Find all positive integers n such that 3x^2-y^2=2018^n has an integer solution

Click For Summary

Homework Help Overview

The discussion revolves around finding all positive integers n such that the equation 3x² - y² = 2018ⁿ has integer solutions. The problem involves exploring properties of the equation under different conditions for n, particularly focusing on even and odd values.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • Participants discuss the implications of n being even or odd, considering modular arithmetic to eliminate impossible cases. There are attempts to relate the general case to simpler instances, particularly n = 1, and to explore the existence of solutions based on specific values of x and y.

Discussion Status

The discussion has progressed with various approaches being explored, including modular considerations and specific examples of solutions. Some participants have provided guidance on how to construct solutions based on known cases, while others have raised questions about the completeness of the solutions presented.

Contextual Notes

There is an ongoing examination of assumptions regarding the parity of n and its impact on the existence of solutions. Participants have noted the need to clarify whether all solutions have been identified or if there are additional solutions beyond those discussed.

littlemathquark
Messages
204
Reaction score
26
Homework Statement
Find all positive integers ##n## such that ##3x^2-y^2=2018^n## has integer solutions.
Relevant Equations
None
I need an idea. Thank you.
 
Physics news on Phys.org
If ##n## is even consider the equation modulo ##3.##

The hard part is to show that odd ##n## actually do have solutions.
 
##3x^2-y^2\equiv 2018^n\mod 3## $$-y^2\equiv 2^n\equiv 1\mod 3$$ $$y^2\equiv 2\mod 3$$ but it is impossible so ##n## must be odd.
 
littlemathquark said:
##3x^2-y^2\equiv 2018^n\mod 3## $$y^2\equiv 2^n\equiv 1\mod 3$$
This is wrong. Use the assumption that ##n## is even.

(For odd ##n## show that ##3x^2-x^2\equiv 2018^n## has always a solution if ##3x^2-x^2\equiv 2018## has a solution. So you will be left with the case ##n=1.##)
 
I mean if n is even ##y^2\equiv 2\mod3## but it is impossible so n must be odd.
 
littlemathquark said:
I mean if n is even ##y^2\equiv 2\mod3## but it is impossible so n must be odd.
Right.

Now show that if ##3a^2-b^2=2018## then there will be a solution for all ##3y^2-x^2=2018^n## if ##n## is odd.
 
$$3x^2=y^2+2018\implies x^2\geq \frac{2018}{3}\implies x\geq 26.$$ For ##x=27## , ##y=13## satisfy equation.
 
littlemathquark said:
$$3x^2=y^2+2018\implies x^2\geq \frac{2018}{3}\implies x\geq 26.$$ For ##x=27## ##y=13##
That's not what I said. This will be the third and last step but first, make the second. What does it mean that ##n## is odd?
 
If n is odd then ##n=2k+1##
 
  • #10
littlemathquark said:
If n is odd then ##n=2k+1##
Yes. Go ahead. How can you corstruct a solution for ##3x^2-y^2=2018^{2k+1}=2018\cdot (2018^{k})^2## if you have a solution ##3a^2-b^2=2018##?
 
  • #11
##3a^2.2018^{2k}-b^2.2018^{2k}=2018^{2k+1}##
 
  • #12
littlemathquark said:
##3a^2.2018^{2k}-b^2.2018^{2k}=2018^{2k+1}##
Yes. I would have written it as
$$
3x^2-y^2=2018^n=2018\cdot 2018^{2k}=
(3a^2-b^2)\cdot 2018^{2k}=3\cdot (a\cdot 2018^k)^2 - (b\cdot 2018^k)^2
$$
but yes.

So what is the answer to your initial question then?
 
  • #13
##3a^2.2018^{2k}-b^2.2018^{2k}=2018^{2k+1}## İf ##a=27## and ##b=13## then ##x=(27.2018^k)## and ##y=(13.2018^k)## are solutions. Thank you very much. But are these all solutions? Are there other solutions?
 
  • #14
littlemathquark said:
##3a^2.2018^{2k}-b^2.2018^{2k}=2018^{2k+1}## İf ##a=27## and ##b=13## then ##x=(27.2018^k)## and ##y=(13.2018^k)## are solutions. Thank you very much. But are these all solutions? Are there other solutions?
There are infinitely many solutions. With ##n,m\geq 0## these are (for ##2018##)
\begin{align*}
3x_n^2-y_n^2&=2018\\[6pt]
x_n &= \pm \dfrac{1}{6} \left(-81 (2 - \sqrt{3})^n + 13 \sqrt{3} (2 - \sqrt{3})^n - 81 (2 + \sqrt{3})^n - 13 \sqrt{3} (2 + \sqrt{3})^n\right)\\
y_n &= \pm \dfrac{1}{2} \left(-13 (2 - \sqrt{3})^n + 27 \sqrt{3} (2 - \sqrt{3})^n - 13 (2 + \sqrt{3})^n - 27 \sqrt{3} (2 + \sqrt{3})^n\right)\\[12pt]
3\hat x_m^2-\hat y_m^2&=2018\\[6pt]
\hat x_m &= \pm \dfrac{1}{6} \left(81 (2 - \sqrt{3})^m + 13 \sqrt{3} (2 - \sqrt{3})^m + 81 (2 + \sqrt{3})^m - 13 \sqrt{3} (2 + \sqrt{3})^m\right)\\
\hat y_m &= \pm \dfrac{1}{2} \left(13 (2 - \sqrt{3})^m + 27 \sqrt{3} (2 - \sqrt{3})^m + 13 (2 + \sqrt{3})^m - 27 \sqrt{3} (2 + \sqrt{3})^m\right)\\[12pt]
x_0&=\hat x_0=27\\
y_0&=\hat y_0=13\\[6pt]
x_1&=67\\
y_1&=107\\[6pt]
\hat x_1&=41\\
\hat y_1&=55\\[6pt]
\ldots \; etc.
\end{align*}
But I used Wolfram Alpha for it. I haven't found them myself. But your idea to find ##(27,13)## was nice. The solutions for higher odd powers of ##2018## are as described above:
$$
2018^{2k+1}=3(x_n\cdot 2018^k)^2-(y_n\cdot 2018^k)^2\, , \,2018^{2k+1}=3(\hat x_m\cdot 2018^k)^2-(\hat y_m\cdot 2018^k)^2
$$
 
Last edited:
  • Like
Likes   Reactions: littlemathquark
  • #15
The way we followed here was a typical one:

1.) Eliminate the impossible cases (##n=2k##) by a modulo consideration.
2.) Reduce the general case (##n=2k+1##) to the simple case (##n=1##).
3.) Solve the simple case (##(x,y)=(27,13)##).
 
  • Like
Likes   Reactions: littlemathquark
  • #16
I added the rest of the solutions and another two examples of low values in post #14 for completion.
 

Similar threads

Replies
12
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
Replies
9
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K