Talk:Valence shell

Untitled

The passage beginning

..."For organometallic compounds containing transition metals"

seems odd and out of place, but would make sense if the words

  "organometallic compounds containing" were deleted.

I've not edited the page, as perhaps I'm missing something, but I suspect that someone had an accident cutting and pasting and didn't realise that those words had been inserted.

Richard Jones — Preceding unsigned comment added by 61.88.6.98 (talk) 04:38, 17 August 2004 (UTC)

Target

Related to talk:Period 4 element‎‎ #From zinc to krypton: is it correct that the page redirects to electron shell not to valence electron? Incnis Mrsi (talk) 09:21, 19 August 2019 (UTC)

As the target says, the term "valence shell" is rather ill-defined for groups 3 through 11 and the lanthanides and actinides included, because valence electrons for those elements come from multiple shells (e.g. Sc–Cu can use 3d and 4s). It is true that for groups 3–10, the main contribution is from their d-shells, but the s-shells make some contribution too (there is sd hybridisation, after all). Double sharp (talk) 13:13, 19 August 2019 (UTC)

Let’s put aside transition elements for a while. What should be the “valence shell” for p-elements in period 8? The classical (n-based) definition of a shell works no more. Incnis Mrsi (talk) 13:27, 19 August 2019 (UTC)

@Incnis Mrsi: Element 172 would have a stable octet of 9s²
9p²
_1/2 8p⁴
_3/2 as the valence shell, so we could consider 9s+9p_1/2+8p_3/2 to be the "valence shell" for those p-elements. Double sharp (talk) 14:05, 19 August 2019 (UTC)

There is “electron shell” (having a staunchly n-based definition, although I recently added disclaimer that it frequently fails), there is also an article “valence electron” (giving a lot of examples but no definition), and this page redirects to “electron shell”. The current situation is very poor anyway, waiting for suggestions. We have at least three situations: “classical” p-elements (where n-shells work well), transition (d and f) elements, and “spin–orbit” p-elements. The definition eventually must cover all three cases. Incnis Mrsi (talk) 19:18, 19 August 2019 (UTC)

@Incnis Mrsi: A common-sense definition of valence shells (which lets you talk about transition metals, etc.) would be "the set of all orbitals that can be used for chemistry for that element", even if they might not have the same n, because those will be the orbitals that contain the valence electrons. This should always be some subset of the subshells corresponding to a row of the periodic table, i.e. {1s}, {2s, 2p}, {3s, 3p}, {4s, 3d, 4p}, {5s, 4d, 5p}, {6s, 4f, 5d, 6p}, {7s, 5f, 6d, 7p}, {8s, 8p_1/2, 5g, 6f, 7d, 9s, 9p_1/2, 8p_3/2} (further we have not predicted far enough, but at least 6g, 7f, and 8d ought to be in the next set). This is because of the general rule that you cannot "breach a period" and take electrons past a noble-gas closure if you started after one, though relativity might create an exception for E119 and maybe E120 by destabilising 7p_3/2. As the eighth period is expected to be so long, orbitals with low angular momenta active at the start of it will sink down into the core even before we reach the end (which is why 8s+8p_1/2 get replaced by 9s+9p_1/2 when we get to the d-block and p-block of that period), but the end result of listing all the valence orbitals still gives a subset of the set of orbitals that are supposed to correspond to the period. Double sharp (talk) 06:28, 20 August 2019 (UTC)

@Double sharp: I have settled this question in favour of valence electron. Valence "shell" is not a shell. It is a misnomer. To be honest it doesn't have much to do with an electron shell but everything to do with valence electron. Google searches indicate that valence shell and valence electron are mostly frequently discussed together and are the top two searches for valence_something. The word "valence" is primarily a chemistry term for binding power so it makes sense that all orbitals that can bind are counted in the valence shell. Now electron shell is back to describing what it does best: n stands for shell, l stands for subshell.--Officer781 (talk) 05:54, 26 May 2020 (UTC)