Calculus I: Rigorous Treatment of Derivative Rules

Moving from simply applying derivative rules to understanding why they work is the transition from computational proficiency to mathematical literacy. For engineers, this deeper understanding is not academic luxury; it is critical for modeling complex, non-standard systems where off-the-shelf rules fail and foundational logic is required to build new solutions. This article reconstructs the core rules of differentiation from the ground up, using the limit definition of the derivative as our sole starting point, empowering you to wield these tools with confidence and creativity.

The Foundation: The Limit Definition

Every proof begins with the definition: for a function $f(x)$, its derivative at a point $x$ is defined as the limit of the difference quotient: $$f^{\prime }(x)=\underset{h\to 0}{\lim }\frac{f(x+h)-f(x)}{h}$$ This definition captures the instantaneous rate of change as the slope of the tangent line. All derivative rules are theorems that follow logically from this definition and the properties of limits. The proofs often involve algebraic manipulation to massage the difference quotient into a form where the limit can be evaluated directly. Mastering this process is essential for verifying rules and, more importantly, for adapting to functions that don't fit standard molds.

Proof of the Power Rule

The Power Rule, stated as $\frac{d}{dx}[x^n]=n{x}^{n-1}$ for any real number $n$, is often taken for granted. For positive integer exponents, we can prove it rigorously using the limit definition and the binomial theorem. Consider $f(x)=x^n$ where $n$ is a positive integer.

We set up the difference quotient: $$f^{\prime }(x)=\underset{h\to 0}{\lim }\frac{(x+h{)}^n-x^n}{h}$$

Expanding $(x+h{)}^n$ using the binomial theorem gives: $$(x+h{)}^n=x^n+n{x}^{n-1}h+(\genfrac{}{}{0ex}{}{n}{2})x^{n-2}{h}^2+...+h^n$$

Substituting this expansion back into the difference quotient: $$f^{\prime }(x)=\underset{h\to 0}{\lim }\frac{[x^n+n{x}^{n-1}h+(\genfrac{}{}{0ex}{}{n}{2})x^{n-2}{h}^2+...+h^n]-x^n}{h}$$

The $x^n$ terms cancel. We then factor an $h$ out of every remaining term in the numerator: $$f^{\prime }(x)=\underset{h\to 0}{\lim }\frac{h[n{x}^{n-1}+(\genfrac{}{}{0ex}{}{n}{2})x^{n-2}h+...+h^{n-1}]}{h}$$

Canceling the common factor of $h$ (valid since $h\to 0$ but $h\ne 0$), we get: $$f^{\prime }(x)=\underset{h\to 0}{\lim }[n{x}^{n-1}+(\genfrac{}{}{0ex}{}{n}{2})x^{n-2}h+...+h^{n-1}]$$

Now, as $h\to 0$, every term except the first contains a factor of $h$ and therefore goes to zero. Thus: $$f^{\prime }(x)=n{x}^{n-1}$$

This elegant proof showcases the power of algebraic structure and the limit process. For rational and real exponents, the proof requires more advanced tools like implicit differentiation or logarithmic differentiation, but the core concept remains.

Derivation of the Product and Quotient Rules

The Product Rule states that the derivative of a product is not simply the product of the derivatives. For functions $u(x)$ and $v(x)$, the rule is: $\frac{d}{dx}[u(x)v(x)]=u^{\prime }(x)v(x)+u(x)v^{\prime }(x)$.

We start from the limit definition for $f(x)=u(x)v(x)$: $$f^{\prime }(x)=\underset{h\to 0}{\lim }\frac{u(x+h)v(x+h)-u(x)v(x)}{h}$$

The key strategy is to subtract and add a helpful term, $u(x+h)v(x)$, in the numerator: $$f^{\prime }(x)=\underset{h\to 0}{\lim }\frac{u(x+h)v(x+h)-u(x+h)v(x)+u(x+h)v(x)-u(x)v(x)}{h}$$

Now, we group terms strategically: $$f^{\prime }(x)=\underset{h\to 0}{\lim }\left[u(x+h)\cdot \frac{v(x+h)-v(x)}{h}+v(x)\cdot \frac{u(x+h)-u(x)}{h}\right]$$

Using the limit laws, we can split this into two limits. As $h\to 0$, we note that $\frac{v(x+h)-v(x)}{h}\to v^{\prime }(x)$, $\frac{u(x+h)-u(x)}{h}\to u^{\prime }(x)$, and critically, $u(x+h)\to u(x)$ because a differentiable function is continuous. Therefore: $$f^{\prime }(x)=u(x)v^{\prime }(x)+v(x)u^{\prime }(x)$$

The Quotient Rule for $f(x)=\frac{u(x)}{v(x)}$ can be derived similarly with clever algebra or by combining the product rule and the chain rule. A direct proof from the limit definition involves a common denominator manipulation. The result is: $$\frac{d}{dx}\left[\frac{u(x)}{v(x)}\right]=\frac{u^{\prime }(x)v(x)-u(x)v^{\prime }(x)}{[v(x){]}^2}$$ Understanding this derivation helps you remember the correct order of terms in the numerator ("derivative of the top times the bottom, minus the top times the derivative of the bottom").

