Sciencemadness Discussion Board

THP protection of amines?

forsh - 12-11-2013 at 03:25

Hi there,
Apologies for the long post,I hope you will bear with me. I'm trying to protect the phenolic group in 3-aminophenol, but i'm wondering if my protecting group will be unspecific and react with the amine group too.

Background
I decided to go for 2,3-dihydro-3H-pyran a I thought that would be quite selective for the phenolic group. The mechanism appears to be that the dihydropyran is partially protonated on the C=C double bond, leaving the other carbon slightly positively charged and so susceptible to nucleophilic attack from something like the lone pair on an oxygen (LINK). Obviously an amine also has that lone pair, but I was thinking that it is a much weaker nucleophile and also the reaction occurs under an acid catalyst, so the amine would be protonated and the lone pair no longer free to act as a nucleophile. Am I wrong though? Would you expect the amine to also react with dihydropyran with an acid catalyst?

Experimental
1 equiv aminophenol and the acid catalyst (0.1 equiv pyridinium p-toluenesulphonate) was dissolved in acetone, before 1.5 equiv hydropyran was added with strong stirring. This was reacted for 4 hours. Over the time the colour changed from pale yellow, then stronger yellow, then orange red, then dark green - I don't think this is a good sign.
I foolishly pumped off the solvent under vacuum and obtained a sticky tar, which redissolved (eventually) in acetone. I've extracted with several solvents and washed with DI and brine which has given me a few different coloured solvents, reds and yellows. I'm guessing i've made a few strongly coloured polymers. The IR is promising, but that could be wishful thinking. I don't have access to anything better than IR for a few weeks.

[Edited on 12-11-2013 by forsh]

[Edited on 12-11-2013 by forsh]

Nicodem - 12-11-2013 at 09:45

Quote: Originally posted by forsh  
Hi there,
Apologies for the long post,I hope you will bear with me. I'm trying to protect the phenolic group in 3-aminophenol, but i'm wondering if my protecting group will be unspecific and react with the amine group too.

Yes, anilines can be alkylated by dihydropyran under the same conditions as phenols. You can find plenty of examples in the literature.
Quote:
Obviously an amine also has that lone pair, but I was thinking that it is a much weaker nucleophile and also the reaction occurs under an acid catalyst, so the amine would be protonated and the lone pair no longer free to act as a nucleophile. Am I wrong though? Would you expect the amine to also react with dihydropyran with an acid catalyst?

An amine, or aniline in this case, is nonucleophilic only when fully protonated and when the equilibrium concentration of the unprotonated form is insignificant. In your case only a very small part of the substrate was protonated. And even if there would be an equivalent of a strong enough acid, the substrate is still an aniline rather than an aliphatic amine, so there is always a significant concentration of the unprotonated species around even under acidic conditions.
Quote:
The IR is promising, but that could be wishful thinking. I don't have access to anything better than IR for a few weeks.

You won't be able to tell anything with certainty only from the IR, unless you also have the spectra of the desired product.
You should at minimum check with TLC (or HPLC), to see what kind of a mixture you have and if it is worth bothering with the isolation (e.g., doing a column). The N-alkylated side product could be potentially removed by dissolving the mixture in methyl t-butyl ether and washing it several times with 2M NaOH(aq). This does not necessarily work (some phenols are not extractable into aq. NaOH. Furthermore, it would not remove any of the N,O-dialkylated product (or any other non-acidic side product).

Do you only need an O-protection of the aminophenol? If so, describe the planed synthesis and we can suggest you an alternative to THP.

forsh - 12-11-2013 at 10:20

Quote: Originally posted by Nicodem  

Yes, anilines can be alkylated by dihydropyran under the same conditions as phenols. You can find plenty of examples in the literature.

