Hey! I have not encountered this problem I can tell you how I would approach the situation....

In most writeups you got uniform blackish mess you simply extract with toluene or something non-polar.
In reality, you got 2 non-miscible layers you can't extract with toluene since darn thing mixs freely with layer that contains goods!

The best thing to do will be to pour the top phase off, or separate in a sep funnel. And save the good phase and discard the sludge.

I have found that alcohols can pull xylene, toluene into aqueous phase. If you have any alcohol present, let it sit somewhere under warm heat until it is pretty much all gone. Then add water to replace what has just evaporated, fix pH again with NaOH if it needs it and extract with toluene.

If both of those phases contain product you need to distill the alcohol off first, then add water and NaOH if needed to fix pH. But I am sure you will need to once you get to the stage where you say "you got 2 non-miscible layers" decant the alcohol solution off. Then distill the alcohol down, and add water/NaOH to fix pH which can then be extracted with toluene.

[edit; If you are brave or short of time, you could just try diluting the alcohol down. To a point where it will not pull the xylene in, but I have found the aqueous phase always still smells of toluene when you do that....You could try applying heat and monitor the temp and try to get as much MeOH out as possible (the temp will stop at 65*C when just methanol is leaving) which may let the toluene out to form its own phase if you can remove enough.]

You are trying to isolate the amine, which is what hopefully results from the reductive alkylation of the imine    You need to basify what's left, then distill the alcohol out and extract the reaction product with a non-polar solvent or preferably add water and steam distill the amine directly from the aluminum hydroxide sludge    An acid base extraction will further purify one's amine product, then gas it with HCl to form the salt    One can then recrystallize the salt if desired  

Most write ups aren't complete, that's why these problems aren't mentioned    One can also use Celite and extract the sludge with methanol; but steam distillation produces the best results since it eliminates the mechanical losses   It's important to understand what's going on, and then one can come up with solutions 

You should be able to steam distill by increasing the temperature and adding water after the methanol/methylamine is distilled out   You can recycle the methylamine/methanol    Clearly you need to study organic laboratory techniques more seriously, then you will understand what you have to do    The results come from the effort applied  

I found no sane scientific explanation for this!

So obviously you were not happy with my first post which if read would have fixed your problem and answered your question? There is no mystery here for me anyway, as I have recrystallized compounds before from ethanol/xylene 60:40 and that is a homogenous solution (no phases) mixes together. I have also tried to work up products from ethanol before and the exact same thing happened, if you were not going to accept the answers why did you post the question? Lugh just gave you the best advice you will get regarding working up this reaction.....

Ethanol is a versatile solvent, miscible with water and with many organic solvents, including acetic acid, acetone, benzene, carbon tetrachloride, chloroform, diethyl ether, ethylene glycol, glycerol, nitromethane, pyridine, and toluene.
http://en.wikipedia.org/wiki/Ethanol

Ok to prove this to yourself. Take a small test tube, add 10ml of methanol. Then add 5ml of toluene and stir, you will note the two solutions mix together in all proportions (miscible).

Another with 7ml methanol, 3ml water and 5ml toluene, surprise surprise they mix together. And you have just emulated your work up solution!

Now take another test tube and add 10ml of water and add 5ml of toluene, the solution will form two phases. Magic? not really.

The differences in techniques will be the amount of water added to the solution, the people that suffer this problem are adding more methanol or less water with their NaOH than the write-ups they are following suggest. This is the most logical conclusion for why different people are experiencing different outcomes, also the grade of methanol would play big part. The people with better (lab) grades of methanol would be more likely to encounter this problem. As there methanol would be of a higher concentration, if they were then to use less water when making the NaOH solution they could encounter this problem.

As I stated in the first post, you can simply add more water until two phases separate. But I have found in those cases the aqueous phase always holds on to some toluene unless you add a lot of water which is not usually ideal. So reducing the methanol down about half and replace what evaporated with water, methanol  boils under 100*C so you can literally just put the beaker on warm heat and come back a couple of hours later and its done. Will return a slightly better yield.

Answer; The concentration of methanol or ethanol is too high, and is pulling the toluene into the aqueous phase (as stated). If you read the first post I made, I have already suggested an appropriate work up and now have provided a sane explanation....

