Ligand-enabled Ni-catalysed dicarbofunctionalisation of alkenes with diverse native functional groups Authors: Dao-Ming Wang, Hui-Mei Shan, Li-Qin She, Yu-Qing He, Yichen Wu, Yong Tang, Li-Ping Xu, Peng Wang Published: 2024 ABSTRACT AbstractThe transition metal-catalysed dicarbofunctionalisation of unactivated alkenes normally requires exogenous strong coordinated directing groups, thus reducing the overall reaction efficiency. Here, we report a ligand-enabled Ni(II)-catalysed dicarbofunctionalisation of unactivated alkenes with aryl/alkenyl boronic acids and alkyl halides as the coupling partners with a diverse range of native functional groups as the directing group. This dicarbofunctionalisation protocol provides an efficient and direct route towards vicinal 1,2-disubstituted alkanes using primary, secondary, tertiary amides, sulfonamides, as well as secondary and tertiary amines under redox-neutral conditions that are challenging to access through conventional methods. The key to the success of this reaction is the use of a bulky β-diketone ligand, which could enable the insertion of alkene to aryl-Ni(II) species, stabilize the alkyl-Ni(II) species and inhibit the homolytic alkyl-Ni(II) cleavage, supporting by both experimental and computational studies. This dicarbofunctionalisation reaction features the use of native directing group, a broad substrate scope, and excellent scalability.Subject terms: Synthetic chemistry methodology, Homogeneous catalysis Introduction Transition metal-catalysed dicarbofunctionalisation of alkenes has emerged as a powerful methodology that allows the simultaneous formation of two new carbon–carbon bonds across an unsaturated carbon–carbon double bond in a single step1–10. This strategy significantly enhances molecular complexity and diversity, making it a valuable tool in in the synthesis of natural products, pharmaceutical molecules and functional materials. One major challenge with this approach is precisely controlling the regioselectivity during the alkene migratory insertion, particularly with innately less reactive unactivated alkenes that lack electronic and steric biases. Moreover, undesired processes such as alkene isomerization, classic Heck or reductive Heck-type coupling, and direct coupling of carbon partners etc. are inevitable in those three-component reactions, complicating the reaction outcome and reducing selectivity. To address these challenges, several elegant approaches, including directing group assistance11–24, substrate control25–31, and catalyst control32–35, have been developed for dicarbofuctionalisation of unactivated alkenes, with the majority of advances in the field being dominated by directing group-assisted processes (Fig. 1a). However, the use of external strong coordinated directing groups (DGs), including 8-aminoquinoline11,12,22,24 and 2-picolinic acid23, pyrimidine14, pyridine15, imine13,16,17,20, and sulfonamide18,19,21 etc. normally requires extra steps for installation and removal, reducing the step/atom-economy and overall reaction efficiency. Synopsis for nickel-catalysed dicarbofunctionalisation of unactivated alkenes. a Directing group-assisted dicarbofunctionalisation of unactivated alkenes. b Challenges for metal-catalysed arylalkylation of unactivated alkenes with native functional groups. c Ligand-enabled nickel-catalysed dicarbofunctionalisation of alkenes with diverse functional groups. Recently, the native functional group (such as acid, amide, ketone, amine, etc.) directed alkene difunctionalisation offers tremendous opportunities for constructing molecular architectures with improved overall efficiency and atom-economy by eliminating the need for extraneous directing groups36–46. However, achieving such alkene dicarbofunctionalisations with high efficiency and regioselectivity using native functional groups as directing groups remain a big challenge, due to weaker coordination ability to transition metals compared to artificially installed directing groups. Despite of those challenges, the Engle group reported nickel-catalysed diarylation of unactivated alkenes using various simple amides as the directing group36. Later on, the diarylation, and aryl-alkenylation has been realized with acid38 or ketone39,41 as native directing group (Fig. 1a), respectively. To date, the native functional group-directed aryl/alkenyl-alkylation initiated by the insertion of carbon-metal species with alkyl halides as