login
A387363
The number of decompositions of 2*n into ordered sums of two cyclic numbers.
2
1, 3, 3, 4, 3, 4, 5, 6, 8, 8, 7, 8, 7, 6, 9, 8, 11, 12, 11, 10, 12, 12, 13, 16, 12, 14, 16, 12, 13, 14, 13, 16, 19, 14, 19, 20, 19, 20, 20, 20, 21, 26, 19, 24, 26, 22, 25, 26, 24, 26, 33, 26, 27, 30, 26, 28, 32, 26, 29, 38, 25, 30, 34, 26, 33, 34, 29, 30, 41, 28
OFFSET
1,2
COMMENTS
Analogous to A002372 with cyclic numbers (A003277) instead of odd primes.
Pomerance (2025) proved that a(n) > 0 for every sufficiently large n.
LINKS
Joel E. Cohen, Conjectures about Primes and Cyclic Numbers, arXiv:2508.08335 [math.NT], 2025.
Carl Pomerance, Patterns for cyclic numbers, 2025.
FORMULA
a(n) ~ C_2 * n / (exp(gamma) * log(log(log(n))))^2 * Product_{p | n, p odd prime < log(log(n/2))} (p-1)/(p-2), where C_2 = A005597, and gamma = A001620 (Pomerance, 2025).
EXAMPLE
a(1) = 1 since 2*1 = 1 + 1.
a(2) = 3 since 2*2 = 1 + 3 = 2 + 2 = 3 + 1.
a(3) = 3 since 2*3 = 1 + 5 = 3 + 3 = 5 + 1.
MATHEMATICA
cyclicQ[n_] := cyclicQ[n] = CoprimeQ[n, EulerPhi[n]]; a[n_] := Count[Range[2*n], _?(And @@ cyclicQ[{#, 2*n-#}] &)]; Array[a, 100]
PROG
(PARI) iscyclic(k) = gcd(k, eulerphi(k)) == 1;
a(n) = sum(k = 1, 2*n, iscyclic(k) * iscyclic(2*n-k));
CROSSREFS
KEYWORD
nonn,easy
AUTHOR
Amiram Eldar, Aug 27 2025
STATUS
approved