( ± )- trans -1,2-Cyclohexanediamine-Based Bis(NHC) Ligand for Cu-Catalyzed Asymmetric Conjugate Addition Reaction

: Bis(NHC) ligand precursors, L1 , based on trans -1,2-diaminocyclohexane were designed and synthesized. , Cu-catalyzed asymmetric conjugate addition reactions of cyclic and acyclic with 2 to evaluate the performance of L1 as a chiral ligand. bis(NHC) ligand precursor, ( rac ; S , S )- L1 , which was prepared from ( S )-leucinol, was the most e ﬀ ective. Thus, treating 2-cyclohexen-1-one ( 3 ) with Et 2 Zn in the presence of catalytic amounts of Cu(hfacac)(btmsa) and ( rac ; S , S )- L1 a ﬀ orded ( R )-3-ethylcyclohexanone (( R )- 4 ) with 97% ee . Similarly, use of ( rac ; R , R )- L1 , which was prepared from ( R )-leucinol, produced ( S )- 4 with 97% ee . Conversely, for the asymmetric 1,4-addition reaction of the acyclic enone, optically pure ( − )- trans -1,2-cyclohexanediamine-based bis(NHC) ligand precursor, ( R , R ; S , S )- L1 , worked e ﬃ ciently. For example, 3-nonen-2-one ( 5 ) was reacted with Et 2 Zn using the CuOAc / ( R , R ; S , S )- L1 catalytic system to a ﬀ ord ( R )-4-ethylnonan-2-one (( R )- 6 ) with 90% ee . Furthermore, initially changing the counterion of the Cu precatalyst between an OAc and a ClO 4 ligand on the metal reversed the facial selectivity of the approach of the substrates. Thus, the conjugate addition reaction of 5 with Et 2 Zn using the Cu(ClO 4 ) 2 / ( R , R ; S , S )- L1 catalytic system, a ﬀ orded ( S )- 6 with 75% ee .


Catalytic Asymmetric Conjugate Addition Reaction of Cyclic Enone
Enantiomerically pure L1 was synthesized from enantiopure trans-1,2-diaminocyclohexane ((R,R)-1 or (S,S)-1) [81]. To introduce chirality at the hydroxyamide side arm at L1, we chose (S)leucinol ((S)-2) or (R)-leucinol ((R)-2) as a chiral source. In this study, we wanted to determine the influence of the chirality of the cyclohexanediamine skeleton of L1 on the catalytic asymmetric transformation. Thus, four sets of bis(NHC) azolium ligands were synthesized (Scheme 2). Scheme 2. Preparation of a series of chiral ligands L1 used in this study.
The previously reported bis(NHC) azolium ligand (rac; S,S)-L1 consists of a 1:1 diastereoisomeric mixture of (R,R; S,S)-L1 and (S,S; S,S)-L1 (Scheme 1) [81]. With a set of diastereomerically and enantiomerically pure azolium salts in hand (Scheme 2), we performed an NMR study of these compounds (Figures 2-4). Initially, temperature dependence was observed in the 1 H NMR spectrum of the bis(NHC) azolium precursor ( Figure 2). For example, 1 H NMR signals of (R,R; R,R)-L1 at 25 °C were extremely broad, indicating that several conformers, slowly interconverting on the NMR timescale, were present at room temperature. Conversely, increasing the temperature (to 80 °C) provided sharp, well-defined signals in the NMR spectra ( Figure 2). Thus, a fully assignable NMR spectrum was obtained.
Furthermore, the methyl group signals of the isobutyl substituent in the 1 H NMR spectra of a series of azolium salts L1 was considered. Figure 3 shows the expanded spectra of the methyl group regions of (rac; S,S)-L1, (R,R; S,S)-L1 and (S,S; S,S)-L1. Their C-H resonances appeared at δ = 0.87/0.82 ppm (J = 6.9 Hz) for (R,R; S,S)-L1 and δ = 0.88/0.83 ppm (J = 6.9 Hz) for (S,S; S,S)-L1. Thus, the methyl group signals in (rac; S,S)-L1 could be assigned, as shown in Figure 3.
Similarly, in the 13 C NMR spectrum of (rac; S,S)-L1, two signals were observed in the methine carbon adjacent to the isobutyl substituent ( Figure 4). This is because of the presence of a diastereoisomeric mixture. The 13 C NMR spectra of the diastereoisomerically pure isomers revealed Scheme 2. Preparation of a series of chiral ligands L1 used in this study.
