Suppose a and b are positive integers satisfying 1 ≤ a ≤ 3 1 , 1 ≤ b ≤ 3 1 such that the polynomial P ( x ) = x 3 − a x 2 + a 2 b 3 x + 9 a 2 b 2 has roots r , s , and t .
Given that there exists a positive integer k such that ( r + s ) ( s + t ) ( r + t ) = k 2 , compute the maximum possible value of a b .
This problem is posed by Samir 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.
This is a very well written solution. Instead of using Vieta's formula for P ( a − x ) , some correct solutions used symmetric algebraic identities.
The most common small mistake was to assume, explicitly or implicitly, that a b = ( l + 3 ) ( l − 3 ) implies that the factors ( a , b ) are, in some order, ( l + 3 , l − 3 ) .
Using Vièt theorem, we get: ⎩ ⎨ ⎧ r + s + t = a r s + s t + r t = a 2 b 3 r s t = − 9 a 2 b 2 .
Hence, k 2 = ( r + s ) ( s + t ) ( r + t ) = ( r + s + t ) ( r s + s t + r t ) − r s t = a 3 b 3 + 9 a 2 b 2 = a 2 b 2 ( a b + 9 ) .
This implies to a b + 9 is a perfect square, put: a b + 9 = n 2 .
Since 1 ≤ a ≤ 3 1 ; 1 ≤ b ≤ 3 1 , n 2 = a b + 9 ≤ 9 7 0 ⇒ n ≤ 3 1 .
If n = 3 1 , a b = n 2 − 9 = 9 5 2 , there are not exist numbers a , b such that.
If n = 3 0 , a b = n 2 − 9 = 8 9 1 , there are not exist numbers a , b such that.
If n = 2 9 , a b = n 2 − 9 = 8 3 2 , there are not exist numbers a , b such that.
If n = 2 8 , a b = n 2 − 9 = 7 7 5 , we can choose: a = 2 5 , b = 3 1 .
So, the maximum possible value of a b is 7 7 5 .
We know that r + s + t = a , so ( r + s ) ( s + t ) ( r + t ) = ( a − r ) ( a − s ) ( a − t ) = P ( a ) since P ( a ) is monic, P ( a ) = a 3 − a 3 + a 3 b 3 + 9 a 2 b 2 = a 2 b 2 ( a b + 9 ) = k 2 which means that a b + 9 is a perfect square, a b = c 2 − 9 for positive integer c . Since a , b have a upper bound of 31, we try c = 3 1 , 3 0 , 2 9 … , since 3 2 2 − 9 > 3 1 2 . Without loss of generality, let a ≥ b . If we choose the divisor d ≥ c 2 − 9 of c 2 − 9 that is closest to c 2 − 9 . If d > 3 1 , obviously there are no solutions since 3 1 ≥ a ≥ c 2 − 9 . For c = 3 1 , d = 3 4 . For c = 3 0 , d = 3 3 . For c = 2 9 , d = 3 2 . For c = 2 8 , d = 3 1 . Therefore a = 3 1 , b = 2 5 , a b = 7 7 5 .
Change (r+s)(s+t)(t+r)=k^2 as (a-r)(a-t)(a-s)=k^2 From polinomial we get: r+s+t=a ...(1) rs+st+rs=a^2b^2 ...(2) rst=9a^2b^3 ...(3) from 1,2,3 and also we get (a-r)(a-t)(a-s)=(ab)^2(ab+9)=k^2 ab+9 must be quadratic number... Now suppose that ab+9=l^2 now we strart from l=31 because 0<=a,b<=31. for l=31, there is not a,b for l=30, there is not a,b for l=29, there is not a,b and for l=28, there is a,b=25,31. If l maximum value then ab maximum value... and we get 775
( r + s ) ( s + t ) ( r + t ) = ( a − r ) ( a − s ) ( a − t ) = P ( a )
P ( a ) = a 3 − a 3 + a 3 b 3 + 9 a 2 b 2 = a 2 b 2 ( 9 + a b ) = k 2
thus 9 + a b is a perfect square
9 + a b = n 2
a b = ( n + 3 ) ( n − 3 )
n = 2 8 works, n = 2 9 3 1 doesn't.
