Sciencemadness Discussion Board - Interesting benzyl chloride synth I discovered

Interesting benzyl chloride synth I discovered

Melgar - 25-6-2010 at 20:08

I discovered this reaction a while ago, and don't think it's been documented, so here goes.

Try this: bubble chlorine gas into a flask of toluene until it's bright yellow-green and more or less saturated. Now add a catalytic amount of lithium bromide. That's it. It seems this reaction needs some UV light to initiate it, but a fluorescent light works fine. Once bubbles start to appear, move the flask away from the light. The bubbling should become more vigorous, and the color should change from yellow to clear. Once all the chlorine is used up, the bubbling should stop, and there should be some salt at the bottom.

The bubbles are HCl gas, so presumably you would need two Cl atoms for each toluene molecule that gets chlorinated. The toluene can be boiled off and collected and reused.

It might be possible to continue bubbling chlorine into the solution in order to chlorinate more of the toluene, or chlorinate it further. I never tried this though.

I think sodium or potassium bromide should work, but I only had lithium bromide, so that's what I used. Perhaps someone else can test sodium or potassium bromide? I don't have access to much in the way of chemicals or facilities at the moment.

I think it works because bromine is a lot more susceptible to free radical formation than chlorine is. The bromine forms radicals, brominates the toluene, then the chlorine replaces it, forming another radical, which brominates another molecule of toluene, and so on. The reaction eventually dies out as the bromine radicals find each other, but UV light splits them apart again. Anyway, it sure beats leaving a flask of toluene and a chlorine generator out in the sun all day. You can have your reaction complete in under a half hour.

psychokinetic - 25-6-2010 at 20:30

Wouldn't the halogen radicals be attracted to the ring more than the methyl group?
Though, as it seems it's the same basic reaction as the chlorine generator way you described, I'm quite likely wrong.

not_important - 25-6-2010 at 21:01

The halogen radicals preferentially attach the benzyl hydrogens, ring substitution is electrophilic and is added by a number of metal salts or other Lewis acids. This is why industrial production uses the vapour phase, it greatly reduces the Lewis acid added ring chlorination.

Missing from this report seems to be any establishing of the product as benzyl chloride. And, yes, it's pretty similar to the standard method except that the amount of chlorination is limited by the amount of chlorine dissolved in the toluene, which is likely rather low.

497 - 25-6-2010 at 21:08

Quote:

And, yes, it's pretty similar to the standard method except that the amount of chlorination is limited by the amount of chlorine dissolved in the toluene, which is likely rather low.

I think he meant the advantage was that it didn't need continuous UV illumination. He mentions continuing to add Cl2 after the beginning of the reaction. Yes a quantified yield would be nice..

Melgar - 25-6-2010 at 23:02

Err. I meant it might be possible to get other chlorides like benzal chloride. It is definitely possible to dissolve more chlorine in the partially-chlorinated toluene, then use UV light to start the reaction off again. The amount of bromide determines how much free radical formation needs to be done by the UV light. Even a very small amount of bromide GREATLY reduces the UV needed though. More bromide allows the reaction to be self-sustaining in ambient light.

As far as what it is, I'd recognize that nasty smell anywhere. It is hard to figure out what yield is, especially when reaction vessels are test tubes, but you can see the chlorine bubbles shrinking as they dissolve in the toluene, and the reaction releases a surprisingly large amount of HCl gas when it's bubbling at top speed. I don't have the equipment to do an accurate assessment though.

zed - 26-6-2010 at 20:52

Guys like Vogel, suggest weighing the whole shebang.....Glassware included. But, the standard method usually involves gassing the solution until a certain weight gain is attained. When you've logged enough weight gain, you are done.

But, those instructions are for macroscale, and you are working microscale. Not so easy to log weight gain, when the whole project only weighs a few grams.

Personally, I think I would prefer adding the LiBr first, AND THEN bubbling Cl2 gas into the toluene. Mixing the reactants together, and then adding the catalyst, seems a little perilous to me. Especially on a larger scale.

I like the idea a lot, it could make the chlorination a lot easier to run.

Melgar - 27-6-2010 at 09:51

Quote: Originally posted by zed

Guys like Vogel, suggest weighing the whole shebang.....Glassware included. But, the standard method usually involves gassing the solution until a certain weight gain is attained. When you've logged enough weight gain, you are done.

But, those instructions are for macroscale, and you are working microscale. Not so easy to log weight gain, when the whole project only weighs a few grams.

Personally, I think I would prefer adding the LiBr first, AND THEN bubbling Cl2 gas into the toluene. Mixing the reactants together, and then adding the catalyst, seems a little perilous to me. Especially on a larger scale.

I like the idea a lot, it could make the chlorination a lot easier to run.

Well, weighing it can be difficult for a bunch of reasons, not the least of which is the fact that you're constantly losing gases. HCl does dissolve in toluene and slowly evaporates from it.

The reason I mentioned adding the LiBr after saturating the toluene, is because it makes the reaction a lot cooler-looking and more dramatic -- you can easily see that there's a reaction happening what with all the bubbles and the color change. Admittedly, once you've got the reaction down this is no longer a concern, so the LiBr could indeed be added at the beginning.

Incidentally, if someone had access to elemental bromine, (something I've never been able to generate with any sort of purity) that could be used instead, but this reaction will generate bromine in situ from a bromide salt, producing LiCl as a byproduct. Or NaCl, or KCl, if using a sodium or potassium salt. I don't think iodine would work though, since iodine radicals probably wouldn't be reactive enough to pull a hydrogen off of toluene, but I could easily be wrong.

Panache - 27-6-2010 at 12:46

nice post, however as NI said verifying you have a significant amount of benzyl chloride is essential even though you can smell the product it doesn't give one any indication of how successful the synth was, other products may simply not be odorous or weakly so, and benzyl chloride is quite distinct.
What made you do the things you did, some detail as to your musings prior to trying it would be of interest, but best of all distill off the chloride and give a yield of some sort.
But again, nice job!
the libr is simply acting as a halogen carrier, nice to know it works so well though.

Melgar - 27-6-2010 at 19:01

Well, I was familiar with the reaction between toluene and sulfuryl chloride, where it's initiated by a little organic peroxide. Unfortunately, sulfuryl chloride is kind of a bitch to make, so I tried just dissolving chlorine in toluene and initiating that with organic peroxide. That actually did work, but once the reaction finished you couldn't just dissolve more chlorine in it and have it start again. The key to this chain reaction seemed to be free-radical formation by a weaker radical than chlorine, so I thought it might be useful to try bromine, which I knew could generate radicals more easily than chlorine when exposed to light. Not only can bromine radicals be easily generated by exposure to light, but bromine can be generated from bromide salts in a chlorine solution. Initially, I thought the displacement of the bromide ions by chlorine would generate the free radicals that would start the reaction, but it turned out to just add orange elemental bromine to the solution. Of course, my disappointment didn't last long, since the anticipated reaction started as soon as I stepped outside.

Anyway, part of my reason for posting this was because I moved recently and all my chemistry equipment is packed up and not with me right now, so I was hoping maybe someone else might take interest in this reaction. I know plenty of people here have expressed interest in synthesizing benzyl chloride, so I was pretty sure someone would be interested, especially since this one seems a lot easier than any other I know of. I do know for sure that if you bubble chlorine into toluene with some bromine/bromide in it, most of the gas coming out of it will be HCl, not Cl2, so it's definitely getting chlorinated. I'm hoping that this reaction, like the sulfuryl chloride/organic peroxide synth, chlorinates the methyl group exclusively and not the benzene ring. Since the bromine radical isn't as reactive, I'm hoping it only reacts with the methyl group before being displaced by chlorine, but I can't verify that this is the case.

Melgar - 27-6-2010 at 20:27

Huh. Apparently bromine is so damn easy to make form radicals, that it can be done with an incandescent light bulb. And then, if there's chlorine around, it'll substitute the bromine, thus ensuring that only the methyl group's hydrogens get chlorinated. So according to my theory, this method should generate fairly pure benzyl chloride. Check this out:

http://dx.doi.org/10.1016%2Fj.tetlet.2006.07.109

I'm a grad student at the moment, but in mechanical, not chemical, engineering. Still, I ought to take this up with one of the chemical engineering professors at my university. It certainly seems to have potential value.

Anders Hoveland - 27-6-2010 at 21:33

Wow, I am fairly familiar with chemistry, but this seems unusual.
Are you saying that Bromine will react with toluene, if initiated by sunlight?
I would think the radicals would be consumed faster than they are made.
Let me think, SO2Cl2 and toluene, I would not think that would be reactive with toluene. What are the details of this reaction?

PhCH3 + Cl. --> PhCH2. + HCl
PhCH2. + Cl2 --> PhCH2Cl + Cl. --> PhCHCl. + HCl
PhCH2CHCl. + Cl2 --> PhCHCl2 + Cl. --> PhCCl3 + HCl

This reaction does not seem to generate excess Cl. radicals.
How is it self sustaining?

Also, in case some of you did not know, the nitronium cation preferentially attacks the ring because the positive charge is initially transfered to the ring, creating a sort of charged radical, then the NO2 reacts with it. Otherwise, obviously the methyl group would be more reactive.

