Chemistry Corner

Breaking new ground with boron

Several research groups are exploring the reactivity of boron and finding new complexes with real value for organic synthesis, reports Dr Cynthia Challener

A valuable reagent in organic synthesis, boron serves many functions in a variety of different reactions. Boron hydrides can act as both electrophilic and nucleophilic hydride donors. Enol, allyl and propargyl boranes serve as carbanion donors and transfer their substituents to electrophiles. Boron compounds are also recognised Lewis acids, which participate in transmetallation reactions for C-C bond formation.

Despite this wide variety of recognised reactivities, researchers continue to discover both improvements to existing transformations and new structures with interesting potential for application in novel reactions. Transmetallation reactions have received a great deal of attention in recent years. The Suzuki cross-coupling method for C-C bond formation has, in particular, proved of immense value in the commercial synthesis of many fine and speciality chemical intermediates.

Most reactions with aryl boronates, however, involve a two-step process that requires the use of pre-formed lithium or magnesium organometallic reagents and excess boronic acid. Researchers at Ludwig-Maximilians University (LMU) in Munich have developed an alternative process that provides the coupling products via a reaction of inexpensive aryl bromides, magnesium, a trialkylborate and aryl halide (or pseudo-halide) receptor.

In addition to being a one-pot reaction, the process also only requires 0.5 equivalents of B(OBu)₃. The diarylboronates formed transfer both aryl groups under Suzuki-Miyaura cross-coupling conditions (Figure 1) to other aryl halides or pseudo-halides, triflates and tosylates for example. "We were looking to find a more atom-economical method and were therefore quite pleased to see that the bis-substituted compound works effectively in the process," notes LMU's Professor Paul Knochel.

The diarylboronates are prepared at 25°C and the coupling proceeds in overall yields of 70-92%. Ester, cyanide, Boc, (thio)methoxy, amino and silyl functional groups are all tolerated. In some cases, less expensive aryl chlorides can be used in place of the corresponding bromides. Scale-up should not be an issue. The researchers successfully carried out the reaction on a 1 litre scale with no issues.

The procedure also works for alkenyl halides and both electron-rich and electron-poor heterocyclic bromides. In the case of sterically hindered aryl bromides or bromides bearing strongly electron-withdrawing groups, however, Knochel and his team found that the monoboronate provides better results. Otherwise, he says, "we believe this chemistry offers a general and low cost one-step process for the synthesis of polyfunctional magnesium diorganoboronates that tolerates a wide range of functionalities and can be used directly in C-C bond forming reactions".

Figure 1 - One-pot Suzuki-Miyaura cross-coupling

Currently, the group is looking to see if they can expand the range of tolerated functionalities during the formation of the boronate intermediate and also improve the yields of reactions with substrates containing sensitive substituents. Additional investigations are ongoing to identify other possible applications of the diarylboronates. One such possibility is the Hayashi reaction, in which rhodium catalysts are applied to the asymmetric conjugate addition of arylboronic acids to C=C bonds.

"Higher yields and minimised waste are key advantages," Knochel observes. "Boronic acids have been shown to mimic phosphate ions and questions are being raised about possible human toxicity concerns based on this similarity. Whilst no regulations are in place or expected in the near term, there could be at some time in the future. Therefore reactions that minimise the amount of boron that ends up in wastewater could be quite important."

At the University of Mississippi, meanwhile, Assistant Professor Takashi Tomioka is investigating the utility of diaminoboryl compounds in synthetic organic transformations. In particular, his group has developed a new bifunctional diaminoboryl acetonitrile reagent for the synthesis of substituted acrylonitriles.

"We proposed that an α-boryl carbanion, a potentially versatile species, could be more readily accessible if its Lewis acidity could be reduced and this formation of '-ate' complexes be inhibited," Tomioka says. To achieve this goal, ligands with appropriate steric and electronic properties would be necessary to control the behaviour of the overall boron compound.

"Our solution was to use diamino groups to reduce the Lewis acidity through back donation of electron density from the lone pairs of the nitrogen atoms. The diaminoboryl group is mildly Lewis acidic and fairly advantageous to prevent the formation of the '-ate' complex during the course of deprotonation."

"However," he adds, "such an electron-rich boryl group cannot stabilise the α-carbanion well, as a result of lowering the resonance contribution of the boron. Therefore, we attached an electron-withdrawing nitrile group functionality on the carbon α to the nitrogen to increase the acidity of the α-hydrogen and stabilise the generated α-carbanion species." In addition, bulky substituents protect the boron from attack by bases or nucleophiles.

Their first success was the development of a one-pot procedure for the Z-stereoselective synthesis of β-monosubstituted acrylonitriles via the reaction of two equivalents of the lithium salt of acetonitrile with bis(diisopropylamino)chloroborane to generate an in situ α-diaminoboryl carbanion species, which was then reacted with both aromatic and aliphatic aldehydes to produce the substituted acrylonitriles in 80-98% yield with up to 96:4 Z:E selectivity.

"This reaction was surprising because the boron intermediate displays unusual compatibility, even in the presence of a nucleophilic base," Tomioka notes. It is also attractive because the process provides a simple route for selective preparation of the Z-isomer, which is difficult to obtain with existing methods.

