What is Additional Mathematics | The Nested Function That Explains the Chain Rule explains chain rule differentiation through a complete mathematical example, the conditions that make each step valid, independent checks and targeted revision.
Recognise the function inside a function
For y=(3x²+1)^5, the inner expression is u=3x²+1 and the outer operation is taking the fifth power. Changing x changes u, and changing u changes y. The chain rule accounts for both stages. Treating the entire bracket as if it were just x leads to a missing multiplier.
Derive the chain-rule structure
If y=F(u) and u=g(x), then dy/dx=(dy/du)(du/dx) wherever the derivatives exist. The derivative of the outer function measures how y responds to u; the inner derivative measures how u responds to x. Multiplying gives the overall rate of change of y with respect to x.
Differentiate the worked example
Let u=3x²+1. Then y=u^5, so dy/du=5u^4. Also du/dx=6x. Multiply: dy/dx=5u^4·6x=30x(3x²+1)^4. The factor 6x is essential. A result of 5(3x²+1)^4 is only the derivative with respect to u, not with respect to x.
Check at x=0
At x=0, the inner derivative 6x is zero. Therefore y′(0)=0. The graph has a horizontal tangent at the point (0,1). The function is even because it depends on x², so the left and right sides are symmetric about the y-axis. This symmetry is consistent with a horizontal tangent at the centre.
Use a second derivative to classify
From y′=30x(3x²+1)^4, differentiate by the product and chain rules: y″=30(3x²+1)^4+30x·4(3x²+1)^3·6x. Factor: y″=30(3x²+1)^3[(3x²+1)+24x²]=30(3x²+1)^3(27x²+1). This is positive for every real x. Thus the stationary point at x=0 is a strict local minimum, consistent with y≥1.
A nested square-root function
Take y=√(2x+5)=(2x+5)^(1/2), with domain x≥−5/2. For x>−5/2, the chain rule gives y′=(1/2)(2x+5)^(−1/2)·2=1/√(2x+5). The derivative formula is not finite at the left endpoint. The original function is defined there, but differentiability at a boundary needs separate attention.
A nested exponential
For y=e^(4x−3), the outer derivative of e^u is e^u, and the inner derivative of u=4x−3 is four. Therefore y′=4e^(4x−3). The exponential remains positive, so y′>0 everywhere. The function is strictly increasing. Omitting the four would produce a derivative that is one-quarter of the correct rate.
A nested trigonometric function
For y=sin(3x), the derivative is cos(3x)·3=3cos(3x). For y=cos(2x+1), the derivative is −sin(2x+1)·2=−2sin(2x+1). The angle arguments must be in radians when using these standard differentiation formulas. These examples show that the same inner-multiplier idea applies to powers, exponentials and trigonometric functions.
Two layers of nesting
Consider y=[1+(2x−1)³]^4. Let v=2x−1, u=1+v³ and y=u⁴. Then dy/du=4u³, du/dv=3v² and dv/dx=2. Multiply all three: y′=24(2x−1)²[1+(2x−1)³]³. Each layer contributes a derivative factor. Working from the outside inward and keeping brackets intact reduces algebraic errors.
Chain rule versus product rule
For y=x(3x²+1)^5, two functions are multiplied. Use the product rule first: y′=(3x²+1)^5+x·30x(3x²+1)^4. Factor to get (3x²+1)^4[(3x²+1)+30x²]=(3x²+1)^4(33x²+1). The second term needs the chain rule because one product factor is itself a composite function. Identify the outermost operation before choosing a differentiation rule.
Chain rule versus quotient rule
For y=1/(3x²+1), rewrite y=(3x²+1)^(−1). The chain rule gives y′=−(3x²+1)^(−2)·6x=−6x/(3x²+1)². This is equivalent to applying the quotient rule with numerator one and denominator 3x²+1. Rewriting a reciprocal as a negative power often shortens the calculation.
Reverse chain rule as an integration link
Since d[(3x²+1)^5]/dx=30x(3x²+1)^4, we can reverse the result: ∫30x(3x²+1)^4 dx=(3x²+1)^5+C. This works because the integrand contains the derivative of the inner expression multiplied by the outer derivative. The chain rule is therefore useful not only for differentiation but also for recognising simple substitution structures in integration.
