Sciencemadness Discussion Board

Intentional runaway test?

Fyndium - 28-2-2021 at 08:09

Doing a new exothermic oxidation I thought that what if a synthesis was ran to runaway to test its limits? Has anyone ever done this? Of course this kind of testing would need a suitable place.

In general, is there a risk for explosive thermal runaway or is it just the good'ol geysir and messy aftermath?

I'm usually very conservative(scared) on pushing reactions and take more time than risk runaways if that is an option. Another thing I love in large waterbaths: they allow temp control within 5C over several hours with manual adjustment and will turn from heat to cool at the moment the reaction temp changes even a degree. Stirring of the bath has huge impact on thermal transmission rate and it can easily allow halving some things time like additions etc.

[Edited on 28-2-2021 by Fyndium]

zed - 28-2-2021 at 19:46

Why yes.... Fyndium, such tests ARE occasionally run.

Seems to me, the events at Three Mile Island and possibly Chernobyl, were triggered by such tests.

But, perhaps I am mistaken. Maybe that was part of the plot of the now long forgotten movie "The China Syndrome".

[Edited on 1-3-2021 by zed]

DraconicAcid - 28-2-2021 at 20:45

I know I spent many trials trying to find a way to get the nitric acid oxidation of cyclohexanol to adipic acid to run without always turning into a volcano.

Texium - 1-3-2021 at 08:50

Quote: Originally posted by Fyndium  
Doing a new exothermic oxidation I thought that what if a synthesis was ran to runaway to test its limits? Has anyone ever done this? Of course this kind of testing would need a suitable place.

In general, is there a risk for explosive thermal runaway or is it just the good'ol geysir and messy aftermath?
I don't think there is a generalization that can be made. What is the specific reaction that you are wanting to test? That is necessary information for knowing whether it will explode or just boil over.

The utility of such a test would be pretty marginal anyway. Testing it on a smaller scale than you intend to run it wouldn't be helpful, since reactions are more sensitive to runaway at larger scales, and if you're going to test it at the scale that you actually want to run it at, you may as well just try and do it right.

Fyndium - 2-3-2021 at 09:14

On hobby scale that would be more for curiosity, but for any production purposes I see there would be a need to find out how rigorous safety measures must be placed and what is the practical control margin of a reaction. If it is run at 60C and goes runaway at 65C, it needs very precise control and powerful enough scram mechanism to stop it from going all out.

For example once I run an oxidation reaction with large water heating bath, I had a large siphon and a cold water supply near in case I had to quickly cool it, as the device itself was mounted so it could not be moved. I don't know how close the thermal runaway threshold is, so I took extra caution. Lowering the bath with a jack would remove the thermal ballast and probably cause the reaction to go even hotter, as air is a very bad thermal conductor. This is another reason heating mantles are not good for any reactions that may need to be cooled, unless they're small enough that you can just pick the flask and immerse it in water, although losing stirring.

ManyInterests - 28-9-2021 at 05:11

I'm looking to make one of these tests for runaway nitrations. This is especially important since I had a mishap with making Nitrocellulose (I think it was caused by the contamination of my latest batch of KNO3, which is why I am glad I didn't add it to my cleaner supply) that caused really, REALLY noxious fumes from my nitration cup.

I was also using pure sulfuric acid, which was a first for me since I normally use the drain cleaner grade for my previous batches of nitrocellulose.

Any advice on performing a runaway nitration safely?

Jenks - 28-9-2021 at 07:58

Quote: Originally posted by DraconicAcid  
I know I spent many trials trying to find a way to get the nitric acid oxidation of cyclohexanol to adipic acid to run without always turning into a volcano.

This reminds me of a similar reaction using isopropanol entitled "Violence of a Nitration Runaway": https://woelen.homescience.net/science/chem/exps/raw_materia...

S.C. Wack - 28-9-2021 at 14:29

Chemistry is full of hidden variables, nonconformity, and irreproducible results (same person, scale, chemicals, etc.). It was surprising to see that Bretherick's mentions diazomethane explosions due to the addition of "undiluted nitroso precursors" to base, rather than distillation, as it's rather benign IME with nitrosomethylurea...what is the cause; technique, impurities, rumors spread by Al Bader?...hidden variables.

Some reactions that produce heat and gas are totally reliable, but in the main it doesn't seem that exotherm and reactivity are a recipe for predictability.

[Edited on 28-9-2021 by S.C. Wack]

B(a)P - 28-9-2021 at 16:16

Quote: Originally posted by ManyInterests  
I'm looking to make one of these tests for runaway nitrations. This is especially important since I had a mishap with making Nitrocellulose (I think it was caused by the contamination of my latest batch of KNO3, which is why I am glad I didn't add it to my cleaner supply) that caused really, REALLY noxious fumes from my nitration cup.

I was also using pure sulfuric acid, which was a first for me since I normally use the drain cleaner grade for my previous batches of nitrocellulose.

Any advice on performing a runaway nitration safely?


Are you looking to do a run away test for the notation of cellulose?
Just do it on a small scale, starting with a temperature that you know won't cause a runaway, then try again at a slightly higher temp, repeat until runaway. Make sure this is done in a well ventilated area, with heaps of freeboard in your reaction vessel.
On a side note you need to chill your acid before adding nitrate. There is a huge wealth of info about nitrocellulose on this forum in posts and the library that are well worth a read

karlosĀ³ - 28-9-2021 at 18:29

Quote: Originally posted by Fyndium  
Doing a new exothermic oxidation I thought that what if a synthesis was ran to runaway to test its limits? Has anyone ever done this? Of course this kind of testing would need a suitable place.

In general, is there a risk for explosive thermal runaway or....

You really do not want to give any details, no? :D
At least to know the oxidants name would help immensely ;)
And the conditions actually matter much too, there are TCCA based oxidations for example, who will be able to go off violently when they runaway, due to NCl3 formation(its done in aqueous ammonia and a cosolvating ether), which forms when too hot but also, well, NCl3 isn't even stable when in water and while the reaction without that "feature" probably would be at most at medium-large volcano-style mess, makes this "bonus"(malus rather) the experience to potentially costing you a hand and few fingers.
A variation with iodine instead of TCCA exists, and indeed NI3 is formed even regularly, but in that case, its neither exothermic, nor is NI3 in such a wet environment at RT of any concern, opposed to NCl3.

That specific example is not an oxidation, but I know that the leuckart reaction, especially when used under clandestine conditions at scale, tends to remove the roof occasionally, but detailed experimentals, or any at all, are nowhere to find to understand why exactly :P
I guess, all the formic acid, the high temperatures, and most importantly, people of the kind who see a cake recipe and genuinely think, twice as hot for half as long will also work :o
We long suspect that the instructions used are also not much differently written :D

So don't be shy and tell us what you are using, at least that is necessary for a vague opinion... but details for details! :P

I start to guess:
1. is it organic?
2. does it contain haloges?
3. more than one different halogen?
4. pool supplies?
5. anorganic oxidant, organic nitrile, the reaction is driven forward and kept at its temperature by careful, if necessary dropwise, addition of the oxidant, but cooled if necessary, cooled to strong, it dies down, and probably will not really kick in again even if everything else is ideal again?

If you answer no on everything except the fifth, then I can tell you that much: it will be smelly and bad and probably not explosively, but almost, since stronger hydrogen peroxides tends to do that at the likely reached final temperature of that reaction running away/doing a runaway.
Depends on the scale of course, and all the other details you forgot to present :P