Prime power, prime exponent

Find all positive integers x , y x,y and positive prime numbers p p such that

p x y p = 1. p^x-y^p = 1.

Enter your answer as the sum ( p i + x i + y i ) , \sum(p_i+x_i+y_i), where the sum runs over the solutions ( p , x , y ) (p,x,y) to the equation.


The answer is 11.

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.

2 solutions

Ivan Koswara
Jan 31, 2016

Note that p > 1 p > 1 because it's a prime number. We divide into two cases.

Case 1 : x > 1 x > 1

Since x , p > 1 x,p > 1 , this means p x , y p p^x, y^p are perfect powers. By Mihăilescu's theorem , the only pair of perfect powers that differ by 1 is 2 3 = 8 2^3 = 8 and 3 2 = 9 3^2 = 9 . This gives ( p , x , y ) = ( 3 , 2 , 2 ) (p,x,y) = (3,2,2) .

Case 2 : x = 1 x = 1

Then p = y p + 1 p = y^p + 1 . If p p is odd, then y p + 1 y^p + 1 factorizes to ( y + 1 ) ( y p 1 y p 2 + y + 1 ) (y+1) (y^{p-1} - y^{p-2} + \ldots - y + 1) , which means y p + 1 y^p + 1 is not prime. Thus p p must be even, giving p = 2 p = 2 . This leads to y = 1 y = 1 , so the solution is ( p , x , y ) = ( 2 , 1 , 1 ) (p,x,y) = (2,1,1) .


Thus the only two solutions are ( p , x , y ) = ( 2 , 1 , 1 ) , ( 3 , 2 , 2 ) (p,x,y) = (2,1,1), (3,2,2) , giving sum of 11.

Moderator note:

Nice usage of Mihailescu's theorem, though that is considerably a high power theorem.

Hmm, Mihailescu's theorem makes this a bit too easy. I had intended it to be an application of the LTE lemma , applied to p x = y p + 1 p^x = y^p+1 .

Patrick Corn - 5 years, 4 months ago

Log in to reply

[hide= "solution "] X =$V p(y^p+1)$ =$v p(y+1)+v_p(P)$ It forces to solution

Trishan Mondal - 1 year, 10 months ago
Barack Clinton
Apr 8, 2017

Case p = 2 p=2 : If x > 1 x>1 then 2 x 0 m o d 4 2^x\equiv0\mod4 and since y y is odd, its square is then 1 m o d 4 \equiv1\mod 4 , so the difference is 0 1 1 m o d 4 0-1\equiv-1\equiv\mod4 but on the RHS we have 1 m o d 4 \equiv 1\mod4 , hence whenever p = 2 p=2 we have x 1 x\le1 , by trial and error one finds the only solution: x = 1 , y = 1 , p = 2 x=1,y=1,p=2

Case p > 2 p>2 : If p y p\mid y then y = k p y=k\cdot p and hence p x ( k p ) p = 1 p^x-(kp)^p=1 is never reached. (consider modulo p p ), so p ∤ y p\not\mid y and the LTE conditions are satisfied. Then we write the original equation as y p + 1 p = p x y^p+1^p=p^x and obtain:

v p ( y p + 1 p ) = v p ( p x ) v p ( p ) + v p ( y + 1 ) = x v p ( y + 1 ) = x 1 v_p(y^p+1^p)=v_p(p^x)\iff v_p(p)+v_p(y+1)=x\iff v_p(y+1)=x-1

We then use the inequality; p v p ( y + 1 ) y + 1 p^{v_p(y+1)}\le y+1 to get

p ( x 1 ) y + 1 p x ( y + 1 ) p p^{(x-1)}\le y+1\iff p^x\le(y+1)p

Finally we arrive at the following inequality;

p(y+1)-y^p\ge 1\tag 1

We now use Fermat's little theorem;

y p y m o d p y^p\equiv y\mod p and y p + 1 = p x y^p+1=p^x gives that p ( y + 1 ) p\mid(y+1) and since p > 2 p>2 this yields y 2 y\ge 2 and then ( 1 ) (1) cannot hold for large p p (this is left to you), namely only for x = 2 , y = 2 , p = 3 x=2,y=2,p=3

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...