Deje $p_1 < p_2 < \dots$ ser el de los números primos, y deje $a_n = 0$ si $n$ no es primo y $a_n = \frac{1}{k}$ para $n=p_k^2$.
Tomar cualquier $\frac{l}{q} \in \mathbb{Q}$. Nota, por el teorema de los números primos para progresiones aritméticas, que $$\lim_{K \to \infty} \frac{1}{K}\sum_{k \le K} e^{2\pi i p_k^2\frac{l}{q}} = \frac{1}{\phi(q)}\sum_{\substack{a \le q \\ (a,q) = 1}} e^{2\pi i a^2 \frac{l}{q}}.$$ By a standard summation by parts argument, it follows that $$\lim_{K \to \infty} \frac{1}{\log K} \sum_{k \le K} \frac{1}{k}e^{2\pi i p_k^2\frac{l}{q}} = \frac{1}{\phi(q)}\sum_{\substack{a \le q \\ (a,q) = 1}} e^{2\pi i a^2 \frac{l}{q}}.$$ Therefore, if we show that the sum $\sum_{un \le q \\ (a,q) = 1} e^{2\pi i a^2\frac{l}{q}}$ is nonzero for any $q \ge 1$ and $(l,q) = 1$, it follows that $\sum_{k \K le} \frac{1}{k} e^{2\pi i p_k^2\frac{l}{q}}$ golpes en magnitud y, en particular, diverge. Se demuestra que la suma es cero debajo de la tapa.
Ahora tomar cualquier $\alpha \in \mathbb{R}\setminus\mathbb{Q}$. Es bien conocido que $$\lim_{K \to \infty} \sum_{k \le K} \frac{1}{k}e^{2\pi i p_k^2\alpha}$$ exists, which clearly implies that $\sum_{n \le N} a_ne^{2\pi i \alpha n}$ converges as $N \to \infty$. Esto completa el argumento.
\begin{align}
&\cup_{i=1}^{\infty} \cap_{n=1}^{\infty} \{i \in B_n\} \\
& \cap_{n=1}^{\infty} \cup_{i=1}^{\infty} \{i \in B_n\}
\end---------------------------------------------------------
El uso de la identidad de $\sum_{d \mid n} \mu(d) = 1_{n=1}$, vemos $$\sum_{a \le q \\ (a,q) = 1} e^{2\pi i a^2\frac{l}{q}} = \sum_{a \le q} e^{2\pi i a^2\frac{l}{q}}\sum_{d \mid (a,q)} \mu(d) = \sum_{a \le q} e^{2\pi i a^2\frac{l}{q}} \sum_{\substack{d \mid a \\ d \mid q}} \mu(d)$$ $$ = \sum_{d \mid q} \mu(d) \sum_{d \mid a \le q} e^{2\pi i a^2\frac{l}{q}} = \sum_{d \mid q} \mu(d)\sum_{j \le q/d} e^{2\pi i j^2\frac{dl}{q/d}}$$ where the last equality was obtained by substituting $a = jd$. We have a linear combination of quadratic Gauss sums. Each $\sum e^{2\pi i j^2\frac{dl}{p/d}}$ is either $0$ or $c\sqrt{p/d}$ for $c \in \{\pm 1, \pm i, \pm (1+i), \pm i(1+i)\}$. If $q \no \equiv 2 \pmod{4}$, then the term corresponding to $d=1$ is $c\sqrt{q}$ which cannot cancel with any of the other terms and we're done. If $q \equiv 2 \pmod{4}$, then the term $d=2$ will give a $-c\sqrt{q/2}$ plazo que no puede cancelar con cualquiera de los otros términos y hemos terminado.