A complex number is z = a + b i z = a + bi z = a + bi where a , b ∈ R a, b \in \mathbb{R} a , b ∈ R and i 2 = − 1 i^2 = -1 i 2 = − 1 . We call a = R e ( z ) a = \mathrm{Re}(z) a = Re ( z ) the real part and b = I m ( z ) b = \mathrm{Im}(z) b = Im ( z ) the imaginary part.
Arithmetic: ( a + b i ) + ( c + d i ) = ( a + c ) + ( b + d ) i (a + bi) + (c + di) = (a + c) + (b + d)i ( a + bi ) + ( c + d i ) = ( a + c ) + ( b + d ) i and ( a + b i ) ( c + d i ) = ( a c − b d ) + ( a d + b c ) i (a + bi)(c + di) = (ac - bd) + (ad + bc)i ( a + bi ) ( c + d i ) = ( a c − b d ) + ( a d + b c ) i .
Proposition 1.1 (Properties of Complex Arithmetic). For all z , w ∈ C z, w \in \mathbb{C} z , w ∈ C :
z + w = w + z z + w = w + z z + w = w + z and z w = w z zw = wz z w = w z (commutativity)( z + w ) + u = z + ( w + u ) (z + w) + u = z + (w + u) ( z + w ) + u = z + ( w + u ) and ( z w ) u = z ( w u ) (zw)u = z(wu) ( z w ) u = z ( w u ) (associativity)z ( w + u ) = z w + z u z(w + u) = zw + zu z ( w + u ) = z w + z u (distributivity)There exist additive identity 0 0 0 and multiplicative identity 1 1 1 . Every z ≠ 0 z \neq 0 z = 0 has a multiplicative inverse 1 z = z ˉ ∣ z ∣ 2 \frac{1}{z} = \frac{\bar{z}}{|z|^2} z 1 = ∣ z ∣ 2 z ˉ . Remark. The complex field C \mathbb{C} C cannot be ordered: there is no total ordering on C \mathbb{C} C Compatible with the field operations. In particular, i 2 = − 1 i^2 = -1 i 2 = − 1 precludes any such ordering.
Definition. The complex conjugate of z = a + b i z = a + bi z = a + bi is z ˉ = a − b i \bar{z} = a - bi z ˉ = a − bi .
Proposition 1.2. For all z , w ∈ C z, w \in \mathbb{C} z , w ∈ C :
z + w ‾ = z ˉ + w ˉ \overline{z + w} = \bar{z} + \bar{w} z + w = z ˉ + w ˉ and z w ‾ = z ˉ w ˉ \overline{zw} = \bar{z}\bar{w} z w = z ˉ w ˉ z z ˉ = ∣ z ∣ 2 z\bar{z} = |z|^2 z z ˉ = ∣ z ∣ 2 z + z ˉ = 2 R e ( z ) z + \bar{z} = 2\,\mathrm{Re}(z) z + z ˉ = 2 Re ( z ) and z − z ˉ = 2 i I m ( z ) z - \bar{z} = 2i\,\mathrm{Im}(z) z − z ˉ = 2 i Im ( z ) z ˉ ˉ = z \bar{\bar{z}} = z z ˉ ˉ = z Definition. The modulus (or absolute value) of z = a + b i z = a + bi z = a + bi is ∣ z ∣ = a 2 + b 2 |z| = \sqrt{a^2 + b^2} ∣ z ∣ = a 2 + b 2 .
Proposition 1.3 (Modulus Properties). For all z , w ∈ C z, w \in \mathbb{C} z , w ∈ C :
∣ z ∣ ≥ 0 |z| \geq 0 ∣ z ∣ ≥ 0 with equality iff z = 0 z = 0 z = 0 ∣ z w ∣ = ∣ z ∣ ∣ w ∣ |zw| = |z||w| ∣ z w ∣ = ∣ z ∣∣ w ∣ ∣ z + w ∣ ≤ ∣ z ∣ + ∣ w ∣ |z + w| \leq |z| + |w| ∣ z + w ∣ ≤ ∣ z ∣ + ∣ w ∣ (triangle inequality)∣ ∣ z ∣ − ∣ w ∣ ∣ ≤ ∣ z − w ∣ \bigl||z| - |w|\bigr| \leq |z - w| ∣ z ∣ − ∣ w ∣ ≤ ∣ z − w ∣ (reverse triangle inequality)Proof of (3). ∣ z + w ∣ 2 = ( z + w ) ( z ˉ + w ˉ ) = ∣ z ∣ 2 + z w ˉ + z ˉ w + ∣ w ∣ 2 = ∣ z ∣ 2 + 2 R e ( z w ˉ ) + ∣ w ∣ 2 ≤ ∣ z ∣ 2 + 2 ∣ z ∣ ∣ w ∣ + ∣ w ∣ 2 = ( ∣ z ∣ + ∣ w ∣ ) 2 |z + w|^2 = (z + w)(\bar{z} + \bar{w}) = |z|^2 + z\bar{w} + \bar{z}w + |w|^2 = |z|^2 + 2\,\mathrm{Re}(z\bar{w}) + |w|^2 \leq |z|^2 + 2|z||w| + |w|^2 = (|z| + |w|)^2 ∣ z + w ∣ 2 = ( z + w ) ( z ˉ + w ˉ ) = ∣ z ∣ 2 + z w ˉ + z ˉ w + ∣ w ∣ 2 = ∣ z ∣ 2 + 2 Re ( z w ˉ ) + ∣ w ∣ 2 ≤ ∣ z ∣ 2 + 2∣ z ∣∣ w ∣ + ∣ w ∣ 2 = ( ∣ z ∣ + ∣ w ∣ ) 2 . The inequality follows from R e ( z w ˉ ) ≤ ∣ z w ˉ ∣ = ∣ z ∣ ∣ w ∣ \mathrm{Re}(z\bar{w}) \leq |z\bar{w}| = |z||w| Re ( z w ˉ ) ≤ ∣ z w ˉ ∣ = ∣ z ∣∣ w ∣ . ■ \blacksquare ■
Every non-zero complex number can be written in polar form :
z = r ( cos θ + i sin θ ) = r e i θ z = r(\cos\theta + i\sin\theta) = re^{i\theta} z = r ( cos θ + i sin θ ) = r e i θ
Where r = ∣ z ∣ = a 2 + b 2 r = |z| = \sqrt{a^2 + b^2} r = ∣ z ∣ = a 2 + b 2 is the modulus and θ = arg ( z ) \theta = \arg(z) θ = arg ( z ) is the argument .
Definition. The principal argument A r g ( z ) \mathrm{Arg}(z) Arg ( z ) is the unique θ ∈ ( − π , π ] \theta \in (-\pi, \pi] θ ∈ ( − π , π ] Such that z = ∣ z ∣ e i θ z = |z|e^{i\theta} z = ∣ z ∣ e i θ . The argument arg ( z ) \arg(z) arg ( z ) is multi-valued: arg ( z ) = A r g ( z ) + 2 π k \arg(z) = \mathrm{Arg}(z) + 2\pi k arg ( z ) = Arg ( z ) + 2 π k for k ∈ Z k \in \mathbb{Z} k ∈ Z .
Proposition 1.4. If z 1 = r 1 e i θ 1 z_1 = r_1 e^{i\theta_1} z 1 = r 1 e i θ 1 and z 2 = r 2 e i θ 2 z_2 = r_2 e^{i\theta_2} z 2 = r 2 e i θ 2 Then z 1 z 2 = r 1 r 2 e i ( θ 1 + θ 2 ) z_1 z_2 = r_1 r_2 e^{i(\theta_1 + \theta_2)} z 1 z 2 = r 1 r 2 e i ( θ 1 + θ 2 ) and z 1 / z 2 = ( r 1 / r 2 ) e i ( θ 1 − θ 2 ) z_1/z_2 = (r_1/r_2)\, e^{i(\theta_1 - \theta_2)} z 1 / z 2 = ( r 1 / r 2 ) e i ( θ 1 − θ 2 ) .
Solution Problem. Convert z = − 1 + 3 i z = -1 + \sqrt{3}\,i z = − 1 + 3 i to polar form and find all arguments.
∣ z ∣ = ( − 1 ) 2 + ( 3 ) 2 = 1 + 3 = 2 |z| = \sqrt{(-1)^2 + (\sqrt{3})^2} = \sqrt{1 + 3} = 2 ∣ z ∣ = ( − 1 ) 2 + ( 3 ) 2 = 1 + 3 = 2 .
R e ( z ) = − 1 < 0 \mathrm{Re}(z) = -1 \lt 0 Re ( z ) = − 1 < 0 and I m ( z ) = 3 > 0 \mathrm{Im}(z) = \sqrt{3} \gt 0 Im ( z ) = 3 > 0 So z z z is in the second quadrant.
θ = arctan ( 3 − 1 ) = 2 π 3 \theta = \arctan\!\left(\frac{\sqrt{3}}{-1}\right) = \frac{2\pi}{3} θ = arctan ( − 1 3 ) = 3 2 π (adjusting to second quadrant).
