Bholanath
Garai
,
Abir
Das
,
Doppalapudi Vineet
Kumar
and
Basker
Sundararaju
*
Department of Chemistry, Indian Institute of Technology Kanpur, Kanpur, Uttar Pradesh – 208016, India. E-mail: basker@iitk.ac.in
First published on 27th February 2024
While progress in enantioselective C–H functionalization has been accomplished by employing 4d and 5d transition metal-based catalysts, the rapid depletion of these metals in the earth's crust poses a serious threat to making these protocols sustainable. On the other hand, because of their unique reactivity, low toxicity, and high earth abundance, newer strategies utilizing affordable 3d transition metals have come to the forefront. Among the first-row transition metals, high-valent cobalt has recently attracted a lot of attention for catalytic C–H functionalization with mono and bidentate directing groups. This approach was extended for asymmetric catalysis due to a fairly thorough knowledge of its catalytic cycles. Four major themes have been investigated as a result of this insight: (1) rational design of a chiral Cp#Co(III)-catalyst, (2) chiral carboxylic acid with achiral Cp*Co(III)-catalysts using monodentate directing groups, (3) cobalt/salox-based systems, and (4) cobalt/chiral phosphoric acid-based hybrid systems with bidentate directing groups. Herein, we highlight the recent developments in high-valent cobalt-catalyzed enantioselective C–H functionalization up to October 2023, with the strong belief that the current state-of-the-art can attract considerable interest in the synthetic community, encouraging discoveries in the evolving landscape of asymmetric catalysis.
In 2021, the same group reported a chiral CpiPrCo(III)-complex [Co-8] for enantioselective intermolecular carbo-amidation of N-phenoxy acetamides with alkenes, and achieved the target products with excellent ee in moderate-to-good yields (11a, 12b) (Scheme 3a).15 The protocol was amenable to various N-phenoxy acetamides and a wide variety of alkenes like acrylates and bicyclic olefins. Subsequently, Cramer's group further extended the scope of the chiral Cp#Co(III) catalysts for three-component coupling of arylpyrazole (13) with aldehyde (14), and activated olefins (15) (Scheme 3b).16
This strategy provided access to molecules with contiguous stereogenic centres and the expected molecules were isolated with diastereoselectivity ranging from 7:1 to 10:1, besides high enantioselectivity. Such multicomponent reaction did not provide any desired results with the Cp#Rh(III)-catalyst highlighting the unique reactivity associated with a chiral Cp#Co(III)-catalytic system.16
Later, You and co-workers demonstrated the utilization of the chiral Cp#Co(III)-catalytic system for asymmetric ring-opening of strained bicyclic 7-oxa (or aza) benzonorbornadienes (18) with pyrimidyl indoles (17). The scope of indoles and 7-oxabenzonorbornadienes was excellent and the products were isolated in good-to-excellent yields with ee up to 99% (19a and b) (Scheme 4).17 They proposed that the migratory insertion of olefins followed by enantiodetermining cis-β-heteroatom elimination furnishes the expected product with complete cis-selectivity. However, the replacement of the chiral Cp#Co(III)-catalyst [(R)-Co-7] by the racemic Cp*Co(III)-catalyst provided the trans-selective product, indicating that the elimination pathway may depend on the coordination environment. The ring-opening pathway through the nucleophilic attack of [Co]–C to the strained C–O bond could be responsible for the trans-selectivity in the case of the latter (Scheme 4).17
In this regard, chiral carboxylic acids (CCAs) in combination with achiral Cp*M(III)-complexes could be exploited for enantioselective transformations to generate conformationally flexible chiral geometry around the metal centre, which can induce chirality to the target molecules.18,20,33a
To validate this concept, Chang's group explored the use of O,O-dipivaloyl-L-tartaric acid as a chiral carboxylic acid along with Cp*Ir(III) for asymmetric C–H bond functionalization with the observed enantioselectivity of 66:34.21 Subsequently, Cramer and co-workers reported enantioselective C–H bond activation with chiral Cp# ligands in cooperation with chiral carboxylic acids, where the presence of chiral Cp-ligands was essential in achieving the high enantioselectivity.22
A closer look at the coordination environment of the Cp*M(III) complexes revealed that only one coordination site is available for the chiral carboxylate to bind to the metal during the C–H cleavage step.23 Hence, the choice of chiral carboxylic acids and their steric environment around the metal is crucial in achieving high enantioinduction using Cp*M(III) with chiral carboxylic acid catalytic systems.
