Diastereoselective Spiroannulation of Phenolic Substrates: Synthesis of (±)-2-tert-Butyl-6-methoxy-1-oxaspiro[4,5]deca-6,9- diene-8-one.

The synthesis of a new spiroether is described. The title compound is obtained as a diastereomeric mixture in 45% yield.


Introduction
We recently published our results related to the diastereoselective spiroannulation of phenolic substrates producing spiroethers 1-3, as shown in Figure 1 [1,2].

Figure 1
We had observed that increasing the size of the alkyl group on the side chain of the phenol increased the diastereoselectivity from 64/36, when the alkyl group was a methyl, to 81/19 for a tertiary butyl group. Furthermore, the oxidant used to carry out these transformations also played a role in the outcome of the reactions. For  chemical yields as well as higher diastereoselectivity than either [bis(trifluoroacetoxy)iodo]benzene (PIFA) or (Diacetoxyiodo)benzene (PIDA). We speculated that the nature and the location of the substituent on the aromatic ring of the phenol had a strong influence on the reaction outcomes. We now wish to report the synthesis of the isomeric spiroether 4 and show that indeed, the position of the methoxy substituent does affect the results of this reaction.

Results and Discussion
The synthesis of phenol 8 was performed as described in Scheme 1. We have previously reported the synthesis of aldehyde 5, which we prepared from the corresponding catechol aldehyde in 78% yield [3]. Following its synthesis, aldehyde 5 was condensed with pinacolone to afford the α,βunsaturated ketone 6 in 65% yield after purification by chromatography. Hydrogenation of 6 was carried out under a slight positive pressure of hydrogen to afford 7 as an oil in 98% yield. Reduction of the ketone carbonyl, followed by acid workup, provided the necessary deprotected phenol 8 as a racemate in 82% yield (Scheme 1).

Scheme 1
We treated phenol 8 under reaction conditions seemingly identical to those previously described in the synthesis of 1-3 ( Figure 1) [1,2], and we obtained racemic 4 in 45% yield with a 50/50 diastereomeric ratio. This ratio was determined by analysis of the 1 H-NMR spectrum of the crude reaction mixture. The signals for H 10 located at 6.55 and 6.63 ppm for each of these diastereomers were used to determine the diastereomeric ratio, while signals for H 7 (5.44 and 5.46) and H 9 (6.01 and 6.05) as well as the signals for the methoxy group (3.74 and 3.77) were used as a confirmation of the stereoselectivity observed. Contrary to compounds 1-3, which were partially separable by column chromatography [1,2], the diastereomers of 4 were not separable and characterization of the products was carried out from the diastereomeric mixture. In this case, only one spot was observed by thin layer chromatography. We did not attempt the spiroannulation with either PIFA and PIDA since results with these oxidants did not previously show any diastereoselectivity [1,2].
While the lower yield of 4 (45%) compared to 3 (69%) can be partially attributed to the greater steric factors near the reactive site in 8 [1,2], the diastereoselectivity observed does suggest that the position of the methoxy group has an influence on the approach of the nucleophile in this spiroannulation reaction, since it is the only difference between the two starting phenols. It has been shown that oxidation of phenolic substrates produces a phenoxonium ion (9a) such as the one shown in Figure 2 [4].

Figure 2
When the methoxy group is located at the 3-position on the phenol, as in the case previously studied for 1-3 [1,2], the phenoxonium ion can be stabilized by resonance giving rise to structure 9b as shown in Figure 2. On the other hand, when the methoxy group is located at the 2-position on the phenol, as seen in 8, the second resonance structure (equivalent to 9b) for the phenoxonium ion 10 cannot be formed and the intermediate is hence less stable and most likely less reactive as well. This lower reactivity may also partially explain the poor yield obtained. Furthermore, assuming that 10 is the reacting intermediate, one would not expect any stereoselectivity in the formation of the spirocenter since the steric factors influencing the approach of the nucleophilic hydroxyl would be identical from either side of the cyclic carbocation.

Conclusions
We have reported the synthesis of a new spiroether 4 and have shown that the location of the methoxy group on the phenol influences the diastereoselectivity of this spiroannulation reaction, most likely via the resonance stabilization of the reactive intermediate. It is expected that other electron donating groups located at the 3-position on the phenol will have a similar effect in terms of stereoselectivity. We are presently investigating this possibility, as well as the influence the size of the 3-subtituent may have on the diastereoselectivity of this reaction.

Acknowledgements
We acknowledge the financial contribution of the University of Northern British Columbia in support of this work.

General
Melting points were determined on a hot stage instrument and are uncorrected. Infrared spectra were recorded either as KBr pellets or neat on a Perkin Elmer System 2000 FTIR. 1  OMe ppm using TMS as internal standard. 13 C-NMR spectra were recorded on a Bruker AMX300 spectrometer at 75.4MHz and chemical shifts are expressed in ppm using chloroform as internal standard. Mass spectra were recorded on a Hewlett Packard 5898B spectrometer. Elemental analysis was performed at the Central Equipment Laboratory of the University of Northern British Columbia.