[Edited on 28-6-2010 by Anders Hoveland]

psychokinetic - 27-6-2010 at 21:50

I believe he said the sunlight (UV) would radicalise the bromine, not initiate the reaction with toluene.

/I'm starting to feel like I'm feeding a new kind of troll.

[Edited on 28-6-2010 by psychokinetic]

mnick12 - 27-6-2010 at 22:00

Well if I understand it correctly it is pretty simple. It is a free-radical halgenation. When any halogen (except iodine) is hit with UV it splits to the reactive radical, which will attack one of the bezylic hydrogens forming HX. The newly formed benzyl radical is highly reactive so it to combines with the halogen forming the benzyl halide. Also the reaction is not only initiated by UV but also sustained by it.

Anyways I think I got that right, but if anything is wrong let me know.

not_important - 27-6-2010 at 22:15

This is covered in Len's fine writeup here http://www.sciencemadness.org/talk/viewthread.php?tid=10490

X2 + hv => 2 X ·

X· + H-C(stuff) => XH + ·C(stuff)

X2 + ·C(stuff) => X· + X-C(stuff)

For X2 == Cl2 you need blue light or shorter wavelengths, for Br2 green and even yellow light will generate radicals. The radical chain propagates for quite a few cycles, interaction with the walls, collision with another radical, and various other things eventually terminate a chain. So some ongoing level of new radical generation is needed, generally sort sort of illumination; incandescent lamps, work, especially quartz-halogen, but the generate a lot of heat relative the the number of effective photons. Fluorescent, mercury, and metal-halid lamps all worh, clear bulb Hg lamps are the cheapest high intensity source.

For liquid phase halogenations of aromatics in more than a small volume, visible light is better than UV as the aromatic (toluene) absorbs the UV strongly enough that much of the volume is "shaded" and the UV photons wasted. Again, see Len's prepub linked to above for some numbers.

Nicodem - 28-6-2010 at 00:24

Quote: Originally posted by Melgar

I do know for sure that if you bubble chlorine into toluene with some bromine/bromide in it, most of the gas coming out of it will be HCl, not Cl2, so it's definitely getting chlorinated. I'm hoping that this reaction, like the sulfuryl chloride/organic peroxide synth, chlorinates the methyl group exclusively and not the benzene ring. Since the bromine radical isn't as reactive, I'm hoping it only reacts with the methyl group before being displaced by chlorine, but I can't verify that this is the case.

Unfortunately, your experiment involves LiBr and in situ formed LiCl which both being relatively strong acids could catalyse the electrophilic chlorination. So without analysing the reaction products there is no way to say that such a reaction selectively gives benzyl chloride. Chlorine does not need much of a catalysis for electrophilic attack on toluene's aromatic ring, even traces of moisture catalyse this to some extent, so I would tend to believe that LiCl might be quite efficient as well. If the benzylic chlorination truly involves Br2/light as the radical initiator, then you should get the same result by just adding a drop of Br2 instead of LiBr and this way improve the chances of having a more chemoselective chlorination.

Melgar - 28-6-2010 at 10:33

Quote: Originally posted by Nicodem

Quote: Originally posted by Melgar

Unfortunately, I don't have any pure Br2, but you are certainly right that it should produce the same result. Of course, the reaction also produces large amounts of HCl, so it seems kind of pointless to go through the trouble of acquiring Br2 to avoid the formation of acidic salts. The reaction needs to be exposed to light in order to happen; dissolving more chlorine in it in the dark will turn it yellow, but as soon as it's exposed to light, the bubbles start again and it turns clear. It cannot be turned yellow by bubbling chlorine through it while the solution is exposed to bright light. I would not be surprised if the LiCl or H2O (LiBr is one of the most hygroscopic salts that exists, and there's bound to be some water in it) or especially the large amount of HCl that's generated by the reaction, was catalyzing ring chlorination, but the rate at which this happens is certainly much slower than the rate at which benzyl chloride forms. Ring chlorination would not be initiated photochemically, right?

Melgar - 28-6-2010 at 10:44

Quote: Originally posted by not_important

This is covered in Len's fine writeup here http://www.sciencemadness.org/talk/viewthread.php?tid=10490

X2 + hv => 2 X ·

X· + H-C(stuff) => XH + ·C(stuff)

X2 + ·C(stuff) => X· + X-C(stuff)

For X2 == Cl2 you need blue light or shorter wavelengths, for Br2 green and even yellow light will generate radicals. The radical chain propagates for quite a few cycles, interaction with the walls, collision with another radical, and various other things eventually terminate a chain. So some ongoing level of new radical generation is needed, generally sort sort of illumination; incandescent lamps, work, especially quartz-halogen, but the generate a lot of heat relative the the number of effective photons. Fluorescent, mercury, and metal-halid lamps all worh, clear bulb Hg lamps are the cheapest high intensity source.

For liquid phase halogenations of aromatics in more than a small volume, visible light is better than UV as the aromatic (toluene) absorbs the UV strongly enough that much of the volume is "shaded" and the UV photons wasted. Again, see Len's prepub linked to above for some numbers.

This is all true, however the reaction I've apparently discovered barely needs any light to sustain it compared to Len's reaction. A single LED from a flashlight can generate enough photons to sustain this reaction, if incandescent bulbs are too inefficient.

Len also apparently has access to WAY more equipment than I ever did. I'd really like to analyze the products with a spectrometer, but since I don't have access to one, my only recourse is to try and persuade other people to do it for me.

mnick12 - 28-6-2010 at 22:35

Another thing that speeds up free radical halogenations is heat, one of my books mentions the free radical chlorination of methane using UV or simply heating the mix to 100c. Also Meglar how dow make sure you only have benzyl chloride, and not benzal chloride or benzotrichloride?

not_important - 29-6-2010 at 00:17

Something to consider is that AFAIK the solubility of Cl2 in toluene is 1-2 moles/liter, and a mole of toluene is very roughly 100 cc, so you aren't going to halogenate more that 10% to 20% of the toluene (which does reduce the chances of making the di or tri chlorides)

If you were to take say 20 cc of toluene and run this procedure on it, then stir it with NaHCO3 solution for awhile, then warm to say 40 C and dropwise add KMnO4 solution with stirring until the pink colour persists, separate the aqueous and toluene layers, wash the toluene with a little cold water. Toluene that was halogenated on the methyl, assuming it's mostly monochlorinated, will have been hydrolysed to benzyl alcohol and then oxidised to benzoic acid, which sticks with the alkaline aqueous layer. Distill the toluene layer, chlorotoluenes have boiling points around 160 so it's pretty easy to distill just toluene. The amount of toluene collected gives you an estimate of how much was halogenated, the distillation residue with be chlorotoluenes and and likely any benzotrichloride that was formed. You can verify the presense of halogen in that fraction using a copper wire flame test; benzotrichloride could be checked for by adding some to alcoholic AgNO3 and heating and possibly letting sit for several days. The benzoic acid can be recovered from the aqueous solution by evaporation and treatment with HCl.

You could just fractionate the entire reaction mass, if you have a decent rig. Benzyl chloride boils about 18 C above the chlorotoluenes, so it has to have reasonable fractionation capabilities.

A GC plus known samples of the possible halogenated products might do a better job of detailing results than an IR, it can be difficult to discern low levels of other compounds if their spectra overlap too much.

Nicodem - 29-6-2010 at 03:26

I did a simple experiment regarding the possible role of bromides as (pre)catalysts for benzylic chlorination of toluene. For simplicity I used trichloroisocyanuric acid (TCCA) instead of chlorine. TCCA can be weighted and does not need to be generated externally like chlorine. Besides it is just as efficient at chlorinating the benzylic radicals so it should not make much difference. Instead of LiBr I used benzyltriethylammonium bromide (TEBAB) as the source for the bromide ion, because its cation should not be able to coordinate with TCCA (which would increase its electrophilicity and promote the undesired ring chlorination). I prepared a 0.5M solution of TCCA in toluene/ethyl acetate (1:1): 0.5M TCCA in PhMe/EA (thus a large excess of the substrate, while EA is only added to increase TCCA solubility). The reaction mixtures and conditions tested were:

A) negative control for radical reaction conditions:
4 mL 0.5M TCCA in PhMe/EA on direct sunlight (benzylic chlorination of toluene with TCCA is known by using benzoyl peroxide as radical initiator and heating, JOC, 35, 719–722, so it should work fine with sunlight as well, though not necessarily at room temperature)

B) negative control for test A:
4 mL 0.5M TCCA in PhMe/EA in the dark (electrophilic chlorination of toluene with TCCA does not proceed at any useful rate at room temperature and without acid catalysis, but a control is nevertheless useful)

C) trial reaction:
4 mL 0.5M TCCA in PhMe/EA + 27 mg TEBAB (5 mol%) on direct sunlight (the mixture was shaken for 5 min prior to light exposure, but not all TEBAB dissolved)

D) positive control for radical reaction conditions:
4 mL 0.5M TCCA in PhMe/EA + 37 mg t-butanol (25 mol%) on direct sunlight (t-BuOCl which forms via O-chlorination of t-BuOH is known to be very easily homoliticaly cleaved to the corresponding Cl* and t-BuO* radicals by either heat or light, so we can expect t-BuOH to efficiently catalyze the benzylic chlorination, hence considered a positive control)

E) negative control for test C:
4 mL 0.5M TCCA in PhMe/EA + 27 mg TEBAB (5 mol%) in the dark (to verify the need for light in reaction C)

These reactions mixtures were prepared in tapered glass vials and set either on direct sunlight or complete dark. Unfortunately TEBAB did not all dissolve and thus the reaction mixtures C and E remained heterogeneous. The reaction mixtures C and D soon begun to heat up when placed on sunlight while the reaction A remained at room temperature judging by touch. After 7 minutes cyanuric acid begun to precipitate from C and D practically at the same time and both become hot (estimated to 60-70 °C). At about 9 minutes the reaction mixture A suddenly heated up with such speed that the cap was blown off and 1/3 of the mixture boiled over. At 15 min all the reaction mixtures cooled back to room temperature, cyanuric acid precipitated, and thus the reactions were considered finished. Meanwhile the reactions B and E which were all the time in dark did not show any sign of reaction (no heating and no cyanuric acid precipitate).

