Hongping
Zhao
a and
Weiming
Yuan
*abc
aKey Laboratory of Material Chemistry for Energy Conversion and Storage, Ministry of Education, Hubei Key Laboratory of Bioinorganic Chemistry and Materia Medica, School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology (HUST), 1037 Luoyu Road, Wuhan 430074, PR China. E-mail: yuanwm@hust.edu.cn
bShenzhen Huazhong University of Science and Technology Research Institute, Shenzhen 518000, PR China
cGuangdong Provincial Key Laboratory of Catalysis, Southern University of Science and Technology, Shenzhen 518055, PR China
First published on 10th January 2023
A three-component reductive cross-coupling of aryl halides, aldehydes, and alkenes by nickel/photoredox dual catalysis is disclosed. The key to success for this tandem transformation is to identify α-silylamine as a unique organic reductant, which releases silylium ions instead of protons to prevent unwanted protonation processes, and meanwhile serves as Lewis acid to activate aldehydes in situ. This dual catalytic protocol completes a traditional conjugate addition/aldol sequence that eliminates the requirement of organometallic reagents and metal-based reductants, thus providing a mild synthetic route to highly valuable β-hydroxyl carbonyl compounds with contiguous 1,2-stereocenters.
Aldehydes, in particular, are abundant and versatile building blocks in organic synthesis, while their application in catalytic reductive olefin DCF has been much less explored. An early example is from the Montgomery group who reported Ni(cod)2-catalyzed three-component reaction of acrylates, aryl iodides and aldehydes under net reductive conditions with a stoichiometric amount of ZnMe2 as the reductant (Scheme 1a).4 Later on, Gall and co-workers reported a similar transformation on using CoBr2 as the catalyst and Zn dust as the reductant.5 These reductive processes provide a complementary approach to a classical conjugate addition/aldol tandem sequence6 that requires sensitive arylmetallic reagents. However, the requirement of stoichiometric amounts of metal-based reductants inevitably produces a lot of metal salt waste and results in a certain degree of difficulty for synthetic scalability and practicability.
Taking advantage of the ability of the photoredox-assisted single-electron transfer (SET) process, researchers have recently applied Ni/photoredox dual catalysis7 to the area of cross-electrophile coupling reactions,8 enabling the coupling of organohalides in the absence of a stoichiometric metal reductant. However, most reported photo-redox-assisted reductive cross-couplings are limited to the two-component version9 with few exceptions have been extended to the intermolecular three-component reaction of aryl and alkyl halides across alkenes.10 Giving our longstanding interest in metallaphotoredox catalysis,9e,11 we reported here a Ni/photoredox-catalysed three-component conjunctive cross-electrophile coupling of aryl halides with aldehydes and alkenes (Scheme 1b). We envisioned that the key to success of this tandem transformation is to identify a suitable organic reductant as any organic reductants release protons might be problematic because they can protonate each tentative C–Ni bond species, thus producing unwanted outcomes such as a dehalogenation product and reductive Heck product. As such, the most commonly used organic reductants in literature including tertiary amines and Hantzsch esters (HEs) would be inferior to the reaction because they can release protons when using as reductant. To address this issue, we envisioned that α-silylamines might be a sort of valid organic reductant to furnish this tandem process as they release silylium ions instead of protons during the single electron oxidation stage. Moreover, the in situ generated silylium ion could act as a Lewis acid to activate aldehydes to promote the subsequent aldol reaction. α-Silylamines have been widely used as α-amino radical precursors in photocatalytic radical transformations,12 while their potential as mild organic reductants in reductive cross-couplings has not yet been explored. Here, we disclose the unique functions of α-silylamine in reductive conjugate addition/aldol tandem transformation.
