Sciencemadness Discussion Board

Thiamine catalyzed benzoin condensation failure

SplendidAcylation - 7-7-2020 at 17:50


Today I attempted the much awaited synthesis of benzoin... Well, it is much awaited by me, as I have been planning to try it for quite some time.

Having decided against using cyanide, I went for the thiamine method, and ordered 90 x 100mg thiamine HCl tablets.

I have read many different publications on this reaction, including this awesome write up, and I came to the following conclusions:

1) the reaction is very easy and seemingly foolproof
2) it doesn't matter too much about the order in which the reactants are added (some add the alkali to the thiamine HCl solution, then add alcohol, others add the thiamine HCl solution to alcohol, then add alkali)
3) temperatures don't matter too much (some say to cool the solution of thiamine HCl before adding the alkali, others don't. Some say to maintain the temperature at 60c, others simply say reflux)

So I was quite confident that I would encounter no problems, but I was wrong. Haha.

I basically followed the procedure given in the write up above, I used the same quantities.

The alcohol used was denatured alcohol, which contains a mixture of ketones, no methanol.

The benzaldehyde I made myself by the oxidation of benzyl alcohol with nitric acid.
It was washed with brine, water, and sodium bicarbonate, distilled, and dried, about a year ago, before being put in a brown glass bottle and stored in the fridge until today.
There were no crystals of benzoic acid present.

The thiamine tablets each contained 100mg of thiamine HCl, with 400mg of inert binders, so I first ground up 70 tablets in the pestle and mortar, until a fine powder was obtained.

20mL of water was then added to the 35g of powder, this was obviously insufficient to make a homogeneous mixture suitable for filtering; instead I ended up with a thick paste.

I wasn't sure what to do at this point, as I did not want to deviate from the quantities given, by adding more water, so I decided I would use the alcohol to mix up the powder, followed by vacuum filtration.

This seemed to work fine, the solids in the Buchner funnel were then washed with a small quantity of alcohol.

The clear solution thus obtained was placed in a 3-necked 500mL RBF, to which a solution of 1.6g of NaOH in 4mL of water was added.
Immediately the solution turned cloudy yellow (although not the bright yellow pictured in the write up), before turning clear yellow after a few seconds of standing.
This dissolution was aided by stirring.

No cooling was used at this stage, however the temperature of the solution did not exceed 25c.

The flask was fitted with a reflux condenser, placed in a water bath, and the benzaldehyde was added all at once.
The benzaldehyde dissolved immediately, a yellow, clear, homogeneous, monophasic solution thus formed.

The solution was refluxed with stirring for 1.5 hours, during which period the temperature remained between 50 and 70c.

No visual change was observed at any point during this period.

After refluxing, the flask was allowed to cool to room temperature.
No crystals precipitated; The solution remained clear, and smelled strongly of benzaldehyde.

I cooled the mixture in the freezer; no change.

So I'm really not sure what went wrong, or what to try next

I do have a few ideas;

The solubilities for thiamine HCl are;
1g/mL in water
1g/100mL in 95% ethanol.

I initially added 20mL of water to the powdered tablets and mixed it into a paste, before adding 75mL of alcohol and filtering, so theoretically the 7g of thiamine HCl should have all dissolved in the water present, but maybe not?
Maybe most of it remained behind in the Buchner funnel?
Thankfully I kept the funnel with the filtrate still in it, so I could wash it through with a few small quantities of water, add NaOH, then add this to the flask and try again...

I'm not sure what impact adding extra water would have; benzoin is much less soluble in water than in alcohol, so I don't see a problem in adding more water, but I'm not sure.

Since the thiamine is a catalyst, does anyone know why such large quantities are required?

What would be likely to happen if only 1g of thiamine ended up in the flask instead of 7g?

The only other idea I have is that the reaction has taken place, but the benzoin isn't crystallising out, but this just seems so unlikely to me.

Ironically, I would not usually do a reaction on such a large scale (I don't have a lot of benzaldehyde), but I decided to take the chance in this case because I have so many plans for benzoin.

Any suggestions would be very much appreciated!

Thanks in advance

Bromolone - 7-7-2020 at 19:06

From the write-up here:

1. The temperature of the mixture during the addition of NaOH must not rise above 20° C and that during reflux must not rise above 65° C. I'm not sure if differences of 5° C might have destroyed/changed your product.


Constant monitoring of temperature is paramount during this part of the reaction and must be maintained between 60-65oC.

2. On crystals:

If crystals did not form after storage, withdraw a drop of the solution on a stirring rod and let it dry to produce a solid on the glass rod; then, rub it against the inside surface of the flask to induce crystallization.

Thiamine catalyses the reaction in a way similar to the way KCN does, and a smaller amount of thiamine used can imply fewer moles of benzaldehyde condensing to benzoin.
What amount of benzaldehyde did you use?

