4
$\begingroup$

my task is following

Let $I=\{a+b\sqrt{10}:13\mid2a-b\}$ be a subset of the ring $\mathbb{Z}[\sqrt{10}]$. Decide if $I$ is an ideal of $\mathbb{Z}[\sqrt{10}]$ and if so, decide if it is principal, prime or maximal.

I've proved that $I$ is ideal indeed (addition is trivial and it is closed under multiplication because $13\mid2a-b\ \Leftrightarrow\ 13\mid a-20b$).

But I have problem with the properties of $I$. I know that $I$ is prime/maximal iff $\mathbb{Z}[\sqrt{10}]/I$ is an integral domain / a field. I don't know, how to continue -- $I$ is kernel of some ring homomorphism, but I have no idea how to use it. I suppose we would like to use something like $f:\mathbb{Z}[\sqrt{10}]\to\mathbb{Z}_{13},a+b\sqrt{10}\mapsto[2a-b]_{13}$ but is this even a map?... Edit: oh, of course $f$ is a map, since the element $a+b\sqrt{10}$ represents only itself...

$\endgroup$
  • $\begingroup$ What's going on with "decide if $I$ is principal" ? $\endgroup$ – user26857 Nov 30 '16 at 14:18
  • $\begingroup$ @user26857 I have a hint, that I should use map $F:a+b\sqrt{10}\mapsto a^2-10b^2$. Suppose $I=(a+b\sqrt{10})$. If $c+d\sqrt{10}\in I$, then there is $x+y\sqrt{10}\in\mathbb{Z}[\sqrt{10}]$ such, that $c+d\sqrt{10}=(x+y\sqrt{10})(a+b\sqrt{10})$ and applying $F$ I should have $a^2-10b^2\mid c^2-10d^2$, right? $\endgroup$ – byk7 Nov 30 '16 at 14:31
  • 1
    $\begingroup$ Hint $\ 10\equiv 6^2\pmod{13}.\, $ See my answer if it is not clear where that leads. $\endgroup$ – Bill Dubuque Nov 30 '16 at 15:05
  • $\begingroup$ @user26857 I don't know wheter $I$ is principle or not. So I suppose it is. (if it is principle indeed i want to find generator, if not I should obtain contradiction). Let be $I=(a+b\sqrt{10})$. If $c+d\sqrt{10}\in I$ then there are $x,y\in\mathbb{Z}$ such, that $c+d\sqrt{10}=(x+y\sqrt{10})(a+b\sqrt{10})$. Now I would like use $F:a+b\sqrt{10}\mapsto a^2-10b^2$ somehow. $F$ should be an analogy to similar situation, when we researched subsets of rings $\mathbb{Z}[i]$ and $\mathbb{Z}[i\sqrt{5}]$. We used there a map $g:\mathbb{C}\to[0,\infty),z\mapsto|z|^2$, so we obtained divisibility condition $\endgroup$ – byk7 Nov 30 '16 at 20:28
  • $\begingroup$ @user26857 Ok, so $I=\left(13,\sqrt{10}-6\right)=\left\{(13x-6y)+y\sqrt{10}\ |\ x,y\in\mathbb{Z}\right\}$. If $I=\left(a+b\sqrt{10}\right)$, then $a^2+10b^2\mid(13x-6y)^2+10y^2,\forall x,y\in\mathbb{Z}$? If so, could you explain me this a little bit more, please? It's not clear to me... $\endgroup$ – byk7 Nov 30 '16 at 22:05
4
$\begingroup$

If you have proved that $I$ is an ideal, then you have proved that the map $$f:\mathbb{Z}[\sqrt{10}]\to\mathbb{Z}_{13} \mbox{ given by } a+b\sqrt{10}\mapsto[2a-b]_{13} $$ is a ring homomorphism with kernel $I$.

By considering the elements with $a=0$, it is clear that $f$ is surjective.

What does this tell you about $f$ ?

$\endgroup$
  • $\begingroup$ Something like there is a homomorphism g: Z[sqrt(10)] / I → Z/13Z, such that g∘p=f, where p is projection Z[sqrt(10)] → Z[sqrt(10)] / I, and g is surjection as well? $\endgroup$ – byk7 Nov 30 '16 at 13:40
  • 1
    $\begingroup$ @byk7: You're overthinking it a bit. What does the First Isomorphism Theorem let you conclude about $I,$ given that it is the kernel of the ring homomorphism $f$ mapping onto the field $\Bbb Z_{13}$? $\endgroup$ – Cameron Buie Nov 30 '16 at 13:48
  • $\begingroup$ @CameronBuie oh, you mean, that $\mathbb{Z}[\sqrt{10}] / I$ is isomorphic to $\mathbb{Z}_{13}$? Or that $I=\ker f$? $\endgroup$ – byk7 Nov 30 '16 at 13:52
  • 1
    $\begingroup$ @byk7: The isomorphism part. What does that tell you about $I$, since $\Bbb Z_{13}$ is a field? $\endgroup$ – Cameron Buie Nov 30 '16 at 13:58
  • 2
    $\begingroup$ @CameronBuie yea, just got it... since $\mathbb{Z}[\sqrt{10}]/I\cong\mathbb{Z}_{13}$ and $\mathbb{Z}_{13}$ is a field, then $I$ is prime as well as maximal... $\endgroup$ – byk7 Nov 30 '16 at 14:02
1
$\begingroup$

Hint $\,\ 13\mid 2a\!-\!b\iff 13\mid 2(a\!+\!6b)\iff 13\mid a\!+\!6b\iff a+b\sqrt{10}\in (\sqrt{10}\!-\!6,\,13)$

$\endgroup$
  • $\begingroup$ ok, so there are at most two generators of $I$, but what to do now, please? $\endgroup$ – byk7 Dec 1 '16 at 23:52
  • $\begingroup$ @byk7 What part(s) are you stuck on? $\endgroup$ – Bill Dubuque Dec 2 '16 at 0:13
  • 2
    $\begingroup$ @byk7 Let $\,w = 6\!-\!\sqrt{10}.\,$ If $\,(w,13)=(v)\,$ then $\,v\mid w\,$ so $\,v'\mid w'\,$ so $\,vv'\mid ww'=26.\,$ Also $\,v\mid 13\,$ so $\,v'\mid 13'=13\,$ so $\,vv'\mid 13^2,\,$ so $\,vv'\mid (26,13^2)=13(2,13)=13.\, $ Thus if $\,v = a+b\sqrt{10}\,$ then $\,vv'=a^2\!-10b^2 = \pm 13.\,$ But that has no integer solutions, being unsolvable mod $10.\ \ $ $\endgroup$ – Bill Dubuque Dec 2 '16 at 0:34
  • 1
    $\begingroup$ @byk7 Btw, proving the nontrivial direction $(\Leftarrow)$ of the final arrow in the answer is very simple via this criterion. $\endgroup$ – Bill Dubuque Dec 3 '16 at 22:40
  • 1
    $\begingroup$ @byk7 $\, w = vu\,\Rightarrow\,w' = (vu)' = v'u'\,$ since conjugation is a ring hom. $\endgroup$ – Bill Dubuque Dec 7 '16 at 14:32

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.