5

2

This is probably an easy question, but I'm not able to figure it out.

Are the following the same:

  1. Field of fractions of the ring of Witt vectors over the algebraic closure of $\mathbb{F}_p$

  2. Algebraic closure of the field of p-adics (which is the field of fractions of the ring of Witt vectors over $\mathbb{F}_p$)

In other words, does the operation of taking the field of fractions of the ring of Witt vectors commute with the operation of taking the algebraic closure?

flag
I assume "Witt ring" means "ring of Witt vectors". Probably worth a mention, since Witt ring normally means something very different (unfortunately). – darij grinberg Jan 8 2012 at 19:00
Fixed it, thanks! – Vipul Naik Jan 8 2012 at 19:10

1 Answer

9

no: the Witt ring of $\bar{F_p}$ is a complete DVR and so its field of fractions will be a complete local field; but the algebraic closure of $Q_p$ is not complete.

However, take the maximal unramified extension of $Q_p$; this is a non-complete field. Its completion $F$ is the fraction field of the Witt ring of $\bar{F_p}$ and the Witt ring itself is the ring of integers in $F$.

(Serre's Local Fields contains all of this and much more!!)

link|flag
Indeed, one should think of the Witt construction as a machine for producing unramified extensions of ${\mathbb{Q}}_p$. Since ${\mathbb{Q}}_p$ has plenty of ramified extensions, the Witt construction will always give you something far from the alg. closure. – Lubin Jan 9 2012 at 20:31
@Lubin:Indeed! Thanks! I forgot to mention that the Witt ring of $\bar{F_p}$ plays an important role in the theory of Lubin-Tate formal groups!! – SGP Jan 14 2012 at 1:12

Your Answer

Get an OpenID
or

Not the answer you're looking for? Browse other questions tagged or ask your own question.