A Simple Synthesis of the Novel Antihistaminic Drug Olopatadine Hydrochloride­

Abstract

A new alternative route for the synthesis of olopatadine is described. The present strategy involves a Lewis acid mediated ring opening of a cyclic ether to introduce 3-(dimethylamino)propylidene group as the side chain.

Olopatadine hydrochloride is a top selling antihistaminic drug ranked in the ‘Top 200 Brand Name Drugs by total US prescriptions in 2010’.[1] It is a selective histamine H₁ receptor (H₁R) antagonist. Olopatadine (1) is used for the treatment of ocular symptoms of seasonal allergic conjunctivitis. The compound may be administered in a solid oral dosage form such as ‘allelock^®’ tablets, as ophthalmic solution form ‘pataday^®’, ‘patanol^®’ and nasal spray ‘patanase^®’. Olopatadine was developed by Kyowa Hakko Kirin Co. Ltd. and is produced commercially using Wittig reaction as the key step.[2] Structure–activity relationship studies have revealed that the key structural features required for the enhanced antiallergic activities are: a 3-(dimethylamino)propylidene group as the side chain at C-11, a terminal carboxyl moiety at C-2, and a dibenzoxepin ring system. Although both Z- and E-isomers of 1 show similar H₁R affinities,[3] only the Z-isomer is marketed as a drug. Because of the potent antiallergic activity of 1, there is a huge interest for the stereoselective generation of the Z-isomer of olopatadine (Figure [1]).

Literature review reveals a number of synthetic strategies to access olopatadine, mostly in the form of patents.[4] In most of the syntheses, the key side chain has been introduced via ‘Grignard reaction’ or ‘Wittig olefination’.[2a] [4d] In most of the previous approaches, the main drawback has been the low E/Z stereoselectivity. Recently, Bosch et al. have utilized a stereoselective Heck reaction for the selective generation of Z-isomer via the intramolecular cyclization of an E-alkene intermediate.[5] Nishimura et al. also reported a stereospecific route to 1 under palladium catalysis. In their synthetic route, the Z-stereoselectivity was controlled by an intramolecular stereospecific seven-membered ring cyclization from an alkyne intermediate using palladium catalyst.[6] Unfortunately most of the reported syntheses involve expensive reagents and harsh reaction conditions, which are operationally difficult to perform on a large scale. Herein we describe a new and simple synthetic route for olopatadine.

The envisaged retrosynthetic strategy for (1) is delineated in Scheme [1]. A linear synthetic strategy was invoked wherein homoallyl alcohol 2 was conceived as the ideal precursor to 1. Homoallyl alcohol 2 upon mesylation, dimethylamination, and hydrochlorination would lead to olopatadine hydrochloride salt. Homoallyl alcohol 2 in turn could be prepared from the cyclic spiroether 3 by Lewis acid mediated ether ring opening. Cyclic spiroether 3 in turn could be obtained from known intermediate 2-(11-oxo-6,11-dihydrodibenzo[b,e]oxepin-2-yl)acetic acid (5, also known as isoxepac) through routine functional group manipulations such as Barbier reaction to introduce the allyl group, followed by hydroboration and ether formation.

According to the proposed plan, synthesis of olopatadine (1) started from isoxepac (5),[7] which on treatment with thionyl chloride in methanol gave the corresponding isoxepac methyl ester in quantitative yield. Barbier reaction on the methyl ester with allyl bromide in the presence of zinc powder in DMF as the solvent furnished the allylic alcohol 4 in 91% yield[8] (Scheme [2]).

Hydroboration of 4 with 9-BBN followed by quenching with sodium hydroxide and hydrogen peroxide afforded the diol 6.[9] In order to introduce the exocyclic double bond by acid-mediated dehydration of diol 6, several conditions were screened as shown in Table [1]. Almost all the reaction conditions led to the predominant formation of the undesired E-olefin along with the generation of the spiroether 3 (Scheme [2]).

The diol 6 was treated with both protic as well as Lewis acids. It was found that inorganic protic acids led to the formation of cyclic ether 3 as the major or exclusive product (Table [1], entries 1 and 2), while organic acids furnished the olefin as the major product (entries 3 and 4). It was observed that treatment of diol 6 with catalytic p-toluenesulfonic acid (PTSA) at ambient temperature instantly formed a spirotetrahydrofuran ring 3 in almost quantitative yield (entry 5) whereas the treatment of 6 with Lewis acid (Et₂O·BF₃) at room temperature led to the formation of cyclic ether 3 along with the formation of olefin 2 as the minor product (entry 6).

