Secondary 4 Additional Mathematics: Differentiation Is About How a Function Is Changing Now
Advanced differentiation begins when the derivative stops being the final answer and becomes evidence. A derivative can reveal gradient, turning behaviour, monotonicity, rates of change, tangent and normal information, and the location of candidate maxima or minima. A second derivative can then describe how the first derivative itself is changing and provide another layer of structural information.
At Secondary 4, the learner should be able to select the correct derivative rule, preserve algebraic structure during differentiation, solve derivative conditions accurately and return the result to the original problem. The most common failures occur after the calculus rule has already been chosen correctly: simplification breaks, a stationary point is not interpreted, a rate loses its units, or a second derivative is calculated without explaining what it establishes.
The derivative gives a local rate. The rest of the solution explains what that rate means.
The Simple Answer
- First derivative: rate of change of the function with respect to its variable.
- Stationary point: a point where the first derivative is zero.
- Second derivative: rate of change of the first derivative; useful for analysing local curvature and classifying stationary behaviour in suitable cases.
- Related rate: a rate of change derived from relationships between changing quantities.
- Optimisation: build an objective function, differentiate, solve the stationary condition and verify the relevant extremum.
The difficult step is often not differentiation itself. It is identifying which function should be differentiated and what condition should be imposed afterward.
Rule Selection Before Calculation
Before differentiating, read the structure of the function.
- Power structure: xn or a polynomial can be differentiated term by term.
- Composite structure: one function sits inside another, suggesting the chain rule.
- Product structure: two non-constant factors multiply, suggesting the product rule.
- Quotient structure: one changing expression is divided by another, suggesting the quotient rule or a useful algebraic rewrite.
- Exponential/logarithmic structure: preserve the special derivative properties of ex, ln x and composites.
- Trigonometric structure: differentiate the trigonometric function and then apply any chain factor.
The strongest student does not see a formula and immediately begin. The student first asks whether algebraic simplification would expose a cleaner derivative route.
Worked Example 1: Chain Rule With Structure
Differentiate
y = (3x2 + 1)5.
The outer function is u5; the inner function is u = 3x2 + 1. Therefore
dy/dx = 5(3x2 + 1)4 × 6x
= 30x(3x2 + 1)4.
The chain multiplier 6x is not an optional correction. It represents the rate at which the inner function changes.
Worked Example 2: Product Rule and Factor Preservation
Let y = x2ex. Then
dy/dx = 2xex + x2ex
= ex(x2 + 2x)
= xex(x + 2).
The final factorised form is often more useful than the expanded derivative because stationary points are now visible: since ex is never zero, dy/dx = 0 when x = 0 or x = −2.
Differentiation gives the derivative; algebra chooses the form that reveals what to do next.
Stationary Points: Zero Gradient Is Only the First Test
A stationary point occurs where dy/dx = 0. But a stationary point can represent different local behaviour. The derivative equation identifies candidate locations; classification requires additional evidence.
Two useful methods are:
- First-derivative sign test: inspect the sign of dy/dx on either side of the stationary value.
- Second-derivative test: evaluate d2y/dx2 at the stationary value where appropriate.
If the first derivative changes from positive to negative, the function moves from increasing to decreasing, indicating a local maximum. If it changes from negative to positive, the point is a local minimum. If the sign does not change, the stationary point may not be an extremum.
Second-Derivative Control
The second derivative describes how the first derivative is changing. At a stationary point x = a:
- if d2y/dx2 > 0 at x = a, the graph is locally curving upward and the stationary point is typically a local minimum;
- if d2y/dx2 < 0, the graph is locally curving downward and the stationary point is typically a local maximum;
- if d2y/dx2 = 0, the test is inconclusive and another method is required.
The final case is important. A zero second derivative does not prove that nothing special happens. It only means the second-derivative test has not classified the point.
Worked Example 3: Classify a Stationary Point
Let y = x2 − 6x + 4. Then
dy/dx = 2x − 6.
Set dy/dx = 0:
2x − 6 = 0 → x = 3.
The second derivative is
d2y/dx2 = 2 > 0.
Therefore the stationary point is a local minimum. Its y-coordinate is y(3) = 9 − 18 + 4 = −5, so the point is (3, −5).
A complete answer includes both coordinates and the interpretation requested.
When the Second-Derivative Test Is Inconclusive
Consider y = x3. Then dy/dx = 3x2, so x = 0 is stationary. The second derivative is 6x, which is also zero at x = 0. The second-derivative test gives no classification. Inspect the first derivative or graph behaviour instead. Since 3x2 is non-negative on both sides, the function remains increasing through the stationary point rather than changing from increasing to decreasing or vice versa.
The lesson is methodological: a test is a tool with conditions, not an oracle.
Related Rates: Connect the Changing Quantities First
In a related-rate problem, several quantities change with time but are linked by another relationship. The correct route is:
- define the changing quantities;
- write a relationship connecting them;
- differentiate the relationship with respect to time;
- substitute the known state and rates;
- solve for the required rate;
- interpret the sign and units.
