I need your help because I've got a lot of problem with the Ullmann reaction...
200ml of anhydrous HPLC quality EtOH is poured in a 300ml RBF fitted with a thermometer and a reflux condenser.
19.8g (860 mmol) of metallic sodium is cutted to sausage slices and added to the EtOH. Hydrogen is envolved. At the end the solution is much more
viscous and the ethylate solution is heated to allow the rest of the sodium to dissolve. Meanwhile 4.6 g (20.6 mmol) of CuBr2 is dissolved in 10ml H2O
and the dark green solution is heated at 50°C. 1.31g (11 mmol) of Na2SO3 and 0.8 g (20 mmol) of NaOH are dissolved in 25 ml H2O. The sulfite solution
is gently added to the Cu+2. The white precipitate is filtered on a buchner, washed with H2O and EtOH.
Error 1 : I started adding the CuBr (0.5g) in the EtOH before the AcOEt... The solution became red and some copper
precipitated out...
4.6 g (52 mmol, 3/5 eq) AcOEt is added to the mixture, then the rest of the CuBr. 20.0 g (87 mmol) of 5-Bromovanillin (beige powder) is added with a
spatula. But the sodium phenolate is not really soluble in the ethanol. 50ml of Ethanol is added agin but there still is a small amount of
bromovanillin at the bottom of the flask.
The heater is replaced and the mixture is refluxed overnight.
22h after, the brown solution is poured on 500ml ice/water solution containing 500 mmol of H2SO4. There is a lot of black gunk in the flask...
Error 2 : I think let the solution being basic would be better because phenolate are soluble in water therefore
Impurities and ethyl acetoacetate can be removed with DCM.
I need to point out that the new water mixture smells nice. It remember the smell of syringaldehyde, it gives the hint the reaction worked. The
solution is extracted with 5x 100ml of DCM.
Error 3 Ethyl acetate would be a better choice IMO
The DCM extracts are filtered because of the copper salt and metallic copper contaminent. DCM is boiled off and the residue is put in the fridge to
give a black oil which dont cristallise. This oil is heated with 100ml 10% MeOH in H2O. After reflux the white methanol solution is separated from the
residual black gunk and put in the fridge. But the Ethoxyvanillin didnt cristallise it was just a sticky amber oil at the bottom...
The gunk was refluxed with 300ml petroleum (bp around 120°C). The light orange petroleum solution is separated and put in the fridge. Same orange oil
is recovered... This oil is very sticky like honey. I dicided to dissolved it in 10% NaOH - very soluble ! 2 ml of this white phenolate solution is
pipetted to a test tube and conc. HCl is added. After cooled to -5°C green crystals appeared ! i'll try to do the same thing with the whole solution
but I only recovered the same sticky oil...
When I started to clean the glassware i added acetone to the beaker where there is a rest of the orange oil green crystal appered !!! WTF ! So I
immediatly tried to add acetone to a bigger amout of the oil but acetone simply dissolved it ...
What happend with my reaction ? Did copper metal fucked it up ? What does black gunk contain ? Polyphenolates ?
Another question :
Could the aldehyde be destroyed via Cannizarro reaction ? I've thought the benzaldehyde can be condensed with NaCH2-COEt, no ?
thanks for the help...
turd - 24-7-2010 at 08:29
Hi Methyl.Magic,
did you have success with this method for syringaldehyde? I believe(!) that Nicodem said that the EtOAc method probably does not work. Why not try
with DMF at higher temperatures?gardenvariety - 24-7-2010 at 09:06
Would DMSO be an equivalent solvent for DMF in this reaction? High bp, dipolar aprotic, solutes benzaldehydes, etc.zed - 24-7-2010 at 10:46
Strange territory for me.
Perhaps if you started by Brominating Ethyl-Vanillin....things would go smoothly.
From that point, the reactions could be run in the usual manner (via methylation), using the usual solvents.
Same end product, but you get to travel down a much more familiar path.
Nicodem - 25-7-2010 at 07:31
Start using references! How is anybody to know what is being discussed if the reference to the pertaining paper is not included?
did you have success with this method for syringaldehyde? I believe(!) that Nicodem said that the EtOAc method probably does not work. Why not try
with DMF at higher temperatures?
