uemura (Hive Photographer)
03-25-02 14:16
No 287617
      Melting point extrapolations  Bookmark   

Suppose a bee runs a reaction which produces two (exactly two) known products A and B in unknown percentages pa and pb (pa+pb=100). For both substances the melting points ma and mb are known.
Is it now valid - or when would it bee valid - to linear extrapolate from the measured melting point mp of the mixture to the percentage of the substances received, ie. pb = (mp-ma)/(mb-ma).

To bee more specific, say you get a mix of vanillin (mp=81) and hydroxyvanillin (mp=131) with an melting point of 124. Could you assume that you have now (124-81)/(131-81) = 86% hydroxyvanillin in the mixture?

Thanks in advance
Carpe Diem
(Hive Bee)
03-25-02 15:06
No 287627
      No, It's not that easy The most importent  Bookmark   

No, It's not that easy

The most importent thing abouth meltingpoints is not when but how it melts. It has to bee sharp, Iff all the xtals melts within 0.2c (say 81-81.2C) its werrye pure, but iff you have a broad meltingpoint (say 75-103C) your product is quite dirty. You will always get a mp. DEpresion when you mix two products, so if you mix vaniline and hydroxyvanilin you wil get a mp starting at app. 75C and ending at say 100C. So meltingpoints tell you two things. 1: is it the right stuff I have (melts at the given temp) and 2:how pure is it (sharp or broad mp.) Personal i prefer TLC at anny time (but you will ofcource need som of your target molecule for that)
(Hive Photographer)
03-25-02 15:33
No 287632
      More details  Bookmark   

Thanks so far hest.

Now Uemura gives more details for comments. The product was prepared from bromovanilline, refluxing with fresh precipated copper and NaOH for 20 hours. A first sample was xtalised from DCM and an mp was determined.

Uemura got a melting point range from 123-125 which should indicate a reasonable pure substance. The literatue says the above rxn produces something like 85% hydroxyvanillin and 15% vanillin. Uemura wants to know if he has hydroxyvanillin produced (he has no reference material for TLC). What do you think is the substance xtalised?

Carpe Diem
(Distinctive Doe)
03-25-02 16:54
No 287645
      3,4-Dihydroxy-5-methoxybenzaldehyde  Bookmark   

Is this the procedure you were using?
Patent US3855306
Bromination of vanillin with Br in 48% HBr gave 99.4% 5-bromovanillin (I).  Hydrolysis of I in NaOH-H2O in the presence of powd. Cu gave 83.3% 5-hydroxyvanillin (II).  Methylation of II with Me2SO4 in Me2CO in the presence of powd. Na2CO3.H2O gave 90% 3,4,5-trimethoxybenzaldehyde of purity 99.3%, useful as intermediate for pharmaceuticals.

Well the melting point of 5-OH-vanillin is 132.5-134.0C so i would bet your produce is fairly pure. 

this is the patents extraction procedure
The reaction product, 5-hydroxyvanillin, is suitably isolated from the reaction mixture by a procedure of hot liquid-liquid extraction. In this procedure, the hot reaction mixture, after acidification with a strong acid, preferably a mineral acid such as concentrated hydrochloric acid, is continuously extracted with a water immiscible organic solvent at an elevated temperature, preferably between about 60 DEG and about 100 DEG. Suitable solvents for the hot liquid-liquid extraction include hydrocarbons, e.g., benzene or toluene; esters, e.g., ethyl acetate; and so forth. An especially preferred solvent for the hot liquid-liquid extraction is toluene.

The reaction product extracted into the organic solvent in the above step is then purified by recrystallization. An especially preferred solvent for recrystallization is toluene. In a preferred procedure, toluene is used as the extracting solvent in the hot liquid-liquid extraction and also serves by cooling of the hot extract, as a recrystallization solvent.

Here is a procedure for "next time"

Zikan, Viktor; Roubinek, Frantisek; Eichler, Jaroslav.
Czech.  (1984),     5 pp.  CODEN: CZXXA9 
CS  216633  B  19841001  Patent  written in Czech.  
Alk. solvolysis of 3,4,5-Br(HO)(MeO)C6H2CHO (I) yielding the title compd. (II), is catalyzed with CuO which is optionally prepd. in situ.  Thus, a soln. of 122.4 g NaOH in 750 mL water was treated dropwise at 50 with 4 g CuSO4 in 40 mL water, 100 g I was added at 70 and the mixt. was refluxed 27 h under H.  The resulting brown-green soln. was made acidic, cleared with C and worked up as usual to give 91.38% II of 99.5% purity. 

Those who give up essential liberties for temporary safety deserve neither liberty nor safety
(Official Hive Translator)
03-25-02 17:11
No 287647
      An idea.  Bookmark   

Uemura, i have no clue about mp and all that stuffsmile, however, i want to suggest to - just for the fun of it (and the future generations' sake) - try to dissolve the aldehyde mixture in 50% K2CO3 solution, as suggested in Post 199293 (Antoncho: "What about separation?", Chemistry Discourse), referencing the data from the article found this summer by Half-a-Pint - searching for a way to separate vanillin from syringaldehyde, the additional info we stumbled on a way to separate p-hydroxyBA from vanillin:

T   K2CO3 g/l   Vanill.,g/l  p-OH-BA,g/l  V:pHBA ratio

6    320         <5            114         >23

12   400         <5            114         >23

20   320         10            171         17

22   500          2            114         57

28   400          <10          171         >17

36   500           3           171         57

So - Why don't you try this, if you got some potash (and pioneering mood) at hand? Of course, it's hard to say exactly how OH-vanilin's gonna beehave at those circumstances, but could bee worth a try.

