2.3 - Second derivatives in other forms
A second derivative measures how a curve's gradient changes with its horizontal coordinate. You will obtain it from implicit and parametric equations, then use its sign at points where the curve can be treated locally as a function of .
Implicit second derivatives
Can two points with horizontal tangents on the same implicit curve be a maximum and a minimum? To decide, we need a second derivative without first making the subject.
Recall that depends on . On , the first differentiation gives
Differentiate the unsimplified equation again. The derivative of needs the product rule:
One factor changes at rate , and the other changes at rate . Consequently,
This equation can be evaluated at a point after the first derivative there has been found.
Implicit second differentiation
Treat , and higher derivatives as functions of . In particular,
Differentiate every term in the first-derivative equation, then solve for where its coefficient is nonzero.
These implicit second-derivative steps are not listed as a formula in the formula booklet; they use the supplied product rule and ordinary chain rule.
Keeping the first differentiated equation often makes the second step shorter than applying a quotient rule to a fraction containing .
Worked example 1: a circle
Question. Find at on .
Plan. Find the first derivative at the point, differentiate the implicit equation again, then substitute.
The point is on the circle because . From ,
Differentiate once more and substitute:
The point has , so solving for a finite derivative is valid.
Answer. at .
Check. Near this point, the upper branch is . Direct differentiation of this explicit function gives , which is at .
Change the coefficient of from to . The product-rule structure stays the same, but its coefficient must survive both differentiations.
Worked example 2: an ellipse
Question. Find at on .
Plan. Differentiate the full implicit equation twice, retaining the factor .
The first derivative equation is
at the given point. Differentiating this equation gives
At ,
Answer. .
Check. The explicit upper branch is . Its second derivative is half the circle's, so at it is . This also checks the changed coefficient geometrically: the ellipse is vertically compressed by a factor of .
Dividing by was justified at these points. At a point with , the circle has a vertical tangent, so the same expression cannot provide a finite .
Watch out
Writing loses the term . Both factors depend on ; use the product rule before substituting a point.
Retain the new coefficient through both differentiations.
Parametric second derivatives
A parametric derivative first measures change with respect to the parameter. Converting that rate into change with respect to requires a division each time.
For example, , gives
Differentiating this gradient with respect to gives . But that is the gradient's rate of change per unit of , not per unit of . Divide by to get the second derivative .
Parametric second derivative
If , are twice differentiable and , then
The first-derivative ratio is in the formula booklet. The second-derivative step is not listed there: apply the same chain-rule conversion again.
Do not replace the numerator by : the object to differentiate is the whole first derivative .
Worked example 3: differentiating a gradient
Question. A curve is defined by , . Find and at .
Plan. Form the first-derivative ratio as a function of , differentiate it, then divide by .
For ,
Therefore
At , , so both calculations apply.
Answer. and .
Check. Near positive , eliminate the parameter to get . Directly, and . At , these give and .
Stretch only the horizontal coordinate by a factor of . The first derivative halves, while the second derivative is divided by : both differentiations are with respect to the stretched coordinate.
Continue this lesson.
You’re previewing NovaLearn without an account. Create a free account to read the rest.
Create free account