Skip to Content

Things I Won’t Work With: Dioxygen Difluoride

The latest addition to the long list of chemicals that I never hope to encounter takes us back to the wonderful world of fluorine chemistry. I’m always struck by how much work has taken place in that field, how long ago some of it was first done, and how many violently hideous compounds have been carefully studied. Here’s how the experimental prep of today’s fragrant breath of spring starts:

The heater was warmed to approximately 700C. The heater block glowed a dull red color, observable with room lights turned off. The ballast tank was filled to 300 torr with oxygen, and fluorine was added until the total pressure was 901 torr. . .

And yes, what happens next is just what you think happens: you run a mixture of oxygen and fluorine through a 700-degree-heating block. “Oh, no you don’t,” is the common reaction of most chemists to that proposal, “. . .not unless I’m at least a mile away, two miles if I’m downwind.” This, folks, is the bracingly direct route to preparing dioxygen difluoride, often referred to in the literature by its evocative formula of FOOF.
Well, “often” is sort of a relative term. Most of the references to this stuff are clearly from groups who’ve just been thinking about it, not making it. Rarely does an abstract that mentions density function theory ever lead to a paper featuring machine-shop diagrams, and so it is here. Once you strip away all the “calculated geometry of. . .” underbrush from the reference list, you’re left with a much smaller core of experimental papers.
And a hard core it is! This stuff was first prepared in Germany in 1932 by Ruff and Menzel, who must have been likely lads indeed, because it’s not like people didn’t respect fluorine back then. No, elemental fluorine has commanded respect since well before anyone managed to isolate it, a process that took a good fifty years to work out in the 1800s. (The list of people who were blown up or poisoned while trying to do so is impressive). And that’s at room temperature. At seven hundred freaking degrees, fluorine starts to dissociate into monoatomic radicals, thereby losing its gentle and forgiving nature. But that’s how you get it to react with oxygen to make a product that’s worse in pretty much every way.
FOOF is only stable at low temperatures; you’ll never get close to RT with the stuff without it tearing itself to pieces. I’ve seen one reference to storing it as a solid at 90 Kelvin for later use, but that paper, a 1962 effort from A. G. Streng of Temple University, is deeply alarming in several ways. Not only did Streng prepare multiple batches of dioxygen difluoride and keep it around, he was apparently charged with finding out what it did to things. All sorts of things. One damn thing after another, actually:

“Being a high energy oxidizer, dioxygen difluoride reacted vigorously with organic compounds, even at temperatures close to its melting point. It reacted instantaneously with solid ethyl alcohol, producing a blue flame and an explosion. When a drop of liquid 02F2 was added to liquid methane, cooled at 90°K., a white flame was produced instantaneously, which turned green upon further burning. When 0.2 (mL) of liquid 02F2 was added to 0.5 (mL) of liquid CH4 at 90°K., a violent explosion occurred.”

And he’s just getting warmed up, if that’s the right phrase to use for something that detonates things at -180C (that’s -300 Fahrenheit, if you only have a kitchen thermometer). The great majority of Streng’s reactions have surely never been run again. The paper goes on to react FOOF with everything else you wouldn’t react it with: ammonia (“vigorous”, this at 100K), water ice (explosion, natch), chlorine (“violent explosion”, so he added it more slowly the second time), red phosphorus (not good), bromine fluoride, chlorine trifluoride (say what?), perchloryl fluoride (!), tetrafluorohydrazine (how on Earth. . .), and on, and on. If the paper weren’t laid out in complete grammatical sentences and published in JACS, you’d swear it was the work of a violent lunatic. I ran out of vulgar expletives after the second page. A. G. Streng, folks, absolutely takes the corrosive exploding cake, and I have to tip my asbestos-lined titanium hat to him.
Even Streng had to give up on some of the planned experiments, though (bonus dormitat Strengus?). Sulfur compounds defeated him, because the thermodynamics were just too titanic. Hydrogen sulfide, for example, reacts with four molecules of FOOF to give sulfur hexafluoride, 2 molecules of HF and four oxygens. . .and 433 kcal, which is the kind of every-man-for-himself exotherm that you want to avoid at all cost. The sulfur chemistry of FOOF remains unexplored, so if you feel like whipping up a batch of Satan’s kimchi, go right ahead.
Update: note that this is 433 kcal per mole, not per molecule (which would be impossible for even nuclear fission and fusion reaction (see here for the figures). Chemists almost always thing in energetics in terms of moles, thus the confusion. It’s still a ridiculous amount of energy to shed, and you don’t want to be around when it happens.
So does anyone use dioxygen difluoride for anything? Not as far as I can see. Most of the recent work with the stuff has come from groups at Los Alamos, where it’s been used to prepare national-security substances such as plutonium and neptunium hexafluoride. But I do note that if you run the structure through SciFinder, it comes out with a most unexpected icon that indicates a commercial supplier. That would be the Hangzhou Sage Chemical Company. They offer it in 100g, 500g, and 1 kilo amounts, which is interesting, because I don’t think a kilo of dioxygen difluoride has ever existed. Someone should call them on this – ask for the free shipping, and if they object, tell them Amazon offers it on this item. Serves ’em right. Morons.

142 comments on “Things I Won’t Work With: Dioxygen Difluoride”

  1. RB Woodweird says:

    1. I had to go through regulatory hell just to use some TMSCN. I think I will apply to use some FOOF and watch the guys in Safety pass out from the stress.
    2. FOOF is Satan’s kimchi would be a good name for a band.
    3. In conjuction with 1. above, I would like to order a couple of kilos of FOOF from Hangzhou Sage Chemical just to see the crater on Google Maps.
    4. One of the links in the commentary leads eventually to “Galvanic Experiments on the Dead Body of a Criminal”:

  2. Seminymous Coward says:

    Streng also has a paper on Dioxygen Chlorine Trifluoride (O2ClF3) and several dealing with other oxygen fluorides, including everything from 1 to 6 oxygens with 2 fluorides.

