Mechanochemical Asymmetric Transfer Hydrogenation of Diketones to Access Chiral 1,3-Diols under Solvent-Free Conditions

Abstract

A mechanochemical asymmetric transfer hydrogenation (ATH) of diketones in the presence of a ruthenium complex under solvent-free conditions was developed to provide chiral 1,3-diol derivatives. This protocol benefits from rapid reaction kinetics, no use of solvents, and excellent enantioselectivity. In addition, the mechanochemical ATH reaction can easily be performed on a gram scale.

The enantioselective construction of 1,3-diols has attracted considerable interest in synthetic chemistry because 1,3-diols units are widely found in several natural products and biologically active drugs, for instance, rosuvastatin,[1] pravastatin, pitavastatin, and atorvastatin (Scheme [1]),[2] which are all competitive inhibitors of 3-hydroxy-3-methylglutaryl-CoA reductase.[3] However, it is a continuing challenge for chemists to mimic Nature’s behavior in efficiently synthesizing flexible chiral diols bearing two stereogenic centers.[4] A landmark in this area was reported in 1995 by Noyori and Ikariyai and their co-workers,[5] who synthesized a chiral [RuCl(η⁶-arene)(N-arylsulfonyl-DPEN)] (DPEN = 1,2-diphenylethane-1,2-diamine) catalyst for the hydrogenation of ketones. Following this contribution, a series of Ir, Ru, and Rh-based TsDPEN complexes have been developed and applied in syntheses of interesting chiral alcohols.[6] [7]

Despite several outstanding benefits, transition-metal-catalyzed asymmetric transfer hydrogenation (ATH) reactions suffer from the use of environmentally damaging organic solvents, for instance, 1,2-dichloroethane (DCE),[6] or high-boiling-point solvents, such as isopropanol, which pose a huge challenge to green and sustainable chemistry.[8]

Recently, the emergence of mechanochemistry offers an enticing alternative for solvent-free reactions.[9] The use of mechanical force (grinding or milling) to induce collision and activation of reactants can provide powerful advantages, mainly reflected in rapid reaction kinetics, lower reaction temperatures, avoidance of the need for a solvent, and even reaction under aerobic conditions. For example, in 2010, Stolle and co-workers established a Sonogashira coupling reaction in the absence of copper or an additional ligands under ball-milling conditions.[10] The Kim group recently reported a direct C–H amidation of aromatic amides through the mechanochemical activation of acyl azides.[11] The Pilarski group developed a highly regioselective mechanochemical C–H methylation of (het)arenes in the presence of a rhodium complex under solvent-free conditions.[12] This protocol requires shorter reaction times than with previous solution-based reactions, and no external heating is necessary. The Geneste group realized a hydroxylation of aryl or hetaryl fluorides by using KOH as base with the assistance of ball milling.[13] Furthermore, mechanochemistry permits the formation of C–N bonds at room temperature.[14] Although great progress has been achieved in recent decades, the application of mechanochemistry in ATH reactions is unusual.

As part of our ongoing pursuits in green and sustainable chemistry,[15] we aimed to develop an efficient method for the chiral reduction of ketones in the absence of a reaction solvent, with a short reaction time. In this report, we describe a mechanochemical ATH reaction of diketones to access chiral 1,3-diols. The benefit of this protocol includes the fast reaction kinetics and no use of toxic organic solvents.

At the beginning of our study, we concentrated on screening the reaction conditions, as shown in Table [1]. These reactions were performed by using a planetary ball mill (25 mL internal volume vessel) equipped with stainless steel balls 5 mm in diameter. Examination of several widely used diamine-based metal complexes indicated that (η⁶-mesitylene)RuCl(TsDPEN) [C; TsDPEN = (S,S)-N-(p-toluenesulfonyl)-1,2-diphenylethane-1,2-diamine] exhibited a superior catalytic performance (82% yield, 97% ee, and 95:5 dr) to other iridium or rhodium complexes (Table [1], entries 1–6). We next optimized the hydrogen source under ball-milling conditions. HCOOH proved to be ineffective, as 2a was not obtained, suggesting a low pH is detrimental to the catalytic ATH reaction of 1a (entry 7). Based on this hypothesis, more-basic hydrogen sources (HCOONa and HCOONH₄) were examined. However, the observed yields and enantiomeric excesses were not as good as those of the reaction using HCOOH/NEt₃ (entries 8 and 9 vs entry 3). Further optimization showed the best ratio of HCOOH to NEt₃ was 1:1 when C was used as the catalyst (entries 10 and 11). Furthermore, optimization of the milling balls indicated that the reaction milled with zirconia milling balls gave a superior yield (96%) and ee (99%) (entries 12 and 13 vs entry 14). To confirm the superiority of ball-milling conditions, a solution-based catalytic reaction of 1a was performed using 1:1 HCOOH/NEt₃; however, only 42% of 2a was isolated after three hours at 35 °C (entry 15), indirectly demonstrating the excellent catalytic capability of the ball-milling conditions.

