The prime-counting Copeland-Erdős constant (Q6607758)

For a sequence \((a(n):n\in\mathbb{N})\) of nonnegative integers, let \(0.a(1)a(2)\ldots\) denote the real value given by concatenating the digit expansions of consecutive entries of this sequence. Many researches on digit expansions of this form has to do with the normality of \(0.a(1)a(2)\ldots\) for a given base. It is well-known that the Copeland-Erdős constant \(0.2357111317\ldots\) is normal in base \(10\), where \(a(n)=p_n\) (the \(n\)th prime number). In this paper, the author introduces a constant related to the prime-counting function \(\pi\) is that may be seen as something of ``inverse'' construction given by concatenating the decimal digits of \((\pi(n):n\in\mathbb{N})\). The author considers the constant, called the prime-counting Copeland-Erdős (PCCE) constant, \[0.012233444455666677888899999910101111\dots\] by setting \(a(n)=\pi(n)\) given by the prime-counting function. A serious problem has to do with how the distribution of groupings of digits in the PCCE constant would depend on the behaviour of the sequence \((g_n=p_{n+1}-p_n: n\in\mathbb{N})\) of prime gaps, since the number of times \(n\in\mathbb{N}\) appears in \((\pi(n):\in\mathbb{N})\) is equal to \(g_n\). In relation to this constant, using a combinatorial method due to \textit{P. Szüsz} and \textit{B. Volkmann} [Forum Math. 6, No. 4, 399--414 (1994; Zbl 0806.11034)], the author proves, under the assumption of Cramér's conjecture, that this new constant is normal in a given base \(g\ge 2\), for \(a(n)=\pi(n)\).