  3. CanChem says:

    I really like in the J. Fluor. Chem. paper how much of the reaction progress is gauged by “pings” in the reaction vessel.
    And Hangzhou? Man, those guys really do have everything.

  4. Henrik Vanger says:

    Some days, I addictively return to this blog just to have phrases like “whipping up a batch of Satan’s kimchi” put a smile on my face. Hunter Thompson would be proud.

  5. A Nonny Mouse says:

    The Chinese would probably ship it in a padded envelope like everything else.

  6. A Nonny Mouse says:

    PS Makes that carbonyl difluoride, which I have been releasing in a reaction, look decidedly tame!

  7. edward says:

    Hangzhou apparently has discontinued this reagent, and no other company offers to sell a kilo.

  8. p says:

    Scifinder should eliminate Hangzhou from their commerical supplier list. Has anyone ever tried to order anything from them? They pop up on pretty much every compound and really pollute searches on commercially available substances.
    Or, we could do as RB Woodweird says and order some of this stuff – that might end the company right there.

  9. okemist says:

    There was some 1930’s German papers doing fluorinations using liquid Cl2 as a solvent, you may be able to bubble O2 into the dissolved F2 to make FOOF but most likely it will just blow up the tube like EVERYTHING else we added, Hmmm…maybe try reverse addition.

  10. processchemist says:

    About pollution of supplier list, some companies (chinese, of course) list almost ALL the products described in the Organic Synthese Collective. And when you ask for quotation and delivery time, price can high and dt 1-2 months from the order (and I’m not talking about 1 ton of material…)

  11. Dave says:

    I’m surprised that FOOF even exists. Then, again, F will latch onto just about everything, even things that nothing else will; otherwise, how could you explain the existance of XeF8?

  12. CanChem says:

    Yeah, in the first paper Derek lists the authors talk about KrF2 forming too. All these things really give me the shivers. I’m glad I’ve never had to work with anything more fluorinated than C6F5OH.

  13. Sili says:

    If I have not said this before, I’ll say it now: Turn this series into a book! There must be a publisher for it.
    You have a coupla fans over at Pharyngula already (some of them non-chemists). I wasn’t even the first to link you.
    Speaking of links: more XMRV whining from WPI. As usual destroyed by her ERVness.

  14. David P says:

    This was an awesome read. The man Streng is a marvel, not in the least part because he was able to run multiple different series of reactions with the stuff. Either that or he had a supply of remarkably unquestioning grad students.

  15. Micah S. says:

    ‘Audible “pings” and pressure excursions occurred when liquid O2F2 dripped onto uncooled portions of the apparatus.’ lol fun times

  16. As a formally trained analytical chemist who was forced to live a more biologically oriented existence due to career demands, reading about basic chemistry principles brings back feelings of nostalgia. Do biologists ever really think about thermodynamics for practical purposes? Is the day of the chemist long gone?

  17. Thomas McEntee says:

    Speaking of KrF2 and fluorine dissociating at 700 deg C, Bezmelnitsyn, Legasov, and Chaivanov reported in 1976 the continuous preparation of KrF2 by thermal cracking of fluorine on a nickel wire at 600-700 C followed by reaction with Kr on the reactor wall at -196 C ([Dokl Chem. (Eng Trans), 235, 365ff (1977)]. Readers interested in oxidative fluorinators should check out Karl O. Christie and David A. Dixon, “A Quantitative Scale for the Oxidizing Strength of Oxidative Fluorinators”, JACS, 114, 2978-85 (1992)

  18. You're Pfizered says:


  19. milkshake says:

    whatever they needed the stuff for, it was not made for improving human welfare.
    This reminds me, when I was in college in Prague we had iodine trichloride preparation in freshman inorganic lab (!) that involved condensing chlorine gas into a flask on a colling bath, spooning solid iodine into that golden liquid and then carefully evaporating the produced goodness until ICl3 crystallized. I did it myself but I hear that the procedure is no longer given to freshmen after one very ugly spill…

  20. Rhenium says:

    The question is whether Streng is still alive. 🙂

  21. Bored says:

    Sili is right, Derek. Turn this series into a book. It would sell, man. “Things I Won’t Work With” is already a perfect title.

  22. Katherine says:

    My day gets better anytime I see “Things I Won’t Work With” in my RSS feed. And I never got past Chem 111.

  23. coprolite says:

    Derek, they’re right. The first time I read your blog was when someone (not a scientist) sent me a link to Sand Won’t Save You This Time, and I pass that one along to people now to weed out the weak-minded.

  24. cicely says:

    I’m one of those non-chemist Pharyngulists of whom Sili speaks, and I would buy the book in a hot second.

  25. Cellbio says:

    Great piece Derek! Who knew volatile chemistry could be so funny.

  26. widget says:

    This very entertaining piece has made me enormously glad I’m a biologist…a total wuss maybe, but at least I still have all my fingers! Thanks for the laughs, Derek…you definitely have a flair! Totally agree with Sili and Bored–you’ve got a best seller here. The Dave Barry of science!

  27. Tim McDaniel says:

    My only chemistry class was in one year of in high school, and I remember little of it. “Sand Won’t Save You This Time” hooked me on this blog.

  28. metaphysician says:

    Put me down as another person hooked by “Sand Won’t Save You This Time”, and another non-chemist to boot. And I concur, really, you could turn this into a book.
    As for this entry. . . 433 kilocalories?! From only five molecules?! I didn’t know you could *put* that much binding energy into a molecule. . .