Thanks Nicodem, I searched and searched for literature examples of the n-alkylation, but I could only seem to find the o-alkylation - I guess I need to work on my searching techniques ;)
Quote:

An amine, or aniline in this case, is nonucleophilic only when fully protonated and when the equilibrium concentration of the unprotonated form is insignificant. In your case only a very small part of the substrate was protonated. And even if there would be an equivalent of a strong enough acid, the substrate is still an aniline rather than an aliphatic amine, so there is always a significant concentration of the unprotonated species around even under acidic conditions.

That makes sense, thanks for clarifying. I repeated the reaction today, but using EtOAc as a solvent and 10 mL conc HCl and the acid catalyst. The reaction produced a dark brown aqueous layer and pale purple organic which I have separated. Now you've explained the mechanism to me i'm not so confident that there will be much of the o-alkylated only product, but I will see. A GC showed 3 non-starting material peaks and I will run a GC-MS this week.
Quote:

You won't be able to tell anything with certainty only from the IR, unless you also have the spectra of the desired product.
You should at minimum check with TLC (or HPLC), to see what kind of a mixture you have and if it is worth bothering with the isolation (e.g., doing a column). The N-alkylated side product could be potentially removed by dissolving the mixture in methyl t-butyl ether and washing it several times with 2M NaOH(aq). This does not necessarily work (some phenols are not extractable into aq. NaOH. Furthermore, it would not remove any of the N,O-dialkylated product (or any other non-acidic side product).

I ran a column of the initially described PPTS catalysed reaction, eluting with Hex:EtOAc and this gave me a brown oil which only has a single broad peak at 3400 cm-1 on the IR, which I guess tells me that the N-alkylation is a major product.
Quote:

Do you only need an O-protection of the aminophenol? If so, describe the planed synthesis and we can suggest you an alternative to THP.


My plan is to do the O-protection so I can form an amide with an acid chloride, then I want to remove the o-protection and form an ether using allyl bromide, which I then plan to rearrange to the ortho position using a Claisen rearrangement. An alternative route which I've come up with is instead selectively doing an N-protection (Demiratas (2002) Turk J Chem) and doing the Claisen first, and then forming the amide afterwards.

[Edited on 12-11-2013 by forsh]

Nicodem - 12-11-2013 at 11:23

Quote: Originally posted by forsh  
My plan is to do the O-protection so I can form an amide with an acid chloride, then I want to remove the o-protection and form an ether using allyl bromide, which I then plan to rearrange to the ortho position using a Claisen rearrangement.

I'm not sure I understand. Why do you need a O-THP protection when all you want to is an anilide of m-aminophenol?

forsh - 12-11-2013 at 11:56

Quote: Originally posted by Nicodem  
Quote: Originally posted by forsh  
My plan is to do the O-protection so I can form an amide with an acid chloride, then I want to remove the o-protection and form an ether using allyl bromide, which I then plan to rearrange to the ortho position using a Claisen rearrangement.

I'm not sure I understand. Why do you need a O-THP protection when all you want to is an anilide of m-aminophenol?


I want to make an amide of aminophenol by reacting with the acyl chloride, but if I did this without the O-THP protection then i'd get a mix of products as the acyl chloride with also (preferably) react with the hydroxy group


Dr.Bob - 12-11-2013 at 12:18

In many cases, amines will react faster than alcohols with acid chlorides, so in those cases, it is unnecessary to protect the alcohol, as most alcohols will not react with acyl chlorides without DMAP or harsher conditions than for the amine. With anilines and phenols, there may not be as much a difference, but still, if you add 1.1 eq of the acyl chloride, I would expect to see mostly acylation of the aniline.

The allyl ether could be then be formed and you could do the Claisen. In some cases, you might just be able to add 1.1 eq. of allyl alcohol and DMAP and do both steps in one pot, not sure in this case.

Don't know how selective the regiochemistry of the Claisen reaction will be.

Many times, I just use a nitro group as a protected amine once reduced, but in this case, the allyl will make that difficult.

