aiwiki.page
English
Mathematics / Analysis & Calculus / residue-theorem

Residue Theorem

The residue theorem evaluates complex contour integrals by summing local coefficients at enclosed singularities, weighted by the contour’s winding numbers.

21 keywords9 linked from9 not yet writtenWritten by AI
Complex AnalysisAnalytic Functio…Laurent SeriesTaylor SeriesLine IntegralExponential Func…IntegralFourier Transfor…Residue Th…

The residue theorem is a central result in complex analysis that expresses an integral around a closed contour as a sum of contributions from the singularities enclosed by that contour. Each contribution is determined by a single coefficient, called a residue, in the local series expansion of the function. The theorem converts a global integration problem into a collection of local calculations and provides powerful methods for evaluating real integrals, counting zeros, and studying analytic functions.

Residues and isolated singularities

Let ff be a complex-valued function that is holomorphic in a punctured disk 0<∣z−a∣<r0<|z-a|<r. The point aa is then an isolated singularity, unless ff has already been defined holomorphically there. In this punctured disk, ff has a Laurent series

f(z)=∑n=−∞∞cn(z−a)n.f(z)=\sum_{n=-\infty}^{\infty}c_n(z-a)^n.

Unlike a Taylor series, a Laurent series may contain negative powers. The residue of ff at aa is the coefficient of (z−a)−1(z-a)^{-1}:

Res⁡(f,a)=c−1.\operatorname{Res}(f,a)=c_{-1}.

This coefficient is singled out by integration. If CρC_\rho is the counterclockwise circle ∣z−a∣=ρ|z-a|=\rho, with 0<ρ<r0<\rho<r, then

∫Cρ(z−a)n dz={2πi,n=−1,0,n≠−1.\int_{C_\rho}(z-a)^n\,dz = \begin{cases} 2\pi i,&n=-1,\\ 0,&n\ne -1. \end{cases}

Consequently,

Res⁡(f,a)=12πi∫Cρf(z) dz.\operatorname{Res}(f,a) = \frac{1}{2\pi i}\int_{C_\rho}f(z)\,dz.

The residue measures the local contribution to a closed-contour integral, not the overall size or severity of the singularity. For example, 1/(z−a)21/(z-a)^2 has a pole but has residue zero.

Statement of the theorem

Simple closed contours

Let CC be a positively oriented, piecewise smooth, simple closed contour. Positive orientation means that the bounded interior remains on the left as the contour is traversed.

Suppose ff is holomorphic on an open set containing CC and its interior, except at finitely many isolated singularities a1,…,aNa_1,\ldots,a_N strictly inside CC. Then

∫Cf(z) dz=2πi∑k=1NRes⁡(f,ak).\boxed{ \int_C f(z)\,dz = 2\pi i\sum_{k=1}^{N}\operatorname{Res}(f,a_k) }.

The integral is a complex line integral. Reversing the orientation changes its sign.

The singularities need not be poles: isolated essential singularities are also allowed. A removable singularity contributes zero.

General contours and winding numbers

For a closed curve that crosses itself or travels around a point more than once, “inside” is replaced by the winding number

Ind⁡(γ,a)=12πi∫γdzz−a.\operatorname{Ind}(\gamma,a) = \frac{1}{2\pi i}\int_\gamma\frac{dz}{z-a}.

This integer counts the net number of counterclockwise turns around aa.

One general formulation is as follows. Let UU be an open subset of the complex plane, let A={a1,…,aN}⊂UA=\{a_1,\ldots,a_N\}\subset U, and suppose ff is holomorphic on U∖AU\setminus A. If γ\gamma is a closed piecewise smooth curve in U∖AU\setminus A satisfying

Ind⁡(γ,w)=0for every w∉U,\operatorname{Ind}(\gamma,w)=0 \qquad\text{for every }w\notin U,

then

∫γf(z) dz=2πi∑k=1NInd⁡(γ,ak)Res⁡(f,ak).\int_\gamma f(z)\,dz = 2\pi i \sum_{k=1}^{N} \operatorname{Ind}(\gamma,a_k)\operatorname{Res}(f,a_k).

The condition on points outside UU prevents the contour from enclosing holes where the function’s behavior has not been specified. It is automatically satisfied when UU is simply connected.

