CHAPTER 1: Introduction to New Trends in Cross-Coupling

Transition metal-catalyzed reactions play a vital role in the production of many industrially important chemicals, where homogeneous catalysis (reactions that take place in the same phase as the catalyst) is rapidly growing, as evidenced by the awarding of three distinct Nobel Prizes in Chemistry during the last decade – chiral catalysis (2001; Noyori, Sharpless and Knowles),¹  olefin metathesis (2005; Grubbs, Chauvin and Schrock)²  and cross-coupling [2010; Heck (Figure 1.1), Suzuki (Figure 1.2) and Negishi (Figure 1.3)].³  The field of cross-coupling, well dominated by homogeneous catalysis, has undoubtedly turned into an area appreciated by all synthetic chemists, irrespective of their prominence in academia or industry.

Recently, heterogeneous catalysis (reactions that take place in a different phase than the catalyst) has also been used for simple cross-coupling reactions, relying on metal leaching to mediate the desired reaction. However, the leached metal must subsequently be readsorbed in order not to contaminate the final product (“release-and-catch” strategy).⁴  This is not always ideal, depending on the target use of the product and from a reproducibility point of view. In addition, with reactions catalyzed heterogeneously, it is difficult to carry out reactions with high selectivity, in terms of stereo-, regio- or, in some cases, chemoselectivity. Pd-catalyzed cross-coupling has enriched the area of homogeneous catalysis, where rapid growth has been taking place in the past several years, as evidenced by the growing total number of publications/patents¹ in the area (Table 1.1). Thus, Pd-catalyzed cross-coupling reactions comprise one of the most important classes of synthetic transformations in modern organic chemistry, providing chemists with an exceptionally powerful tool for the construction of carbon–carbon (C–C) and carbon–heteroatom (C–X) bonds. These and many related transformations have become ubiquitous in both industry and academia. Indeed, as mentioned above, the 2010 Nobel Prize in Chemistry was a monumental accomplishment for the assiduous contributions of Professors Richard F. Heck (University of Delaware), Akira Suzuki (University of Hokkaidō) and Ei-ichi Negishi (Purdue University) for their achievements within the area of Pd-catalyzed C–C bond-forming reactions.⁵  According to Negishi, the roots of cross-coupling can be traced all the way back to Victor Grignard (Figure 1.4). These early studies laid the foundation of what would become one of the most important and most studied classes of catalytic reactions. Intense research efforts would soon spawn several new C–C coupling reactions in addition to C–X coupling reactions as the chemistry evolved into what it has become today.

The aim of this book is to serve both academia and industry. In the following sections, some key parameters and basic concepts are introduced.

The turnover number is defined as the absolute number of passes through the catalytic cycle before the catalyst becomes deactivated. In general, industrial chemists are interested in both TON and turnover frequency (TOF) (see the next section). A large TON (e.g., 10⁶–10¹⁰) indicates a stable, very long-lived catalyst. The TON can be calculated by dividing the amount of reactant (moles) by the amount of catalyst (moles):

Authors often report mole % of catalyst used. This refers to the fraction of catalyst used relative to the amount of limiting reactant present.

Turnover frequency is defined as the number of passes through the catalytic cycle per unit time (typically seconds, minutes or hours). This number is usually determined by dividing the TON by the time required to produce the given amount of product.

However, as with the TON, the actual yield of the product also needs to be taken into account. Continuing the example above, if the reaction in question was run for 7 h to obtain the 94% yield, the TOF is

The generally accepted simplified catalytic cycle for cross-coupling reactions is shown in Scheme 1.1, where L_nPd(0), the active catalytic species, acts as a “matchmaker”. In Japanese language, “catalyst” is pronounced shoku bai and in Chinese it is chu mei (the same character as for matchmaker).⁶ 

In C–C bond-forming cross-coupling, there are two coupling partners: an aryl/vinyl halide or pseudohalide and an organometallic reagent such as a Grignard reagent. There are three basic steps in the catalytic cycle: oxidative addition, transmetallation and reductive elimination.