The previously reported bis(NHC) azolium ligand (rac; S,S)-L1 consists of a 1:1 diastereoisomeric mixture of (R,R; S,S)-L1 and (S,S; S,S)-L1 (Scheme 1) [81]. With a set of diastereomerically and enantiomerically pure azolium salts in hand (Scheme 2), we performed an NMR study of these compounds (Figures 2-4). Initially, temperature dependence was observed in the 1 H NMR spectrum of the bis(NHC) azolium precursor ( Figure 2). For example, 1 H NMR signals of (R,R; R,R)-L1 at 25 • C were extremely broad, indicating that several conformers, slowly interconverting on the NMR timescale, were present at room temperature. Conversely, increasing the temperature (to 80 • C) provided sharp, well-defined signals in the NMR spectra ( Figure 2). Thus, a fully assignable NMR spectrum was obtained.
Furthermore, the methyl group signals of the isobutyl substituent in the 1 H NMR spectra of a series of azolium salts L1 was considered. Figure 3 shows the expanded spectra of the methyl group regions of (rac; S,S)-L1, (R,R; S,S)-L1 and (S,S; S,S)-L1. Their C-H resonances appeared at δ = 0.87/0.82 ppm (J = 6.9 Hz) for (R,R; S,S)-L1 and δ = 0.88/0.83 ppm (J = 6.9 Hz) for (S,S; S,S)-L1. Thus, the methyl group signals in (rac; S,S)-L1 could be assigned, as shown in Figure 3.
Similarly, in the 13 C NMR spectrum of (rac; S,S)-L1, two signals were observed in the methine carbon adjacent to the isobutyl substituent ( Figure 4). This is because of the presence of a diastereoisomeric mixture. The 13 C NMR spectra of the diastereoisomerically pure isomers revealed that the signal at δ 48.6 ppm could be attributed to the methine carbon on (R,R; S,S)-L1 and that at δ 48.5 ppm, it could be attributed to the methine carbon on (S,S; S,S)-L1.
It is noteworthy that the asymmetric catalytic 1,4-addition reaction occurred efficiently when using a 1:1 mixture of (R,R; S,S)-L1 and (S,S; S,S)-L1 ((rac; R,R)-L1). To highlight the importance of both components in the catalytic reaction, we then varied the ratio of (R,R; S,S)-L1 and (S,S; S,S)-L1 (Table 2). Table 2. Influence of the ratio of chiral ligands L1 in the conjugate addition reaction of 3.
It is noteworthy that the asymmetric catalytic 1,4-addition reaction occurred efficiently when using a 1:1 mixture of (R,R; S,S)-L1 and (S,S; S,S)-L1 ((rac; R,R)-L1). To highlight the importance of both components in the catalytic reaction, we then varied the ratio of (R,R; S,S)-L1 and (S,S; S,S)-L1 ( Table 2). Table 2. Influence of the ratio of chiral ligands L1 in the conjugate addition reaction of 3.
The same experiment was conducted using (S,S; R,R)-L1 and (R,R; R,R)-L1 to provide (S)-4 in the Cu-catalyzed reaction of 3 with Et 2 Zn (Table 2). In a similar manner, a 9:1 or 1:9 mixture facilitated the highly enantioselective conjugate addition reaction to afford (S)-4 with 90% or 94% ee, respectively. These results confirm the importance of the presence of both stereoisomers of the cyclohexanediamine moiety in the chiral ligands for reaction efficiency and stereoselectivity.
The same experiment was conducted using (S,S; R,R)-L1 and (R,R; R,R)-L1 to provide (S)-4 in the Cu-catalyzed reaction of 3 with Et2Zn (Table 2). In a similar manner, a 9:1 or 1:9 mixture facilitated the highly enantioselective conjugate addition reaction to afford (S)-4 with 90% or 94% ee, respectively. These results confirm the importance of the presence of both stereoisomers of the cyclohexanediamine moiety in the chiral ligands for reaction efficiency and stereoselectivity.
