Erdős Problem #1050 — Irrationality of a shifted binary Lambert series

About 38 years old · traced to

Is ∑n=1∞1/(2n−3)\sum_{n=1}^{∞}1/(2^n-3) irrational?

References

Additional references

P. Erdős, On the irrationality of certain series: problems and results, in New Advances in Transcendence Theory (1988), 102–109.

Progress summary

Refreshed
Claimed solved

A general theorem reported in 2024 implies that this particular series is irrational, although this scan does not independently verify the proof.

Erdős recorded the question in 1988: determine whether the series with denominator 2n−32^n-3 is irrational.

Known result reported in 2024

A 2024 arXiv paper states that Borwein proved the more general result that ∑n=1∞1/(tn+q)\sum_{n=1}^{\infty}1/(t^n+q) is irrational for integer t≥2t\ge2 and rational q≠0,−tnq\ne0,-t^n for every nn. Substituting t=2t=2 and q=−3q=-3 gives the Erdős problem, since −3≠0-3\ne0 and −3≠−2n-3\ne-2^n for every nn; the paper also mentions an alternative proof elsewhere.

Current status (as of September 2026): The sum is reported irrational by Borwein's general theorem, so the problem is claimed solved; this report does not independently verify the theorem.

Sources

Solutions 0

No solutions have been posted yet.