ReviewsREVIEW

Transition metal–catalyzed alkyl-alkyl bond formation: Another dimension in cross-coupling chemistry

See allHide authors and affiliations

Science  14 Apr 2017:
Vol. 356, Issue 6334, eaaf7230
DOI: 10.1126/science.aaf7230

Stitching one alkyl group to another

Chemical reactions such as Heck and Suzuki coupling facilitate access to an enormous range of relatively flat molecules. This geometrical constraint is associated with the comparative ease of linking together aryl and alkenyl carbons. Choi and Fu review recent developments in forming bonds between the more abundant alkyl carbon centers that underlie diverse molecules with complex three-dimensional structures. Nickel catalysis in particular has emerged as a powerful method to access individual mirror-image isomers selectively and thereby tune the biological properties of the targeted products.

Science, this issue p. eaaf7230

Structured Abstract

BACKGROUND

The development of useful new methods for the construction of carbon-carbon bonds has had an impact on the many scientific disciplines (including materials science, biology, and chemistry) that use organic compounds. Tremendous progress has been made in the past several decades in the creation of bonds between sp2-hybridized carbons (e.g., aryl-aryl bonds), particularly through the use of transition metal catalysis. In contrast, until recently, advances in the development of general methods that form bonds between sp3-hybridized carbons (alkyl-alkyl bonds) had been rather limited. A variety of approaches, such as classical SN2 reactions and transition metal catalysis, typically led to side reactions rather than the desired carbon-carbon bond formation. With transition metal catalysis, the unwanted but often facile β-hydride elimination of alkylmetal complexes presented a key impediment to efficient cross-coupling of alkyl electrophiles.

In the case of many alkyl-alkyl bonds, there is an additional challenge beyond construction of the carbon-carbon bond itself: controlling the stereochemistry at one or both carbons of the new bond. It is important to control the stereochemistry of organic molecules because of its influence on properties such as biological activity.

Each of these two challenges is difficult to solve individually; addressing them simultaneously is even more demanding. Until recently, the methods for achieving alkyl-alkyl bond formation were comparatively limited in scope, typically involving the use of unhindered (e.g., primary) electrophiles and unhindered, highly reactive nucleophiles (e.g., Grignard reagents, which have relatively poor functional group compatibility). With respect to enantioconvergent reactions, there were virtually no examples.

ADVANCES

In recent years, it has been established that, through the action of an appropriate transition metal catalyst, it is possible to achieve a broad range of alkyl-alkyl bond-forming processes; nickel-based catalysts have proved to be especially effective. With respect to the electrophilic coupling partner, a wide range of secondary alkyl halides are now suitable. This has enabled the development of enantioconvergent reactions of readily available racemic secondary electrophiles. In view of the abundance of tertiary stereocenters in organic molecules, this is a noteworthy advance in synthesis.

With respect to the nucleophilic partner, alkylboron and alkylzinc reagents (Suzuki- and Negishi-type reactions, respectively) can now be used in a wide variety of alkyl-alkyl couplings, which greatly increases the utility of such processes, as these nucleophiles are more readily available and have much improved functional group compatibility relative to Grignard reagents. These new methods for alkyl-alkyl bond formation have been applied to the synthesis of natural products and other bioactive compounds.

OUTLOOK

A number of major challenges remain. For example, with regard to the electrophilic coupling partner, there is a need to develop general methods that are effective for tertiary alkyl halides, including enantioconvergent processes. With regard to the nucleophilic partner, there is a need to discover more versatile catalysts that can use a wide range of hindered (e.g., secondary and tertiary) alkylmetal reagents, as well as to achieve a broad spectrum of enantioconvergent couplings of racemic nucleophiles. These advances can enable the doubly stereoconvergent coupling of a racemic electrophile with a racemic nucleophile.

The synthesis of alkyl-alkyl bonds is arguably the most important bond construction in organic synthesis. The ability to achieve this bond formation at will, as well as to control the product stereochemistry, would transform organic synthesis and empower the many scientists who use organic molecules. Recent work has provided evidence that transition metal catalysis can address this exciting challenge.

Alkyl-alkyl bond formation, including control of stereochemistry: an ongoing challenge in organic synthesis.

From top to bottom: sp2- versus sp3-hybridized carbon-carbon bonds; the difficulty of stereochemical control; and enantioconvergent reactions of racemic secondary electrophiles and racemic nucleophiles. X, leaving group; M, metal.

