Chiral Cobalt(II) Complex Catalyzed Asymmetric [2,3]-Sigmatropic Rearrangement of Allylic Selenides with α-Diazo Pyrazoleamides

Abstract

Open AccessCCS ChemistryCOMMUNICATION1 Apr 2021Chiral Cobalt(II) Complex Catalyzed Asymmetric [2,3]-Sigmatropic Rearrangement of Allylic Selenides with α-Diazo Pyrazoleamides Xiaobin Lin, Zheng Tan, Wenkun Yang, Wei Yang, Xiaohua Liu and Xiaoming Feng Xiaobin Lin Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu 610064. Google Scholar More articles by this author , Zheng Tan Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu 610064. Google Scholar More articles by this author , Wenkun Yang Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu 610064. Google Scholar More articles by this author , Wei Yang Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu 610064. Google Scholar More articles by this author , Xiaohua Liu *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu 610064. Google Scholar More articles by this author and Xiaoming Feng *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu 610064. Google Scholar More articles by this author https://doi.org/10.31635/ccschem.020.202000345 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail Organoselenium compounds, due to their high structural diversity, special function, and biological activities, have drawn attention in synthetic chemistry. Herein, a novel example of chiral N,N′-dioxide/cobalt(II) complex catalyzed asymmetric [2,3]-sigmatropic rearrangement of allylic selenides with α-diazo pyrazoleamides is disclosed, which represents a highly efficient approach to optically active selenides bearing a quaternary C–Se stereocenter. Most of the reactions proceed with 0.5–2 mol % catalyst loading in an inert-free gas atmosphere, and a wealth of chiral selenides are obtained in up to 99% yield and 97% enantiomeric excess (ee). The control experiments demonstrate the high reactivity of allylic selenides, as well as the conspicuous superiority of chiral N,N′-dioxide ligand and α-diazo pyrazoleamide in [2,3]-sigmatropic rearrangement. The mechanism studies reveal that the key to asymmetric rearrangement of allylic selenium ylides is the transfer of chirality from the stable chiral selenium to the carbon of the product. A feasible catalytic cycle is proposed as well. Download figure Download PowerPoint Introduction Organoselenium compounds have emerged as powerful reagents, intermediates, and even catalysts in organic synthesis.1–14 For instance, significant achievements of chiral selenium catalysts in asymmetric catalysis have been made by Yeung’s15,16 and Zhao’s group17–22 recently. Moreover, a plethora of synthetic organoselenium compounds have drawn much attention in the medicinal domain due to their distinctive biological activities.9,12 Despite the great efforts that have been devoted to the construction of C–Se stereocenters with chiral starting materials,1–8,15–22 the enantioselective catalytic route to novel optically active organoselenium compounds is less known and investigated.23–30 The asymmetric catalytic methods are appealing in terms of versatile reactivity and interesting stereochemistry of selenium during the process. Asymmetric α-selenenylation of aldehydes and β-keto esters have been reported by the use of electrophilic selenium sources (Scheme 1a).24–28 The kinetic resolution of selenocarbonates via a chiral palladium-catalyzed decarboxylative coupling to provide optically active allylic selenides was achieved (Scheme 1b).29 Recently, an asymmetric hydroselenation of heterobicyclic alkenes was developed to afford chiral selenol-incorporated adducts.30 Scheme 1. | (a–c) Examples of asymmetric catalytic construction of C–Se stereocenter. Download figure Download PowerPoint The [2,3]-sigmatropic rearrangement of allylic chalcogen ylides allows efficient