I hear that Google tests prospective employees with weird vaguely-science-related riddles. If I were in charge of this, here's what I would ask:
You're an American spy in Cuba. The CIA has gotten you a position refilling the water coolers in Castro's presidential palace, hoping you can poison him. But Castro's security is pretty good. Every time you enter the palace, they search you so exhaustively that you're sure you can't smuggle anything in. And you're sure you can't access any poisons within the palace. And every time he drinks water, Castro calls in a chemist to test it for any impurity first; the tests can detect any contaminant at any concentration. On the plus side, you're completely unsupervised within the palace, and have access to a kitchen with all the usual kitchen appliances. And the CIA has given you a time manipulation gizmo, so you can take literally as long as you want, even if it’s thousands of years. How do you kill Castro?
One answer: Start with an amount of water several thousand times Castro's usual daily consumption. Put it in the freezer until it's half frozen. Dump out the unfrozen half, melt the frozen half, then repeat this process with the meltwater. After some very large number of cycles, put the result in Castro's water cooler every day. He'll be dead within a year.
The hydrogen in water is a combination of normal hydrogen (only a proton in the nucleus) and deuterium (a proton and neutron in the nucleus). These have slightly different chemical properties, so you can do various types of distillation to enrich for one or the other, including repeated freezing (realistically freezing works very slowly; our hypothetical spy would need an unrealistic amount of time, water, and patience). Normal water is about 99.9% H2O, 0.1% HDO, and negligible amounts of D2O. Water with more D2O than normal is called heavy water, water with more HDO than normal is called semiheavy water, and water with more H2O than normal (ie not even the usual tiny amounts of the other two constituents) is called light water.
Heavy or semiheavy water is pretty toxic. It's not radioactive or anything. It's just that your body is 70% water, and if the water has even slightly different chemical properties than usual, that screws up a lot of stuff. It's the perfect poison - except that you need to replace a significant proportion of someone's body water with semiheavy water before they feel any worse; realistically they would have to consume gallons and gallons. Probably beyond the reach of most poisoners, which is why the question specified you're refilling Castro's whole water cooler for months on end.
If heavy water makes you feel worse and eventually kills you, naturally light water should make you feel better and eventually make you immortal. That's how logic works, right?
Maybe! This is actually a real alternative medicine thing. You can buy light water for about $20/liter (though, like Castro, you would have to drink nothing but light water for months before you replaced enough of your body water to matter).
I love this theory. It's so much better than homeopathy. If you tell the homeopaths that their products are just ordinary water, they'll get pissy. If you tell these people, they'll admit right away that it's even more water than regular water is, the only water which you can be really confident is perfectly normal H2O. I love this theory so much.
Start with the evolutionary argument. During the Ice Age, a lot of the world's water froze. Just as your freezing concentrated deuterium in Castro's freezer, so Ice Age freezing concentrated it in the glaciers. Also, rainwater naturally has less deuterium when it's colder, because heat preferentially increases the evaporation of heavier water. So drinking water was probably a bit lighter during the Ice Ages, when humans were evolving. The 10,000 years since the last Ice Age wasn't enough time to re-evolve, so human bodies are probably adapted for slightly lighter water than we get.
Move on to the molecular biology argument. The basis of cellular energy generation is the proton pump in the mitochondria, which moves hydrogen nuclei around to create a charge gradient. If the hydrogen nuclei have unexpected neutrons, that could jam the pump or change how the gradient works. I'm not entirely able to follow this, but maybe then there is some kind of cellular signaling cascade? Basov, Fedulova, Baryshev, and Dzhimak:
Deuterium reduction in matrix water can regulate not only functional activity of mitochondria, but also speed of other cellular processes caused by transfer deuterium-depleted water to the cytosol, and as result of the foregoing, cell ATP-production [78,79] and interfacial protein interactions are changed and, consequently, cellular rate of growth can be various due to deuterium loaded in molecular structures of living cell.
See also this paper.
Then there's the anthropological argument. The Hunza people of Pakistan drink mostly glacial meltwater. Because its freezing-and-melting cycle replicated Castro's freezer in reverse, their water is naturally lighter than usual. They also seem to be unusually long-lived and healthy.
And there's the oncological argument (that's the St. Anselm thing, right?). You can read a review paper here, and it's pretty impressive. Light water halves the growth rate of cancer cells in vitro, changes the ratio of mitotic stages in vitro, and seems to help treat cancer in rats. In a human double-blind randomized controlled trial, light water significantly increased survival rates. There are case studies of otherwise untreatable cancers remitting or stabilizing in the face of light water treatment.
Is this all bunk?
Well, at least most of it is. For example, the Hunza people of Pakistan were deemed supernaturally long-lived and healthy by the same sort of early British explorers who repeated legends of Shangri-La and the Abominable Snowman. According to reputable sources, the average Hunza life expectancy is fifty-something, same as most other groups in poor parts of Pakistan. So probably light glacial meltwater isn't a very good longevity secret. Also, assuming that we're evolved to Ice Age conditions limits evolution to a pretty specific speed - any faster and we'd evolve to present conditions; any slower and we'd still be pre-Ice Age. I'm not too impressed with the rat studies, and other studies on plants suggest they grow better in high-deuterium conditions, eg this thesis suggesting that "deuterium oxide has a stabilizing force on biomolecules". Most of the positive studies were done by the same Hungarian team, or by people with sufficiently-Hungarian-sounding names that I suspect they're affiliated. The anticancer studies are better quality than the others, but some other groups find deuterium causes less cancer, so who knows?
This probably sounds like faint condemnation. I don't have good responses to all the arguments in favor. The problem is, every alternative medicine fad has evidence approximately this good. Some are better; there are dozens of positive RCTs for homeopathy. This is why the replication crisis sucks so much - just because a few converging lines of evidence support a theory and it has lots of positive studies including an RCT or two, doesn't mean it's any good. Where do you draw the line?
I find myself pretty unconvinced, but I wouldn't mind someone looking through the cancer studies to tell me if I'm missing something. At least a couple of different groups seem interested, and some of the cell line studies seem to show major effects. Not sure what's going on there.
This probably doesn't have enough medical benefits to have been worth my time to research or yours to read. I still find it fascinating. I keep being amazed at how many dimensions things can vary along. You think you know what kind of things medicine has to investigate - how different chemicals interact, the effects of food and smoking and sleep and so on - and unless some weird Hungarians remind you, you would never in a million years remember that there are multiple different isotopes of water and this seems to have some effect on living cells. You would never think to check whether attempts to mine the Martian icecaps for drinkable water will result in dangerous water that could sicken the unfortunate astronauts who drink it (answer: it might! Martian water has five times more deuterium than Earthly water and seems to kill shrimp). You would never think that you could buy something called "deuterium depleted water" on Amazon, or that it would be completely safe to drink. But here we are!