Sciencemadness Discussion Board

A TCCA chlorination safety warning

Polverone - 5-8-2008 at 09:23

I recently received the message I'm quoting here via email:

Quote:

Dear Moderator,

I greatly admire your forum and spend many hours browsing on it on a regular basis.
I would like to become a member but cannot as I have no email. This is not due to any shortcoming but of choice: I find life has more than enough stresses without adding to them. Please note this is not technophobia-snail mail is equally repugnant to me. (This is being conveyed on my wife’s email. I would ask that this fact be respected and while one or two communications are fine, I don’t want this to become my email by proxy.)

Following a mishap with a prep, I feel obliged to make this communication as it might prevent your members or guests being injured or killed.

Desiring a small quantity of chlorobenzene and having pure benzene and TCCA to hand, I decided to follow the prep outlined by ‘axt’ in the pentryl thread. This prep is somewhat short on detail in the following areas, stirring and reflux. At 60 - 80 degrees C I felt some of the benzene would be lost reducing the yield and creating a health hazard so I fitted a reflux condenser. Since 50:50 sulphuric acid/ water, benzene and TCCA are basically immiscible, I gave the mix a vigorous stir about once per hour. Everything went as expected.
When distilling I used an all glass quickfit set and I divided the reaction mix in 3 parts as I expected some foaming and the distillation flask was only 150 mls. As expected a cloudy yellow oil passed over, settling beneath an aqueous layer. 2 such runs were collected and the final run started. I do not know how far the final run had progressed ( I did not stay with it for reasons explained later) but my attention was demanded by a massive blast. Fearing a fire I tried to enter the room but was driven back by dense acrid smoke but I could see most of the glassware was gone, the only remnant being the distillation flask, bubbling away serenely on its oil bath as if nothing happened.
I waited about one hour for the extractor to clear the smoke and then went in to survey the damage. It was quickly apparent that the explosion was in the receiver and lower end of the condenser (an expensive Davies double surface type). Some time later I found the thermometer, still head and adapter, all intact! I estimate the explosion to be similar to 5 - 10g lead azide
I examined the existing distillate and I am certain it contains nitrogen trichloride. After washing with distilled water (excess) 10 times it’s still impossible to detect the odour of chlorobenzene due to its intense acrid odour. (Not at all like chlorine, more like hydrazoic acid but not exactly the same.)

Despite finding plenty of research papers to back up this prep, I believe it is inherently unsafe. I mentioned that I did not stay with the distillation: this is because I witnessed an industrial accident with TCCA in the early ‘70s when I working in a chemical plant.
A 1 ton powder mixer had just been emptied of its load which consisted of 20% W/W TCCA in anhydrous sodium carbonate with a drop of light liquid paraffin to control the dust. The operator was instructed to throw in a bucket of sodium hydroxide solution and then hose out the mixer as it was needed clean for a batch of different product.
I was doing something else and not involved with the powder mixer but I looked over when I heard a crackling sould. (The operator had thrown in the bucket of hydroxide but had not applied the hose at this stage.) The crackling died down, a short silence fell, then a truly massive Brisant Detonation occurred. The sheet metal mixer covers flew off, nearly decapitating the operator and spraying residual mix over a wide area. (when empty there would be normally 20 - 50 lbs product left behind, depending on consistency and particle size.)
Since then I have had a very healthy respect for TCCA. I do not believe it is safe to heat it with strong bases or acids (other than the hydrogen halides with which it reacts smoothly and immediately.)
With bases one gets high concentration hypochlorite which is explosively unstable.
With OXO acids one gets high concentration hypochlorous acid ( explosive) which can do 2 things: attack the ring to produce Nitrogen trichloride or simply dehydrate to Dichlorine Monoxide (hypochlorous anhydride), Also explosive.
Even damp TCCA has been known to explode.
In retrospect ( the only exact science! ) I had 2 signs that all was not well. The still head thermometer had not reached 70C when this yellow oil passed over and the intense acrid odour that I have never come across before. This filled the room despite having a powerful centrifugal blower on extraction duties.
I do have some chlorobenzene, the smell is easily detectible, just not in the distillate.
I also have some P-Dichlorobenzene crystals in the distillate flask.