carbon–carbon formation.31–35 For example, asymmetric catalytic Doyle–Kirmse reaction via the sulfonium ylides, catalyzed by various catalysts, including the high-profile chiral copper(I) and dirhodium(II) complexes, as well as our chiral N,N′-dioxide/nickel(II) complex, has been well studied in recent years.36–49 In principle, the stereochemistry in arrangement of allylic selenium ylides is similar to the sulfur analogs, and the discrimination of the heterotopic lone pairs on chalcogen nucleophiles, as well as the stereogenic center at chalcogen ylides, is crucial to the newly formed carbon–carbon bond via a rearrangement.49 Although selenium and sulfur have similar electronegativity and radii, in contrast, the stereoselective arrangement of the allylic selenium ylides remains quite limited since the seminal work by Nishibayashi et al. in 1995.36,50 In relation to the bond energies, the carbon–selenium bond is weaker than the carbon–sulfur bond, and the stability of the related selenium ylides enhances difficulties of asymmetric [2,3]-sigmatropic rearrangement of allylic selenides.1 Recently, our group has demonstrated that, in combination with the chiral N,N′-dioxide/nickel(II) complex and a type of α-diazo pyrazoleamide, three types of [2,3]-sigmatropic rearrangements of sulfur ylides could be achieved efficiently and enantioselectively.49,51–58 The catalytic system shows excellent stereocontrol for the discrimination of the heterotopic lone pairs on sulfur and [2,3]-sigmatropic rearrangement process. We conjectured whether the construction of a quaternary C–Se stereocenter could be achieved by the introduction of allylic selenides into the catalytic system (Scheme 1c). Surprisingly, it was later found that chiral N,N′-dioxide–cobalt(II) complex was appropriate for this process. In most previous cases, metal catalysts including Rh2, Cu, Ir, and Fe complexes are favored for carbene formation,36–48,59–63 but Co(II) complex was unique in the cyclopropanation of α-diazo carbonyl compounds and others, which was disclosed as a metalloradical catalyst by Dzik et al.64 and Lu et al.65 Herein, we report a highly enantioselective [2,3]-sigmatropic rearrangement of selenium ylides generated from allylic selenides and α-diazo pyrazoleamides. The potentiality of chiral cobalt(II) complex in the ylide formation and rearrangement was developed. Various selenides bearing a quaternary C–Se stereocenter were obtained in decent yields and excellent enantioselectivities. Experimental Methods A dry volumetric flask (2.0 mL) was charged with N,N′ -dioxide (0.02 mmol) and Co(II) (0.02 mmol). Then tetrahydrofuran (THF; 2.0 mL) was added and the mixture was stirred at room temperature for more than 2 h before use. Then the solution (50–500 μL, 0.5–5 mol %) was transferred to a dry test tube according to the catalyst loading of each reaction and THF was removed in vacuum. Subsequently, CH2Cl2 (1.0 mL), selenide 2 (0.1–0.15 mmol, 1.0–1.5 equiv), and diazo compound 1 or 4 (0.1 mmol, 1.0 equiv) were added successively at the corresponding temperature. After the diazo compound 1 or 4 was consumed [detected by thin-layer chromatography (TLC)], the residue was purified by column chromatography on silica gel to afford the product (Pet/EtOAc = 80/1–50/1 as eluent). The racemic products were prepared with race -L2-PiPr2/Ni(OTf)2 complex as catalyst. The purification processes were the same as those for the chiral products. Results and Discussion Initially, we selected vinyl-substituted α-diazo pyrazoleamide 1aa and allyl(phenyl)selane 2a as the suitable substrates. The mixture of chiral N,N′-dioxide L2-PiPr2, NiCl2, and AgNTf2 was used as the catalyst, which was the best one for asymmetric catalytic Doyle–Kirmse reaction of allylic sulfides in our previous study (Table 1, entry 1).49 The desired