(Hive Photographer)
04-16-02 13:56
No 297806
      How to prepare hydroxyvanillin
(Rated as: excellent)

This is a write-up of a succesfully performed two-step rxn from vanillin to 5-hydroxy-vanillin. It doesn't provide real new stuff, but it does provide a report of a real performed and several times applied procedure to arrive at a valuable precursor namely the 5-hydroxy-vanillin starting with vanillin. Everything, except the soxhlet apparatus is OTC and easily acquired.

5-Bromo-vanillin from vanillin
60 grms vanillin (0.4 mol) are dissolved with stirring in 300ml GAA. 96grms KBr (0.8 mol) are dissolved in another flask in 150ml dH2O. When all solids are dissolved and both solutions have reached room temperature they are mixed together forming a still water clear solution. In parallel 22ml H2SO4 are diluted with 40ml GAA carefully. The H2SO4/GAA mix is added to the KBr/Vanillin solution from which now solid KBR may precipitate [1]. This still stirrable mixture is put into a 1000ml beaker sitting in a water bath at room temperature. Under moderate mag-stirring 44ml 32% H2O2 (0.8 mol) are added during 60-80 minutes. When the first 30% of the H2O2 has been added, ice cubes are put into the water bath to keep the whole rxn temp at about 20DegC [2]. The 5-Bromo-vanillin appears as a yellow precipitate when approx. 40% of the H2O2 has been added. After finishing the addition of the H2O2, the rxn is stirred further for 1.5hrs in the ice water bath keeping the rxn temp between 15 and 20DegC to complete the rxn. The yellow mix is now put into a 2l beaker containing 1200ml HO with ice cubes and stirred for another 20mins. The 5-Bromo-vanillin is vac filtered and washed with ice water several times. It is not worth to keep the mother liquors since overnight a dark crust is building up at the bottom of the beaker. Total yield of crude 5-bromo-vanillin about 80grsm (approx 90%) with a mp of 160-162 DegC. The crude product is used for the next step. Pure 5-Bromo-vanillin has a mp of 165DegC and forms white! crystals [3].

1: The separation of solid KBr after the addition of the H2SO4/GAA mix doesn't cause a problem as long as at the beginning of the H2O2 addition the rxn temp is about 20DegC.  Later the rxn should be cooled, the KBr dissolves and the 5-Bromo-vanillin precipitates.

2: The bromation of vanillin can be done without elemental bromine using the above described in-situ bromine generation. At no time any smell or vapor of bromine appears and Uemura can't frighten any bees (Thanks to Rhodium).

3: This rxn has been performed several times, it always gives high yields between 85% and 92%. It is however important to keep the temp less than 25DegC (say) to avoid di-bromation of the vanillin. Warming up the rxn does not improve but decrease yields and provides dark colored impure products.

See ../rhodium/chemistry /mmda.mescaline.html, Post 266929 (Rhodium: "Re: Alkali metal alkoxides: finally, OTC!", Novel Discourse), Post 27874 (Osmium: "Re: 2CB for the LAH impaired", Methods Discourse)

5-Hydroxyvanillin from 5-Bromo-vanillin
70grms NaOH (1.75 mol) are dissolved in 375ml H2O in a 2 (better 3) neck 1l RB flask equipped with a condensor together with a gas-inlet for inert gas (e.g. argon), inside a magnetic stirr bar. Into this freshly prepared NaOH solution 2grm CuSO4 dissolved in 30ml H2O are added droppwise at 50DegC under stirring. A deep blue solution builds up. The flask is warmed up by means of an heating mantel to 70DegC+, the inert gas stream is activated before 45grsm of bromo-vanillin (0.2 mol) are added slowly under stirring. After each addition of a spatula complete solution of the bromo-vanillin should occur before adding the next amount. Temperature is continuously increased to keep the Na-salt of the bromo-vanillin in solution.

When all bromo-vanillin has been added, the mixture is refluxed under stirring for 20hrs. The inert gas stream should be continued when sufficient inert gas is available
to avoid oxidation. After the reflux time has elapsed, the brown-green rxn mixture is cooled down to 10DegC [3]. It is then made acidic to ph 3-4 by slowly addition of conc. muriatic acid (about 160ml needed) keeping the temperature less than 25DegC.

