Abstract

2-Alkyl cyclohexenones are useful intermediates for organic synthesis. The Wittig reaction of a series of aldehydes with (cyclopropylmethyl)triphenylphosphonium bromide delivered the corresponding alkenyl cyclopropanes. UV irradiation in the presence of Fe(CO)5 converted the alkenyl cyclopropanes to the 2-substituted cyclohexenones. This approach enabled a three-step synthesis of the tricyclic core of estrone methyl ether.
Introduction
Estrone 1 and estradiol 2 (Eq. 1), important hormones that control the menstruation cycle of mammals, are an attractive research area due to their high biological activity and many pharmaceutical applications. A variety of synthetic strategies have been published for the synthesis of estrone and its derivatives.1 The first total synthesis of estrone was published in 1948 by Anner and Miescher.1a The tricyclic phenanthrenone core 3 played a key role in that synthesis and is a potentially useful intermediate for steroid syntheses.2
![]() |
(1) |
We envisioned (Scheme 1) that aldehyde 4a could be homologated to the corresponding alkenyl cyclopropane by Wittig reaction with the commercially available (cyclopropylmethyl)-triphenylphosphonium bromide 5.3 UV irradiation of the alkenyl cyclopropane 6a so prepared in the presence of Fe(CO)5 under a CO atmosphere4 would then lead to the corresponding cyclohexenone 7a. Acid catalyzed cyclization would then deliver the tricyclic ketone 3, in just three steps from the commercial aldehyde 4a.
Scheme 1.
Results and Discussion
2-Alkyl cyclohexenones are important intermediates for target-directed synthesis.5 While several methods have been developed to prepare such cyclohexenones, these routes tend to be either low yielding or multi-stepped.6 There were two challenges to be overcome in developing the route to 2-alkylcyclohexenones outlined in Scheme 1: Development of the Wittig reaction, and optimization of the Fe-mediated cyclocarbonylation.
Development of the Wittig Reaction
We initially performed the Wittig reaction utilizing one equivalent of commercial (cyclopropylmethyl)triphenylphosphonium bromide and one equivalent of potassium t-butoxide at cryogenic temperatures. This protocol led to modest yields, with some unreacted aldehyde. During optimization, we found that increasing both base and phosphonium salt to a slight excess increased conversion. While maintaining the slight excess of phosphonium bromide, we doubled the amount of base, which led to complete conversion and high yields. Finally, we found that the reaction could be carried out at room temperature. The products (Table 1) were isolated as a mixture of both the Z and E isomers of the alkenyl cyclopropane, which was of no consequence for the subsequent cyclocarbonylation.
Table 1. Wittig Reaction and Carbonylation.
| Entry | Aldehyde | Wittig Yield (%) | Cyclohexenone | Carbonylation Yield (%) |
|---|---|---|---|---|
| 1 | ![]() |
93 | ![]() |
89 |
| 2 | ![]() |
84 | ![]() |
89 |
| 3 | ![]() |
78 | ![]() |
86 |
| 4 | ![]() |
98 | ![]() |
72a,b |
| 5 | ![]() |
89 | ![]() |
86 |
| 6 | ![]() |
88 | ![]() |
85 |
Autocooling in Rayonet was turned off.
For a previous preparation of 3d, see ref 8.
Optimization of the Fe-Mediated Cyclocarbonylation
We initiated carbonylation studies using the conditions that we had previously developed.6 We found that using less than one equivalent of Fe(CO)5 reduced the conversion rate, while adding greater than two equivalents did not lead to significantly higher conversions. Increasing the concentration of the solution typically lowered conversion. We did find that the conversion and yield improved when the mixture was agitated periodically, providing for an exchange between the product and starting material forming the thin film that was being irradiated. We maintained a CO atmosphere in each of the runs, even though CO is formed during the irradiation of the Fe(CO)5.
