Consider all pairs ( p , q ) of primes such that p 2 + q 3 is a perfect cube. Find the sum of all possible values of p + q + p q as p and q run over all these pairs.
This problem is a part of Tessellate S.T.E.M.S. (2019)
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.
p 2 + q 3 = x 3 , where x is a positive integer.
p 2 = x 3 − q 3 ⟹ p 2 = ( x − q ) ( x 2 + q 2 + x q )
there are two possibilities now (considering x 2 + q 2 + x q > x − q ).
1- x − q = p and x 2 + q 2 + x q = p
we have
x 2 + q 2 + x q = x − q ⟹ ( x − q ) 2 + 3 x q = x − q ⟹ ( x − q ) ( x − q − 1 ) = − 3 x q
since the RHS is negative and the LHS can be either zero or positive, the first possibility would not lead to any solution.
2- x − q = 1 and x 2 + q 2 + x q = p 2
then we substitute x = q + 1 in x 2 + q 2 + x q = p 2 and, after few simplifying steps, we get
3 q ( q + 1 ) = ( p − 1 ) ( p + 1 )
now we can do the rest as the following.
if p − 1 is a multiple of three, then
q ( q + 1 ) = 3 ( p − 1 ) ( p + 1 )
Now, for this part I could not come up with a nice proof, but intuitively, we need to take factors k from p + 1 and multiply to 3 p − 1 so we make two consecutive integers q and q + 1 . The factor cannot be 3 . If the factor is greater than or equal to 5 , then the difference 3 k ( p − 1 ) − k p + 1 would have a value grater than one (this is the part that I'm not sure about). Therefore, we are only left with k = 2 .
3 k ( p − 1 ) − k ( p + 1 ) = 3 2 ( p − 1 ) − 2 ( p + 1 ) = 1
solve for p then. the solution is p = 1 1 , however, there would be no suitable q in this case.
If p + 1 is a multiple of three, then we get the solution.
3 k ( p + 1 ) − k ( p − 1 ) = 3 2 ( p + 1 ) − 2 ( p − 1 ) = 1 ⟹ p = 1 3 , q = 7
When you take a factor of k from p + 1 and multiply to 3 p − 1 to get q and q + 1 respectively, perhaps you should also consider the possibility that k can be small enough for you to have q + 1 = k p + 1 , q = 3 k ( p − 1 ) so that k p + 1 − 3 k ( p − 1 ) = 1 so you should work on (and mention, if the problem was subjective) this case separately (even though it is easy to deal with (notice that the coefficient of p is non-positive, so p can't be very large...) but you may get some additional "boundary-solutions"). The same goes for elaborating on the case when p + 1 is a multiple of 3 .
"If the factor is greater than or equal to 5 then the difference 3 k ( p − 1 ) − k p + 1 would have a value greater than one (this is the part that I'm not sure about)": Well if k ≥ 5 then k p + 1 ≤ 5 p + 1 and 3 k ( p − 1 ) − k p + 1 ≥ 3 5 ( p − 1 ) − 5 p + 1 = 1 5 2 2 p − 2 8 ≥ 1 for all p ≥ 2 > 2 2 4 3 and you may now be sure about this.
Cheers, Tessellate S.T.E.M.S. 2019 Math Committee.
Problem Loading...
Note Loading...
Set Loading...
We claim that ( p , q ) = ( 1 3 , 7 ) is the only solution to the given Diophantine Equation, whence the desired value is 1 4 . 8 − 1 = 1 1 1 .
We have p 2 + q 3 = a 3 for some positive integer a which yields p 2 = a 3 − q 3 = ( a − q ) ( a 2 + a q + q 2 ) and by the Unique Factorization Theorem for integers we may have one of the following two possibilities:
Case 1 : p = a − q = a 2 + a q + q 2 ⟹ p = ( q + p ) 2 + q ( q + p ) + q 2 > p 2 , a contradiction as p > 1 .
Case 2 : 1 = a − q , p 2 = a 2 + a q + q 2 ⟹ p 2 = ( q + 1 ) 2 + q ( q + 1 ) + q 2 = 3 q 2 + 3 q + 1 = 3 q ( q + 1 ) + 1 ⟹ ( p − 1 ) ( p + 1 ) = 3 q ( q + 1 )
Now since q is prime we can have either q ∣ ( p − 1 ) or q ∣ ( p + 1 ) .
In the former case, p = q k + 1 for some k ∈ N and we have 3 q ( q + 1 ) = ( p − 1 ) ( p + 1 ) = q k ( q k + 2 ) ⟺ 3 ( q + 1 ) = k ( k q + 2 ) ≥ 2 ( 2 q + 2 ) unless k = 1 in which case we have 3 ( q + 1 ) = q + 2 which has no solution.
Hence we must have q ∣ ( p + 1 ) which yields, since k ∈ N , 3 ( q + 1 ) = k ( q k − 2 ) ⟺ 2 k + 3 = q ( k 2 − 3 ) ≥ 2 ( k 2 − 3 ) ⟹ 2 k 2 − 2 k − 9 ≤ 0 ⟹ 1 ≤ k ≤ \floor 2 1 + 1 9 = 2 As checked immediately, we then must have k = 2 which yields q = 7 , p = 1 3 as the only solution.