Actually it was Ullmann who said he had no success with it:
Personal attemps by using the EtOAc protocol failed.
Assuming the authors of the paper were indeed honest, it is still important to note that for the putative acetylacetonate ligand to form, the reaction
mixture must be completely dry, because only a small amount of the preligand is used. If there is any water/hydroxide present it will hydrolyse the
ethyl acetate instead. Ligands such as N,N-dimethylglycine or various diamines that are more commonly used in the Ullmann condensation are insensitive
to hydrolysis so it might be best to try any of these for the low temperature Ullmann condensations (like at the bp of ethanol).Methyl.Magic - 26-7-2010 at 14:07
Thanks for your help.
Nicodem, you talked about the presence of water in the reaction but my reaction was water-free because I used metalic sodium. By the way, water could
be rest in the CuBr when I made it ...
Nicodem, do you think adding ethylenediamine as copper ligant would help it to be soluble in EtOH ? And (en) complexes would be less prone to
dismutation. Although MeONa is a better nucleophile, is there any risk to add an (2-aminoethyl)amino instead of methoxy to the phenol ortho position ?
thanksFleaker - 26-7-2010 at 15:10
Methyl.Magic,
The water will come from not drying your cuprous bromide over P2O5 and possibly from your ethyl acetate. Also, you must use anhydrous (m)ethanol.
Anyway, I've been told by quite a competent chemist that the reaction as done in ethyl acetate does not work, as they had attempted it many times. Do
not despair!
This chemist also told me this much:
You can brominate vanillin in nigh on quantitative yield (90%+) in methanol, glacial acetic acid or in methanol-water with bromine added to it.
Methanol is easiest to remove. After you're done, the powder should be a very slight yellow--the beige powder you have is impure and contains residual
bromine. The pure 5-bromovanillin will still smell of faintly vanillin but have a spicier note and a more burning sensation upon the tongue. This
substrate is hard to dry, you must spread it thin on a large baking dish and place in a vacuum oven, and let dry for two or three days. When you
dissolve it into DMF, it will be quite endothermic and you will get a deep purple-orange solution.
This chemist went on to say that this reaction using absolutely anhydrous 3A-dried DMF as the solvent does indeed work and that they've done it many
times, with yields either being 30% or below if air leaked in (with workup being a nightmare), or 80-85% if done right. Purportedly it scales well
into the kilograms. The chief issues that I remember that chemist mentioning are: 1.) phenolates are exceedingly delicate at the reaction temperature
and conditions with respect to oxygen so you must use a heavy argon atmosphere lest you deal with a horrible purple tar, 2.) if there is excess water
in your DMF, it will hydrolyze and when you go to distill it you will get many byproducts, namely fishy dimethylamine and your labmates will dislike
you, 3.) it is advisable to remove half to two-thirds the portion of DMF by high vacuum distillation to make for easier workup, 4.) it is wise to
cool everything to 0*C and add nitrogen-degassed ice water, and keep the mixture under argon whilst slowly adding the HCl solution. Do NOT ever expose
it to air until the whole thing has stirred and been made acidic. 5.) This chemist claimed after HCl-neutralization of the (m)ethoxide, the whole
reaction flask will be filled with a grainy mocha/coffee-coloured mass, and that the pH should be about 4. This mixture is then vacuum filtered and
rinsed with ice cold water till the acid is rinsed off. The filtrate is yellow, and the product is a crystalline beige-yellow powder. Do NOT attempt
to extract or dissolve anything with ethyl acetate; instead, dissolve this product in boiling toluene (the toluene-water azeotrope conveniently dries
your product!) and filter through a fine frit to remove copper oxides (or leave them, it will not affect further steps). Then evaporate/ice bath the
toluene and your product will crystallize out leaving a purplish supernatant containing polyphenolics/tar (which may be concentrated to give an
inferior quality second crop). The product then has the literature melting point and is a beautiful yellow needle mass that supposedly smells quite
good and is quite prone to stain your hands nitric acid-yellow! This is true for the ethyl or methyl substituted product--toluene is the best
crystallization solvent for this reaction product.
