Random Diophantine Equation

Find the number of positive integers n n such that there exists an integer m n 2 m \neq \frac{n}{2} such that n 3 2 m 3 6 m \frac{n^3-2m^3}{6m} is a perfect square.


Details and Assumptions:

  • All perfect squares are integers. For example, 8 2 , 4 2 , 8^2, 4^2, etc. are perfect squares, but ( 2 3 ) 2 \left( \frac{2}{3} \right) ^2 isn't.


The answer is 0.

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

Let n 3 2 m 3 6 m = x 2 \dfrac{n^3-2m^3}{6m} = x^2 . This implies n 3 2 m 3 = 6 m x 2 n 3 = 2 m 3 + 6 m x 2 = ( m + x ) 3 + ( m x ) 3 . n^3 - 2m^3 = 6m x^2 \implies n^3 = 2m^3 + 6mx^2= (m+x)^3 + (m-x)^3. By Fermats' Last Theorem, this has no integer solutions unless m = x m = x . However, m = x m = x implies n 3 = 8 m 3 n = 2 m n^3 = 8m^3 \implies n = 2m , contradicting the fact that m n 2 m \neq \dfrac{n}{2} . Hence, the answer is 0 \boxed{0} .

Nice use of Fermat's last theorem.

Note that FLT usually refers to Fermat's Little Theorem instead.

Calvin Lin Staff - 6 years, 4 months ago

Log in to reply

Thanks, sir! I've edited my solution now.

Sreejato Bhattacharya - 6 years, 4 months ago

Very nice! Fermat's last theorem is one of the most interesting theorems of old, in my opinion. I could prove it but it would be too large to fit in the margin of this comment.

Caleb Townsend - 6 years, 4 months ago

Log in to reply

I see what you did there! :D

Prasun Biswas - 6 years, 4 months ago

We can solve this question by seeing the parity of the numerator and the denominator. k 3 k^3 will always be odd because k k is odd and 2 m 3 2m^3 will always be even. If we subtract an even number from an odd number, we will always get an odd number. So, our numerator will always be an odd number.

But our denominator would always be an even number because it is a factor of 6 6 . So, their division will not result in a whole number and consequently will not be a perfect square.

Yash Singhal - 6 years, 4 months ago

Come back quickly , I've heard a lot about you and want to see you in action. :)

A Former Brilliant Member - 6 years, 2 months ago
Patrick Corn
Feb 12, 2015

Using Sreejato's notation below, if we let X = 6 n / m X = 6n/m , Y = 36 x / m Y = 36x/m , we get Y 2 = X 3 432 Y^2 = X^3-432 . This is an elliptic curve with rank 0 0 and torsion subgroup of order 3 3 . So the only rational points are the point at infinity and ( 12 , ± 36 ) (12,\pm 36) , which leads to x = ± m , n = 2 m x = \pm m, n = 2m as the only possible solutions.

(Obviously, the facts about the elliptic curve require some work to prove!)

Let's have a "who does the most overkill" contest. :)

Sreejato Bhattacharya - 6 years, 3 months ago
Figel Ilham
Feb 12, 2015

Let n 3 2 m 3 6 m = p 2 \frac{n^3-2m^3}{6m}=p^2 . It implies that:

n 3 2 m 3 = 6 m p 2 n 3 = 6 m p 2 + 2 m 3 n^3-2m^3=6mp^2 \Rightarrow n^3=6mp^2+2m^3

The RHS must be even for all integers ( m , p ) (m,p) . Since the RHS even, the LHS must equal to the value of RHS and even. In other words, all positive integers for n n must be even. Thus, n 3 n^3 must be even too.

If n n is odd, then n 3 n^3 must be odd and the RHS must be odd and equal to the LHS. Since the odd number is not divisible by 2, then it contradicts both sides for the RHS is divisible by 2. Thus, there are no solutions for all odd positive integers n n

The problem statement has been edited now to make sure the parity argument doesn't work. :)

Sreejato Bhattacharya - 6 years, 4 months ago

i forgot the n m 2 n\neq \frac{m}{2}

Figel Ilham - 6 years, 4 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...