Upgrading ketone synthesis direct from carboxylic acids and organohalides

The ketone functional group has a unique reactivity in organic chemistry and is associated with a number of useful reactions. Catalytic methods for ketone synthesis are continually being developed. Here, we report a photoredox, nickel and phosphoranyl radical synergistic cross-electrophile coupling of commercially available chemicals, aromatic acids and aryl/alkyl bromides. This allows for concise synthesis of highly functionalized ketones directly, without the preparation of activated carbonyl intermediates or organometallic compounds, and thus complements the conventional Weinreb ketone synthesis. Use of the appropriate photocatalyst, ligand amount and solvents can match the reaction rate required by any simple catalytic cycle. The practicality and synthetic robustness of the reaction are illustrated by the facile synthesis of complex ketones from readily available feedstock chemicals. Due to their abundance and importance in organic chemistry, development of methods for ketone synthesis is essential. Here, the authors report a photoredox, nickel and phosphoranyl radical synergistic cross-electrophile coupling of aromatic acids and aryl/alkyl bromides to directly synthesise ketones.

K etones play a prominent role in organic chemistry. The ketone moiety is extremely common in natural products and pharmaceuticals 1 and in dyes, fragrancies and flavors 2 . It is also a versatile reaction center in organic synthesis 3 . Many frequently used reactions, including the Mannich reaction, Wittig reaction, Grignard reaction, Passerini reaction, Baeyer-Villiger oxidation, and Wolff-Kishner-Huang reduction describe a wide array of transformations of ketones. The development of a practical route to ketones from feedstock chemicals has long been a subject of interest [4][5][6][7][8][9][10][11][12] . Carboxylic acids and organohalides are commercially abundant and structurally diverse, bench-stable feedstock chemicals commonly used in organic synthesis (Fig. 1a). When producing ketones from carboxylic acids and organohalides, the stoichiometric approach requires preparation of necessary intermediates such as amides or aldehydes and Grignard reagents 13,14 . If aldehydes are employed, reoxidation is necessary 15 . Catalytic strategies for production of ketones rely on transition metal-catalyzed carbon-carbon coupling between activated carbonyls such as acid chlorides or anhydrides with organometallic reagents (Fig. 1b) [16][17][18][19] . However, activated carbonyls are generally prepared in as many as three steps from carboxylic acids and organometallics are obtained typically by metalation of organohalides 20 , which can often lead to poor functional group compatibility or a lengthy functional group protection/deprotection process.
In recent years, nickel-catalyzed cross-electrophile coupling has attracted considerable attention [21][22][23][24][25][26][27][28][29][30] . The carbon-carbon coupling arises from two different electrophiles in the presence of stoichiometric reductant. We posited that a carboxylic acid could be directly used as a latent electrophile (Cterminus) rather than a nucleophile (O-terminus) in crosscoupling. This could simplify and upgrade ketone synthesis from carboxylic acids and organic halides 13 . Very recently, our group and Doyle et al. reported an elegant photoredoxpromoted mild deoxygenation of carboxylic acids generating acyl radicals [31][32][33][34][35] . Since photoredox and nickel-catalyzed C-O bond formation between carboxylic acids and aromatic bromides has been reported 36 to achieve the desired crosselectrophile coupling, acyl radical oxidative addition by a metallaphotoredox pathway 37-41 is essential to suppress the C-O bond formation (Fig. 1c). Alternatively, a judicious strategy reported by Gong et al. is conversion of carboxylic acid into anhydrides in situ to suppress the C-O coupling 22,23 . However, the use of free carboxylic acids as acyl radical precursors is a challenge in the oxidative addition step as a result of the strong bond dissociation energy of the C-O bond (106 kcal mol −1 ) 33 .
A proposed mechanism for the designed metallaphotoredox cross-electrophile coupling is shown in Fig. 2 Fig. 2 is a subject of continuing research [46][47][48][49][50][51][52][53][54][55][56][57][58] , and several challenges remain in this cross-electrophile coupling and must be addressed: (1) Efficient control of the matching of the C-O bond cleavage and the radical addition to the nickel center; (2) weakening of the interference of stoichiometric triphenylphosphine in the nickel catalytic unit; and (3) use of appropriate ligands and solvents to significantly suppress the C-O bond formation. Herein, we report a formal cross-electrophile coupling of carboxylic acids with aryl or alkyl halides enabled by photoredox and nickel catalysis, and phosphoranyl radical synergistic chemistry, leading to concise synthesis of ketones (Fig. 1b).