That very same chemist also said that one could substitute NMP for DMF and the yields are 90%+ if care is taken to avoid atmosphere exposure and use a
trapped rotary vane to pull off the NMP. That chemist advised against storing your DMF for long over sieve, as zeolite is somewhat acidic and will
slightly decompose it; your DMF should never smell fishy if quite pure.
Keep in mind that this was many years ago and I'm no organic chemist. All I can recall is what gave that person trouble when they did this.
Good luck with it!
[Edited on 26-7-2010 by Fleaker]
[Edited on 26-7-2010 by Fleaker]Methyl.Magic - 27-7-2010 at 00:30
Hey Fleaker, thank you very much for your well-described report. It really helped me to understand the errors I've done.
I have a few more question for you :
1) what is the best R-ONa/Bromovanilline ratio ?
2) what is the reaction time ?
3) what is the temperature when NMP is used ? Because I've heard high temperature decrease the reaction time. And NMP has a higher bp, therefore we
can work at higher temperatue...
4) What about DMSO ? It will be less prone to hydrolise than amides.
Tomorrow I'll do a secon attempt with NMP, I think I've got a rest of this precious solvent, something like 100 ml ...
Thanks again Fleaker. Now I know what was my enemy : Oxygen !
Nicodem, you talked about the presence of water in the reaction but my reaction was water-free because I used metalic sodium. By the way, water could
be rest in the CuBr when I made it ...
I guess water would generate hydroxyl ions according to
Na + H2O --> Na+ + OH- + 1/2H2
or
H2O + MeO- <--> OH- + MeOH
And those hydroxyl ions would hydrolize your ester or amide.
Actually it works quite good (at least with MeOH). I've done it under very uncontrolled conditions (solvents out of the bottle, no inert atmosphere)
and got very reasonable yields. Not the >80% of hest, but still. See also Ullmann's excellent posts on the matter: https://www.sciencemadness.org/whisper/viewthread.php?tid=70...
Actually my problem was that I got unreacted 5-bromo-compound and for some substrates it has a really nasty tendency to co-crystallize with the
5-methoxy-compound in later steps and cannot easily be removed by distillation. I'm not making this up - I have irrefutable proof. So probably better
use the workup proposed by Ullmann.Fleaker - 27-7-2010 at 01:46
Sorry methyl.magic,
I've never personally attempted it before so I can't give specifics. I suppose you could post up the original procedure/reference and people here
could have a crack at it. Sadly, I don't remember any other details, I merely thought it was an interesting and colourful reaction but I do not even
know if what this chemist said is the truth.
The old fellow I knew that did this particular chemistry worked/works for a flavour company. I met that person perchance at a bar and we had a long
talk about many fine fragrances and related compounds and their potential misuses. Funny story indeed how we got on that topic. Anyway, years later,
I was making a phenoxide (phenol and NaH) to attach via nucleophilic substitution to a phosphazene and I inadvertently let air in, and sure enough, it
made a weird mauve/purple colour and polymerised to gunk. I then remembered that I'd been warned once about excluding air at all costs when working
with phenolic substrates and the purple tar I saw in my flask reminded me of that evening.
Anyway, good luck with it! I always enjoy reading your procedures. I wish I could be of more help, but I think that the present advice will suffice. Picric-A - 2-4-2011 at 13:29
What would the yields be like if i did the following;
-Dissolve 5 moles of NaOH in excess methanol (anhydrous) and heat to dissolve.
-Dissolve 1 mole of 5-bromovanillin in the methanoic NaOH, add around 1g of CuCl catalyst
-Reflux for 2-3hours (or more?) under inert gas (butane) and with moisture trap (P2O5)
Would i get an appreciabe amount of the 5-methoxy product? i dont have any DMF or other high b.p. solvents so i need to find a way thi rxn can work at
standard pressure and common solvents.
/thankschemrox - 3-4-2011 at 10:01
@methyl-magic what is the original ref? It could be helpful to let us read it. I assume you have...
