[gavindor]

Why is Mg2Cl2 described as a linear molecule on wikipedia, if it is an ionic compound?

I understand  Mg2Cl2 to have a mercurous ion and so be ionic.

But why does this wikipedia page
https://en.wikipedia.org/wiki/Mercury(I)_chloride
say
"Hg2Cl2 is a linear molecule"

I notice also it doesn't mention about the cation and anion either.

[Corribus]

I suppose you mean Hg, not Mg.

The rough answer is that as you go down the periodic table, the distinction between ionic bonds and covalent bonds blurs particularly when heavy transition metals are involved. There are a some different ways you may explain this. One is in terms of electronegativity values. So, for example, the difference in EN between Hg and Cl is around 1.16, less than the difference between silicon and oxygen (EN difference 1.54), which forms SiO2 glass - most people would consider covalent. You may think of mercury as positive charged and chloride as negatively charged, but they also retain some covalent character. This is evident because in Hg2Cl2, the two Hg atoms share a covalent bond.  Certainly the Hg2Cl2 lattice is held together by strong electrostatic forces just like any other ionic compound, but compared to sodium chloride (EN difference >2.15) they just aren't as strong.

The fact of the matter is ionic and covalent bonds really shouldn't be viewed a completely different things. Rather, it is better visualize them along a continuous spectrum of potential interactions between adjacent nuclei, differentiated mostly by the degree to which electrons are shared. The ionic case is simply the extreme case where electrons are almost completely localized around one of the two nuclei (in a diatomic pair). There are a lot of things that determine the equality (or lack thereof) of electron sharing, and therefore the covalent/ionic character of a bond. EN is a good guide, although it's more of an empirical concept. To really dig down, you need a good understanding of orbital structure, quantum mechanical effects, and - particularly at the bottom-most levels of the perdiodic tables - even relativistic effects. Transition metals especially tends to form more covalent-type structures, a lot of it because they have access to d- and f- orbital sets that open up a lot of potential avenues for complex bonding modes. Also note that transition metals and other elements deep in the periodic table are simply bigger, which has its own effects on the strength of electrostatic interactions between the positively charged nuclei and valence electrons.

[gavindor]

I suppose you mean Hg, not Mg.
....The fact of the matter is ionic and covalent bonds really shouldn't be viewed a completely different things. Rather, it is better visualize them along a continuous spectrum of potential interactions between adjacent nuclei, differentiated mostly by the degree to which electrons are shared. The ionic case is simply the extreme case where electrons are almost completely localized around one of the two nuclei (in a diatomic pair).

Thanks, yes I mean Hg not Mg, I can't see an edit button for earlier posts in a thread.
 
And I understand that covalent and ionic is a spectrum, with purely covalent on one end (with something like Cl2). And going to non polar covalent, to polar covalent, to predominently ionic, and with nothing being 100% ionic / purely ionic.

When I say ionic I mean the bond between the Mercurous cation, and each of the Chlorine anions, being predomimently ionic.

Maybe it could be modelled covalently without writing ionic charges and treating it as a cation and two anions?

The way I see things is.. If we were to take NaCl, and boil it very hot, so it split into ionic monomers and ionic dimers..  So  clusters some are  just an NaCl pair, a molecule.    Some are  NaClNaCl(ionic dimer) Na2Cl2. But let's say we get it to all just discrete NaCl pairs.   Then if we let it cool so it forms crystals, we'd then say it's not molecular,  presumably because the bond between the Na and Cl, is ionic and the bonds between one NaCl and another, is ionic. And so it's not seen to be distinguishable entities.. discrete molecules.  In contrast, if we take O2 gas and freeze it, then we have a molecular solid.   There's VDW forces between the O2 molecules (and within them). But the O2 molecules while they have covalent bonds within the molecule.  Between the molecules they don't. So the O2 molecules are all considered very much separate.

Now, if I look at this picture on the wikipedia page for Hg₂Cl₂



What I am seeing is a structure.                 Cl---Hg₂---Cl

(and that has a covalent bond between the Hg atoms, and an ionic bond between the Hg2 and each Cl)

So that three ion entity is ionic - ionic meaning composed of ions. And Hg2 like any polyatomic ion is internally covalent.

And those three ion entities(as in, those entities each consisting of three ions), are not connected to each other with any covalent or ionic bond.

Just like with the O2 molecules.

So those "three ion entities", those ionic entities (ionic as in composed of ions), are technically molecules. 'cos they are all separated/distinct.

And I guess like in the O2 case, what is between one molecule and another, is just VDW forces.

So it seems to me what we have is a crystal that is molecular, where each molecule is ionic(composed of ions).

(and one of the ions has a covalent bond - which isn't unusual).

Is that right?

Thanks

[Borek]

I think the wikipedia image is misleading, as it draws ionic bonds as if they were directional and connected Cl^- with a particular cation.

[gavindor]

Well an ionic bond can be drawn with either a ball and stick diagram, or a spherical ions packed together diagram.

It seems to me that surely (as wikipedia shows), each Cl-  is meant to only be connected to particular cation.

Bear in mind that the wikipedia article says "linear molecule" (as mentioned in post #1).