During the carbonylation of 6d, we observed a significantly lower conversion of the alkenyl cyclopropane to the cyclohexenone 7d. It was our hypothesis that the steric bulk of the molecule hindered the carbonylation. When we ran the photolysis without autocooling and so warmer, the conversion improved significantly. This two-step conversion appears to be a versatile and convenient way to prepare 2-alkyl cyclohexenones from the corresponding aldehydes.7
Cyclization of the Enone
We envisioned (Scheme 2) that a variety of Lewis acids could effect the cyclization of the enone 3.5b,9 We initiated our studies with BF3·OEt2 and TiCl4. While both Lewis acids were efficient in forming the tricyclic ring, BF3·OEt2 was milder in its reaction with the cyclohexenone, leading to less decomposition and thus higher yield. We found that by adding the Lewis acid slowly at 0°C, we could also minimize decomposition.
Scheme 2.
The tricyclic ketone was approximately a 1:1 mixture of trans and cis diastereomers 3 and 8. As had been described2, the trans diastereomer 3 could be separated by fractional crystallization. Alternatively, the two diastereomers could be efficiently separated by column chromatography. The isolated cis diastereomer 8 could then be epimerized to the 1:1 mixture by heating to reflux with K2CO3 in THF. Using this approach, one could prepare gram quantities of the crystalline trans ketone 3.
Conclusion
We expect that the aldehyde to 2-alkyl cyclohexenone conversion described here will make such cyclohexenones more readily available as intermediates for target-directed synthesis. We also envision that this facile new preparation of the ketone 3 will facilitate future steroid syntheses.
Experimental
1-(4-Cyclopropylbut-3-enyl)-3-methoxybenzene 6a
Potassium t-butoxide (1.0 M solution in THF, 50 mL, 50 mmol) was added over 20 min to a suspension of (cyclopropylmethyl)triphenylphosphonium bromide (10.0 g, 25 mmol) in 25 mL of dry THF at 0°C. The external cooling was removed and the mixture was stirred for 30 min. The aldehyde 4a (3.28 g, 20 mmol) was added, and the solution was stirred at rt for 1 h. The mixture was quenched with 1N HCl (50 mL) and then partitioned between water and EtOAc (3 × 100 mL). The combined organic extract was dried (MgSO4) and concentrated. The residue was chromatographed to give the alkenyl cyclopropane 6a (3.76 g, 93% yield, 77:23 Z/E) as an oil. TLC Rf = 0.62 (9:1 hexanes/EtOAc), 1H NMR (Z) δ 7.19 (t, J = 7.8 Hz, 1H), 6.72-6.84 (m, 3H), 5.33-5.38 (m, 1H), 4.76 (t, J = 10.8 Hz, 1H), 3.79 (s, 3H), 2.61-2.72 (m, 2H), 2.44-2.52 (m, 2H), 1.44-1.53 (m, 1H), 0.63-0.72 (m, 2H), 0.27-0.31 (m, 2H); 13C NMR (Z) δ CH3: 55.1 CH2: 36.2, 29.4, 6.8 CH: 134.5, 129.2, 127.0, 120.9, 114.2, 111.0, 9.5 C: 159.6, 143.8; 1H NMR (E) δ 7.19 (t, J = 7.8 Hz, 1H) 6.72-6.84 (m, 3H), 5.52-5.57 (m, 1H), 5.01 (dd, J = 15.2, 8.4 Hz, 1H), 3.79 (s, 3H), 2.61-2.72 (m, 2H), 2.25-2.32 (m, 2H), 1.30-1.36 (m, 1H), 0.63-0.72 (m, 2H), 0.27-0.31 (m, 2H); 13C NMR (Z) δ CH3: 55.1 CH2: 36.1, 29.4, 6.4 CH: 134.4, 129.2, 127.2, 120.9, 114.2, 111.0, 9.5 C: 159.6, 143.8; IR 2935, 1598, 1487, 1455, 1258 cm-1; MS m/z (%) 202 (M, 30), 137 (65), 121 (100); HRMS: Calcd for C14H18O 202.1358, obsd 202.1360.