Table 1 Optimization of the Asymmetric Transfer Hydrogenation of 1a ^a

Entry	Catalyst	H source	Yieldb (%)	eec (%)	drc
1	A	HCOOH/NEt3 (5:2)	47	90	88:12
2	B	HCOOH/NEt3 (5:2)	58	71	87:13
3	C	HCOOH/NEt3 (5:2)	82	97	95:5
4	D	HCOOH/NEt3 (5:2)	65	67	97:3
5	E	HCOOH/NEt3 (5:2)	90	82	98:2
6	F	HCOOH/NEt3 (5:2)	72	93	95:5
7	C	HCOOH	–	–	–
8	C	HCOONa	12	96	95:5
9	C	HCOONH4	25	97	97:3
10	C	HCOOH/NEt3 (1:1)	89	98	97:3
11	C	HCOOH/NEt3 (2:5)	78	86	97:3
12	C	HCOOH/NEt3 (1:1)d	72	72	96:4
13	C	HCOOH/NEt3 (1:1)e	69	90	96:4
14	C	HCOOH/NEt3 (1:1)f	96	99	98:2
15	C	HCOOH/NEt3 (1:1)g	42	97	97:3

^a Reaction conditions: 1a (0.20 mmol), catalyst (0.01 mmol), H source (0.56 mmol), 35 °C, number of balls: 8.

^b Isolated yield.

^c The ee and dr were determined by HPLC.

^d With 3 mm diameter stainless steel milling balls.

^e With 7 mm diameter stainless steel milling balls.

^f With 5 mm diameter zirconia milling balls.

^g No ball milling.

With the best ball-milling conditions in hand [5 mol% C, HCOOH/NEt₃ (1:1), 5 mm zirconia milling balls], we investigated the scope of the present protocol (Scheme [2]). As expected, all the tested reactions completed within 45 minutes, which is much shorter than the comparable solution-based reactions, which generally require 8–12 hours. In the case of linear diketones, electron-withdrawing groups and electron-donating groups were well tolerated, as 2b–l were all isolated in better than 92% yield and 96% ee. Interestingly, besides acyclic diketones, the fused diol 2n was also obtained in 93% yield with 98% ee and 98:2 dr, demonstrating the broad applicability of the mechanochemical-force-promoted ATH reaction. We then checked the behaviors of heterocyclic diketones. Under ball-milling conditions, the starting di-2-thienyl ketone 1o was quickly transformed into diol 2o in 94% yield with 98% ee and 99:1 dr. In addition, the present ball-milling protocol is also compatible with trifluoromethylated diketones, as 2p could be prepared in 92% yield with 99% ee and 94:6 dr. However, we failed to convert aliphatic cyclohexane-1,3-dione into cyclohexane-1,3-diol under the optimal ball-milling conditions, as only a small amount of cyclohexane-1,3-diol was obtained, along with a complex mixture. In the case of the bulky diketone 1q, diol 2q was not obtained and the starting diketone was fully recovered.

To confirm the absolute configuration of the 1,3-diols, 2a was crystallized from dichloromethane (DCM), and its structure was determined by X-ray diffraction (Figure [1]).[16]

A prominent feature of mechanochemistry is large-scale synthesis. We therefore performed the catalytic reaction of 1a on a two-gram scale under ball-milling conditions (Scheme [3]). Before addition of catalyst C, the H source and 1a were initially mixed in the planetary ball mill for three minutes to improve the efficiency of the catalyst. The desired product 2a was obtained in 96% yield with 98% ee and 98:2 dr. This experiment confirmed that the mechanochemical ATH reaction of diketones is not limited by the quantity of the starting material, implying potential applications in practical synthesis and industrial fields.