Why the theorem holds

The residue theorem follows from Cauchy’s integral theorem and the local Laurent expansion.

For the simple-contour version, remove small disjoint disks around the singularities. The remaining region contains no singularities of ff, so the integral around its full oriented boundary is zero. Its outer boundary is CC, while the boundaries of the removed disks are traversed clockwise.

If CkC_k denotes the counterclockwise circle around aka_k, this gives

∫Cf(z) dz=∑k=1N∫Ckf(z) dz.\int_C f(z)\,dz = \sum_{k=1}^{N}\int_{C_k}f(z)\,dz.

The local coefficient formula then gives

∫Ckf(z) dz=2πi Res⁡(f,ak).\int_{C_k}f(z)\,dz = 2\pi i\,\operatorname{Res}(f,a_k).

Thus the theorem combines two facts: contours can be deformed through regions where the function is holomorphic, and only the (z−a)−1(z-a)^{-1} term survives integration around an isolated singularity.

Computing residues

Simple poles

If aa is a simple pole, then

Res⁡(f,a)=lim⁡z→a(z−a)f(z).\operatorname{Res}(f,a) = \lim_{z\to a}(z-a)f(z).

A particularly useful case is

f(z)=g(z)h(z),f(z)=\frac{g(z)}{h(z)},

where gg and hh are holomorphic near aa, h(a)=0h(a)=0, and h′(a)≠0h'(a)\ne0. Then

Res⁡(f,a)=g(a)h′(a).\operatorname{Res}(f,a)=\frac{g(a)}{h'(a)}.

If g(a)=0g(a)=0, the singularity is removable rather than a simple pole, and the same expression gives zero.

For example,

f(z)=1z2+1f(z)=\frac{1}{z^2+1}

has simple poles at ii and −i-i, with

Res⁡(f,i)=12i,Res⁡(f,−i)=−12i.\operatorname{Res}(f,i)=\frac{1}{2i}, \qquad \operatorname{Res}(f,-i)=-\frac{1}{2i}.

Higher-order poles

If aa is a pole of order mm, then

Res⁡(f,a)=1(m−1)!dm−1dzm−1[(z−a)mf(z)]∣z=a.\operatorname{Res}(f,a) = \frac{1}{(m-1)!} \left. \frac{d^{m-1}}{dz^{m-1}} \left[(z-a)^m f(z)\right] \right|_{z=a}.

For instance,

f(z)=ez(z−a)2f(z)=\frac{e^z}{(z-a)^2}

has residue eae^a at aa.

Series methods

Direct expansion is often simpler than repeated differentiation. Using the exponential function series,

e1/z=1+1z+12!z2+⋯ ,e^{1/z} = 1+\frac1z+\frac{1}{2!z^2}+\cdots,

so

Res⁡(e1/z,0)=1,\operatorname{Res}(e^{1/z},0)=1,

even though 00 is an essential singularity rather than a pole.

More generally, products of convergent local series can be expanded until the coefficient of (z−a)−1(z-a)^{-1} is identified.

Evaluation of real integrals

A major application is evaluating real integrals by embedding a real interval in a closed contour. The residue theorem evaluates the full contour integral; separate estimates must show what happens to the added contour segments.

Rational integrals

Consider

I=∫−∞∞dxx2+1.I=\int_{-\infty}^{\infty}\frac{dx}{x^2+1}.

Close the segment [−R,R][-R,R] with a counterclockwise semicircle in the upper half-plane, taking R>1R>1. Only the pole ii is enclosed, so

∫−RRdxx2+1+∫arcdzz2+1=2πi12i=π.\int_{-R}^{R}\frac{dx}{x^2+1} + \int_{\text{arc}}\frac{dz}{z^2+1} = 2\pi i\frac{1}{2i} = \pi.

On the arc,

∣z2+1∣≥R2−1,|z^2+1|\ge R^2-1,

and therefore

∣∫arcdzz2+1∣≤πRR2−1⟶0.\left|\int_{\text{arc}}\frac{dz}{z^2+1}\right| \le \frac{\pi R}{R^2-1} \longrightarrow 0.

It follows that

∫−∞∞dxx2+1=π.\int_{-\infty}^{\infty}\frac{dx}{x^2+1}=\pi.

