Researchers Discover Hydrogen Can Form Multicenter Bonds

[phantom power]

I blame quantum mechanics for bullshit like this.

Researchers at the University of California, Santa Barbara have shown that, under the right circumstances, hydrogen can form multicenter bonds, where one hydrogen atom simultaneously bonds to as many as four or six other atoms. Tested for hydrogen in metal oxides, the discovery could have a broad range of technological impact. The research is available today in the advance online publication of Nature Materials.

Professor Chris G. Van de Walle and Project Scientist Anderson Janotti, both of the Materials Department of the College of Engineering at UC Santa Barbara, have shown that multi-coordinated hydrogen is a likely explanation for electronic conductivity in metal oxides. Metal oxides are widely used in everything from sunscreen to sensors.

Hydrogen, the simplest of the elements (consisting of one proton and one electron) is typically expected to exhibit simple chemistry when forming molecules or solids. Hydrogen atoms almost always form a single bond to just one other atom, leading to a two-center bond with two electrons. Exceptions to the rule are rare; there are only a few cases when hydrogen bonds simultaneously to two other atoms, forming a three-center bond.

Hydrogen can replace an oxygen atom and form a multicenter bond with adjacent metal atoms. For example, in ZnO, hydrogen equally bonds to the four surrounding Zn atoms, becoming fourfold coordinated. These multicenter bonds are highly stable and explain previously puzzling variations in conductivity as a function of temperature and oxygen pressure. The results suggest that hydrogen can be used as a substitutional dopant in oxides, a concept that is counterintuitive and should be of wide interest to researchers.

http://www.sciencedaily.com/releases/2006/12/061204081945.htm

[mike_c]

1. is this really a "multicenter" bond or is it a two center covalent bond...

...with the second position oscillating between multiple nuclei? I mean, hydrogen only has a single electron to share. Or is this a matter of that electron entering some meta-orbital that is an emergent property of the molecule rather than an orbital of any individual nucleus? That's the only way I can envision a proton "sharing" an electron with multiple atoms simultaneously.

On edit: I'm not sure that last bit makes sense though. It certainly doesn't fit in with the physical chem I learned-- but I'm not a chemist so that's no surprise!

[phantom power]

5. Exactly. It hurts my brain. All I know is, quantum physics is to blame.

If you think you're confused, I'm just an algorithm guy.

[whistle]

2. What exactly are the implications of this, can hydrogen thus be...

...formed into numerous new molecules yet unknown on earth? Is that good or bad for the environment and the continued survival of life on earth?

[phantom power]

4. Good old fashioned molecules are infinitely more dangerous: CO2 & CH4

[bananas]

3. So maybe ice-9 is possible after all... nt

[Occulus]

7. At least we know it wasn't the Gaijin n/t

[Odin2005]

6. OK, my head hurts, now...

[NNadir]

8. Multiple co-ordinate hydrogen is not really that exotic.

Hydrogen bonds on which all life depends are multicentered in a sense. Ice is an array of complexed hydrogen with each hydrogen co-ordinated to two oxygens.



Many boron hydrides are bridged compounds.

What is different here is the co-ordination number for hydrogen.