4
$\begingroup$

Let $A$ be a finitely generated $K$ algebra and $S \subset A$ be any multiplicative closed set. Then can I say $S^{-1}A$ is also finitely generated $K$ algebra ? In particular my interest is when $A$ is Artinian ring and $S$ is complement of a maximal ideal.

My Try: If $A=K[X_1,\ldots,X_n]/I$ and $S \cap I=\phi $ then $S^{-1}A \cong S^{-1}K[X_1, \ldots,X_n]/ S^{-1}I$ after that I cannot conclude !

$\endgroup$
  • 1
    $\begingroup$ Try $\Bbb C[x]$ and it's field of fractions $\Bbb C(x)$ $\endgroup$ – xsnl Mar 28 '18 at 9:44
  • 1
    $\begingroup$ I guess since there are infinitely many non associate primes in $\mathbb{C}[X],$ $\mathbb{C}(X)$ is not finitely generated $\mathbb{C}[X]$ algebra. $\endgroup$ – user371231 Mar 28 '18 at 9:51
  • 1
    $\begingroup$ Is it true if $A$ is Artinian ring and $S$ is complement of a maximal ideal ? $\endgroup$ – user371231 Mar 28 '18 at 9:52
3
$\begingroup$

Yes, if $A$ is a finitely generated Artinian algebra over $K$ and $S=A\setminus \frak m$ is the complement of a maximal ideal, then $S^{-1}A$ is finitely generated over $K$. Here is why:
If the maximal ideals of $A$ are $\mathfrak m_1=\mathfrak m,\cdots,\mathfrak m_r$ we have $A\cong A/\mathfrak m_1^N\times \cdots A/\mathfrak m_r^N$, where $N$ is a sufficiently large integer [Atiyah-Macdonald, Theorem 8.7], so that $$S^{-1}A\cong S^{-1}(A/\mathfrak m_1^N)\times \cdots S^{-1}(A/\mathfrak m_r^N)\quad (\bigstar) $$ a) Now, for $i\geq 2, S^{-1}(A/\mathfrak m_i^N)=S^{-1}A/S^{-1}\mathfrak m_i^N=0$ since $S^{-1}\mathfrak m_i^N=S^{-1}A$ because $\mathfrak m_i^N$ meets $S$ [ Atiyah-Macdonald, Proposition 3.11 ii)].
b) On the other hand $S^{-1}(A/\mathfrak m_1^N)=A/\mathfrak m_1^N$ because every $s\in S$ has an image $\bar s\in A/\mathfrak m_1^N$ which is already invertible as a consequence of not belonging to $\mathfrak m_1/\mathfrak m_1^N$, the only maximal ideal of the local ring $A/\mathfrak m_1^N$.
Conclusion
In view of a) and b), $(\bigstar)$ implies $S^{-1}A\cong A/\mathfrak m_1^N$, thus proving that $S^{-1}A$ is finitely generated over $K$, because it is a quotient of $A$.

$\endgroup$

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.