Consider in $(\mathbb{R} ,\tau_e)$ ($\tau_e$ is the Euclidean topology) the following subset:

$$X=\{x \in \mathbb{R}:x= \frac{p}{10^q},\; p,q \in \mathbb{Z}\}.$$

Decide if:

(i) $X$ is open in $(\mathbb{R} ,\tau_e)$ and find its interior, and

(ii) $X$ is closed in $(\mathbb{R} ,\tau_e)$ and find its closure.

$X$ is the set formed by rational non-periodic numbers... I think it is not open because every interval in the Euclidean topology contains a periodic rational number, so $\operatorname{Int}(X)=\emptyset$.

To see if $X$ is closed: $C_{\mathbb{R}}(X)=$ periodic rational numbers $\cup$ $\mathbb{R} \setminus \mathbb{Q}$... and is this open? I don't think but I can't show it... so $\operatorname{cl}(X)=X$.


Surely $X$ is not open as $X$ is only countable. Hence its interior is empty.

$X$ is also not closed because it does not contain all its limit points. You can find a sequence of points in $X$ converging to $\pi$, say: $$ 3/1,\,31/10,\,314/100,\,3141/1000,\dots $$

For the closure, note that any dyadic rational number of the form $m/2^n$ for $m,n\in\Bbb Z$ can be expressed as an element of $X$ since $2^n = 10^n/5^n$, so $m/2^n = 5^nm/10^n$. Since the dyadic rational numbers $D$ are dense in $\Bbb R$, and $D\subset X$, we have $\operatorname{cl}(X) = \Bbb R$, i.e., the closure of $X$ is $\Bbb R$.

  • $\begingroup$ Thanks for the answer. Is my motivation for the "interior" right? I didn't know that property of the dyadic ration numbers... can I show that $C_{\mathbb{R}}(X)$ is not open? (I can't do this..) $\endgroup$ – VoB Jul 13 '17 at 17:34
  • $\begingroup$ @feddy What do you mean by $C_\Bbb{R}(X)$? I'm not familiar with the notation. I am not sure how you could prove that every interval in $\Bbb{R}$ contains a periodic rational number, but if you could, and supposing that $X$ truly does consist of non-periodic rationals, then you would have shown that the interior is empty, yes. $\endgroup$ – Alex Ortiz Jul 13 '17 at 17:36
  • $\begingroup$ With that I mean the complementar of $X$ in $\mathbb{R}$. The fact is that I don't know how to prove it. And for the closure... is there a way to show that $C_{\mathbb{R}}(X)=$ is an open set? $\endgroup$ – VoB Jul 13 '17 at 17:39
  • $\begingroup$ @feddy Ok, I understand now. There is no way to show that $C_\Bbb{R}(X)$ is open because this would imply $X$ is closed, which it is not. $\endgroup$ – Alex Ortiz Jul 13 '17 at 19:24

Concerning closure: consider any real number x = b_n...b0 . a1...am... (b's are the digits of the integral part, and a's make the decimal part, possibly an infinite sequence). By picking q = m, p = b_n ... b0 a1 ... am (integer of n+m+1 digits) you can approximate x arbitrarily precisely. Ergo, the closure of the set is the whole R.


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.