Polar form: z = 2 e 2 π i / 3 z = 2\,e^{2\pi i/3} z = 2 e 2 π i /3 .
All arguments: arg ( z ) = 2 π 3 + 2 π k \arg(z) = \frac{2\pi}{3} + 2\pi k arg ( z ) = 3 2 π + 2 π k for k ∈ Z k \in \mathbb{Z} k ∈ Z .
Problem. Convert z = 3 e − i π / 4 z = 3e^{-i\pi/4} z = 3 e − iπ /4 to rectangular form.
z = 3 ( cos ( − π 4 ) + i sin ( − π 4 ) ) = 3 ( 2 2 − i 2 2 ) = 3 2 2 − 3 2 2 i z = 3\left(\cos\!\left(-\frac{\pi}{4}\right) + i\sin\!\left(-\frac{\pi}{4}\right)\right) = 3\left(\frac{\sqrt{2}}{2} - i\,\frac{\sqrt{2}}{2}\right) = \frac{3\sqrt{2}}{2} - \frac{3\sqrt{2}}{2}\,i z = 3 ( cos ( − 4 π ) + i sin ( − 4 π ) ) = 3 ( 2 2 − i 2 2 ) = 2 3 2 − 2 3 2 i .
Problem. Express z = − 3 − 4 i z = -3 - 4i z = − 3 − 4 i in polar form.
∣ z ∣ = 9 + 16 = 5 |z| = \sqrt{9 + 16} = 5 ∣ z ∣ = 9 + 16 = 5 .
Both real and imaginary parts are negative, so z z z is in the third quadrant.
θ = arctan ( 4 / 3 ) + π = π + arctan ( 4 / 3 ) \theta = \arctan(4/3) + \pi = \pi + \arctan(4/3) θ = arctan ( 4/3 ) + π = π + arctan ( 4/3 ) .
z = 5 e i ( π + arctan ( 4 / 3 ) ) z = 5\,e^{i(\pi + \arctan(4/3))} z = 5 e i ( π + a r c t a n ( 4/3 )) .
Euler’s formula: e i θ = cos θ + i sin θ e^{i\theta} = \cos\theta + i\sin\theta e i θ = cos θ + i sin θ .
De Moivre’s theorem: ( e i θ ) n = e i n θ (e^{i\theta})^n = e^{in\theta} ( e i θ ) n = e in θ So
( cos θ + i sin θ ) n = cos ( n θ ) + i sin ( n θ ) (\cos\theta + i\sin\theta)^n = \cos(n\theta) + i\sin(n\theta) ( cos θ + i sin θ ) n = cos ( n θ ) + i sin ( n θ )
Proposition 1.5. De Moivre’s theorem holds for all integers n n n Including negative values.
Proof. For n ≥ 0 n \geq 0 n ≥ 0 It follows by induction from the multiplication law e i α e i β = e i ( α + β ) e^{i\alpha}e^{i\beta} = e^{i(\alpha + \beta)} e i α e i β = e i ( α + β ) . For n < 0 n \lt 0 n < 0 Write n = − m n = -m n = − m with m > 0 m \gt 0 m > 0 : ( cos θ + i sin θ ) n = 1 ( cos θ + i sin θ ) m = 1 cos ( m θ ) + i sin ( m θ ) = cos ( − m θ ) + i sin ( − m θ ) = cos ( n θ ) + i sin ( n θ ) (\cos\theta + i\sin\theta)^n = \frac{1}{(\cos\theta + i\sin\theta)^m} = \frac{1}{\cos(m\theta) + i\sin(m\theta)} = \cos(-m\theta) + i\sin(-m\theta) = \cos(n\theta) + i\sin(n\theta) ( cos θ + i sin θ ) n = ( c o s θ + i s i n θ ) m 1 = c o s ( m θ ) + i s i n ( m θ ) 1 = cos ( − m θ ) + i sin ( − m θ ) = cos ( n θ ) + i sin ( n θ ) . ■ \blacksquare ■
Example. Compute ( 1 + i ) 20 (1 + i)^{20} ( 1 + i ) 20 .
1 + i = 2 e i π / 4 1 + i = \sqrt{2}\,e^{i\pi/4} 1 + i = 2 e iπ /4 So ( 1 + i ) 20 = ( 2 ) 20 e 20 π i / 4 = 2 10 e 5 π i = 1024 e π i = − 1024 (1 + i)^{20} = (\sqrt{2})^{20}\, e^{20\pi i/4} = 2^{10}\, e^{5\pi i} = 1024\,e^{\pi i} = -1024 ( 1 + i ) 20 = ( 2 ) 20 e 20 π i /4 = 2 10 e 5 π i = 1024 e π i = − 1024 .
Solution Problem. Express cos ( 5 θ ) \cos(5\theta) cos ( 5 θ ) in terms of cos θ \cos\theta cos θ using de Moivre.
By de Moivre: cos ( 5 θ ) + i sin ( 5 θ ) = ( cos θ + i sin θ ) 5 \cos(5\theta) + i\sin(5\theta) = (\cos\theta + i\sin\theta)^5 cos ( 5 θ ) + i sin ( 5 θ ) = ( cos θ + i sin θ ) 5 .
Expanding the right side by the binomial theorem and equating real parts:
cos ( 5 θ ) = cos 5 θ − 10 cos 3 θ sin 2 θ + 5 cos θ sin 4 θ \cos(5\theta) = \cos^5\theta - 10\cos^3\theta\sin^2\theta + 5\cos\theta\sin^4\theta cos ( 5 θ ) = cos 5 θ − 10 cos 3 θ sin 2 θ + 5 cos θ sin 4 θ .
Using sin 2 θ = 1 − cos 2 θ \sin^2\theta = 1 - \cos^2\theta sin 2 θ = 1 − cos 2 θ :
cos ( 5 θ ) = cos 5 θ − 10 cos 3 θ ( 1 − cos 2 θ ) + 5 cos θ ( 1 − cos 2 θ ) 2 \cos(5\theta) = \cos^5\theta - 10\cos^3\theta(1 - \cos^2\theta) + 5\cos\theta(1 - \cos^2\theta)^2 cos ( 5 θ ) = cos 5 θ − 10 cos 3 θ ( 1 − cos 2 θ ) + 5 cos θ ( 1 − cos 2 θ ) 2 = cos 5 θ − 10 cos 3 θ + 10 cos 5 θ + 5 cos θ − 10 cos 3 θ + 5 cos 5 θ = \cos^5\theta - 10\cos^3\theta + 10\cos^5\theta + 5\cos\theta - 10\cos^3\theta + 5\cos^5\theta = cos 5 θ − 10 cos 3 θ + 10 cos 5 θ + 5 cos θ − 10 cos 3 θ + 5 cos 5 θ = 16 cos 5 θ − 20 cos 3 θ + 5 cos θ = 16\cos^5\theta - 20\cos^3\theta + 5\cos\theta = 16 cos 5 θ − 20 cos 3 θ + 5 cos θ .
Problem. Show that ∑ k = 0 n − 1 cos ( k θ ) = sin ( n θ / 2 ) sin ( θ / 2 ) cos ( ( n − 1 ) θ 2 ) \sum_{k=0}^{n-1} \cos(k\theta) = \frac{\sin(n\theta/2)}{\sin(\theta/2)}\cos\!\left(\frac{(n-1)\theta}{2}\right) ∑ k = 0 n − 1 cos ( k θ ) = s i n ( θ /2 ) s i n ( n θ /2 ) cos ( 2 ( n − 1 ) θ ) For θ ∉ 2 π Z \theta \notin 2\pi\mathbb{Z} θ ∈ / 2 π Z .
Consider S = ∑ k = 0 n − 1 e i k θ = 1 − e i n θ 1 − e i θ S = \sum_{k=0}^{n-1} e^{ik\theta} = \frac{1 - e^{in\theta}}{1 - e^{i\theta}} S = ∑ k = 0 n − 1 e ik θ = 1 − e i θ 1 − e in θ (geometric series with r = e i θ ≠ 1 r = e^{i\theta} \neq 1 r = e i θ = 1 ).
S = e i n θ / 2 ( e − i n θ / 2 − e i n θ / 2 ) e i θ / 2 ( e − i θ / 2 − e i θ / 2 ) = e i ( n − 1 ) θ / 2 ⋅ sin ( n θ / 2 ) sin ( θ / 2 ) S = \frac{e^{in\theta/2}(e^{-in\theta/2} - e^{in\theta/2})}{e^{i\theta/2}(e^{-i\theta/2} - e^{i\theta/2})} = e^{i(n-1)\theta/2} \cdot \frac{\sin(n\theta/2)}{\sin(\theta/2)} S = e i θ /2 ( e − i θ /2 − e i θ /2 ) e in θ /2 ( e − in θ /2 − e in θ /2 ) = e i ( n − 1 ) θ /2 ⋅ s i n ( θ /2 ) s i n ( n θ /2 ) .
Taking real parts gives the result.
Definition. An n n n -th root of w ∈ C w \in \mathbb{C} w ∈ C is a complex number z z z such that z n = w z^n = w z n = w .
