I am working on a question:

Find all the ideals of norm $10$ in $\mathcal{O}_K$ where $K=\mathbb{Q}(\sqrt{35})$.

I am given the hint:

Observe that $(2)=(2,\alpha)^2, (5)=(5,\alpha)^2, (\alpha)=(2,\alpha)(5,\alpha)$ where $\alpha=5+\sqrt{35}$.

So I can tell that $(2,\alpha)$ is of norm 2, $(5,\alpha)$ is of norm 5, $(\alpha)$ is of norm 10. I think, to find all ideals of norm 10 I first should find all ideals of norm 2 and 5 since any ideal of norm 10 factorizes into prime ideals of norms 2 and 5.

But in general, how do I find all ideals of norm 2, 5?

  • $\begingroup$ Notice that all ideals of prime norm is a prime ideal. Thus it amounts to the same to finding the prime ideal decompositions of $2\mathbb Z$ and $5\mathbb Z$ in the given extension. $\endgroup$ – awllower Mar 4 '13 at 12:46
  • $\begingroup$ Why is finding a prime ideal of norm $2$ the same as factorizing $(2)$? $\endgroup$ – Spook Mar 4 '13 at 12:48
  • $\begingroup$ If the prime ideal is of norm $2$, then it divides $2$. But you already know that $2$ is totally ramified, so that means all prime ideals dividing $2$ must be of norm $2$.In shorter terms, look at the unique factorisation of ideals, and notice that having norm $p$ means dividing $p$. $\endgroup$ – awllower Mar 4 '13 at 12:51
  • $\begingroup$ In which case, the only ideal of norm $2$ is $(2,\alpha)$? $\endgroup$ – Spook Mar 4 '13 at 12:56
  • $\begingroup$ An ideal f norm $2$ is prime, and it divides $2$, so that it must be $(2,\alpha)$. $\endgroup$ – awllower Mar 4 '13 at 12:57

You can also approach this the following way:

If $I$ is an ideal of norm $10$, the additive group of $\mathcal O_K/I$ is an abelian group with $10$ elements, hence isomorphic to $\mathbb Z/10\mathbb Z$. We obtain $\mathcal O_K/I \cong \mathbb Z/10\mathbb Z$ as rings, as the ring structure of a cyclic group is uniquely determined by $1 \cdot 1 = 1$.

So ideals of norm $10$ are in bijective correspondence to surjective ring maps $\mathcal O_K \to \mathbb Z/10\mathbb Z$. By the homomorphism theorem such maps correspond to maps $\mathbb Z[X] \to \mathbb Z/10\mathbb Z$, which map $X^2-35$ to zero. By the universal property of the polynomial ring, such maps correspond to elements $a \in \mathbb Z/10\mathbb Z$ with $a^2-35=a^2-5=0$. We get $a=5$ as unique solution. Going reverse in our arguments, this gives us the map $\mathbb Z[\sqrt{35}] \to \mathbb Z/10\mathbb Z, a+b\sqrt{35} \mapsto a+5b$ with kernel $(5+\sqrt{35})$. This is the unique ideal of norm $10$.

After "We get $a=5$ as the unique solution", you could stop and be like "Using abstract arguments, I showed there is a unique such ideal. The hint gives me one such ideal. So I am done".


The ideal $(p)$ is the product of all of the prime ideals of norm a power of $p$ (with multiplicity equal to the ramification index).

  • 1
    $\begingroup$ Can you give me a proof or this? I know that the ideal $(p)$ is the product of prime ideal with norm a power of $p$, but how i prove that they are all? $\endgroup$ – Yecabel May 20 '18 at 14:00
  • $\begingroup$ @Yecabel If $\mathfrak{p}$ is a nonzero prime ideal, then $\mathcal{O}_K/\mathfrak{p}$ is a finite field. Suppose that the order of this field is $p^r$. Then $p+\mathfrak{p}=0+\mathfrak{p}=\mathfrak{p}\in \mathcal{O}_K/\mathfrak{p}$ and $p\in \mathfrak{p}$ and $\langle p\rangle\subseteq \mathfrak{p}$. Write $\langle p\rangle$ as the prime ideals decomposition $\mathfrak{p}_1^{r_1}\mathfrak{p}_2^{r_2}\cdots \mathfrak{p}_s^{r_s}$. $\endgroup$ – bfhaha May 31 '18 at 9:30
  • $\begingroup$ Then $\mathfrak{p}_1^{r_1}\mathfrak{p}_2^{r_2}\cdots \mathfrak{p}_s^{r_s}=\langle p\rangle \subseteq \mathfrak{p}$ and $\mathfrak{p}_i\subseteq \mathfrak{p}$ for some $i\in \{1, 2, ..., s\}$ because $\mathfrak{p}$ is a prime ideal. In the Dedekind domain $\mathcal{O}_K$, $\mathfrak{p}_i$ is maximal and the prime ideal $\mathfrak{p}\neq \mathcal{O}_K$ (by the definition of a prime ideal). It follows that $\mathfrak{p}=\mathfrak{p}_i$. That is, every prime ideal $\mathfrak{p}$ must appear in the prime ideals decomposition of some $\langle p\rangle$. $\endgroup$ – bfhaha May 31 '18 at 9:30

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.