A patent was issued last week that has several interesting aspects. The reaction is shown in the diagram and is a cycloaddition of carbon dioxide, aniline derivatives and propylene oxide to form oxazolidinone.
The mechanism (shown in the second diagram) utilizes the iodide (or other halide) from tetrabutylammonium iodide (TBAI) for the nucleophilic attack on the epoxide to open the ring, aided by coordination of the oxygen to the nickel catalyst, to form the 1-iodopropyl-2-alkoxide coordinated to the nickel. The alkoxide oxygen attacks the electron-deficient carbon of carbon dioxide to form the iodopropyl carbonate anion, that performs the intramolecular displacement of the iodide to form the cyclic carbonate that also liberates the iodide leaving group to regenerate the TBAI catalyst.
In parallel, the aniline derivative reacts with the propylene oxide, aided by coordination with the nickel, to form N-hydroxypropyl aniline that couples with the cyclic carbonate to form the oxazolidinone while liberating the nickel catalyst for another cycle and forming propylene glycol.
One of the interesting finds in the patent is that tetrabutylammonium iodide performs much better (85% yield) than the corresponding bromide (45% yield), chloride (32% yield) or fluoride (21% yield). This is due to the fact that BOTH the order of nucleophilicity and the order of leaving group ability are iodide > bromide > chloride > fluoride. That makes the both the ring opening of the propylene oxide faster and the ring closing to cyclic carbonate faster.
In addition, it was found that the nickel catalyst is activated (notated in the patent by the addition of the letter A in Tables I and II) by removing the water and solvent from the catalyst complex. Activation of the nickel catalyst by removing solvent and water increases when using TBAI as co-catalyst from 22% to 85%. This is consistent with conventional thought processes for activating nucleophilicity of the anions such as iodide and the alkoxide.
Table III in the patent shows the application of this reaction to a wide variety of aniline derivatives. Electron withdrawing groups give higher yields (p-nitroaniline: 95% yield, p-chloroaniline: 90% yield) than electron donating groups (p-methylaniline: 81% yield, p-methoxyaniline: 79% yield).
In summary, the effects of solvent, hydration and halide counterion introduced with the quat salt, deliver the performance we expect in phase-transfer catalysis reactions, even though the tetrabutylammonium salts are acting as soluble sources of halide catalyst, not as true phase-transfer catalysts.
We teach these effects of hydration, solvent, choice of anions and other process parameters in our 2-day course “Industrial Phase-Transfer Catalysis.” Inquire now about bringing this course in-house to your company.