product 3aa was given in 84% yield and 84% enantiomeric excess (ee) within 5 min. No obvious effect on the reactivity and enantioselectivity was observed when various silver(I) salts were examined ( Supporting Information Table S1). Subsequently, we switched to optimize the Ni(OTf)2 complexes of chiral N,N′-dioxide ligands derived from different amino acids and anilines ( Supporting Information Table S2, entries 2–4). The reactions could be finished in 10 min, affording the rearranged product 3aa in good yields. Notably, the introduction of two alkoxy substituents into the N,N′-dioxide motifs offers an additional dimension for diversification and fine-tuning. The chiral ligand L2-Pi(O i Pr)2, bearing 2,6-isopropoxy groups on the amide unit, exhibited excellent reactivity (2 min, 84% yield) and enantioselectivity (94% ee, entry 4). Then, we embarked on the screening of various metal salts using L2-Pi(O i Pr)2 as the chiral ligand. Unexpectedly, it was found that cobalt(II) salts, including Co(ClO4)2·6H2O, Co(BF4)2·6H2O, and Co(OTf)2, showed better performance than the nickel salts, yielding 3aa in 89% yield and 97% ee within 1 min, regardless the counterion of the cobalt salts (entries 5–7). Additionally, there is an obvious ligand-accelerating effect because only 31% yield of the desired product was detected 40 min later without any ligand. Naturally, we settled on Co(BF4)2·6H2O/ L2-Pi(O i Pr)2 complex,b one of the optimal catalysts, to investigate other reaction parameters, such as other metal salts, additives, and reaction concentration, but no better results were obtained ( Supporting Information Tables S3–S5). It is worth noting that Lewis acids, such as Ca(OTf)2, Mg(OTf)2, Al(OTf)3, and Zn(OTf)2, proved ineffective in this transformation ( Supporting Information Table S3, entries 10–13). To our delight, we found that the reaction completed within 3 min even with as low as 0.5 mol % catalyst loading, delivering the final product 3aa with almost unchanged results (89% yield and 96% ee; entry 8). Further reduction of the catalyst loading to 0.1 mol % resulted in drastic loss of reactivity (entry 9). Then, we turned attention to optimize the reaction condition between phenyl-substituted α-diazo pyrazoleamide 4a and allylic selenide 2a ( Supporting Information Tables S6–S8). The Co(OTf)2/ L3-PiPr2 complexc proved to be an optimal catalyst in view of enantioselectivity, and the desired product 5aa was obtained in 87% yield with 87% ee after 2 h (entry 10). Table 1 | Optimization of Asymmetric [2,3]-Sigmatropic Rearrangements of Allylic Selenium Ylidesa Entry Metal Salt(s) (mol %) Ligand Time Yield (%)b ee (%)c 1d NiCl2 (5), AgNTf2 (10) L2-PiPr2 <5 min 84 84 2 Ni(OTf)2 (10) L2-PiEt2 <10 min 73 94 3 Ni(OTf)2 (10) L2-Pi(OMe)2 5 min 76 93 4 Ni(OTf)2 (10) L2-Pi(O i Pr) 2 2 min 83 94 5 Co(OTf)2 (10) L2-Pi(O i Pr)2 1 min 89 97 6 Co(BF4)2·6H2O (10) L2-Pi(O i Pr)2 1 min 89 97 7 Co(ClO4)2·6H2O (10) L2-Pi(O i Pr)2 1 min 89 97 8 Co(BF4)2·6H2O (0.5) L2-Pi(O i Pr)2 3 min 89 96 9 Co(BF4)2·6H2O (0.1) L2-Pi(O i Pr)2 10 h — — 10e Co(OTf)2 (2) L3-PiPr2 2 h 87 87 aUnless otherwise noted, the reactions were carried out with 2a (0.1 mmol), 1a (1.0 equiv), metal salt/ ligand (1∶1, 0.1–10 mol %), and CH2Cl2 (0.1 M) at 40 °C. bIsolated yields. cDetermined by high-performance liquid chromatography (HPLC) on a chiral stationary phase. d 1a (1.15 equiv). eThe reaction was carried out with 4a (0.1 mmol), 2a (1.0 equiv), Co(OTf)2/ L3-PiPr2 (1∶1, 2 mol %), and CH2Cl2 (0.1 M) at 35 °C. With the best reaction conditions in hand, we first investigated the variations of the vinyl-substituted α-diazo pyrazoleamides, keeping allyl(phenyl)selane 2a as the reaction partner. As depicted in Table 2, a wide range of vinyl-substituted α-diazo pyrazoleamides were found to be suitable substrates in this transformation. Among