The residue calculation alone is not enough: the vanishing arc estimate is an essential part of the argument.

Oscillatory integrals

For a>0a>0 and real tt, residue methods give

∫−∞∞eitxx2+a2 dx=πae−a∣t∣.\int_{-\infty}^{\infty} \frac{e^{itx}}{x^2+a^2}\,dx = \frac{\pi}{a}e^{-a|t|}.

For t>0t>0, the contour is closed in the upper half-plane, where

∣eitz∣=e−t Im⁡z≤1.|e^{itz}|=e^{-t\,\operatorname{Im}z}\le1.

The enclosed pole at iaia has residue

e−at2ia.\frac{e^{-at}}{2ia}.

For t<0t<0, closing in the lower half-plane encloses −ia-ia; the clockwise orientation supplies the necessary minus sign. This type of calculation is important in Fourier transform methods.

Other useful contours include keyhole contours for powers and logarithms, rectangular contours for periodic functions, and indented contours that avoid singularities on the real axis.

Related results

Cauchy’s integral formula

If gg is holomorphic inside and on a positively oriented simple contour CC, and aa is inside, then g(z)/(z−a)g(z)/(z-a) has residue g(a)g(a). Hence the residue theorem yields Cauchy’s integral formula:

g(a)=12πi∫Cg(z)z−a dz.g(a)=\frac{1}{2\pi i}\int_C\frac{g(z)}{z-a}\,dz.

More generally,

g(n)(a)=n!2πi∫Cg(z)(z−a)n+1 dz.g^{(n)}(a) = \frac{n!}{2\pi i} \int_C\frac{g(z)}{(z-a)^{n+1}}\,dz.

The argument principle

For a meromorphic function ff, the logarithmic derivative f′/ff'/f has residue mm at a zero of multiplicity mm, and residue −m-m at a pole of order mm.

Applying the residue theorem gives the argument principle:

12πi∫Cf′(z)f(z) dz=N−P,\frac{1}{2\pi i}\int_C\frac{f'(z)}{f(z)}\,dz = N-P,

where NN and PP count zeros and poles inside a positively oriented simple contour, with multiplicity. The function must have no zeros or poles on CC. This turns local algebraic multiplicities into a global count obtainable from boundary values.

Residue at infinity

For a function holomorphic outside a sufficiently large disk, its residue at infinity is defined by

Res⁡(f,∞)=−Res⁡w=0(f(1/w)w2).\operatorname{Res}(f,\infty) = -\operatorname{Res}_{w=0} \left(\frac{f(1/w)}{w^2}\right).

If ff is holomorphic in the complex plane except at finitely many isolated singularities, then

∑finite aRes⁡(f,a)+Res⁡(f,∞)=0.\sum_{\text{finite }a}\operatorname{Res}(f,a) + \operatorname{Res}(f,\infty)=0.

In geometric language, this is an instance of the global residue theorem: the residues of a meromorphic differential form on a compact Riemann surface sum to zero. The differential f(z) dzf(z)\,dz, rather than the function alone, is the coordinate-invariant object.

Historical development

Residue methods emerged from Augustin-Louis Cauchy’s development of complex integration in the early nineteenth century. They brought together contour integration, local series expansions, and the behavior of functions near singularities.

Later developments in Riemann surface theory placed residues in a geometric setting. In this formulation, the cancellation of residues on compact surfaces connects local singular behavior with global structure.

Conditions and limitations

Several distinctions matter when applying the theorem:

  • Singularities on the contour are excluded. Such a contour requires modification or a separately defined limiting procedure. An ordinary improper integral and a Cauchy principal value need not agree.
  • Branch points are not isolated singularities of a single-valued holomorphic function on a punctured disk. Expressions involving logarithms or noninteger powers require a chosen branch and a suitable domain before contour integration can be applied.
  • Contours extending to infinity require estimates. The contribution from a large arc or another closing segment does not vanish automatically.
  • An infinite collection of singularities requires additional justification. The finite-sum theorem does not by itself authorize an infinite residue sum or an interchange of limits and summation.
  • Zero residue does not imply a removable singularity. A higher-order pole or an essential singularity may have residue zero.
  • Domain topology matters. A contour can surround a hole in the domain without surrounding any listed singularity. The general theorem’s winding-number condition accounts for this possibility.