Satz von Wolstenholme

Der Satz von Wolstenholme (nach Joseph Wolstenholme) ist eine Aussage aus dem mathematischen Teilgebiet der Zahlentheorie. Er lautet:

Ist p 5 {\displaystyle p\geq 5} eine Primzahl, so hat die harmonische Zahl

H ( p 1 ) = 1 + 1 2 + 1 3 + 1 4 + + 1 p 1 {\displaystyle H(p-1)=1+{\frac {1}{2}}+{\frac {1}{3}}+{\frac {1}{4}}+\ldots +{\frac {1}{p-1}}}

einen durch p 2 {\displaystyle p^{2}} teilbaren Zähler (in vollständig gekürzter und daher auch in jeder anderen Darstellung als Quotient zweier ganzer Zahlen).[1][2]

Beispiele, andere Formulierungen, Folgerungen

Zur Veranschaulichung einige Beispiele:

  • p = 7 : 1 + 1 2 + 1 3 + 1 4 + 1 5 + 1 6 = 49 20 , {\displaystyle p=7{\text{:}}\quad 1+{\tfrac {1}{2}}+{\tfrac {1}{3}}+{\tfrac {1}{4}}+{\tfrac {1}{5}}+{\tfrac {1}{6}}={\tfrac {49}{20}},} der Zähler 49 = 1 7 2 {\displaystyle 49=1\cdot 7^{2}} ist durch 7 2 {\displaystyle 7^{2}} teilbar.
  • p = 13 : 1 + 1 2 + 1 3 + 1 4 + 1 5 + 1 6 + 1 7 + 1 8 + 1 9 + 1 10 + 1 11 + 1 12 = 86021 27720 , {\displaystyle p=13{\text{:}}\quad 1+{\tfrac {1}{2}}+{\tfrac {1}{3}}+{\tfrac {1}{4}}+{\tfrac {1}{5}}+{\tfrac {1}{6}}+{\tfrac {1}{7}}+{\tfrac {1}{8}}+{\tfrac {1}{9}}+{\tfrac {1}{10}}+{\tfrac {1}{11}}+{\tfrac {1}{12}}={\tfrac {86021}{27720}},} der Zähler 86021 = 509 13 2 {\displaystyle 86021=509\cdot 13^{2}} ist durch 13 2 {\displaystyle 13^{2}} teilbar.

Der Satz von Wolstenholme ist äquivalent zu der Aussage, dass der Zähler von

1 + 1 2 2 + 1 3 2 + 1 4 2 + + 1 ( p 1 ) 2 {\displaystyle 1+{\frac {1}{2^{2}}}+{\frac {1}{3^{2}}}+{\frac {1}{4^{2}}}+\ldots +{\frac {1}{(p-1)^{2}}}}

durch p {\displaystyle p} teilbar ist.[3]

Eine Folgerung aus dem Satz ist die Kongruenz

( 2 p p ) 2 mod p 3 , {\displaystyle {\binom {2p}{p}}\equiv 2\mod {p^{3}},}

die auch in der Form

( 2 p 1 p 1 ) 1 mod p 3 {\displaystyle {\binom {2p-1}{p-1}}\equiv 1\mod {p^{3}}}

geschrieben werden kann.

Wolstenholme-Primzahlen

Eine Wolstenholme-Primzahl p ist eine Primzahl, die eine stärkere Fassung des Satzes von Wolstenholme erfüllt, genauer: die eine der folgenden äquivalenten Bedingungen erfüllt:[4]

  • Der Zähler von
1 + 1 2 + 1 3 + + 1 p 1 {\displaystyle 1+{\frac {1}{2}}+{\frac {1}{3}}+\dots +{\frac {1}{p-1}}}
ist durch p 3 {\displaystyle p^{3}} teilbar.
  • Der Zähler von
1 + 1 2 2 + 1 3 2 + + 1 ( p 1 ) 2 {\displaystyle 1+{\frac {1}{2^{2}}}+{\frac {1}{3^{2}}}+\dots +{\frac {1}{(p-1)^{2}}}}
ist durch p 2 {\displaystyle p^{2}} teilbar.
  • Es gilt die Kongruenz
( 2 p p ) 2 mod p 4 . {\displaystyle {\binom {2p}{p}}\equiv 2\mod {p^{4}}.}
  • Es gilt die Kongruenz
( 2 p 1 p 1 ) 1 mod p 4 . {\displaystyle {\binom {2p-1}{p-1}}\equiv 1\mod {p^{4}}.}
  • Der Zähler der Bernoulli-Zahl B p 3 {\displaystyle B_{p-3}} ist durch p {\displaystyle p} teilbar.

