Why does lemon sink in diet cola? I have made a funny homemade experiment, repeatable, and with a high success rate over 6 bottles of cola. I haven't found any trace on internet.
It reminds me the Mentos-Diet Coke experiment, but I am focusing on the lemon.
Take a bottle of sugarfree cola, e.g. diet Pepsi in this case. Cut a lemon and put some through the bottle's hole to give some taste to the beverage (we are in the process of making Lemon Pepsi...). Seal the bottle. Put it in the fridge.
As soon as you do it, the lemon floats on top the cola. The next day, however, the lemon disappears. Apparently disappears. You can find it on the bottom of the bottle.
Now open the bottle. It obviously fizzes. And then the lemon almost instantly appears floating, with a huge amount of foam trying to break through the top. This is the part where this experiment reminds me about Mentos.
But the Mentos experiment differs because the reaction is instantaneous and the bottle is left open.
Question
What does happen inside the bottle so that

*

*The lemon sinks when the bottle is closed and refrigerated (I haven't trying air temperature)

*After opening the bottle, the lemon starts to float again, and won't apparently sink anymore again

*When opening the bottle, a large amount of foam is produced and a small eruption occurs

Number #3 is likely to be the same reason (nucleation) of the Mentos experiment, but of course a lemon has different physical features than the Mentos candy.And when the bottle is shut the air pressure inside the sealed bottle may be increased.
I don't know the rest. Enjoy your drink if you like to try yourself!
 A: What happens is similar to what happens to a slice of lemon in a fresh glass of gin and tonic (or gin and any other carbonated 'mixer', like coke, lemonade etc).
When the cocktail has been freshly prepared it contains plenty of $\text{CO}_2$. Some bubbles invariably stick to the lemon slice, creating buoyancy (it floats)
But as the drink goes 'stale', either over time or due to intense stirring or shaking, all $\text{CO}_2$ has been removed and no more bubbles can adhere to the lemon slice.
Lemon has a density that's lightly higher than water, so that without the buoyancy provided by the $\text{CO}_2$ bubbles the lemon slice invariably sink to the bottom of the glass (just drop a slice of lemon in a glass of tap water)

Number #3 is likely to be the same reason (nucleation) of the Mentos
experiment, but of course a lemon has different physical features than
the Mentos candy.And when the bottle is shut the air pressure inside
the sealed bottle may be increased.

As pressure increases bubble generation stops and no more bubbles can adhere to the lemon slice, which is why in a closed bottle it eventually sinks to the bottom.