Proposition 1.6. Every non-zero w ∈ C w \in \mathbb{C} w ∈ C has exactly n n n distinct n n n -th roots. If w = ρ e i ϕ w = \rho\, e^{i\phi} w = ρ e i ϕ Then
z k = ρ 1 / n e i ( ϕ + 2 π k ) / n , k = 0 , 1 , … , n − 1 z_k = \rho^{1/n}\, e^{i(\phi + 2\pi k)/n}, \quad k = 0, 1, \ldots, n - 1 z k = ρ 1/ n e i ( ϕ + 2 π k ) / n , k = 0 , 1 , … , n − 1
Where ρ 1 / n > 0 \rho^{1/n} \gt 0 ρ 1/ n > 0 is the positive real n n n -th root of ρ \rho ρ .
Proof. If z n = w z^n = w z n = w Write z = r e i θ z = r\,e^{i\theta} z = r e i θ . Then r n e i n θ = ρ e i ϕ r^n e^{in\theta} = \rho\, e^{i\phi} r n e in θ = ρ e i ϕ So r = ρ 1 / n r = \rho^{1/n} r = ρ 1/ n and n θ = ϕ + 2 π k n\theta = \phi + 2\pi k n θ = ϕ + 2 π k . For k = 0 , 1 , … , n − 1 k = 0, 1, \ldots, n-1 k = 0 , 1 , … , n − 1 these give distinct Values of θ \theta θ ; for k ≥ n k \geq n k ≥ n they repeat. ■ \blacksquare ■
Remark. The n n n -th roots of w w w lie equally spaced on a circle of radius ρ 1 / n \rho^{1/n} ρ 1/ n Forming a Regular n n n -gon.
The n n n -th roots of unity are the solutions of z n = 1 z^n = 1 z n = 1 :
z k = e 2 π i k / n , k = 0 , 1 , … , n − 1 z_k = e^{2\pi i k / n}, \quad k = 0, 1, \ldots, n - 1 z k = e 2 π ik / n , k = 0 , 1 , … , n − 1
They form a regular n n n -gon on the unit circle in the complex plane.
Proposition 1.7. If ω = e 2 π i / n \omega = e^{2\pi i/n} ω = e 2 π i / n is a primitive n n n -th root of unity, then ∑ k = 0 n − 1 ω k = 0 \sum_{k=0}^{n-1} \omega^k = 0 ∑ k = 0 n − 1 ω k = 0 and ∑ k = 0 n − 1 ω j k = 0 \sum_{k=0}^{n-1} \omega^{jk} = 0 ∑ k = 0 n − 1 ω j k = 0 for any j j j not divisible by n n n .
Proof. The sum ∑ k = 0 n − 1 ω k = 1 − ω n 1 − ω = 1 − 1 1 − ω = 0 \sum_{k=0}^{n-1} \omega^k = \frac{1 - \omega^n}{1 - \omega} = \frac{1 - 1}{1 - \omega} = 0 ∑ k = 0 n − 1 ω k = 1 − ω 1 − ω n = 1 − ω 1 − 1 = 0 Provided ω ≠ 1 \omega \neq 1 ω = 1 . For j j j not divisible by n n n , ω j \omega^j ω j is a non-trivial root of unity, So the same argument applies. ■ \blacksquare ■
Solution Problem. Find all cube roots of − 8 -8 − 8 .
− 8 = 8 e i π -8 = 8\,e^{i\pi} − 8 = 8 e iπ . The cube roots are: z k = 8 1 / 3 e i ( π + 2 π k ) / 3 = 2 e i ( π + 2 π k ) / 3 z_k = 8^{1/3}\, e^{i(\pi + 2\pi k)/3} = 2\, e^{i(\pi + 2\pi k)/3} z k = 8 1/3 e i ( π + 2 π k ) /3 = 2 e i ( π + 2 π k ) /3 for k = 0 , 1 , 2 k = 0, 1, 2 k = 0 , 1 , 2 .
z 0 = 2 e i π / 3 = 2 ( 1 2 + i 3 2 ) = 1 + i 3 z_0 = 2\,e^{i\pi/3} = 2\left(\frac{1}{2} + i\,\frac{\sqrt{3}}{2}\right) = 1 + i\sqrt{3} z 0 = 2 e iπ /3 = 2 ( 2 1 + i 2 3 ) = 1 + i 3 . z 1 = 2 e i π = − 2 z_1 = 2\,e^{i\pi} = -2 z 1 = 2 e iπ = − 2 . z 2 = 2 e i 5 π / 3 = 2 ( 1 2 − i 3 2 ) = 1 − i 3 z_2 = 2\,e^{i5\pi/3} = 2\left(\frac{1}{2} - i\,\frac{\sqrt{3}}{2}\right) = 1 - i\sqrt{3} z 2 = 2 e i 5 π /3 = 2 ( 2 1 − i 2 3 ) = 1 − i 3 .
Problem. Find all fourth roots of z = 16 i z = 16i z = 16 i .
16 i = 16 e i π / 2 16i = 16\,e^{i\pi/2} 16 i = 16 e iπ /2 . The fourth roots are: z k = 16 1 / 4 e i ( π / 2 + 2 π k ) / 4 = 2 e i ( π / 8 + π k / 2 ) z_k = 16^{1/4}\, e^{i(\pi/2 + 2\pi k)/4} = 2\, e^{i(\pi/8 + \pi k/2)} z k = 1 6 1/4 e i ( π /2 + 2 π k ) /4 = 2 e i ( π /8 + π k /2 ) for k = 0 , 1 , 2 , 3 k = 0, 1, 2, 3 k = 0 , 1 , 2 , 3 .
z 0 = 2 e i π / 8 z_0 = 2\,e^{i\pi/8} z 0 = 2 e iπ /8 , z 1 = 2 e i 5 π / 8 z_1 = 2\,e^{i5\pi/8} z 1 = 2 e i 5 π /8 , z 2 = 2 e i 9 π / 8 z_2 = 2\,e^{i9\pi/8} z 2 = 2 e i 9 π /8 , z 3 = 2 e i 13 π / 8 z_3 = 2\,e^{i13\pi/8} z 3 = 2 e i 13 π /8 .
Problem. Show that the n n n -th roots of any non-zero w w w are in geometric progression.
The roots are z k = ρ 1 / n e i ( ϕ + 2 π k ) / n = z 0 ⋅ ( e 2 π i / n ) k = z 0 ⋅ ω k z_k = \rho^{1/n}\, e^{i(\phi + 2\pi k)/n} = z_0 \cdot \left(e^{2\pi i/n}\right)^k = z_0 \cdot \omega^k z k = ρ 1/ n e i ( ϕ + 2 π k ) / n = z 0 ⋅ ( e 2 π i / n ) k = z 0 ⋅ ω k Where ω = e 2 π i / n \omega = e^{2\pi i/n} ω = e 2 π i / n is a primitive n n n -th root of unity. This is a geometric sequence With ratio ω \omega ω .
A complex function is a function f : D ⊆ C → C f : D \subseteq \mathbb{C} \to \mathbb{C} f : D ⊆ C → C . We can write f ( z ) = u ( x , y ) + i v ( x , y ) f(z) = u(x, y) + iv(x, y) f ( z ) = u ( x , y ) + i v ( x , y ) where z = x + i y z = x + iy z = x + i y and u , v u, v u , v are real-valued functions.
Example. f ( z ) = z 2 = ( x + i y ) 2 = ( x 2 − y 2 ) + i ( 2 x y ) f(z) = z^2 = (x + iy)^2 = (x^2 - y^2) + i(2xy) f ( z ) = z 2 = ( x + i y ) 2 = ( x 2 − y 2 ) + i ( 2 x y ) . Here u = x 2 − y 2 u = x^2 - y^2 u = x 2 − y 2 and v = 2 x y v = 2xy v = 2 x y .
Example. f ( z ) = z ˉ = x − i y f(z) = \bar{z} = x - iy f ( z ) = z ˉ = x − i y . Here u = x u = x u = x and v = − y v = -y v = − y .
Example. f ( z ) = ∣ z ∣ 2 = x 2 + y 2 f(z) = |z|^2 = x^2 + y^2 f ( z ) = ∣ z ∣ 2 = x 2 + y 2 . Here u = x 2 + y 2 u = x^2 + y^2 u = x 2 + y 2 and v = 0 v = 0 v = 0 .
The limit lim z → z 0 f ( z ) = L \lim_{z \to z_0} f(z) = L lim z → z 0 f ( z ) = L means: for every ε > 0 \varepsilon \gt 0 ε > 0 There exists δ > 0 \delta \gt 0 δ > 0 Such that 0 < ∣ z − z 0 ∣ < δ 0 \lt |z - z_0| \lt \delta 0 < ∣ z − z 0 ∣ < δ implies ∣ f ( z ) − L ∣ < ε |f(z) - L| \lt \varepsilon ∣ f ( z ) − L ∣ < ε .
Unlike the real case, z z z can approach z 0 z_0 z 0 from any direction in C \mathbb{C} C . This makes limits More restrictive.