Frequently asked questions
What is the chain rule? Differentiate the outer function and multiply by the derivative of the inner expression. Why does the extra factor appear? Because the inner input changes with x. Does it work for exponentials and trigonometric functions? Yes. What if there are three layers? Multiply the derivatives from every layer. Can I check a result? Differentiate step by step using a clearly labelled inner variable.
Independent practice
Differentiate (5x−2)^4: 20(5x−2)^3. Differentiate e^(−3x): −3e^(−3x). Differentiate sin(5x): 5cos(5x). Differentiate (x²+4)^(−2): −4x(x²+4)^(−3). Differentiate √(x²+1): x/√(x²+1). For each question, identify the outer function and the inner derivative before multiplying.
Why this matters in G3 A-Math
The chain rule is a central differentiation method in G3 Additional Mathematics. It connects polynomial, exponential and trigonometric derivatives, tangent gradients and simple reverse-chain integration. A useful solution labels the layers, preserves domain restrictions and checks the final derivative against the structure of the original function.
How to read the problem before calculating
Begin by stating exactly what is given and what the question asks. Identify which expressions represent inputs, outputs, constants and unknowns. Look for restrictions before transforming an equation: excluded denominators, positive logarithm arguments, restricted inverse-function branches and positive exponential substitutions are not optional. When a question asks for all solutions, one plausible candidate is not enough. Every candidate must be tested against the original conditions.
Why the first incorrect step matters
A learner can perform several lines of correct arithmetic after making an invalid assumption. That later accuracy does not repair the earlier mistake. The most effective correction finds the first step that changes the solution set without justification. Examples include dividing by an expression that might be zero, dropping the sign condition of a modulus branch, choosing the wrong square-root sign or forgetting the permitted domain. The repair is to restore the condition and redo the argument from that point.
Exact answers and numerical approximations
Use exact fractions, surds, powers and logarithms where possible. A decimal approximation is helpful for checking reasonableness but should not replace an exact answer when the problem asks for one. Round only after the algebraic conditions have been checked. A rounded candidate near a strict boundary can appear valid even when the exact value is excluded. The final answer should state whether a value is exact or approximate and should include angle or length units where relevant.
Independent checking and transfer
An answer is stronger when it can be reproduced in a second way. Substitute candidates into the original equation, use an equivalent algebraic form, compare graph behaviour or check a derivative by a simpler representation. Then change one feature of the example and solve again. Transfer matters: a student who understands the method should handle a different coefficient, sign, branch or interval rather than depend on the original numbers.
A practical revision routine
First read the question and underline its conditions. Second write the defining identity or method. Third carry out each transformation on a separate line. Fourth check every candidate against its branch and domain. Fifth verify at least one result independently. Sixth explain the solution in ordinary language. Seventh attempt a changed problem without looking at the worked example. This routine builds accuracy and makes errors easier to diagnose in a small-group tutorial.
Parents and learners: what to ask
Instead of asking only whether the final number is correct, ask why each step is allowed. What input values are excluded? Why is this branch chosen? What happens at the boundary? Can the answer be checked in the original expression? Is there another method? A learner who can answer these questions has begun to connect procedures with mathematical meaning. That understanding is more durable than copying a formula without knowing its conditions.
Frequently asked questions about the method
Do I need to state restrictions? Yes, whenever the expression or inverse operation has a restricted domain. Is a calculator answer enough? Not if the question asks for all solutions, exact values or justification. Can two algebraic forms give different domains? Simplification must preserve the original domain. How should I revise? Work one complete example, explain each step and solve a changed example independently. What should I do if two methods disagree? Find the first point where their assumptions or arithmetic differ.
Sources and further learning
Consult the official SEAB 2027 G3 syllabus listing to confirm examination requirements. The worked examples here are original educational illustrations, not official exam questions. The immutable Clementi Mathematics teaching reference provides the small-group learning framework. Continue with the Additional Mathematics overview.
The outermost operation decides the first step
Before differentiating, read an expression from the outside inward. In y=(3x²+1)^5, the outermost operation is raising a quantity to the fifth power. Inside that quantity is the polynomial 3x²+1. Write u=3x²+1 and y=u⁵. Differentiating the outside gives 5u⁴, but u itself changes with x at rate 6x. The full derivative is therefore 5u⁴×6x=30x(3x²+1)⁴. This explanation prevents the common mistake of copying the ordinary power rule without the inner multiplier.