A gray precipitate forms which is vac filtered off when the mix has been cooled down to 10DegC again. The precipitate is dried in an evacuated excicator to constant weight (35 grms). Instead of a standard solvent extraction or as indicated in some references, applying a continuous solvent extraction over 20hrs! it was found easier to evaporate all water under reduced pressure and apply a Soxhlet extraction on the dried mix of inorganic salts and the hydroxy-vanillin [2]. Therefore the filtered and slightly acidic solution of cooper/sodium salts and hydroxyvanillin was vacuum de stilled at 50-100mbar until the solution starts to bump heavily due to the separation of solids. The solids consisting of Na-salt crystals and organic material are again vac filtered and weighted 98 grms after drying. The remaining almost black oily filtrate was left in a petri dish until all water has evaporated. The black solid left weighted 20 grms.

DCM was found to be a proper solvent choice for the soxleth extraction. It doesn't dissolve any inorganic salts and the hydroxy-vanillin -which is not very good soluble in DCM - crystallizes already during reflux as a coffee cream crust in the flask. Soxhlet extraction was stopped, when the DCM extract became water clear without any yellow touch. The flask was cooled down in the freezer to -20DegC to complete crystallization. The crude hydroxy-vanillin was filtered and dried. A total amount of 26 grms crude product with a mp of 124-127 DegC has been yielded (approx 80%).

1: Yields are less than reported in the literature. Uemura however did not apply inert gas protection all the time and some parts of the product may have been oxidised further. His work-up strategy is perhaps not optimal, but has some advantages, like minimal amount of solvents and use of standard equipment.

2: Attempts to do classic extraction with e.g. ethyl acetate ended in painful emulsions with the final result of a broken piece of an expensive separation funnel. The hydroxy vanillin seems further to be not very soluble in organic solvent, water seems to bee the preferred solvent!

3: It should be worth to consider direct processing of the still alkaline rxn mix for methylation with DMS (say) to the 3,4,5 trimethoxybenzaldehyde to avoid the picky work-up. The 3,4,5 trimethoxybenzaldehyde may be easier to purify than the 5-Hydroxyvanillin.

See Post 287645 (foxy2: "3,4-Dihydroxy-5-methoxybenzaldehyde", Chemistry Discourse), Post 287652 (uemura: "'3,4-Dihydroxy-5-methoxybenzaldehyde'", Chemistry Discourse) and last not least the big failure via the syringaldehyde route, see Post 275171 (uemura: "Re: Vanillin", Chemistry Discourse)

Carpe Diem
(Official Hive Translator)
04-17-02 05:46
No 298138
      Awesome!  Bookmark   

Dear Uemura!

It was a grand joy for me to read your post - your work is truly great!

I am certain, that your post will inspire many a bee to repeat SWIU's experience.

Thank you. It was excellent. Keep us informed of the consequenceswink

(Pioneer Researcher)
04-17-02 06:15
No 298143
      Nice  Bookmark   

Beautiful to read your works !!!
(Official Hive Translator)
04-17-02 09:44
No 298183
(Rated as: excellent)

Take a look at somewhat relevant Patent US4933498:

3,4,5-Trialkoxybenzaldehydes, notably 3,4,5-triamethoxybenzaldehyde, are prepared by hydrolyzing 5-bromo-4-hydroxy-3-methoxybenzaldehyde with an alkali metal hydroxide, in water and in the presence of a copper catalyst, and then directly sequentially (i.e., without separating any reaction intermediates) etherifying the 4,5-dihydroxy-3-methoxybenzaldehyde thus produced with a lower alkyl halide, in an aqueous medium and at a pH maintained at a value of from 6 to 12, optionally in the presence of a catalyst. - the catalyst being a PTC, or an amine (which forms ptc in situ)
Of course,they use pressure conditions in the patent (CH3Cl as the alkyl halide), but using MeI instead would allow avoiding this problem.

All DMS patents i've seen involve either acetone or solvtless conditions - so, maybee, smth's wrong w/performing it in water - although, vanillin can bee methylated in this fashion quite smoothly.


P.S. Oh, yeah - Patent US4065504 - solventless methylation of vanillin, hydroxyvanillin, etc. - w/DMS and Na2CO3, - short rxn time, quantitative yield. At least they say so smile. Would bee SWIM's choice of all DMS alkylations.
(Hive Bee)
04-17-02 10:11
No 298193
      NaOH??  Bookmark   

with an alkali metal hydroxide, in water and in the presence of a copper catalyst,

Hmm that sounds resonable (and whupty you got NaOH, wather and methanol).
Nice to know that mixin DMS and an phenol works, but I think I still prefer to ad some DMF

(Hive Photographer)
04-17-02 12:29
No 298243
      Methylation  Bookmark   

Antoncho and Sunlight,

Uemura is glad that you like the write-up. Now, the idea of conversion of the hydroxyvanillin to another aldehyde without its isolation has already been published in a french patent mentioned in Post 282519 (3base: "MMDA: bromovanillin -> myristicinaldehyde", Methods Discourse) provided by 3base. There the hydroxyvannilin is converted with a dihalomethane in a two-phase reaction medium to myristicinaldehyde  without isolation.