Besides ketones, the mechanochemically forced ATH reactions of imines are also of interest. As reported previously,[17] transfer hydrogenation of imines requires an acidic reaction medium. Therefore, HCOOH was initially selected as a hydrogen source for the ATH reaction of quinaldine (3). However, no reaction occurred, and 3 was fully recovered. Therefore, the less acidic hydrogen source HCOOH/Et₃N (1.1:1) was next used (Scheme [4]). The milling-induced reaction then gave an 82% yield of (2S)-2-methyl-1,2,3,4-tetrahydroquinoline (4) within one hour, which is a far shorter time than that for the solution-based reaction (12 h). The low ee (47%) was attributed to the rigid plane of quinoline.

To demonstrate the superiority of mechanochemistry for the ATH reaction of ketones, time-course experiments were performed under ball-milling conditions and under solution conditions with 2 mL of DCM as the solvent (Figure [2]). Dione 1a was converted into diol 2a within 45 minutes under ball-milling conditions whereas the solution-based reaction required 12 hours.[18] The reason for the rapid reaction kinetics under ball-milling conditions can be attributed to the ultrahigh concentration of 1a that is present under solvent-free conditions.[19]

In summary, we have developed a practical method for the synthesis of diols derivatives by using (η⁶-mesitylene)RuCl(TsDPEN) as a catalyst and HCOOH/NEt₃ (1:1) as a hydrogen source under ball-milling conditions.[20] Both diketones and imines are well tolerated in this protocol. Studies on the synthetic applications of mechanochemical asymmetric transfer hydrogenation are underway in our laboratory.

Conflict of Interest

The authors declare no conflict of interest.

