soc., Chem. Commun., 1972, 927; (b) M. F. Lappert, J. Organomet. Chem., 1984, 264, 217; (c) J. H. Hill and T. A. Nile, J. Organomet. Chem., 1977, 137, 293.
5. (a) A. J. Arduengo III, R. L. Harlow and M. Kline, J. Am. Chem. Soc., 1991, 113, 361; (b) A. J. Arduengo III, H. V. R. Dias, R. L. Harlow and M. Kline, J. Am. Chem. Soc., 1992, 114, 5530.
6. W. A. Herrmann, Angew. Chem., Int. Ed., 2002, 41, 1290.
7. D. S. McGuinness, N. Saendig, B. F. Yates and K. J. Cavell, J. Am. Chem. Soc., 2001, 123, 4029.
8. D. S. McGuinness, W. Mueller, P. Wasserscheid, K. J. Cavell, B. W. Skelton, A. H. White and U. Englert, Organometallics, 2002, 21, 175.
9. M. J. Green, K. J. Cavell, B. W. Skelton and A. H. White, J. Organomet. Chem., 1998, 554, 175.
10. D. S. McGuinness, M. J. Green, K. J. Cavell, B. W. Skelton and A. H. White, J. Organomet. Chem., 1998, 565, 165.
11. A. M. Magill, D. S. McGuinness, K. J. Cavell, G. J. P. Britovsek, V. C. Gibson, A. J. P. White, D. J. Williams, A. H. White and B. W. Skelton, J. Organomet. Chem., 2001, 617–618, 546.
12. D. S. McGuinness, K. J. Cavell, B. W. Skelton and A. H. White, Organometallics, 1999, 18, 1596.
13. S. Caddick, F. G. N. Cloke, P. B. Hitchcock, J. Leonard, A. K. de K. Lewis, D. McKerrecher and L. R. Titcomb, Organometallics, 2002, 21, 4318.
14. D. S. McGuinness and K. J. Cavell, Organometallics, 2000, 19, 4918.
15. D. J. Nielsen, A. M. Magill, B. F. Yates, K. J. Cavell, B. W. Skelton and A. H. White, Chem. Commun., 2002, 2500.
16. D. C. Graham, K. J. Cavell and B. F. Yates, Dalton Trans., 2005, 1093.
17. D. C. Graham, K. J. Cavell and B. F. Yates, Dalton Trans., 2006, 1768.
18. A. M. Magill, B. F. Yates, K. J. Cavell, B. W. Skelton and A. H. White, Dalton Trans., 2007, 3398.
19. D. J. Nielsen, K. J. Cavell, B. W. Skelton and A. H. White, Inorg. Chim. Acta, 2006, 359, 1855.
20. D. J. Nielsen, K. J. Cavell, B. W. Skelton and A. H. White, Inorg. Chim. Acta, 2002, 327, 116.
21. J. C. C. Chen and I. J. B. Lin, J. Chem. Soc., Dalton Trans., 2000, 839.
22. E. Peris, J. A. Loch, J. Mata and R. H. Crabtree, Chem. Commun., 2001, 201.
23. Z. Freixa and P. van Leeuwen, Dalton Trans., 2003, 1890.
24. A. T. Normand and K. J. Cavell, Eur. J. Inorg. Chem., 2008, 18, 2781.
25. (a) P. J. Frazer, W. R. Roper and F. G. A. Stone, J. Organomet. Chem., 1973, 50, C54; (b) P. J. Frazer, W. R. Roper and F. G. A. Stone, J. Chem. Soc., Dalton Trans., 1974, 102.
26. (a) D. S. McGuinness, B. F. Yates and K. J. Cavell, Chem. Commun., 2001, 355; (b) D. S. McGuinness, K. J. Cavell, B. F. Yates, B. W. Skelton and A. H. White, J. Am. Chem. Soc., 2001, 123, 8317.
