Let's prepare for RMO #3

Find all the ordered pars of primes p , q p,q such that ( 5 p 2 p ) ( 5 q 2 q ) p q \dfrac{(5^{p}-2^{p})(5^{q}-2^{q})}{pq} is an integer.

Submit your answer as sum of all ( p , q ) (p,q) .


Note: p p and q q doesn't need to be necessarily distinct.


The answer is 38.

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.

4 solutions

Md Zuhair
Sep 23, 2017

Nice one!

By Fermat's Little Theorem we get

5 p = 5 m o d p 5^p=5 \mod p & 2 p = 2 m o d p 2^p=2 \mod p

So Subtracting 5 p 3 p = 3 m o d p 5^p-3^p = 3 \mod p

Which means ( p , q ) = ( 3 , 3 ) (p,q)=(3,3) is a solution.

Now Either Case I ( p = 3 a n d q = q ) (p=3 and q=q) or Case 2 ( q = 3 a n d p = p ) (q=3 and p=p)

For case I, When we put p=3, We get q=13.

The opposite is the solution for q=3.

So ( 3 + 3 + 13 + 3 + 3 + 13 ) = 38 (3+3+13+3+3+13)=\boxed{38}

That's just what I have done, but I'm still not conviced - can't there be a case where p ( 5 q 2 q ) p|(5^q-2^q) and q ( 5 p 2 p ) q|(5^p-2^p) ?

me myself - 3 years, 8 months ago

Log in to reply

Ya. There is . 13 is one of them. See it carefully.

Md Zuhair - 3 years, 8 months ago

Log in to reply

39 divides 5 3 2 3 5^3-2^3 , regardless of the value of 5 13 2 13 5^{13}-2^{13} . But are there p\=q\=3 such that p ( 5 q 2 q ) p|(5^q-2^q) and q ( 5 p 2 p ) q|(5^p-2^p) ? , and if not, why?

me myself - 3 years, 8 months ago

I agree with @me myself that the solution is incomplete--you have to show that at least one of p p and q q is 3.

Patrick Corn - 3 years, 8 months ago

Log in to reply

Sir can you help me out with that? Actually I am number theory Beginner. Actually I need a lot of help from you as you are number theory expert.

Can we contact somewhere? Like Whatsapp?

Md Zuhair - 3 years, 8 months ago

Log in to reply

@Md Zuhair I just posted a solution. Let me know if it makes sense.

Patrick Corn - 3 years, 8 months ago

Log in to reply

@Patrick Corn Yes sir. It seems nice :)

Md Zuhair - 3 years, 8 months ago
Patrick Corn
Sep 26, 2017

Suppose for the moment that p q . p \ge q. Now either p ( 5 p 2 p ) p | (5^p-2^p) or q ( 5 p 2 p ) . q|(5^p-2^p). If the former, then by Fermat's little theorem we get 0 5 p 2 p 5 2 0 \equiv 5^p - 2^p \equiv 5-2 mod p , p, so p = 3. p=3. Since q p , q\le p, the only solution in this case is ( 3 , 3 ) (3,3) by inspection.

If the latter, then get ( 5 / 2 ) p 1 (5/2)^p \equiv 1 mod q . q. (We know q 2 q\ne 2 so it's all right to divide through by 2 p . 2^p. )

This means that the order of 5 / 2 5/2 mod q q divides p . p. Since p p is prime, the order is either 1 1 or p . p. But Fermat's little theorem says that the order is at most q 1 < p , q-1 < p, so it has to be 1. 1. Hence 5 / 2 1 5/2 \equiv 1 mod q q and q = 3. q=3.

Now we get that ( 5 p 2 p ) 117 3 p \frac{(5^p-2^p)117}{3p} is an integer, and if p 3 p \ne 3 then p p doesn't divide 5 p 2 p , 5^p-2^p, so it must divide 117 , 117, and the only possibility is p = 13. p=13.

Putting these all together and dropping the condition that p q , p \ge q, we get the three solutions ( 3 , 3 ) , ( 3 , 13 ) , ( 13 , 3 ) . (3,3),(3,13),(13,3). The answer is 38 . \fbox{38}.