HPLC analysis was performed on A, C, D and E reaction mixtures (MeCN/H2O (70:30 v/v), Nucleosil C18 reverse phase column, UV detection at 254 nm). Chromatograms A, C and D were practically identical: essentially two peaks, one smaller at 5.30 min and one more intense at 6.12 min. Some minor peaks could only be observed at retention times shorter than 4 min, but in that region only relatively polar compounds show up. All these reaction mixtures had the characteristic benzyl chloride smell. Reaction mixture E showed only one peak at 6.12 min even after 1:20 h of standing in the dark. Reaction B was thus not even analysed as it showed no sign of reaction either. HPLC analysis of pure toluene and benzyl chloride standards at the same conditions gave retention times of 6.12 min and 5.32 min respectively. Reaction mixture C will be analysed with 1H NMR when time will allow in order to additionally confirm the identity of the product and absence of ring chlorinated products (these could have the same retention time on HPLC).

Conclusions:

TEBAB and t-BuOH appear to catalyze the benzylic chlorination with TCCA already at room temperature. In the absence of catalyst the reaction mixture goes to a runaway mode which is an indication that the reaction rate increases more rapidly at higher temperatures than the catalyzed reactions. In all cases the only product detected chromatographically is benzyl chloride. In the absence of light no reaction occurs either in the presence or absence of TEBAB. Extreme care should be taken when designing a preparatory version of any of these reactions, particularly of the noncatalysed one (TCCA should be added in small portions to an excess of toluene!).

DJF90 - 29-6-2010 at 05:38

Nice experiment Nicodem!

Sedit - 29-6-2010 at 08:48

Thank you Nicodem very nice.
So in its essence Meglar suspicions have been confirmed correct and the addition of a Bromine ion does indeed promote the chlorination of toluene under better conditions then one normally deals with in such a reaction. I don't know if you remember sometime back myself having and issue with runaways using hypochlorite as the chlorine source and this lead to runaways that made me deam the reaction worthless from my point of view. But with this new infomation and the ability to avoid that runaway when a certine temperature is reached means I may once again have a go at reproducing a variation of the conditions you just laid out but using a bromine salt as the halogen carrier instead out of accessablity.

Magpie - 29-6-2010 at 11:51

Thank you, Nicodem, for such a nice and well thought out experiment. This sets a standard for us all. A proper experiment with controls and results is just so much more meaningful than endless speculation.

Anders Hoveland - 29-6-2010 at 12:00

Let me get this straight: heating to 100degC will generate chlorine radicals? (mnick12) Can you please give a reference?

"benzotrichloride could be checked for by adding some to alcoholic AgNO3 and heating and possibly letting sit for several days."
(not_important)
I thought that only organic bromine and organic iodine reacted in a precipitation reaction, not chloride. However, 3 chlorines on the same carbon is fairly electron withdrawing, so perhaps an electron could be donated to a chlorine through one of the double bonds/ delocalized bonding on the benzene, somewhat similar to hydrolysis of 1,1,1 trichlo acetone.

"Chlorine does not need much of a catalysis for electrophilic attack on toluene's aromatic ring, even traces of moisture catalyse this to some extent" (Nicodem) I thought chlorine was soluble in toluene, are you saying adding a drop of water to the mix will induce a reaction? Is the reaction rate very slow?

This is a fascinating reaction, though I am not sure it is exceptionally useful.

rrkss - 29-6-2010 at 12:49

Thanks Melgar for sharing, now I have another project to consider for my grignard reagent work that I am currently doing. Since this reaction makes benzyl chloride accessable to me, I am going to give it a shot.

woelen - 29-6-2010 at 13:00

Melgar, I have followed this thread and I can only say that you have excellent observation capabilities combined with the ability to make a connection with theory. Nicodem has done a nice experiment in a more controlled environment using the right equipment. This is the kind of hobby chemistry I really like. Keep up the good work. I also consider doing this reaction now, even although I am not really into organics.

Nicodem - 29-6-2010 at 13:43

Thanks.

Sedit, if you want to do a preparative reaction using TCCA, then just follow the literature procedure (the JOC paper cited above and attached in a few posts already) and thus use benzoyl peroxide as radical initiator. This is the classical Wohl-Ziegler reaction on toluene (interestingly, TCCA was used for the Wohl-Ziegler reaction already in their seminal paper - the example was allylic chlorination of cyclohexene). If instead of benzoyl peroxide you want to use sunlight or artificial light to initiate the reaction, then just heat up toluene to reflux and slowly add either very small portions of TCCA or use an addition funnel to slowly drip in an ethyl acetate solution of TCCA while irradiating with strong (sun)light. This way you can not have any runaways as you never have too much of the chlorinating reagent present (unless you get stupidly impatient with the addition rate). Perhaps it is optimal to use an addition funnel with TCCA in ethyl acetate so that while the addition progresses, the reflux temperature lowers down due to the lower bp of ethyl acetate. Use a large excess of toluene or else you will end up with a thick slurry of cyanuric acid (CA) and furthermore potentially get some PhCHCl2 and PhCCl3 side products (remember also that the stoichiometry of the reaction is 3PhMe + TCCA => 3PhCH2Cl + CA). Isolate by filtering off the precipitate, wash it with some toluene (or ethyl acetate or petroleum ether), and then fractionate the filtrate using a good distillation column.

If you are careful enough you might actually do all this without tears. Be careful especially when dismantling and cleaning the distillation apparatus. Otherwise, ammonia is good enough to quench benzyl chloride, though it takes quite some time.

Using quat bromides or t-BuOH is hardly of any use in a preparative reaction given the reaction proceeds also in their absence. The only benefit would be in decreasing the required reaction temperature, for example if one would want to do the reaction at room temperature. But since the reaction is highly exothermic, it will heat up anyway, so instead of removing heat via reflux condenser you would have to use a cold bath in order to maintain low temperatures. A few quat bromides are sold OTC as algaecides. Perhaps it would be of use to those who do not have a reflux condenser or any proper glassware, but otherwise I'm not sure if using any such catalysis has any other special advantage (like I said the chromatograms were identical for all reaction mixtures exposed to sunlight). This is however of much interest for research, yet from my experiments I am not able to conclude anything about the catalysis mechanism of TEBAB or of the LiBr catalysed reaction described by Melgar.

Melgar - 29-6-2010 at 13:51

Wow, thanks Nicodem! I've been trying to get access to the chem labs at my university so I could get enough data to potentially publish a paper, and this makes me much more hopeful. Plus, the professors may be more likely to hear me out if there's some (albeit informal) verification of my results.

I have actually done the reaction with TCCA and LiBr once already, but it generated all this white precipitate which I thought was probably cyanuric acid, but couldn't test to be sure. Thus, I used chlorine gas since it would make the products easier for me to identify. Still, it's great to know that the reaction can be done just as easily with TCCA.

My only test so far for more chlorinated toluene derivatives was to mix the result with an alcohol solution of NaOH. The benzyl chloride smell gradually disappeared, but there was no benzaldehyde smell that would have been indicative of the presence of benzal chloride. Since there was benzyl chloride, but no benzal chloride, I figured it was a safe bet to assume there was also no benzotrichloride.

As far as uses for this reaction, it could probably be used to make grignard reagents. Toluene wouldn't interfere with a Grignard reaction, so it wouldn't even need to be distilled beforehand. The HCl and chlorine would have to be given time to evaporate though. Although, if Nicodem's process is used, the cyanuric acid could just be allowed to settle out, thus minimizing the production of noxious gases.