the alkylating reagents has not been disclosed, probably due to the feasible generation of alkyl radicals via single-electron pathway36,39 and the generation of metal-hydride species via β-H elimination that initiates undesired alkyl-functionalization or hydrofuctionalisation (Fig. 1b). More importantly, the regioselective dicarbofucntionalization of unactivated alkenes that tolerates a wide range of native functional groups is lacking due to the absence of efficient catalysts. Here, we report a native functional group-directed Ni(II)-catalysed 1,2-dicarbofunctionalisation of unactivated alkenes using alkenyl amides, sulfonamides and amines as substrates, alkyl halides as alkylating reagents and arylboronic acids as aryl sources under redox-neutral conditions (Fig. 1c). A wide range of primary, secondary and tertiary alkenyl amides, sulfonamides, as well as secondary and tertiary amines serve as efficient native directing group for this nickel-catalysed aryl/alkenylalkylation reaction. The key to the success is the use of a bulky β-diketone ligand that accelerates the alkene migratory insertion, stabilizes the alkyl-Ni(II) species, and suppresses the homolytic cleavage of alkyl-Ni(II) species. This protocol features a broad substrate scope, excellent functional group and heterocycle tolerance, thus paves a new avenue for the expedient enlargement of molecular complexity in a highly efficient manner. Compared to the existing methodologies for transition metal-catalysed dicarbofunctionalization of unactivated alkenes with exogenous directing group, this catalytic process obviates the need for pre-installed strong monodentate or bidentate directing group, offering notable advantages in practical synthesis. Leveraging this process, several bioactive compounds including a CBP/EP300 bromodomain inhibitor, an atypical antipsychotic compound and a key intermediate for the synthesis of potential antipsychotic compound, were prepared via shorter synthetic routes with high step-economy. Results and discussion Reaction optimization Recently, we have demonstrated a ligand-enabled Ni(II)-catalysed hydroxyarylation of unactivated alkenes by using molecular oxygen as the oxidant and oxygen source46, in which a series of native functional groups could serve as efficient directing group, including various amines, amides, and alcohols etc. Although the reaction mechanism for the oxidation of alkyl-Ni(II) species is unclear, we have proven that the alkyl-Ni(II) species generated by the migratory insertion of alkene into aryl-Ni(II) species is stable in the presence of β-diketone ligand, and the radical process via homolytic cleavage of alkyl-Ni(II) species is unlikely. Hence we envision that the Ni(II)-catalysed arylalkylation reaction might be feasible with alkyl halides as alkylating reagent by reacting with the alkyl-Ni(II) species. Following this hypothesis, we first attempted to verify the feasibility of this hypothesis by checking the reaction of well-define alkyl-Ni(II) complex with n-butyl iodide, and found that the desired alkylated product S1 was obtained in 30% yield along with the 12% of protonation product S2 and 5% of dimerized product S3 (Fig. 2a). Following this lead, we turned to systematically investigate the reaction parameters for catalytic arylalkylation of β, γ-unsaturated amides by using phenylboronic acid as the arylating reagent and n-butyl iodide as the alkylating reagent. As the ligands play a crucial role in our Ni-catalysed alkene difunctionalisation reactions, we firstly evaluated the ligand effects on this reaction by using Ni(OTf)2 as the catalyst precursor (Fig. 2b). Not surprisingly, the dinitrogen ligands including bipyridine ligand (L1), 1,10-phenanthroline ligand (L2), diimine ligand (L3), pyridine-oxazoline (L4), bis-oxazoline ligand (L5), diamine ligand (L6), and NHC ligand (L7) didn’t afford the desired arylalkylated product. Instead, all those ligands resulted in many side-reactions, including direct Heck-type reaction (5), direct isomerization of the unactivated alkene (6), hydroarylation (7) etc. The use of phosphine ligands including DPPF (L8) and DPPH (L9), only gave the isomerization products (6). These results further confirmed the challenge of the aryalkylation of unactivated alkene without a strong bidentate directing group. Given the bulky β-diketone ligands have the capability of enabling the Ar-Ni(II) species-initiated functionalizations of unactivated