However, when the reaction was carried out at elevated temperature the olefin 2 was the only product formed (Table [1], entry 7). In both the cases the E/Z ratio was 3:2, in favor of the unwanted isomer. From this it was clear that diol 6 was transformed to spiroether 3 under the influence of Lewis acid, and then the spiroether 3 opened to give the olefin 2. Since the olefin 2 obtained from diol 6 gave at best a ratio of E/Z = 3:2, it was decided to study the behavior of spiroether 3 under the influence of a variety of Lewis­ acids.

In order to improve the Z-selectivity, we screened different Lewis acids and their combination with base as summarized in Table [2]. Amongst various Lewis acids screened, we observed that AlCl₃ gave the best result. Thus, spiroether 3 on treatment with 2.5 equivalents of AlCl₃ resulted in the formation of homoallyl alcohol 2 in 95% yield with an E/Z ratio of 2:3 (Scheme [3]). Attempts to separate the isomers by AgNO₃ column chromatography failed. As compound 2 was solid, we attempted to separate the isomers by preferential recrystallization. The E/Z ratio could be improved to 1:9 after two recrystallizations.[10]

Although the exact mechanism of this reaction remains to be elucidated, the plausible mechanism for the stereoselectivity towards Z-isomer (major) could be explained by considering the metal coordination. The Lewis acid first chelates with the 1′-oxygen of furan ring and the metal coordinates with oxygen of C-13 (Figure [2]). Opening of the furan ring results in the formation of Z-isomer as the major isomer.

To complete the synthesis, the isomeric mixture (E/Z = 1:9) of homoallyl alcohol 2 was subjected to mesylation and dimethylamination resulting in compound 7 in 84% yield (Scheme [3]). Ohshima et al.[2a] have reported the saponification of 7 and E/Z-isomer separation of the corresponding acid. The spectroscopic data of 7 was in good agreement with the reported values.[2] [5] [6]

In summary, we have developed a novel and concise synthetic route to the antihistaminic drug, olopatadine hydrochloride, in seven steps with 59% overall yield. Lewis acid mediated ring opening of the cyclic ether 3 resulted in the formation of the desired Z-isomer of 2 (E/Z = 2:3) as the major component, which could be improved to E/Z = 1:9 after two recrystallizations. The present method is operationally simple and has several advantages compared to the conventional methods of manufacturing olopatadine hydrochloride,[2a] where a large excess of the reagent is required and the reaction has to be performed under strict anhydrous conditions.

All reagents and solvents were used as received from the manufacturer. Petroleum ether (PE) used refers to the fraction boiling in the range of 60–80 °C. Melting points are recorded using Büchi B-540 melting point apparatus in capillary tubes and are uncorrected. IR spectra were recorded on a PerkinElmer IR spectrophotometer Model 68B or on a PerkinElmer 1615 FT IR spectrophotometer. ¹H (200, 400, and 500 MHz) and ¹³C (50, 100, and 125 MHz) NMR spectra were recorded on Bruker and Bruker Avance 400 spectrometers, using a 1:1 mixture of CDCl₃ and CCl₄ as solvent. The chemical shifts (δ) and coupling constants (Hz) are reported in the standard fashion with reference to CHCl₃ (δ = 7.26, for ¹H) or the central line (δ = 77.0 of CDCl₃ for ¹³C). In the ¹³C NMR spectra, the nature of the carbons (C, CH, CH₂, or CH₃) was determined by recording the DEPT-135 spectra. HRMS (ESI) were taken on Orbitrap (quadrupole plus ion trap) and TOF mass analyzer. Standard abbreviations were used to explain the multiplicities of the signals. The reaction progress was monitored by the TLC analysis on TLC plates precoated with silica gel 60 F₂₅₄ (Merck) and visualized by fluorescence quenching or I₂ or by charring after treatment with p-anisaldehyde and also 2,4-DNP. Merck’s flash silica gel (230–400 mesh) was used for column chromatography.

