I am asked to determine if the double integral: $$I=\iint_{\mathbb{R}^2} \frac{x^2}{(1+x^2)(x^2 + y^2)^{3/2}}\,dx\,dy$$ converges or not.
My attempt: I let $$x = r\cos\phi.$$ $$y = r\sin\phi.$$ $$E_k : (\frac{1}{k} \leq r \leq k, 0 \leq \phi \leq 2\pi).$$
I notice that the integrand is positive which means the value of the integral is independent of the suite. I calculate the integral:
$$\lim_{k \to \infty}\iint_{E_k} \frac{r^2\cos^2\phi\cdot r}{(1+r^2\cos^2\phi)(r^3 )}\,dr\,d\phi $$
$$= \lim_{k \to \infty}\iint_{E_k} \frac{\cos^2\phi}{1+r^2\cos^2\phi}\,dr\,d\phi$$
However, I get stuck here and can't compute the integral. The answer is that the integral converges.
What can I change about my setup to make the integral easier to compute?
Is my setup flat out wrong?
Is there an easier way of determining if the integral converges?
To compute the integral $I$ on $(x,y)$ in $\mathbb R^2$, first note that, by symmetry of the integrand, $I$ is four times the integral on $x>0$, $y>0$, then define $(x,y)=(u,uv)$ and note that $dx\,dy=u\,du\,dv$ hence $$I=4\iint_{u>0,v>0}\frac{u^2}{(1+u^2)u^3(1+v^2)^{3/2}}\,u\,du\,dv=4KL$$ with $$K=\int_0^\infty\frac{du}{1+u^2}\qquad L=\int_0^\infty\frac{dv}{(1+v^2)^{3/2}}$$ One sees that $K$ and $L$ both converge hence, by Tonelli(-Fubini), $I$ converges as well, and it suffices to compute the values of $K$ and $L$ to deduce the value of $I$ by the formula above. Can you do that?