What happens to Clifford algebra structure and periodicity when the field is weird?

50 Views Asked by At

The classification of the structure of real and complex Clifford algebras is well-known. Most sources concentrate only upon that case. Another resource. Also, I'm only interested in nondegenerate forms, at this point, and I'd prefer to skip characteristic $2$ if it complicates the answer.

The interesting thing is the periodicity exhibited in their structures as the dimension and signature changes.

But the theory of bilinear forms is well-developed over all fields, so I am very curious what happens outside of the idyllic cases of $\mathbb R$ and $\mathbb C$. I believe I've seen somewhere how to prove how Clifford algebras over nondegenerate spaces are all semisimple rings (at least, outside of characteristic $2$.)

Correct me if I'm wrong, but if I remember right, the only thing you need to reduce quadratic forms to a "signature" is to either use an ordered which contains the square roots of all its positive members, or to use a field that's quadratically closed altogether.

Is the structure/periodicity of Clifford algebras of algebras over fields described in the last paragraph known?

Alternatively, we could stop caring about general bilinear forms and just ask for a classification of those arising from a matrix with $1$'s and $-1$'s on the diagonal.

Is the structure/periodicity of Clifford algebras for these particular forms over fields known?

I haven't had a lot of luck with the literature. I could believe it if everything had to exhibit periodicity $1, 2, 4,$ or $8$, because that seems to be the way geometry likes to roll.