The oxidation and coupling of readily accessible feedstock chemicals constitute a powerful strategy for synthesis of structurally intricate molecules in modern organic chemistry. Within this paradigm, the catalytic oxyalkylation of alkenes has become a cornerstone methodology, enabling simultaneous elevation of oxidation states and molecular complexity from simple precursors [
1–
6]. While significant progress has been achieved in this area, the pursuit of practical and sustainable protocols characterized by the use of earth-abundant metals, cost-effective ligands, and green oxidants remains a pivotal frontier.
One noteworthy application of alkene oxyalkylation is the synthesis of 1,4-diketones
via a sequence involving the generation of
α-keto radicals from ketones, addition to alkenes, and oxidation of the resulting
γ-keto radicals (
Fig. 1A) [
7–
13]. In 2014, Klussmann’s group developed a
para-toluenesulfonic acid-catalyzed alkylation-peroxidation reaction of styrene with ketones to afford
γ-peroxyketones, which was then converted to the 1,4-diketones [
7]. The groups of Huang [
8], Wan [
9], Xing [
10], and Thakur [
11] independently developed direct oxidative couplings of terminal vinylarenes with ketones to furnish 1,4-diketones directly, employing Cu/Mn co-catalysis or cobalt salts in conjunction with TBHP. Maity [
12] and Senadi [
13] reported an alternative strategy involving photoredox-catalyzed 1,2-oxoalkylation of vinyl arenes with 1,3-diketones, employing air as the terminal oxidant, sequentially. Inspired by these advances on 1,4-diketone synthesis through
α-keto radicals, we envisioned that
β-keto radicals generated by oxidation of cyclopropanols, the most prominent type of the homo-enol equivalents [
14–
21], might undergo an addition reaction with olefins to form
δ-keto radicals, which could then be oxidized by molecular oxygen to afford 1,5-diketones straightforwardly [
22–
24].
The challenge of achieving this aerobic radical transformation lies in the intrinsic instability of
β-keto radicals, which are less stable than
α-keto counterparts and readily quenched by molecular oxygen [
25–
27]. Overcoming this limitation requires a catalytic system capable of efficiently generating
β-keto radicals and directing their regioselective addition to alkenes. Recently, our group developed an intramolecular homo-Mannich reaction of cyclopropanols to construct azabicyclo[3.3.1]nonane skeletons [
28–
31]. In this reaction,
β-keto radicals were generated
via oxidative ring-opening of cyclopropanols and then efficiently trapped by the C=N double bonds of imines with CuCl
2 as the catalyst. Inspired by the advances in cyclopropanols chemistry made by our group [
28–
31] and others [
32–
47], we surmised that a copper-based catalytic system could be adapted to oxyalkylation of the C=C double bonds of alkenes with cyclopropanols, to furnish diverse 1,5-diketones (
Fig. 1B). Herein, we report a copper-catalyzed oxyalkylation of alkenes with cyclopropanols under an atmosphere of air. This operationally simple protocol accommodates both inter- and intramolecular variants [
48–
55], providing rapid access to 1,5-diketones bearing linear, fused, spirocyclic, or bridged scaffolds-motifs commonly found in natural products such as spiroetherone A, siccanin, and dysidavarone A (
Fig. 1C) [
56–
63].
As shown in
Table 1, we first commenced our study by screening various metal salts using cyclopropanol
1a as the model substrate (entries 1–4), and CuCl
2 was identified as the optimal copper salt, yielding product
2a in 32% yield (entry 3). Investigation of ligands (entries 5–9) indicated that 1,10-phenanthroline (Phen,
L5) was the ligand of choice, increasing the yield to 69% (entry 9). Examination of solvents indicated that substitution of THF with 1,4-dioxane or MeCN was detrimental to the product yield (entries 10 and 11). When the reaction time was shortened to 2 h, the yield remained at 69%, underscoring the efficiency of this reaction protocol (entry 12). Further improvement was achieved using the preformed CuCl
2·Phen complex as the catalyst, which delivered
2a in 79% yield at a 10 mol% catalyst loading (entries 13 and 14).
With the optimized reaction conditions established, we first explored the intramolecular oxyalkylation of alkenes with cyclopropanols (
Scheme 1). Styryl-substituted substrates, bearing electron-donating or -withdrawing groups on the aryl ring, underwent a smooth transformation to afford the corresponding 4-benzoylcyclohexanone derivatives
2a–
2h in moderate to good yields. Other olefins with aryl groups including 1-naphthyl, 3-pyridyl and 2-thienyl were well tolerated, delivering products
2i–
2k in good yields. Importantly, the 1,1-diphenylethene-derived substrate
1l also furnished the hydroxylated cyclohexanone
2l. Aliphatic olefins proved to be compatible as well, affording
2n and
2o in good yields. Significantly, the methodology enabled the synthesis of a diverse array of bridged, spiro, and fused polycyclic products
2q–
2y. For instance, cyclopropanols
1q and
1r produced functionalized bicyclo[3.3.1]nonanes
2q and
2r, structural analogues of Dysidavarone A [
53]. Likewise, fused products
2s–
2v were successfully generated, which are the core scaffold of the succinate dehydrogenase inhibitor Siccanin [
54]. Importantly, the product
2u could be obtained in 48% yield when the reaction was carried out on a 5 mmol scale, and the structure of
2t was confirmed by X-ray crystallography (see details in Supporting information). Furthermore, cyclopropanols with dihydronaphthalene and cyclohexene substituents yielded spirocyclic products
2w–
2y. Among them,
2w represents the spirocyclic core structure found in the natural product Spiroetherane [
55].
