# Math Help - Closed sets

1. ## Closed sets

Find a counterexample of:

the union of the closure of A_i = the closure of the union of A_i

I hope its understandable

2. Originally Posted by kezman
Find a counterexample of:

the union of the closure of A_i = the closure of the union of A_i

I hope its understandable
How is closure defined here?

This is my understanding.

Definition: Given a set $S\subseteq \mathbb{R}^2$ we define $\bar S$ the "closure of $S$" to be the set of all points such that $\forall s\in S \ \exists \delta>0 \ \ N(s,\delta)\cap S\not = \{ \}$.

Note: The notation $N(s, \delta)$ is the open disk cented at $s$ with radius $\delta$.

3. Consider the set $A_n = \left( {\frac{1}{{1 + n}},1} \right]$ the closure of which is $\overline {A_n } = \left[ {\frac{1}{{1 + n}},1} \right]$.

But consider this: $\overline {\bigcup\limits_{n = 1}^\infty {A_n } } = \left[ {0,1} \right]\quad \mbox{but}\quad \bigcup\limits_{n = 1}^\infty {\overline {A_n } } = \left( {0,1} \right]$.

4. Originally Posted by Plato
Consider the set $A_n = \left( {\frac{1}{{1 + n}},1} \right]$ the closure of which is $\overline {A_n } = \left[ {\frac{1}{{1 + n}},1} \right]$.

But consider this: $\overline {\bigcup\limits_{n = 1}^\infty {A_n } } = \left[ {0,1} \right]\quad \mbox{but}\quad \bigcup\limits_{n = 1}^\infty {\overline {A_n } } = \left( {0,1} \right]$.
Beautiful conterexample.

I was trying to come up with a conterexample in $\mathbb{C}$ using the definition I posed above (same idea) but I used finite number of sets. And I forgot the important rule of analysis (or topology here): what works for finite sets does not necessarily work for infinite sets.

So I am guessing if $S=\{A_i | i \in I\}$ was such that $|S|<\aleph_0$ then the above would actually be true. Correct? Maybe this is consequence of Heine-Borel theorem, no?

5. The example works great thanks.
Now im trying to find a counterexample of finite unions (that was a problem of my latest exam).

6. Originally Posted by kezman
Now im trying to find a counterexample of finite unions.
The nature of a standard metric topology makes it impossible to find such an example. However, I quite sure that one could construct a general (probability a finite) topological space where that property is true. I just not have thought about it in a very long time. Good Luck!