PPA via alanine and BzH, rxn end indicators?

[DRIVEN]

Hello everyone,

D was wondering if anyone out there has practical experience with the Akabori reaction alanine and BnZ to form PPA. In particular, how long to run the reaction is in question.  IOC had performed a run and mentioned that the reaction be terminated after "fizzing" (or release of CO2) had stopped. 

At 140 degrees C the reaction boils so it is difficult for D to distinguish between "fizzing" and boiling. D figured that by perhaps removing the heat, the interference of boiling could be removed however this will not work as  the reaction depends on the addition of heat.

The reaction has run about 3 hrs. The solid in the mixture, that was once alanine (fine and granular)in the beginning of the reaction, has now become a fluffy white solid which does not resemble the original alanine appearance. Is it possible to regard this as another indicator that the reaction has completed? As well the solution has progressed from clear, to pee yellow, to orange, to red-orange, to deep red.

Any input is much appreciated!

Take care

DRIVEN   

[Aurelius]

They will want this information, so I'm saving you the time. 



"PPA via Akabori of benzaldehyde and alanine.

1) 100g of benzaldehyde was placed in a 250 ml rbf and 40g of dl-alanine (molecular bio grade) was added with stirring.
2) The mixture refluxed at 140 degrees C with heavy stirring for 3 hrs.  The reaction mixture went through notable characterisitics.   When the temperature of the liquid reached 115 deg, the mixture evolved CO2, emanating particularily around the alanine and condensate was apparent in the condenser.  At 140 deg, T=0  the solution was lime green.  At T=15min, pee yellow, at T=30 min, deep orange, T=40 min, red-orange,
3) At T= 3hrs, the solution was a deep red colour. Another solution phase was apparent and was of fluffy consistency and crème in colour. It was unclear at this point whether the reaction was finished.  Previous reports have suggested ceasing of gas evolution as an end point however this was difficult to tell since the solution was boiling as well. 
4) The post-reaction was cooled to 4 degrees and slowly with good stirring 30% HCl was added until the solution was ph= 2.
5) The mixture was vacuum  filtered twice to remove insolubles. There was about 1 tablespoon. (unreacted alanine?).
6) The filtrate was washed with 2 X 100ml of DCM. Much of the red colour present in the filtrate moved into the DCM. The aqueous took on a light, clear yellow.
7) The aqueous was basified to ph=13 slowly with 30% NaOH.  Solution progressed from yellow to cloudy white with a deep orangy oil falling to the bottom of the beaker. Only about 5-10 ml of this oil was present.
The basified solution was extracted with 3 X 50 ml of DCM, extracts combined and fractionally distilled at atmospheric pressure.
9) After the DCM had come off, the heat had to be cranked up quite a bit for the next fraction (PPA?) to come over at 85 deg C. ( quite a bit off from bp of PPA being 100 deg C. as published?). It condensed much like water and had a strong amine smell to it. Only 3g of distillate came over.
9) Upon continued heating the remaining liquid in the boiling flask sublimed into a deep red chunk. No other fractions besides the one at 85 deg C. came over.
10) As a check, the original DCM washes were back extracted with acidic aqueous, basified, extracted with DCM and distilled. The 85 Deg C fraction was absent and once again the remnants in the boiling flask sublimed. Urgh!"  --Driven


These are the details for the specific reaction that Driven is referring to.

[DRIVEN]

Commentary on boiling point of PPA produced from above experimental.

The BP of norpseudoephedrine base : 77.5-78°
The BP of norephedrine base: 100-101

The bp of the PPA from the above experimental came over at a range of 75-85 deg C.  Perhaps its possible that the product is a mixure of norpseudoephedrine and norephedrine.

This would make sence because the carbonyl of the benzaldehyde is planar therefore attack would be 50/50 from either side.  d,l(R,S)-alanine is used therefore the possible isomers of PPA are (R,S) (R,R),(S,R),(S,S) and:

(R,R),(S,S): norpseudoephedrine
(R,S),(S,R): norephedrine

At around T=3hrs their is a distinctive second phase (aqueous) produced in the reaction mixture. D is not sure of the reaction mechanism of the akabori reaction but if PPA is produced by an intermediate that requires reaction with the other phase then this could account for the low yield of the reaction. Especially since the phase separation occurred around the time the reaction was stopped.

Moreover SPISSHAK

Post 356191

(SPISSHAK: "Possible 'easy' PPA", Methods Discourse) reffered to a patent that used a variation of an akabori type rxn where alanine and benzaldehyde are supposed to give a small yield of PPA with the use of a phase transfer catalyst. Is the translated version of the original patent out there?

D is contemplating a low scale run with some HTAB (hexadecyl trimethyl ammonium bromide )as a PTC, killer stirring and a long reaction time (8hrs).

Any thoughts, anyone? Wouter, where art thou?

Driven

[Aurelius]

Post 422067 (missing)

(Aurelius: "Ephedrine Compilation", Stimulants)

This has an entire section devoted to the Akabori.  I suggest you follow ALL the links and read them carefully.  Print out each of the patents and read them multiple times.  One of the patents includes the mechanism for this reaction type.  It will also give the reaction conditions you're looking for.

[DRIVEN]

Ok lets see if I can get this.

According to post

Post 423224 (missing)

(Aurelius: "US Patent 4501919 (Akabori)", Stimulants) benzaldehyde can be condensed with glycine in an aqueous sodium hydroxide solution yielding first a benzylidene phenylserine and, after acid hydrolysis, the desired phenyl serine.

Since this reaction proceeds through the formation of a schiff base does this ultimately mean that only the alpha carbon (as opposted to the carbon of the carboxyl group) of glycine will form a bond with the carbon of the aldehyde group of the benzaldehyde.

If this is the case then would a condensation of benzaldehyde with alanine under the same conditions indicated above result in benzylidene alpha-methyl phenylserine which when acid hydrolyzed would result in alpha-methyl phenylserine.  If this is the case, to produce phenylpropanolamine from this compound a decarboxylation would have to be performed? Do I have this right or am I way off?

Thank you for your patience

Driven

[roger2003]

[roger2003]

@ Aurelius

I think, your Akabori synthesis with benzaldehyde and alanine must first perform a “Schiffsche Base”  ( 1:1 mol ratio benzaldehyde and alanine)  and than with a further mol of benzaldehyde  to complete.

Older syntheses (DE Pat 960722) are carried out in the first step in  solutions of CaO in water.
The “Schiffsche Base” is than filtert off and in the next step was formed the Phenylserine in methanol /water/CaO

[Aurelius]

I don't remember exactly, but yes, I believe the Schiff base is formed first, but the procedure above was not mine- it was written by DRIVEN to me in a pm.  I just thought it would be easier for me to copy/paste than to have him re-type that mess.  After all, everybody knows rxn details are demanded around here for a good response. 

one of the patents ( don't rememer which one) has the mechanism- it think my compilation points it out though.