Boffis - 9-7-2020 at 00:56

@ Splendidacylation; When I did that write-up and several subsequent preparations I had a few failures for no apparent reason at the time. However, I have noted a few thing; firstly when making up the catalyst from a thiamine salt it turns deep yellow and then fades to pale yellow fairly quickly but if you add the other reactants before the colour fades yields are poor. If this happens simply prepare a fresh batch of catalyst after filtering off any product that has separated. This tends to lead though to a discoloured product.

When I did my original work it was mostly done over winter and my lab is then typically cold <10 C so I didn't worry about destroying the catalyst at start up. However, subsequently I had problem that might have been due to overheating of the catalyst or leaving too long before use. On the whole though I have found the thiamine method to be very reliable and often gives better yields of much purer product than the cyanide method.

Also, always purify your aldehyde/s before use unless they are new unopened bottles from a reliable supplier. I recently ran an experiment with some old benzaldehyde and found it to contain about 30% benzoic acid! Furfuraldehyde always contains lots of resin unless its new and was sealed under nitrogen or freshly distilled. The problem with benzoic acid is not only that it dilutes the active reactants but it also neutralises the base.

SplendidAcylation - 11-7-2020 at 11:35

Thanks for the replies!

I have got it to work.
I left it overnight after the initial (failed) attempt, and by the morning, 2g of needle crystals had formed in the solution.

I thought that maybe I had failed to extract enough thiamine from the tablets, so I extracted the (previously extracted) tablet residue again, except this time I did it 4 times, and with copious amounts of water.
The filtrate from the first extraction tasted bitter, but the fourth tasted like water, so I figured four was enough.
I ended up with about 100mL of water, which I then vacuum distilled down to around 10mL of slightly cloudy solution.

To this solution was added a solution of 2g of NaOH in 4g of water, and the solution quickly turned cloudy yellow, before quickly turning less cloudy (but not clear).

The resultant solution was then added to the failed benzoin condensation solution; The solution then darkened quite rapidly (not sure if this is supposed to happen)

It was then refluxed for 2 hours, crystals were visible even when hot
Upon cooling, the solution solidified into a slushy mass of yellow crystals!

The yield of crude yellow damp crystals (vacuum filtered) was 22g
After recrystallization from the minimum quantity of denatured ethanol, the yield was a slightly disappointing 12g.

I evaporated down the filtrate to about a third of the initial volume, in the hope that more benzoin would crystallize out; something did, but I'm not sure its actually benzoin; the crystals are much smaller, like grains of salt, compared with the needle crystals of benzoin.

The recrystallized benzoin still smells of benzaldehyde, but not too much, so that will probably suffice in terms of purification.

@Boffis: it seems that thiamine hydrochloride is fairly thermally stable, I had it heated to 70c for about an hour while vacuum distilling, also the Merck index says it decomposes at 258c, still, the catalyst formed when the NaOH is added is probably much less stable?

That's interesting about the benzaldehyde impurities, I would be quite disappointed if my precious aldehyde were 30% benzoin acid! :P

Tsjerk - 11-7-2020 at 12:05

Benzoic acid is not soluble in benzaldehyde, so it would clearly show up as an impurity.

SplendidAcylation - 11-7-2020 at 12:50

Quote: Originally posted by Tsjerk  
Benzoic acid is not soluble in benzaldehyde, so it would clearly show up as an impurity.

That's what I thought!
Well in that case, mine didn't have any, as there was absolutely nothing at the bottom of the bottle.

My benzaldehyde was washed with water, stirred with CaCO3 to get rid of acid impurities, then dried over CaCl2, filtered, and distilled, before storing in a sealed bottle

So I suppose, logically, it can't contain much benzoic acid, because it has only been in contact with a few milligrams of oxygen in the bottle.

DavidJR - 11-7-2020 at 19:25

Quote: Originally posted by Tsjerk  
Benzoic acid is not soluble in benzaldehyde, so it would clearly show up as an impurity.

Do you have data to support that assertion? I somehow doubt that it is completely insoluble.

karlos³ - 12-7-2020 at 05:29

It is definitely partially soluble, obvious when you put an older bottle of benzaldehyde in the freezer, as benzoic acid will crash out then.

Tsjerk - 12-7-2020 at 07:25

Ok, apparently at 30 degrees benzoic acid is 5% soluble in benzaldehyde, but seeing how rapidly solubility goes up with temperature I guess at 20 degrees it is only a couple percent soluble. I think indeed the fridge or freezer would be a nice way to clean benzaldehyde from benzoic acid.

Wang 2015

DraconicAcid - 8-11-2022 at 13:39

I actually want to try this reaction- anyone got tips for better extraction (and possibly isolation) of the thiamine?

UC235 - 8-11-2022 at 18:14

Quote: Originally posted by DraconicAcid  
I actually want to try this reaction- anyone got tips for better extraction (and possibly isolation) of the thiamine?

Just buy some? There are several online distributors of bulk dietary supplements. I'm seeing $5.75 for 25g of thiamine HCl on Purebulk, whom I've purchased from quite a few times with no issues. They also have the nitrate.

DraconicAcid - 8-11-2022 at 21:11

I just bought some pills- if I bought online, I'd have to wait for shipping.