x 2 ( y 3 + z 3 ) = 3 1 5 ( x y z + 7 )
Let ( x 1 , y 1 , z 1 ) , ( x 2 , y 2 , z 2 ) , … , ( x n , y n , z n ) be all of the solutions to the equation above, where x , y , z are positive integers. Find k = 1 ∑ n ( x k + y k + z k ) .
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.
the other solutions containing negative integers are . (x,y,z) = (-525,-1/5,-1/5) , (-525,-1/5,2/5) , (7,1,-1) , (7,-1,1)
Log in to reply
Thanks for pointing out this. I'll try to figure out how to obtain theses solutions.
By the same arguments as Kazem, we can deduce that x = 7 , 2 1 .
If x = 7 then, putting z = U − y , we obtain the equation 3 ( U + 1 5 ) y 2 − 3 U ( U + 1 5 ) y + ( U 3 − 4 5 ) = 0 Since this quadratic in y has integer roots, its discriminant 8 1 0 0 + 5 4 0 U + 2 0 2 5 U 2 + 9 0 U 3 − 3 U 4 must be a perfect square. Solving this equation numerically, the discriminant is only positive for integer U when − 1 5 ≤ U ≤ 4 5 , and the discriminant is a perfect square for an integer U just when U = 4 5 . Solving the quadratic equation for U = 4 5 yields y = 2 2 , 2 3 . Thus we obtain solutions ( 7 , 2 2 , 2 3 ) and ( 7 , 2 3 , 2 2 ) .
Similarly, if x = 2 1 then, putting z = U − y , we obtain the equation 3 ( U + 5 ) y 2 − 3 U ( U + 5 ) y + U 3 − 5 = 0 The discriminant 3 0 0 + 6 0 U + 2 2 5 U 2 + 3 0 U 3 − 3 U 4 is only positive for integer U when − 5 ≤ U ≤ 1 5 , and the discriminant is a perfect square for integer U precisely when U = 5 . Solving the quadratic equation for U = 5 yields y = 1 , 4 . Thus we obtain solutions ( 2 1 , 1 , 4 ) and ( 2 1 , 4 , 1 )
Thus the solution is 7 + 2 2 + 2 3 + 7 + 2 3 + 2 2 + 2 1 + 1 + 4 + 2 1 + 4 + 1 = 1 5 6 .
It is worth noting that this method can be used to show that there are no solutions when x = 1 or x = 3 , so the condition that x > 3 could be omitted.
Thanks for the nice solution Mark.
Thanks! I've removed the condition that x > 3 .
Log in to reply
Thanks for bringing up this point. I'll update my solution to cover the cases x = 1 and x = 3 .
Problem Loading...
Note Loading...
Set Loading...
I have always liked to use big numbers in my Diophantine equations, specially 3 1 5 , which is my favorite number. When you work on a equation with big numbers in it, you will end up learning a lot about the ways of bounding variables. Now we have the following equation x 2 ( y 3 + z 3 ) = 3 1 5 ( x y z + 7 )
There is a constant term in the equation, which is a good point to start. Note that x divides the constant term, which is 3 1 5 × 7 = 3 2 × 5 × 7 2 . If 5 ∣ x then the LHS of the equation is divisible by 2 5 , and it follows that 2 5 ∣ 3 1 5 ( x y z + 7 ) or 5 ∣ x y z + 7 . This is false as 5 ∣ x . Therefore x must be in the form 3 r 7 s .
If r = 2 then the LHS is divisible by 3 4 , and, thus, 3 4 ∣ 3 1 5 ( x y z + 7 ) . But 3 1 5 ( x y z + 7 ) = 3 2 × 3 5 ( 3 m + 1 ) and the power of 3 in 3 1 5 ( x y z + 7 ) will be 2 . Therefore r ≤ 1 . Using the same reasoning, one can conclude that s ≤ 1 . Hence possible values for x are 1 , 3 , 7 , 2 1 , which results in the following four equations y 3 + z 3 = 3 1 5 ( y z + 7 ) ( 1 ) y 3 + z 3 = 3 5 ( 3 y z + 7 ) ( 2 ) y 3 + z 3 = 4 5 ( y z + 1 ) ( 3 ) y 3 + z 3 = 5 ( 3 y z + 1 ) , ( 4 ) respectively.
Observe that , for example, in the first equation 3 ∣ y 3 + z 3 and 5 ∣ y 3 + z 3 . In the second equation 5 ∣ y 3 + z 3 . Now we will prove for primes p ∈ { 3 , 5 } that if p ∣ y 3 + z 3 then p ∣ y + z . Consider the following more general lemma:
Lemma : For every odd prime number p , where p − 1 ≡ 3 0 , if p ∣ y 3 + z 3 then p ∣ y + z .
