Photocatalytic Synthesis of (Hetero)biaryls via Palladium-Catalyzed Hybrid-Radical Cross-Coupling of (Hetero)aryl Halides and C–H Functionalization of Heteroarenes

Abstract

Light activates palladium catalysts to enable radical C–H (hetero)­arylation of heterocycles using simple (hetero)­aryl halides. Without the need of an external photocatalyst, the reaction proceeds under mild conditions with broad functional-group tolerance and high site selectivity, expanding photoactivated palladium catalysis for sustainable C–C bond formation in complex bi­(hetero)­aryl scaffolds.

Introduction

Transition-metal catalysis under visible-light irradiation has emerged as a powerful strategy for promoting synthesis methods that are inaccessible under traditional thermal conditions. It features the use of a transition metal complex that acts both as a photosensitizer and catalyst simultaneously, eliminating the need for separate photosensitizer and catalyst components. Among the various metals explored, palladium has recently attracted particular attention owing to its ability to engage in photoinduced single-electron processes that complement classical two-electron Pd(0)/Pd­(II) catalysis (Scheme a). These excited state pathways of palladium complexes can trigger unconventional bond activations, including radical-type C–H functionalizations, cross-couplings, and cascade transformations under exceptionally mild conditions. Since the pioneering studies by the Gevorgyan group, the field of photoinduced palladium catalysis has expanded rapidly. It has been shown that photoexcited Pd complexes can activate a wide range of electrophiles, including alkyl halides or redox-active esters, enabling diverse radical-type bond-forming reactions under mild conditions with a main focus on alkyl radicals as coupling partner (Scheme a). In contrast, the direct use of aryl halides in intermolecular arylation reactions under photoinduced palladium catalysis remains far less developed. Most reported examples involving aryl radicals proceed via intramolecular HAT resulting in radical translocation and formation of less reactive alkyl radicals (Scheme a). The direct application of (hetero)­aryl halides in radical coupling reactions, however, continues to present a major challenge.

1. Pd-Catalyzed Cross-Coupling Reactions, (a) Ground State vs Excited State Reactivity; (b) Conventional Cross-Coupling; (c) Photoinduced Cross-Coupling of Aryl Halides.

The direct C–H arylation of heterocycles is a transformation of high synthetic value. Conventional palladium catalysis, however, typically requires harsh conditions, strong bases, elevated temperatures, and often suffers from issues of site-selectivity. When coupling two distinct (hetero)­aryl fragments, further complications arise from the limited stability of coupling reagents such as boronic acids or Grignard reagents and their poor compatibility with sensitive functional groups. The development of a mild, functional-group-tolerant, and broadly applicable protocol for the cross-coupling of (hetero)­aryl and heteroaryl building blocks would therefore represent a significant advance, providing a sustainable and general route to complex heterobiaryl scaffolds that are key in pharmaceuticals and natural products.

Results and Discussion

Herein, we describe a photoinduced palladium-catalyzed radical C–H (hetero)­arylation of heterocycles using simple (hetero)­aryl halides as readily available aryl radical sources. This transformation harnesses the excitation of a palladium catalyst with light to generate (hetero)­aryl radicals that undergo direct coupling with heteroarenes through a Pd-radical mechanism. The reaction proceeds under mild, operationally simple conditions, exhibits a broad substrate scope, and demonstrates excellent site selectivity alongside outstanding functional group tolerance. Together, these features provide a versatile and sustainable approach to constructing C–C bonds within complex heteroaromatic frameworks (Scheme c).

Against this background, we set out our investigations by studying the reaction of azauracil 1a with phenyl iodide 2a in the presence of different palladium complexes and ligands (for details see the Supporting Information), among which the combination Pd­[(o-Tol)₃P]₂Cl₂ and RuPhos gave the best yield (Table , entry 1). Other ligands such as XantPhos or SPhos gave lower yields (Table , entry 2). Further evaluation of the reaction conditions comprised different bases, solvents and the reaction stoichiometry to give the desired arylation product 3a in 71% yield (for the details of optimization, see the Supporting Information and Table entry 3). It is important to note that only trace amounts of product were obtained in the absence of base, the palladium catalyst, under dark reaction conditions or in air (Table , entry 4), which shows that the Pd catalyst, base and light are required for a productive coupling reaction. Bromobenzene instead of iodobenzene gave a significantly diminished yield of the arylation product (Table , entry 5). Finally, we carried out the reaction under irradiation with 370 nm light, which significantly improved the yield of arylation product 3a to 84% yield (Table , entry 6), which we assume to be related to more facile photoexcitation of the Pd catalyst at shorter wavelengths.

