Frustrated Lewis pair

The discovery that some FLPs are able to split, and therefore activate, H₂[4] triggered a rapid growth of research into this area. The activation and therefore use of H₂ is important for many chemical and biological transformations. Using FLPs to liberate H₂ is metal-free, this is beneficial due to the cost and limited supply of some transition metals commonly used to activate H₂ (Ni, Pd, Pt).[6] FLP systems are reactive toward substrates that can undergo heterolysis (e.g hydrogen) due to the "unquenched" reactivity of such systems. For example, it has been previously shown that a mixture of tricyclohexylphosphine (PCy₃) and tris(pentafluorophenyl)borane reacts with H₂ to give the respective phosphonium and borate ions:

${\displaystyle {\ce {PCy3 + B(C6F5)3 + H2 -> [HPCy3]+ [HB(C6F5)3]-}}}$

In this reaction, PCy₃ (the Lewis Base) and B(C₆F₅)₃ (the Lewis Acid) cannot form an adduct due to the steric hindrance from the bulky cyclohexyl and pentafluorophenyl groups. The proton on the phosphorus and hydride from the borate are now ‘activated’ and can subsequently be ‘delivered’ to an organic substrate, resulting in hydrogenation.

The mechanism for the activation of H₂ by FLPs has been discussed for both the intermolecular and intramolecular cases. Intermolecular FLPs are where the Lewis Base is a separate molecule to the Lewis Acid, it is thought that these individual molecules interact through secondary London dispersion interactions to bring the Lewis Base and Acid together (a pre-organisational effect) where small molecules may then interact with the FLPs. The experimental evidence for this type of interaction at the molecular level is unclear. However, there is supporting evidence for this type of interaction based on computational DFT (Density Functional Theory) studies. Intramolecular FLPs are where the Lewis Acid and Lewis Base are combined in one molecule by a covalent linker. Despite the improved ‘pre-organisational effects’, rigid intramolecular FLP frameworks are thought to have a reduced reactivity to small molecules due to a reduction in flexibility.

The activation of H₂ using FLPs was first reported in 2006.[4] In this report a phosphonium-borate species was observed to undergo the thermal loss of a H₂ molecule, to generate the phosphine and the borane. The phosphonium-borate species 1 was prepared from B(C₆F₅)₃ and (C₆H₂Me₃)₂PH, due to the steric demands of both of these molecules a traditional adduct could not be formed and the zwitterionic phosphonic-boronate was prepared [Figure 1]. A “H-for-F exchange” was carried out to give 2 which was stable to moisture and air. This salt will release molecular H₂ when heated above 100 °C, a colour change was also observed from the colourless 2 to the orange-red 3 (both in THF). This reaction is reversed by reacting 3 with H₂ at 25 ℃ where the colour change is reversed and 2 is reformed. The reaction of H₂ with 3 was successful for temperatures as low as 25 ℃. Recrystallisation of 3 from THF afforded colourless crystals of the adduct 4.

Synthetic methodology for the first FLP, (1) reported by Stephan et. al. formed from dimesityl phosphine and tris(pentafluorophenyl)borane. After reaction with Me₂SiHCl to give (C₆H₂Me₃)₂PH(C₆F₄)BH(C₆F₅)₂ (2), the resulting dihydride will thermally decompose to release hydrogen gas and the neutral compound 3. Using THF as a donor solvent afforded white colourless crystals of the adduct molecule 4.

FLPs are also reactive toward many unsaturated substrates beyond H₂. Some FLPs react with CO₂, specifically in the deoxygenative reduction of CO₂ to methane.[7]

Ethylene also reacts with FLPs:[8]

${\displaystyle {\ce {PCy3 + B(C6F5)3 + C2H4 -> Cy3P+CH2CH2B^{-}(C6F5)3}}}$

For acid-base pairs to behave both nucleophilically and electrophilically at the same time offers a method for the ring-opening of cyclic ethers such as THF, 2,5-dihydrofuran, coumaran, and dioxane.[9]

Catalytic cycle for reduction of imine to an amine using an FLP

Reduction of imines, nitriles, and aziridines to primary and secondary amines traditionally is effected by metal hydride reagents, e.g. lithium aluminium hydride and sodium cyanoborohydride. Hydrogenations of these unsaturated substrates can be effected by metal-catalyzed reactions. Metal-free catalytic hydrogenation was carried out using the phosphonium borate catalyst (R₂PH)(C₆F₄)BH(C₆F₅)₂ (R = 2,4,6-Me₃C₆H₂) 1. This type of metal-free hydrogenation has the potential to replace high cost metal catalyst.

The mechanism of imine reduction is proposed to involve protonation at nitrogen giving the iminium salt. The basicity of the nitrogen centre determines the rate of reaction. More electron rich imines reduce at faster rates than electron poor imines. The resulting iminium center undergoes nucleophilic attack by the borohydride anion to form the amine. Small amines bind to the borane, quenching further reactions. This problem can be overcome using various methods: 1) Application of elevated temperatures 2) Using sterically bulky imine substituents 3) Protecting the imine with the B(C₆F₅)₃group, which also serves as a Lewis acid promoter.[10]

Chiral boronate Lewis acid derived from (1R)-(+)-camphor form a frustrated Lewis pair with ^tBu₃P, which is isolable as a salt. This FLP catalyses the enantioselective hydrogenation of some aryl imines in high yield but modest ee (up to 83%).