DJF90 - 12-11-2013 at 14:17

Reduction of nitrobenzenes should be effected chemoselectively using iron (either Fe/HCl or Fe/FeCl3 or Fe/NH4Cl or any other known iron based reduction) Dr Bob, so that should allow your preferred route to be usuable. Alternatively, selective acylation at nitrogen ought to be afforded using Schotten-Baumann conditions using NaHCO3 as the base (personally I prefer aq. K2CO3, but that'll be "incompatible" with the phenol).

forsh - 13-11-2013 at 01:51

Well I can definitely see the difference between the long experienced chemists and the young researcher ... Thanks very much for your help!

Dr Bob, I always thought that the alcohol was more reactive than the amine, hence the decision for o-protection. I have buckets of aminophenol and acyl chloride so I might try this reaction. I was planning the N-acylation as DJF suggested anyway, just with O-protection, so will try it without. I'll run a TLC/GC and see if I can separate out any mixtures on a column

With regards to the regiospecificity of the Claisen - i've found a few papers which say that with 2 substituents in meta positioning then the allyl group tends to rearrange para to a electron donating group and ortho to a EWG (e.g. 1,2,4 over 1,2,3) which is what I want.

Dr.Bob - 13-11-2013 at 06:26

I know that the Fe or Sn based reductions should work, but having run some I try to stick to hydrogenations when I can. The advantage of iron or tin reductions for home chemists are clearly there, however, but they can be messy and create a lot of metal salt wastes. The soidum dithionate reductions work in some cases as well, but are not always as clean or quick either, but easy to do. And you used to be able to buy de-rusting agents (rust stain removers) with it in them OTC.

But I am impressed with doing Claisens. I have only done a few so far, with some mixed success, but they are a great reaction when they work. And some of them provided compounds not easily made any other way.

Nicodem - 13-11-2013 at 06:50

Quote: Originally posted by forsh  
I have buckets of aminophenol and acyl chloride so I might try this reaction. I was planning the N-acylation as DJF suggested anyway, just with O-protection, so will try it without. I'll run a TLC/GC and see if I can separate out any mixtures on a column

The N-acylation of aminophenols is quite selective if done properly. I prepared many mono- and dihydroxyanilides by the reaction of a few aminophenols with diverse acyl and aroyl chlorides in dichloromethane or ethyl acetate in the presence of about 1.5 eq. pyridine (ethyl acetate is better for the work up and don't use Et3N). You add the acyl chloride slowly while stirring and cooling in the ice bath. The reactions are rapid. Either use a 1 : 1 amiphenol vs. acyl chloride ratio or a slight excess of the aminophenol (especially when commercial like 3-aminophenol). You wash the reaction mixture with 1M HCl(aq), water, dry and rotavap. The crude should be pure enough for what you need, else recrystallize. All the starting aminophenol and any other aniline side product is washed away during the HCl wash. If you don't use the aminphenol in excess, you can sometimes observe a few % of the N,O-diacylated sideproduct by NMR, but this will not interfere with the allylation (it can get O-allylated to the same product under proper conditions).

DJF90 - 13-11-2013 at 12:59

Or you could save yourself a step and buy a pack of paracetamol...

forsh - 14-11-2013 at 05:36

Tried both DFJ and Nicodem's methods today, both worked superbly with similar yields. Thank you very much! I've ran an IR of the crude and they both look great, i'm sending them off for GC-MS and NMR tomorrow so i'll get confirmation in a few days.

Meanwhile i'm going to work on the allyl ether synthesis and subsequent Claisen. I will keep this thread updated for anyone interested :)

DJF90 - 14-11-2013 at 13:18

Cool. Glad I could be of help.

forsh - 28-11-2013 at 10:00

Hi all. I've moved onto the next few steps of this synthesis, but i've hit a bit of a wall now.

To recap, I am reacting aminophenol with an acyl chloride to get the amide, then reacting this with allyl bromide to get an ether, and finally doing a Claisen rearrangement (I realise this will be non-specific).




I did the amide synth as recommended in this thread, the GC-MS just came back and it is very high purity, there were no other peaks on the GC.

