Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

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

The book I am studying uses the notation $w_k$ to denote the number of roots of unity contained in $K$, an imaginary quadratic number field.

Supposedly, this is the same as the size of the unit group in $O$, the ring of integers in $K$. I am well familiar with the standard arguments for finding the units in $O$.

Im not really sure exactly what the phrase "the number of roots of unity in $K$" means precisely. Does anyone know?

Also, I was thinking that it is probibly the case that the only elements of $K$ with finite multiplicative order are the elements of the unit group. Is this correct? If so, then is this the correct line of reasoning to show that $w_k=|U(O)|$?

share|cite|improve this question
up vote 5 down vote accepted

A root of unity is a complex number $a$ such that $a^n = 1$ for some positive integer $n$. The "number of roots of unity" is the cardinality of the set of roots of unity.

For example, if $K=\mathbb{Q}(i)$, then the roots of unity in $K$ are $1$, $-1$, $i$, and $-i$, so the number of roots of unity contained in $K$ is $4$.

Yes: a complex number is a root of unity if and only if it has finite multiplicative order in $\mathbb{C}^*$. In fact, the group of roots of unity is the torsion subgroup of $\mathbb{C}^*$.

To prove that in the imaginary quadratic case the unit group of $\mathcal{O}_K$ coincides with the group of roots of unity in $K$ you could use Dirichlet's Unit Theorem (as Timothy Wagner suggests), but you don't have to. You can simply show that every unit in $\mathcal{O}_K$ must in fact be a root of unity. This is not hard, since you (probably) know exactly what $\mathcal{O}_K$ for $K=\mathbb{Q}(\sqrt{d})$ ($d$ square free, $d\neq 1$), and you can use the norm map. For example, to show it in the case of $K=\mathbb{Q}(i)$, note that an element in $\mathcal{O}_K = \mathbb{Z}[i]$ is a unit if and only if $N(a+bi) = a^2+b^2 = 1$ (since $N((a+bi)(c+di)) = N(a+bi)N(c+di)$). But since $a$ and $b$ are integers, that requires $a=\pm 1$ and $b=0$ or $a=0$ and $b=\pm 1$, giving you roots of unity.

(In fact, imaginary quadratic fields are the only number fields in which the group of units in the number field is finite and torsion; this follows from the aforementioned Dirichlet Unit Theorem)

share|cite|improve this answer
First off, thanks. This helps a lot. Is it easy to show that a complex number is a root of unity iff it has finite multiplicative order? Clearly, one direction is trivial. – Jason Smith Dec 7 '10 at 23:48
@Jason: both directions look trivial to me. Isn't this true by definition? – Qiaochu Yuan Dec 8 '10 at 0:03

You need Dirichlet's unit theorem.

share|cite|improve this answer
+1, You are fast – Bryan Yocks Dec 7 '10 at 23:08
You can use it, but to show that a unit in the ring of integers of an imaginary quadratic field must be a root of unity you don't need to use it. You can get it out of the norm map. – Arturo Magidin Dec 7 '10 at 23:09
@Arturo: Thanks, that is better. Note to self: If you hear hoofbeats think horses not zerbas. – Timothy Wagner Dec 7 '10 at 23:15
Oh, I'm a big fan of Dirichlet's Unit Theorem. Only a hair behind the Finiteness of the Class Number in my estimation. – Arturo Magidin Dec 7 '10 at 23:16
Yes, Timothy you are right, the Unit Theorem reduces this one to a triviality. Thanks. – Jason Smith Dec 7 '10 at 23:55

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.