• References and Notes

• 1a Vempala N, Venkateswara Rao S, Shree AJ, Pradhan BS. Synthesis 2016; 48: 4167
  Thieme ConnectSearch in Google Scholar
• 1b Gamba-Sánchez D, Prunet J. Synthesis 2018; 50: 3997
  Thieme ConnectSearch in Google Scholar
• 2 Thakker D, Nair S, Pagada A, Jamdade V, Malik A. Pharmacoepidemiol. Drug Saf. 2016; 25: 1131
  CrossrefPubMedSearch in Google Scholar
• 3 Chakravarti R, Sahai V. Appl. Microbiol. Biotechnol. 2004; 64: 618
  CrossrefPubMedSearch in Google Scholar
• 4 Hetzler BE, Volpin G, Vignoni E, Petrovic AG, Proni G, Hu CT, Trauner D. Angew. Chem. Int. Ed. 2018; 57: 14276
  CrossrefPubMedSearch in Google Scholar
• 5a Ohkuma T, Ooka H, Hashiguchi S, Ikariya T, Noyori R. J. Am. Chem. Soc. 1995; 117: 2675
  CrossrefSearch in Google Scholar
• 5b Hashiguchi S, Fujii A, Takehara J, Ikariya T, Noyori R. J. Am. Chem. Soc. 1995; 117: 7562
  CrossrefSearch in Google Scholar
• 6 Wang D, Astruc D. Chem. Rev. 2015; 115: 6621
  CrossrefPubMedSearch in Google Scholar
• 7 Dub PA, Gordon JC. Dalton Trans. 2016; 45: 6756
  CrossrefPubMedSearch in Google Scholar
• 8a Pan H.-J, Zhang Y, Shan C, Yu Z, Lan Y, Zhao Y. Angew. Chem. Int. Ed. 2016; 55: 9615
  CrossrefPubMedSearch in Google Scholar
• 8b Caleffi GS, Demidoff FC, Nájera C, Costa PR. R. Org. Chem. Front. 2022; 9: 1165
  CrossrefSearch in Google Scholar
• 9a Stolle A, Szuppa T, Leonhardt SE. S, Ondruschka B. Chem. Soc. Rev. 2011; 40: 2317
  CrossrefPubMedSearch in Google Scholar
• 9b Egorov IN, Santra S, Kopchuk DS, Kovalev IS, Zyryanov GV, Majee A, Ranu BC, Rusinov VL, Chupakhin ON. Green Chem. 2020; 22: 302
  CrossrefSearch in Google Scholar
• 9c Friščić T, Mottillo C, Titi HM. Angew. Chem. Int. Ed. 2020; 59: 1018
  CrossrefPubMedSearch in Google Scholar
• 9d Pérez-Venegas M, Juaristi E. ACS Sustainable Chem. Eng. 2020; 8: 8881
  CrossrefSearch in Google Scholar
• 9e Perona A, Hoyos P, Farrán Á, Hernáiz MJ. Green Chem. 2020; 22: 5559
  CrossrefSearch in Google Scholar
• 9f Ying P, Yu J, Su W. Adv. Synth. Catal. 2021; 363: 1246
  CrossrefSearch in Google Scholar
• 10 Thorwirth R, Stolle A, Ondruschka B. Green Chem. 2010; 12: 985
  CrossrefSearch in Google Scholar
• 11 Yoo K, Hong EJ, Huynh TQ, Kim B.-S, Kim JG. ACS Sustainable Chem. Eng. 2021; 9: 8679
  CrossrefSearch in Google Scholar
• 12 Ni S, Hribersek M, Baddigam SK, Ingner FJ. L, Orthaber A, Gates PJ, Pilarski LT. Angew. Chem. Int. Ed. 2021; 60: 6660
  CrossrefPubMedSearch in Google Scholar
• 13 Rodrigo E, Wiechert R, Walter MW, Braje W, Geneste H. Green Chem. 2022; 24: 1469
  CrossrefSearch in Google Scholar
• 14 Kubota K, Takahashi R, Uesugi M, Ito H. ACS Sustainable Chem. Eng. 2020; 8: 16577
  CrossrefSearch in Google Scholar
• 15a Wu L, Jin R, Li L, Hu X, Cheng T, Liu G. Org. Lett. 2017; 19: 3047
  CrossrefPubMedSearch in Google Scholar
• 15b Jin R, Zheng D, Liu R, Liu G. ChemCatChem 2018; 10: 1739
  CrossrefSearch in Google Scholar
• 15c Yang D, Wang C, Wang Y, Liu G, Cheng T, Liu R. Org. Chem. Front. 2022; 9: 102
  CrossrefSearch in Google Scholar
• 16 CCDC 2156357 contains the supplementary crystallographic data for compound 2a. The data can be obtained free of charge from The Cambridge Crystallographic Data Centre via. www.ccdc.cam.ac.uk/structures
• 17 Liu R, Cheng T, Kong L, Chen C, Liu G, Li H. J. Catal. 2013; 307: 55
  CrossrefSearch in Google Scholar
• 18 Solvent-Based Reaction: Experimental Procedure A 25 mL Schlenk tube was charged with ketone 1a (224.3 mg, 1 mmol), catalyst C (31.8 mg, 0.1 mmol), 1:1 HCOOH/NEt₃ (0.28 mmol), and DCE (10 mL), and the mixture was stirred at 35 °C. During the first hour of the reaction, 0.5 mL aliquots of the solution were sampled every 15 min and then 0.5 mL aliquots were sampled every hour. Each sample of the solution was evaporated and the residue was dissolved in CDCl₃ (1 mL) containing 0.05 mmol of 1,3,5-trimethoxybenzene. [NOTE: To eliminate experimental error, the deuterated solvent should be prepared on a large scale by dissolving 1,3,5-trimethoxybenzene (5 mmol) in CDCl₃ (100 mL).] Finally, the yield of 2a was determined by ¹H NMR spectral analysis of the crude sample solution.
• 19 Takahashi R, Seo T, Kubota K, Ito H. ACS Catal. 2021; 11: 14803
  CrossrefSearch in Google Scholar
• 20 Diols 2a–p: General Procedure A 25 mL stainless steel milling vessel was charged the appropriate ketone 1 (0.2 mmol), catalyst C (0.01 mmol), 1:1 HCOOH/NEt₃ (0.056 mmol), and eight 5 mm diameter zirconia milling balls. Then the milling vessel was then placed on a planetary ball mill (900 rpm) and the mixture was milled for 45 min at approximately 35 °C. Upon completion of the reaction, the organic compounds were taken up in Et₂O. The solution was concentrated and the residue was purified by column chromatography (silica gel). (1S,3S)-1-(4-Methoxyphenyl)-3-phenylpropane-1,3-diol (2k) Purple oil; yield: 95% (99% ee, 99:1 dr) [α]_D ²⁵ = -50.2 (c 1.0, CHCl₃). HPLC [IC; hexane–i-PrOH (93:7), 1.1 mL/min, 25 °C, λ = 215 nm]: t ₁ = 18.2 min (major), t ₂ = 28.2 min, t ₃ = 29.4 min, t ₄ = 37.3 min (minor). ¹H NMR (400 MHz, DMSO-d ₆): δ = 7.32 (d, J = 2.4 Hz, 4 H), 7.28–7.18 (m, 3 H), 6.88 (d, J = 8.7 Hz, 2 H), 5.26 (d, J = 4.9 Hz, 1 H), 5.18 (d, J = 4.9 Hz, 1 H), 4.75 (m, 2 H), 3.73 (s, 3 H), 1.79 (dt, J = 8.2, 3.8 Hz, 2 H). ¹³C NMR (100 MHz, DMSO-d ₆): δ = 158.5 (C), 147.2 (C), 139.1 (C), 128.5 (CH), 127.3 (CH), 127.0 (CH), 126.1 (CH), 113.9 (CH), 69.6 (CH), 69.1 (CH), 55.5 (CH₃), 50.2 (CH₂).

