^ That's quite interesting. Mind if I have a go at it?
At first glance, the mass spectrum of the impurity doesn't really resemble that of MDPV, but I believe this substance is in fact very closely related to MDPV. The base peak at m/z 124 is probably a butylpyrrolinium ion. It is less abundant than the corresponding butylpyrrolidinium (m/z 126) of MDPV, which is consistent with a lower theoretical stability.
m/z 121 and 149 could be assigned to methylenedioxyphenyl and methylenedioxybenzenecarbonyl ions, respectively. They indicate that the impurity contains both the methylenedioxy and beta-keto moieties of MDPV.
m/z 273 is a probable molecular ion. m/z 230 results from the loss of a propyl radical from m/z 273. The propyl loss is quite characteristic of pyrovalerone type structures.
m/z 82, albeit at low abundance, could be assigned to the methylenepyrrolinium ion resulting from the loss of propene from m/z 124. The corresponding methylpyrrolidinium (m/z 84) of MDPV is more stable, and therefore more prominent in the spectrum of MDPV.
m/z 70 should be the pyrrolinium (dehydropyrrolidinium) ion.
I have a slight problem explaining m/z 65, but it's also present in the spectrum of MDPV according to at least one publication. I'm also not sure about m/z 187 and 204. The spectrum has apparently not been subtracted, so they could be due to background or interfering compounds.
Of course, the identification is far from complete, and my peak assignments may be off, but this is a not a PhD dissertation.
Based on the above interpretation, I believe that the mass spectrum is that of an unsaturated analogue of MDPV, or dehydro-MDPV if you will, in which the additional double bond is located in the pyrrolidine substructure (making it a pyrroline). Now, I'm not entirely convinced that this is the impurity actually present in the sample material - it may very well be an artefact resulting from the loss of water in the GC injector. If this is the case, its origin is most likely 2'-hydroxy-MDPV (1-(benzo[1,3]dioxol-5-yl)-2-( 2-hydroxypyrrolidin-1-yl)pentan-1-one). This particular substance also strikes me as a likely oxidation product and human metabolite of MDPV.
Could you perhaps shed any light on the abundance and retention time of the impurity relative to MDPV in the sample? Is the GC peak symmetrical and well defined? Water loss artefacts tend to display broad or tailing peaks.
As to whether this impurity might be implicated in the epic story of the tan MDPV, I really don't know. It seems unlikely if the abundance is relatively low, but as I have stated before, we should keep an open mind and there are some very interesting potential targets other than DA/NE/5-HT receptors.
