Talk:Crystal structure of boron-rich metal borides

Electron deficiency

There seems to be an implication that electron-deficient compounds are unstable. This is misleading. Some reference to the various structures of elemental boron, particularly those containing icosahedral fragments, would be helpful. An electron deficient compound is one in which there are not enough electrons to form localized 2-centre, 2-electron bonds. The formation of delocalized multi-centre 2-electron bonds can result in quite stable compounds such as in B₁₂ H₁₂^2-. Petergans (talk) 08:43, 8 September 2010 (UTC)

Not sure I understand - despite the electron deficiency of most borides (not only boron rich ones), they are all rather hard and thermally stable, i.e. they tend to take electrons from metal, and restructure if such electrons are not available (pure boron). By the way, caesium dodecaborate is not that stable. Materialscientist (talk) 09:16, 8 September 2010 (UTC)

What I'm saying is that electron deficiency is not that significant. Simple compounds of boron are Lewis acids (electron pair acceptors) by virtue of the fact that a boron atom has only 3 electrons of its own. The polyhedra in the borides, considered by themselves, are still electron-deficient structures. I just thought that the way electron deficiency was treated was a little confusing. BTW, DYK that, for many years, Norman Greewood was my boss and Alan Earnshaw my friend and colleague? NNG, together with John Kennedy, did a lot of work on boranes, so I'm quite familiar with them. Petergans (talk)

"B₁₂ cuboctahedron and B₁₂ icosahedron, lack two valence electrons per polyhedron to complete the polyhedron-based framework structure. " This is the sentence that aroused my suspicion. To quote from G&E in relation to α rhombohedral boron: " In terms of MO theory, the 36 valence electrons of each B₁₂ unit are distributed as follows: 26 electrons just fill the the 13 available bonding MOs within the icosahedron and 6 electrons share with 6 other electrons from 6 neighbouring icosahedra in adjacent planes to for 6 rhombohedrally directed normal 2c-2e bonds. This leaves 4 electrons which is just the number required for contribution to the 6 equatorial 3c-2e bonds." There is no lack of electron there.

The simplified math is like this: 26 electrons per icosahedron + 12 electrons per its every corner = 38 = 36 + 2, thus 2 are lacking. Materialscientist (talk) 12:15, 12 September 2010 (UTC)

We agree on the closo- 2n+2 skeletal e^- (Wade's rules Wade was a Ph.D. student with NNG) The notion that each vertex should also carry a non-skeletal e^- is a simplification I had not heard of before. Petergans (talk) 15:18, 12 September 2010 (UTC)

Sorry for belated reply. Electron deficiency is just a model, which was established by calculation of solid borides and which fits well to various experimental results, same as the notion that an icosahedron needs 26 electrons. By the way, B₁₂ H₁₂^2-. structure you quoted does confirm that 2 electrons are required per icosahedron, if we assume that every vertex is either bonded, or passivated by hydrogen. Materialscientist (talk) 07:29, 17 September 2010 (UTC)

How does this square with the RE borides? As it stands, there does not appear to be any need for the RE atoms to contribute electrons, unless the number of bridging borons is reduced. The other thing which struck me as odd was the statement that "If both metal elements are trivalent ions then 3.99 electrons can be transferred to the boron framework." If two RE atoms become trivalent, won't they lose 6 electrons? Maybe I'm finding these questions because I'm looking at the article from a different perspective, but there are certainly things that I find confusing. Petergans (talk) 11:20, 12 September 2010 (UTC)

The actual composition is Y_0.62Al_0.71B₁₄, not YAlB₁₄, and every metal ion is 3+, thus the math here is (0.62+0.71)*3=3.99 Materialscientist (talk) 12:15, 12 September 2010 (UTC)

In that case the formula should not be given as YAlB₁₄. As the composition is somewhat variable Y_xAl_yB₁₄ (or Y_xAl_1.33-xB₁₄ ) would be preferable. This also applies to other compounds of variable composition. Petergans (talk) 15:18, 12 September 2010 (UTC)

Because of conventions, people keep calling materials as YAlB₁₄, B₄C, etc., even when most of them do believe those compounds are intrinsically non-stoichiometric. Materialscientist (talk) 07:29, 17 September 2010 (UTC)

It was obvious that the incorrect formula was used for historical reasons. I can accept that, but it must be made clear at the beginning of the section that this is so. Otherwise, the statement "REAlB₁₄ and REB₂₅ compounds have the MgAlB₁₄ structure " does not make sense because the electron count is apparently different in the two cases. I urge you, as the expert, to clarify the text. The structural information is not sufficient by itself. The chemical formula is also needed in order to understand the electron count. Petergans (talk) 16:02, 18 September 2010 (UTC)