Here is an analogy: one of the partners with a family member or friend (R–X) establishes a connection with the matchmaker [L_nPd(0)] with the profile of “R”. This is called the oxidative addition of an organic halide/pseudohalide, R–X, to L_nPd(0) to generate an R–Pd(Ar)(X)(II) intermediate. In the second step, the other partner (R¹) in the form of R¹–M also forms a connection with the matchmaker so that R and R¹ can communicate with each other through the Pd. This is the second step, called transmetallation, where M (a friend or family member of R¹) forms a “bond” with X. In the third step, R and R¹ are united and detach from the matchmaker (Pd catalyst) in an event called reductive elimination. The success of a matchmaker depends on how many challenging coupling partners are successfully coupled (get married) without any deleterious incidents, within a short time frame. This is related to the TON and TOF of the catalyst. Although Heck shared the 2010 Nobel Prize for Pd-catalyzed cross-coupling reactions with Suzuki and Negishi, some argue that the Heck–Mizoroki reaction (often shortened to the Heck reaction) is not a true cross-coupling reaction as it does not involve a transmetallation step. In the Heck reaction, the Pd(II)–R species undergoes a migratory insertion with the alkene substrate, followed by a syn-periplanar β-hydride elimination event to give the product. This step was well established by the work of Fu and Hartwig.⁷  Base is necessary to turn over palladium catalyst by inducing the reductive elimination of HX in the last step. Depending on the nature of substituents on the olefin, linear or branched coupled products are obtained, as these olefin substituents can influence the regioselectivity of the product. The general rule of thumb is that electron-withdrawing groups on the olefin favor linear products with neutral Pd complexes. Bidentate ligands such as dppf [1,1′-bis(diphenylphosphino)ferrocene] under cationic conditions⁸  and dnpf [1,1′-bis(dinaphthylphosphino)ferrocene] in presence of a polar solvent and TBAC (tetrabutylammonium chloride) additive⁹  produce branched products for electron-rich and electron-neutral olefins.

Since the original discoveries of cross-coupling reactions, there has been a great deal of effort in this area to better understand the reaction mechanism, where the role of the ligand is important. The electronic and steric nature of the ligand (L) and the coordination number of Pd can significantly influence two important steps of the cycle; oxidative addition and reductive elimination (Figure 1.5).¹⁰  The role of ligands in the transmetallation step is not as well understood; however, the groups of Hartwig, Amatore and Lloyd-Jones have carried out some impressive work in the area of Suzuki–Miyaura coupling.¹¹  The groups of Beletskaya¹²  and Buchwald¹³  have shown that more electron-deficient ligands can increase the rate of C–N cross-coupling reactions involving ureas and amides, respectively, likely reflecting an increased rate of “transmetallation” (amide binding).¹³  Oxidative addition was considered to be the rate-limiting step, where the choice of the ligand is important. For example, it is proposed that electron-rich ligands make the Pd basic enough to do the oxidative addition of challenging aryl chlorides, while with aryl iodides and bromides oxidative addition is relatively facile, even with less electron-rich ligands such as Ph₃P. Figure 1.5 shows the valence bond (VB) representations for the two components L₂Pd and Ar–X and for a concerted, three-centered transition state of the oxidative addition process.¹⁴  The energy (ΔG) required to excite one electron into the antibonding (σ*) orbital of the Ar–X bond decreases in the series Ar–Cl>Ar–Br>Ar–I.

The low reactivity of more challenging substrates such as unactivated aryl chlorides was often attributed to the large bond dissociation energy of the C–Cl bond (95 kcal mol⁻¹) in comparison with Ar–Br (79 kcal mol⁻¹) or Ar–I (64 kcal mol⁻¹),¹⁵  which highlights the difficulty for an aryl chloride to add oxidatively to a less electron-rich L_nPd(0) species.¹⁶ 

Interestingly, in the transmetallation step, recent evidence suggests that the trend is the opposite: chloride complexes are transmetallated faster than those of bromides and iodides.^13,17  The size of the ligand also plays an important role in the reductive elimination,¹⁰  in addition to stabilizing the coordinatively unsaturated L_nPd(0).

