New sequence derived from the totient function

Benoît Jubin benoit.jubin at gmail.com
Thu May 15 02:44:53 CEST 2008


>>  a(n) is the least m such that k>=m implies phi(k)>=n

I will submit this sequence, but I'd like to find an asymptotic first.
 There might be a constant c such that a(n) is equivalent to
c.n.ln(ln(n)) (?)  Or maybe a(n)/(n.ln(ln(n))) has finite positive
limsup and liminf ?

> 1, 2, 7, 7, 13, 13, 19, 19, 31, 31, 31, 31, 43, 43, 43, 43, 61, 61,

Thanks for the computation, Max.  Shouldn't the second term be 3 instead of 2 ?

Regards,
Benoit





More information about the SeqFan mailing list