Proposition 2.1. lim z → z 0 f ( z ) = L \lim_{z \to z_0} f(z) = L lim z → z 0 f ( z ) = L if and only if lim ( x , y ) → ( x 0 , y 0 ) u ( x , y ) = a \lim_{(x,y) \to (x_0, y_0)} u(x, y) = a lim ( x , y ) → ( x 0 , y 0 ) u ( x , y ) = a And lim ( x , y ) → ( x 0 , y 0 ) v ( x , y ) = b \lim_{(x,y) \to (x_0, y_0)} v(x, y) = b lim ( x , y ) → ( x 0 , y 0 ) v ( x , y ) = b where L = a + b i L = a + bi L = a + bi .
Definition. f f f is continuous at z 0 z_0 z 0 if lim z → z 0 f ( z ) = f ( z 0 ) \lim_{z \to z_0} f(z) = f(z_0) lim z → z 0 f ( z ) = f ( z 0 ) .
Solution Problem. Show that lim z → 0 z ˉ z \lim_{z \to 0} \frac{\bar{z}}{z} lim z → 0 z z ˉ does not exist.
Let z = r e i θ z = re^{i\theta} z = r e i θ . Then z ˉ z = e − 2 i θ \frac{\bar{z}}{z} = e^{-2i\theta} z z ˉ = e − 2 i θ . As z → 0 z \to 0 z → 0 along different Rays (θ = 0 , π / 2 , π / 4 \theta = 0, \pi/2, \pi/4 θ = 0 , π /2 , π /4 Etc.), the ratio takes different values (1 , − 1 , − i 1, -1, -i 1 , − 1 , − i Etc.). Since the limit depends on the direction of approach, it does not exist.
Problem. Determine whether f ( z ) = z 2 − 1 z − 1 f(z) = \frac{z^2 - 1}{z - 1} f ( z ) = z − 1 z 2 − 1 is continuous at z = 1 z = 1 z = 1 .
For z ≠ 1 z \neq 1 z = 1 : f ( z ) = z + 1 f(z) = z + 1 f ( z ) = z + 1 . The limit as z → 1 z \to 1 z → 1 is 2 2 2 But f ( 1 ) f(1) f ( 1 ) is undefined (division by zero). If we define f ( 1 ) = 2 f(1) = 2 f ( 1 ) = 2 Then f f f becomes continuous at z = 1 z = 1 z = 1 .
Definition. f f f is differentiable at z 0 z_0 z 0 if
f ′ ( z 0 ) = lim h → 0 f ( z 0 + h ) − f ( z 0 ) h f'(z_0) = \lim_{h \to 0} \frac{f(z_0 + h) - f(z_0)}{h} f ′ ( z 0 ) = lim h → 0 h f ( z 0 + h ) − f ( z 0 )
Exists (and is independent of how h → 0 h \to 0 h → 0 in C \mathbb{C} C ).
Remark. The requirement that the limit be the same for all directions of approach of h h h is what Makes complex differentiability far more restrictive than real differentiability.
Definition. A function f f f is analytic (or holomorphic ) on an open set U ⊆ C U \subseteq \mathbb{C} U ⊆ C if f f f is differentiable at every point of U U U . A function that is analytic On all of C \mathbb{C} C is called entire .
Examples of entire functions: z n z^n z n , e z e^z e z , sin z \sin z sin z , cos z \cos z cos z Polynomials.
Example of a non-analytic function: f ( z ) = z ˉ f(z) = \bar{z} f ( z ) = z ˉ is nowhere differentiable (except at z = 0 z = 0 z = 0 if we define it, but still not analytic there).
Solution Problem. Show that f ( z ) = ∣ z ∣ 2 f(z) = |z|^2 f ( z ) = ∣ z ∣ 2 is differentiable only at z = 0 z = 0 z = 0 .
f ( z ) = x 2 + y 2 f(z) = x^2 + y^2 f ( z ) = x 2 + y 2 So u = x 2 + y 2 u = x^2 + y^2 u = x 2 + y 2 and v = 0 v = 0 v = 0 . u x = 2 x u_x = 2x u x = 2 x , u y = 2 y u_y = 2y u y = 2 y , v x = 0 v_x = 0 v x = 0 , v y = 0 v_y = 0 v y = 0 . The Cauchy-Riemann equations require 2 x = 0 2x = 0 2 x = 0 and 2 y = 0 2y = 0 2 y = 0 So x = y = 0 x = y = 0 x = y = 0 . Thus f f f satisfies CR only at z = 0 z = 0 z = 0 .
At z = 0 z = 0 z = 0 : f ′ ( 0 ) = lim h → 0 ∣ h ∣ 2 h = lim h → 0 h ˉ = 0 f'(0) = \lim_{h \to 0} \frac{|h|^2}{h} = \lim_{h \to 0} \bar{h} = 0 f ′ ( 0 ) = lim h → 0 h ∣ h ∣ 2 = lim h → 0 h ˉ = 0 So f f f is Differentiable at 0 0 0 but not analytic anywhere (no neighbourhood of 0 0 0 is analytic).
Problem. Show that f ( z ) = z z ˉ + z f(z) = z\bar{z} + z f ( z ) = z z ˉ + z is differentiable only at z = 0 z = 0 z = 0 .
f ( z ) = ∣ z ∣ 2 + z = ( x 2 + y 2 + x ) + i y f(z) = |z|^2 + z = (x^2 + y^2 + x) + iy f ( z ) = ∣ z ∣ 2 + z = ( x 2 + y 2 + x ) + i y . u x = 2 x + 1 u_x = 2x + 1 u x = 2 x + 1 , u y = 2 y u_y = 2y u y = 2 y , v x = 0 v_x = 0 v x = 0 , v y = 1 v_y = 1 v y = 1 . CR equations: 2 x + 1 = 1 ⇒ x = 0 2x + 1 = 1 \Rightarrow x = 0 2 x + 1 = 1 ⇒ x = 0 And 2 y = 0 ⇒ y = 0 2y = 0 \Rightarrow y = 0 2 y = 0 ⇒ y = 0 . At ( 0 , 0 ) (0, 0) ( 0 , 0 ) : f ′ ( 0 ) = lim h → 0 h h ˉ + h h = lim h → 0 ( h ˉ + 1 ) = 1 f'(0) = \lim_{h \to 0} \frac{h\bar{h} + h}{h} = \lim_{h \to 0} (\bar{h} + 1) = 1 f ′ ( 0 ) = lim h → 0 h h h ˉ + h = lim h → 0 ( h ˉ + 1 ) = 1 . So f f f is differentiable at z = 0 z = 0 z = 0 only, hence nowhere analytic.
Many important functions in complex analysis are inherently multi-valued. To work with them as Single-valued functions, we must restrict the domain.
Definition. A branch of a multi-valued function f f f is a single-valued analytic function g g g Defined on a domain D D D such that g ( z ) ∈ f ( z ) g(z) \in f(z) g ( z ) ∈ f ( z ) for all z ∈ D z \in D z ∈ D .
The Complex Logarithm. We define log z = ln ∣ z ∣ + i arg ( z ) \log z = \ln|z| + i\arg(z) log z = ln ∣ z ∣ + i arg ( z ) Which is multi-valued because arg ( z ) = A r g ( z ) + 2 π k \arg(z) = \mathrm{Arg}(z) + 2\pi k arg ( z ) = Arg ( z ) + 2 π k for k ∈ Z k \in \mathbb{Z} k ∈ Z . The principal branch is
L o g z = ln ∣ z ∣ + i A r g ( z ) \mathrm{Log}\, z = \ln|z| + i\,\mathrm{Arg}(z) Log z = ln ∣ z ∣ + i Arg ( z )
Defined on C ∖ ( − ∞ , 0 ] \mathbb{C} \setminus (-\infty, 0] C ∖ ( − ∞ , 0 ] . The negative real axis is called the branch cut .
Proposition 2.2. The principal branch L o g z \mathrm{Log}\, z Log z is analytic on C ∖ ( − ∞ , 0 ] \mathbb{C} \setminus (-\infty, 0] C ∖ ( − ∞ , 0 ] and d d z L o g z = 1 z \frac{d}{dz}\,\mathrm{Log}\, z = \frac{1}{z} d z d Log z = z 1 .
Complex Powers. For z , α ∈ C z, \alpha \in \mathbb{C} z , α ∈ C with z ≠ 0 z \neq 0 z = 0 :
z α = e α log z z^\alpha = e^{\alpha \log z} z α = e α l o g z
This is multi-valued . When α \alpha α is rational with reduced form p / q p/q p / q There are exactly q q q distinct values.
Solution Problem. Find all values of ( − 1 ) i (-1)^i ( − 1 ) i .
( − 1 ) i = e i log ( − 1 ) = e i ( i π + 2 π i k ) = e − π − 2 π k (-1)^i = e^{i \log(-1)} = e^{i(i\pi + 2\pi i k)} = e^{-\pi - 2\pi k} ( − 1 ) i = e i l o g ( − 1 ) = e i ( iπ + 2 π ik ) = e − π − 2 π k for k ∈ Z k \in \mathbb{Z} k ∈ Z .
