Hi,
I am looking to do some line fitting of Lorentzian/Cauchy functions to a type of chromatogram. It is working fine, but I need an analytical expression to the area of the curve. The function is given as:
f(x) = I*y^2/((x-x0)^2+y^2)
where I,y,x0 are constants. Using Wolfram|Alpha to integrate, I get the following:
F(x) = Iyatan((x-x0)/y) + C
The integral should be taken from -inf to inf. As atan approaches a constant value, the integral equals 0. Where am I going wrong in this? Is the integral undefined for this problem, or are my math skills just lacking? The integral should be dependent of I and y, but not x0.
I’m going to assume that I is not the imaginary number but a positive real constant. Using Mathematica I get the following (after changing I to i which might be the issue with Wolfram|Alpha):
Integrate[i*y^2/((x - x0)^2 + y^2), {x, -\[Infinity], \[Infinity]}, Assumptions -> {x0 \[Element] Reals, y > 0, i > 0}]
(* i \[Pi] y *)
Thanks for the reply! I found similar last night, when setting y=1 and I=1 and taking the numerical integral, thus getting pi, and then doubling y / I.
Now I also realized my mistake… Missing a minus sign from the result from Wolfram|Alpha.
F(x) = Iyatan((x-x0)/y) + C
In interval -inf…inf
A = Iy(atan(inf)-atan(-inf))
A = Iypi