Calculus › Section C · Derivatives › Topic C.2
Basic differentiation rules and rates of change
The limit definition is the truth, but it is slow. A handful of rules, each proved once from that definition, turns differentiation from a three-minute puzzle into a three-second reflex.
Why we build rules
Every derivative is, at bottom, the limit from the last topic. But nobody wants to run a limit every time they need a slope, any more than you re-derive multiplication before doubling a recipe. The way forward is to apply the definition once to each basic kind of function, bottle the result as a rule, and then combine rules to handle anything built from the basics. This topic bottles the first set: constants, powers, sums, and the two workhorse trig functions. Learn them to the point of reflex; everything in Sections C and D assumes you can do this part without thinking.
Constants and the power rule
The derivative of a constant function is \(0\): its graph is a flat line, and a flat line has no slope. The powerhouse of the topic is the power rule: for any exponent \(n\),
bring the exponent down in front and drop it by one. So \(x^5 \to 5x^4\), \(x \to 1\), and \(x^{-2} \to -2x^{-3}\). The rule works for fractional and negative powers alike (wherever \(x^{n-1}\) is defined), which is exactly why it handles roots and reciprocals. For whole-number \(n\), the proof is the limit definition plus the binomial expansion of \((x + h)^n\), the same move you made for \(x^2\) in C.1.
Rewrite before you differentiate
The power rule speaks the language of powers, so translate roots and fractions into that language first. \(\frac{1}{x^3} = x^{-3}\), so its derivative is \(-3x^{-4} = -\frac{3}{x^4}\). \(\sqrt{x} = x^{1/2}\), whose derivative is \(\frac{1}{2}x^{-1/2} = \frac{1}{2\sqrt{x}}\). And \(\sqrt[3]{x^2} = x^{2/3}\) gives \(\frac{2}{3}x^{-1/3}\). Most power-rule mistakes are actually translation mistakes. Make the rewrite the literal first line of your working, every time, and the rule itself becomes the easy part.
Sums and constant multiples
Two structural rules close the basic set. The derivative of a sum is the sum of the derivatives: \((f + g)' = f' + g'\), and likewise for differences. And multiplying by a constant multiplies the derivative: \((cf)' = c f'\). Together they say differentiation respects the way polynomials are assembled, so a polynomial differentiates term by term:
Each term faces the power rule alone, the \(-9\) vanishes as a constant, and the whole machine takes seconds.
Worked example: a tangent line from the rules
Find the tangent line to \(f(x) = x^3 - 3x^2 + 2x\) at \(x = 1\). Term by term, \(f'(x) = 3x^2 - 6x + 2\). The point: \(f(1) = 1 - 3 + 2 = 0\). The slope: \(f'(1) = 3 - 6 + 2 = -1\). Point-slope form finishes it: \(y - 0 = -(x - 1)\), or \(y = -x + 1\). One general observation now becomes available: the tangent is horizontal exactly where \(f'(x) = 0\), and peaks and valleys live on horizontal tangents. Keep that idea in your pocket; the worked example with the thrown ball puts it to work.
Sine and cosine
The special trig limits from B.3 are exactly what the definition needs here, and they deliver the two most elegant rules in the course:
Both demand \(x\) in radians, without exception. Check them against the graphs: \(\sin x\) climbs steepest at \(x = 0\), where \(\cos x = 1\); at its peak \(x = \frac{\pi}{2}\) the sine is momentarily flat, and \(\cos\frac{\pi}{2} = 0\). The cycles fit together: the graph of sine's slope is the cosine curve, which is one reason these two functions describe so many oscillations in science.
A preview: the natural exponential
One more rule is worth meeting now, even though Section F tells its full story. There is a special base, the number \(e \approx 2.71828\), chosen so that \(y = e^x\) has slope exactly \(1\) where it crosses the \(y\)-axis, at \((0, 1)\). The beautiful consequence: \(\frac{d}{dx} e^x = e^x\). The exponential is its own derivative: its slope always equals its height, which is precisely why it models anything whose growth rate is proportional to its size, from bacteria to bank balances. For this topic you only need the rule itself. The remaining trig functions arrive in C.3, and where \(e\) comes from, with the rules for \(\ln\) and other bases, arrives in Section F.