These are all positive real numbers: … , e 3 π , e π , e − π , e − 3 π , … \ldots, e^{3\pi}, e^{\pi}, e^{-\pi}, e^{-3\pi}, \ldots … , e 3 π , e π , e − π , e − 3 π , … . The principal value (using the principal branch) is e − π e^{-\pi} e − π .
Problem. Find all values of i 1 / 2 i^{1/2} i 1/2 .
i 1 / 2 = e ( 1 / 2 ) log i = e ( 1 / 2 ) ( i π / 2 + 2 π i k ) = e i π / 4 + i π k i^{1/2} = e^{(1/2)\log i} = e^{(1/2)(i\pi/2 + 2\pi i k)} = e^{i\pi/4 + i\pi k} i 1/2 = e ( 1/2 ) l o g i = e ( 1/2 ) ( iπ /2 + 2 π ik ) = e iπ /4 + iπ k .
For k = 0 k = 0 k = 0 : e i π / 4 = 2 2 ( 1 + i ) e^{i\pi/4} = \frac{\sqrt{2}}{2}(1 + i) e iπ /4 = 2 2 ( 1 + i ) . For k = 1 k = 1 k = 1 : e i 5 π / 4 = − 2 2 ( 1 + i ) e^{i5\pi/4} = -\frac{\sqrt{2}}{2}(1 + i) e i 5 π /4 = − 2 2 ( 1 + i ) . These are the two square roots of i i i .
Problem. Find the domain of analyticity of f ( z ) = L o g ( z 2 + 1 ) f(z) = \mathrm{Log}(z^2 + 1) f ( z ) = Log ( z 2 + 1 ) .
L o g w \mathrm{Log}\, w Log w is analytic on C ∖ ( − ∞ , 0 ] \mathbb{C} \setminus (-\infty, 0] C ∖ ( − ∞ , 0 ] So we need z 2 + 1 ∉ ( − ∞ , 0 ] z^2 + 1 \notin (-\infty, 0] z 2 + 1 ∈ / ( − ∞ , 0 ] .
z 2 + 1 ≤ 0 z^2 + 1 \leq 0 z 2 + 1 ≤ 0 when z 2 ≤ − 1 z^2 \leq -1 z 2 ≤ − 1 I.e., z ∈ [ − i , 0 ] ∪ [ 0 , i ] z \in [-i, 0] \cup [0, i] z ∈ [ − i , 0 ] ∪ [ 0 , i ] (the imaginary axis Segment from − i -i − i to i i i ). Also z 2 + 1 = 0 z^2 + 1 = 0 z 2 + 1 = 0 at z = ± i z = \pm i z = ± i .
Domain: C ∖ { z : z = i y , y ∈ [ − 1 , 1 ] } \mathbb{C} \setminus \{z : z = iy,\, y \in [-1, 1]\} C ∖ { z : z = i y , y ∈ [ − 1 , 1 ]} .
Theorem 3.1 (Cauchy-Riemann Equations). If f ( z ) = u ( x , y ) + i v ( x , y ) f(z) = u(x, y) + iv(x, y) f ( z ) = u ( x , y ) + i v ( x , y ) is differentiable at z = x + i y z = x + iy z = x + i y Then
∂ u ∂ x = ∂ v ∂ y , ∂ u ∂ y = − ∂ v ∂ x \frac{\partial u}{\partial x} = \frac{\partial v}{\partial y}, \quad \frac{\partial u}{\partial y} = -\frac{\partial v}{\partial x} ∂ x ∂ u = ∂ y ∂ v , ∂ y ∂ u = − ∂ x ∂ v
Proof. Compute the limit along the real axis (h ∈ R h \in \mathbb{R} h ∈ R , h → 0 h \to 0 h → 0 ):
f ′ ( z ) = lim h → 0 u ( x + h , y ) − u ( x , y ) h + i lim h → 0 v ( x + h , y ) − v ( x , y ) h = ∂ u ∂ x + i ∂ v ∂ x f'(z) = \lim_{h \to 0} \frac{u(x+h, y) - u(x, y)}{h} + i\lim_{h \to 0} \frac{v(x+h, y) - v(x, y)}{h} = \frac{\partial u}{\partial x} + i\frac{\partial v}{\partial x} f ′ ( z ) = lim h → 0 h u ( x + h , y ) − u ( x , y ) + i lim h → 0 h v ( x + h , y ) − v ( x , y ) = ∂ x ∂ u + i ∂ x ∂ v
Compute along the imaginary axis (h = i k h = ik h = ik , k ∈ R k \in \mathbb{R} k ∈ R , k → 0 k \to 0 k → 0 ):
f ′ ( z ) = lim k → 0 u ( x , y + k ) − u ( x , y ) i k + i lim k → 0 v ( x , y + k ) − v ( x , y ) i k = − i ∂ u ∂ y + ∂ v ∂ y f'(z) = \lim_{k \to 0} \frac{u(x, y+k) - u(x, y)}{ik} + i\lim_{k \to 0} \frac{v(x, y+k) - v(x, y)}{ik} = -i\frac{\partial u}{\partial y} + \frac{\partial v}{\partial y} f ′ ( z ) = lim k → 0 ik u ( x , y + k ) − u ( x , y ) + i lim k → 0 ik v ( x , y + k ) − v ( x , y ) = − i ∂ y ∂ u + ∂ y ∂ v
Equating real and imaginary parts: ∂ u ∂ x = ∂ v ∂ y \frac{\partial u}{\partial x} = \frac{\partial v}{\partial y} ∂ x ∂ u = ∂ y ∂ v And ∂ v ∂ x = − ∂ u ∂ y \frac{\partial v}{\partial x} = -\frac{\partial u}{\partial y} ∂ x ∂ v = − ∂ y ∂ u . ■ \blacksquare ■
Theorem 3.2. If u u u and v v v have continuous first partial derivatives on an open set U U U and Satisfy the Cauchy-Riemann equations on U U U Then f = u + i v f = u + iv f = u + i v is analytic on U U U .
Proof. Since u x , u y , v x , v y u_x, u_y, v_x, v_y u x , u y , v x , v y are continuous on U U U , u u u and v v v are (real) differentiable. Let Δ z = Δ x + i Δ y \Delta z = \Delta x + i\Delta y Δ z = Δ x + i Δ y . By real differentiability:
u ( x + Δ x , y + Δ y ) − u ( x , y ) = u x Δ x + u y Δ y + ε 1 u(x + \Delta x, y + \Delta y) - u(x, y) = u_x\,\Delta x + u_y\,\Delta y + \varepsilon_1 u ( x + Δ x , y + Δ y ) − u ( x , y ) = u x Δ x + u y Δ y + ε 1 v ( x + Δ x , y + Δ y ) − v ( x , y ) = v x Δ x + v y Δ y + ε 2 v(x + \Delta x, y + \Delta y) - v(x, y) = v_x\,\Delta x + v_y\,\Delta y + \varepsilon_2 v ( x + Δ x , y + Δ y ) − v ( x , y ) = v x Δ x + v y Δ y + ε 2
Where ε 1 , ε 2 = o ( ∣ Δ z ∣ ) \varepsilon_1, \varepsilon_2 = o(|\Delta z|) ε 1 , ε 2 = o ( ∣Δ z ∣ ) . Therefore
f ( z + Δ z ) − f ( z ) Δ z = ( u x + i v x ) Δ x + ( u y + i v y ) Δ y + ε 1 + i ε 2 Δ x + i Δ y \frac{f(z + \Delta z) - f(z)}{\Delta z} = \frac{(u_x + iv_x)\Delta x + (u_y + iv_y)\Delta y + \varepsilon_1 + i\varepsilon_2}{\Delta x + i\Delta y} Δ z f ( z + Δ z ) − f ( z ) = Δ x + i Δ y ( u x + i v x ) Δ x + ( u y + i v y ) Δ y + ε 1 + i ε 2
By CR: u y + i v y = − v x + i u x = i ( u x + i v x ) u_y + iv_y = -v_x + iu_x = i(u_x + iv_x) u y + i v y = − v x + i u x = i ( u x + i v x ) . Substituting:
= ( u x + i v x ) Δ x + i Δ y Δ x + i Δ y + o ( ∣ Δ z ∣ ) Δ z → u x + i v x = (u_x + iv_x)\frac{\Delta x + i\Delta y}{\Delta x + i\Delta y} + \frac{o(|\Delta z|)}{\Delta z} \to u_x + iv_x = ( u x + i v x ) Δ x + i Δ y Δ x + i Δ y + Δ z o ( ∣Δ z ∣ ) → u x + i v x
As Δ z → 0 \Delta z \to 0 Δ z → 0 . ■ \blacksquare ■
When the Cauchy-Riemann equations hold:
f ′ ( z ) = ∂ u ∂ x + i ∂ v ∂ x = ∂ v ∂ y − i ∂ u ∂ y f'(z) = \frac{\partial u}{\partial x} + i\frac{\partial v}{\partial x} = \frac{\partial v}{\partial y} - i\frac{\partial u}{\partial y} f ′ ( z ) = ∂ x ∂ u + i ∂ x ∂ v = ∂ y ∂ v − i ∂ y ∂ u
Definition. A real-valued function ϕ ( x , y ) \phi(x, y) ϕ ( x , y ) is harmonic if ϕ x x + ϕ y y = 0 \phi_{xx} + \phi_{yy} = 0 ϕ xx + ϕ y y = 0 (Laplace’s equation).
