◀ Back to Note

Tony's Intertwingled Memex • Reference Manual

Concavity and the Second Derivative Test

Formulas & Definitions
Core Definition

Let fff be twice-differentiable on an open interval III:

  1. Concave Upward (∪\cup∪): fff is concave upward on III if f′(x)f'(x)f′(x) is strictly increasing on III. Algebraically, this occurs whenever f′′(x)>0f''(x) > 0f′′(x)>0 for all x∈Ix \in Ix∈I. Geometrically, the curve curves upward like a cup, and all tangent lines lie below the curve.
  2. Concave Downward (∩\cap∩): fff is concave downward on III if f′(x)f'(x)f′(x) is strictly decreasing on III. Algebraically, this occurs whenever f′′(x)<0f''(x) < 0f′′(x)<0 for all x∈Ix \in Ix∈I. Geometrically, the curve curves downward like a cap, and all tangent lines lie above the curve.
  3. Point of Inflection: A point (c,f(c))(c, f(c))(c,f(c)) on a continuous curve where the concavity strictly changes sign (from concave upward to concave downward, or vice versa). A necessary (but not sufficient) condition for an inflection point is that f′′(c)=0f''(c) = 0f′′(c)=0 or f′′(c)f''(c)f′′(c) does not exist.

Concept Explanation

1. Geometric Intuition and Tangent Line Dynamics

While the first derivative f′(x)f'(x)f′(x) measures the instantaneous rate of change (the slope of the tangent line), the second derivative f′′(x)=ddx[f′(x)]f''(x) = \frac{d}{dx}[f'(x)]f′′(x)=dxd​[f′(x)] measures the rate of change of that slope.

  • When f′′(x)>0f''(x) > 0f′′(x)>0, the slopes of the tangent lines are increasing as xxx moves from left to right (e.g., slopes go from −3→0→+4-3 \to 0 \to +4−3→0→+4). The tangent line rotates counterclockwise, resting entirely beneath the graph.
  • When f′′(x)<0f''(x) < 0f′′(x)<0, the slopes of the tangent lines are decreasing (e.g., slopes go from +4→0→−3+4 \to 0 \to -3+4→0→−3). The tangent line rotates clockwise, resting entirely above the graph.
  • At an inflection point (c,f(c))(c, f(c))(c,f(c)), the tangent line slices directly through the curve.

2. The Second Derivative Test for Local Extrema

The Second Derivative Test allows us to classify critical numbers where f′(c)=0f'(c) = 0f′(c)=0 without having to construct a full 1D sign chart for f′(x)f'(x)f′(x):

The Second Derivative Test

Suppose f′′(x)f''(x)f′′(x) is continuous near ccc and f′(c)=0f'(c) = 0f′(c)=0:

  1. If f′′(c)>0f''(c) > 0f′′(c)>0, then f(x)f(x)f(x) is concave up at ccc, meaning f(c)f(c)f(c) is a local minimum.
  2. If f′′(c)<0f''(c) < 0f′′(c)<0, then f(x)f(x)f(x) is concave down at ccc, meaning f(c)f(c)f(c) is a local maximum.
  3. If f′′(c)=0f''(c) = 0f′′(c)=0 (or does not exist), the test is inconclusive. The point could be a local minimum (e.g., f(x)=x4f(x) = x^4f(x)=x4), a local maximum (e.g., f(x)=−x4f(x) = -x^4f(x)=−x4), or an inflection point (e.g., f(x)=x3f(x) = x^3f(x)=x3). In this case, revert to the First Derivative Test.

3. Inflection Point Caveats: Necessary vs. Sufficient Conditions

A common error in differential calculus is assuming that solving f′′(x)=0f''(x) = 0f′′(x)=0 automatically produces an inflection point:

  • For f(x)=x4f(x) = x^4f(x)=x4, we have f′(x)=4x3f'(x) = 4x^3f′(x)=4x3 and f′′(x)=12x2f''(x) = 12x^2f′′(x)=12x2. Setting f′′(x)=0  ⟹  x=0f''(x) = 0 \implies x = 0f′′(x)=0⟹x=0. However, 12x2≥012x^2 \ge 012x2≥0 for all real numbers. Because the concavity never changes sign (+→++ \to ++→+), x=0x = 0x=0 is a local minimum, not an inflection point.
  • An inflection point strictly requires a concavity sign transition across the test coordinate, and the coordinate must belong to the domain of f(x)f(x)f(x). Vertical asymptotes (such as x=0x = 0x=0 for f(x)=1/xf(x) = 1/xf(x)=1/x) can flip concavity but are never inflection points because the function is undefined there.

Engineering Applications

  • Mechanical & Structural Engineering (Beam Bending & Neutral Axis): In Euler-Bernoulli beam theory, the bending moment M(x)M(x)M(x) satisfies M(x)=EId2wdx2M(x) = E I \frac{d^2w}{dx^2}M(x)=EIdx2d2w​, where w(x)w(x)w(x) is beam deflection. A point of inflection (w′′(x)=0w''(x) = 0w′′(x)=0) marks zero internal bending moment, identifying the exact cross-section where surface fibers switch between mechanical tension and compression.
  • Chemical Engineering (Thermal Runaway Boundaries): In exothermic continuous stirred-tank reactors (CSTR), the reaction heat release profile follows a sigmoidal curve. The inflection point marks the maximum acceleration of temperature growth (d2Tdt2=0\frac{d^2T}{dt^2} = 0dt2d2T​=0), identifying the operational threshold where secondary cooling jacket valves must open to prevent explosion.
  • Civil & Transportation Engineering (Highway Transition Curves): Highway exit ramps use Euler spirals (clothoids) where curvature transitions linearly (f′′(x)=0→κf''(x) = 0 \to \kappaf′′(x)=0→κ) to eliminate sudden lateral jerk on vehicle occupants.

