and Rebecca L.
a School of Chemistry, Cardiff University, Cardiff, Cymru/Wales, CF10 3AT, UK. E-mail: MelenR@cardiff.ac.uk
One field of organometallic chemistry that has seen great advancements over the last 20 years is that of main-group chemistry, in particular boron chemistry, that has led to a wealth of new discoveries. In this review, we will focus on modern advancements in this growing field, such as interesting uses of firmly established reagents, such as tris(pentafluorophenyl)borane, B(C6F5)3, which has demonstrated extensive applications in a wide variety of chemistry. In addition to this, a number of novel Lewis acidic boranes and borocations have been recently synthesised, which are often structurally tailored for a specific role such as borylation reactions or use in main-group catalysis. The reactions of these compounds are broad in scope, inclusive of borylation substitution reactions, addition of B–E across π-bonds and applications in pharmaceuticals and materials science. In addition, boron reagents often constitute the Lewis acid moiety of frustrated Lewis pairs (FLPs), an area of main-group chemistry that has also expanded rapidly, with numerous applications notably in main-group catalysis. Newly discovered Lewis acidic boron reagents and their implementations are proving to be an appealing and exciting applications-based field as more advances are discovered.
Boron reagents are often employed as Lewis acids due to their strongly electrophilic nature granted by a vacant p-orbital into which electrons can be received. Many neutral boranes have been synthesised and utilised, such as trialkyl-, triaryl- and trihalo-boranes, although as the field of boron chemistry has grown, more complex and structurally diverse boron reagents have been reported. One of the key features of neutral borane species is that the Lewis acidity can be attenuated by variation of the three substituents bound to boron. An example of this is tris(pentafluorophenyl)borane [B(C6F5)3], a powerful Lewis acid due to the electron withdrawing effects of the three perfluorinated aryl rings, which was first synthesised in the 1960s.1 Since this discovery, other strongly Lewis acidic boranes have been reported, select examples of which are included herein.
A useful tool when considering Lewis acidic boranes is the ability to determine experimentally their Lewis acidity, allowing them to be placed on a scale, such as in Fig. 1.2 The most well-known procedures for this use NMR spectroscopic analysis. The Gutmann–Beckett method involves the coordination of triethylphosphine oxide (Et3PO) to a Lewis acid and recording the chemical shift in the 31P NMR spectrum.3 The Lewis basic oxygen atom of Et3PO can form an adduct with boron reagents, causing deshielding of the adjacent phosphorus atom, the degree of which can be measured to ascertain the Lewis acidity of the boron reagent. Childs method instead uses crotonaldehyde and 1H NMR spectroscopy.4
The chemistry of boranes is well documented, with numerous examples of borylation reactions which involve the formation of new B–E bonds with a π-nucleophile, such as arenes, alkenes and alkynes.5 The first reported examples of borylation reactions were dependent upon transition-metal catalysts, which can be problematic due to potentially high catalyst cost and more difficult purification of products.6 More recently, new methodologies have sought to increase the reactivity of boranes, in order to avoid the necessity of metal catalysts. One potential answer to this has been the study of borocations,7 which have seen recent advancements in synthesis and applications. These compounds have the capacity to have very high Lewis acidity, due to the cationic nature of the boron centre.
Boranes such as B(C6F5)3 are commonly used as the Lewis acid component of frustrated Lewis pairs (FLPs), with a few recent exceptions involving silylium,8 phosphonium,9 aluminium10 and carbon11 Lewis acids reported. FLPs are composed of sterically encumbered Lewis acids and bases that, due to the high levels of steric obstruction, are unable to form classical adducts, but have been observed to undergo unique reactivity with various other reagents. For example, the steric bulk and electron-withdrawing nature of the three C6F5 groups of B(C6F5)3 are the reason it is a commonly used Lewis acid in FLP chemistry.12 Applications of FLPs include small molecule activation, such as H2, as shown in Scheme 1, along with many others.13
This review aims to cover recent reports of emergent boron chemistry. This will begin by examining new methodologies for the synthesis of novel Lewis acidic boranes and borocations, to demonstrate how this field has evolved with these developments. This review will then demonstrate the utilisation of boron reagents by discussing a series of reactions that use a variety of boron-based compounds. These range in scope from relatively simple borylation reactions, including additions and substitutions, to more intricate methodologies, such as boron induced cyclisations, to create a range of complex cyclic products. Modern developments in FLP chemistry and boron-based catalysts, will also be probed, covering advancements in catalytic chemistry. The scope of this review will focus on the most recent examples from the last 5–10 years, although some historical examples are included for context.