Do not substitute numerical values too early if doing so destroys the variable relationship that needs to be differentiated.
Worked Example 4: Area Changing With Radius
For a circle, A = πr2. Suppose r changes with time. Differentiate with respect to t:
dA/dt = 2πr · dr/dt.
If r = 4 and dr/dt = 0.5 units per second, then
dA/dt = 2π(4)(0.5) = 4π square units per second.
The chain-rule structure is visible: area changes because radius changes.
Rates, Signs and Units
A rate must be interpreted. If dV/dt is negative, volume is decreasing. If dx/dt is positive under a chosen coordinate direction, x is increasing. Units should change appropriately: a length rate may be centimetres per second, an area rate square centimetres per second, and a volume rate cubic centimetres per second.
Units are a powerful diagnostic. If differentiating an area with respect to time produces a result still labelled centimetres per second, inspect the interpretation.
Optimisation: Build the Right Function Before Differentiating
Optimisation questions frequently fail before calculus begins. The learner differentiates the wrong expression or keeps too many variables. A reliable route is:
- define the objective quantity;
- use the constraint to express it in one independent variable;
- state the meaningful domain;
- differentiate the objective;
- solve the stationary condition;
- classify or otherwise justify the extremum;
- return to the requested quantity.
Worked Example 5: Maximum Product Under a Sum Constraint
Two positive numbers have sum 20. Let one be x, so the other is 20 − x. Their product is
P = x(20 − x) = 20x − x2.
Differentiate:
dP/dx = 20 − 2x.
Set to zero: x = 10. The second derivative is −2, so the stationary point is a maximum. Therefore the maximum product is 10 × 10 = 100.
The key structure is not the derivative formula. It is the one-variable model created from the constraint.
Differentiation as a Graph Reader
The sign of the first derivative tells whether a differentiable function is locally increasing or decreasing. The sign of the second derivative can help describe how the gradient is changing. Together they allow a learner to reason about graph shape even when a full graph has not been drawn.
| Evidence | Interpretation |
|---|---|
| dy/dx > 0 | Function increasing locally. |
| dy/dx < 0 | Function decreasing locally. |
| dy/dx = 0 | Stationary candidate. |
| d²y/dx² > 0 | Gradient increasing; locally curving upward. |
| d²y/dx² < 0 | Gradient decreasing; locally curving downward. |
These are local statements and should be applied within the actual function and domain being considered.
The Advanced Differentiation Decision Tree
- What is the function structure? Choose derivative rules accordingly.
- Can algebra simplify the structure first?
- What is the target? Gradient, stationary point, rate, tangent, normal or optimum?
- What condition should be imposed after differentiating?
- Does classification require a first-derivative sign test or second derivative?
- What units and domain conditions must survive?
- Can the answer be checked against graph shape or original constraints?
Common Secondary 4 Differentiation Errors
- Forgetting the inner derivative in the chain rule.
- Applying the product rule to only one factor.
- Losing brackets during quotient-rule algebra.
- Finding x-values of stationary points but omitting y-coordinates.
- Assuming every stationary point is a maximum or minimum.
- Using d²y/dx² = 0 as if it proves a classification.
- Substituting numerical values before differentiating a related-rate relationship.
- Dropping units from a rate.
- Optimising a two-variable expression without first using the constraint to reduce it.
- Completing the calculus but failing to answer the physical or geometric quantity requested.
A Seven-Stage Training Sequence
- Differentiate mixed power, exponential, logarithmic and trigonometric forms.
- Identify chain, product and quotient structure before writing.
- Factor derivatives and solve stationary conditions accurately.
- Classify stationary points using sign or second-derivative evidence.
- Solve related-rate problems with explicit units.
- Build one-variable optimisation models from constraints.
- Complete mixed timed questions where differentiation is only one stage of a longer route.
Checkpoint: Advanced Differentiation Control
- What does dy/dx = 0 identify?
- If d²y/dx² = 0 at a stationary point, what should you conclude?
- Why should related-rate relationships usually be differentiated before substituting all numerical values?
- What do opposite signs of velocity and acceleration imply about speed?
- What is the first modelling job in an optimisation question?
Checkpoint Answers
- A stationary candidate.
- The second-derivative test is inconclusive; use another method such as a first-derivative sign test.
- The variable relationship must remain visible so the chain of rates is preserved.
- The speed is decreasing at that instant.
- Define the objective and use the constraint to express it in one independent variable.
Wintour House V1.0 Learning Standard
Wintour House V1.0 treats differentiation as rate-based evidence. CivDJ routing identifies the current function structure, selects the derivative rule, tests whether the resulting expression is in the best algebraic form, applies the condition required by the problem, classifies the resulting state and returns the answer to its graph, geometry, motion or optimisation meaning.
Differentiate only after you know what is changing, and interpret before you declare the job finished.