them, the α-diazo pyrazoleamides ( 1a– 1j) bearing an electron-donating or electron-withdrawing substituent at various positions on the phenyl group were tolerated well, delivering the expected rearranged products ( 3aa– 3ja) in good yields (83–96%) and high enantioselectivities (94–97% ee). To our delight, all the reactions could be finished within 10 min in the presence of 0.5–2 mol % catalyst. Moreover, other vinyl-substituted α-diazo substrates, fused with 2-naphthyl ( 1k), 2-furyl ( 1l), and 3-thienyl ( 1m) groups, were also readily converted smoothly within 10 min in the catalytic system, affording the targeted products ( 3ka– 3ma) in 86–96% yields with 96% ee in the presence of 2 mol % catalyst. Substrates bearing longer or shorter conjugated group ( 1n and 1o) were also conducted well in less than 5 min with high stereocontrol (93% ee for 3na; 95% ee for 3oa). Diphenylvinyl-substituted α-diazo compound 1p, which was more sterically demanding, provided the product 3pa in quantitative yield but significantly reduced enantioselectivity (74% ee). Table 2 | Substrate Scope of Vinyl-Substituted α-Diazo Pyrazoleamidesa aUnless otherwise noted, the reactions were carried out with diazo compound 1 (0.1 mmol), allyl(phenyl)selane 2a (1.0 equiv), Co(BF4)2·6H2O/ L2-Pi(O i Pr)2 (1∶1, 0.5–5 mol %), and CH2Cl2 (0.1 M) at 40 °C. Isolated yields. The ee values were determined by HPLC on a chiral stationary phase. b 2a (1.5 equiv). Subsequently, the [2,3]-sigmatropic rearrangement with aryl-substituted α-diazo pyrazoleamides 4 was evaluated by using 2 mol % of chiral Co(OTf)2/ L3-PiPr2 complex as the optimal catalyst (Table 3). The substrates bearing an electron-withdrawing or electron-donating group at meta- and para-positions on the phenyl group were tolerated well in the catalytic system, affording the rearranged products ( 5ba– 5ga) in 76–94% yields with 80–91% ee (entries 2–8). Comparatively, α-diazo pyrazoleamides 4 containing an electron-deficient aryl group gave the desired products in slightly higher enantioselectivities than those with an electron-rich aryl group and nonsubstituted phenyl group, but longer reaction time was required ( 4b– 4f vs 4a and 4g). Furthermore, α-(2-naphthyl)diazo pyrazoleamide 4g could also deliver the corresponding product 5ga in 76% yield and 85% ee (entry 9). To our disappointment, α-diazo pyrazoleamide 4i bearing an ortho-substituent on the phenyl group was not compatible with this transformation (<10% yield, entry 10), which might be due to the high steric hindrance between 4i and the catalyst. To date, we have no access to α-alkyl-substituted α-diazo pyrazoleamide, although various synthetic methods have been tried. Table 3 | Substrate Scope of Aryl-Substituted α-Diazo Pyrazoleamidesa Entry Ar Time (h) Yield (%)b ee (%)c 1 C6H5 ( 4a) 2 87 ( 5aa) 87 2 3-FC6H4 ( 4b) 16 94 ( 5ba) 88 3 3-ClC6H4 ( 4c) 12 88 ( 5ca) 91 4 3-BrC6H4 ( 4d) 16 89 ( 5da) 90 5 4-ClC6H4 ( 4e) 10 86 ( 5ea) 82 6 4-IC6H4 ( 4f) 16 85 ( 5fa) 89 7 4-MeC6H4 ( 4g) 1.5 79 ( 5ga) 80 8 2-Naphthyl ( 4h) 10 76 ( 5ha) 85 9 2-ClC6H4 ( 4i) 72 Trace ( 5ia) – aUnless otherwise noted, the reactions were carried out with diazo compound 4 (0.1 mmol), allyl(phenyl)selane 2a (1.0 equiv), Co(OTf)2/ L3-PiPr2 (1∶1, 2 mol %), and CH2Cl2 (0.1 M) at 35 °C. bIsolated yields. cDetermined by HPLC on a chiral stationary phase. Whereafter, variations of the allylic selenides 2 were evaluated, using α-diazo pyrazoleamide 1a as the reaction partner in the presence of 0.5–5 mol % of Co(BF4)2·6H2O/ L2-Pi(O i Pr)2 complex (Table 4). When the R substituent on the arylvinyl selenides varied, substrates with a chloride or methyl substituent at different positions of aryl group were transformed into the corresponding products ( 6ab– 6af) with high reactivity and enantioselectivity. In comparison, a slightly