alkenes efficiently46,47, we next turned our attention to systematically investigate the effects of β-diketone ligands. To our delight, the simple acetylacetonate (acac) ligand (L10) could provide the desired product in 28% yield, and the efficiency of this reaction could be further enhanced by increasing the steric hindrance on the β-diketone ligands (L11-14). The diketone ligands bearing cyclohexyl (L12), isopropyl (L13), tert-butyl (L14) gave similar reactivities, affording the arylalkylated product in 98% yield. The β-diketone ligand bearing 1-adamantyl group (L15), which is the optimal ligand in our previous nickel-catalysed hydroarylation reaction47, resulted in 32% yield. Although dibenzoylmethane (L16) is also reactive for our reaction, electron-deficient hexafluoroacetylacetone (L17) did not result in an arylalkylation reaction. Choosing iPr-acac ligand (L13) as the optimal ligand, the desired product 4a can be obtained in 88% isolated yield using Ni(OTf)2 as the precatalyst and CsHCO3 as the base in t-AmylOH. Control experiments indicated that all reaction parameters were indispensable. Both alkyl bromide and alkyl iodide showed similar reactivities in our protocol, while alkyl chloride was less reactive (Fig. 2c). Ni(II)-catalysed dicarbofunctionalization of unactivated alkenes. a The reaction of alkyl-Ni(II) complex with n-butyl iodide; b Ligand evaluation, reaction conditions: 1 (0.1 mmol, 1.0 equiv), 2a (0.2 mmol, 2.0 equiv), 3a (0.2 mmol, 2.0 equiv), Ni(OTf)2 (10 mol%), Ligand (20 mol%), CsHCO3 (0.2 mmol, 2.0 equiv), t-AmylOH (0.5 mL), N2, 80 oC, 10 h; ND = not detected. the yield was determined by 1H NMR using dibromomethane as the internal standard. See Supplementary Information for experimental details. c Optimal conditions and control experiments. w/o, without. Substrate scope With the optimized reaction conditions in hand, we set out to evaluate the substrate scope of the ligand-enabled aryalkylation reaction using N-phenylbut-3-enamide 1 as the model substrate with n-butyl iodide (Fig. 3a). Arylboronic acids containing electron-donating or electron-deficient substituent on the ortho-, meta-, or para-position of the aromatic ring served as effective substrates, providing the corresponding products in moderate to excellent yields. Excellent functional group compatibility was observed, and this protocol could tolerate methyl (2b, 2l, 2t), methoxy (2c), trifluoromethoxy (2d), fluoro (2e, 2n), chloro (2f, 2o), bromo (2g, 2p), trifluoromethyl (2h, 2q), acetyl (2i), ester (2j), nitrile (2k), methylthio (2m). Multisubstituted aryl boronic acids (2u–y, 2aa) and 2-naphthyl boronic acid (2z) gave the desired products in 57–83% yields. The arylboronic acid bearing a native phenolic hydroxy group could also be tolerated, providing the arylalkylating product 4r in 55% yield. Notably, this protocol features excellent heteroatom and heterocycle tolerance, which might find applications in pharmaceutical discovery and synthetic chemistry. For instance, the arylboronic acids bearing morpholine (2s), quinoline (2ab), benzofuran (2ac) and dibenzothiophene (2ad) derivatives are all suitable arylating reagents. Notably, the alkenylboronic acids could also been employed in our reaction, giving desired alkenylalkyated products (4ae-4ag) in 30–63% yields, while no desired dialkylation product was obtained with alkylboronic acid (2ah). The scope of arylboronic acids and alkyl halides. The values under each structure indicate isolated yields. a Scope of aryl boronic acids; b, Scope of alkyl halides. Reaction conditions: 1 (35.0 mg, 0.2 mmol), 2 (0.4 mmol, 2.0 equiv), 3 or 8 (0.4 mmol, 2.0 equiv), Ni(OTf)2 (7.1 mg, 10 mol %), L13 (6.3 mg, 20 mol %), CsHCO3 (77.6 mg, 2.0 equiv.), t-AmylOH (1.0 mL), 80 oC, 10 h. For 9i and 9j, alkyl bromides were used. Next, we turned our attention to investigate the scope of aliphatic alkyl halides (Fig. 3b). A wide range of alkyl iodides with various functional groups proceeded smoothly, yielding the desired products in 47–94% yields. The reactions proceeded smoothly in the presence of a methyl (8a), long-chain alkyl (8c), ester (8d), nitrile (8e), chloride (8f), trifluoromethyl (8g), amides with free N-H (8h), phthalimide (8i), hydroxyl (8j), acetal (8k), organoboronate (8l), phosphate ester (8 m) to afford the expected products in 57–94% yield. Notably, the deuterated iodomethane can also obtain desired product (9b) in 86% yield. Moreover, the alkyl iodides