LECTURE 14

Finding Extrema

Where are the maxima and minima? Optimization in one variable.

The Hook

Two argon atoms. At what distance is the energy minimized?

$$V(r) = 4\varepsilon \left[ \left(\frac{\sigma}{r}\right)^{12} - \left(\frac{\sigma}{r}\right)^{6} \right]$$

You could plot it and eyeball the minimum. But chemistry demands precision.

The derivative tells you exactly where. At a minimum, the function stops decreasing and starts increasing. At that instant, the rate of change is zero: V'(r) = 0.

Why Extrema Matter in Chemistry

ProblemWhat You're Optimizing
Equilibrium bond lengthMinimize potential energy
Transition stateSaddle point (max in one direction)
Equilibrium constantMinimize Gibbs free energy
Most probable speedMaximize Maxwell-Boltzmann distribution
Crystal structureMinimize lattice energy
Protein foldingMinimize free energy

Finding extrema is how nature finds equilibrium.

Critical Points

Critical point: A value x = c where f'(c) = 0 (horizontal tangent) or f'(c) does not exist (corner, cusp).

Critical points are candidates for extrema, but not every critical point is an extremum.

Local vs Global Extrema

Local maximum: f(c) ≥ f(x) for all x near c.

Global maximum: f(c) ≥ f(x) for all x in the entire domain.

Key insight: Every global extremum is a local extremum, but local extrema are not necessarily global. A critical point may not be an extremum at all (inflection point).

Interactive: Critical Point Classifier

Explore Critical Points and Extrema
Select a function to analyze its critical points.

The First Derivative Test

First Derivative Test

Let c be a critical point where f is continuous.

  1. If f' changes from + to − at c, then f has a local maximum.
  2. If f' changes from − to + at c, then f has a local minimum.
  3. If f' does not change sign, then c is not an extremum.

Why it works: f' > 0 means increasing, f' < 0 means decreasing. If f goes from increasing to decreasing, it peaked (maximum). If from decreasing to increasing, it bottomed out (minimum).

The Second Derivative Test

Second Derivative Test

Let f'(c) = 0 and f''(c) exist.

  1. If f''(c) > 0, then f has a local minimum at c. (concave up)
  2. If f''(c) < 0, then f has a local maximum at c. (concave down)
  3. If f''(c) = 0, the test is inconclusive. (use first derivative test)

Why it works: f'' > 0 means the curve bends upward (like a cup) — minimum. f'' < 0 means it bends downward (like a cap) — maximum.

Chemistry: Lennard-Jones Minimum

Finding the Equilibrium Distance
Equilibrium: r = 2^(1/6)σ with V = -ε

Derivation

Setting V'(r) = 0:

$$\frac{12\sigma^{12}}{r^{13}} = \frac{6\sigma^{6}}{r^{7}} \implies r^6 = 2\sigma^6 \implies r_{eq} = 2^{1/6}\sigma \approx 1.122\sigma$$

At the minimum: V(req) = −ε

Chemistry: Maxwell-Boltzmann Most Probable Speed

Where Does the Distribution Peak?
Most probable speed for N₂ at 300 K

Derivation

The distribution: f(v) ∝ v²e−av² where a = m/(2kBT)

Setting d/dv[v²e−av²] = 0:

$$2ve^{-av^2}(1 - av^2) = 0 \implies v_{mp} = \sqrt{\frac{1}{a}} = \sqrt{\frac{2RT}{M}}$$

Global Extrema on Closed Intervals

Extreme Value Theorem

If f is continuous on [a, b], then f attains both a global maximum and minimum on [a, b].

Closed Interval Method

  1. Find all critical points in (a, b)
  2. Evaluate f at each critical point
  3. Evaluate f at endpoints: f(a) and f(b)
  4. Compare all values: largest is global max, smallest is global min
Closed Interval Method
f(x) = x³ - 3x² on the interval

Extrema-Finding Procedure

Summary: How to Find Extrema

  1. Find critical points: Solve f'(x) = 0, find where f'(x) doesn't exist
  2. Classify each: Use first or second derivative test
  3. Check boundaries: Endpoints for [a,b], limits for open/infinite domains
  4. Compare values: Global max = largest, global min = smallest

Common Pitfalls

Summary

TestConditionResult
First Derivativef' changes + to −Local maximum
First Derivativef' changes − to +Local minimum
First Derivativef' no sign changeNot an extremum
Second Derivativef''(c) > 0Local minimum
Second Derivativef''(c) < 0Local maximum
Second Derivativef''(c) = 0Inconclusive