Find the largest integer N < 1 0 0 0 such that the polynomial f N ( x ) = x 6 + x 5 + x 4 + N x 3 + x 2 + x + 1 is reducible over the integers.
Definition: A polynomial that is reducible over the integers can be expressed as a product of two non-constant polynomials with integer coefficients.
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.
In fact we have the nice (and symmetric) identity: ( 1 + n x + x 2 ) ( 1 − ( n − 1 ) x + n ( n − 1 ) x 2 − ( n − 1 ) x 3 + x 4 ) = = 1 + x + x 2 + ( n − 1 ) ( n 2 − 2 ) x 3 + x 4 + x 5 + x 6 , which means that the polynomial is reducible when N = ( n − 1 ) ( n 2 − 2 ) for some integer n .
nice solution
Do you think there is a way to make use of the symmetry of the polynomial? For example, that strongly suggests that a = b , c = 1 , d = e , f = 1 in case 2 and b = 1 , c = e , f = 1 in case 3.
Log in to reply
How is it so for case 2? With N = 3 , it factors to ( x 3 + x + 1 ) ( x 3 + x 2 + 1 ) . I got a hunch that the costants for the cases 2 and 3 can't be − 1 because of the symmetry but I can't directly prove it, So I took an extra step just to make sure no such solution exist.
Log in to reply
Oh, I misread your solution. I thought the pairs of ( a , d ) and ( b , e ) had to line up with each other, but I see that is not the case.
I'll try to write a solution that makes use of the symmetry, but I think it misses some cases so someone else can try to modify it. For these kinds of problems, the first step you should do is factor out x 3 . x 3 ( x 3 + x 2 + x + N + x 1 + x 2 1 + x 3 1 ) Then rewrite the polynomial in terms of x + x 1 . x 3 ( x 3 + x 3 1 + x 2 + x 2 1 + x + x 1 + N ) x 3 ( ( x + x 1 ) 3 − 3 ( x + x 1 ) + ( x + x 1 ) 2 − 2 + ( x + x 1 ) + N ) x 3 ( ( x + x 1 ) 3 + ( x + x 1 ) 2 − 2 ( x + x 1 ) + N − 2 ) You can substitute x + x 1 with y . x 3 ( y 3 + y 2 − 2 y + N − 2 ) Factoring this polynomial can translate into a factorization of the original polynomial. The x 3 can be ignored since it can be distributed into each factor of the polynomial such that each term in each factor has a degree of at least 0.
We want to find the largest integer N such that the cubic can be factored. At least one factor has to be linear, so you can let the other factor be a quadratic. Assuming one of the factors is y + a where a is an integer, the other factor must be y 2 + ( 1 − a ) y + a 2 − a − 2 to keep the first 3 coefficients the same. ( y + a ) ( y 2 − ( 1 − a ) y + a 2 − a − 2 ) = y 3 + y 2 − 2 y + a 3 − a 2 − 2 a y 3 + y 2 − 2 y + a 3 − a 2 − 2 a = y 3 + y 2 − 2 y + N − 2 a 3 − a 2 − 2 a = N − 2 N = ( a − 1 ) ( a 2 − 2 ) The maximum value of a such that N is less than 1000 is 1 0 , so N = 9 ⋅ 9 8 = 8 8 2 .
When I solved this problem I did something similar to Pi Han Goh's solution, with the 3 cases.
I have also done in same way, I also find some mistakes in my solution
So Nice!
Problem Loading...
Note Loading...
Set Loading...
There's 3 possible ways f N ( x ) can be factored.
Case 1 : f N ( x ) has linear factor(s), then by rational root theorem, f N ( x ) = 0 has possible rational roots of ± 1 , solving f N ( ± 1 ) = 0 gives N = − 6 , 2 , so max ( N ) = 2
Case 2 : f N ( x ) has only cubic factors, so f N ( x ) = ( x 3 + a x 2 + b x + c ) ( x 3 + d x 2 + e x + f ) for integers a , b , c , d , e , f . Expand and compare different powers of x gives the following equations
a + d b + a d + e c d + b e + a f c e + b f c f = = = = = 1 1 1 1 1
Since they are all integers, we have only two possible subcases: c = f = 1 and/or c = f = − 1
For c = f = 1 , we get a + d = 1 , b + a d + e = 1 , d + b e + a = 1 , b + e = 1 , eliminates one equation from another gives ( a , d ) = ( 1 , 0 ) , ( 0 , 1 ) , and ( b , e ) = ( 1 , 0 ) , ( 0 , 1 ) , so N = c + b d + a e + f = 2 , 3
For c = f = − 1 , we get a + d = 1 , a d = 2 , ( a , d ) satisfy the equation y 2 − y + 2 = 0 , but it has negative discriminant, so c = f = − 1 is infeasible.
Therefore, max ( N ) = 3
Case 3 f N ( x ) has no linear factors nor cubic factors but have quadratic factor(s), so f N ( x ) = ( x 2 + a x + b ) ( x 4 + c x 3 + d x 2 + e x + f ) for integers a , b , c , d , e , f . (Note that these values of a , b , c , d , e , f are not necessarily similar to the ones in Case 2). Likewise
a + c b + a c + d b d + a e + f b e + a f b f = = = = = 1 1 1 1 1
Again, there's two possible subcase: b = f = 1 and/or b = f = − 1
For b = f = 1 , simplify the equations, and let a = x for some integer x , then state a , c , d , e in terms of x gives a = x , c = 1 − x , d = x 2 − x , e = 1 − x , so N = b c + a d + e = x 3 − x 2 − 2 x + 2 . Because of the constraint N < 1 0 0 0 , we have x 3 − x 2 − 2 x + 2 < 1 0 0 0 , it's easy to see that for x = 1 0 yields the largest N , so N = 1 0 3 − 1 0 2 − 2 ⋅ 1 0 + 2 = 8 8 2
For b = f = − 1 , we have a + c = 1 , a c + d = 2 , − d + a e = 2 , − e − a = 1 . The second equation plus the third equation gives c + e = a 4 ; while the first equation minus the fourth equation gives 2 a + c + e = 0 ⇒ 2 a + 4 / a = 0 ⇒ a 2 = − 2 which is absurd because a is real. Thus b = f = − 1 is infeasible.
Hence, with all the these 3 cases, max ( N ) = 8 8 2