Asymmetric transfer hydrogenation of γ-aryl α,γ-dioxo-butyric acid esters

Abstract

The asymmetric transfer hydrogenation (ATH) of a series of γ-aryl-α,γ-dioxo-butyric acid esters has been accomplished smoothly. Six ferrocene-based chiral ligands have been prepared and applied in the reactions respectively. Simultaneously, enantiopure Ts-DPEN's utilization in the ATH also has been investigated and the products were obtained in 30–85% chemical yields with 37–95.5% ee.

---

Introduction

Chiral α-hydroxy-γ-keto-butyric acid ethyl esters are key building blocks for some important bioactive compounds, such as Croomia,^1,2 Citrafungins³ and other natural extracts.^4–10 The synthesis of the chiral pharmaceutical α-hydroxy-γ-keto intermediates is of great interest in organic chemistry. However, only a few methods have been reported on their preparation such as asymmetric reduction via enzymatic catalysis,^11–13 enantioselective aza-ene-type reactions^14,15 as well as Brønsted acid catalyzing asymmetric aldol reactions¹⁶ and heterogeneous enantioselective hydrogenations.^17,18 Compared with the above reported methods, transition metal-catalyzed asymmetric transfer hydrogenation (ATH) should be a more facile access because of concise procedure and easy operation. As a kind of powerful method for asymmetric transformation,¹⁹ ATH is widely used in both academics and industrials. Since it was reported in early 1980s,²⁰ many excellent chiral ligands have been found and various substrates have also been investigated.²¹ Nowadays the pursuit for novel chiral ligands²² and the new applications of the existed ligands^23–25 for ATH were still attractive. As far as the substrates in ATH were concerned, the class of chemicals with several carbonyl moieties in one molecule^26–29 were very interesting because multiple-carbonyl groups mean many synthetic applications. But there were limited papers on the ATH of multi-carbonyl compounds.³⁰ It was worthy of notice that the ATH of γ-aryl-α,γ-dioxo-butyric acid esters is seldom reported to date (Scheme 1).

Scheme 1 Asymmetric transfer hydrogenation (ATH) of γ-aryl α,γ-dioxo-butyric acid ester.

Ferrocene was an outstanding skeleton for ligand design.³¹ There were lots of excellent ferrocene-based chiral ligands successfully used in many asymmetric transformations.³² Our group have also developed a series of ferrocene-based chiral ligands. They possessed excellent enantioselectivities in many reactions.^33–36 Herein, we have chose two of them (L₁, L₂) as well as other four new prepared ligands (L₃–L₆) (Fig. 1) to carry out the research of the ATH of α,γ-dioxo-butyric acid esters. Simultaneously, the classical Noroyi's catalysts RuCl(p-cymene)[Ts-DPEN] (Fig. 2) have also been examined in this reaction to prepare chiral α-hydroxy-γ-keto-butyric acid ethyl esters in detail.

Fig. 1 The structure of ferrocene-based chiral ligands used in the ATH.

Fig. 2 The structure of RuCl(p-cymene)[Ts-DPEN].

Results and discussion

The synthesis of ferrocene-based chiral ligands

The preparation of the six ferrocene-based chiral ligands (Fig. 1) was straightforward and as shown in Scheme 2. According to literature,³⁶ ortho-lithiation of (R)-Ugi's amine and subsequent treatment with ClPAr₂, Ac₂O and a large excess of ammonia, intermediate 1a could be obtained. Then the reductive amination of 1a with picolinaldehydes resulted in the formation of ligands L₁–L₃ in 30–54% isolated yields.³⁷ Similar process also could lead to L₄ (50% yield). If (R)-Ugi's amine was stirred with Ac₂O at 100 °C under nitrogen conditions, acetate 2 was furnished and could be used in the next step without further purification.³⁶ After amination of 2 by (S,S) or (R,R)-1,2-diphenyl-1,2-ethanediamine in CH₃OH, L₅ or L₆ was produced directly.³⁶

Scheme 2 The route to ferrocene-based chiral ligands.

As to the structure of the above ligands, L₁, L₂ and L₃ had both central chirality and planar chirality while L₄, L₅ and L₆ only possessed stereogenic carbon. They were designed for detection of the planer chirality and the match of multicentre chirality on the effect of the ATH reaction.

The ATH catalyzed by RuCl(p-cymene)(Ts-DPEN)

Noroyi's catalysts, RuCl(p-cymene)[(S,S)-Ts-DPEN] and its enantiomer have been star catalysts in ATH and they can be got in a commercial way. So their applications in ATH of γ-aryl-α,γ-dioxo-butyric acid esters (Scheme 1) have been carried out firstly.

A survey of reaction media was finished with γ-phenyl-α,γ-dioxo-butyric acid ester (3a)^38,39 as model substrate. Different hydrogen sources, various solvents and temperature were probed.