The development of novel boranes is an area that has shown considerable growth, often with an applications-based methodology. Many techniques are focused on increasing the Lewis acidity of the boron centre, historically achieved with fluorinated boranes such as the aforementioned B(C6F5)3. The borane tris[3,5-bis(trifluoromethyl)phenyl]borane (BArF3) (1, Fig. 2) was synthesised and found to be a more powerful Lewis acid than B(C6F5)3via the Gutmann–Beckett method.14 Studies also indicated that BArF3 can form FLPs with select Lewis bases, and is capable of activating molecular hydrogen.
A different approach was to synthesise boranes featuring cationic substituents, which can cause a strong negative inductive effect, thus increasing the Lewis acidity of the boron centre as a result.15 A recent example of this involved the synthesis of cationic analogues of trimesitylborane (2–4, Fig. 3).16 These boranes were air and moisture stable, postulated to be a result of the steric protection of the boron centre afforded by six ortho-methyl groups. Cyclic voltammetry was used to measure the reduction potential of these boranes, where it was found that there was almost a linear trend in their reduction potential, as determined by the number of ammonium substituents present. This strategy provides an approach that allows control of the redox properties of the boranes.
Wagner et al. synthesised a trio of novel mono-haloboranes featuring 3,5-bis(trifluoromethyl)phenyl-groups.17 These electron-withdrawing groups were designed to prevent steric crowding at the boron centre by forgoing ortho-substituents on the phenyl rings, potentially increasing achievable reactivity. Beginning with (3,5-(CF3)2C6H3)Li and BH3·SMe2, sequential reactions afford 5, the precursor to the desired haloboranes. Reactions of 5 with KHF2/Me3SiCl, BCl3, and BBr3, generate fluoro, chloro, and bromo haloboranes 6, respectively. Each of these species was isolated and fully characterised (Scheme 2).
Wildgoose et al. have designed a novel triaryl borane (7, Fig. 4), the first structurally characterised 1:1:1 hetero-tri(aryl)borane to be reported.18 This compound acts as the Lewis acid component in FLPs to cleave H2 heterolytically, and has been demonstrated to be compatible with a number of Lewis bases, namely P(tBu)3, 2,6-lutidine and 2,2,6,6-tetramethylpiperidine (TMP). It was observed that the degree of conversion to the cleaved product over time was dependant on the Lewis base, with P(tBu)3, and 2,6-lutidine providing higher conversions than TMP. This synthetic methodology, namely step-wise addition of groups to a borane, represents a remarkably useful way to access this family of boranes, allowing greater modification of future triarylboranes in order to suit the desired reactivity.
In addition to neutral boranes, considerable interest has been directed at borocations, compounds that were defined over 30 years ago, but more recently have developed into an interesting new field of boron chemistry.7,19 These compounds often possess high Lewis acidity due to the formal positive charge on the boron centre.20 Borocations are often defined and characterised by their coordination number, which corresponds to a general trend in relative Lewis acidity, as shown in Fig. 5. The 2-coordinate boriniums are generally the most reactive, but are often highly unstable precluding widespread application. The 4-coordinate boroniums, on the other hand, exhibit high levels of structural stability, but suffer poor reactivity due to the fully occupied coordination sphere at boron. Indeed, it is often the 3-coordinate borenium cations that are most reported in the literature, as they offer a compromise between the other borocations; they possess the added stability of an L-type donor ligand which allows much easier manipulation (under an inert atmosphere) whilst reactivity is enhanced due to the unsaturated coordination sphere.
The synthesis of borenium cations typically follows the same general pathway.19 Beginning with a Lewis basic neutral borane, a donor ligand is coordinated, often an amine, to generate an adduct. This is followed by either halide or hydride abstraction from the boron adduct, forming the positively charged borocation. In order to facilitate borenium formation, the donor ligand must bind strongly enough to boron that it does not dissociate during the halide/hydride abstraction step. Additionally, the B–X bond (X=hydride or halide) should be weakened by ligand binding so as to induce the abstraction. This approach allows for great diversity in the structure of borocations, and hence their reactivity through variation of the substituents (R-groups) at boron and the donor ligand. For example, electron-donating groups can be used to stabilise borenium cations, whereas the Lewis acidity can be increased with electron-withdrawing groups. More recently, borocations that feature N-heterocyclic carbenes (NHCs) have been synthesised.21 One such example featured an NHC-stabilised borenium 8 which could be isolated and structurally characterised (Scheme 3).
