Thursday, September 4, 2014

Methanol "tornado" experiment goes awry

Courtesy of my local TV news habit, I heard about a science experiment that sent a few kids to the hospital in Reno, NV. Here's the report from the Reno Gazette-Journal:
A science experiment went awry at the Terry Lee Wells Discovery Museum, creating a chemical flash that injured up to 13 Wednesday in Reno, officials said. 
Of those 13, eight children and an adult were transported to Renown Regional Medical Center, the city said in a statement. 
Primary injuries include minor burns and minor smoke inhalation, the statement said.
Four people were treated at the scene and released, Reno police officer Tim Broadway said earlier. 
Reno Fire Chief Michael Hernandez said it was a routine experiment involving the simulation of a tornado that is conducted daily at the museum, 490 S. Center St. 
The officials said a methyl alcohol and boric acid mixture caught fire during a routine exhibition that is conducted each day to create a whirling tornado effect. 
City of Reno spokesman Matthew Brown said that a preliminary investigation indicates it was not an explosion but a chemical flash, which is "similar to if someone threw gasoline on a fire."
Here's a video showing the experiment by a demonstrator in Britain (using copper instead of boric acid):

When you look at the video that the parents were taking (it's about 30 seconds into the report), as well as the pictures from the parents, I think I have a theory as to what happened:

The demonstration requires you to use a mesh garbage can and a spinning turntable. I suspect that the garbage can tipped over while spinning and it lit the jug (or its fumes) on fire. Maybe I'm wrong, but I suspect that's what happened. Nope, see below.

The Reno fire chief has an opinion on the matter, based on his conversations with museum staff:
Terry Lee Wells Discovery Museum staff reacted quickly after an accident that injured up to 13 late Wednesday afternoon, Reno Fire Chief Michael Hernandez said.
"According to our surveillance video within seconds they already had extinguishers out," Hernandez said. "They were well-trained and assisted our first responders."
About a dozen children were seated 6 feet to 10 feet away from an experiment done routinely at the museum to simulate a dust devil or tornado, Hernandez said. The experiment uses methyl alcohol and boric acid and the order got switched and it caused a flash that lasted two to four seconds. 
"It's sort of a one, two, three process and they went one, three, two," Hernandez said. He described it as being like being too close to a campfire. 
"There was very minimal damage within the immediate area of where the flash occurred," Hernandez said. "There was an easel immediate adjacent to the demonstration and it did not even get knocked over." 
Given the smaller quantity of chemicals they used in the demonstration, there wasn't much of a chance of a more serious flash, the chief said. Nothing that would have blown out windows, for instance. 
One child suffered second-degree burns and was going to be kept overnight at Renown Regional Medical Center for observation, Hernandez said. He visited all the children taken to the hospital Wednesday night and most seemed shaken by the incident.
The fire department will recommend the museum review its procedures. 
"I think at the end of the of the day it's going to come down to a simple accident in procedure," Hernandez said. "As the fire chief, I'm not going to call them and say, 'Please stop doing this procedure.' The fact is, they've been doing this for quite some time. This is probably an isolated event." 
Museum spokeswoman Meagan Noin said they had no new comment this morning.
"We are still investigating what happened so we don't have any new information at this time," Noin said.
I'd like to know what the different order of operation was -- was the methanol added after the flame was already lit? Odd. I'm sort of weirded out by the fire chief's initial judgment, but hey, maybe he knows something I don't.

I am tempted to say: it's time for people to stop setting fire to methanol around kids for demonstrations. Matter of fact, I did say that on Twitter last night. After a night's sleep, I think that the right answer might be that people need to take many, many, many, many, many more precautions than were taken in Reno. What could have been done better?:
  • Assuming that I'm correct that the methanol jug was part of the problem (and seeing as how it continues to be a problem with the rainbow flame demonstration), you can't have a larger source of methanol in the same vicinity as the lit demonstration. 
  • Assuming I'm right about the tipping of the flame, the turntable needed secondary containment. Nope, see below.
  • The kids needed to be much further away from the experiment. 
Obviously, we don't have all the facts yet and my interpretations of the photos/video could be entirely incorrect. That said, I think the need extreme caution with experiments involving burning methanol has been demonstrated one more time.

UPDATE: This other video from ABC News shows the exact moment that the flash occurred, which seems to suggest that the instructor was adding methanol while another flame was lit?

Watch more news videos | Latest world news

Well, that would explain it. Yeesh.


  1. I'm with you on what could have been done better. It was a very bad idea to have a full jug of methanol nearby. The amounts necessary for the demo should have been pre measured into small aliquots and clearly labeled. That way, even if the "experiment" went wrong, the resulting fire would be small. That is just good lab practices.

    Also, not only were the children too close, but they were below the demo, so if anything spilled, it quickly spilled into children.

  2. Ahh. Remember pouring ethyl ether down the drain trough behind the lab bench with a couple of Bunsen burners going when you were a kid?

    Wfffffff. :-)

    Those were the days...

  3. Huh. did not know that you get increased flame height from spinning it. Going to make this one of my usual demos now.

  4. Thank you for this CJ. Im doing General Chem demos in an adjunct position. I will never do a demo, or a lab, that includes MeOH.

  5. If they were using boric acid and methanol, the flame should have been green due to trimethyl borate. To me it sounds like they were using Li/Cu like in the youtube video where it is more red/orange. Only way a flash that big could have happened is if methanol was spilled all over the counter-top so that the vapors were nearby.

  6. You can see the green flame on the floor at ~45 s into the news video.

  7. Anon130p:

    Congratulations, you're my first deleted comment of 2014! You may think that the magic word that caused your deletion was a slang word for female genitalia, but you'd be wrong. The magic word was "you."

    Enjoy! Cheers, CJ

    1. I'm surprised you haven't deleted mine. Perhaps you might find me entertaining--if you were half drunk.

    2. How about if i make a snarky comment about Patrick Harran and his performing demos for public service?

    3. I've never really set out to figure out my blog comment policy, but I want to enforce minimum standards for interactions between commenters and I found Anon130p to be below those standards.

      One would sincerely hope and imagine that Professor Harran would be the first to talk about the importance of PPE when dealing with flammables.

    4. Sorry, I'm apparently in Super Earnest Mode today.

  8. Having an open container of methanol close to a flame doesn't look like a good idea. Why did she have so much of it on the bench in the first place? For a demo like that you only need about 50ml.

  9. The lower video provides the clue for the upper video. Pouring the methanol on last led to ignition of the methanol in the 'quart' container. Internal pressure forced the flaming methanol across the counter-top (see beginning of upper parent video) which then ignited the fumes from the open gallon container at the left of the counter. That container then toppled from the counter as it melted, giving the material that spread across the floor to the kids. Too dangerous for amateur scientists.