During the same period, Matsunaga and co-workers demonstrated the use of BINOL-derived chiral carboxylic acid (CCA 2) in the presence of the achiral Cp*Co(III)-catalyst for asymmetric addition of an alkenyl–[Co] bond to maleimides (24) with indole in a moderately enantioselective manner.
The stereoselective insertion of a [Co–C] moiety of N-protected indole (23) to maleimides (24) was controlled by the CCA. The reported catalytic conditions provided moderate enantiomeric excess, which may be likely due to the enolization of the α-methylene hydrogen; however, the protocol afforded excellent yields of the corresponding product (Scheme 5b).25
In 2021, Shi's group reported a Cp*Co(III)-catalyzed highly enantioselective hydroarylation of un-activated terminal alkenes (26) with N-protected indoles (17). They displayed a wide range of substrate tolerability in moderate-to-excellent yields and ee of up to 98%. They demonstrated that N-phthaloyl-protected non-canonical bulky amino acids were effective in inducing high enantiomeric excess in product (CCA 3). The developed protocol was effective for the intramolecular annulation of N-protected pyrroles and indole in moderate-to-good yields with good enantiomeric excess (27b). They reasoned that the π–π stacking between the electron-deficient directing group and electron-rich amino acid can stabilize the enantio-determining transition state and create a specific chiral pocket for high enantio-induction (TS-2) (Scheme 5c).26
Recently, Ackermann and co-workers introduced a data-driven model to identify efficient CCAs for enantio- and diastereoselective C–H alkylation of indoles under achiral cobalt catalysis. The machine learning approach assisted them in identifying the favourable transition state with effective CCAs in a short time and subsequent validation of those results led to high enantio-induction. They provided a library of C-central and C–N axial chiral molecules in excellent yields and ee of up to 95% (29a–d) (Scheme 5d).27
In this regard, in 2019, Matsunaga and co-workers illustrated the first example of a hybrid catalytic system for the enantioselective C(sp3)–H amidation of thioamides under the Cp*Co(III)/CCA catalytic system.28 An achiral Cp*Co(III)-catalyst in combination with amino acid-derived CCA 5 leads to the expected amidated product in good enantiomeric excess. The scope of the reaction was very general providing the target molecules in 50–99% yields and up to 88% ee (32a to 32d). Also, sterically bulkier CptBuCo(III) with CCA 5 turned out to be the most effective combination to deliver the expected molecules with moderate-to-excellent yields and good ee (Scheme 6a). The scope of the reaction was extended to a variety of thioamides (33) and dioxazolone derivatives (31). Moreover, non-cyclic amides were also amenable resulting in good ee and good yield (32b) (Scheme 6a).28 In addition to the exploration of C–N bond formation, a parallel effort was put forth to apply for asymmetric C–C bond formation.
In 2019, Matsunaga et al. exploited the inherent chirality associated with ferrocenes. Rationally designed new chiral carboxylic acids based on 1,2-disubstituted chiral ferrocene carboxylic acids that ensured only enantioselective C–H activation with the achiral Cp*Co(III)-complex were also effective in achieving high enantiomeric excess under mild conditions. The systematic study allowed the authors to discover a series of new planar chiral carboxylic acids CCA 6–9, which can be easily accessed using the reported C–H arylation strategy (Scheme 6b). Among the CCAs screened, CCA 9 was found to be the best candidate for both reactivity and enantioselectivity (Scheme 6b).29
In 2019, Shi's group reported a thioamide-directed C(sp2)–H amidation for the construction of planar chiral molecules under Cp*Co(III)-catalysis using commercially available monoprotected amino acid (D)-Bz-Hpg–OH (CCA 10) as a chiral carboxylic acid. The isolated products were obtained in excellent yields with observed enantio-induction of up to 77.5:22.5 er (Scheme 7a).30 The coordination of thioamide-directed C–H activation is crucial, and it turns out to be the enantio-determining step. Among the solvents screened, ethanol was found to be the best solvent to eliminate the other background reactions.