Proposition 3.3. If f = u + i v f = u + iv f = u + i v is analytic, then u u u and v v v are harmonic.
Proof. From the Cauchy-Riemann equations: u x = v y u_x = v_y u x = v y and u y = − v x u_y = -v_x u y = − v x . Differentiating: u x x = v y x u_{xx} = v_{yx} u xx = v y x and u y y = − v x y u_{yy} = -v_{xy} u y y = − v x y . By equality of mixed partials, u x x + u y y = v y x − v x y = 0 u_{xx} + u_{yy} = v_{yx} - v_{xy} = 0 u xx + u y y = v y x − v x y = 0 . Similarly for v v v . ■ \blacksquare ■
Definition. If u u u and v v v are harmonic on U U U and satisfy the Cauchy-Riemann equations, then v v v is the harmonic conjugate of u u u .
Proposition 3.4. If U U U is a connected domain and u u u is harmonic on U U U Then u u u has A harmonic conjugate on U U U Unique up to an additive constant.
Proof. Define v ( x , y ) = ∫ ( x 0 , y 0 ) ( x , y ) ( − u y d x + u x d y ) v(x, y) = \int_{(x_0, y_0)}^{(x, y)} (-u_y\, dx + u_x\, dy) v ( x , y ) = ∫ ( x 0 , y 0 ) ( x , y ) ( − u y d x + u x d y ) . The integrand is closed (since ( − u y ) y = − u y y = u x x = ( u x ) x (-u_y)_y = -u_{yy} = u_{xx} = (u_x)_x ( − u y ) y = − u y y = u xx = ( u x ) x ) and since U U U is Connected, v v v is well-defined (path-independent) by Green’s theorem. Then v x = − u y v_x = -u_y v x = − u y and v y = u x v_y = u_x v y = u x Which are the CR equations. ■ \blacksquare ■
Solution Problem. Find the harmonic conjugate of u ( x , y ) = x 3 − 3 x y 2 u(x, y) = x^3 - 3xy^2 u ( x , y ) = x 3 − 3 x y 2 .
Verify u u u is harmonic: u x x = 6 x u_{xx} = 6x u xx = 6 x , u y y = − 6 x u_{yy} = -6x u y y = − 6 x So u x x + u y y = 0 u_{xx} + u_{yy} = 0 u xx + u y y = 0 . ✓ \checkmark ✓
By CR: v y = u x = 3 x 2 − 3 y 2 v_y = u_x = 3x^2 - 3y^2 v y = u x = 3 x 2 − 3 y 2 So v = 3 x 2 y − y 3 + g ( x ) v = 3x^2 y - y^3 + g(x) v = 3 x 2 y − y 3 + g ( x ) . Also v x = − u y = 6 x y v_x = -u_y = 6xy v x = − u y = 6 x y So 6 x y = 6 x y + g ′ ( x ) 6xy = 6xy + g'(x) 6 x y = 6 x y + g ′ ( x ) Giving g ′ ( x ) = 0 g'(x) = 0 g ′ ( x ) = 0 So g ( x ) = C g(x) = C g ( x ) = C .
Harmonic conjugate: v ( x , y ) = 3 x 2 y − y 3 + C v(x, y) = 3x^2 y - y^3 + C v ( x , y ) = 3 x 2 y − y 3 + C .
Note: f ( z ) = u + i v = x 3 − 3 x y 2 + i ( 3 x 2 y − y 3 ) = ( x + i y ) 3 = z 3 f(z) = u + iv = x^3 - 3xy^2 + i(3x^2 y - y^3) = (x + iy)^3 = z^3 f ( z ) = u + i v = x 3 − 3 x y 2 + i ( 3 x 2 y − y 3 ) = ( x + i y ) 3 = z 3 .
Problem. Show that u ( x , y ) = ln ( x 2 + y 2 ) u(x, y) = \ln(x^2 + y^2) u ( x , y ) = ln ( x 2 + y 2 ) is harmonic on R 2 ∖ { 0 } \mathbb{R}^2 \setminus \{0\} R 2 ∖ { 0 } but Has no harmonic conjugate on R 2 ∖ { 0 } \mathbb{R}^2 \setminus \{0\} R 2 ∖ { 0 } .
u x = 2 x x 2 + y 2 u_x = \frac{2x}{x^2 + y^2} u x = x 2 + y 2 2 x , u x x = 2 ( y 2 − x 2 ) ( x 2 + y 2 ) 2 u_{xx} = \frac{2(y^2 - x^2)}{(x^2 + y^2)^2} u xx = ( x 2 + y 2 ) 2 2 ( y 2 − x 2 ) . u y = 2 y x 2 + y 2 u_y = \frac{2y}{x^2 + y^2} u y = x 2 + y 2 2 y , u y y = 2 ( x 2 − y 2 ) ( x 2 + y 2 ) 2 u_{yy} = \frac{2(x^2 - y^2)}{(x^2 + y^2)^2} u y y = ( x 2 + y 2 ) 2 2 ( x 2 − y 2 ) . Δ u = 0 \Delta u = 0 Δ u = 0 . ✓ \checkmark ✓
However, ∮ ∣ z ∣ = 1 ( − u y d x + u x d y ) = ∮ ∣ z ∣ = 1 − y d x + x d y x 2 + y 2 = ∫ 0 2 π 1 d θ = 2 π ≠ 0 \oint_{|z|=1} (-u_y\, dx + u_x\, dy) = \oint_{|z|=1} \frac{-y\, dx + x\, dy}{x^2 + y^2} = \int_0^{2\pi} 1\, d\theta = 2\pi \neq 0 ∮ ∣ z ∣ = 1 ( − u y d x + u x d y ) = ∮ ∣ z ∣ = 1 x 2 + y 2 − y d x + x d y = ∫ 0 2 π 1 d θ = 2 π = 0 .
Since R 2 ∖ { 0 } \mathbb{R}^2 \setminus \{0\} R 2 ∖ { 0 } is not connected and this integral is non-zero, no Harmonic conjugate exists on this domain.
Solution Problem. Verify that f ( z ) = e z f(z) = e^z f ( z ) = e z satisfies the Cauchy-Riemann equations and find f ′ ( z ) f'(z) f ′ ( z ) .
Solution. e z = e x + i y = e x ( cos y + i sin y ) e^z = e^{x+iy} = e^x(\cos y + i\sin y) e z = e x + i y = e x ( cos y + i sin y ) . So u = e x cos y u = e^x \cos y u = e x cos y and v = e x sin y v = e^x \sin y v = e x sin y .
u x = e x cos y u_x = e^x \cos y u x = e x cos y , u y = − e x sin y u_y = -e^x \sin y u y = − e x sin y , v x = e x sin y v_x = e^x \sin y v x = e x sin y , v y = e x cos y v_y = e^x \cos y v y = e x cos y .
Cauchy-Riemann: u x = e x cos y = v y u_x = e^x \cos y = v_y u x = e x cos y = v y and u y = − e x sin y = − v x u_y = -e^x \sin y = -v_x u y = − e x sin y = − v x . Both satisfied.
f ′ ( z ) = u x + i v x = e x cos y + i e x sin y = e z f'(z) = u_x + iv_x = e^x \cos y + ie^x \sin y = e^z f ′ ( z ) = u x + i v x = e x cos y + i e x sin y = e z . ■ \blacksquare ■
Problem. Verify CR for f ( z ) = sin z f(z) = \sin z f ( z ) = sin z and find f ′ ( z ) f'(z) f ′ ( z ) .
sin z = sin ( x + i y ) = sin x cosh y + i cos x sinh y \sin z = \sin(x + iy) = \sin x \cosh y + i\cos x \sinh y sin z = sin ( x + i y ) = sin x cosh y + i cos x sinh y .
u = sin x cosh y u = \sin x \cosh y u = sin x cosh y , v = cos x sinh y v = \cos x \sinh y v = cos x sinh y .
u x = cos x cosh y u_x = \cos x \cosh y u x = cos x cosh y , u y = sin x sinh y u_y = \sin x \sinh y u y = sin x sinh y . v x = − sin x sinh y v_x = -\sin x \sinh y v x = − sin x sinh y , v y = cos x cosh y v_y = \cos x \cosh y v y = cos x cosh y .
CR: u x = cos x cosh y = v y u_x = \cos x \cosh y = v_y u x = cos x cosh y = v y ✓ \checkmark ✓ and u y = sin x sinh y = − v x u_y = \sin x \sinh y = -v_x u y = sin x sinh y = − v x ✓ \checkmark ✓ .
f ′ ( z ) = u x + i v x = cos x cosh y − i sin x sinh y = cos z f'(z) = u_x + iv_x = \cos x \cosh y - i\sin x \sinh y = \cos z f ′ ( z ) = u x + i v x = cos x cosh y − i sin x sinh y = cos z . ■ \blacksquare ■
Problem. Show f ( z ) = 1 z f(z) = \frac{1}{z} f ( z ) = z 1 satisfies CR on C ∖ { 0 } \mathbb{C} \setminus \{0\} C ∖ { 0 } .
