This is a front-end for the Online Encyclopedia of Integer Sequences, made by Christian Perfect. The idea is to provide OEIS entries in non-ancient HTML, and then to think about how they're presented visually. The source code is on GitHub.
%I A383872 #43 May 20 2025 00:27:51 %S A383872 9,12,26,56,76,122,332,992,2042,3344,4336,8186,16256,32762,227744, %T A383872 266176,269072,299576,856544,2097146,5385812,8388602,16580864, %U A383872 17895664,19173944,33554426,61008020,67100672,201931760,1074789376,1108378592,17179738112,62472251540,68700578816 %N A383872 Nonprime numbers whose sum of proper divisors is a power of 4. %C A383872 Includes 2*p for p in A135535. - _Robert Israel_, May 13 2025 %C A383872 From _David A. Corneth_, May 13 2025: (Start) %C A383872 If a(n) = m*p where p is the largest prime divisor and has multiplicity 1 and s = sigma(m) then p = (4^k - s) / (s - m). Using this, a(44), a(45) and a(46) are at most 4971704751572, 44088037271892 and 44358570286896 respectively. %C A383872 If a(n) is odd then a(n) is a perfect square. Proof: Suppose a(n) is not a perfect square. Then sigma(a(n)) is even and so sigma(a(n)) - a(n) = 4^k. As sigma(a(n)) - a(n) = 4^k then sigma(a(n)) - a(n) = 1. As a(n) is composite this has no solutions. (End) %C A383872 For the first 43 terms, a(1) is the only square term and for the other terms, a(n) has a squarefree odd part. However, this is not always true as 44088037271892 (see above) is a term and its odd part is not squarefree. - _Chai Wah Wu_, May 19 2025 %H A383872 Amiram Eldar, <a href="/A383872/b383872.txt">Table of n, a(n) for n = 1..43</a> (calculated from the b-file at A279731) %e A383872 12 is not prime; 12 has proper divisors 1, 2, 3, 4, and 6, with a sum of 16. This is a square number as well as a power of 2. %p A383872 filter:= proc(n) local s; %p A383872 s:= numtheory:-sigma(n)-n; %p A383872 s > 1 and s = 4^padic:-ordp(s,4) %p A383872 end proc: %p A383872 select(filter, [$4..10^7]); # _Robert Israel_, May 13 2025 %t A383872 Zweierpotenzen = {}; %t A383872 Quadratzahlen = {}; %t A383872 Beides = {}; %t A383872 For[k = 1, k <= 50000000, k++, %t A383872 SumET = Total[Divisors[k]] - k; %t A383872 If[IntegerQ[Log[2, SumET]] && PrimeQ[k] == False, %t A383872 AppendTo[Zweierpotenzen, k]]; %t A383872 If[IntegerQ[Sqrt[SumET]] && PrimeQ[k] == False, %t A383872 AppendTo[Quadratzahlen, k ]]]; %t A383872 Beides = Intersection[Zweierpotenzen, Quadratzahlen]; %t A383872 Beides %o A383872 (PARI) isok(k) = if (!ispseudoprime(k), my(s=sigma(k)-k, z); issquare(s) && (ispower(s, , &z) && (z==2))); \\ _Michel Marcus_, May 13 2025 %Y A383872 Intersection of A048699 and A279731. %Y A383872 Cf. A001065, A135535. %K A383872 nonn %O A383872 1,1 %A A383872 _Hans Ulrich Keller_, May 13 2025 %E A383872 a(27)-a(29) from _Michel Marcus_, May 13 2025 %E A383872 a(30)-a(34) from _Amiram Eldar_, May 13 2025