Can you just tell me what ratio of H2O:EtOH and H2O:MeOH will prevent two layers to mix at all?!

I actually cant! That is something you would have to determine in a small test tube with the reagents you are using. I can say if you use EtOH it can be removed pretty quickly under gentle heat, sitting it aside on the heat for 1/2 hour to an hour will remove enough so is not like its a very big hassle to do that. I would just be careful not heat the reaction mixture too much, so separating the alcohol layer first would be better then reduce volume.

To be clear; When you said

"In most writeups you got uniform blackish mess you simply extract with toluene or something non-polar.
In reality, you got 2 non-miscible layers"

I would separate those two layers, keep the alcohol layer with the goods in it and discard the other! Then reduce alcohol right down to the ratio you worked out above, before adding NaOH.

I was a bit thrown, when you said "I found no sane scientific explanation for this!" I felt as if you were talking about my explanation! But you obviously were not, and I do wish you the best of luck with this procedure. You know we also like a good write-up around here, so if you find the time we will always be interested in seeing/hearing about your procedure and how it went. Once again good luck!

For God sake no Naf1,of course not!
I was referring to the mine 3-day long browsing on the net.
Your explanation actually helped a lot!
To be fair,some of your explainations on other threads helped so much
I did't need to open new threads on some puzzling situations!

And of course I will try to do complete write-up of mine own.
I will try to cover all the aspects and details other people didn't in one write-up,
especially synthesis of hard to come by chemicals.

LOL now we misunderstood again, it is probably because my lousy English grammar...
It is becoming quite a Monty Python situation...

I would separate those two layers, keep the alcohol layer with the goods in it and discard the other!
Then reduce alcohol right down to the ratio you worked out above, before adding NaOH.

You probably misunderstood, layers I was referring to separated because of NaOH addition after Al/Hg step was over
and xylene freely mixed with upper layer.
So I can't reduce this alcohol layer with the goods prior adding the NaOH when this layer
formed BECAUSE of NaOH addition in the first place.

As long as I understood,this is same as saying you are going to make your own father when you grow older.
(Don't want to sounds offensive at all, but it is same logical situation!)

Another thing that confuses me is:

...so separating the alcohol layer first would be better then reduce volume.

I was trying to say it is best to evaporate off some EtOH BEFORE basification that resulted in layer separation,
and "volume reduction" = "EtOH evaporation" in this case, right!?
So it hopefully wont separate if I reduce volume of EtOH and replace it with water in before NaOH addition in the first place.

To make it less complicate to understand here is what I plan to do:

"Make mixture of 150ml EtOH,80mg HgCl2 and 8g of Al,
ketone/nitro/EtOH mixture was added drop vise after amalgation started,
after addition reaction was left to stir for 3h under reflux (actually no need since temp.
was under 50°C all the time and no reflux was noticed, it just helped keep nasty smell off...)
and than cca. 120ml of EtOH was vacuum distilled of rxn contest and cca. 120ml of dH2O was added instead.
Now this rxn contest is basified with same volume of 30% NaOH, hopefully it won't separate in two
layers. (unless I am enchanted in some way) Only separation that will occur (hopefully) is separation of this uniform
basified layer and toluene containig goods when I add it."

You understand now,is this O.K.?!

Lugh, unfortunately this link says "forbidden 404", I tried with 2 browsers.

The link worked for me!

You understand now,is this O.K.?!

Yes I like that a lot better

Now this rxn contest is basified with same volume of 30% NaOH, hopefully it won't separate in two
layers.

Just quickly if you have a lot of product in there you will see it separate as an oil (freebase), even if you have not got much you will still see oil beads of freebase separate. Once you add the toluene it will be taken up (dissolved) in the toluene, and then you will have two layers after adding toluene = water/toluene.

separating the upper layer discarding the bottom, then evaporate methanol to impure amine, could be done like Ritter?
 dissolve the impure amine with some DCM then wash several times with dh2o, nacl, sodium carbonate ?? would need A/B before crystalzation? i ask that since Ritter, MM, Gonzo, dont do A/B.

peace