The intent of this chapter is not to provide an exhaustive review of the history of cross-coupling reactions,¹⁸  but to identify the most notable milestones (Figure 1.6) and the genesis of some of the topics of the chapters presented here.

Some argue that the history of the use of metals as catalysts to accomplish organic transformations was initiated by Fittig, who recorded sodium-mediated alkylations of halogenated arenes in 1862.¹⁹  In the early 1900s, Ullmann and Goldberg carried out extensive studies on copper-catalyzed C–C, C–O and C–N bond-forming reactions.²⁰  Noteworthy is that the first person to combine successfully organometallic reagents with catalysis, in this case NiCl₂, was the French chemist André Job.²¹  He reported that PhMgBr, in the presence of NiCl₂, was able to absorb CO, NO, C₂H₄, C₂H₂ and H₂. Since Job's underappreciated revolutionary discovery, nickel has been overshadowed by palladium in similar transformations. Since this early discovery, carbonylation has become an industrially important process and its modern version, carbonylative cross-coupling, is reviewed in detail by Xiao-Feng Wu and Christopher Barnard in Chapter 10.

Following Job's discoveries, the next notable milestone would be the reports by Kharasch on the metal-catalyzed homo-couplings of organomagnesium reagents.²²  More specifically, he employed catalytic amounts of CoCl₂, MnCl₂, FeCl₃ or NiCl₂ in the presence of Grignard reagents and organic halides to affect this homo-coupling reaction (Scheme 1.2, top).

The use of vinyl bromide in place of bromobenzene, under the same conditions, resulted not in the expected homo-coupling of the Grignard reagent, but in the first-ever reported catalytic cross-coupling reaction (Scheme 1.2, bottom).²³  These findings, to some extent, make Kharasch (Figure 1.7) the “grandfather” of cross-coupling reactions.

More than 20 years later came the next breakthrough in the Pd-catalyzed cross-coupling area. Heck reported in 1968 that arylations of alkenes could be achieved by using an organomercury arylating reagent and a palladium catalyst (Scheme 1.3).²⁴ 

A modification of this Pd-catalyzed reaction was subsequently published by Moritani and Fujiwara. They disclosed the direct coupling between arenes and alkenes, first in the presence of stoichiometric amounts of palladium compounds²⁵  and later using catalytic amounts (Scheme 1.4).²⁶  This finding can be classified as one of the first direct C–C bond formations via C–H activation chemistry.

Most of the early developments involved the use of prefunctionalized coupling partners in terms of organometallic reagents as nucleophiles and aryl halides as electrophiles. An alternative attractive approach would be (as Fujiwara and Moritani showed) the direct functionalization of arene C–H bonds, without the need for prefunctionalization. In addition to Fujiwara and Moritani's disclosure, a few examples of C–H activation were reported in the 1980s by Ames²⁷  (intramolecular) and Ohta²⁸  (intermolecular). During the past two decades, the development of palladium-catalyzed direct arylations has progressed enormously and these advances are discussed by Upendra Sharma, Atanu Modak, Soham Maity, Arun Maji and Debabrata Maiti in Chapter 12.

Building on Kharasch's cobalt-catalyzed cross-coupling reaction, Kochi accomplished an iron-catalyzed reaction between C(sp²)–Br electrophiles and Grignard reagents (Scheme 1.5).²⁹ 

In the same year, Mizoroki and co-workers presented a related reaction to the one reported by Heck in 1968 that importantly did not require the use of toxic arylmercury, -tin or -lead reagents. The C–C bond formation between ethylene or monosubstituted alkenes and iodobenzene could be achieved using catalytic amounts of PdCl₂ or heterogeneous Pd black (Scheme 1.6).³⁰  Concurrently, Heck demonstrated independently the Pd-catalyzed reaction of aryl halides with alkenes in the presence of a hindered amine base.³¹  Heck's work on aryl and vinyl halide substrates led to the second most practiced reaction in cross-coupling, namely the Mizoroki–Heck reaction.³²  Irina Beletskaya and Andrei Cheprakov in Chapter 9 discuss the role of modern Heck reactions in organic synthesis.

