Ru-Catalyzed Asymmetric Addition of Arylboronic Acids to Aliphatic Aldehydes via P-Chiral Monophosphorous Ligands

Chiral alcohols are among the most widely applied in fine chemicals, pharmaceuticals and agrochemicals. Herein, the Ru-monophosphine catalyst formed in situ was found to promote an enantioselective addition of aliphatic aldehydes with arylboronic acids, delivering the chiral alcohols in excellent yields and enantioselectivities and exhibiting a broad scope of aliphatic aldehydes and arylboronic acids. The enantioselectivities are highly dependent on the monophosphorous ligands. The utility of this asymmetric synthetic method was showcased by a large-scale transformation.


Results and Discussions
At the outset of the research, initial experimental results show that the monophosphine ligand L1 effectively promoted the Ru (II)-catalyzed asymmetric addition of phenylpropanal 1a with 4-methylphenylboronic acid 2a to access chiral alcohol 3aa in excellent yields and with good enantioselectivity; only a trace amount of 4aa was obtained as a byproduct (Table 1, entry 1). Further ligand investigations indicated that the ligands had a remarkable influence on enantioselectivity (Table 1, entries 2~5 and entry 11). For instance, good enantioselectivities can be afforded by using L3 or L5 as chiral phosphorous ligands (Table 1, entries 3 and 5), while a racemic or nearly racemic product was formed by employing L2, L4 and L11 (Table 1, entries 2, 4 and 11), indicating that the substituents on the carbon between the phosphorus and oxygen atom appear to have a negative influence on enantioselectivity.
Similar trends were also observed by using chiral bisphosphine ligands L6~L7 or L9~L10,which were connected via α-position of a P-chiral center ( Table 1, entries 6~7 and 9~10). For example, only low enantioselectivity was shown when bisphosphine ligands L6~L7 and L9~L10 were utilized in the reaction, while a moderate enantioselective product could be afforded by using L8 as a bisphosphine ligand, which was coupled at the β-position ( Despite the fact that these remarkable achievements were created in the asymmetric addition of carbonyl compounds with organoboron reagents, efforts have been mainly devoted to the active ketones [55][56][57][58][59][60][61][62][63][64][65][66][67][68][69], inactive ketones [70][71][72], and aromatic aldehydes [73][74][75], but studies on aliphatic aldehydes have been less reported [30]. In this regard, the development of new, high-efficient catalytic systems for the enantioselective addition of aliphatic aldehydes with arylboronic acids still is desirable. L1~L8 (Scheme 1) represent one kind of phosphorus ligand featured with a P-chiral center, which is successfully utilized in asymmetric synthesis [76][77][78][79][80][81][82]. For example, the ligands L3 and L9 were discovered to realize the asymmetric addition of aromatic aldehydes and active ketones with arylboronic acids with excellent enantioselectivities [67,83]. Based on this, we envisioned to further extend the utility of these chiral phosphorous ligands in the asymmetric catalysis. Herein, we reported the Ru-catalyzed enantioselective addition of aliphatic aldehydes with arylboronic acids via P-chiral monophosphorous ligands to access chiral secondary alcohols (Scheme 2), delivering the desired aryl alkyl alcohols in excellent yields and enantioselectivities. The application for gram-scale synthesis of 3da was also disclosed.

Results and Discussions
At the outset of the research, initial experimental results show that the monophosphine ligand L1 effectively promoted the Ru (II)-catalyzed asymmetric addition of phenylpropanal 1a with 4-methylphenylboronic acid 2a to access chiral alcohol 3aa in excellent yields and with good enantioselectivity; only a trace amount of 4aa was obtained as a byproduct (Table 1, entry 1). Further ligand investigations indicated that the ligands had a remarkable influence on enantioselectivity (Table 1, entries 2~5 and entry 11). For instance, good enantioselectivities can be afforded by using L3 or L5 as chiral phosphorous ligands (Table 1, entries 3 and 5), while a racemic or nearly racemic product was formed by employing L2, L4 and L11 (Table 1, entries 2, 4 and 11), indicating that the substituents on the carbon between the phosphorus and oxygen atom appear to have a negative influence on enantioselectivity.