Initially, at room temperature (r.t.), two most common hydrogen source: both i-PrOH-KOH system and HCOOH/Et₃N (5 : 2) system have been tested. But after monitored by TLC, no transformation has happened in i-PrOH system. On the other hand, reaction in HCOOH/Et₃N (5 : 2) system have proceeded smoothly and afforded α-hydroxy ester with the yield of 85% and 84% ee (Table 1, entry 1). So HCOOH/Et₃N (5 : 2) was selected as hydrogen source for further experiments.

Table 1 The ATH reactions catalyzed by Ru-(S,S)-Ts-DPEN complex

Entry^a	Temp.	Solvent	Yield (%)	ee^e (%) (conf.)
a 1 mmol 3a with 0.0025 mmol of Ru(p-cymene)Cl and 0.005 mmol of (S,S)-Ts-DPEN for each entry.b Dissolution in 4 mL HCOOH/Et3N (5 : 2) and 1 mL solvent.c Using 5 mL of HCOOH/Et3N (5 : 2) without solvent.d 0.005 mmol of (R,R)-Ts-DPEN instead of (S,S)-Ts-DPEN was employed.e The enantiomeric excess (ee) and configuration (conf.) of product were determined by chiral HPLC with Chiralpak OD-H column and according to literature or X-ray crystal.
1^b	r.t.	DMF	85	84(S)
2^b	r.t.	MeOH	30	73(S)
3^b	r.t.	Et2O	Trace	ND
4^b	r.t.	DCM	Trace	ND
5^b	r.t.	THF	73	50(S)
6^b	r.t.	EtOAc	80	23(S)
7^b	r.t.	Dioxane	77	60(S)
8^b	r.t.	t-BuOMe	82	81(S)
9^c	r.t.	—	75	79(S)
10^b	0 °C	DMF	80	84(S)
11^b	−20 °C	DMF	68	94(S)
12^b	−40 °C	DMF	Trace	ND
13^b^,^d	−20 °C	DMF	68	94(R)

---

To optimize the reaction efficiency, several solvents have been examined. We have observed that the ee in proton solvent (MeOH) was moderate (73% ee), but the yield was low (30%) (Table 1, entry 2). Moreover, the results in nonproton solvents, such as DMF, dioxane, DCM, EtOAc and t-BuOMe, were better than that in proton solvent (Table 1, entry 2–8). Especially, polar nonproton solvent DMF gave the highest yield (85%) and the highest ee (84% ee). The solvent influence on the reaction may related to the formation of transition state. Bigger steric hindrance solvent corresponded to better enantioselectivity.

In order to investigate temperature effect on the reaction, we have decreased the temperature from r.t. to 0 °C. But the ees didn't change (Table 1, entry 1 vs. entry 10). At −20 °C, much higher optical yield (94% ee) accompanied with lower chemical yield (68%) was observed (Table 1, entry 11). However, at more lower temperature (−40 °C), the reaction only had trace conversion (Table 1, entry 12). Lastly, −20 °C was determined as the optimal temperature.

We have also found that the configuration of the product was controlled by the ligand. Switching the configuration of the Ts-DPEN from (S,S) to (R,R), the product configuration also changed to (R)-configuration (Table 1, entry 13).

Under the optimized conditions, a wide range of substrates have been put into the reaction and RuCl(p-cymene)[(R,R)-Ts-DPEN] has been employed (Table 2). Form these reaction, we noticed that substituent on the γ-phenyl had not effect on the results. 3b–3d, which separately possessed electron withdrawing group (F, Cl, Br), afforded almost the same results as 3e did (Table 2, entry 1–4). They all provided products with the ee exceeded 91% and the chemical yields were moderate. Moreover, for all substrates, there only α-carbonyl group could transfer into hydroxyl and got corresponding chiral α-hydroxy-γ-keto-butyric acid ethyl esters. To our delight, furan and thiophene derivatives (3f or 3g) also gave satisfactory optical selectivity results (Both ees were more than 96%). It's should known that this two kind of compounds were widely used in pharmacy. When more sterically crowded substrate 3h or 3i has been tested, no product was detected (Table 2, entry 7–8). This indicated that steric-hindrance on the β-site influenced coordination between substrate and metal. All the products' configurations were consistent with that of the ligand. They were (R)-configuration. Product 4b and 4e were also characterized by X-ray single crystal diffraction analysis. The structure of 4e was shown in Fig. 3.

Table 2 Different substrates of the ATH catalyzed by RuCl [(R,R)-TsDPEN](p-cymene)

Entry^a	Substrate	Ar	R	Product	Yield (%)	ee^b (%) (conf.)
a 1 mmol of substrate with 0.0025 mmol of Ru(p-cymene)Cl and 0.005 mmol of (R,R)-Ts-DPEN in 4 mL HCOOH/Et3N (5 : 2) and 1 mL DMF at −20 °C stir 4 days for each entry.b The ee and configuration of product were determined by chiral HPLC with Chiralpak OD-H, OJ-H or AD-H column and according to literature or X-ray crystal.
1	3b	4-F-Ph	H	4b	61	91(R)
2	3c	4-Cl-Ph	H	4c	58	91(R)
3	3d	4-Br-Ph	H	4d	60	91(R)
4	3e	4-OMe-Ph	H	4e	58	94.5(R)
5	3f	2-Furyl	H	4f	55	96(R)
6	3g	2-Thienyl	H	4g	71	96(R)
7	3h	Ph	Me	—	—	ND
8	3i	Ph	Ph	—	—	ND

---

Fig. 3 X-ray crystal structure of (R)-4e.