Sedit - 29-6-2010 at 16:36

Quote:

If instead of benzoyl peroxide you want to use sunlight or artificial light to initiate the reaction, then just heat up toluene to reflux and slowly add either very small portions of TCCA or use an addition funnel to slowly drip in an ethyl acetate solution of TCCA while irradiating with strong (sun)light. This way you can not have any runaways as you never have too much of the chlorinating reagent present (unless you get stupidly impatient with the addition rate

I did manage to quell the runaway by slow addition AFTER the PhMe or PhMe2 in other test reached critical temperatures(PhMe2 has a lower runaway temperature) in the past but unless I read your post wrong you did seem to suggest that the bromine carrier did indeed allow the reaction to proceed at low temperatures with out the run away is this correct? I was also using impure Ca(OCl)2 as the hypochlorite so next attempts will all be TCCA since I have sourced it cheeply now and its purer.

I would like a set it and forget it method if you get my drift. Something I would be able to leave out in the sun for a few days with little worry of runaway or over chlorination. The halogen carrier appears as though it may quell the worrys of exploading bottles of toluene when things get hairy.

I feel IpOH may also work in place of t-BuOH as well when using hypochlorites as the chlorine source and there was a thread a while back discussing the rapid chlorintion of Toulene using Alkylhypochlorites but I was turned off by the sensitivity to explosion on exposure to light. If they work catalyticaly in this process then a small amount of IpOH could go along way and avoid the hazards of the infamous "chlorine bombs" You tube is littered with.

I eagerly away your data on ring chlorination products if any are formed.

not_important - 29-6-2010 at 16:50

Umm - aldehydes react in the presence of strong base. In the case of an aldehyde such as benzaldehyde, with no alpha hydrogens, you get the Cannizzaro reaction yielding benzyl alcohol and sodium benzoate. If the water content is low enough you could get the Tishchenko reaction, giving the corresponding ester.

I'd base the presence or absence of polyhalo compouds on the ratio of chlorine to toluene, in your case around 1:10 which with the reactions and reactants makes polyhalogenation unlikely.

Bubbling CO2 or N2 through the mix would remove HCl, as would adding a toluene solution of a tertiary base or better a tert-base ion exchange resin as it would be easier to filter off.

Melgar - 29-6-2010 at 20:18

Woelen - thanks! I started trying to teach myself chemistry about six months ago. Fascinating stuff! I like how everything we can see is made out of chemicals, and is capable of any number of reactions.

Sedit - if my past experience holds true for you, you'll only have to leave your vessel out in the sun for about 15 minutes.

not_important - I thought the Cannizzaro reaction was really slow at room temperature. Anyway, in that experiment, the ratio was probably higher than 1:10, since I continued chlorinating it after the reaction finished the first time. Nicodem's results seem to confirm that the polychlorinated products are negligible, since he used more chlorine than I did.

Nicodem - 30-6-2010 at 03:53

I was left with some 20 mL of the 0.5M TCCA solution in PhMe/EA and since I had no chlorotoluenes standards to verify their presence via HPLC, I added 0.25 mL CF3COOH to catalyse the electrophilic chlorination and left this reaction mixture standing overnight. The next day there was cyanuric acid precipitate and the HPLC analysis showed that the product(s) formed (presumably o- and p-chlorotoluenes) have the retention time of 7:46 min (there was only one broad peak which might be o- and p-chlorotoluene overlapping; traces of BnCl could also be seen but that was because I did not wrap the flask in aluminium foil to protect from light). I went back to check if there is anything at 7:46 min in the chromatograms I made yesterday and there was nothing obviously rising beyond the background, except for a negligible peak at 8:00 min in the reaction mixture A. So if any chlorotoluenes formed they formed in negligible amounts only.

1H NMR analysis of the intact reaction mixture (minus the insoluble CA) also confirmed this, as there were no unambiguously identifiable singlets for the o-chlorotoluene's and p-chlorotoluene's methyl groups (lit. values of the chemical shifts taken from SDBS are 2.34 and 2.29 ppm respectively). The two singlets that might belong to them were partially overlaping with the PhMe singlet so their integrals are misleading. But even if taken as they are, there should be less than some 5% of chlorotoluenes in relation to BnCl. No signals for PhCHCl2 could be found at all. (Edit: Actually, I just realized the tiny singlet at 6.67 ppm is from PhCHCl2. It molar ratio in relation to BnCl is 1 : 100. So the amount of dichlorination is pretty much negligible when using a threefold excess of toluene.)

The reaction conversion to benzyl chloride was estimated from the BnCl/PhMe ratio which was found to be 1 : 4.3 from 1H NMR spectra. The initial ratio of reagent vs. substrate was 0.5 mmol TCCA per 0.5 mL toluene (4.7 mmol). According to my calculation, the conversion was about 60%, but estimating conversion with NMR is not particularly accurate, especially this way and without any internal standard.

An interesting thing observed in the 1H NMR spectra is the set of signals that in my opinion belong to ethyl acetate chlorinated at the alpha ethyl position: CH3COOCHClCH3. This can only occur via radical chlorination, because the electrophilic chlorination of ethyl acetate could only give ethyl chloroacetate (ClCH2COOCH2CH3). I should go find the NMR spectral data in the literature to confirm the identity of this side product, but don’t have the will to do so now. So, for all those interested in the synthesis of acetaldehyde and its derivatives, here you have a potentially new photochemical route starting from a common solvent.

toluene_radical_chlorination_C.gif - 38kB

toluene_radical_chlorination_C.gif - 38kB

[Edited on 1/7/2010 by Nicodem]

Nicodem - 1-7-2010 at 01:19

Quote: Originally posted by Sedit

What is sold as calcium hypochlorite is actually Ca(OCl)Cl or CaCl2/Ca(OCl)2 in varying proportions. Ca(OCl)2 is not easy to come by. Anyway, "calcium hypochlorite" in this context is something else altogether when compared to chloroimides. You have a heterogenous mixture of an oxidant and a fuel that will only react once they reach a certain temperature and once they do they practically blow up. On the other hand, with TCCA you have a way more controlled media for the Wohl-Ziegler reaction as long as you do it properly.
What you seem to not properly understand is that a reaction has the same reaction enthalpy regardless if it is catalysed or not. The only thing that changes in the presence of a catalyst is the activation energy. In short, this means, that yes you can have the reaction run at a lower temperature when using a catalyst, but the amount of heat liberated will be the same as in the one without the catalyst. If you compare my experiments A with C and D you can see that already on a 2 mmol reaction scale the heat liberated was enough for a boil over in A. The only reason why C and D did not reach the boiling point is because they started earlier and slowly so that the heat of the reaction had 9 minutes more to dissipate. Now, imagine you would scale up by 100 times to an amount of 200 mmol TCCA! The surface to volume ratio certainly does not increase by a factor of 100, so in such a case where you would have a reaction scale of 200 mmol you would have the same runaway regardless of the presence or absence of whatever catalyst. You simply need to remove the heat of the reaction!

Quote:

I would like a set it and forget it method if you get my drift. Something I would be able to leave out in the sun for a few days with little worry of runaway or over chlorination. The halogen carrier appears as though it may quell the worrys of exploading bottles of toluene when things get hairy.

No, you would still get an exploding bottle, unless you would be able to inversely regulate the light exposure with the reaction temperature. In no case don't you just put it on direct sunlight! You keep forgetting that you are dealing with a highly exothermic radical chain reaction. These reaction can not be upscaled without severe modifications or run safely unless you have a slow addition of the limiting reagent. You can't just mix them together on a large scale, furthermore concentrated, and then lit with a radical initiator / light. Radical chain reactions don't have stable kinetics because the rate of the reaction depends not only on the activation energy, but is terribly dependent on the length of radical chain propagation. This very strongly depends on many conditions: temperature, light intensity, radical initiator concentration and properties, and even some things you can not control in any way or are pretty much unknown. So is it strange that they end up with a runaway if you just "set it and forget it"?
That's why the safest way to do it is by using Cl2. You limit the reaction by the chlorine gas inflow and don't have to worry about such things as flasks blowing up with lachrymogenic benzyl chloride.

Quote:

I feel IpOH may also work in place of t-BuOH as well when using hypochlorites as the chlorine source and there was a thread a while back discussing the rapid chlorintion of Toulene using Alkylhypochlorites but I was turned off by the sensitivity to explosion on exposure to light. If they work catalyticaly in this process then a small amount of IpOH could go along way and avoid the hazards of the infamous "chlorine bombs" You tube is littered with.

iPrOH might work but it would soon die out as it gets oxidized rapidly under the reaction conditions (forming acetone and HCl). I would avoid it. Why don't you try with just any tetraalkylammonium bromide? As far as I know, some are available OTC. Or just buy some tetrabutylammonium bromide (TBAB). It is cheap and you might need it for other experiments as well. But like I said, don't live in the illusion that any such additive will prevent a runaway if you premix all the reagents and expose them to sunlight. You need to use some way to remove the reaction heat, either via the reflux condenser in an "non-catalysed" reaction, or via a cold bath using a "catalyst". But in any case always have TCCA as the reaction limiting reagent to be added slowly.

kmno4 - 1-7-2010 at 04:07

Quote: Originally posted by Nicodem

This is however of much interest for research, yet from my experiments I am not able to conclude anything about the catalysis mechanism of TEBAB or of the LiBr catalysed reaction described by Melgar.

