Equation of the Normal Line

The normal line equation is the line perpendicular to the tangent at the point of tangency. Its slope is the negative reciprocal of the derivative at that point: m_normal = −1 / f'(a), provided f'(a) ≠ 0. Find it step by step with worked examples and the special cases: horizontal tangent gives vertical normal, vertical tangent gives horizontal normal.

The normal line is the straight line that runs perpendicular to the tangent line at the point of tangency. The tangent line best approximates the curve at that single point. The normal line points directly away from the curve, at a right angle to the direction of travel. If you imagine a curve and draw its tangent, the normal is the line that stands upright from it, like a flagpole on a slanted roof.

This matters because the normal line is used in optics (the angle of reflection equals the angle of incidence measured from the normal), in curvature calculations, and in closest-point problems (finding the point on a curve nearest to a given external point).

The most common confusion is mixing up the normal line with the perpendicular bisector. The normal line always passes through the point of tangency, and its slope comes directly from the derivative at that point. A perpendicular bisector, by contrast, passes through the midpoint of a segment and has no connection to a curve.

From the OpenStax Calculus exercises (Volume 1, section 3.1, problems 57-66), you can see the pattern: for any differentiable function, compute f'(a) first, then flip and negate it to get the normal slope.

The slope of the normal line is the negative reciprocal of the slope of the tangent line. If the tangent slope at x = a is m_tan = f'(a), then the normal slope is m_norm = −1 / f'(a). This rule holds whenever f'(a) ≠ 0.

If f'(a) = 0, the tangent is horizontal, so the normal line is vertical (equation x = a). If f'(a) is undefined (vertical tangent), the normal line is horizontal (equation y = f(a)).

Example from OpenStax Exercise 57: f(x) = 2x + 3 at x = 0. f'(0) = 2, so normal slope = −1/2. The normal line equation is y − 3 = (−1/2)(x − 0), or y = (−1/2)x + 3.

Example from OpenStax Exercise 60: f(x) = x² − 2x at x = 1. f'(1) = 0, so the normal line is vertical: x = 1.

Common Mistake: Forgetting the Negative Sign

Tutors report that students often compute the reciprocal but forget the negative sign. The normal line is perpendicular, which requires a slope that is the negative reciprocal. A plain reciprocal gives a line that is symmetric but not perpendicular, it will intersect the tangent at an angle of about 26.6° instead of 90°.

The formula for the normal line equation uses point-slope form: y − y₁ = m(x − x₁), where (x₁, y₁) is the point of tangency and m is the normal slope.

1. Find the point of tangency. You usually know the x-coordinate a. Compute y = f(a) to get (a, f(a)).
2. Compute the tangent slope. Find f'(x), then evaluate f'(a). This is the slope of the tangent line.
3. Compute the normal slope. If f'(a) ≠ 0, m_norm = −1 / f'(a). If f'(a) = 0, the normal is vertical. If f'(a) is undefined, the normal is horizontal.
4. Write the equation. Use point-slope form: y − f(a) = m_norm(x − a).
5. Convert to slope-intercept (optional). Simplify to y = m_normx + b for graphing.

Check your work: the product of the tangent slope and the normal slope should equal −1 (for non-zero, finite slopes).

Three worked examples, each using the formula above. All come from the OpenStax Calculus Vol. 1 exercises.

Example 1: Polynomial (OpenStax Exercise 64)

f(x) = x³ − x, at x = 1.

• Point: f(1) = 1³ − 1 = 0, so (1, 0).
• Derivative: f'(x) = 3x² − 1. f'(1) = 3(1)² − 1 = 2.
• Normal slope: m_norm = −1/2.
• Equation: y − 0 = (−1/2)(x − 1) → y = (−1/2)x + 1/2.

Check: tangent slope 2 times normal slope −1/2 equals −1. Correct.

Example 2: Trigonometric (OpenStax-style, not in the numbered set)

f(x) = sin x, at x = π/3.

• Point: f(π/3) = sin(π/3) = √3/2, so (π/3, √3/2).
• Derivative: f'(x) = cos x. f'(π/3) = cos(π/3) = 1/2.
• Normal slope: m_norm = −1/(1/2) = −2.
• Equation: y − √3/2 = −2(x − π/3).

For a trig function, the derivative rule is the same, no special handling needed beyond knowing the derivative of sine.

Example 3: Rational With A Vertical Normal (OpenStax Exercise 62, adapted)

f(x) = 1/x, at x = 2.

• Point: f(2) = 1/2, so (2, 1/2).
• Derivative: f'(x) = −1/x². f'(2) = −1/4.
• Normal slope: m_norm = −1/(−1/4) = 4.
• Equation: y − 1/2 = 4(x − 2) → y = 4x − 15/2.

Two cases break the negative reciprocal rule:

• Horizontal tangent (f'(a) = 0): The normal line is vertical. Equation: x = a. Example: f(x) = x² − 2x at x = 1 (OpenStax Exercise 60). The tangent is y = −1, the normal is x = 1.
• Vertical tangent (f'(a) undefined): The normal line is horizontal. Equation: y = f(a). Example: f(x) = x^(1/3) at x = 0 has a vertical tangent (the derivative approaches infinity). The normal is y = 0.

These cases are not rare. On the AP Calculus exam, expect at least one free-response question where you must identify a zero derivative and give the vertical normal line. The failure mode here is treating “undefined” as “zero”, a vertical tangent means the normal slope is 0, not undefined.

The normal line is not just a textbook exercise. It shows up in three real applications:

• Optics: The law of reflection states that the angle of incidence equals the angle of reflection, both measured from the normal line. When light hits a curved mirror, the normal to the mirror surface at the point of contact determines the reflected path. This is why parabolic mirrors focus light to a single point, the normal lines all point toward the focus.
• Curvature: The radius of curvature at a point on a curve is defined using the normal line. The circle of best fit (osculating circle) has its center on the normal line, at a distance equal to the radius of curvature. This appears in differential geometry and in computer graphics for smooth curve rendering.
• Closest-point problems: Given a point P outside a curve, the point on the curve closest to P lies on a line from P that is normal to the curve. This is because the distance function has a minimum where the vector from P to the curve is perpendicular to the tangent. Example: find the point on y = x² closest to (0, 5). Set up the distance squared, differentiate, and the optimal point satisfies that the line from P to the curve is normal.

On a calculator (TI-84 Plus), you can check normal slopes numerically using nDeriv(, but beware: nDeriv( uses a symmetric difference quotient and can give a false slope at non-differentiable points (like a corner). Always verify with the analytic derivative.

Three errors cost students points on exams:

1. Forgetting the negative sign. The normal slope is the negative reciprocal, not just the reciprocal. Double-check: multiply the tangent slope and the normal slope; the product must be −1.
2. Mixing up vertical and horizontal. When f'(a) = 0, the normal is vertical (x = a). When f'(a) is undefined, the normal is horizontal (y = f(a)). Memorize: zero → vertical, undefined → horizontal.
3. Using the wrong point. The normal line must pass through the point of tangency, not through any other point on the curve. Always plug a into f(x) to get the y-coordinate before writing the equation.

If you are using a calculator, cross-check: compute the normal line equation by hand, then graph both the curve and the normal line to see if they look perpendicular at the point of tangency. If they don't, you've made one of these errors.