A chain of small changes
Suppose x changes by a small amount Δx. Then u=3x²+1 changes approximately by 6xΔx. The output y=u⁵ changes approximately by 5u⁴Δu. Combining the approximations gives Δy≈5u⁴(6xΔx), so Δy/Δx≈30xu⁴. Taking the limit as Δx approaches zero produces the chain rule. The idea is not simply multiplying two unrelated formulas: one rate describes how the intermediate variable responds to x, and the other describes how the final output responds to that intermediate variable.
Why the chain rule uses multiplication rather than addition
If the outer output changes by roughly five units for each unit of intermediate input, and that intermediate input changes by roughly three units for each unit of x, the final output changes by roughly fifteen units for each unit of x. Multiplication composes the two rates. Adding five and three would mix quantities with different meanings. This dimensional perspective helps students remember the structure dy/dx=(dy/du)(du/dx). The notation also hints at cancellation of the intermediate change, although a rigorous proof uses limits.
A worked quadratic bracket with a negative coefficient
Differentiate y=(5−2x²)^4. The outer derivative is 4(5−2x²)³. The inner derivative is −4x. Multiply to obtain y′=−16x(5−2x²)³. The negative sign is not optional: the inner expression decreases for positive x. At x=1, the derivative is −16(3³)=−432. A student who writes +16x(5−2x²)³ has used the outer power correctly but lost the direction of change of the inside.
Differentiate a negative power correctly
Let y=(x²+3)^(−2). The outer derivative is −2(x²+3)^(−3), and the inner derivative is 2x. Thus y′=−4x(x²+3)^(−3)=−4x/(x²+3)³. Because x²+3 is positive for every real x, the original expression and derivative are defined everywhere. The negative power is handled by the ordinary power rule, followed by the same chain multiplier as a positive power.
Differentiate a fractional power with its domain
For y=(x²+1)^(1/2)=√(x²+1), the inside is strictly positive for all real x. Differentiate: y′=(1/2)(x²+1)^(−1/2)×2x=x/√(x²+1). There is no derivative-domain difficulty because the denominator never vanishes. By contrast y=√(x−2) is defined for x≥2 but has derivative 1/[2√(x−2)] only for x>2. Being defined at a boundary is not the same as having a finite two-sided derivative there.
The inner derivative can be a constant
For y=(7x−4)^9, the inner expression is 7x−4, whose derivative is seven. Therefore y′=9(7x−4)^8×7=63(7x−4)^8. The multiplier seven remains even though the inner expression is linear. For y=(x−4)^9, the multiplier is one and may be omitted. Comparing these two expressions makes the source of the extra coefficient visible.
A power of a trigonometric function
Differentiate y=[sin(2x)]³. The outside is a cube, so begin with 3[sin(2x)]². The next layer is sine, whose derivative is cosine. The innermost angle 2x has derivative two. Therefore y′=6sin²(2x)cos(2x). The result contains three factors corresponding to the three nested operations. Trigonometric differentiation formulas assume angles measured in radians.
A trigonometric function of a power
Now consider y=sin(x³). The outer function is sine, not a cube. Its derivative is cos(x³), multiplied by the inner derivative 3x². Hence y′=3x²cos(x³). This is different from [sin x]³, whose derivative is 3sin²x cos x. Brackets and the position of the exponent determine the nesting. Before calculating, ask whether the power acts on the angle x or on the sine output.
Differentiate a nested cosine
For y=cos(5x²−1), differentiate the outer cosine to obtain −sin(5x²−1), then multiply by the derivative 10x of the inside. Thus y′=−10x sin(5x²−1). The two common errors are dropping the negative sign from the cosine derivative and forgetting the factor 10x. A quick structural check asks whether the result still contains the unchanged inner expression as the argument of the sine function.
Differentiate an exponential of a quadratic
Let y=e^(x²−4x). The exponential is the outer function, so its derivative with respect to its own input is the same exponential. The inner derivative is 2x−4. Thus y′=(2x−4)e^(x²−4x). At x=2, the derivative is zero even though the exponential is positive. The stationary point arises because the inner quadratic is stationary there. The chain rule makes that relationship transparent.
