Problem with divergent integral

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
2 replies · 1K views
parton
Messages
79
Reaction score
1
I'm confused with the following integral.

Let a > 1.

[tex]\int_{a}^{\infty} \left( \dfrac{1}{t} - \dfrac{1}{t-1} \right) ~ dt = \left[ \log \left| \dfrac{t}{t-1} \right| \right]_{a}^{\infty} = - \log \left| \dfrac{a}{a-1} \right|[/tex]

This should be the correct result. But I could also decompose the integral into two parts (because the integrand is a sum) and compute:

[tex]\int_{a}^{\infty} \left( \dfrac{1}{t} - \dfrac{1}{t-1} \right) ~ dt = \left| \log \vert t \vert \right|_{a}^{\infty} - \left[ \log \vert t - 1 \vert \right]_{a}^{\infty} = - \log \left| \dfrac{a}{a-1} \right| + \infty - \infty[/tex]

But [tex]\infty - \infty[/tex] is of course not defined!

Where did I make a mistake? I don't find it.
 
Physics news on Phys.org
Decomposing the integral into two parts is justified only if both integrals are finite.