About $\mathbb Z_{p}[\sqrt{k}]$, when is it a field?

957 Views Asked by At

I give up. I'm new in the fields world, and

I'm trying to give a sufficient and necessary condition for $\mathbb{Z}_{p}[\sqrt{k}]=\{a+b\sqrt{k}:a,b\in \mathbb{Z}_{p}\}$ to be a field ($p$ is a prime and $k$ is a positive integer).

I claim that the condition is that $p$ doesn't divide $k$, but I don't know if this is true, I've tried to prove this but I've got stuck in the "sufficient" part.

It would be great if some of you can help me, thanks!.

1

There are 1 best solutions below

6
On BEST ANSWER

If $F$ is a field, then $F[\sqrt{a}]$ is always a field.

Case I: $a$ is a square in $F$, $b^2 = a$. Then $F[\sqrt{a}] = F[b] = F$.

Case II: $a$ is not a square in $F$. Then the polynomial $X^2 - a$ is irreducible, so $F[\sqrt{a}] \cong F[X]/(X^2-a)$ is a field.

More generally, if $R$ is a ring with no zero divisors that contains $F$, and it is finite-dimensional as a vector space over $F$, then it is a field.

The issue raised in the comments is that we might define $F[\sqrt{a}]$ to be $F[X]/(X^2-a)$, rather than the smallest subring of $\overline{F}$ containing $F$ and $\sqrt{a}$ (this appears to be the definition you intend). But when $a$ is a nonzero square, this is isomorphic to $F\times F$, which is not a field, and when $a=0$, it is isomorphic to $F[\epsilon]/\epsilon^2$ (there is a very minor exception when $\operatorname{char}K = 2$, in which case nonzero squares also yield $F[\epsilon]/\epsilon^2$).