Abstract

Because the backbone of most organic molecules is composed primarily of carbon-carbon bonds, the development of efficient methods for their construction is one of the central challenges of organic synthesis. Transition metal–catalyzed cross-coupling reactions between organic electrophiles and nucleophiles serve as particularly powerful tools for achieving carbon-carbon bond formation. Until recently, the vast majority of cross-coupling processes had used either aryl or alkenyl electrophiles as one of the coupling partners. In the past 15 years, versatile new methods have been developed that effect cross-couplings of an array of alkyl electrophiles, thereby greatly expanding the diversity of target molecules that are readily accessible. The ability to couple alkyl electrophiles opens the door to a stereochemical dimension—specifically, enantioconvergent couplings of racemic electrophiles—that substantially enhances the already remarkable utility of cross-coupling processes.

The construction of carbon-carbon bonds is central to organic chemistry. During the past several decades, a wide array of powerful new methods for carbon-carbon bond formation have been developed, including two transition metal–catalyzed processes that have recently been recognized with Nobel Prizes in Chemistry [olefin metathesis in 2005 (1) and cross-coupling in 2010 (2)]. Such methods have an impact not only on synthetic organic chemistry, but also on the many other disciplines that involve organic compounds, including biology and materials science.

Metal-catalyzed cross-coupling can provide a particularly straightforward, modular approach to carbon-carbon bond formation through the union of two coupling partners, an organic electrophile and an organometallic nucleophile, which may be either commercially available or readily synthesized (Fig. 1A) (2). Early studies of such processes were dominated by the use of palladium catalysts to accomplish couplings that generate a bond between two sp2-hybridized carbons, and these methods have found application in industry (Fig. 1; R and R1 = aryl or alkenyl).

Fig. 1 Transition metal–catalyzed cross-coupling to form carbon-carbon bonds.

(A) General scheme. (B) Application of a Suzuki cross-coupling to form a Embedded Image bond in an industrial synthesis of BASF’s agricultural fungicide Boscalid (>1000 tons/year).

Although methods to construct carbon-carbon bonds between sp2-hybridized carbons (e.g., “aryl-aryl” bonds) are exceptionally powerful tools in organic synthesis, bonds between sp3-hybridized carbons (“alkyl-alkyl” bonds) are much more common. Figure 2A provides illustrative examples of bioactive compounds that include a variety of alkyl-alkyl bonds. The development of effective cross-coupling catalysts that could generate such bonds at will would have a substantial impact on the retrosynthetic analysis (3) and, in turn, the synthesis of a broad array of organic molecules. Until recent years, progress in addressing this challenge had been limited; thus, Denmark and Sweis observed in 2004 that, despite the pervasiveness of alkyl-alkyl bonds, “alkyl-alkyl cross-coupling reactions have historically been the most difficult to realize” (4).

Fig. 2 Stereochemistry as an added dimension in cross-coupling reactions of alkyl electrophiles.

(A) Bioactive compounds that include an array of alkyl-alkyl bonds. (B) Aryl electrophiles (top) versus alkyl electrophiles (bottom). (C) Use of a chiral catalyst to control stereochemistry: enantioconvergent cross-couplings of racemic alkyl electrophiles. Me, methyl; ee, enantiomeric excess.

Examination of the bioactive molecules depicted in Fig. 2A serves as a reminder that, in addition to constructing the carbon-carbon bond itself, another challenge—controlling the stereochemistry at the carbon derived from the electrophile (Fig. 2B)—can arise when an alkyl, rather than an aryl, electrophile is used as the cross-coupling partner; in the case of cross-couplings that generate biaryl compounds, the opportunities for enantioselective catalysis are limited (5). Stereochemistry can, of course, play a key role in determining properties such as biological activity (6), and selective control of product stereochemistry independent of the stereochemistry of the starting material (Fig. 2C) is advantageous.

Here, we describe recent progress in addressing these objectives, with a focus on initial breakthroughs (new families of coupling partners and new transition metal catalysts) in alkyl-alkyl cross-couplings of unactivated alkyl electrophiles (Box 1) (79); except as noted, we do not include reactions of activated electrophiles that bear a π system adjacent to the leaving group. In view of the importance of this objective, it is not surprising that a number of related strategies for metal-catalyzed alkyl-alkyl bond construction are now being pursued that build on the discoveries described below, including reductive couplings (10, 11) and decarboxylative couplings (12, 13).

Box 1

Recent progress in transition metal–catalyzed alkyl-alkyl coupling.