  29. Hap says:

    I was hooked at the “How Not To Do It: Liquid Nitrogen Tanks” post, though “Things I Won’t Do: Isolating Ciguatoxin” was also pretty good.
    433 kcal/mol is about five C-C single bonds, so it’s in the realm of (unpleasant) possibility. 280 g of FOOF and 34 g H2S gets you…about a stick and half of TNT (I think that a stick of dynamite gives off about 1 MJ), with the bonus that the products aren’t all that nice (I think SF6 is bulletproof, but HF gas can’t be too fun), though I guess if the shock wave containing their highest concentrations is blowing by you at above the speed of sound, you don’t have to worry for long.
    Sounds like a good idea to pass.

  30. Actually fluorine does not need much encouragement to become atomic at any temperature, which is probably why CF3OF and CF2(OF)2 are so easily made from CO and CO2. No abstractions here, any number of kilos, if Messrs Huanzhou are not too busy making everything else let’s watch them try. The legendary George Cady pretty much could not decompose CF3OF up to when CF3OF started reacting with the cylinder (it will ignite happily any organics and detonate C2F4 though). CF2(OF)2 now, you’re in the far upcountry of the proverbial creek if you don’t know what you’re doing and never safe even if you do. Example of the beauty of the beast: my former boss DesMarteau made CF3OOCF2OF (which is the Rock of Gibraltar compared to most hypofluorites) from CF2(OF)2. It’s a wonder: CF2(OF)2 oxidizes COF2 to CF3OF and is ultimately reduced to CF3OOCF2OF. I spent some time on that clean reaction back in the days and otherwise handled fluorine for a living. I miss the eerie elegance of the science and the good memories, but tell you what, do not volunteer anywhere near CF2(OF)2. Let Los Alamos handle that by remote control if they need.

  31. David Marjanović says:

    Let me just jump on the “please turn it into a book” bandwagon.
    Oh, and, streng is German for “strict”.

  32. David Marjanović says:

    Messrs Huanzhou

    Hángzhōu is an ancient city in China, known for its fine silk and its violently hideous…

  33. Karl says:

    “Things I Won’t Work With” would certainly have a place of honor on the bookshelf, right next to my treasured copy of Clark’s “Ignition!” (which has to be the funniest technical book I’ve ever read).
    If you got together with milkshake to include some of his stories, the result might even take over the prime spot.

  34. Hap says:

    I don’t know the language, but I am guessing that Huanzhou is Chinese for “shoebox funeral”, or “the graveyard of fluorine chemists”. Maybe someone can request a site tour, as soon as the wreckage setlles to earth and the clouds of HF pass by.

  35. Doug says:

    Second (or third or fourth) for the book. I was thinking it would be exactly the right thing to use in a high school to get kids interested in chemistry.
    Great stuff!

  36. Anonymous BMS Researcher says:

    widget on February 23, 2010 4:23 PM wrote…
    > This very entertaining piece has made me
    > enormously glad I’m a biologist…a total wuss
    > maybe, but at least I still have all my fingers
    Oh, biology has its thrills too, like BSL-4 pathogens (such as Ebola), not that I ever worked on anything that exotic! I last used a pipette in the bio labs maybe 20 years ago, now I work with computers so I am at risk of carpal tunnel.

  37. Josh says:

    I have a minor in chemistry (major in zoology) but my main focus as my first job out of college was working in a BL-3 and high containment areas with anthrax, bot, serin, soman, and VX. FOOF scares me more than those….

  38. drjim says:

    Duh….makes me glad I’m an EE!

  39. Curt Fischer says:

    It’s amazing that FOOF does not react with CO2.
    When added to Dry Ice, dioxygen difluoride did not react and was only absorbed by the solid. Addition of acetone to this mixture resulted in sparking accompanied by an explosion.
    Maybe he was pissed that the CO2 didn’t blow and wanted to see some pretty colors, and that a bottle of acetone was close at had.
    I was also touched by the acknowledgements section:
    The author wishes to thank[…]Mrs. L. V. Streng […]for their contributions in the experimental work.
    That sounds like love.

  40. milkshake says:

    I think Mrs Steng stood nearby with a fire extinguisher in one hand and first-aid kit in the other. She fluffed her husband’s ego after each round of bloodcurdling mishaps with words “at least you are not a nuclear scientist”

  41. iridium says:

    “Sand Won’t Save You This Time” rules!
    … it does not need to be a big book …. 🙂

  42. Fungus says:

    A book dedicated to stories of fires and explosions in the lab (aka “war stories”) would be a fantastic read. I have about 5 or 6 to share myself.

  43. Stephanie says:

    Always love seeing “Things I won’t work with” in my RSS feed. I’m a librarian, I’d make sure a book of these hit the bestseller list.
    This post just made my day. I am not a chemist (although I spent most of my college years studying it), I always enjoyed the more, um…. explosive aspects of the discipline.
    Now, I just have to figure out how to explain my giggling to my cubemates.

  44. What a great read! Thanks!
    I like how you mentioned Hangzhou Sage Chemical. This is actually a very serious problem. For ordinary chemicals it is more difficult to catch suppliers that play these sorts of tricks. And we have seen a fair share of these. In our company we learn which supplier is relaible and which is not, avoid the ones that have a poor service. But there is no place all chemists can share this experience so that other don’t make the same mistakes.
    And I wonder how they decide what to put in their catalog. Random Wikipedia selection?

  45. Dave_n says:

    In another life, many years ago and not in the USA, I used to work in organometallic chemistry, when “men were men and women were women” and did not worry about thermodynamics, only if it reacted well!! When someone ended their graduate career (with a PhD not a “bang”), we used to surround their reaction train with blast shields and give the newly minted PhD a switch to activate a break seal that would let air into the train. Then just before s/he pressed the button, an 8 mm camera would start rolling. Some pretty spectacular effects when fluorinated, and other energetic organ-ometallic compounds met oxygen.
    In that same life time, I used to have (legal) target practice with 0.22 rifles on the “Winchester quarts” of reactants that had to be disposed of. Spectacular colors and explosions at a safe distance. Those were the days when you could own rifles and pistols in the UK. Today Health and Safety would never allow such enjoyable sport. Dave_n

  46. Sili says:

    But there is no place all chemists can share this experience so that other don’t make the same mistakes.