Following this, Robinson et al. synthesised a novel borenium stabilised by an N-heterocyclic olefin (NHO).22 The NHO was reacted with the strong Lewis acid BBr3, forming an isolable adduct 9. It was observed that in the presence of THF, 9 is capable of cleaving the C–O bond of the solvent, ring-opening two equivalents whilst simultaneously delivering a bromine atom, generating borenium species 10 with Br− as the counter ion (Scheme 4).
Melen et al. recently reported a new methodology for synthesising borenium compounds, using diimines and dichlorophenylborane.23 When select diaryldiimines were reacted with dichlorophenylborane (PhBCl2), it was observed that 11 was formed, with the by-product HCl being formed also. Subsequent addition of aluminium trichloride abstracted the chloride from HCl causing addition of the proton to compound 11 forming the borenium compound 12 (Scheme 5). Through attempts with a range of diamine precursors, it was found that the steric properties of the diimine were important to achieve borenium ion formation, as an absence of substitution in the ortho-position of the aryl rings resulted in more complex, unclean reactivity. Conversely, too much steric bulk in this position precluded reaction all together. This work represents a new methodology for the synthesis of borocations.
A large proportion of novel boranes and borocations are utilised in borylation reactions, such as dehydroborylation, hydroboration, carboboration and haloboration. Due to the different requirements of each of these reactions (most notably the nature of the groups bound to the boron reagent), as well as the variety of substrates, the ability to functionalise the structure and attenuate the reactivity of the boron reagent is of vital importance. The products of these reactions can be thought of as intermediates en route to more complex, and more valuable products, as the addition of a boron species into the molecular structure allows greater functionalisation through subsequent cross-coupling reactions.
The dehydroborylation reaction allows for the direct insertion of boron into a molecule by transforming a C–H bond into a C–B bond. Historically, Brown et al. reported that trialkoxyboranes were suitable borylation reagents when combined with lithiated organic species, including alkynes, generating alkynyl boronates 13, as seen in Scheme 6.24 The lithiated reagents were synthesised from alkyne precursors, hence requiring a multi-step reaction to acquire the borylated products.
Modern approaches avoid the necessity of metallation by using (commonly) an amine base to deprotonate the substrate, which can either be added separately or, as has been reported with certain borocations, incorporated into the structure of the reagent. Recent advances have shown that certain borenium species undergo selective dehydroborylation of arenes and heteroarenes. Ingleson et al. have probed this area considerably,25,26 and have shown that highly reactive dihalo-boreniums of the general formula [Cl2B(L)][AlCl4] trigger dehydroborylation of a range of arenes, as shown in Scheme 7, such as N-TIPS-pyrrole 14.27,28 The enhanced electrophilicity of the borenium cations was generated by the use of two covalently bound halide atoms, which allowed the reactions to proceed quickly at ambient temperature. In addition, modulation of the Lewis base ligand allowed access to several diborylated heteroaryl species 15 (Scheme 8). Isolation of these compounds as air-stable pinacol boronate esters increases the usefulness of the products, as they can then be employed as critical scaffolds to more complex molecules via cross-coupling reactions, allowing the products to be used as synthetic ‘building blocks’ towards desirable synthetic targets.
Recently, Fontaine et al. demonstrated a metal-free catalytic approach to arene borylation, utilising borane (1-TMP-2-BH2-C6H4)2 (TMP=2,2,6,6-tetramethylpiperidine) 16 as a main-group catalyst.29 As depicted in Scheme 9, borane 16 initially reacts with the arene, in this case N-methyl pyrrole, activating the C–H bond in the 2-position, generating H2. The intermediate is subsequently reacted with H–BPin, causing catalyst regeneration and producing the desired borylated arene as the pinacol boronate ester 17. The substrate scope was expanded beyond N-substituted pyrroles, to include indoles, furans and electron rich thiophenes, often producing yields in excess of 80%. A catalyst loading of 2.5 mol% of 16 achieved optimal results.