So it's consistent.

Also, Look at this paper..

I found this paper 10.1143/JPSJ.49.554  that also mentions linear molecules. So I think maybe that supports that idea too, consistent with wikipedia.



I don't know if one would say it's an ionic compound.  (And we'd agree it's certainly not a covalent one)

It is Molecular.

Each molecule is ionic. (ionic as in composed of ions, and one of them being polyatomic).

If you try and say that it's arbitrary what the Cl connects to then I think you're also saying that it's all close packed together, but I don't think it is 'cos that loses the molecular nature that the diagram shows.

What it says in that paper "linear molecules of Hg2Cl2" totally confirms what wikipedia says and shows.

[Borek]

Well an ionic bond can be drawn with either a ball and stick diagram, or a spherical ions packed together diagram.

When it comes to pure bonds it is either ionic, or directional. Just because it can be drawn with sticks doesn't mean anything. I wouldn't call "linear molecule" ionic.

But as Corribus told you earlier - distinction between ionic and covalent gets fuzzy for transition metals, way too many electrons and orbitals looking for an energy minimum for things to be simple.

[gavindor]

I wouldn't call "linear molecule" ionic.
.

Are you suggesting that the geometry of the molecule affects whether or not it is ionic?

Or do you think the geometry of the molecule is irrelevant and ionic means not molecular?

When we say Ionic, do you agree that means composed of ions, and doesn't mean that a thing actually is an ion?


https://en.wikipedia.org/wiki/Mercury(I)_bromide



^^ that's from wikipedia.

That formula shows  Hg₂²⁺  Isn't that an ion?

Or the Cl^-  Isn't that an ion?

Why are they represented as ions in the formula if they are not ions?

[Borek]

Are you suggesting that the geometry of the molecule affects whether or not it is ionic?

Or do you think the geometry of the molecule is irrelevant and ionic means not molecular?

Geometry is secondary.

Ionic bonds are not directional, if you state something is ionic and has a well defined shape at the same time you are contradicting yourself.

When we say Ionic, do you agree that means composed of ions, and doesn't mean that a thing actually is an ion?

Composed of ions. But the arrangement of these ions is not set in stone.

As it was mentioned earlier in one of your threads, even in the crystal situation doesn't have to be obvious, as some compounds can crystallize in more than one way, creating crystals with different arrangement of ions.

To be honest with at the moment I an not sure if you are really confused or just trolling, I feel like we are going in circles.

[gavindor]

I am confused but i'll put my confusion related to this question on hold for now as i'm puzzling over another thing that might give me some insight into this thing. Thanks

[gavindor]

ok thanks.

My current understanding, and from the fact that it's a linear molecule.. Is that in the molecule  Cl---Hg2---Cl
The two Hg2---Cl bonds, are  polar covalent

A strange thing is that while the top of the page says Hg2Cl2, it has this image



I guess that image is Hg2Cl5?

Also, back to Hg2Cl2, it has this picture which is fine, given the understanding of it as a linear molecule, polar covalent bonds there between Hg2 and Cl.  And not ions / not an ionic compound.

Let's call that diagram "A" (The covalent one)


Though it also has this diagram , let's call that diagram "B" (The ionic one)



In Post #3 you suggested that  diagram "A" was misleading. Though I suppose it could be it's fine and it's Diagram B that's misleading?

[Corribus]

What makes you think the two diagrams depict something different?

Also, just because molecules form a regular lattice does not mean they are "ionic". As example, sucrose forms a crystalline lattice.

[gavindor]

What makes you think the two diagrams depict something different?

Also, just because molecules form a regular lattice does not mean they are "ionic". As example, sucrose forms a crystalline lattice.

Well I agree that to say lattice / crystal lattice,  doesn't imply molecular or ionic.. I know there are molecular crystal lattices, and ionic crystal lattices.

And I can see the view that either of those diagrams could represent either a molecular or an ionic crystal  lattice.

Infact when I was asking about it I was wondering whether shall I ask if those diagrams are consistent or if they are inconsistent.  But yeah them being consistent makes sense. So a space-filling model can be used for a molecular lattice.

I've seen ionic represented with space-filling or "ball and stick".  I'd not seen molecular crystal represented with "space-filling" though I hadn't seen that many molecular crystals represented. But it makes sense, what you say.

Thanks

[gavindor]

... the two diagrams ..

I would edit this into my post #11 but I can't see an edit/modify option. (maybe the modify option disappears after x amount of time)

Have you ever seen any examples(or any other examples), of a space-filling model used for a molecular crystal lattice?

I've seen both types of diagram (Ball and stick, and "space-filling"), for ionic crystals. But (aside from that Hg₂Cl₂ wikipedia page),  Whenever i've seen  molecular crystal lattices modelled, it has always been with ball and stick, never with space-filling.

[Borek]

Whenever i've seen  molecular crystal lattices modelled, it has always been with ball and stick, never with space-filling.

That's because they serve a different purpose, google for "molecular crystal with space filling models" though.

I feel like you are mistaking map with the territory though. All models are just that - models, limited in their mimicry of the real world. They do help in learning, but they are not designed to be exact, yet you try to treat them as such.