    That’s actually odd. In this age of ‘Web 2.0’, there should be an Ebay/Amazon of chemicals, where buyers can add feedback and reviews, that in turn can be voted on by other buyers. Of course such systems can be gamed, but they’re not wholly useless.

  47. hell to the chief says:

    Of course in the old days when this stuff was published, it was common for reviewers to try and repeat the observations in the paper before publication.
    So you can add “review the paper” to the list of things you would not do.

  48. Hap says:

    I’m not a fan of less stringent review requirements, but I’d make an exception for that, because there is no way in hell (or anyplace else) that I would try to reproduce anything in that paper (nor in Klapotke’s or Christe’s work, to take two good examples) – even if I had the position and the equipment, I sure don’t have the experience to do so, and I would guess that few people in chemistry do. Very few people would thus actually be able to attempt to reproduce the work, and thus review it (and likely most of those would be competitors). The work won’t be published very much at all, and that would be a loss if for nothing else than the loss in entertainment value.
    I like good chemical literature, but I’m not willing to die for it (and in a way that probably would remove me from the organ donor lists, thus assuring that my life would only be useful as a particularly pointed warning on what not to do.)

  49. Rustler says:

    If you are ever in southern New Mexico, give me a ring. We’ll go out on the outskirts of the Jornada Del Muerto and we’ll sling some lead. One of my favorite targets is a 55 gallon drum into which I’ve put about one gallon of gasoline. On a hot July evening it makes a most satisfying fireball. Of course, you have to use the right ammunition. As a kid, I’d stretch plastic over the drum and cover it with powdered magnesium. We would pretend to re-enact the Trinity test, which happened just a few dozen miles away. It was really quite amazing. We would get the blinding flash and a miniature mushroom cloud. We’d laugh our heads off then do it again. And no radioactivity!

  50. Spectrochimico says:

    The very first project I worked on in graduate school involved synthesizing halooxides for spectroscopic analysis. The senior graduate student working on the project was pregnant and wouldn’t even go into the room where we ran the synthesis.

  51. BenP says:

    I’m a bit confused. You quoted 433kcal per 4 ATOMS of FOOF reacted with HS?
    That would be like 10^25 per mol. That exceeds the energy of splitting a uranium atom by like, a trillion.
    Please tell me its 433kcal per mol.

  52. Nile says:

    Makes you wonder what species exist in ‘Fluorox’ propellants – both in the exhaust stream, and in the storage tank. It isn’t used often, for some or other reason, but it has the distinction of being the most powerful oxidiser available for bipropellants, so it’s not quite ‘left on the shelf’.
    Yes, I’d buy a copy of ‘Sand won’t save you…’ in book form. Would you care to do a limited-run signed special binding with scorch marks?
    Having just read a samizdat copy of ‘Ignition!’, I can assure you that there’s a market for the book. Ignore all asmonitions from the publisher that you should ‘dumb it down’.
    Remind me, sometime, to blog a distressing story about a large vat of Hydrofluoric acid in a British civil nuclear installation.

  53. UC says:

    The 433kcal is for the reaction stoichiometry given in moles.
    If I had a better table of enthalpies of formation, it would be trivial to back calculate the enthalpy of formation of FOOF, assuming it’s not already in the paper (which I can’t access). I’d also be guessing at the state of the H2S (probably solid).

  54. Nile says:

    If I was an unscrupulous supplier (say, somewhere near Huanzhou), and I’d taken your money for an order I know damn’ well I can’t synthesise, I’d examine my conscience at length and take the ‘What would Satan do?’ approach.
    The answer? Mail out a small vial of HF, lots of powdered Teflon, and a kilogram of plastique with an anti-tampering detonator. Who’s to know?
    Any chemist stupid enough to order the stuff can claim his Darwin award, and the world’s a safer place.
    A footnote, and a cover-my-ass disclaimer:
    I would not normally advocate the use of letterbombs – it is very unfair on the delivery guy and illegal to boot, and you might not take the above suggestion entirely seriously – but you might want to sit down and try writing out a packaging & despatch procedure for something as hazardous as a cryogenic fluorine superoxidiser. Then ask yourself: what if some dishonest half-assed laboratory supply company tried to deliver something a tenth as dangerous? How many people died to make it?
    They have *some* good suppliers in China, and some very, very bad ones.

  55. Joe says:

    Sili is right, Derek. Turn this series into a book. It would sell, man. “Things I Won’t Work With” is already a perfect title.
    Better still, “Things that make you go BOOM!”

  56. Paul says:

    Many of the compounds Streng added FOOF to are well known in the rocket propellant field. Tetrafluorohydrazine, for example, is a storable oxidizer. I imagine he was seeing if FOOF could be made to stably dissolve in another propellant component, to give it extra oomph.

  57. +1 for the book! And I note that I’m also a fan of Ignition!, though I do not own a copy.
    I come by now and then, mostly when something points me here again — which has happened repeatedly.
    I’m a software engineer and photographer, trained as a mathematician, and took a course in chemistry once in highschool; and I enjoy reading this sort of thing, and know just barely enough to follow along (I even guessed that the 433kcal must relate to reactions larger than the number of atoms you seemed to be describing, and what that larger quantity probably was).
    Will now forward this to my friend the science fiction author who is the daughter of a chemist :-).

  58. milkshake says:

    the problem with rocket fuels is that the oxidizer is always by far the heaviest component, and in terms of thrust/weight efficiency we get pretty much to a limit with elementary oxygen and fluorine as a oxidizer. Ignition! gives a good summary of various nasties that were tested, including ozone and ClF3 but even if they were less hazardous they would give you only a marginal gain. There is only so much that the re-configuring valence electrons can give you. (The same goes for conventional explosives.)