The dehydroborylation of arenes, heteroarenes and alkenes was reported by Repo et al. who demonstrated that 2-aminophenylboranes can be used to generate the borylated products (18, Scheme 10).30 This reaction was reported to proceed via a C–H insertion in an FLP-type fashion, wherein the boron and amine heterolytically split the C–H bond. This reactivity is promoted by the close structural proximity of the Lewis acid and base moieties, resulting in a relatively low kinetic barrier to reaction (ΔG=21.0 kcal mol−1 for thiophene).
In addition to substitution reactions such as dehydroborylations, boranes and borocations can be used in addition reactions, such as hydro-, carbo-, and halo-boration. These elementoborations feature the addition of boron and another group to a π-system. Discoveries of 1,n-addition (n=1, 2 or 3) have been reported of boron reagents to alkynes,31 allowing the incorporation of a boron unit whilst generating a vinyl species. Likewise, vinyl species themselves can be targeted, in order to generate borylated alkyl-compounds.32 Again, many historical examples require a metal catalyst.33 More recent examples have harnessed main-group reagents such as borocations to access these borylated alkenes in a metal-free fashion. These products are desirable as they have been shown to be useful building blocks to more complex compounds, often utilising cross-coupling reactions through the boron unit.34
Lappert et al. were amongst the first to explore haloboration reactions, wherein boron and a halogen are added to a π-nucleophile, such as an alkyne, generating a vinylboron species.35 Although the resulting products are similar to those of carboboration reactions, which are discussed later, the addition of a halogen provides an additional site of functionalisation via cross-coupling reactions such as the Suzuki reaction.36 However, until recently this work was limited to using haloboranes, such as BCl3, and terminal alkynes.35,37
Recent advancements have shown that borocations featuring halides are able to exceed the reactivity observed with trihaloboranes.38 The 1,2-haloboration of a range of internal alkynes was achieved by using the borenium [Cl2B(lut)][AlCl4] (lut=2,6-lutidine) 19, allowing access to several new vinyl boron species which were isolated as pinacol boronate esters. This borenium compound was shown to be compatible with a range of internal alkynes, featuring a variety of substituents such as aliphatic and/or aromatic moieties, thiophenes and anisoles (Scheme 11). When enynes were employed as targets, haloboration occurred exclusively at the alkyne site, leaving the alkene functionality untouched. These products were isolated as single regio- and stereo-isomers, showing the proclivity of this reaction to create highly functionalisable substituted alkenes. The utility of these products was demonstrated by sequential cross-coupling reactions of the vinyl boron and chloro-substituents, with subsequent Suzuki cross-coupling reactions. The highly functionalised alkene 20 shown in Scheme 12 was the end result of this.14 This methodology was further expanded by demonstrating that the reaction was possible in a one-pot fashion, from the alkyne to the fully substituted alkene. These tetra-substituted alkenes were shown to be structural analogues and precursors to several drug molecules, such as Tamoxifen, a powerful anti-cancer agent, and Zuclomiphene, a selective oestrogen receptor modulator, further demonstrating the utility of this reaction pathway.39
The carboboration reaction allows for the facile formation of both new C–B and C–C bonds across a π-system.40 A variety of trialkyl boranes have traditionally been used for carboboration reactions with certain activated alkynes, as demonstrated by Wrackmeyer.41 Modern advances in this field have shown that a range of heteroleptic boranes and borocations can be synthesised and applied in various carboboration reactions. Ingleson et al. reported that neutral borane species of the general formula (Cl2B–aryl), as well as borocations derived from RBCl2 boranes [RBCl(2-DMAP)][AlCl4], can be utilised for the 1,1-carboboration of trimethylsilyl (TMS) substituted alkynes.42 Commercially available PhBCl2 was combined with a variety of TMS-alkynes, producing vinylborane products with excellent stereo- and regio-selectivity which were isolated as the pinacol boronates 21 (Scheme 13). In addition to PhBCl2, other dichloroboranes were synthesised featuring p-chlorobenzene, triphenylamine, thiophene and furan. Whilst borocations 22 did not surpass the effectiveness of the boranes, they did allow access to an alternate reaction pathway to 2-bora-1,3-dienes 23 (Scheme 14). These species have been synthesised with either a phenyl group or a 2-methyl thiophene in the 3-position of the diene.