The ATH catalyzed by ferrocene-based chiral catalysts

With ferrocene-based chiral ligands at hand, our initial survey has focused on evaluating the possibility of using these ligands L₁–L₆ for the chiral induction in ATH of 3a. Employing L₂ as ligand, we have tested the solvent and temperature of the reaction. They were shown in Table 3. Lastly, DMF and −20 °C proved to be the optimized condition.

Table 3 The ATH reactions catalyzed by L₂

Entry^a	Solvent	Temp.	Yield (%)	ee^b (%) (conf.)
a 1 mmol of 3a with 0.0025 mmol of Ru(p-cymene)Cl and 0.005 mmol of L2 in 4 mL HCOOH/Et3N (5 : 2) and 1 mL DMF at −20 °C stir 4 days for each entry.b The ee and configuration of product were determined by chiral HPLC with Chiralpak OD-H column and according to literature or X-ray crystal.
1	DMF	r.t.	72	5(R)
2	MeOH	r.t.	Trace	ND
3	THF	r.t.	Trace	ND
4	CH2Cl2	r.t.	Trace	ND
5	CH3CN	r.t.	Trace	ND
6	DMF	−20 °C	60	65(R)

---

Then screen of these chiral ligands was necessary. The results were described in Table 4. All the six chiral ligands induced moderate chemical yield while their optical results were much different. L₁ had P, N, N three special elements and achieved the best optical yield (90% ee) (Table 4, entry 1). Comparing with L₁, L₂'s result was much worse though it had little structure difference with L₁. The ee was only 65% (Table 4, entry 2). In order to check pyridine unit's impact on the reaction, L₃ was prepared and the ee from L₃ was a little lower than that of L₂ (Table 4, entry 3). This implied that the N atom on pyridine of ligand influenced stereoselectivity and chemical yield but not too much. On the other hand, L₄ only had one chiral element and the chiral centre was far from the metal center. Then its racemic result was not surprising (Table 4, entry 4). What’ more, L₅ and L₆'s behaviours were almost the same. Their ees were 40% and 37%, but the configurations were different (Table 4, entry 5 and 6). So we can know that the chiral carbon centre of Ugi's amine almost had not attribution to the stereocontrol in the ATH. The planer chiral elements and the P unit on the ferrocene ring were essential for the high enantioselectivity. At the same time, the Ar groups on P provides bulkiness which caused better enantiocontrol in the reaction (Table 4 entry 1 vs. entry 2). N atom on pyridine also helped to improve this kind of selectivity (Table 4, entry 2 vs. entry 3). It's noteworthy that these six ligands' behavior were not as good as that of chiral RuCl(p-cymene)[Ts-DPEN]. Maybe it was due to the activity of H on N in these six ligands was lower than RuCl(p-cymene)[Ts-DPEN], which resulted in establishing Ru–H–C–O–H–N–Ru transition state ring harder.

Table 4 The ATH reactions catalyzed by ferrocene-based chiral catalysts

Entry^a	Ligand	Yield (%)	ee^b (%) (conf.)
a 1 mmol of 3a with 0.0025 mmol of Ru(p-cymene)Cl and 0.005 mmol of ligand in 4 mL HCOOH/Et3N (5 : 2) and 1 mL DMF at −20 °C stir 4 days for each entry.b The ee and configuration of product were determined by chiral HPLC with Chiralpak OD-H column and according to literature or X-ray crystal.
1	L1	57	90(R)
2	L2	60	65(R)
3	L3	55	50(R)
4	L4	47	Rac
5	L5	50	40(S)
6	L6	52	37(R)

---

Of course, L₁ would applied to more substrates' ATH reactions. As Table 5 showed, at −20 °C and in HCOOH/Et₃N (5 : 2) system, L₁ exhibited moderate to good inducing ability. The ees ranged from 67% to 87% with the yield of 50% or so. Like the results in Table 2, the substituent, F, Cl, Br or OMe on the γ-phenyl of the substrate had no influence on the ATH (Table 5, entry 1–4). Furthermore, furan derivative 3f and thiophene derivative 3g also gave corresponding α-hydroxy-γ-keto-butyric acid ethyl ester with 75% ee and 87% ee respectively (Table 5, entry 5 and 6).

