Distribution of arithmetic functions on certain subsets of integers (Q2478038)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Distribution of arithmetic functions on certain subsets of integers
scientific article

    Statements

    Distribution of arithmetic functions on certain subsets of integers (English)
    0 references
    0 references
    0 references
    14 March 2008
    0 references
    Let \(d(n)\) denote the number, and \(\sigma(n)\) the sum, of the positive divisors of \(n\), \(\varphi\) denote Euler's totient function, and \(d_3(n)\) be the number of solutions of \(a_1a_2a_3= n\). For \(n\geq 2\), let \(p(n)\) be the least prime factor of \(n\) and \(\omega(n)\) the number of distinct primes dividing \(n\). In [Int. J. Math. Anal., to appear] the authors proved that \[ \#\{n\leq x: (nd(n),\sigma(n) = 1\}= (C+ o(1))\sqrt{{x\over\log x}} \] for a certain constant \(C\), thus improving a result of H. J. Kanold. The first three results of the current paper concern the analogous problem for the properties \[ \text{(i) }(nd(n),\varphi(p(n))=\text{ a power of 2}; \;\;\text{(ii) }(nd_3 (n), (p(n))= 1; \;\;\text{(iii) }U(n):= (nd_3(n),\sigma(n))= 1; \] the size of the corresponding sets are, respectively, \[ \text{(i) }(c_1+ o(1)){x\over \log\log\log_x};\;\text{(ii) }(c_2+ o(1)){x\over \sqrt{\log x\log\log\log x}}; \;\text{(iii) }(c_3+ o(1)){x\over \sqrt{\log x\log\log\log x}} \] as \(x\to\infty\) for certain positive constants \(c_1\), \(c_2\), \(c_3\). Theorem 4 establishes that, given any integer \(a\neq 0\), there exists a constant \(c_4> 0\) such that as \(x\to\infty\) \[ \sum_{\substack{ p\leq x\\ U(p+ a)= 1}} d(p+ a)= (c_4+ o(1)){li(x)\over \log\log\log x}. \] The final theorem states that, uniformly for \(r= (1+ o(1))\log\log x\), \[ \#\{n\leq x: (n,(p (n))= 1,\;\omega(n)= r\}\sim\#\{n\leq x: p(n)> \log\log x,\;\omega(n)= r\} \] as \(x\to\infty\), and an asymptotic formula for the right side is obtained. P. Erdős had derived in 1948 a corresponding result without the condition \(\omega(n)= r\). The proofs employ some classical methods including sieve arguments, known results and, in establishing the error terms, some intricate manipulations of various sums over primes.
    0 references
    0 references
    Euler function
    0 references
    sum of divisors
    0 references
    distribution of arithmetic functions
    0 references
    0 references