General process for upscaling constructions of the aryl iodine precatalyst library. Intrigued by this outlook, we commenced our program on catalyst syntheses. The establishment of aryl iodine precatalyst library started with the previously known and facilely obtained industrial waste (1S,2S)-ANP 5, as summarized in Fig. 2. Thus, amino alcohol 5 was converted to its tert-butyl carbamate 6 in 91% yield and the latter was transformed to the corresponding silyl ether 7 through selective protection at the primary hydroxyl group. Subsequently, a reported Mitsunobu reaction utilizing 2-iodobenzene-1,3-diol 8 and silyl ether 7 enabled approach to the related diether Cat-1 accompanied by an intramolecular substituent aziridine 9. In this step, PPh3 and DIAD were used in excess, and the separation of Cat-1 should be facilitated by the column isolation due to the complicated mixture of triphenylphosphine oxide, DIAD and its reduced byproduct. Therefore, we attempted to perform a concise SN2 substitution to afford Cat-1.63 Based on our previous work and experiment phenomena, we found that the scaffold of ANP and its derivates possessed the unique solubility (e.g., slightly soluble in methanol), leading to access a convenient route without column chromatography isolation. The revised procedure began with a protection to afford a cyclic N-Boc sulfamidite, which was subsequently converted to cyclic sulfamidate 10 in 86% yield over three steps from the product 6. An SN2 substitution of the sodium anion of phenol 8 to the sulfamidate 10 led to a facile nucleophilic cleavage to furnish the intermediate N-sulfate which could be smoothly transformed to the Cat-1 via an acid-quenched step. A subsequent N-Boc deprotected process was readily achieved, resulting in the formation of a key amine structure 11 (> 99.9% ee) up to 106 grams scale. With the high-quality achievement of this upscaling operation, the pivotal scaffold 11 could be successfully transferred to an aryl iodine library bearing abundant of precatalysts with different hydrogen bonds, such as amides Cat-2 to Cat-4 and Cat-6 to Cat-11, ureas Cat-14 to Cat-17, and sulfamides Cat-18 to Cat-21. Moreover, the catalyst bearing the OTBDPS moiety could readily convert to other kind of catalysts, such as the ester (Cat-25 to Cat-34) and ether (Cat-37 to Cat-40) substituted precatalysts (for details, see the Supplementary Information page of P15). In particular, the column chromatograph isolation was unnecessary during the whole purification processes for synthesizing the ANP scaffold derived precatalyst which could be obtained via the recrystallization or precipitation; therefore, our route demonstrated the potential for industrialization. Furthermore, (1R,2R)-ANP could be smoothly utilized to construct the corresponding precatalyst (Cat-24) in high efficiency without column isolation. Notably, D- and L-threonine were also facilely applied to the corresponding aryl iodines (Cat-5, Cat-12 to Cat-13, and Cat-22 to Cat-23) through the similar procedures, and 5-Me/CO2Me substituted 2-iodobenzene-1,3-diols smoothly reacted with the (1S,2S)-ANP 5 to furnish the related catalysts Cat-35 to Cat-36 (for details, see the Supplementary Information page of P16). The structures of 11 and Cat-8 were confirmed by the single crystal X-ray analysis (for details, see the Supplementary Information Figs. 5 and 6).
Applications of the new conformationally flexible aryl iodine catalyst library in asymmetric oxidations. With the new versatile class of precatalysts in hand, various asymmetric oxidations were consequently explored to verify the utility of the catalyst library. Firstly, the oxidative spirolactonization (Kita reaction)40,45,54,64−68 of phenol derivatives, considered as a benchmark reaction for the examination of the catalytic activity of hypervalent iodines, was investigated (Table 1). A series of aryl iodines were loaded to the oxidative conditions, in which m-chloroperbenzoic acid (mCPBA) was used as oxidant at -30 ℃ (Table 1, Entries 1–7). These demonstrated the 2,4,6-trimethylbenzoyl amide substituted catalyst Cat-8 was the best catalyst leading to access the spirocyclic chiral product 13a in 77% yield with 98% ee (Table 1, Entry 4). Temperature investigations found elevating the temperature to -20 ℃ would increase the yield to 82% without erosion of the ee value (Table 1, Entries 4, 8–9). When the TBDPS substituent on the side arm of the catalyst was changed to the acetyl (Cat-33) with less steric hindrance, the efficiency to access desired product 13a would be decreased (Table 1, Entries 9–10). Solvent investigations demonstrated CH2Cl2 was the best solvent in this oxidative transformation (Table 1, Entries 8, 11–12). When five equivalent amounts of ethanol were added as additive to stabilize the hypervalent iodine(III) intermediate,54 the product 13a would be furnished in 92% yield with 98% ee (Table 1, Entry 13). The fewer amount of precatalyst was loaded, the lower yield of compound 13a was afforded (Table 1, Entries 13–15). Increasing the solvent concentration led to slight reduction of the enantioselectivity (Table 1, Entry 16).
