What's The Most Dangerous Chemical In The World?

This question originally appeared on Quora. Answer by Keith Shannon.

Other answers including Franklin Veaux’s very good one mention dimethylmercury. That’s a really good candidate for “most dangerous substance”, but I got one better.

One row up in the same group as mercury, you’ll find cadmium. Most people’s experience with cadmium is in rechargeable batteries, where nickel-cadmium battery chemistry is still kicking around. The batteries are sealed to within an inch of their lives, and we’re all thankful for that, and even more so that the chemistry is dying out in favor of NiMH and various lithium ion chemistries. Artists, especially painters, will probably be familiar with the Italian-like red shade known as cadmium red; paints of this color used to actually contain the red-orange-brown cadmium oxide as a pigment, but thankfully most modern formulations of the hue have moved on.

Cadmium’s bad news. It’s acutely toxic, much like mercury, though again like mercury, swallowing the pure metal isn’t a very bioavailable way to poison yourself. No, to do that, you’ll need to breathe it in. The original way that happened was when cadmium miners raised a bunch of dust while working without masks, and that was bad enough. Unlike various forms of “miner’s lung” like silicosis, asbestosis, coal pneumoconiosis (black lung) etc, there’s not really a classification of “cadmiosis” as a disease, because inhaling anywhere near enough cadmium dust to cause a chronic degenerative lung disease will kill you in hours from the acute effects, not years. Cadmium’s mechanism is very similar to mercury’s, running around collecting electrons off the organic molecules in your cells in turn causing those compounds to self-destruct, and there’s only so long you can keep that up. It’s also, for much the same reason, extremely carcinogenic, so if you survive the initial poisoning it’ll most likely take you down with aggressive multisystem cancer just to spite you.

However, cadmium is just the elemental building block of my contender, which is the cadmium homolog of dimethylmercury, known as dimethylcadmium.

Dimethylcadmium is bad. How bad? Bad enough that chemists switched to using dimethylmercury in reactions wherever they could, just so they didn’t have to deal with dimethylcadmium.

Because the base metal is lighter, dimethylcadmium is more volatile and reactive than dimethylmercury; it has a lower boiling point and a higher vapor pressure at room temp, both of which means the stuff readily evaporates into the air. You don’t want it there; 3 millionths of a gram of dimethylcadmium per cubic meter of air volume is the safe limit for exposure, and not for long. Dimethylcadmium is an even more bioavailable inhalant than cadmium oxide, doing much the same damage with the kicker of methylating your lung tissues on the way in.

It also yearns to simplify its structure (translated: highly reactive), so anything it touches that’s more willing to share electrons than carbon, it will readily and exothermically react with. It’s not truly pyrophoric; that distinction belongs to much lighter methyl-metallics like dimethylzinc and trimethylaluminum. Those will burst into vigorous flames on contact with oxygen in the air at room temp, and that’s actually a good thing because the byproducts are fairly benign zinc and aluminum oxides, so any toxic effects the dimethyls might have are purified in fire. Dimethylcadmium, however, requires a fairly large footprint of a spill before its natural oxidation in air generates enough heat to hit autoignition, and when it does, assuming you survived the blast, the combustion byproduct of cadmium oxide dispersed through the air in a haze is almost as bad as the dimethyl.

Smaller spills are just as dangerous. The chemical penetrates latex and other organic barrier materials just as easily as dimethylmercury, which is how Karen Wetterhahn died handling that fine example of hand sanitizer. There aren’t too many notable examples of dimethylcadmium poisoning, but that’s largely because of what I said previously; if there’s anything else, even dimethylmercury, that does the same job in a reaction, chemists will use it instead.

Assuming you didn’t spill it on yourself, you still have the vaporization/inhalation hazard, but then you also have the question of how to clean it up. Maybe you rinse it down the drain? Well, water will cause a series of decompositions producing a bunch of delightfully flammable compounds including hydrogen gas, methane and methanol, plus more than enough heat to ignite it all, so yeah, bad idea. Sweeping it up generates friction which can cause it to ignite, and also could bring it into contact with things the components of the molecule might like better (like the organics in dust, or a droplet of water on the lab bench etc), producing exothermic reactions generating heat, so no-go there. Maybe you just ignore it? Well, it will oxidize in air without releasing the dimethyl, producing a powder of dimethyl cadmium peroxide, which is an even more friction-sensitive explosive, so when this red-brown dust settles on the floor you’re just one shoe scuff from a harp and halo.

All of this wonderful reactivity is why the number one preference of chemists the world over, given the prospect of working with a dimethylcadmium reaction, is obviously to quit the field and spend your remaining years hand-dipping hungry wolverines into vats of cold Worcestershire sauce. At least you’ll be able to enjoy the workman’s comp money. In all seriousness, newer and (relatively) safer methylating reagents such as organo-zinc, organo-lithium and organo-aluminum compounds, as well as their much tamer halides such as Grignard reagents (alkyl-magnesium bromides) are available to the modern organic chemist, so as long as you don’t need the mercury or cadmium itself in the product of your labors (and why in God’s name would you want that), you would be much better served making your reaction work with something a little more user-friendly if you can. Not that the alternatives are that much easier to handle (see the aforementioned dimethylzinc), but usually, the primary danger of these lighter organometallics to the chemist is their tendency to burst into flame, and fire is quite honestly one of the least lethal things in the average organic chemistry lab.

Here’s a wonderful YouTube from the SciShow vlog with more on this fine chemical, and four others, each dangerous in their own way:

… And except for the last entry, all of these chemicals (and a lot more) are found on the blog of Dr. Derek Lowe, in the category very pointedly named “Things I Won’t Work With.” When you consider he is a drug discovery research chemist charged with creating complex compounds using any means he can think of, and you read some of the related stories about things he has worked with, a statement that these chemicals are firmly in the territory of “noli me tangere” should give pause for consideration.