What is a Cross-Coupling? An Argument for a Universal Definition

Abstract

Despite amazing advances in cross-coupling technologies over the past several decades, there is not a consistent definition of what a cross-coupling reaction is. Often, definitions rely on comparison to “traditional” palladium-catalyzed cross-couplings pioneered in the 1970s by chemists such as Suzuki, Negishi, and Heck. While these reactions provide a basis for a cross-coupling definition, they do not define this type of transformation, originally described by Linstead almost 20 years prior. Rather than modify and compartmentalize modern transformations to categorize them into either a synthetic or mechanistic definition, we make an argument for broadening the cross-coupling definition to the union of two distinct molecular entities in a covalent-bond-forming process, to encourage discussion around exploring novel reactivity and disconnections. In addition to making a case for a universal cross-coupling definition, we cite specific examples of reactions that break the mold of prior cross-coupling definitions. We believe this perspective will stimulate dialog around what it means to be a cross-coupling and in turn inspire future developments within this field.

Over the past 65 years, Tetrahedron has established itself as one of the premier scientific journals serving the organic community. Among its most notable contributions to the field are the numerous pioneering Pd-catalyzed cross-couplings that were in great part introduced and highlighted on this platform.¹

Indeed, transition metal-catalyzed cross-coupling has become one of the most versatile and important tools in the modern synthetic chemist’s arsenal. The ability of palladium, in particular, to construct carbon─carbon and carbon─heteroatom bonds both selectively and under mild conditions has revolutionized the fields of synthetic chemistry,² materials science,³ and pharmaceutical development,⁴ among numerous others. The 2010 Nobel Prize, awarded to Richard Heck, Ei-ichi Negishi, and Akira Suzuki “for palladium-catalyzed cross-couplings in organic synthesis" highlights a paradigm shift in how synthetic chemists construct organic molecules.⁵ However, the definition of cross-couplings as "a transition metal-mediated reaction between a nucleophile and an electrophile” remained an unchallenged idea for almost thirty years. As the scope of organometallic partners have expanded into a litany of named reactions, mechanistic hypotheses originally proposed at the advent of palladium-catalyzed coupling methods have become an implicit standard to which all future cross-couplings would be compared.

In reality, the term “cross-coupling” was first used in the 1950’s by Linstead to describe the heterodimerization of two different fatty acids through Kolbe electrolysis.⁶ In the 1970’s and beyond, this term has been used to describe a series of reactions, typically using palladium as a catalyst, for C(sp²)─C(sp²) bond formation. Beginning in the mid-1990s, a new wave of methods challenged the limited definition established in the 1970’s, but have been largely categorized based on their resemblance to the established palladium-catalyzed methods of the 20^th century rather than the individual elements of the reaction, such as transition metals, coupling partners, mechanisms, or types of bond formation.⁵ As a result, transition metal-catalyzed reactions have developed a subjective taxonomy, with boundaries between subfields governed by historical context and surface-level observations, which often create contradictory standards for the inclusion and exclusion of certain reaction manifolds. By re-examining how new reactions are categorized, we hope to clarify and broaden what constitutes a “cross-coupling” to facilitate intellectual advances through an unadulterated approach similar to those that stimulated the initial development of palladium-catalyzed cross-coupling reactions.

Attempts to revise the modern concept of cross-couplings have been hampered by the lack of a universally recognized definition, with most criteria falling into one of two major categories: 1) the mechanistic definition, and 2) the synthetic definition. Contemporary chemists often use mechanistic definitions to classify reactions based on elementary steps, and cross-coupling reactions are often described by a general catalytic cycle of oxidative addition, transmetalation, and reductive elimination (Scheme 1A).^6g While the first and final steps are generally conserved, reactions that utilize migratory insertion processes (e.g. the Heck Reaction or Fujiwara-Moritani Reaction) or nucleophilic substitution processes (e.g. the Buchwald–Hartwig coupling or Ullmann Reaction) in place of transmetalation are usually included in this realm.⁷ Other classification systems rely on synthetic definitions based on the overall chemical transformation (Scheme 1B). Tamao and Miyaura, for example, define these processes as the generation of carbon─carbon (C─C) or carbon─heteroatom (C─X) bonds from organohalide electrophiles and organometallic nucleophiles in the presence of catalytic quantities of Group 8–10 metals.⁸ Echavarren and Homs further limit this definition to reactions based on transmetalation of organometallic nucleophiles to form exclusively C─C bonds.⁹

Scheme 1.

Traditional classifications of cross-couplings: mechanistic (A) and synthetic (B).

