A NOTE ON THREE CONSECUTIVE POWERFUL NUMBERS

Tsz Ho Chan · 2025

Read A NOTE ON THREE CONSECUTIVE POWERFUL NUMBERS by Tsz Ho Chan (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-21485.txt.

Introduction

Tsz Ho Chan presents a 2025 number theory note on powerful numbers, whole numbers whose every prime factor appears at least squared. Runs of such numbers are rare curiosities, and whether three can stand consecutively has drawn attention. Chan attacks one structured case of the problem.

He uses Pell equations, elliptic curves, and second-order recurrences. With them, he proves no such triplet exists in the case he studies. There, the middle number is a perfect cube. Each neighbor is a single prime raised to an odd power. The note is short and purely theoretical. Results of this kind map the fine structure of consecutive special numbers, a corner of arithmetic where simple questions hide deep machinery.

Catalog record

CREATORTsz Ho Chan
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE12 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-21485

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A NOTE ON THREE CONSECUTIVE POWERFUL NUMBERS Tsz Ho Chan Abstract Received: , Revised: , Accepted: , Published: Abstract This note concerns the non-existence of three consecutive powerful numbers. We use Pell equations, elliptic curves, and second-order recurrences to show that there are no such triplets with the middle term a perfect cube and each of the other two having only a single prime factor raised to an odd power. 1 Introduction and Main Results A positive integer n is powerful or squareful if p^{2}\mid n for all primes p such that p\mid n . It is well-known that any powerful number can be factored uniquely as n=a^{2}b^{3} for some positive integer a and squarefree number b . Here a number n is squarefree if p^{2}\nmid n for all primes p . The following is an initial list of powerful numbers: 1,\,4,\,8,\,9,\,16,\,25,\,27,\,32,\,36,\,49,\,64,\,72,\,\ldots. Notice that 8 and 9 are consecutive powerful numbers. Indeed, there are infinitely many such pairs. For example, the solutions of the Pell equation x^{2}-2y^{2}=1 give consecutive powerful pairs 2y^{2},x^{2} as y is even by considering perfect squares modulo 4 . The interested readers may consult [ 5 ] and [ 8 ] for more discussions on this topic. Next, one can ask if there are three consecutive powerful numbers. Conjecture 1 (Erdős, Mollin, Walsh) . No three consecutive powerful numbers exist. This appears to be very hard. Some relevant references are [ 4 ] , [ 7 ] and [ 6 ] . Conditional on the abc -conjecture, one can show that there are only finitely many triples of consecutive powerful numbers. In this note, we consider the special case of three consecutive powerful numbers of the form x^{3}-1 , x^{3} , x^{3}+1 and prove the following. Theorem 1 . There are no three consecutive powerful numbers of the form x^{3}-1=p^{3}y^{2},\;\;x^{3},\;\;x^{3}+1=q^{3}z^{2} (1) with primes p,q and positive integers x,y,z . Corollary 1 . The Diophantine equation 64x^{6}-1=p^{3}q^{3}y^{2} has no solution with integers x,y and primes p,q . The author hopes that this note will stimulate further studies of powerful numbers and Conjecture 1 . Throughout this paper, all variables are integers. The letters p and q stand for prime numbers. The symbol a\mid b means that a divides b , and a\nmid b means that a does not divide b . The symbol p^{k}\,\|\,a means that p^{k}\mid a but p^{k+1}\nmid a . 