containing heterocycles (8n, 8p, 8q) were also well-tolerated, obtaining desired products in 79–92% yields. In addition to primary alkyl halides, the secondary alkyl halides (8r-8t) were also reactive, giving the desired products in moderate yields, albeit the tertiary alkyl halides were not tolerated probably due to the steric hindrance. Bioactive complexes containing alkyl iodides including citronellol (8u), ciprofibrate (8v), indometacin (8w), ibuprofen (8x) were well tolerated in our current protocol. We have also investigated aryl halides under our optimal conditions. Unfortunately, no desired diarylated products were formed under current catalytic system, which might be explained by the week oxidative potentials of aryl iodides in comparison to alkyl iodides31. The generality of alkenes was evaluated after the scope of arylboronic acids and alkyl halides was systematically evaluated. As listed in Fig. 4a, various primary, secondary and tertiary amide-substituted unactivated olefins could smoothly be converted to the corresponding arylalkylation products in moderate to high yields. In general, the primary alkenyl amide and secondary alkenyl amides were more reactive than the tertiary alkenyl amides. Both aniline- (10a, 10b) and primary alkyl amine-derived secondary alkenyl amides (10c-l) were all suitable substrates, delivering the desired product in 65–98% yields. A variety of functional groups including phenyl (11c, 11d), ether (11e), ester (11f), Boc-protected amino group (11g), indolyl (11h), and cyclohexyl (11i) were all suitable in this reaction. In comparison to the secondary alkenyl amides, the tertiary alkenyl amides(10m-p) gave the desired products with a slightly lower yield (49–74%). It is noteworthy that the Weinreb amide (10n) was also compatible with our protocol, affording desired product in 49% yield, which provides a new route for the synthesis of related ketones. This protocol features excellent heterocycle tolerance, the alkenyl amides containing unprotected indole (10h), tetrahydro-2H-pyran (10j), piperidine (10k, 10o), and morpholine (10p) were all suitable substrates. The substrates with α-substituents (10r, 10s) were also reactive, giving the desired products in moderate yields. Furthermore, internal alkenes bearing methyl (10t) and ethyl (10u) at the γ-position could be converted to corresponding products with excellent regioselectivities and diastereoselectivities. A series of bioactive complex molecule-derived alkenyl amides, including aminoglutethimide (10v), atomoxetine (10w), an apixaban intermediate (10x), proved to be compatible with our protocol. The scope of unactivated alkenes. The values under each structure indicate isolated yields. a Scope of alkenyl amides; b Scope of alkenyl amines. Reaction conditions: 10 (35.0 mg, 0.2 mmol), 2a (0.4 mmol, 2.0 equiv.), 3 (0.4 mmol, 2.0 equiv), Ni(OTf)2 (7.1 mg, 10 mol%), L13 (6.3 mg, 20 mol%), CsHCO3 (77.6 mg, 2.0 equiv.), t-AmylOH (1.0 mL), 80 oC, 10 h. For 11t and 11 u, the reaction was conducted at 90 oC. Encouraged by the success with alkenyl amides, we envisioned that our protocol could also be applied in other unactivated alkenes bearing other native directing group, such as secondary amines and tertiary amines (Fig. 4b). To our delight, the secondary alkenyl amine (12a) gave the desired arylalkylation product in 64% yield. Following this positive lead, other secondary amines (12b-c) were evaluated, providing the corresponding products in moderate yields. In addition to secondary amines, this reaction possesses excellent compatibility with various tertiary amines, such as N-methylaniline (12 d), piperidine (12e), and morpholine (12f-o). Again, a large range of alkyl halides (13g-j) and aryl boronic acids (13k-o) are all tolerate with alkenyl amine substrates with moderate to excellent yields, which highlighted the generality of our reaction system. To demonstrate the synthetic utility of this nickel-catalysed dicarbofunctionalization reaction, we performed synthetic derivatizations employing Weinreb amide derivative 11n (Fig. 5a). It is noteworthy that the Weinreb amide is widely utilized in synthetic chemistry and pharmaceutic industry, making it an appealing choice for the late-stage derivatization. Treatments of 11n with various nucleophilic reagents, including phenylmagnesium bromide, trimethylsilylacetylene in the presence of n-BuLi, and TMSCHCF2 in the presence of potassium tert-amylate, gave