[airplanet56]

Since I'm founding out that so few elements follow the Octet Rule, what are some strategies for guessing how many valence electrons are wanted for atoms in higher periods? Whenever I look up exceptions to the Octet Rule, all I see is Phosphorus and Sulfur in compounds with Chlorine or Fluorine. This isn't really helpful. For elements such as Silicon or Selenium, how do I know how many valence electrons they want?

[magician4]

chemistry isn't as "precise" as physics with respect to straight laws that never can be violated

there is one "law" (let's better say: a rule for calculating energies) which will apply in chemistry without exception nonetheless: the quantum electrodynamics [this one  alone includes the explanation for the strange behaviour of for example Mercury]
... but unfortunately we didn't even beginn*⁾ to seriously calculate systems with this tool: the equations almost always are too complicated to be solved
the little sister of QED (in fact, it is more like it's mother, as it's been developed a generation  earlyer) is quantum mechanics. it is not as "flawless"**⁾as QED , but you can at least do ab initio calculations the results of which will match reality, i.e. the molecules will do what you calculated to be the best for them.
but let's not be too cocky about this: still we can't calculate ab initio even the behaviour of a single drop of pure water -  the sheer numbers involved will eat up that much computerization time, it usually can't be done.
so, we're down to semiempirical approaches to the problem, introducing part of real values measured into the equations, to facilitate calculations: this will work, still timeconsuming, but will work.
at least for simple problems.

nevertheless, the results even of those calculations often are very cryptic, and can't be easily morphed into something like "chemical principles": they're more of a mathematical solution than something you can "grasp" as being part of a more general rule.


so, the more precise you want to be, i.e. the less "exceptions" you want to allow for, the less comprehensible (in a human , psychological sense )  the situation becomes.

as a result, we prefer to live with a patchwork of "local rules" in an ocean of uncertainties and in-betweens, sometimes as isolated from each other as the Easter Islands, sometimes more like the Hebrides, and every now and than more like a small and comfortable continent of knowledge .

the octet rule is such an "continent", and often will be presented to beginners in a way that sometimes embezzles that there's an ocean around, even with a continent.

your question is like standing at the shore of such an continent, and wondering what the situation with the islands out there might be.
Yes, there are , again, local rules out there which will be usefull: "stability of the half-filled shell" , "18 electrons rule" and thatlike

... but to learn them, and to learn how to apply them and how to predict what will happen, usually requires advanced chemical education.
this, on the other hand, should be accompanied by a lot of factual knowledge about what which substance will do, is like and how to understand it

so, in the beginning you'd better accept those exceptions from the octet rule and learn them by heart - at least for the time being

btw.: esp. sulfur is an element the understanding of it's bonding situation will chance over the years several times for you: first you will learn that often it's an exception to the octet rule  (like for example in sulfuric acid)
then you will learn that the octet rule isn't violated at all for sulfuric acid, if you regard it as being formally charged at the sulphur as ^(+II)S(-O^-)₂(OH)₂
... and you will remain with the revelation, that only for SF₆ and thatlike there is octet-rule violation.
then you will learn about modern bonding theory, and strange bondings like "2 electrons-3-center-bonding" and thatlike, and suddenly SF₆ again becomes a molecule obeying the octet rule (scroll up to p. 160 to study pictures of those bonds)

... and when you stay around in chemistry long enough for your PhD, this again will chance,  when someone shows up with an article regarding QM-calculations of SF₆, claiming that at least a certain - if only smaller - amount of d-orbital-density is involved nevertheless

so , for the time being: just learn what higher elements would behave like, IMHO there is no other way

regards

Ingo



*⁾
if memory serves, first usefull calculation from Feynmann diagrams was done by the end of the 90'th -  and there hasn't been much additional advancement since

**⁾
quantum mechanics is incompatible with Einsteins theory of relativity

[Big-Daddy]

but let's not be too cocky about this: still we can't calculate ab initio even the behaviour of a single drop of pure water -  the sheer numbers involved will eat up that much computerization time, it usually can't be done.

I thought it had been done for water? (Full CI, which I take it is what you mean.) This is hardly solving the n-body problem ... water is treated as a single species, right? We're not trying to model the microscopic behaviour of each particle, just the macroscopic behaviour of the particles on average, so it should be doable for small molecules?

[magician4]

no, that's not exactly what I meant

the idea was to calculate ab initio all system values for a single drop of water , which by the end of the day would result in a solution showing for example phases (i.e melting point incl. density anomaly, ice-phase diagram, triple point and thatlike) , dynamic smallsized structures (liquid-ice insulae) , pH properties...
... i.e. the full scale of all known properties of liquid water

to the best of my knowledge this hasn't been done yet, though I am aware of semi-empirical calculations for some of the subquestions (i.e. the dynamic structure, if memory serves)

but if you know of any new development, I'd be happy to learn about it, and I'd be glad if you could bring some literature I've missed to my attention

regards

Ingo

[curiouscat]

to the best of my knowledge this hasn't been done yet, though I am aware of semi-empirical calculations for some of the subquestions (i.e. the dynamic structure, if memory serves)

I think you are right. Till a few years ago there was a regular flow of papers to Nature-Science simulating H2O molecules in ice.

Don't think everything's been resolved yet.

[Babcock_Hall]

The notion that sulfur in SF₆ actually breaks the octet rule has been challenged by Reed and Weinhold in a paper in JACS circa 1986.  Frank Weinhold has written some articles on the subject since then.

[magician4]

The notion that sulfur in SF₆ actually breaks the octet rule has been challenged by Reed and Weinhold in a paper in JACS circa 1986.  (...)

If you're referring to this paper:

A.E.Reed , F.Weinhold On the role of d-Orbitals in SF₆  J.Am.Chem.Soc. 108 (1986) , 3586

my reading here is, that by NLMO - analysis it has been shown that the hybridization at the sulfur in SF₆ should be described as sp^1.7d^0.16

... and that is an example of what i called

nevertheless, the results even of those calculations often are very cryptic, and can't be easily morphed into something like "chemical principles": they're more of a mathematical solution than something you can "grasp" as being part of a more general rule.

neither does this support a "clean" sp³d²  , nor is it something that matches octet-rule obeying type hybridization in my opinion

as I said with this article in mind:

... and when you stay around in chemistry long enough for your PhD, this again will chance,  when someone shows up with an article regarding QM-calculations of SF₆, claiming that at least a certain - if only smaller - amount of d-orbital-density is involved nevertheless

.. and I intentionally didn't elaborate about what those results might mean in categories of "hypervalent or not" - because I don't see how this should fit in either one

maybe I just don't understand how to morph above broken hybridization into something meaningfull with respect to our human categories of "hypervalent" or "octet".
on the other hand, maybe those manmade categories all by themselves are the problem, because mother nature once again choose to give a dam 'bout what we think she should be organised like and just did it , no matter how systematic it looks to us from the outside.

my personal choice is, to just accept the facts, no matter how psychologically disturbing they might be for someone who is looking for order in chaos

there's an ocean of things still to be discovered and to be understood around our islands of "knowledge"...

regards

Ingo

[Babcock_Hall]

I think you have done a good job explaining some of the complexities involved.  Thanks.