Differentiate a logarithm of a positive expression
For y=ln(x²+1), the inner argument is positive for every real x. The derivative of ln u is 1/u, multiplied by u′. Therefore y′=2x/(x²+1). For y=ln(3x−2), the domain is x>2/3 and y′=3/(3x−2). The logarithm’s domain must be stated before differentiation. An algebraically written derivative does not make the original logarithm meaningful at forbidden inputs.
A logarithm of an absolute value
For y=ln|x−2|, the domain is x≠2. On either side of two, the derivative is 1/(x−2). This follows by differentiating the positive branch ln(x−2) for x>2 and the negative branch ln(2−x) for x<2. In the latter case, the chain rule gives [1/(2−x)]×(−1)=1/(x−2). The absolute value extends the real logarithm’s domain to both sides, but does not permit x=2.
A triple-nested power
Take y={1+[2x−1]³}⁴. Define v=2x−1, u=1+v³ and y=u⁴. Then dy/du=4u³, du/dv=3v² and dv/dx=2. Multiplying gives y′=24(2x−1)²{1+(2x−1)³}³. The calculation has three derivative stages. A practical way to avoid missing a stage is to write each inner layer on its own line before differentiating.
When a product contains a composite function
For y=x(3x²+1)^5, the outermost operation is multiplication of x and the composite bracket power. Apply the product rule: y′=(3x²+1)^5+x×30x(3x²+1)^4. Factor (3x²+1)^4 to get y′=(3x²+1)^4[(3x²+1)+30x²]=(3x²+1)^4(33x²+1). The product rule handles the two multiplied factors; the chain rule handles the composite second factor. Both are needed.
When a quotient contains a composite function
Let y=(x²+1)/(2x−3)^2, with x≠3/2. Write the denominator as a negative power: y=(x²+1)(2x−3)^(−2). The product rule gives y′=2x(2x−3)^(−2)+(x²+1)(−2)(2x−3)^(−3)×2. Collecting terms over (2x−3)³ yields [2x(2x−3)−4(x²+1)]/(2x−3)³=(−6x−4)/(2x−3)³. Rewriting a quotient can simplify differentiation, but the original denominator restriction must remain.
Find the tangent line to a composite curve
Consider y=(x²+1)³ at x=1. The point on the curve is (1,8). The derivative is y′=3(x²+1)²×2x=6x(x²+1)². At x=1 the gradient is 6×4=24. The tangent equation is y−8=24(x−1), or y=24x−16. Verify that x=1 gives y=8. The chain rule supplies the gradient; the point-gradient equation supplies the tangent. A correct derivative alone is not the final answer to a tangent-line question.
Find the normal line at the same point
The normal to a nonvertical tangent has gradient equal to the negative reciprocal of the tangent gradient. At x=1 for y=(x²+1)³, the tangent gradient is 24, so the normal gradient is −1/24. Through (1,8), the normal equation is y−8=−(x−1)/24. Do not confuse the reciprocal of a derivative with the derivative of an inverse function. The normal-gradient rule concerns perpendicular straight lines at a point.
Stationary points of a composite power
For y=(x²−4)², the derivative is 2(x²−4)×2x=4x(x²−4). Stationary points occur where 4x(x²−4)=0, giving x=−2, 0 or 2. The corresponding y-values are 0, 16 and 0. The graph is nonnegative and symmetric. At x=±2 the function attains minima of zero, while at x=0 it has a local maximum of sixteen. The derivative factorisation reveals three stationary candidates, not just the one arising from the inner derivative.
Why stationary does not always mean a turning point
For y=x³, the derivative is 3x², which is zero at x=0. However, y=x³ increases through the origin and does not turn there. Likewise, a composite derivative may vanish because one factor is zero without changing sign. To classify a stationary point, inspect the derivative’s sign on either side or use an appropriate second-derivative test. Merely setting the derivative equal to zero finds candidates; it does not by itself establish maxima or minima.
Reverse the chain rule to recognise an integral
Because d[(3x²+1)^5]/dx=30x(3x²+1)^4, we can immediately state ∫30x(3x²+1)^4 dx=(3x²+1)^5+C. If the integrand is 6x(3x²+1)^4 instead, it is one-fifth of that derivative, so its integral is (3x²+1)^5/5+C. This recognition is a useful bridge to substitution-based integration. The inner derivative must be present with the correct factor; otherwise a coefficient adjustment or a different method is needed.
