Known bounds on |π(x) − Li(x)| imply p_n lies between Li⁻¹(n(1 − a f(n ln n))) and Li⁻¹(n(1 + a f(n ln n))), but the stated n* is too small for large-a bounds.
Effective exponential bounds on the prime gaps
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Over the last 50 years a large number of effective exponential bounds on the first Chebyshev function $\vartheta(x)$ have been obtained. Specifically we shall be interested in effective exponential bounds of the form \[ |\vartheta(x)-x| < a \;x \;(\ln x)^b \; \exp\left(-c\; \sqrt{\ln x}\right); \qquad (x \geq x_0). \] Herein we shall convert these effective bounds on $\vartheta(x)$ into effective exponential bounds on the prime gaps $g_n = p_{n+1}-p_n$. Specifically we shall establish a number of effective exponential bounds of the form \[ {g_n\over p_n} < { 2a \;(\ln p_n)^b \; \exp\left(-c\; \sqrt{\ln p_n}\right) \over 1- a \;(\ln p_n)^b \; \exp\left(-c\; \sqrt{\ln p_n}\right)}; \qquad (x \geq x_*); \] and \[ {g_n\over p_n} < 3a \;(\ln p_n)^b \; \exp\left(-c\; \sqrt{\ln p_n}\right); \qquad (x \geq x_*); \] for some effective computable $x_*$. It is the explicit presence of the exponential factor, with known coefficients and known range of validity for the bound, that makes these bounds particularly interesting.
fields
math.NT 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
The n-th prime exponentially
Known bounds on |π(x) − Li(x)| imply p_n lies between Li⁻¹(n(1 − a f(n ln n))) and Li⁻¹(n(1 + a f(n ln n))), but the stated n* is too small for large-a bounds.