Using the inexpensive and readily available carbon monoxide as a C1 synthon to construct amide-containing skeletal compounds with a wide range of functionalities is a facile synthetic method [1-4]. To date, aminocarbonylation strategies for aryl and benzyl halides have been successfully established and even industrially applied [5-7]. However, the aminocarbonylation of alkyl halides has been slow to develop compared to aryl halides. The primary reason is that alkyl halides are highly susceptible to β-H elimination from alkyl-metal intermediates generated by metal-dependent activation [8-11]. This has changed in recent years with the rapid development of photocatalysis. In 1997, Sonoda's group reported the light-mediated, metal-free catalyzed esterification of alkyl iodide substituents [12]. Although the reaction was limited by product, high-energy light sources, and high-pressure carbon monoxide, it provides a direction for the further development on carbonylative transformation of alkyl halides. As photocatalysis has experienced impressive progresses during past decades, catalytic systems for the amidation of alkyl halides under visible light have been developed using iridium, manganese, nickel, palladium, and other metals (Fig. 1A) [13-16]. Although these strategies have been reported to improve the utilization of alkyl halides for carbonylation, transition metal catalysts are still required in general.

The photo-mediated conversion of alkyl halides by transition metals involves a single electron transfer process in which the transition metal plays a dual role as a reductant and traps the subsequently generated radicals so that the presence of the metal cannot be eliminated. In this sense, we need a scheme that bypasses the SET reduction of alkyl halides. In 2019, MacMillan's group reported an example of a reaction in which photo-oxidative reduction produces a silicon radical to activate an alkyl bromide substituent [17, 18]. In 2020, Leonori's group discovered that α-aminoalkyl radicals enable the activation of alkyl iodides [19, 20]. Immediately afterward, in 2021, Liu's group discovered that aryl radicals also have the effect of activating alkyl iodides [21]. The method reported above uses the XAT (halogen atom transfer) strategy, where the transition metal is used in the reaction only as a trapping radical effect and the substrate activation step can be split with the metal cycle. Based on the XAT strategy, the reaction does not need to overcome the strong reduction potential of the C(sp³)-halogen bond (E_red ≪ −2.0 V vs. SCE), which makes the reaction much less dependent on the transition metal (Fig. 1B).

Considering the growing need for green and sustainable chemical synthesis and the important role of amide-containing backbone compounds, there is an urgent need to develop a low-energy and metal-free catalytic system for the aminocarbonylation of alkyl halides. Along these lines, we learn that in 2023, Procter's group reported an example of a reaction in which an aryl sulfonium salt and an aryl amine yielded an aryl radical under visible light irradiation in the absence of an exogenous photocatalyst [22]. In the same year, Chen's group reported a similar reaction in which trifluoromethyl sulfonium salt and aniline formed trifluoromethyl and nitrogen radicals, respectively, only under visible light irradiation [23]. The above reaction leads to the formation of electron donor-acceptor complexes (EDA) by molecular aggregation through the binding of an electron acceptor substrate and an electron donor substrate [24, 25]. In this reaction, the EDA complex can form a pair of free radicals respectively after being exposed to light. Based on the above, we have attempted to combine the aryl sulfonium salts EDA complexes with the XAT strategy to establish a light-only, non-metallic, exogenous catalyst-free aminocarbonylation reaction of alkyl halides. The design has the following advantages: (1) High efficiency of inter-radical coupling, avoiding reverse decomposition of acyl radicals, overcoming the use of high-pressure CO and special equipment; (2) aryl sulfonium salts as a stable source of XAT reagents and recyclable degradation products. To demonstrate the advantages of the method, we have attempted to explore the aminocarbonylation of a range of alkyl iodides in a non-metallic, exogenous catalyst-free, and atmospheric pressure carbon monoxide atmosphere, using phenyl sulfonium salts, aniline, and alkyl iodide substituents as substrates for the reaction, using a combination of EDA and XAT reactions at room temperature (Fig. 1C).

We set out to explore the feasibility of this light-mediated exogenous catalyst-free reaction design using 4-iodooxane 1a and aniline 2a as reaction substrates, and after systematic collation of the reaction transformation data, the reaction conditions are summarised in Table 1 (for more details, see Supporting information). Initially, in order to better enable the reaction to occur, we used 2 equiv. of aniline and phenyl sulfonium salts as electron donors and acceptors of the EDA complex, acetone (2 mL), and irradiated with a 15 W blue light lamp under a 60 bar CO atmosphere, and successfully detected the target product of the reaction (Table 1, entry 1). To investigate the effect of changing the reaction environment on the reaction, we replaced acetone with other solvents and found a slight decrease in yield (Table 1, entry 2). Immediately afterwards, we replaced the cesium fluoride with sodium carbonate and, to our surprise, obtained a significant increase in yield (Table 1, entry 3). Based on this, we then carefully screened sodium ion bases, carbonate ion bases and organic bases, but did not obtain better results (Table 1, entries 4–6). To reduce energy consumption, we chose to reduce the reaction time to 24 h and observed no significant change in yield (Table 1, entry 7). To verify the initial idea, we also reduced the carbon monoxide pressure to 1 bar and adjusted the dosage of aniline and phenyl sulfonium salts to 1.2 times and found that the reaction gave the target product in a high yield of 76% (Table 1, entries 8 and 9). This is a good demonstration of the high efficiency of the XAT strategy and the inter-radical coupling.