The 1,2-carboboration reaction is far rarer, with some examples of metal-catalysed synthesis reported.43 However, in the last few years examples have been reported that use novel borocations or boranes in the absence of a metal. Ingleson et al. synthesises quinolato-boreniums 24 which undergo 1,2-carboboration with 3-hexyne, with the ability to transfer either a phenyl or thiophenyl group. Subsequent esterification with pinacol allowed access to vinyl boronates (25, Scheme 15).44 In addition to this, Bourissou et al. utilised a phosphorus-stabilised borenium (26) for 1,2-carboboration, which reacted readily with 3-hexyne (Scheme 16).45 They showed that a mesityl group could be transferred selectively, generating the vinyl boron species 27. The reagent represented another novel borenium species, which uses the naphthalene scaffold to create a strong intramolecular B–P interaction, hence demonstrating the rich structural variety this family of molecules can possess.
Another metal-free methodology for 1,2-carboboration was reported by Melen and co-workers utilising allenes and B(C6F5)3. The 1,2-carboboration of allenyl ketones afforded the selective formation of α,β-unsaturated ketones (28, Scheme 17). The C6F5 group was transferred to the β-position, and the alkene itself was exclusively of the E-configuration.
Other 1,n-carboboration reactions have also been reported. The 1,1-carboboration of propargyl esters can undergo a 1,3-allyl shift, resulting in formal 1,3-carboboration products.46 In addition, the reaction between B(C6F5)3 and 1,6-enynes generate novel borylated cyclic species that are the result of a net 1,4-carboboration reaction.47
In addition to the aforementioned borylation reactions, main-group boron reagents are useful in a number of more complex reactions. These often involve the boron compound inducing further reactivity within the structure of the substrate, causing new intramolecular bonds to form and cyclisation reactions to occur. Manipulation of the structure of the boron reagent can again be implemented to achieve the desired reactivity, showing the potential broad utility of these molecules. In addition, examples of FLP-mediated reactions have also been reported, with select applications featured herein.
Work in this field has introduced a variety of novel cyclisation mechanisms, most notably with diyne precursors.48 One such example used an FLP approach to induce cyclisation, with B(C6F5)3 as the Lewis acid and P(o-tol)3 as the base (Scheme 18). The observed reactivity and subsequent structure of the product was found to vary depending on the carbon–carbon chain length between the two alkyne moieties, as shown in Scheme 18.49 For 1,6-heptadiyne (n=1), the reaction proceeded as a 1,1-carboboration of the first alkyne group, followed by 1,2-FLP addition to the other, generating a eight-membered heterocyclic zwitterionic phosphonium borate 29. With 1,7-octadiyne (n=2), both boron and phosphorus groups are added at the terminal end of the alkynes, inducing cyclisation to produce a six-membered cyclic compound 30.
More complex diynes featuring heteroatoms, such as silicon, can be used to generate heterocycles, (i.e. siloles) via the use of B(C6F5)3.50 In these reactions, 1,1-carboboration of the first alkyne is followed by 1,1-vinylboration of the second, generating the heterocyclic product, in this example a silole 31, shown in Scheme 19.51 This report stands alongside similar reactivity observed with other group 14 elements, with reports of the synthesis of germanium-, tin- and lead-containing heterocycles synthesised with trialkyl boranes.52 As heterocyclic compounds have wide applications throughout chemistry, this methodology that produces boron-substituted heterocyclic compounds stands out as a useful synthetic tool.
A similar reaction can be used to access phospholes, for which few synthetic routes have been reported.53 These five-membered heterocycles have applications in material science,54 particularly when boron-based acceptor substituents are in conjugation with the phosphorus donor.55 Hence, the 3-boryl substituted phospholes 33 shown in Scheme 20 are of significant interest for the aforementioned reasons. They are synthesised from the respective bis(alkynyl)phosphine 32 in combination with B(C6F5)3, resulting in a 1,1-carboboration reaction sequence generating the desired product.56
Stephan et al. has demonstrated the application of boranes in the synthesis of tellurium containing heterocycles from readily synthesised alkynyl-telluride reagents.57 These products are desirable as both catalysts58 and for their optoelectronic properties.59 Tellurium acetylides were the target of 1,1-carboboration reactions with B(C6F5)3 and other boranes, generating boron-containing vinylic telluroethers 34. Subsequent addition of phenylacetylene generated the zwitterionic heterocyclic products 35, seen in Scheme 21, and this vinyl species can be thought of as an intramolecular FLP.
