2022-08-21

116: Some Properties About Adjunction Topological Space When Inclusion to Attaching-Origin Space from Subset Is Closed Embedding

<The previous article in this series | The table of contents of this series | The next article in this series>

A description/proof of some properties about adjunction topological space when inclusion to attaching-origin space from subset is closed embedding

Topics


About: topological space

The table of contents of this article


Starting Context



Target Context


  • The reader will have a description and a proof of some properties concerning any adjunction topological space when the inclusion to the attaching-origin space from the subset is a closed embedding.

Orientation


There is a list of definitions discussed so far in this site.

There is a list of propositions discussed so far in this site.


Main Body


1: Description


For any topological spaces, T1,T2, any subset, ST1, any continuous map, f:ST2, the adjunction topological space, T2fT1, f1:ST1, f2:T1T2fT1, f3:T2T2fT1, and f4:T1+T2T2fT1, when f1 is a closed embedding, these properties hold: 1) f3 is a closed embedding; 2) f2|T1S:T1ST2fT1 is an open embedding; 3) if T1 and T2 are spaces such that 3-1) each 1 point subset is closed or 3-2) T1 and T2 are normal, T2fT1 satisfies 3-1) or 3-2); 4) if f is a quotient map, f2 is a quotient map.


2: Proof


As for 1), by the proposition that for any adjunction topological space, the canonical map from the attaching-destination space to the adjunction space is a continuous embedding, f3 is an embedding. For any closed set, CT2, C=CC where C=Cf(S) and C=CC. f3(C)=f3(C)f3(C)={[p]T2fT1|pC,[p]=f1(p){p}}C where the last C is of course not really exactly C but the set of equivalent classes such that each equivalent class consists of the corresponding single point in C, which is [p]={p}, but is denoted so because introducing another expression seems to complicate notations. Note that similar expressions are used henceforth.

Is f3(C) closed? By the definition of quotient topology, its closed-ness is nothing but the closed-ness of f41(f3(C)), whose closed-ness is nothing but the closed-ness of f41(f3(C))T1 and the closed-ness of f41(f3(C))T2, by the definition of topological sum. But f41(f3(C))=f41({[p]T2fT1|pC,[p]=f1(p){p}})f41(C)=f1(C)CC=f1(C)C by the proposition that for any map, the map preimage of any union of sets is the union of the map preimages of the sets. f41(f3(C))T1=f1(C) is closed on S because f is continuous; f1(C)=f1(f1(C)) is closed on T1 because f1 is closed. f41(f3(C))T2=C is closed.

As for 2), f2 is continuous, because f2:T1T1+T2T2fT1 is a compound of continuous maps (the former is an inclusion and the latter is f4). By the proposition that any restriction of any continuous map on the domain and the codomain is continuous, f2|T1S is continuous, and is obviously injective. f2|T1S:T1Sf2|T1S(T1S) is obviously a continuous bijection. For any open set, UT1S, f2|T1S(U)=U, which is open on f2|T1S(T1S) if there is an open set, UT2fT1, such that U=Uf2|T1S(T1S), but U=U will do, because U is open on T2fT1, because the openness of U is nothing but the openness of f41(U)T1=U and the openness of f41(U)T2=, which hold true. So, f2|T1S is a homeomorphism, so, f2|T1S is an embedding. f2|T1S(U)=U, which is open on T2fT1, as has been already shown. So, it is an open embedding.

As for 3) with 3-1), suppose that each 1 point subset on T1 and on T2 is closed. Any 1 point set on T2fT1 is {{p}} where pT1S or pT2f(S) or {[p]} where pf(S) and [p]=f1(p){p}. The former is obviously closed. As for the latter, f41([p])=f1(p){p}, so, f41([p])T1=f1(p) is closed on S as f is continuous, but as f1 is closed, S=f1(S) is closed on T1, so, f41([p])T1 is closed on T1 by the proposition that any closed set on any closed subspace topological space is closed on the base topological space. f41([p])T2={p}, closed.

As for 3) with 3-2), suppose that T1 and T2 are normal. By the reverse of Tietze extension theorem, if for any closed set, CT2fT1, and any continuous map, g:CR, there is a continuous extension, g:T2fT1R,g|C=g, T2fT1 will be normal.

S is closed on T1, because S=f1(S) and f1 is closed while S is closed on S. f21(C)=f41(C)T1 and f31(C)=f41(C)T2 are closed by the definition of quotient topological space and the definition of topological sum.