Similar trends were also observed by using chiral bisphosphine ligands L6~L7 or L9~L10,which were connected via α-position of a P-chiral center (Table 1, entries 6~7 and 9~10). For example, only low enantioselectivity was shown when bisphosphine ligands L6~L7 and L9~L10 were utilized in the reaction, while a moderate enantioselective product could be afforded by using L8 as a bisphosphine ligand, which was coupled at the β-position (Table 1, entry 8). In general, the ruthenium precursors have no influence Scheme 2. Asymmetric addition of aliphatic aldehydes with arylboronic acids.

Results and Discussions
At the outset of the research, initial experimental results show that the monophosphine ligand L1 effectively promoted the Ru (II)-catalyzed asymmetric addition of phenylpropanal 1a with 4-methylphenylboronic acid 2a to access chiral alcohol 3aa in excellent yields and with good enantioselectivity; only a trace amount of 4aa was obtained as a byproduct (Table 1, entry 1). Further ligand investigations indicated that the ligands had a remarkable influence on enantioselectivity (Table 1, entries 2~5 and entry 11). For instance, good enantioselectivities can be afforded by using L3 or L5 as chiral phosphorous ligands (Table 1, entries 3 and 5), while a racemic or nearly racemic product was formed by employing L2, L4 and L11 (Table 1, entries 2, 4 and 11), indicating that the substituents on the carbon between the phosphorus and oxygen atom appear to have a negative influence on enantioselectivity. The reaction was further assessed under various reaction conditions ( Table 2). Solvent screening showcased that xylene/H2O was the optimal reaction medium, providing the corresponding asymmetric addition product 3aa in full conversion and 91:9 er (Table 2, entry 7). Interestingly, full conversion and 93:7 er were obtained even when the reaction was conducted within 4 h (  Similar trends were also observed by using chiral bisphosphine ligands L6~L7 or L9~L10,which were connected via α-position of a P-chiral center ( Table 1, entries 6~7 and 9~10). For example, only low enantioselectivity was shown when bisphosphine ligands L6~L7 and L9~L10 were utilized in the reaction, while a moderate enantioselective product could be afforded by using L8 as a bisphosphine ligand, which was coupled at the β-position (Table 1, entry 8). In general, the ruthenium precursors have no influence on the yields and enantioselectivities of this reaction (Table 1, entries 12 and 13). However, only low yields and enantioselectivities are accomplished using rhodium and iridium complexes as precursors (Table 1, entries 14 and 15).
The reaction was further assessed under various reaction conditions ( Table 2). Solvent screening showcased that xylene/H 2 O was the optimal reaction medium, providing the corresponding asymmetric addition product 3aa in full conversion and 91:9 er (Table 2,  entry 7). Interestingly, full conversion and 93:7 er were obtained even when the reaction was conducted within 4 h (  Solvent screening showcased that xylene/H2O was the optimal reaction medium, providing the corresponding asymmetric addition product 3aa in full conversion and 91:9 er (Table 2, entry 7). Interestingly, full conversion and 93:7 er were obtained even when the reaction was conducted within 4 h (Table 2, entry 8). Corresponding various bases also delivered the desired chiral alcohol product 3aa in good enantioselectivities (  Having established the optimal conditions, we next investigated the scope of the reaction substrates (Scheme 3).
Generally, parasubstituted arylboronic acids bearing electron-donating or electronwithdrawing groups (2a~2h) reacted with phenyl propanal (1a) smoothly to form the relevant alcohol products (3aa, 3ac and 3ae~3ah) in excellent yields and enantioselectivities. However, only moderate yields and good enantioselectivities were observed when arylboronic acids bearing alkoxy substituents in the para position were introduced (3ab and 3ad). On the other hand, the position of the substituent onaryl boronic acid has significant influence on enantioselectivity and yield. For instance, arylboronic acids bearing substituents in the ortho position resulted in low yields and enantioselectivities (3ai and 3aj), presumably due to ortho effects [83]. Gratifyingly, the metasubstituted phenylboronic acids are also well compatible under the standard conditions, giving the desired addition products excellent enantioselectivities and yields (3ak~3am). In addition, except for thiophene boronic acid (3ao), other different boronic acids, including 2-naphthalene boronic acid, phenyl boronic acid and disubstituted boronic acid, worked well in this catalytic system to yield the alcohols in excellent yields and enantioselectivities (3an, 3ap and 3aq). Unlike aryl boronic acid bearing alkoxy, phenyl propanal bearing 4-methoxy substituent on the phenyl group resulted in a low yield but with excellent enantioselectivities (3ba and 3be). To our satisfaction, excellent yields and enantioselectivities resulted when furan-substituted propionaldehyde was loaded in this Ru-monophosphine catalytic system (3ca and 3ce). Interestingly, aldehydes with electron-withdrawing substituents (1j) are also compatible with this catalytic system, giving desirable product 3ja in the yield of 74% and 91:9 er. Generally, parasubstituted arylboronic acids bearing electron-donating or electronwithdrawing