Subsequently, Matsunaga et al. reported a sulfoximine-directed enantioselective synthesis of benzothiadiazine-1-oxides via C–H activation in a combination of achiral Co(III)-complex and CCA 11. The newly designed (S)-BINOL-derived pseudo-C2-symmetric H8-binaphthyl CCA provided the target products in high yields and with an enantiomeric ratio of up to 98:2 (38) (Scheme 7b).31 The protocol was applied to various diarylsulfoximines (37) and dioxazolone derivatives (31) and provided moderate enantioselectivity (38a and 38b). Control experiments and mechanistic investigations suggested that the C–H bond cleavage was the rate-limiting step and the DFT study further confirmed that it was also the enantio-determining step (TS-5) (Scheme 7b).31
Later, Shi's group showcased the use of amide-hydrogen bonded, carboxylate-assisted enantioselective C–H activation of sulfoximines in the presence of achiral Cp*Co(III) and CCA 12. The DFT calculations suggested that the C–H activation was the rate-determining step and free –NH was crucial for the reaction as it involved hydrogen bonding with the amide at the proximal position (TS-6) (Scheme 7c).32
In 2014, Daugulis’ group reported a cobalt-catalyzed C–H annulation of alkynes using a bidentate directing group, starting from simple and commercially available cobalt(II) salts.7c Since then a plethora of examples have been reported over the last ten years including the isolation of an active octahedral cobaltacycle [Co-12] involved in the catalytic cycle.36 The close examination of the octahedral cobaltacycle revealed that cobalt was in the +3 oxidation state and bound with one tridentate N, N, C type coordination and N,N-coordination of 8-aminoquinoline as L, and one X-type bidentate spectator ligand along with the solvent at the axial position (Scheme 8a).
Since the spectator ligands in the octahedral complex bind to the metal to stabilize the 6-coordinate geometry, also the spectator ligand stabilizes the Co(I)-intermediate upon reductive elimination.36 These coordination sites can potentially be replaced by external bidentate ligands to enhance the reactivity and selectivity including asymmetric induction (Scheme 8b).
In 2022, Shi's group demonstrated an elegant work of replacing the N,N-spectator ligand of the substrate with an external chiral monoanionic bidentate N,O-ligand based on salicyloxazoline (salox), which can mimic the coordination nature of in situ- generated monoanionic 8-aminoquinoline, and at the same time induce a rigid chiral environment around cobalt (Scheme 8b).35 Shi and co-workers demonstrated that the asymmetric C–H bond annulation was able to achieve chiral molecules using simple, commercially available, inexpensive cobalt(II) salts along with (S)-salox ligand L1 using phosphinamide and alkyne. Shi et al. proposed that the C–H activation is the enantio-determining step and provides the chiral annulated compounds in up to 99% ee. The reaction tolerated various terminal and internal alkynes with electron-donating and withdrawing groups, and the resultant products were obtained with high regio- and enantioselectivities besides high yields (43a–d). Alkynes containing ferrocene, estrone, and pharmaceutical moieties like ezetimibe were amenable to the reaction. Various electronically diverse diaryl phosphinamides were tolerated. Furthermore, a gram-scale reaction was performed to result in product formation with a 97% yield and > 99% ee with 1 mol% loading of the cobalt source (Scheme 9a).35