Die beiden bisher einzigen bekannten Wolstenholme-Primzahlen sind 16843 (Selfridge und Pollack 1964)[5] und 2124679 (Buhler, Crandall, Ernvall und Metsänkylä 1993).[6] Jede weitere Wolstenholme-Primzahl müsste größer als 109 sein.[7] Es wurde die Vermutung aufgestellt, dass unendlich viele Wolstenholme-Primzahlen existieren, und zwar etwa log ( log ( x ) ) {\displaystyle \log(\log(x))} unterhalb x {\displaystyle x} (McIntosh 1995).[8]

Verwandter Begriff

Betrachtet man nur Summanden mit ungeradem Nenner, also die Summe

1 + 1 3 + 1 5 + + 1 p 2 {\displaystyle 1+{\frac {1}{3}}+{\frac {1}{5}}+\dots +{\frac {1}{p-2}}}

für eine Primzahl p 3 {\displaystyle p\geq 3} , so ist der Zähler genau dann durch p {\displaystyle p} teilbar, wenn die stärkere Form

2 p 1 1 mod p 2 {\displaystyle 2^{p-1}\equiv 1\mod {p^{2}}}

des Satzes von Euler-Fermat gilt.[9] Derartige Primzahlen heißen Wieferich-Primzahlen.

Geschichte

Aus dem Satz von Wilson folgt die Kongruenz

( n p 1 p 1 ) 1 ( mod p ) {\displaystyle {\binom {np-1}{p-1}}\equiv 1{\pmod {p}}}

für jede Primzahl p {\displaystyle p} und jede natürliche Zahl n . {\displaystyle n.}

Charles Babbage bewies 1819[10] die Kongruenz

( 2 p 1 p 1 ) 1 ( mod p 2 ) {\displaystyle {\binom {2p-1}{p-1}}\equiv 1{\pmod {p^{2}}}}

für jede Primzahl p > 2. {\displaystyle p>2.}

Joseph Wolstenholme bewies 1862[1] die Kongruenz

( 2 p 1 p 1 ) 1 ( mod p 3 ) {\displaystyle {\binom {2p-1}{p-1}}\equiv 1{\pmod {p^{3}}}}

für jede Primzahl p > 3. {\displaystyle p>3.}

Literatur

  • G. H. Hardy, E. M. Wright: An introduction to the theory of numbers. 6. Auflage. Oxford University Press, Oxford 2008, ISBN 978-0-19-921985-8 (englisch; revidiert von D. R. Heath-Brown und J. H. Silverman).
  • The Prime Glossary: Wolstenholme prime (englisch)
  • Eric W. Weisstein: Wolstenholme’s Theorem. In: MathWorld (englisch).

Einzelnachweise

  1. a b J. Wolstenholme: On certain properties of prime numbers. In: The quarterly journal of pure and applied mathematics 5. 1862, S. 35–39 (englisch).
  2. Hardy, Wright: An introduction to the theory of numbers. 2008, S. 112 (englisch; Theorem 115).
  3. Hardy, Wright: An introduction to the theory of numbers. 2008, S. 114 (englisch; Theorem 117).
  4. Anthony Gardiner: Four problems on prime power divisibility. In: The American Mathematical Monthly 95. Dezember 1988, S. 926–931 (englisch).
  5. J. L. Selfridge, B. W. Pollack: Fermat’s last theorem is true for any exponent up to 25,000. In: Notices of the AMS 11. 1964, S. 97 (englisch; nur Zusammenfassung; 16843 nicht ausdrücklich angegeben).
  6. J. Buhler, R. Crandall, R. Ernvall, T. Metsänkylä: Irregular primes and cyclotomic invariants to four million. In: Mathematics of Computation 61. Juli 1993, S. 151–153 (englisch).
  7. Richard J. McIntosh, Eric L. Roettger: A search for Fibonacci-Wieferich and Wolstenholme primes. (PDF; 151 kB). In: Mathematics of Computation, 76, Oktober 2007, S. 2087–2094 (englisch).
  8. Richard J. McIntosh: On the converse of Wolstenholme’s theorem. (PDF; 190 kB). In: Acta Arithmetica, 71, 1995, S. 381–389 (englisch).
  9. Hardy, Wright: An introduction to the theory of numbers. 2008, S. 135 (englisch; Theorem 132).
  10. Charles Babbage: Demonstration of a theorem relating to prime numbers. In: The Edinburgh philosophical journal 1. 1819, S. 46–49 (englisch; „n+1.n+2.n+3...“ bedeutet „(n+1)(n+2)(n+3)…“; die Umkehrung wird auch behauptet: „otherwise it is not“, aber nicht bewiesen und ist falsch für Quadrate von Wolstenholme-Primzahlen).