To understand the substrate compatibility of this design, we reacted a series of amines with different substituents (Fig. 2). When different electron-withdrawing and electron-donating substituents were used to modify the aniline counterparts, the amide target products were obtained in high yields and in essentially unchanged yields (3a-3g). By adding substituents at different positions on the benzene ring to change the spatial resistance of the substrate, the reaction also gave the target products (3h-3l) in good yields. In addition, aniline with two substituents on the aromatic ring was also compatible with the reaction and the desired products 3m and 3n were obtained in 78% and 49% yields respectively. Surprisingly, the reaction is also compatible with heterocyclic substrates (3o-3p), which provides great potential for subsequent application of the reaction. Secondary amines were also well compatible in the reaction and the established products (3q-3r) were obtained in good yields of 68% and 60% together. In addition to aromatic amines, alkyl amines are likewise well compatibilized by the reaction, yielding alkylamides (3s-3t) in over 70% yields in general. In summary, the overall experimental results showed that the reaction has a good compatibility for the electronic as well as spatial site resistance effects of the substrate amines. Finally, we attempted the bioactive molecular backbone modification of the product and successfully obtained the target products (3u-3z) in good to excellent yields.

The compatibility test of alkyl iodides with different substituents was carried out immediately after understanding the substrate properties of anilines (Fig. 3). Under optimal conditions, straight-chain alkyl iodides were added to the reaction to give the target products (4a-4c) in moderate yields. When the reaction was carried out with secondary carbon iodide compounds, the reaction yields were significantly higher (4d-4j), indicating that the stability of the secondary carbon radicals was preferred to that of the primary carbon radicals. Modifying the substrate with heteroatoms still yielded the products (4i-4j) in high yields. Fortunately, the presence of both chlorine and iodine halogens in the substrate was observed to selectively activate iodine to give the chlorinated amide 4k, and the retained chlorine atom facilitated further modification of the product. The substrate of the dual reaction site was successfully incorporated into the reaction to give the desired bis-amide compound 4l in 60% yield, and the reaction was equally well tolerated when bioactive molecules were introduced into the substrate (4m-4n). Additionally, tert‑butyl iodide was also tested with aniline under our standard conditions, but less 10% of the desired product was detected.

Inspired by the above reaction results, we also learned that o-benzoyl oxime can generate phenyl radical and imine radical respectively through energy transfer process under light induction [26, 27]. We envisage that o-benzoyl oxime could act as both a XAT promoter and a radical coupling agent, and most importantly, the product obtained is a primary amide, since primary amides play an important role as starting reagents in many reactions. We are excited that the reaction is successful under non-metallic catalysis and low carbon monoxide pressure. After a series of condition optimisations to determine the optimum conditions, we started to test the compatibility of the reaction with different alkyl iodides (Fig. 4). Like the reaction results above, both primary and secondary alkyl iodides can react well. However, secondary carbon iodides are more reactive than primary carbon iodides, and the corresponding primary amide products (6e-6l) were obtained in greater than 50% yields. It is worth mentioning that when the iodide was modified with heteroatoms it has essentially no effect on the reaction outcome.

To understand the specific transformation mechanism of the above reaction, we performed some control experiments (Fig. 5). First, the radical scavengers TEMPO, DPE and BHT were added to the system under standard reaction conditions. As a result, the formation of the target product could not be detected, but radical scavenging intermediates were detected by HRMS (Fig. 5A). Through the intermediates obtained by the radical scavenger, it is clearly observed that phenyl radical, alkyl carbon radical and acyl radical were generated in the system. The reaction was performed under standard conditions, we detected the target product and observed the formation of a large amount of iodobenzene. This result confirmed the presence of phenyl radical and its role as XAT promoter (Fig. 5B). When the carbon monoxide pressure of the reaction was increased to 60 bar, the formation of dicarbonyl product could be detected from the reaction mixture. This result confirmed that alkyl carbon radical and nitrogen radical were successfully generated by the EDA and XAT strategies, and radical coupling occurred. Competition experiments confirmed that more electron-rich alkylamines were more reactive and could serve as better electron donors for EDA (Fig. 5C).

Following the results of the controlled experiments described above and the relevant literatures [22-33], we proposed a specific transformation mechanism for the occurrence of the reaction (Fig. 6). We believe that the reaction is started initially by the EDA complex, which forms a complex between the phenyl sulfonium salts and aniline, and that under visible light irradiation the EDA complex is excited to form the phenyl radical and the aniline cation Ⅰ. Subsequent activation of alkyl iodides by phenyl radicals to form alkyl carbon radicals Ⅱ. In an atmosphere of carbon monoxide, the alkyl carbon radical Ⅱ traps carbon monoxide to form the acyl radical Ⅲ. Finally, the aniline cation was deprotonated to give nitrogen radical Ⅳ, which then rapidly coupled with acyl radical Ⅲ to afford the desired amide product.

In summary, we have developed a green and low-energy carbonylation reaction that effectively combines the EDA and XAT strategies to convert alkyl iodides to amides in a metal-free, exogenous catalyst-free, atmospheric carbon monoxide atmosphere at room temperature. The reaction has good substrate compatibility and enriches the use of alkyl iodides. Most importantly, the use of high-pressure carbon monoxide in metal-free catalysis has been overcome by efficient free radical coupling, leading to further liberation of the carbonylation reaction limits.

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

Hefei Yang: Writing – original draft, Methodology, Formal analysis, Data curation. Le-Cheng Wang: Formal analysis, Data curation. Xiao-Feng Wu: Writing – review & editing, Writing – original draft, Supervision, Project administration, Funding acquisition, Conceptualization.

Acknowledgments

This work was supported by National Key R & D Program of China (No. 2023YFA1507500) and the International Partnership Program of Chinese Academy of Sciences (No. 028GJHZ2023045FN).

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.110843.