corresponding ketones 14-16 in 69–89% yields. The Weinreb amide could be also converted to the corresponding aldehyde 17 using LiAlH4 as a reductant. To further demonstrate the synthetic utility of this methodology, the efficient synthesis of several bioactive compounds was carried out by using our current method as the key step. For instance, starting from the product 9a, 4-methyl tetrahydro-2H-benzo[b]azepin-2-one 20, a CBP/EP300 bromodomain inhibitor48, could be efficiently synthesized via hydrolysis of amide, polyphosphoric acid-facilitated Friedel-Crafts cyclization, condensation with hydroxylamine hydrochloride, and Beckman rearrangement (Fig. 5b). The αvβ3 inhibitor 24 could be efficiently prepared in 3 steps (Fig. 5c) with a synthetic sequence containing arylalkylation of N-phenylbut-3-enamide 1 with (4-hydroxyphenyl)boronic acid and methyl iodide, Mitsunobu reaction and further hydrolysis. In comparison to known 6-step synthetic route49, this synthetic route is more efficient and concise, showcasing the potential synthetic value of our current method. In addition, the formal synthesis of an atypical antipsychotic 28 could also be achieved by using unmasked 2-bromoethanol 25 as the alkylating reagent, the desired arylalkylated product 26 was generated in 42% yield, which could be transformed to lactone 27 in 70% yield via lactonization in the presence of HBr (Fig. 5d). Notably, the lactone 27 is the key intermediate for the synthesis of a potential antipsychotic compound50. Derivatization of Weinreb amide and synthetic applications. a Derivatization of the Weinreb amide 11n. b Synthesis of CBP/EP300 bromodomain inhibitor. c Synthesis of the αvβ3 inhibitor. d Formal synthesis of a potential antipsychotic compound. PPA polyphosphoric acid, ADDP 1,1’-(azodicarbonyl)-dipiperidine. Mechanistic studies To gain mechanistic insights into this reaction, a series of mechanistic experiments were carried out. The control experiments in the absence of n-butyl iodide 3 gave the hydroarylation product 7, indicating that alkyl halides were not involved in the migratory insertion of alkenes to aryl-Ni(II) (Fig. 6a). The fact that no arylalkylation products were observed by using alkyl trifluoromethanesulfonate (29) instead of alkyl halides under standard conditions, excludes a nucleophilic substitution pathway with in-situ generated alkyl-Ni(II) species and alkyl iodides. As the alkyl-Ni(II) species could undergo a homolytic process to produce alkyl radicals and corresponding Ni(I) species, we next carried out the radical capturing experiments to explore if our reaction might undergo a metal-radical relay process51,52. The addition of radical scavenger BHT and TEMPO could not shut down the reaction, providing the desired product in 85% and 32% yield, respectively (Fig. 6b). In addition, the reactions with 1,1-diphenylethylene (30) or (E)-(2-(phenylsulfonyl)vinyl)benzene (32) instead of n-butyl iodide only resulted in the hydroarylation product 7 in high yield, rather than the formation of arylalkenylation product (31 or 33). This control experiment might also imply that the Ni(I)/Ni(II)/Ni(III) mechanism is unlikely in our reaction. Given those substrates are reactive in Ni-catalysed 1,1-arylalkylation of alkenes using bipyridine-type ligand via organometallic-radical relay35, these experiments suggested that our reaction might not involve a radical process and our β-diketone ligand might be capable of stabilizing the alkyl-Ni(II) species and inhibiting the homolytic alkyl-Ni(II) cleavage. In addition, replacement of alkyl halides with redox-active ester 34 resulted in no reaction, which further indicated the Ni-catalysed coupling via single-electron pathway is unlikely to be involved in our bulky β-diketone ligand-enabled dicarbofunctionalisation process, given the coupling of redox active ester with Ni(I)/Ni(II)/Ni(III) process is feasible and widely proposed in the literature53. Although the radical clock experiment with vinyl cyclopropylamide (36) mainly gave the ring-opening product 37 in 72% yield, the observation of the 1,2-arylalkylation product 38 (5% yield) further supports that the free alkyl radical species might not be involved in our reaction (the ring-opening rate of cyclopropylcabinyl radical is 1.3 × 108 s−1)54. The ring-opening product 37 might be formed though the β-carbon elimination of corresponding alkyl-Ni(II) species (Fig. 6c). Notably, the in-situ EPR experiment was carried