A Novel Synthesis of 1-Acetyl-4-Isopropenyl-1-Cyclopentene by Chemoselective Cyclization of 4-Methyl-3-( Oxobutyl )-4-Pentenal : An Important Intermediate for Natural Product Synthesis

Neste artigo apresentamos a oxidação direta do epóxi-limoneno 1 com KIO4 em água como uma melhor alternativa para se obter o ceto-aldeído 3, um importante intermediário na síntese de produtos naturais. O ceto-aldeído 3 é ciclizado com Al2O3-ácido de forma quimosseletiva produzindo, preferencialmente, a cetona 4. Estas duas reações elevam o rendimento total da síntese da cetona 4 para 70% em comparação com 8% pelo método citado na literatura.


Introduction
Terpenes are abundant chiral compounds present in nature and an important source of intermediaries for natural product synthesis [2][3] .
Wolinsky by two routes [4][5] prepared ketone 4 (Fig. 2), a natural product first isolated from Spanish Eucalyptus globulus in 1947 5 and aldehyde 5 (Fig. 4), an important intermediate in several natural product syntheses [6][7][8][9][10][11] (Fig. 1), from limonene-oxide 1 (Scheme 1).The keto-aldehyde 3, intermediate in the Wolinsky approach has also been used in the study of asymmetric cyclizations 12 and chemoselective reduction 13 .Only two other papers reported the formation of ketone 4: First, in 1977, Wolinsky 14 reported the formation of ketone 4 from α-Pinene (Fig. 2) and then, in 1996, Jones and Kover 15 demonstrated the formation of the enantiomer of ketone 4 by hydrolysis of epoxy-tosylates prepared from (-)-R-Carvone (Fig. 3), but the first Wolinsky synthesis still remained the route of choice to prepare the ketone 4. In the original Wolinsky approach which gave a low overall yield for formation of ketone 4, the diol 2 was obtained in 39% yield and the subsequent cyclization of keto-aldehyde 3, proceeded in only 10% yield.
Diol 2 is soluble in water, and thus the reaction is best conducted at low temperature and with occasional shaking to induce the formation of crystals.The isolation of diol 2, as an oil, from water is difficult.Diol 2 is also obtained in its hydrated which complicates its characterization and quantification.Drying the diol adds a tedious but necessary task to the whole process in order to avoid use of excess NaIO4 13 , even though the periodate cleavage is conducted in aqueous medium.
The cyclization of keto-aldehyde 3 leads to a mixture of ketone 4 and aldehyde 5, which is very difficult to separate by conventional methods.Attempts to purify 4 by distillation led to substantial losses of material even under reduced pressure.Preparative chromatography was not an alternative since 4 and 5 elute together in all solvent mixtures tested.The chemoselective preparation of aldehyde 5 16 , via formation of an enamine (Fig. 4), was reported in the literature 4 , but a chemoseletive preparation for ketone 4 was still necessary 17 .

Discussion and Results
We wish to report here two modifications of Wolinsky's approach that solve the problems outlined above and enhance the yields of keto-aldehyde 3 and the ketone 4 17 .
First we oxidized limonene-oxide 1 directly with KIO4 leading to keto-aldehyde 3 in one step and high yield (85%).This avoids preparation of intermediate diol 2, represents economy of time and reagent and solves the problems associated with isolation of diol 2. Although oxidation of epoxides with HIO4 was reported in the literature 18 , it has never been applied to limonene-oxide 1.
Second, continuing our study of cyclization of 1,6 dicarbonyls we cyclized keto-aldehyde 3 with acidic Al2O3 19 and verified that the process was highly chemoselective resulting in ketone 4 in high yield (95%).
(+)-R-limonene 7 was initially epoxidized with peracid prepared in situ by mixing ethyl chloroformate and H2O2 in dichloromethane (Scheme 2).The desired limonene-oxide 1 was isolated by distillation, as a mixture of dias-   tereomers, in 73% yield.This mixture of epoxides was then treated with an equimolar quantity of KIO4 in water.
We observed the formation of at least three side products (by HRGC analysis) during the oxidation of limoneneoxide 1 with KIO4.They were not characterized but could be ascribed to epoxide ring opening products from attack of various nucleophiles present in the reaction medium.Keto-aldehyde 3 was refluxed in hexane over acidic Al2O3 while the reaction was monitored by HRGC.Analysis of the product mixture after the total consumption of keto-aldehyde 3 shows a ketone 4 to aldehyde 5 ratio of 4:1 (by integration of peak area).When the reaction was conducted at reflux for prolonged times the ratio of 4 to 5 increased to 8:1 (HRGC) After isolation, analysis of the crude light yellow oil revealed a ratio of 4 to 5 of 14:1.This ratio was determined by comparison of the integrated 1 H-NMR signal of the aldehyde proton present in compound 5 (δ 9.95 ppm), and the vinylic proton present in compound 4 (δ 6.68 ppm).
It is not clear if chemoselectivity arises from a simple equilibration of products due to catalysis on the surface of the alumina or if the selectivity arises during the desorption from alumina 20 .