Actually, IIRC, it is iodovanillin method that produces 85:15 product/vanillin ratio - hydroxyV made via bromoV route should bee much purer.

Apart from all that, i am immensely glad to see you succeed!! Excellent! My sincere congrats!cool


(Hive Photographer)
03-25-02 17:17
No 287652
      '3,4-Dihydroxy-5-methoxybenzaldehyde'  Bookmark   

Hi foxy!

Uemura used bits and pieces from Rhodiums site and from threads on this topic. The bromovanilline rxn was done with insitu bromine generation, works very easy and high yielding.

The hydrolysis is a bit 'lengthy'. And Uemura hasn't enough argon to protect the reflux mix for 20hours. More difficult is the workup! Extraction with EtOAc gives a very bad emulsion. Also, hot liquid extraction. No idea of the proper equipment needed. What Uemura did was vacuum evap of a part of rxn mix and soxleth the drak-grey powder he got. Works fine with DCM, which shouldn't dissolve any unorganic salts present. Physical appearance of the product is exactly as described in the Merck.

BTW: what is the 'H' in ...the mixt. was refluxed 27 h under H.  The resulting...

Carpe Diem
(Hive Photographer)
03-25-02 19:27
No 287694
      Seperation idea  Bookmark   

Hello Antoncho,

almost as usual, our posts crossed. Thanks for congrat. Uemura took now this route because of the massive failure on the copper catalised methoxylation. This one IS NOT watersensitive smile.

Disadvantage going via the 5-Hydroxyvanilline is:

1) You have to seperate from vanillin, but Uemura may consider your idea. Column chromatography isn't setup easily either.
2) You need twice the amount of a methylation agent to arrive at the 345 TMbenz.
3) Workup is a pain in the a.. neck! Uemura hasn't still determined the yields, because he tried different extraction techniques of part of the rxn mix. In addition, the 5-hydroxyvanillin seems to be badly soluable in DCM and perhaps other solvents as well. The product xtalised during soxleth in the boiling DCM. After cooling very few additional cristalls appeared.

Uemura will report on further experiences, at the moment he is happy beeing a step further.
Carpe Diem
(Distinctive Doe)
03-26-02 00:24
No 287807
      One advantage of this route  Bookmark   

You can also make 3-methoxy-4,5-methylenedioxybenzaldehyde!!

In the patent they stress recrystallization from Toluene.

As to the H, question.  I can only assume they mean Hydrogen or it could bee a typo and they meant Nitrogen.  Who knows?  That patent is not online so its tough to verify.

Those who give up essential liberties for temporary safety deserve neither liberty nor safety
(Official Hive Translator)
03-26-02 07:49
No 287977
      H=Helium??  Bookmark   

(Chief Bee)
03-26-02 09:25
No 288011
      Protecting gas  Bookmark   

H2, N2, He, Ar - whatever. The point to stress is to use an inert gas, at least one that is not oxidizing like the atmospherical oxygen. The reason a protective gas blanket is used is the risk of oxidation of the aldehydes to the benzoic acids by oxygen. Using inerts like He/Ar/N2 is great, and H2 should work too as it is reducing and not oxidizing, but the other gases are cheaper, easier to use and constitutes no explosion hazard.
(Distinctive Doe)
03-26-02 11:06
No 288045
      Antoncho your right  Bookmark   

I'm pretty sure it is helium that the H stands for.

There is one nice thing about Hydrogen, its really easy to generate it yourself with H2SO4 + metal.  It could get a bit old generateing for 20 hours, however a good answer to that is a syringe pump.  Those things are SUPER handy for controling your addition rates!  And really old used ones can bee had cheap since everyone wants a nice new digital.

Those who give up essential liberties for temporary safety deserve neither liberty nor safety
(Chief Bee)
03-26-02 11:13
No 288048
      syringe pump  Bookmark   

How much does a used syringe pump go for? PM me with suggested sources.
(Hive Photographer)
03-27-02 11:04
No 288475
      Protecting atmosphere  Bookmark   


it seems we are discussing about the building block of the universe: hydrogen and helium. But aren't they also used to fill up baloons because the are so light, that the easily escape. Does it make sense to use them as a protective atmosphere? Perhaps, if you blow in enough hydrogen, but where does it go? Out of the cendenser into your OTC-non explosive protected vent...

Isn't a mixture under fully reflux not self-protected by the solvent vopour? You can do Grignards without protecting atmosphere because of the ether blanket (even at rt).

Wouldn't it bee sufficient to do:
1) Fill in the stuff. Then push out the air in your apparatur with the heavy argon, keep argon stream up the reflux start. Shut down argon
2) When rxn is finished, release again argon stream up the rxn is cooled and you do the workup.

Other idea: Connect consender with two wash-bottles (so no suck-back can occur). Fill in a solvent which sucks up all oxygen (e.g. alkalic pyrrogallol or tin-oxy??? (*)) Then you have kind of closed apparatus. Pressure instabilities are handled by the two bottles.

What about these two ideas?

(*) Forget this one.
(Official Hive Translator)
03-27-02 14:20
No 288523
      Butane  Bookmark   

Well, idunno. SWIM just uses a lighter-refill ballon, which, combined w/a valve, torn out from a gas lighter and a piece of PVC tubing, makes a perfect flushing kit!

Butane is flammable, but that's not an issue if you're careful. It's heavier than air. And inert to nearly everything. And you can see when it reaches the mouth of the flask.

SWIM's not very confident butane's perfect - but for him, it works fine.


03-28-02 21:18
No 289141
      re: butane  Bookmark   

antoncho, you're a genius!

butane is indeed a protective gas that fits all requirements, except, that its flammable. otto had this idea too, wanted to post it and scrolled down to your post. you are actually using it, thats unbelievable cool.