The relationship between the catalyst ee and the product ee was also investigated ( Figure 5). Various mixtures of (rac; S,S)-L1 and (rac; R,R)-L1 were carefully prepared. The enantioselective conjugate addition reaction of cyclic enone 3 with Et2Zn provided sufficient chiral amplification ( Figure 5). As shown in Table 2, we have clearly demonstrated that a 9:1 or 1:9 mixture of (R,R; S,S)-L1 and (S,S; S,S)-L1 promotes the enantioselective catalytic reaction. These reactions were performed in the presence of 6.0 mol % of Cu salt and 4.5 mol % of the chiral ligand mixtures. We assumed that an excess of (R,R; S,S)-L1 would not be needed in the catalytic reaction with a 9:1 mixture of (R,R; S,S)-L1 and (S,S; S,S)-L1 and that a small amount of a 1:1 mixture of these ligands, with respect to the Cu precatalyst, would be sufficient to facilitate the catalytic reaction. This assumption prompted us to study the asymmetric 1,4-addition reaction of 3 with Et2Zn catalyzed by Cu(hfacac)(btmsa), with a reduced amount of the chiral ligand mixtures ( Table 3).
As expected, the 1,4-addition reactions with Cu/ligand ratios of 1:0.75, 1:0.3, 1:0.225, and 1:0.15 occurred in a similar manner to afford the corresponding adduct with 97%, 93%, 96%, and 97% ee, respectively (entries 1-4). These results indicate that the same catalytic active species were probably formed. It is also noteworthy that the loading of chiral ligand could be dramatically reduced.
No reaction was observed in the absence of L1. Additionally, two bis(NHC) ligands, L2 and L3, were synthesized to investigate the effect of the ligand structure on the catalytic performance. Changing the alkyl substituent at the chiral carbon center of the ligand from the isobutyl group ((rac; As shown in Table 2, we have clearly demonstrated that a 9:1 or 1:9 mixture of (R,R; S,S)-L1 and (S,S; S,S)-L1 promotes the enantioselective catalytic reaction. These reactions were performed in the presence of 6.0 mol % of Cu salt and 4.5 mol % of the chiral ligand mixtures. We assumed that an excess of (R,R; S,S)-L1 would not be needed in the catalytic reaction with a 9:1 mixture of (R,R; S,S)-L1 and (S,S; S,S)-L1 and that a small amount of a 1:1 mixture of these ligands, with respect to the Cu precatalyst, would be sufficient to facilitate the catalytic reaction. This assumption prompted us to study the asymmetric 1,4-addition reaction of 3 with Et 2 Zn catalyzed by Cu(hfacac)(btmsa), with a reduced amount of the chiral ligand mixtures ( Table 3).
As expected, the 1,4-addition reactions with Cu/ligand ratios of 1:0.75, 1:0.3, 1:0.225, and 1:0.15 occurred in a similar manner to afford the corresponding adduct with 97%, 93%, 96%, and 97% ee, respectively (entries 1-4). These results indicate that the same catalytic active species were probably formed. It is also noteworthy that the loading of chiral ligand could be dramatically reduced.
No reaction was observed in the absence of L1. Additionally, two bis(NHC) ligands, L2 and L3, were synthesized to investigate the effect of the ligand structure on the catalytic performance. Changing the alkyl substituent at the chiral carbon center of the ligand from the isobutyl group ((rac; S,S)-L1) to the more sterically hindered tert-butyl group (L2) led to the conjugate adduct, (S)-4, in lower yield and enantioselectivity with the opposite configuration (Table 3, entry 6). This might be due to the highly hindered tert-butyl group blocking the approach of the reagents. Furthermore, L3 was synthesized from 1-aminoethanol. However, it was difficult to react 3 with Et 2 Zn in the presence of Cu(hfacac)(btmsa) in combination with L3 (entry 7).
The catalytic performance of the Cu(hfacac)(btmsa)/(rac; R,R)-L1 system in the asymmetric 1,4-addition reaction of various cyclic enones was investigated, as described in our previous publication [81]. S,S)-L1) to the more sterically hindered tert-butyl group (L2) led to the conjugate adduct, (S)-4, in lower yield and enantioselectivity with the opposite configuration (Table 3, entry 6). This might be due to the highly hindered tert-butyl group blocking the approach of the reagents. Furthermore, L3 was synthesized from 1-aminoethanol. However, it was difficult to react 3 with Et2Zn in the presence of Cu(hfacac)(btmsa) in combination with L3 (entry 7). The catalytic performance of the Cu(hfacac)(btmsa)/(rac; R,R)-L1 system in the asymmetric 1,4addition reaction of various cyclic enones was investigated, as described in our previous publication [81].