Table 5 Different substrates of the ATH reactions catalyzed by L₁

Entry^a	Substrate	Ar	Product	Yield (%)	ee^b (%) (conf.)
a 1 mmol of substrate with 0.0025 mmol of Ru(p-cymene)Cl and 0.005 mmol of L1 in 4 mL HCOOH/Et3N (5 : 2) and 1 mL DMF at −20 °C stir 4 days for each entry.b The ee and configuration of product were determined by chiral HPLC with Chiralpak OD-H, OJ-H or AD-H column and according to literature or X-ray crystal.
1	3b	4-F-Ph	4b	55	69(R)
2	3c	4-Cl-Ph	4c	58	67(R)
3	3d	4-Br-Ph	4d	53	75(R)
4	3e	4-OMe-Ph	4e	50	69(R)
5	3f	2-Furyl	4f	48	75(R)
6	3g	2-Thienyl	4g	50	87(R)

---

Conclusion

This paper presented a convenient method toward chiral α-hydroxy-γ-keto-butyric acid ethyl esters. The preparation of six chiral ferrocene-based ligands and their utilization in ATH of multiple-carbonyl compounds were discussed. Several optical α-hydroxy-γ-keto-butyric acid ethyl esters were achieved with moderate to excellent ees and moderate chemical yields. Enantiopure RuCl(p-cymene)[Ts-DPEN] catalysts' application in the asymmetric transformation were also developed. Our six ferrocene-based ligands and the reaction for further utilization are both on going.

Experimental

General

All reactions involving air- or moisture-sensitive species were finished under N₂ atmosphere. High-resolution mass spectra were performed on a Bruker smartapex II CCD Mass Spectrometer with ES ionization (ESI). The ¹H and ¹³C NMR spectra were recorded on a Bruker AV-400 spectrometer with TMS as an internal reference. Coupling constant (J) values were given in Hz. Multiplicities are designated by the following abbreviations/s, singlet; d, doublet; t, triplet; q, quartet; br, broad; m, multiplet. Melting points were uncorrected and expressed in °C by MRS-2 melting point apparatus from Shanghai Apparatus Co., Ltd. An Agilent 1200 series apparatus and Chiralpak AD-H, OD-H and OJ-H columns, purchased from Daicel Chemical Industries, were used in Chiral High Performance Liquid Chromatography (HPLC) analyses. All commercially available reagents were used as received. Thin layer chromatography on silica (with GF254) was used to monitor all reactions. Products were purified by flash column chromatography on silica gel purchased from Qingdao Haiyang Chemical Co., Ltd. Optical rotations were measured on a Perkin Elmer 343 polarimeter. The configuration of the products had been assigned by comparison to the literature data or single crystal diffraction.

The synthesis of ferrocene-based chiral ligand

The synthesis of L₁. Corresponding 1a (1.284 g, 2 mmol), which could be prepared from (R)-Ugi's amine,³⁶ and 6-methyl-2-pyridylaldehyde (0.3025 g, 2.5 mmol) were dissolved in 10 mL MeOH. At r.t., the system was stirred for 6 hours under N₂ atmosphere. With NaBH₄ (0.19 g, 5 mmol) added, the reaction was proceeded overnight. After that, 10 mL H₂O was added and subsequently extracted with CH₂Cl₂ (10 mL) for three times. The organic layers were dried over Na₂SO₄. L₁ can be purified by column chromatography (n-hexane/EtOAc/Et₃N = 8/1/0.03, v/v).

Yield 54%; yellow foam; [α]²⁵_D = −184.6° (c = 0.25, CH₂Cl₂); ¹H NMR (400 Hz, CDCl₃) δ 7.42–7.37 (m, 3H), 7.24 (s, 2H), 7.23–7.21 (m, 1H), 7.17–7.10 (m, 1H), 6.81 (d, J = 7.5 Hz, 1H), 5.89 (d, J = 8 Hz, 1H), 4.51 (s, 1H), 4.30–4.23 (m, 2H), 4.06 (s, 5H), 3.75 (s, 1H), 3.54 (d, J = 14.5 Hz, 1H), 3.48 (d, J = 14.5 Hz, 1H), 2.39 (s, 3H), 1.55 (d, J = 6.5 Hz, 3H), 1.29 (s, 18H), 1.13 (s, 18H); ³¹P NMR (202 Hz, CDCl₃) δ −24.98 (s); ¹³C NMR (101 Hz, CDCl₃) δ 159.4 (d, J = 5.9 Hz), 157.1, 150.5 (d, J = 6.6 Hz), 150.0 (d, J = 7.4 Hz), 138.5 (d, J = 7.7 Hz), 136.3, 135.6 (d, J = 7.3 Hz), 129.1 (d, J = 21.4 Hz), 127.4 (d, J = 20.7 Hz), 125.3, 122.7 (d, J = 20.5 Hz), 120.8, 117.8, 96.9, 75.1 (d, J = 14.1 Hz), 71.1 (d, J = 4.2 Hz), 69.6, 69.4 (d, J = 3.8 Hz), 68.6, 51.8, 51.1 (d, J = 10.1 Hz), 34.9, 34.8, 31.5, 31.3, 24.3, 19.2; HRMS (ESI) calcd for C₄₇H₆₃FeN₂P [M + H]⁺ = 743.4157, found/743.4148.