An integral that looks similar but is different
Compare ∫2x(x²+1)^4 dx with ∫(x²+1)^4 dx. In the first, the derivative of the inner expression x²+1 is 2x, so the antiderivative is (x²+1)^5/5+C. In the second, that inner derivative factor is absent. One cannot simply write the same power expression divided by five. Differentiating (x²+1)^5/5 gives 2x(x²+1)^4, not (x²+1)^4. Reverse differentiation is the safest check on a proposed integral.
A finite-difference check for a derivative
For y=(3x²+1)^5, the derivative at x=1 is 30×1×4⁴=7680. A small forward difference [f(1+h)−f(1)]/h should approach 7680 as h approaches zero. It will not equal 7680 exactly for a finite nonzero h because higher-order terms remain. This is a conceptual numerical check, not a substitute for the symbolic derivative. Using a moderate h can give a noticeably different number; reducing h usually improves the approximation until numerical rounding becomes significant.
An error diagnosis for chain-rule questions
If a learner writes 5(3x²+1)^4, the inner derivative has been omitted. If they write 30x(3x²+1)^5, the outer exponent was not reduced. If they write 30x(3x²+1)^4+6x, the two rates were added rather than multiplied. If they differentiate sin(x³) as 3sin²x cos x, they have misread the nesting. The remedy is to label the outer and inner operations before using a formula. Each mistake has a distinct cause.
A structured independent practice set
Differentiate (4x+1)^6: 24(4x+1)^5. Differentiate (x²−2x+3)^4: 4(2x−2)(x²−2x+3)^3. Differentiate e^(−2x²): −4xe^(−2x²). Differentiate cos(3x): −3sin(3x). Differentiate ln(x²+4): 2x/(x²+4). Differentiate [sin(x²)]²: 4x sin(x²)cos(x²). For each, state the outermost operation, identify the inner derivative and verify the final structure.
A mixed-rule practice challenge
Find the derivative of y=(x+1)e^(x²). The product rule gives y′=e^(x²)+(x+1)×2x e^(x²)=e^(x²)[1+2x(x+1)]. For y=(x²+1)sin(3x), the product rule gives y′=2x sin(3x)+3(x²+1)cos(3x). The challenge is to identify both levels: the top-level multiplication and the nested function within one factor. A student who can solve these without a chapter label has begun to choose methods independently.
Parent-friendly questions about nested functions
Ask what happens first to x and what happens last to the resulting quantity. For (3x²+1)^5, x is squared, multiplied by three, increased by one and raised to the fifth power. When differentiating, the chain rule works backward through the layers. Ask the learner to explain why the factor 6x appears. If the learner can describe the intermediate quantity and its rate of change, the formula has meaning rather than being a string of symbols.
Frequently asked questions about the chain rule
What is a composite function? One function applied to the output of another. What is the chain rule? Differentiate the outside with respect to the inside, then multiply by the inside derivative. Must the inside be a polynomial? No; it can be exponential, logarithmic or trigonometric. Can the chain rule be combined with the product rule? Yes, when a product contains a composite factor. What about a constant inner derivative? It still multiplies the outer derivative. Does a stationary derivative always mean a maximum or minimum? No; classification requires further reasoning.
Syllabus scope and final checklist
The chain rule is a standard differentiation technique relevant to Singapore G3 Additional Mathematics. Confirm the precise examinable functions and notation in the official 2027 SEAB K341 syllabus: https://www.seab.gov.sg/secondary-education-certificate-sec/g3-syllabuses-for-school-candidates-2027/. This guide also includes illustrative extensions, so the official syllabus remains authoritative. Before submitting a derivative, identify the outermost operation, preserve brackets, multiply by every inner derivative, state restrictions when needed, simplify without changing the domain and check the result through a related form or numerical behaviour. The immutable Clementi reference is https://edukatesg.com/portfolio/secondary-1-mathematics-tutor-clementi-small-groups-tutorials/.
