Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm was browsing this question, where it is proven the quotient field of $\mathbb{Q}[X,Y]/(X^2+Y^2-1)$ is isomorphic to the rational function field $\mathbb{Q}(t)$ under the isomorphism $$ (x,y) \mapsto \left( \frac{2t}{t^2+1}, \frac{t^2-1}{t^2+1} \right). $$ How can this isomorphism be used to show $\mathbb{Q}[X,Y]/(X^2+Y^2-1)$ is normal in its quotient field?

I identified $\mathbb{Q}[X,Y]/(X^2+Y^2-1)$ with $\mathbb{Q}\left[\frac{2t}{t^2+1}, \frac{t^2-1}{t^2+1} \right]$, and was trying to show this is integrally closed in $\mathbb{Q}(t)$.

If $f/g\in\mathbb{Q}(t)$, with $f,g$ coprime, is integral over $\mathbb{Q}[2t/(t^2+1),(t^2-1)/(t^2+1)]$, then $$ (f/g)^n+h_{n-1}(f/g)^{n-1}+\cdots+h_1(f/g)+h_0=0,\qquad h_i\in \mathbb{Q}[2t/(t^2+1),(t^2-1)/(t^2+1)]. $$ Multiplying through by $g^n$, I found the relation $$ f^n=g(-h_{n-1}f^{n-1}-\cdots-h_0g^{n-1}) $$ so any prime dividing $g$ divides $f$, so $g$ is constant, and in fact $f/g\in\mathbb{Q}[t]$. I was hoping to see that is is in fact in $\mathbb{Q}[2t/(t^2+1),(t^2-1)/(t^2+1)]$, but I think I've made a mistake somewhere.

share|cite|improve this question
up vote 3 down vote accepted

Let's prove that $R=\mathbb{Q}\left[\frac{2t}{t^2+1},\frac{t^2-1}{t^2+1}\right]$ is integrally closed.

We have $\mathbb{Q}\left[\frac{2t}{t^2+1},\frac{t^2-1}{t^2+1}\right]=\mathbb{Q}\left[\frac{1}{t^2+1},\frac{t}{t^2+1}\right]$. It's obvious that $\mathbb Q(t)$ is its field of fractions. Let $f(t)/g(t)\in\mathbb Q(t)$ integral over $R$ with $\gcd(f(t),g(t))=1$. For any element $h\in R$ there is $r\in\mathbb N$ such that $(t^2+1)^rh\in\mathbb Q[t]$. Consider an integral dependence relation $$(f/g)^n+h_{n-1}(f/g)^{n-1}+\cdots+h_1(f/g)+h_0=0$$ with $h_i\in R$ and multiply this by $(t^2+1)^{nr}$ with $r$ large enough in order to get an integral dependence relation for $(t^2+1)^rf(t)/g(t)$ over $\mathbb Q[t]$. Since $\mathbb Q[t]$ is integrally closed we get $(t^2+1)^rf(t)/g(t)\in\mathbb Q[t]$, and therefore $g(t)\mid(t^2+1)^r$. Since $t^2+1$ is irreducible over $\mathbb Q$ we have $g(t)=(t^2+1)^s$. Now it's clear that $f(t)/g(t)\in R$.

share|cite|improve this answer
It might be a silly question. But how do you see that $\frac{1}{t^2+1} \in \mathbb{Q}[\frac{2t}{t^2+1},\frac{t^2-1}{t^2+1}]?$ – Youngsu Apr 9 '14 at 17:13
@Youngsu: $\frac{t^2-1}{t^2+1} = 1 - 2\left(\frac{1}{t^2+1}\right)$ – zcn Apr 9 '14 at 19:16
@zcn: Thank you. – Youngsu Apr 9 '14 at 20:41

Due to both denominator's being the same that ring is isomorphic to the ring $\mathbf{Q}[2t, t^2-1]$. That is, the subring of $\mathbf{Q}[t]$ generated by $2t$ and $t^2-1$. But in the presence of $2t$ we do not need $t^2-1$. So this subring can be generated by just $2t$. So this is isomorphic to the PID $\mathbf{Q}[2t]$, hence normal.

share|cite|improve this answer
$\mathbb Q[2t]=\mathbb Q[t]$ and $\mathbb{Q}[2t/(t^2+1),(t^2-1)/(t^2+1)]$ can't be isomorphic to $\mathbb Q[t]$ since it's not a UFD. – user26857 Apr 9 '14 at 10:13

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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