Let A be an irrational number and let N is an integer > 1.
Is N√(A + √(A2-1)) + N√( A - √(A2-1)) always an irrational number?
Give reasons for your answer.
I think I'm most of the way to a proof for all N>1, however I just can't seem to get past the last hurdle. I'm posting it here in hopes someone else might be able to finish it
lemma 1:
if x is irrational and n is an integer > 1, then x^(1/n) is irrational
proof:
assume x^(1/n)=r a rational number
then x=r^n, however r^n is rational and thus contradicts x being irrational
lemma 2:
if x is irrational then x-sqrt(x^2-1) is also irrational
proof:
assume x-sqrt(x^2-1)=r a rational number
sqrt(x^2-1)=(x-r)
x^2-1=(x-r)^2
x^2-1=x^2-2xr+r^2
r^2-2xr+1=0
2xr=r^2+1
x=(r^2+1)/2r
if r is rational then so is (r^2+1)/2r and thus contradicts x being irrational
lemma 3:
if x is irrational then x+sqrt(x^2-1) is also irrational
proof:
assume x+sqrt(x^2-1)=r a rational number
sqrt(x^2-1)=(r-x)
x^2-1=(x-r)^2
x^2-1=x^2-2xr+r^2
r^2-2xr+1=0
2xr=r^2+1
x=(r^2+1)/2r
if r is rational then so is (r^2+1)/2r and thus contradicts x being irrational
combining lemmas 1,2, and 3 we can show that if A is irrational then so is both
(A+sqrt(A^2-1))^(1/n) and (A-sqrt(A^2-1))^(1/n)
and this is where I am stuck. I can show the two parts remain irrational, however I can not guarantee that their sum is also irrational. For a simple example, imagine if they somehow simplified to 1+sqrt(5) and 1-sqrt(5) then they would both be irrational but their sum would be the rational value of 2.
|
|
Posted by Daniel
on 2015-11-17 15:26:00 |
most of a general proof