The ether synthesis seems to have gone less well. I dissolved 40 mmol amide starting material in dry DMF and added 60 mmol K2CO3 and 45 mmol allyl bromide, and stirred the suspension for 3 hours at RT. I then added ether and water, with the intent of making the DMF more polar and forcing the product into the ether layer. Then I dried the ether and stripped the solvent which gave me about 60% yield of a crude material. I ran a IR on this and it appeared that the O-H peak vanished (it wasn't very big to begin with TBH) and a medium peak at 1650 cm-1 appeared which I think is the C=C bond. I did not have a GC-MS to confirm, but went on to the next step.

I then attempted the Claisen rearrangement twice. The first time I stirred the ether compound in dry DCM with zinc as a catalyst - at RT for 3 hours. This just gave me starting material. I followed a paper's method to the letter, so I think they may have been telling porkies - a Zn catalyst is just too good to be true. Next I tried refluxing the ether compound in DCM with AlCl3 for 3 hours. The reaction was quenched at the end with water. This gave me something different, but I wasn't sure what. The IR shows a shift in some of the peaks, but nothing solid. The O-H appears to have come back but its very weak, and the fingerprint region is very messy.

I have just received a GC-MS of both the 'ether compound' and my 'unknown possible claisen product'. They are both pretty much the same thing, but with a slightly different ratio of constituents in the GC - one peak is marginally larger in one GC spec to another. The retention times are the same. Both have a significant amount of the amide starting material in there, so obviously not everything reacted to become the ether product. There are then 3 more GC peaks, they all have the same molecular ion. The rest of the MS doesn't give many other clues, any difference between the intensities is small and its hard to tell what is real.

My questions:
How is my method for the ether compound synthesis? Would you expect a mix of isomer products and what could they be?

Any ideas would be great, thank you.

[EDIT] I should clarify - i'm wondering if as there are 3 forms of the 'ether product' that maybe some of the allyl bromide has attacked the oxygen and bonded there, while in some cases it attached the benzene ring and directly bonded there .. though I wouldn't expect that in these conditions that is the kind of impression i'm getting from the GC-MS. I'm not sure how many more isomers of the ether I can think of.

[EDIT2] After a good night sleep I had a good idea - what if the compound is rearranging in the GC column?

[EDIT3] Yep - had a chat with the GC-MS technician and the inlet is 300oC, which could be hot enough to cause a Claisen rearrangement before analysis. Its unlikely that this happened inside the column cause the peaks are quite sharp. I ran a HPLC with MeOH:H20, with the methanol conc increasing from 50% to 80% gradually and I only got 2 peaks. One i'm assuming is starting material since that came up in the GCMS, and one of my ether NOT rearranging.

[Edited on 28-11-2013 by forsh]

[Edited on 29-11-2013 by forsh]

[Edited on 29-11-2013 by forsh]

Nicodem - 29-11-2013 at 12:22

Quote: Originally posted by forsh  
How is my method for the ether compound synthesis? Would you expect a mix of isomer products and what could they be?

The method is OK, but you don't mention how you monitored your reaction. 3 h at rt might or might not be enough. If you didn't monitor, you would have done better leaving it overnight at 40 °C. That should be enough for a complete conversion with that method. Your work-up however is not exemplary. When you do such reactions and do an extractive work-up you should exploit it to purify the product as much as possible. This means that you should wash with 1M NaOH to remove the unreacted starting material and the phenolic side products. You then wash a couple of times with water to remove residual solvent and water soluble stuff. A proper extractive work-up can save you recrystallizations and columns, if you take enough care.

As for the selectivity of the allylation of phenols, that was just recently discussed in another thread.
Quote:
The first time I stirred the ether compound in dry DCM with zinc as a catalyst - at RT for 3 hours. This just gave me starting material.