  59. Make it a book and I will buy it. And I’m not even a chemist. I just know enough to read the experiments and go, “OH MY FREAKING G…!!!”
    And that’s just this post.

  60. Tom says:

    John D. Clark. IGNITION! An informal history of liquid rocket propellants. Rutgers University, USA 1972.
    Perchloryl Fluoride was used as an experimental liquid fuel oxidizer for rockets, and even has its own entry in the glossary as PF. If you haven’t checked this out yet, it’s full of cringeworthy chemistry.

  61. Mark Gibbs says:

    Good heavens! Have you checked out the price of a copy of Ignition? On Amazon there are two used copies; $595 and $1,000! I can’t remember the last time I posted a reply with so many exclamation marks.
    Anyone know of an online facsimile?

  62. David Moisan says:

    I never got beyond HS chemistry; I’m just a CS grad. But whenever I read this series, I want to: 1) ROFLMAO, and 2) Review my evacuation plans. Preferably to the next country over.

  63. Hap says:

    On the original “Sand Won’t Save You This Time”, a commenter pointed to a UMich website that would print copies of Ignition – I assume they were legal, and that they 1) still exist and 2) don’t charge more than $600 a copy.

  64. Ben says:

    By the way, if you google “bonus dormitat Strengus” (without the quotes works fine too) you get a number of sites that are blatantly ripping off your content.

  65. Duane says:

    It isn’t just the enthalpy. Butane combustion produces an enthalpy change of about 790 kcal (if I’ve done my sums correctly) yet we carry butane lighters in our pockets and light them without fear. It is the kinetics combined with the large enthalpy change that produces the uncontrolability of the FOOF chemistry. Thanks for high activation energies in hydrocarbon chemistry!

  66. Hap says:

    The link referenced in #63 is dead – it doesn’t point to anywhere to get a copy of Ignition but to a library database group called ProQuest.

  67. DTL says:

    Yet another comment for the book. It would be completely fantastic and utterly awesome.

  68. Tom says:

    Re ignition: There was a bittorrent copy on thepiratebay as of a couple of weeks ago. I assume someone still holds the copyright though – it’s a great shame it’s not been reprinted.
    And yes, I was led here by “Sand won’t save you this time” and would probably buy the collected “Things I won’t work with”!

  69. Re Ignition, the bittorrent copy on thepiratebay isn’t very good quality – it’s a PDF.
    I got one of the UMich copies – nasty facsimile, but it’s a great book!
    I’m about midway through typing Ignition up in HTML ready to convert to ePub. Natch, it’s still (c) the estate of John D Clark until 2058. It will have reverted to his estate from Rutgers since it’s been out of print for so long, but I have no idea who his heirs and assigns are for the copyright to get a release. Been trying to track them down to get a copyright release for a nice eBook version of Ignition – they’d get a small income from it, which beats the nothing that copyright’s been worth these last three decades.

  70. InfMP says:

    I tried to order a pyridine from Hangzhou last week. I asked when, how much g and $, and they responded by asking me how much i want (broken english).
    Then I said like 5 g, and they never replied again.

  71. Richard Gadsden says:

    Re: Ignition There is an image-only PDF online at
    Yes, you need to know something to get that link to work, no, I’m not telling you what that something is.

  72. Birgit says:

    Your entries are ADORABLE. I studied chemistry myself, including a promotion in physical chemistry which involved mostly computers and some high vacuum equipment. Whenever one of us broke out an OMG WHITE COAT it meant REAL CHEMISTRY might happen, so everyone tried to keep away as far as possible 🙂
    The worst compound ever used was methylamine (condensed as solvent) – the poor guy who worked with it smelled like a fish factory for days.
    Your articles made my ex-chemist heart very happy and your writing style is adorable. Thank you!

  73. Mike says:

    Just saw that link, thank you Richard!

  74. Nerf says: Shows a FOOF apparatus. Remarkably intact, too.

  75. Fred says:

    My chemistry teacher read some of your “Things I won’t work with” blogs. We all thought they were awesome and written in a hilarious tone.
    Thank you! Write more!

  76. Cassady says:

    Yet another vote in favor of turning this into a book. I’d buy one for me and a pile of copies for gifts.

  77. I am not a chemist. I have no education in chemistry whatsoever. All I know is that excessive amounts of “N” in a formula tend to spell disaster, and a teaspoon worth of low-grade black powder can reduce a can of paint to centimetre-wide shrapnel if encased (im)properly.
    ..And yet I find myself coming here time and time again. “Satan’s Kimchi” indeed. You sir, are a genius of the comic variety. Also, I’m interested in a kilo of FOOF, as I have a few rooms that need a fresh coat of paint.

  78. caspian says:

    there is a copy of the Ignition! e-book online at if anyone wishes it.

  79. random websurfer says:

    Derek, you should write an entry about fluorine gas itself… especially about people who like it to come in transparent specimen tubes (yes, you heard that right: *transparent*, specimen tubes, that you put up on display, say in a, oh, children’s museum or something?!):
    From the description, it seems that these tubes would have a lifetime of about 50 years or so, ASSUMING that there is only one or two molecules of HF or H2O among the, oh, 10^23 or so molecules of pure fluorine gas? (That’s 1 part per 10^23 purity, y’know.)
    The lengths to which some people would go…

  80. Anonymous says:

    I find it interesting that the safetygram referenced in the original post has this to say…
    “Pressure relief devices are not permitted on chlorine trifluoride cylinders.”
    Cylinders sans pressure releif devices are a no-no according to OSHA. I bet telling them that you had ClF3 in your facility would be grounds for an instant audit/investigation.

  81. Anonymous says:

    Correction…any EH&S officer will tell you that pressure releases are forbidden on canisters of chlorine or other toxic gases. That is in accordance with OSHA requirements.

