6.3 Implicit Differentiation, Parametric Equations & Higher-Order Derivatives

Implicit Differentiation, Parametric Equations & Higher-Order Derivatives

Many functional relationships encountered in physical science and coordinate geometry cannot be conveniently expressed in the explicit form $y = f(x)$. Instead, they appear as implicit algebraic equations $F(x, y) = 0$ (such as conics) or as parametric trajectories $x = f(t), y = g(t)$ (such as projectile and orbital paths). On the Pakistan ECAT, questions frequently test implicit slopes, second derivatives of parametric curves, and $n$-th order derivatives. Applying procedural shortcuts like the partial derivative technique and Leibniz's Theorem is crucial to solving these multi-tiered problems within the 1-minute time window.

---

1. Implicit Differentiation Technique

An equation relating $x$ and $y$ of the form $F(x, y) = 0$ is said to define $y$ implicitly as a function of $x$. To calculate $\frac{dy}{dx}$ without solving explicitly for $y$:

1. Differentiate both sides of the equation with respect to $x$ term by term.
2. Whenever differentiating an expression in $y$, apply the chain rule: $\frac{d}{dx}[g(y)] = g'(y) \frac{dy}{dx}$.
3. Collect all terms containing $\frac{dy}{dx}$ on the left-hand side and move all terms without $\frac{dy}{dx}$ to the right-hand side.
4. Factor out $\frac{dy}{dx}$ and divide by its coefficient.

Classic Textbook Example

Find $\frac{dy}{dx}$ for the circle $x^2 + y^2 = r^2$: $\frac{d}{dx}(x^2) + \frac{d}{dx}(y^2) = \frac{d}{dx}(r^2) \implies 2x + 2y \frac{dy}{dx} = 0 \implies \frac{dy}{dx} = -\frac{x}{y}$

---

2. The Partial Derivative Shortcut for Implicit Relations

When faced with elaborate algebraic relations (such as polynomials or mixed trigonometric-exponential implicit equations), standard multi-line factoring is too slow. The multivariable chain rule provides an instant shortcut.

The Theorem

If an implicit relation is expressed as $F(x, y) = 0$, the total differential is: $dF = \frac{\partial F}{\partial x} dx + \frac{\partial F}{\partial y} dy = 0$ Dividing through by $dx$ and solving for $\frac{dy}{dx}$ yields:

$\frac{dy}{dx} = -\frac{\frac{\partial F}{\partial x}}{\frac{\partial F}{\partial y}} = -\frac{F_x}{F_y}$

where:

• $F_x = \frac{\partial F}{\partial x}$ is the partial derivative of $F(x, y)$ with respect to $x$, treating $y$ as an absolute constant.
• $F_y = \frac{\partial F}{\partial y}$ is the partial derivative of $F(x, y)$ with respect to $y$, treating $x$ as an absolute constant.

Application to General Conic Equation

Consider the general second-degree conic curve: $F(x, y) = ax^2 + 2hxy + by^2 + 2gx + 2fy + c = 0$

1. Differentiate w.r.t $x$ treating $y$ as constant: $F_x = 2ax + 2hy(1) + 0 + 2g(1) + 0 + 0 = 2(ax + hy + g)$
2. Differentiate w.r.t $y$ treating $x$ as constant: $F_y = 0 + 2hx(1) + 2by + 0 + 2f(1) + 0 = 2(hx + by + f)$
3. Apply the shortcut formula: $\frac{dy}{dx} = -\frac{F_x}{F_y} = -\frac{2(ax + hy + g)}{2(hx + by + f)} = -\frac{ax + hy + g}{hx + by + f}$ This universal formula gives the tangent slope for any parabola, ellipse, hyperbola, or circle instantly!