Corresponding Authors

Publication History

Received: 11 March 2022

Accepted after revision: 21 July 2022

Accepted Manuscript online:
21 July 2022

Article published online:
19 August 2022

© 2022. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

• References and Notes

• 1a Vempala N, Venkateswara Rao S, Shree AJ, Pradhan BS. Synthesis 2016; 48: 4167
  Thieme ConnectSearch in Google Scholar
• 1b Gamba-Sánchez D, Prunet J. Synthesis 2018; 50: 3997
  Thieme ConnectSearch in Google Scholar
• 2 Thakker D, Nair S, Pagada A, Jamdade V, Malik A. Pharmacoepidemiol. Drug Saf. 2016; 25: 1131
  CrossrefPubMedSearch in Google Scholar
• 3 Chakravarti R, Sahai V. Appl. Microbiol. Biotechnol. 2004; 64: 618
  CrossrefPubMedSearch in Google Scholar
• 4 Hetzler BE, Volpin G, Vignoni E, Petrovic AG, Proni G, Hu CT, Trauner D. Angew. Chem. Int. Ed. 2018; 57: 14276
  CrossrefPubMedSearch in Google Scholar
• 5a Ohkuma T, Ooka H, Hashiguchi S, Ikariya T, Noyori R. J. Am. Chem. Soc. 1995; 117: 2675
  CrossrefSearch in Google Scholar
• 5b Hashiguchi S, Fujii A, Takehara J, Ikariya T, Noyori R. J. Am. Chem. Soc. 1995; 117: 7562
  CrossrefSearch in Google Scholar
• 6 Wang D, Astruc D. Chem. Rev. 2015; 115: 6621
  CrossrefPubMedSearch in Google Scholar
• 7 Dub PA, Gordon JC. Dalton Trans. 2016; 45: 6756
  CrossrefPubMedSearch in Google Scholar
• 8a Pan H.-J, Zhang Y, Shan C, Yu Z, Lan Y, Zhao Y. Angew. Chem. Int. Ed. 2016; 55: 9615
  CrossrefPubMedSearch in Google Scholar
• 8b Caleffi GS, Demidoff FC, Nájera C, Costa PR. R. Org. Chem. Front. 2022; 9: 1165
  CrossrefSearch in Google Scholar
• 9a Stolle A, Szuppa T, Leonhardt SE. S, Ondruschka B. Chem. Soc. Rev. 2011; 40: 2317
  CrossrefPubMedSearch in Google Scholar
• 9b Egorov IN, Santra S, Kopchuk DS, Kovalev IS, Zyryanov GV, Majee A, Ranu BC, Rusinov VL, Chupakhin ON. Green Chem. 2020; 22: 302
  CrossrefSearch in Google Scholar
• 9c Friščić T, Mottillo C, Titi HM. Angew. Chem. Int. Ed. 2020; 59: 1018
  CrossrefPubMedSearch in Google Scholar
• 9d Pérez-Venegas M, Juaristi E. ACS Sustainable Chem. Eng. 2020; 8: 8881
  CrossrefSearch in Google Scholar
• 9e Perona A, Hoyos P, Farrán Á, Hernáiz MJ. Green Chem. 2020; 22: 5559
  CrossrefSearch in Google Scholar
• 9f Ying P, Yu J, Su W. Adv. Synth. Catal. 2021; 363: 1246
  CrossrefSearch in Google Scholar
• 10 Thorwirth R, Stolle A, Ondruschka B. Green Chem. 2010; 12: 985
  CrossrefSearch in Google Scholar
• 11 Yoo K, Hong EJ, Huynh TQ, Kim B.-S, Kim JG. ACS Sustainable Chem. Eng. 2021; 9: 8679
  CrossrefSearch in Google Scholar
• 12 Ni S, Hribersek M, Baddigam SK, Ingner FJ. L, Orthaber A, Gates PJ, Pilarski LT. Angew. Chem. Int. Ed. 2021; 60: 6660
  CrossrefPubMedSearch in Google Scholar
• 13 Rodrigo E, Wiechert R, Walter MW, Braje W, Geneste H. Green Chem. 2022; 24: 1469
  CrossrefSearch in Google Scholar