Entryb | Photocatalyst | [Ni] | Ligand | Solvent | GC yielda | ||
---|---|---|---|---|---|---|---|
4 | 5 | 6 | |||||
a Ir[dF(CF3)ppy]2(dtbbpy)PF6 (1 mol%), NiBr2·DME (10 mol%), and 6,6′-di(Me)bpy (10 mol%). Yields are determined by GC with n-tridecane as the internal standard. b Reaction conditions: photocatalyst (1 mol%), Ni catalyst (10 mol%), and ligand (10 mol%). c The molar ratio of 1a:2a:3a = 2:2:1. d No light irradiation. e 4-Bromo-1,1′-biphenyl instead of 1a. f [1,1′-biphenyl]-4-yl trifluoromethanesulfonate instead of 1a. | |||||||
1 | Ir(ppy)3 | NiBr2·DME | L1 | DMAc | 34 | 23 | 21 |
2 | Ir(ppy)2bpyPF6 | NiBr2·DME | L1 | DMAc | 21 | 11 | 43 |
3 | Ru(bpy)3(PF6)2 | NiBr2·DME | L1 | DMAc | 32 | 16 | 5 |
4 | 4-CzIPN | NiBr2·DME | L1 | DMAc | 15 | 0 | 62 |
5 | Ir(ppy)3 | NiBr2·DME | L2 | DMAc | 62 | 20 | 5 |
6 | Ir(ppy)3 | NiBr2·DME | L3 | DMAc | 44 | 22 | 1 |
7 | Ir(ppy)3 | NiBr2·DME | L4 | DMAc | 19 | 9 | 3 |
8 | Ir(ppy)3 | NiBr2·DME | L5 | DMAc | 21 | 8 | 4 |
9 | Ir(ppy)3 | NiBr2·DME | L6 | DMAc | 24 | 16 | 51 |
10 | Ir(ppy)3 | NiBr2 | L2 | DMAc | 45 | 13 | 2 |
11 | Ir(ppy)3 | Ni(cod)2 | L2 | DMAc | 55 | 4 | 10 |
12 | Ir(ppy)3 | NiBr2·DME | L2 | DMF | 3 | 0 | 4 |
13 | Ir(ppy)3 | NiBr2·DME | L2 | MeCN | 24 | 43 | 2 |
14 | Ir(ppy)3 | NiBr2·DME | L2 | THF | 0 | 3 | 0 |
15 | Ir(ppy) 3 | NiBr 2 ·DME | L2 | DMAc | 78 | 43 | 0 |
16 | — | NiBr2·DME | L2 | DMAc | 0 | 0 | 0 |
17d | Ir(ppy)3 | NiBr2·DME | L2 | DMAc | 0 | 0 | 0 |
18 | Ir(ppy)3 | — | L2 | DMAc | 0 | 0 | 0 |
19 | Ir(ppy)3 | NiBr2·DME | — | DMAc | 44 | 13 | 0 |
20e | Ir(ppy)3 | NiBr2·DME | L2 | DMAc | 22 | 4 | 50 |
21f | Ir(ppy)3 | NiBr2·DME | L2 | DMAc | 15 | 6 | 8 |
With the optimized reaction conditions in hand, we started to evaluate the scope generality of the three-component reaction (Table 2). First, a series of aryl iodides were examined. Aryl iodides with electron-donating and electron-withdrawing groups including phenyl (4), OMe (9), ester (11), ketone (12), SMe (13), and Cl (14) were well tolerated, delivering the corresponding products in moderate to good yields. A sterically hindered substrate such as ortho-methyl substituted iodobenzene (15) also reacted with a good yield. Heteroaryl iodide was tolerated as well (16). The structure of the trans-configuration of product 4 was determined by X-ray diffraction analysis. Regarding the scope of alkenes, except for different substituted acrylates (4, 17, and 18), various N-substituted acrylamides were investigated. Acyclic acrylamides (19, 20, 25, and 26) and cyclic acrylamides (21–24) all reacted smoothly and afforded a series of structurally abundant β-hydroxyl amide products. Particularly, α-methyl substituted methyl acrylate was also tolerated, enabling a product with an all-carbon quaternary stereocenter (27). Finally, the scope generality of aldehydes was evaluated. Again, not only electron-donating groups, but electron-withdrawing groups were also compatible (28–34). Sterically ortho-substituted benzaldehydes have no impact on the reactivity (35, 36). Of note, more reactive naphthaldehyde (37), furfural (38) and 2-thenaldehyde (39) successfully underwent conjugative coupling. Pleasingly, aliphatic aldehydes irrespective of containing an α-primary, or secondary, or tertiary alkyl substituent were smoothly incorporated (40–43). Both structures of trans- and syn-configuration of product 42 were determined by X-ray diffraction analysis. Regrettably, the reaction could only give a poor stereocontrol with diastereoisomeric ratios ranging from 1:1 to 2:1. The low level of diastereoselectivity might have contributed to the poor selectivity of the formed Z- vs. E-enolate in the conjugate addition step or the absence of a Zimmerman–Traxler-type transition state in the second aldol reaction step.13 We also added some metal-based Lewis acids such as ZnCl2 and MgCl2 to improve diastereoselection, but no further improvement was obtained. Anyway, both diastereoisomers can be separated by column chromatography on silica gel. The reaction could be scaled-up to 1 mmol (60%) and 3 mmol (48%) with an acceptable yield obtained.