With the optimized conditions in hand, we examined the scope of the Kita reaction (Table 2). We initiated our efforts with substituted 3-(1-hydroxynaphthalen-2-yl)propanoic acids 12 to this transformation. 4-Halogeno substituted substrates were smoothly converted to the desired products in high yields with excellent enantioselectivities (Table 2, 13b-13c). The feedstocks bearing electron-withdrawing groups (EWGs), such as benzoyl and acetyl, were successfully transformed to the products 13d-13e in good chemical yields with high enantioselectivities. The 1-naphthol derivatives substituted with both 4-phenyl and 4-benzyl were tolerant of the standard conditions leading to access the desired products with 96% ee and good chemical yields (Table 2, 13f-13g). The utilization of other substrates bearing 4-alkyl and 6-methoxyl gave the corresponding products 13h-13l in moderate to high yields with excellent enantioselectivities. Subsequently, 3-(2-hydroxynaphthalen-1-yl)propanoic acids containing different kinds of functional groups were well tolerated to generate the counterparts 13m-13o in good yields with excellent ee values. Gratefully, gram-scale operation was successfully achieved by using lower catalyst loading of Cat-8 (5 mol%), leading to access 13a in 1.19 g with 80% yield and 99% ee via recrystallization along with recovery of the precatalyst in 96% yield (for details, please see the Supplementary Information page of P50). Moreover, the corresponding catalyst Cat-24 and Cat-22 prepared from (1R,2R)-ANP and L-threonine could facilely be utilized to this reaction system, leading to the generation of products 13g and 13i with opposite configuration in high efficiency, respectively. Importantly, our designed conformationally flexible chiral organoiodine(III) catalyst could also be applied to the oxidative cyclization of alcohols. Because of substrates incorporating EWGs presenting higher reactivity in Ciufolini’s work,68 we focused on investigation of the lower reactive compounds bearing electron-donating groups (EDGs). All these substrates with EDGs could produce the desired products 14b-14f and 14j in moderate to high yield with excellent enantioselectivities. Compared with the reported catalysts, our catalyst presented the highest efficiency in this transformation (14g-14i) illustrating the proposed transition state would be more stable to inhibit its dissociation, which might result from its peculiar chiral pocket and intramolecular H-bond interactions.
Encouraged by the high efficiency of our developed catalyst in the oxidative dearomatization of substituted naphthols, we continued to expand applications of our catalyst library to other type of transformations (Table 3). Firstly, the C-attacked asymmetric spirocyclization of 1-hydroxy-N-aryl-2-naphthamide derivatives 15 was further employed to access functionalized spirooxindoles 16 bearing an all-carbon stereogenic center (Table 3 − 1). In this reaction, the selected precatalyst Cat-9 displayed the outstanding reactivity to furnish the desired products 16a-16d in moderate to good yields with excellent enantioselectivities under milder conditions (-10 oC) by using mCPBA as oxidant, trifluoroethanol (TFE) and water as additives, and nitromethane as solvent (for details, please see the Supplementary Information Table 2). Therefore, our catalyst possessed higher stereoselective efficiency than the previous catalyst reported by Gong and coworkers,69 which further demonstrated the H-bond donor and the multiple tunable site installed on the chiral pocket were necessary. To facilitate the diversified application of our catalyst library, the direct C(sp2)−H/C(sp3)−H cross-coupling was subsequently engaged (Table 3 − 2), in which the anilide derivatives 17 was smoothly transformed to the spirooxindoles 18 by utilizing pivalamido-substitued precatalyst Cat-3, mCPBA as oxidant, CF3CO2H and H2O as additives, and acetonitrile as solvent at room temperature (for details, please see the Supplementary Information Table 3). In this scenario, the more secure and available oxidant mCPBA was employed to replace CH3CO3H that was loaded in the previous work.22 The confused spirooxindoles 18a-18f could be furnished in moderate to high yields (41%-80%) with high enantioselectivities (82%-90% ee). Hence, our developed iodine catalyst further possessed high reactivity and practicability.