1 z = z ˉ ∣ z ∣ 2 = x − i y x 2 + y 2 \frac{1}{z} = \frac{\bar{z}}{|z|^2} = \frac{x - iy}{x^2 + y^2} z 1 = ∣ z ∣ 2 z ˉ = x 2 + y 2 x − i y .
u = x x 2 + y 2 u = \frac{x}{x^2 + y^2} u = x 2 + y 2 x , v = − y x 2 + y 2 v = \frac{-y}{x^2 + y^2} v = x 2 + y 2 − y .
u x = y 2 − x 2 ( x 2 + y 2 ) 2 u_x = \frac{y^2 - x^2}{(x^2 + y^2)^2} u x = ( x 2 + y 2 ) 2 y 2 − x 2 , v y = y 2 − x 2 ( x 2 + y 2 ) 2 v_y = \frac{y^2 - x^2}{(x^2 + y^2)^2} v y = ( x 2 + y 2 ) 2 y 2 − x 2 . So u x = v y u_x = v_y u x = v y . ✓ \checkmark ✓
u y = − 2 x y ( x 2 + y 2 ) 2 u_y = \frac{-2xy}{(x^2 + y^2)^2} u y = ( x 2 + y 2 ) 2 − 2 x y , v x = 2 x y ( x 2 + y 2 ) 2 v_x = \frac{2xy}{(x^2 + y^2)^2} v x = ( x 2 + y 2 ) 2 2 x y . So u y = − v x u_y = -v_x u y = − v x . ✓ \checkmark ✓
f ′ ( z ) = u x + i v x = − ( x 2 − y 2 + 2 i x y ) ( x 2 + y 2 ) 2 = − 1 z 2 f'(z) = u_x + iv_x = \frac{-(x^2 - y^2 + 2ixy)}{(x^2 + y^2)^2} = \frac{-1}{z^2} f ′ ( z ) = u x + i v x = ( x 2 + y 2 ) 2 − ( x 2 − y 2 + 2 i x y ) = z 2 − 1 . ■ \blacksquare ■
A contour (or piecewise smooth path) in C \mathbb{C} C is a continuous function γ : [ a , b ] → C \gamma : [a, b] \to \mathbb{C} γ : [ a , b ] → C that is differentiable except at finitely many points, with a Continuous derivative everywhere it exists.
A simple closed contour is a contour with γ ( a ) = γ ( b ) \gamma(a) = \gamma(b) γ ( a ) = γ ( b ) and no other Self-intersections.
Definition. For a contour γ \gamma γ and a continuous function f f f on γ \gamma γ :
∫ γ f ( z ) d z = ∫ a b f ( γ ( t ) ) γ ′ ( t ) d t \int_{\gamma} f(z)\, dz = \int_a^b f(\gamma(t))\gamma'(t)\, dt ∫ γ f ( z ) d z = ∫ a b f ( γ ( t )) γ ′ ( t ) d t
Proposition 4.1. The complex integral is linear:
∫ γ ( a f + b g ) d z = a ∫ γ f d z + b ∫ γ g d z \int_\gamma (af + bg)\, dz = a\int_\gamma f\, dz + b\int_\gamma g\, dz ∫ γ ( a f + b g ) d z = a ∫ γ f d z + b ∫ γ g d z
Proposition 4.2. Reversing orientation changes the sign:
∫ − γ f d z = − ∫ γ f d z \int_{-\gamma} f\, dz = -\int_\gamma f\, dz ∫ − γ f d z = − ∫ γ f d z
Proposition 4.3. Additivity over contours:
∫ γ 1 + γ 2 f d z = ∫ γ 1 f d z + ∫ γ 2 f d z \int_{\gamma_1 + \gamma_2} f\, dz = \int_{\gamma_1} f\, dz + \int_{\gamma_2} f\, dz ∫ γ 1 + γ 2 f d z = ∫ γ 1 f d z + ∫ γ 2 f d z
Proposition 4.4 (ML Inequality). If ∣ f ( z ) ∣ ≤ M |f(z)| \leq M ∣ f ( z ) ∣ ≤ M for all z z z on a contour γ \gamma γ of length L L L Then
∣ ∫ γ f ( z ) d z ∣ ≤ M L \left|\int_\gamma f(z)\, dz\right| \leq ML ∫ γ f ( z ) d z ≤ M L
Proof. ∣ ∫ a b f ( γ ( t ) ) γ ′ ( t ) d t ∣ ≤ ∫ a b ∣ f ( γ ( t ) ) ∣ ∣ γ ′ ( t ) ∣ d t ≤ M ∫ a b ∣ γ ′ ( t ) ∣ d t = M L \left|\int_a^b f(\gamma(t))\gamma'(t)\, dt\right| \leq \int_a^b |f(\gamma(t))||\gamma'(t)|\, dt \leq M \int_a^b |\gamma'(t)|\, dt = ML ∫ a b f ( γ ( t )) γ ′ ( t ) d t ≤ ∫ a b ∣ f ( γ ( t )) ∣∣ γ ′ ( t ) ∣ d t ≤ M ∫ a b ∣ γ ′ ( t ) ∣ d t = M L . ■ \blacksquare ■
Solution Problem. Evaluate ∫ γ z 2 d z \int_\gamma z^2\, dz ∫ γ z 2 d z where γ \gamma γ is the line segment from 0 0 0 to 1 + i 1 + i 1 + i .
Solution. Parameterize γ ( t ) = t ( 1 + i ) \gamma(t) = t(1 + i) γ ( t ) = t ( 1 + i ) for 0 ≤ t ≤ 1 0 \leq t \leq 1 0 ≤ t ≤ 1 . Then γ ′ ( t ) = 1 + i \gamma'(t) = 1 + i γ ′ ( t ) = 1 + i .
∫ γ z 2 d z = ∫ 0 1 ( t ( 1 + i ) ) 2 ( 1 + i ) d t = ( 1 + i ) 3 ∫ 0 1 t 2 d t = ( 1 + i ) 3 ⋅ 1 3 \int_\gamma z^2\, dz = \int_0^1 (t(1+i))^2 (1+i)\, dt = (1+i)^3 \int_0^1 t^2\, dt = (1+i)^3 \cdot \frac{1}{3} ∫ γ z 2 d z = ∫ 0 1 ( t ( 1 + i ) ) 2 ( 1 + i ) d t = ( 1 + i ) 3 ∫ 0 1 t 2 d t = ( 1 + i ) 3 ⋅ 3 1
( 1 + i ) 3 = ( 1 + i ) ( 1 + i ) 2 = ( 1 + i ) ( 2 i ) = 2 i + 2 i 2 = 2 i − 2 = − 2 + 2 i (1+i)^3 = (1+i)(1+i)^2 = (1+i)(2i) = 2i + 2i^2 = 2i - 2 = -2 + 2i ( 1 + i ) 3 = ( 1 + i ) ( 1 + i ) 2 = ( 1 + i ) ( 2 i ) = 2 i + 2 i 2 = 2 i − 2 = − 2 + 2 i .
∫ γ z 2 d z = − 2 + 2 i 3 \int_\gamma z^2\, dz = \frac{-2 + 2i}{3} ∫ γ z 2 d z = 3 − 2 + 2 i . ■ \blacksquare ■
Problem. Evaluate ∫ γ z ˉ d z \int_\gamma \bar{z}\, dz ∫ γ z ˉ d z where γ \gamma γ is the unit circle traversed once Counterclockwise.
γ ( t ) = e i t \gamma(t) = e^{it} γ ( t ) = e i t , 0 ≤ t ≤ 2 π 0 \leq t \leq 2\pi 0 ≤ t ≤ 2 π , γ ′ ( t ) = i e i t \gamma'(t) = ie^{it} γ ′ ( t ) = i e i t . z ˉ = e − i t \bar{z} = e^{-it} z ˉ = e − i t on γ \gamma γ .
∫ γ z ˉ d z = ∫ 0 2 π e − i t ⋅ i e i t d t = ∫ 0 2 π i d t = 2 π i \int_\gamma \bar{z}\, dz = \int_0^{2\pi} e^{-it} \cdot ie^{it}\, dt = \int_0^{2\pi} i\, dt = 2\pi i ∫ γ z ˉ d z = ∫ 0 2 π e − i t ⋅ i e i t d t = ∫ 0 2 π i d t = 2 π i .
Note: Since z ˉ \bar{z} z ˉ is not analytic, this result is non-zero, as expected.
Problem. Evaluate ∫ γ d z z \int_\gamma \frac{dz}{z} ∫ γ z d z where γ \gamma γ is the upper semicircle z = e i θ z = e^{i\theta} z = e i θ , 0 ≤ θ ≤ π 0 \leq \theta \leq \pi 0 ≤ θ ≤ π .
