Dual visible-light and NHC-catalyzed radical relay trifunctionalization of unactivated alkenes

visible-light and NHC-catalyzed


INTRODUCTION
The assembly of molecular frameworks by multi-step relay carbon-carbon bond cleavage and formation in a controlled and efficient manner has been at the heart of organic synthesis.Alkenes are readily available and inexpensive feedstocks that have been widely used for that purpose.Difunctionalization of alkenes presents numerous mature methods for accessing diverse-ranging molecules with high added value and structural complexity [1][2][3][4][5][6] .In comparison, the trifunctionalization of alkenes, particularly unactivated alkenes, is less − Scheme 1. Radical-mediated trifunctionalization of unactivated alkenes.FGM: Functional group migration; NHC: N-Heterocyclic carbene.
explored and still poses a challenge.Recently, the appearance of radical functionalization of unactivated alkenes through the strategy of remote functional group migration (FGM) [7][8][9][10] provides an exceptional chance for the construction of trifunctionalized derivatives that are not accessible by other reactions [Scheme 1A].Therefore, the discovery of new protocols for FGM reactions remains an important target and continues to be highly desirable for molecular assembly.
As a continuation of our studies on radical NHC-catalysis, our purpose herein is to realize the first reductive radical organocatalytic FGM reactions to fulfill trifunctionalization of hexenenitriles 3 [70,71] via remote cyano migration [72][73][74][75] .Compared to the previous work using the NHC oxidative radical strategy, the substrate scope can enlarge to in-stock carboxylic acids and much more radical precursors.Thus, this protocol features a compatible dual catalytic system, mild reaction conditions, readily available substrates, excellent

RESULTS AND DISCUSSION
To start this work, commercially available sodium trifluoromethanesulfinate 1a was used as the trifluoromethyl radical precursor to investigate the feasibility of the reaction with 4-chlorobenzoic acid 2a and hexenenitrile 3a using DCE as the solvent, Cs 2 CO 3 as a base, and Blue LEDs as light sources [Figure 1].It is worth noting that the acid can be activated to produce benzoylimidazole 2a' in situ with 1,1'-carbonyldiimidazole (CDI).The combination of NHC catalyst and photocatalyst was crucial for this transformation with otherwise trace formation of product 4a (entries 1 and 2).
Other NHCs B-D did not improve the reaction yields (entries 8-10).The nitrogen atmosphere protection was also essential for this reaction.The yield dropped to 45% when the reaction was conducted in the air (entry 11).Moreover, a scale-up (1 mmol) reaction was taken to afford product 4a in maintained yield (entry 12).Remarkably, only a trace of byproduct 4a' was observed along the optimization process.At last, bases were screened, and a slightly decreased yield was observed with K 2 CO 3 .Organic bases were not suitable for this transformation.
With the optimal condition in hand, the generality of substrates was explored.We initially tested different acids 2 [Scheme 2].Substituted benzoic acids bearing either halide or electron-donating groups at para positions of the benzoic acid were well compatible with the optimal condition, giving the products 4a-e in comparable yields.Slightly decreased yields were tracked when benzoic acids bearing electron-withdrawing groups 4f-h.Benzoic acids bearing meta substitution were also well tolerated under the standard conditions 4i-j.However, the more sterically hindered 2-methylbenzoic acid was not suitable for this reaction.Either 1-naphthoic acid or 2-naphthoic acid was applicable for the optimal condition 4k-l.A decreased yield of 1-naphthoic acid was probably due to the steric effect.This protocol could also accommodate a variety of heterocyclic aromatic acids (4m-q).The reaction of picolinic acid almost did not work, perhaps due to the electronic effect of this substrate.To demonstrate the synthetic potential of this protocol, the late-stage functionalization of benzoic acid derived from menthol was tested, which worked smoothly to afford the corresponding product 4af in 52% yields.After that, the scope of hexenenitriles was screened.In terms of substituents (Y) at the phenyl ring, the CN migration products 4r-4z were exclusively produced in satisfying yields and excellent selectivity.The substituents on the phenyl ring have little impact on the yield.Even the more sterically hindered hexenenitriles were found to be suitable substrates for cyano migration, furnishing the expected products 4y and 4z, both in a 67% yield.
The feasibility of installation of other radical precursors to the hexenenitrile 3d was also studied [Scheme 2, bottom].Several sulfinate salts bearing a fluorinated alkyl group displayed good compatibility with the present system 4aa-ad.Sulfinate salts bearing a phenylsulfonyl substituted methyl group were also used as effective substrates, enabling the trifunctionalization of 3d to give product 4ae in an acceptable yield.However, sulfinate salts, such as sodium methanesulfinate, sodium ethanesulfinate, and other listed aliphatic sulfinates, did not deliver the desired products.Density functional theory (DFT) calculations were then taken into account for the possible reasons [Scheme 3].To our interests, the activating energy of the corresponding radical added to the hexenenitrile significantly contributes to the success of the reaction.The reactions failed to yield the desired product when the activating energy was above 14.1 Kcal/mol.These findings might be helpful for prediction of the reactivity of other sulfinate salts.
To highlight the utility of this transformation, we undertook the derivatization of the ketone 4a [Scheme 4].The alcohol 5 was synthesized using LiAlH 4 as redundant in the yield of 83%.The cyano group within 4a can transform to amide to yield compound 6 with a mixed acid system (HOAc and H 2 SO 4 ).In addition, the reaction of 4a with hydrazine could form hydrazone 6 in 80% yield.
Next, DFT calculations have been conducted to further prove the potential reaction mechanism and origin of the regioselectivity for 1,4-cyano migration [Figure 2].facilitating the cyano migration process.Therefore, high regioselectivity of this protocol can be rationalized by these calculation results.
Based on these results, it can be concluded that the dual photocatalytic and NHC-promoted SET between sodium trifluoromethanesulfinate 1a and acyl azolium intermediate SA could simultaneously produce the NHC-derived radicals C3 and CF 3 radicals.

CONCLUSIONS
To sum up, we have described a novel protocol for trifunctionalization of unactivated hexenenitriles via merged NHC organocatalytic and photocatalytic radical relay alkylacylation.This reaction offers a generalizable and efficient strategy for molecular framework assembly, which involves multi-step C−C bond cleavage and formation by remote 1,4-cyano migration and alkylacylation.Theoretical calculations were used to both support the possible mechanism and figure out the origin of the regioselectivity for 1,4-CN migration.

Figure 2 . 5 .
Figure 2. Relative Gibbs free energy profiles of the reaction.
CF 3 radicals were subsequently triggered by hexenenitriles 3 to obtain S2.The generated radical intermediate S2 may undergo remote 1,4-cyano migration via a cyclic intermediate S3 to yield radicals S4.The radical-radical cross-coupling with the persistent ketyl radical C3 and S2 or S4 would produce intermediate S5 and S7, respectively, which finally afford adducts 4a and byproduct 4a' with the loss of NHC for the next catalytic cycle [Scheme 5].