When medicinal chemists use NaH/DMF for milligram scale etherifications, we understand that they need confidence to achieve high yield from their very expensive highly specialized molecules, so they must use non-optimal reaction conditions to be efficient with time and raw materials.
The O-methylation reported here this month was performed on a 14 kg scale and the reaction conditions chosen are surprising (actually cringeworthy for a PTC expert).
The inventors knew about PTC since they used catalytic tetrabutylammonium iodide as the phase-transfer catalyst. Chemists who know enough phase-transfer catalysis to use a tetrabutylammonium salt almost always know that hydroxide is certainly a strong enough base to deprotonate an alcohol (pKa of secondary alcohol ~ 17) under PTC conditions. Many chemists also know that PTC excels in etherification, especially methylation, more than any other method of etherification.
Sometimes, PTC-hydroxide is not suitable for performing etherifications due to the presence of functional groups on the molecule that may be sensitive to hydroxide such as esters or acid chlorides that could hydrolyze, acidic chiral C-H groups that could racemize, etc. But in this case, the entire molecule other than the hydroxyl to be O-methylated is made up of hydrocarbons and ethers, none of which are sensitive to hydroxide.
The procedure reported in Example 4 in this patent was to dissolve 14.2 kg of the starting material in 71 L DMF, cooling to 0 C, then adding 3.68 kg sodium hydride and 1.31 kg tetrabutylammonium iodide. Then 18.12 kg methyl iodide were added slowly and stirred for 2 hours at 20-30 C. Workup used 284 L of water or aqueous NH4Cl and 71 L of ethyl acetate. It is not known in which phase(s) the DMF wound up but it is clear that this process and workup generated a lot aqueous waste and organic stream, that in combination contained 71 L of DMF that had to be recovered. Even though the yield after workup was an attractive 95.9%, this procedure unnecessarily suffers from solvent, waste and safety issues on this scale.
The most obvious PTC conditions for such a reaction that most chemists (not expert in PTC) would choose would be to use toluene as the solvent, NaOH as the base and almost any phase-transfer catalyst should work. The obvious major advantages of NaOH over sodium hydride include much better safety, much lower cost and much easier handling. The use of toluene would likely require no more than two aqueous washes instead of four, cutting the aqueous waste volume AT LEAST by 50% (likely more) and the toluene would be easily and fully recoverable without a problem, unlike DMF. The choice of methyl iodide over an alternative methylating agent such as dimethyl sulfate would depend mostly on safety setup (chemical cost being less important than safety).
Tetrabutyl ammonium iodide is a very expensive tetrabutylammonium salt and is not needed since there is an iodide leaving group. Tetrabutylammonium bromide (TBAB) is the least expensive tetrabutylammonium salt and would be a better first choice. However, if there would be a concern using TBAB for decomposing the methyl iodide in-situ to the less reactive and lower boiling point methyl bromide, then one could choose tetrabutylammonium hydrogen sulfate which is much less expensive than TBAI and does not contain a nucleophilic anion that could consume the methyl iodide.
Overall, the choice of reaction conditions for this 14 kg scale reaction handling nearly 500 L of liquids, generating hydrogen as a byproduct instead of water as a byproduct, for PTC O-methylation that enjoys thousands of much easier literature precedents, appears to be surprising (cringeworthy) due to the many disadvantages of safety, handling, cost and waste.
This is a classic case in which expert PTC Process Consulting by PTC Organics should have been engaged to develop a much lower cost, much greener process. If your company finds itself in a similar unfortunate situation, now contact Marc Halpern of PTC Organics to develop low-cost high-performance green chemistry, all while minimizing the number of experiments to achieve high R&D efficiency.