larger amount of catalyst, lower yield, and lower enantioselectivity were observed for the substrates bearing a chloride group than for those of the substrates with a methyl group. Moreover, products containing disubstituted phenyl group ( 6ag), 1-naphthyl ( 6ah), and 2-naphthyl ( 6ai) could be obtained within 15 min with 91–95% yield and 96–97% ee. However, less than 10% yield of the desired product ( 6aj) was given when allyl(benzyl)selane 2j was tested under the best reaction conditions, and unidentified byproducts were detected as an inseparable mixture. It is particularly noteworthy that when cinnamyl(phenyl)selane 2k was tested with Co(BF4)2·6H2O/ L2-Pi(OPh)2 complex, the rearranged product 6ak was afforded in 73% yield, greater than 95∶5 dr, and with 92% ee of major diastereoisomer. The stereoselectivity is much better than the reactions of α-diazoesters catalyzed by chiral copper(I) or dirhodium(II) complex.44,46 Table 4 | Substrate Scope of Allylic Selenidesa aUnless otherwise noted, the reactions were carried out with diazo compound 1a (0.1 mmol), allylic selenide 2 (1.0 equiv), Co(BF4)2·6H2O/ L2-Pi(O i Pr)2 (1∶1, 0.5–5 mol %), and CH2Cl2 (0.1 M) at 40 °C. Isolated yields. The ee values were determined by HPLC on a chiral stationary phase. b 2 (1.2 equiv). To evaluate the synthetic potential of the catalytic system, gram-scale reactions of α-diazo pyrazoleamides 1a and 4d were conducted (Scheme 2a). Upon treatment of 3.0 mmol of 1a and equivalence of 2a with 1 mol % of Co(BF4)2·6H2O/ L2-Pi(O i Pr)2 complex, the desired product 3aa was generated in 96% yield (1.30 g) with 96% ee within 1 min (Scheme 2, eq. 1). Furthermore, the gram-scale reaction between 4d and 2a was also performed smoothly in the presence of 2 mol % of Co(OTf)2/ L3-PiPr2 complex, giving the final product 4da in 92% yield (1.39 g) with 90% ee (Scheme 2, eq. 2). It is noteworthy that all cases of the asymmetric [2,3]-σ rearrangements were performed in an air atmosphere without inert gas protection, although selenium-containing molecules are sometimes sensitive toward oxidant or light. Furthermore, the asymmetric [2,3]-sigmatropic rearrangement of allyl(phenyl)sulfane 9a was also compatible using chiral cobalt(II) complex as the catalyst (Scheme 2, eq. 3). It was found that the N,N′-dioxide ligands were critical to the reactivity and enantioselectivity, and a 92% yield with 90% ee of 10a was afforded in connection with L-proline-derived ligand L3-PrPr2. Scheme 2. | (a and b) Gram-scale reactions and control experiments. Download figure Download PowerPoint Next, we compared the reactivity between allylic selenide 2a and allylic sulfide 9a. When equivalent 2a and 9a were subjected to various identified chiral catalytic systems (Scheme 2, eqs. 4 and 5), it turned out that the amount of selenide products ( 3aa and 5aa) exceeded that of the sulfur-containing products ( 10a and 11a), albeit the ratio varied with the diazo compounds, the chiral ligands, and metal salts. The results suggested that the selenides possessed higher reactivity than sulfides in this rearrangement, due to the lower energy to form a carbon–selenium bond than a carbon–sulfur bond. A previous study has declared that the racemization by pyramidal inversion of selenium ylide is sluggish in comparison with similar sulfur ylides,66 and it manifests diverse enantio-determining factors between these two rearrangements. Noteworthily, when the symmetric diallylselane 2l was used to react with α-diazo pyrazoleamide 4a catalyzed by Co(OTf)2/ L3-PiPr2 complex (Scheme 2, eq. 6), the desired product 6al was generated as racemates. This result is quite different from that of the reaction of the diallylsulfane in our previous study.49 We proposed that in current [2,3]-sigmatropic rearrangements via allylic selenium ylides, the discrimination