Interestingly, during this study, we found that simply changing the copper catalyst precursor with the same ligand reversed the stereochemistry. The reaction of 5 with Et2Zn using the CuCl2/(rac; S,S)-L1 catalytic system afforded (R)-6 in 50% yield and with 39% ee (entry 7). In contrast, Cu(OTf)2
Interestingly, during this study, we found that simply changing the copper catalyst precursor with the same ligand reversed the stereochemistry. The reaction of 5 with Et 2 Zn using the CuCl 2 /(rac; S,S)-L1 catalytic system afforded (R)-6 in 50% yield and with 39% ee (entry 7). In contrast, Cu(OTf) 2 in place of CuCl 2 led to the 1,4-adduct, (S)-6, with the opposite configuration with 33% ee (entry 8). Developing a synthetic methodology for switching enantioselectivity is an essential and challenging  in place of CuCl2 led to the 1,4-adduct, (S)-6, with the opposite configuration with 33% ee (entry 8). Developing a synthetic methodology for switching enantioselectivity is an essential and challenging research topic. Recently, many papers have been published to provide a comprehensive overview of the importance of stereodivergent catalytic transformations [94][95][96][97][98][99][100]. Cu(ClO4)2 ( S)-6 31 50 a To a solution of Cu salt (6.0 mol%) and (rac; S,S)-L1 (4.5 mol%) in THF (9 mL), Et2Zn (3 mmol) was added first, then 5 (1 mmol). The reaction mixture was stirred at room temperature for 3 h. b Determined by GC analysis using the internal standard technique. c Determined by GC analysis of a chiral stationary phase. d Previously reported data [81].
Based on this finding, we decided to evaluate several Cu salts. CuOTf•0.5C6H6 and Cu(NO3)2 resulted in racemic 6 ( Table 4, entries 9 and 10). Cu(ClO4)2 led to a marked increase in the enantioselectivity of the 1,4-addition reaction to provide (S)-6 as a major product (entry 11). Previously, we showed that the reversal of enantioselectivity was achieved in the Cu-catalyzed conjugate addition reaction of cyclic enone using a mono-NHC azolium ligand L4 with a chiral hydroxyamide side-arm ( Figure 6) [76]. For example, 3 was reacted with Et2Zn catalyzed by Cu(OTf)2/L4 to afford (S)-4, whereas (R)-4 was obtained in the same reaction with the Cu(acac)2/L4 catalytic system [74]. However, no reversal of enantioselectivity was observed in the Cu-catalyzed 1,4-addition reaction of acyclic enone when changing the Cu precatalyst in the presence of L4 [71]. Therefore, to the best of our knowledge, this is the first example of switchable enantioselectivity in a catalytic conjugate addition reaction of acyclic enone, with the same chiral ligand. CuOTf·0.5C 6 H 6 (S)-6 57 2 10 Cu(NO 3 ) 2 (S)-6 62 9 11 Cu(ClO 4 ) 2 (S)-6 31 50 a To a solution of Cu salt (6.0 mol%) and (rac; S,S)-L1 (4.5 mol%) in THF (9 mL), Et 2 Zn (3 mmol) was added first, then 5 (1 mmol). The reaction mixture was stirred at room temperature for 3 h. b Determined by GC analysis using the internal standard technique. c Determined by GC analysis of a chiral stationary phase. d Previously reported data [81].
Based on this finding, we decided to evaluate several Cu salts. CuOTf•0.5C6H6 and Cu(NO3)2 resulted in racemic 6 ( Table 4, entries 9 and 10). Cu(ClO4)2 led to a marked increase in the enantioselectivity of the 1,4-addition reaction to provide (S)-6 as a major product (entry 11). Previously, we showed that the reversal of enantioselectivity was achieved in the Cu-catalyzed conjugate addition reaction of cyclic enone using a mono-NHC azolium ligand L4 with a chiral hydroxyamide side-arm ( Figure 6) [76]. For example, 3 was reacted with Et2Zn catalyzed by Cu(OTf)2/L4 to afford (S)-4, whereas (R)-4 was obtained in the same reaction with the Cu(acac)2/L4 catalytic system [74]. However, no reversal of enantioselectivity was observed in the Cu-catalyzed 1,4-addition reaction of acyclic enone when changing the Cu precatalyst in the presence of L4 [71]. Therefore, to the best of our knowledge, this is the first example of switchable enantioselectivity in a catalytic conjugate addition reaction of acyclic enone, with the same chiral ligand.