Uemura heads however still for the 3,4,5 trimethoxy pattern - mostly for some historical reasons - and therefore he thought methylation with DMS should work on the crude alkaline hydroxyvanillin solution. Looking in e.g. Organikum or Houbden-Weyl he does not see good reasons why phenolic groups shouldn't be methylated in the classical way by adding DMS to the alkaline phenolic solution (O.K. inert gas protection is of a big advantage, but Uemura methylated hydrochinon with trimethylphosphate exactly this way).

He would and will first try to apply trimethylphosphate under waterfree conditions on the hydroxyvanillin and see what comes out.

Antoncho, do you have any evidence why hydroxyvanillin would be a special beast in this respect?

Hest, Uemura knows you are lucky having arbitrary access to DMF. When you say you would add DMF when you methylate in alkaline water with DMS, is it to increase the solubility of the DMS? If yes, wouldn't formamide do the job as well?

Carpe Diem
(Hive Photographer)
04-17-02 12:49
No 298252
      P.S. Oh, yeah - Patent US 4,065,504  Bookmark   


what a beautiful reference!

If this works with trimethylphosphate as well and if halfapint's sodiumhypophosphite/CuCl2/MeOH -> trimethylphosphate rxn also works as explained, the doors are open to almost unlimited methylation experiences  smile

Have a good night
(Hive Bee)
04-18-02 04:24
No 298620
      To uemura, not only for disolving the DMS but ...  Bookmark   

To uemura, not only for disolving the DMS but also for dissolvin the hydroxy-vanilin.
In my younger dayes I used acetone as a solvent, reflux for 24houers, still with CaCO2 as the base works nice (>80%).
The only feture with the DMF is that you can let it stir at room. temp. for 2 houers and then your done.
Make it all a bit faster but that's all (ans you can make the reactio in a mason jarr if you wants smile)
(Official Hive Translator)
04-18-02 07:01
No 298654
      Oh. Oh.  Bookmark   

Hest, again you surprize me!

can you, please , elaborate - what you used as a base in dmf/dms methylation?


such a useful solvent, dmf. You can also use it for methylenations!

(Hive Bee)
04-18-02 07:20
No 298662
      Base  Bookmark   

I use CaCO3 Usual 1-1.5g for 1g phenol(no matter witch on).
And yes DMF is my favorit solvent, run the reaction, pour it over ice, let it melt and filter off your product.
(Hive Bee)
04-18-02 11:00
No 298744
      ottos low yields in methylation  Bookmark   

hi uemura,

congratulation to your good results! some weeks ago otto did some trials on the same matter, but unfortunately had not enough time to spent on it. maybe otto can give you some hints.

otto was too cheap to buy acetic acid, so he made the 5-bromovanillin by stirring vanillin in a solution of KClO3, H2SO4 and KBr in water at r.t. the vanillin didn't dissolve, reaction was almost complete the next day. the product obtained thus was only ~80% pure (vanillin and the corresponding acids).

the 5-hydroxyvanillin was made by heating the 5-bromovanillin (1.25 g) in aqueous NaOH solution (1.3 g in 15 ml, a little CuSO4 was added) in a pressure tube (no inert gas available) for 20 hours at 110C.

this was repeated twice and the resulting combined solutions were acified w/ HCl and evaporated to dryness. the resulting powder was extracted with acetone, MeI (2 ml) was added along with K2CO3 (3 g) and held at gentle reflux overnight.
evaporation of the acetone solution gave 0.6 g of a yellow product with a smell reminiscent of vanillin. main constituents 3,4,5-TMBA and 3,4-DMBA.

an attempt to directly methylate in the alkaline reaction mixture failed:

to the raw solution of (same quantities as above, only one p.t. filling!) there was added 3 ml of 85% MeONO2 and the mixture was stirred overnight at 60C. twice extraction w/ EtOAc yielded only 70 mg of product, which constitutes of 3,4,5-TMBA and 3,4-DMBA.

otto is looking for a way to skip isolation of 5-OH-vanillin as mentioned in the patents. otto is convinced, that there will be a simplicistic and OTC road to 3,4,5-TMBA in the near future.