I do not believe this

Formation of BrCl (from Br2, formed from ionic bromide and Cl2) is rather obvious and it looks like Br radical is only chain initator. Something more about radical reactions Cl2/Br2/BrCl can be found here.
However, it seems that it is not "real" catalyst because it should be consumed during reaction.

[Edited on 1-7-2010 by kmno4]

Melgar - 1-7-2010 at 19:47

I had to research proton NMR before I could understand that last post, but now that I get most of it, it's quite fascinating. I am curious as to why you used ethyl acetate? Also curious as to how you'd use this reaction to get acetaldehyde. Personally, I'm partial to using chlorine gas, that way you can't have too much of the limiting reagent (since only so much chlorine will dissolve in toluene) and there's nothing to separate out afterwards.

I did notice a mist forming above my vial that I believe was the gas-phase version of this reaction whereby chlorine is radicalized by the action of UV light. I'm not sure if this would happen with TCCA, but if so, that could be where the benzal chloride is coming from. Or perhaps it just forms in small quantities anyway.

Anders Hoveland - 1-7-2010 at 22:30

I made bromo toluene in the way described, and reacted with AgNO3 in toluene to make hydroxy,nitrotoluene ( and some traces of PhCH2ONO2).

I think all the organic people are missing out by not checking my reactions in the Energetic section of this forum.
Check out "explosives from quinone". All the commentators there seem more interested in dangerous acids and explosions, than the actual organic chemistry required to make compounds that have any more complexity than basic TNT or nitroglycerine.

[Edited on 2-7-2010 by Anders Hoveland]

Satan - 2-7-2010 at 00:20

Quote: Originally posted by Anders Hoveland

I made bromo toluene in the way described, and reacted with AgNO3 in toluene to make nitrotoluene ( and some traces of PhCH2ONO2).

You mean AgNO2? Or is this some novel process of your invention?

[Edited on 2-7-2010 by Satan]

Anders Hoveland - 2-7-2010 at 00:29

I probably got methyl,hydoxy, nitro benzene. Where the hydroxy could be anywhere in relation the methyl group, and the nitro group is adjacent to either the hydroxy, or the methyl group. Obviously this creates several combination possibilities. Phenyl nitrate is unstable. Toluene nitrate would also be if the nitrate was no on the methyl.

I did not use silver nitrite. There are no nitronium ions that form either.

[Edited on 2-7-2010 by Anders Hoveland]

Nicodem - 2-7-2010 at 00:44

Quote: Originally posted by kmno4

Quote: Originally posted by Nicodem

This is however of much interest for research, yet from my experiments I am not able to conclude anything about the catalysis mechanism of TEBAB or of the LiBr catalysed reaction described by Melgar.

I do not believe this

Nice find! I was already afraid nobody would take the time to check the literature for anything related to the topic.
BTW, I did not say I can not come up with a hypothesis (I can always do that for just about anything, but that is cheap

), just that I'm "not able to conclude anything" (that is to prove any such hypothesis based on those experiments alone). One surprising observation from the NMR is that there is no BnBr formed (well, there is "something" at 4.44 ppm where PhCH2Br should be, but that something could be anything and it integrates for even less than the PhCHCl2 peak). If all the bromide from TEBAB got consumed into benzylic bromination, there should be some more BnBr formed (theoretically 1.6% in relation to BnCl, which is a bit too little for NMR and calls for a GC analysis). So, for some reason the intermediate species that allows the reaction to smoothly run already at room temperature is either highly selective for chlorination (like BrCl or more?) or something else happens. On the other hand it appears as if TEBAB is either consumed or made insoluble (note that not all dissolved!) as its cation can not be detected either (its NMR is also available at SDBS). So, there are still questions to be answered before giving a conclusion. One obvious thing that should be done is to try using some quat bromide that would completely dissolve and in a higher amount so to see if there is some BnBr formed or whatever is the fate of the bromide.

Quote: Originally posted by Melgar

I had to research proton NMR before I could understand that last post, but now that I get most of it, it's quite fascinating.

You better read about NMR as much as you can, because if you will ever choose organic chemistry, this analytical method will become your daily routine. About 20% of the time in the lab (or more) is spent running NMR scans and interpreting spectra. It is annoying but you can't do much without it. Chromatographic methods only give you products separation, but without standards they can not be used for their identification. It is the spectroscopic methods which give the answer, and single crystal XRD the one that gives the final answer (if needed).

Quote:

I am curious as to why you used ethyl acetate? Also curious as to how you'd use this reaction to get acetaldehyde.

TCCA does not dissolve very well in plain toluene, so I wanted to use a cosolvent to get a homogeneous 0.5M solution. If you want to draw conclusions from experiments and compare different conditions, it is best if the reaction mixtures are homogeneous. It is also important to know the amount of reactants and that is why I could not use Cl2. Of course, for a preparative reaction it does not matter if TCCA has poor solubility in toluene, because you always can and must use stirring to compensate. I had only few choices of which cosolvent to use: acetonitrile, ethyl acetate or acetone, as these are the ones which dissolve TCCA very well and do not easily react with it. Alcohols get O-chlorinated by TCCA and the so formed hypochlorite esters are extremely photolabile (see experiment with t-BuOH added!). Standard polar aprotic solvents like DMF, DMSO or NMP react with TCCA (DMSO extremely rapidly and exothermically) while inert chlorinated solvents like CH2Cl2 do not dissolve TCCA any better than toluene. Acetone was out of play because it is the more reactive than MeCN or EtOAc toward TCCA, and because I was afraid it would quench the radical chain propagation. For some reason I expected MeCN is more prone toward radical chlorination than EtOAc so I chose this later (luckily I was wrong

). In fact there are very few reports of such radical chlorination of ethyl acetate in the literature and such a synthesis of MeCOOCHClMe is very attractive. This compound easily hydrolyses to acetaldehyde and it can also be used as a masked acetaldehyde (a vinylidene synthon) in several reactions where acetaldehyde or paraldehyde is otherwise used. This means that the radical chlorination of ethyl acetate with TCCA gives an excellent opportunity to prepare this useful reagent that might be used in several interesting reactions or as a starting material for the preparation of acetaldehyde (hydrolysis), acetyl chloride (retro-acylation), vinylidene acetals (alcoholysis) or vinyl acetate (elimination). It also opens a new potential route to aldehydes starting from esters. Several other compounds could also be radically chlorinated this way to give useful compounds. Ethers are already known to be very easily alpha-chlorinated by TCCA, but many other compounds could be as well, perhaps even acetonitrile to give chloroacetonitrile and so on.

Quote:

Personally, I'm partial to using chlorine gas, that way you can't have too much of the limiting reagent (since only so much chlorine will dissolve in toluene) and there's nothing to separate out afterwards.

Yes, like I explained to Sedit, if you are unsure what you are doing, it is best to use chlorine on larger scale preparative reactions. Though, generating large amounts of chlorine is also hazardous, but again only if you don't know what you are doing. The bottom line is, that in every case you should know what you are doing and here lies the problem. I have read here about people designing experiments on a large scale (why?) without even doing a literature search first, let alone thinking of what can go wrong, so I'm afraid someone will do something stupid again.

Quote:

I did notice a mist forming above my vial that I believe was the gas-phase version of this reaction whereby chlorine is radicalized by the action of UV light. I'm not sure if this would happen with TCCA, but if so, that could be where the benzal chloride is coming from. Or perhaps it just forms in small quantities anyway.

Where the benzylic proton of benzylidene chloride should be (at 6.695 ppm according to SDBS) there is almost nothing as you can see above, so only trace amount of PhCHCl2 formed. But you have to keep in mind that I used a threefold excess of toluene and that TCCA could be more selective toward monochlorination than Cl2.

[Edited on 2/7/2010 by Nicodem]

Melgar - 2-7-2010 at 14:20

Check out this paper:

http://dx.doi.org/10.1016/0039-9140(58)80019-3

Apparently, BrCl acts exclusively as a brominator. I suspect this reaction produces only chlorinated versions of organobromides that are formed via free-radical halogenation. Thus, the bromine forms radicals, brominates the hydrocarbon, and is replaced by chlorine, which forms more BrCl. For example:

RH + Br* + Cl* -> RBr + HCl
RBr + Cl2 -> RCl + BrCl

Fleaker - 2-7-2010 at 16:32

@ Nicodem

Forgive me if this sounds like shameless sycophancy but I'm seriously convinced that you direct some pharma company's research division. Your posts set a standard for what should be reported on this board.

I agree with kmno4 that you must look for bromide present in the product via GC; that would settle whether or not bromide had a catalytic role.

psychokinetic - 2-7-2010 at 19:15

Quote: Originally posted by Fleaker

@ Nicodem

Your posts set a standard for what should be reported on this board.

While we all can't quite achieve such a level of awesome, those who could and can sure do make an awesome contribution to our knowledge. Thanks Nicodem.

jon - 7-7-2010 at 15:12

hey i have a question for you guys seeing how i'm dumber than a box of rocks, i was curious, if one were to do the same chlorination on a toulene with a ring activating group meta to the benzylic carbon for instance 2-hydroxy toulene for example.
would this increase the likelihood of side reactions such as ring halogenation to any great extent?
seeing how the ring is a bit more activated and there is greater electron density locant to the benzylic carbon would this also increase the possibility of dichlorination to the benzal chloride?
excuse my stupidity i'm just looking to get spoonfeed.

manimal - 19-8-2010 at 22:32

I will chime in my appreciation for Nic's experimentalism.