To validate the proposed proof-of-concept, the authors isolated the octahedral complex featuring (S)-L1 and two acetylacetonate (acac) ligands. Among the many possible diastereomers, the authors isolated the octahedral [(Λ/Δ) Co(acac)2((S)-L1)] complex in Λ- and Δ-diastereomers in a 15:1 ratio and both ΛCo and ΔCo complexes were structurally characterized (Scheme 9c). Solvents such as tBuOH and tAmylOH provided the highest ratio of 15:1(Λ/Δ) and they reported that the ratio may vary depending on the solvent employed. Replacement of one acac ligand with phosphinamides featuring 8-aminoquinoline did not change the ratio of the Λ/Δ diastereomers (Scheme 9b) indicating that ΛCo-(SC)-Co-13 is a major diastereomer formed before C–H activation. However, both ΛCo and ΔCo complexes were catalytically active and provided the same annulated molecules without change in the absolute configuration of the isolated chiral product indicating that upon C–H activation, only the major diastereomer formed among other possible diastereomers generated in situ in the reaction. To further ascertain the hypothesis based on the catalytic outcome with ΛCo and ΔCo as catalysts, the authors further explored the isolation of the cobalt complex with the 8-aminoquionline bound phosphinamide as shown in Scheme 9. It is important to highlight that the ligand exchange with acac in Co(acac)2 first occurs with salox L1 and subsequent cyclometallation leads to a ΔCo-(SC)-Co-13 octahedral complex in contrast to the expected retention of the configuration with rac-phosphinamide that leads to ΛCo-(SC)-Co-14. The highly diastereoselective cobaltacycle mer-(SC, SP)-Co-15 (Scheme 9e) (at this stage, it was not clear whether the isolated cobalt complex was Δ or Λ) was isolated with sodium pivalate as the carboxylate source and 4-MeO-Py as the stabilizing ligand starting from racemic phosphinamide, (S)-L1 and Co(II) precursors. According to them, enantio-determining C–H activation results in a mer-(SC, SP)-Co-15 complex as a major diastereomer. However, the energy differences between cobaltacycles such as mer-ΔCo-(SC, SP)-Co and mer-ΛCo-(SC, SP)-Co and mer-ΔCo-(SC, SP)-Co and mer-ΔCo-(SC, RP)-Co are not discussed in detail. Also, Shi and co-workers proposed that major diastereomer (mer-(SC, SP)-Co-15) was stabilized through the π–π interactions between the phenyl groups of L1 and 8- aminoquinoline, and one of the phenyl groups of the phosphinamide upon cyclometallation.
Furthermore, they explained that the steric repulsion between the phenyl group of alkynes and the aryl group bound to phosphorus forced the sterically bulkier aryl group away, resulting in one single regio-isomer as the major one over the other isomers, especially with unsymmetrical internal and terminal alkynes (Scheme 9).35
At that time Niu's group independently discovered an elegant strategy for the synthesis of a C–N axially chiral compound containing an isoquinolinone motif in a single step in combination with Co/salox (L2) catalytic conditions with benzamides bearing a C–7 substituent of quinoline ring with alkynes. The reaction conditions were tolerated for a variety of internal and terminal alkynes and a variety of benzamides with excellent yields of up to 98% and enantioselectivity of up to 98% ee. Interestingly various substituents at C–7 of the quinoline ring allowed access to chiral isoquinolinones with good to excellent yields and enantiomeric excess (46). In addition to that, a mechanistic study indicates that C–N reductive elimination is the stereo-determining step. Also, the resulting product shows a high-temperature resistance for atropostability. The obtained product (46a) was used for Pd-catalyzed C–H functionalization of indole (Scheme 10a).34b
