Improper integrals are limits, not unusual notation
Replace infinite bounds or unbounded integrands with limits before evaluating, then decide whether the result converges.
What the idea means, why its conditions matter, and where it connects.
Reviewed July 11, 2026An improper integral converges only when every required defining limit exists and is finite. Otherwise it diverges, even if symbolic antiderivatives can be written.
Two sources of impropriety
An interval may extend to infinity, or the integrand may become unbounded at an endpoint or inside the interval. Each issue must be replaced by a one-sided limit.
If a singularity lies inside the interval, split the integral there and require both pieces to converge independently.
Antiderivatives do not settle convergence
The Fundamental Theorem applies first on finite intervals where the integrand behaves appropriately. Only after evaluation do you take the defining limit.
An expression such as infinity minus infinity is not a cancellation; the pieces must be analyzed separately.
Comparison can avoid hard antiderivatives
For nonnegative functions, comparison with a known p-integral or another benchmark can prove convergence or divergence without an exact formula.
Evaluate ∫₁∞ 1/x² dx.
Replace the infinite bound.
Evaluate on a finite interval.
Now take the limit.
Common mistakes
- Plugging infinity into an antiderivative as if it were a number.
- Failing to split at an interior singularity.
- Declaring convergence because the integrand approaches zero.
Three takeaways
- Rewrite first as one or more limits.
- Every improper piece must converge.
- Comparison is often more useful than exact integration.