Sciencemadness Discussion Board - Issues with reduction of vanillin

Issues with reduction of vanillin

Hexavalent - 19-1-2013 at 06:35

Recently, I have attempted the reduction of vanillin in basic medium using borohydride, but have run into some difficulties.

I weighed out 380 mg (2.5 mmol) of food-grade vanillin, obtained online, and transferred into a small, clean, test tube. 2.5 mL of 1M sodium hydroxide solution was then added to the vanillin, but not all was dissolved. The volume was increased to 6ml with DI water before all the solid dissolved fully. The result was a medium-intense yellow solution, which was cooled to ~5*C in an ice bath.

75 mg of NaBH₄ was then weighed out, and added cautiously, with stirring, to the vanillin solution over 3 minutes. Upon addition, large volumes of a gas - presumably hydrogen - was released. A small temperature rise (~3-4*C) was also noted during the addition. The mixture was then removed from the ice bath and allowed to sit at RT for half an hour, before replacing in the ice bath and adding 2M hydrochloric acid dropwise until the solution was slightly acidic to pH paper (pH 6.5). The sides of the flask were then stratched to induce crystallization, but to no avail. At this point, the solution was pale yellow in colour.

Assuming that the excess water present was holding the alcohol in solution, the mixture was gently heated with a hot air gun until the volume reached 2.5 mL. The mixture clouded significantly and a colour change to pink was observed. What appeared to be a fine precipitate was formed, and vacuum filtration was attempted. The filtrate was clear, but no solid was recovered from the filter paper, which, upon closer inspection, did not appear to have any solids upon it. To verify this, the filter paper used was immersed in a small quantity of ethanol, stirred, and left to sit for 3 hours. The ethanol was evaporated but no solid was left behind.

Any ideas?

The original idea for this synthesis came from "Microscale Reactions of Vanillin" by Rosemary G Fowler, which is available online but exceeds the file size limits to be included as an attachment here.

Boffis - 19-1-2013 at 07:28

I have looked at the original paper and it seems to me that you have simply used too much NaOH. I suggest you try again using gentle warming of the mixture to affect solution and then cool. Don't make it too hot as that will cause side reactions, try 40-50 C and patience.

Vanillyl alcohol is probably soluble in excess NaOH because of its phenolic OH group so perhaps you need to make it more strongly acid. The cloudiness but nothing on the filter suggests that the product has come out of soution as an oil too but then it should have been simply drawn through the paper. What size funnel did you used? You should be using a small Hirsh funnel with a 21mm plate for this amount of material.

Let us know how you get on.

Hexavalent - 19-1-2013 at 07:38

Thanks for the suggestion, but, alas, I lack a Hirsch funnel: for all vacuum filtrations, I use a 50mm diameter Buchner funnel.

I shall try again tomorrow or later, and I concur about the oil issue. Perhaps I should have left it overnight again to allow it to crystallize?

Also, what pH should I acidify to this time, and how much NaOH to use for the 380 mg aliquot of vanillin?

Adas - 19-1-2013 at 08:19

The yellow color might have been caused by aldol condensation.

Hexavalent - 19-1-2013 at 08:26

No, I don't think so. I've used vanillin a few times in various solvents, and it has almost always gone yellow, as the paper referenced decscribes.

Besides, what would the vanillin condense with? It is an aromatic aldehyde, does not have an enolizable hydrogen and therefore I believe cannot condense with itself.

kavu - 19-1-2013 at 08:34

You might like to add a bit more NaBH4 just to be on the safe side if some decomposes before reduction. Otherwise the procedure seems viable. Stirring more than 30 min might also help (you can monitor the reaction using TLC-plates if you have them). Product should crystallize by cooling the quenched reaction mixture in an ice bath. If this does not do the trick then extraction with EtOAc etc. should work out fine.

[Edited on 19-1-2013 by kavu]

Hexavalent - 19-1-2013 at 08:47

No, unfortunately I do not have access ATM to TLC plates

Yes, I expected the product to crystallize after cooling in the ice bath, but an excess of water was used to completely dissolve the vanillin, as seen here. Perhaps when I re-try the synthesis, the elevated temperatures (40*C suggested) will encourage solvation.

I shall use 85 mg of the borohydride next time, too, based upon your suggestion.

Can anyone suggest what pH to acidify to with the hydrochloric acid?

Thanks.

Prometheus23 - 19-1-2013 at 13:06

You should be good if you just use enough acid to destroy any excess borohydride and bring the pH down low enough to ensure the phenolic OH stays protonated. But I wouldn't worry too much about your pH not being low enough during workup. Since vanillyl alcohol's pKa is about 9.9 then even a slightly acidic solution (<7) shouldn't contain any appreciable amount of phenolate ion.

Hexavalent - 19-1-2013 at 13:44

Okay, thanks. BTW, is a staggered addition of the borohydride initially better, or should one simply add it all at once?

killer_lapin - 19-1-2013 at 14:54

No you were right initially, the borohydride has to be added in small portion, otherwise the rate of decomposition is faster than the rate of reduction, so you'll end up with alot of starting material.