groups (2a~2h) reacted with phenyl propanal (1a) smoothly to form the relevant alcohol products (3aa, 3ac and 3ae~3ah) in excellent yields and enantioselectivities. However, only moderate yields and good enantioselectivities were observed when arylboronic acids bearing alkoxy substituents in the para position were introduced (3ab and 3ad). On the other hand, the position of the substituent onaryl boronic acid has significant influence on enantioselectivity and yield. For instance, arylboronic acids bearing substituents in the ortho position resulted in low yields and enantioselectivities (3ai and 3aj), presumably due to ortho effects [83]. Gratifyingly, the metasubstituted phenylboronic acids are also well compatible under the standard conditions, giving the desired addition products excellent enantioselectivities and yields (3ak~3am). In addition, except for thiophene boronic acid (3ao), other different boronic acids, including 2-naphthalene boronic acid, phenyl boronic acid and disubstituted boronic acid, worked well in this catalytic system to yield the alcohols in excellent yields Scheme 3. Substrates of 3-arylpropanal for asymmetric addition with arylboronic acids a,b . a Performed with 1 (0.5 mmol), 2 (2.0 eq.), [RuCl 2 (cymene)] 2 (1.0 mol%), L1 (2 mol%), K 2 CO 3 (2.0 eq.), xylene (1.5 mL) and H 2 O (0.5 mL), at 80 • C under N 2 atmosphere for 4 h. b The yield is isolated yield and ee is determined by CHIRAL column.
We next investigated the utility of this catalytic system for the substrate scope of aliphatic aldehydes bearing no aryl substituents (Scheme 4). Various common aliphatic aldehydes including cyclohexylformaldehyde, n-heptanal, isovaleraldehyde and propionaldehyde were also well tolerated in this catalytic system. Notably, the electron-withdrawing and electron-donating aryl boronic acids could be used as nucleophiles, generating chiral alcohols in 82%~93% yields and 93:7~95:5 ers (3da~3ge). Due to the steric hindrance, only a moderate yield and good enantioselectivity were afforded (3ha) when t-BuCHO was employed as a substrate under standard conditions. In addition, cinnamaldehyde can also react with 2a, giving 3ia in 90% ee. aldehydes including cyclohexylformaldehyde, n-heptanal, isovaleraldehyde and propionaldehyde were also well tolerated in this catalytic system. Notably, the electronwithdrawing and electron-donating aryl boronic acids could be used as nucleophiles, generating chiral alcohols in 82%~93% yields and 93:7~95:5 ers (3da~3ge). Due to the steric hindrance, only a moderate yield and good enantioselectivity were afforded (3ha) when t-BuCHO was employed as a substrate under standard conditions. In addition, cinnamaldehyde can also react with 2a, giving 3ia in 90% ee. Additionally, in order to gain more insight into the versatility of nucleophilic reagents, other organoboron reagents were utilized under standard conditions. To our delight, potassium trifluoro(phenyl)borate (2s) could be compatible with this catalytic system, providing the desirable product in a moderate yield and with good ee. However, no desired product 3ar was afforded when 4,4,5,5-tetramethyl-2-phenyl-1,3,2dioxaborolane (2r) was severed as an organoboron reagent in this system (Scheme 5). Additionally, in order to gain more insight into the versatility of nucleophilic reagents, other organoboron reagents were utilized under standard conditions. To our delight, potassium trifluoro(phenyl)borate (2s) could be compatible with this catalytic system, providing the desirable product in a moderate yield and with good ee. However, no desired product 3ar was afforded when 4,4,5,5-tetramethyl-2-phenyl-1,3,2-dioxaborolane (2r) was severed as an organoboron reagent in this system (Scheme 5). In order to illustrate the practicability of the catalytic system, we carried out a largescale reaction of cyclohexyl formaldehyde (1d) and p-methylphenyl boronic acid (2a). Indeed, the corresponding addition product 3da was successfully afforded on a 6.0 mmol scale from 1d in a 91% yield and 94:6 er (Scheme 6). In order to illustrate the practicability of the catalytic system, we carried out a largescale reaction of cyclohexyl formaldehyde (1d) and p-methylphenyl boronic acid (2a). Indeed, the corresponding addition product 3da was successfully afforded on a 6.0 mmol scale from 1d in a 91% yield and 94:6 er (Scheme 6). In order to illustrate the practicability of the catalytic system, we carried out a largescale reaction of cyclohexyl formaldehyde (1d) and p-methylphenyl boronic acid (2a). Indeed, the corresponding addition product 3da was successfully afforded on a 6.0 mmol scale from 1d in a 91% yield and 94:6 er (Scheme 6). The model reaction was measured over time to know about details of this reaction (Scheme 7). To our surprise, the treatment of 1a with 2a under standard conditions within 4 min afforded the desirable alcohol product 3aa in the yield of 96% and 92:8 er, albeit with trace amounts of a ketone byproduct. The best enantioselectivity of product 3aa was observed in 4 h, after which the enantioselectivities of alcohol decreased slightly, and the byproduct ketone increased slightly as time went by. In addition, the racemization of enantioenriched product 3aa was monitored over time under the standard system, which showcased that the yield of 4aa increased gradually, and a slight loss of enantioselectivity was observed (Scheme 8).