The installation of fluorine atom on organic molecules is a significant approach towards the adjustment of molecular properties, such as lipophilicity, membrane permeability, biokinetics, and biodynamics, within agrochemicals, pharmaceuticals, and materials.70–75 Furthermore, hypervalent iodine catalysis has been successfully utilized in asymmetric fluorination, which provides a concise and inexpensive approach to access potential pharmaceuticals. Therefore, the development of new catalyst to achieve the formation of chiral fluorinated molecules is unquestionably beneficial. Our catalyst library was subsequently loaded to the enantioselective fluorination of keto esters 19.46,76 To our delight, the aryl iodine Cat-38 bearing large steric hindrance groups was the optimized catalyst to access the asymmetric fluorination with high efficiency, in which 2,4,6-tribromobenzoyl and triphenylmethyl (Trt) were substituted on the amine and ether moieties, respectively (for details, please see the Supplementary Information Table 4). Hence, the excellent ee values and moderate yields of desired products 20a-20k could be obtained.
Catalyst recovery and recycling experiments in the oxidative dearomatization. Inspired by the concise and utility procedure of catalyst construction, the recovery and recycle of our developed catalyst might be feasible. A significant step in the evaluation of the sustainability characteristics of the precatalyst Cat-8 was the assessment of its recyclability (Table 5). To this end, a sample of Cat-8 was repeatedly utilized in the reaction of substrate 12a in the presence of mCPBA, ethanol, and CH2Cl2. At the end of each reaction of the cycle, the reaction mixture was quenched in the sequence of saturated Na2S2O3 and NaHCO3 aqueous solution. The organic layer was then extracted with CH2Cl2, washed with brine, dried over anhydrous Na2SO4. The mixture was concentrated in vacuo, the residue was subsequently dissolved by 30 mL of MeOH, H2O (10 mL) was finally added to precipitate the catalyst which could enter the next cycle. The filtrate was concentrated in vacuo to get the crude product which could be further recrystallized by the mixture (10 mL, 1:300 volume ratio) of ethyl acetate and petroleum ether (for details, please see the Supplementary Information page of P48). Delightfully, Cat-8 presented greatly high reactivity and selectivity in 10 consecutive cycles, therefore indicating high robustness. Notably, aryl iodine Cat-8 was easily and efficiently recovered from reaction mixture with high efficiency (all cycles > 96% yield) over the whole operation period, while enantiospecificity of product 13a steadily maintained in 97% ee. Some 5.37 g (combined weight) of pure 13a was isolated over ten-cycle operations, which represented an accumulated TON of 83.6. These implements demonstrated our designed catalysts possessed high potential for industrial applications.
Investigation of the significance for H-bond interactions and a tunable chiral pocket. Based on our designed strategy to construct chiral aryl iodine catalyst, both the H-bond interaction and the tunable chiral pocket are essential (Fig. 1b, Int-C). To validate our hypothesis, the controlled experiments were conducted in turn (Table 6). Firstly, when the N−H bonds of the amide moiety on the optimized catalysts (Cat-41, Cat-43, Cat-45, and Cat-47) were changed to N−Me bonds, the yield and enantioselectivity of the corresponding products (13a, 16a, 18a, and 20a) would sharply decline, indicating the H-bond interactions in the key species were fundamental. D-threoninol derived aryl iodine catalysts (Cat-12, Cat-13, Cat-5, and Cat-48) were then applied to replace the (1S,2S)-ANP substituted analogues, leading to slight or dramatic decrease of the reactivities and enantioselectivities. These demonstrate that multi chiral centers were significant for the new catalyst scaffold. Finally, when the substituent R2 (CH2OTBDPS or CH2OTrt) on the side arm of related catalysts (Cat-8, Cat-9, Cat-3, and Cat-38) were replaced by H atom (Cat-42, Cat-44, Cat-46, and Cat-49), the ee values of desired products would be reduced dramatically in most cases, which indicated that the chiral pocket forming from the structure of multi chiral centers via self-assembly was essential. Hence, both the H-bond interactions and tunable chiral pocket were indispensable for these types of catalyst skeleton.