The synthesis L₂. L₂ is prepared from 1a through the same procedure as described above for L₁.

Yield 50%; red foam; [α]²⁵_D = −243.8° (c = 0.65, CH₂Cl₂); ¹H NMR (400 MHz, CDCL₃) δ 7.56–7.50 (m, 2H), 7.40–7.33 (m, 3H), 7.28–7.20 (m, 3H), 7.17–7.10 (m, 3H), 6.84 (d, J = 7.5 Hz, 1H), 6.33 (d, J = 7.5 Hz, 1H), 4.53 (s, 1H), 4.33–4.28 (m, 1H), 4.25–4.17 (m, 1H), 4.00 (s, 5H), 3.85–3.80 (m, 1H), 3.62 (s, 1H), 2.41 (s, 3H), 1.55 (d, J = 6.5 Hz, 3H); ³¹P NMR (162 MHz, CDCL₃) δ −25.03 (s); ¹³C NMR (101 MHz, CDCl₃) δ 159.3, 157.1, 140.1 (d, J = 9.8 Hz), 137.4 (d, J = 9.2 Hz), 136.3, 135.0 (d, J = 21.3 Hz), 132.6 (d, J = 18.6 Hz), 131.5 (d, J = 10 Hz), 129.1, 128.3 (d, J = 5.9 Hz), 128.3, 120.9, 118.3, 97.8 (d, J = 24.3 Hz), 75.1 (d, J = 8.2 Hz), 71.3 (d, J = 4 Hz), 69.6, 69.5 (d, J = 4.4 Hz), 69.1, 52.1, 51.3 (d, J = 9.2 Hz), 24.4, 19.4; HRMS (ESI) calcd for C₃₁H₃₁FeN₂P [M + H]⁺ = 519.1653, found/519.1645.

The synthesis of L₃. Corresponding 1a (0.828 g, 2 mmol) and 6-methyl-2-benzaldehyde (0.3 g, 2.5 mmol) were dissolved in 10 mL MeOH. The system was stirred at r.t. for 6 hours under N₂ atmosphere. With 5 g Pd/C was added, the reaction was carried out under 20 atm of H₂ in autoclave overnight. Then filtered and the filtrate was dried over Na₂SO₄. L₃ can be purified by column chromatography (n-hexane/EtOAc/Et₃N = 8/1/0.03, v/v).

Yield 30%; red foam; [α]²⁵_D = −139° (c = 0.1, CH₂Cl₂); ¹H NMR (400 MHz, CDCl₃) δ 7.89–7.75 (m, 2H), 7.73–7.60 (m, 2H), 7.59–7.47 (m, 3H), 7.42–7.30 (m, 3H), 6.99 (t, J = 7.5 Hz, 1H), 6.92 (d, J = 7.5 Hz, 1H), 6.67 (s, 1H), 6.59 (d, J = 7.1 Hz, 1H), 4.68 (d, J = 27.6 Hz, 1H), 4.39 (d, J = 2.3 Hz, 1H), 4.24 (s, 5H), 3.96 (s, 1H), 3.37 (s, 2H), 3.30 (dd, J = 7.1, 5.9 Hz, 1H), 2.23 (s, 3H), 1.53 (d, J = 6.6 Hz, 3H); ³¹P NMR (162 MHz, CDCl₃) δ −25.36 (s); ¹³C NMR (101 MHz, CDCl₃) δ 169.99 (s), 137.41 (s), 135.55 (s), 134.38 (d, J = 26.6 Hz), 133.19 (s), 131.61 (d, J = 2.7 Hz), 131.44 (d, J = 9.9 Hz), 131.19 (d, J = 9.9 Hz), 128.65 (s), 128.46 (d, J = 12.0 Hz), 128.31–127.84 (m), 127.19 (s), 126.97 (s), 125.04 (s), 98.53–94.93 (m), 73.54 (d, J = 15.2 Hz), 71.39 (s), 71.00 (d, J = 9.9 Hz), 70.30 (s), 70.04 (d, J = 11.5 Hz), 69.70–69.59 (m), 50.77 (s), 23.37 (s), 19.38 (s); HRMS (ESI) calcd for C₃₂H₃₂FeNP [M + H]⁺ = 518.1700, found/518.1739.

The synthesis of L₄. L₄ was prepared from 1b (0.458 g, 2 mmol), which was also obtained from (R)-Ugi's amine,³⁶ and 6-methyl-2-pyridylaldehyde through the same procedure as described above for L₁.

