Finding Squares in a Product involving Squares
Thang Pang Ern · 2024
Read Finding Squares in a Product involving Squares by Thang Pang Ern (2024) 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-2411-00012.txt.
Description
Abstract . We wish to discuss positive integer solutions to the Diophantine equation ∏ k = 1 n ( k 2 + 1 ) = b 2 . \displaystyle\prod_{k=1}^{n}(k^{2}+1)=b^{2}. Some methods in analytic number theory will be used to tackle this problem. 1. Introduction We motivate our discussion with a simple geometry problem mentioned on Numberphile in 2014.
Catalog record
| CREATOR | Thang Pang Ern |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2411-00012
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Finding Squares in a Product involving Squares Thang Pang Ern Abstract . We wish to discuss positive integer solutions to the Diophantine equation ∏ k = 1 n ( k 2 + 1 ) = b 2 . \displaystyle\prod_{k=1}^{n}(k^{2}+1)=b^{2}. Some methods in analytic number theory will be used to tackle this problem. 1. Introduction We motivate our discussion with a simple geometry problem mentioned on Numberphile in 2014. Arrange three congruent squares ABFE , BCGF , and CDHG in the following manner (Figure 1 ). From vertex D , construct a line segment from it to the bottom-left vertex of each square — the vertices are namely E , F , and G . Let \angle DEF=\alpha , \angle DFG=\beta , and \angle DGH=\gamma . It can be proven using elementary geometry, that \alpha+\beta+\gamma=\pi/2 . \alpha \beta \gamma A B C D E F G H Figure 1. The three square geometry problem As \triangle DGH is isosceles, it follows that \gamma=\pi/4 . Thus, this identity is reduced to \alpha+\beta=\pi/4 . In contrast, using complex numbers makes the problem far simpler. In particular, using arguments, it is easy to show that (1) \displaystyle\arg\left((1+i)(2+i)(3+i)\right)=\arg(1+i)+\arg(2+i)+\arg(3+i)=\frac{\pi}{2}. Naturally, we can extend the problem to n squares. As such, which values of n , where n is a positive integer, yield the sum of angles being some rational multiple of \pi ? That is to say, does there exist p,q\in\mathbb{Z} , where q\neq 0 , such that (2) \displaystyle\sum_{k=1}^{n}\theta_{k}=\sum_{k=1}^{n}\operatorname{arctan}\left(1/k\right)=\frac{p}{q}\cdot\pi? If the sum of angles \theta_{1}+\ldots+\theta_{n} converges, then we have many cases to consider (other than the case when n=3 ). However, it turns out that the sum of angles diverges [ 1 ] . That is, (3) \displaystyle\lim_{n\rightarrow\infty}\sum_{k=1}^{n}{{{\arctan}\left(\frac{1}{k}\right)}}\quad\text{is a divergent series}. One can use the integral test to verify this. Moreover, attempts to tackle the problem by constructing a recurrence relation involving the addition formula of inverse tangent are futile as there are no nice properties of the following formula: (4) \displaystyle\arctan u+\arctan v=\arctan\left(\frac{u+v}{1-uv}\right) 2. The Number-Theoretic Connection This question can be regarded as a number theoretic one. Note that for 1\leq k\leq n , the modulus of the complex number k+i is \sqrt{k^{2}+1} , and every z\in\mathbb{C} can be expressed as z=a+bi , where a,b\in\mathbb{R} . If (5) \displaystyle\prod_{k=1}^{n}(k+i) is purely imaginary, then z=bi (i.e. a=0 ), and so |z|=b . Consider the product (6) \displaystyle\prod_{k=1}^{n}{\sqrt{k^{2}+1}} which is the product of the moduli of the complex numbers k+i for all 1\leq k\leq n . Hence, the problem is equivalent to finding all pairs of natural numbers \left(b,n\right) such that (7) \displaystyle\prod_{k=1}^{n}(k^{2}+1)=b^{2}. We wish to find all solutions to the Diophantine equation in (7), and we assert that (b,n)=(10,3) is the only solution. This would conclude that the product in (5) is purely imaginary if and only if n=3 . In this paper, we will discuss and expand on a proof given by Cilleruelo. It aims to make his proof more accessible by explaining the parts that were deemed trivial. Moreover, as shown in (79), we will obtain a better bound for our problem, reducing the time taken to seek all positive integer solutions to the Diophantine equation. We first discuss the following problem from the 17th China Western Mathematical Invitational Competition in 2017. It is from Problem 1 of Day 1 of the competition [ 3 ] . For any prime p , this result would help us find a bound for p in terms of n . Theorem 1 . Let p be a prime and