Based on this finding, we decided to evaluate several Cu salts. CuOTf·0.5C 6 H 6 and Cu(NO 3 ) 2 resulted in racemic 6 ( Table 4, entries 9 and 10). Cu(ClO 4 ) 2 led to a marked increase in the enantioselectivity of the 1,4-addition reaction to provide (S)-6 as a major product (entry 11). Previously, we showed that the reversal of enantioselectivity was achieved in the Cu-catalyzed conjugate addition reaction of cyclic enone using a mono-NHC azolium ligand L4 with a chiral hydroxyamide side-arm ( Figure 6) [76]. For example, 3 was reacted with Et 2 Zn catalyzed by Cu(OTf) 2 /L4 to afford (S)-4, whereas (R)-4 was obtained in the same reaction with the Cu(acac) 2 /L4 catalytic system [74]. However, no reversal of enantioselectivity was observed in the Cu-catalyzed 1,4-addition reaction of acyclic enone when changing the Cu precatalyst in the presence of L4 [71]. Therefore, to the best of our knowledge, this is the first example of switchable enantioselectivity in a catalytic conjugate addition reaction of acyclic enone, with the same chiral ligand. Catalysts 2019, 9, x FOR PEER REVIEW 10 of 21 The combination of transition metal complex and chiral ligand can provide a chiral environment for the metal, where high enantioselectivity requires only small differences in transition state energies in the catalytic reaction. The transition metal complex consists of a cationic metal and an achiral counter anion. While the chiral ligand design is critical to achieving a highly stereoselective catalytic reaction, the choice of the achiral counter anion of the transition metal complex is also an important factor. In the literature, it appears that changing the achiral counter anion of the metal catalyst precursor has sometimes switched the stereochemistry of the catalytic reaction [76,[101][102][103][104][105][106][107]. Table 5 summarizes the switching of enantioselectivity in the asymmetric 1,4-addition reaction of 5 with Et2Zn catalyzed by CuOAc or Cu(ClO4)2 using the chiral bis(NHC) ligand precursor, (rac; S,S)-L1, under selected reaction conditions. To optimize the reaction conditions, various reaction parameters, including the ratio of Cu salt/chiral ligand, solvents and reaction temperatures were screened with CuOAc (entries 1-8) or Cu(ClO4)2 (entries 9-16). Further evaluations of the reaction parameters with both catalytic systems are presented in Tables S1 and S2, respectively (see Supplementary Materials). a To a solution of Cu salt and (rac; S,S)-L1 in solvent (9 mL), Et2Zn (3 mmol) was added first, then 5 (1 mmol). b CuOAc was used as a copper precatalyst. c The same data entry is shown in Table 4, as entry 5. d Cu(ClO4)2 was used as a copper precatalyst. e The same data entry is shown in Table 4, as entry 11. f DME (6 mL) was used.