Asymmetric imine hydrogenation by an FLP

Although conceptually interesting, the protocol suffers from lack of generality. It was found that increasing steric bulk of the imine substituents lead to decreased yield and ee of the amine product. methoxy-substituted imines exhibit to superior yield and ee's.[10]

A group of catalysts, Frustrated Lewis pairs of chiral alkenylboranes and phosphines are beneficial for asymmetric Piers-type hydrosilylations of 1,2-dicarbonyl compounds and alpha-keto Esters, resulting in products of high yield enantioselectivity. However, in comparison conventional Piers-type hydrosilyation, asymmetric Piers-type hydrosilylations are not as well developed

In the following example, the chiral alkenylborane is formed in situ from chiral diyne and the HB(C₆F₅)₂. Heterolytic cleavage of Si-H bond from PhMe₂SiH by the FLP catalyst, forms a silylium and hydridoborate ionic complex.[11]

Asymmetric Hydrosilylation of a Diketone by an FLP

Metal free hydrogenation of unactivated internal alkynes to cis-alkenes is readily achieved using FLP-based catalysts.[12] The condition for this reaction were relatively mild utilising 2 bar of H₂. In terms of mechanism, the alkyne material is first hydroborated and then the resulting vinylborane-based FLP can then activate dihydrogen. A protodeborylation step releases the cis-alkene product, which is obtained due to the syn-hydroborylation process, and regenerating the catalyst. While active for alkyne hydrogenation the FLP-based catalysts do not however facilitate the hydrogenation of alkenes to alkanes.

Mechanism for hydrogenation of alkynes using FLPs

The reaction is a syn-hydroboration, and as a result a high cis selectivity is observed. At the final stage of the catalytic cycle the C₆F₅ group is cleaved more easily than an alkyl group, causing catalyst degradation rather than alkane release. The catalytic cycle has three steps:

• Substrate binding (the hydroboration of alkyne)
• H₂ cleavage with vinylborane, followed by intramolecular protodeborylation of vinyl substituent, recovering N,N-Dimethyl-2-[(pentafluorophenyl)boryl]aniline
• Release of the cis-alkene

Binding of terminal alkyne to the FLP catalyst

With internal alkynes, a competitive reaction occurs where the proton bound to the nitrogen can be added to the fluorobenzenes. Therefore, this addition does not proceed that much, the formation of the alkene seems favoured.

But terminal alkynes do not bind to the boron through hydroboration but rather through C-H activation. Thus, the addition of the proton to the alkyne will result in the initial terminal alkyne. Hence this hydrogenation process is not suitable to terminal alkynes and will only give pentafluorobenzene.

The metal free hydrogenation of terminal alkynes to the respective alkenes was recently achieved using a pyridone borane based system.[13] This system activates hydrogen readily at room temperature yielding a pyridone borane complex.[14] Dissociation of this complex allows hydroboration of an alkyne by the free borane. Upon protodeborylation by the free pyridone the cis alkene is generated. Hydrogenation of terminal alkynes is possible with this system, because the C-H activation is reversible and competes with hydrogen activation.

Amine-borane FLPs catalyse the borylation of electron-rich aromatic heterocycles (Scheme 1).[15] The reaction is driven by release of hydrogen via C-H activation by the FLP. Aromatic borylations are often used in pharmaceutical development, particularly due to the abundance, low cost and low toxicity of boron compounds compared to noble metals.,

Scheme 1: Mechanism for borylation catalysed by FLP

The substrate for the reaction has two main requirements, strongly linked to the mechanism of borylation. First, the substrate must be electron rich, exemplified by the absence of a reaction with thiophene, whereas its more electron rich derivatives - methoxythiophene and 3,4-ethylenedioxythiophene - can undergo a reaction with the amino-borane. Furthermore, substitution of 1-methylpyrrole (which can react) with the strongly electron withdrawing tertbutyloxycarbonyl (Boc) group at the 2-position completely inhibits the reaction. The second requirement is for the absence of basic amine groups in the substrate, which would otherwise form an unwanted adduct. This can be illustrated by the lack of a reaction with pyrrole, whereas both 1-methyl and N-benzylpyrrole derivatives are able to react.

Further work by the same authors revealed that simply piperidine as the amine R group (as opposed to tetramethylpiperidine, pictured above) accelerated the rate of reaction. Through kinetic and DFT studies the authors proposed that the C-H activation step was more facile than with larger substituents.[16]

Dearomatisation can also be achieved under similar conditions but using N-tosyl indoles. Syn-hyrdoborylated indolines are obtained.[17]

Deromatisation of N-tosyl indole by HBpin

Borylation of S-H bonds in thiols by a dehydrogenative process has also been observed. Alcohols and amines such as tert-Butanol and tert-Butylamine form stable products that prevent catalysis due to a strong π-bond between the N/O atom’s lone pair and boron, whereas the same is not true for thiols, thus allowing for successful catalysis. In addition, successful borylation of Se-H bonds has been achieved. In all cases, the formation of H₂ gas is a strong driving force for the reactions.[18]