We have a project in my group on bitopic ligands targeting the adrenergic receptors. We recently published a perspective paper on the topic that you can check out here. Anyway, today we had a project meeting regarding the synthesis of bitopic salbutamol analogues.
To this end, one of the guys wanted to synthesise an epoxide as outlined above. However, to his (and everyone else surprise) even under mild conditions he managed to loose both of his benzylic alcohols in the process. The crude product is of excellent purity and he isolated the over-reduced phenol in >90% yield.
This is a common problem but it really is trial and error. I would say that the case above is the most extreme example I have seen considering that it was run at atmospheric pressure and room temperature. It brought memories back from when I was a PhD student. The very last reaction I did in the lab was an attempt to reduce an alkene, however, as you can probably guess what I got out as the major product was the deoxygentaed molecule. However, this reaction was somewhat more messy than that above and the desired molecule may have been hiding in the mixture as well. D!