We set up the condensation reactions with the same molar quantities and under the same conditions as we had done the thiamine catalyzed benzoin reactions as few weeks before. It soon became apparent that the condensation of anisaldehyde was not occurring to any extent and the condensation of tolualdehyde was occurring much slower than it had with benzaldehyde. Heating the tolualdehyde reaction mixture to 60 °C seemed to speed up the reaction enough so that about a 50% conversion could be realized in a week. The anisaldehyde reactions were a bust and everyone switched to tolualdehyde. It ended up that the tolualdehyde reactions were incubated from 5 to 14 days at 60 °C. The toluoin did not crystallize from the reaction mixture in the same way that benzoin had, so it was extracted with ethyl acetate. The thiamine remained in the aqueous phase but the unreacted tolualdehyde crossed to the non-aqueous along with the toluoin product. We are able to crystalize toluoin from the resulting extract with ethanol and methanol. Crystalizing a product that contained significant amounts of impurity was a challenge. In most cases the crystallization had to be done several times and the (re)crystallization solution had to be cooled to less than zero Celsius to form crystals. The resulting dimethylbenzoin had a lambda max of 256 nm in ethanol. You will notice in the IR spectrum that the alcohol peak of 4-4′-dimethylbenzoin is quite sharp. Wait, we are not done with this yet! GC revealed that the recovered solid products contained a mixture of up to three products.