Low-Energy Light-Driven Excited-State Palladium Catalysis: Cross-Coupling of Pyridine N‑Oxides with Unactivated Alkyl Bromides Authors: Jisun Kim, Minyoung Ju, Chaerin Baek, Eun Joo Roh, Jeongcheol Shin, Jung Tae Han Published: 2026 ABSTRACT Excited-state Pd catalysis has emerged as a new platform for cross-coupling reactions. Nonetheless, such catalysis typically requires high-energy near-ultraviolet (NUV) or blue (380–450 nm) light. In contrast, the use of lower-energy light (>520 nm) in excited-state Pd catalysis remains largely unexplored. Herein, we report a low-energy light-driven excited-state Pd-catalyzed cross-coupling of pyridine N-oxides with unactivated alkyl bromides. Mechanistic studies reveal the formation of a Pd(0)–pyridine N-oxide complex that absorbs low-energy light. Pd-catalyzed cross-coupling reactions are fundamental transformations in organic synthesis, enabling the construction of C–C, C–N, C–O, C–X, and other C–heteroatom bonds. Traditionally, these reactions are initiated by ground-state Pd catalysts and proceed through two-electron redox processes. Recently, excited-state Pd catalysts have emerged and found applications in cross-coupling reactions that proceed through single-electron transfer (SET) processes. Representative examples include the Mizoroki–Heck reaction, the Negishi reaction, the Suzuki–Miyaura reaction, Miyaura borylation, carbonylation, and C–H functionalization. However, these reactions typically require high-energy near-ultraviolet (NUV) or blue light (380–450 nm) and thus suffer from issues such as limited scalability, functional group incompatibility, and competitive photoexcitation of substrates (Scheme a). Alternatively, the use of lower-energy green, orange, or red light (520–740 nm) could address the issues associated with NUV or blue light; however, such light lacks the energy to excite the ground-state Pd precatalyst (Scheme b). Despite this significant challenge, Iwasawa has recently demonstrated that low-energy light-driven excited-state Pd catalyzes a C–O cross-coupling reaction, albeit with only a single example in which 4-tert-butyl iodobenzene is coupled with cesium pivalate (Scheme c). Overview of Excited-State Pd Catalysis Alkylation of pyridine N-oxides provides efficient access to alkylated pyridine N-oxides and pyridines, which are found in many biologically active molecules and pharmaceuticals. Thus, numerous catalytic and stoichiometric methods have been developed for the alkylation of pyridine N-oxides. However, only a few methods employing Pd catalysts have been reported, despite significant advances in Pd-catalyzed cross-coupling reactions. In 2013 and 2014, Fu and Zhou independently reported Pd-catalyzed cross-coupling of pyridine N-oxides with unactivated alkyl bromides and iodides under thermal conditions (100–110 °C). More recently, Yu and Fu independently demonstrated that blue light-driven excited-state Pd catalyzes the same transformation, although they reported only a single example in which 2-methylpyridine N-oxide reacts with 1-adamantyl bromide. , Inspired by prior work by Iwasawa, we hypothesized that in the presence of an appropriate ligand, a ground-state Pd precatalyst could be photoexcited by low-energy light and the resulting excited-state Pd catalyst could promote the C–C cross-coupling reaction. We herein report a low-energy light-driven excited-state Pd-catalyzed cross-coupling of pyridine N-oxides with unactivated alkyl bromides (Scheme d). Mechanistic studies reveal that pyridine N-oxide substrates act as ligands to Pd(0), forming Pd(0)–pyridine N-oxide complexes that absorb low-energy light. Our investigation began with the cross-coupling of 2-methylpyridine N-oxide (1a) with cyclohexyl bromide (2a) (Table ). Following an extensive study of reaction parameters, we found that a combination of Pd­(PPh3)4 (10 mol %), Cs2CO3 (1.5 equiv), and TBAB (1.5 equiv) in THF at 25 °C under green light (525 nm) irradiation without an external ligand provides product 3aa in 60% yield (entry 1). The use of Pd­(OAc)2 in conjunction with PPh3 slightly decreased the yield of 3aa (entry 2); the use of Pd­(PPh3)2Cl2 delivered 3aa in only 8% yield (entry 3). Notably, external ligands (BINAP, Xantphos, DPEphos, and DPPF) resulted in lower yields (entries 4–7, respectively). Amine bases such as diisopropylethylamine (DIPEA) showed reduced performance (entry 8). Altering the light sources led to either decreased yields or no product formation (entries 9–12). Whereas cyclohexyl iodide afforded a yield comparable to that of cyclohexyl bromide (2a), cyclohexyl chloride was shown to be unreactive (entries 13 and 14). In the absence of Pd­(PPh3)4 or light, 3aa was not formed (entry 15 or 16, respectively); in the absence of TBAB, which has been shown to enhance the reactivity in blue light-driven excited-state Pd catalysis, the yield of 3aa decreased slightly (entry 17). The role of TBAB in this transformation is unclear at this time. When the reaction was conducted under an ambient atmosphere, no product was formed, presumably due to quenching of the active Pd catalyst by oxygen (entry 18). Lastly, increasing the Pd­(PPh3)4 loading to 15 mol % allowed product 3aa to be obtained in 75% yield (entry 19). 1Optimization, aReactions were conducted on a 0.2 mmol scale.bYields were determined by 1H NMR analysis of the unpurified reaction mixture with dibromomethane as an internal standard. Optimization, Reactions were conducted on a 0.2 mmol scale. Yields were determined by 1H NMR analysis of the unpurified reaction mixture with dibromomethane as an internal standard. With the optimized reaction conditions in hand, we investigated the scope of unactivated alkyl bromides (Scheme ). Cyclic and acyclic secondary alkyl bromides reacted efficiently to provide the desired products (3aa–3ag). Remarkably, when exo-2-bromonorbornane (2e) was employed, complete diastereoselectivity was observed. In contrast to secondary alkyl bromides, primary and tertiary alkyl bromides tended to furnish the products in lower yields (3ah and 3ai). We next examined the scope of pyridine N-oxides. 2-Phenylpyridine N-oxide (1b) was also a suitable substrate for the transformation, but in this case, 1,4-dioxane was needed as the solvent. When quinoline N-oxide (1c) was subjected to the optimized reaction conditions, only a trace of the desired product was observed. After extensive reoptimization, we were pleased to find that the reaction required BINAP as an external ligand to provide the desired product in 76% yield. Under these modified reaction conditions, quinoline N-oxide derivatives (1d–1g) were transformed into the corresponding products in satisfactory yields (70–96%). Finally, we sought to use lower-energy light in our catalyst system. Under orange and red light irradiation, quinoline N-oxide (1c), lepidine N-oxide (1d), and benzo­[h]­quinoline N-oxide (1g) were all reactive. These results are notable because low-energy orange or red light-driven excited-state transition-metal catalysis has rarely been reported. Finally, an 8 mmol scale reaction was conducted, and 0.93 g of product 3aa was obtained in 61% yield (see the Supporting Information for details). Substrate Scope aReactions were conducted on a 0.2 mmol scale. Yields were determined by 1H NMR analysis of the unpurified reaction mixture with dibromomethane as an internal standard. Yields in parentheses are those of isolated products. bFor 24 h. cFor 48 h. d1,4-Dioxane was used as a solvent instead of THF. eWith 15 mol % BINAP used as a ligand. fThe reaction was conducted under an orange LED (595 nm). gThe reaction was conducted under a red LED (660 nm). Reactions were conducted on a 0.2 mmol scale. Yields were determined by 1H NMR analysis of the unpurified reaction mixture with dibromomethane as an internal standard. Yields in parentheses are those of isolated products. For 24 h. For 48 h. 1,4-Dioxane was used as a solvent instead of THF. With 15 mol % BINAP used as a ligand. The reaction was conducted under an orange LED (595 nm). The reaction was conducted under a red LED (660 nm). To gain mechanistic insights into the low-energy light-driven excited-state Pd-catalyzed cross-coupling reaction, we carried out radical trapping and control experiments (Scheme ). When 2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO), a radical trap, was