The goal of this contribution is to appeal to the broadest community of chemists in an attempt to universally define the term “cross-coupling.” We propose to define a cross-coupling reaction as the union of two distinct molecular entities in a covalent-bond-forming process (Scheme 2). In this way, the terminology returns to the root of the words “cross-coupling,” which implies nothing about mechanism, reactants, or reagents. Although mechanistic studies increase understanding and, therefore, the synthetic utility of these reactions, it should not be necessary to know the mechanism or the specific roles played by each reactant (nucleophile or electrophile) in a particular transformation to deem it a “cross-coupling.” In turn, descriptors may easily be added to more exactly describe a particular reaction (e.g. oxidative Heck reaction, reductive cross-electrophile coupling, etc.)

Scheme 2.

A new way to conceptualize cross-couplings.

By returning to the root of the word, we acknowledge the major obstacle to its adoption: utility. Language and nomenclature are, by design, practical, and the community will not adopt a new definition unless it serves a clear purpose and is more advantageous than previous preconceptions. Cross-couplings of today, while loosely defined, help chemists to organize new advances in the literature. Therefore, such a broad definition risks diluting impact of research if every bond-forming process can be included. New chemistries have been developed even in the presence of these poorly defined subdivisions, so will challenging this paradigm truly add much more to the discussion? Perhaps it may stimulate dialog around novel disconnections and reaction development in the way palladium-catalyzed cross-couplings discovered in the 1970s have revolutionized the manner in which synthetic chemists assemble biaryl compounds.

Breaking the mold: Modern cross-couplings with novel mechanisms and reactants

In recent years, nickel catalysis has developed into a powerful complement to palladium, as facile oxidative addition, slower β-hydride elimination, and readily accessible +1/+3 oxidation states facilitate the transformation of traditionally challenging substrates such as alkyl electrophiles.¹⁰ Though these reactions conform to most synthetic cross-coupling definitions, they differ greatly from the general catalytic cycles proposed by mechanistic descriptions. Fu’s C(sp³)─C(sp²) “Suzuki” coupling, for example, is believed to proceed through transmetalation of a Ni(I) precatalyst, followed by two consecutive single-electron transfers (SET) before reductive elimination occurs from an alkylnickel(III) complex (Scheme 3).¹¹ In the case of reductive cross-couplings, Weix and coworkers suggest an amalgamation of polar oxidative addition and radical chain reaction processes which are used to modulate the oxidation state of their nickel catalyst using stoichiometric manganese, and Reisman has since rendered these reactions enantioselective.¹² Similar reactions have been achieved by metallaphotoredox catalysis,¹³ which enables previously inaccessible transformations by leveraging the ability of a photocatalyst to transfer energy or electrons to a transition metal. Furthermore, these processes are not restricted to heterobimetallic catalytic systems. Homobimetallic catalysts, such as those investigated by Uyeda,¹⁴ have afforded novel reactivity between 1,1-dichloroalkenes and olefins to form products that would otherwise be challenging to conceive using a mechanistic classification of cross-coupling (Scheme 4).

Scheme 3:

General mechanism for Ni-catalyzed alkyl-alkyl/aryl cross-coupling

Scheme 4.

Novel Reactivities Expanding the Scope of the “Cross-Coupling”

The emergence of nickel and other base metals in cross-couplings has enabled a variety of transformations, many of which do not conform to the mechanistic standards of palladium chemistry. As a result of this paradigm shift, other mechanistically similar reactions that do not adhere to the synthetic definition are now included under this moniker. The scope of functional handles that participate in transition-metal-catalyzed couplings has expanded to include carboxylic acids,¹⁵ amides,¹⁶ oxalates,¹³ N-hydroxyphthalimide (NHP) esters,¹⁷ epoxides,¹⁸ aziridines,¹⁹ and numerous others. Reductive couplings have eliminated the need to discriminate functional groups based on electronic properties, as cross-selectivity can originate from hybridization, which disposes substrates to selectively undergo either oxidative addition (C(sp²)─X bonds) or radical formation (C(sp³)─X bonds).²⁰ Finally, advances in C─H and C─C bond functionalization have demonstrated that previously considered “inert” bonds can participate in couplings through judicious choice of catalyst, ligand, and directing group.²¹

Despite such drastic expansion of the field, many chemists maintain stolid ideological barriers between cross-couplings and other classical subdivisions of synthetic methodology. This produces contradictory standards where inclusion and exclusion of reactions are assigned by analogy rather than anatomy. For example, transition metal-catalyzed arylation of enolates with aryl halides are frequently grouped with cross-couplings²² while the corresponding catalytic α-alkylations—specifically allylic alkylations— with allyl alcohols, acetates, and carbonates are considered substitution reactions.⁶ Cross-coupling of enolates with other nucleophiles in an umpolung fashion is further denoted as an oxidative coupling. This might be a vestige of pedagogy since alkyl electrophiles are often associated with nucleophilic substitution reactions while reaction using aryl electrophiles resemble the products of first-generation Pd-catalyzed cross-couplings.

