Sciencemadness Discussion Board - benzidine

benzidine

Magpie - 24-3-2016 at 16:19

Benzidine_200.svg.png - 3kB

Today I synthesized benzidine HCl using a 2X scale version of the procedure in Brewster (forum library). This synthesis uses nitrobenzene as precursor. The reductant is hydrogen generated via Mg turnings and methanol. Although the nitrobenzene charge was only 5.2 ml the sequential hydrogenations (-->azobenzene -->hydrazobenzene) generated a lot of heat. Lastly, benzidine is formed as a rearrangement of hydrazobenzene. As advised I used a tub of cold water for cooling when it looked like my reflux condenser was being overwhelmed. By restraining the 500ml RBF/reflux condenser at just one point it was easy to keep the reactants mixed by swirling the whole system.

Tomorrow I will freebase (I guess that's the term) my product to benzidine using 5% NaOH, crystalize, filter, dry, weigh, then calculate a yield.

Below are some pictures I took along the way:

Attachment: php6umGgL (90kB)
This file has been downloaded 1206 times
hydrazobenzene heated to 75-85°C

acidified hydrazobenzene

aqueous benzidine HCl

[Edited on 25-3-2016 by Magpie]

[Edited on 25-3-2016 by Magpie]

[Edited on 25-3-2016 by Magpie]

UC235 - 24-3-2016 at 17:30

Very nice. I've wanted to make a small amount of benzidine for some time (given the uniqueness of the rearrangement reaction and for production of congo red).

If this goes to prepub, I think a significant disclaimer about the product's carcinogenicity, proper handling, and waste management is in order.

Magpie - 24-3-2016 at 17:48

Quote: Originally posted by UC235

Very nice. I've wanted to make a small amount of benzidine for some time (given the uniqueness of the rearrangement reaction and for production of congo red).

If this goes to prepub, I think a significant disclaimer about the product's carcinogenicity, proper handling, and waste management is in order.

Yes, that's why I'm making it: as a precursor for Congo Red

The carcinogenicity issue kept me from making this for some time. It's everywhere with the Congo Red precursors. I feel I have good control of my lab hygiene so decided to go for it. After all, benzene is carcinogenic. I don't know how you can do much organic chemistry without it, IMO.

But if I decide to place this in prepub I would add the warnings.

UC235 - 24-3-2016 at 18:14

Quote: Originally posted by Magpie

Yes, that's why I'm making it: as a precursor for Congo Red

It's because all the old Orgsyn preps use it as an indicator and you can't buy it easily, isn't it. Because that's why I want to make it and it seems like we have the same list of plans usually.

Magpie - 24-3-2016 at 18:21

yes - that's it.

Plus it looks like a neat synthesis.

gdflp - 24-3-2016 at 20:08

I guess a lot of us are working on the same synthesis. I currently have some recrystallized 1-nitronaphthalene from last weekend's work sitting on my lab bench. I was planning to convert it to naphthionic acid this weekend if I have the time, though I haven't prepared the requisite benzidine yet.

Pumukli - 26-3-2016 at 05:04

Not to mention, that benzidine is insanely overpriced as a chemical. At least I found 2-3x prices on a mass to mass basis compared to other "similarly not too complicated" compounds. (What can be considered similar is a personal taste though.) :-)

unionised - 26-3-2016 at 05:25

So, rather than use a different indicator like bromophenol blue or methyl orange, you decided to make one of just a handful of chemicals that have been banned from (most) commercial manufacture.
Well- it's not the decision I'd have made...

Did it occur to you that there was a reason the price was so high?

DJF90 - 26-3-2016 at 06:30

As unionised has mentioned, there are alternatives to congo red as an indicator with very similar range and transition pH. I'm not sure how bad congo red itself is, but benzidine (and its derivatives) is/are one of the COSHH banned substances, at least here in the UK.

Perhaps the warnings are overexaggerated as in the case with benzene (the IARC monographs deal with exposure in an industrial setting), but its still worthwhile avoiding where safer alternatives exist, and taking due care if you decide to persist.

