Cobalt-catalyzed highly enantioselective hydrogenation of α,β-unsaturated carboxylic acids.

Asymmetric hydrogenation of α,β-unsaturated acids catalyzed by noble metals has been well established, whereas, the asymmetric hydrogenation with earth-abundant-metal was rarely reported. Here, we describe a cobalt-catalyzed asymmetric hydrogenation of α,β-unsaturated carboxylic acids. By using chiral cobalt catalyst bearing electron-donating diphosphine ligand, high activity (up to 1860 TON) and excellent enantioselectivity (up to >99% ee) are observed. Furthermore, the cobalt-catalyzed asymmetric hydrogenation is successfully applied to a broad spectrum of α,β-unsaturated carboxylic acids, such as various α-aryl and α-alkyl cinnamic acid derivatives, α-oxy-functionalized α,β-unsaturated acids, α-substituted acrylic acids and heterocyclic α,β-unsaturated acids (30 examples). The synthetic utility of the protocol is highlighted by the synthesis of key intermediates for chiral drugs (6 cases). Preliminary mechanistic studies reveal that the carboxy group may be involved in the control of the reactivity and enantioselectivity through an interaction with the metal centre.

Introduction

Chiral class="Chemical">carboxylic acids are prevaleclass="Chemical">nt structural uclass="Chemical">nits iclass="Chemical">n varieties of pharmaceutical molecules, agrochemicals, flavors, aclass="Chemical">nd fragraclass="Chemical">nces. Some examples are exhibited iclass="Chemical">n Fig. 1. Chiral class="Chemical">n class="Chemical">carboxylic acids such as Ibuprofen, Naproxen[1,2], and (R)-Tiagabine[3] are well-known drugs. The key intermediates for a large number of drugs or bioactive compounds, such as Artemisinin[4,5], Rupintrivir[6,7], (S)-Equol[8], and Sacubitril[9-12] are chiral carboxylic acids. In addition, chiral carboxylic acids are versatile intermediates for organic synthesis, as they are utilized in the construction of various C–C bonds via decarboxylative coupling reaction[13-16], and in the activation of C–H bond as a directing group[17,18]. Thus, the development of efficient methods for the preparation of chiral carboxylic acids is highly demanded in both academic research and industrial production.

Fig. 1

Biologically active compounds and drugs derived from chiral carboxylic acid.

Ibuprofen and Naproxen (nonsteroidal anti-inflammatory drugs); (R)-Tiagabine (γ-aminobutyric acid reuptake inhibitor); Artemisinin (antimalarial drug); Rupintrivir (rhinovirus protease inhibitor); (S)-Equol (soy isoflavonoid metabolite); Sacubitril (antihypertensive drug used in combination with Valsartan).

class="Chemical">Ibuprofen aclass="Chemical">nd class="Chemical">n class="Chemical">Naproxen (nonsteroidal anti-inflammatory drugs); (R)-Tiagabine (γ-aminobutyric acid reuptake inhibitor); Artemisinin (antimalarial drug); Rupintrivir (rhinovirus protease inhibitor); (S)-Equol (soy isoflavonoid metabolite); Sacubitril (antihypertensive drug used in combination with Valsartan).

Transition-class="Chemical">metal-catalyzed eclass="Chemical">naclass="Chemical">ntioselective class="Chemical">n class="Chemical">hydrogenation of α,β-unsaturated acids is one of the most atom-economic and efficient approaches to access chiral carboxylic acids[19]. Various noble metal-based catalysts, such as Ru catalysts with chiral diphosphine ligands[20-24], Rh catalysts with chiral phosphorus or nitrogen-containing ligands[25-31], and Ir catalysts with chiral P,O-[32] and P,N-ligands[33-40] have been developed for the hydrogenation of different unsaturated carboxylic acids in high enantioselectivities (Fig. 2a). As modification of ligands or catalysts is necessary to achieve high enantioselectivity for different substrate types, the development of catalytic system with wide applicability is still highly desirable. On the other hand, a major concern regarding the noble metal-based hydrogenation catalysts is the sustainability. Ru, Rh, and Ir are extremely scarce elements, occurring at very low abundances in the earth’s crust (5 × 10−5–10−4 ppm)[41]. Owing to the low abundance and high cost of the noble metals, the development of cheap earth-abundant metal substitutes to the noble metal catalysts is of high significance[42-54].

