Exponentials and Logarithms

Core ≈ 55 min Chapter 9 8 exercises
Why this matters
Natural growth — money, populations, isotopes, cooling coffee — is exponential. Calculating with it demands fluency in both eˣ and its inverse ln x, and the strange-but-wonderful fact that eˣ is its own derivative. This chapter builds that fluency from the ground up.

1 · Exponentials and logarithms as inverses

Exponential laws: b^(x+y) = bˣbʸ, b^(x−y) = bˣ/bʸ, (bˣ)^y = b^(xy), b⁰ = 1. The logarithm undoes the exponent: log_b x is the power to which b must be raised to produce x. They cancel each other both ways — log_b(bˣ) = x and b^(log_b x) = x — and ln x is simply log_e x. Domains matter: y = ln x exists only for x > 0.

Three log rules do most of the work: the product rule log_b(MN) = log_b M + log_b N, the quotient rule log_b(M/N) = log_b M − log_b N, and the power rule log_b(Mʳ) = r·log_b M. Change-of-base (log_b M = ln M/ln b) converts anything into ln, your calculator's native tongue.

2 · Birth of e: compound interest

Invest P at annual rate r and compound n times per year: A = P(1 + r/n)^(nt). Compounding yearly, quarterly, monthly… the factor (1 + r/n)^n creeps upward toward a ceiling. That ceiling defines the number e: e = lim_{n→∞}(1 + 1/n)^n ≈ 2.71828, equivalently e = lim_{x→0}(1 + x)^(1/x). e is irrational and transcendental, and it is the base of natural growth — the only base whose curve rises at the rate of its own height.

3 · Differentiating eˣ, aˣ, ln x, log_a x

From the limit definition of e one shows (eˣ)′ = eˣ: the slope equals the height, at every point, exactly. Everything else follows by the chain or inverse-function rule: (aˣ)′ = aˣ·ln a, (ln x)′ = 1/x for x > 0, and (log_a x)′ = 1/(x ln a). Finally, with the power rule extended from integers to all real exponents, (xᵃ)′ = a x^(a−1) — now valid for fractional and irrational a, proved via xᵃ = e^(a ln x).

4 · Limits involving e

Infinite compounding is the launchpad: (1 + 1/n)^n → e as n → ∞, and more generally (1 + k/n)^n → eᵏ. The mirrored form lim_{x→0}(1 + x)^(1/x) = e shows up in the definition of eˣ's derivative. Know the asymptotics cold: eˣ → ∞ as x → ∞, eˣ → 0 (positive) as x → −∞, ln x → ∞ as x → ∞, and ln x → −∞ as x → 0⁺. So eˣ has a horizontal asymptote at y = 0 while ln x has a vertical one at x = 0.

Worked example 9.1 — xˣ, the exponential-power monster
Differentiate y = xˣ (x > 0) — neither the power rule nor the exponential rule fits alone.
Step 1. Take ln: ln y = x ln x.
Step 2. Differentiate implicitly: y′/y = ln x + x·(1/x) = ln x + 1.
Step 3. Multiply by y: y′ = xˣ(ln x + 1).
The same recipe handles x^(sin x), giving x^(sin x)·(cos x·ln x + sin x/x).

5 · Logarithmic differentiation

The trick generalises: whenever the variable sits in both base and exponent (xˣ, x^(sin x)) or in sprawling products, quotients and powers, slap ln on both sides first. The ln power rule flattens exponents into multipliers, and implicit differentiation finishes. This is also the cleanest proof that (xᵃ)′ = a x^(a−1) for arbitrary real a.

6 · Exponential growth and decay

If a quantity's rate of change is proportional to itself, dy/dt = ky, then y = y₀e^(kt). Positive k models growth (bank balances, bacteria), negative k decay (isotopes, drug clearance). From one datum find k, then predict. Doubling time: t₂ = ln 2/k; half-life: t½ = ln 2/|k|. Example: a 200 g isotope with half-life 4 days retains 200·(½)^(t/4) grams after t days — after 12 days, 25 g remain.

7 · Hyperbolic functions

Combine eˣ and e⁻ˣ symmetrically: sinh x = (eˣ − e⁻ˣ)/2, cosh x = (eˣ + e⁻ˣ)/2, tanh x = sinh x/cosh x. They mirror circular trig: cosh²x − sinh²x = 1, the derivatives are sinh′ = cosh, cosh′ = sinh (no minus sign!), and tanh′ = sech²x. sinh is odd, cosh is even with minimum value 1 — the catenary shape of a hanging chain.

Watch out
eˣ is not differentiated by the power rule: d/dx eˣ = eˣ, not x·e^(x−1). The two rules apply to different signatures — exponential base xᵃ versus base-with-exponent-variable aˣ — and mixing them is the classic silent error.
Chapter formula sheet
log_b(MN) = log_b M + log_b N
log_b(M/N) = log_b M − log_b N
log_b(Mʳ) = r log_b M
e = lim_{n→∞} (1 + 1/n)^n
(eˣ)′ = eˣ, (aˣ)′ = aˣ ln a
(ln x)′ = 1/x, (log_a x)′ = 1/(x ln a)
y = y₀e^(kt), t½ = ln 2/|k|
sinh x = (eˣ − e⁻ˣ)/2, cosh x = (eˣ + e⁻ˣ)/2, cosh²x − sinh²x = 1

Common mistakes

  • Applying the power rule to eˣ — the derivative is eˣ, never x·eˣ⁻¹.
  • Taking ln of a non-positive number without checking the domain; y = ln x needs x > 0.
  • Mis-signing the chain rule in (aˣ)′ = aˣ ln a — that extra ln a is mandatory for a ≠ e.
  • Confusing log_b M with log_b N in inversion: log₅25 = 2, not 5.
  • Writing the decay model with k positive — decay needs k < 0 inside y = y₀e^(kt).
  • Forgetting the implicit step in logarithmic differentiation: always multiply by y at the end.

✎ Practice — marked instantly

ln a + ln b equals:
  • A. ln(a·b)
  • B. ln a / ln b
  • C. ln a · ln b
  • D. ln(a + b)

Multiple choice · interactive marking is enabled when JavaScript loads.

Evaluate lim n→∞ (1 + 2/n)^n. Round to two decimals.

Numeric answer · interactive marking is enabled when JavaScript loads.

Differentiate y = e^(3x).

Short answer · interactive marking is enabled when JavaScript loads.

Differentiate y = ln(x).

Short answer · interactive marking is enabled when JavaScript loads.

Compute the derivative of eˣ at the point x = ln 2.

Numeric answer · interactive marking is enabled when JavaScript loads.

Using logarithmic differentiation, y = xˣ (x > 0) has derivative:
  • A. x·xˣ⁻¹
  • B. xˣ(ln x + 1)
  • C. xˣ ln x
  • D. x ln x

Multiple choice · interactive marking is enabled when JavaScript loads.

A 200 g radioactive sample has a half-life of 4 days. How many grams remain after 12 days?

Numeric answer · interactive marking is enabled when JavaScript loads.

Differentiate y = sinh(x).

Short answer · interactive marking is enabled when JavaScript loads.