I hope you mean zinc chloride? Zinc can't catalyse this rearrangement and if you read an article saying that it does, it is extremely likely just bullshit (and I could pinpoint the country from which the authors published it with quite some accuracy).
Try first with the normal Claisen rearrangement conditions. You can use a high boiling solvent, if you wish, or you can do it solventless provided that you use inert atmosphere and use a bath that can dissipate heat efficiently enough. The thermal Claisen is quite sensitive and the temperature should be controlled to about 200 °C, which is not so trivial at a scale larger than few grams due to the exotherm. Your substrate might not be suitable for the thermal variant in which case the best acid catalsyst is BCl3 in CH2Cl2 as far as I remember (see some review article for more reliable information). Yet, the acid catalysed variant can also lead to a series of side products, so in any case first check the literature.

DJF90 - 29-11-2013 at 14:53

I second most of what Nicodem has to say:

1) You should always monitor the reaction by some means to identify if it has gone to completion or not. I like TLC personally because its quick, relatively cheap and it gives me an idea of a) whether starting material is all gone, b) how many compounds are now present and c) is it possible to separate by column chromatography. At other times you may find HPLC or NMR is more informative, if you have access to these instruments. IR is only really any good if you're reacting a strongly absorbing functionality such as a ketone or azide or nitro group; its fairly evident when theres none of these left. By monitoring the reaction you can make an informed decision as to whether its ready for work up or if it needs to be given more time/heat/reagent.

2) As Nicodem says, zinc metal will not effect the claisen rearrangement. You'll need a lewis acid for that, or attempt it thermally (which would be my first approach). A great solvent for this is supposedly o-dichlorobenzene, which has bp 180 *C.

forsh - 2-12-2013 at 01:57

Thank you for your suggestions.
The paper I read that told me that zinc would catalyst a Claisen rearrangement was Indian, and it wasn't the first such paper to lie to me ...

I'm going to try the reaction again using dichlorobenzene or dimethylaniline as a solvent. Refluxing with rapid stirring, keeping an ice/salt bath nearby. I know TLC is important for monitoring reactions and I will make sure to incorporate this into my technique.

I'm not sure how to get compound out form these high boiling organics though. Normally I would wash and dry the organic and then take it off on rotovap, and I doubt I could get DCB/DMA off to dryness. Would it be a good idea to wash the organic a few times with 1M NaOH to get the salt product in the aqueous, and then slightly acidify it with HCl. I'd get a precipitate of the compound (and probably a lot of NaCl) which I could filter and take up in something like ether? Alternatively I could try adding a solvent that my compound is insoluble in, but that is soluble in DCB - though I don't think there will be very many of these.

If the non-catalysed reaction doesn't work then I have some BF3 someone has given me, and I could try and order in some BCl3. After the reaction I would add water to deactivate the catalyst, separate and dry the organic and then remove the solvent. This would contain some unreacted starting material (allyl ether), but i'm not sure how to remove it. HCl is too weak to cleave the ether. I could try that idea of making the salt of my product, collecting it in aqueous and precipitating it out with HCl before collecting in organic solvent ... but i'm not sure how good of an idea that is.

[Edited on 2-12-2013 by forsh]

forsh - 7-1-2014 at 03:57

Hi,

Just a short update. I am now getting excellent yields in the earlier steps in my synthesis using the methods described and advised here.
The amide is synthesised by stirring the anilide and 1.5 eq pyridine catalyst in ethyl acetate, cooling to -10oC and then slowly adding 0.95 eq acyl chloride. The reaction is left for about half an hour, but I think this is unnecessary as it is very rapid. This is washed with 1M HCl, water and brine, before the solvent is evaporated to give about 80% yield of a 99%+ pure product (GC).
The aromatic allyl ether is formed by refluxing this compound in acetone with 1.5 eq K2CO3 (non-ionising base) and 1.5 eq allyl bromide for 5 hours. To this is added excess water and the product extracted with ether. This is washed with 1M NaOH, water and brine before the solvent is evapourated to give 90% yield of a 99%+ pure product (HPLC, as GC causes Claisen rearrangement).