In the CuOAc/(rac; S,S)-L1 catalytic system, a 1:1 ratio of Cu/ligand was most effective (entries 1-3). We thus investigated the solvent effect on the reaction in the presence of 4 mol% of a 1:1 mixture of CuOAc and (rac; S,S)-L1 (entries 3-5; also see Table S1). Although slightly lower enantioselectivity was observed in the catalytic reaction with Et2O, a promising result was achieved with 1,2dimethoxyethane (DME) (entries 4 and 5). Furthermore, a decrease in the reaction temperature to 0 °C improved the enantioselectivity (73-78% ee), even though a longer reaction time was needed (entries 6 and 7). The desired product (R)-6 was obtained in 69% yield with 75% ee when the reaction The combination of transition metal complex and chiral ligand can provide a chiral environment for the metal, where high enantioselectivity requires only small differences in transition state energies in the catalytic reaction. The transition metal complex consists of a cationic metal and an achiral counter anion. While the chiral ligand design is critical to achieving a highly stereoselective catalytic reaction, the choice of the achiral counter anion of the transition metal complex is also an important factor. In the literature, it appears that changing the achiral counter anion of the metal catalyst precursor has sometimes switched the stereochemistry of the catalytic reaction [76,[101][102][103][104][105][106][107]. Table 5 summarizes the switching of enantioselectivity in the asymmetric 1,4-addition reaction of 5 with Et 2 Zn catalyzed by CuOAc or Cu(ClO 4 ) 2 using the chiral bis(NHC) ligand precursor, (rac; S,S)-L1, under selected reaction conditions. To optimize the reaction conditions, various reaction parameters, including the ratio of Cu salt/chiral ligand, solvents and reaction temperatures were screened with CuOAc (entries [1][2][3][4][5][6][7][8] or Cu(ClO 4 ) 2 (entries 9-16). Further evaluations of the reaction parameters with both catalytic systems are presented in Tables S1 and S2, respectively (see Supplementary Materials). a To a solution of Cu salt and (rac; S,S)-L1 in solvent (9 mL), Et 2 Zn (3 mmol) was added first, then 5 (1 mmol). b CuOAc was used as a copper precatalyst. c The same data entry is shown in Table 4, as entry 5. d Cu(ClO 4 ) 2 was used as a copper precatalyst. e The same data entry is shown in Table 4, as entry 11. f DME (6 mL) was used.
In the Cu-catalyzed conjugate addition reaction of acyclic enone using the bis(NHC) ligand, we described successful enantioselectivity reversal by simply changing of the counter anion of the Cu salt. We then assumed that combining copper chloride with a silver salt facilitates the catalytic reaction and that a silver salt additive would control the setereoselectivity of the 1,4-addition reaction ( Table 6). a To a solution of Cu salt (4 mol%) and (rac; S,S)-L1 (4 mol%) in DME (9 mL), Et 2 Zn (3 mmol) was added first, then 5 (1 mmol). The reaction mixture was stirred at 0 • C for 24 h. b The same data entry is shown in Table 5, as entry 6. c To a solution of Cu salt (6 mol%) and (rac; S,S)-L1 (3 mol%) in THF (9 mL), Et 2 Zn (3 mmol) was added first, then 5 (1 mmol). The reaction mixture was stirred at 0 • C for 24 h. d The same data entry is shown in Table 5, as entry 14.
First, we tested the reaction of 5 with Et 2 Zn catalyzed by CuCl (4 mol%) in the presence of (rac; S,S)-L1 (4 mol%). This reaction afforded (R)-6 in lower enantioselectivity, compared with the reaction using the CuOAc/(rac; S,S)-L1 catalytic system (entries 1 and 2). As expected, adding AgOAc (4 mol%) to the CuCl-catalyzed reaction led to a remarkable increase in the stereoselectivity (86% ee) (entry 3). This result was comparable to that of the CuOAc-catalyzed reaction, meaning that the acetate anion would play an important role through interaction with a copper ion having the (rac; S,S)-L1 ligand. Based on this assumption, an excess of AgOAc additive, with respect to the CuCl catalyst, was employed. The catalytic 1,4-addition reaction with a 1:1.5 molar ratio of CuCl/AgOAc provided (R)-6 in 54% yield with 85% ee, whereas a greater excess (8 mol%) of AgOAc decreased its catalytic activity (entries 4 and 5).
To evaluate the substrate scope and limitations of the developed catalytic reactions, several α,β-unsaturated enones were investigated ( Table 7).
The performance of both catalytic systems for in the Cu-catalyzed conjugate addition of cyclic enone were evaluated (Table 7, entries 11 and 12). As mentioned above, we previously reported successful enantioselectivity reversal of the Cu-catalyzed 1,4-addition reaction of 3 with Et 2 Zn using the mono-NHC azolium ligand, L4 (Figure 6), by changing the Cu precatalyst from Cu(OTf) 2 to Cu(acac) 2 [76]. The bis(NHC) azolium ligand, (rac; S,S)-L1, was suitable for achieving the enantioselectivity switch in the 1,4-addition reaction of both acyclic and cyclic enones. The combination of CuOAc and (rac; S,S)-L1 catalyzed the reaction of 3 with Et 2 Zn to afford the corresponding adduct (R)-4 in 38% yield with 57% ee (entry 11). The facial selectivity of the 1,4-addition reaction using the Cu(ClO 4 ) 2 /(rac; S,S)-L1 catalytic system was reversed compared to that of the reaction using CuOAc (entry 12).