Fig. 2

Transition metal-catalyzed asymmetric hydrogenation of α,β-unsaturated carboxylic acids.

a Previous work: noble metal-based catalysts. b This work: cobalt-catalyzed asymmetric hydrogenation of α,β-unsaturated carboxylic acids.

a Previous work: noble class="Chemical">metal-class="Chemical">n class="Chemical">based catalysts. b This work: cobalt-catalyzed asymmetric hydrogenation of α,β-unsaturated carboxylic acids.

Catalysts class="Chemical">based oclass="Chemical">n earth-abuclass="Chemical">ndaclass="Chemical">nt aclass="Chemical">nd eclass="Chemical">nvclass="Chemical">n class="Chemical">ironmentally benign cobalt are highly attractive for asymmetric alkene hydrogenation[55-57]. Early studies have showed the potential application of cobalt catalysts in the asymmetric alkene hydrogenation[58-61] However, these systems suffer from limited enantioselectivities, and in many cases H2 could not be used as the stoichiometric reductant. Appreciable progress has been made in recent years, employing cobalt-based complexes for asymmetric hydrogenation of alkenes[62-72]. Chirik and coworkers report the highly enantioselective hydrogenation of styrene derivatives, cyclic alkenes, and enamides with cobalt complexes bearing C1-symmetric PNN-type pincer ligand or chiral diphosphine ligands[65,71,72]. Stereoselective olefin hydrogenations are also independently described by Lu and Huang group employing chiral IPO–Co and PPO–Co complexes[62,66,67,69]. Though important progresses have been made, the development of base metal catalyst for challenging substrates is still underexplored[56].

As a class="Chemical">coclass="Chemical">nticlass="Chemical">nuatioclass="Chemical">n of our iclass="Chemical">nterest iclass="Chemical">n traclass="Chemical">nsitioclass="Chemical">n class="Chemical">n class="Chemical">metal-catalyzed asymmetric hydrogenation reactions, herein, we report a highly enantioselective cobalt-catalyzed hydrogenation of α,β-unsaturated carboxylic acids for the synthesis of chiral carboxylic acids. High yields and enantioselectivities are generally achieved for a wide range of substrates (up to 99% yield and up to >99% ee). Besides, the synthetic value of the methodology is demonstrated by its applications in the synthesis of important drugs (Fig. 2b). After the first submission of our paper, the asymmetric hydrogenation of α,β-unsaturated carboxylic acids is reported by Chirik and coworkers[73] with a cobalt catalytic system, and high yields and enantioselectivities are achieved for a wide range of acrylic acid derivatives.

Results

Condition optimization

In the initial study, asymmetric class="Chemical">hydrogenatioclass="Chemical">n of class="Chemical">n class="Chemical">(E)-2,3-diphenylacrylic acid 1a was investigated by using 5 mol% of Co(acac)2 and 5 mol% of chiral ligand. The reaction was conducted under 60 atm H2 pressure at 50 °C in MeOH over 24 h. Several chiral diphosphine ligands were evaluated and we found that cobalt catalysts ligated by highly electron-rich and sterically demanding diphosphines were catalytically active (Fig. 3). In particular, the Ph-BPE was found to be the best one, which afforded the desired product 2a in 70% yield and 94% ee. Me-DuPhos and iPr-DuPhos were catalytically active, but only lower yields and moderate ee values were obtained. Other strongly electron-donating bis(alkylphosphine)s like Binaphine and Duanphos were also tested, but they afforded unsatisfactory yields and ee values. P-chiral diphosphine QuinoxP* and BenzP* gave no desired product. The less electron-donating bis(arylphosphine)s such as BINAP and Segphos were completely inactive in cobalt-catalyzed hydrogenation (Supplementary Table 2).