2 Some Basic Observations Lemma 1 . The difference between any two perfect squares cannot be 2 . Proof. One simply observes that (n+1)^{2}-n^{2}=2n+1>2 when n\geq 1 , and (0+1)^{2}-0^{2}=1\neq 2 . ∎ Lemma 2 . If x\equiv 1\pmod{3} , then 3\,\|\,x^{2}+x+1 . Proof. Suppose x\equiv 1\pmod{3} . Say x=3x^{\prime}+1 for some integer x^{\prime} . Then x^{2}+x+1=(3x^{\prime}+1)^{2}+(3x^{\prime}+1)+1=9x^{\prime 2}+9x^{\prime}+3=3(3x^{\prime 2}+3x^{\prime}+1) which is divisible by 3 but not 9 . ∎ Lemma 3 . If x\equiv 2\pmod{3} , then 3\,\|\,x^{2}-x+1 . Proof. This follows from Lemma 2 by the substitution x\mapsto-x . ∎ Lemma 4 . Suppose a\neq 0 , c and e are some fixed integers. If y^{2}=ax^{4}+cx^{2}+e has an integer solution (x,y) with x\neq 0 , so does the elliptic curve Y^{2}=X^{3}+cX^{2}+aeX . Proof. One simply multiplies both sides of y^{2}=ax^{4}+cx^{2}+e by a^{2}x^{2} and gets (axy)^{2}=(ax^{2})^{3}+c(ax^{2})^{2}+ae(ax^{2}) . This yields a non-zero integer solution for the above elliptic curve with X=ax^{2}\neq 0 and Y=axy . ∎ Lemma 5 . The Pell equation x^{2}-3y^{2}=1 has all positive integer solutions (x_{k},y_{k}) generated by x_{k}+y_{k}\sqrt{3}=(2+1\cdot\sqrt{3})^{k} for k\in\mathbb{N} . Moreover, the solutions satisfy the recursions: x_{1}=2 , x_{2}=7 , x_{k}=4x_{k-1}-x_{k-2} ; y_{1}=1 , y_{2}=4 , y_{k}=4y_{k-1}-y_{k-2} for k>2 . Proof. This is a standard result in the theory of Pell equation that all integer solutions are generated by some fundamental (minimal) solution. See [ 1 , Theorem 5.3] for example. The recursions easily follow from the observation (2+\sqrt{3})^{k}-4(2+\sqrt{3})^{k-1}+(2+\sqrt{3})^{k-2}=(2+\sqrt{3})^{k-2}[(2+\sqrt{3})^{2}-4(2+\sqrt{3})+1]=0 as 2+\sqrt{3} is a root to the quadratic equation x^{2}-4x+1=0 . ∎ Lemma 6 . The generalized Pell equation x^{2}-3y^{2}=-2 has all positive integer solutions (x_{k},y_{k}) generated by x_{k}+y_{k}\sqrt{3}=(1+1\cdot\sqrt{3})(2+1\cdot\sqrt{3})^{k} for k\in\mathbb{N} . Moreover, the solutions satisfy the recursions: x_{1}=1 , x_{2}=5 , x_{k}=4x_{k-1}-x_{k-2} ; y_{1}=1 , y_{2}=3 , y_{k}=4y_{k-1}-y_{k-2} for k>2 . Proof. One can easily see that x=1=y is the smallest positive integer solution (i.e., x_{1}+y_{1}\sqrt{3}=1+1\cdot\sqrt{3} ). Then one can generate all the integer solutions by combining this with the solutions in Lemma 5 . See [ 2 , Theorem 3.3] for example. The recursions follow from a similar observation as in Lemma 5 . ∎ 3 Proof of Theorem 1 We are now ready to prove Theorem 1 . Proof of Theorem 1 . First, note that any three consecutive powerful numbers must be of the form 4n-1 , 4n , 