← All problems

Erdős Problem #1049 — Irrationality of a divisor Lambert series

38 years old · traced to

For each integer n≥1n\geq 1, let τ(n)=#{d∈N:d∣n}\tau(n)=\#\{d\in\mathbb{N}:d\mid n\} be the number of positive divisors of nn. For rational t>1t>1, both series below converge, and expansion into positive geometric series gives

∑n=1∞1tn−1=∑n=1∞∑k=1∞t−nk=∑m=1∞τ(m)tm.\sum_{n=1}^{\infty}\frac{1}{t^n-1} =\sum_{n=1}^{\infty}\sum_{k=1}^{\infty}t^{-nk} =\sum_{m=1}^{\infty}\frac{\tau(m)}{t^m}.

Is this common value irrational for every rational t>1t>1; that is, does ∑n=1∞(tn−1)−1∉Q\sum_{n=1}^{\infty}(t^n-1)^{-1}\notin\mathbb{Q} hold for all such tt? Erdős Problems #1049

Background

This is an irrationality question for a Lambert series: expanding each denominator as a geometric series and collecting terms with the same exponent makes the coefficient of t−mt^{-m} equal to the number of divisors of mm. The rational-parameter assertion is attributed to Chowla; Erdős’s 1948 paper places it among irrationality questions for Lambert series. Erdős Problems #1049 Erdős, On arithmetical properties of Lambert series (1948)

References

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.