Compactness: Proving [a, B] Is Compact In Real Numbers

Hey guys! Let's dive into a fundamental concept in real analysis and general topology: compactness. Specifically, we're going to explore why every closed and bounded interval $[a, b]$ in the real numbers $\mathbb{R}$ is compact. This is a crucial result, and while you might know it, the proofs can sometimes feel a bit elusive. So, I'm going to walk you through a few ways to demonstrate this important theorem. Get ready to boost your understanding of compactness! So, what does it even mean for a set to be compact? In the context of real numbers (and more generally, metric spaces), compactness is often defined using open covers. A set $K$ is compact if every open cover of $K$ has a finite subcover. An open cover of $K$ is a collection of open sets whose union contains $K$, and a finite subcover is a finite subset of that collection whose union still contains $K$. This definition might sound a bit abstract, but don't worry, we'll make it concrete with our proofs. Another way to think about compactness, especially in $\mathbb{R}$, is through the Heine-Borel theorem, which directly connects closedness, boundedness, and compactness. The Heine-Borel theorem states that a subset of $\mathbb{R}^n$ is compact if and only if it is closed and bounded. So, for our interval $[a, b]$, being closed and bounded is equivalent to being compact. This gives us a powerful tool for proving compactness, but we still need to understand why closed and bounded implies compactness. There are several approaches to proving that closed bounded intervals are compact, and each offers a slightly different perspective. Let's explore some of the most common and insightful methods.

Proof 1: Using the Bolzano-Weierstrass Theorem

The Bolzano-Weierstrass Theorem is a cornerstone of real analysis, and it provides a neat way to prove the compactness of closed bounded intervals. The theorem states that every bounded sequence in $\mathbb{R}$ has a convergent subsequence. How does this help us prove compactness? Let's break it down. Suppose we have an open cover of $[a, b]$, say $\{U_i\}_{i \in I}$, where $I$ is some index set (possibly uncountable). We want to show that we can find a finite subset of these $U_i$ that still covers $[a, b]$. Now, assume, for the sake of contradiction, that no finite subcover exists. This means that for any finite collection of open sets from our cover, their union will not completely cover $[a, b]$. We can use this assumption to construct a sequence in $[a, b]$ that will lead us to a contradiction. Here’s how we do it: Divide the interval $[a, b]$ into two equal subintervals, $[a, \frac{a+b}{2}]$ and $[\frac{a+b}{2}, b]$. Since no finite subcover of $\{U_i\}_{i \in I}$ covers $[a, b]$, at least one of these subintervals must also lack a finite subcover. Let's call that subinterval $I_1$. Now, divide $I_1$ into two equal subintervals. Again, at least one of these must lack a finite subcover. Call that subinterval $I_2$. We continue this process, creating a sequence of nested intervals $I_1 \supseteq I_2 \supseteq I_3 \supseteq \dots$, where each $I_n$ has length $\frac{b-a}{2^n}$ and cannot be covered by any finite subcollection of the $U_i$. Now, choose a point $x_n$ from each interval $I_n$. This gives us a sequence $(x_n)$ in $[a, b]$. Since $[a, b]$ is bounded, the sequence $(x_n)$ is also bounded. By the Bolzano-Weierstrass Theorem, there exists a convergent subsequence $(x_{n_k})$ that converges to some limit $x$ in $\mathbb{R}$. Because the intervals are nested and their lengths shrink to zero, this limit $x$ must actually lie within $[a, b]$. Thus, $x \in [a, b]$. Since $x \in [a, b]$ and $\{U_i\}_{i \in I}$ is an open cover of $[a, b]$, there must exist some open set $U_j$ in the cover such that $x \in U_j$. Because $U_j$ is open, there exists some $\epsilon > 0$ such that the open interval $(x - \epsilon, x + \epsilon)$ is contained in $U_j$. Since $x_{n_k}$ converges to $x$, there exists some $K$ such that for all $k > K$, $|x_{n_k} - x| < \frac{\epsilon}{2}$. Also, since the length of the interval $I_{n_k}$ shrinks to zero, we can choose $k$ large enough such that the length of $I_{n_k}$ is less than $\frac{\epsilon}{2}$. This means that $I_{n_k}$ is contained in $(x - \epsilon, x + \epsilon)$, and therefore, $I_{n_k} \subseteq U_j$. But this contradicts our assumption that $I_{n_k}$ cannot be covered by any finite subcollection of the $U_i$. We have shown that if no finite subcover exists, we arrive at a contradiction. Therefore, our initial assumption must be false, and every open cover of $[a, b]$ must have a finite subcover. Hence, $[a, b]$ is compact.