27. M. A. Duin, N. D. Clement, K. J. Cavell and C. J. Elsevier, Chem. Commun., 2003, 400.
28. N. D. Clement, K. J. Cavell, C. Jones and C. J. Elsevier, Angew. Chem., Int. Ed., 2004, 43, 1277.
29. S. Gründemann, M. Albrecht, A. Kovacevic, J. W. Faller and R. H. Crabtree, J. Chem. Soc., Dalton Trans., 2002, 2163.
30. E. Mas-Marzá, M. Sanaú and E. Peris, Inorg. Chem., 2005, 44, 9961.
31. D. Kremzow, G. Seidel, C. W. Lehmann and A. Furstner, Chem.–Eur. J., 2005, 11, 1833.
32. K. J. Hawkes, D. S. McGuinness, K. J. Cavell and B. F. Yates, Dalton Trans., 2004, 2505.
33. D. C. Graham, K. J. Cavell and B. F. Yates, Dalton Trans., 2007, 4650.
34. N. D. Clement and K. J. Cavell, Angew. Chem., Int. Ed., 2004, 43, 3845.
35. A. T. Normand, K. J. Hawkes, N. D. Clement, K. J. Cavell and B. F. Yates, Organometallics, 2007, 26, 5352.
36. A. T. Norman, S. K. Yen, H. V. Huynh, T. S. Andy Hore and K. J. Cavell, Organometallics, 2008, 27, 3153.
37. (a) S. Gründemann, A. Kovacevic, M. Albrecht, J. W. Faller and R. H. Crabtree, Chem. Commun., 2001, 2274; (b) S. Gründemann, A. Kovacevic, M. Albrecht, J. W. Faller and R. H. Crabtree, J. Am. Chem. Soc., 2002, 124, 1047.
38. (a) H. Lebel, M. K. Janes, A. B. Charette and S. P. Nolan, J. Am. Chem. Soc., 2004, 126, 5046; (b) A. Danopoulos, N. Tsoureas, J. A. Wright and M. E. Light, Organometallics, 2004, 23, 166; (c) N. Stylianides, A. A. Danopoulos and N. Tsoureas, J. Organomet. Chem., 2005, 690, 5948.
39. (a) P. L. Arnold and S. Pearson, Coord. Chem. Rev., 2007, 251, 596; (b) M. Albrecht, Chem. Commun., 2008, 3601; (c) M. Albrecht and K. J. Cavell, Specialist Periodical Reports, Organometallic Chemistry, Royal Society of Chemistry, Cambridge, 2008, vol. 35.
40. D. Bacciu, K. J. Cavell, I. A. Fallis and L.-l. Ooi, Angew. Chem., Int. Ed., 2005, 44, 5282.
41. D. Bacciu, PhD Thesis, Abnormally Bound N-Heterocyclic Carbene Ligands at Group 10 Metal Centres: Synthesis, Structure, Reaction Chemistry and Catalysis, Cardiff University, 2007.
42. (a) K. L. Tan, R. G. Bergman and J. A. Ellman, J. Am. Chem. Soc., 2001, 123, 2685; (b) K. L. Tan, R. G. Bergman and J. A. Ellman, J. Am. Chem. Soc., 2002, 124, 3202.
43. K. L. Tan, R. G. Bergman and J. A. Ellman, J. Am. Chem. Soc., 2002, 124, 13964.
44. (a) K. L. Tan, A. Vasudevan, R. G. Bergman, J. A. Ellman and A. J. Souers, Org. Lett., 2003, 5, 2131; (b) K. L. Tan, S. Park, J. A. Ellman and R. G. Bergman, J. Org. Chem., 2004, 69, 7329; (c) S. H. Wiedemann, R. G. Bergman and J. A. Ellman, Org. Lett., 2004, 6, 1685.
45. (a) M. Tamm and F. E. Hahn, Coord. Chem. Rev., 1999, 182, 175; (b) F. E. Hahn, V. Langenhahn, N. Meier, T. Lügger and W. P. Fehlhammer, Chem.–Eur. J., 2003, 9, 704; (c) F. E. Hahn, V. Langenhahn, T. Lügger, T. Pape and D. L. Van, Angew. Chem., Int. Ed., 2005, 44, 3759.
46. S. H. Wiedemann, J. C. Lewis, J. A. Ellman and R. G. Bergman, J. Am. Chem. Soc., 2006, 128, 2452.
47. (a) J. C. Lewis, S. H. Wiedemann, R. G. Bergman and J. A. Ellman, Org. Lett., 2004, 6, 35; (b) J. C. Lewis, A. M. Berman, R. G. Bergman and J. A. Ellman, J. Am. Chem. Soc., 2008, 130, 2493.
48. J. C. Lewis, R. G. Bergman and J. A. Ellman, J. Am. Chem. Soc., 2007, 129, 5332.
49. R. M. Wilson, R. K. Thaji, R. G. Bergman and J. A. Ellman, Org. Lett., 2006, 8, 1745.
50. S. H. Wiedemann, J. A. Ellman and R. G. Bergman, J. Org. Chem., 2006, 71, 1969.
51. J. C. Rech, M. Yato, D. Duckett, B. Ember, P. V. LoGrasso, R. G. Bergman and J. A. Ellman, J. Am. Chem. Soc., 2007, 129, 490.

---