1. Reaction Optimization .

#	changes from above	% yield (3a)
1	none	60
2	XantPhos/SPhos instead of RuPhos	44/57
3	2 equiv of PhI and Cs2CO3	71
4	no base/no Pd-catalyst/no light/in air	trace
5	2 equiv of bromobenzene and Cs2CO3	22
6	2 equiv of 2a and Cs2CO3, 370 nm, 24 h	84
7	2 equiv of PhBr and Cs2CO3, 370 nm, 24 h	48

^aReaction conditions: 1a (0.2 mmol, 1 equiv), 2a (0.3 mmol, 1.5 equiv), Cs₂CO₃ (0.2 mmol, 1 equiv), Pd­[(o-Tol)₃P]₂Cl₂ (5 mol %), RuPhos (6 mol %), MeCN (2 mL) were irradiated under argon atmosphere with 2 × 40 W at RT. Yields determined by ¹H NMR spectroscopy using mesitylene as internal standard.

With the optimized reaction conditions in hand, we next explored the scope of aryl halide coupling partners (Scheme ). The reaction displayed broad compatibility across a wide range of aryl halides bearing both electron-donating and electron-withdrawing substituents (Scheme a). Electron-rich aryl iodides gave the corresponding coupling products (3b,c, 3e,f) in high yields. In contrast, an ortho- fluoro-substituted aryl iodide gave reduced yield (3d), presumably due to the related slower oxidative addition step or less favorable aryl-radical formation. Substrates bearing reactive functional groups, such as ester, ketone, basic amine, or nitrile, similarly afforded the desired products (3g–j, 3m) in high isolated yields. Unprotected functional groups that typically challenge Pd-catalyzed systemssuch as hydroxyl and amino substituentswere well tolerated, affording the corresponding heterobiaryl products (3k, 3l) in synthetically useful yields without the need for protecting groups. This exceptional functional group tolerance further underlines the mild nature of the photoinduced process. Finally, the scope was extended to a series of azauracil derivatives, which underwent smooth coupling under the standard conditions to furnish the desired products (3n–r) in good yields (Scheme b). In addition, several other heterocycles, such as cinnoline, indole, benzoxazole, quinoxaline and piperazine, successfully underwent C–H arylation, providing the corresponding arylated products (4a–h) in moderate to good yields (Scheme c).

2. Coupling of Aryl Iodides and Bromides .

^a If not stated otherwise, X = I. Reaction conditions: azauracil (0.2 mmol, 1 equiv), aryl halide (0.4 mmol, 2 equiv), Cs₂CO₃ (0.4 mmol, 2 equiv), Pd­[(o-Tol)₃P]₂Cl₂ (5 mol %), RuPhos (6 mol %), MeCN (2 mL) were irradiated under Ar atmosphere with 2 × 40 W at RT. All yields refer to isolated yields.

To further evaluate the versatility of the developed protocol, we next investigated the C–H heteroarylation of heterocyclic scaffolds (Scheme ). This heterobiaryl coupling reaction proved highly general, enabling efficient coupling of structurally and electronically diverse heteroarenes with readily available heteroaryl halides. Pyridine derivatives were efficiently heteroarylated, affording the corresponding heterobiaryl products (5a–5e) in 31–94% yield, with both electron-neutral and electron-withdrawing substituents performing well. Importantly, halide variation (I vs Br) had only a moderate effect on reactivity and we therefore focused our efforts on simple and/or commercially available heteroaryl halides. More electron-rich systems such as methoxy-substituted pyridines also underwent smooth coupling (5e, 42%). An extension to quinoline and isoquinoline derivatives provided the corresponding heterobiaryl products (5h, 52%; 5i, 66%) in good yield.

3. (a) Evaluation of the hetero­(biaryl) Coupling Reaction; (b) Studies on the Compatibility of Drug Molecules .

^a Reaction conditions: see Scheme . *48 h reaction time. All yields refer to isolated yields.

The methodology also proved applicable to other N-heterocycles, including pyrimidines, pyrazoles, pyridazines, and quinoxalines (5j–5t). Substituents such as amide, CF₃, and halogen groups were well tolerated, and both iodinated and brominated heteroaryl halides served as suitable coupling partners. Further examples include the coupling of a carbazole, thiazole, and benzoxazole heterocycle (5n, 5s, 5t). Importantly, nitrogen-containing heterocycles that are notoriously challenging due to their strong coordination tendency toward transition metals, were well tolerated.

To demonstrate the synthetic practicality and late-stage utility of this transformation, we further applied the method to complex molecules derived from nonsteroidal anti-inflammatory drugs (ibuprofen, naproxen), fibrates (gemfibrozil, ciprofibrate), steroids (cholesterol, pregnenolone), phenylalanine, menthol, borneol, and vanillin (6–17). Across this set, arylation with the corresponding iodo-arenes proceeded smoothly, delivering the corresponding products in 28–71% yield without the need for protecting groups. These substrates contain multiple reactive functionalities (free alcohols, carboxylates, phenols, carbamates, enones, and sterically encumbered frameworks), underscoring the method’s functional-group tolerance, mildness, and applicability to medicinally relevant structures.