Embedded Image

An impediment to alkyl-alkyl cross-coupling: β-hydride elimination

As mentioned above, most early studies of cross-couplings used palladium catalysts and focused on reactions of aryl electrophiles (2). The typical mechanism for such processes (Fig. 3A) involves a sequence of oxidative addition of the organic electrophile (R–X) to a Pd(0) complex (1) to generate an organopalladium(II) complex (2), transmetalation by the nucleophilic coupling partner (M–R1) to furnish a diorganopalladium(II) complex (3), and reductive elimination to form the carbon-carbon bond (R–R1) and regenerate a Pd(0) complex (1) (2).

Fig. 3 Mechanistic details of metal-catalyzed cross-couplings, illustrated for palladium.

(A) An outline of a catalytic cycle. (B) β-Hydride elimination as a side reaction.

Organopalladium(II) complex 2 is a key intermediate in this catalytic cycle (Fig. 3A). Organometallic compounds that bear a hydrogen in the β position have the potential to undergo β-hydride elimination, an intramolecular process that generates a metal hydride (Fig. 3B). In the case of arylpalladium(II) complexes, there is no precedent for β-hydride elimination to form a palladium-aryne complex; correspondingly, cross-couplings of aryl electrophiles are not diverted by this undesired side reaction. In contrast, β-hydride elimination of an alkylpalladium(II) complex to generate a palladium-alkene complex is a common pathway, the efficiency of which is critical for important palladium-catalyzed processes such as the Wacker (14) and Heck reactions (15). The unwanted, but often facile, β-hydride elimination of alkylmetal complexes presents a key impediment to efficient cross-coupling of alkyl electrophiles.

Primary alkyl electrophiles

In early studies, primary alkyl electrophiles were coupled with alkylmagnesium reagents (Grignard reagents) in the presence of transition metal catalysts such as copper (Fig. 4A) (1619). Although such methods have found application in the total synthesis of natural products (Fig. 4B) (20), Grignard reagents can be incompatible with many functional groups, such as carbonyl compounds, that are commonly encountered in organic chemistry (21). More recently, palladium and nickel catalysts have achieved cross-couplings of primary alkyl halides with alkylboron and alkylzinc reagents (Fig. 4C) (22, 23), which have improved functional group compatibility (2).

Fig. 4 Cross-couplings of primary alkyl electrophiles: Early (pre-2000) methods.

(A) Examples. (B) An application in the total synthesis of a natural product. (C) Use of nucleophiles with improved functional group compatibility. n-Hex, n-hexyl; THF, tetrahydrofuran; n-Pr, n-propyl; n-Bu, n-butyl; Tf, trifluoromethylsulfonyl; Et, ethyl; 9-BBN, 9-borabicyclo[3.3.1]nonane; Ph, phenyl; Ac, acetyl; acac, acetylacetonate; NMP, 1-methyl-2-pyrrolidone.

Since 2000, the scope of methods for the cross-coupling of primary alkyl electrophiles with mild organometallic nucleophiles has increased considerably. For example, palladium complexes that bear a bulky, electron-rich phosphine have proved to be versatile catalysts (Fig. 5A) (24), enabling the coupling of an array of alkyl electrophiles with alkylboron reagents; in contrast to the earlier palladium/PPh3-based method, which was only applied to primary alkyl iodides (Fig. 4C), palladium/trialkylphosphine catalysts are effective for cross-couplings of alkyl bromides, chlorides, and tosylates. More recently, copper and iron catalysts have proved useful for cross-couplings of primary alkyl electrophiles with alkylboron reagents (Fig. 5A) (25, 26). The palladium-based method has been applied to late-stage fragment couplings in the total synthesis of natural products such as (+)-spirolaxine methyl ether and (+)-pyranicin (Fig. 5B) (2729).

Fig. 5 Cross-couplings of primary alkyl electrophiles: Recent (post-2000) methods that use alkylboron and alkylzinc reagents as nucleophiles.

(A) Examples with alkylboron reagents. (B) Applications of alkylboron reagents in the total synthesis of natural products. (C) Examples with alkylzinc reagents. (D) An application of an alkylzinc reagent in the total synthesis of a natural product. Cy, cyclohexyl; r.t., room temperature; Ts, p-toluenesulfonyl; t-Bu, tert-butyl; DMF, N,N-dimethylformamide; i-Pr, isopropyl; xantphos, 4,5-bis(diphenylphosphino)-9,9-dimethylxanthene; TBDPS, tert-butyldiphenylsilyl; TBS, tert-butyldimethylsilyl; dba, dibenzylideneacetone.