  82. Anonymous says:

    … I can’t believe all this stuff actually happened.
    (This stuff was compared to hellfire by a certain site. I won’t mention the name of the site, but I THINK you can guess what site I mean.) It comes right below chlorine trifluoride in teh real lfie section of the article.)

  83. Brooks Moses says:

    Derek, you might be entertained by one of the other things Streng worked with: Solid ozone:

  84. James says:

    In high school in the mid 1980s my “friends” on phreaker BBSs wanted to trade explosive synthesis instructions for hacked telephone long distance codes (which I needed to reach such BBSs) so I took a textbook nitroglycerin synthesis and embellished it with all kinds of hyperbolic safety warnings. A decade later I found it referenced in the Congressional Record by Diane Feinstein when she was trying to outlaw bomb making instructions on the internet. I always wondered whether the instructions I produced were a net safety benefit because they would discourage almost everyone from actually attempting them, or so I hoped, and the regular syntheses were so easy to find in any university library. Who knows?

  85. loupgarous says: is the latest online .pdf copy of John C. Clark’s Ignition! (which I was fortunate enough to find at a K-Mart for fifty cents during my college career, and is quite out of print aside from photocopy reprints that cost $200/copy from
    I don’t have a particular problem circulating this link, given that people have made a good faith effort to get publication rights from the Clark estate for Ignition! and the only people profiting from paid reprints at this point aren’t the Clark heirs, but a reprint company. And Clark is one of the funnier raconteurs of science history in print; it’s a delightful bonus that he and his friends made some incredibly energetic compounds, fired them in rocket test stands and (mostly) lived to tell about it.
    And the book’s foreword is by Isaac Asimov, an early colleague of Clark’s when both men were active in doing chemistry for our side in World War II.

  86. jim says:

    I’m a chemical engineer. We could scale this up. How about a million tons a year at 900 dollars a ton. The cost of crude oil. No greenhouse gases.

  87. theoryguy says:

    Looking at Prof. Streng’s acknowledgements, it appears that one of the fearless experimenters may have been his wife.

  88. Esko says:

    FOOF to react with sodium azide? “Nice” reaction, I’d guess.

  89. fifiste says:

    Hydrogen sulfide, for example, reacts with four molecules of FOOF to give sulfur hexafluoride, 2 molecules of HF and four oxygens. . .and 433 kcal
    Am I off by something but should it really mean that a mole of HS+4molesFOOF will produce around 10e25 joules? Wiki says that 10e24 joules are enough to heat all water on earth by 1 C.

  90. John says:

    Having read your hilarious blurb on chlorine trifluoride, I thought I had found the ne plus ultra of chmistry, but now I realize there are all kinds of “Things I won’t Work With,” combining F, Cl, O, etc. I plan to read them all. As a kid, I once bought a container of sodium (Na) in Atlanta and proceeded immediately to carve it up and throw the pieces in the creek next to our home. WOW! I also experimented with homemade rockets/bombs, etc., and once set our beautiful oak tree on fire—from the TOP! (Yes, it was fall. My ad looked helplessly on and no doubt wondered, “What am I RAISING?!” Lucky for me, it sooon went out. Years later, when I was taking organic chemistry at SMU (Dallas), I tried to make nitroglycerin at home. All I got was dilute nitric/sulfuric acid and water. Does God love me or WHAT? Keep up the good work, bro!

  91. heteromeles says:

    Congratulations for inspiring Charlie Stross’ story “Tall Tail.”

  92. John Dallman says:

    @fifiste, I think you’ll find that 433kcal is per mole. Certainly your energy figure is considerably in excess of the total conversion of mass to energy for the reactants. FOOF is not as lively as that.
    On the subject of moles:

  93. Oldnuke says:

    > Looking at Prof. Streng’s acknowledgements, it appears that one of the fearless experimenters may have been his wife.
    Could you imagine what dinner conversation at the Streng household must have been like? Or if they had any neighbors downwind of their kitchen range hood?

  94. Daniel says:

    Lucia Streng’s obituary is here:
    There’s a little on A R (Alex) Streng as well.

  95. Anonymous says:

    What if first!

  96. David says:

    Prepare for the XKCD army.

  97. 852derek852 says:

    If you ordered 1 kg of F2O2 from Hangzhou, you’d probably end up with a 1kg ziplock bag of sodium floride if you were lucky, or nothing if you were unlucky. If you were REALLY unlucky, it turns out they weren’t bluffing, and you die.

  98. XKCDer says:

    So after the A-bomb and H-bomb, we’re yet to see a FOOF-bomb?

  99. Michael says:

    I am not a chemist, or even particularly good with science (I was a liberal arts major in undergrad and grad school). The last chemistry course I had was in the 11th grade.
    And yet, I found this post pretty darn funny. You have a gift for writing. And even in its technicality, I was able to appreciate your humor and follow the post. That’s a gift, sir.

  100. trowa says:

    Can you imagine what happens when a police officer says “What’s in your tanker, son?” to the truck driver waiting for the recovery van that will change his tyre for him and let him get on his way, and the driver replies “HF”?
    The policeman closes quite a lot of lanes on the motorway, that’s what!

  101. Sci_File says:

    All I can think of is a couple of my Chem students who ask me everyday if we can blow something up. Ruff and Menzel were probably those kids.

  102. Alsadius says:

    I must say, it’s kind of weird seeing XKCD linking a site I already read. Best of luck with the new traffic, Derek.

  103. Dan says:

    #101: Alarming as FOOF is, it’s got nothing on nuclear reactions for sheer destructive power. One fission event produces about 200 MeV of energy, according to Wikipedia; at 433 kcal/mol, the reaction of H2S + 4 FOOF -> SF6 + 4 O2 + 2 HF is around 18.8 eV (not MeV, not even keV) per molecule of H2S reacting.
    That’s not to say that finding a few moles of those compounds suddenly mixed in close proximity to you isn’t going to ruin your day. I’m pretty sure explosions on the scale of a kilo of TNT, with the added bonus of spectacularly toxic and corrosive byproducts, are frowned upon by most labs. But a similar amount of fissile material going off would obliterate a small city.