So far, only simple metal salts had been employed as catalysts. Corriu and Masse³³  and Tamao and Kumada³⁴  independently described the nickel-catalyzed coupling reaction of Grignard reagents with aryl halides. Tamao and Kumada (Figure 1.8) thereby pioneered the area of cross-coupling by showing the effects of adding phosphine ligands to the catalysts.^33–35 

The benefit of using phosphine ligands was particularly striking in reactions with less reactive aryl chlorides. Chapter 2, authored by Andrew DeAngelis and Thomas Colacot, covers the emergence of the development and use of ligands in Pd-catalyzed cross-coupling reactions in detail, with some theoretical background in choosing the right ligands for specific reaction types. Only during the past 10–15 years has the importance of the steric and electronic properties of the ligands used been fully recognized and evaluated.

A few years after Tamao and Kumada's disclosure of the importance of added phosphine ligands, four groups made independent reports on the cross-coupling reactions between alkynes and C(sp²)–halide reagents.³⁶  Cassar made use of Pd(PPh₃)₄ as catalyst, whereas Heck and Sonogashira (Figure 1.9) used Pd(PPh₃)₂(OAc)₂ and Pd(PPh₃)₂Cl₂, respectively. In addition, Sonogashira's studies showed that the presence of catalytic amounts of CuI enabled the reaction to proceed efficiently under significantly milder reaction conditions (Scheme 1.7). Recent work from a few laboratories, including ours, demonstrated that Cu is not required; in fact, it is detrimental for alkynylation reactions of aryl chlorides and unactivated aryl bromides where electron-rich Pd catalysts are required.³⁷  An insight into this phenomenon has been proposed.^37a 

Until 1976, the development of cross-coupling reactions had focused on the use of magnesium-based Grignard reagents as nucleophiles, with the exceptions of the discoveries of Mizoroki, Heck and Sonogashira. However, in 1976, Negishi showed that other organometallic reagents could be used as coupling partners.³⁸  First, Ni- or Pd-catalyzed cross-coupling reactions with organoaluminium reagents as nucleophiles were disclosed,³⁹  followed by the use of arylzinc reagents (Negishi reaction) (Scheme 1.8).⁴⁰ 

Continuing the investigations of the use of alternative nucleophiles, Kosugi and Migita,⁴¹  followed by Stille,⁴²  reported cross-coupling reactions involving organotin reagents (Scheme 1.9).

Extending the range of possible organometallic coupling partners, following the report by Murahashi's on a transmetallation strategy for using trialkylboranes,⁴³  Suzuki and Miyaura disclosed their findings on the beneficial effect of added bases in cross-coupling reactions of alkenylboranes with aryl halides catalyzed by palladium (Suzuki–Miyaura reaction) (Scheme 1.10).⁴⁴ 

Since Suzuki and Miyaura's seminal report, this cross-coupling reaction has been extensively developed and its mechanism studied in great detail. Chapter 8, authored by Alastair Lennox and Guy Lloyd-Jones, describes the discovery, development and deployment of boron reagents in cross-coupling reactions, including mechanistic analysis. In Chapter 11, the topic of stereospecific and stereoselective Suzuki–Miyaura coupling reactions is discussed by Ben Glasspoole, Eric Keske and Cathleen Crudden.

Subsequent to the introduction of organoboron reagents as cross-coupling partners, Hiyama reported efficient cross-coupling reactions of arylsilanes by the use of fluoride additives (Scheme 1.11).⁴⁵ 

By 1989, it was already possible to achieve C–C bond formation through cross-coupling reactions using a wide variety of organometallic reagents. The next breakthrough came with the independent discoveries by Buchwald and Hartwig that C–N bond formation could also be efficiently accomplished through palladium-catalyzed cross-coupling reactions of free amines with aryl halides (Buchwald–Hartwig amination) (Scheme 1.12).⁴⁶  This methodology was also extended to cover C–O and C–S bond formation. Chapter 6, authored by James Stambuli, provides a specific discussion of the development of C–O and C–S bond-forming reactions using cross-coupling methodologies.

