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”).

a(n) = isigma(n): sum of infinitary divisors of n.
89

%I #107 Sep 21 2024 08:42:00

%S 1,3,4,5,6,12,8,15,10,18,12,20,14,24,24,17,18,30,20,30,32,36,24,60,26,

%T 42,40,40,30,72,32,51,48,54,48,50,38,60,56,90,42,96,44,60,60,72,48,68,

%U 50,78,72,70,54,120,72,120,80,90,60,120,62,96,80,85,84,144,68,90

%N a(n) = isigma(n): sum of infinitary divisors of n.

%C A divisor of n is called infinitary if it is a product of divisors of the form p^{y_a 2^a}, where p^y is a prime power dividing n and sum_a y_a 2^a is the binary representation of y.

%C Multiplicative: If e = Sum_{k >= 0} d_k 2^k (binary representation of e), then a(p^e) = Product_{k >= 0} (p^(2^k*{d_k+1}) - 1)/(p^(2^k) - 1). - _Christian G. Bower_ and _Mitch Harris_, May 20 2005 [This means there is a factor p^2^k + 1 if d_k = 1, else the factor is 1. - _M. F. Hasler_, Oct 20 2022]

%C This sequence is an infinitary analog of the Dedekind psi function A001615. Indeed, a(n) = Product_{q in Q_n}(q+1) = n*Product_{q in Q_n} (1+1/q), where {q} are terms of A050376 and Q_n is the set of distinct q's whose product is n. - _Vladimir Shevelev_, Apr 01 2014

%H Amiram Eldar, <a href="/A049417/b049417.txt">Table of n, a(n) for n = 1..10000</a> (terms 1..7417 from R. J. Mathar)

%H Graeme L. Cohen, <a href="http://dx.doi.org/10.1090/S0025-5718-1990-0993927-5">On an integer's infinitary divisors</a>, Math. Comp. 54 (189) (1990) 395-411.

%H Steven R. Finch, <a href="/A007947/a007947.pdf">Unitarism and Infinitarism</a>, February 25, 2004. [Cached copy, with permission of the author]

%H J. O. M. Pedersen, <a href="http://amicable.homepage.dk/tables.htm">Tables of Aliquot Cycles</a> [Broken link]

%H J. O. M. Pedersen, <a href="http://web.archive.org/web/20140502102524/http://amicable.homepage.dk/tables.htm">Tables of Aliquot Cycles</a> [Via Internet Archive Wayback-Machine]

%H J. O. M. Pedersen, <a href="/A063990/a063990.pdf">Tables of Aliquot Cycles</a> [Cached copy, pdf file only]

%H Tomohiro Yamada, <a href="https://arxiv.org/abs/1705.10933">Infinitary superperfect numbers</a>, arXiv:1705.10933 [math.NT], 2017.

%F Let n = Product(q_i) where {q_i} is a set of distinct terms of A050376. Then a(n) = Product(q_i + 1). - _Vladimir Shevelev_, Feb 19 2011

%F If n is squarefree, then a(n) = A001615(n). - _Vladimir Shevelev_, Apr 01 2014

%F a(n) = Sum_{k>=1} A077609(n,k). - _R. J. Mathar_, Oct 04 2017

%F a(n) = A126168(n)+n. - _R. J. Mathar_, Oct 05 2017

%F Multiplicative with a(p^e) = Product{k >= 0, e_k = 1} p^2^k + 1, where e = Sum e_k 2^k, i.e., e_k is bit k of e. - _M. F. Hasler_, Oct 20 2022

%F a(n) = iphi(n^2)/iphi(n), where iphi(n) = A091732(n). - _Amiram Eldar_, Sep 21 2024

%e If n = 8: 8 = 2^3 = 2^"11" (writing 3 in binary) so the infinitary divisors are 2^"00" = 1, 2^"01" = 2, 2^"10" = 4 and 2^"11" = 8; so a(8) = 1+2+4+8 = 15.

%e n = 90 = 2*5*9, where 2, 5, 9 are in A050376; so a(n) = 3*6*10 = 180. - _Vladimir Shevelev_, Feb 19 2011

%p isidiv := proc(d, n)

%p local n2, d2, p, j;

%p if n mod d <> 0 then

%p return false;

%p end if;

%p for p in numtheory[factorset](n) do

%p padic[ordp](n,p) ;

%p n2 := convert(%, base, 2) ;

%p padic[ordp](d,p) ;

%p d2 := convert(%, base, 2) ;

%p for j from 1 to nops(d2) do

%p if op(j, n2) = 0 and op(j, d2) <> 0 then

%p return false;

%p end if;

%p end do:

%p end do;

%p return true;

%p end proc:

%p idivisors := proc(n)

%p local a, d;

%p a := {} ;

%p for d in numtheory[divisors](n) do

%p if isidiv(d, n) then

%p a := a union {d} ;

%p end if;

%p end do:

%p a ;

%p end proc:

%p A049417 := proc(n)

%p local d;

%p add(d, d=idivisors(n)) ;

%p end proc:

%p seq(A049417(n),n=1..100) ; # _R. J. Mathar_, Feb 19 2011

%t bitty[k_] := Union[Flatten[Outer[Plus, Sequence @@ ({0, #1} & ) /@ Union[2^Range[0, Floor[Log[2, k]]]*Reverse[IntegerDigits[k, 2]]]]]]; Table[Plus@@((Times @@ (First[it]^(#1 /. z -> List)) & ) /@ Flatten[Outer[z, Sequence @@ bitty /@ Last[it = Transpose[FactorInteger[k]]], 1]]), {k, 2, 120}]

%t (* Second program: *)

%t a[n_] := If[n == 1, 1, Sort @ Flatten @ Outer[ Times, Sequence @@ (FactorInteger[n] /. {p_, m_Integer} :> p^Select[Range[0, m], BitOr[m, #] == m &])]] // Total;

%t Array[a, 100] (* _Jean-François Alcover_, Mar 23 2020, after Paul Abbott in A077609 *)

%o (PARI) A049417(n) = {my(b, f=factorint(n)); prod(k=1, #f[,2], b = binary(f[k,2]); prod(j=1, #b, if(b[j], 1+f[k,1]^(2^(#b-j)), 1)))} \\ _Andrew Lelechenko_, Apr 22 2014

%o (PARI) isigma(n)=vecprod([vecprod([f[1]^2^k+1|k<-[0..exponent(f[2])], bittest(f[2],k)])|f<-factor(n)~]) \\ _M. F. Hasler_, Oct 20 2022

%o (Haskell)

%o a049417 1 = 1

%o a049417 n = product $ zipWith f (a027748_row n) (a124010_row n) where

%o f p e = product $ zipWith div

%o (map (subtract 1 . (p ^)) $

%o zipWith (*) a000079_list $ map (+ 1) $ a030308_row e)

%o (map (subtract 1 . (p ^)) a000079_list)

%o -- _Reinhard Zumkeller_, Sep 18 2015

%o (Python)

%o from math import prod

%o from sympy import factorint

%o def A049417(n): return prod(p**(1<<i)+1 for p, e in factorint(n).items() for i, j in enumerate(bin(e)[-1:1:-1]) if j=='1') # _Chai Wah Wu_, Jul 11 2024

%Y Cf. A037445, A004607, A091732, A127661, A293355.

%Y Cf. A049418 (3-infinitary), A074847 (4-infinitary), A097863 (5-infinitary).

%Y Cf. A000079, A030308, A027748, A124010.

%K nonn,mult

%O 1,2

%A _Yasutoshi Kohmoto_, Dec 11 1999

%E More terms from _Wouter Meeussen_, Sep 02 2001