Exactly one square

Number theory
Post Reply
Demetres
Former Team Member
Former Team Member
Posts: 77
Joined: Mon Nov 09, 2015 11:52 am
Location: Limassol/Pyla Cyprus
Contact:

Exactly one square

#1

Post by Demetres »

Let \(x\) and \(y\) be positive integers such that \(x+xy\) and \(y+xy\) are both perfect squares.
(a) Show that exactly one of \(x\) and \(y\) is a perfect square.
(b) Can you characterise all such pairs?


Edit after the solution: This was problem 10213 from the American Mathematical Monthly proposed by P. G. Walsh from the University of Waterloo. The solution appeared in the issue of February 1995.
Vangelis Mouroukos
Former Team Member
Former Team Member
Posts: 4
Joined: Wed Nov 11, 2015 6:18 am

Re: Exactly one square

#2

Post by Vangelis Mouroukos »

First of all, we note that \(x\) and \(y\) cannot be both squares. Indeed, suppose that \(x\) is a square. Since \(xy+x=x(y+1) \) is a square, \(y+1\) must also be a square. But then (since \(y >0\) ) \(y \) cannot be a square.

Suppose now that \(x\) and \(y\) are positive integers such that both \( x(y+1) \) and \( y(x+1) \) are squares. We can write \( x = au^2\) and \(y= bv^2\) for some positive integers \(a, b, u, v\), where \(a\) and \(b\) are square-free. We need to show that \(a =1 \) or \(b =1 \).

Since \(au^2(bv^2 +1) \) and \(bv^2(au^2 +1) \) are perfect squares, there exist positive integers \(s \) and \(t \) such that \(a\left( {b{v^2} + 1} \right) = {\left( {as} \right)^2}\) and \(b\left( {a{u^2} + 1} \right) = {\left( {bt} \right)^2}\). It follows that \(\boxed{a{s^2} - b{v^2} = 1}\) and \(\boxed{a{u^2} - b{t^2} = - 1}\). Note that \(\gcd \left( {a,b} \right) = 1\).

Let \(z = st - uv\). Then, we have:

\(\left( {b{v^2}} \right)\left( {a{u^2}} \right) = ab{\left( {st - z} \right)^2} \Leftrightarrow \left( {a{s^2} - 1} \right)\left( {b{t^2} - 1} \right) = ab{\left( {st - z} \right)^2} \Leftrightarrow \)

\( \Leftrightarrow ab{s^2}{t^2} - a{s^2} - b{t^2} + 1 = ab{s^2}{t^2} - 2abstz + ab{z^2} \Leftrightarrow \)

\( \Leftrightarrow a{s^2} + b{t^2} + ab{z^2} - 1 = 2abstz\) \( \bf \color{red} {(1)} \).

If we set \(m = as^2 \), \( n = bt^2 \) and \(p = abz^2 \), then \( \bf \color{red} {(1)} \) is equivalent to
\[\boxed{m + n + p - 1 = 2\sqrt {mnp} } \bf \color{red} {(\bigstar)}.\]
Claim : At least one of the numbers \(m, n \) and \( p\) is a perfect square.

It follows from the Claim that at least one of the numbers \(a, b\) and \(ab\) is a square. But \(a\) and \(b\) are square-free and \(\gcd \left( {a,b} \right) = 1\), so \(a =1 \) or \( b = 1 \), answering the first part of the problem.

It remains to prove the Claim. Rewrite equation \(\bf \color{red} {(\bigstar)}\) as

\({\left( {m + n + p - 1} \right)^2} = 4mnp \bf \color {red} {(2)}\).

Let \(\alpha = 2m - 1\), \( \beta = 2n -1 \) and \( \gamma = 2p-1\). Then, equation \(\bf \color {red} {(2)}\) is equivalent to

\({\left( {\alpha + \beta + \gamma + 1} \right)^2} = 2\left( {\alpha + 1} \right)\left( {\beta + 1} \right)\left( {\gamma + 1} \right) \Leftrightarrow \)

\( \Leftrightarrow {\alpha ^2} + {\beta ^2} + {\gamma ^2} - 2\alpha \beta \gamma = 1 \bf \color {red} {(3)}\).

Now equation \(\bf \color {red} {(3)}\) is equivalent to each of the equations

\(\left( {{\alpha ^2} - 1} \right)\left( {{\beta ^2} - 1} \right) = {\left( {\alpha \beta - \gamma } \right)^2}\)

and

\(\left( {{\beta ^2} - 1} \right)\left( {{\gamma ^2} - 1} \right) = {\left( {\beta \gamma - \alpha } \right)^2}\),

from which we infer that there exist non-negative integers \(d, e, f, g\), with \( d \) not a perfect square, such that

\({\alpha ^2} - 1 = d{e^2}\), \({\beta ^2} - 1 = d{f^2}\), \({\gamma ^2} - 1 = d{g^2}\),

\(\left| {\alpha \beta - \gamma } \right| = def\) and \(\left| {\beta \gamma - \alpha } \right| = dfg\).