Proof : This is clear if p divides one of y or z , so let p ∤ y and p ∤ z . Then y p − 1 ≡ p z p − 1 by FLT . On the other hand y 3 ≡ p − z 3 .
We claim that there are positive integers a and b such that 3 a = ( p − 1 ) b + 1 . This is clear, because whatever p might be (except primes that are congruent to 1 modulo 3 ), we can find a positive integer b such that ( p − 1 ) b + 1 is divisible by 3 . And for an odd prime, a will be an odd number.
Now using a and b as powers, from y p − 1 ≡ p z p − 1 we get y ( p − 1 ) b ≡ p z ( p − 1 ) b and from y 3 ≡ p − z 3 we get y 3 a ≡ p − z 3 a . Sum two relations y 3 a + y ( p − 1 ) b ≡ p z ( p − 1 ) b − z 3 a ⇒ y ( p − 1 ) b ( y + 1 ) ≡ p z ( p − 1 ) b ( 1 − z ) ≡ p y ( p − 1 ) b ( 1 − z ) and y + 1 ≡ p 1 − z ⇒ y + z ≡ p 0
We are going to use this lemma for decreasing coefficients and also for lowering the order of main variables in the four obtained equations.
Equation (1) Using Lemma for the equation ( 1 ) , we get y + z = 1 5 k , where k is a positive integer and the equation becomes 1 5 k ( 2 2 5 k 2 − 3 y z ) = 3 1 5 ( y z + 7 ) ⇒ k + 7 7 5 k 3 − 4 9 = y z Applying ( y + z ) 2 ≥ 4 y z , gives k + 7 7 5 k 3 − 4 9 = y z ≤ 4 ( y + z ) 2 = 4 2 2 5 k 2 Rearranging this inequality gives 7 5 k 2 ( k − 2 1 ) ≤ 1 9 6 . It is easy to see that k ≤ 2 1 . Once again write y z = k + 7 7 5 k 3 − 4 9 = k + 7 7 5 k 3 + 7 5 × 7 3 − 7 5 × 7 3 − 4 9 = 7 5 ( k 2 − 7 k + 4 9 ) − k + 7 2 × 7 2 × 2 6 3 Thus, k + 7 ∣ 2 × 7 2 × 2 6 3 . But note that 8 ≤ k + 7 ≤ 2 8 and the only divisor of 2 × 7 2 × 2 6 3 that lies in this range is 1 4 . Hence k + 7 = 1 4 and k = 7 . Therefore ( 1 0 5 − z ) z = 1 4 7 5 × 7 3 − 4 9 = 1 8 3 4 , which results in z 2 − 1 0 5 z + 1 8 3 4 = 0 and no integer solutions.
Equation (2) In an exact similar manner to equation ( 1 ) , one can prove that there is no integer solution for equation ( 2 ) .
Equations (3) and (4) Using Lemma for the equation ( 3 ) we get y + z = 1 5 k and for the equation ( 4 ) we get y + z = 5 l , where k and l are some positive integers. Using the identity y 3 + z 3 = ( y + z ) ( ( y − z ) 2 + y z ) equations ( 3 ) and ( 4 ) take the following forms: k ( y − z ) 2 + ( k − 3 ) y z = 3 ( 5 ) l ( y − z ) 2 + ( l − 3 ) y z = 1 ( 6 ) In equation ( 5 ) for y = z we get ( k − 3 ) y 2 = 3 which implies y = 1 and y + z = 2 = 1 5 k , thus y = z and y z ≥ 2 and ( y − z ) 2 ≥ 1 . Hence if k > 3 then LHS of ( 5 ) will be greater than 3 . It follows that k ≤ 3 . For k = 1 , ( y − z ) 2 − 2 y z = 3 and y + z = 1 5 , which fail to have integer solutions. For k = 2 , 2 ( y − z ) 2 − y z = 3 and y + z = 3 0 , which fail to have integer solutions. For k = 3 , ( y − z ) 2 = 1 and y + z = 4 5 , which give two solutions: ( x , y , z ) = ( 7 , 2 2 , 2 3 ) , ( 7 , 2 3 , 2 2 ) .
In equation ( 6 ) , if l > 3 then LHS of ( 3 ) will be greater than 1 . It follows that l ≤ 3 . Examining l = 1 , 2 , 3 with combining equation ( 6 ) and y + z = 5 l gives two solutions again: ( x , y , z ) = ( 2 1 , 1 , 4 ) , ( 2 1 , 4 , 1 ) .