---

3. Parametric Equations & First Derivative

When a curve is described parametrically by expressing both coordinates in terms of a third independent variable $t$ (the parameter): $x = f(t), \quad y = g(t)$ Applying the chain rule $\frac{dy}{dt} = \frac{dy}{dx} \cdot \frac{dx}{dt}$ yields:

$\frac{dy}{dx} = \frac{\frac{dy}{dt}}{\frac{dx}{dt}} = \frac{g'(t)}{f'(t)}, \quad \text{provided } f'(t) \neq 0$

Common Parametric Representations Tested on the ECAT

• Circle ($x^2 + y^2 = a^2$): $x = a\cos\theta, y = a\sin\theta \implies \frac{dy}{dx} = \frac{a\cos\theta}{-a\sin\theta} = -\cot\theta$.
• Ellipse ($\frac{x^2}{a^2} + \frac{y^2}{b^2} = 1$): $x = a\cos\theta, y = b\sin\theta \implies \frac{dy}{dx} = \frac{b\cos\theta}{-a\sin\theta} = -\frac{b}{a}\cot\theta$.
• Hyperbola ($\frac{x^2}{a^2} - \frac{y^2}{b^2} = 1$): $x = a\sec\theta, y = b\tan\theta \implies \frac{dy}{dx} = \frac{b\sec^2\theta}{a\sec\theta\tan\theta} = \frac{b}{a}\csc\theta$.
• Parabola ($y^2 = 4ax$): $x = at^2, y = 2at \implies \frac{dy}{dx} = \frac{2a}{2at} = \frac{1}{t}$.

---

4. Second Derivative of Parametric Curves: The Classic ECAT Trap

Evaluating the second derivative of a parametric function is notorious for catching students unprepared.

The Common Falsehood

WARNING — ECAT TRAP: $\frac{d^2y}{dx^2} \neq \frac{\frac{d^2y}{dt^2}}{\frac{d^2x}{dt^2}}$ Simply differentiating the numerator and denominator with respect to $t$ is mathematically invalid because the second derivative is the derivative of $\frac{dy}{dx}$ with respect to $x$, not $t$.

Correct Mathematical Formulation

By definition, $\frac{d^2y}{dx^2} = \frac{d}{dx}\left(\frac{dy}{dx}\right)$. To differentiate a function of $t$ with respect to $x$, we must apply the chain rule:

$\frac{d^2y}{dx^2} = \frac{d}{dx}\left(\frac{dy}{dx}\right) = \frac{d}{dt}\left(\frac{dy}{dx}\right) \cdot \frac{dt}{dx} = \frac{\frac{d}{dt}\left(\frac{dy}{dx}\right)}{\frac{dx}{dt}}$

Algorithmic 3-Step Procedure

1. Find $\frac{dy}{dx} = \frac{dy/dt}{dx/dt}$.
2. Differentiate this resulting function of $t$ with respect to $t$: $\frac{d}{dt}\left(\frac{dy}{dx}\right)$.
3. Divide by $\frac{dx}{dt}$ a second time!

Worked Example: Find $\frac{d^2y}{dx^2}$ for $x = a\cos\theta$, $y = a\sin\theta$.

1. First derivative: $\frac{dx}{d\theta} = -a\sin\theta$, $\frac{dy}{d\theta} = a\cos\theta \implies \frac{dy}{dx} = \frac{a\cos\theta}{-a\sin\theta} = -\cot\theta$.
2. Differentiate w.r.t $\theta$: $\frac{d}{d\theta}(-\cot\theta) = -(-\csc^2\theta) = \csc^2\theta$.
3. Divide by $\frac{dx}{d\theta}$: $\frac{d^2y}{dx^2} = \frac{\csc^2\theta}{-a\sin\theta} = -\frac{1}{a\sin^3\theta} = -\frac{1}{a}\csc^3\theta$

---

5. Higher-Order Derivatives & Standard $n$-th Derivatives

Successive differentiation produces higher-order derivatives: $y' = \frac{dy}{dx}, \quad y'' = \frac{d^2y}{dx^2}, \quad y''' = \frac{d^3y}{dx^3}, \quad \dots, \quad y^{(n)} = \frac{d^n y}{dx^n} = D^n y$

Closed-Form Expressions for $n$-th Order Derivatives

By identifying induction patterns, the $n$-th derivatives of standard elementary functions can be memorized for instantaneous recall:

1. Exponential Function ($y = e^{ax}$): $y' = a e^{ax}, \quad y'' = a^2 e^{ax} \implies y^{(n)} = a^n e^{ax}$

2. General Exponential ($y = b^{ax}$): $y^{(n)} = a^n (\ln b)^n b^{ax}$

3. Sine Function ($y = \sin(ax + b)$): Each differentiation shifts the phase by $\frac{\pi}{2}$ radians: $y' = a\cos(ax+b) = a\sin\left(ax + b + \frac{\pi}{2}\right)$ $y^{(n)} = a^n \sin\left(ax + b + \frac{n\pi}{2}\right)$

