It is incredibly hard to replicate with proof so I can only offer anecdotes and experience from hours upon hours of testing both in a heavily modded survival world designed to stress test 0.1.0 and a creative world with just the essentials installed (where it has happened far more frequently thanks to more extensive testing).
Whenever there is too much chemistry going on, the world TPS freezes out and eventually the game crashes. I am assuming that general chemistry stability is already planned, so that is not my focus. However, I have observed that in worlds in which a lot of chemistry has taken place, the game would often freeze up and crash in a similar way to when too much chemistry is occurring, even when there is only a single active vat at once, this means some reactions cannot take place in the older worlds without causing a freezing event but can easily take place in a fresh world.
What leads me to believe it is the vats not properly deleting their chemistry instances upon the vats destruction (when I say vat destruction, I mean one of the casing blocks being broken, I will refer to vat breaking and vat controller breaking separately) is that thanks to some unfortunate timing, my survival world saved on one of these freezing events, essentially making the world unplayable, so I went into the world's files with MCA selector and manually deleted the vat. This worked as expected, but when I created a new vat with the same controller that had been edited and subsequently destroyed said new vat, its contents did not leave along with it.
Since then, in creative, I have started habitually destroying vat controllers instead of just the vat itself, and have not noticed any further freezing events despite using chemistry just as, if not more taxing than previously.
I have included two of my worlds created in snapshot 10, with one being stable and the other being unstable.
It is incredibly hard to replicate with proof so I can only offer anecdotes and experience from hours upon hours of testing both in a heavily modded survival world designed to stress test 0.1.0 and a creative world with just the essentials installed (where it has happened far more frequently thanks to more extensive testing). Whenever there is too much chemistry going on, the world TPS freezes out and eventually the game crashes. I am assuming that general chemistry stability is already planned, so that is not my focus. However, I have observed that in worlds in which a lot of chemistry has taken place, the game would often freeze up and crash in a similar way to when too much chemistry is occurring, even when there is only a single active vat at once, this means some reactions cannot take place in the older worlds without causing a freezing event but can easily take place in a fresh world. What leads me to believe it is the vats not properly deleting their chemistry instances upon the vats destruction (when I say vat destruction, I mean one of the casing blocks being broken, I will refer to vat breaking and vat controller breaking separately) is that thanks to some unfortunate timing, my survival world saved on one of these freezing events, essentially making the world unplayable, so I went into the world's files with MCA selector and manually deleted the vat. This worked as expected, but when I created a new vat with the same controller that had been edited and subsequently destroyed said new vat, its contents did not leave along with it. Since then, in creative, I have started habitually destroying vat controllers instead of just the vat itself, and have not noticed any further freezing events despite using chemistry just as, if not more taxing than previously.
I have included two of my worlds created in snapshot 10, with one being stable and the other being unstable.
Minecraft username: FlynnGuy
unstable.zip stable.zip