2-(3-Methoxyphenethyl)cyclohex-2-enone 7a
To the alkenyl cyclopropane 6a (202 mg, 1.0 mmol) in 15 mL of 2-propanol (0.075 M) was added Fe(CO)5 (392 mg, 2.0 mmol). The reaction vessel was purged with CO, a CO balloon was attached, and the mixture was photolyzed for 8 h at room temperature in a Rayonet apparatus (300 nm) set for autocooling. The reaction was halted every 2 hours to agitate the tube inside the larger tube, after which photolysis was restarted. At the end of the irradiation, DBU (304 mg, 2.0 mmol) was added, and the mixture was stirred at room temperature for 1 h under nitrogen. The mixture was diluted with 40 mL of EtOAc and filtered through a small pad of packed silica gel and then subsequently concentrated. The residue was chromatographed to give 2 mg of unreacted 2a and 202 mg of 7a (89% yield based on 6a not recovered) as an oil. TLC Rf = 0.17 (9:1 hexanes/EtOAc), 1H NMR δ 7.16 (t, J = 8.4 Hz, 1H), 7.07 (d, J = 9.2 Hz, 1H), 6.81 (t, J = 9.2 Hz, 1H), 6.73 (m, 1H), 6.61 (m, 1H), 3.78 (s, 3H), 2.68 (m, 2H), 2.45 (m, 4H), 2.28 (m, 2H), 1.98 (m, 2H); 13C NMR δ CH3: 55.4 CH2: 38.7, 35.1, 31.7, 26.1, 23.1 CH: 149.6, 129.2, 121.0, 114.1, 111.2 C: 199.4, 159.5, 143.7, 139.2; IR 3442, 2960, 1734, 1206, 1099 cm-1; MS m/z (%) 230 (M, 8), 221 (7), 163 (15), 135 (95), 107 (100); HRMS: Calcd for C15H18O2 231.1385, obsd 231.1391.
7-Methoxy-2,3,4,4a,10,10a-hexahydrophenanthren-1(9H)-one 3 and 8
BF3·OEt2 (1.0 M solution in CH2Cl2, 2 mL, 2 mmol) was added over 5 min to cyclohexenone 7a (230 mg, 1.0 mmol) in 5 mL of dry CH2Cl2 (0.20 M) at 0°C. The solution was stirred at rt for 3 h, quenched with water (5 mL), and then partitioned between water and methylene chloride. The combined organic extract was dried (MgSO4) and concentrated. The residue was chromatographed using 80:20 PE:MTBE as an elution solvent to give the phenanthrenone 3 (92 mg, 40% yield) as a solid and phenanthrenone 8 (90 mg, 38% yield, 78% yield overall) as an oil. The cis diastereomer 8 (85 mg, 0.37 mmol) was then heated to reflux with a suspension of K2CO3 (690 mg, 5 mmol) in dry THF (2 mL) overnight to epimerize the phenanthrenone to a 1:1 mixture of diastereomers. For phenanthreone 3: mp = 106-109°C, TLC Rf = 0.45 (8:2 hexanes/EtOAc (double elution)). 1H NMR δ 7.23 (d, J = 9 Hz, 1H), 6.75 (dd, J = 9, 3 Hz, 1H), 6.66 (d, J = 3 Hz, 1H), 3.78 (s, 3H), 2.84 (m, 2H), 2.73 (td, J = 12, 3 Hz, 1H), 2.61 (m, 1H), 2.41-2.53 (m, 2H), 2.35 (td, J = 12, 3 Hz, 1H), 2.18-2.30 (m, 2H), 1.86 (tdd, J = 13, 5, 4 Hz, 1H), 1.61-1.77 (m, 2H); 13C NMR δ CH3: 55.2 CH2: 44.3, 30.5, 29.3, 26.3, 21.7 CH: 138.0, 113.8, 112.0, 52.7, 44.3; C: 212.0, 157.8, 138.0, 131.2, 126.9; IR 1720, 1615 cm-1; MS m/z (%) 230 (M, 100), 213 (55), 187 (30), 160 (25), 135 (43); HRMS: Calcd for C14H18O 230.1307, obsd. 230.1302; For phenanthrenone 8: TLC Rf = 0.42 (8:2 hexanes/EtOAc (double