In addition to this, it was reported that exposing a diethynyl-tellurium compound to various boranes is a viable pathway to similar heterocyclic products.60 The reaction between the tellurium diyne and the borane begins with a 1,1-carboboration, forming 36, which can then undergo either a second intramolecular 1,1-carboboration forming 37, or alternatively proceed via an intermolecular FLP addition to generate (38, Scheme 22). The products of the former reaction 37 are currently being investigated as novel tellurium heterocycles containing an electrophilic unit within the ring, giving rise to attractive future applications in optoelectronic materials.
Similar work involving bis(dialkynyl)sulfides has been reported, and their reactions with borane reagents documented.61 When bis(phenylethynyl)sulfide 39 and B(C6F5)3 are combined, a 2 : 1 molar ratio reaction occurs to generate the benzothiophene derivative 40. Conversely, when bis(tbutylethynyl)sulphide 41 is used, boranes of the formula R–B(C6F5)2 trigger cyclisation to form the five-membered thiophenes (42, Scheme 23). This reactivity is in contrast to that seen with the tellurium diynes, as under these conditions no boron insertion into the ring is observed.
Although the strong Lewis acid BCl3 has been used in a myriad of reactions, modern examples of its utility continue to emerge. For example, BCl3 can be applied in borylative cyclisations of a range of 4-aryl-butynes, producing 1-aryl-2-boro-3,4-dihydronapthalenes 43, as shown in Scheme 24.62 Although only select examples are shown here, more than 20 products are reported, showing the versatility of this reaction. By using a relatively inexpensive reagent in the form of BCl3 and avoiding the use of transition metals, this work has shown a readily accessible pathway to synthetically useful C(sp2)-boronate esters. In order to demonstrate the utility of these products, two additional reactions were undertaken, as shown in Scheme 25. The first of these confirmed that cross-coupling with 4-bromotoluene proceeds readily to give 44 in good yields (75%). In addition, the borylated products can undergo dehydrogenation to give the borylated naphthalene analogue 45. The ability to create highly functionalised molecules from commercially available starting materials is of great importance in synthetic chemistry, with the products synthesised here posing as potential intermediates in the synthesis of pharmaceuticals such as Nafoxidine.63
Borylated lactones can be readily synthesised from alkynyl ester starting materials using B-chlorocatecholborane (CatBCl).64 It was shown that trans-oxyboration of the alkyne moiety occurs in the presence of CatBCl in the absence of a catalyst, only elevated temperatures were required (100 °C). From readily available methyl esters, the borylated isocoumarin and 2-pyrone products were isolated as pinacol boronate esters 46. The substrate scope for this reaction was expanded to include functional groups incompatible with previous borylation techniques,65 such as esters, cyanides, aryl halides and thiophenes, providing a diverse range of isolable products (Scheme 26). Similar reactions using B(C6F5)3 with enynoate precursors generated a number of pyrylium borates 47via a 6-endo-dig cyclisation, as shown in Scheme 26.66
A novel boron induced cyclisation reaction was reported wherein 2-alkynylthioanisoles are combined with CatBCl, forming a series of borylated benzothiophenes (48, Scheme 27).67 The proposed mechanistic pathway involves the borane acting as an electrophilic Lewis acid by activating the alkyne bond whilst the sulfur undergoes nucleophilic attack. This generates a sulfonium intermediate, whilst subsequent loss of methylchloride forms the borylated product. A similar reaction was reported by Ingleson et al. using BCl3 and 2-alkynyl-anisoles to produce borylated benzofurans 49 as pinacol boronate esters (Scheme 28).68 The products were then used in cross-coupling reactions to demonstrate the utility of this reaction pathway, generating synthetically useful 2,3-disubstituted heteroarenes.
One area of main group chemistry that has flourished in recent years is that of catalysis, with reactions featuring boranes, borocations and FLPs gaining traction in the literature. These reagents have led to the advancement of carbon–carbon bond forming reactions, hydrogenation and hydroboration, which are traditionally metal-dominated pathways. The key advantages often cited of main-group catalysis is the generally lower cost of materials, making these developments more attractive to industry, and the lower toxicity of these compounds. The latter precludes the need for rigorously stringent purification of products, due to the absence of trace heavy-metals.
