Thermite (?) incident

[tommj]

I've got a friend that uses a timesaver WET belt sander/deburr machine. It has a belt followed by a scotchbrite type drum to deburr laser cut steel and aluminum parts. Below the conveyor is a filter fabric fed from a roll through a trough with a float switch. The filter fabric is advanced exposing clean material when the float switch detects that water is building up in the filter trough. Coolant (with 3% corrosion inhibitor) drains thru the filter into a 75 gal or so tank on casters under the machine, which is in an approx 600sq ft metal building. One day this past week someone noticed smoke coming from the corners of the metal building. The building has been closed and the machine not used for 3 days prior. When the door was opened the entire building was full of dense steam fog, zero visibility. Roll up door was opened and the floor and Everything inside was completely wet as if hosed down. The coolant tank was boiling and had boiled off about 1/2 its volume. It was not localized boiling but evenly distributed. The tank was upended on a gravel lot and there was no localized chunks of anything, but there was a large soup of metal particulates in an angry swirling active reaction that looked like it would have continued to run if the tank was not dumped. The coolant tank had not been cleaned in a long time, I speculated that aluminum and steel particles in the tank had built up in chunks and oxidized leading to a thermite reaction which I think generates its own O2 so could work under water, but I'm not a chem e so not sure how this could actually START under water- the particles would be wet in process and almost all removed by the filter What actually happened here?

 

[IRstuff]

Thermite ignition is extremely difficult to obtain; you'd need a roadflare or blowtorch to get it started

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Heaviside1925]

OP,
Personal experience. Bicycle rim factory, steel and aluminum filings in water-based coolant. When the coolant tank was full, the coolant was filtered out to be reused and the steel and aluminum filings were placed in a basement. They'd been doing this way for years. 1 dead, 10 injured in the ensuing explosion.

Typical aluminum metal will not typically react with water because it immediately creates a protective oxide film on its surface that protects the aluminum from the water. IF that oxide layer is removed aluminum in its pure form, Will react with water exothermically and create hydrogen gas. What Could have happened is steel filling / powder or corrosion inhibiter or a combination of both, reacted with the aluminum oxide layer and allowed for the pure aluminum to react with the water. The fog you encountered was likely steam and everything being wet was the condensed steam. What you didn't see was a room full of hydrogen gas.

A spontaneously boiling tank inside a building is a potentially very dangerous condition. This implies a very energetic exothermic reaction. You have no way of knowing what gasses and vapor you or your employees were exposed to, let alone, possible explosive environment. Situations like this are highly abnormal and should NOT be taken lightly by your management, owners, nor employees. The work practices at the company need to change immediately. At a minimum, any metal filings / powder need to be stored outside, in a non-combustible container and away from other combustibles, employees or public. Your coolant may need filtered, and particles removed on a daily / shift or even hourly basis depending on how much are being produced.

Your company should hire an engineering firm knowledgeable in metal, polishing processes and dust / powder remediation and control. If you are in the US, the NFPA has very good guidance on this. If you are elsewhere, most recognized standards address these manufacturing processes.

 

[tommj]

Not my company...in any case, the coolant is filtered before it gets to the reservoir, but obviously the filtration was not 100% effective. Procedure is now to clean the reservoir tank frequently and there will be a wifi temp sensor thermocouple in the tank with alarm and text notification, and a smoke detector in the building assuming they can be triggered by steam. Re: the rim factory, were the filings stored dry or wet? I can see how damp material might react, but not how this would occur with such a large amount of water. The coolant is a Castrol product but I'm waiting to hear which formulation. Anxious to know because there are other (cutting tool, not abrasive, so larger particles in chip form) processes in the same facility which may be using the same coolant. I'm having trouble understanding exactly how this happened because it appeared when the tank was rolled out that that the reaction would have continued until all the water boiled off, then what? Thermite explosion? Is it possible that some oil in the mix covered the water and allowed some sort of anerobic or bacterial process to get the ball rolling with some clumps of powdered material?

 

[Heaviside1925]

Re: the rim factory, were the filings stored dry or wet?

Wet, damp basement. I say personal not professional because I was a kid when it happened.

smoke detector in the building assuming they can be triggered by steam.

Need a combined smoke and rate of rise detector. I would strongly consider a fixed LEL meter in the vicinity as well. Consider you can't see or smell hydrogen nor can you many other flammable and toxic gasses.

I'm doubtful of a thermite reaction since the initiation temp is so high. Thats not to say impurities can't lower that threshold but that's a stretch. The aluminum-water reaction would die off when it ran out of water and also when the aluminum is exposed to air, the protective film will immediately be reestablished stopping the reaction. This is likely what happen when it was dumped into the gravel.

For specificities, these are questions for an industrial chemist, maybe there's one on here who can chime in. I am pleased to hear that the company seems to be taking this seriously and has taken steps, but I would again suggest that unless the company has someone on staff that can professionally evaluate this occurrence, that an appropriate engineering firm be consulted.