I’m working on an in-depth video game that uses procedural generation for 3-dimensional meshes, which got me thinking about how the laws of chemistry might be imitable on a macroscopic scale.

My goal is to start small with a testing grounds standalone model, that can be later added to or implemented into another game for it’s features. The testing grounds being a moderate Laboratory equipped with Bunsen burners, graduated cylinders, beakers, flasks, autoclaves, storage cabinets, reaction chambers, kilns, etc. Where a player can experiment with making composites, alloys, reagents, solvents, and other things that will be useful later.

I have an intermediate understanding of physics and chemistry, so I know it’s possible to do, but I’m wondering if I need to have a property library for every possible molecule or can I do this procedurally? Say I had a library of ALL of the elemental properties and applicable laws for combining and breaking molecular structures, will I need a separate properties Library for water, vinegar, etc.?

My main concern is trivial, since I know the probability is slim, but let’s say someone gets Hollywood on my game and creates an extremely sophisticated polymer chain that has a surface hardness above 10 Mohs, melting point of 2000 Kelvin, and molecular weight of 1 Dalton … I know extreme, but should I even worry or is having this as trivial as the possibility of a player “transmuting” lead into gold? (which is possible BTW, just not feasible in “normal” circumstances since particle acceleration isn’t used for such non-sense).
I only used the above scenarios because like any video game, there is always that one person who has to break things, so whether it’s a plastic-boy suit or irradiated gold I want to avoid them.

I’m not worried about the graphics to the point of 8k or whatever it is ATM, or even tracking individual atoms or molecules.
Just simple: if I have a library of elemental and chemical properties/laws, do I need a Library of every molecule possible as a control/constant to test the results and their validity/accuracy?

Please don’t spam the comments section with irrelevant answers, if you have a question ask a question don’t assume I’m a teenager asking questions too big for her britches, else your wasting this forum’s time and space with non-sense that could have had a teaching point that’s glossed over with you garbage. So please teach us all, if we’re ambitious enough and have the resources to do this and do it what are some possible issues that may arise outside of “bugs”.
Thank you for your time. All serious answers are extremely appreciated.

Created an abstract with documentation of how the system will work.
Basically the library will act as the rules for how the 3D meshes interact and transform one another on a macroscopic chemical level, visible to the naked-eye. It is already possible to use statistical data for the specific material in question to be “broken” at “weak-points” that are generated for a mesh upon impact, for dynamic destruction of meshes.
I’m looking to remove the cookie-cutter material data, with procedurally generated material data using approximate molecular properties that would result from a given scenario; where the only libraries are for the periodic elements and laws of chemistry and physics, to define the material properties of the 3D mesh.

I’m not expecting this system to be easily exploitable, but Murphy’s Law says otherwise.