Photoredox catalysis has been considered a highly effective strategy for generating active radical intermediates under mild conditions over the past few decades, causing widespread attention from synthetic chemists [1-7]. Currently, precious transition-metal catalysts such as iridium and ruthenium play important roles in photocatalytic organic reactions [8-13]. However, these precious transition-metal catalysts have some boundedness, such as high cost and sensitivity to oxygen, which might limit their application in organic synthesis. In this context, the advancement of cost-effective photocatalytic systems has been an enduring objective within the field of photocatalytic organic chemistry.

Recently, photoinduced ligand-to-metal charge transfer (LMCT), an electronic transition from the filling orbit of the ligand to the empty orbit of the metal center, has been a powerful photocatalytic platform to achieve the construction of organic compounds [14-19]. Compared to classical iridium and ruthenium complexes, LMCT employed inexpensive and readily available metal complexes (Cu, Fe, Ce, etc.) as photocatalysts, the high-valent metal complexes were excited with the irradiation of light to obtain active radicals, further participating in the relevant transformation (Scheme 1A) [20-25]. While many research groups have made significant contributions in this field [26-32], the recycling and reutilization of homogeneous photocatalysts still face significant challenges.

Indeed, over the past few years, our group has developed several heterogeneous photocatalytic systems for organic reactions based on semiconductor catalysis [33-35]. However, in these processes, photocatalytic reactions occur on the surface of heterogeneous photocatalysts, often resulting in relatively low efficiency. Therefore, we aim to develop a novel photocatalytic system that combines the high efficiency of homogeneous catalysis with the recyclability characteristic of heterogeneous catalysts.

Ionic liquids (ILs) are one of the most promising green solvents and catalysts of the 21^st century due to their potential advantages including non-volatile, good chemical and thermal stability, recyclable, and environmentally friendly [36-42]. The designability of ILs, changing the combination of anions and cations to synthesize the required functionalized ILs, provides us with ideas for designing the required photocatalysts [43-47]. Based on the mechanism of photocatalysts in the LMCT process, the high-valent metal halides could be regarded as the anionic portion of ILs. Additionally, iron, as the most abundant transition metal element in the earth's crust, has become one of the most concerned transition metal catalysts due to its inexpensiveness, easy availability, good biocompatibility, and environmental friendliness [15,48-50]. Consequently, we herein report the first example of ionic liquids as recyclable photocatalysts for organic reactions. By using the iron(Ⅲ)-based IL photocatalyst, a light-induced hydroacylation of alkenes with various aldehydes was realized via the LMCT process (Scheme 1B). Notably, the IL photocatalyst exhibits outstanding recyclability without notable decomposition after reaction, suggesting its great potential for the sustainable development of organic synthetic chemistry. Additionally, the remarkable advantages of our developed universal methodology are broad substrate scope, mild reaction conditions, and easy scale-up.

The experiments were started using p-tolualdehyde 1a, and benzyl acrylate 2a as the model substrates, as summarized in Table 1. The model reaction was carried out by employing 1-butyl-3-methylimidazolium tetrachloroferrate (C₄mim-FeCl₄) (IL-1) as a photocatalyst, and DCM as reaction solvent under N₂ atmosphere with the irradiation of 40 W purple LED (390 nm) for 24 h. Fortunately, the desired product 3a was obtained in a 78% isolated yield (entry 1). Subsequently, various solvents including DCE, CH₃CN, EtOAc, acetone, and PhCl were screened, and the results indicated that the best solvent was DCM (entries 2–6). To further enhance the reaction efficiency, a series of ILs were evaluated, it is regrettable that the yield of the target product did not further increase (entries 7–14). As the light intensity decreases, the yield of the target product also further decreases (entries 15 and 16). The results of the control experiments revealed that when there was an absence of light or IL in the photocatalytic reaction, the reaction could not proceed smoothly (entries 17 and 18), suggesting that light or IL played a vital role in this photocatalytic transformation. Consequently, the optimal reaction conditions were illustrated as follows: 1a (0.6 mmol), 2a (0.2 mmol), C₄mim-FeCl₄ (20 mol%), and DCM (2 mL) were stirred under N₂ atmosphere with the irradiation of 40 W purple LED (390 nm) for 24 h.

With the optimal reaction conditions in hand, the scope of aldehydes was investigated (Scheme 2). Benzaldehydes bearing electron-donating groups (−Me, −iPr, −^tBu, −CH₂OPh, −OAc) and electron-withdrawing substituents (−F, −Cl, −Br, −CF₃, −CN) at the para-position all reacted smoothly with 2a, leading to the corresponding products 3a-3j in 20%−92% yields. Additionally, the meta- and ortho-substituted benzaldehydes were suitable for this photocatalytic system to afford the target products 3k-3p in moderate to good yields (44%−79%). When benzaldehyde without any substituents was employed as a reaction substrate, the product 3q could be generated in an isolated yield of 87%. It was worth mentioning that 2-thiophenecarboxaldehyde and 2-naphthaldehyde were compatible with the reaction system, obtaining the target products 3r and 3s in 38% and 51% yield, respectively. More importantly, various alkyl aldehydes including linear chains and rings could be also considered as feasible substrates to generate the products 3t-3x in 41%−52% yields.

Afterward, the scope of alkenes was also explored. As shown in Scheme 3, the reactivity of diverse substituted alkenes derived from acrylate was surveyed, both aryl acrylates and alkyl acrylates could react smoothly with 1a to deliver the products 3y-3ab in 38%−89% yields. Then, the products 3ac-3ad were successfully prepared in moderate to good yields (60%−76%), suggesting α, β-unsaturated ketones could also be applied to this photocatalytic methodology. To further highlight the diversity and efficiency of our strategy, other alkenes including alkenylnitriles and alkenylsulfones were also amenable, obtaining the desired products 3ae-3ah in 21%−71% yields. However, some unactivated alkenes, including styrene, and 1-pentene, were not suitable for the reaction system (Fig. S2 in Supporting information).