As shown in Table 2, a mixture of (R,R; S,S)-L1 and (S,S; S,S)-L1 promoted the Cu-catalyzed conjugate addition reaction of cyclic enone with Et 2 Zn, and using (R,R; S,S)-L1 or (S,S; S,S)-L1 alone led to a poor stereoselectivity. Based on this finding, we next investigated the catalytic reaction of acyclic enone using a set of diastereomerically and enantiomerically pure azolium salts, (R,R; S,S)-L1 and (S,S; S,S)-L1 (Table 8).  a To a solution of CuOAc (4 mol%) and L1 in DME (9 mL), Et 2 Zn (3 mmol) was added first, then 5 (1 mmol). The reaction mixture was stirred at 0 • C for 24 h. b The same data entry is shown in Table 5, as entry 6. c To a solution of Cu(ClO 4 ) 2 (6 mol%) and L1 in THF (9 mL), Et 2 Zn (3 mmol) was added first, then 5 (1 mmol). The reaction mixture was stirred at 0 • C for 24 h. d The same data entry is shown in Table 5, as entry 14.
Compound 5 was treated with Et 2 Zn in the presence of catalytic amounts of CuOAc and (R,R; S,S)-L1 in DME at 0 • C for 24 h to afford (R)-6 in good yield and stereoselectivity (79% yield, 88% ee, entry 1). This is in contrast to the catalytic reaction of cyclic enone, where the Cu/(R,R; S,S)-L1 catalytic system was inert (see Table 2). It is noteworthy that (R,R; S,S)-L1 showed a superior catalytic performance compared a diastereomeric mixture of ligands, such as (rac; S,S)-L1, in the 1,4-addition reaction of acyclic enone (entry 1 vs. entry 2). In contrast to the catalytic conjugate addition reaction of 5 using (R,R; S,S)-L1, the reaction using (S,S; S,S)-L1 proceeded, with difficulty, to produce the desired (R)-6 (17% yield and 48% ee, entry 3). These results suggest that (S,S; S,S)-L1 acts as a catalyst poison in the catalytic reaction with (rac; S,S)-L1. Additionally, an excellent ee value of 91% of (R)-6 was obtained in the conjugate addition reaction of 5 with Et 2 Zn using 4 mol% of CuOAc and 3 mol% of (R,R; S,S)-L1 (entry 4).
Next, in the switchable enantioselective reaction using Cu(ClO 4 ) 2 precatalyst to provide (S)-6, the performances of (R,R; S,S)-L1 and (S,S; S,S)-L1 were examined (Table 8, entries 5-7). With 6 mol% of Cu(ClO 4 ) 2 , 3 mol% of chiral ligand containing (R,R; S,S)-L1 and/or (S,S; S,S)-L1 was employed in the conjugate addition reaction. These experiments showed the formation of the desired (S)-6 with almost the same yields and enantioselectivities (66-75% ee) in each of the catalytic reactions (entries 5-7). These results differed from those of the CuOAc/L1 catalytic system (entries 1-3). It could be assumed that the chiral hydroxyamide side arm of the bis(NHC) ligand would play an important role in the Cu(ClO 4 ) 2 -catalyzed asymmetric conjugate addition reaction.

General Procedures
All chemicals were obtained from commercial sources and were used as received. 1 H and 13 C NMR spectra were recorded on spectrometers at 400 and 100 MHz, respectively. Chemical shifts were reported in ppm relative to TMS for 1 H and 13 C NMR spectra. CDCl3 and (CD3)2SO were used as the NMR solvent. Thin-layer chromatography (TLC) analysis was performed with glass-backed plates, pre-coated with silica gel and examined under UV (254 nm) irradiation. Flash column chromatography was executed on silica gel 60 (230-400; particle size: 0.040-0.063 nm).