I am still on about the topic of styrene. What would be the result of free-radical chlorination of styrene with TCCA? I note that there is no Cl- ion produced in chlorinations w/ TCCA since all Cl atoms are in the +1 state, so a vacinal chloride would not form under anhydrous conditions, but this could be inconsequential to the addition of chlorine at all. Maybe the chlorine would add to the β carbon first forming chlorostyrene, then additional additions on either the α or β carbon forming 1,2 dichlorostyrene or 2,2 dichlorostyrene.

manimal - 19-8-2010 at 22:37

Quote: Originally posted by jon

hey i have a question for you guys seeing how i'm dumber than a box of rocks, i was curious, if one were to do the same chlorination on a toulene with a ring activating group meta to the benzylic carbon for instance 2-hydroxy toulene for example.
would this increase the likelihood of side reactions such as ring halogenation to any great extent?
seeing how the ring is a bit more activated and there is greater electron density locant to the benzylic carbon would this also increase the possibility of dichlorination to the benzal chloride?
excuse my stupidity i'm just looking to get spoonfeed.

Cresols do not undergo free radical chlorination easily. 30 seconds with google turned up at least one patent for a workaround chlorination of cresol esters: http://www.google.com/patents/about?id=E4guAAAAEBAJ.

Nicodem - 20-8-2010 at 00:26

Quote: Originally posted by manimal

I am still on about the topic of styrene. What would be the result of free-radical chlorination of styrene with TCCA? I note that there is no Cl- ion produced in chlorinations w/ TCCA since all Cl atoms are in the +1 state, so a vacinal chloride would not form under anhydrous conditions, but this could be inconsequential to the addition of chlorine at all. Maybe the chlorine would add to the β carbon first forming chlorostyrene, then additional additions on either the α or β carbon forming 1,2 dichlorostyrene or 2,2 dichlorostyrene.

TCCA is too electrophilic for radical halogenation of styrene (is radical chlorination of styrene even possible?), or styrene is too nucleophilic for TCCA, depends on the perspective. It is the same problem as in trying to radically chlorinate cresols. In principle electrophilic chlorination of styrene with chloroimides should give beta-chlorostyrene, but in practice and with CH2Cl2 as solvent, the reaction of styrene and TCCA results in styrene oligomerization. The oligomerization is most likely electrophilic: in the sense that the initially formed PhCH<sup>+</sup>-CH2Cl adds on a styrene and so on until carbocations dye out by H<sup>+</sup> elimination. But I was never that interested to run a GC-MS on the product mixture to see if the oligomers start with PhCH=CH- and end with -CH2Cl groups. I only did a few analyses to see if my desired product was there as the major product or not - oligomers don't interest me (though I remember that they nicely separated on TLC, so they could be isolated). But this was years ago when I still worked on halogenations and I do not remember much of the details.

Melgar - 24-8-2010 at 05:41

Yeah, free radicals tend to cause styrene to polymerize. Plus, the evolved HCl can add to alkenes.

manimal - 26-8-2010 at 23:46

Quote: Originally posted by Nicodem

Quote: Originally posted by manimal

That makes sense, since both polymerization and chain chlorination are radical-promoted.

US2803675 (http://www.google.com/patents/about?id=6VpoAAAAEBAJ) makes use of in-situ chlorine generated from HCl and a chlorine-furnishing substance to side-chain chlorinate styrene in an inert solvent. I imagine it works by using a highly acidic(excess HCl) aqueous phase that ensures that HClO+HCl <--> Cl2+H2O stays far to the right therefore minimizing the amount of HClO floating around that might hydroxychlorinate or ring chlorinate the styrene.

Nicodem - 27-8-2010 at 00:39

I did not say that styrene polymerized. I said it oligomerized and I did not say it was a radical promoted oligomerization. I don't really remember much as this was a long time ago (relative to my poor memory capabilities at least) and I don't know where my lab diary of that period is or on which backup CD did I put the spectroscopic data, but the major oligomers were of n<7 or thereabout, which is more consistent with the carbocationic polymerization. I took care of preventing any radical polymerization (low temperature, absence of light, highly electrophilic source of "positive chlorine", etc.). I do not know what would happen and what would be the major products in the radical chlorination of styrene as radical halogenations were never part of my research or interest. I just described what happened in my attempt in reacting styrene with TCCA under the described conditions (my goal was an efficient synthesis of beta-chlorostyrene). I would assume that for an efficient radical chlorination of styrene the temperature would have to be high enough for the polymerization to be reversible enough and radical chlorination not (I would guess something like about 200 °C or more!), else the major product would be terminally chlorinated polystyrenes and not much else. However, I'm not really able to conceive a reagent being non-electrophilic at such conditions, yet being a source of Cl* radicals (perhaps CCl4?). But since I do not know enough about radical halogenations, don't take my words too seriously.

Edit: Quickly checked the patent you linked. That is not a radical chlorination of styrenes! It just "normal" electrophilic chlorination. The electrophilic addition of Cl2 to styrenes gives 1-aryl-1,2-dichloroethanes and the electrophilic atack followed by the deprotonation of the chloronium ion gives the other type of products, beta-chlorostyrenes. And yes, the biphasic system prevents the formation of the halohydrins and the presence of HCl keeps the concentration of HClO and other oxychloro species low (as it reduces them).

[Edited on 27/8/2010 by Nicodem]

manimal - 30-8-2010 at 17:51

Quote: Originally posted by Nicodem

I did not say that styrene polymerized. I said it oligomerized and I did not say it was a radical promoted oligomerization. I don't really remember much as this was a long time ago (relative to my poor memory capabilities at least) and I don't know where my lab diary of that period is or on which backup CD did I put the spectroscopic data, but the major oligomers were of n<7 or thereabout, which is more consistent with the carbocationic polymerization. I took care of preventing any radical polymerization (low temperature, absence of light, highly electrophilic source of "positive chlorine", etc.). I do not know what would happen and what would be the major products in the radical chlorination of styrene as radical halogenations were never part of my research or interest. I just described what happened in my attempt in reacting styrene with TCCA under the described conditions (my goal was an efficient synthesis of beta-chlorostyrene). I would assume that for an efficient radical chlorination of styrene the temperature would have to be high enough for the polymerization to be reversible enough and radical chlorination not (I would guess something like about 200 °C or more!), else the major product would be terminally chlorinated polystyrenes and not much else. However, I'm not really able to conceive a reagent being non-electrophilic at such conditions, yet being a source of Cl* radicals (perhaps CCl4?). But since I do not know enough about radical halogenations, don't take my words too seriously.

Edit: Quickly checked the patent you linked. That is not a radical chlorination of styrenes! It just "normal" electrophilic chlorination. The electrophilic addition of Cl2 to styrenes gives 1-aryl-1,2-dichloroethanes and the electrophilic atack followed by the deprotonation of the chloronium ion gives the other type of products, beta-chlorostyrenes. And yes, the biphasic system prevents the formation of the halohydrins and the presence of HCl keeps the concentration of HClO and other oxychloro species low (as it reduces them).

OK, Nic.

What do you suppose is the point of the solvent in the "normal" chlorination of styrene with Cl? Most procedures use CCl4 or similar inerts, unlike in chlorinations of toluene.

Nicodem - 31-8-2010 at 00:54

Quote: Originally posted by manimal

What do you suppose is the point of the solvent in the "normal" chlorination of styrene with Cl? Most procedures use CCl4 or similar inerts, unlike in chlorinations of toluene.

Can't say anything about what solvent would be ideal for your application since you told absolutely nothing about it. Solvents are certainly not something you chose on a general basis.

If you want to prepare 1-phenyl-1,2-dichloroethane by using Cl2 via electrophilic addition (I guess this would be the most "normal" chlorination possible), then CH2Cl2 would be the most practical choice, but you can also use other saturated halogenated solvents, ethyl acetate or other relatively non-nucleophilic aprotic solvents. I would suggest reverse addition: adding styrene to a dichloromethane solution of Cl2 to minimize carbocationic oligomerization. Some beta-chlorostyrene will nevertheless form as the reaction is not very selective, but a good distillation column should separate these as well as leave the oligomers behind.

I find it surprising that you say CCl4 used as that would be the last choice of a solvent for an electrophilic chlorination (CCl4 is avoided in synthesis unless required). Time ago someone else on this forum also said that he saw CCl4 being used for electrophilic chlorinations, but it turned out he confused radical chlorinations with electrophilic ones (CCl4 is used as a solvent only for radical halogenations as it is inert toward Cl* and Br* radicals).

manimal - 31-8-2010 at 11:40

Quote: Originally posted by Nicodem

Yes, that is apparently true in general, but styrene is an exception in that it chlorinates well in CCl4. I have also seen a procedure that uses hexane.

I don't really have an 'application', I'm merely curious about the reaction dynamics, e.g., whether the purpose of the inert solvent is to act as a chlorine carrier, to minimize polymerization/ring chlorination, or what have you.