Conclusion
Ketone 4 was prepared efficiently, selectively, and in high overall yield by comparison with previous routes.Cyclization of aldehyde 3 with acidic Al2O3 is shown to be very chemoselective.The direct oxidation of epoxy-limonene 1 with KIO4 is a novel and better way to obtain keto-aldehyde 3, an important intermediate for natural product synthesis.

Limonene-oxide (1)
To a 1 liter round-bottom flask equipped with a magnetic stirrer were added 25 g (180 mmol) of (R)-(+) limonene, 29 mL (180 mmol) of ethyl chloroformate, 193 mL of 30% H2O2, 193 mL of dichloromethane and 178 g of NaHPO4.The flask and contents were vigorously stirred at room temperature for 48 h (limonene consumption was followed by gas chromatography).Upon consumption of starting material the mixture was separated in a separatory funnel.The aqueous phase was extracted with dichloromethane (3 x 50 mL).The combined organic phases were washed with 1% aqueous sodium sulfite solution until complete decomposition of excess H2O2.The presence of H2O2 was verified with a standard starch-I2 test.The organic phase was then washed with water (3 x 75 mL), dried over anhydrous Na2SO4, and the solvent was evaporated in a rotatory evaporator.The resulting oil was distilled under reduced pressure (3 mmHg) and a fraction corresponding to compound 1 was collected at 40 °C (20 g, 131.5 mmol, 73% yield).The identity of this fraction was confirmed by analysis by GC/MS and IR, and by comparison to an authentic sample.

4-Methyl-3-(oxobutyl)-4-pentenal (3)
A 250 mL round-bottom flask equipped with a magnetic stirrer was charged with 13 g (85 mmol) of 1,2-limonene oxide (1), followed by addition of a suspension of 19.6 g (85 mmol) of KIO4 in 200 mL of water.The flask and contents were heated on a steam bath to 55 °C and vigorously stirred for 8 h.The limonene-oxide consumption was followed by gas chromatography.Upon consumption of starting material the mixture was saturated with NaCl and was extracted with ether (3 x 60 mL).The layers were separated and the organic phase was dried over anhydrous Na2SO4.The solvent was evaporated in a rotatory evaporator at room temperature.Compound 3 was isolated as an oil (obtained 12 g, 71.4 mmol, 85% yield) that was pure enough to use in the next reaction.This oil was analyzed by GC/MS, IR, 1 H and 13 C-NMR, and then it was compared with an authentic sample.(10), 135 (10), 107(30), 67(30), 43(100), (molecular ion m/z 168, not found).

1-Acetyl-4-isopropenyl-1-cyclopentene (4)
To a 500 mL round-bottom flask equipped with a magnetic stirrer and a reflux condenser were added 12 g (71 mmol) of aldehyde 3, 300 mL of hexane and 50 g of acidic Al2O3.The flask and contents were heated on a steam bath to 65 °C with vigorous agitation.Aldehyde consumption was accompanied by gas chromatography.Upon consumption of starting material the mixture was filtered in a Buchner funnel and washed with CHCl3 (4 x 50 mL).The organic solution was dried over anhydrous Na2SO4 and the solvent was evaporated in a rotatory evaporator at room temperature.Compound 4 was obtained as an oil (10.5 g, 70 mmol, 90% pure by HRGC, 95% yield).This oil was analyzed by GC/MS, IR, 1 H and 13 C-NMR, and then compared with an authentic sample.

Figure 1 .
Figure 1.All the compounds above have been prepared from keto-aldehyde 3.