of the heterotopic lone pairs on selenium atom is critical. To gain an insight into the enantioselective [2,3]-sigmatropic rearrangements of allylic chalcogen ylides catalyzed by the chiral N,N′-dioxide-cobalt(II) complexes, we conducted a series of control experiments (see the Supporting Information for details). First, other chiral ligands coordinated with Ni(OTf)2 or Co(BF4)2·6H2O, such as (S)- BOX and (S)- BINAP, were applied for the reaction between α-diazo pyrazoleamides 1a and allyl(phenyl)selane 2a, but the final product 3aa was detected in low yields (<10%) with less than 6% ee in all cases. It shows the superiority of chiral N,N′-dioxide ligands over other chiral ligands in current reactions. Subsequently, different diazo carbonyl compounds were subjected to the reaction with chiral cobalt(II) catalyst of L2-Pi(O i Pr)2 for comparison (Table 5). The rearrangement reaction of α-diazo pyrroleamide 7a could not proceed, accompanied by the decomposition of diazo compound 7a (Table 5, entry 1). The result demonstrated the necessity of the pyrazole group in the formation of selenium ylide intermediate. The reactivity of diazoester 1z was also negative, and only trace amount of the related product was detected (Table 5, entries 2 and 3). The methyl substituent on the pyrazole unit also affected both the reactivity and enantioselectivity (entries 4–7). The 3,5-methyl substituents on the pyrazole unit ( 1a and 1w) are critical to the reactivity and enantioselectivity (entries 4 and 5). Both the reactivity and enantioselectivity dropped when 3-methyl or unsubstituted pyrozole units were introduced ( 1x and 1y; entries 6 and 7). We compared the 13C NMR spectra of these diazo compounds, and it was found that the chemical shifts of diazo-unit-attached carbon changed dramatically, and high-field shift was observed when pyrazoleamide was introduced instead of ester (around 66 vs 77 ppm). Note that the chemical shift of diphenyldiazomethane is 62.6 ppm.68 It indicates that the diazo pyrazoleamides might be more nucleophilic than the corresponding diazoester, and its reactivity is intermediate between diazoester and aryldiazoalkanes, but the stability of the related carbenoid is enhanced due to the flanked acceptor group (pyrazoleamide) and donor group (vinyl or aryl group).69 Thus the formation of metal carbene is easier with metal salts such as nickel(II) or cobalt(II), leading to better performance than diazoesters in current catalytic system. Table 5 | The Effect of the Acceptor Group of α-Diazo Carbonyl Compoundsa Entry PG Time Yield (%)b ee (%)c C=N2δ (ppm) 1 2 h – ( 8a) – 67.2 2 OEt ( 1z) 8 h Trace ( 3az) – 77.367 3d OEt ( 1z) 8 h Trace ( 3az) – 77.367 4 1 min 89 ( 3aa) 97 65.9 5 1 min 88 ( 3aw) 97 66.1 6 30 min 75 ( 3ax) 81 65.4 7 50 min 68 ( 3ay) 33 65.6 aUnless otherwise noted, the reactions were carried out with diazo compound (0.1 mmol), allyl(phenyl)selane 2a (1.0 equiv), Co(OTf)2/ L2-Pi(O i Pr)2 (1∶1, 10 mol %), and CH2Cl2 (0.1 M) at 40 °C. bIsolated yields. cDetermined by HPLC on a chiral stationary phase. dCoCl2/AgNTf2/ L2-Pi(O i Pr)2 (1∶2∶1 10 mol %). In addition, the steric hindrance of pyrazoleamides might also affect the enantioselectivity, which represents a characteristic of donor–acceptor diazo compounds upon the classical carbenoids.70 On one hand, the sterically hindered 3,5-methyl pyrazoamide unit will increase the interaction between the carbenoid and the nucleophile, discriminating the lone-pair electrons of Se or S. On the other hand, the ready coordination ability of this functional group enables the quick shift of the metal ion species from the dipolar transition state, accelerating the rearrangement of the vinyl group. The use of typical Lewis acids, such as Zn(OTf)2, Sc(OTf)3, Al(OTf)3, and as well the combination with B(C6F5)3,71 led to