Methyl 2-(11-Allyl-11-hydroxy-6,11-dihydrodibenzo[b,e]oxepin-2-yl)acetate (4)

2-(11-Oxo-6,11-dihydrodibenzo[b,e]oxepin-2-yl)acetic acid (5; 5.00 g, 18.65 mmol) was dissolved in MeOH (100 mL) and cooled to 0 °C. SOCl₂ (2.06 mL, 27.98 mmol) was added dropwise over 30 min and the resulting solution was stirred at r.t. for 24 h. The solvent was evaporated almost to dryness and the residue was partitioned between CH₂Cl₂ (50 mL) and sat. aq NaHCO₃ (50 mL). The organic layer was separated, dried (Na₂SO₄), filtered, and concentrated under reduced pressure to give isoxepac ester, which was used without further purification. To a stirred mixture of isoxepac ester (5.00 g, 17.66 mmol) and Zn (3.44 g, 53 mmol) in DMF (50 mL) was added allyl bromide (1.66 mL, 19.43 mmol) at 0 °C. After 2 h, the mixture was filtered to remove the excess Zn. Aq 10% HCl (20 mL) was added and the organic layer was separated. The aqueous layer was extracted with small portions of EtOAc (3 × 20 mL), the combined organic extracts were dried (Na₂SO₄), filtered, and concentrated under reduced pressure. The resulting liquid was purified by column chromatography (PE–EtOAc, 8:2) to give 4 (5.2 g, 91% over two steps) as a thick colorless liquid; R_f = 0.5 (PE–EtOAc, 8:2).

IR (CHCl₃): 3482, 3073, 2951, 1735, 1490 cm^–1.

¹H NMR (200 MHz, CDCl₃ + CCl₄): δ = 2.86–2.97 (m, 1 H), 3.34–3.44 (m, 1 H), 3.60 (s, 2 H), 3.68 (s, 3 H), 5.04 (d, J = 15.5 Hz, 1 H), 5.09–5.18 (m, 2 H), 5.47 (d, J = 15.5 Hz, 1 H), 5.35–5.56 (m, 1 H), 6.90–7.00 (m, 1 H), 7.06 (d, J = 8.1 Hz, 1 H), 7.15–7.31 (m, 3 H), 7.56 (d, J = 2.2 Hz, 1 H), 7.94–7.84 (m, 1 H).

¹³C NMR (50 MHz, CDCl₃ + CCl₄): δ = 40.5, 48.7, 51.8, 73.6, 75.7, 119.4, 121.3, 125.8, 125.9, 126.8, 127.0, 127.5, 129.5 (2 C), 133.5, 134.5, 139.0, 142.1, 154.7, 171.9.

HRMS (ESI): m/z [M + H]⁺ calcd for C₂₀H₂₁O₄: 325.1434; found: 325.1433.

Methyl 2-[11-Hydroxy-11-(3-hydroxypropyl)-6,11-dihydro­dibenzo[b,e]oxepin-2-yl]acetate (6)

9-BBN (1.80 g, 14.81 mmol) was added to a well-stirred solution of olefin 4 (4.00 g, 12.34 mmol) in anhydrous THF (40 mL) at r.t. and the reaction mixture was stirred for 24 h at 70 °C. The mixture was quenched with aq 3 M NaOH (0.54 g, 13.50 mmol) at 0 °C, followed by the dropwise addition of 30% H₂O₂ (3.50 mL, 37.03 mmol). The resulting solution was stirred for 6 h at r.t. to cleave the boron complex. The organic phase was separated and the aqueous layer extracted with EtOAc (3 × 20 mL). The combined organic phases were washed with brine (30 mL), dried (Na₂SO₄), filtered, and concentrated under reduced pressure. The crude product was purified by column chromatography (PE–EtOAc, 7:3) to afford diol 6 (3.54 g, 84%) as a colorless liquid; R_f = 0.4 (PE–EtOAc, 1:1).

IR (CHCl₃): 3502, 2949, 1735, 1491 cm^–1.

¹H NMR (200 MHz, CDCl₃ + CCl₄): δ = 1.20–1.56 (m, 2 H), 2.08–2.28 (m, 1 H), 2.64–2.89 (m, 1 H), 3.36–3.64 (m, 4 H), 3.67 (s, 3 H), 4.84 (br s, 1 H), 5.01 (d, J = 15.4 Hz, 1 H), 5.43 (d, J = 15.4 Hz, 1 H), 6.93 (d, J = 7.2 Hz, 1 H), 6.98–7.32 (m, 4 H), 7.62 (d, J = 2.2 Hz, 1 H), 7.97 (d, J = 7.5 Hz, 1 H).

