A277173 - OEIS

%I #20 Jan 02 2017 02:00:12

%S 1,2,6,12,24,60,120,126,240,420,480,504,672,780,1248,1260,2340,2520,

%T 3360,4680,5040,5460,6240,6552,8160,8736,9360,10080,11424,16380,21216,

%U 26208,27360,32760,38304,43680,57120,65520,71136,74592,106080,131040,147168,148512,171360,191520,202464,325920,355680,372960

%N Numbers m such that b^sigma(m) == b^phi(m) == b^numdiv(m) == b^m (mod m) for every integer b.

%C Are terms products products of primes of the form 2^i*3^j + 1, A058383, for some nonnegative i and j? This is true for all terms up to 7.6*10^6. 7600320 is divisible by 29, which isn't of the form 2^j*3^i+1. Up to 10^8, all of the terms are divisible by only 16 distinct prime factors. That is: omega(lcm(all terms up to 10^8)) = 16.

%C Subsequence of A124240.

%H Robert G. Wilson v, <a href="/A277173/b277173.txt">Table of n, a(n) for n = 1..510</a>

%e 6 is a term because for the primes up to 6, (2, 3 and 5), b^sigma(6) == b^phi(6) == b^numdiv(6) == b^6 (mod 6). This is sufficient to prove for all values b up to 6.

%t fQ[n_] := Block[{b = 2, s = DivisorSigma[1, n], e = EulerPhi[n], d = DivisorSigma[0, n]}, While[b < n && PowerMod[b, s, n] == PowerMod[b, e, n] == PowerMod[b, d, n] == PowerMod[b, n, n], b = NextPrime@ b]; b >= n]; lst = {1}; k = 2; While[k < 400000, If[ fQ@ k, AppendTo[lst, k]]; k ++]; lst (* _Robert G. Wilson v_, Nov 04 2016 *)

%o (PARI) isk(n, k) = {Mod(k, n)^sigma(n)==Mod(k, n)^n && Mod(k, n)^eulerphi(n)==Mod(k, n)^n && Mod(k, n)^numdiv(n)==Mod(k, n)^n}

%o is(n) = my(i);forprime(i=2, n, if(isk(n, i)==0,return(0))) ; 1

%o upto(lim) = my(l=List());for(n=1, lim, if(is(n), listput(l,n))); l

%Y Cf. A124240.

%K nonn

%O 1,2

%A _David A. Corneth_ and _Altug Alkan_, Oct 02 2016