Finally lest you think I may have done something silly, I have been at this game for 45 yrs+. I have proper lab and industry experience and especially in my teens I was obsessed with forbidden preps, ie. Manganese heptoxide. So I’m well familiar with dangerous operations. Ironically Nitrogen Trichloride is probably the only substance the preparation of which I baulked at.

Yours faithfully

Wicked Wolf

IrC - 5-8-2008 at 10:14

That was interesting. The last prep he mentioned I see Woelen has on his site, link below.

Manganese(VII) oxide



EDIT by woelen: Changed link to my website, so that it works again.

[Edited on 12-6-12 by woelen]

Sauron - 5-8-2008 at 10:34

Frankly he sounds like a crank. No email by choice, but happy to convey the above by wifey's email. Right. A flake.

I'll side with axt on this one, and the literature is supportive.

woelen - 5-8-2008 at 10:50

OK, don't start off with such a negative attitude. Just regard this message as a warning, nothing more, nothing less.

I take it serious and if I do experiments with TCCA, then I'll keep in mind what this guy has written. This story will not keep me away from TCCA, but it is good to be more cautious, of course without irrational fear. I'll try some heating of TCCA with NaOH or H2SO4 on a small scale (a few 100's of mg) and see what happens.

evil_lurker - 5-8-2008 at 11:16

Speaking as somone who has ha a rather nastry runaway with TCCA, I don't doubt the story one bit.

@ Wicked Wolf

franklyn - 5-8-2008 at 20:52

Thanks for the advisory, but 19 out of 20 of these incidents, as in your case, occur
when the reaction is not observed or monitored, as such your conjecture of what
may have happened is no more informed than anything members of this forum might
speculate. Merely boiling water as anything else in a sealed container will produce
the same result you are reporting. You may as well be warning to watch out for
glassware because it is prone to shatter. TCCA as any other reagent has innate
safety concerns that one needs to be aware of and avoid.

Sax's Dangerous Properties 0f Industrial Materials
pg 3530
A powerful oxidizer. Forms an explosive product with cyanuric acid + sodium hydroxide.
Potentially violent reaction with combustible materials. When heated to decomposition
it emits very toxic fumes of Cl and NOx.

Bretherick's Handbook of Reactive Chemical Hazards
pg 376
an aqueous slurry [ with ], cyanuric acid and sodium hydroxide is accompanied by
evolution of nitrogen trichloride. Nitrogen purging will keep the concentration of the
explosive gas below the lower explosive limit of 5–6%.
If mixed with a small amount of water, the conc. solution (with pH around 2) may explode,
owing to evolution of nitrogen trichloride. It is believed that hydrolysis leads to formation
of hypochlorous acid and dichloro-s-triazinetrione, and the protonated acid then attacks
the C=N bonds in the triazine ring leading to formation of chloramines and nitrogen trichloride.

http://www.advancechemicals.ca/pdf/msds/StabilizedChlorinati...
" UNUSUAL FIRE AND EXPLOSIVE HAZARD: Nitrogen trichloride can be generated slowly by
the reaction of small quantities of water with a high concentration of this product."

http://cameochemicals.noaa.gov/chemical/12852
" Prolonged exposure to heat may cause vigorous decomposition with rupture of containers."
* My note; decomposes at > 225º C

http://www.cdc.gov/niosh/ipcsneng/neng1675.html
The substance decomposes on heating producing toxic fumes. May explode on heating.
The substance is a strong oxidant and reacts violently with combustible and reducing materials.
Reacts violently with ammonia, ammonium salts and amines, sodium carbonate (soda ash)
causing fire and explosion hazard.

.

crazyboy - 5-8-2008 at 21:09

I agree with franklyn I will take your warning into consideration if/when I do a chlorination with TCCA as it, like virtually all chemicals has the potential to be dangerous however I plan on carefully monitoring the reaction as long as I can which I suggest you do next time.

