Post by Matt GarveyMore seriously, I don't think LiBr2 would work either, in terms of
valence electrons
Well, using a molecular model of what is (essentially, but certainly
not totally) an ionic compound, LiBr2 can be seen as a semi-stable
resonance between:
Br:Li Br
and
Br Li:Br
If this seems dodgy, that's because a resonance model can't really
express half bonds well (as opposed to one-and-a-half, etc.).
A molecular orbital model is more satisfactory, but also much, much
more complicated.
Suffice to say, bromide's large ionic radius makes it less observant
of the octet rule. In other words, Br (NOT Br2, Br) is not insanely
reactive, as is something like F. F will almost immediately react
with anything to be reduced to F-, but Br can be said to be "more
tolerant."