# Will molecules containing heavy isotopes tend to rise up in a liquid?

When you have a bag of chips or nuts or something else with big and small pieces and you shake it, the bigger pieces will rise up and the smaller pieces will rise down. This is known as the brazil nut effect.

I'm curious, does something similar happen with molecules in a liquid? For example: say you have a glass filled with water. Is the concentration of $\ce{HDO}$ and $\ce{D2O}$ larger at the top in comparison to the bottom?

• I'm quite sure that random thermal Brownian motion at just about any temperature above absolute zero completely mixes the isotopes . The drive towards isotope segregation would be due to a decrease in overall gravitational potential of the system, but gravity is an extremely weak force on the scale of molecules, and this tiny "signal" would be completely swamped by the "noise" of thermal fluctuations. The answer would end up being similar to this. Jun 28 '16 at 22:14
• The only system with isotopic segregation I know of is the phase separation of He-3 and He-4 at a fraction above $\mathrm{0\ K}$, which may meet the requirement in part, but this is a very special case. Jun 28 '16 at 22:16
• Molecules don't behave like different sized lumps in a cereal packet. And the behaviour in a pack of nuts, for example isn't about weight but size. Jun 28 '16 at 22:25
• Why, of course they would, but the effect would be ridiculously weak. Think of centrifugal isotope separation. Jun 29 '16 at 6:03
• As a side not, the Brazil nut effect is not well understood.
– bon
Jun 29 '16 at 7:53