Nicodem - 6-8-2008 at 00:07

I issued warnings about TCCA many times and still accidents like this happen. I really do not understand why people underestimate the dangers of such powerful oxidants. For example, the accident that happened to Hinz was the consequence of his own carelessness and his unthoughtful "modification" of the acetone chlorination otherwise nicely described in the patent. Things like this are common. I saw other examples of members posting their use of TCCA in irresponsible way (and doing their first experiments on a huge scale!). In the particular case described in this thread it is quite obvious what happened. The guy forgot to quench the reaction with sulfite and distilled the whole reaction mix still containing the oxidant and the acid thus causing Cl2O formation and disproportionation to chlorine dioxide which together with organic vapors form a terribly light sensitive explosive gas phase. While working in the lab only thinking about what you are doing can save you from injuries or even death. I'm glad he described his accident as this might give some thought to all those who start their first reactions on a huge scale without thinking much what they are doing. And of course it is nice to hear he got trough it without glass shrapnels in his face.

Sauron - 6-8-2008 at 00:47

There was another "warning" about a TCCA chlorination of a ketone, which the fellow carried out in his house without a hood, and which went badly wrong (but no explosion.)

But in that instance, he had totally misread the preparation instructions and was heating a reaction he ought to have been cooling, as I recall.

So the lessons are:

Get the prep right
2. Take appropriate precautions against such events.

I still discount the merits of the "warning" in this thread because of the nonmember status of the source and his eccentric stance regarding email etc. We are all anonymous, but at least we are answerable for our postings in the sense that we can be communicated with. Someone who acts as this guy has is the equivalent of grafitti.

len1 - 6-8-2008 at 02:19

I find this story bizarre for several reasons.

First the distillation was done 3 times, and the explosion occured only on the third - with the power of 5-10gms lead azide, whcih is no joke. So what was in the previous two flasks where the distillation went 'well', why wasnt that analysed for the NCl3?

The conditions in the distillation flask, where superheating, bumping and condensation all occur, are generally more condusive to an explosion than the bottom of a condenser and receiver flask. This almost sounds like an expolsion on mixing of two different components.

The story about the email is bordering on paranoia. An email is worth nothing these days - it is just a way to send something to someone - just create an account on yahoo, send, and never log in again. On the other hand the guy admits he sent the mail from his wife's account - thats no safety - he can be identified through her. If he's worried about stress caused by reactions if he posted directly (and some people here are in that business) then he says he regularly visits this forum and will see all this stuff which is equally likely to cause this stress.

So quite a few suspisions - I guess one has to try the exp. to get to the bottom of this, but the source doesnt seem trustworthy, so I shant be in a hurry to do this.

[Edited on 6-8-2008 by len1]

Nicodem - 6-8-2008 at 03:01

Why don't you all cool down with all these insinuations and calling the guy a crank. So what if he refuses to use email? Perhaps he simply has no use of penis enlargement and viagra for which 95% of email is used for. Is that so hard to respect? I don't have a TV and barely ever drive a car. Does that makes me a crank?
And Len, trust me, I know from experience that heating TCCA with an organic and water can lead to the vapor phase blowing up. It is documented in the literature that heating TCCA with water, not to mention diluted H2SO4, creates Cl2O and ClO2. The fact that he only experienced the accident distilling the third aliquot means nothing but that the thing is unpredictable. That he did not quench the reaction says a lot about how careless he is, but he certainly does not sound like someone trying to bullshit anyone. Of course, the explanation that it might have been due to NCl3 is unreasonable, but chlorine oxides + organic vapors is more than reasonable.

len1 - 6-8-2008 at 03:08

Yes - I read what you wrote, but there would not be the equivalent of 5-10gms lead azide in the vapour phase - this needs the presence of a substantial amount of explosive - which as it occured inside the apparatus couldonly have been in the liquid phase


My point about the email is not that only paranoics dont have one - rather that he used it to explain why he didnt write directly to the forum - and this explanation doesnt hold water, which only adds suspicion. I rather suspect he used it as an excuse to write to the admin something he thought would big note himself, and the NCl3 and 5-10gms leads azide also come from there. As for the explosion, maybe there was one maybe there wasnt - maybe his condenser got blocked with TCCA, when a person doesnt tell the full truth its hard to fathom what really happened

[Edited on 6-8-2008 by len1]

Sauron - 6-8-2008 at 09:40

Nicodem, getting rid of viagra spam, Louis Vuitton spam, etc is just a matter of a spam filter, built into all popular email accounts, quite efficient and transparent. The few pieces of spam that end up in Inbox are quite obvious, just check and delete unread. Maybe a few years ago email was a source of stress but not now. No.