At this stage, it may seem like most of the monumental discoveries had already been made; however, there is by no means an end to the advancement of cross-coupling reactions. Reports continue to appear in the literature regarding novel reactivity, enhanced substrate compatibility, new trends and solutions to previously problematic transformations.

Following the seminal discoveries made before 2000, the first decade of the new millennium would see the expansion of cross-coupling reactions to include the use of the previously very challenging aryl chlorides as coupling partners. This was achieved by the groups of Buchwald,⁴⁷  Fu⁴⁸  and Koie⁴⁹  by means of using bulky, electron-rich monophosphines as ligands. In addition, enormous progress has been made to be able to employ C(sp³)-based coupling partners, such as alkyl–BR₂ or alkyl halide reagents.⁵⁰  The intent of the chapters in this book is to highlight and discuss the most recent advances and discoveries within the field of cross-coupling.

Chapter 3 by Carin Johansson Seechurn, Hongbo Li and Thomas Colacot discusses the advantages of the use of well-defined preformed Pd catalysts in cross-coupling reactions. This is an area that has only been recognized for its importance in the past 10 years.⁵¹ 

Although Chapter 2 briefly touches the subject of ligands other than phosphine, Chapter 4, by Anthony Chartoire and Steven Nolan, details the advances in C–X coupling that has been made using NHC (N-heterocyclic carbene) ligands in conjunction with palladium.

Specific examples of the importance of the structural design of the ligands and its impact on the outcome of cross-coupling reactions are demonstrated in Chapter 5 by Mark Stradiotto. Here, selective monoarylation reactions of amines and carbonyl compounds are discussed.

In addition to more thorough mechanistic investigations and realizations and the continuous expansion of substrate scope for the named cross-coupling reactions, there are also examples of novel reactions appearing in the literature. An example is highlighted and discussed in Chapter 7 by David Petrone, Christine Le, Stephen Newman and Mark Lautens, where the recently disclosed Pd(0)-catalyzed carboiodination reaction, a reaction fundamentally rooted in the Heck reaction, is also discussed.

The importance of cross-coupling reactions is reflected in their frequent use within the chemical industry. To meet the criteria of the economic demands for cheaper and more efficient processes, the challenges within industry are somewhat different from those met within the academic community. In recent years, continuous flow technology has provided an attractive alternative to the more traditional batch preparations usually carried out in process chemistry. This technology has also reached the field of cross-coupling, an area which is discussed by Timothy Noël and Volker Hessel in Chapter 13. The chemical industry is constantly faced with stricter regulations regarding the environmental impact of the processes being carried out. Kevin Shaughnessy reviews the area of green chemistry via cross-coupling reactions in Chapter 14. Javier Magano and Joshua Dunetz demonstrate in Chapter 15 some recent applications of cross-coupling reactions in the large-scale synthesis of pharmaceuticals and showcase the problems encountered in a process chemistry environment. Chapter 16, authored by Kazunori Koide, highlights the problem with residual Pd in final APIs (active pharmaceutical ingredients) and discusses detection methods.

In addition to all the improvements developed to expand the field of cross-coupling involving the introduction of novel ligands, precatalysts, chemical engineering designs and ever-more in-depth understanding of reaction mechanisms, the area continues to grow as challenges in terms of incorporating problematic nucleophilic and electrophilic reaction partners are met with innovative solutions. As an example, shown in Scheme 1.13, Feringa and co-workers recently demonstrated the previously difficult⁵²  direct coupling reaction of organolithium reagents with aryl halides (Murahashi coupling),⁵³  demonstrating how important problems can be solved through the utilization of the ever-growing technology made available by the numerous research groups across the globe who are active within the field of cross-coupling.

Substantial discoveries within the field of cross-coupling have not reached an end; new reactions are waiting to be discovered, as chemists will never run out of problems to solve. Even fundamental questions such as why Pd is significantly better than Ni or Pt in most cases are still not fully understood.