subjected to the standard reaction conditions, the formation of product 3aa was inhibited and TEMPO adduct 4 was detected by electrospray ionization mass spectrometry (ESI-MS) (Scheme a). This is consistent with our hypothesis that the reaction involves alkyl radical intermediates. Under otherwise identical conditions in the absence of 1a, TEMPO adduct 4 was not formed; however, upon blue light irradiation, TEMPO adduct 4 was detected (Scheme b). These results indicate that 1a is essential for the low-energy light-driven Pd-catalyzed cross-coupling reaction and suggest the formation of a putative Pd(0)·1a complex that absorbs low-energy light. Radical Trapping and Control Experiments Next, we investigated the formation of the Pd(0)·1a complex by UV–vis–NIR absorption spectroscopy (Figure ). None of the individual reaction components [1a, 2a, Pd­(PPh3)4, DIPEA, and TBAB] absorbed at around the operational wavelength (525 nm) (Figure a). In contrast, upon mixing Pd­(PPh3)4 with 20 equiv of 1a, a diagnostic absorption at 505 nm was observed, providing evidence that the Pd(0)·1a complex absorbs green light (Figure b). However, addition of 2a, DIPEA, and TBAB resulted in an only negligible bathochromic shift. To assign the observed absorption, time-dependent density functional theory (TD-DFT) calculations were conducted. First, the absorption spectra of Pd­(PPh3)4, Pd­(PPh3)3, and Pd­(PPh3)2 were simulated, as PPh3 is known to reversibly dissociate from Pd­(PPh3)4 in THF. However, none of these species exhibited absorbance above 500 nm (Figure c). Further calculations revealed that Pd­(PPh3)4 and 1a associate to form a linear Pd­(PPh3)·1a complex with a highly exergonic ΔG (−16.6 kcal/mol) (Figure d,e). Importantly, the simulated absorption spectrum of the Pd­(PPh3)·1a complex was in close agreement with the experimental spectrum (Figures b,c). The possibility that the Pd­(PPh3)3·1a or Pd­(PPh3)2·1a complex is the green light-absorbing species was ruled out, as geometry optimization resulted in spontaneous dissociation of 1a from the Pd complex. Frontier molecular orbital (FMO) analysis of the Pd­(PPh3)·1a complex suggests that the low-energy light absorption corresponds primarily to a metal-to-ligand charge transfer (MLCT) from the Pd 4d z 2 -based MO to the 1a π*-based MO (Figure f). The resulting excited-state species, Pd­(I)­(PPh3)­(1a •–), could promote subsequent steps such as halogen atom transfer (XAT) of 2a or C–H activation of 1a. Experimental absorption spectra of (a) individual reaction components and (b) their mixture. (c) TD-DFT-simulated absorption spectra of Pd complexes. DFT-calculated (d) formation thermodynamics and (e) geometry of the Pd­(PPh3)·1a complex. (f) Frontier molecular orbital (FMO) analysis of the Pd­(PPh3)·1a complex. On the basis of radical trapping and control experiments, UV–vis–NIR absorption studies, and TD-DFT calculations along with generally accepted mechanisms for excited-state Pd catalysis, ,,, a catalytic cycle for this transformation is proposed (Scheme ). The association of Pd(0) with pyridine N-oxide 1 forms Pd(0)–pyridine N-oxide complex I, which is photoexcited to I* under low-energy light irradiation. Subsequent XAT of unactivated alkyl bromide 2 to I* generates hybrid Pd­(I) alkyl radical II. Radical insertion of II into 1 then affords hybrid Pd­(I) nitrogen-centered radical III. Finally, single-electron transfer of III followed by deprotonation provides product 3 and regenerates I. Alternative mechanisms such as preferential C–H activation of 1 with I* over the XAT process cannot be ruled out at this time. Proposed Catalytic Cycle In conclusion, we have developed a low-energy light-driven excited-state Pd-catalyzed cross-coupling of pyridine N-oxides with unactivated alkyl bromides. This catalytic protocol exhibits a broad substrate scope, providing alkylated pyridine N-oxides in moderate to high yields. Mechanistic studies reveal that the reaction proceeds through the formation of a low-energy light-absorbing Pd(0)–pyridine N-oxide complex. The isolation and characterization of this complex are the subjects of ongoing research in our laboratory. Supplementary Material