×

# Can this be the solution for composites?

Propositions:

It is well known by the standard Riemann Zeta function that

$\zeta (s) = \displaystyle \sum_{n=1}^{\infty} \dfrac{1}{n^s}$

For $$\mathfrak{R} (s) > 1$$.

Also, the Prime Zeta function gives us

$P(s) = \displaystyle \sum_{p} \dfrac{1}{p^s}$

For $$p \in \{ \text{primes} \}$$ and $$\mathfrak{R} (s) > 1$$.

Conclusion:

This means that the sum of $$s^{\text{th}}$$ powers of the reciprocals of all composite numbers can be given by the formula

\begin{align} P'(s) &= \displaystyle \sum_{p'} \dfrac{1}{{p'}^s} \\ &= \sum_{n=1}^{\infty} \dfrac{1}{n^s} - \sum_{p} \dfrac{1}{p^s} - 1 \\ &= \zeta (s) - P(s) - 1 \end{align}

For $$p' \in \{ \text{composite numbers} \}$$ and $$\mathfrak{R} (s) > 1$$.

Knowing that infinite sums are risky to handle. I would like your opinions on whether the conclusion drawn from the two given true propositions made above are correct or not.

Note by Tapas Mazumdar
4 months, 3 weeks ago

Sort by:

if $$u_n,v_n$$ are convergent sequences then so is their difference. Provided $$\zeta(s),P(s)$$ converges which is obvious ,your conclusions are true enough · 4 months, 1 week ago