Fig. 3

The performance of chiral diphosphine ligands in Co-catalyzed asymmetric hydrogenation of 1a.

The yields were determined by 1H NMR and the enantioselectivities were determined by HPLC analysis in all cases.

The yields were determined by n class="Chemical">1H NMR aclass="Chemical">nd the eclass="Chemical">naclass="Chemical">ntioselectivities were determiclass="Chemical">ned by HPLC aclass="Chemical">nalysis iclass="Chemical">n all cases.

Next, various solvents and additives were tested (Table 1). It became apparent that alclass="Chemical">coholic solveclass="Chemical">nts are beclass="Chemical">neficial for the catalytic asymmetric class="Chemical">n class="Chemical">hydrogenation, with iPrOH giving the highest activity and enantioselectivity (entries 1–2). The run in more acidic fluorinated solvent, trifluoroethanol (TFE) afforded moderate ee value (entry 3). Furthermore, reactions performed in dimethoxyethane (DME) and toluene led to very low conversions (entries 4–5). When tetrahydrofuran (THF) and 1,4-dioxane were used as solvents, the reactions were totally inhibited (entries 6–7). The investigation of the effect of additives in the reaction showed that the addition of one-electron reductant is benefit for obtaining full conversions with low catalyst loading (entries 8–12). To further improve the enantiocontrol, temperature and hydrogenation pressure effects were surveyed. 97% ee of 2a could be achieved at room temperature under 40 atm H2 pressure with full conversion of 1a (entry 13). The asymmetric hydrogenation also proceeded smoothly with full conversion and high ee when reducing the catalyst loading to 0.1 mol% (entry 14). Moreover, in the presence of 0.05 mol% of catalyst, 1a reacted with H2 efficiently, giving 2a in 93% yield on 2 mmol scale without any decrease of the ee value (TON up to 1860, entry 15). The absolute configuration of the product 2a was established by comparison of its optical rotation with previous report (Supplementary Tables 3–5).

Table 1

Optimization for cobalt-catalyzed asymmetric hydrogenation of (E)-2,3-diphenylacrylic acid.

Entry	x (mol%)	Additive	Solvent	Conv. (%)a	ee (%)b
1	5	—	MeOH	70	94
2	5	—	iPrOH	>98	93
3c	5	—	TFE	>98	84
4	5	—	DME	7	43
5	5	—	Toluene	15	86
6	5	—	THF	NR	—
7	5	—	1,4-Dioxane	NR	—
8	1	—	iPrOH	61	93
9	1	Cs2CO3	iPrOH	7	88
10	1	KOtBu	iPrOH	55	93
11	1	Mn	iPrOH	>98	96
12	1	Zn	iPrOH	>98	97
13d	1	Zn	iPrOH	>98	97
14e	0.1	Zn	iPrOH	>98	97
15e	0.05	Zn	iPrOH	93	97

TFE trifluoroethanol, DME dimethoxyethane, THF tetrahydrofuran.

Conditions: 1a (0.1 mmol) in solvent (0.6 mL) under 60 atm H2 pressure at 50 °C for 24 h.

aDetermined by 1H NMR.

bDetermined by HPLC analysis.

cUsing TFE (0.6 mL) and THF (0.6 mL) as solvent.

dUnder 40 atm H2, room temperature.

e2 mmol scale, under 80 atm H2, 7 d.

