I'm not as impressed as I guess I was supposed to be with this news, and certainly not as impressed as the New York Times headline-writer.
However, poultry farms feed arsenic to chickens, which seems to improve their growth. Lots of things, like selenium, which had a run in health-food stores a while back, may be essential to life in small amounts and damaging in larger amounts. That's even true of iron. So the "poison" label on arsenic isn't that important chemically, although it has a lot of resonance in literature.
But the reason I'm not impressed is that I've seen something very much like this before, up close and personal, my very first introduction to chemical research. Or official chemical research, anyway. I did a bunch of things earlier on my own.
I worked in Harold Strain's laboratory at Argonne National Laboratory. Harold was one of the most skillful separator of plant pigments in the world. He used big (a couple of feet high by a few inches across) columns of packed confectioner's sugar. The plants were mashed up, extracted with light petroleum ether (boils in the palm of your hand), and adsorbed on the column, "developed" with additions of alcohols to the solvent, and then sculpturally scraped, sugar and all, out of the column by color and re-extracted from the sugar. The process typically gave a few milligrams of each pigment. Spinach was a good starter material for the standard pigments.
Those were the days when differences among isotopes were a big deal, nuclear magnetic resonance was new, and it looked like it might give more information about the processes by which chlorophyll converts light into chemical energy. If you had normal chlorophyll and chlorophyll with deuterium everywhere that normal chlorophyll had a hydrogen. Then you could use NMR to watch particular deuterium atoms exchanging with hydrogen atoms and get some of the kinetics.
Deuterium is an isotope of hydrogen. Whereas arsenic is below phosphorus on the periodic chart, deuterium shares hydrogen's box. But, because hydrogen is the lightest element, deuterium is approximately twice as heavy as hydrogen, a much bigger mass difference than between arsenic and phosphorus. This mass difference makes some difference in the chemical reactions of hydrogen and deuterium, although phosphorus and arsenic have greater differences. Further, deuterium is very uncommon, about 150 atoms in a million hydrogen atoms. So how does one go about obtaining chlorophyll with 72 deuterium atoms in place of its hydrogen?
The way it was done by Joseph Katz at Argonne was to grow algae in deuterated water, also known as heavy water, commonly used in certain types of nuclear reactors, and certainly, at that time, used in nuclear research. As Joe told it, the algae didn't like it at first, and got sort of fat and bulgy, a few died, but most eventually settled down and looked pretty much normal. But their chlorophyll was deuterated.
So Joe's lab grew the algae, and then dried and processed them to get a separable extract, which Harold then put on his sugar columns. The deuterated pigments were sealed into glass tubes under vacuum, to be opened for the appropriate experiments. We speculated on how much per pound those pigments were worth. A lot, but not many potential buyers.
Algae are bigger, with more enzyme systems and biochemical conversions than the arsenic-based bacteria. It's probably even up as to difficulty for the little guys to survive in such alien media, though. Which aren't nearly as alien as what other planets have to offer. They're still carbon based. And some of us have seen something very much like that before, although it was a long time ago.
If you google "algae deuterium oxide", you'll find a fair number of publications from Katz's group. Apparently Argonne is still growing stuff in deuterated water. And I think there's a black and white video of me talking about what I was doing in Harold's lab somewhere out there on the intertubes, although I couldn't find it with a quick google.
Update: And we have to add in XKCD's contribution. (h/t to @jfleck)
Further update: More here.