login

Year-end appeal: Please make a donation to the OEIS Foundation to support ongoing development and maintenance of the OEIS. We are now in our 61st year, we have over 378,000 sequences, and we’ve reached 11,000 citations (which often say “discovered thanks to the OEIS”).

A320146
a(n) = 2*prime(n) modulo (prime(n-1) + prime(n+1)).
1
6, 0, 14, 2, 26, 2, 38, 46, 4, 62, 2, 2, 86, 94, 0, 4, 122, 2, 2, 146, 2, 166, 178, 4, 2, 206, 2, 218, 226, 10, 262, 4, 278, 8, 302, 0, 2, 334, 0, 4, 362, 8, 386, 2, 398, 0, 8, 2, 458, 466, 4, 482, 4, 0, 0, 4, 542, 2, 2, 566, 586, 10, 2, 626, 634, 8, 674, 8, 698, 706, 718, 2, 0, 2, 766, 778, 4, 802, 818, 8, 842
OFFSET
2,1
COMMENTS
This sequence has to do with the relative position of primes with respect to their adjacent primes:
(i) if prime(n) is closer to its predecessor than to its successor, then a(n) = 2*prime(n);
(ii) if prime(n) is closer to its successor than to its predecessor, then a(n) = 2*prime(n) - prime(n-1) - prime(n+1); and
(iii) if prime(n) is equidistant from its predecessor and its successor, then a(n) = 0.
Is lim_{n -> infinity} (Sum_{i=1..n} a(i))/(Sum_{i=1..n} prime(i)) finite? If so, what is its value?
LINKS
MAPLE
seq(modp(2*ithprime(n), (ithprime(n-1)+ithprime(n+1))), n=2..90); # Muniru A Asiru, Oct 07 2018
MATHEMATICA
Table[Mod[2*Prime[n], Prime[n-1] + Prime[n+1]], {n, 2, 120}]
Mod[2#[[2]], #[[1]]+#[[3]]]&/@Partition[Prime[Range[90]], 3, 1] (* Harvey P. Dale, Jan 03 2019 *)
PROG
(PARI) a(n) = 2*prime(n) % (prime(n-1) + prime(n+1)); \\ Michel Marcus, Oct 18 2018
CROSSREFS
KEYWORD
nonn
AUTHOR
Andres Cicuttin, Oct 06 2018
STATUS
approved