Example Problems

Example 1: Concavity and Inflection Points of a Cubic Polynomial

Determine the intervals of concavity and find all inflection points of f(x)=2x3−9x2+12x−3f(x) = 2x^3 - 9x^2 + 12x - 3f(x)=2x3−9x2+12x−3.

Solution
  1. Differentiate:
    f′(x)=6x2−18x+12f'(x) = 6x^2 - 18x + 12f′(x)=6x2−18x+12
    f′′(x)=12x−18=6(2x−3)f''(x) = 12x - 18 = 6(2x - 3)f′′(x)=12x−18=6(2x−3)
  2. Locate Candidate Points:
    6(2x−3)=0  ⟹  2x−3=0  ⟹  x=326(2x - 3) = 0 \implies 2x - 3 = 0 \implies x = \frac{3}{2}6(2x−3)=0⟹2x−3=0⟹x=23​
  3. Construct Second Derivative Sign Chart:
    • Interval (−∞,32)\left(-\infty, \frac{3}{2}\right)(−∞,23​): Test x=0  ⟹  f′′(0)=−18<0  ⟹  x = 0 \implies f''(0) = -18 < 0 \impliesx=0⟹f′′(0)=−18<0⟹ Concave Downward (∩\cap∩).
    • Interval (32,∞)\left(\frac{3}{2}, \infty\right)(23​,∞): Test x=2  ⟹  f′′(2)=12(2)−18=6>0  ⟹  x = 2 \implies f''(2) = 12(2) - 18 = 6 > 0 \impliesx=2⟹f′′(2)=12(2)−18=6>0⟹ Concave Upward (∪\cup∪).
  4. Evaluate Inflection Point Coordinate: Because f′′(x)f''(x)f′′(x) changes sign across x=3/2x = 3/2x=3/2, an inflection point occurs:
    f(32)=2(278)−9(94)+12(32)−3=274−814+18−3=−544+15=−272+302=32f\left(\frac{3}{2}\right) = 2\left(\frac{27}{8}\right) - 9\left(\frac{9}{4}\right) + 12\left(\frac{3}{2}\right) - 3 = \frac{27}{4} - \frac{81}{4} + 18 - 3 = -\frac{54}{4} + 15 = -\frac{27}{2} + \frac{30}{2} = \frac{3}{2}f(23​)=2(827​)−9(49​)+12(23​)−3=427​−481​+18−3=−454​+15=−227​+230​=23​
    The inflection point is (32,32)\left(\frac{3}{2}, \frac{3}{2}\right)(23​,23​).

Example 2: Classifying Stationary Points with the Second Derivative Test

Find and classify all local extrema of g(x)=x4−8x2+5g(x) = x^4 - 8x^2 + 5g(x)=x4−8x2+5 using the Second Derivative Test.

Solution
  1. First Derivative and Stationary Points:
    g′(x)=4x3−16x=4x(x2−4)=4x(x−2)(x+2)g'(x) = 4x^3 - 16x = 4x(x^2 - 4) = 4x(x - 2)(x + 2)g′(x)=4x3−16x=4x(x2−4)=4x(x−2)(x+2)
    Setting g′(x)=0g'(x) = 0g′(x)=0 yields three critical numbers: x=−2x = -2x=−2, x=0x = 0x=0, and x=2x = 2x=2.
  2. Second Derivative:
    g′′(x)=12x2−16g''(x) = 12x^2 - 16g′′(x)=12x2−16
  3. Evaluate at Critical Numbers:
    • At x=0x = 0x=0: g′′(0)=−16<0  ⟹  g''(0) = -16 < 0 \impliesg′′(0)=−16<0⟹ Concave Down   ⟹  \implies⟹ Local Maximum at (0,5)(0, 5)(0,5).
    • At x=2x = 2x=2: g′′(2)=12(4)−16=32>0  ⟹  g''(2) = 12(4) - 16 = 32 > 0 \impliesg′′(2)=12(4)−16=32>0⟹ Concave Up   ⟹  \implies⟹ Local Minimum at (2,−11)(2, -11)(2,−11).
    • At x=−2x = -2x=−2: g′′(−2)=12(4)−16=32>0  ⟹  g''(-2) = 12(4) - 16 = 32 > 0 \impliesg′′(−2)=12(4)−16=32>0⟹ Concave Up   ⟹  \implies⟹ Local Minimum at (−2,−11)(-2, -11)(−2,−11).

Connections & References

  • Parent Concepts:
    • Definition of the Derivative
    • First Derivative Test
  • Related Optimization & Curve Concepts:
    • Critical Numbers and Critical Points
    • Summary of Curve Sketching
  • Course Sessions:
    • Calc 1 Session 13 - Critical Points and Extrema
    • Calc 1 Session 16 - Curve Sketching