Worked example: mixing the rules
Differentiate \(f(x) = 3x^4 + 2\sin x - \dfrac{5}{x}\). Translate the fraction first: \(-\frac{5}{x} = -5x^{-1}\). Now term by term: the power rule gives \(12x^3\); the trig rule gives \(2\cos x\); the translated term gives \(-5 \cdot (-1)x^{-2} = +5x^{-2}\), watching the double minus. Altogether
No limits, no expansions, only rules. The full routine (translate, go term by term, watch signs) becomes the default first move of differentiation, and you will use it inside every bigger rule that follows.
Rates of change in the real world
The whole point of the rules is speed of interpretation. If \(s(t)\) is a position in metres after \(t\) seconds, then \(v(t) = s'(t)\) is the velocity in metres per second, and \(|v(t)|\) is the speed. Near Earth's surface, an object thrown straight up from ground level at \(v_0\) metres per second has height \(s(t) = -4.9t^2 + v_0 t\) metres (ignoring air resistance), where \(-4.9\) is half of gravity's \(-9.8\) metres per second squared. The average velocity over an interval is a secant slope; the instantaneous velocity is the derivative, an exact number instead of an approximation. Whatever the quantity (prices, populations, temperatures), its derivative reads as a rate with units of output per input.
Worked example: the top of a throw
A ball thrown straight up at \(20\) m/s has height \(s(t) = -4.9t^2 + 20t\) metres. Its velocity is \(v(t) = s'(t) = -9.8t + 20\). How fast is it moving after \(1\) second? \(v(1) = -9.8 + 20 = 10.2\) m/s, still climbing. When does it stop rising and start falling? At the top, velocity passes through zero: \(-9.8t + 20 = 0\), so \(t = \frac{20}{9.8} \approx 2.04\) seconds. Notice what we just did: found where the derivative equals zero to locate a peak. That single idea, \(f' = 0\) at the summit, grows into the entire optimisation toolkit of Section D.
Common mistakes to avoid
The recurring slips: differentiating a constant into itself instead of \(0\); applying the power rule directly to \(\frac{1}{x^2}\) without rewriting it as \(x^{-2}\), which produces nonsense exponents; lowering the exponent but forgetting to bring the old one down; losing the minus sign in \(\frac{d}{dx}\cos x = -\sin x\); using degrees instead of radians with the trig rules; and, soon, the great trap of the next topic, C.3: assuming the derivative of a product is the product of the derivatives. It is \(x \cdot x = x^2 \to 2x\), not \(1 \cdot 1 = 1\): keep that counterexample handy.
Key takeaways
Constants differentiate to \(0\). The power rule sends \(x^n\) to \(nx^{n-1}\) for any exponent, so translate roots and fractions into powers first. Sums, differences and constant multiples split and carry. In radians, sine becomes cosine and cosine becomes negative sine, and \(e^x\) is its own derivative. The payoff is interpretation: velocity is the derivative of position, and \(f'(x) = 0\) locates level tangents, the seeds of peaks, valleys and optimisation. Master these reflexes now; the product, quotient and chain rules of the next topics all stand on them.
Flashcards
Flashcard
0
Test your understanding
Question 1 of 5
What is the derivative of 4x³ − 2x + 7?
Question 2 of 5
What is the derivative of the cube root of x, that is, x^(1/3)?
Question 3 of 5
What is the slope of the tangent to y = sin x at x = π/6?
Question 4 of 5
The curve y = x³ − 12x has horizontal tangents at which x-values?
Question 5 of 5
A ball has height s(t) = −4.9t² + 24.5t metres. What is its velocity at t = 2 seconds?
Standards: 2