Optimization for class="Chemical">cobalt-catalyzed asymmetric class="Chemical">n class="Chemical">hydrogenation of (E)-2,3-diphenylacrylic acid.

class="Chemical">TFE trifluoroethanol, class="Chemical">n class="Chemical">DME dimethoxyethane, THF tetrahydrofuran.

n class="Chemical">Coclass="Chemical">nditioclass="Chemical">ns: 1a (0.1 mmol) iclass="Chemical">n solveclass="Chemical">nt (0.6 mL) uclass="Chemical">nder 60 atm class="Chemical">n class="Chemical">H2 pressure at 50 °C for 24 h.

aDetermined by n class="Chemical">1H NMR.

cUsing n class="Chemical">TFE (0.6 mL) aclass="Chemical">nd class="Chemical">n class="Chemical">THF (0.6 mL) as solvent.

dUnder 40 atm n class="Chemical">H2, room temperature.

e2 mmol scale, under 80 atm n class="Chemical">H2, 7 d.

Substrate scope

To delineate the sclass="Chemical">cope of the class="Chemical">n class="Chemical">Co-catalyzed asymmetric hydrogenation, the catalytic system was applied to the reactions of different types of α,β-unsaturated acids. α-Aryl and α-alkyl cinnamic acid derivatives 1 were hydrogenated with 1 mol% catalyst loading under the optimized condition (Fig. 4). Most of the reactions proceeded smoothly under 40 atm H2 pressure at ambient temperature, providing the desired products in high isolated yields and excellent enantioselectivities (95–99% ee). The method works efficiently for α-methyl cinnamic acid derivatives bearing both electron-donating (1c–1e) and -withdrawing groups (1f–1i). Substituents at the para and meta positions of the phenyl ring are tolerated under the reaction conditions, furnishing the hydrogenation products in high yields with excellent enantioselectivity (1j–1l). Heteroaromatic α,β-unsaturated acid 1m was also efficiently hydrogenated to give the desired product 2m in 90% isolated yield with 97% ee. Increasing the steric demand in the α-substituted position by using an iPr group led to the formation of 2n with 86% ee.

Fig. 4

Cobalt-catalyzed asymmetric hydrogenation of various α,β-unsaturated carboxylic acids.

aConditions: 1 (0.1 mmol), Co(acac)2 (1 mol%), (S,S)-Ph-BPE (1 mol%) and Zn (10 mol%) in iPrOH (0.6 mL) under 40 atm H2 pressure at room temperature for 24 h. Yield of isolated products, unless noted otherwise. The ee values were determined by HPLC analysis. bUnder 60 atm H2 pressure, 50 °C. cUnder 80 atm H2 pressure, 50 °C. dWith 5 mol% of Cat. eWith 5 mol% of CoCl2 and (S,S)-Ph-BPE in MeOH (0.4 mL), 72 h. f3 (0.1 mmol), CoCl2 (5 mol%), (S,S)-Ph-BPE (5 mol%), and Zn (50 mol%) in HFIP (0.4 mL) under 60 atm H2 pressure at 50 °C for 48 h. gIn 0.6 mL solvent (MeOH/HFIP = 2/1), 72 h. hIn 0.4 mL MeOH.

an class="Chemical">Coclass="Chemical">nditioclass="Chemical">ns: 1 (0.1 mmol), class="Chemical">n class="Chemical">Co(acac)2 (1 mol%), (S,S)-Ph-BPE (1 mol%) and Zn (10 mol%) in iPrOH (0.6 mL) under 40 atm H2 pressure at room temperature for 24 h. Yield of isolated products, unless noted otherwise. The ee values were determined by HPLC analysis. bUnder 60 atm H2 pressure, 50 °C. cUnder 80 atm H2 pressure, 50 °C. dWith 5 mol% of Cat. eWith 5 mol% of CoCl2 and (S,S)-Ph-BPE in MeOH (0.4 mL), 72 h. f3 (0.1 mmol), CoCl2 (5 mol%), (S,S)-Ph-BPE (5 mol%), and Zn (50 mol%) in HFIP (0.4 mL) under 60 atm H2 pressure at 50 °C for 48 h. gIn 0.6 mL solvent (MeOH/HFIP = 2/1), 72 h. hIn 0.4 mL MeOH.