n be a positive integer such that p^{2} divides (8) \displaystyle\prod_{k=1}^{n}(k^{2}+1). Then, p<2n . Proof. Define P_{n} to be the above product. Since p^{2}\mid P_{n} , then p\mid P_{n} . One case to consider is p^{2}\mid(k^{2}+1) for some 1\leq k\leq n . Then, p^{2}<n^{2}+1 and hence, p<\sqrt{n^{2}+1}<2n . Suppose otherwise, then there exists 1\leq a<b\leq n such that p\mid(a^{2}+1) and p\mid(b^{2}+1) , and so, taking the difference, p\mid(b+a)(b-a) . As a+b<2n and b-a>0 , b^{2}-a^{2}<2n . Thus, p is a divisor of some number less than 2n and the result follows. \Box Theorem 2 . For n>3 , (9) \displaystyle P_{n}=\prod_{k=1}^{n}(k^{2}+1)\quad\text{is non-square}. Suppose on the contrary that for n>3 , P_{n} is square. Then, we can write it as a product of some primes p , where p<2n , in the following form: (10) \displaystyle P_{n}=\prod_{p<2n}p^{\alpha_{p}},\text{ where }\alpha_{p}\in\mathbb{N} Note that (11) \displaystyle P_{n}=\prod_{k=1}^{n}(k^{2}+1)>\prod_{k=1}^{n}{k^{2}}=(n!)^{2}. Consider writing n! as (12) \displaystyle n!=\prod_{p\leq n}p^{\beta_{p}} since n! is a product of all the natural numbers from 1 to itself inclusive, and there are some numbers which are composite and hence, can be written as the product of primes. This expression is apt when we compare it with the expression for P_{n} in terms of \alpha_{p} claimed in (10). We have the following as we make use of the property that the logarithm of a product is the sum of that logarithm: (13) \displaystyle{{\left(\prod\limits_{p\leq n}{{{p}^{{{\beta}_{p}}}}}\right)}^{2}} \displaystyle<\prod\limits_{p<2n}{{{p}^{{{\alpha}_{p}}}}} (14) \displaystyle 2\log\left(\prod\limits_{p\leq n}{{{p}^{{{\beta}_{p}}}}}\right) \displaystyle<\log\left(\prod\limits_{p<2n}{{{p}^{{{\alpha}_{p}}}}}\right) (15) \displaystyle\sum\limits_{p\leq n}{{{\beta}_{p}}\log p} \displaystyle<\frac{1}{2}\sum\limits_{p<2n}{{{\alpha}_{p}}\log p} Corollary 3 . For n\geq 1 , \alpha_{2}=\lceil{n/2}\rceil . Proof. Since k^{2}\equiv 0\text{ or }1\text{ }(\operatorname{mod}4) , then k^{2}+1\equiv 1\text{ or }2\text{ }(\operatorname{mod}4) . The initial result can be easily derived by considering the cases where k is odd or even. Note that \alpha_{2} denotes the number of times P_{n} can be divided by 2. If k is odd, then k^{2}+1 is even, which is why for all odd k , P_{k} contributes an additional multiplicity of 2 (other than odd multiples) as compared to P_{k-1} . Hence, for odd k , consider the pair \left(P_{k},P_{k+1}\right) , where both their values of \alpha_{2} are the same. As there are n/2 such pairs, then \alpha_{2}=\lceil{n/2}\rceil . \Box Before we state and prove an important lemma (Lemma 6 ), we define the Legendre symbol and state a major theorem known as the law of quadratic reciprocity. Definition 4 (Legendre symbol) . Let p be an odd prime. An integer a is a quadratic residue modulo p if it is congruent to a perfect square modulo p and is a quadratic non-residue modulo p otherwise. The Legendre symbol is a function of a and p defined as \left(\frac{a}{p}\right)=\left\{\begin{array}[]{*{35}{l}}1&\text{if }a\text{ is a quadratic residue modulo }p\text{ and }a\not\equiv 0\text{ }\left(\operatorname{mod}p\right),\\ -1&\text{if }a\text{ is a non-quadratic residue modulo }p,\\ 0&\text{if }a\equiv 0\text{ }\left(\operatorname{mod}p\right).\\ \end{array}\right. Theorem 5 (law of quadratic reciprocity) . For distinct odd primes p and q , (16) \displaystyle\left({\frac{p}{q}}\right)\left({\frac{q}{p}}\right)=(-1)^{{\frac{p-1}{2}}\cdot{\frac{q-1}{2}}}. Lemma 6 . Let p be an odd prime. If p\text{ }|\text{ }(k^{2}+1) for 1\leq k\leq n , then p\equiv 1\text{ }(\operatorname{mod}4) . Proof. As k^{2}\equiv-1\text{ }(\operatorname{mod}p) , -1 is a quadratic residue modulo p . Here, we set a=-1 and so \left(\frac{-1}{p}\right)=1. By the law of quadratic reciprocity (Theorem 5 ), p\equiv 1\text{ }(\operatorname{mod}4) . \Box Lemma 7 . For any prime p , x^{2}\equiv 0\text{ }(\operatorname{mod}p) has the unique solution x\equiv 0\text{ }(\operatorname{mod}p) . Proof. The only zero divisor in the ring \mathbb{Z}/p\mathbb{Z} is 0. As such, if a product is 0, one of the factors must be 0 as well, and the result follows. \Box Theorem 8 . For any prime p , each interval of length p^{j} contains two solutions to the congruence x^{2}\equiv 1\text{ }\left(\operatorname{mod}p^{j}\right) [ 4 ] . Proof. We proceed by induction.