Now we invoke the theory of the Pell equation. Let \(\left( {{x_1},{y_1}} \right)\) be the fundamental solution of the Pell equation \({X^2} - d{Y^2} = 1\). If we set \({z_1} = {x_1} + {y_1}\sqrt d \), then it is known that all its solutions \(\left( {{x_k},{y_k}} \right)\) (where \( k \ge 0\) ) are given by
\[{x_k} = \frac{1}{2}\left( {z_1^k + z_1^{ - k}} \right), \mbox{ } {y_k} = \frac{1}{{2\sqrt d }}\left( {z_1^k - z_1^{ - k}} \right).\]
It follows that there exist non-negative integers \(k_1, k_2\) and \(k_3\), such that
\[\alpha = \frac{1}{2}\left( {z_1^{{k_1}} + z_1^{ - {k_1}}} \right),\beta = \frac{1}{2}\left( {z_1^{{k_2}} + z_1^{ - {k_2}}} \right),\gamma = \frac{1}{2}\left( {z_1^{{k_3}} + z_1^{ - {k_3}}} \right).\]
Assume, without loss of generality, that \(m \ge n \ge p\). Then, we have \(k_1 \ge k_2 \ge k_3 \) and \(\alpha \beta - \gamma = def\). Hence,
\[\gamma = \alpha \beta - def =\]
\[ =\frac{1}{4}\left( {z_1^{{k_1}} + z_1^{ - {k_1}}} \right)\left( {z_1^{{k_2}} + z_1^{ - {k_2}}} \right) - \frac{1}{4}\left( {z_1^{{k_1}} - z_1^{ - {k_1}}} \right)\left( {z_1^{{k_2}} - z_1^{ - {k_2}}} \right) = \]
\[ = \frac{1}{2}\left( {z_1^{{k_1} - {k_2}} + z_1^{ - \left( {{k_1} - {k_2}} \right)}} \right).\]
We conclude that \(k_3 = k_1 - k_2\), so not all of the numbers \(k_1, k_2\) and \(k_3\) are odd. Suppose that \(k_1\) is even. Then, we can write
\[m = \frac{{\alpha + 1}}{2} = {\left[ {\frac{1}{2}\left( {z_1^{\frac{{{k_1}}}{2}} + z_1^{ - \frac{{{k_1}}}{2}}} \right)} \right]^2},\]
so \(m\) is a perfect square and the Claim follows.

To answer the second question. i.e. to characterize all pairs \( (x,y) \) of positive integers with the property that both \(x(y+1) \) and \( y(x+1) \) are squares, we may proceed as follows: Without loss of generality, we may assume that \( x \) is a square. Then, \(y+1 \) will also be a square, so there exist positive integers \( r\) and \(s\) such that \(x = r^2 \) and \(y+1 = s^2\). We also have that \(y(x+1) = (r^2 +1)(s^2 -1) \) is a square. Let \(d = \gcd \left( {{r^2} + 1,{s^2} - 1} \right)\). Then, there exist positive integers \(t\) and \(u\) such that

\({r^2} + 1 = d{t^2} \Longleftrightarrow r^2 - dt^2 =-1\)

and

\({s^2} - 1 = d{u^2} \Longleftrightarrow s^2-du^2 =1\).

Thus, the pair \((r,t) \) of positive integers is a solution of the negative Pell equation \(X^2 - dY^2 = -1 \). Depending on \(d\), this has either no solutions or infinitely many solutions.

According to the theory of the Pell equation, if \(d \) is such that the above negative Pell equation is solvable in the positive integers, then we can find positive integers \(r,s,t \) and \(u\) such that \({r^2} + 1 = d{t^2}\) and \({s^2} - 1 = d{u^2} \). Then the pair \[\left( {x,y} \right) = \left( {{r^2},d{u^2}} \right)\] has the desired property.

One could possibly characterize (via continued fractions) all \(d\) such that the negative Pell equation \(X^2 - dY^2 = -1 \) has (infinitely many) solutions and therefore obtain a more precise answer to the second part of the problem.
Demetres
Former Team Member
Former Team Member
Posts: 77
Joined: Mon Nov 09, 2015 11:52 am
Location: Limassol/Pyla Cyprus
Contact:

Re: Exactly one square

#3

Post by Demetres »

This was problem 10213 from the Monthly. I will update my original post with the details. In the proposed solution by Robin Chapman it shows that the negative Pell equation $X^2 - abY^2 = -1$. has a solution. For example, with Vangelis' notation $(aus+bvt,uv+st)$ is such a solution. Since now $(2x+1)^2 - ab(2cd)^2 = 1$ it follows by the theory of Pell's equation that $2x+1 + 2cd\sqrt{ab} = (p+q\sqrt{ab})^2$ for some $p, q$ with $p^2 - abq^2 = \pm 1.$. With the choice of $-1$ this gives $x=p^2$ and with the choice of $+1$ this gives $x+1=p^2$ in which case $y$ is a perfect square.
Post Reply

Create an account or sign in to join the discussion

You need to be a member in order to post a reply

Create an account

Not a member? register to join our community
Members can start their own topics & subscribe to topics
It’s free and only takes a minute

Register

Sign in

Who is online

Users browsing this forum: No registered users and 3 guests