Rational Linear Combinations of Integral Basis
If \(\{\alpha_1, \dots, \alpha_n\}\) is an integral basis of \(\mathcal{O}_\mathbb{K}\), then \(\{\alpha_1, \dots, \alpha_n\}\) is a basis for \(\mathbb{K}\) over \(\mathbb{Q}\). Moreover, if
then \(x \in \mathcal{O}_\mathbb{K}\) if and only if all \(m_i\) are in \(\mathbb{Z}\).
Proof
Let \(\{\alpha_1, \dots, \alpha_n\}\) is an integral basis for \(\mathcal{O}_\mathbb{K}\), and as such, they are \(\mathbb{Z}\) linearly independent. Now let \(m_1, \dots, m_n, k_1, \dots, k_n \in \mathbb{Z}\) with \(k_1, \dots, k_n \neq 0\) and consider the \(\mathbb{Q}\) linear combination
Now we have that
which means that \(m_i k_1 \dots k_{i - 1} k_{i + 1} \dots k_n = 0\) for each \(i\) since \(\{\alpha_1, \dots, \alpha_n\}\) are \(\mathbb{Z}\) linearly independent. However \(k_j \neq 0\) for all \(j\), and so \(m_i = 0\) for all \(i\). This means that \(\frac{m_i}{k_i} = 0\) for all \(i\) and thus \(\{\alpha_1, \dots, \alpha_n\}\) is \(\mathbb{Q}\) linearly independent in \(\mathbb{K}\). Since this is a set of \(n\) vectors in an \(n\) dimensional space, linear independence implies it is a basis.
It is important to note that a converse result does not hold. That is, given a field basis of \(\mathbb{K}\) comprised entirely of integral elements, this does not necessarily have to be an integral basis.
\(\mathbb{Q}(\sqrt{5})\) has field basis \(\{1, \sqrt{5}\}\) but this is not an integral basis.
See ring of integers of quadratic field for details. In short, \(5 \equiv 1 \pmod 4\) and so \(\mathcal{O}_{\mathbb{Q}(\sqrt{5})} \neq \mathbb{Z}[\sqrt{5}]\).
As such, we need more sophisticated methods, namely the integral basis algorithm.