I need to calculate ∫∞0lnx1+xn ,n≥2 using complex analysis. I probably need to use the Residue Theorem.
I use the function f(z)=lnz1+zn in the normal branch.
I've tried to use this contour.
Where θ is an angle so that only z0 will be in the domain (I hope this is clear from the drawing)
I estimated |∫ΓR| and |∫Γϵ| and showed that they must tend to 0 when ϵ→0 and R→∞. (Is it true?)
However I'm having trouble calculating ∫Γ2 . Does it have something to do with choosing the "right" θ?
Any ideas?
Thanks!
UPDATE
After Christopher's comment I chose θ=2πn which gives, after the paramatrization Γ(t)=te2πin, t∈(ϵ,R):
∫Γ2lnz1+zndz=∫Rϵln(te2πin)1+tne2πindt=e2πin∫Rϵlnt1+tndt+ie2πin∫Rϵ2πn1+tndt= =e2πin∫∞0lnt1+tndt+ie2πin∫∞02πn1+tndt
But I have no idea how to deal with the second integral.
Answer
Let us define, for r>0, Γr={reiα : 0≤α≤θ} The circular arcs of your contour are then Γϵ and ΓR.
We note that, for z∈Γϵ |lnz1+zn|≤|lnz|≤2lnϵ if ϵ is small enough. So |∫Γϵf(z)|≤2θϵlnϵ which goes to 0 with ϵ. Also, for R large enough, |lnz|≤2lnR when z∈ΓR. So |lnz1+zn|≤2lnRRn−1 hence |∫ΓRf(z)|≤θR2lnRRn−1 which goes to 0 when R→∞, if n>1. (So, yes, these two pieces go to zero)
Now take Γ1={r : ϵ≤r≤R} and Γ2=eiθΓ1; define the contour Γ by walking along Γ1, then ΓR, then backwards along Γ2 and Γϵ.
The roots of 1+zn=0 are the n-th roots of −1, that is zk=ei(2k+1)πn so, if we take θ∈(πn,3πn), Γ surrounds exactly one pole of f.
Now, following the hint of Christopher A. Wong, we set θ=2πn; therefore, if z∈Γ2, we get zn=|z|neinθ=|z|n and lnz=ln|z|+i2πn.
So ∫Γ2f(z)=∫Rϵlnx+i2πn1+xnei2πndx
We recall that ∫∞011+xndx=π/nsin(π/n) with the same method we used now (integrating on Γ!).
Call J=∫∞0lnx/(1+xn)dx and L the residue of f(z)=lnz/(1+zn) at z=eiπn, then 2iπL=J−ei2πn(J+i2πnπ/nsin(π/n)) that is J=2iπL+iei2πn2π2/n2sin(π/n)1−ei2πn
Now it is enough to perform the computations.
No comments:
Post a Comment