Homepage Solution manuals Ivan Niven An Introduction to the Theory of Numbers Exercise 5.4.8 (Diophantine equation with $18$ variables)

Exercise 5.4.8 (Diophantine equation with $18$ variables)

Let f ( x ) = f ( x 1 , x 2 , x 3 ) = x 1 4 + x 2 4 + x 3 4 x 1 2 x 2 2 x 2 2 x 3 2 x 3 2 x 1 2 x 1 x 2 x 3 ( x 1 + x 2 + x 3 ) . Show that f ( x ) 1 ( mod 4 ) unless all three variables are even. Deduce that if f ( x ) + f ( y ) + f ( z ) = 4 ( f ( u ) + f ( v ) + f ( w ) ) for integral values of the variables, then all 18 variables are 0 .

Answers

Proof. If all three variables are even, then f ( x 1 , x 2 , x 3 ) 0 ( mod 4 ) . It remains 56 = 64 8 verifications with Sagemath

A = set()
for x1 in Integers(4):
    for x2 in Integers(4):
        for x3 in Integers(4):
            r = x1^4 + x2^4 + x3^4 - x1^2 * x2^2 - x2^2 * x3^2 -
                   x3^2 * x1^2 - x1 * x2 * x3 * (x1 + x2 + x3)
            if x1.lift() % 2 != 0 or x2.lift() % 2 != 0 or x3.lift() % 2 != 0:
                A.add(r)
A
    {1}

This shows that f ( x 1 , x 2 , x 3 ) 1 ( mod 4 ) unless all three variables are even.

Suppose for the sake of contradiction that

f ( x 1 , x 2 , x 3 ) + f ( y 1 , y 2 , y 3 ) + f ( z 1 , z 2 , z 3 ) = 4 ( f ( u 1 , u 2 , u 3 ) + f ( v 1 , v 2 , v 3 ) + f ( w 1 , w 2 , w 3 ) ) , (1)

where ( x 1 , x 2 , x 3 , y 1 , y 2 , y 3 , z 1 , z 2 , z 3 , u 1 , u 2 , u 3 , v 1 , v 2 , v 3 , w 1 , w 2 , w 3 ) ( 0 , 0 , , 0 ) .

If we divide by d 4 , where d is the gcd of these 18 values, we obtain a similar equation, where the new 18 values are relatively prime. So we can assume that in (1),

x 1 x 2 x 3 , y 1 y 2 y 3 z 1 z 2 z 3 u 1 u 2 u 3 v 1 v 2 v 3 w 1 w 2 w 3 = 1 .

(We use the same Fermat’s method of descent as in the previous problems.)

Reducing modulo 4, we obtain

f ( x 1 , x 2 , x 3 ) + f ( y 1 , y 2 , y 3 ) + f ( z 1 , z 2 , z 3 ) 0 ( mod 4 ) .

Since f ( x 1 , x 2 , x 3 ) 0  or  1 ( mod 4 ) , this congruence implies

f ( x 1 , x 2 , x 3 ) f ( y 1 , y 2 , y 3 ) f ( z 1 , z 2 , z 3 ) 0 ( mod 4 ) .

By the first part,

x 1 x 2 x 3 y 1 y 2 y 3 z 1 z 2 z 3 0 ( mod 2 ) .

Then 2 4 ff ( x 1 , x 2 , x 3 ) + f ( y 1 , y 2 , y 3 ) + f ( z 1 , z 2 , z 3 ) , thus by (1), 2 4 4 ( f ( u 1 , u 2 , u 3 ) + f ( v 1 , v 2 , v 3 ) + f ( w 1 , w 2 , w 3 ) ) . This shows that

f ( u 1 , u 2 , u 3 ) + f ( v 1 , v 2 , v 3 ) + f ( w 1 , w 2 , w 3 ) 0 ( mod 4 ) .

By the same reasoning,

u 1 u 2 u 3 v 1 v 2 v 3 w 1 w 2 w 3 0 ( mod 2 ) .

Therefore

2 x 1 x 2 x 3 , y 1 y 2 y 3 z 1 z 2 z 3 u 1 u 2 u 3 v 1 v 2 v 3 w 1 w 2 w 3 = 1 .

This contradiction 2 1 shows that the equation f ( x 1 , x 2 , x 3 ) + f ( y 1 , y 2 , y 3 ) + f ( z 1 , z 2 , z 3 ) = 4 ( f ( u 1 , u 2 , u 3 ) + f ( v 1 , v 2 , v 3 ) + f ( w 1 , w 2 , w 3 ) ) has no nontrivial integral solution. □

User profile picture
2025-04-16 11:11
Comments