Since we have that $\tan^{-1} \left( \dfrac{1}{s} \right) + \tan^{-1}(s) = \dfrac{\pi}{2}$, does this not imply that $s > 0$?

60 Views Asked by At

I'm currently studying the textbook An Introduction to Laplace Transforms and Fourier Series, second edition, by Phil Dyke. Chapter 2.2 Derivative Property of the Laplace Transform says the following:

The function $$Si(t) = \int_0^t \dfrac{\sin(u)}{u} \ du$$ defines the Sine Integral function which occurs in the study of optics. The formula for its Laplace Transform can now be easily derived as follows.
Let $F(t) = \sin(t)$ in the result $$\mathcal{L} \left( \dfrac{F(t)}{t} \right) = \int_s^\infty f(u) \ du$$ to give $$\begin{align*} \mathcal{L} \left( \dfrac{\sin(t)}{t} \right) &= \int_s^\infty \dfrac{du}{1 + u^2} \\ &= \left[ \tan^{-1}(u) \right]_s^\infty \\ &= \dfrac{\pi}{2} - \tan^{-1}(s) \\ &= \tan^{-1} \left( \dfrac{1}{s} \right). \end{align*}$$

So, in the last part, we have that $\dfrac{\pi}{2} - \tan^{-1}(s) = \tan^{-1} \left( \dfrac{1}{s} \right) \Rightarrow \tan^{-1} \left( \dfrac{1}{s} \right) + \tan^{-1}(s) = \dfrac{\pi}{2}$. Researching this identity, I found it said that

$$\tan^{-1}(x) + \tan^{-1} \left( \dfrac{1}{x} \right) = \begin{cases} \dfrac{\pi}{2}, & x > 0 \\ -\dfrac{\pi}{2}, & x < 0 \end{cases}$$

And we know that, in this context of the Laplace transform, $s$ is a complex number. But I don't think the general Laplace transform requires that $s > 0$, right? But, in the above case, since we have that $\tan^{-1} \left( \dfrac{1}{s} \right) + \tan^{-1}(s) = \dfrac{\pi}{2}$, does this not imply that $s > 0$?