ABOUT VERY PERFECT NUMBERS Mihály Bencze and Florin Popovici Department of Mathematics Áprily Lajos College Braşov, Romania Florentin Smarandache, Ph D Professor Chair of Department of Math & Sciences University of New Mexico 200 College Road Gallup, NM 87301, USA E-mail: smarand@unm.edu Abstract. In this short paper we prove that the square of an odd prime number cannot be a very perfect number. Introduction. A natural number n is called very perfect if σ (σ (n)) = 2n (see [1]), where σ ( x ) means the sum of all positive divisors of the natural number x. We now prove the following result: Theorem. The square of an odd prime number cannot be a very perfect number. Proof: Let’s consider n = p 2 , where p is an odd prime number, then σ (n ) = 1 + p + p 2 , σ (σ (n )) = σ (1 + p + p 2 )= 2 p 2 . We decompose σ (n ) in canonical form, from where 1 + p + p 2 = p1α1 p2α 2 ... pαk k . Because p ( p + 1) + 1 is odd, in the canonical decompose there must be only odd numbers. σ (σ (n) ) = (1 + p1 + ... + p1α ) ... (1 + pk + ... + pkα ) = k Because p1α1 +1 − 1 pkα k +1 − 1 ... = 2 p2 p1 − 1 pk − 1 pα k +1 − 1 p1α1 +1 − 1 > 2,..., k >2 p1 − 1 pk − 1 one obtains that 2 p 2 cannot be decomposed in more than two factors, then each one is > 2 , therefore k ≤ 2 . Case 1. For k = 1 we find σ (n) = 1 + p + p 2 = p1α1 , from where one obtains 1 p1α1 +1 = p1 (1 + p + p 2 ) and σ (σ (n)) = p1α1 +1 − 1 = 2 p 2 , p1 (1 + p + p 2 ) − 1 = 2 p 2 ( p1 − 1) , p1 − 1 p1 − 1 = p (pp1 − 2 p − p1 ) . The right side is divisible by p , thus p1 − 1 is a p multiple. Because p1 > 2 it results that 2 p1 ≥ p − 1 and p12 ≥ ( p + 1) > p 2 + p + 1 = p1α1 , from where thus α1 = 1 and σ (n) = p 2 + p + 1 = p1 , σ (σ (n)) = σ ( p1 ) = 1 + p1 . If n is very perfect then 1 + p1 = 2 p 2 or p 2 + p + 2 = 2 p 2 . The solutions of the equation are p = −1 , and p = 2 which is a contradiction. Case 2. For k = 2 we have σ (n ) = p 2 + p + 1 = p1α1 p2α 2 . p1α1 +1 − 1 p2α 2 +1 − 1 σ (σ (n) ) = (1 + p1 + ... + p1 ) (1 + p2 + ... + p2 ) = ⋅ = 2 p2 . p1 − 1 p2 − 1 Because pα 2 + 1 − 1 p1α1 +1 − 1 > 2 and 2 > 2, p1 − 1 p2 − 1 it results p2α 2 +1 − 1 p1α1 +1 = p and = 2p p1 − 1 p2 − 1 (or inverse), thus p1α1 +1 − 1 = p ( p1 − 1) , p2α 2 +1 − 1 = 2 p (p2 − 1), α2 α then thus p1α1 +1 p2α 2 +1 − p1α1 +1 − p2α 2 +1 + 1 = 2 p 2 (p1 − 1)(p2 − 1), σ (n) = p 2 + p + 1 = p1α +1 p2α 1 and or 2 +1 p1 p2 ( p 2 + p + 1) = 2 p 2 ( p1 − 1)( p2 − 1) + p1α1 +1 + p2α 2 +1 − 1 p1 p2 p ( p + 1) + p1 p2 − 1 = 2 p 2 ( p1 − 1)( p2 − 1) + ( p1α1 +1 − 1) + ( p2α 2 +1 − 1) = = 2 p 2 (p1 − 1)(p2 − 1) + p (p1 − 1) + 2 p (p2 − 1) accordingly p divides p1 p2 − 1 , thus p1 p2 > p + 1 and p12 p22 ≥ ( p + 1) > p 2 + p + 1 = p1α1 p2α 2 . 2 Hence: Π1 ) If α1 = 1 and n = 2 p 2 , 2 pα 2 + 1 − 1 p12 − 1 = p , and 2 = 2p , p2 − 1 p1 − 1 thus p1 + 1 = p which is a contradiction. Π2 ) If α 2 = 1 and n = 2 p 2 , then p α1 + 1 − 1 σ (n) = p 2 + p + 1 = p1α1 p2 and 1 = p , and p1 − 1 then σ (n) = p 2 + p + 1 = p1 p2α and 2 p22 − 1 = 2p, p2 − 1 thus p2 + 1 = 2 p , p2 = 2 p − 1 and σ (n) = p 2 + p + 1 = p1α (2 p + 1) , 1 4σ (n) = (2 p − 1)(2 p + 3) + 7 = 4 p1α1 (2 p − 1) , accordingly 7 is divisible by 2 p − 1 and thus p is divisible by 4 which is a contradiction. from where Reference: [1] Suryanarayama – Elemente der Mathematik – 1969. [Published in “Octogon”, Braşov, Vol. 5, No. 2, 53-4, October 1997.] 3