It was found that 1,5-enynes, when exposed to catalytic amounts of B(C6F5)3, are structurally suited to undergo cyclisation reactions to produce functionalised indenes.69 The reaction was reported to proceed via π-activation of the alkyne moiety by B(C6F5)3 followed by 5-endo-dig cyclisation with the alkene, generating (50) (Scheme 29). Triphenylphosphine was required to regenerate the boron catalyst, as the protodeboronation step is key to achieving a catalytic turnover, hence the reaction is an FLP-type system. The observed reactivity was analogous to that reported with transition metals,70 further alluding to the ubiquitous nature of main-group catalysts mimicking the reactivity of their more expensive counterparts.
Propargyl amides can be readily cyclised when exposed to B(C6F5)3, which was shown to generate a number of differing cyclic products, such as 5-alkylidene-4,5-dihydrooxazolium borate species (51, Scheme 30).71 Of particular interest is that when N–H substrates are targeted, protodeboronation of species (51) can occur to give the oxazole derivative following isomerisation (52) and regenerating the borane. Hence this methodology represents a catalytic pathway to these species, which are useful synthetic building blocks as numerous natural products, drugs and pharmaceuticals contain this heterocycle.72
Wagner et al. have developed cyclic phosphonium bis(fluoroaryl)boranes 53, which were successfully employed as catalysts for the [4+2] cycloaddition reaction between 2,5-dimethyl-1,4-benzoquinone and cyclopentadiene (Scheme 31).73 Analysis of 53via the Gutmann–Beckett method revealed that these compounds are highly Lewis acidic.
Although the hydroboration reaction has been systematically reported, many historic and indeed recent examples rely on metal catalysts,74 whereas main group alternatives have been presented recently that bypass the necessity of removing trace metal impurities. In order to increase atom efficiency, modern hydroboration reactions often employ hydroboranes such as catechol borane (CatBH) or pinacol borane (HBPin), precluding a second reaction to esterify the products. In addition, the vast majority of reported reactions are cis-1,2-addition and are usually selective, due to syn-addition across the π-bond.
The hydroboration of alkenes was found to be successfully catalysed by BArF3 (1, Fig. 2) as opposed to B(C6F5)3 which was found to be unsuitable.75 The reaction proved to be selective, generating several cis-1,2-hydroboration products (54, Scheme 32). Substrate scope expansion showed that this reaction has wide utility, with a range of substituted styrenes and aliphatic alkenes successfully reacted, often with excellent anti-Markovnikov regioselectivity being observed.
Alkynes also represent a viable target for hydroboration reactions, and it was found that Piers’ borane [HB(C6F5)2] was well suited to act as a catalyst.76 The products of this hydroboration were (E)-vinyl pinacol boronate esters 55, which formed with excellent regioselectivity. In addition, this system was shown to be compatible with a range of terminal and internal alkynes, although the latter required elevated temperatures to achieve full conversion (Scheme 33). Selectivity is somewhat diminished when unsymmetrically substituted alkynes were targeted, with a mixture of Markovnikov and anti-Markovnikov products being formed, although the former are more prevalent.
Although the cis-hydroboration of alkynes is well documented, the trans-analogue of this reaction is noticeably under reported. Ingleson et al. recently reported the main-group catalysed trans-hydroboration of alkynes using B(C6F5)3 and borenium cations.77 The borenium compound used was designed for this particular reaction, and during development it was observed that both NHCs and chelating dialkyl substituents were necessary for the catalyst to function as desired, as without these features the molecule was prone to undergo dehydroboryation reactions. A non-basic anion was also required to prevent this. A catalytic amount of B(C6F5)3 is required to activate the hydroborane precursor to form the borenium, which subsequently generates the Z-alkene products (56, Scheme 34). NMR spectroscopy was used to confirm the product, whilst deuterium labelling experiments established that this was a formal 1,2-trans-hydroboration 57.
Beyond the hydroboration of C–C multiple bonds, Fontaine et al. have developed a system for the hydroboration of CO2 using a phosphinoborane intramolecular FLP.78 The catalyst used was again custom designed, in this case a highly ambiphilic 1-catecholbora-2-diphenylphosphabenzene (58, Scheme 35), which was found to be involved in each step of the reduction of CO2 to methoxyboranes 59. This advancement in main-group catalysis was proposed as a way to turn a harmful greenhouse gas into a useful fossil fuel substitute.
