The controlled functionalization of a single fluorine in a CF
3 group is difficult and rare. Photochemical C–F bond functionalization of the sp
3-C–H bond in trifluorotoluene, PhCF
3, is achieved using catalysts made from earth-abundant lanthanides, (Cp
Me4)
2Ln(2-
O-3,5-
tBu
2-C
6H
2)(1-C{N(CH)
2N(
iPr)}) (Ln = La, Ce, Nd and Sm, Cp
Me4 = C
5Me
4H). The Ce complex is the most effective at mediating hydrodefluorination and defluoroalkylative coupling of PhCF
3 with alkenes; addition of magnesium dialkyls enables catalytic C–F bond cleavage and C–C bond formation by all the complexes. Mechanistic experiments confirm the essential role of the Lewis acidic metal and support an inner-sphere mechanism of C–F activation. Computational studies agree that coordination of the C–F substrate is essential for C–F bond cleavage. The unexpected catalytic activity for all members is made possible by the light-absorbing ability of the redox non-innocent ligands. The results described herein underscore the importance of metal–ligand cooperativity, specifically the synergy between the metal and ligand in both light absorption and redox reactivity, in organometallic photocatalysis.The controlled functionalization of a single fluorine in a CF
3 group is difficult and rare. Photochemical C–F bond functionalization is achieved using catalysts made from a range of earth-abundant lanthanides by using a ligand that enables M–L cooperativity.
Photoredox catalysis is a powerful synthetic method for the functionalization of inert molecules using single electron transfer (SET) reactivity
1–3 under irradiation with visible light.
4 This has enabled challenging transformations under mild conditions including C–H activation,
5–7 radical cross-coupling,
8–11 and the valorization of lignin.
12,13 However, detailed mechanistic studies of photoredox systems are difficult due to their inherent complexity and the short lifetimes of photoexcited intermediates.Many lanthanides are more abundant in the environment than copper and their salts are less toxic than those of iron, so their potential for applications in catalysis merits exploration.
14–17 In 1990, divalent Sm, Eu, and Yb complexes Ln(Cp*)
2 (Cp* = C
5Me
5), were shown to more efficiently cleave vinylic C–F bonds when photolyzed, stoichiometrically forming Ln(
iii) halide complexes, and suggesting the value of increasing the reducing power of the Ln
II excited state.
18 Subsequently, analogous reactions to cleave the weaker C–Cl and C–Br bonds could be made catalytic in Ln(
ii) halide (Ln = Sm, Eu, Yb), under near UV-photolysis conditions, by the addition of sacrificial reductant such as Zn or Al.
19,20 The addition of simple donor ligands enabled benzylic C–Cl cleavage by Eu
II under blue light irradiation.
21 The addition of an organic photocatalyst or a photo-absorbing substrate to Lewis acidic LnX
3 salts (X = halide, triflate) has also been used to enhance the catalysis.
22 Ln centers (Ln = Nd, Dy, Lu) with light-absorbing ligands such as porphyrins or phthalocyanins have been used to stoichiometrically dechlorinate phenols.
23Few reports of lanthanide photoredox catalysis exist with Ce
III complexes receiving the most attention. Ce possesses both an accessible III/IV redox couple and an allowed excitation from the 4f
1 ground state to the 5d
1 excited state, which can give rise to luminescent behaviour. It is also the cheapest and most readily isolated of the rare earths, offering a promising alternative to current precious metal photocatalysts.Building on the pioneering work on stoichiometric photoluminescent Ce chemistry,
24,25 in 2015 Schelter and co-workers demonstrated the utility of Ce
III in photocatalysis.
26,27 Their Ce
III amido complexes were catalysts for chlorine atom abstraction from benzyl chloride (, top), with both NaN(SiMe
3)
2 and additional Ce
0 required for turnover.
28Open in a separate windowPrevious examples of photocatalytic C–X (X = halide) bond cleavage, and this work.They proposed an inner-sphere mechanism involving Ce⋯ClCR
3 adduct formation that provides an additional thermodynamic driving force to a bond cleavage that was otherwise out of range of the reducing power of the Ce excited state. A more sterically congested Ce
III tris(guanidinate) operates
via an outer-sphere single electron transfer (SET) mechanism to cleave aryl iodides,
28 highlighting the mechanistic diversity that is possible in these systems.
29To date, ligands that support lanthanide-centered photocatalysts have been limited to halides, pseudohalides, and simple N-donors.
30,31 No organometallic lanthanide photocatalyst has yet been reported that combines the photoexcitable Ce cation with multidentate, tunable ligands. We have developed organometallic lanthanide complexes as sustainable catalysts,
16,17,32,33 and considered that those capable of forming an inner-sphere adduct, and absorbing light, could achieve the unusual and difficult, selective catalytic conversion of strong sp
3 C–F bonds.Fluorine forms the strongest single bond to carbon and the C–F bond is
ca. 25 kcal mol
−1 stronger than the C–Cl bond in monohaloalkanes, and the C–H bond in alkanes.
34 The selective activation and functionalization of C–F bonds is important, both due to the high bioaccumulation and toxicity of many perfluorinated compounds,
35 and the utility of fluorinated pharmaceuticals.
36 However, stoichiometric C(sp
3)–F bond activation reactions are rare.
37–40 In particular, it is difficult to facilitate the controlled cleavage of a single C–F bond as the C(sp
3)–F bond strength decreases as each F is removed and the remaining C–F bonds lengthen.
41,42This obstacle makes a radical methodology more attractive.
43–49 Jui and co-workers have demonstrated that some common photocatalysts can selectively activate a single C–F bond to form the putative ArCF
2˙ radical, which can either be quenched directly
via H atom transfer (HAT), or coupled with an alkene followed by HAT to generate difluoroalkanes (, middle).
50,51 Gschwind and König have shown the photochemical functionalization of electron-poor trifluoromethylarenes.
52 Nishimoto and Yasuda have described related C–F coupling protocols of perfluoroalkylarenes using tin reagents and an iridium photocatalyst.
53Here we show how selective, catalytic C–F bond functionalization can be achieved using a new family of Ln
III compounds supported by a light-absorbing aryloxide-tethered N-heterocyclic carbene, Cp
Me4, and pseudohalide ligands (, lower). We show that visible light-irradiated Ce complexes can selectively abstract a single fluoride from PhCF
3 and catalyze its alkylation by MgR
2 to afford PhCF
2R. The PhCF
2˙ can also be quenched to selectively form PhCF
2H or further alkylated
via coupling with an alkene or other metal alkyls. We use combined experiment and density functional theory (DFT) computations to show the importance of coordination of the fluorinated substrate to the Lewis acidic metal in C–F activation, and the utility of the ligand in enabling photoredox catalysis for other lanthanide congeners.
相似文献