It is however an interesting synthesis once used as an undergraduate practical, and the product is indeed used alot as an indicator in OrgSyn preps. For those interested I have attached an old J. Chem. Ed. paper detailing the preparation

Attachment: Preparation of congo red.pdf (2.3MB)
This file has been downloaded 877 times

[Edited on 26-3-2016 by DJF90]

Magpie - 26-3-2016 at 09:09

This whole issue of carcinogeneity and what chemicals we should or should not be using needs some vetting, IMO. I hope we can do this on this board.

FYI it is interesting to look at the International Agency for Research on Cancer (IARC) list 1 carcinogens. Note that ethanol and UV radiation (which I assume includes the UV of sunlight) are on the list, along with benzidine: IARC 1

[Edited on 26-3-2016 by Magpie]

S.C. Wack - 26-3-2016 at 09:57

"IARC MONOGRAPHS VOLUME 99" is on the internet and it mentions all the benzidines research. It shows a link to bladder cancer. Similar perhaps to popular boogeyman hexavalent chromium, in the effect of dust on unprotected workers shoveling in the packaging department for 10 years being considered as dangerous by the regulation/litigation culture as looking at or even thinking of a 0.1% solution.

BTW treatment of the dye with dithionite regenerates benzidine; this would be a convenient synthesis if the dye was as available as it was.

Magpie - 26-3-2016 at 16:03

The aqueous benzidine HCl described above in the OP was freebased to benzidine. This required taking the slurry to a litmus paper blue endpoint using 210 ml of 5% aqueous NaOH.

aqueous benzidine slurry

The somewhat gummy product was then caught on a 7cm Buchner funnel and dried. It was drying very slowly at room temp so I placed it in a drying oven at 80°C for a couple hours.

This solid product was then dissolved in a little (10-15ml) 95% ethanol heated to boiling. The flask was then cooled over ice to recrystallize. The dried benzidine product is shown below:

benzidine

Results
The yield was 1.9g for a %yield of 41.2%. mp was 121.5°C. The Wiki mp range is 122-125°C.

Discussion
I suspected that the yield would not be too good based on some reading. That is why I doubled the scale of the Brewster procedure. I wanted to get at least 1g.

As mentioned above benzidine is an IARC list 1 carcinogen. Therefore this procedure should not be performed without adequate safety precautions and PPE.

The mp was hard to determine because the solid is a dark gray. It may well have melted over the Wiki range indicated. The 121.5°C is just the first indication of darkening and slumping that I noted. At any rate I am confident it is pure enough for my purposes.

Questions, comments, and suggestions are welcomed.

[Edited on 27-3-2016 by Magpie]

[Edited on 27-3-2016 by Magpie]

[Edited on 27-3-2016 by Magpie]

blogfast25 - 26-3-2016 at 17:23

Very nice write-up, as usual Magpie.

I'm quite intrigued by:

Quote: Originally posted by Magpie

This synthesis uses nitrobenzene as precursor. The reductant is hydrogen generated via Mg turnings and methanol.

Mg turnings and methanol generate H₂ at significant rates? Or did I miss something...

Also, there seems to be a C-C coupling reaction going on. Any reaction mechanism known?

[Edited on 27-3-2016 by blogfast25]

Magpie - 26-3-2016 at 17:48

Thanks for your interest, blogfast.

Mg and methanol produce H2 like gangbusters. Another advantage of Mg (over Fe or Zn) is that my turnings were very thin and light so almost floated. But the azobenzene-hydrazobenzene and methoxide products become very viscous during the reaction. I kept them fairly well mixed by vigorous manual swirling. At the same time I was keeping the slurry at 75-85°C, so heavy gloves are required.

I really don't know what was generating all the heat. It's hard to believe it was the reaction of the 5.2 ml of nitrobenzene to azobenzene then to hydrazobenzene. Could it have been the heat of formation of methoxide and H2?

I haven't seen the mechanisms for the first two products. The mechanism for the rearrangement to benzidine (a classic organic puzzle) is shown in the Wiki for benzidine: rearrange

UC235 - 26-3-2016 at 17:50

Quote: Originally posted by blogfast25

Mg turnings and methanol generate H₂ at significant rates? Or did I miss something...

