1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Showing a Set is not Connected

  1. Oct 21, 2016 #1
    1. The problem statement, all variables and given/known data
    I am trying to show that ##X = \{z : |z| \le 1 \} \cup \{z : |z-2|<1\} \subset \mathbb{C}## is a connected set.

    2. Relevant equations
    Definition of connectedness that I am working with: A metric space ##(X,d)## is connected if the only subsets of ##X## that are both open and closed are ##\emptyset## and ##X##. If ##A \subseteq X##, then ##X## is connected if ##(A,d)## is connected

    3. The attempt at a solution
    I have been staring at this problem for quite some time and I don't really have much:

    Let ##C \subseteq X## be nonempty and both closed and open. I want to show that ##X \subseteq C##. Let ##z \in X = C \cup (X-C)##. Then either ##z \in C##, in which case we are done, or ##z \in (X-C)##...I would like to show that the latter case implies a contradiction, but I am having difficulty.
     
    Last edited: Oct 21, 2016
  2. jcsd
  3. Oct 21, 2016 #2
    According to my book, they have to be disjoint open sets in order to conclude that ##X## is disconnected. The first set in the union is not open, if I am not mistaken.
     
  4. Oct 21, 2016 #3

    Mark44

    Staff: Mentor

    You are correct -- both sets have to be open. I have deleted my earlier post.
     
  5. Oct 21, 2016 #4

    Mark44

    Staff: Mentor

    Shouldn't this be " A metric space ##(X,d)## is connected if the only subsets of ##X## that are both open and closed are ##\emptyset## and ##X##."?
    X clearly has two subsets: {z : |z| ≤ 1} and {z | |z - 2| < 1}, but the first set is closed and the second set is open. Neither set is both open and closed.

    I'm sure you already know this -- I'm just trying to clarify your statement above.
     
  6. Oct 21, 2016 #5
    Whoops! I fixed it.
     
  7. Oct 21, 2016 #6

    Mark44

    Staff: Mentor

    Your set C is clopen iff its boundary is empty. Can you use this to show that ##z \in X - C## is a contradiction?
     
  8. Oct 21, 2016 #7

    fresh_42

    Staff: Mentor

    The usual definition goes by separation. Two sets ##A\, , \,B\subseteq X## are called separated in ##X## if ##\overline{A} \cap B = \emptyset ## and ##A \cap \overline{B} = \emptyset ##. Now ##X## is connected, if it cannot be written as union of two non-empty, separated sets.

    So what you actually want to use is the equivalence of these two definitions: Your ##X## is connected, since the obvious choice of ##A## and ##B## are not separated (the second condition does not hold) and all other choices clearly neither.
    ???

    If ##A \subseteq X##, then ##A## is connected if ##(A,d)##, i.e. ##A## with the induced topology from ##X## is connected.
    However, one doesn't really need to use the detour of the induced topology.

    Anyway, back to your proof. I cannot see, how the consideration of a single element ##z## could help, since we are talking about global properties.
    Wouldn't it be more natural to consider ##A \cap C## and ##B \cap C## with the assumed clopen set ##C## and ##A=\{z:|z|≤1\}## and ##B=\{z:|z−2|<1\}##? Are they clopen as well? And what does this mean for ##X=A \cup B \;##?
     
  9. Oct 23, 2016 #8
    I don't see the contradiction. Let ##\partial C := \overline{C} \cap (\overline{X-C})## be the boundary (this is how my book defines it). If ##C## is clopen, then ##\overline{C} = C## and ##int ~C=C##, and therefore ##\partial C = C \cap (\overline{X-C})##. Since ##z \in X-C##, then ##z \in \overline{X-C}##; and since ##z \notin C = \overline{C}##, we can say ##z \notin \partial C##. But can conclude neither that ##\partial C## is empty nor nonempty.
     
  10. Oct 23, 2016 #9
    Hold on! Wouldn't that say no point ##z## in ##X## is in ##\partial C##, and this would be the contradiction sought after?

    What worries me is that the proof doesn't use any specific information about the metric space ##X## and the subsets ##A## and ##B##.
     
  11. Oct 23, 2016 #10

    fresh_42

    Staff: Mentor

    Probably because one doesn't need neither a metric nor the induced topology. Both are only used to describe the sets ##A## and ##B## in this example. For the properties in question, they aren't necessary.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Showing a Set is not Connected
  1. Connected Sets. (Replies: 12)

  2. Connected sets (Replies: 3)

Loading...