The hydrogenation of a wide range of substrates using molecular hydrogen has been rigorously reported, although until recently the vast majority of examples required metal-catalysis.79 It is another area that main-group catalysis has expanded to, notably using boron in FLP catalysis.12
A metal-free pathway using FLPs to activate H2, and subsequently for the hydrogenation of activated alkenes, was discovered by Stephan et al.80 This work was the culmination of the discovery of low-temperature H2 activation by an FLP composed of B(C6F5)3 and phosphine Lewis bases. NMR spectroscopy showed no difference when the FLP was charged with 5 bar H2, however upon gradual cooling to −80 °C the phosphonium-borate salt was observed. It was concluded that H2 activation was reversible at low temperatures, providing a methodology for reversible metal-free hydrogen activation with a low energy barrier. Hence, this FLP system was used to hydrogenate successfully a range of alkenes, as it is reasoned that in the presence of a substrate, the transient H2 activation product is intercepted by the alkene, generating the hydrogenated species (60, Scheme 36).
The FLP-catalysed hydrogenation of electron-poor alkenes was achieved by finely tuning the boron Lewis acid.81 Previous studies of the catalytic reduction of alkylidene malonate suggest that activation of the substrate is necessary to carry out the desired reduction, therefore hydride transfer from the borohydride moiety is most likely the rate-determining step. Hence it was postulated that boranes with weaker Lewis acidity than B(C6F5)3 would be favourable. However, weaker Lewis acids might not be able to induce H2 cleavage, which is necessary for the reduction to take place. Therefore range of fluoronated triphenylboranes were probed using Childs method, in order to find a borane that could activate the substrate (alkylidene malonate), yet was still able to cleave H2. It was shown that an FLP consisting of tris(2,4,6-trifluorophenyl)borane 61 and DABCO provided optimal reactivity with a range of alkylidene malonates, and was extended to several examples of electron-poor alkenes and nitroalkenes (Scheme 37).
Whilst great strides have been made in the field of FLP-catalysed reactions, such as hydrogenation, almost all reported systems exhibit sensitivity to moisture, and hence require anhydrous conditions. This presents a potential roadblock which hinders widespread uptake of this methodology. Recently, Ashley and Wildgoose et al. have demonstrated that 1,4-dioxane solutions of B(C6F5)3 can be used in FLP catalysed hydrogenations of weakly basic carbonyl and olefin substrates, without the need for inert conditions.82 It was initially shown that the H2O-adduct of B(C6F5)3 can successfully catalyse the hydrogenation of acetone, showing that catalyst inhibition by H2O is reversible, although it is still a significant catalyst poison, as the rate of reaction was reduced compared to anhydrous reactions. This was overcome with increased H2 pressure, with an increased substrate scope (Scheme 38). These reactions were all performed in undried ‘bench’ solvents, without inert atmosphere techniques, potentially opening up this process to commercial or industrial scale, by lowering the costs of the materials involved.
In addition, Soós et al. reported another instance of moisture tolerant FLP catalysis.83 In this example, they developed a novel borane to act as the Lewis acid in their FLP system, tailoring the structural and electronic properties to achieve the desired stability and reactivity. Starting from B(C6F5)3, two key strategies were employed in developing a new borane; a size exclusion approach, which used enhanced steric bulk to inhibit binding to the Lewis acid centre, and the reduction of the electron-deficiency at the boron centre. With these two factors in mind, the novel borane (62) was readily synthesised and paired with ethereal solvents to generate moisture tolerant FLPs (Scheme 39). It was found that a range of carbonyl substrates could be successfully hydrogenated, showing again that main-group FLP systems have the capacity to be used without stringent anhydrous conditions.
Throughout the field of organometallic chemistry, main-group elements including Lewis acidic boranes are increasingly being utilised in reactions which are traditionally dominated by d-block compounds. The benefits of this are often proposed as the potentially reduced cost of materials in comparison to some expensive transition metals, and the lower toxicity of the compounds. Using these reagents to introduce a boron-group into a molecule creates a molecular scaffold, immediately suitable for further cross-coupling reactions. The expansion of this field of chemistry has resulted in new synthetic pathways to novel, useful main-group compounds, such as boranes and borocations, whilst their applications have been studied in depth. This extends to FLPs, which have risen to prominence over the last 20 years, and become synonymous with small molecule activation and further catalytic reactions. More recently, work has been conducted to overcome some of the limitations of this field, and as this area of chemistry continues to evolve, the scientific community will surely follow these developments with great interest.
© The Royal Society of Chemistry 2017 (2017)