Also, there seems to be a C-C coupling reaction going on. Any reaction mechanism known?

Reaction of Mg with alcohol can indeed be aggressive, but may need initiation. See Klute's prep of salicylaldehyde for a photo (http://www.sciencemadness.org/talk/viewthread.php?tid=10124)

That being said, I think it is extraordinarily unlikely that the hydrogen itself is the reducing agent but rather that the magnesium reduces nitrobenzene through a single electron transfer mechanism and the coupling to azobenzene or azobenzene-N-oxide occurs between (radical?) intermediates which are further reduced.

The strong-acid catalyzed rearrangement of hydrazobenzene to benzidine (Benzidine Rearrangment) is a classic mechanistic mystery, thought likely to be a concerted [5,5] sigmatropic reaction.

blogfast25 - 26-3-2016 at 18:06

Thanks, Magpie.

Dr.Bob - 27-3-2016 at 17:37

Very interesting thread, the chemistry is interesting, the mechanism is, and the cancer risk discussion are all topics I am interested in.

As for the cancer risk, I have always thought that many risks were overrated, but still am careful with some compounds due to common sense. Max Gergal made at least 8 of the 10 top known carginogens in bulk and he is now well over 90. Many scientists I know used benzene, chloroform, and others for years and are fine, but that is not a scientific fact or a real study. But should anyone be interested, I have many IARCpublications, as part of the books that I recently got. So if anyone wants to study cancer risks, I can help you find the data, although some may be online now, it appears.

Darkstar - 10-4-2016 at 22:30

Quote: Originally posted by UC235

That being said, I think it is extraordinarily unlikely that the hydrogen itself is the reducing agent but rather that the magnesium reduces nitrobenzene through a single electron transfer mechanism and the coupling to azobenzene or azobenzene-N-oxide occurs between (radical?) intermediates which are further reduced.

You're right about the reduction proceeding through an SET mechanism; however, the coupling to azobenzene most likely occurs through nitroso and hydroxylamine intermediates, not collisions between nitrogen radicals. I've always found these dissolving-metal reductions rather interesting mechanistically. And since I had some time to kill, I drew a possible mechanism for the reduction of nitrobenzene to nitroso and hydroxylamine intermediates, along with their coupling to azobenzene and subsequent reduction to hydrazobenzene. The mechanism for the final rearrangement of hydrazobenzene to benzidine can be found on Wikipedia, as already mentioned.

Since the reduction is done in absolute methanol, I assume that the methanol is functioning as the proton source. I will say that this reduction was a bit trickier to figure out than those done in acidic media. Anyway, if anyone's interested, here's what I came up with:

Reduction of nitrobenzene to nitroso and hydroxylamine intermediates:

Coupling to azobenzene and reduction to hydrazobenzene:

Magpie - 12-4-2016 at 07:44

@Darkstar: Very nice. Thanks for this.

Are the e^- running loose in the solution?

Darkstar - 14-4-2016 at 03:29

Quote: Originally posted by Magpie

Are the e^- running loose in the solution?

It's certainly possible. Unfortunately, there haven't been a great deal of studies done on the actual electron-transfer mechanism of these magnesium-alcohol reduction systems, so it's difficult to really say one way or another at the moment. With that said, it was recently discovered that at least sodium in ethanol does in fact generate solvated electrons, and polar solvents are known to readily trap free electrons. Solvated electrons have also been found to exist in water as well, sometimes referred to as "hydrated electrons."

[Edited on 4-14-2016 by Darkstar]

Magpie - 14-4-2016 at 07:49

I understand that Na dissolved in liquid ammonia produces solvated electrons that cause the solution to take on a blue tint.
I was wondering if one could expect this same color to result from the reaction of Mg with methanol. The solutions in my benzidine synthesis are so colored that any faint blue would be masked, I assume.