The fellow shuns email except when he wants to play Cassandra, then he uses his frau's account.

Sorry, he's a crank, and if he were selling 10 pound notes for half a crown I would pass.

len1 - 6-8-2008 at 20:56

Its interesting to have a look at exactly how much NCl3 or ClO2 must have been present to be equivalent to the explosive power of 5 - 10gms lead azide.

SInce all three compounds decompose into their elements, one can use the formation energy as a rough guide to explosive power.

Pb(N3)2 = 470kJ/mol, MW 291gms -> 10gms

NCl3 = 232kJ/mol, 120gms -> 8.4gms

ClO2 = 105kJ/mol, 67gms -> 10.3gms

So we can rule out ClO2 exploding with that force as there could hardly have enough free volume in the apparatus to hold that much gas. As for NCl3 - out of say 75ml reagent that must be pretty close to theoretical yield

Nicodem - 6-8-2008 at 23:59

Come on! You surely do not believe that silly estimation of his "explosion" being the equivalent of 5-10g lead azide? If that would be the case his distilling flask and all the surroundings would certainly have not been left intact. As if he was some military explosive expert to know what such a large amount of a primary explosive does. If only his receiver flask exploded than this is nothing but the usual ClO2/organic vapors explosion that is completely normal for the conditions he described, or else things would have been way more dramatic. Let it be a lesson to others. Always quench reactions involving, not only TCCA, but also other oxidants like peracids, H2O2, NBS and others before any workup involving heat. Sometimes what is the most logical and normal laboratory practice seems not to be equally logical to those who do experiments at home.

Sauron, apparently according to your viewpoint I'm a crank as well, but I don't take offense. I look at someone who decides not be exposed to any technological mean of dehumanization as a hero of the postmodern age rather than a crank.

stateofhack - 7-8-2008 at 03:03

All of you need chill the fuck out, just take that as a bit of advice! No need to ramble on about him having an email or not;) That is for him, we are talking about what he "claims" to have happened..god so of you or angry on you keyboards!

Sauron - 7-8-2008 at 03:04

Nicodem, I do not regard you as a crank. You don't watch TV, neither do I. You don't (or rarely) drive a car, neither do I. But you certainly do not eschew email, and the Internet, do you? So skip the psychobabble about technological means of dehumanization. As far as I am concerned the fellow is a crank. Not you. Him. As long as his idea of communication is unidiorectional, he is the same as any other author of a poison-pen letter. He lacks any standing. He IMO is devoid of credibility. So, to the rubbish bin.

len1 - 7-8-2008 at 21:11

Quote:
Originally posted by Nicodem
Come on! You surely do not believe that silly estimation of his "explosion" being the equivalent of 5-10g lead azide? If that would be the case his distilling flask and all the surroundings would certainly have not been left intact. As if he was some military explosive expert to know what such a large amount of a primary explosive does. If only his receiver flask exploded than this is nothing but the usual ClO2/organic vapors explosion that is completely normal for the conditions he described, or else things would have been way more dramatic. Let it be a lesson to others. Always quench reactions involving, not only TCCA, but also other oxidants like peracids, H2O2, NBS and others before any workup involving heat. Sometimes what is the most logical and normal laboratory practice seems not to be equally logical to those who do experiments at home.

Sauron, apparently according to your viewpoint I'm a crank as well, but I don't take offense. I look at someone who decides not be exposed to any technological mean of dehumanization as a hero of the postmodern age rather than a crank.


Yes its clear now he isnt. But more seriously he was actually just following the procedure in the pentryl thread which recommends 'distilling the slurry'. The benzene and TCCA are used roughly in stoichiometric proportion there, but the yield of chlorobenzene is claimed at only 40%. This allows the possibility of a large amount of available chlorine to still be present either as unreacted TCCA or DCCA, and heating this with conc H2SO4 can certainly lead to a minor explosion.