out with our catalytic system, no EPR signal was observed, which might further comply with our hypothesis that the Ni(I) species might be not involved in our reaction (Supplementary Fig. S5). Mechanistic studies. a Control experiments. b Radical trapping experiments. c Radical clock experiments. d Kinetic experiments. e Proposed catalytic cycle. Starting from the alkyl-Ni(II) species, the arylalkylating product 38 could be formed by two-electron oxidation pathway or single electron pathway. To further understand this process, the reaction with (iodomethyl)cyclopropane (39) was carried out, which afforded a 61:39 mixture of the ring-opening product 40 and cyclopropylmethylated 41 in 55% total yield. To further clarify this process, we next performed the reaction with 6-bromo-1-hexene 42 as an alkylating reagent under standard conditions, affording both acyclic (43) and cyclic products (44) in 43% total yield and a ratio of 74/26 (the rate constant for the rearrangement of 5-hexene radicals to cyclopentyl methyl radicals is 1.0 × 105 s−1)54. The observation of both ring-opening products and normal alkylated products in the reaction with (iodomethyl)cyclopropane and 6-bromo-1-hexene indicated that the alkyl-Ni(II) species might be oxidized by alkyl halides through XAT (halogen atom transfer). Preliminary kinetic studies were also carried out to shed light on the roles of each component (alkene 1, arylboronic acid 2a, alkyl iodide 3 and nickel catalyst). Results shown in Fig. 6d indicate that this reaction is first order rate dependence on both arylboronic acid and Ni catalyst. Moreover, altering the concentration of alkyl iodide had no significant effect on the reaction rate. Along with the increase of the concentration of alkene, the rate gradually decreases. These kinetic experiments unveiled that the transmetalation step might be the rate determining step in the catalytic cycle. The apparent negative rate dependence on alkene concentration could be rationalized by the onset of competitive coordination to the intermediate alkyl-nickel species. Based on aforementioned mechanistic studies, we speculate a plausible catalytic cycle, as depicted in Fig. 6e. Firstly, the intermediate Int I was obtained by coordinating the olefin with the nickel catalyst, which underwent transmetalation with the arylboronic acid to deliver an Ar–Ni(II) species Int II. The alkyl-Ni(II) species Int III could be efficiently formed by alkene migratory insertion with the Int II. Next, the alkyl-Ni(II) with alkyl iodine underwent XAT (halogen atom transfer) to produce the alkyl-Ni(III) species Int IV. Next, the arylalkylated product might be formed by a SH2 process, according to the observation by previous reports55–57. However, the formation of high-valent Ni(IV)58–60 species via radial recombination and further reductive elimination could not be fully ruled out by current mechanistic experiments. DFT calculations In order to further elucidate the reaction mechanism, density functional theory (DFT) studies were performed using the B3LYP-D3BJ functional (Supplementary Computational Details). We commenced our studies by searching the possible Ni(II) active catalytic species. With Ni(OTf)2, CsHCO3 base, and L13 ligand, computations show that the formation of Ni(HCO3)(L13’) (where L13’ is the deprotonated enolate form of L13) is highly exergonic by 30.0 kcal/mol (Supplementary Fig. S6). Thus, the Ni(HCO3)(L13’) species, denoted as IN1 (Fig. 7), is taken as the active catalytic species in the calculations. Starting from IN1, coordination of the alkene substrate 10m to Ni through its C-C π bond is exergonic by 7.5 kcal/mol to form a more stable intermediate IN2 (Supplementary Fig. S7). Next, the exchange between arylboronic acid 2a and 10 m would generate IN3, which is 10.3 kcal/mol and 17.8 kcal/mol higher in free energy than IN1 and IN2, respectively. In comparison to the starting material IN1, the formation of a more stable intermediate IN2 indicates the inhibition effect of alkene substrate in the reaction. Sequentially, transmetalation takes place via TS1 from IN3 to give the aryl-Ni(II) species IN4. This step has a free energy barrier of 27.2 kcal/mol and is endergonic by 13.6 kcal/mol with respect to IN2. The exchange between alkene 10m and the formed (HCO3)B(OH)2 produces a stable π-coordinated intermediate IN5, from where the migratory insertion occurs through TS2 to provide the alkyl-Ni(II) species IN6. The