¹³C NMR (50 MHz, CDCl₃ + CCl₄): δ = 27.3, 40.4, 41.9, 52.0, 62.7, 73.7, 76.6, 121.2, 125.8, 126.2, 126.9 (2 C), 128.3, 129.3 (2 C), 134.4, 139.7, 143.4, 154.9, 172.6.

HRMS (ESI): m/z [M + Na]⁺ calcd for C₂₀H₂₂O₅ + Na: 365.1359; found: 365.1360.

Methyl 2-[4′,5′-Dihydro-3′H,6H-spiro(dibenzo[b,e]oxepine-11,2′-furan)-2-yl]acetate (3)

To a solution of the diol 6 (3.00 g, 2.50 mmol) in anhydrous CH₂Cl₂ (30 mL) in an oven-dried flask under a N₂ atmosphere was added PTSA (83 mg, 0.43 mmol) at r.t. The solution was stirred for an additional 10 min and then the reaction was quenched by the addition of H₂O (20 mL). The resulting organic mass was extracted with CH₂Cl₂ (3 × 20 mL), and the combined organic layers were washed with brine (30 mL), dried (Na₂SO₄), filtered, and concentrated under reduced pressure. The crude product was column-purified over silica gel (PE–EtOAc, 9:1) to furnish the spiroether 3 (2.81 g, 99%) as a colorless liquid; R_f = 0.7 (PE–EtOAc, 8:2).

IR (CHCl₃): 2926, 1738, 1488 cm^–1.

¹H NMR (400 MHz, CDCl₃ + CCl₄): δ = 1.87–1.99 (m, 2 H), 2.56–2.72 (m, 2 H), 3.59 (s, 2 H), 3.69 (s, 3 H), 4.33–4.17 (m, 2 H), 5.02 (d, J = 15.3 Hz, 1 H), 5.56 (d, J = 15.3 Hz, 1 H), 6.97 (d, J = 7.2 Hz, 1 H), 7.03 (d, J = 8.03 Hz, 1 H), 7.10–7.26 (m, 3 H), 7.51 (s, 1 H), 7.74 (d, J = 7.5 Hz, 1 H).

¹³C NMR (100 MHz, CDCl₃ + CCl₄): δ = 25.9, 40.7, 42.6, 51.9, 68.9, 72.9, 85.1, 121.4, 124.5, 126.2 (2C), 126.9, 127.0, 129.21, 129.27, 133.6, 140.2, 143.4, 153.9, 172.0.

HRMS (ESI): m/z [M + H]⁺ calcd for C₂₀H₂₁O₄: 325.1434; found: 325.1437.

Methyl 2-[11-(3-Hydroxypropylidene)-6,11-dihydrodibenzo[b,e]oxepin-2-yl]acetate (2)

To an ice cold (0 °C), magnetically stirred solution of spiroether 3 (2.00 g, 6.12 mmol) in anhydrous CH₂Cl₂ (25 mL) was added anhydrous crystalline AlCl₃ (2.05 g, 15.43 mmol) in one portion under N₂. The resulting mixture was warmed to r.t. and the red-orange reaction mixture was stirred at r.t. until the completion of reaction (7 h, by TLC, eluent: PE–EtOAc, 6:4). The mixture was then poured into an ice cooled 10% aq HCl (20 mL) and the aqueous layer was extracted with CH₂Cl₂ (3 × 20 mL). The combined organic extracts were dried (Na₂SO₄), filtered, and concentrated under reduced pressure. The crude mixture was purified by silica gel column chromatography (PE–EtOAc, 7:3) to afford 2 (1.9 g, 95%) as a white solid; mp 102–104 °C; R_f = 0.4 (PE–EtOAc, 6:4).

IR (CHCl₃): 3446, 2921, 1736, 1463 cm^–1.

¹H NMR (200 MHz, CDCl₃ + CCl₄): δ = 2.38–2.49 (m, 0.8 H, E-isomer), 2.63–2.73 (m, 1.2 H, Z-isomer), 3.53 (s, 2 H), 3.68 (s, 3 H), 3.69–3.75 (m, 0.8 H, E-isomer), 3.81 (t, J = 6.3 Hz, 1.2 H, Z-isomer)_, 5.19 (br s, 2 H), 5.73 (t, J = 7.8 Hz, 0.6 H, Z-isomer), 6.06 (t, J = 7.8 Hz, 0.4 H, E-isomer), 6.70 (d, J = 8.2 Hz, 0.4 H, E-isomer), 6.79 (d, J = 8.2 Hz, 0.6 H, Z-isomer), 7.00–7.34 (m, 6 H).

