From the archives: Chemists lose hands from peroxide explosions

Regarding the inadvertent synthesis of TATP at the University of Bristol, someone commented at “In The Pipeline”:

I recall a C&EN story from the early 1980s about a group at K (Kansas or Kentucky?) preparing a batch of 100% H2O2. It exploded during purification and blew off a corner of the building. I vaguely recall a picture of the lab walls completely blown out. I believe they (Kansas? Kentucky?) shut down their chemistry program after that incident before restoring it after a couple of years.

I dug into our archives to see if I could find the incident in question. I haven’t been able to find it, but I did dig up some other interesting stories:

From July 21, 1952:

Chemist Loses Hand in Performic Acid Explosion
performic acidFive milliliters of performic acid exploded recently at Laval University, Quebec, Canada, tearing off the right hand of a graduate student and smashing all glassware in a radius of 2 to 3 feet. Numerous glass slivers were driven into his skin and into one of his eyes.

According to information from the student, A. Weingartshofer-Olmos, and Paul A. Giguere, professor of physical chemistry at Laval, a small receiving flask containing the 5 ml of approximately 90% performic acid was being removed from the still when it detonated for no apparent reason. The acid had been prepared by the addition of 25 grams of 99% hydrogen peroxide to 20 grams of 99% formic acid in the presence of 6.5 grams of concentrated sulfuric acid as catalyst. After two hours for reaching equilibrium, the mixture was distilled under reduced pressure (5 to 10 mm Hg) at 30° to 35° C.

This preparation had been performed several times before in the same manner without any mishap. The material was known to be dangerous and adequate precautions were taken. All glassware was thoroughly cleaned in fuming sulfuric acid. The distillation apparatus was entirely assembled through ground glass joints and no lubricant of any sort was used. The still was connected to a dry-ice trap, manometer, and vacuum pump through a length of Tygon tubing. Only 5 to 10 milliliters of the acid was prepared at a time. As nothing unusual had happened while the material was heated for distillation and as the distillate was kept at —10° to — 15° C , the operator felt that the danger period was over. He removed his face shield, pushed aside the two safety screens, and reached for the receiving flask. As he was about to touch the discharge tube to collect a pendant drop, the flask exploded.

Like all peroxides and ozonides, performic acid is unstable, since it is so extremely endothermic. It can decompose in three ways: to give formic acid and oxygen, to give formic acid and hydrogen peroxide, and to give water and carbon dioxide. The last one is thought to be predominant in the explosion. A number of compounds are known to bring about the decomposition of 90% acid, but none of these were thought to be present. The only explanation that has been presented for the unfortunate accident is that the material was more concentrated, and that as the remaining water was removed, the peracid became sensitive to shock or dust particles. A somewhat similar accident with acetyl peroxide was reported in C&EN, Oct. 25, 1948, page 3197.

The residue in the distillation flask consisted approximately of 30 ml of a dilute mixture of formic and performic acids, with some hydrogen peroxide also present. It was impossible to determine if this mixture had also detonated, since the entire apparatus was shattered. It was also noted that glass Dewar flasks constitute a potential danger not always realized in laboratories. In the explosion at Laval, the detonation wave was sufficient to shatter two or three flasks, some located several feet away. Even those which were wrapped with adhesive tape broke and the crown at the top was blown off.

As for that Oct. 25, 1948 reference:

A Warning to Users of Acetyl Peroxide
acetyl peroxideIn view of the considerable amount of work that is being done with acetyl peroxide, we should like to describe an explosion of this material which occurred here recently, with the hope that such accidents may be avoided in the future.

On a Tuesday afternoon a worker prepared acetyl peroxide according to the method of Gambarjan [Berichte, 42, 4010 (1909)] using 20 g of acetic anhydride and 100 ml. of ether and 10 g of sodium peroxide. The reaction appeared to proceed smoothly and at the end of the day the acetyl peroxide was in ether solution and had been dried over calcium chloride. The solution was placed in the electric refrigerator over night. On the next day, Wednesday, the solution was concentrated under water pump pressure – no heat was applied. Because he had other things to do the worker was unable to use the material that day so he put the concentrated acetyl peroxide solution back in the refrigerator.

On Thursday morning he took it out and put it in the dry ice chest for about 30 minutes and the material crystallized. The crystals were filtered off on a sintered glass funnel. About half of the material was then used in an experiment and the other half was put into a small bottle and put back into the dry-ice chest. About a half hour later the worker took the bottle out of the chest and was carrying it back to his lab when the material detonated with sufficient force to tear off both his hands. There was only about 5 g of material in the bottle and the worker did not recall having jarred the bottle, nor having done anything else that might have caused it to explode.

It would be wise to take two precautions in the future : (1) The preparation and use of the acetyl peroxide should be carried out without interruption. The material should not be allowed to sit around for 24 hours before being used. (2) Since the destructive effect of a detonation varies with approximately the inverse third power of the distance, a container of acetyl peroxide should be held in a device with a handle so that it may be held some distance from the hands and body of the worker.

Lester P. Kuhn, Ballistic Research Laboratories, Aberdeen Proving Ground, Md.

Author: Jyllian Kemsley

Share This Post On


  1. One phrase from the first article is truly terrifying:

    “his preparation had been performed several times before in the same manner without any mishap.”

    This phrase scares me more than any other part of that article, because it seems to accompany every safety mishap that makes headlines, even 64 years later.

  2. FYI the 1909 reference in the second report is Berichte, not Berichle.

    Years ago I dug up everything I could find (I only found one) on the infamous ‘bottle of peroxidizable solvent explodes when someone unscrewed the cap because peroxide crystals were in the cap threads’ incident. I found only newspaper reports. I vaguely recall that the individual died from the results, it was pointed out that the bottle was held at bench level when the attempt to open it was made, that the word ‘eviscerate’ was used, and that it was a bottle of diisopropyl ether. I no longer have those articles, but if someone is able to search old newspaper articles (guessing 30’s-60’s, in the US) they should be able to find it.

    Finally, at Indiana University ca. 1979-80 a graduate student had just finished distilling ethyl perchlorate when the receiving flask exploded in his hand. He lost some fingers. There were reports (I don’t know if true) that an NMR magnet on the floor above moved from the explosion. This was another example of a repeated procedure that had been done “in the same manner without any mishap” – but should never have been done in the first place. As far as I know this incident never showed up in the news.

  3. @Kenrod–Berichte fixed, thank you! It was an OCR mistake when our old issues were scanned, but I should’ve noticed it when I added the link.

Submit a Comment

Your email address will not be published. Required fields are marked *