In attempting to form a lithiated dianion then C-alkylate with prenyl bromide an unusual product was observed in up to 50% yield which I cannot mechanistically account for. The structure was confirmed by X-ray crystallography and is consistent with the NMR of the bulk material on a 100mg scale. The reaction was worked up with aqueous ammonium chloride. The reaction was reproducible with different halides. Traces of water or oxygen are possible but unlikely to be major contributors as the reaction was carried out under N2, in oven dried glassware.
Any input would be appreciated as I am currently writing up my thesis and this is sitting as a glaring unanswered question and the proposal I have lacks any literature support.
Thanks,
Eric Harris
To test Elisa's hypothesis, after generating the dianion, I would simply try to add (freshly distilled) isoprene instead of prenyl bromide.
The hydroxyl group in the product could come from the reaction of an organolithium species with oxygen introduced at the stage of the hydrolysis, when you open your flask before pouring the NH4Cl solution. Sometimes such reactions can be extremely fast.
Just a very wild guess. Since you have 1 equivalent excess of BuLi, would it be possible that this extra equivalent abstracts a proton of your alkylating agent forming a diene (eliminating BuH and LiX) that is then attacked by the anion and after hydrolysis you obtain that product? Did you reduced the amount of BuLi (to i.e. 2.1 equiv.) in your reaction and observed any difference? Maybe is a good experiment to quickly try...Good luck with your writing!
I know you're at the writing stage, but one or two more experiments wont hurt and its a chance to get back in the lab :o) This does look like you have some oxidation taking place. I agree with Elisa that there is a high probability of an t-BuLi mediated 1,4-elimination. Addition of the aryl anion followed by hydrolysis would give the protonated benzyl species not the benzyl alcohol. This suggests there is an oxidation taking place. I suggest rather than simply having a nitrogen atmosphere you thoroughly deaereate your solvent and rerun the reaction under argon or nitrogen and see if the yields are affected. Then, repeat the reaction but under dry air and with a gentle stream of air running through your reaction mixture to see if the yield pops up. You have quite an electron rich arene and its the benzylic position thats hydroxylating. Adding a radical scavenger could also kill this side product from forming. I wonder whether your prenyl bromide can also be a source of radicals? How does this rxn behave in the dark? Have you swapped out the prenyl bromide for a chloride or a tosylate/mesylate/triflate etc?
Nice contributions from Elisa and Panayiotis. I would offer that THF can sometimes be hard to deoxygenate completely, maybe make super-sure it is oxygen free, and use as little solvent as possible. Also... that prenyl bromide.... I had a bottle once that was very unstable, and the alcohol (reaction with H2O) was being formed quite fast...
Can you generate the di-anion in a two stage process?
Can you play a little with the temperature? (make it -78C at the addition stage and afterwards slowly warm-up to whatever you feel will work there?
To test Elisa's hypothesis, after generating the dianion, I would simply try to add (freshly distilled) isoprene instead of prenyl bromide.
The hydroxyl group in the product could come from the reaction of an organolithium species with oxygen introduced at the stage of the hydrolysis, when you open your flask before pouring the NH4Cl solution. Sometimes such reactions can be extremely fast.
I agree to Yvan Six and Elisa opinion.
Hope to do the related experiment, in this regards.
Hi Eric,
I will be working with similarly substituted aromatics soon, I am wondering what kind of precautions you took when purifying those products chromatographically? Addition of water or triethylamine to your silica? I am interested in not losing a MOM group while keeping my other relatively acidic functionality in the neutral form. Thanks
Thank you all for your input.
3 equivalents of t-BuLi are required to form the dianion, the third equivalent rapidly undergoes an E2 reaction with the t-Bu halide to generate isobutene. n-BuLi does not work and butylated products from butylhalide are observed, t-BuLi is not present in excess. All other methods attempted to generate the dianion failed to undergo the lithium halogen exchange. The negative 40 temperature was optimal for dianion generation.
The mesylate and chloride didn't give any reaction product. Solvent was freshly distilled from sodium wire then degassed. The same product was observed when quenched under N2 via syringe and no other products were isolated. Light was minimised with tinfoil. Prenyl bromide and prenyl iodide were freshly prepared and distilled as both are quite light sensitive. Prenyl iodide ceases to be useful within a few minutes if not cautious.
Although a 1,4 elimination is possible with the dianion acting as a base (also explaining the limited yield), the resulting diene would be completely unfunctionalised, addition of a lithium anionic species to such a system is not thermodynamically viable at such low temperatures in the absence of a catalyst or other metal, unless someone can provide me with a reference suggesting otherwise. Addition to the diene would also result in any anion being in the wrong place and only the E-alkene is observed in the product.
Further experimental work on this is not possible at this time for numerous reasons. Otherwise I would make partially deuterated prenyl bromide, or try cyclic equivalents to determine the mechanism and where each methyl group is coming from along with a range of other experiments.
I am moderately sure that the alcohol is coming from the work up as a hydrolysis product, potentially from a bromide or oxetane which would be highly labile. If a lithiated reaction product can form at the benzylic position then lithium halogen exchange with another molecule of prenyl bromide is more likely than oxidation by oxygen.
The proposal I have is ipso deprotonation of prenyl bromide to generate a reactive carbene, which rearranges. Very little literature exists on cis-substituted allyl carbenes, so it has been hard to justify.