HRMS (ESI): m/z [M + H]⁺ calcd for C₂₀H₂₁O₄: 325.1434; found: 325.1437.

Methyl 2-{11-[3-(Dimethylamino)propylidene]-6,11-dihydro­dibenzo[b,e]oxepin-2-yl}acetate (7)

To homoallyl alcohol 2 (1.00 g, 3.08 mmol) in pyridine (16 mL) was added MsCl (0.95 mL, 11.72 mmol) gradually at 0 °C. The reaction mixture was allowed to come to r.t. and stirred for further 2 h. The mixture was quenched with H₂O (5 mL) and then extracted with EtOAc (2 × 20 mL). The combined organic layers were washed with brine (10 mL) and concentrated. To a solution of the resulting oil in MeOH (20 mL) was added 50% aq Me₂NH (5.20 mL, 18.0 equiv) and the mixture was stirred under reflux for 3 h. The solvent was evaporated and the residue extracted with EtOAc (2 × 20 mL). The combined organic layers were washed with brine (10 mL), dried (Na₂SO₄), filtered, and concentrated under reduced pressure. The crude reaction mixture was purified by silica gel column chromatography (CH₂Cl₂–MeOH, 97:3) to give 7 (0.91 g, 84% over two steps) as a pale yellow liquid; R_f = 0.4 (MeOH).

IR (CHCl₃): 1740, 1495, 1221 cm^–1.

¹H NMR (200 MHz, CDCl₃ + CCl₄): δ = 2.15 (s, 2 H), 2.24 (s, 4 H), 2.31–2.62 (m, 4 H), 3.51 (s, 2 H), 3.67 (s, 3 H), 4.73 (br s, 1 H), 5.45 (br s, 1 H), 5.69 (t, J = 7.1 Hz, 0.6 H, Z-isomer), 6.02 (t, J = 6.9 Hz, 0.4 H, E-isomer), 6.69 (d, J = 8.3 Hz, 0.4 H, E-isomer), 6.78 (d, J = 8.3 Hz, 0.6 H, Z-isomer), 6.98–7.37 (m, 6 H).

Acknowledgment

P.L. would like to thank CSIR, New Delhi, India, for a fellowship. S.P.C. is grateful to the Council of Scientific and Industrial Research (CSIR)-New Delhi for research funding under the ACT programme of the 12th five year plan. We are thankful to Dr. U. R. Kalkote and Dr. H. B. Borate for helpful discussions and Centaur Pharmaceuticals Pvt. Ltd. for providing the starting material.

• References

• 1 McGrath NA, Njardarson JT. J. Chem. Educ. 2010; 87: 1348
  CrossrefSearch in Google Scholar
• 2a Ohshima E, Otaki S, Sato H, Kumazawa T, Obase H, Ishii A, Ishii H, Ohmori K, Hirayama N. J. Med. Chem. 1992; 35: 2074
  CrossrefSearch in Google Scholar
• 2b Sano T, Sugaya T, Kasai M. Org. Process Res. Dev. 1998; 2: 169
  CrossrefSearch in Google Scholar
• 3 Shimamura T, Shiroishi M, Weyand S, Tsujimoto H, Winter G, Katritch V, Abagyan R, Cherezov V, Liu W, Han G, Kobayashi T, Stevens R, Iwata S. Nature 2011; 475: 65
  CrossrefSearch in Google Scholar

