Abstract:Let $s_1, s_2, s_3, \cdots$ be the set of squarefree numbers in ascending order. In this paper, we prove that the following asymptotic on moments of gaps between squarefree numbers \[ \sum_{s_{k+1} \le x} (s_{k+1} - s_k)^\gamma \sim B(\gamma) x \; \; \mbox{ with some constant} \; \; B(\gamma) > 0 \] is true for $0 \le \gamma < 3.75$. This improves the previous best range $0 \le \gamma < 3.6875$.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about the asymptotic behavior of the moments of the gaps between square - free numbers. Specifically, the author tries to prove that for \(0\leq\gamma < 3.75\), the following asymptotic formula holds:
\[
\sum_{s_{k + 1}\leq x}(s_{k + 1}-s_k)^{\gamma}\sim B(\gamma)x
\]
where \(B(\gamma)\) is a positive constant and is defined by:
\[
B(\gamma)=\sum_{h = 1}^{\infty}h^{\gamma}\alpha(h)
\]
Here \(\alpha(h)\) is a constant independent of \(x\) and satisfies:
\[
\log\alpha(h)\leq-\frac{5}{4}h\log\log h+O(h)
\]
### Background and Motivation
A square - free number is a positive integer that is not divisible by the square of any prime number. For example, 6 is a square - free number, while 12 is not. The first few square - free numbers are:
\[
s_1 = 1, s_2 = 2, s_3 = 3, s_4 = 5, s_5 = 6, s_6 = 7, s_7 = 10,\ldots
\]
It is known that the asymptotic density of the set of square - free numbers is \(\frac{6}{\pi^2}>\frac{1}{2}\), which means that there are infinitely many consecutive pairs of square - free numbers (i.e., \(s_{k + 1}-s_k = 1\) occurs infinitely many times). In addition, Mirsky proved the following asymptotic formula:
\[
\sum_{s_{k+1}\leq x, s_{k+1}-s_k = h}1=\alpha(h)x+O\left(\frac{hx}{\log x\log\log x}\right)
\]
Erdős proved that for \(0\leq\gamma\leq 2\), the above asymptotic formula for the moment holds. Subsequently, Hooley, Filaseta, Huxley and others gradually improved this result, extending the range to \(0\leq\gamma < 3.6875\).
### Main Contributions of the Paper
This paper further extends the effective range to \(0\leq\gamma < 3.75\) through more refined analysis and the use of results of higher - order derivatives. Specifically, the author proves the following theorem:
**Theorem 1**: For \(0\leq\gamma < 3.75\),
\[
\sum_{x/2 < s_{k+1}\leq x}(s_{k+1}-s_k)^{\gamma}\sim\frac{1}{2}B(\gamma)x
\]
This theorem significantly improves the previous best result and provides a new direction for further research.