Yield 50%; red foam; [α]²⁵_D = −8° (c = 0.25, CH₂Cl₂); ¹H NMR (400 MHz, CDCl₃) δ 7.53 (t, J = 7.6 Hz, 1H), 7.11 (d, J = 7.6 Hz, 1H), 7.03 (d, J = 7.6 Hz, 1H), 4.26 (s, 1H), 4.20 (s, 6H), 4.13 (s, 2H), 3.92 (m, 2H), 3.57 (q, J = 6.4 Hz, 1H), 2.56 (s, 3H), 1.42 (d, J = 6.4 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 158.94 (s), 158.01 (s), 136.60 (s), 121.42 (s), 119.26 (s), 94.06 (s), 68.55 (s), 68.51 (d, J = 6.8 Hz), 67.29 (s), 66.69 (s), 66.26 (s), 52.95 (s), 52.02 (s), 24.47 (s), 21.65 (d, J = 69.8 Hz); HRMS (ESI) calcd for C₁₉H₂₂FeN₂ [M + H]⁺ = 335.1211, found/335.1224.

The synthesis of L₅. 2 (1.088 g, 4 mmol), which was prepared from (R)-Ugi's amine,³⁶ and (S,S)-DPEN (2.12 g, 10 mmol) were dissolved in the mixture of 30 mL MeOH, 30 mL THF and 3 mL H₂O. The reaction was stirred at 70 °C overnight under N₂ atmosphere. After vacuum distillation, the system was extracted with Et₂O (30 mL) three times and the organic layers were dried over Na₂SO₄. L₅ can be purified by column chromatography (n-hexane/EtOAc/Et₃N = 8/1/0.03, v/v).

Yield 40%; orange foam; [α]²⁵_D = −17.8° (c = 0.5, CH₂Cl₂); ¹H NMR (400 MHz, CDCl₃) δ 7.33–7.15 (m, 10H), 4.24 (m, 1H), 4.14–4.06 (m, 4H), 4.01 (m, 5H), 3.34–3.20 (m, 2H), 1.23 (d, J = 6.3 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 170.18 (s), 143.20 (s), 141.45 (s), 128.20 (s), 127.92 (s), 127.14 (s), 127.11 (s), 127.05 (s), 94.93 (s), 68.27 (s), 67.29 (s), 66.44 (s), 66.06 (s), 61.53 (s), 47.71 (s), 23.29 (s); HRMS (ESI⁺) calcd for C₂₆H₂₈FeN₂ [M + H]⁺ = 425.1680, found/425.1686.

The synthesis of L₆. L₆ was prepared from (R,R)-DPEN through the same procedure as described above toward L₅.

Yield 40%; orange foam; [α]²⁵_D = 34.8° (c = 1, CH₂Cl₂); ¹H NMR (400 MHz, CDCl₃) δ 7.32–7.15 (m, 10H), 4.23 (s, 1H), 4.14–4.06 (m, 4H), 4.02 (s, 5H), 3.30 (m, 2H), 1.22 (d, J = 6.3 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 169.91 (s), 143.41 (s), 141.56 (s, J = 69.1 Hz), 128.13 (d, J = 1.6 Hz), 127.91 (s), 127.11 (s), 127.03 (s), 126.97 (s), 94.99 (s), 68.22 (s), 67.22 (s), 66.42 (s), 66.07 (d, J = 6.5 Hz), 61.60 (s), 47.73 (s), 23.35 (s); HRMS (ESI⁺) calcd for C₂₆H₂₈FeN₂ [M + H]⁺ = 425.1680, found/425.1679.

The general procedure for ATH in HCOOH/Et₃N (5 : 2)

1 mmol of substrate with 0.0025 mmol of Ru(p-cymene)Cl and 0.005 mmol of chiral ligand were dissolved in 1 mL solvent and stirred at r.t. for 4 hours under N₂ atmosphere. Then 4 mL HCOOH/Et₃N (5 : 2) was injected by syringe. The mixture was stirred at −20 °C for 4 days under N₂ atmosphere. Saturated NaHCO₃ (5 mL) and H₂O (5 mL) were added and then extracted with EtOAc (10 mL) for three times and dried over Na₂SO₄. The product was purified by column chromatography (n-hexane/EtOAc = 4/1, v/v).

Product 4a. Yellow oil; ¹H NMR (400 MHz, CDCl₃) δ 7.97 (d, J = 7.3 Hz, 2H), 7.64–7.57 (m, 1H), 7.49 (t, J = 7.7 Hz, 2H), 4.68 (m, 1H), 4.29 (q, J = 7.1 Hz, 2H), 3.52 (m, 2H), 1.30 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 197.57 (s), 173.83 (s), 136.42 (s), 133.63 (s), 128.70 (s), 128.18 (s), 67.20 (s), 61.85 (s), 42.19 (s), 14.11 (s).

Product 4b. Yellow solid; m.p. 71–72 °C; ¹H NMR (400 MHz, CDCl₃) δ 8.01 (m, 2H), 7.21–7.13 (m, 2H), 4.67 (t, J = 5.7, 3.9 Hz, 1H), 4.30 (q, J = 7.1 Hz, 2H), 3.49 (qd, J = 17.4, 4.9 Hz, 2H), 3.33 (d, J = 5.6 Hz, 1H), 1.31 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 195.90 (s), 173.72 (s), 130.94 (s), 130.84 (s), 115.99 (s), 115.77 (s), 67.18 (s), 61.96 (s), 42.06 (s), 14.13 (s).