Nicodem - 7-9-2010 at 07:38

Quote: Originally posted by manimal

Yes, that is apparently true in general, but styrene is an exception in that it chlorinates well in CCl4. I have also seen a procedure that uses hexane.

It is not about styrene being an exception, it is about solvent issues changing with time. CCl4 does indeed dissolve Cl2 and is inert under the reaction conditions, so it can be used for the electrophilic addition of chlorine on alkenes. The problem is thus not in the reaction not working in CCl4, but the properties and availability of this solvent. Once it used to be a fairly standard and relatively commonly used solvent (though it is too non-polar for most uses as it poorly dissolves many organic compounds, at least relative to CHCl3 and CH2Cl2). Times changed a lot in the last few decades. Just as an illustration, officially I'm not allowed to have a bottle of CCl4 in the lab where I work (though I unofficially do), even though nobody cares if much more toxic, carcinogenic, hazardous or whatever chemicals are there. Ordering it from chemical suppliers also become a challenge in bureaucracy, or so I was told by the person who does the ordering. Believe me that using CCl4 for something as trivial as the addition of Cl2 on alkenes, where there are plenty of solvents that can be used instead, is just terribly weird and inappropriate (that's what I meant with "CCl4 is avoided in synthesis unless required"). Just check the date of publication of the paper where you read about styrene chlorination in CCl4. I don't know the reference, but I bet it is at least two or three decades old.

Even for the radical chlorination other solvents will soon put CCl4 obsolete. There is a number of solvents that could be used instead for specific methods of radical halogenations, though none is equally general as CCl4. As you saw upthread, in the experiment of radical chlorination of toluene with TCCA, ethyl acetate was uses as the cosolvent. You can look at this experiment also as some sort of a competitive reaction experiment. Taking it this way, the results indicate the rate of toluene radical chlorination is about 7 times faster than the rate for ethyl acetate. So, as you see, in this case even ethyl acetate is somewhat inert though still useless as a true solvent (if using a more diluted toluene in ethyl acetate, then the ratio between BnCl to MeCOOCHClMe would drop to useless levels).

chemrox - 15-11-2012 at 17:18

I did a search for chlorination using TCCA and ended up here. My project is to make phenethyl chloride and I have a number of chlorinating agents including TCCA, SOCl2, PCl3, PCl5 and HCl. I don't seem to have ZnCl and don't want to make any. I was wondering if TCCA had been used by someone here for chlorinating an alcohol and if so how well it worked? I think Sauron was a big fan of the stuff and dropped off some papers on it. They got corrupted on an older drive and haven't been replaced yet in my system.

tetrahedron - 15-11-2012 at 17:30

if i'm not mistaken TCCA is a source of chlorine, like SO₂Cl₂, thus no good for your purpose.

SM2 - 15-11-2012 at 18:03

so you guys didn't know that chlorinating toluene is the usual (old) bread and butter, standard synth for benzyl chloride? really?!

edit (next day),

there MUST be something I'm missing because a lot of you on this thread are way more knowledgeable in the field than I.

Saw the catalyst, kinda gleaned through everything. It's just that, to me, it looked like the standard route.

[Edited on 16-11-2012 by Fennel Ass Ih Tone]

Nicodem - 1-4-2017 at 05:32

Mild Aliphatic and Benzylic Hydrocarbon C–H Bond Chlorination Using Trichloroisocyanuric Acid
Sascha H. Combe, Abolfazl Hosseini, Alejandro Parra, and Peter R. Schreiner
J. Org. Chem., 2017, 82, pp 2407–2413
DOI: 10.1021/acs.joc.6b02829

Quote:

Abstract: We present the controlled monochlorination of aliphatic and benzylic hydrocarbons with only 1 equiv of substrate at 25–30 °C using N-hydroxyphthalimide (NHPI) as radical initiator and commercially available trichloroisocyanuric acid (TCCA) as the chlorine source. Catalytic amounts of CBr4 reduced the reaction times considerably due to the formation of chain-carrying ·CBr3 radicals. Benzylic C–H chlorination affords moderate to good yields for arenes carrying electron-withdrawing (50–85%) or weakly electron-donating groups (31–73%); cyclic aliphatic substrates provide low yields (24–38%). The products could be synthesized on a gram scale followed by simple purification via distillation. We report the first direct side-chain chlorination of 3-methylbenzoate affording methyl 3-(chloromethyl)benzoate, which is an important building block for the synthesis of vasodilator taprostene.

Attachment: phpnl49TW (317kB)
This file has been downloaded 898 times

AvBaeyer - 1-4-2017 at 17:48

Nicodem,

Thanks for the link to the paper. It has some interesting mechanistic discussion but does not seem to add much to synthetic methodology. It would have been much more useful to have information about the material balance from the various reactions. Despite it having some interesting information, I am surprised the paper made it into JOC. I say this as a former JOC referee.

AvB

Melgar - 1-4-2017 at 22:07

Huh. This is just my reaction again, but instead of using a bromide salt as the bromine carrier, it uses an organobromide. The exact same mechanisms would be in place still; the only difference is that chlorine is replacing bromine in a covalent rather than ionic bond. It would still form bromine monochloride, which would be capable of forming radicals in the presence of longer-wavelength light (in my experience, blue light works) than chlorine on its own. And the bromine atoms would still diffuse throughout the solution, enabling the initiation of radical chains throughout a much larger volume, rather than just in the first few mm.

I'm curious what your thoughts on this are, Nicodem, because they're obviously greatly overcomplicating this reaction. Heck, the reason Cu(OAc)2 seems to help, could be that it selectively lets blue light through, There doesn't seem to be any indication this reaction was done in darkness, or that there was any attempt to control ambient light. The researchers seem to be surprised by the behavior of the system in several places, but it's not surprising at all if it's just an overly-complex variant on the reaction that I stumbled on and you tested.

I wonder if there would be any value in trying to publish a paper on this reaction now?

CuReUS - 2-4-2017 at 03:42

Quote: Originally posted by Nicodem

If you want to prepare 1-phenyl-1,2-dichloroethane by using Cl2 via electrophilic addition,then CH2Cl2 would be the most practical choice.I would suggest reverse addition: adding styrene to a dichloromethane solution of Cl2 to minimize carbocationic oligomerization. Some beta-chlorostyrene will nevertheless form as the reaction is not very selective, but a good distillation column should separate these as well as leave the oligomers behind.

I didn't know that adding halogen across the double bond could give b-haloalkenes as side products,so I checked it out.Apparantly,this is a significant side reaction(41%) in this type of reaction

http://www.tandfonline.com/doi/abs/10.1080/00397919808007018 (the reaction conditions are not same though)
also TCCA chlorinations aren't as easy as they seem

-https://youtu.be/khfYvY_Gczk?t=511 (although nurdrage didn't use ethyl acetate as co-solvent like nicodem suggested)

clearly_not_atara - 4-4-2017 at 12:32

Quote:

TCCA chlorinations aren't as easy as they seem

-https://youtu.be/khfYvY_Gczk?t=511 (although nurdrage didn't use ethyl acetate as co-solvent like nicodem suggested)

NurdRage is trying to perform a Friedel-Crafts chlorination of the arene whereas Nicodem linked a radical chlorination of the side-chain. Different reactions.

I don't know if there's a good way for FeCl3 to coordinate to TCCA to generate the electrophile needed for a Friedel-Crafts reaction. I'm having trouble picturing how such a complex would form. In Nicodem's paper above, using GAA as solvent with TCCA promoted Friedel-Crafts reactions at the expense of radical chlorination, but the yields of chlorotoluenes were still poor. IIRC TCCA decomposes in GAA to generate acetyl hypochlorite which is the true halogenating agent. This explains why F-C reactions only occur in GAA and not in other solvents. I'm tempted to say that to perform a F-C halogenation with TCCA, GAA is the recommended solvent.

However, TCCA can certainly form radicals, since the N-Cl bond is weak and attacked by light. Also, once chlorination has started, TCCA reacts with the byproduct HCl to generate Cl2, which is a much better radical former and which is probably the real chlorinating agent in the usual TCCA/toluene reaction. When F-C chlorination is tried in an aprotic solvent, the formation of Cl2 is a bad thing because it is less electrophilic than the chloroimine.

[Edited on 4-4-2017 by clearly_not_atara]

Σldritch - 5-4-2017 at 01:30

TCCA should inhibit Friedel-Crafts chlorination by complexing to Ferric chloride. NaDCC complexes with copper to form sodium copper dichloroisocyanurate so i do not see why TCCA would not do something similar with Ferric Chloride which is a much stronger lewis acid.

Also the cyanuric acid formed in the reaction might precipitate as ferric cyanurate but im not so sure about that.

CuReUS - 5-4-2017 at 10:17

Quote: Originally posted by clearly_not_atara

NurdRage is trying to perform a Friedel-Crafts chlorination of the arene whereas Nicodem linked a radical chlorination of the side-chain. Different reactions.