In order to gain insight into the reaction mechanism, some control experiments were conducted. The reaction was completely inhibited by adding radical scavenger TEMPO [(2,2,6,6-tetramethylpiperidin-1-yl)oxyl] (Scheme 2(1)). Besides, all reactions failed to deliver any desired product when various different fluoride sources added (Scheme 2(2)). We speculated that the in situ generated silylium ion species (see the mechanistic discussion) was trapped by a fluoride source to form the useless TMSF. These interesting results indicate the significant importance of silyl groups for this reaction. We deduced that the TMS group plays at least two roles: (1) may serve as a Lewis acid to activate a carbonyl group for aldol reaction; (2) re-generate the active nickel catalyst via transmetallation of the O–Ni bond to the stronger O–Si bond. A control experiment without aldehyde was performed where no three-component conjunctive addition product was detected, and only a large amount of reductive Heck product 5 was afforded (Scheme 2(3)), suggesting that an α-amino radical might not be generated11a under this condition. The intermediate 5 may possibly react with aldehyde in the presence of a base under thermal conditions to get the final product. However, the reaction cannot produce 4, thus excluding this possibility (Scheme 2(4)). To better understand the behavior of the photoexcited step, Stern–Volmer fluorescence quenching experiments14 were performed (see the ESI† for details). The results revealed that the oxidative quenching of the excited state of the photocatalyst by the nickel(II) catalyst is much more efficient over other components (Scheme 2(5)), suggesting that a single-electron-transfer (SET) event occurred between the nickel(II) catalyst and *Ir(ppy)3 that initiates the photocatalytic process. To exclude the possibility of the nickel complex affecting the emission intensity of the photocatalyst, the absorption spectra of each component were recorded and the results showed that all components have no absorption in this region (Scheme 2(6)).
Based on the above mechanistic studies and our previous reports,11 a plausible mechanism was proposed for this reductive conjugate addition/aldol tandem sequence (Scheme 3). First, successive SET process between the photoexcited *IrIII (EIV/*III1/2 = −1.73 V vs. SCE)15 and NiII [E1/2(NiII/Ni0) = −1.2 V vs. SCE]16 generates Ni0 and IrIV. IrIV possessing a strong reductive potential (EIV/III1/2 = +0.77 V vs. SCE)15 can oxidize α-silylamine (Eox1/2 = +0.71 V vs. SCE)17,18 to get iminium ion by successive SET. Meanwhile, IrIV is reduced back to the ground state of IrIII. On the other hand, Ni0 species undergoes oxidative addition with ArX to afford ArNiIIX, followed by migration insertion into CC bond to get α-carbonyl-NiII intermediate (IV). An equilibration between nickel O-enolate (V) and C-enolate (IV) tautomer is possible,19 which could be further transformed into a more stable silyl ketene acetal (VI) via a Ni/Si transmetallation. Final aldol reaction promoted by Lewis acidic silylium ion delivers products and regenerate NiII catalyst.
Footnote |
† Electronic supplementary information (ESI) available: Experimental procedures, characterization data, and copy of NMR spectra. CCDC 2217184, 2217195 and 2217198. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d2sc06303d |
This journal is © The Royal Society of Chemistry 2023 |