Maths Olympiad Prep

Library / /58 of 92

Number theory Difficulty 6.6 National olympiad Prove it Iran

Let nn be a natural number. Permutation a1,a2,,ana_1, a_2, \dots, a_n of numbers 1,2,,n1, 2, \dots, n is called square (cubic), if for each natural number 1in11 \le i \le n - 1, aiai+1+1a_i a_{i+1} + 1 is a perfect square (cube).

a) Prove that for infinitely many natural numbers nn there exists at least one square permutation of numbers 1,2,,n1, 2, \dots, n.

b) Prove that for no natural number nn there exists a cubic permutation of numbers 1,2,,n1, 2, \dots, n.

Solution

a) Let a1=2,a2=4,,ak=2k,ak+1=1,ak+2=3,,a2k=2k1a_1 = 2, a_2 = 4, \dots, a_k = 2k, a_{k+1} = 1, a_{k+2} = 3, \dots, a_{2k} = 2k-1. We can easily check that aiai+1+1a_i a_{i+1} + 1 is a perfect square for 1i2k1 \le i \le 2k except i=ki = k, which can be repaired if 2k+12k+1 is a perfect square which is possible for infinitely many values of kk.

b) Let a1,a2,,ana_1, a_2, \dots, a_n be a cubic permutation. Let 2k2^k be the largest power of 22 less than or equal to nn. By the definition of the cubic permutation we know that 2ku+1=x32^k u + 1 = x^3, where uu is an element of the permutation. So we have 2ku=(x1)(x2+x+1)2^k u = (x-1)(x^2+x+1). Hence we conclude that 2k(x1)2^k | (x-1). Because of the way that kk is chosen, we have n<2k+1n < 2^{k+1}. So we have 2kx1n23<223(k+1)2^k \le x-1 \le n^{\frac{2}{3}} < 2^{\frac{2}{3}(k+1)} which is a contradiction. Hence no such permutation exists.

Want a route through all this instead of an archive? The track puts 2,000 problems in a working order, from AMC 10 level to the IMO shortlist.

Source: MathNet, licensed CC-BY-4.0. Statement and solution reproduced as published; topic and difficulty added by this site.