Why gold is yellowish?

Question

This question -- Why are most metals gray/silver? -- is partly about why gold is yellowish.

The most popular answer (and thus I assume the right one) -- https://physics.stackexchange.com/a/72412/340807 -- states that normally the transition from 4d to 5s would need too much energy for visible light. But electrons in s orbitals are likely to be closer to nucleus. So, in gold, since the nucleus has a lot of protons and thus charge, electrons in 5s need a lot of energy to avoid falling on the nucleus, so because of relativity and Lorentz transformations the difference in energy between 5s and 6s is lower. It becomes low enough that it goes into the bluish range, so gold's electrons absorb blue and thus gold is yellowish

I have three questions about this still (I don't have enough rep to comment):

1. If the electrons in gold are at relativistic speeds because gold have a high atomic number, then why aren't metals with a higher atomic number than gold yellow (like mercury is silvery)? Is it because the Lorentz transformation is so much that it overshoots and the transition energy is now below visible light frequency?

2. In the graph it shows that at lower wavelengths silver reflects less. So why isn't silver yellow (besides the name)?

3. In the graph for silver, why does the reflectance apporach 0 around 350 nm?

Answer 1

The electronic structure of heavy atoms is fiendishly complicated because we have a lot of electrons all interacting with each other and all interacting with the nucleus, and the result is a system with a large number of degrees of freedom.

To add further complication we rarely see the colours of atoms because with the exception of the noble gases we are seeing the colours of molecules. Metals are even more complicated because in metals the atomic states interact to produce energy bands. So when we see that gold is coloured we are observing transitions between energy bands not simply between atomic orbitals.

The point of all this is that while it is certainly true that relativistic effects play a role in the colour of gold, to say that gold is coloured because of relativistic effects is probably taking this a bit too far. Gold is coloured because that's just the way the band structures in the metal happened to work out. For example copper is also coloured, but no-one invokes relativity to explain that, and compounds of metal ions are frequently coloured despite the electronic configuration of the metal differing by only one or two electrons.

The bottom line is that to look for simple patterns in this is futile because it is so immensely complicated that the colours of all solids (not just metals) are effectively random.

To return to our original point, it is certainly true that without relativistic effects gold wouldn't be coloured. However any small change, like changing the mass of an electron slightly, would also change the spectra enough to probably remove the colour as well. So to say that gold is coloured because of relativity is true but not useful.