The use of a bulky, electron-rich ligand has enabled palladium-catalyzed cross-couplings not only of alkylboron reagents, but also of alkylzinc reagents; this has opened the door to carbon-carbon bond formation with a wide range of primary alkyl electrophiles, including iodides, bromides, chlorides, and tosylates (Fig. 5C) (30, 31). A subsequent report showed that copper can catalyze alkyl-alkyl cross-couplings of primary alkyl bromides with alkylzinc reagents (Fig. 5C) (32). These methods have found application in the synthesis of bioactive compounds such as MaR1n–3 DPA (Fig. 5D) (33).

Stereochemical studies of a palladium/trialkylphosphine-catalyzed alkyl-alkyl coupling were consistent with an SN2 pathway for oxidative addition under these conditions (34). This mechanism can account for the inability of this catalyst to accomplish alkyl-alkyl cross-couplings of secondary alkyl electrophiles.

Secondary alkyl electrophiles

As in the case of metal-catalyzed alkyl-alkyl cross-coupling reactions of primary alkyl electrophiles, early proof-of-principle studies showed that couplings of secondary alkyl electrophiles are indeed possible (Fig. 6A) (35, 36). As with primary electrophiles (Fig. 4A), these early methods used reactive Grignard reagents as the nucleophilic coupling partner. Despite this limitation, such cross-couplings have found application in the total synthesis of natural products (Fig. 6B) (37).

Fig. 6 Cross-couplings of secondary alkyl electrophiles: Early (pre-2000) methods.

(A) Examples. (B) An application in the total synthesis of a natural product. HMPA, hexamethylphosphoramide.

More recently, the first methods for coupling secondary alkyl electrophiles with mild organometallic nucleophiles (alkylboron and alkylzinc reagents) have been described [Fig. 7A (26, 38, 39) and Fig. 7B (40)]. To date, nickel-based complexes have proved to be the most versatile, enabling alkyl-alkyl couplings of a range of secondary alkyl electrophiles (iodides, bromides, and chlorides), although copper-catalyzed methods (limited to allylboron reagents) and iron-catalyzed methods have also been reported. Nickel-catalyzed cross-couplings have been applied, for example, to the diastereoselective synthesis of C-alkyl glycosides, an important family of bioactive molecules (Fig. 7C) (41).

Fig. 7 Cross-couplings of secondary alkyl electrophiles: Recent (post-2000) methods that use nucleophiles with improved functional group compatibility.

(A) Alkylboron reagents. (B) Alkylzinc reagents. (C) An application in synthesis. TMEDA, N,N,N′,N′-tetramethylethylenediamine; glyme, 1,2-dimethoxyethane; i-Bu, isobutyl; pin, pinacolato; Cbz, carboxybenzyl; cod, 1,5-cyclooctadiene; s-Bu, sec-butyl; DMA, N,N-dimethylacetamide; Bn, benzyl; DMI, 1,3-dimethyl-2-imidazolidinone.

Catalytic asymmetric carbon-carbon bond formation

The ability to use secondary electrophiles as partners opened the door to an additional dimension in cross-coupling chemistry: catalytic enantioselective carbon-carbon bond formation starting with racemic alkyl electrophiles (Fig. 2C) (42). Preliminary mechanistic data indicated that the nickel-catalyzed cross-coupling methods described above proceed via the formation of a radical intermediate from the electrophile, which is ideal for an enantioconvergent process (Fig. 8A). Thus, both enantiomers of the electrophile could generate the same secondary radical upon homolytic cleavage of the C–X bond, thereby ablating the original stereochemistry and enabling a chiral catalyst to react with the alkyl radical and transform both enantiomers of the electrophile into a single enantiomer of the product. Because catalytic asymmetric alkyl-alkyl cross-couplings are still in a relatively early stage of development [with the exception of allylic electrophiles (43)], we describe couplings not only of unactivated alkyl electrophiles, but also of several activated alkyl electrophiles.

Fig. 8 Catalytic asymmetric carbon-carbon bond formation.

(A) Enantioconvergent cross-coupling via a radical intermediate. (B) Methods for activated electrophiles. (C) An application in the total synthesis of a natural product. diglyme, diethylene glycol dimethyl ether.

Activated racemic alkyl halides, specifically α-bromoamides and benzylic halides, served as the electrophilic partner in early examples of catalytic asymmetric alkyl-alkyl cross-coupling (Fig. 8B) (4446). In the presence of a chiral nickel catalyst, an array of alkylzinc reagents can be used as the nucleophilic partner. These methods have found application in the total synthesis of natural products such as (–)-daphenylline (Fig. 8C) (47).