That raises a good question, what carbene or radical traps would be stable to 3 equivalents of t-BuLi?
Prenyl bromide gives dibromide and nonsubstituted, dibromide gives with PhLi bromo derivative in alpha position, water wark up gives corresponding hidroxy/ I think 3 eq of organolithium is too much.
The first equivalent of t-BuLi deprotonates the phenol and a normal SN2 reaction follows with the prenyl bromide to give the O-prenyl ether. The second equivalent of t-ButylLi then deprotonates the ether alfa to the oxygen. The resulting anion then debrominates the ring to leave the aryl carbonion and put the bromine on the carbon with oxygen in the preny group. Finally the aryl carbanion does a nucleohilic displacement of the bromine on the carbon to form an oxetane ring that is hydrolyzed on workup back to the phenol, now alkylated with the hydroxy-substituted prenyl group. I'm not saying that this is a slam dunk, but its my best idea for the 10 minutes I have to work on this.
Oxetane formation explains the hydroxyl group well but it doesn't explain the prenyl group rearranging at low temperatures.
I wonder if the reaction goes through a radical mechanism. I assume the formation of the dianion was confirmed by trapping the intermediate with D2O or something? The formed dianion species is so electron-rich that it could undergo a single-electron-transfer to prenyl bromide forming a radical anion (a radical on the carbon). At the same time, this dianion is basic enough to cause 1,4-elimination to generate some isoprene inside the flask. The radical anion and isoprene should react readily, and the addition should happen in a way that it generates more stable tertiary allylic radical. Double bond isomerization, followed by a C-H abstraction could give a benzylic allylic radical, which could abstract bromide from prenyl bromide to form a benzyl bromide species. It's a highly reactive benzyl-allylic bromide, so it could get hydrolyzed upon work-up to provide what you have isolated.
I am not sure if there is any radical inhibitor that is compatible with this highly basic condition, but you could use a radical clock-type compound instead of prenyl bromide to rule out this possibility.
The reaction of metallation of 1,3 dicyanobenzene in 2 position followed by CH3J interception leads to 2-Ethyl-1,3-dicyanobenzene but not methyl on the same reason. Methyl derivative is deprotonated and alkylated by CH3J.
For starters, the addition of phenyllithium to isoprene is well-known ... it's a basic method of initiating anionic polymerization (see pdf.) This would neatly assemble the carbon skeleton.
What's difficult is getting the newly-introduced oxygen into place. I agree that a halogen at that location would solvolyze under the workup conditions, but the requisite sequence of proton abstractions and halogen transfers is scarcely credible.
OK, I think I have it. At least in the absence of further data, this seems to be the most plausible explanation with the most precedence that I will be able to find references for.
1. The dianion forms as expected with three equivalents of t-BuLi. One to deprotonate the phenol, one for lithium halogen exchange and one to eliminate t-butyl bromide.
2. The high basicity of the dianion results in a vinylogous E2 elimination of HBr from prenyl bromide in a rate limiting step, explaining the observation of ~50% recovered debrominated/iodinated starting material.
3. The isoprene formed rapidly reacts with any dianion present undergoing addition forming the n-3 allylic anion which has precedence.
4. The n-3 allyl anion undergoes a suprafacial anionic 1,4-H-shift to generate the aromatic n-5 anion, further stabilised by the pendant MOM-group.
5. The terminal methyl group rotates to relieve steric strain in the n-5 anion resulting in the observed stereoselectivity, allylic lithium species have low barriers to such isomerisation, precedented by the inability to selectively generate E or Z crotyl lithiums even at low temperatures.
6a. The least substituted position of the n-5 anion reacts with oxygen encountered in work up (when the flask is opened). The labile peroxy group is hydrolysed on work up to give the observed product.
6b. The least substituted position of the n-5 anion undergoes lithium halogen exchange with remaining prenyl bromide in equilibrium and is driven forward due to oxetane formation via an SN2 cyclisation. The corresponding halide or oxetane is then hydrolysed on work up to give the observed product.
Thank you for everyone's valuable input.
f you make a hydrolysis workup without traces of oxygen, and obtain normal product, you can exclude a hypotesis of oxygenation by athmospheric oxygen.
I don't think I can reasonably exclude 6a. or 6b. without further experimentation.
6a. I am sceptical of, as no protonated equivalent was found, I'm also fairly sure the reaction was quenched under nitrogen, although that was a few years ago now. I tend to be extremely cautious when working with 3 equivalents of t-Butyl lithium.
6b. Really relies on the oxetane formation pushing the lithium halogen exchange reaction to completion as I would expect the n-5 anion to be more stable than prenyl lithium.
If the addition of a phenyl lithium species to isoprene results in an n-5 aromatic anion that may be worth researching in its own right.
Most of the steps seem reasonable to me. A few final observations:
Oxidation of the anion seems less likely than halogen exchange with prenyl bromide, the fate of half of which is otherwise unexplained - and you certainly would not expect it to be inert to the anionic species.
Whether atmospheric oxidation is reasonable depends on the scale of the reactions: what volume of air is needed to explain the yield -- and how reasonable is an effective exposure to such a volume?
If 6b does anything, it's much more likely to convert to the o-quinone methide than to the benzooxetane.
- Badges
- Science topic
- Similar topics
- Chemistry
- Organic Chemistry