Implicit differentiation: derivatives, tangents and stationary points
Implicit differentiation finds derivatives when and are related by an equation that has not, or cannot easily, be rearranged into . The central idea is simple: differentiate every term with respect to , remembering that is itself a function of .
Before you begin
Section titled “Before you begin”You should be confident with:
- differentiating standard functions
- the product, quotient and chain rules
- rearranging equations and substituting coordinates
Why does appear?
Section titled “Why does dydx\dfrac{dy}{dx}dxdy appear?”Suppose depends on . Differentiating with respect to requires the chain rule:
The factor records the fact that a change in causes a change in . More generally,
Useful cases include
and
By contrast, because is already the variable with respect to which we are differentiating.
The method
Section titled “The method”For an equation involving and :
- Differentiate both sides with respect to .
- Apply the chain rule to every expression involving .
- Use the product rule if and are multiplied together.
- Collect every term containing .
- Factor out and rearrange.
Example 1: a circle
Section titled “Example 1: a circle”Find for
Differentiate each term with respect to :
Now isolate :
This result makes geometric sense. At the radius from the origin has gradient , so the tangent, which is perpendicular to the radius, has gradient .
Example 2: collect several derivative terms
Section titled “Example 2: collect several derivative terms”Given
find .
The middle term needs the product rule:
Therefore
Collect the terms containing :
Hence
Do not treat as though were constant. Both factors vary with .
Example 3: a chain inside a chain
Section titled “Example 3: a chain inside a chain”Find if
Differentiate:
Collect and factor:
Therefore
Tangents and normals
Section titled “Tangents and normals”Once is known, substitute the given point to obtain the tangent gradient. A line through with gradient has equation
If the tangent gradient is , the normal gradient is
Example 4: tangent and normal equations
Section titled “Example 4: tangent and normal equations”The curve
passes through . From Example 2,
At ,
The tangent is
The normal gradient is , so the normal is
Keep the equations in point-gradient form unless another form is requested. It reduces avoidable algebra.
Stationary points
Section titled “Stationary points”At a stationary point, the tangent is horizontal, so
If
then a stationary point normally satisfies , provided . You must combine this condition with the original curve equation.
Example 5: locating stationary points
Section titled “Example 5: locating stationary points”Find the stationary points of
We have
For a stationary point,
so . Substitute this into the original equation:
Thus
Since , the stationary points are
At these points , so the derivative is defined.
Second derivatives
Section titled “Second derivatives”For an implicitly defined curve, differentiate the first derivative equation again. Do not assume becomes constant on the second differentiation.
Example 6: finding and using
Section titled “Example 6: finding and using d2ydx2\dfrac{d^2y}{dx^2}dx2d2y”For the circle
the first differentiated equation is
Differentiate this equation again:
The square appears because the product rule gives
Rearranging,
At , , so
Therefore is a local maximum of the upper branch of the circle. At , the second derivative is , so the lower branch has a local minimum there.
You can also differentiate an explicit formula for , but this often introduces a quotient rule. Differentiating the simpler first-derivative equation is usually cleaner.
A compact general result
Section titled “A compact general result”If a curve is written as
then, where ,
Here means differentiate with respect to while holding constant, and means differentiate with respect to while holding constant. This formula explains why the numerator and denominator in an implicit derivative often resemble partial derivatives. At A level, the step-by-step method is normally clearer and earns the method marks.
Common mistakes
Section titled “Common mistakes”Missing the chain-rule factor
Section titled “Missing the chain-rule factor”Incorrect:
Correct:
Forgetting the product rule
Section titled “Forgetting the product rule”Incorrect:
Correct:
Substituting the point too early
Section titled “Substituting the point too early”Differentiate and rearrange symbolically first. Early substitution can hide terms and makes the method harder to check.
Setting the whole fraction equal to zero without checking it
Section titled “Setting the whole fraction equal to zero without checking it”A fraction is zero when its numerator is zero and its denominator is non-zero. A zero denominator may represent a vertical tangent or a point at which the derivative formula is undefined.
Using the negative reciprocal in the wrong place
Section titled “Using the negative reciprocal in the wrong place”gives the tangent gradient. Take the negative reciprocal only when finding the normal gradient.
Self-check
Section titled “Self-check”- Given , find .
- Given , find .
- Find the tangent to at .
- For , find the stationary points and classify them.
Answers
Section titled “Answers”- .
- .
- Since , the gradient at is . The tangent is .
- , so stationary points require . The curve gives and . Differentiating again shows at a stationary point. Hence is a local maximum and is a local minimum.
What to learn next
Section titled “What to learn next”- Use implicit derivatives in tangents and normals.
- Compare this method with parametric differentiation.
- Apply related derivatives to connected rates of change.
- Develop classification skills in stationary points and curve sketching.