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 dirichlet series for the Vonmangoldt function, $\Lambda(n)$, which is equal to zero when $n$ is not a prime a power, and $\ln(p)$ when it is a prime power say, $n=p^j$, is


Where $\zeta(s)$ is the zeta function, and $\zeta'(s)$ is the derivative of the zeta function with respect to $s$,

This can be re-written as $$-\frac{\zeta'(s)}{\zeta(s)}=\sum_{k=1}^{\infty}\frac{\Lambda(k)}{k^s}=\sum_{p}\frac{\ln(p)}{p^s-1},$$ with the last sum ranging over all primes p,

Can someone help me brake this sum, up into prime congruence sums similarly $$\sum_{k=0}^{\infty}\frac{\Lambda(5k+1)}{(5k+1)^s}$$, ie re-write it as prime sums, where the primes are congruent to some b modulo $5$. I have done it before modulo $4$, and $3$ so I know it can be done, I am just having trouble restricting the powers appropietly to account for cases when $p^a\equiv1$ mod $5$, has no solutions.

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

See Greg Martin's answer on this thread: Von mangoldt function dirichlet series

In general,

$$\sum_{n=0}^\infty \frac{\Lambda(qn+b)}{(qn+b)^s}=\frac{1}{\phi(q)}\sum_{\chi \pmod{q}}\overline{\chi}(b)\sum_{p}\frac{\chi(p)\log p}{p^{s}-\chi(p)}.$$ We may also write this as $$\frac{-1}{\phi(q)}\sum_{\chi\pmod{q}}\overline{\chi}(b)\frac{L^{'}}{L}(s,\chi).$$ Looking at a specific example, this yields $$\sum_{n\equiv1\ (3)}^{\infty}\frac{\Lambda(n)}{n^{s}}=\sum_{p\equiv1\ (3)}\frac{\log p}{p^{s}-1}+\sum_{p\equiv2\ (3)}\frac{\log p}{p^{2s}-1} .$$

To see this, notice that $$\sum_{n=0}^\infty \frac{\Lambda(qn+b)}{(qn+b)^s}=\sum_{n\equiv b\pmod{q}} \frac{\Lambda(n)}{n^s}.$$ Using the orthogonality relations of the Dirichlet Characters, that is the fact that $$\frac{1}{\phi(q)}\sum_{\chi \pmod{q}} \chi(a)=\left\{ \begin{array}{c} 1\ \text{when } a\equiv 1 \pmod{q} \\ 0\ \text{otherwise} \end{array}\right\},$$ it follows that

$$\sum_{n=0}^\infty \frac{\Lambda(qn+b)}{(qn+b)^s}=\frac{1}{\phi(q)}\sum_{\chi\pmod{q}}\overline{\chi}(b)\sum_{n=1}^{\infty}\frac{\Lambda(n)\chi(n)}{n^{s}}$$ Recalling the function $L(s,\chi)$, this last sum then equals $$\frac{-1}{\phi(q)}\sum_{\chi\pmod{q}}\overline{\chi}(b)\frac{L^{'}}{L}(s,\chi).$$

From here, we obtain the first equation by expanding we obtain the first equation by using the fact that $$L(s,\chi)=\prod_{p}\left(1-\frac{\chi(p)}{p^{s}}\right)^{-1}.$$

Key Idea: Use Dirichlet characters to isolate an arithmetic progression. This is how most major results on primes in arithmetic progressions are proven.

share|cite|improve this answer
I appreciate the work you spent, but Im not trying to evaluate in terms of L functions, I want to re-write it as a series of primes. – Ethan Dec 17 '12 at 23:02
@Ethan: $L$-functions allow you to do exactly that. I'll expand a bit more. – Eric Naslund Dec 17 '12 at 23:17
@Ethan: Using the formula with Dirichlet characters I wrote above, if you look at the case $p=3k+1$, for example, we get that $$\sum_{n\equiv1\ (3)}^{\infty}\frac{\Lambda(n)}{n^{s}}=\sum_{p\equiv1\ (3)}\frac{\log p}{p^{s}-1}+\sum_{p\equiv2\ (3)}\frac{\log p}{p^{2s}-1} .$$ There will be a similar formula for all other moduli if you want to write it over sums of primes based on congruence class. – Eric Naslund Dec 22 '12 at 23:01
I do, but I want a general way of doing so, its a pain to check which individula congruences are satisfied between 1 and a, and then using that to re-write the sum – Ethan Dec 23 '12 at 1:57
@Ethan: For a general modulus $b$, we have that $$\sum_{a\in\left(\mathbb{Z}/b\mathbb{Z}\right)^{\times}}\sum_{p\equiv a\ (b)}\frac{\log p}{p^{s\cdot \text{ord}_{b}(a)}-1}$$ where $\text{ord}_{b}(a)$ is the order of $a$ in the multiplicative group $\left(\mathbb{Z}/b\mathbb{Z}\right)^{\times},$ that is the smallest positive integer $r$ such that $a^r\equiv 1\text{ mod }b$. I doubt there will be any further simplification since $\text{ord}_b(a)$ is a rather complicated function. – Eric Naslund Dec 23 '12 at 4:57

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.