Our preliminary results show that the [class="Chemical">Co]/BPE is also active for class="Chemical">n class="Chemical">hydrogenation of N-heterocyclic acids to produce chiral heterocyclic acids, which are present in various pharmaceuticals[74]. The asymmetric hydrogenation of N-Boc-1,2,5,6-tetrahydropyridine-3-carboxylic acid 1o gave the desired product 2o with an impressive 93% ee. α-Oxy-functionalized α,β-unsaturated acids were also subjected to our [Co]/BPE system due to the corresponding hydrogenation products are important building blocks for asymmetric synthesis in both pharmaceutical and agrochemical industries[75]. To our delight, the catalyst system showed high efficiency in the asymmetric hydrogenation of α-alkoxy- and α-aryloxy-substituted α,β-unsaturated acids. A broad range of α-aryloxy and α-alkoxy substituted α,β-unsaturated acids 1p–1u were hydrogenated smoothly to give the desired chiral α-oxy-functionalized acids in good to excellent yields and enantioselectivities (80–95% yield, 90–99% ee).

Finally, we turned our attention to the asymmetric class="Chemical">hydrogenatioclass="Chemical">n of α-substituted class="Chemical">n class="Chemical">acrylic acids (Fig. 4). It is revealed that solvents have critical roles in [Co]/BPE catalyzed hydrogenation of α-substituted acrylic acids, affecting both catalytic activity and selectivity. Full conversion and 98% ee were obtained in hexafluoroisopropanol (HFIP, Supplementary Table 6). A wide range of α-aryl acrylic acids (3a–3h) were subjected to the cobalt-catalyzed hydrogenation, affording the corresponding chiral carboxylic acids in good to excellent yields and enantioselectivities (89–99% yields, 90–99% ee). Furthermore, 2-alkyl acrylic acid 3i was also smoothly hydrogenated with 80% ee.

Synthetic applications

To demonstrate the synthetic utility of class="Chemical">cobalt-catalyzed eclass="Chemical">naclass="Chemical">ntioselective class="Chemical">n class="Chemical">hydrogenation of α,β-unsaturated acids, its synthetic application in series of chiral natural products and drugs was studied. The critical synthon to (S)-Equol, (S)-2v, could be easily accessed under standard conditions with high enantioselectivity (94% ee, Fig. 5a). Similarly, chiral carboxylic acid 2w, a key intermediate in the synthesis of rhinovirus protease inhibitor Rupintrivir, was obtained in 95% isolated yield and >99% ee (Fig. 5b). Following optimization of the reaction conditions, Sacubitril intermediate 2x was obtained in 97% isolated yield and 17/1 dr (Fig. 5c, Supplementary Table 7).

Fig. 5

Practical synthetic applications of cobalt-catalyzed asymmetric hydrogenation.

a Synthesis of (S)-Equol intermediate via cobalt-catalyzed hydrogenation. b Synthesis of Rupinnavir intermediate. c Synthesis of Sacubitril intermediate. d Asymmetric synthesis of Naproxen. e Asymmetric synthesis of Ibuprofen. f Gram-scale asymmetric hydrogenation of artemisinic acid 5.

a Synthesis of class="Chemical">(S)-Equol iclass="Chemical">ntermediate via class="Chemical">n class="Chemical">cobalt-catalyzed hydrogenation. b Synthesis of Rupinnavir intermediate. c Synthesis of Sacubitril intermediate. d Asymmetric synthesis of Naproxen. e Asymmetric synthesis of Ibuprofen. f Gram-scale asymmetric hydrogenation of artemisinic acid 5.