Additionally, a scale-up reaction between 1a and 2a was carried out under standard conditions (Fig. S3 in Supporting information). The desired product 3a could be isolated in 66% yield (372.2 mg) (Scheme 4A), exhibiting great potential in practical applications. Considering the synthetic versatility of 1, 4-dicarbonyl compounds, the transformations of 3a were further performed (Scheme 4B). Condensation of 3a with hydrazine hydrate led to obtaining the target 4, 5-dihydropyridazin-3(2H)-one product 4 in 80% isolated yield. 3a could also undergo reduction by KBH₄, followed by condensation to form five-membered cyclic lactone 5 in 90% yield.

To evaluate the sensitivity of this photocatalytic system, a series of parameters including concentration, water level, reaction time, light intensity, air atmosphere, and scale were tested based on standard conditions (Table S5 in Supporting information). It can be seen from Fig. 1 that the N₂ atmosphere and light intensity were a vital point for the smooth progress of the photoreaction.

Interestingly, in comparison to traditional heterogeneous catalysts, the homogeneous C₄mim-FeCl₄ exhibits superior solubility within the reaction system and can be easily recovered after the reaction. As illustrated in Fig. 2A, the C₄mim-FeCl₄ and the target product 3a could be easily separated by removing the reaction solvent (DCM) and washing with diethyl ether from the homogeneous catalytic system. Additionally, the recycling experiments of C₄mim-FeCl₄ were performed to verify the stability and reusability of the photocatalytic system. The catalytic reactivity of C₄mim-FeCl₄ could still be maintained after 5 runs (Fig. 2B). Meanwhile, the UV–vis of fresh C₄mim-FeCl₄ and recovered C₄mim-FeCl₄ was displayed in Fig. 2C, the recovered C₄mim-FeCl₄ remains almost unchanged. These results indicated that C₄mim-FeCl₄ was an effective and stable photocatalyst for transformation.

Subsequently, a series of radical trapping experiments were performed. When the radical scavengers 2, 2, 6, 6-tetramethylpiperidin-1-yl-oxidanyl (TEMPO) or 2, 6-di-tert-butyl-4-methylphenol (BHT) were respectively added to the model reaction, the photoreaction was significantly suppressed, suggesting that this may involve a radical pathway (Fig. 3A). Additionally, the radical adducts A and B were detected by high-resolution mass spectroscopy (HRMS), suggesting the generation of acyl radicals. Chlorine radicals were confirmed by observing the radical adducts C (Fig. S10 in Supporting information).

To further explore the mechanism of photoreaction, we proceed with the UV–vis absorbance experiments by controlling the irradiation time of C₄mim-FeCl₄. We found that [Fe^ⅢCl₄]^– could be converted into [Fe^ⅡCl₄]^2– under the irradiation of the purple light (390 nm), indicating the generation of chlorine radicals in the photoreaction (Fig. 3B).

Based on the above detailed experimental results and relevant literature reports [51], a possible mechanism was proposed (Fig. 3C). Initially, the high-valent metal complex [Fe^ⅢCl₄]^– was excited to *[Fe^ⅢCl₄]^– under the irradiation of purple light, undergoing a LMCT process to form [Fe^ⅡCl₃]^– and key chlorine radical. Then, the chlorine radical could react with aldehydes 1 via hydrogen atom transfer (HAT), leading to the acyl radical 4. The radical 4 was added to alkenes to obtain the radical intermediate 5, which was further reduced by [Fe^ⅡCl₃]^– via a single-electron transfer (SET) process to form the anion 6 and [Fe^ⅢCl₄]^–. Finally, 6 was protonated to generate the desired products 3.

In conclusion, our study presents an unprecedented example of utilizing an ionic liquid as a recyclable photocatalyst for the light-induced hydroacylation of alkenes through the photoinduced LMCT process. Building upon the principles of homogeneous catalysis and heterogeneous separation, the ionic liquid photocatalyst demonstrates excellent catalytic efficiency and functional group compatibility, enabling easy separation and recycling through phase behavior. Various target products including ketones, 1, 3-dicarbonyl compounds, and 1, 4-dicarbonyl compounds could be synthesized under mild conditions, wherein 1, 4-dicarbonyl compounds could serve as significant intermediates in organic transformation. Notably, the iron-based ionic liquid exhibits good recyclability and stability without obvious decomposition in this photoreaction. Our discovery will pave the way for new approaches to photocatalytic organic synthesis using ionic liquids as recyclable and versatile photocatalysts.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Hao-Cong Li: Writing – original draft, Methodology, Investigation. Ming Zhang: Methodology, Investigation. Qiyan Lv: Writing – review & editing, Supervision, Data curation. Kai Sun: Methodology. Xiao-Lan Chen: Writing – review & editing, Supervision, Investigation. Lingbo Qu: Data curation. Bing Yu: Writing – review & editing, Supervision, Investigation, Data curation.

Acknowledgments

We acknowledge the financial support from the National Natural Science Foundation of China (Nos. 22071222, 22171249), the Natural Science Foundation of Henan Province (Nos. 232300421363, 242300420526), Key Research Projects of Universities in Henan Province (No. 23A180010), Science & Technology Innovation Talents in Universities of Henan Province (No. 23HASTIT003), Science and Technology Research and Development Plan Joint Fund of Henan Province (No. 242301420006).

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110579.