2
$\begingroup$

Could someone please verify whether my solution is okay? The problem with my answer is that it isn't rigorous, meaning I just assumed a lot of properties to be true, since the professor did not prove these in class. Additionally, I am not sure what the question is asking and so this is the best idea I came up with.

Show that $S=\left \{ 2^{m}3^{n}:m,n\in \Bbb{Z} \right \}\leq \Bbb{Q}\setminus \left \{0\right \}$ is the internal direct product isomorphic to $\Bbb{Z}\times \Bbb{Z}$.

Let $H=\left \{ 2^{m}:m\in \Bbb{Z} \right \}$ and $K=\left \{ 3^{n}:n\in \Bbb{Z} \right \}$.

Claim: $S$ is the internal direct product of $H$ and $K$.

$S=HK=\left \{ 2^{m}3^{n}:2^{m}\in H, 3^{n}\in K \right \}$, which is true.

If $m=n=0$, then $2^{0}=3^{0}=1$. Since $2^{m}\neq 3^{n}$ for any $m,n>0$, it must be that $H\cap K=\left \{1 \right \}$.

I am honestly not even certain whether the above statement is correct. I just wrote it to make the condition hold in my favor.

Let $2^{m}\in H$ and $3^{n}\in K$. Then $2^{m}3^{n}=3^{n}2^{m}$, since integers are abelian.

Claim: $S\cong \Bbb{Z}\times \Bbb{Z}$

Define $\phi:\Bbb{Z}\times \Bbb{Z}\to S$ by $\phi(m,n)=2^{m}3^{n}$.

Well-defined: If $(m,n)=(x,y)$, then $\phi(m,n)=2^{m}3^{n}=2^{x}3^{y}=\phi(x,y)$ by coordinate equality.

Operation-preserving: $\phi((m,n)+(x,y))=\phi(m+x,n+y) = 2^{m+x}3^{n+y}=2^{m}2^{x}3^{n}3^{y}=2^{m}3^{n}2^{x}3^{y}=\phi(m,n)\phi(x,y)$

Injective: If $\phi(m,n)=\phi(x,y)$, then $2^{m}3^{n}=2^{x}3^{y}$. Then $m=x$ and $n=y$ by countability of $\Bbb{Z}\times \Bbb{Z}$ and laws of exponents.

Surjective: $\phi$ is onto iff for all $2^{m}3^{n}\in S$, there exists an $(m,n)\in \Bbb{Z}\times \Bbb{Z}$ such that $\phi(m,n)=2^{m}3^{n}$. By construction of $\phi$, it is surjective.

$\endgroup$
2
$\begingroup$

Your proof seems fine to me.

If you want to justify that $2^m\neq 3^n$ for $m, n\ge 1$, note that the LHS is even whereas the RHS is odd. It also follows by uniqueness of factorisation.

$\endgroup$
  • 1
    $\begingroup$ Thank you for taking the time to look at my long solution. And thanks, that is actually very helpful! $\endgroup$ – numericalorange Mar 27 '18 at 0:54
  • $\begingroup$ You're welcome! :) $\endgroup$ – Shaun Mar 27 '18 at 0:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.