MathLabs

Open problem, Arithmetic and number theory, Analysis, posed 1734

Irrationality of the Euler–Mascheroni constant

Open

Is the Euler–Mascheroni constant γ=lim⁡n→∞(∑k=1n1k−ln⁡n)≈0.5772156649\gamma = \lim_{n \to \infty} \left( \sum_{k=1}^n \frac{1}{k} - \ln n \right) \approx 0.5772156649 an irrational number?

Research frontier as of 2026

As of 2026, it remains unknown whether γ\gamma is irrational. Unlike ζ(3)\zeta(3) or ln⁡2\ln 2, γ\gamma is not known to be a period in the sense of Kontsevich and Zagier, because it arises as the regularized constant term at the pole s=1s = 1 of ζ(s)\zeta(s) rather than as an integral of an algebraic differential form over a domain defined by algebraic inequalities. Padé approximants and linear forms in logarithms of rational numbers produce rational approximations pn/qnp_n/q_n to γ\gamma, but their denominators qnq_n grow too fast relative to the remainder ∣qnγ−pn∣|q_n \gamma - p_n| to cross the threshold ∣qnγ−pn∣→0|q_n \gamma - p_n| \to 0 required for an irrationality proof.

Best known results

  • Aptekarev (2009) and Rivoal (2012): at least one of the Euler–Mascheroni constant γ\gamma and the Euler–Gompertz constant δ\delta is transcendental (and hence irrational).
  • Papanikolaou (1997): if γ=p/q\gamma = p/q in lowest terms, then q>10242080q > 10^{242080}.

Tools and where they stop

ToolAchievedWhere it stops
Simultaneous Padé approximations and hypergeometric integrals (Aptekarev–Rivoal)Proves disjunctive irrationality and transcendence for pairs such as (γ,δ)(\gamma, \delta)Linear forms built from Laplace-type integrals inevitably mix γ\gamma with e−1e^{-1} or δ\delta, preventing isolation of γ\gamma alone
High-precision continued-fraction expansionEstablishes astronomical lower bounds (q>10242080q > 10^{242080}) on any hypothetical rational denominator qqAny finite computation of digits or partial quotients can only rule out denominators up to a finite bound

Open questions

  • Is the Euler–Mascheroni constant γ\gamma irrational, and more strongly, is it transcendental?
  • Is γ\gamma an exponential period, or not a classical Kontsevich–Zagier period at all?

References

  1. Jeffrey C. Lagarias (2013). Euler's constant: Euler's work and modern developments · DOI:10.1090/S0273-0979-2013-01423-X · arXiv:1303.1856
  2. Julian Havil (2003). Gamma: Exploring Euler's Constant
  3. Tanguy Rivoal (2012). On the arithmetic nature of the values of the gamma function, Euler's constant, and Gompertz's constant · DOI:10.1307/mmj/1339011525