4. Cosine Function ($y = \cos(ax + b)$): $y^{(n)} = a^n \cos\left(ax + b + \frac{n\pi}{2}\right)$

5. Reciprocal Linear Function ($y = \frac{1}{ax + b} = (ax+b)^{-1}$): $y' = -a(ax+b)^{-2}, \quad y'' = 2a^2(ax+b)^{-3}, \quad y''' = -6a^3(ax+b)^{-4}$ $y^{(n)} = \frac{(-1)^n n! a^n}{(ax + b)^{n+1}}$

6. Logarithmic Function ($y = \ln(ax + b)$): Since $y' = \frac{a}{ax+b}$, taking $n-1$ further derivatives gives: $y^{(n)} = \frac{(-1)^{n-1} (n-1)! a^n}{(ax + b)^n}$

---

6. Leibniz's Theorem for Product Differentiation

Leibniz's Theorem generalizes the product rule to the $n$-th derivative of the product of two functions $u(x)$ and $v(x)$. Its algebraic structure is isomorphic to the Binomial Theorem.

Theorem Statement

If $u(x)$ and $v(x)$ possess derivatives up to order $n$:

$(uv)^{(n)} = \sum_{r=0}^n \binom{n}{r} u^{(n-r)} v^{(r)}$

$(uv)^{(n)} = u^{(n)} v + n u^{(n-1)} v' + \frac{n(n-1)}{2!} u^{(n-2)} v'' + \dots + \binom{n}{r} u^{(n-r)} v^{(r)} + \dots + u v^{(n)}$

Application to Polynomial-Transcendental Products

On the ECAT, Leibniz's Theorem is most potent when one factor is a low-degree polynomial $P(x)$, because its derivatives vanish identically after degree $k$ ($v^{(k+1)} = 0$).

Example: Find the 4th derivative of $y = x^2 e^{2x}$. Let $u = e^{2x}$ and $v = x^2$. Then $v' = 2x$, $v'' = 2$, and $v''' = v^{(4)} = 0$. For $u = e^{2x}$, the $k$-th derivative is $u^{(k)} = 2^k e^{2x}$. Using Leibniz's Theorem with $n = 4$: $y^{(4)} = \binom{4}{0} u^{(4)} v + \binom{4}{1} u^{(3)} v' + \binom{4}{2} u^{(2)} v'' + 0$ $y^{(4)} = 1(2^4 e^{2x})(x^2) + 4(2^3 e^{2x})(2x) + 6(2^2 e^{2x})(2)$ $y^{(4)} = 16x^2 e^{2x} + 64x e^{2x} + 48 e^{2x} = 16 e^{2x}(x^2 + 4x + 3)$

---

7. Maclaurin and Taylor Series Expansions

If $f$ has derivatives of all orders at $x = 0$, its Maclaurin series is

$f(x) = f(0) + f'(0)\,x + \frac{f''(0)}{2!}x^2 + \frac{f'''(0)}{3!}x^3 + \dots + \frac{f^{(n)}(0)}{n!}x^n + \dots$

and its Taylor series about $x = a$ replaces powers of $x$ with powers of $(x - a)$:

$f(x) = f(a) + f'(a)(x - a) + \frac{f''(a)}{2!}(x - a)^2 + \dots$

Standard Expansions

Typical ECAT Uses

• Coefficient extraction: the coefficient of $x^n$ is $\frac{f^{(n)}(0)}{n!}$. For $e^{2x}$, the coefficient of $x^4$ is $\frac{2^4}{4!} = \frac{16}{24} = \frac{2}{3}$.
• Quick approximations: $e^{0.1} \approx 1 + 0.1 + 0.005 + 0.000167 = 1.105167$ (true value $1.105171$).
• Limits: $\lim_{x \to 0}\frac{\sin x - x}{x^3} = \lim_{x \to 0}\frac{-x^3/6 + x^5/120 - \dots}{x^3} = -\frac{1}{6}$.
• Small-angle results: $\sin x \approx x$ and $\cos x \approx 1 - \frac{x^2}{2}$ for small $x$ (in radians) explain why $\lim_{x \to 0}\frac{\sin x}{x} = 1$ and $\lim_{x \to 0}\frac{1 - \cos x}{x^2} = \frac{1}{2}$.