[Big-Daddy]

Why does S₂O₃^2- adopt the structure it does rather than something like this:



Where both of the negative formal charges are on the more electronegative atoms (oxygen), whereas in the real structure of S₂O₃^2- one of the negative formal charges is held by an S atom?

Whereas in every other S_xO_y structure I know of (S₂O₆^2-, S₂O₈^2-, S₄O₆^2-, S₃O₆^2-) the structures are both more similar to the one I wrote above, and only have negative charges held on or delocalized over oxygen atoms?

[Hunter2]

The thiosulfate ion is built like a sulfate ion, where one oxygen is replaced to sulfur.

You can draw mesomere forms in Thiole and Thion form.

^-S-SO₃^- or ^-O-SSO₂^-

http://en.wikipedia.org/wiki/Thiosulfate

[sjb]

Why does S₂O₃^2- adopt the structure it does rather than something like this:



Where both of the negative formal charges are on the more electronegative atoms (oxygen), whereas in the real structure of S₂O₃^2- one of the negative formal charges is held by an S atom?

Whereas in every other S_xO_y structure I know of (S₂O₆^2-, S₂O₈^2-, S₄O₆^2-, S₃O₆^2-) the structures are both more similar to the one I wrote above, and only have negative charges held on or delocalized over oxygen atoms?

Because it does. Sorry if that sounds facetious, but I'd imagine various studies have been carried out to show that the isomer that exists is the one that has an S-S bond. You can probably generate crystals or similar, maybe not NMR to show that the sulfurs are not equivalent; though this isomer may exist elsewhere I imagine it would be a bit unstable (bond strengths?). Even the dithionate and diothionite ions have a S-S bond...

[Big-Daddy]

Because it does. Sorry if that sounds facetious, but I'd imagine various studies have been carried out to show that the isomer that exists is the one that has an S-S bond.

Hmm, ok. Given this was an exam situation without access to energy-estimation by computation, do you think I was just expected to know this as the experimental and "true" thiosulphate structure, or is there some obvious-ish rationale to choose it over the Lewis structure in the OP?

though this
isomer may exist elsewhere I imagine it would be a bit unstable (bond strengths?).

This seems like a clever reason. Maybe, to phrase more accurately, delocalization of a negative formal charge over several O atoms adds to stability, even more than having S^- rather than O^- detracts from it?

Even the dithionate and diothionite ions have a S-S bond...

It's not the S-S bond that surprised me but really having the second negative formal charge on an S rather than an O, which is not repeated in any other SxOy species I know of (neither dithionate, nor dithionite nor any of the others I mentioned in OP). But all those species do also have delocalization over several O atoms, so maybe when it comes down to one or the other, the preference for delocalization wins (adds more to the stability)?

[sjb]

Hmm, ok. Given this was an exam situation without access to energy-estimation by computation, do you think I was just expected to know this as the experimental and "true" thiosulphate structure, or is there some obvious-ish rationale to choose it over the Lewis structure in the OP?

Depending on your level, I'd think you may just "have to know" that this is the structure, possibly through the name? I'd imagine that the linear form would be called something different.

It's not the S-S bond that surprised me but really having the second negative formal charge on an S rather than an O, which is not repeated in any other SxOy species I know of (neither dithionate, nor dithionite nor any of the others I mentioned in OP). But all those species do also have delocalization over several O atoms, so maybe when it comes down to one or the other, the preference for delocalization wins (adds more to the stability)?

Ahhh, I see. Of course, you can push an electron pair from the S- back through the central sulfur to end up at 2 x O-, and as this is pseudotetrahedral there is no steric or other argument to detract. So you can have , rather than , and these are related as mesomers. I'd guess the former is slightly better given electronegativity, but it's a tough call - maybe with hard and soft acids the second may be a better descriptor.

[Big-Daddy]

Depending on your level, I'd think you may just "have to know" that this is the structure, possibly through the name? I'd imagine that the linear form would be called something different.

Roughly end of undergraduate. (Or at least that is where I would like to be!)

Ahhh, I see. Of course, you can push an electron pair from the S- back through the central sulfur to end up at 2 x O-, and as this is pseudotetrahedral there is no steric or other argument to detract. So you can have , rather than , and these are related as mesomers. I'd guess the former is slightly better given electronegativity, but it's a tough call - maybe with hard and soft acids the second may be a better descriptor.

Of course this is possible but I thought pi bonds between Period 3 atoms are generally strongly discouraged by energetics? In which case we are still forced to choose between holding 2O^- rather than S^-, and delocalizing one formal negative charge over all the O atoms. Question on the side: how many delocalized electrons are there in this system (of "true" thiosulphate)? Is it 1 (the negative formal charge electron held) or 3 or 5, and how does one tell?

Also, by mesomer you must mean "resonance structure". These are the same, really, aren't they? One molecule can alternate between its mesomers, as benzene does. Incidentally, I have to clarify, that a mesomer is nothing to do with a meso-compound (where potential sources of optical activity cancel each other out in certain isomers due to symmetry making them equivalent?

[antimatter101]

The solution is simple.

While the first row elements have the ability to form strong pi bonds, the second, third, and fourth row elements do not. In fact their ability to form pi bonds decreases down the group. Thus, while the first row nonmetals easily form double and triple bonds, second row elements such as sulfur and phosphorous do not.

eg. The molecule S₂ and P₂ are non-existant due to the high energy of the pi bonds, while O₂ and N₂ are common.

Thus, if the sulfur atom in thiosulphate ion does not own an extra electron, it must form a double bond with p-orbital overlap - which is highly unstable. Thus, the negative charge is more heavily localised on the terminal sulfur atom to increase stability.

Happy now?

[antimatter101]

For example, silicon dioxide SiO₂ does not consist of individual molecules, but crystal lattices of Silicon-Oxygen single bonds.

[Big-Daddy]

The solution is simple.

While the first row elements have the ability to form strong pi bonds, the second, third, and fourth row elements do not. In fact their ability to form pi bonds decreases down the group. Thus, while the first row nonmetals easily form double and triple bonds, second row elements such as sulfur and phosphorous do not.

I assume you mean "ability to form strong pi bonds with themselves or each other" rather than general ability to form strong pi bonds, as S=O is a strong enough bond indeed.

Thus, if the sulfur atom in thiosulphate ion does not own an extra electron, it must form a double bond with p-orbital overlap - which is highly unstable. Thus, the negative charge is more heavily localised on the terminal sulfur atom to increase stability.

Happy now?

You've corroborated my point that sjb's resonance form is much less energetically likely than the true form. But you haven't answered why this true structure appears in nature rather than my structure proposed in the OP (which contains no S=S double bonds).