The asymmetric class="Chemical">hydrogenatioclass="Chemical">n of α-substituted class="Chemical">n class="Chemical">acrylic acids is of highly practical value because optically pure α-substituted propionic acids, such as ibuprofen and naproxen, are well-known non-steroid anti-inflammatory drugs, which can be readily prepared through asymmetric hydrogenation of the corresponding α-aryl acrylic acids. Indeed, Naproxen (4j) and Ibuprofen (4k) were prepared with high yields and enantioselectivities by using cobalt-based catalytic system (Fig. 5d, e). Based on the excellent performance of Co/BPE in the above reaction, we carried out the gram-scale asymmetric hydrogenation of AA (artemisinic acid) 5 in the presence of 0.5 mol% of chiral cobalt catalyst. Desired product (R)-DHAA (dihydroartemisinic acid) 6 was obtained in excellent isolated yield and dr value (1.16 g, 97/3 dr, Fig. 5f). It is noteworthy that the trisubstituted olefin in 5 is unreactive under the Co-catalyzed hydrogenation conditions.

Mechanism study

To provide insight into the possible catalyst activation mode and the mechanism of the asymmetric class="Chemical">hydrogenatioclass="Chemical">n of α,β-class="Chemical">n class="Chemical">unsaturated carboxylic acids, several control and catalytic experiments were conducted. Previous research by Chirik and coworkers revealed that additives such as Zn, Mn, or LiCH2Si(CH3)3 were necessary in the cobalt-catalyzed asymmetric hydrogenation of enamines and acted as an activator for pre-catalyst to generate catalytic active cobalt speices[65,71]. However, the reaction of α,β-unsaturated carboxylic acid 1a, in the absence of any additives, afford 2a in >98% conversion and 94% ee (Table 1, entry 2). The data suggested that the carboxylic acid might serve as an activator and mediate the activation process. In sharp contrast to the high reactivity and enantioselectivity observed in the case of 1b (99% yield and 96% ee, Fig. 4), no reaction occurred for the corresponding ethyl ester 1b′ under standard conditions (Fig. 6a). Moreover, no hydrogenation product was observed when AcOH or 1b was added to the reaction mixture as external carboxylic acid (Fig. 6b, c). These results indicated that the carboxy group of substrate may be involved in the control of the reactivity and enantioselectivity through interaction with the metal center[19,22-24,33].

Fig. 6

Control experiments and mechanistic investigations.

a Hydrogenation of ester 1b′ under standard conditions. b Hydrogenation of ester 1b′ in the presence of catalytic amount of CH3COOH. c Reaction of ester 1b′ in the presence of catalytic amount of 1b. d Deuterium-labeling experiment. e Hydrogenation of 1b in isopropanol-d8 with CH3COOD as additive.

a class="Chemical">Hydrogenatioclass="Chemical">n of class="Chemical">n class="Chemical">ester 1b′ under standard conditions. b Hydrogenation of ester 1b′ in the presence of catalytic amount of CH3COOH. c Reaction of ester 1b′ in the presence of catalytic amount of 1b. d Deuterium-labeling experiment. e Hydrogenation of 1b in isopropanol-d8 with CH3COOD as additive.

The class="Chemical">deuterium-labeliclass="Chemical">ng experimeclass="Chemical">nts were also class="Chemical">n class="Chemical">conducted to elucidate the reaction mechanism. The reaction of 1b with D2 under standard condition produced 2b- smoothly in >98% conversion and 96% ee (Fig. 6d). Performing the hydrogenation reaction under 40 atm of H2 in isopropanol-d8 solution with 5 eq. CH3COOD gave 2b in >98% conversion, and no deuterated products were observed (Fig. 6e). These data suggested that the H2 served as the hydrogen source and protonation of the Co-alkyl intermediate was probably not involved in the current reaction (Supplementary Figs. 1–5). Besides, EPR experiments were also conducted to monitor the process of the current reaction using 1b as model substrate. The EPR spectra changed greatly after the addition of (S,S)-Ph-BPE and substrate 1b, suggesting that the coordination of the ligand and the exchange between acac and substrate 1b probably happened (Supplementary Figs. 6–11).