Diversity in Reactivity included by a Universal Definition

One advantage to a broader definition is that it may inspire chemists to exploit non-traditional coupling partners in reaction development. By removing restrictive classifications that delineate traditional cross-couplings and other classical modes of reactivity (e.g. S_N1,²³ aldols,²⁴ 1,2-additions²⁵), chemists are encouraged to explore nonconventional functionalities, mechanisms, and catalysis to facilitate new disconnection strategies.

For instance, metal carbenes are traditionally employed in cyclopropanations, C─H/X─H insertions, and olefin metathesis⁷ (all of which could be individually considered cross-couplings). Recent advances in rhodium and copper catalysis²⁶ allow carbene precursors to serve as ambiphilic partners, reacting with classic electrophiles or nucleophiles (when external oxidants are used) and generating sterically congested moieties like tetrasubstituted olefins and all-carbon quaternary centers (Scheme 5). Another Cu-catalyzed transformation that should be reclassified as a cross-coupling reaction is the ubiquitous “click reaction,” which constitutes the union of organic azide and alkynyl coupling partners in a concerted fashion.²⁷ Despite its incredible importance to polymer synthesis, chemical biology, and other industrial processes, Cu-catalyzed azide-alkyne cycloaddition (CuAAC) has not been classified as a traditional cross-coupling due to its distinct mechanism. However, these reactions closely fit the synthetic definition of cross-couplings, generating products that combine distinct fragments with new carbon─heteroatom bonds.

Scheme 5.

Non-Traditional Disconnection Strategies Redefined as Cross-Couplings

Recent advances in C─C bond activation have highlighted its utility as a cross-coupling reaction. In the presence of rhodium, strained C─C bonds can be exploited to access acyclic or ring-expanded cyclic products, and recently Dong and co-workers demonstrated that unstrained C─C bonds in simple ketones could be employed in Suzuki-Miyaura couplings to produce functionalized aryl ketones.²⁸ We believe this reconceptualization will inspire further advancements in identifying and developing novel coupling partners.

Application of the universal definition of “cross-coupling” enables inclusion of direct arylations, in which an unfunctionalized arene is coupled with either an aryl halide or a main group organometallic nucleophile.^10,29 Direct arylations proceed through a variety of mechanistic pathways depending on the substrate, reaction conditions, and catalyst system, but mechanisms generally differ from the traditional sequence of oxidative addition, transmetalation, and reductive elimination. They have been known to proceed through electrophilic aromatic substitution (S_EAr),³⁰ σ-bond metathesis,³¹ concerted–metalation–deprotonation (CMD),³² arene carbometalation,⁶ or oxidative addition of the arene C─H bond.³³ Indeed, advances in C─H functionalization research have rendered direct arylations³⁴ and oxidative cross-couplings²⁶ viable and efficient options in synthetic planning. Although these reactions follow neither the synthetic nor mechanistic definition of a cross-coupling, they are often discussed alongside traditional cross-coupling reactions due to the resemblance of the biaryl products to those of first-generation cross-couplings. Fully oxidative couplings between two arene C─H bonds are also frequently referred to as “oxidative arene cross-couplings” due to the familiarity of the biaiyl products.

Beyond the coupling of two reaction partners, the field has expanded to include multicomponent reactions. For instance, the performance of well-established cross-couplings such as the Stille,³⁵ Suzuki,³⁶ or Heck³⁷ reactions under carbon monoxide atmosphere results in carbonylative cross-coupling reactions.³⁸ The mechanisms of these cross-couplings include the usual oxidative addition, transmetalation, and reductive elimination steps, but also include migratory insertion(s) prior to closure of the catalytic cycle. Additionally, structurally complex ring systems such as trans-decalins³⁹ and trans-fused hydrindanes⁴⁰ can be assembled using coupling cascades between alkynes and 1,7-enynes or 1,6-enynes, respectively, as illustrated by the efforts of Micalizio and co-workers (Scheme 6).⁴¹ These reactions resemble first-generation cross-coupling reactions only sparingly, yet the formation of C─C bonds mediated by a transition metal species parallels the functional utility of traditional cross-coupling reactions. As such, extending the principles that guided initial cross-coupling development to a broader range of substrate classes and synthetic transformations offers vast opportunities for further methodological growth and exploration of novel reactivity.