At the time, the Shi group further extended their initial discovery to other enantioselective C–H activation reactions. For example, C–H alkoxylation and amination was achieved using 8-aminoquinoline-derived diphenylphosphinamide in the presence of chiral salox ligands (L1, 3 or 4) in combination with a Co(II)-salt (Scheme 10b).37 The substrate scope of alcohol, amine, and different diphenylphosphinamides was very general with various functional groups and producing the target product in excellent yields with very high enantio-induction (47 and 49). Mechanistic investigation showed that the chiral octahedral cobaltacycle stabilized with a 4-acetylpyridine ligand was catalytically inactive under oxidant-free conditions even at 120 °C. In the presence of air, the mono alkoxylated product was obtained in 15% yield and 99% ee and also they isolated the bis-alkoxylation product using Ag2CO3 as an oxidant in good yield with 99% ee. These experiments supported the hypothesis that the reaction was proceeding through oxidation-induced reductive elimination pathways and high-valent Co(IV) might be involved, which was earlier proposed by Ackermann and Hong through computational studies (Scheme 10b).38,39
Di-axial or multi-axial chirality was not much explored in the literature and only a few examples were documented in the literature. In 2022, Shi and co-workers disclosed an enantioselective synthesis of atropisomers with vicinal C–N and C–C chiral biaxiality in a single molecule. Judicious choice of benzamide and chiral salox L1 in combination with suitable cobalt salts afforded sterically congested axially chiral molecules in 99% ee with a dr of >7:1 (51a–d). This approach holds a high potential for accessing sterically crowded benzamide derivatives (Scheme 10c).40
In 2023, Niu and co-workers demonstrated the asymmetric synthesis of dihydroisoquinolones using benzamides and alkenes in the presence of a Co/salox (L6 or 7) hybrid catalytic system. The fast kinetics of the reaction afforded the products in just 10 min and the applicability of the protocol was extended to aliphatic, aromatic, symmetric and un-symmetric alkenes, cyclic dienes, and benzamides having –CO2Me, –CF3 and –OMe functionalities in high yields and enantioselectivity. Also, heterocyclic (53a), and aliphatic amides (53b) were amenable to delivering the corresponding enantioenriched products (Scheme 10d).41
Recently, our group demonstrated a reversing of the regioselectivity of asymmetric C–H and N–H bond annulation of diaryl phosphinamides with bromo-alkyne, and a carboxylate nucleophile with the Co/salox (L5) catalytic system. The developed reaction conditions were amenable to a variety of bromo-alkynes (54), and diaryl phosphinamides (41), and the expected products were isolated in good-to-excellent yields with ee up to 99% (55). The catalytically active chiral Oh-cobaltacycle (S)-[Co-16] was isolated and a stoichiometric reaction with alkynyl-bromide resulted in an annulated product with carbon attached to bromine, alpha to the nitrogen atom (56). This vinyl bromide was further subjected to the standard reaction conditions yielding the expected OPiv product (55e) suggesting the involvement of low-valent [CoI] species and the in situ-generated [CoI] species after the annulation underwent oxidative addition with vinyl bromide; subsequently, ligand exchange and reductive elimination gave the expected product (Scheme 10e).42 Furthermore, DFT studies suggested that the energy difference between cobaltacycles such as mer-ΔCo-(SC, SP)-Co and mer-ΛCo-(SC, SP)-Co was found to be 12.4 kcal mol−1T. The difference in energy between TS-mer-ΔCo-(SC, SP)-Co orientations is found to be 5.44 kcal mol−1 lower in energy compared to the TS-mer-ΔCo-(SC, RP)-Co of the phosphorus suggesting that the (S) isomer is more favored and the theoretically predicted ee is found to be 99.9%.