I am now having trouble with the Claisen rearrangement. As advised i've tried heating both 'neat' and in bromobenzene (the 'best' solvent available at the time) to 180oC and 156oC respectively for 6 hours - both times there was evidence that the rearrangement was occurring, but it is a minor product. I have also tried refluxing in MeCl2 and EtCl3 with AlCl3 which gave me a nasty tar with no desired product. I have been banned from using BF3 and BCl3 for safety reasons :/

I have found light at the end of the tunnel however! White & Folfarth (1969) performed several kinetic studies on the effects of various solvents on the aromatic Claisen rearrangement and have written (with a method) how more polar solvents give a much faster rate of reaction (300 fold faster for p-chlorophenol). I have chosen to do the reaction in ethylene glycol because out of the solvents which are liquid at room temp (making the workup easier), it has the highest k value. The workup will involve cooling to 0oC and adding excess (2x vol.) cold water, then extracting the product in diethyl ether or toluene (I will experimentally determine the best in advance). I know this won't give me the highest purity product, but as long as the reaction proceeds to 90%+ completion then I can lose the unreacted material in the later steps of my synthesis. I'm planning on monitoring the reaction by taking small samples and running them on a UV/Vis spec to monitor the disappearance of one peak from the allyl ether, and appearance of another for the ortho-olefin. I will also use TLC, but it would be nice to find out at the same time if UV/Vis is just as sensitive.

I should be doing this tomorrow (Wednesday) and will have analysed the results by Thursday afternoon (GMT) so will update this thread then. I think this might be a good aromatic Claisen method because ethylene glycol is cheap, readily available and a lot less harmful than some solvents/compound that could be used.

If anyone has anything to add please let me know.

Thanks

[EDIT: SPELLING]

[Edited on 7-1-2014 by forsh]

Dr.Bob - 8-1-2014 at 13:08

Other solvents to consider might be DMSO and DMF. DMSO is more readily available OTC from some "health" stores. Try to keep the EG to a minimum, as it will extract into ether moderately well. Glad to hear that the reactions are working. When you alkylate in acetone, you might want to consider simplifying a bit by removing some acetone first, before workup, as that should make the extraction easier. Even just blowing some air or N2 through the flask will remove a surprising amount of solvent for lower boiling ones.

forsh - 15-1-2014 at 14:23

Hi,
I've had some success with this Claisen method now. I heated 1.5g of my aromatic allyl ether in 50 mL ethylene glycol and heated to between 180-200oC for 5 hours, with a reflux adaptor on top. I didn't run the reaction under N2 in the end cause it was actually cooling down the solution! But I did flush the flask periodically, especially when taking a sample.

The solution started off golden brown/yellow and after about an hour turned red/orange, then continued to get darker so that by 5 hours it was deep red. I also ran a UV/vis spec at the beginning of the reaction and every hour, this showed a big shoulder peak at around 260 nm which reduced in intensity over time.

I ran an HPLC (MeOH/water) and this showed the disappearance of my starting material peak over time, and the appearance of two other peaks, which i'm guessing are the two different rearrangements as I have two substituents on my benzene ring meta to each other. Though on the HPLC the wavelengths which the two peaks absorb are quite different, which usually indicates two very different compounds - I wonder if one of the peaks is Et(OH)2 and the other my compound selective for one rearrangement, OR it could be that one of the rearrangements formed a ring.

I haven't extracted the product yet. I added 150mL water and a precipitate rapidly formed, but this was too fine to filter even with my finest paper. I ran out of time so didn't have time to progress, but I'm planning on extracting my product with ether, then just washing my ether several times with water to remove the Et(OH)2. I was wondering - if I add salt to the water/Et(OH)2 to try and force more of my product into the ether layer, will it also move some of the Et(OH)2 because that also is a compound dissolved in water? I guess it would, but to a much lesser degree so I should just slowly add salt until all of the colour is in the ether?

Cheers.