3.3.1 Rules of Differentiation

Motivation for Differentiation Rules

  • Up to now, derivatives were computed directly from the definition (limits) – correct but time-consuming.
  • A collection of rules streamlines differentiation, making calculus practical for everyday and advanced work.
  • Knowing the rules well is “the good stuff” that students keep using in later courses.

Constant Rule

  • Statement: If cc is any real number, then ddx(c)=0\frac{d}{dx}(c)=0.
    • Geometric reasoning: y=cy=c is a horizontal line; slope at every point is 00.
  • Example
    • ddx(28)=0\frac{d}{dx}(2^8)=0 (even though 28=2562^8=256 looks large, it’s still just a constant).

Power Rule (MOST-USED)

  • Let nn be a non-negative integer.
    ddx(xn)=nxn1\frac{d}{dx}(x^{n}) = n x^{n-1}
  • Procedure: “Pull the exponent down, subtract 1 from it.”
  • The text gives a formal proof (worth reading if curious).
  • Examples
    • ddx(x9)=9x8\frac{d}{dx}(x^9)=9x^8.
    • ddx(x)=ddx(x1)=1x0=1\frac{d}{dx}(x)=\frac{d}{dx}(x^1)=1\cdot x^{0}=1 (agrees with the fact that the line y=xy=x has slope 1).

Constant Multiple Rule

  • If cc is a constant and ff differentiable, then
    ddx(cf(x))=cf(x)\frac{d}{dx}\bigl(c\,f(x)\bigr)=c\,f'(x).
  • In practice: drag the constant outside → apply other rules to the variable part.
  • Example
    • f(x)=78x11f(x)=-\tfrac78 x^{11}
      f(x)=7811x10=778x10f'(x)=-\tfrac78\cdot 11 x^{10}=-\tfrac{77}{8}x^{10}.

Rewriting Roots for the Power Rule

  • Square root t=t1/2\sqrt{t}=t^{1/2}, etc.
  • Example
    • g(t)=38t=38t1/2g(t)=\tfrac38\sqrt{t}=\tfrac38 t^{1/2}
      g(t)=3812t1/2=316t1/2=316tg'(t)=\tfrac38\cdot\tfrac12 t^{-1/2}=\tfrac{3}{16}t^{-1/2}=\tfrac{3}{16\sqrt{t}}.

Sum (and Difference) Rule

  • If ff and gg are differentiable,
    ddx(f(x)±g(x))=f(x)±g(x)\frac{d}{dx}\bigl(f(x)\pm g(x)\bigr)=f'(x)\pm g'(x) (same sign).
  • Justified because derivatives are limits, and limit laws allow term-by-term work.
  • Example – Polynomial
    • h(w)=3w3+9w2+6w8h(w)=3w^3+9w^2+6w-8
    • Apply rules termwise:
      h(w)=6w2+18w+6h'(w)=6w^2+18w+6 (constant −8 vanishes).

The Number ee and Its Special Exponential Function

  • e2.7182818e\approx2.7182818\dots is to exponentials what π\pi is to circles.
  • Characterized by limh0eh1h=1\displaystyle\lim_{h\to0}\frac{e^{h}-1}{h}=1.
  • Forms the base of the natural exponential exe^{x}, whose inverse is the natural logarithm lnx\ln x.
Unique Derivative Property
  • ddx(ex)=ex\displaystyle \frac{d}{dx}(e^{x}) = e^{x} — the only elementary function equal to its own derivative.
  • Proof outline is in the textbook; for now we employ the fact directly.

Worked Application: Tangent Line & Horizontal Tangents

Given f(x)=2xex2f(x)=2x-\dfrac{e^{x}}{2} and the point (0,12)(0,-\tfrac12).

Step 1 – Derivative
  • Use difference, constant multiple, and exe^{x} rules:
    f(x)=ddx(2x)ddx(12ex)=212exf'(x)=\frac{d}{dx}(2x)-\frac{d}{dx}\Bigl(\tfrac12 e^{x}\Bigr)=2-\tfrac12 e^{x}.
Step 2 – Tangent Line at (0,12)(0,-\tfrac12)
  • Slope: m=f(0)=212e0=212=32m=f'(0)=2-\tfrac12 e^{0}=2-\tfrac12=\tfrac32.
  • Point-slope form:
    y+12=32(x0)y+\tfrac12=\tfrac32(x-0) → y=32x12y=\tfrac32x-\tfrac12.
Step 3 – Horizontal Tangent(s)
  • Horizontal means slope 00, so set derivative to zero:
    212ex=0ex=4.2-\tfrac12 e^{x}=0 \quad\Longrightarrow\quad e^{x}=4.
  • Solve with natural log:
    x=ln4.x=\ln4.
  • Corresponding point:
    (ln4,  f(ln4)=2ln412eln4)=(ln4,  2ln42).\bigl(\,\ln4,\;f(\ln4)=2\ln4-\tfrac12 e^{\ln4}\bigr)=\bigl(\ln4,\;2\ln4-2\bigr).

Practical Tips & Connections

  • Most derivative computations are combinations of the Constant, Power, Constant-Multiple, and Sum Rules.
  • Rewriting radicals and reciprocals into power form is essential to leverage the Power Rule.
  • The derivative rules mirror limit laws; understanding limits deepens your grasp of why these shortcuts work.
  • exe^{x}’s self-derivative property underlies continual growth models, compound interest, differential equations, and later, solving linear ODEs.
  • Recognize that “horizontal tangent” questions translate to “solve f(x)=0f'(x)=0,” a key concept for optimization problems.