It is easy to show that $\mathbb{Q}(\sqrt[3]{2}+i)\subseteq \mathbb{Q}(\sqrt[3]{2},i)$.
But how I can show that $\mathbb{Q}(\sqrt[3]{2},i)\subseteq\mathbb{Q}(\sqrt[3]{2}+i)$?
I can't find a way to express $\sqrt[3]{2}$ in terms of $\sqrt[3]{2}+i$.
It is easy to show that $\mathbb{Q}(\sqrt[3]{2}+i)\subseteq \mathbb{Q}(\sqrt[3]{2},i)$.
But how I can show that $\mathbb{Q}(\sqrt[3]{2},i)\subseteq\mathbb{Q}(\sqrt[3]{2}+i)$?
I can't find a way to express $\sqrt[3]{2}$ in terms of $\sqrt[3]{2}+i$.
On
Consider numbers of the form $$x_0+x_1a+x_2a^2+x_3i+x_4ai+x_5a^2i$$ where $\sqrt[3]2=a$ and $x_i\in\mathbb Q$. It is easily shown that these numbers form a vector space $V$ under addition and are closed under multiplication.
Now consider the powers of $\sqrt[3]2+i=a+i=z$ from $z^0=1$ to $z^5$. Certainly all these numbers are of the form above, and the six powers of $z$ are linearly independent in $V$, so they are a basis. By inverting this basis, it is possible to derive an expression for $\sqrt[3]2$ in terms of powers of $z$, showing that it is in $\mathbb Q(\sqrt[3]2,i)$, and similarly for $i$.
Specifically, $$\sqrt[3]2=\frac1{22}(91+100z-78z^2+40z^3-9z^4+12z^5)$$
On
If you are lazy and do not want to compute by hand, or you want to check a greater number of similar examples and be sure there is no slip of the pen in the solution, you can do it with Macaulay2. Note:
1) iJ is generated by the irreducible polynomial, that $z = i + 2^{1/3}$ has to fulfill.
2) The primaryDecomposition calculations below give the representations of $v = p(z)$ for $v = i$ ($v^2 + 1 = 0$) and $v = 2^{1/3}$ ($v^3 - 2 = 0$). You can read them off from the factors of the primary decomposition. If you have several possibilities (several factors contain an $p_i(z) = v$, here the case at $v^2+1=0$) then you must embed everything into (a floating point approximation) of the complex (or real) numbers and calculate which $p_i(z) = v$ is fulfilled. Increase precision, until only one $p_{i_0}(z) = v$ is floating point zero.
Macaulay2, version 1.12
with packages: ConwayPolynomials, Elimination, IntegralClosure, InverseSystems, LLLBases,
PrimaryDecomposition, ReesAlgebra, TangentCone
i1 : R=QQ[x,y]
o1 = R
o1 : PolynomialRing
i2 : iI = ideal(x^2+1, y^3-2)
2 3
o2 = ideal (x + 1, y - 2)
o2 : Ideal of R
i3 : R1=R/iI
o3 = R1
o3 : QuotientRing
i4 : S=QQ[z]
o4 = S
o4 : PolynomialRing
i5 : phi=map(R1,S,{x+y})
o5 = map(R1,S,{x + y})
o5 : RingMap R1 <--- S
i6 : iJ = ker phi
6 4 3 2
o6 = ideal(z + 3z - 4z + 3z + 12z + 5)
o6 : Ideal of S
i7 : S1 = S/iJ
o7 = S1
o7 : QuotientRing
i8 : S11 = S1[v]
o8 = S11
o8 : PolynomialRing
i9 : primaryDecomposition ideal (S11/ideal(v^2 + 1))
2 5 4 3 2 2 5 4
o9 = {ideal (v + 1, 22v + 12z - 9z + 40z - 78z + 78z + 91), ideal (v + 1, 22v - 12z + 9z -
-----------------------------------------------------------------------------------------------
3 2
40z + 78z - 78z - 91)}
o9 : List
i10 : primaryDecomposition ideal (S11/ideal(v^3-2))
3 5 4 3 2 3 2 5 4
o10 = {ideal (v - 2, 22v - 12z + 9z - 40z + 78z - 100z - 91), ideal (v - 2, 22v + (12z - 9z
----------------------------------------------------------------------------------------------
3 2 5 4 3 2
+ 40z - 78z + 100z + 91)v - 18z + 8z - 60z + 106z - 106z - 142)}
o10 : List
Follows a sequence of maneuvers which shows how to express $i$ and $\sqrt[3]2$ in terms of $\sqrt[3]2 + i$, and hence that $\Bbb Q(\sqrt[3]2, i) = \Bbb Q(\sqrt[3]2 + i)$:
$\alpha = \sqrt[3]2 + i; \tag 1$
$\alpha - i = \sqrt[3]2; \tag 2$
$(\alpha - i)^3 = 2; \tag 3$
$\alpha^3 - 3 \alpha^2 i - 3 \alpha + i = 2; \tag 4$
$i(1 - 3\alpha^2) = 2 + 3\alpha - \alpha^3; \tag 5$
$i = \dfrac{\alpha^3 - 3\alpha - 2}{3\alpha^2 - 1} = \dfrac{2 + 3\alpha - \alpha^3}{1 - 3\alpha^2} \in \Bbb Q(\alpha) = \Bbb Q(\sqrt[3]2 + i); \tag 6$
$\sqrt[3]2 = \alpha - i \in \Bbb Q(\alpha) = \Bbb Q(\sqrt[3]2 + i); \tag 7$
from (6), (7) and the fact that $\Bbb Q(\alpha) = \Bbb Q(\sqrt[3]2 + i) \subset \Bbb Q(\sqrt[3]2, i)$ we affirm
$\Bbb Q(\sqrt[3]2, i) \subset \Bbb Q(\alpha) \Longrightarrow \Bbb Q(\sqrt[3]2, i) = \Bbb Q(\alpha) = \Bbb Q(\sqrt[3]2 + i). \tag 8$
N.B. We may in fact derive from (6) a sixth degree polynomial satisfied by $\alpha$:
$\alpha^3 - 3\alpha - 2 = i(3\alpha^2 - 1), \tag 9$
$(\alpha^3 - 3\alpha - 2)^2 = -(3\alpha^2 - 1)^2, \tag{10}$
$\alpha^6 + 9\alpha^2 + 4 - 6\alpha^4 - 4\alpha^3 + 6\alpha = -9\alpha^4 + 6\alpha^2 - 1, \tag{11}$
$\alpha^6 + 3\alpha^4 - 4\alpha^3 + 3\alpha^2 + 6\alpha + 1 = 0, \tag{12}$
consistent with the results of Parcly Taxel.