Product 4c. Yellow oil; ¹H NMR (400 MHz, CDCl₃) δ 7.95–7.88 (m, 2H), 7.51–7.44 (m, 2H), 4.67 (dd, J = 5.5, 1.6 Hz, 1H), 4.30 (q, J = 7.1 Hz, 2H), 3.48 (qd, J = 17.4, 4.9 Hz, 2H), 3.32 (d, J = 5.5 Hz, 1H), 1.31 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 196.26 (s), 173.69 (s), 140.15 (s), 134.80 (s), 129.86 (s), 129.51 (d, J = 18.2 Hz), 129.13 (d, J = 14.8 Hz), 67.13 (s), 61.98 (s), 42.13 (s), 14.13 (s).

Product 4d. Yellow oil; ¹H NMR (400 MHz, CDCl₃) δ 7.81 (m, 2H), 7.61 (m, 2H), 4.70–4.61 (m, 1H), 4.26 (q, J = 7.1 Hz, 2H), 3.46 (dd, J = 17.2, 4.8 Hz, 2H), 1.28 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 196.47 (s), 173.76 (s), 135.20 (s), 132.02 (s), 129.68 (s), 128.85 (s), 67.05 (s), 61.91 (s), 42.16 (s), 14.12 (s).

Product 4e. Yellow solid; m.p. 65–66 °C; ¹H NMR (400 MHz, CDCl₃) δ 7.93 (d, J = 8.8 Hz, 2H), 6.94 (d, J = 8.8 Hz, 2H), 4.65 (dd, J = 5.8, 4.1 Hz, 1H), 4.26 (q, J = 7.1 Hz, 2H), 3.87 (s, 3H), 3.45 (qd, J = 17.3, 5.0 Hz, 2H), 1.28 (t, J = 7.2 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 196.16 (s), 173.83 (s), 163.95 (s), 130.54 (s), 129.55 (s), 113.87 (s), 67.42 (s), 61.81 (s), 55.52 (s), 41.75 (s), 14.14 (s).

Product 4f. Black oil; ¹H NMR (400 MHz, CDCl₃) δ 7.63 (d, J = 1.0 Hz, 1H), 7.26 (s, 1H), 6.58 (dd, J = 3.6, 1.7 Hz, 1H), 4.66 (s 1H), 4.29 (q, J = 7.1 Hz, 2H), 3.45–3.31 (m, 2H), 1.30 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 186.16 (s), 173.65 (s), 152.40 (s), 146.84 (s), 117.85 (s), 112.50 (s), 67.06 (s), 61.98 (s), 42.02 (s), 14.09 (s).

Product 4g. Purple oil; ¹H NMR (400 MHz, CDCl₃) δ 7.77 (d, J = 3.8 Hz, 1H), 7.71 (d, J = 4.9 Hz, 1H), 7.21–7.14 (m, 1H), 4.67 (dd, J = 9.8, 5.6 Hz, 1H), 4.30 (q, J = 7.1 Hz, 2H), 3.53–3.36 (m, 2H), 3.34 (d, J = 5.6 Hz, 1H), 1.31 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 190.14 (s), 173.57 (s), 143.67 (s), 134.47 (s), 132.66 (s), 128.24 (s), 67.34 (s), 61.99 (s), 42.81 (s), 14.10 (s).

Acknowledgements

Financial support from the National Natural Science Foundation of China (NSFC, No. 21102175, 21172262) are gratefully acknowledged. In addition, special thanks are due to Professor Wei Ping Chen.