(Hive Photographer)
04-19-02 03:14
No 299063
      methylation procedures  Bookmark   

Thanks a lot Otto for your report.
It is helpful to avoid the same failures.
It a shame that DMS is so difficult to get. Uemura is sure DMS would be the preferred methylation agent. Shortly  Uemura will make some small scale investigations once he has prepared the trimethylphosphate. Reports will follow.

Carpe Diem
(Hive Bee)
04-19-02 13:55
No 299247
      patent 4065504
(Rated as: excellent)

So I have some questions about this. Please help me!!!


A mixture of 100g of syringealdehyde (0.55 mol), 85.0g of sodium carbonate (0.80 mol) and 106.0g of dimethyl sulfate (0.84 mol) was placed in a round-bottomed flask equipped with a reflux condenser, a stirrer and a dropping funnel. The syringealdehyde was of about 99% purity and was obtained, by distillation, from a crude mixture of vanillin and syringealdehyde produced by alkaline oxidation of waste Kraft liquor. The dimethyl sulfate was of commercial quality (B.P. 75.degree.-77.degree./15 mm) and the sodium carbonate was of reagent grade in some experiments and of commercial grade in others. The flask was kept in a silicone oil bath on a hot plate. The mixture as prepared at room temperature was a thick paste which could be kneaded but could not be mixed by the stirrer driven by an electric motor. When the temperature of the mixture reached 75.degree. C the mixture became fluid and was easily stirred by the blades of the stirrer. The temperature was further raised to about 85.degree. C and maintained at this temperature with stirring for a total of about 2 hours. After the initial 35 minutes of this period the reaction mixture began to thicken due to the consumption of dimethyl sulfate and, in order to maintain it fluid, water in 5 gram portions was added from time to time through the funnel while the stirring continued. In total about 60g of water were added to the mixture during the reaction period.

At the end of two hours, the heating was discontinued and about 500 ml of hot water was added to the mixture. The mixture was acidified with concentrated hydrochloric acid and was then extracted three times with about 250 ml of benzene and the combined extract was washed with water. The benzene was removed by distillation and the solid product was dried in a vacuum oven and weighed. The yield was 106.9 g of 3,4,5-methoxybenzaldehyde, representing a yield of 99.3% of the theoretical. Analysis by gas-liquid chromatography (g.l.c.) showed a purity of 99.85% with syringealdehyde as the only detectable impurity.


50g of syringealdehyde (0.275 mol) 40g of sodium carbonate (0.38 mol) and 53.4g of dimethyl sulfate (0.42 mol) were placed in a round-bottom flask and heated substantially as in Example 1. When the temperature reached 75.degree. C the reaction mixture became fluid but after 30 minutes at a temperature between 80.degree. and 87.degree. it began to thicken again. 30 ml. of water was then added dropwise while the temperature was maintained for 1.2 hours. On completion of the reaction 250 ml of hot water was added to the mixture. Then the mixture was cooled, the solidified organic material separated by filtration, washed with water and vacuum dried. The yield of 3,4,5-trimethoxybenzaldehyde was 99.6% with about 0.4% syringealdehyde remaining.


The conditions of Example 2 were repeated with the difference that the reaction mixture was maintained at a temperature between 85.degree. C and 95.degree. and that it was maintained at this temperature for 3.6 hours. The yield of 3,4,5-trimethoxybenzaldehyde was substantially quantitative with not more than 0.02% syringealdehyde remaining.


25.0g syringealdehyde (0.137 mol), 26.6g of dimethyl sulfate (0.211 mol), and 27.7 g. of potassium carbonate (0.201 mol) were heated with stirring as in Example 1. When the temperature reached 45.degree. C the mixture became fluid and carbon dioxide was given off. The temperature was slowly increased to 75.degree. C and held at this value for 1 hour. Addition of water, acidification, and benzene extraction gave a quantitative yield of 3,4,5-trimethoxybenzaldehyde which contained no detectable syringealdehyde or other impurities as determined by g.l.c. analysis.

The examples which produce 3,4,5-TMBA all start with syringealdehyde.
Will ya get also 3,4,5-TMBA if ya use hydroxyvanillen?

This is what the patent says but why isn't there a example with Hydroxyvanillen?

A method of making veratraldehyde from vanillin is described by J. S. Buck in "Organic Syntheses" (Vol. II, Wiley, 1943, p. 619) and the same method was used for the methylation of 5-hydroxyvanillin to 3,4,5,-trimethoxybenzaldehyde (Journal of American Chemical Society 74, 1952 p. 4263) and for the methylation of syringealdehyde to 3,4,5,-trimethoxybenzaldehyde (Journal of American Chemical Society 76, 1954, p. 5555).