Procedure for Preparation of Azolium Salt L1
Azoles were synthesized from trans-1,2-cyclohexanediamine (1) according to the literature procedure [47,59]. The reaction mixture of azole (485 mg, 1.5 mmol) and α-chloroacetamide (580 mg, 3.0 mmol), derived from chloroacetyl chloride and luecinol, was heated to 110 °C and stirred for 2 days. After the reaction, the solvent was removed under reduced pressure. The desired product was isolated from the crude residue by column chromatography (SiO2, CH2Cl2/CH3OH = 8/2) to yield the light yellow solution after removing the solvent. The residue was dissolved in methanol, and then activated carbon (ca. 1 g) was added. After 3 h, the activated carbon was removed by filtration. After removing the CH3OH in vacuo from the filtrate, the azolium salt L1 was purified by reprecipitation using CH3OH and CH3CO2C2H5 affording white solid (yield: 710 mg, 62%). Compounds (rac; S,S)-L1, (rac; R,R)-L1, L5 and L6 were reported in our previous publication [81].

General Procedures
All chemicals were obtained from commercial sources and were used as received. 1 H and 13 C NMR spectra were recorded on spectrometers at 400 and 100 MHz, respectively. Chemical shifts were reported in ppm relative to TMS for 1 H and 13 C NMR spectra. CDCl 3 and (CD 3 ) 2 SO were used as the NMR solvent. Thin-layer chromatography (TLC) analysis was performed with glass-backed plates, pre-coated with silica gel and examined under UV (254 nm) irradiation. Flash column chromatography was executed on silica gel 60 (230-400; particle size: 0.040-0.063 nm).

Procedure for Preparation of Azolium Salt L1
Azoles were synthesized from trans-1,2-cyclohexanediamine (1) according to the literature procedure [47,59]. The reaction mixture of azole (485 mg, 1.5 mmol) and α-chloroacetamide (580 mg, 3.0 mmol), derived from chloroacetyl chloride and luecinol, was heated to 110 • C and stirred for 2 days. After the reaction, the solvent was removed under reduced pressure. The desired product was isolated from the crude residue by column chromatography (SiO 2 , CH 2 Cl 2 /CH 3 OH = 8/2) to yield the light yellow solution after removing the solvent. The residue was dissolved in methanol, and then activated carbon (ca. 1 g) was added. After 3 h, the activated carbon was removed by filtration. After removing the CH 3 OH in vacuo from the filtrate, the azolium salt L1 was purified by reprecipitation using CH 3 OH and CH 3 CO 2 C 2 H 5 affording white solid (yield: 710 mg, 62%). Compounds (rac; S,S)-L1, (rac; R,R)-L1, L5 and L6 were reported in our previous publication [81].

General Procedure for Cu-Catalyzed Asymmetric Reaction of Enone with Et 2 Zn
Cu salt and azolium salt were added to anhydrous THF. After stirring at room temperature for 1 h, the mixture was cooled to 0 • C. Then, Et 2 Zn (3 mmol, 1 M in hexanes, 3 mL) was added to the reaction vessel. After the resulting mixture was stirred at room temperature for 30 min, a solution of enone (1 mmol) in anhydrous THF was added dropwise over a period of 10 min. After stirring at room temperature for 3 h, the reaction was quenched with 10% HCl aq. The product yield was determined by GC using internal standard technique. The enantiomeric excess was measured by the chiral GC (see Supplementary Materials). The conjugate adduct was isolated as follows: After quenching the reaction mixture with 10% HCl aq., the resulting mixture was extracted with diisopropyl ether (3 × 10 mL) and dried over Na 2 SO 4 . The product was purified by silica gel column chromatography (hexane/Et 2 O).

Conclusions
We demonstrated that a chiral bis(NHC) ligand precursor L1, which is a dibenzimidazolium salt having a hydroxyamide side arm, efficiently performs Cu-catalyzed asymmetric conjugate addition of Et 2 Zn to cyclic or acyclic enones. In the catalytic 1,4-addition reaction of cyclic enone 3 with Et 2 Zn, an excellent enantioselectivity (up to 97% ee) was achieved with the Cu(hfacac)(btmsa)/(rac; S,S)-L1 catalytic system. Further investigations revealed that a 9:1 or 1:9 mixture of (R,R; S,S)-L1 and (S,S; S,S)-L1 promoted the highly enantioselective catalytic reaction. A potential application of the Cu/L1 catalytic system was investigated in the enantioselective 1,4-addition reaction of acyclic enones with Et 2 Zn. Interestingly, the stereoselectivity switching was observed in the Cu-catalyzed 1,4-addition reaction of 5 with Et 2 Zn in the presence of L1 when changing the counter anion on the Cu precatalyst.