
Number Theory Number Theory Math Forum 
 LinkBack  Thread Tools  Display Modes 
March 13th, 2009, 10:28 AM  #1 
Member Joined: Nov 2006 Posts: 54 Thanks: 0  Positive Integer Exponent Pair Puzzle
Problem Determine all possible pair(s) (p, q) positive integers that satisfy this equation: 2^p – 5^q = 3 My unsuccessful attempt is furnished hereunder as follows: p =1 yields then 5^q =1, which is a contradiction. p=2 yields 5^q = 1, so that q = 0., a ontradiction. If p>=3, then it follows that: 3^q = 3 (Mod 8), so that q must be odd If q =1, 2^p = 8, giving p = 3 Recalling that q is odd whenever q>=3, we substitute q = 2s+ 1, and observe that:: 2^p  5^(2s+1) = 3 Or, 2^p = 3 (Mod 5) or, p= 4t+3, where t is a non negative integer. So we have: 2^(4t+3) – 5^(2s+1) = 3 Or, 8*(16^t)  5*(25^s) = 3 For t =1, we obtain s =1, so that: (p, q) = (7, 3) Accordingly, till now, we have : (p,q) = (7, 3); (3,1) as valid solutions to the given problem. (Of course, if p and q were nonnegative integers rather than positive integers, then we would have obtained the additional pair (2,0) ) **** From this point onwards, my attempts to proceed any further has proved futile, and consequently, I am looking for a methodology giving any further valid solution(s) or any procedure conclusively proving that no further solutions can exist for the given problem. 
March 16th, 2009, 01:46 AM  #2 
Senior Member Joined: Sep 2008 Posts: 150 Thanks: 5  Re: Positive Integer Exponent Pair Puzzle
I don't see how to prove it, but you should not look for additional solutions: If we believe the ABCconjecture http://en.wikipedia.org/wiki/Abc_conjecture in the form "$q$ is always smaller than 2" (the biggest found up to date is 1,6...) then this gives that there can't be any solution with $t\geq 2$. 
March 17th, 2009, 06:21 AM  #3 
Senior Member Joined: Dec 2008 Posts: 160 Thanks: 0  Re: Positive Integer Exponent Pair Puzzle
Problem looks like modified Catalan's conjecture: that has only solutions: . Analytical proof of it was done just in 2002 by Mihailescu and involved theory of cyclotomic fields. Unless given problem is so special that has elementary solution, you, probably, should look at Mihailescu proof first. 
March 17th, 2009, 08:21 AM  #4 
Member Joined: Nov 2006 Posts: 54 Thanks: 0  Re: Positive Integer Exponent Pair Puzzle
Ty Peter and zolden, for your invaluable assistance and for your considered views in the foregoing matter. This problem was inspired by dansmath problem #253 (Powers Five Apart), in the following location. http://www.dansmath.com/probofwk/probar ... chor377644 The premise of the other problem possessing an elementary solution, led me to surmise a corresponding existence of an elementary solution in the current instance. As I mentioned in the original post that all my endeavours in solving the current exercise in terms of an elementary methodology failed to reach fruition. Having regard to your views, the existence of an elementary solution in the current instance does seem highly unlikely. 
March 18th, 2009, 08:56 AM  #5 
Global Moderator Joined: Nov 2006 From: UTC 5 Posts: 16,046 Thanks: 938 Math Focus: Number theory, computational mathematics, combinatorics, FOM, symbolic logic, TCS, algorithms  Re: Positive Integer Exponent Pair Puzzle
I understand that Mihailescu's proof uses the special nature of 1 deeply, so that the proof cannot (easily?) be generalized with greater numbers.

March 18th, 2009, 11:25 AM  #6 
Senior Member Joined: Dec 2008 Posts: 160 Thanks: 0  Re: Positive Integer Exponent Pair Puzzle
As of my knowledge it is a case of Pillai's conjecture that concerns a difference of perfect powers. Michailescu's proof, most probably, cannot be extended to the general case. I mentioned it just to outline the level of complexity given problem has. 
March 18th, 2009, 04:13 PM  #7 
Global Moderator Joined: Nov 2006 From: UTC 5 Posts: 16,046 Thanks: 938 Math Focus: Number theory, computational mathematics, combinatorics, FOM, symbolic logic, TCS, algorithms  Re: Positive Integer Exponent Pair Puzzle
Agree and agree. I was showing that it seems to be even harder than that.


Tags 
exponent, integer, pair, positive, puzzle 
Search tags for this page 
Click on a term to search for related topics.

Thread Tools  
Display Modes  

Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Integer Exponent Problem  Kiphyl  Algebra  3  March 4th, 2014 09:24 AM 
Smallest positive integer...  Alann  Number Theory  7  November 7th, 2012 08:39 AM 
Positive integer, perfect square and divisibility  K Sengupta  Math Events  4  July 1st, 2012 02:03 AM 
Perfect Square and Positive Integer Pair Determination  K Sengupta  Number Theory  0  March 8th, 2010 05:32 AM 
A base14 positive integer and divisibility puzzle  K Sengupta  Number Theory  3  April 26th, 2009 10:33 AM 