This is easily fixed though, just decant/filter prior to distillation. I think that should be incorporated into the procedure, else this site might be held responsible

A violent reaction on the third distillation actually makes more sense if you consider that what he was doing is pouring successive portions from the reaction vessel into the distilation flask. The last 'batch' would then have mainly consisted of unreacted TCCA and H2SO4 and he would have been distilling that .. by that time the distillation vessel heated up enough to make ClO2 most of the organic would have been in the receiver flask, so yes that makes sense

[Edited on 8-8-2008 by len1]

woelen - 7-8-2008 at 22:35

I did an experiment in which I heated TCCA in appr. 20% H2SO4. The TCCA did not react nor dissolve. Even when it was boiled, no reaction occurred. When a small amount of NaCl was added, then the usual formation of Cl2 occurred. So, at least in aqueous systems, even at low pH and high temperature, there is nothing special.

Nicodem - 7-8-2008 at 23:46

Quote:
Originally posted by len1
This is easily fixed though, just decant/filter prior to distillation. I think that should be incorporated into the procedure, else this site might be held responsible

Rather than just decant the cyanuric acid slurry, I would suggest to also wash the organic phase with sulfite to quench absolutely all the oxidant (at least that is what is commonly done in a lab).

Woelen, you had no organic vapors in your test. Try heating toluene, TCCA and water to reflux for a longer period in a semi dark place. If you are "lucky" the chlorine oxides+toluene vapors built up in the reflux condenser will suddenly blow up with a red flame crating some soot. I had this happening once, but no glass breaking accidents happened since I never do such potentially exothermic experiments on a scales higher than 10 mmol on first times.

another email followup

Polverone - 15-8-2008 at 14:25

I'm still not sure why this person communicates by email only. I do recall, though, that a certain Mr. Anonymous used to read the forum but only post through email sent to me until he joined under his more familiar current name. So this is not without precedent.

Quote:
Some very good incisive comments from Nicodem and Len1.
They are a credit to the forum. The answer I needed is obviously quenching but I would make the following points.
1. I was concerned with the stoichometry of the reaction in view of the relatively low yield and would have thrown in something like thiosulphate but I had none.
2. AXT used no quench. Vogel lists two preps with NBS and one with NCS - no quench.
3. The paper by Tilsam and Weinmann, although not a prep doc doesn't suggest any need to quench the reaction. Incidentally under the heading "a safe and efficient oxidant" they warn of the dangers of TCCA in strongly acid aqueous media - a bit of a contradiction.
4. I don't have access to the J.O.C paper by Juenge, Beal and Duncan.
5. Decantation would not be the answer; at the end of the chlorination I found most of the organic phase had soaked into the cyanuric solid phase.
6. Chlorine Dioxide was not the cause of the explosion; I am very familiar with its odour but not that of Nitrogen Trichloride and Dichlorine Monoxide.
7. Len1 was totally correct re final run concentrations. After the first ceased to distil significant amounts I topped up the distillation flask with the slurry and repeated the same sequence on the third run.
8. AXT used a copper kit, I wonder did the presence of copper catalyse the decomposition of whatever energetic compound was present before a sufficient concentration built up.
9. Perhaps I misled you on one point. The distillation was not unattended; I checked it approx every 15 mins, but did not stay with it. There was 10 - 15 mls liquid in the receiver on the last check, previous quantities having been bottled due to the acrid
HN3-like odour it was emitting.
10. I have been making lead and sometimes silver azide since 1968. I am more familiar with lead azide than any other H.E. My best estimate of the damage to the kit and surroundings is 5-10grams.
I am now of the opinion that heating TCCA with strong oxo acids is inherently dangerous.

Formatik - 29-8-2008 at 13:32

The yellow oil and tear gas properties are telling of NCl3, though its formation conditions are sensitive as acids and bases can destroy it. The mentioned heating temperature of 60 to 80 deg.C is right in the explosion temperature of the nitride, Gmelin says NCl3 can be distilled at 71º, but further heating causes explosion. It could cause similar damage to Pb(N3)2 because it's a primary explosive. As to the claimed amount equivalent to azide, it doesn't take much at all to blow glassware to pieces, e.g. 3 mL NCl3 is more than far enough.

[Edited on 29-8-2008 by Schockwave]

AndersHoveland - 18-5-2011 at 09:02

Site about accidental industrial explosions:
http://www.chemaxx.com/explosion16a.htm

(interestingly it also claims that liquified nitric oxide is a very sensitive and unstable explosive)