How do you prove that algebraically closed fields of characteristic $p$ exist?

I have also read:

For a finite field of prime power order $q$, the algebraic closure is a countably infinite field that contains a copy of the field of order $q^n$ for each positive integer $n$ (and is in fact the union of these copies).

Why would the algebraic closure have to be characteristic $p$?

  • 3
    $\begingroup$ If adding $1$ added to itself a prime number of times yields zero in the finite field $F$, then it also does so in any field $K$ containing the finite field $F$! $\endgroup$ – anon Jun 24 '15 at 2:36
  • $\begingroup$ Oh, I answered the title, not the body. Those are two entirely different things. Can you please find a title that is closer to the question? $\endgroup$ – Thomas Andrews Jun 24 '15 at 2:48
  • $\begingroup$ If you are interested: Keith Conrad carried out Artin's classical construction of an algebraic closure (of any field) and made it available online. $\endgroup$ – Stefan Mesken Jun 24 '15 at 2:49
  • $\begingroup$ The characteristic of a subfield is the same as the characteristic of the field of which it is a subfield. $\endgroup$ – Michael Hardy Jun 24 '15 at 2:52
  • $\begingroup$ @ThomasAndrews, I meant for the questions to be different. $\endgroup$ – user240033 Jun 24 '15 at 3:01

This only answers the main question.

Just take a sequence of inclusions:

$$\mathbb F_p\to\mathbb F_{p^2}\to\mathbb F_{p^6}\to\cdots\to \mathbb F_{p^{n!}}\to\cdots$$

Then the direct limit (essentially the union) is algebraically closed and a field.

That's because any polynomial in elements of this field has coefficients contained in one $\mathbb F_{p^{k!}}$. Thus, it splits in $\mathbb F_{p^{d\cdot k!}}$ for some $d$, and then note that $d\cdot k!\mid(dk)!$. So the polynomial splits in $\mathbb F_{p^{(dk)!}}$.


Let $F$ be the field of prime power order. Let $p$ be that prime. The field $F$ is a subfield of it algebraic closure, which let us call $G$. your question seems to be: why must $G$ have characteristic $p$? We could as why $F$ must have characteristic $p$. And then, supposing $F$ does have characteristic $p$, why must $G$ have characteristic $p$?

Let's answer the second question first: Since $F\subseteq G$, the "$1$" in $F$ is the same as the "$1$" in $G$. It satisfies $\underbrace{1+\cdots+1}_p= 0$. Since this is the "$1$" in $G$, the field $G$ has characteristic $p$.

Now the first question: If $F$ has $p^n$ elements, why must $F$ have characteristic $p$? If you know the characteristic is some prime number $r$, then you have $\underbrace{1+\cdots+1}_r=0$, and $\{0, 1, 1+1, 1+1+1, \ldots, \underbrace{1+\cdots+1}_{r-1}\,\}$ is a subgroup of the additive group. The order of the subgroup divides the order of the whole group, which is $p^n$. If $p$ and $r$ are prime and $r$ divides $p^n$, then $r=p$.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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