I have another important question. I don't have excess to this kind of books and references and I can't find it. Can somebody send the text or post the text of this reference, Pleace.

(Journal of American Chemical Society 74, 1952 p. 4263)


You've gotta love me.
(Hive Photographer)
04-19-02 22:33
No 299419
      Why not...  Bookmark   

This is what the patent says but why isn't there a example with Hydroxyvanillen?

Uemura thinks there is no clue why it should not work with hydroxyvanillin, just adjusting the amount of DMS for 2 instead of 1 phenolic group. He thinks the patent author had just around the syringaldehyde and not the hydroxyvaillin. And, Uemura saw elsewhere (Rhod's site?) this waterfree methylation method applied to hydroxyvanillin with DMS (Trimethylphosphate resp.)

Carpe Diem
(Official Hive Translator)
04-21-02 23:18
No 300096
      In situ methylation  Bookmark   

Dear Uemura -

i just recalled suddenly a patent that i once posted  - about production of TMBA from phenol by triple hydroxymethylation followed by oxidation of all methylol groups to aldehydes and selectively splitting off two formyls leaving 3,4,5-trihydroxy-BA - which is then methylated in situ w/DMS!

Sorry for reposting the stuff, i cant UTFSE now - in addition, here is precisely the part you want - note that after you remove CuO from your rxn w/bromovanillin, the solution you get is virtually identical to the one described beelow (alkali concentration?):

Stage 3. Oxidation of the triformylphenol with hydrogen peroxide at pH<7

450 mg of triformylphenol (2.5 mmols) were introduced into a 500 ml five-necked glass reactor fitted with a thermometer, an electrode, a condenser and two dropping funnels. 12.5 ml of deaerated water were added, the magnetic stirrer was started and the temperature was raised to 45.degree. C., everything being carried out under nitrogen.

The pH was raised to 4.5 with 10% strength by weight sodium hydroxide solution contained in the first funnel, and 1.96 g of 10% strength by weight hydrogen peroxide (5.8 mmols) were then dropwise introduced, over the course of 10 minutes, from the second funnel.

The solution, which was heterogeneous at the beginning, gradually became homogeneous. The reaction mixture was maintained at 45.degree. C. for:

50 minutes at pH 4.5, and

45 minutes at pH 5-5.5.

4.16 mmols of sodium hydroxide had been added in total.

Stage 4. Methylation of the 3,4,5-trihydroxybenzaldehyde

In order to more easily measure the yield of gallic aldehyde obtained during the previous step, all products obtained were methylated in the following manner:

In the apparatus described above, the dropping funnel containing 10% strength by weight sodium hydroxide solution was replaced by a funnel containing degassed 30% strength by weight sodium hydroxide solution. The funnel containing hydrogen peroxide was replaced by a funnel containing dimethyl sulfate (DMS).

The pH of the mixture was raised to 8, while maintaining a temperature of 45.degree. C. and a nitrogen atmosphere. 4.75 g of DMS (37.5 mmols) were then introduced over a period of 40 minutes, while maintaining the pH at 8-8.5. The experiment was terminated when the pH stabilized, which was after 1 hour, the amount of 30% strength sodium hydroxide solution used then being 4.2 g (31.5 mmols).

After cooling to 20.degree. C., the reaction mixture was acidified to pH 3.5 with 0.25 ml of 50% strength by weight H.sub.2 SO.sub.4 and extracted three times with 20 ml of dichloroethane. The organic extracts were washed with 10 ml of water, dried over sodium sulfate and analyzed by gas phase chromatography.

The hydroxyls had been totally methylated because neither gallic aldehyde, nor 5-hydroxy-p-vanillin, nor seringaldehyde (3,5-dimethoxy-4-hydroxybenzaldehyde) and/or its isomer, 3,4-dimethoxy-5-hydroxybenzaldehyde, were observed.
    Yield of 3,4,5-trimethoxybenzaldehyde,
                          216 mg = 44%
    relative to the triformylphenol introduced
    in step 3
    Yield of 1,3,4,5-tetramethoxybenzene
                           24 mg =  5%

of course, 44% yield doesn't sound that good, but i am positive that most of the loss occurs at the Dakin stage, quite naturally. "All of the hydroxyls have been methylated" - now that does sound good.

In all aq. DMS methylations pH should bee precisely monitored to get high yield.

hope that it helps,