Does anyone know of a proof for the following?
If the nth root of X (n and X are integers>1) is rational, then
X = Yn (where Y is also an integer).
I am aware of the proof that sqrt(2) is irrational, but is there a
general case for the nth root of X?
The appropriate theorem here is:
Theorem If a is a root of a monic polynomial with integer
coefficients, then a is either irrational or an integer.
A monic polynomial is a polynomial whose leading coefficient (the
coefficient of the highest power of x) is 1. This proves all of
these sorts of problems. Specifically, in your case, consider the
polynomial xn-X=0. This is a monic polynomial, so the
roots must be either irrational or an integer, which is the proof
you are looking for.
To prove the theorem above, suppose the solution is p/q in lowest
terms, and q>1 (so that it is not an integer). If the polynomial
is
xn+c1xn-1+...+cn-1x+c
n=0. Substitute p/q into this equation, and multiply by
qn, to get
pn+c1pn-1q+...+cn-1pq
n-1+cnqn=0. q divides 0, and q divides
c1pn-1q+...+cn-1pqn-1+c
nqn, because there is a factor of q in each term.
Therefore, q must also divide pn (because if q divides
a+b and q divides b, q must also divide a). Therefore q divides p,
but this contradicts p/q in lowest terms. Hence the result.
Thanks Dan, that is very helpful.
But is it not also true that if X is an integer then the nth root
of X is also equal to Yn (where Y is an integer). So in
other words the 5th root of 10 is irrational because 10 is not
equal to Y5? Maybe the Theorem you gave demonstrates
this and I am missing it, but from what I can tell it only says
that the root can be any integer.
Thanks for your help, Joe
Yup, the 5th root of 10 is irrational, because 10 is not equal to Y5 for any integer Y (15=1, 25=32, ...). And since x5-10=0 must have roots that are either integers or irrational, and it cannot have integer roots by the above, so it must have irrational roots. Furthermore, the 5th root of 10 is a root, and so must be irrational! I'm not entirely sure what you don't understand.
Sorry, I got it after I posted the last message. I was slightly confused about xn-X=0 and if the second X was the same as the first x, but after working it out it makes sense. Thanks again for your help.