These results underscore the remarkable generality and functional-group compatibility of this photoinduced palladium-catalyzed C–H (hetero)­arylation reaction, which provides a general strategy for (hetero)­biaryl cross-coupling directly from readily available heterocycles and simple (hetero)­aryl halides and overcomes classical limitations of Pd catalysis using conventional thermal methods.

To elucidate the reaction mechanism, a series of control experiments was conducted (Scheme ). The addition of TEMPO resulted in significantly reduced product formation, and the corresponding TEMPO-aryl adduct was detected by HRMS (Scheme c), confirming the involvement of aryl-radical intermediates. A kinetic isotope effect (KIE) experiment using a deuterated azauracil substrate gave a value of k _H/k _D ≈ 1.0, indicating that C–H bond cleavage is not rate-determining. In light on/off experiments, product formation occurred only under continuous irradiation, demonstrating that persistent photoactivation of the palladium catalyst is required to sustain the reaction (Scheme b). This is further underlined by a quantum yield of Φ = 0.249. A radical clock experiment with 2b provides further evidence of the intermediacy of an aryl radical (for details, see the Supporting Information). Here, palladium-catalyzed radical formation from 2b results a cascade of 1,5-HAT and subsequent cyclopropane ring opening to give the coupling product 19.

4. Combined Computational and Experimental Studies on the Mechanism of the Arylation Reaction .

^a (a) Computational studies on the mechanism (energies in kcal mol^–1). (b) On–off experiment. (c) Control experiments.

On the basis of the control experiments and computational studies, a photoinduced radical-type catalytic cycle is proposed (Scheme a). Under 370 nm light irradiation, photoexcitation of the Pd(0) complex C0 promotes ligand dissociation and allows for the formation of monoligated Pd(0) complex C1. This complex can undergo oxidative addition of the aryl iodide to form a Pd­(II)-aryl species C2. A second photoexcitation, followed by ISC furnishes triplet species ³ C3, which releases the aryl radical intermediate INT1 in a slightly uphill fashion (ΔG = 4.0 kcal mol^–1). Next, the radical addition to 1a takes place via low-lying TS1 with an activation energy of 6.9 kcal mol^–1. The formed radical intermediate INT2 can now engage in two distinct pathways depending on the interplay of INT2 and C4. If both species are in close proximity a barrierless HAT process seems feasible, giving 3a and C5. Otherwise, deprotonation with Cs₂CO₃ furnishes INT3 followed by an SET (ΔG _MH = 1.7 kcal mol^–1) leading to the formation of the desired product 3a and C6. In both cases, the ground-state Pd(0) catalyst C1 is regenerated by Cs₂CO₃, thereby closing the catalytic cycle.

This catalytic cycle is further supported by ³¹P NMR, which show that light is needed in two separate reaction steps: (a) to facilitate ligand exchange of Pd­[(o-Tol)₃P]₂Cl₂ and the external RuPhos ligand (for details, see the Supporting Information) and (b) to facilitate the oxidative addition. The energy required for the transition from ground state to photoexcited state was calculated as 71.6 kcal mol^–1 (transition C0 to C1) and irradiation with light allows for both steps to proceed.

Conclusion

In summary, we have developed a photoinduced palladium-catalyzed C–H (hetero)­arylation that enables the direct synthesis of (hetero)­biaryls from simple (hetero)­aryl halides and heteroarenes under irradiation with 370 nm light. This transformation proceeds under mild conditions without the need of an external photocatalyst and it tolerates a wide range of functional groups and delivers excellent site-selectivity across structurally diverse heterocycles. The ability to engage aryl halides as radical precursors in coupling reactions through photoexcited Pd catalysis expands the reactivity landscape of palladium far beyond conventional two-electron cross-coupling. Mechanistic studies support a hybrid Pd-radical pathway involving light-driven aryl-radical generation and subsequent C–H functionalization. We anticipate that these findings will inspire the design of further light-driven Pd processes for C–C and C–heteroatom bond formation, deepening the connection between radical chemistry and transition-metal catalysis.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.5c01634.

• Description of all experiments, optimization data, copies of all NMR spectra (PDF)

⊥.

S.C. and L.K. are equal contribution. CRediT: Sudip Senapati data curation, formal analysis, investigation, methodology; Sneha Chandra formal analysis, investigation; Lennard Kloene data curation, formal analysis, investigation, visualization; Claudia Poehner data curation, formal analysis, supervision, visualization, writing - review & editing; Claire Empel conceptualization, data curation, project administration, supervision, validation, visualization, writing - original draft, writing - review & editing; Sandip Murarka conceptualization, data curation, formal analysis, funding acquisition, project administration, supervision, validation, visualization, writing - original draft, writing - review & editing; Rene Michael Koenigs conceptualization, data curation, formal analysis, funding acquisition, project administration, resources, supervision, validation, visualization, writing - original draft, writing - review & editing.

The authors declare no competing financial interest.