elution)), 1H NMR δ 7.05 (d, J = 9 Hz, 1H), 6.75 (dd, J = 9, 3 Hz, 1H), 6.62 (d, J = 3 Hz, 1H), 3.78 (s, 3H), 2.84 (m, 2H), 2.73 (td, J = 12, 3 Hz, 1H), 2.61 (m, 1H), 2.41-2.53 (m, 2H), 2.35 (td, J = 12, 3 Hz, 1H), 2.18-2.30 (m, 2H), 1.86 (tdd, J = 13, 5, 4 Hz, 1H), 1.61-1.77 (m, 2H); 13C NMR δ CH3: 55.6 CH2: 38.8, 30.7, 29.0, 24.5, 22.6 CH: 129.7, 113.6, 112.7, 50.6, 40.4; C: 214.4, 157.8, 137.0, 131.2, 126.9; IR 1720, 1615 cm-1; MS m/z (%) 230 (M, 100), 213 (4), 187 (40), 159 (35), 147 (31), 121 (19); HRMS: Calcd for C14H18O 230.1307, obsd 230.1302.
Supplementary Material
General experimental procedures, experimental procedures for 2b-2f and 3b-3f, details of the photochemical apparatus and spectroscopic data. This material is available free of charge via the Internet at http://pubs.acs.org.
Acknowledgments
This work is dedicated to E. E. Schweizer, in recognition of his many contributions to organic chemistry. We thank DuPont Agricultural Products and the NIH (GM060287) for financial support of this work.
References
- 1.For the Bachmann-Miescher estrone synthesis, see Anner G, Miescher K. Helv Chim Acta. 1948;31:2173. doi: 10.1002/hlca.19480310739.Bachmann WE, Cole W, Wilds AL. J Am Chem Soc. 1940;62:824.Bachmann WE, Kushner AC, Stevenson AC. J Am Chem Soc. 1942;64:976.. For a review of synthetic routes to estrone, see Quinkert G, Stark H. Angew Chem Int Ed. 1983;22:637. For more recent references, see Vollhardt KPC. Pure Appl Chem. 1985;57:1819.Lecker SH, Nguyen NH, Vollhardt KPC. J Am Chem Soc. 1986;108:856.Taber DF, Raman K, Gaul MD. J Org Chem. 1987;52:28.Michellys PY, Pelliser H, Santelli M. Tetrahedron Lett. 1993;34:1931.Trost BM, Shil Y. J Am Chem Soc. 1993;115:9421.Sakamoto Y, Yasuharu H, Takahashi T. Synlett. 1995:231.Quinkert G, Del Grosso M, Doring A, Doring W, Schenkel RI, Bauch M, Dambacher GT, Bats JW, Zimmerman G, Durner G. Helv Chim Acta. 1995;78:1345.Sugahara T, Ogasawara K. Tetrahedron Lett. 1996;37:7403.Ouellet L, Langlois P, Deslongchamps P. Synlett. 1997:689.Tietze LF, Nobel T, Spescha M. J Am Chem Soc. 1998;120:8971.Hu QY, Rege PD, Corey EJ. J Am Chem Soc. 2004;126:5984. doi: 10.1021/ja048808x.Pattenden G, Gonzalez MA, McCulloch S, Walter A, Woodhead SJ. Proc Nat Acad Sci. 2004;101:12024. doi: 10.1073/pnas.0401925101.Soorukram D, Knochel P. Org Lett. 2007;9:1021. doi: 10.1021/ol063052n.
- 2.For previous preparations of the tricyclic ketones 3 and 8, see Birch AJ, Smith H, Thornton RE. Chem Ind. 1956:1310.Birch AJ, Smith H, Thornton RE. J Chem Soc. 1957:1339.Ansell MF, Ducker JW. J Chem Soc. 1961:206.Wiley RH, Crawford TH, Bray NF. J Polym Sci, Part B: Polym Lett. 1965;3:99.Nagata W, Terasawa T, Tori K. J Am Chem Soc. 1964;86:3746.Sundar NS, Subba Rao GS. J Chem Res, Synop. 1982:1381.Varech D, Lacombe L, Jacques J. New J Chem. 1984;8:445.Thompson HW, Long DJ. J Org Chem. 1988;53:4201.
