Can we predict the properties of a molecule?

Answer

Dear Nelson,

That’s a very interesting question. With what we know today about how nature works, it is possible to predict the properties of a newly invented molecule. It should be noted, however, that not all properties are equally easy to predict.

Let’s start simply with the molecule you are proposing. The behavior of the atoms and their electrons is described by the laws of quantum mechanics. While we could write down the equations describing the quantum mechanical behavior of the molecule you give here, solving them is far from easy. Fortunately, we don’t have to do this by hand, but there are specialized computer programs that can do this for us. On the English Wikipedia you can even find a list of the most famous of these programs (some are free, others are paid). These programs solve the same equations, but with a different goal (focus on a different property) or with a different method. For a simple molecule as you write here you can calculate some properties fairly quickly (less than a few hours on your own computer), for other systems and properties the computation time can take many weeks or months on a supercomputer.

One of the first things one will try to predict is whether the molecule is stable (ie whether it can exist or will disintegrate). This is done by calculating the energy of the molecule and comparing it with the energy constituent components. Knowing whether something is a gas, liquid or solid at room temperature is also possible, but that is already a bit more difficult. Gas and solid are the easiest to simulate, and allow you, back by comparing the energy of the gas-phase and solid-phase molecule, to see which one will be the most stable.

The color of a molecule could be derived from the energies of the electrons. From this you can determine the spectrum. A similar calculation for the solid-state version of your molecule tells you something about the conductivity. The elasticity of a solid can also be calculated.

As you can see, a lot is possible. However, there are two important comments to make:

* The calculations you perform are usually for perfect solids (without cracks, holes, or misaligned atoms), which you will never actually have in an experiment. These errors can greatly alter the properties. Take, for example, diamonds. This is an insulator. However, in real diamond there is always a small amount of nitrogen which alters the electrical structure (and can cause a yellowish color). If you put a little bit of Boron in the diamond (it will turn bluish in color) your diamond will become a conductor, or even a superconductor (if it is a lot of Boron).

* Although you can calculate these things, it can sometimes be faster to make them in a lab. Calculations for a realistic material become very complex very quickly because you have to make all possible combinations by hand and try them out to see which one is the most stable. Nature (in the lab) does that difficult work for you. The disadvantage in the latter case is that you do not know which combination it is. Are you going to combine both things; on the one hand making a material in the lab, and on the other hand calculating with the computer, then you can combine the strength of both. The lab work can easily give you information about properties that are difficult to calculate (eg solubility) while the computer work can tell you exactly what the atomic structure is: Your H2FeOH is that 1 molecule or rather HFe + H2O? or H2+ FeOH?

Regards,
Danny