Proof 2: Using the Lebesgue Number Lemma

The Lebesgue Number Lemma provides another elegant way to prove the compactness of $[a, b]$. To use this lemma, we first need to understand what a Lebesgue number is. Given an open cover $\{U_i\}_{i \in I}$ of a metric space $X$, a Lebesgue number $\delta > 0$ is a number such that any subset of $X$ with diameter less than $\delta$ is contained entirely within one of the open sets $U_i$. The Lebesgue Number Lemma states that if $X$ is a compact metric space, then every open cover of $X$ has a Lebesgue number. Conversely, if a metric space has the property that every open cover has a Lebesgue number, then the space is compact. So, how do we use this to prove that $[a, b]$ is compact? Let $\{U_i\}_{i \in I}$ be an open cover of $[a, b]$. We want to show that this cover has a Lebesgue number. Assume, for the sake of contradiction, that no such Lebesgue number exists. This means that for every $\delta > 0$, there exists a subset of $[a, b]$ with diameter less than $\delta$ that is not contained in any single $U_i$. In particular, for each positive integer $n$, consider $\delta_n = \frac{1}{n}$. Since $\frac{1}{n} > 0$, there must exist a subset $A_n$ of $[a, b]$ with diameter less than $\frac{1}{n}$ that is not contained in any single $U_i$. Choose a point $x_n$ from each $A_n$. This gives us a sequence $(x_n)$ in $[a, b]$. Since $[a, b]$ is bounded, the sequence $(x_n)$ is also bounded. By the Bolzano-Weierstrass Theorem, there exists a convergent subsequence $(x_{n_k})$ that converges to some limit $x$ in $\mathbb{R}$. Because $[a, b]$ is closed, $x$ must lie within $[a, b]$. Thus, $x \in [a, b]$. Since $x \in [a, b]$ and $\{U_i\}_{i \in I}$ is an open cover of $[a, b]$, there must exist some open set $U_j$ in the cover such that $x \in U_j$. Because $U_j$ is open, there exists some $\epsilon > 0$ such that the open interval $(x - \epsilon, x + \epsilon)$ is contained in $U_j$. Since $x_{n_k}$ converges to $x$, there exists some $K$ such that for all $k > K$, $|x_{n_k} - x| < \frac{\epsilon}{2}$. Also, since the diameter of $A_{n_k}$ is less than $\frac{1}{n_k}$, we can choose $k$ large enough such that $\frac{1}{n_k} < \frac{\epsilon}{2}$. Now, consider the set $A_{n_k}$. We know that its diameter is less than $\frac{\epsilon}{2}$, and $x_{n_k}$ is a point in $A_{n_k}$. Therefore, every point in $A_{n_k}$ is within a distance of $\frac{\epsilon}{2}$ of $x_{n_k}$. Since $x_{n_k}$ is within a distance of $\frac{\epsilon}{2}$ of $x$, every point in $A_{n_k}$ is within a distance of $\epsilon$ of $x$. This means that $A_{n_k}$ is contained in $(x - \epsilon, x + \epsilon)$, and therefore, $A_{n_k} \subseteq U_j$. But this contradicts our assumption that $A_{n_k}$ is not contained in any single $U_i$. We have shown that if no Lebesgue number exists, we arrive at a contradiction. Therefore, our initial assumption must be false, and every open cover of $[a, b]$ must have a Lebesgue number. Since every open cover of $[a, b]$ has a Lebesgue number, $[a, b]$ is compact.

