cp's OEIS Frontend

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.

Showing 1-3 of 3 results.

A069220 Denominator of Sum_{1<=k<=n, gcd(k,n)=1} 1/k.

Original entry on oeis.org

1, 1, 2, 3, 12, 5, 20, 105, 280, 63, 2520, 385, 27720, 6435, 8008, 45045, 720720, 85085, 4084080, 2909907, 3695120, 1322685, 5173168, 37182145, 118982864, 128707425, 2974571600, 717084225, 80313433200, 215656441, 2329089562800
Offset: 1

Views

Author

Sharon Sela (sharonsela(AT)hotmail.com), Apr 12 2002

Keywords

Crossrefs

Cf. A093600 (numerator of this sum).
Also denominators of row sums of A111879/A111880. For the numerators see A111881.

Programs

  • Mathematica
    Table[s=0; Do[If[GCD[i, n]==1, s=s+1/i], {i, n}]; Denominator[s], {n, 1, 35}]
    Table[Denominator[Total[1/Select[Range[n],GCD[n,#]==1&]]],{n,40}] (* Harvey P. Dale, Jun 07 2020 *)
  • PARI
    for(n=1,40,print1(denominator(sum(k=1,n,if(gcd(k,n)==1,1/k))),","))

Formula

G.f. A(x) (for fractions) satisfies: A(x) = -log(1 - x)/(1 - x) - Sum_{k>=2} A(x^k)/k. - Ilya Gutkovskiy, Mar 31 2020

Extensions

More terms from Jason Earls, Apr 14 2002

A290815 Numbers m such that the numerator of Sum_{k=1..m, gcd(k,m) = 1} 1/k is divisible by m^3.

Original entry on oeis.org

1, 39, 78, 155, 310, 465, 546, 793, 798, 930, 1092, 1586, 1638, 1860, 2170, 2379, 2394, 3172, 3276, 3965, 4340, 4758, 4914, 5219, 6045, 6510, 7137, 7182, 7930, 9516, 9828, 10374, 10438, 11102, 11895, 12090, 13020, 14274, 15657, 15860, 16843, 16891, 18135
Offset: 1

Views

Author

Amiram Eldar, Aug 11 2017

Keywords

Comments

A generalization of Wolstenholme primes (A088164) for composite number.
Leudesdorf proved in 1888 that the numerator of Sum_{k=1..n, gcd(k,n)=1} 1/k is divisible by n^2 for all (but not only) numbers n with gcd(n,6)=1, which is a generalization of Wolstenholme's theorem.
Terms that are coprime to 6: 1, 155, 793, 3965, 5219, 16843, 16891, 51305, ...
a(41) = A088164(1) = 16843.
A general conjecture: if, for some e > 0, m^e | Numerator(Sum_{k=1..m, gcd(k,m)=1} 1/k), then m^(e-1) | Numerator(Sum_{k=1..m, gcd(k,m)=1} 1/k^2). Note: in this case, the exponent e = 3. Problem: are there numbers m > 1 such that m^4 | Numerator(Sum_{k=1..m, gcd(k,m)=1} 1/k)? - Thomas Ordowski, Aug 10 2019
This general conjecture was checked up to 10^4. This problem has no solution up to 10^5. - Amiram Eldar, Aug 10 2019
It appears that all odd terms of this sequence are odd numbers m such that the numerator of Sum_{k=1..m, gcd(k,m)=1} 1/k^2 is divisible by m^2. - Thomas Ordowski, Aug 12 2019

Examples

			Sum_{k=1..39, gcd(k,39)=1} 1/k = 46855131783993/15222026943200, and 46855131783993 = 39^3 * 789884047, thus 39 is in the sequence.
		

References

  • G. H. Hardy and E. M. Wright, Introduction to the theory of numbers, 5th edition, Oxford, England: Clarendon Press, 1979, pp. 100-102.

Crossrefs

Programs

  • Mathematica
    seqQ[n_] :=  Module[{}, g[m_] := GCD[n, m] == 1; Divisible[Numerator[Plus @@ (1/Select[Range[n], g])], n^3]]; Select[Range[10^5], seqQ]
  • PARI
    isok(n) = numerator(sum(k=1, n, if (gcd(n, k)==1, 1/k))) % n^3 == 0; \\ Michel Marcus, Aug 11 2017
    
  • PARI
    upto(n) = {my(v = vector(n), d = divisors(n), res = List(), squarefreepart(n) = factorback(factorint(n)[, 1])); v[1] = 1; for(i = 2, n, v[i] = v[i-1] + 1/i; ); for(j = 1, n, fr = v[j]; d = divisors(squarefreepart(j)); for(i = 2, #d, fr += 1/d[i] * v[j/d[i]] * (-1)^omega(d[i]) ); if(numerator(fr) % j^3 == 0, listput(res, j); ) ); res } \\ David A. Corneth, Aug 23 2019

A099940 a(n) = 2*(A056855(n)) /(phi(n)*n), where phi() is the Euler phi function.

Original entry on oeis.org

2, 1, 1, 1, 5, 1, 84, 11, 184, 15, 193248, 23, 19056960, 833, 33740, 64035, 520105017600, 2473, 130859579289600, 203685, 963513600, 23748417, 16397141420298240000, 645119, 555804546402631680, 8527366575
Offset: 1

Views

Author

Leroy Quet, Nov 12 2004

Keywords

Comments

Conjecture: this sequence consists completely of integers.
From Leudesdorf's theorem this is an integer sequence. - Benoit Cloitre, Nov 13 2004

Examples

			a(6) = 2*(1 + 1/5)*1*5/(6*2) = 1.
		

References

  • G. H. Hardy and E. M. Wright, Introduction to the theory of numbers, fifth edition, Oxford Science Publication, pp. 100-102

Crossrefs

Programs

  • Mathematica
    f[n_] := Block[{k = Select[Range[n], GCD[ #, n] == 1 &]}, 2Plus @@ (Times @@ k*Plus @@ 1/k)/EulerPhi[n]/n]; Table[ f[n], {n, 26}] (* Robert G. Wilson v, Nov 16 2004 *)

Extensions

More terms from Don Reble, Nov 12 2004, who remarks that the conjecture is true for n <= 5000.
Showing 1-3 of 3 results.