Can you find them all?

m 2 n n 2 + m \large \dfrac{m^2-n}{n^2+m}

Find the number of all positive integers pairs of ( m , n ) (m,n) for which the above expression will be an integer.

4 2 5 0 1 Infinitely many

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

3 solutions

Chew-Seong Cheong
Feb 21, 2017

For m 2 n n 2 + m \dfrac {m^2-n}{n^2+m} to be an integer,

m 2 n n 2 + m for m , n > 0 and m 2 n 0 m 2 n n 2 + m m 2 m n 2 n 0 A quadratic equation of m m = 1 ± 1 + 4 n 2 + 4 n 2 = 1 ± ( 2 n + 1 ) 2 m = n + 1 for m > 0 \begin{aligned} |m^2-n| & \ge n^2 + m & \small \color{#3D99F6} \text{for } m, n > 0 \text{ and } m^2 - n \ge 0 \\ m^2 - n & \ge n^2 + m \\ m^2 - m - n^2 - n & \ge 0 & \small \color{#3D99F6} \text{A quadratic equation of }m \\ \implies m & = \frac {1 \pm \sqrt{1+4n^2+4n}}2 \\ & = \frac {1 \pm (2n+1)}2 \\ \implies m & = n+1 & \small \color{#3D99F6} \text{for } m>0 \end{aligned}

Putting m = n + 1 m=n+1 , then we have m 2 n n 2 + m = \dfrac {m^2-n}{n^2+m} = ( n + 1 ) 2 n n 2 + n + 1 = \dfrac {(n+1)^2-n}{n^2+n+1} = n 2 + 2 n + 1 n n 2 + n + 1 = \dfrac {n^2+2n+1-n}{n^2+n+1} = n 2 + n + 1 n 2 + n + 1 = 1 \dfrac {n^2+n+1}{n^2+n+1} = 1 . Therefore, ( n + 1 , n ) (n+1,n) is a solution for any n n and there are infinitely many .

Jon Haussmann
Mar 12, 2017

Given a positive integer m m , let n = m 2 n = m^2 . Then m 2 n n 2 + m = 0. \frac{m^2 - n}{n^2 + m} = 0. Thus, there are infinitely many pairs ( m , n ) (m,n) .

Yeah did the same way

I Gede Arya Raditya Parameswara - 4 years, 2 months ago
Kushal Bose
Feb 19, 2017

Put n = 1 n=1 in the given expression we will get

m 2 1 1 + m = ( m + 1 ) ( m 1 ) m + 1 = m 1 \dfrac{m^2-1}{1+m}=\dfrac{(m+1)(m-1)}{m+1}=m-1

Here m m can be ant positive integer.SO, there are infinitely many solutions

In general we can also have m = n + 1 m = n + 1 , in which case both the numerator and denominator simplify to n 2 + n + 1 n^{2} + n + 1 .

I wonder how many solutions there are with both n > 1 n \gt 1 and m n > 1 m - n \gt 1 . Just a quick check gives ( m , n ) = ( 10 , 2 ) (m,n) = (10,2) as one such solution.

Brian Charlesworth - 4 years, 3 months ago

Log in to reply

Yes but I also can not find all solutions where n > 1 n>1

Kushal Bose - 4 years, 3 months ago

Log in to reply

Suppose we have n = 4 n = 4 , and let m 2 4 m + 16 = k \dfrac{m^{2} - 4}{m + 16} = k .

Then after some algebra we require that m = k + k 2 + 64 k + 16 2 m = \dfrac{k + \sqrt{k^{2} + 64k + 16}}{2} .

For m m to be an integer, we then require that k 2 + 64 k + 16 = d 2 k^{2} + 64k + 16 = d^{2} , or

( ( k + 32 ) d ) ( ( k + 32 ) + d ) = 1008 ((k + 32) - d)((k + 32) + d) = 1008 .

Then by looking at the factorings of 1008 1008 we can determine the possible integer values for k k , and thus for m m . Solutions that pop out for ( m , n ) (m,n) are then ( 20 , 4 ) , ( 47 , 4 ) , ( 110 , 4 ) (20,4), (47,4), (110,4) and ( 236 , 4 ) (236,4) , among others. So if we can follow the same process for all n > 1 n \gt 1 and k 1 k \ge 1 and always find solutions we would be able to conclude that there are an infinite number of solutions with n > 1 n \gt 1 . I still have to prove this, though.

Brian Charlesworth - 4 years, 3 months ago

Log in to reply

@Brian Charlesworth Nice observation

Thanks a lot sir

Kushal Bose - 4 years, 3 months ago

@Brian Charlesworth If we are given n n , an easier approach is that we want n 2 + m n 4 n n^2 + m \mid n^4 - n .

For n = 4 n = 4 , this gives us 16 + m 252 16 + m \mid 252 . Then, by subtracting 16 from the valid factors of 252, the corresponding values of m m are 2, 5, 12, 20, 26, 47, 68, 110, 236,


The first line follows by the standard "remainder-factor theorem" / "partial fraction" approach, where we have m 2 n m + n 2 = m n 2 + n 4 n m + n 2 \frac{ m^2 - n } { m + n^2 } = m - n^2 + \frac{ n^4 - n } { m+ n^2 } .

Calvin Lin Staff - 4 years, 3 months ago

Log in to reply

@Calvin Lin Now we can vary m m and find the values of n n

Kushal Bose - 4 years, 3 months ago

Log in to reply

@Kushal Bose If m m is fixed and n n is allowed to vary, then I would take a different approach.

Namely, determine the solution set of x 2 + m > x m 2 0 |x^2 + m | > |x - m^2 | \neq 0 and then we know that n n has to lie in the complement.

Calvin Lin Staff - 4 years, 3 months ago

Log in to reply

@Calvin Lin Then how to solve the second part of your expression ??

Kushal Bose - 4 years, 3 months ago

Log in to reply

Log in to reply

@Calvin Lin Yeah got it

Kushal Bose - 4 years, 3 months ago

Nice way of solving it.

Hana Wehbi - 4 years, 3 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...