I know they are different reactions,but the separation step has to be done in both,right ? I was referring to the separation step which is the difficult part of a TCCA reaction(from the nurdrage video)

clearly_not_atara - 5-4-2017 at 10:28

Quote: Originally posted by Σldritch

TCCA should inhibit Friedel-Crafts chlorination by complexing to Ferric chloride. NaDCC complexes with copper to form sodium copper dichloroisocyanurate so i do not see why TCCA would not do something similar with Ferric Chloride which is a much stronger lewis acid.

Also the cyanuric acid formed in the reaction might precipitate as ferric cyanurate but im not so sure about that.

All Friedel-Crafts byproducts (well, almost all) inhibit the Lewis acids responsible for reaction. For example, when methyl chloride reacts with benzene, the byproduct is HCl, which reacts with AlCl3 to form H+ and AlCl4-, the latter ion being inactive as a Lewis acid. When trichloroisocyanuric acid transfers Cl+ to benzene, the byproduct is dichloroisocyanurate, which should react with AlCl3 to give an aluminium complex.

Note that NaDCC complexes with copper ions primarily through the negative charges on the N which is "missing" its chlorine relative to TCCA.

However, in order for that to happen, the AlCl3 has to form the FC-electrophile by complexing to TCCA and weakening the N-Cl bond to generate [Cl+], which reacts with benzene in the first place. It's this step that I'm not so sure about (but I guess it could be ok). That the byproduct inhibits AlCl3 is a foregone conclusion; that's why such high catalytic loadings are always used for FC reactions.

If instead you use GAA as a cosolvent, the AlCl3 can form a complex with acetyl hypochlorite which is an active FC electrophile (in fact AcOCl is so reactive it may not need a Lewis acid), but as usual the byproduct acetate will form a complex and inhibit the catalyst.

EDIT: searching tells me H2SO4 is the preferred solvent for ring chlorination, in fact. I assume this works by a similar mechanism.

Quote:

I know they are different reactions,but the separation step has to be done in both,right ?

Oh, yeah. The papers with TCCA use a solvent, whereas NurdRage uses neat toluene and tries to chlorinate most of it, which means that the TCCA-solvent ratio is large enough to make distillation a pain.

The paper uses CH2Cl2 as a cosolvent. I think that in their case the DCCA (byproduct) dissolves in the CH2Cl2 and precipitates as slightly larger and less colloid-like particles when the CH2Cl2 is distilled off which makes it easier for them to distill the products. Since CH2Cl2 will distill off much faster than toluene, it doesn't seem likely that it's still there at the end of the distillation. so it must make a noticeable difference in the consistency of the post-reaction mixture. It's also possible that the presence of FeCl3 serves to make distillation harder as this will complex with other ions and these complexes could increase the viscosity of the solution.

EtOAc cosolvent has been reported. Acetone is also apparently a solvent for TCCA, although the production of even small amounts of chloroacetone is very undesirable.

[Edited on 5-4-2017 by clearly_not_atara]

[Edited on 5-4-2017 by clearly_not_atara]

Nicodem - 10-4-2017 at 07:08

Quote: Originally posted by clearly_not_atara

I don't know if there's a good way for FeCl3 to coordinate to TCCA to generate the electrophile needed for a Friedel-Crafts reaction. I'm having trouble picturing how such a complex would form. In Nicodem's paper above, using GAA as solvent with TCCA promoted Friedel-Crafts reactions at the expense of radical chlorination, but the yields of chlorotoluenes were still poor. IIRC TCCA decomposes in GAA to generate acetyl hypochlorite which is the true halogenating agent. This explains why F-C reactions only occur in GAA and not in other solvents. I'm tempted to say that to perform a F-C halogenation with TCCA, GAA is the recommended solvent.

First of all, calling an electrophilic aromatic chlorination a "F-C halogenation" makes no sense. Friedel-Crafts reactions are something else.
I don't know why you say that acetic acid is "the recommended solvent". I would expect a diminished regioselectivity when using it as a solvent. Certainly an inert solvent like dichloromethane is a better choice, as in principle, direct electrophilic chlorination with TCCA should give better para-selectivity (though obviously, even under best of circumstances, you can't get much more than 1 : 1 para/ortho selectivity on toluene). Using sterically less demanding electrophiles (such as AcOCl) is going to reduce para selectivity. See the solvent effect evaluation in DOI: 10.1007/s00706-014-1383-6 to see how the nature of the in situ electrophile formed from TCCA affects the regioselectivity on anisole (not really comparable to toluene, but it gives you some idea).

Quote:

However, TCCA can certainly form radicals, since the N-Cl bond is weak and attacked by light. Also, once chlorination has started, TCCA reacts with the byproduct HCl to generate Cl2, which is a much better radical former and which is probably the real chlorinating agent in the usual TCCA/toluene reaction. When F-C chlorination is tried in an aprotic solvent, the formation of Cl2 is a bad thing because it is less electrophilic than the chloroimine.

There is no HCl as a byproduct in the Wohl-Ziegler reactions with TCCA. And TCCA was among the haloimides used in the seminal article where this reaction was first reported (almost a century ago). If I remember correctly, it was reported for a reaction with cyclohexene. It was only reported much later for a benzylic chlorination of toluene.

Where do you get that Cl₂ is a "much better radical former" than TCCA?

And that Cl₂ is less electrophilic than the chloroimide? On the contrary. TCCA will not significantly react with toluene in the absence of acid catalysis (I used CF₃COOH in an example somewhere in this thread, but any non-oxidizable acid of suitable strength will do - using Al or Fe chlorides does not sound like the first option to try, given that they are likely to be oxidized by TCCA). Chlorine is much more reactive.

Σldritch - 5-5-2017 at 05:14

I decided to sloppily attempt the reaction using BCDMH as a catalyst. I left it out in the sun for a while and the solution of TCCA in toluene took on a slightly brown color and some, what i presume is cyanuric acid precipitated, but the smell only changed a tiny bit. Benzyl chloride is supposed to be a strong lachrymator, so what happened?

My TCCA is pretty impure, it is supposed to contain a few percent boric acid and Copper Sulfate but i do not see why that would affect it.

Melgar - 6-5-2017 at 05:06

Since this is a chain reaction, it can be very sensitive to certain contaminants. If your TCCA has significant impurities, I recommend using it as a source of chlorine gas instead, and bubbling that into toluene.

Metacelsus - 10-6-2017 at 03:55

I would assume so, since it's used as an initiator for polymerization. Of course, you should be careful with it.

OrganoLeptic - 4-9-2017 at 12:53

I would like to thank the posters in this thread for providing useful information (especially Nicodem).

Based on your comments and work I prepared benzyl chloride.
Toluene served both as the solvent and reactant.
Finely powdered trichloroisocyanuric acid (TCCA) was the source of the chlorine radicals.
Sunlight was used as the radical initiator.
My final yield was 28% based on the TCCA used.

You can check out the synthesis here:
https://www.youtube.com/watch?v=G_4kfY0dR2A

I have a two part question I was not able to find an answer to:
1) Do toluene and benzyl chloride form an azeotrope?
2) If they do what is its composition and boiling point?

[Edited on 4-9-2017 by OrganoLeptic]

Melgar - 15-10-2017 at 15:22

Quote: Originally posted by OrganoLeptic

I would like to thank the posters in this thread for providing useful information (especially Nicodem).

Based on your comments and work I prepared benzyl chloride.
Toluene served both as the solvent and reactant.
Finely powdered trichloroisocyanuric acid (TCCA) was the source of the chlorine radicals.
Sunlight was used as the radical initiator.
My final yield was 28% based on the TCCA used.

You can check out the synthesis here:
https://www.youtube.com/watch?v=G_4kfY0dR2A

I have a two part question I was not able to find an answer to:
1) Do toluene and benzyl chloride form an azeotrope?
2) If they do what is its composition and boiling point?

[Edited on 4-9-2017 by OrganoLeptic]

I typed "toluene benzyl chloride azeotrope" into Google, and discovered that indeed they do have an azeotrope, and if you want to separate them, your best bet is to add isopropanol, distill the toluene/isopropanol azeotrope, then distill benzyl chloride once the toluene is all gone. You really ought to vacuum distill this stuff, since impure benzyl chloride can polymerize when heated, presumably due to some sort of Friedel-Crafts alkylation.

I did this reaction again recently, and I can add some more information for anyone attempting to replicate this. First, the reaction goes much better if you dry your chlorine before bubbling it into the toluene. Silica gel and calcium chloride have both been confirmed to work. Otherwise, the HCl that's produced as a byproduct won't ever leave the vessel, because of its attraction to the water. This tends to manifest itself as slight cloudiness in the reaction vessel, and a sluggish reaction.

Second, if you have elemental bromine and prefer to use that, you don't need to add it in liquid form, Even a drop is probably too much. Rather, you can just remove the cap from your bottle of bromine, and tip it sideways a bit directly over the toluene. The orange vapors should fall directly down and be absorbed. 3-5 seconds of this should be enough.

I've gotten this to work with a "soft white" LED lightbulb, even. With bromine present as a catalyst, it seems as though the important wavelength range of light would be whatever can cause bromine monochloride to disassociate, but is transmitted by chlorine.