A closer look at the high-valent cobalt-catalyzed C–H bond functionalization using the bidentate directing group revealed that the use of Mn(III) or Ag(I) as an oxidant in a stoichiometric amount was unavoidable. This was circumvented in recent times either by using electricity as an oxidant or a cobalt catalyst combined with photosensitizers and oxygen under light irradiation. While the former method has been extensively explored by the groups of Ackermann, Lei, and others for C–H bond functionalization, an extension of such strategies for asymmetric C–H bond functionalization, it is in its infancy due to challenges such as (a) conformational stability in [Co]-complexes associated with the ligand, (b) electrolyte interference in the enantio-determining transition state, and (c) unexpected oxidation of functional groups or the substrate in the presence of electricity that may occur in asymmetric electrocatalysis.51,52
With the above challenges in mind, first Ackermann and co-workers reported the electro-oxidative asymmetric C–H activation using a [Co]/salox (L1, 8, or 9) system with various π-coupling partners, amines, and alcohols (Scheme 11).50 The reaction was shown to be highly regio-, stereo-, and chemo- selective to produce P-stereogenic (43, 47 and 58), atroposelective (59), and C-stereogenic (61), molecules in high yields with excellent enantio-induction. Benzamides derived from C7-substituted quinolines (44) were amenable to annulation with alkynes and the expected atropisomeric products were obtained in excellent yields with up to 99% ee.
Besides, various benzamides (52) were also annulated with cyclic alkenes to furnish a group of central chiral molecules (61) in good yields and up to 99.5:0.5 enantiomeric ratio (Scheme 11).50 Around the same time, Shi's group disclosed an electro-oxidative annulation of benzamides and alkenes in the presence of sterically bulky salox (L10) and Co(II)-salts at room temperature. The scope of various α-aryl olefins was suitable for chiral dihydroisoquinolone (c-DHIQO) skeleton synthesis (53).51 Challenging and unbiased α-aliphatic alkenes were not amenable, but a judicious choice of pyridine ligands generated a highly regio- and enantiocontrolled reaction outcome. Notably, in the presence of a 3,4,5-trichloropyridine ligand, a π–π interaction between the benzene of benzamide and benzene of the ligand was crucial to get stable catalytically active organometallic species. The corresponding catalytically active chiral octahedral cobaltacycles were isolated (a) without any stabilizing ligand, (b) with 3,4,5-trichloropyriine, and (c) with MeOH (Scheme 12a).51
The same group further extended their work on electrochemical enantioselective C–H oxygenation of diarylphosphinamides using Co(II)-salt and chiral salox (L1) ligand. A range of alkoxylated chiral phosphinamides were isolated with good yields and up to >99% enantiomeric excess (49c, d). This method obviated the use of a stoichiometric silver oxidant producing hydrogen gas, thereby rendering the process more environmentally friendly. The involvement of the [CoIV] intermediate upon binding with OMe was further supported by cyclic voltammetry (Scheme 12b).52
In continuation of the earlier work, Ackermann's group reported an electrocatalytic enantioselective synthesis of C–N axially chiral amides in combination with Co(II)-salt/chiral salox (L2 or L7) in an undivided cell. The catalytic conditions could be used to access atroposelective amides through C–N axial chirality in a single step (63a–c). Furthermore, the same methodology was applied for the synthesis of a variety of P-stereogenic compounds in excellent yields (up to 99%) and 99% ee (Scheme 12c). This protocol was amenable to diverse electronically and sterically biased benzamides and allenes (64a–c). Also, the protocol was extended to late-stage diversification of drug molecules. The synthetic potential was further extended to a continuous flow reactor setup without any supporting electrolyte yielding the same stereo-electronic product with a slight modification of the reaction conditions (Scheme 12c).53
In 2023, Ling and co-workers documented the electrochemical synthesis of P-stereogenic cyclic phosphinamides (65) with the aid of chiral salicyloxazoline (L1) and cobalt acetate tetrahydrate as an inexpensive catalyst with moderate-to-excellent yields and ee up to 99%. The reported reaction conditions displayed a diverse range of phosphinamides and alkynes with good functional group tolerance. In addition, a good kinetic resolution with an s-factor of up to 347.8 was obtained (Scheme 12d).54
