Splitting Field Proof for f(x) = x^3-5: Vector Space Basis B for K over Q

Join the discussion
Registration is free. Start your own thread to ask a follow-up.
2 replies · 2K views
AlexChandler
Messages
281
Reaction score
0

Homework Statement


Consider [tex]f(x) = x^3-5[/tex]
and its splitting field [tex]K = Q(5^{1/3}, \omega)[/tex]
where [tex]\omega = e^{2 \pi i/3}[/tex]
Show that [tex]B = \{1, 5^{1/3}, 5^{2/3}, \omega, \omega 5^{1/3} , \omega 5^{2/3} \}[/tex]
is a vector space basis for K over Q.

The Attempt at a Solution



I am just a bit confused. Since [tex]5^{1/3}[/tex] and [tex]\omega[/tex] are in K, and K is a field, then [tex]B'= \{ \omega ^2, \omega ^2 5^{1/3}, \omega ^2 5^{2/3} \} \subseteq K[/tex]
But how can we get any of these elements using only the shown basis B with scalars in Q? I would think that B+B' would be the vector space basis.
 
Physics news on Phys.org
try building up K in 2 steps:

first adjoin 51/3. what is [Q(51/3):Q]?

what does a basis of E = Q(51/3) over Q look like?

now K = E(ω). what is [K:E]? what is a basis for K over E?

if dimQ(E) = m, and dimE(K) = n,

what must dimQ(K) be?

this tells you how many basis elements you have to have.

then it's just a matter of proving linear independence over Q.

*******

as for how we get the elements of B', i'll show you for ω2:

ω satisfies the polynomial x2 + x + 1 = (x3 - 1)/(x - 1).

so ω2 = (-1)1 + (-1)ω, see?
 
Yes, thanks so much I was confusing myself.