• 14 Kubota K, Takahashi R, Uesugi M, Ito H. ACS Sustainable Chem. Eng. 2020; 8: 16577
  CrossrefSearch in Google Scholar
• 15a Wu L, Jin R, Li L, Hu X, Cheng T, Liu G. Org. Lett. 2017; 19: 3047
  CrossrefPubMedSearch in Google Scholar
• 15b Jin R, Zheng D, Liu R, Liu G. ChemCatChem 2018; 10: 1739
  CrossrefSearch in Google Scholar
• 15c Yang D, Wang C, Wang Y, Liu G, Cheng T, Liu R. Org. Chem. Front. 2022; 9: 102
  CrossrefSearch in Google Scholar
• 16 CCDC 2156357 contains the supplementary crystallographic data for compound 2a. The data can be obtained free of charge from The Cambridge Crystallographic Data Centre via. www.ccdc.cam.ac.uk/structures
• 17 Liu R, Cheng T, Kong L, Chen C, Liu G, Li H. J. Catal. 2013; 307: 55
  CrossrefSearch in Google Scholar
• 18 Solvent-Based Reaction: Experimental Procedure A 25 mL Schlenk tube was charged with ketone 1a (224.3 mg, 1 mmol), catalyst C (31.8 mg, 0.1 mmol), 1:1 HCOOH/NEt₃ (0.28 mmol), and DCE (10 mL), and the mixture was stirred at 35 °C. During the first hour of the reaction, 0.5 mL aliquots of the solution were sampled every 15 min and then 0.5 mL aliquots were sampled every hour. Each sample of the solution was evaporated and the residue was dissolved in CDCl₃ (1 mL) containing 0.05 mmol of 1,3,5-trimethoxybenzene. [NOTE: To eliminate experimental error, the deuterated solvent should be prepared on a large scale by dissolving 1,3,5-trimethoxybenzene (5 mmol) in CDCl₃ (100 mL).] Finally, the yield of 2a was determined by ¹H NMR spectral analysis of the crude sample solution.
• 19 Takahashi R, Seo T, Kubota K, Ito H. ACS Catal. 2021; 11: 14803
  CrossrefSearch in Google Scholar
• 20 Diols 2a–p: General Procedure A 25 mL stainless steel milling vessel was charged the appropriate ketone 1 (0.2 mmol), catalyst C (0.01 mmol), 1:1 HCOOH/NEt₃ (0.056 mmol), and eight 5 mm diameter zirconia milling balls. Then the milling vessel was then placed on a planetary ball mill (900 rpm) and the mixture was milled for 45 min at approximately 35 °C. Upon completion of the reaction, the organic compounds were taken up in Et₂O. The solution was concentrated and the residue was purified by column chromatography (silica gel). (1S,3S)-1-(4-Methoxyphenyl)-3-phenylpropane-1,3-diol (2k) Purple oil; yield: 95% (99% ee, 99:1 dr) [α]_D ²⁵ = -50.2 (c 1.0, CHCl₃). HPLC [IC; hexane–i-PrOH (93:7), 1.1 mL/min, 25 °C, λ = 215 nm]: t ₁ = 18.2 min (major), t ₂ = 28.2 min, t ₃ = 29.4 min, t ₄ = 37.3 min (minor). ¹H NMR (400 MHz, DMSO-d ₆): δ = 7.32 (d, J = 2.4 Hz, 4 H), 7.28–7.18 (m, 3 H), 6.88 (d, J = 8.7 Hz, 2 H), 5.26 (d, J = 4.9 Hz, 1 H), 5.18 (d, J = 4.9 Hz, 1 H), 4.75 (m, 2 H), 3.73 (s, 3 H), 1.79 (dt, J = 8.2, 3.8 Hz, 2 H). ¹³C NMR (100 MHz, DMSO-d ₆): δ = 158.5 (C), 147.2 (C), 139.1 (C), 128.5 (CH), 127.3 (CH), 127.0 (CH), 126.1 (CH), 113.9 (CH), 69.6 (CH), 69.1 (CH), 55.5 (CH₃), 50.2 (CH₂).

• Supplementary Material