  104. Anonymous says:

    Haven’t laughed this hard in a while. Now I’m almost sad I’ve given up on being a chemist years ago …

  105. Anonymous says:

    How would you ship FOOF? I bet it costs like 500$ a gram but that doesn’t mattter
    | |
    ____/ \_____

  106. Kevin says:

    North Korea would love this stuff!

  107. Andrew says:

    #106 Dan, don’t forget that plutonium is “spectacularly toxic” in its own right, too. It is quite effective at killing you with its electron,s as well as its nucleus. I think I recall reading somewhere that a sand-sized grain of Pu was a thoroughly lethal dose.

  108. Pete O'Brien says:

    FOOF is indeed an unstable compound. It’s cousin, Phosphorus Dioxygen Flouride is also extremely volatile and disappears instantly when exposed to the atmosphere.

  109. flymousechiu says:

    xkcd people would like 1 kg sample of FOOF along with 1kg of HSSH for testing. combined shipping preferred.

  110. metaphysician says:

    While plutonium is nothing fun, my understanding is that its toxicity is overblown. A sand sized grain of Pu will kill you, but because of its radiation, not its chemical properties.

  111. zathael says:

    I’m looking at this Hydrogen Sulfide reaction and wondering if this is at all possible that 4 molicules of each would result in 433 kcals. I’ll admit I’m not the best at math, but when you scale it up to a gram… well. 2.06*10^26 kcals is… ridiculous. I can’t concieve of a chemical reaction releasing that much energy all at once.
    I really hope that I’m making some sort of mistake here, or that the units are getting mixed up

  112. Alsadius says:

    zathael, yes, you’re getting the units mixed up. Measurements of that size are always per mole, because not even annihilation reactions produce 433 kcal per molecule. The given amount FOOF and H2S is about 314 grams. Interestingly, that means its actual energy density is significantly less than ordinary food(314 grams of fat is over 2500 kcal, for comparison), it’s just much better at releasing it with a bang.

  113. Jbo says:

    Alsadius- yeah, if it were molecules instead of moles one kilo of the stuff reacting with excess H2S would be equivalent to a 10e+14 megatonne bomb, which is far more than the energy that would be released by converting all the mass to energy (using E = mc2), which would yield about 20 megatonnes. It would also be about a hundred million times more powerful than the Chicxulub impact that wiped out the dinosaurs and would presumably annihilate a considerable chunk of the earth…

  114. Gridlock says:

    If you want another scary thought, some nuclear engineering types out there want to run thorium reactor with a cooling system using liquid fluoride salts.
    I cannot imagine a worse design unless you coated the thing with arsenic and had cooling rods made of pure sodium metal.

  115. anonymouser says:

    Not that I want to alarm you or anything, but one incarnation of the LFTR uses blocks of carbon as moderator.

  116. will says:

    Didn’t “Dr” Gosnell use this to do abortions?

  117. Bryan says:

    The description of A. G. Streng reminds me of Cave Johnson, just doing stuff for science!

  118. Anonymous says:

    Then there’s the ever popular HOOOCCH. Note the dateline…

  119. treeowl says:

    I don’t know much about chemistry, but I’m a bit puzzled how you can make a chemical at 700C if it tears itself apart immediately at anything but absurdly low temperatures. Can someone explain?

  120. sveingold says:

    The comment on having FOOF mailed to you in an envelope by Hangzhou (way up in these comments) reminds me of my time as a chemist at a large elemental phosphorus producer. Not as immediately fierce as F2 or FOOF, but of course very oxygen sensitive; P4 can sit on a lab bench for hours after being lifted out of its protective water cover, but once it starts burning it tends to stick to you (literally – it sprays all over the place once it melts) causing deep almost incurable burns. A viper rather than a tiger. Apart from many legit applications it’s used as a reduction agent in amphetamin synthesis. One day some dope lab bozo asked me the price of 2 kg of the stuff mailed to him in an envelope. To which I replied “six years to life arson charges for setting fire to the post office”. I didn’t see anybody in this thread actually ordering FOOF via mail order, any news since 2010?
    BTW phosphorus and fluorine make for for an interesting reaction enthalpy. We tried it once, but unsurprisingly it didn’t exactly lend itself to scaleup.

  121. William o' WiSp says:

    Hang on,
    433kcal is given off for a reaction involving 5 MOLECULES of O2F2 and 1 of H2S?
    I may have studied biology, but if wikipedia is anything to go by 1g of TNT gives off 4.1->4.6kj of energy on exploding, so roughly 1kcal, for a GRAM…. Is there perhaps a typo at play here, how is this possible?

  122. Derek Lowe says:

    There’s been so much confusion on the thermodynamics that I’ve updated the post. That’s 433 kcal/mol (per mole), not per molecule. Deuterium fusion won’t even give you that much energy per molecule. But for a chemical reaction, this is still an alarming figure, believe me.

  123. G.R.L. Cowan says:

    If you want another scary thought, some nuclear engineering types out there want to run thorium reactor with a cooling system using liquid fluoride salts.

    Already done twice: the aircraft nuclear powerplant prototype and the molten salt reactor experiment.
    The quoted comment is analogous to someone expressing fear of rocks because he heard about the rock that killed the dinosaurs.

  124. Anonymous says:

    433 Kcal wouldn’t be possible even with antimatter let alone nuclear reactions. I quick calculation finds the amount of energy represents about 0.02 micrograms of matter.

