The divergence test is a quick check used in calculus to determine whether an infinite series diverges. The rule is simple: if the terms of a series don’t approach zero as you go further and further out, the series diverges. It’s typically the first test you apply when analyzing any infinite series, and it takes only a few seconds to use.
How the Test Works
An infinite series is what you get when you add up infinitely many terms: a₁ + a₂ + a₃ + a₄, and so on forever. The divergence test asks one question: what happens to those individual terms as you move deeper into the series? Specifically, you take the limit of the nth term as n approaches infinity.
If that limit is anything other than zero, the series diverges. That’s the entire test. If the terms settle toward 3, or oscillate between 1 and negative 1, or grow without bound, then there’s no way the infinite sum can land on a finite value. The series blows up.
The logic behind this is intuitive. For an infinite sum to converge to a finite number, the terms you’re adding must get smaller and smaller, eventually contributing almost nothing. If each term stays close to some nonzero value, you’re essentially adding that value over and over, which makes the sum grow without limit, much like an arithmetic series that never settles down.
The Zero-Limit Trap
Here’s where most students make a critical mistake: if the limit of the terms does equal zero, you cannot conclude that the series converges. A limit of zero means the divergence test is inconclusive. It tells you nothing either way.
The classic example is the harmonic series: 1 + 1/2 + 1/3 + 1/4 + 1/5 + … The individual terms (1/n) clearly approach zero as n gets large. Yet the harmonic series diverges. It grows past any number you can name if you add enough terms. So the fact that terms shrink toward zero is necessary for convergence but not sufficient. Other tests, like the integral test, are needed to prove the harmonic series diverges.
Think of it this way. The divergence test is a one-way gate. It can prove divergence, but it can never prove convergence. If the limit isn’t zero, the series definitely diverges. If the limit is zero, you need a different tool.
Applying the Test Step by Step
Using the divergence test comes down to three steps:
- Identify the general term. Write down the formula for the nth term of the series, usually called aₙ.
- Take the limit. Compute the limit of aₙ as n approaches infinity. Use whatever algebra or limit techniques you normally would: dividing by the highest power, L’Hôpital’s rule, or simplification.
- Interpret the result. If the limit is nonzero (or doesn’t exist at all), the series diverges. If the limit equals zero, the test is inconclusive, and you move on to another test.
Examples That Clarify the Boundaries
Consider the series where the nth term is n/(2n + 1). As n grows large, this fraction approaches 1/2. Since 1/2 is not zero, the divergence test immediately tells you the series diverges. You’re done. No further work needed.
Now consider a series where the nth term is (n² + 5n + 6)/n³. Dividing top and bottom by n³, you get terms that simplify to roughly 1/n for large n, and the limit as n approaches infinity is zero. The divergence test gives no verdict here. You’d need to apply a different test to figure out whether this series converges or diverges.
A subtler case: if the terms of a series oscillate without settling down, say bouncing between 1 and negative 1, the limit doesn’t exist. The divergence test still applies. A nonexistent limit counts as “not equal to zero,” so the series diverges.
Why the Proof Works
The formal reasoning uses a contradiction. Suppose you have a series whose terms don’t approach zero, but assume for a moment that the series converges to some finite sum S. If the series converges, then the partial sums (the running totals as you add more and more terms) must approach S. But each individual term equals the difference between two consecutive partial sums. If both partial sums approach S, then their difference approaches zero, which would force the terms to approach zero. That contradicts the starting assumption. So the series can’t converge.
This argument also covers the case where the limit of the terms simply doesn’t exist. If the partial sums converged to a finite value, the terms would have to approach zero. Since they don’t, the partial sums can’t converge, and the series diverges.
What to Do When the Test Is Inconclusive
When the divergence test gives you a limit of zero, you’re back to square one for determining convergence. Several other tests can pick up where the divergence test leaves off, depending on the structure of the series. The integral test works well for series whose terms come from a function you can integrate. The comparison test and limit comparison test are useful when your series resembles a known convergent or divergent series. The ratio test handles series with factorials or exponential terms effectively. The alternating series test applies when terms switch between positive and negative.
Many calculus courses recommend using the divergence test as a first filter on every problem. It’s fast, requires minimal computation, and can save you from wasting time on a more complex test when the answer is sitting right on the surface. If the terms don’t shrink to zero, you’re finished. If they do, you pick the next appropriate test from your toolkit based on the structure of the series.