The primorial up to a real bound is the product of all prime numbers not exceeding :For example, , while . Each prime factor occurs once, so a primorial is a squarefree integer. It stays constant between successive primes, and for by the empty product convention.
Its natural logarithm is the Chebyshev theta function, . This converts multiplicative bounds on primorials into additive estimates over primes. In particular, the elementary primorial bound gives for .
The Chebyshev theta function is one of the two Chebyshev functions and is the natural logarithm of the primorial:where both the sum and the product range over prime numbers.
The Chebyshev theta function satisfies for every sufficiently large real number . This follows without the Prime number theorem.
Write for the second Chebyshev function. The largest binomial coefficient in row is the central one, and that row sums to , soBy the Legendre formula, the prime valuation of this central binomial coefficient is a sum of terms , each zero or one. Hence the coefficient divides , whose natural logarithm is . Taking logarithms gives , the central-coefficient argument of Chebyshev estimate from central binomial coefficients.
For real , take . Monotonicity gives . Higher prime powers contribute at most to , as proved for the second Chebyshev function. ThereforeAfter division by , the right side tends to , proving the stated lower bound.
For every real ,For integer , strong induction proves this by splitting according to parity. If , every prime in divides , soIf , the valuation of a near-central binomial coefficient givesThe binomial estimates follow from the binomial theorem; in the odd case the two equal central coefficients together are at most . The result for real follows by replacing with .