migratory insertion step bears a free energy barrier of 13.0 kcal/mol and is exergonic by 4.8 kcal/mol relative to IN5. After that, the binding of the n-butyl iodine to Ni leads to a meta-stable intermediate IN7, which could undergo via two pathways to provide the C–C coupling product. Firstly, IN7 proceeds through twice single-electron oxidation (i.e., via the Ni(III)-nBu radical pair, IN7’) to yield the Ni(IV) intermediate IN8. This step has a free energy barrier of 19.2 kcal/mol and is endergonic by 13.1 kcal/mol with respect to IN6. The free energy profile for the Ni-catalysed alkene dicarbofunctionalisation reactions. Calculations were performed at the SMD//M06/6-311 + + G(d,p)/SDD//B3LYP-D3BJ/6-31 G(d,p)/LANL2DZ level of theory. It is noted that IN8 could also be obtained feasibly through two-electron oxidative addition process via TS3. Next, the reductive elimination of Ni(IV) complex occurs via TS4 to produce the Ni(II) species IN9, with the C–C bond being formed. This step bears a free energy barrier of 25.9 kcal/mol and is exergonic by 24.6 kcal/mol relative to IN6. Finally, ligand exchange is highly exergonic to release the product 11 m and regenerate IN2, from where another catalytic cycle opens. Alternatively, IN7 could undergo through XAT (halogen atom transfer) to produce the Ni(III)-nBu radical pair species, IN7’, followed by the outsphere SH2 process via TS4’ to afford IN9. This step has a free energy barrier of 24.6 kcal/mol with respect to IN6. Thus, calculations show that the SH2 mechanism (via TS4’) is slightly more favorable than those processes involved Ni(IV) species for the C–C coupling step. However, the relatively small energetic preference for the free energy barrier (i.e., 1.3 kcal/mol) could not rule out the possibility of reductive elimination via Ni(IV) intermediate IN8. In summary, computations show that the reaction might proceed via substrate coordination and ligand exchange, transmetalation, migratory insertion, XAT, and sequential SH2 process, among which the transmetalation is the rate-determining step. This is also in agreement with the kinetic experiments. The overall activation-free energy barrier (∆G≠ = 27.2 kcal/mol) indicates that the proposed mechanism is reasonable under the utilized reaction conditions. We have also performed preliminary calculations to explore the possibility of other mechanisms. For the Ni(I)/Ni(III) mechanism, it requires the formation of Ni(I) species to initiate the reaction. Firstly, we have considered the disproportionation process to form Ni(I) and Ni(III) intermediates using two Ni(II) species IN1. However, calculations shows that the reaction free energy is as high as 35.5 kcal/mol, which means that the disproportionation is unlikely (Supplementary Fig. S8). Then, we thought that the aryl-Ni(II)-aryl species, generated through twice of transmetalation, could undergo reductive elimination to yield Ni(0) intermediate, followed by comproportionation with Ni(II) to provide Ni(I) species. However, our calculations show that the second transmetalation step to provide the aryl-Ni(II)-aryl species has a dramatically high free energy barrier (44.7 kcal/mol), which makes it unlikely (Supplementary Fig. S9). We have also considered the direct C-O reductive elimination from the aryl-Ni(II) species, IN4, to form the Ni(0) species. However, this mechanism bears an even higher free energy barrier of 56.7 kcal/mol61,62, ruling out this possibility (Supplementary Fig. S9). Therefore, based on our calculations, the formation of Ni(I) species is not practical. At last, we have investigated an anionic Ni(I)/Ni(III) mechanism63, in which the Ni(II) species IN1 is treated as a Ni(I) species bearing an anionic ligand (Supplementary Fig. S10). Nevertheless, computation on the anionic Ni(I)/Ni(III) pathway shows a high free energy barrier of 48.6 kcal/mol for the migratory insertion step, which also rules out this mechanism. In addition, the homolysis of the Ni–C bond in the Ni(II)-alkyl intermediate is feasible in the presence of bipyridine-type of ligands56, our experimental studies indicates our reaction system with bulky β-diketone ligand might inhibit this process, thus enabling a process via Ni(IV) species rather than a Ni(I)/Ni(III) pathway. To clarify this process, the calculation of homolysis of IN6 was carried out. The preliminary computational results gave a high free energy barrier of 33.8 kcal/mol (via TSd, Fig. 7), and therefore