I consider, for $s>\frac{1}{2}$, the space $L^{2,s}(\mathbb{R}^3)=\bigg\lbrace{f: \int_{\mathbb{R}^3}f(x)^2(1+x^2)^{s}<\infty\bigg\rbrace}$ and I have to show that the function $$f(x)=\frac{e^{i\sqrt{\lambda+i\varepsilon}x}}{x}$$ converges in the norm of the considered space to$$g(x)\frac{e^{i\sqrt{\lambda}x}}{x}$$when $\varepsilon\to 0$ where $\lambda>0$ and from complex analysis $\sqrt{\lambda+i\varepsilon}=\sqrt{\lambda_\varepsilon}+i\frac{\varepsilon}{2\sqrt{\lambda_\varepsilon}}$ with $\lambda_\varepsilon=\frac{\sqrt{\lambda^2+\varepsilon^2}+\lambda}{2}$, so $\lambda_\varepsilon\to\lambda$ and $$e^{i\sqrt{\lambda+i\varepsilon}}=e^{i\sqrt{\lambda_\varepsilon}}e^{\frac{\varepsilon}{2\sqrt{\lambda_\varepsilon}}}\to e^{i\sqrt{\lambda}}$$. Then I consider the integral $$\int_{\mathbb{R}^3}\frac{e^{i\sqrt{\lambda+i\varepsilon}}e^{i\sqrt{\lambda}}^2}{x^2}\frac{1}{(1+x^2)^s}dx $$ The integrand tend to $0$ when $\varepsilon\to 0$ almost everywhere and it is majorized by $$\frac{1}{x^2(1+x^2)^s}$$ integrable. So by the theorem of dominated convergence I have the thesis. Is it right?
Take the 2minute tour
×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.
Yes, it is correct, as $$\int_{\{a\leqslant x\leqslant b\}}\frac 1{x^2(1+x^2)^s}dx=C\int_a^b\frac{r^{31}}{r^2(1+r^2)^s}ds=C\int_a^b\frac{ds}{(1+r^2)^s},$$ where $C$ is an universal constant. 