The model reaction was measured over time to know about details of this reaction (Scheme 7). To our surprise, the treatment of 1a with 2a under standard conditions within 4 min afforded the desirable alcohol product 3aa in the yield of 96% and 92:8 er, albeit with trace amounts of a ketone byproduct. The best enantioselectivity of product 3aa was observed in 4 h, after which the enantioselectivities of alcohol decreased slightly, and the byproduct ketone increased slightly as time went by. In addition, the racemization of enantioenriched product 3aa was monitored over time under the standard system, which showcased that the yield of 4aa increased gradually, and a slight loss of enantioselectivity was observed (Scheme 8). Based on the experimental results, we proposed a possible mechanistic cycle described in Scheme 9 [84]. Firstly, the catalyst Ru-L1 was generated in situ with an Ru precursor and L1, followed by a transmetalation with aryl boronic acid to form Int-I under the condition of base. Subsequently, there was the coordination of aliphatic aldehydes with Int-I to generate the Int-II, followed by carbonyl insertion to produce Int-III. The β-H elimination of Int-III produced byproduct 4. Finally, the addition of product 3 was released, and the catalyst was regenerated under the action of water and aryl boronic acid. Based on the experimental results, we proposed a possible mechanistic cycle described in Scheme 9 [84]. Firstly, the catalyst Ru-L1 was generated in situ with an Ru precursor and L1, followed by a transmetalation with aryl boronic acid to form Int-I under the condition of base. Subsequently, there was the coordination of aliphatic aldehydes with Int-I to generate the Int-II, followed by carbonyl insertion to produce Int-III. The β-H elimination of Int-III produced byproduct 4. Finally, the addition of product 3 was released, and the catalyst was regenerated under the action of water and aryl boronic acid. Scheme 8. Plots for relationship of time, enantioselectivity and ketone formation.
Based on the experimental results, we proposed a possible mechanistic cycle described in Scheme 9 [84]. Firstly, the catalyst Ru-L1 was generated in situ with an Ru precursor and L1, followed by a transmetalation with aryl boronic acid to form Int-I under the condition of base. Subsequently, there was the coordination of aliphatic aldehydes with Int-I to generate the Int-II, followed by carbonyl insertion to produce Int-III. The β-H elimination of Int-III produced byproduct 4. Finally, the addition of product 3 was released, and the catalyst was regenerated under the action of water and aryl boronic acid.

Scheme 9. Proposed mechanism.
In summary, we described a Ru-catalyzed asymmetric addition of aliphatic aldehydes with aryl boronic acids based on a monophosphorous ligand, providing the corresponding alcohol products in excellent yields and enantioselectivities. A large-scale experiment showcased the utility of this catalytic system, which provides a supplementary method on acquiring chiral aryl alky alcohols. In summary, we described a Ru-catalyzed asymmetric addition of aliphatic aldehydes with aryl boronic acids based on a monophosphorous ligand, providing the corresponding alcohol products in excellent yields and enantioselectivities. A large-scale experiment showcased the utility of this catalytic system, which provides a supplementary method on acquiring chiral aryl alky alcohols.