A chain rule with a rational inner expression
Differentiate y=[(x+1)/(x−2)]³, where x≠2. Let u=(x+1)/(x−2). The outer derivative is 3u². The quotient rule gives u′=[(x−2)−(x+1)]/(x−2)²=−3/(x−2)². Multiply to obtain y′=−9[(x+1)/(x−2)]²/(x−2)²=−9(x+1)²/(x−2)⁴. The negative sign reflects the fact that the inner rational function decreases on either side of its vertical asymptote. The derivative is not defined at x=2 because the original function is not defined there. This example demonstrates that the chain rule can wrap around a quotient-rule calculation.
A quotient rule wrapped around a chain rule
Consider y=(x²+1)³/(x−1), where x≠1. Write the numerator as u=(x²+1)³ and denominator as v=x−1. Then u′=6x(x²+1)² by the chain rule, and v′=1. The quotient rule gives y′=[6x(x²+1)²(x−1)−(x²+1)³]/(x−1)². Factoring (x²+1)² produces y′=(x²+1)²[6x(x−1)−(x²+1)]/(x−1)²=(x²+1)²(5x²−6x−1)/(x−1)². The factorised answer makes both the algebra and the original denominator restriction visible.
A composite function involving a variable exponent
The expression y=x^x for x>0 is not an ordinary fixed-power function, because both its base and exponent depend on x. Take logarithms: ln y=x ln x. Differentiate implicitly: y′/y=ln x+1, using the product rule on x ln x. Hence y′=x^x(ln x+1). This is an extension illustrating logarithmic differentiation, rather than a reason to apply the simple power rule as though the exponent were constant. The domain x>0 matters for this real-valued derivation.
Implicit differentiation and the chain rule
Suppose x²+y²=25 describes a circle, and y depends on x along a smooth branch. Differentiating with respect to x gives 2x+2y(dy/dx)=0. The factor dy/dx appears because y² is a composite function of x. Rearranging gives dy/dx=−x/y where y≠0. At (3,4), the tangent gradient is −3/4. At (5,0), the formula has zero denominator and the circle has a vertical tangent rather than a finite derivative dy/dx. This example shows that the chain rule also appears when the dependent variable is not explicitly isolated.
Related rates as a chain-rule application
Let the area of a square be A=s², where the side length s changes with time t. Then dA/dt=(dA/ds)(ds/dt)=2s(ds/dt). If the side is 5 cm and growing at 0.2 cm per second, the area is increasing at 2×5×0.2=2 cm² per second. The units confirm the multiplication: centimetres times centimetres per second becomes square centimetres per second. This is an illustrative application; it does not imply that every related-rates problem is prescribed by the examination syllabus.
Check a derivative by expanding first
For y=(x+2)³, the chain rule gives y′=3(x+2)². Expanding first gives y=x³+6x²+12x+8, whose derivative is 3x²+12x+12. Expanding the chain-rule answer gives 3(x²+4x+4)=3x²+12x+12, so both methods agree. Expansion is practical for a low power and offers an independent check. For (3x²+1)^9, expansion would be long and error-prone; the chain rule is far more efficient.
A sign analysis without expanding the derivative
Take y=(x²+2)^5. Its derivative is 10x(x²+2)^4. Because (x²+2)^4 is strictly positive for every real x, the sign of y′ is determined entirely by x. The function decreases for x<0 and increases for x>0, with a minimum at x=0. This conclusion needs no expansion of the fifth power. Factorised derivatives often communicate more about graph behaviour than fully expanded polynomials.
A chain-rule expression with a stationary inflection
Consider y=(x−1)³. The derivative is 3(x−1)², which is zero at x=1 but positive on both sides. The curve is increasing before and after that point, so it does not have a turning point there. The second derivative is 6(x−1), which changes sign at x=1; the point (1,0) is a stationary inflection. This is why finding y′=0 is only the first step in classifying a stationary point. Sign changes and curvature provide the rest of the argument.
How to choose a useful substitution label
For y=[1+sin(3x)]⁴, write u=1+sin(3x). Then y=u⁴, so dy/du=4u³. The inner derivative is du/dx=3cos(3x). Thus y′=12[1+sin(3x)]³cos(3x). A student can also introduce v=3x, but an extra label is only useful if it clarifies the layers. The goal is to see the composition, not to create more notation than the problem requires.
