Paul Leisher
who showed there are no solutions for n < 15,000, t
< 10,000, k < 1,000. Partial solutions were also received
from Nathan Pauli, Ray Kremer. Solutions were submitted by Ivan Lisac,
Alan Zimmermann, William Troxel, Osama Alkam, Emanuele Macrì, Lorenzo
Pozzoli, Massimo Brignone.
Note that n and n+1 differ by only 1, so can have no common prime factor. Let p1, p2,..., pk be the primes dividing n, and q1, q2,..., qj be the primes dividing n+1. Since each prime factor of t appears a multiple of k times, it follows that n is of the form rk and n+1 is of the form sk; here the primes p1, p2,..., pk are the prime divisors of r, while the primes q1, q2,..., qj are the prime divisors of s. But
| 1 | = (n + 1) - n = sk - rk |
| = (s - r)(sk-1 + sk-2r + sk-3r2 + sk-4r3 + ... + s2rk-3 + srk-2 + rk-1). |
Since both factors on the right-hand side are integers,
and the second one is clearly positive, each must equal 1.
But the second factor is clearly greater than 1 whenever k >
1.
You are visitor number 2963
to
this page.
Page last updated 16 November 1999.