Darkstar - 14-4-2016 at 08:06

I would imagine that the methanol is too reactive for the concentration of solvated electrons to ever really get high enough to give the characteristic blue color. Ammonia works much better as a solvent because it's fairly resistant to reduction by the electrons, allowing for a much more stable and concentrated solution.

clearly_not_atara - 15-4-2016 at 15:14

^^^ It seems like the azoxy compound should be an intermediate:

http://orgsyn.org/demo.aspx?prep=CV2P0057

In particular when you consider that hydroxylamines are the reduction product using acids (Zn/NH4Cl or borane) and azo coupling products are the products in strongly basic conditions, one possible explanation is that the radical anion:

Ph-N*-O(-) <<--->> Ph-N(-)-O* <<--->> Ph*=N-O-

is very resistant to reduction when deprotonated, but is reduced to the hydroxylamine when protonated. The anion can dimerize to azoxybenzene by eliminating the equivalent of O2-, possibly involving a H+ abstraction from the solvent.

Alternatively, if k1*[PhN*O-][PhN*OH] > k2*[e-][PhN*OH], you get azo coupling, otherwise the reverse, for some rate constants k1, k2.

[Edited on 15-4-2016 by clearly_not_atara]

Darkstar - 15-4-2016 at 22:17

Quote: Originally posted by clearly_not_atara

^^^ It seems like the azoxy compound should be an intermediate

I agree. I was even going to show an azoxy intermediate in my condensation mechanism as well. In the end, though, I chose to omit it to save steps. But in reality, I imagine the intermediate shown in the fourth step of my second mechanism would immediately get deprotonated to an azoxy given the strongly basic conditions. The excess magnesium would then transfer an electron to the azoxy to give a negative charge on the left nitrogen and a radical on the right nitrogen, with the radical center being stabilized by the two adjacent negative charges and the aromatic ring (in addition to resonance):

The left nitrogen then abstracts a proton from methanol to give a radical nitroxide, followed by a second electron transfer to the radical nitrogen to give an anion. The oxygen finally grabs a proton and gets eliminated as hydroxide, along with the extra proton on nitrogen to give azobenzene:

Quote:

In particular when you consider that hydroxylamines are the reduction product using acids (Zn/NH4Cl or borane) and azo coupling products are the products in strongly basic conditions, one possible explanation is that the radical anion:

Ph-N*-O(-) <<--->> Ph-N(-)-O* <<--->> Ph*=N-O-

is very resistant to reduction when deprotonated, but is reduced to the hydroxylamine when protonated. The anion can dimerize to azoxybenzene by eliminating the equivalent of O2-, possibly involving a H+ abstraction from the solvent.

While I don't deny the possibility of dimerization, another explanation for the lack of azo coupling under acidic conditions is the protonation of the hydroxylamine's nitrogen. This would prevent the hydroxylamine intermediates from being able to attack the nitroso intermediates, effectively inhibiting their condensation to azo products. Thus strongly basic conditions would promote condensation to an azoxy, while strongly acidic conditions would hinder it. Also, acidic reductions only tend to stop at the hydroxylamine stage when weak acids are used. When strong acids are used, the reductions will usually go all the way to an amine. (that, or an oxime that gets hydrolyzed to a ketone, depending on the substituents on the nitro group carbon)

In the case of phenylhydroxylamine, the nitrogen is much more basic than the oxygen. The reason nitrobenzene reductions using zinc in aqueous NH₄Cl give phenylhydroxylamine and not aniline is likely because the conditions simply aren't acidic enough to protonate the oxygen over the nitrogen, especially if the nitrogen is already protonated. Reduction of hydroxylamines to amines under acidic conditions is thought to involve an initial protonation of the hydroxyl group, which then leaves as water, and a subsequent reduction of the resulting nitrenium ion.

But as I said, I don't discount the possibility of dimerization through radical coupling, either as a competing mechanism alongside condensation, or as the primary pathway to azoxybenzene itself.

[Edited on 4-16-2016 by Darkstar]

Nicodem - 16-4-2016 at 00:06

I posted here an article about the study of rearrangements of N,N'-diarylhydrazines. It is mostly about the rearrangement of these into semidines, but might be interesting nevertheless.