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

Let $l$ be an odd prime number and $\zeta$ be a primitive $l$-th root of unity in $\mathbb{C}$. Let $K = \mathbb{Q}(\zeta)$. Let $A$ be the ring of algebraic integers in $K$. Let $G$ be the Galois group of $\mathbb{Q}(\zeta)/\mathbb{Q}$. $G$ is isomorphic to $(\mathbb{Z}/l\mathbb{Z})^*$. Hence $G$ is a cyclic group of order $l - 1$. Let $\sigma$ be one of its generators. Then $\sigma(\zeta) = \zeta^r$, where $r$ is a primitive root mod $l$. Let $f$ be a positive divisor of $l - 1$. Let $e = (l - 1)/f$. There exists a unique subgroup $G_f$ of $G$ whose order is $f$. Let $K_f$ be the fixed subfield of $K$ by $G_f$. Then $e = [K_f : \mathbb{Q}]$. Let $A_f$ be the ring of algebraic integers in $K_f$.

We introduce the following algebraic integers.

\begin{align*} \eta_0 &= \zeta + \sigma^e(\zeta) + \sigma^{2e}(\zeta) + \dots + \sigma^{(f-1)e}(\zeta) \\ \eta_1 &= \sigma(\zeta) + \sigma^{e+1}(\zeta) + \dots + \sigma^{(f-1)e + 1}(\zeta) \\ &\vdots \\ \eta_{e-1} &= \sigma^{e-1}(\zeta) + \sigma^{2e-1}(\zeta) + \dots + \sigma^{fe - 1}(\zeta) \end{align*}

$\eta_0, \ldots, \eta_{e-1}$ are called periods of length $f$.

Applying $\sigma$ to $\eta_0$, we get $\eta_1$. Similarly,

$\eta_0 \to \eta_1 \to \dots \to \eta_{e-1} \to \eta_0$

My question: Is the following proposition true? If yes, how would you prove this?

Proposition The set $\{\eta_0, \ldots, \eta_{e-1}\}$ is an integral base of $A_f$.

share|cite|improve this question
I played around with the $\TeX$ a bit. Take a look — there are some things in there that might save you time. Interesting series of questions, by the way. I wish I knew the answers offhand. – Dylan Moreland Jul 23 '12 at 22:25
@DylanMoreland Thanks, Dylan. – Makoto Kato Jul 23 '12 at 22:26
Doesn't Washington discuss this in the first or second chapter of his book on cyclotomic fields, at the place where he defines periods? – KCd Jul 24 '12 at 0:45
@KCd I don't have the book at hand. – Makoto Kato Jul 24 '12 at 3:17
What's the reason for the downvote? Unless you make it clear, I cannot improve my question. – Makoto Kato Aug 13 '12 at 2:24

The $\eta_i$ form a basis for $A_f$, because the algebraic integers of $K_f$ are the algebraic integers of $K$ that are in $K_f$.

Since $l$ is an odd prime, the family $\{\sigma^i(\zeta)\}$ is the same as $\{\zeta^k,k=1 \ldots l-1\}$, and $1 = - (\zeta^1 + \ldots \zeta^{l-1})$, which should convince you that this family form a basis for the algebraic integers of $K$. Because it is also an orbit of the Galois group, we say it is an integral normal basis.
The elements of $K$ that are elements of $K_f$ are those fixed by $\sigma^e$. Thus, an element of $A_f$ is of the form $x = \sum_{i=0}^{l-1} a_i \sigma^i(\zeta)$ where $a_i \in \mathbb Z$, such that $\sigma^e(x) = x$, which means $a_{i+e} = a_i$, so regrouping the terms you get $x = a_0\eta_0 + a_1 \eta_1 + \ldots a_{e-1} \eta_{e-1}$

This works because we have an integral normal basis. Some number fields, such as $\mathbb Q(i)$, don't have such an integral basis for their ring of integers.

share|cite|improve this answer
On the topic of cyclotomic fields without an integral normal basis, see… – KCd Jul 24 '12 at 0:46

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.