Proof 3: Direct Proof Using the Definition of Compactness

This approach directly uses the definition of compactness, which, as we said earlier, states that every open cover has a finite subcover. While more involved, it provides a solid understanding of the concept. Suppose $\{U_i\}_{i \in I}$ is an open cover of $[a, b]$. Let $S$ be the set of all $x \in [a, b]$ such that the interval $[a, x]$ can be covered by finitely many of the $U_i$. In other words, $S = \{x \in [a, b] : [a, x] \text{ has a finite subcover from } \{U_i\}_{i \in I}\}$. Our goal is to show that $b \in S$, which would mean that $[a, b]$ has a finite subcover and is therefore compact. First, note that $S$ is non-empty. Since $a \in [a, b]$, and $\{U_i\}_{i \in I}$ is an open cover, there exists some $U_j$ such that $a \in U_j$. Since $U_j$ is open, there exists some $\epsilon > 0$ such that $(a - \epsilon, a + \epsilon) \subseteq U_j$. Therefore, for any $x$ such that $a \le x < a + \epsilon$, the interval $[a, x]$ is contained in $U_j$, and thus has a finite subcover (just $U_j$ itself). This means that all such $x$ are in $S$, so $S$ is non-empty. Next, we show that $S$ is bounded above. Since $S$ is a subset of $[a, b]$, it is bounded above by $b$. Now, let $c = \sup S$, the least upper bound of $S$. Since $S$ is bounded above by $b$, we know that $c \le b$. We want to show that $c \in S$ and that $c = b$. Since $c \in [a, b]$ (because $c$ is the supremum of a subset of $[a, b]$), there exists some $U_k$ in the open cover such that $c \in U_k$. Because $U_k$ is open, there exists some $\epsilon > 0$ such that $(c - \epsilon, c + \epsilon) \subseteq U_k$. Since $c$ is the supremum of $S$, there must exist some $x \in S$ such that $c - \epsilon < x \le c$. Because $x \in S$, the interval $[a, x]$ can be covered by finitely many of the $U_i$, say $U_{i_1}, U_{i_2}, \dots, U_{i_n}$. Now, consider the interval $[a, c]$. We can cover it by the open sets $U_{i_1}, U_{i_2}, \dots, U_{i_n}, U_k$. The union of these finitely many open sets covers $[a, x] \cup (c - \epsilon, c + \epsilon)$, which includes the entire interval $[a, c]$. Therefore, $[a, c]$ has a finite subcover, which means that $c \in S$. Finally, we show that $c = b$. Suppose, for the sake of contradiction, that $c < b$. Then, consider the interval $(c, c + \epsilon)$. Since $(c - \epsilon, c + \epsilon) \subseteq U_k$, any $y$ in $(c, c + \epsilon)$ is also in $U_k$. This means that we can extend our finite subcover of $[a, c]$ to cover $[a, y]$ for some $y > c$. But this contradicts the fact that $c$ is the least upper bound of $S$, because we've found an element $y > c$ that is also in $S$. Therefore, our assumption that $c < b$ must be false, and we must have $c = b$. Since $c = b$ and $c \in S$, we know that $b \in S$. This means that $[a, b]$ can be covered by finitely many of the $U_i$, and therefore, $[a, b]$ is compact.

Conclusion

So there you have it, folks! We've explored three different ways to prove that every closed and bounded interval $[a, b]$ in $\mathbb{R}$ is compact. Each proof leverages different tools from real analysis and topology, giving us a deeper appreciation for the concept of compactness. Whether you prefer the elegance of the Bolzano-Weierstrass Theorem, the power of the Lebesgue Number Lemma, or the direct approach using the definition of compactness, you now have a solid arsenal of techniques to tackle this fundamental result. Keep exploring, keep questioning, and keep proving! You're now well-equipped to understand and appreciate why closed bounded intervals are compact. Keep this knowledge in your toolkit—it's sure to come in handy!