Scheme 6.

Titanium-mediated couplings of alkynes and enynes.

Future Outlook

With a re-clarification of terms, practitioners in the field might enjoy an unrestricted outlook on the types of cross-couplings that might be possible. With a clear, broad, and demarcated definition, one could envision a comprehensive table illustrating the products formed from cross-couplings between any types of organic reactants. If more descriptors are included, a large multi-dimensional matrix could be generated to account for other subtopics of interest (i.e. base-metal catalysis, stereoselectivity, carbonylation), as illustrated in Figure 1. Such a broad collection of potential cross-reactants would provide a structured and visual way to conceptualize unexplored reactivity and identify potential avenues for future research.

Figure 1.

Dimensions of cross-coupling research, represented as a matrix.

Having recognized the power of cross-coupling reactions in synthesis, the chemical community has appropriately poured a great deal of research effort into improving upon first-generation discoveries and expanding the scope and applicability of these transformations. These advances have transformed the field significantly beyond its original inception regarding the types of catalysts and coupling partners employed and the products generated. In line with the increased synthetic utility of cross-coupling reactions, these modifications prompt synthetic chemists to re-conceptualize transition-metal-catalyzed reactions not traditionally considered cross-coupling reactions. Ultimately, updating a poorly defined and unnecessarily narrow definition of "cross-coupling" may reveal new angles to important research topics that could fuel the next major synthetic breakthroughs.

Biographies

Christopher Reimann is from Middletown, New Jersey. He earned his B.S. in biochemistry from the University of Southern California (2015) and his Ph.D. in organic chemistry at the California Institute of Technology (2020) under the mentorship of Professor Brian Stoltz. His dissertation research focused on the total synthesis of dimeric indole alkaloids and polyoxygenated diterpenoids through convergent strategies. Since 2020, he has been a Senior Scientist in medicinal chemistry with AstraZeneca Oncology R&D in Waltham, MA.

Kelly E. Kim grew up in Fort Myers, FL, USA, and earned her B.S. degree in 2011 from Yale University, where she double majored in chemistry and music. She completed her Ph.D. in 2016 at the California Institute of Technology in the research group of Professor Brian M. Stoltz, where her dissertation focused on the synthesis and late-stage diversification of the cyanthiwigin natural product core. After continuing her training as an NIH postdoctoral fellow studying the reactivity of iridium complexes with Professor Karen I. Goldberg at the University of Washington in Seattle, Kelly relocated to the University of Washington Tacoma campus in 2019 to begin a position as an Assistant Professor of Organic Chemistry. Her research focuses on the design and synthesis of molecular libraries from complex organic scaffolds of biological interest.

Alexander W. Rand was born in Phoenix, AZ, USA, and earned a B.A. in chemistry in 2015 from Occidental College. He completed his Ph.D. in 2020 at the University of Michigan under the guidance of Professor John Montgomery where he focused on mechanistic investigations and development of new methods in nickel-catalyzed C─N and C─H functionalizations. Since completing his graduate studies, he has pursued the total synthesis of bisindole alkaloids as a postdoctoral scholar in the lab of Professor Brian M. Stoltz.

Farbod A. Moghadam was born in Saratoga, CA, USA and earned his B.S. degree in Chemistry from the University of California, Santa Barbara. At Caltech, he is a current third year graduate student in the Stoltz Lab performing research in primarily transition metal catalysis.

Brian M. Stoltz was born in Philadelphia, PA, USA and earned his B.S. degree from the Indiana University of Pennsylvania. After graduate studies at Yale University in the labs of John L. Wood and an NIH postdoctoral fellowship at Harvard with E. J. Corey, he took a position at the California Institute of Technology in 2000. At Caltech, he is currently the Victor and Elizabeth Atkins Professor of Chemistry and an Investigator of the Heritage Medical Research Institute. His research interests lie in all areas of synthetic organic chemistry. He is the recipient of numerous award and accolades for his research and teaching including most recently the Feynman Teaching Prize from Caltech and the American Chemical Society Award for Creative Work in Synthetic Organic Chemistry. He is a fellow of the American Association for the Advancement of Science and the American Chemical Society. Finally, Stoltz is Editor-in-Chief of Tetrahedron and the Chair of the Executive Committee for Tetrahedron Publications.