Next, this method was adapted to the intermolecular oxyalkylation under slightly modified reaction conditions (
Scheme 2). A broad range of cyclopropanols were compatible, affording the corresponding linear 1,5-dicarbonyl compounds
5a–
5m in moderate to good yields. Notably, cyclopropanol
4k with a free hydroxyl group delivered product
5k in 62% yield. Substrates bearing two cyclopropanol units (
4l and
4m) also participated in the transformation, providing products
5l and
5m with two 1,5-dicarbonyl fragments. Furthermore, the utility of this protocol for drug derivatization was demonstrated. Cyclopropanols derived from ibuprofen, naproxen, and lithocholic acid reacted well with styrene to furnish products
5n–
5q in 45%–64% yields. Remarkably, coupling of an ibuprofen-derived cyclopropanol with a styrene derived from fenofibrate generated a conjugate of two drugs (
5q), which may be of pharmacological interest in drug discovery.
Additionally, this versatile synthetic strategy was further applied to the synthesis of natural products or their analogues (
Fig. 2) [
56–
63]. First, the natural product analogue 4-dehydroxydiversonol (
7) [
56,
57] with multiple oxygen functionalities was synthesized from
2t by taking advantage of the two ketone groups generated by our new method. The linear 1,5-dicarbonyl compounds can be employed in generating diverse range of heterocycles (
9 and
10). For instance, a lasubine Ⅱ analogue
9 was successfully prepared from oxyalkylation product
5k through a sequence of reductive amination and intramolecular alkylation [
60,
61]. (±)-Centrolobine (
10), which was isolated from the heartwood of
Centrolobium robustum and exhibits biological activity against
Leishmania amazonensis promastigotes, was synthesized in two steps from
5i [
62,
63].
Control experiments were conducted to study the reaction mechanism (
Fig. 3A). No product formation was observed in the presence of radical scavengers TEMPO or BHT, the corresponding adducts of the
β-carbonyl radical with TEMPO or BHT were detected by HRMS, indicating the involvement of a radical process. To probe the source of the incorporated oxygen atom, isotopic labeling studies were performed. When
18O
2 was employed as the oxidant in place of air, the resulting
2a was obtained in 50% yield with complete
18O incorporation; while 10 equivalents of H
218O were added to the standard reaction conditions, the product
2a displayed 51%
18O incorporation as determined by high-resolution mass spectrometry. To further examine the possibility of water as the oxygen source, compound
11 was prepared and subjected to the standard conditions, furnishing product
2a in 49% yield. These findings suggest that both molecular oxygen and water may contribute as sources of the incorporated oxygen atom.
Based on the results of control experiments and previous reports [
28–
31],[
64–
70], a plausible reaction mechanism is proposed (
Fig. 3B). Initially, association of Cu
Ⅱ-
L5 with cyclopropanol
1a yields intermediate
IM1, which undergoes ring opening and radical-type 6-
exo-trig cyclization onto the alkene to afford
δ-ketone radical
IM2 [
68]. Subsequently, two possible pathways may account for the formation of the final product
2a. The dominant route is Path A, in which radical
IM2 is intercepted by molecular oxygen to generate peroxyl radical
IM4, which undergoes reduction by [
CuⅠ] to afford
IM5. Then,
IM5 go through a elimination reaction to ultimately afford product
2a and release [
CuⅡ-(OH)]
+ species [
66,
67]. In this pathway, molecular oxygen serves a dual role as both the oxidant and oxygen source. A plausible alternative mechanism (Path B) involves the further oxidation of radical
IM2 to form carbocation
IM3, which is subsequently attacked by water to yield hydroxylated intermediate
11, which is then followed by oxidation to deliver the final diketone
2a [
69,
70].
In summary, we have developed an efficient method for the synthesis of 1,5-diketones via Cu-catalyzed oxyalkylation of alkenes with cyclopropanols under ambient air. The use of CuCl2·Phen complex as the catalyst proved critical to the success of this transformation. This protocol offers notable synthetic advantages, including operational simplicity, mild conditions, and broad substrate scope, enabling the conversion of unactivated alkenes into structurally diverse 1,5-diketones. Importantly, the intramolecular variant of this reaction grants access to valuable fused, spiro, bridged polycyclic scaffolds.