Well, I'm really drunk right now, but I've been reading some really cool shit on the wacker. I spent a few days at the library reading journal articles and shit and found out the following info:
1) Pd+2 catalyzes not one reaction but two. The first one is of course the conversion of an olefin to a ketone. The second, however, is the isomerization of the olefin. It turns out that this isomerization is about 300 times faster than the oxidation. This goes to prove the next point.
2) Within minutes of addind safrole to the rxn. mixture, it is converted to the equilibrium mixture that consists of 99% isosafrole. In other words, the reaction doesn't give a flying fuck whether you use safrole or isosafrole as the starting reagent. This transformation occurs by a pi comlex formed from the safrole and the Pd+2. This complex can rapidly shift between the 1-2 carbons and the 2-3 carbons. These two pi complexes are also in equilibrium soon after the rxn starts.
3) The carbonyl products are derived from these pi complexes. The 1-2 complex produces MDP-1-P and MDP-2-P. The 2-3 complex produces MDP-2-P and MDP-3-P (the aldehyde). The oxidation rxn from the pi complex is irreversible and is governed only by rate (thus the product that is formed fastest is the major product, not the equilibrium product which would be MDP-1-P -- this product is the most stable due to a conjucated double bond with the aromatic ring). Thus from the starting material (safrole or isosafrole) you get a mixture of all three carbonyl compounds. From experience, the major product is MDP-2-P.
4) MDP-2-P and MDP-1-P have almost the same boiling point (within 5C) at 20 mm Hg. Therefore, it is virtually impossible to separate them by distillation. This may be the reason why many bees report getting product in the correct distillation range that doesn't aminate. I am wondering, though, will MDP-1-P aminate with the MeNO2 amalgamation? My opinion is no, due to steric hendrance and stability of the ketone due to conjugated double bonds.
5) In one article, some scientists discovered that using palladium calalysts with ammine ligands with result in a selectivity of >95% methyl ketone (MDP-2-P in our case). I am not sure of the cause of this phenomenon, though. It could be due to one of two factors. The first is that the ammine ligands slow down the isomerization process, so that it is much slower than the rate of oxidation. The other is that the ligands slow down the rate of formation of MDP-1-P while not affecting the formation of the desired ketone. There is, however, a descrepancy in the literature. One article states that an equilibrium mixture of alkene is formed within minutes of starting the rxn. Another article (the one that used the ammine ligands) stated that after 24 hrs of rxn, the concentration of 1-alkene was >80% with PdCl2, having no ligands. This just doesn't add up. The only difference is that the first article used 1-hexene as the staring alkene, while the second use 1-octene. Shouldn't the final concentration of 1-octene be <1%? Both procedures used similar amounts of catalyst.
6) For the O2 wacker, PO2 only needs to be about 0.5 atm. Higher pressures actually slow down the oxidation process. This is for the oxidation of 1-hexene, which is probably faster than the oxidation of safrole anyway. Therefore, all this crap about 3+ atm of O2 seems unnecessary, if not detrimental to yields. The O2 is only needed to reoxidize the Pd, which is independant of the actual oxidation rxn -- we have two rxns going on simultaneously.
This is all the shit that I found out. If you have any questions about is, please feel free to post them. Again, I am really fucked up now, so I may not have explained everything clearly. Hope this shit helps.