An exact derivative evaluation at a negative input
Let y=(2x²−3x+1)^4. Differentiate to obtain y′=4(2x²−3x+1)³(4x−3). At x=−1, the inner expression equals 2+3+1=6, while its derivative is −4−3=−7. Hence y′(−1)=4×6³×(−7)=−6048. Keeping the two calculations separate reduces sign errors. A negative gradient is plausible because the inner quadratic is decreasing at that input and the outer fourth-power function is increasing for positive inner values.
A step-by-step mini assessment
Question A: Differentiate (2x−5)^7. Answer 14(2x−5)^6. Question B: Differentiate √(4x+1) for x>−1/4. Answer 2/√(4x+1). Question C: Differentiate ln(5x²+2). Answer 10x/(5x²+2). Question D: Differentiate e^(sin x). Answer e^(sin x)cos x. Question E: Differentiate (x²+1)^3 sin x. Answer 6x(x²+1)² sin x+(x²+1)³ cos x. Each question should be solved by first naming the outer operation, then identifying the inner derivative and finally checking the signs.
A correction exercise: locate the first wrong line
A student differentiates y=(4x−1)^3 and writes y′=3(4x−1)². The first wrong step is omission of the derivative four of the inner expression; the corrected answer is 12(4x−1)². Another student writes d[cos(2x)]/dx=2sin(2x). The chain multiplier two is present, but the derivative of cosine should be negative: the correct result is −2sin(2x). In a third case, y=ln(x²+1) becomes y′=1/(x²+1); the missing factor is 2x. These errors look similar on a marked page but require distinct corrections.
A study sequence from recognition to independence
First, identify the inner and outer functions without calculating. Second, differentiate a bracket power with a linear inner expression. Third, differentiate one with a quadratic inner expression. Fourth, mix powers with exponential, logarithmic and trigonometric functions. Fifth, combine the chain rule with a product or quotient. Sixth, apply the derivative to a tangent, stationary-point or integration question. Finally, attempt a mixed set without topic labels. At every stage, require an explanation of why each multiplier appears. This sequence builds transfer rather than dependence on the immediately preceding example.
Why a derivative should be checked against the original domain
Consider y=ln(2x−3). The derivative is 2/(2x−3), but the original real logarithm exists only for x>3/2. The rational expression 2/(2x−3) can be evaluated for x<3/2, yet it is not the derivative of the original real logarithm there because that original function is undefined. Similarly, y=√(x−1) is defined for x≥1, while the ordinary derivative formula applies only for x>1. Algebraic simplification must never silently enlarge the differentiability domain.
A final explanation for parents and students
The chain rule is the mathematics of a process inside another process. It is easier to remember when a learner can name the stages: x changes an intermediate quantity, and that intermediate quantity changes the final output. Encourage the learner to ask whether the outside function has been differentiated, whether the inside derivative has been included, and whether a product or quotient is also present. A correct final derivative is useful; an explanation that survives a changed example is stronger evidence of mastery.
How to use the chain rule in a graph question
Suppose a graph is defined by y=(x²−1)³. Its derivative is 6x(x²−1)². This factorisation reveals stationary candidates x=−1, 0 and 1. The squared factor is nonnegative, so away from those three points the sign of the derivative follows the sign of x. Consequently the curve decreases for x<0 and increases for x>0. The stationary points at x=−1 and x=1 do not reverse the sign of the derivative, while x=0 does. The point x=0 is therefore a local minimum with y=−1. This is a stronger explanation than merely listing all three stationary coordinates.
A full worked tangent-and-normal practice question
Let y=(2x+1)^4 and find the tangent and normal at x=0. First find the point: y(0)=1, so the point is (0,1). Differentiate using the chain rule: y′=4(2x+1)³×2=8(2x+1)³. At x=0 the tangent gradient is eight, giving y−1=8x, or y=8x+1. The normal gradient is −1/8, so the normal is y−1=−x/8. Check that both lines pass through (0,1). The tangent and normal gradients multiply to −1, confirming that the lines are perpendicular.
What mastery looks like after the lesson
A learner who has mastered this topic can recognise composite functions without a chapter label, explain why the derivative is a product of rates, differentiate several layers in the correct order and combine the method with product or quotient rules when needed. The learner should also know that a derivative formula is valid only where the original function is differentiable. Finally, they should be able to interpret the derivative as a gradient or rate of change and check the result by substitution, expansion where practical, or a numerical difference. These habits make the chain rule a reusable tool rather than a memorised trick.