Applications of Integration
The definite integral is more than an antiderivative bookkeeping tool — it is a machine for summing infinitely many infinitesimal contributions. Every application in this chapter follows one recipe: cut a quantity into thin slices, approximate each slice by a simple shape, sum with a Riemann sum, and pass to the limit to recover an integral.
Read 6.1–6.3 as one continuous skill, not three. The whole game is choosing the slicing direction that makes the height (or radius) easy to write. If solving for x in terms of y is ugly, slice the other way or switch to shells before you commit to an integral.
For 6.4 and 6.5, watch your units and your reference point. Most "wrong" work answers come from measuring distance from the wrong end, not from a calculus error.
Areas Between Curves
In Chapter 5, area meant the region under a single curve. Here we measure the region trapped between two curves by treating the gap between them as the height of a thin approximating rectangle.
Region between two curves
Given continuous functions with \(f(x)\ge g(x)\) on \([a,b]\), the region \(S\) lies above \(y=g(x)\), below \(y=f(x)\), and between the vertical lines \(x=a\) and \(x=b\). Each thin vertical strip has width \(\Delta x\) and height equal to the vertical gap \(f(x_i^*)-g(x_i^*)\).
From Riemann sum to integral
Summing the rectangle areas gives \(\sum_{i=1}^{n}\bigl[f(x_i^*)-g(x_i^*)\bigr]\Delta x\), an approximation that sharpens as \(n\to\infty\). The limit is exactly the definite integral of the difference of the two functions.
When the upper curve switches
If neither curve stays on top across the whole interval, split the region where the curves cross and integrate the absolute gap. In practice: integrate \(\bigl|f(x)-g(x)\bigr|\), breaking the integral at each intersection.
Integrating with respect to y
Some regions are far simpler when \(x\) is treated as a function of \(y\). Horizontal strips of height \(\Delta y\) span from a left boundary \(x_L\) to a right boundary \(x_R\).
Decision rule I want you to internalize: choose the slicing variable that lets you write each boundary as a single function. If both top and bottom are clean functions of \(x\), slice vertically. If left and right are clean functions of \(y\), slice horizontally. Example 6 is the poster child — vertical slicing there needs two integrals, horizontal needs one.
Volumes
To find a volume, slice the solid into thin slabs perpendicular to an axis. Each slab is approximately a cylinder whose volume is cross-sectional area times thickness. Summing and taking the limit turns the sum into an integral of the cross-sectional area.
Cylinder volume (the building block)
A cylinder is a region (the base) swept straight to a congruent region in a parallel plane. Its volume is base area times height, \(V=Ah\). Special cases: a circular cylinder gives \(V=\pi r^{2}h\), and a rectangular box gives \(V=lwh\).
Cross-sectional area function
Slice a general solid \(S\) by a plane perpendicular to the \(x\)-axis at position \(x\). The area of that slice is \(A(x)\). As \(x\) runs from \(a\) to \(b\), \(A(x)\) traces how the solid grows and shrinks.
Let \(S\) lie between \(x=a\) and \(x=b\). If the cross-sectional area perpendicular to the \(x\)-axis is \(A(x)\), with \(A\) continuous, then
\[ V=\lim_{n\to\infty}\sum_{i=1}^{n}A(x_i^{*})\,\Delta x=\int_{a}^{b}A(x)\,dx. \]
Solids of revolution: disks
Rotating a region about an axis produces a solid whose cross-sections are disks. The disk radius is the distance from the axis to the curve.
Solids of revolution: washers
When the region does not touch the axis, the cross-section is a washer (an annulus). Subtract the inner disk area from the outer disk area.
Solids that are not solids of revolution
The slicing definition works for any solid with a known cross-section, not just revolutions.
Disk vs. washer is not a separate formula to memorize — a disk is just a washer with inner radius zero. Always ask "does the region touch the axis of rotation?" If yes → disk. If a gap → washer. And when you rotate about a shifted line, write each radius as (the line) minus (the curve), not the bare curve value. That single habit kills most sign errors in Examples 5 and 6.
Volumes by Cylindrical Shells
Some rotations are awkward with washers because solving the curve for the other variable is hard. The shell method slices parallel to the axis of rotation, building the solid from nested thin cylindrical shells.
Volume of one shell
A thin cylindrical shell of average radius \(r\), height \(h\), and thickness \(\Delta r\) unrolls into a flat sheet of length \(2\pi r\) (its circumference), height \(h\), and thickness \(\Delta r\). Its volume is therefore \(2\pi r h\,\Delta r\) — circumference times height times thickness.
Quick mental check for the shell radius when rotating about a shifted line: it is always the perpendicular distance from the line to the strip, written as a positive quantity. About \(x=2\) with strips at \(x\le 2\), that is \(2-x\); about the \(y\)-axis it is just \(x\). Get the radius right and the rest is routine integration.
Work
Work measures effort: force acting through distance. When force is constant, work is just their product. When force varies along the path, we slice the path, treat force as constant on each tiny piece, and integrate.
Force and work (constant force)
By Newton's Second Law, force is mass times acceleration, \(F=ma\). When a constant force \(F\) moves an object a distance \(d\), the work is \(W=Fd\). SI units: force in newtons (N), distance in meters, work in joules (J). US units use foot-pounds (ft-lb).
The force on an object equals its mass times its acceleration:
\[ F=m\,\frac{d^{2}s}{dt^{2}}=ma. \]
A force of 1 N gives a 1 kg mass an acceleration of \(1\ \text{m/s}^{2}\).
The force needed to hold a spring stretched \(x\) units beyond its natural length is proportional to that stretch:
\[ f(x)=kx, \]
where \(k\) is the spring constant. Valid while the stretch \(x\) is not too large.
The pumping problems (Example 5 and its exercises) trip everyone up because two things vary: the amount of water in a layer (a volume/weight factor) and the distance it must travel. Build the layer's weight first, then multiply by the lift distance, then integrate. Decide once where your coordinate's zero sits and keep it consistent — measuring depth from the top vs. height from the bottom both work, but mixing them is the classic mistake.
Average Value of a Function
Averaging finitely many numbers is easy. To average a continuously varying quantity — like temperature over a day — we sample, average the samples, and take the limit, which produces an integral.
From discrete average to continuous average
The average of \(n\) sampled values \(f(x_1^{*}),\dots,f(x_n^{*})\) is their sum divided by \(n\). Writing \(n=(b-a)/\Delta x\) turns the average into \(\dfrac{1}{b-a}\sum f(x_i^{*})\,\Delta x\), whose limit is a definite integral.
If \(f\) is continuous on \([a,b]\), then there exists a number \(c\) in \([a,b]\) such that \(f(c)=f_{\text{ave}}\); equivalently,
\[ \int_{a}^{b}f(x)\,dx=f(c)\,(b-a). \]
Geometrically, some rectangle of height \(f(c)\) over \([a,b]\) has exactly the same area as the region under \(f\).
Two student errors I see every term here. First, people compute \(\int f\,dx\) and forget to divide by the interval width — that gives area, not an average. Second, the Mean Value Theorem for Integrals only guarantees \(c\) exists; it does not say it is unique (Example 2 has two). Use the theorem to justify that an average is actually attained, not to "find the" single point.