In 2021, Shi's group reported a cobalt(II)-catalyzed enantioselective spirocyclization through C–H olefination and spirocyclization in the presence of chiral phosphoric acid (sCPA-1). The reaction was very specific to spirochiral phosphoric acid and the solvent system. The catalytic conditions provided a library of spiro compounds with 41–70% yields and up to 98% ee (67a and b). Also, the methodology was extended to the synthesis of chiral aldose reductase inhibitors in a 60% yield and 91% ee.63 Mechanistic investigation showed that stereochemical 1,4-addition of [Co] might be the enantio-determining step (TS-8) (Scheme 13a).63
Recently, Ackermann and co-workers demonstrated spiro cyclization through electrochemical cobalt-catalyzed C–H activation in the presence of chiral phosphoric acid. Here, (S)-BINOL-derived chiral phosphoric acid was used to carry out the reaction in moderate yield and high enantioselectivity (Scheme 13b).50 Notably, the authors observed the opposite stereoisomer under the combination of electricity and CPA-2 conditions. Furthermore, the ultraviolet-visible spectro-electrochemical analysis provided evidence that the reactions might proceed via the Co(III/IV/II) pathway, and the corresponding intermediate was detected by HRMS (TS-9) (Scheme 13).50
Subsequently, Shi and co-workers developed a new methodology for the desymmetrization of benzyl-protected picolinamides (PAs) through C–H bond annulation in a combination of the Co(II)/salox catalytic system. Fine-tuning of the directing group to picolinyl led to an electronically and sterically similar chiral Oh cobaltacycle [Co-18], which could be exploited for the synthesis of chiral-diaryl methylamine (DAMA) (Scheme 14a).64 A variety of chiral DAMAs were accessed through C–H alkoxylation in the presence of Co(II)-salt and salox (L2) with good-to-excellent ee with moderate-to-excellent yields (69). The reaction proceeds through sequential processes such as desymmetrization, kinetic resolution, and parallel kinetic resolution to afford the expected methoxylated product in an excellent yield (up to 90%) and 99% ee (Scheme 14b). Also, the obtained products were derivatized into value-added molecular architectures (Scheme 14).64
Scheme 14 Synthesis of chiral diarylmethylamines by cobalt-catalyzed enantioselective C–H alkoxylation. |
C–N axially chiral molecules have applications in the agricultural and medicinal areas; therefore, C–N chiral sulphonamide is a unique skeleton in the broad area of organic synthesis and synthesis of such molecules in an asymmetric manner is holding the attention of the scientific community.65–67 Also, the construction of chiral sultam-based scaffolds was not much explored in the literature. In 2023, Niu and co-workers reported a Co/salox (L6) based catalytic system for C–H bond annulation of sulfonamides with allenes/alkynes with excellent enantiomeric excess and moderate-to-good yields (Scheme 15a).68 The developed methodology was applied to access a variety of chiral sultams in one step (70 and 71). At the same time, they developed a C–N axially chiral hetero-biaryl isoquinolinone skeleton construction via cobalt-catalyzed atroposelective C–H activation/annulation of benzamides and allenes with excellent yields and enantiomeric excess under thermal conditions (Scheme 15b).69 The work was further extended without any external oxidant using an environmentally begin enantioselective electro-organic process (63d).69
Scheme 15 C–N axially chiral hetero-biaryl skeleton construction via cobalt-catalyzed atroposelective annulation. |
In continuation of their work, the same group further developed an unprecedented synthesis of C–N axially chiral cyclized products of benzamide (44) and isonitriles (72) as a C1 carbon source under the cobalt/salox (L2) catalytic system. The isolated products were accessed with high yields and ee (Scheme 15c).70
We have discussed the construction of C–N axially chiral five- or six-membered N-heterocycle architectures in an atroposelective manner via thermal as well as under electro-oxidative conditions. The construction of C–C and C–N axially chiral frameworks is relatively easy because the rotational barrier about such bonds is sufficiently high in comparison to others. Atropisomerism around the N–N bond is a long-lasting problem in the scientific community, because of the easy epimerisation of the N–N bond. However, judicious choice of substrates might restrict the rotation around N–N and result in atropisomerism.71
