Williamson's attempted ether synthesis

08 Oct
Published by Alice Williamson


Request for Help

Hello, I’m Alice and I joined Mat Todd’s group as a Postdoc just over a month ago. I have started working on the Malaria open source drug discovery project and this is the first of many posts to come, thanks for reading and please let me know your thoughts. My first task was to synthesise compound Sd a fairly unintimidating target, especially as Paul Ylioja has developed a robust synthesis of the required alcohol 1.

Firstly, I repeated Paul’s synthesis of 1. Paul was away at conferences during my first month in the lab and I think it is worth mentioning how easy it was to find relevant preps and raw data for comparison of spectra. I am certain that it would have been more arduous had I needed to root through paper-based lab notebooks or the imaginative filing methods often found in a laboratory. Anyhow, I soon got my hands on 2 g of the aldehyde, primed for reduction to alcohol (1) and subsequent alkylation with 2-bromoacetamide (2) to yield Sd. The alcohol (1) doesn’t seem to be very stable either on silica or on storage (as also noted by Paul) and displays two spots by TLC soon after synthesis. Therefore, as my lab book details, I either attempted alkylation of the crude material soon after synthesis or in some cases following a rapid attempt at purification.



I quickly repeated two of Paul’s failed attempts to check that nothing had gone awry with his starting material or reagents. The first method employed sodium hydride as the base with anhydrous tetrahydrofuran as solvent, however no conversion was observed. I next attempted conditions featuring in situ Finkelstein with potassium iodide, potassium carbonate as base and acetonitrile as solvent with no synthetic joy.

Increasing reaction temperature did not help with either of these methods, resulting in levels of decomposition that varied in severity, from matching the decomposition products found in the alcohol co-spot, to formation of a black tar. The alkylation was no more amenable when DMF was employed as solvent in conjunction with NaH. I also tried a new batch of NaH to try and rule out problems with the base, but to no avail.

My next attempt employed microwave heating and also saw no product formation.

Next, I decided to select potassium tert-butoxide as base, which brought back good memories from my PhD of my fumehood bud’s frequent large-scale sublimations of this pesky base. It’s probably worth mentioning here that all of the solvents I used were anhydrous and if necessary reagents (e.g. Et3N) were freshly distilled, recrystallised (e.g. KI) or sublimed (e.g. KOtBu). The reactions using KOtBu in THF both in the presence and absence of 18-crown-6 both afforded trace amounts of a ‘new product’ by TLC. Unfortunately, careful attempts to purify the reaction mixture were unsuccessful and the only spot(s) isolated corresponded to impure alcohol starting material.

I came across a patent that detailed alkylation of a hindered tertiary alcohol with a primary bromide, in the presence of silver carbonate and triethylamine in dichloromethane. I applied these conditions to this reaction, but even after three days, only decomposition spots present in the alcohol could be detected by TLC and attempts to isolate any of these spots were unsuccessful.


So what’s going wrong?

Honestly, I’m not sure. The alcohol starting material (1) doesn’t seem to be very stable to acidic conditions but doesn’t completely decompose in the reactions I have tried, unless subjected to heating for prolonged reaction periods. We have started to question the stability of Sd and wonder whether this compound is less stable than the alcohol and thus limits our ability to detect the desired compound. I am in the process of attempting alkylation of 1 with other alkylating agents (e.g. benzyl bromide), although early indications are not so promising.


And what’s next?

I found a rather nice paper that details the selective reduction of esters to ethers (even in the presence of amides) using an InBr3, Et3SiH system. I therefore synthesised ester 5 and hope to obtain Sd via reduction of the ester carbonyl.

I have also resynthesised Matin’s compound Sa and hope that this selective reduction might be applied to this tertiary amide and also to the previously synthesis ethyl ester. We need to ascertain whether Sd and analogues are inherently unstable, or whether its just the alkylation that’s problematic. If the reductive method works, I think it would be interesting to compare the biological results of the three compounds and also to synthesise mono and dialkyated analogues of both amides.



It may be that the NH2 on the amide needs a protecting group. It might be possible to use a nitrile (start from bromoacetonitrile)  as a masked form of the amide and then gently hydrolyse the nitrile to the amide at the end.

I am no pyrrole expert, but I was wondering could your alcohol and other alkylated products be decomposing along these lines?