Let $(R,m)$ be a local commutative integral domain. Are the residue field $k= R/m$ and the quotient field of $R$ are the same? I would think they are since $m$ is the only maximal ideal. What if $R$ is not an integral domain?


Consider a commutative local ring $(R, \mathfrak m).$ By definition, the residue field of $R$ is given by $k = R / \mathfrak m.$ On the other hand, the total ring of fractions $Q(R)$ is given by $$Q(R) = S^{-1} R = \biggl\{\frac r s \,\bigg|\, r \in R, \, s \in S, \text{ and } \frac r s = \frac{r'}{s'} \text{ iff } \exists \, s'' \in S \text{ s.t. } s''(s'r - r's) = 0 \},$$ where we have that $S = R - \{r \in R \,|\, rs = 0 \text{ for some nonzero } s \in R \}.$ Observe that $S$ is the multiplicatively closed set that consists of the elements of $R$ that are neither zero divisors nor $0.$

Purely in terms of their definitions, these two rings are not the same as sets: the residue field consists of equivalence classes of elements modulo the unique maximal ideal $\mathfrak m,$ and the total ring of fractions $Q(R)$ consists of fractions of elements in $R$ whose denominator is neither a zero divisor nor $0.$ But there are more properties that distinguish these two rings in general.

1.) Of course, the residue field is always a field, but the total ring of fractions need not be a field. Given that $R$ is Artinian (i.e., $R$ is Noetherian and $\dim R = 0$) but not a field, the maximal ideal $\mathfrak m$ consists of all zero divisors of $R.$ (Indeed, every prime ideal of $R$ is maximal, hence $\mathfrak m$ is the unique prime ideal of $R$ so that $\mathfrak m$ is the unique minimal prime ideal of $R,$ from which it follows that $\mathfrak m$ is precisely the set of zero divisors of $R.$) Consequently, we have that $S = R - \mathfrak m$ so that $Q(R) = S^{-1} R =$ $R_{\mathfrak m} \cong R$ is not a field by assumption.

2). Every field is reduced (i.e., the only nilpotent element of a field is $0$); however, the total ring of fractions need not be reduced. Consider the ring $R = \mathbb Z / 4 \mathbb Z.$ Observe that $2$ is the only zero divisor of $R,$ hence we have $S = \{0, 2 \}.$ We have therefore that $Q(R) = \bigl \{0, 1, 2, 3, \frac 1 3, \frac 2 3 \bigr \}$ is not reduced, as $\frac 2 3$ is nilpotent. Particularly, we have that $\bigl(\frac 2 3 \bigr)^2 = 0.$

3.) Every field is an integral domain; however, the total ring of fractions need not be a domain. Particularly, the zero divisors of $R$ give rise to zero divisors of $Q(R).$ Explicitly, given that $ab = 0$ in $R,$ we have that $0 = ab = 1(1 \cdot ab - 0 \cdot ss'),$ from which it follows that $\frac a s \cdot \frac b {s'} = \frac{ab}{ss'} = \frac 0 1.$

Given that $R$ is a domain, these are non-issues. Explicitly, we have that $S = R - \{0\}$ so that $Q(R)$ coincides with the field of fractions $\operatorname{Frac}(R)$; that immediately does away with points (1.) and (3.). Further, if $R$ is reduced, then $Q(R)$ is reduced, hence (2.) is irrelevant. But it remains to be seen how the residue field $k = R / \mathfrak m$ and the field of fractions $\operatorname{Frac}(R)$ relate in this case.

Like you mentioned, there are a few things we can say.

1.) We have a short exact sequence $R \to k \to 0,$ i.e., $R \to k$ is surjective.

2.) We have a short exact sequence $0 \to R \to \operatorname{Frac}(R),$ i.e., $R \to \operatorname{Frac}(R)$ is injective.

3.) There exists a commutative local domain $(R, \mathfrak m)$ such that $R / \mathfrak m \cong \operatorname{Frac}(R).$


I think you may be confusing two concepts. Take the example $R = \mathbb{Z}_p$, the $p$-adic integers. Then $m = p\mathbb{Z}_p$. The residue field $k = R/m$ is the field of $p$ elements, while the quotient field of $R$ is $\mathbb{Q}_p$, the $p$-adic rationals.

  • $\begingroup$ I see. But it should be the case that there's a surjection from the residue field into $R$ and an injection from $R$ into the quotient field, right? $\endgroup$ – FearfulSymmetry May 10 '20 at 20:55
  • $\begingroup$ There is a surjection from $R$ onto the residue field (not the other way around as you stated), and an injection from $R$ into the quotient field. $\endgroup$ – Ted May 10 '20 at 20:56
  • $\begingroup$ Does the residue field inject into the quotient field? $\endgroup$ – FearfulSymmetry May 10 '20 at 21:00
  • $\begingroup$ Did you think about the above example? Does the field of $p$ elements inject into $\mathbb{Q}_p$ ? $\endgroup$ – Ted May 11 '20 at 19:40

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.