Notes and references

1. S. F. Martin, K. J. Barr, D. W. Smith and S. K. Bur, J. Am. Chem. Soc., 1999, 121, 6990–6997.
2. P. Wipf and S. R. Spencer, J. Am. Chem. Soc., 2005, 127, 225–235.
3. S. B. Singh, D. L. Zink, G. A. Doss, J. D. Polishook, C. Ruby, E. Register, T. M. Kelly, C. Bonfiglio, J. M. Williamson and R. Kelly, Org. Lett., 2004, 6, 337–340.
4. F. A. Macías, R. M. Varela, A. M. Simonet, H. G. Cutler, S. J. Cutlerb and R. A. Hillc, Tetrahedron Lett., 2003, 44, 941–943.
5. C. Zhu, L. Jing, N. Yu, X. Yanga and Y. Zhao, Acta Pharm. Sin. B, 2013, 3, 109–112.
6. K. D. Yoon, D. G. Jeong, Y. H. Hwang, J. M. Ryu and J. Kim, J. Nat. Prod., 2007, 70, 2029–2032.
7. Y. Feng, L. Wang, S. Niu, L. Li, Y. Si, X. Liu and Y. Che, J. Nat. Prod., 2012, 75, 1339–1345.
8. R. F. Angawi, D. C. Swenson, J. B. Gloer and D. T. Wicklow, J. Nat. Prod., 2003, 66, 1259–1262.
9. W. Cheng, C. Zhu, W. Xu, X. Fan, Y. Yang, Y. Li, X. Chen, W. Wang and J. Shi, J. Nat. Prod., 2009, 72, 2145–2152.
10. B. Chen, Li-Q. Fan, J.-H. Xu, J. Zhao, X. Zhang and Li-M. Ouyang, Appl. Biochem. Biotechnol., 2010, 162, 744–756.
11. C.-Y. Chang and T.-K. Yang, Tetrahedron: Asymmetry, 2003, 14, 2239–2245.
12. C. D. F. Milagre, H. M. S. Milagre, P. J. S. Moran and J. A. Rodrigues, J. Org. Chem., 2010, 75, 1410–1418.
13. M. L. Contente, F. Molinari, P. Zambelli, V. De Vitis, R. Gandolfi, A. Pinto and D. Romano, Tetrahedron Lett., 2014, 55, 7051–7053.
14. S. V. Pansare and A. Bhattacharyya, Tetrahedron, 2003, 59, 3275–3282.
15. K. Zheng, X. Liu, J. Zhao, Y. Yang, L. Lin and X. Feng, Chem. Commun., 2010, 46, 3771–3773.
16. G. Pousse, F. Le Cavelier, L. Humphreys, J. Rouden and J. Blanchet, Org. Lett., 2010, 12, 3582–3585.
17. P. Herold, A. F. Indolese, M. Studer, H. P. Jalett, U. Siegrist and H. U. Blaser, Tetrahedron, 2000, 56, 6497–6499.
18. H.-U. Blaser, S. Burkhardt, H. J. Kirner, T. Mössner and M. S. Solvias, Synthesis, 2003, 11, 1679–1682.
19. K. Vandyck, B. Matthys and J. Van der Eycken, Tetrahedron Lett., 2005, 46, 75–78.
20. D. Wang and D. Astruc, Chem. Rev., 2015, 115, 6621–6686.
21. F. Foubelo, C. Najera and M. Yus, Tetrahedron: Asymmetry, 2015, 26, 769–790.
22. S. Rast, B. Modec, M. Stephan and B. Mohar, Org. Biomol. Chem., 2016, 14, 2112–2120.
23. T. Cheng, Q. Ye, Q. Zhao and G. Liu, Org. Lett., 2015, 17, 4972–4975.
24. D. Zhang, T. Cheng, Q. Zhao, J. Xu and G. Liu, Org. Lett., 2014, 16, 5764–5767.
25. M. Wu, T. Cheng, M. Ji and G. Liu, J. Org. Chem., 2015, 80, 3708–3713.
26. J. W. Yang and B. List, Org. Lett., 2006, 8, 5653–5655.
27. D. R. Clay and M. C. McIntosh, Tetrahedron Lett., 2012, 53, 1691–1694.
28. J. Cossy, F. Eustache and P. I. Dalko, Tetrahedron Lett., 2001, 42, 5005–5007.
29. R. Hess, S. Diezi, T. Mallat and A. Baiker, Tetrahedron: Asymmetry, 2004, 15, 251–257.
30. Z. Fang and M. Wills, J. Org. Chem., 2013, 78, 8594–8605.
31. M. J. Palmer and M. Wills, Tetrahedron: Asymmetry, 1999, 10, 2045–2061.
32. A. Zirakzadeh, R. Schuecker, N. Gorgas, K. Mereiter, F. Spindler and W. Weissensteiner, Organometallics, 2012, 31, 4241–4250.
33. X. Zhang, P. Ma, D. Zhang, Y. Lei, S. Zhang, R. Jiang and W. Chen, Org. Biomol. Chem., 2014, 12, 2423–2426.
34. J. Ma, C. Li, D. Zhang, Y. Lei, M. Li, R. Jiang and W. Chen, RSC Adv., 2015, 5, 35888–35892.
35. W. Yao, M. Chen, X. Liu, R. Jiang, S. Zhang and W. Chen, Catal. Sci. Technol., 2014, 4, 1726–1729.
36. H. Nie, G. Zhou, Q. Wang, W. Chen and S. Zhang, Tetrahedron: Asymmetry, 2013, 24, 1567–1571.
37. J.-H. Xie, X.-Y. Liu, J.-B. Xie, Li-X. Wang and Qi-L. Zhou, Angew. Chem., Int. Ed., 2011, 50, 7329–7332.
38. W. Tian, G. Han, J. Zhu, J. Qi, Q. Chen, J. Zhao, C. Zheng, L. Zhang, Y. Zhou and J. Lv, Bioorg. Med. Chem. Lett., 2013, 23, 4177–4184.
39. A. K. Roy and S. Batra, Synthesis, 2003, 2325–2330.

---

Footnote

† Electronic supplementary information (ESI) available: ¹H, ¹³C NMR of products 4a–g, ¹H, ³¹P, ¹³C NMR and HRMS of compounds L₁–L₆, HPLC charts of products 4a–g. X-ray crystal of 4b, 4g. See DOI: 10.1039/c6ra02500e

---