Volumes by Cylindrical Shells
The
disk method
cut a solid into circular slabs perpendicular to the axis of rotation. When you rotate about the
y-axis but your region is described as a function of
x, those slabs become a nuisance — you would have to invert
f. The shell method dodges that entirely: instead of
slicing across, it builds the solid out of thin nested cylindrical shells, like the rings
of a tree.
Each shell is a thin-walled tube of radius x, height
f(x), and wall-thickness dx. Summing them
across [a, b] gives
V = \int_a^{b} 2\pi x\, f(x)\,dx.
Unroll one shell — the derivation
The factor 2\pi x looks mysterious until you slit a shell open and flatten
it.
Step 1 — picture one shell. Rotate a thin vertical strip of the region, at distance
x from the y-axis, about that axis. It sweeps
out a hollow tube of radius x and height
f(x), with wall thickness equal to the strip's width
dx.
Step 2 — slit it and unroll. Cut the tube straight down one side and flatten it.
It becomes a thin rectangular sheet: its width is the circumference of the circle it used to be,
2\pi x; its height is f(x); and its thickness
is still dx.
Step 3 — the shell's volume is that sheet's. Volume = (circumference) × (height) ×
(thickness):
\Delta V = \underbrace{2\pi x}_{\text{circumference}} \cdot \underbrace{f(x)}_{\text{height}} \cdot \underbrace{\Delta x}_{\text{thickness}}.
Step 4 — nest the shells and refine. Sum over all radii from
a to b and let the walls grow infinitely thin:
V = \lim_{\Delta x \to 0} \sum 2\pi x_i\, f(x_i)\,\Delta x = \int_a^{b} 2\pi x\, f(x)\,dx.
Worked example: rotate y = x^2 on [0, 1] about the y-axis
Step 1 — identify radius and height. A shell at position x
has radius x and height f(x) = x^2.
Step 2 — build the integrand 2\pi x\, f(x):
V = \int_0^{1} 2\pi x \cdot x^2\,dx = 2\pi \int_0^{1} x^3\,dx.
Step 3 — integrate and evaluate.
V = 2\pi \left[\frac{x^4}{4}\right]_0^{1} = 2\pi \cdot \frac{1}{4} = \frac{\pi}{2}.
The bowl-shaped solid swept out by y = x^2 around the
y-axis has volume \pi/2. (Notice how the extra
factor of x from the radius nudged the
power from
x^2 up to x^3 — shells integrate a power
higher than disks would.)
Rotate the region under y = f(x) \ge 0 on
[a, b] (with 0 \le a) about the
y-axis. Its volume is
V = \int_a^{b} 2\pi x\, f(x)\,dx,
-
from a typical shell of radius x,
height f(x), and thickness
dx — volume 2\pi x\, f(x)\,dx (circumference
× height × thickness).
-
To rotate about a vertical line x = c instead, replace the radius
x by the distance |x - c|.
Shells and disks compute the same volume; they differ only in which variable you must
integrate. The rule of thumb: slice perpendicular to the axis of rotation for disks,
parallel to it for shells — and pick whichever spares you from inverting a function.
Rotate the region under y = x^2, 0 \le x \le 1,
about the y-axis. With shells you integrate the given
f(x) = x^2 directly — one clean line, as above. With disks
(slabs perpendicular to the y-axis) you must first solve
x = \sqrt{y}, integrate in y, and subtract from
the enclosing cylinder:
V = \pi(1)^2(1) - \pi \int_0^{1} (\sqrt{y})^2\,dy = \pi - \pi \cdot \frac{1}{2} = \frac{\pi}{2}.
Same \pi/2 — but you had to invert the curve and reason about a hole.
The shell setup needed neither. Whenever rotating a "y in terms of
x" region about the y-axis, reach for shells
first.
Unroll the shell yourself
On the left, the curve y = x^2 with a shell drawn as a vertical strip at
radius x. On the right, that same shell unrolled into a
flat rectangle of width 2\pi x (its circumference) and height
f(x) = x^2. Slide the radius and watch the rectangle stretch: its area is
the shell's lateral area 2\pi x\, f(x), and the readout shows it.