1st, suppose that f31(C)=T2, which implies that Sf21(C). h:f21(C)C,h=f2|f21(C), is continuous by the proposition that any restriction of any continuous map on the domain and the codomain is continuous. gh:f21(C)R is continuous as a composition of continuous maps. As T1 is normal, by the Tietze extension theorem, there is a continuous map, i:T1R,i|f21(C)=gh. Let us define g by gf2=i on f2(T1) and gf3=gf3 on f3(T2), which is well-defined because f2(T1) and f3(T2) cover T2fT1; for any pT1f21(C), gf2(p)=g([p])=i(p), but pT1S as Sf21(C), so, g({p})=i(p), and for any pf21(C), gf2(p)=g([p])=i|f21(C)(p)=g(h(p))=g([p]); for any pT2, gf3(p)=g([p])=gf3(p)=g([p]); g does not contradict for S and for f(S), because g=g there anyway.

Is g continuous? That is about that for any open set, UR, whether g1(U) is open. The openness is nothing but the openness of f41(g1(U))T1=f21(g1(U)) and the openness of f41(g1(U))T2=f31(g1(U)). The former is open because gf2=i is continuous. As for the latter, f31(g1(U))=f31(g1(U)C) because f3 maps into C anyway by the supposition, f31(C)=T2, but as g|C=g, g1(U)C=g|C1(U)=g1(U) by the proposition that for any map between sets and its any domain-restriction, the preimage under the domain-restricted map is the intersection of the preimage under the original map and the restricted domain. So, f31(g1(U))=f31(g1(U)), which is open as a preimage of an open set under a composition of continuous maps.

Now, suppose that f31(C)T2. j:f31(C)C,j=f3|f31(C), is continuous by the proposition that any restriction of any continuous map on the domain and the codomain is continuous. gj:f31(C)R is continuous as a composition of continuous maps. As T2 is normal, by the Tietze extension theorem, there is a continuous map, k:T2R,k|f31(C)=gj. f3(T2) is closed because f41(f3(T2))T1=S and f41(f3(T2))T2=T2 are closed on T1 and T2. Let us define g:f3(T2)CR by gf3=k on f3(T2) and g=g on C. It is well-defined because f3(T2) and C cover f3(T2)C; for any pT2, gf3(p)=g([p])=k(p) but p[p] is injective; g does not contradict on f3(T2)C because k|f31(C)=gj.

Is g continuous? {f3(T2),C} is a finite closed cover of f3(T2)C, and if g|f3(T2) and g|C are continuous, g will be continuous by the proposition that any map between topological spaces is continuous if the domain restriction of the map to each closed set of a finite closed cover is continuous. The latter is continuous, as it equals g. For the former, it is the matter of that for any UR, whether g|f3(T2)1(U) is open on f3(T2) (remember that it is not on T2fT1). f41(g|f3(T2)1(U))T1=f1(f31(g|f3(T2)1(U)))=f1(k1(U)), open on S because g|f3(T2)f3=k is continuous and f is continuous, which means that there is an open set, UT1 such that US=f41(g|f3(T2)1(U))T1. S=f41(f3(T2))T1. f41(g|f3(T2)1(U))T2=f31(g|f3(T2)1(U))=k1(U):=U, open on T2. US=f1(U)=f1(Uf(S)). By the proposition that for any adjunction topological space and its any subspace, any subset of the subspace is open if and only if the projections of the preimage of the subset under the quotient map onto the attaching-origin space and the attaching-destination space are open on the projections of the preimage of the subspace under the quotient map with the condition that the attaching-origin space projection accord with the attaching-destination space projection with respect to the attaching map, as U and U satisfy the necessary conditions, g|f3(T2)1(U) is open on f3(T2). So, g is continuous.

Now, f31(f3(T2)C)=T2, the 1st case with g instead of g, so, there is a continuous extension, g:T2fT1R, which is a continuous extension of g.

So, by the reverse of Tietze extension theorem, T2fT1 is normal.

As for 4), suppose that f is a quotient map. f2 is a continuous surjection. For any subset, ST2fT1, suppose that f21(S) is open. Is S open? The openness is nothing but the openness of f41(S)T1 and the openness of f41(S)T2. The former is nothing but f21(S), so, open. As for the latter, f1(f41(S)T2)=f21(S)S, which is open in the S subspace topology as f21(S) is open on T1, so, as f is a quotient map, f41(S)T2 is open.


References


<The previous article in this series | The table of contents of this series | The next article in this series>