  125. anonymus says:

    I need a link to all things he wont work with

  126. D.J. says:

    @anonymus June 7, 9:43: Look at the archives category on the right side of the page, and you will find and

  127. Not a Dr. Dot says:

    OK, I’m about to demonstrate just how much of a doctor I’m not. But I’m going to ask anyway, because I’m sure I’m missing something.
    The initial preparation of FOOF involves cramming a few hundred TORR of O2 and a few more hundred TORR of F2 at several hundred C.
    However, later in the article, it’s mentioned that FOOF will tear itself apart anywhere room temperature. How I’ll admit that my idea of room temperature might differ from the norm (I call Atlanta home, and I didn’t move here for the winters), I am having trouble seeing just how FOOF can be created at 700c, cooled to -70c or so without losing what little stability it may or may not have had.
    I guess I’m asking: How could something that requires over 700c to create and -70 (or so) to save actually exist?
    Love your stuff, Dr. Lowe
    Definitely NOT a Dr. Dot

  128. Collin says:

    The heat and pressure drive the reactants together, but they won’t stay together, so they have to be rapidly cooled. Basically, the molecule isn’t stable at rxn temps, but won’t react at stable temps. Well, won’t react to form itself, at least.

  129. Anonymous says:

    Having read Ignition!, I think this looks like it might be propellant work.
    The first few reactions are making sure it’s hypergolic with all the usual fuels. Then after checking some other compounds people might be interested in, he moves on to the high-energy oxidizers, probably to see whether it might be miscible with any of them. If it is, you might be able to create a new mixture that’s reasonably stable, yet more energetic than previous ones. (chlorine trifluoride for one, was ‘frequently’ mixed with perchloryl fluoride, as the oxygen helps hydrocarbons combust better)

  130. Anonymous says:

    The solution is quite simple…
    Just have tanks of oxygen and fluorine in your rocket. Run the lines down to the rocket bell, where they heat up, mix, and react, then up to the LOX line to cool down (heating the Lox on the way). Then inject into the rocket motor’s chamber.
    This is also a good way to set a new record for the number of rocket motor test stands destroyed.

  131. Anonymous says:

    Thermodynamics say that won’t be an improvement on simply using FLOX.

  132. Getheren says:

    #113: As with most other “toxic metals”, there isn’t a great deal of toxicity worry with plutonium metal because in metallic form its bioavailability is low. (By analogy, a sliver of arsenic metal in the skin is relatively benign — it’s *arsenic trioxide* that used to be known as “inheritance powder” — and lead bullet fragments that can’t be removed don’t cause lead poisoning.)
    It’s the soluble plutonium *salts* that really get down to some toxic rocking and rolling, because once they get inside an ugly bag of mostly water the plutonium ions will enter solution and be distributed throughout the organism.

  133. Helico says:

    It is good to see the clarification on the per mole versus per molecule calculation, although it would have been nice to see when I first read the article! I was originally directed to this article from the “what if” blog a year or so ago… I went through my high school chemistry notes for hours trying to reconcile the figures after I calculated out the energy released from reacting a pound or so of FOOF and some similar quantity of Hydrogen Sulfide. Using 433 kcal per molecule instead of mole yields thermal energy on the order of several hundred times that of the asteroid impact that killed off the dinosaurs… Now that is a scary proposition!

  134. Gregg Eshelman says:

    Anyone have a link to A. G. Streng’s 1962 paper that’s not paywalled?

  135. Mark Fergerson says:

    I don’t think you planned on becoming an Internet meme, but it seems to have happened anyway.
    To combine you with a couple of others-
    While reading your blog I noticed a joke trying to write itself in my mind. It begins “Uncle Al and James Nicoll walk into a chemistry lab…” and ends “…Ohshit KABOOOOOOOOM”. I couldn’t quite get the middle, then I realized that pretty much any of your TIWWW posts fits nicely.

  136. Autolykos says:

    @Gridlock: What molten salt reactors use are fluorides, which are about as benign as it gets. Once fluorine gets the electron it wants it’s happy, content and harmless. The on-site processing involving HF, ThF4 and UF6 is slightly more concerning, but nothing we can’t handle (in fact, we’re doing very similar things already for making conventional nuclear fuel).
    In a boiling (or pressurized) water reactor, on the other hand, you have tons of radioactive junk at a pressure well in excess of 80 atmospheres. And when the pressure drops and the fuel rods aren’t under water anymore… well, you could see that in Fukushima.
    A molten salt reactor sure isn’t something I’d like to have next door. But I wouldn’t mind it a few miles away. With older nuclear power plants however… I’m glad that the closest one is separated from me by a hundred miles and a chain of mountains.

  137. Sam Huff says:

    Larsia from “Sandra and Woo” just ask me to locate
    some FOOOF. Do anyone know a source.
    She just got a flamethrower from the internet.

  138. David Edwards says:

    Just took a look at the Lucia Streng obituary. So she was the person who first synthesised Krypton Fluoride? I’d say she was a spectacularly good chemist in her own right on that basis, which means she was probably well aware of the hazards involved in her husband’s work with Dioxygen Difluoride. If you’re going to have anyone present to help you with a hazardous chemistry experiment, what better than another gifted chemist?

    Oh, and having availed myself of a copy of the A. G. Streng paper, I note with interest the following comment on page 1381, just underneath equation (1):

    “Traces of H2O4 were also formed. No radical or other intermediate products have been found.”

    So now we can look forward, after Derek’s excellent piece on H2O3, to another piece on this presumably lively substance. 😀

    Back to Streng … apparently one of his intermediates, O2ClF3, is described in the paper as having an “intense violet colour”. I could not help but parse that the first time as “intense violent colour” given the nature of the reactions he was investigating! But, he not only worked with hideously violent oxidising agents, but produced a new one that was violet in colour. Trying to extract this for further study occupied a fair amount of the paper.

  139. arjowh says:

    high explosive+substance that burns through anyting = well i don’t know becaus when they did it i ran 5 miles upwind

Comments are closed.