4n+1 as 2\,\|\,4n+2 . Recall that the middle number is a cube. So, 4n=x^{3} and x is even. Suppose that, contrary of Theorem 1 , we have (x-1)(x^{2}+x+1)=p^{3}y^{2}\;\text{ and }\;(x+1)(x^{2}-x+1)=q^{3}z^{2}. (2) Note that x>1 as p,y>0 . By the fact that \gcd(a,b)=\gcd(a,b-a) , we have \gcd(x-1,x^{2}+x+1)=\gcd(x-1,(x-1)(x+2)+3)=\gcd(x-1,3)=1 or 3 . Hence, \gcd(x-1,x^{2}+x+1)=\left\{\begin{array}[]{ll}3,&\text{ if }x\equiv 1\pmod{3},\\ 1,&\text{ otherwise}.\end{array}\right. (3) Suppose x\not\equiv 1\pmod{3} . Then \gcd(x-1,x^{2}+x+1)=1 by Equation ( 3 ). It follows that if p\mid x-1 , then p^{3}\mid x-1 and x^{2}+x+1 is a perfect square by Equation ( 2 ). This is impossible as x^{2}<x^{2}+x+1<(x+1)^{2} . Hence, p^{3}\mid x^{2}+x+1 . In summary, we have x\not\equiv 1\pmod{3}\;\text{ implies }\;p^{3}\mid x^{2}+x+1\text{ and }x-1\text{ is a perfect square}. (4) Applying the substitution x\mapsto-x to the above argument, we also have \gcd(x+1,x^{2}-x+1)=\left\{\begin{array}[]{ll}3,&\text{ if }x\equiv 2\pmod{3},\\ 1,&\text{ otherwise},\end{array}\right. (5) and x\not\equiv 2\pmod{3}\;\text{ implies }\;q^{3}\mid x^{2}-x+1\text{ and }x+1\text{ is a perfect square}. (6) Case 1: x\equiv 0\pmod{3} . From Equations ( 3 ) and ( 5 ), we have \gcd(x-1,x^{2}+x+1)=1=\gcd(x+1,x^{2}-x+1). By Equations ( 4 ) and ( 6 ), both x-1 and x+1 are perfect squares which contradicts Lemma 1 . Case 2: x\equiv 1\pmod{3} . From Equations ( 3 ) and ( 5 ), we have \gcd(x-1,x^{2}+x+1)=3\;\text{ and }\;\gcd(x+1,x^{2}-x+1)=1. Hence, q^{3}\mid x^{2}-x+1 and x+1=u^{2} for some integer u>0 by Equation ( 6 ). Suppose p=3 . We have (x-1)(x^{2}+x+1)=3^{3}y^{2} and 3\,\|\,x^{2}+x+1 by Lemma 2 . Thus, 9\mid x-1 and x-1=9x^{\prime 2}=(3x^{\prime})^{2} for some integer x^{\prime} . This contradicts Lemma 1 . Suppose p\neq 3 . If p\mid x^{2}+x+1 , then p^{3}\mid x^{2}+x+1 by Equation ( 3 ). This forces x-1=3v^{2} or x-1=(3v)^{2} for some integer v . The latter contradicts Lemma 1 . The former yields u^{2}-3v^{2}=2 which is also impossible as u^{2}\equiv 0 or 1\pmod{3} . Therefore, we must have p\mid x-1 and, hence, p^{3}\mid x-1 by Equation ( 3 ). From Equation ( 2 ) and Lemma 3 , we have x^{2}+x+1=3v^{2} and x-1=3p^{3}w^{2} for some integers v,w>0 with \gcd(v,3)=1=\gcd(v,w) . Substituting x=u^{2}-1 into x^{2}+x+1 , we obtain 3v^{2}=u^{4}-u^{2}+1\;\;\text{ or }\;\;y^{2}=3u^{4}-3u^{2}+3 where y=3v . By Lemma 4 , the elliptic curve Y^{2}=X^{3}-3X^{2}+9X has a non-zero integer solution. This contradicts (X,Y)=(0,0) being the only integer solution by the SageMath command \text{E = EllipticCurve([0,-3,0,9,0]); E.integral\_points()}, for example. Case 3: x\equiv 2\pmod{3} . From Equations ( 3 ) and ( 5 ), we have \gcd(x-1,x^{2}+x+1)=1,\;\text{ and }\;\gcd(x+1,x^{2}-x+1)=3. Hence, p^{3}\mid x^{2}+x+1 and x-1=u^{2} for some integer u>0 with 3\nmid u by Equation ( 4 ). Suppose q=3 . We have (x+1)(x^{2}-x+1)=3^{3}y^{2} and 3\,\|\,x^{2}-x+1 by Lemma 3 . Thus, 9\mid x+1 and x+1=9x^{\prime 2}=(3x^{\prime})^{2} for some integer x^{\prime}

Read the complete text