gaffe of the hour

[oonh]

Dodecahedrane is a highly symmetric icosahedral molecule.
huh? it's got 'dodecahedra' in it and yet you call it icosahedral? flark I say. flark.

Comments

[carus_erus]

Nope, makes perfect sense. If I recall correctly, an isocohedron fits inside a dodecahedron perfectly, resting each point on the center of each of the dodecahedron's inner 20 triangles.

12 pentagons * 5 points / 3 points shared between different sides = 20 points on the isocahedron which is why the chemical composition is C20H20. The hydrogen on the outside forms the dodecahedron (the center of the triangles, not the edges), the carbon on the inside is isocohedral.

Take some group theory and some basic organic chemistry... Then look back at the diagram and the chemical struture (C20H20) and it will make perfect sense...

[oonh]

There's a difference between saying "has an icosahedral symmetry point group" and "is icosahedral". Even with hydrogens attached it's dodecahedral.

[carus_erus]

You notice in the diagram the hydrogen molecules are removed to talk about the underlying structure of the carbon. You have to realize that this page is coming very much from an organic chemistry perspective. The hydrogen, while part of the overall structure of the chemical, doesn't actually add on to the basic structure of the carbon.

The isochahedron and dodecahedron are not really different shapes from a symmetry perspective. Remember also that the molecule doesn't really have "faces" either, and that there's no particles joining the points between the carbon molecules (except EM force). Both the carbon and hydrogen are just 20 points floating in space.

There might also be chemistry terminology that means something significantly different than the mathematical terminology.

Dictionary says:

[grayden]

"Icosahedron: A geometrical shape with 20 triangular faces and 12 corners."

"Dodecahedron: A geometrical shape with 12 pentagonal faces and 20 corners."

Looks like they just made a simple slip-up. :-)

There's a picture of the molecule here.

"hedral" refers to faces, not vertices. If they has said "icosa[pointed]" where you substitute the actual suffix which is unknown to me, they would have been correct.

Re: Dictionary says:

[carus_erus]

Except that the carbon molecules DO have an isocahedral structure if looked at alone, while the hydrogen molecules have a dodecahedral structure if looked at alone.

that carbon structre is inside the hydrogen structure, much like an isocohedron can be laid inside the dodecahedral structure (or vice versa).

The two shapes are symmetrical, at least from a group theory point of view.

Re: Dictionary says:

[geekosaur]

From a chemistry perspective, the carbons are more "interesting" than the hydrogens, aren't they? In particular the carbon structure more or less controls its chemical behavior, so it makes sense that chemists would concentrate on its carbon structure instead of its shape when decked out with hydrogen atoms.

Re: Dictionary says:

[carus_erus]

Actaully I made comments similar to this in another comment in this thread.

I just realized that "dodeca" is (I believe) a greek prefix for "20". Probably having a lot to do with the 20 carbon atoms in the molecule.

And a little googling later.. "hedrane" is used elsewhere in achemical context.

So dodecahedrane has nothing to do with shape whatsoever, any more than my mortgage has a dodeca-annuation..

[kilroi]

like the first comment, yeah, icosahedral is a reference to a point group. take a look-see here:

http://mathworld.wolfram.com/IcosahedralGroup.html

[mmcirvin]

I'm not a chemist, just a lapsed particle physics student, but in general, in chemistry and physics the symmetry group of something symmetrical is a really important thing to know and crucial to talking about it. I can easily imagine calling anything with icosahedral symmetry "an icosahedral molecule" regardless of how it looks, e.g. I've talked to people modeling the electronic structure of buckyballs, and I've seen buckyballs called "icosahedral" even though they are soccer-ball-shaped. It's just a more complicated polyhedron with the same symmetry.

Also, just because the page uses a stick model of chemical bonds doesn't mean that's the only way to visualize the molecule. If you showed just the carbons and used a squished-ball model to represent them, they'd look like faces instead of vertices and you'd get a lumpy icosahedron. It looks like a dodecahedron one way and an icosahedron the other, because they're dual shapes; no big deal.