Sign in to save

Bookmark this page so you can find it later.

Sign in to save

Bookmark this page so you can find it later.

Improper integrals extend the idea of definite integration to cases where the interval is unbounded or the function becomes infinite at some point. They matter because many important quantities in physics, probability, and engineering are modeled by areas that stretch forever or blow up near a boundary. A normal definite integral cannot be applied directly in these cases, so calculus replaces it with a limit process.

The key question is whether that limit exists and gives a finite value.

There are two main types of improper integrals. One type occurs on an infinite interval, such as 11x2dx\int_1^{\infty} \frac{1}{x^2}\,dx, and the other occurs when the integrand has a vertical asymptote or discontinuity. In both situations, the integral is defined by taking a limit of ordinary definite integrals over safer intervals.

If the limit exists, the improper integral converges, and if not, it diverges.

Understanding Improper Integrals

The limit process works by placing a temporary stopping point where the problem occurs. For a tail that continues to the right, calculate the accumulated area up to a large number, then observe what happens as that number increases without bound. For example, the accumulated area under one over x cubed from one to a cutoff gets closer and closer to one half.

Each extra stretch contributes less area than the stretch before it. In contrast, the accumulated area under one over x from one to a cutoff keeps growing, although it grows slowly. A graph can make this slow growth look almost flat, so the algebraic limit is the reliable test.

A vertical asymptote needs equal care. A function becoming unbounded does not automatically mean its integral diverges. Near zero, one over the square root of x rises infinitely high, yet the region near zero is narrow enough to have finite area.

One over x squared rises too sharply near zero, so its area is unbounded. This difference is important because height alone does not decide the result. Students should examine how rapidly the function grows as the input approaches the troublesome point.

When the bad point lies inside an interval, the left side and right side must each produce a finite result. A finite value from one side cannot repair divergence on the other side.

Comparison tests turn a difficult formula into a familiar growth problem. They work best when the function stays nonnegative near the endpoint being studied. If a positive function is no larger than another function with a finite total area, then the first function has finite total area too.

If it is no smaller than a function whose area is unbounded, then it must diverge. Powers provide the usual benchmark. Far from zero, a larger power in the denominator causes faster decay.

Close to zero, that same larger power creates a more severe blowup. Limit comparison is useful when two formulas look different but their ratio approaches a positive finite number near the endpoint. In that case, both integrals have the same convergence behavior.

Improper integrals appear whenever a model includes an unlimited range or an extreme boundary. In probability, a density function may extend across all positive values, yet its total probability must be finite. In physics, decaying fields and energy distributions are often integrated over large distances.

Real measurements never reach infinity, but the integral tells whether the unmeasured tail becomes negligible. One important warning applies on an interval extending in both directions forever. Treat the left tail and the right tail separately.

Apparent cancellation between positive and negative parts can hide two divergent pieces. This is why symmetric graphing or plugging in matching cutoffs is not enough to prove convergence.

Key Facts

  • For an infinite upper limit, af(x)dx=limbabf(x)dx\int_a^{\infty} f(x)\,dx = \lim_{b\to\infty} \int_a^b f(x)\,dx.
  • For an infinite lower limit, bf(x)dx=limaabf(x)dx\int_{-\infty}^b f(x)\,dx = \lim_{a\to -\infty} \int_a^b f(x)\,dx.
  • If cc is a discontinuity in [a,b][a,b], split the integral and take one-sided limits from each side of cc.
  • An improper integral converges only if the defining limit exists and is finite.
  • pp-test on [1,)[1, \infty): 11xpdx\int_1^{\infty} \frac{1}{x^p}\,dx converges if p>1p > 1 and diverges if p1p \leq 1.
  • pp-test near 00: 011xpdx\int_0^1 \frac{1}{x^p}\,dx converges if p<1p < 1 and diverges if p1p \geq 1.

Vocabulary

Improper integral
An integral defined by a limit because the interval is \infty or the function is unbounded at some point.
Converges
Describes an improper integral whose defining limit exists and equals a finite number.
Diverges
Describes an improper integral whose defining limit does not exist or is \infty.
Vertical asymptote
A vertical line where a function grows without bound and is not defined in the usual way.
Infinite interval
An interval that extends to infinity, such as [2, infinity) or (-infinity, 5].

Common Mistakes to Avoid

  • Treating \infty like a regular endpoint, which is wrong because \infty is not a number and the integral must be rewritten as a limit first.
  • Ignoring a discontinuity inside the interval, which is wrong because the integral must be split at the point where the function is undefined or unbounded.
  • Combining one-sided limits automatically, which is wrong because each side of a vertical asymptote must converge separately before adding them.
  • Assuming a small-looking graph means a finite area, which is wrong because a region can extend forever and still diverge depending on the rate of decay.

Practice Questions

  1. 1 Evaluate 11x3dx\int_1^{\infty} \frac{1}{x^3} \,dx and state whether it converges or diverges.
  2. 2 Evaluate 011xdx\int_0^1 \frac{1}{\sqrt{x}} \,dx and state whether it converges or diverges.
  3. 3 Explain why 111xdx\int_{-1}^{1} \frac{1}{x}\,dx is not a convergent improper integral even though the areas on the left and right may seem to cancel.