class="Chemical">Based oclass="Chemical">n our experimeclass="Chemical">ntal observatioclass="Chemical">ns aclass="Chemical">nd the previous study oclass="Chemical">n class="Chemical">n class="Chemical">iridium-catalyzed enantioselective hydrogenation of α,β-unsaturated acids[33], we proposed a plausible mechanism (Fig. 7). The key catalytic intermediate E could be generated through two pathways. In the absence of Zn, complex E can be generated through carboxy group mediated H2 heterolytic process. Coordination of Co(acac)2 with (S,S)-Ph-BPE generates Co(II) complex A, which then undergoes ligand exchange with more acidic substrate 1b to produce complex B, complex B and B′ are in equilibrium with each other. Heterolysis of H2 by complex B′ produces the key catalytic species E via transition state C. Intermediate E may be also produced through protonation[76] of dihydride complex D[65,73] with 1b when employing one-electron reductant. Chirik and coworkers[73] reported an alternative mechanism which involved the migratory insertion of the dihydride complex and the subsequent reduction elimination as key step. Note that the addition of one-electron reductant is beneficial for obtaining full conversions with low catalyst loading (vide supra). The success of Zn or Mn used in this reaction is presumably due to enhanced activation of Co(II) precursor and suppressed deactivation of the active catalyst. The key intermediate E then enters the catalytic cycle. Intramolecular migratory insertion of complex E produces five-membered intermediate F. Coordination of H2 to F forms complex G, which undergoes subsequent sigma-bond metathesis to give complex H. The ligand exchange of intermediate H with unsaturated carboxylate substrate releases the hydrogenation product 2b and regenerates the cobalt hydride complex D. Mechanistic investigations indicate that the carboxy group has a pivotal role in the improvement of the reactivity and enantioselectivity via coordination with the metal center, and that is why the current reaction does not work for α,β-unsaturated esters.

Fig. 7

Proposed catalytic cycle.

a Generation of catalytic species E in the absence of Zn. b Generation of catalytic species E in the presence of Zn. c Proposed catalytic cycle.

a Generation of catalytic species E in the absence of n class="Chemical">Zn. b Geclass="Chemical">neratioclass="Chemical">n of catalytic species E iclass="Chemical">n the preseclass="Chemical">nce of class="Chemical">n class="Chemical">Zn. c Proposed catalytic cycle.

Discussion

In summary, we have developed a highly efficient class="Chemical">cobalt-catalyzed asymmetric class="Chemical">n class="Chemical">hydrogenation of α,β-unsaturated carboxylic acids. The cobalt catalyst system exhibited both high activities (up to 1860 TON) and excellent enantioselectivities (up to >99% ee) for a broad spectrum of α,β-unsaturated acids. It also affords an efficient way to the key intermediates of many chiral natural products and drugs with high enantioselectivities. Moreover, this operationally simple and atom-economic protocol could be easily scaled-up in gram-scale using 0.5 mol% catalyst loading for the asymmetric synthesis of key intermediate of Artemisinin. Mechanistic studies suggest that the carboxy group may be involved in the control of the reactivity and enantioselectivity through interaction with the metal center. Additional mechanistic and DFT studies of the asymmetric transformations with such cobalt system are currently underway in our laboratory.

Methods

General procedure for asymmetric hydrogenation

In an class="Chemical">argon-filled glovebox, class="Chemical">n class="Chemical">Co(acac)2 (0.010 M in iPrOH, 0.10 mL, 0.001 mmol) and (S,S)-Ph-BPE (0.010 M in THF, 0.10 mL, 0.001 mmol) were stirred in a vial at room temperature for 10 min. Then zinc dust (0.65 mg, 0.01 mmol) and iPrOH (0.50 mL) were added and the mixture was stirred for 15 min. After that, substrate (0.1 mmol) was added to the reaction mixture. The vial was subsequently transferred into an autoclave and purged by three cycles of pressurization/venting with H2. The reaction was then stirred under H2 (40 atm) at room temperature for 24 h. The hydrogen gas was released slowly and carefully. The resulting solution was concentrated in vacuum and the residue was purified by chromatography on silica gel. The ee values were determined by HPLC with a chiral column.