Results
Reaction optimization. Our investigation of this crosselectrophile coupling began with the reaction of 4-methylbenzoic acid (1) with 5-bromo-2-(trifluoromethyl) pyridine (2), and the representative results are presented in Table 1.
The optimized reaction conditions include 2 mol% [Ir{dF(CF 3 ) ppy} 2 {dtbbpy}]PF 6 , 3 mol% NiBr 2 .dme together with 5 mol% 4,4′-di-tert-butyl-2,2′-bipyridine (L1, Fig. 3) and 1.5 equiv Ph 3 P with a mixed DMF-CH 3 CN solvent (entry 1, Table 1). Under the standard conditions, the desired cross-electrophile coupling product (3) can be obtained in 82% yield while the yield of the C-O coupling process, giving 3′ is suppressed to 16%. We found the loading amount of ligand plays an important role in   Table 1). An increased or decreased loading amount of 4,4′-ditert-butyl-2,2'-bipyridine can facilitate the formation of the C-O coupling by-product (3′). Screening of different ligands and solvents indicated that both a ligand effect and a solvent effect are crucial for a successful cross-electrophile coupling (entries 5-11, Table 1). We speculated that only compatible consecutive steps with well-matched rates would benefit this cross-electrophile coupling. In the absence of either photocatalyst, NiBr 2 ·dme, triphenylphosphine or light irradiation, the model reaction failed to occur. The quantum yield of the model reaction was determined to be 0.35, arguing against a radical chain pathway.
Synthetic application. To further demonstrate the synthetic robustness of the reaction, we applied the strategy for the construction of a series of complex ketones from carboxylic acids and aromatic bromides (Fig. 6). Fenofibrate (50) is a drug used to adjust lipid levels and blood viscosity and it could be prepared in one step in 65% yield. The complex ketones (50)(51)(52)(53)(54)(55) can be obtained in synthetically useful yields. The precise crosselectrophile coupling also allows for introduction of functional groups at an early synthetic stage to limit the number of synthetic steps thus improving the efficiency.
Mechanism of stoichiometric reactions. We performed the stoichiometric reactions of Ar-Ni(II) intermediate (56) with 1.5 equiv. Ph 3 P in DMF/MeCN. Interestingly, no ligand exchange was observed by 31 P NMR analysis (Fig. 7, also see Supplementary Fig. 10). Treatment of Ar-Ni(II) intermediate (56) under the photoredox conditions, led to the desired deoxygenative C-C coupling product (35), which was obtained in 42% yield, further supporting the proposed mechanism (Fig. 7).

Discussion
We have developed a cross-electrophile coupling between aromatic carboxylic acids and organic bromides, inexpensive and abundant feedstock chemicals, enabled by photoredox and a nickel and phosphoranyl radical synergistic combination, affording a wide array of structurally diverse ketones with excellent functional group compatibility. This strategy for ketone synthesis can significantly improve the synthetic efficiency and step-economy, and it also opens a door to construct highly functionalized or complex ketones which are still difficult to prepare by a conventional Weinreb ketone synthesis. We found that the use of appropriate ligand loading amount (5 mol% of 4,4′-di-tert-butyl-2,2′-bipyridine), mixed solvents (DMF/CH 3 CN) and combined inorganic bases (K 3 PO 4 and Cs 2 CO 3 ) is crucial to achieve the desired C-C bond formation reactions, affording the desired ketone products. The employment of more or less of ligand results in a sharply decreased yield of the ketone product and an increased yield of the ester by-product. Use of combined bases and mixed solvents would improve the deprotonation of carboxylic acids to expedite the acyl radical generation and use of the precise amount of a ligand would promote the acyl radical oxidative addition to the arylnickel (II) species. We speculated that a facile C-O bond cleavage and subsequent rapid acyl radical oxidative addition rate can control the selective C-C bond formation. We believe this cross-electrophile coupling strategy of carboxylic acids and organic halides will upgrade the synthesis of ketones with great potential application in organic synthesis, drug discovery and optochemical biology given the importance and ubiquity of ketones.
Photocatalyst Ir[dF(CF 3 )ppy] 2 (dtbbpy)PF 6 (4.5 mg, 2 mol%), aromatic carboxylic acid (0.2 mmol, 1.0 equiv), aryl bromide (0.4 mmol, 2.0 equiv), Ph 3 P (78.6 mg, 0.3 mmol, 1.5 equiv), anhydrous powder K 3 PO 4 (42.4 mg, 0.2 mmol, 1.0 equiv), and anhydrous powder Cs 2 CO 3 (65.0 mg, 0.2 mmol, 1.0 equiv) were added to an oven-dried 10 mL Schlenk tube equipped with a magnetic stirring bar. The tube was evacuated and backfilled with argon three times. Subsequently, the nickel-catalyst solution was transferred into this Schlenk tube under argon. The tube was then sealed and placed~5 cm from 2 × 45 W blue LEDs. The reaction mixture was stirred for 20-36 h at room temperature (air-condition was used to keep the temperature is 25°C or so). After completion, the reaction mixture was removed from the light, diluted with water and the aqueous layer was extracted with EtOAc (3 × 2.0 mL). The combined organic layers were washed with brine, dried over anhydrous Na 2 SO 4 , filtered, and concentrated. The residue was purified by flash chromatography on silica gel to afford the corresponding ketone products.  Fig. 4 Carboxylic acid scope at a 0.2 mmol scale under standard conditions. The isolated yield of ketone is given for product and the GC ratio of ketone and ester is given in parenthesis. a The ratio of ketone and ester is calculated based on isolation.