Totally disconnected space

Topological space that is maximally disconnected

In topology and related branches of mathematics, a totally disconnected space is a topological space that has only singletons as connected subsets. In every topological space, the singletons (and, when it is considered connected, the empty set) are connected; in a totally disconnected space, these are the only connected subsets.

An important example of a totally disconnected space is the Cantor set, which is homeomorphic to the set of p-adic integers. Another example, playing a key role in algebraic number theory, is the field Qp of p-adic numbers.

Definition

A topological space X {\displaystyle X} is totally disconnected if the connected components in X {\displaystyle X} are the one-point sets.[1][2] Analogously, a topological space X {\displaystyle X} is totally path-disconnected if all path-components in X {\displaystyle X} are the one-point sets.

Another closely related notion is that of a totally separated space, i.e. a space where quasicomponents are singletons. That is, a topological space X {\displaystyle X} is totally separated if for every x X {\displaystyle x\in X} , the intersection of all clopen neighborhoods of x {\displaystyle x} is the singleton { x } {\displaystyle \{x\}} . Equivalently, for each pair of distinct points x , y X {\displaystyle x,y\in X} , there is a pair of disjoint open neighborhoods U , V {\displaystyle U,V} of x , y {\displaystyle x,y} such that X = U V {\displaystyle X=U\sqcup V} .

Every totally separated space is evidently totally disconnected but the converse is false even for metric spaces. For instance, take X {\displaystyle X} to be the Cantor's teepee, which is the Knaster–Kuratowski fan with the apex removed. Then X {\displaystyle X} is totally disconnected but its quasicomponents are not singletons. For locally compact Hausdorff spaces the two notions (totally disconnected and totally separated) are equivalent.

Confusingly, in the literature (for instance[3]) totally disconnected spaces are sometimes called hereditarily disconnected,[4] while the terminology totally disconnected is used for totally separated spaces.[4]

Examples

The following are examples of totally disconnected spaces:

Properties

  • Subspaces, products, and coproducts of totally disconnected spaces are totally disconnected.
  • Totally disconnected spaces are T1 spaces, since singletons are closed.
  • Continuous images of totally disconnected spaces are not necessarily totally disconnected, in fact, every compact metric space is a continuous image of the Cantor set.
  • A locally compact Hausdorff space has small inductive dimension 0 if and only if it is totally disconnected.
  • Every totally disconnected compact metric space is homeomorphic to a subset of a countable product of discrete spaces.
  • It is in general not true that every open set in a totally disconnected space is also closed.
  • It is in general not true that the closure of every open set in a totally disconnected space is open, i.e. not every totally disconnected Hausdorff space is extremally disconnected.

Constructing a totally disconnected quotient space of any given space

Let X {\displaystyle X} be an arbitrary topological space. Let x y {\displaystyle x\sim y} if and only if y c o n n ( x ) {\displaystyle y\in \mathrm {conn} (x)} (where c o n n ( x ) {\displaystyle \mathrm {conn} (x)} denotes the largest connected subset containing x {\displaystyle x} ). This is obviously an equivalence relation whose equivalence classes are the connected components of X {\displaystyle X} . Endow X / {\displaystyle X/{\sim }} with the quotient topology, i.e. the finest topology making the map m : x c o n n ( x ) {\displaystyle m:x\mapsto \mathrm {conn} (x)} continuous. With a little bit of effort we can see that X / {\displaystyle X/{\sim }} is totally disconnected.

In fact this space is not only some totally disconnected quotient but in a certain sense the biggest: The following universal property holds: For any totally disconnected space Y {\displaystyle Y} and any continuous map f : X Y {\displaystyle f:X\rightarrow Y} , there exists a unique continuous map f ˘ : ( X / ) Y {\displaystyle {\breve {f}}:(X/\sim )\rightarrow Y} with f = f ˘ m {\displaystyle f={\breve {f}}\circ m} .

See also

Citations

  1. ^ Rudin 1991, p. 395 Appendix A7.
  2. ^ Munkres 2000, pp. 152.
  3. ^ Engelking, Ryszard (1989). General Topology. Heldermann Verlag, Sigma Series in Pure Mathematics. ISBN 3-88538-006-4.
  4. ^ a b Kuratowski 1968, pp. 151.

References

  • Munkres, James R. (2000). Topology (Second ed.). Upper Saddle River, NJ: Prentice Hall, Inc. ISBN 978-0-13-181629-9. OCLC 42683260.
  • Rudin, Walter (1991). Functional Analysis. International Series in Pure and Applied Mathematics. Vol. 8 (Second ed.). New York, NY: McGraw-Hill Science/Engineering/Math. ISBN 978-0-07-054236-5. OCLC 21163277.
  • Willard, Stephen (2004), General topology, Dover Publications, ISBN 978-0-486-43479-7, MR 2048350 (reprint of the 1970 original, MR0264581)
  • Kuratowski, Kazimierz (1968), Topology II: Transl. from French (Revised ed.), New York: Academic Press [u.a.], ISBN 9780124292024