Benzaldehydes with substituents in the ortho, meta and para positions can be converted with Z:E selectivites of 80:20 or better, and even aliphatic aldehydes bearing α-acidic protons were converted to the desired product without undergoing competitive enolisation. Sterically hindered aldehydes did provide improved Z:E ratios.

When substituted acetonitriles are used, the reaction proceeds to give α,β-disubstituted acrylonitriles, but yields and Z:E selectivities vary considerably with the starting substrate. Furthermore, it is not attractive to consume two equivalents of expensive nitriles.

The group therefore developed an alternative divergent method for this transformation. In this route, one equivalent of the lithiated acetonitrile is reacted with the chloroboronate and then an alkyl halide. The boron intermediate is then treated with base and the aromatic or aliphatic aldehyde. Again, a wide range of reagents can be used in the process.

Figure 2 - Preparation of 2-aminoquinolines via a diamonoboryl acetonotrile reagent

One critical step was identified, though, to ensure success. "The presence of excess acetonitrile causes unwanted side reactions upon the addition of the aldehyde. Therefore it is necessary to concentrate the boronate intermediate to remove that excess acetonitrile," says Tomioka. The group has applied this reaction to the synthesis of (E)-2-butyl-2-octenal, the alarm pheromone of the African weaver ant.

More recently, both the original method and the divergent route have been used for the one pot synthesis of 2-aminoquinolines, pharmaceutically important alkaloids that may be potential treatments for Alzheimer's disease. "These compounds are typically obtained via reductive cyclisation of nitrophenyl acrylonitrile derivatives and the mildest processes require the Z isomer, which our chemistry provides access to," comments Tomioka.

The reaction of two equivalents of acetonitrile and various 2-nitrobenzaldehydes gives the desired Z-acrylonitriles, which undergo effective reductive cyclisation upon treatment with acetic acid and zinc powder to yield functionalised aminoquinolines in 40-70% yield (Figure 2). Using the divergent approach with just one equivalent of substituted diaminoboryl acetonitrile, yields of 55-73% were obtained for aminoquinolines with substituents ortho to the amino group.

Even 2'-nitroacetophenone gives the desired (Z)-acrylonitrile with decent stereoselectivity (Z:E = 83:17), but reductive cyclisation occurs slowly, providing a 36% yield of the 4-substituted-2-aminoquinoline. "These results are very encouraging. This approach now enables a rapid divergent synthesis of a variety of substituted 2-aminoquinoline derivatives from acetonitrile in one pot," Tomioka concludes.

Currently his group is studying other related but more functionalised diaminoboryl acetonitriles, such as those with two diaminoboron groups, a boron and trialkylsilyl substituent or a boron and halide substituent. These compounds could serve as novel multi-functionalising reagents for organic synthesis. The group plans to investigate their synthetic utility for the production of alkenyl boranes and silanes and possibly epoxy boranes.

Another group that has focused on finding a way to minimise the Lewis acidity of boron compounds has also turned to nitrogen-based ligands. More specifically, Guy Bertrand, professor of Chemistry at the University of California, Riverside, and his research group used two aminocarbene ligands to stabilise the boron centre.

"Generally boron exists in the +3 oxidation state and it has only been recently that transient low valent borylene derivatives have been characterised spectroscopically. Of these compounds, borylene-carbene complexes appear to have very interesting properties," Bertrand observes.

That led his group to propose that two carbene ligands might be enough to stabilise boron and form a neutral complex in the +1 state. They believed that using a cyclic (alkyl)(amino)carbene (CAAC) would decrease the Lewis acidity and accept enough of the electron density of boron into their empty p orbitals.

Figure 3 - Borylene-bis(CAAC) adduct

The CAAC of choice possessed a bulky 2,6-diisopropylphenyl group at nitrogen and a flexible cyclohexyl group on the cyclopentyl ring. The isolated borylene-bis(CAAC) adduct does indeed have boron in the +1 oxidation state and is electron-rich. It reacts with gallium trichloride to form the boryl radical cation [(CAAC)₂BH]⁺.GaCl₄. It can be readily protonated with BrCH₂CO₂H but is not reactive with PhCO₂H. The protonated complex can, in turn, be easily deprotonated with sodium ethoxide, but no reaction occurs with bulky bases.

Importantly, the parent complex and the radical cation are air-sensitive but remain stable for at least two months if stored under argon. And while the steric properties of the CAAC ligands chosen for these initial experiments clearly reduce the reactivity of the boron compound with electrophiles, Bertrand is confident that this difficulty can be overcome.

"We are working to address this steric issue and I am sure we will solve the problem. We have not identified the ideal CAAC ligands yet, but I have no doubt that we will find a solution," he says. The synthesis of this new type of boron compound is doubly exciting for Bertrand because it is a completely new area of research for him as well as an opportunity to explore the chemistry of a novel system.

"We don't really know yet what types of behaviour these new boron compounds will exhibit. They are isolectronic with amines and phosphines but, with the lower electronegativity of boron, they could potentially be strong electron-donor ligands for transition metals," comments Bertrand.

In addition, they should be usable in any applications where phosphines have been applied. "It would be beneficial to replace phosphines, which are often toxic materials, with boron ligands. There is a whole new family of catalysts to explore. What transformations they might be effective in is not yet known. And it is not possible to say in which applications these new neutral boron compounds will best perform, but it will be interesting to explore the numerous possibilities."
 

 

 

 

From Online Issue: December 2011