The Dark Lord
Feb 21, 2018
  • Case 1 : p 5 p 2 p p \mid 5^{p} - 2^{p} and q 5 q 2 q q \mid 5^{q} - 2^{q} .

  • Case 2 : p 5 q 2 q p \mid 5^{q} - 2^{q} and q 5 p 2 p q \mid 5^{p} - 2^{p} .

Case 1 ) - Suppose p 2 / 5 p \neq 2/5 .Then we have from Fermat 's Little Theorem that 5 p 5 m o d p 5^{p} \equiv 5 \mod p and 2 p 2 m o d p 2^{p} \equiv 2 \mod p .This implies 5 p 2 p 3 m o d p 5^{p} - 2^{p} \equiv 3 \mod p .If p 5 p 2 p p \mid 5^{p} - 2^{p} then this forces p = 3 p = 3 .Similarly we deduce q = 3 q =3 .

Case 2) - Suppose p 5 p \neq 5 , p > 3 p > 3 . Then p 5 q 2 q p \mid 5^{q} - 2^{q} . Or we can say 5 q 2 q m o d p 5^{q} \equiv 2^{q} \mod p .Since we have again from Fermat 's little theorem that 5 p 1 2 p 1 1 m o d p 5^{p-1} \equiv 2^{p-1} \equiv 1 \mod p .These 2 deductions imply 5 g c d ( p 1 , q ) 2 g c d ( p 1 , q ) m o d p 5^{gcd(p-1,q)} \equiv 2^{gcd(p-1,q)} \mod p . But since p , q p,q are primes it means g c d ( p 1 , q ) = 1 gcd(p-1,q) = 1 . Thus 5 2 m o d p 5 \equiv 2 \mod p .This again forces p = 3 p = 3 a contradiction. Hence one out of p , q p,q must be 3 3 always.In case 1 1 both are together 3 3 however in case 2 2 only one is 3 3 .Putting p = 3 p = 3 we get that q 5 3 2 3 = 117 q \mid 5^{3} -2^{3} = 117 and thus q = 13 q =13 .Similarly q = 3 q = 3 renders p = 13 p =13 .

Thus we have 3 ordered solutions ( p , q ) = [ ( 3 , 3 ) ; ( 3 , 13 ) ; ( 13 , 3 ) ] (p,q) = [(3,3) ; (3,13) ; (13,3)]

Bro, I already got the idea that one of them is 3, since 5 p 2 p 5^{p}-2^{p} always divides 3 for any p and also observed that 5 3 2 3 = 117 = 3 2 13 5^3-2^3=117=3^2\cdot13 , so 13 would also work and I got the answer without applying Fermat's little theorum.

Harry Potter - 3 years, 3 months ago

Log in to reply

@The Boy who Lived U have to also prove both of them are 3.

The Dark Lord - 3 years, 3 months ago

Log in to reply

I actually didn't solve it subjectively but, managed to get the correct pairs

Harry Potter - 3 years, 3 months ago

Log in to reply

@Harry Potter Okay here 's a nice problem : Find with proof all n such that 2^n + 12^n + 2011^n is a perfect square.Dont copy frm net. !!

The Dark Lord - 3 years, 3 months ago

J u s t w o r k i n g w i t h t h e p r i m e s f r o m 2 t o 41 o n 50 d i g i t c a l c u l a t o r n o t i n g t h a t p q . ( 3 , 3 ) a n d ( 3 , 13 ) a n d s o ( 13 , 3 ) w e r e t h e r e s u l t s . S o 2 ( 3 + 3 + 13 ) = 38. p = q = 41 , w a s a 53 d i g i t r o u n d n u m b e r . I a m n o t m u c h f a m i l i a r w i t h T h e o r y o f N u m b e r s . Just~ working ~with~ the ~primes~ from~~ 2 ~to~ 41~~ on~ 50~ digit~ calculator~ noting~ that~~~p~\Longleftrightarrow~q.\\ (3,3)~ and ~(3,13) ~and~so~~(13,3) ~were~ the~ results. ~~So~2*(3+3+13)=38.\\ p=q=41, ~was~ a ~53~ digit~ round~ number. ~~~I~ am ~not~ much~ familiar~ with~ Theory~ of~ Numbers.

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...