We have:
r+s+t = a
rs + sr + rt = (a^2)(b^3)
rst = -9(a^2)(b^2)
(r+s)(s+t)(r+t) = (rs + st + rt)(r+s+t) - rst
k^2 = (a^3)(b^3) + 9a^2b^2
Let k = (ab)(m)
m^2 = ab + 9
(m-3)(m+3) = ab
Observe that m < 32 since 32^2 - 3 > 961
m = 31: (28)(34) = ab
m = 30: (27)(33) = ab
m = 29: (26)(32) = ab
m = 28: a = 31, b = 25 is a solution, giving us ab = 775
interestingly, it is easily provable that r,s,t cannot be simultaneously real:
r^2 + s^2 + t^2 + 2rs + 2st + 2rt = a^2
r^2 + s^2 + t^2 - rs - st - rt = 1/2 { (r-s)^2 + (r-t)^2 + (s-t)^2 }, implying
r^2 + s^2 + t^2 >= rs + st + rt
a^2 = r^2 + s^2 + t^2 + 2rs + 2st + 2rt >= 3rs + 3 st + 3rt = 3a^2b^3
a^2 >= 3a^2b^3
1 >= 3b^3; since b is a positive integer, contradiction.
Because P ( x ) = x 3 − a x 2 + a 2 b 3 x + 9 a 2 b 2 , then r + s + t = a .
Hence, ( r + s ) ( s + t ) ( r + t ) = ( a − t ) ( a − r ) ( a − s ) .
Meanwhile, P ( x ) can be wrote as P ( x ) = ( x − t ) ( x − r ) ( x − s ) .
Substituting x with a , now P ( a ) = ( a − t ) ( a − r ) ( a − s ) .
Because ( r + s ) ( s + t ) ( r + t ) = k 2 , then P ( a ) = k 2 , which means a 3 − a ⋅ a 2 + a 2 b 3 ⋅ a + 9 a 2 b 2 = k 2 .
Thus we get a 2 b 2 ( a b + 9 ) = k 2 .
a , b , k are positive integers, so a b + 9 must equal some m 2 .
a b ≤ 3 1 2 , and a b = ( m + 3 ) ( m − 3 ) , just need to verify m when m + 3 ≤ 3 1 .
At last, a b = ( 3 1 ) ( 3 1 − 3 − 3 ) = 3 1 × 2 5 = 3 2 × 2 5 − 2 5 = 8 0 0 − 2 5 = 7 7 5 .
First. From Vieta theory we can get :
1) r + s + t = a ⇒ (1)
2) r s + r t + s t = a 2 b 3 ⇒ (2)
3) r s t = − 9 a 2 b 3 ⇒ (3)
( r + s ) ( s + t ) ( t + r ) = ( r + s ) ( r s ) + ( r + t ) ( r t ) + ( s + t ) ( s t ) + 2 r s t ⇒ (4)
from (1) we can get :
r + s = a − t ,
r + t = a − s ,
s + t = a − r .
so, (4) can be changed into :
( a − r ) ( s t ) + ( a − t ) ( r s ) + ( a − s ) ( r t ) + 2 r s t .
( a ) ( r s + s t + r t ) − 3 r s t + 2 r s t .
( a ) ( r s + s t + r t ) − r s t . ⇒ (5)
from (2) and (3), we can change (5) into :
( a ) ( a 2 b 3 ) − ( − 9 a 2 b 2 ) .
( ( a b ) 2 ) ( a b + 9 ) .
we know that ( r + s ) ( s + t ) ( r + t ) = k 2 . with k is a positive integer.