- 3.For the preparation of (cyclopropylmethyl)triphenylphosphonium bromide, see Maercker A. Angew Chem Int Ed. 1967;6:557.Schweizer EE, Thompson JG, Ulrich TA. J Org Chem. 1968;33:3082.. For the utility of (cyclopropylmethyl)triphenylphosphonium bromide, see: Kataoka F, Shimizu N, Nishida S. J Am Chem Soc. 1980;102:711.Cavasotto CN, Liu G, James SY, Hobbs PD, Peterson VJ, Bhattacharya AA, Kolluri SK, Zhang X, Leid M, Abagyan R, Liddington RC, Dawson MI. J Med Chem. 2004;47:4360. doi: 10.1021/jm030651g.
- 4.For the development of Fe-mediated cyclocarbonylation, see: Taber DF, Kanai K, Jiang Q, Bui G. J Am Chem Soc. 2000;122:6807.Taber DF, Bui G, Chen B. J Org Chem. 2001;66:3423. doi: 10.1021/jo001737+.Taber DF, Joshi PV, Kanai K. J Org Chem. 2004;69:2268. doi: 10.1021/jo0302760.
- 5.For examples of the utility of 2-alkyl cyclohexenones in target-directed synthesis, see Hauser FM, Caringal Y. J Org Chem. 1990;55:555.Majetich G, Liu S, Fang J, Siesel D, Zhang Y. J Org Chem. 1997;62:6928.Iwamatsu S, Matsubara K, Nagashima H. J Org Chem. 1999;64:9625.Coelho F, Diaz G. Tetrahedron. 2002;58:1647.
- 6.For previous preparations of 2-alkyl cyclohexenones, see: Taber DF, Gunn BP, Bruce P. J Org Chem. 1979;44:450.Johnson CR, Braun MP. J Am Chem Soc. 1993;115:11014.Armitage MA, Lathbury DC, Mitchell MB. J Chem Soc, Perkin Trans 1. 1994;12:1551.Abbas AA, Kobayashi Y. Tetrahedron Lett. 2003;44:119.Takeishi K, Sugishima K, Sasaki K, Tanaka K. Chem Eur J. 2005;10:5681. doi: 10.1002/chem.200400679.
- 7.This approach is not useful for 2-aryl cyclohexenones. The iron reactions proceed with >90% yield but only to about 20% conversion. We believe that the lower conversion is due to the competing absorption of UV radiation by the aryl enone. For practical routes to 2-aryl cyclohexenones, see: Bosse F, Tunoori AR, Niestro AJ, Gronwald O, Maier ME. Tetrahedron. 1996;52:9485.Prasad ASB, Knochel P. Tetrahedron. 1997;53:16711.Kim J, Kulawiec R. Tetrahedron Lett. 1998;39:3107.Felpin F. J Org Chem. 2005;70:8575. doi: 10.1021/jo051501b.
- 8.Reetz MT, Heimbach H. Chem Ber. 1983;116:3702. [Google Scholar]
- 9.For previous examples of acid-mediated cyclization of cyclohexenones onto arenes, see: Pepin JJ, Andre-Louisfert J, Bisagni E. Bull Soc Chim Fr. 1970;8-9:3038.Hauser FM, Caringal Y. J Org Chem. 1990;55:555.Bie PY, Zhang CL, Peng XJ, Chen B, Yang Y, Pan XF. Chem J Chin Univ. 2003;24:1219.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
General experimental procedures, experimental procedures for 2b-2f and 3b-3f, details of the photochemical apparatus and spectroscopic data. This material is available free of charge via the Internet at http://pubs.acs.org.