Chain Rule Proof Strategies

The Chain Rule handles derivatives of composite functions: $f(x)=g(u(x))$. The rule is $f^{\prime }(x)=g^{\prime }(u(x))\cdot u^{\prime }(x)$. Its proof is subtle because it involves the change in an intermediate variable. One common strategy defines a new function to elegantly handle a potential division-by-zero issue.

Let $y=u(x)$ and $z=g(y)=f(x)$. The derivative $g^{\prime }(y)$ is defined as ${\lim }_{k\to 0}\frac{g(y+k)-g(y)}{k}$. We want $f^{\prime }(x)={\lim }_{h\to 0}\frac{g(u(x+h))-g(u(x))}{h}$.

Let $k(h)=u(x+h)-u(x)$. Note that as $h\to 0$, $k\to 0$ because $u$ is continuous. If we assume $k\ne 0$ as $h\to 0$, we can write: $$\frac{g(u(x+h))-g(u(x))}{h}=\frac{g(y+k)-g(y)}{k}\cdot \frac{k}{h}=\frac{g(y+k)-g(y)}{k}\cdot \frac{u(x+h)-u(x)}{h}$$ Taking the limit as $h\to 0$ gives $g^{\prime }(y)\cdot u^{\prime }(x)=g^{\prime }(u(x))\cdot u^{\prime }(x)$.

The technical difficulty arises if $u(x+h)-u(x)=0$ infinitely often as $h\to 0$. A rigorous proof circumvents this by defining an auxiliary function. The core insight remains: the rate of change of the composite function is the rate of change of the outer function with respect to its input multiplied by the rate of change of the inner function.

Justification of Implicit Differentiation

Implicit differentiation is not a new rule but a powerful application of the chain rule to an equation that defines a relationship between variables. Consider an equation like $x^2+y^2=25$, where $y$ is implicitly a function of $x$. To differentiate both sides with respect to $x$, we treat $y$ as $y(x)$.

Differentiating term-by-term:

• The derivative of $x^2$ with respect to $x$ is $2x$.
• The derivative of $y^2$ with respect to $x$ requires the chain rule: think of it as $(y(x){)}^2$. The derivative is $2y\cdot \frac{dy}{dx}$.
• The derivative of the constant $25$ is $0$.

This yields: $2x+2y\frac{dy}{dx}=0$. Solving for $\frac{dy}{dx}$ gives $\frac{dy}{dx}=-\frac{x}{y}$. The justification relies entirely on the chain rule and the assumption that the original equation defines a differentiable function $y(x)$. For engineers, this technique is indispensable for working with equations from geometry, thermodynamics, or circuit analysis where explicit solutions are impractical.

Common Pitfalls

1. Misapplying the Product Rule Order: A common memory lapse flips the terms to $u(x)v^{\prime }(x)+u^{\prime }(x)v(x)$. While mathematically identical, consistent order prevents errors in more complex problems. Remember the pattern: differentiate the first, multiply by the second; plus the first times the derivative of the second.
2. Incorrect Quotient Rule Numerator: The most frequent error is writing the numerator in the wrong order: $u(x)v^{\prime }(x)-u^{\prime }(x)v(x)$. A mnemonic like "Lo d-Hi minus Hi d-Lo" helps, but understanding the derivation (seeing the subtraction arise from the added term in the proof) builds a more reliable intuition.
3. Chain Rule "Dropping" the Inner Derivative: It's easy to compute $g^{\prime }(u(x))$ and forget to multiply by $u^{\prime }(x)$. Always remember the chain rule signifies a multiplication of rates. In Leibniz notation, $\frac{dz}{dx}=\frac{dz}{dy}\cdot \frac{dy}{dx}$, the symbolic cancellation of $dy$ provides a useful, intuitive check.
4. Algebraic Errors in Limit Proofs: When proving rules from the definition, missteps in algebraic manipulation (e.g., incorrect binomial expansion, faulty grouping of terms) are common. Work slowly, write each step clearly, and verify that the original expression is equivalently transformed at each stage.

Summary

• Derivative rules are theorems that must be proven from the fundamental limit definition $f^{\prime }(x)={\lim }_{h\to 0}\frac{f(x+h)-f(x)}{h}$. This process relies heavily on algebraic manipulation and the properties of limits.
• The Power Rule for integers follows from the binomial theorem, canceling, and evaluating the limit. The Product Rule and Quotient Rule derivations involve the strategic addition and subtraction of a term to factor the difference quotient.
• The Chain Rule, essential for composites, is proven by considering the change in an intermediate variable. Its logic underpins implicit differentiation, which is an application of the chain rule to equations relating variables.
• For engineers, memorizing the rules is necessary for efficiency in calculation, especially under time constraints like exams or design iterations. However, understanding the proofs is essential for adaptability, enabling you to troubleshoot novel problems, modify methods for custom applications, and grasp why a rule fails when its underlying assumptions are violated. True mastery requires both.