Unactivated electrophiles can also serve as useful partners in enantioconvergent alkyl-alkyl cross-couplings. In this case, alkylboron reagents have proved to be the nucleophiles of choice, coupling with an array of racemic alkyl halides in good enantiomeric excess and yield with the aid of chiral nickel/diamine catalysts (Fig. 9) (4852). For these methods, the presence of a directing group, which likely interacts with the chiral catalyst in the stereochemistry-determining step of the cross-coupling, is essential for high enantioselectivity.

Fig. 9 Enantioconvergent cross-couplings of unactivated alkyl electrophiles, directed by the indicated functional groups.

Enantioconvergent reactions of racemic electrophiles are not the only opportunity in the field of catalytic asymmetric alkyl-alkyl cross-coupling. The umpolung (53) process (i.e., enantioconvergent couplings of racemic nucleophiles) is also conceivable (Fig. 10A); such a reaction with an alkenyl halide has been reported (54). An example of such a process has recently been described, wherein a racemic alkylzinc reagent is coupled with an array of alkyl halides (Fig. 10B) (55). This study sets the stage for addressing an interesting new challenge: the doubly stereoconvergent cross-coupling of two racemic partners (electrophile and nucleophile) to generate each of the four possible stereoisomeric products simply through the appropriate choice of catalyst (Fig. 10C).

Fig. 10 Enantioconvergent alkyl-alkyl cross-couplings.

(A) Use of a racemic electrophile or a racemic nucleophile. (B) Enantioconvergent nickel-catalyzed coupling of a racemic alkylzinc reagent. (C) Doubly stereoconvergent coupling of a racemic electrophile and a racemic nucleophile. Boc, tert-butoxycarbonyl; de, diastereomeric excess.

Tertiary alkyl electrophiles

Although quaternary carbons are less common than tertiary carbons, the development of alkyl-alkyl cross-couplings to generate such fully substituted centers is nonetheless an important objective in organic synthesis. As in the case of primary and secondary alkyl electrophiles, the initial advances in the use of tertiary electrophiles as coupling partners involved the use of Grignard reagents as nucleophiles. For example, cobalt, silver, and copper complexes serve as effective catalysts for such cross-couplings, but only in the case of allylmagnesium and benzylmagnesium reagents (Fig. 11A) (5658). More recently, silver-catalyzed couplings of tertiary alkyl bromides with organozinc reagents have been described (Fig. 11B) (59), although these methods are also limited to allyl and benzyl nucleophiles, and the mechanism of these processes has not been elucidated. In contrast to alkyl-alkyl cross-couplings of secondary electrophiles, general methods that use organozinc or organoboron nucleophiles have not yet been developed for couplings of tertiary electrophiles, nor have highly enantioselective variants.

Fig. 11 Cross-couplings of tertiary alkyl electrophiles.

(A) Early methods. (B) Recent method that uses a nucleophile with improved functional group compatibility. dppp, 1,3-bis(diphenylphosphino)propane.

Outlook and conclusions

Because alkyl-alkyl bonds are commonplace in organic molecules, the development of increasingly powerful methods for their construction (and for the control of any associated stereochemistry) from readily available coupling partners will have a substantial impact on the many disciplines that make use of organic compounds. In the case of a synthesis of a particular target compound, the availability of tools to reliably achieve alkyl-alkyl bond formation will provide versatile options for retrosynthetic analysis and, in turn, the synthesis of the desired molecule.

In recent years, diversity-oriented library synthesis, which is focused on the efficient generation of families of molecules rather than a particular molecule, has become an increasingly important tool in science, especially in drug discovery (60, 61). This strategy depends on reliable reactions that provide ready access to diverse collections of molecules for compound libraries and for lead optimization. Recently, medicinal chemists have observed that many current efforts in drug development may be biased toward compounds that have aromatic subunits, as a consequence of the dependability of (and therefore the reliance on) cross-coupling reactions of readily available aryl electrophiles and nucleophiles (62). On the other hand, an analysis has suggested that a higher percentage of sp3-hybridized (rather than sp2-hybridized) carbons, as well as a larger number of stereogenic centers, can increase the probability of clinical success for a compound (62, 63). Thus, the development of increasingly versatile tools for alkyl-alkyl bond formation, including stereoselective processes, may facilitate an “escape from flatland” (62, 63). A number of challenges remain, including further expanding the scope of coupling partners and of enantioselective processes.

References and Notes

Acknowledgments: Supported by National Institute of General Medical Sciences grant R01-GM62871.
View Abstract

Navigate This Article