( a b + 9 ) = ( a b k ) 2 .
We can assume ( a b k ) 2 = w 2 .
So, we get ( a b + 9 ) = w 2
a b = w 2 − 3 2
a b = ( w − 3 ) ( w + 3 )
Remember that a m a x = 3 1 and b m a x = 3 1 .
a b become max if w + 3 = a m a x or b m a x .
a b m a x = 3 1 ∗ ( 3 1 − 6 ) = 7 7 5 .
Using Vieta we can conclude that
r + s + t = a
r s + s t + t r = a 2 b 3
r s t = 9 a 2 b 2
then ( r + s ) ( s + t ) ( t + r ) = ( a − t ) ( a − r ) ( a − s ) = k 2
so a 3 − ( r + s + t ) a 2 + ( r s + s t + t r ) a + r s t
= a 3 − a 3 + a 3 b 3 + 9 a 2 b 2 = k 2
for k ∈ N
and a 2 b 2 ( a b + 9 ) = k 2 so we find the maximum value of ab
where a b + 9 = l 2 where l ∈ N
a b + 9 = l 2
a b = l 2 − 9 = ( l − 3 ) ( l + 3 )
then the maximum value will be reached when a b = 2 5 × 3 1 = 7 7 5
since r, s and t are roots of x 3 − a x 2 + a 2 b 3 x + 9 a 2 b 2 , we have r + s + t = a ; r s + r t + s t = a 2 b 3 ; r s t = − 9 a 2 b 2 ; hence ( r + s ) ( s + t ) ( r + t ) = ( r s + r t + s t ) ( r + s + t ) − r s t = a 3 b 3 + 9 a 2 b 2 = a 2 b 2 ( a b + 9 ) ; On the other hand, since a 2 b 2 ( a b + 9 ) is a square of a integer, and a b + 9 , a , b are integers, then it can be easily seen that a b + 9 is also a perfect square, which means that a b = u 2 − 9 for some integer u ≤ 3 1 . By a few trials, we find that u = 2 5 is the largest integer that satisfies that u 2 − 9 can be factored as the product of two integers ≤ 3 1 . Hence the maximum possible value of a b is 2 5 2 − 9 = 7 7 5 ;
hee, hee.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 |
|
1 |
|
Problem Loading...
Note Loading...
Set Loading...
r + s + t = a (By the Vieta's formula )
( r + s ) ( s + t ) ( r + t ) = ( a − r ) ( a − s ) ( a − t )
P ( x ) has roots r , s , and t
⟹ P ( a − x ) has roots a − r , a − s and a − t
Let P ( a − x ) = c 3 x 3 + c 2 x 2 + c 1 x 1 + c 0 . By the Vieta's formula , ( a − r ) ( a − s ) ( a − t ) = − c 3 / c 0 , so we only need to calculate c 3 and c 0 .
Because the coefficient of x 3 in P ( x ) is 1 , c 3 = − 1
c 0 = P ( a − 0 ) = P ( a ) = − a 3 b 2 − 9 a 2 b 2
Therefore ( r + s ) ( s + t ) ( r + t ) = ( a − r ) ( a − s ) ( a − t ) = − c 3 / c 0 = a 3 b 2 + 9 a 2 b 2 = a 2 b 2 ( a b + 9 ) = k 2
a and b are positive integers, so a b + 9 is also a square of a positive integer. Let a b + 9 = l 2 . Then a b = ( l + 3 ) ( l − 3 ) . we can easily find that ( a , b ) = ( 3 1 , 2 5 ) can be a solution of the equation. In this case, l = 2 8 .
If l > 2 8 , we can show that there is no solution of a b + 9 = l 2 in the range(If l = 2 9 , 3 0 or 3 1 , use direct computation. If l ≥ 3 2 , a b > 3 1 ⋅ 3 1 so there is no solution).
Therefore, The possible maximum value of a b is 3 1 × 2 5 = 7 7 5 .