indicating that homolysis is less likely. In conclusion, a nickel-catalysed aryl/alkenyl-alkylation of unactivated alkenes has been achieved with a wide range of native functional group as the directing group. The key to achieve this general aryl/alkenyl-alkylation is the identification of bulky β-diketone ligands, which are capable of stabilizing the in-situ generated alkyl-Ni(II) species and supressing the homolytic cleavage. This protocol has been successfully employed in the efficient synthesis of several bioactive compounds, underscoring the broad synthetic utility and transformative potential of our approach. Developing transition metal-catalysed difunctionalisation reactions of unactivated alkenes with native directing group enabled by the anionic ligands are ongoing in our laboratory. Methods General procedure for Ni-catalysed dicarbofunctionalisation of alkenes In a nitrogen-filled glovebox, substrate 1 or 10 or 12 (0.2 mmol, 1.0 equiv), 2 (0.4 mmol, 2.0 equiv), 3 or 8 (0.4 mmol, 2.0 equiv), Ni(OTf)2 (7.1 mg, 10 mol%), L13 (6.3 mg, 20 mol%), CsHCO3 (77.6 mg, 0.4 mmol, 2.0 equiv) were charged in a 10-mL tube. The tube was sealed using an open-top cap with PTFE cap liner, and moved outside of the glovebox, followed by the addition of tert-amyl alcohol (1.0 mL) under nitrogen atmosphere. The tube was sealed again with parafilm and heated to 80 oC for 10 h. After cooling to room temperature, the mixture was passed through a pad of silica gel with EtOAc as the eluent to remove the nickel and the insoluble precipitate. The resulting solution was concentrated. The residue was then purified by silica gel chromatography or preparative thin-layer chromatography to afford the aryl/alkenyl-alkylation product. Notes: this reaction is sensitive to air and humidity. Full experimental details and characterization of new compounds can be found in Supplementary Information. Supplementary information Supplementary Information Description of Additional Supplementary Files Supplementary Data 1 Peer Review file Acknowledgements We gratefully acknowledge National Key R&D Program of China (2021YFA1500200), National Natural Science Foundation of China (22371293, 22171277, 22101291), Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0610000), Program of Shanghai Academic/Technology Research Leader (23XD1424500), Shanghai Institute of Organic Chemistry (SIOC), and State Key Laboratory of Organometallic Chemistry for financial support. We acknowledge the support from the Youth Innovation Team Program in Colleges and Universities of Shandong Province (2022KJ228), the Taishan Scholar Youth Expert Program in Shandong Province (tsqn202312066), and the Qilu Young Scholar of Shandong University. We also thank H.-B. S. at SIOC for verifying the reproducibility of this work. Author contributions D.-M.W. conducted the experiments and developed the reactions. H.-M.S. and L.-P.X. performed the DFT calculations. L.-Q.S, Y.-Q.H., and Y.W. assisted with some substrate preparation. Y.T. and P.W. directed the project. P.W. conceived the concept. P.W. and L.-P.X. prepared this manuscript with feedback from D.-M.W. and H.-M.S. Peer review Peer review information Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available. Data availability All information relating to optimization studies, experimental procedures, mechanistic studies, NMR spectra, high-resolution mass spectrometry, and computational studies are available in the Supplementary Information. Cartesian coordinates of the calculated structures are available from Supplementary Data 1. All other data are available from the corresponding authors upon request. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Li-Ping Xu, Email: xulp@sdu.edu.cn. Peng Wang, Email: pengwang@sioc.ac.cn. Supplementary information The online version contains supplementary material available at 10.1038/s41467-024-54170-9. References Associated Data Supplementary Materials Supplementary Information Description of Additional Supplementary Files Supplementary Data 1 Peer Review file Data Availability Statement All information relating to optimization studies, experimental procedures, mechanistic studies, NMR spectra, high-resolution mass spectrometry, and computational studies are available in the Supplementary Information. Cartesian coordinates of the calculated structures are available from Supplementary Data 1. All other data are available from the corresponding authors upon request.