Recently, Niu and co-workers established a protocol that renders N–N axially chiral isoquinolinones through cobalt-catalyzed atroposelective C–H activation/annulation. They constructed the N–N axially chiral architectures with excellent ee and yields (76). A variety of benzamides (74) and alkynes (75) were tolerated with excellent efficiencies. Moreover, such complex chiral molecules could be synthesized even in the presence of electrochemical conditions for asymmetric synthesis in a combination of cobalt/salox (L10) catalytic system with high ee and reasonably good yields (76c and d) (Scheme 16).71
Dynamic kinetic resolution (DKR) is a powerful process to obtain the racemic molecules with a quantitative yield. The process could proceed via two different diastereomeric transition states [Co-20A and Co-20B], which are interconvertible. However, only one transition state [Co-20] provides the favoured product with an energetically favoured transition state.72
In 2023, Shi and co-workers demonstrated an unprecedented cobalt/salox catalytic system for the DKR of racemic biaryls as shown in Scheme 17. The (rac)-biaryl would coordinate to the [Co]/salox (L11) system and generate two diastereomers [Co-20A] and [Co-20B]. More stable transition state [Co-20B] would lead to a chiral octahedral cobaltacycle [Co-20] (C), which subsequently furnishes the expected chiral product (aS)-3 (Scheme 17a).73 This dynamic kinetic asymmetric transformation (DYKAT) was applied to racemic picolinamide-protected biaryl amines with aryl boronic acids under cobalt/salox(L11) hybrid catalytic conditions. The scope of the biaryl and aryl boronic acid derivatives was versatile, tolerating hydroxy (79a), silyl (79b), and terminal alkynes (79c) with excellent enantiomeric excess and moderate-to-excellent yields of the chiral arylated products (Scheme 17b).73
While research was exploding with numerous transformations utilizing both mono- and bidentate directing groups under two different Co(III) catalytic systems, the development of chiral carboxylic acid combined with Cp*Rh(III) or Cp*Ir(III) for asymmetric C–H bond annulation was reported by Matsunaga. The utilization of such an approach with high valent cobalt was first reported by Ackerman and co-workers using C-2 symmetric imidazole-based chiral carboxylic acids. Later, this approach was further explored by various groups for asymmetric C–H bond functionalization using mono-dentate directing groups, where chiral carboxylic acids were used as the sole chiral source along with the Cp*Co(III)-catalytic system. Around the same time, Cramer reported the C-2 symmetric chiral Cp-ligands and their cobalt complexes in C–H bond annulation and amine transfer reactions with good yields and ee of up to 99%.
On the other hand, very recently, Shi reported an elegant method for enantioselective C–H functionalization utilizing a Co(II) salt, a bidentate directing group, and chiral salox as external ligands. In the same year, Niu and Shi reported several asymmetric C–H bond functionalizations using the same catalytic system. However, the use of stoichiometric amounts of Mn(III) is unavoidable. This was independently circumvented by the groups of Shi and Ackerman through electrochemical enantioselective C–H functionalization via anodic oxidation of Co(II) to Co(III) in the absence of metal oxidants. However, an alternative benign route by combining Co(II) with photocatalysts under oxidant-free conditions is yet to be explored for the asymmetric version. The mechanistic investigation revealed that mer-[(Sc), (Sp)-ΔCo] or mer-[(Sc)-ΔCo] is the major diastereomer among the various diastereomers in the case of phosphinamide- or benzamide-based cobaltacycles. There is a possibility that this strategy will likely be explored for asymmetric C(sp3)–H bond functionalization, perhaps possible to operate under mild conditions.
These studies further provide insights into the design of new directing groups, catalysts, and new cobalt complexes for non-directed asymmetric C–H bond functionalization with external ligands. We hope that this field has tremendous potential to improve more in the coming years and we may witness numerous transformations by judiciously altering the coordination environment of the octahedral cobalt complex.
This journal is © The Royal Society of Chemistry 2024 |