ineffective [2,3]-sigmatropic rearrangement (see the Supporting Information for details), thus we ruled out Lewis acid accelerated dinitrogen-extrusion process. The X-ray diffraction structures of the Co(BF4)2·6H2O/ L2-Pi(O i Pr)2b and Co(ClO4)2·6H2O/ L3-PiPr2c complexes indicate that chiral N,N′-dioxide readily coordinates to Co2+ to form an octahedral 19-electon structure, which is similar to the structure of Ni(BF4)2·6H2O/ L2-PiPr2 complex and others.49 Apart from different outer-shell electronic configurations and ionic radii between Co(II) and Ni(II) ions, the complexes of their chiral N,N′-dioxide exhibit slight changes. Based on the analysis of X-ray structure of chiral Co(II)/ L3-PiPr2 and Ni(II)/ L3-PiPr2 complexes72 (see Supporting Information for details), we found that the length of O-metal bonds varied a little as follows: ON–Co (2.01 Å) versus ON–Ni (1.98 Å); OC–Co (2.10 Å) versus OC–Ni (2.03 Å). At the same time, the length of the ligand also changed as follows: N–O in Co(II) (1.38–1.39 Å) versus N–O in Ni(II) (1.40–1.41 Å). It means that chiral pocket around the metal ion changed with the metal ions, and the pocket of Co(II) complex is larger than Ni(II) complex. The slightly higher electrophilicity of the cobalt(II) catalyst might account for the improved reactivity. Based on these findings, the catalytic cycle of enantioselective [2,3]-sigmatropic rearrangement of allylic selenium ylides was proposed as shown in Scheme 3. Initially, the chiral Co(II) complex catalyzes the N2 extrusion of the diazo compound to generate the carbene species, as demonstrated by infrared absorption measurements ( Supporting Information Section 6.6). The steric environment of the chiral ligand enables the enantioselective attack of the lone pair of electrons of selenide to the carbon of carbene species. As a result, the stable and chiral allylic selenium ylide is formed, followed by a concerted [2,3]-sigmatropic rearrangement to transfer the chirality from Se* to the carbon center, yielding the selenide bearing a quaternary C–Se stereocenter. Scheme 3. | Proposed catalytic cycle. Download figure Download PowerPoint Conclusion We have delineated a novel example of enantioselective [2,3]-sigmatropic rearrangement via allylic selenium ylides. The reactions proceeded efficiently through chiral N,N′-dioxide/cobalt(II) complex initiated metal carbenes generated from α-diazo pyrazoleamides and allylic selenides. A series of selenides with quaternary C–Se stereocenter can be prepared with decent yields and enantioselectivities (up to 99% yield and 97% ee) under operationally expedient and mild conditions. Most of the examples could be finished with low catalyst loading (0.5–2 mol %). Exploring the mechanism manifested the difference between allylic chalcogen ylides in terms of reactivity and enantioselection. It is worth noting that the asymmetric rearrangement of allylic selenium ylides involves the transfer of chirality from the stable chiral selenium to the carbon of the product. More applications of this combination in asymmetric rearrangement reactions are under exploration in our lab. Footnotes a CCDC 1972936 ( 1a) contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. b CCDC 1959861 [Co(BF4)2·6H2O/ L2-Pi(OiPr)2] contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. c CCDC 1977471 [Co(ClO4)2·6H2O/ L3-PiPr2] contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. Supporting Information Supporting Information is available. Conflict of Interest There is no conflict of interest to report. Funding Information The authors acknowledge financial support from the National Natural Science Foundation of China (grant nos. 21625205 and 21772127). Acknowledgments The authors wish to acknowledge Dr. Yuqiao Zhou (Sichuan University) for his assistance in X-ray crystallographic analysis.