[Rutherford]

The reaction sequence below was taken from this publication: http://pubs.acs.org/doi/abs/10.1021/jo026676p. I don't understand the stereoselectivity of this reaction sequence. I thought that epoxide formation will happen from the sterically less hindered face (bottom) and then after trans-diaxial opening at the less sterically hindered center, the other anti product will be produced. Why it isn't the case?

[zoork34]

It might be helpful to build a molecular model.  Sometimes what looks to be the least sterically hindered site on paper is actually the opposite, especially in multi-ring systems.

[clarkstill]

They cite this synlett for that step:

https://www.thieme-connect.com/products/ejournals/abstract/10.1055/s-1997-5766

Which shows the epoxide forming on the same side as the substituents, followed by Furst Plattner ring opening as you said. I agree it's curious that the epoxide forms on that side, but I guess that's chemistry.

[Rutherford]

I agree it's curious that the epoxide forms on that side, but I guess that's chemistry.

I can't possibly accept this as an answer

[Rutherford]

Reaction at the other face would lead to trans-hydrindane which has more 1,3-diaxial interaction than cis-hydrindane. Seems that thermodynamics prevails here.

[clarkstill]

I agree with the argument, but not necessarily the conclusion... do you think the epoxidation is reversible? I'd be surprised if the epoxide were capable of reoxidizing acetone to DMDO. Kinetic control with a product-like transition state, a la Hammond Postulate?

[orgopete]

Reaction at the other face would lead to trans-hydrindane which has more 1,3-diaxial interaction than cis-hydrindane. Seems that thermodynamics prevails here.

I think you are looking at this backwards. Since you already know on which side the epoxide forms, you should try to discern the factors that might favor it. The conformation of the alkene probably has some degree of flexibility, perhaps expoxidation occurs from one of those unexpected conformations. That is, the rate governing factors may be greater for a less favored conformation or some other factor controls epoxide formation.

[Rutherford]

I agree with the argument, but not necessarily the conclusion... do you think the epoxidation is reversible? I'd be surprised if the epoxide were capable of reoxidizing acetone to DMDO. Kinetic control with a product-like transition state, a la Hammond Postulate?

I don't expect it to be reversible, but kinetics would favor the product I thought of.

Reaction at the other face would lead to trans-hydrindane which has more 1,3-diaxial interaction than cis-hydrindane. Seems that thermodynamics prevails here.

I think you are looking at this backwards. Since you already know on which side the epoxide forms, you should try to discern the factors that might favor it. The conformation of the alkene probably has some degree of flexibility, perhaps expoxidation occurs from one of those unexpected conformations. That is, the rate governing factors may be greater for a less favored conformation or some other factor controls epoxide formation.

Any thoughts on that?

[orgopete]

Reaction at the other face would lead to trans-hydrindane which has more 1,3-diaxial interaction than cis-hydrindane. Seems that thermodynamics prevails here.

I think you are looking at this backwards. Since you already know on which side the epoxide forms, you should try to discern the factors that might favor it. The conformation of the alkene probably has some degree of flexibility, perhaps expoxidation occurs from one of those unexpected conformations. That is, the rate governing factors may be greater for a less favored conformation or some other factor controls epoxide formation.

Any thoughts on that?

Stereochemistry has never been a strength of mine. I had seen instances in which the results were the reverse of predictions. Since these often had a chiral auxiliary, one just uses the opposite chirality. This simply suggested to me that making such predictions are very difficult. Look at the current model for Sharpless oxidations and compare them with his early ones.

For me, I don't know if this is an unexpected result. What are the products if a peracid were used or BrOH/NaOH. I recall finding the opposite stereochemistry in substituted cyclopentenes in which it was rationalized that the substituent preferred to be equatorial and this created an axial hydrogen that formed the steric barrier.

Without making models, the structure seems as though a mixture should have resulted. The fact that one isomer was preferred does not convince me that I should have predicted the cis isomer and to be surprised that it was trans. As I recall, the traditional prediction is expoxidation opposite of the more polar substituent. Hydrogen bonding can give cis-expoxidation. For this reason, I suggested that now that you know the product, then see if you can find any factors that would lead to that product. I don't think you have a rigid starting material, so there should be some lattitude in rationalizing the result. The additional isomer related by clarkstill seems to indicate this may be more general and reduces polarity as a factor.