For synthetic processes, see:
• 4a Oshima E, Kumazawa T, Otaki S, Obase H, Ohmori K, Ishii H, Manabe H, Tamura T, Shuto K. US Patent US005116863, 1992 ; Chem Abstr. 1988, 108, 167330.
• 4b Lever W, Leighton H. European Patent EP0214779, 1990 ; Chem Abstr. 1987, 107, 58673.
• 4c Bosch C, Bachs R, Gomez A, Alonso M, Bessa A. Patent WO2006/010459, 2006; Chem Abstr. 2006, 144, 192128.
• 4d Tarur V, Bhise N, Sathe D, Naidu A, Aher U, Patil S, Verma S, Sawant K, Naik T, Amre R. WO2007/105234 A2, 2007 ; Chem Abstr. 2007, 147, 371628.
• 4e Castellin A, Ferrari C, Galvagni M. U.S. Patent US 2011/0065936 A1, 2011 ; Chem Abstr. 2010, 153, 554939.
• 5 Bosch J, Bachs J, Gómez AM, Griera R, Écija M, Amat M. J. Org. Chem. 2012; 77: 6340
  CrossrefSearch in Google Scholar
• 6 Nishimura K, Kinugawa M. Org. Process Res. Dev. 2012; 16: 225
  CrossrefSearch in Google Scholar
• 7 Ueno K, Kubo S, Tagawa H, Yoshioka T, Tsukada W, Tsubokawa M, Kojima H, Kasahara A. J. Med. Chem. 1976; 19: 941
  CrossrefSearch in Google Scholar
• 8 Barbier P.  C. R. Acad. Sci. 1898; 128: 110
  Search in Google Scholar
• 9a Brown HC, Subba Rao BC. J. Am. Chem. Soc. 1956;  78: 5694
  Search in Google Scholar
• 9b Brown HC, Zweifel G. J. Am. Chem. Soc. 1959;  81: 247
  Search in Google Scholar
• 10 Recrystallization details: after two recrystallization using PE–EtOAc (9:1) system, compound 2 with E/Z = 2:3 ratio could be improved to E/Z = 1:9 ratio.

• References

• 1 McGrath NA, Njardarson JT. J. Chem. Educ. 2010; 87: 1348
  CrossrefSearch in Google Scholar
• 2a Ohshima E, Otaki S, Sato H, Kumazawa T, Obase H, Ishii A, Ishii H, Ohmori K, Hirayama N. J. Med. Chem. 1992; 35: 2074
  CrossrefSearch in Google Scholar
• 2b Sano T, Sugaya T, Kasai M. Org. Process Res. Dev. 1998; 2: 169
  CrossrefSearch in Google Scholar
• 3 Shimamura T, Shiroishi M, Weyand S, Tsujimoto H, Winter G, Katritch V, Abagyan R, Cherezov V, Liu W, Han G, Kobayashi T, Stevens R, Iwata S. Nature 2011; 475: 65
  CrossrefSearch in Google Scholar

For synthetic processes, see:
• 4a Oshima E, Kumazawa T, Otaki S, Obase H, Ohmori K, Ishii H, Manabe H, Tamura T, Shuto K. US Patent US005116863, 1992 ; Chem Abstr. 1988, 108, 167330.
• 4b Lever W, Leighton H. European Patent EP0214779, 1990 ; Chem Abstr. 1987, 107, 58673.
• 4c Bosch C, Bachs R, Gomez A, Alonso M, Bessa A. Patent WO2006/010459, 2006; Chem Abstr. 2006, 144, 192128.
• 4d Tarur V, Bhise N, Sathe D, Naidu A, Aher U, Patil S, Verma S, Sawant K, Naik T, Amre R. WO2007/105234 A2, 2007 ; Chem Abstr. 2007, 147, 371628.
• 4e Castellin A, Ferrari C, Galvagni M. U.S. Patent US 2011/0065936 A1, 2011 ; Chem Abstr. 2010, 153, 554939.
• 5 Bosch J, Bachs J, Gómez AM, Griera R, Écija M, Amat M. J. Org. Chem. 2012; 77: 6340
  CrossrefSearch in Google Scholar
• 6 Nishimura K, Kinugawa M. Org. Process Res. Dev. 2012; 16: 225
  CrossrefSearch in Google Scholar
• 7 Ueno K, Kubo S, Tagawa H, Yoshioka T, Tsukada W, Tsubokawa M, Kojima H, Kasahara A. J. Med. Chem. 1976; 19: 941
  CrossrefSearch in Google Scholar
• 8 Barbier P.  C. R. Acad. Sci. 1898; 128: 110
  Search in Google Scholar
• 9a Brown HC, Subba Rao BC. J. Am. Chem. Soc. 1956;  78: 5694
  Search in Google Scholar
• 9b Brown HC, Zweifel G. J. Am. Chem. Soc. 1959;  81: 247
  Search in Google Scholar
• 10 Recrystallization details: after two recrystallization using PE–EtOAc (9:1) system, compound 2 with E/Z = 2:3 ratio could be improved to E/Z = 1:9 ratio.

• Supplementary Material