Hexavalent - 20-1-2013 at 02:51

Ah okay, I thought as much. I guess this is similar to why, when making potassium (via the KOH reduction method) the staggered addition of the t-butyl/amyl alcohol is favoured, in that case to reduce side reactions initially.

zed - 22-1-2013 at 19:45

As I recall, Sodium Borohydride is fairly soluble in Ethanol. It also dissolves Vanillin.

Mercedesbenzene - 22-1-2013 at 20:38

I have actually done this lab recently. I used 1g of vanillin and 1.2g of sodium borohydride, I dissolved the vanillin in 6mL of 1M NaOH and added in the borohydride all at once. I heated the solution for 20minutes at 50-60C. I then cool the reaction mixture down in an ice bath. I slowly added 2M HCl until the pH was 7, if your pH gets any lower than 6 your product will be ruined and you will not be able to extract any, a few of my fellow classmates had this happen to them. I left the solution to cool for a bit in the ice bath, then i vacuum filtered, and recrystallized by first heating up a solution of 2:8 ethanol to water, i dissolved the vanillyl alcohol in 4mL of this hot solution and then let it cool down, this produced a nice lot of crystals which were then vacuum filtered and dried. I obtained at 59% yeild with a MP of 108-114C, the literature value is 110-117C. A second recrystallization would have probably fixed this

[Edited on 23-1-2013 by Mercedesbenzene]

Paddywhacker - 23-1-2013 at 05:32

Most boronhydride reductions that I have seen have been in ethanol, as zed has hinted.

Nicodem - 23-1-2013 at 07:52

Quote: Originally posted by Paddywhacker

Most boronhydride reductions that I have seen have been in ethanol, as zed has hinted.

Only because most substrates are not soluble in aqueous media. Once deprotonated by NaOH, vanillin dissolves nicely in water. Besides, it is methanol in most cases, not ethanol (NaBH₄ is not that well soluble in ethanol, though it can be used as a solvent). Additionally, I'm not so sure that the reduction of vanillin in methanol would work equally well.

Quote: Originally posted by Mercedesbenzene

I slowly added 2M HCl until the pH was 7, if your pH gets any lower than 6 your product will be ruined and you will not be able to extract any, a few of my fellow classmates had this happen to them.

Indeed, ortho- and para-hydroxybenzyl alcohols easily undergo elimination of water to form the corresponding quinone methides. In the absence of suitable nucleophiles (like in acidic media), these easily decompose by oligomerization. Acidifying with acetic acid would obviate the problem of pH.
Elimination to quinone methides can occur also under basic conditions, and this is why I once suggested that vanillin could perhaps be reduced all the way to 4-hydroxy-3-methoxytoluene by using thiourea dioxide.

mnick12 - 23-1-2013 at 14:15

I did this reaction a while ago a on a larger scale, but the procedure was pretty similar to what some of you have posted. I believe I did this reaction at the 100mmol scale, and it was pretty uneventful. Dissolve your vanillin in 1M NaOH cool in a water bath. Once it has cooled down a bit add the borohydride and initiate vigorous stirring (~1000rpm). As the reaction calms down a little bit remove from the water bath and allow it to stir for a few minutes. Next neutralize with GAA (Watch out exothermic!) and soon the whole flask will gel up with small pleasant smelling crystals. I believe my crude yield was around 70%, but I never bothered to check mp or do a recrystalization because I simply did this one for fun.

Hexavalent - 23-1-2013 at 14:53

Quote: Originally posted by Nicodem

In the original article I mentioned above, no ethanol or methanol is added as a solvent during the reaction: it is performed in aqueous medium, which, as you mentioned, allows for the deprotonation of the vanillin and consequently easy solvation.

Quote: Originally posted by Mercedesbenzene

I slowly added 2M HCl until the pH was 7, if your pH gets any lower than 6 your product will be ruined and you will not be able to extract any

In the original article, the writer mentions that the mixture was acidified until "distinctly acidic to litmus paper" - is pH6 sufficient for this? And should one wait a few moments before taking final measurements due to the buffering nature of some boron compounds?

Finally, if acetic acid is used for the acidification, would a slight excess have to be used to compensate for the somewhat basic nature of the sodium acetate formed?

Thanks

[Edited on 23-1-2013 by Hexavalent]

Mercedesbenzene - 23-1-2013 at 15:06

1M NaOH works quite well to dissolve the vanillin and they may say distinctly acidic, but it works quite well going to pH 7, use litmus or a pH meter and go very slowly and you will be fine. If using acetic acid of HCl just go slowly to pH 7 on pH paper

Hexavalent - 23-1-2013 at 15:12

Alas, I lack GAA and have to resort to using vinegar when it is needed in a reaction. I shall use 1M HCl (aq) for the acidification, and this time shall attempt to obtain pH 7 or just below, as shown by narrow-range indicator paper (of which I have a large supply).

zed - 29-1-2013 at 19:46

Methanol and NaBH4? OK, I suppose, but I used to find it wasteful, at least for my purposes.

NaBH4 slowly decomposes in Methanol, liberating Hydrogen.

Since I wanted to use that Hydrogen, for Hydrogenations, it troubled me that my NaBH4 was breaking down before I could get it into the system.

[Edited on 30-1-2013 by zed]