∫ 0 π i e i θ e i θ d θ = ∫ 0 π i d θ = i π \int_0^\pi \frac{ie^{i\theta}}{e^{i\theta}}\, d\theta = \int_0^\pi i\, d\theta = i\pi ∫ 0 π e i θ i e i θ d θ = ∫ 0 π i d θ = iπ .
Problem. Evaluate ∫ γ z d z \int_\gamma z\, dz ∫ γ z d z where γ \gamma γ consists of the line segment from 0 0 0 to 1 1 1 followed by the line segment from 1 1 1 to 1 + i 1 + i 1 + i .
γ 1 ( t ) = t \gamma_1(t) = t γ 1 ( t ) = t , 0 ≤ t ≤ 1 0 \leq t \leq 1 0 ≤ t ≤ 1 : ∫ 0 1 t ⋅ 1 d t = 1 2 \int_0^1 t \cdot 1\, dt = \frac{1}{2} ∫ 0 1 t ⋅ 1 d t = 2 1 .
γ 2 ( t ) = 1 + i t \gamma_2(t) = 1 + it γ 2 ( t ) = 1 + i t , 0 ≤ t ≤ 1 0 \leq t \leq 1 0 ≤ t ≤ 1 : ∫ 0 1 ( 1 + i t ) ⋅ i d t = ∫ 0 1 ( i − t ) d t = i − 1 2 \int_0^1 (1 + it) \cdot i\, dt = \int_0^1 (i - t)\, dt = i - \frac{1}{2} ∫ 0 1 ( 1 + i t ) ⋅ i d t = ∫ 0 1 ( i − t ) d t = i − 2 1 .
Total: 1 2 + i − 1 2 = i \frac{1}{2} + i - \frac{1}{2} = i 2 1 + i − 2 1 = i .
Check: Since z z z is entire, the integral from 0 0 0 to 1 + i 1 + i 1 + i is 1 2 ( 1 + i ) 2 = i \frac{1}{2}(1+i)^2 = i 2 1 ( 1 + i ) 2 = i . Consistent. ■ \blacksquare ■
Solution Problem. Use the ML inequality to show that lim R → ∞ ∫ C R e i z z d z = 0 \lim_{R \to \infty} \int_{C_R} \frac{e^{iz}}{z}\, dz = 0 lim R → ∞ ∫ C R z e i z d z = 0 Where C R C_R C R is the upper semicircle ∣ z ∣ = R |z| = R ∣ z ∣ = R , I m ( z ) ≥ 0 \mathrm{Im}(z) \geq 0 Im ( z ) ≥ 0 .
On C R C_R C R : z = R e i θ z = Re^{i\theta} z = R e i θ , 0 ≤ θ ≤ π 0 \leq \theta \leq \pi 0 ≤ θ ≤ π . ∣ e i z ∣ = ∣ e i R ( cos θ + i sin θ ) ∣ = e − R sin θ |e^{iz}| = |e^{iR(\cos\theta + i\sin\theta)}| = e^{-R\sin\theta} ∣ e i z ∣ = ∣ e i R ( c o s θ + i s i n θ ) ∣ = e − R s i n θ .
∣ ∫ C R e i z z d z ∣ ≤ ∫ 0 π e − R sin θ R ⋅ R d θ = ∫ 0 π e − R sin θ d θ \left|\int_{C_R} \frac{e^{iz}}{z}\, dz\right| \leq \int_0^\pi \frac{e^{-R\sin\theta}}{R} \cdot R\, d\theta = \int_0^\pi e^{-R\sin\theta}\, d\theta ∫ C R z e i z d z ≤ ∫ 0 π R e − R s i n θ ⋅ R d θ = ∫ 0 π e − R s i n θ d θ .
By Jordan’s inequality sin θ ≥ 2 θ π \sin\theta \geq \frac{2\theta}{\pi} sin θ ≥ π 2 θ for θ ∈ [ 0 , π / 2 ] \theta \in [0, \pi/2] θ ∈ [ 0 , π /2 ] :
≤ 2 ∫ 0 π / 2 e − 2 R θ / π d θ = π R ( 1 − e − R ) → 0 \leq 2\int_0^{\pi/2} e^{-2R\theta/\pi}\, d\theta = \frac{\pi}{R}(1 - e^{-R}) \to 0 ≤ 2 ∫ 0 π /2 e − 2 R θ / π d θ = R π ( 1 − e − R ) → 0 as R → ∞ R \to \infty R → ∞ . ■ \blacksquare ■
Problem. Bound ∣ ∫ γ d z z 2 + 4 ∣ \left|\int_\gamma \frac{dz}{z^2 + 4}\right| ∫ γ z 2 + 4 d z where γ \gamma γ is ∣ z ∣ = 3 |z| = 3 ∣ z ∣ = 3 .
On γ \gamma γ : ∣ z 2 + 4 ∣ ≥ ∣ z ∣ 2 − 4 = 9 − 4 = 5 |z^2 + 4| \geq |z|^2 - 4 = 9 - 4 = 5 ∣ z 2 + 4∣ ≥ ∣ z ∣ 2 − 4 = 9 − 4 = 5 (reverse triangle inequality). So ∣ 1 z 2 + 4 ∣ ≤ 1 5 \left|\frac{1}{z^2 + 4}\right| \leq \frac{1}{5} z 2 + 4 1 ≤ 5 1 .
Length of γ \gamma γ : L = 2 π ⋅ 3 = 6 π L = 2\pi \cdot 3 = 6\pi L = 2 π ⋅ 3 = 6 π .
∣ ∫ γ d z z 2 + 4 ∣ ≤ 1 5 ⋅ 6 π = 6 π 5 \left|\int_\gamma \frac{dz}{z^2 + 4}\right| \leq \frac{1}{5} \cdot 6\pi = \frac{6\pi}{5} ∫ γ z 2 + 4 d z ≤ 5 1 ⋅ 6 π = 5 6 π .
When f f f is analytic on a connected domain and has a known antiderivative F F F with F ′ = f F' = f F ′ = f :
∫ γ f ( z ) d z = F ( γ ( b ) ) − F ( γ ( a ) ) \int_\gamma f(z)\, dz = F(\gamma(b)) - F(\gamma(a)) ∫ γ f ( z ) d z = F ( γ ( b )) − F ( γ ( a ))
This follows from the fundamental theorem of calculus applied to F ( γ ( t ) ) F(\gamma(t)) F ( γ ( t )) .
Solution Problem. Evaluate ∫ γ cos z d z \int_\gamma \cos z\, dz ∫ γ cos z d z where γ \gamma γ is any path from 0 0 0 to π + i \pi + i π + i .
Since cos z \cos z cos z is entire with antiderivative sin z \sin z sin z :
∫ γ cos z d z = sin ( π + i ) − sin ( 0 ) = sin ( π + i ) \int_\gamma \cos z\, dz = \sin(\pi + i) - \sin(0) = \sin(\pi + i) ∫ γ cos z d z = sin ( π + i ) − sin ( 0 ) = sin ( π + i ) .
sin ( π + i ) = sin π cosh 1 + i cos π sinh 1 = − i sinh 1 \sin(\pi + i) = \sin\pi\cosh 1 + i\cos\pi\sinh 1 = -i\sinh 1 sin ( π + i ) = sin π cosh 1 + i cos π sinh 1 = − i sinh 1 .
So the integral equals − i sinh 1 -i\sinh 1 − i sinh 1 .
Problem. Evaluate ∫ γ e 2 z d z \int_\gamma e^{2z}\, dz ∫ γ e 2 z d z where γ \gamma γ is any path from 1 1 1 to i i i .
Antiderivative: 1 2 e 2 z \frac{1}{2}e^{2z} 2 1 e 2 z .
∫ γ e 2 z d z = 1 2 ( e 2 i − e 2 ) \int_\gamma e^{2z}\, dz = \frac{1}{2}(e^{2i} - e^{2}) ∫ γ e 2 z d z = 2 1 ( e 2 i − e 2 ) .
Theorem 5.1 (Cauchy’s Theorem). If f f f is analytic on a connected domain D D D and γ \gamma γ Is a simple closed contour in D D D Then
∫ γ f ( z ) d z = 0 \int_\gamma f(z)\, dz = 0 ∫ γ f ( z ) d z = 0
Proof (for f ′ f' f ′ continuous). By Green’s theorem in the plane, writing f = u + i v f = u + iv f = u + i v :
∫ γ f d z = ∫ γ ( u d x − v d y ) + i ∫ γ ( v d x + u d y ) \int_\gamma f\, dz = \int_\gamma (u\, dx - v\, dy) + i\int_\gamma (v\, dx + u\, dy) ∫ γ f d z = ∫ γ ( u d x − v d y ) + i ∫ γ ( v d x + u d y )
Applying Green’s theorem to each integral:
= ∬ D ( − v x − u y ) d A + i ∬ D ( u x − v y ) d A = 0 = \iint_D (-v_x - u_y)\, dA + i\iint_D (u_x - v_y)\, dA = 0 = ∬ D ( − v x − u y ) d A + i ∬ D ( u x − v y ) d A = 0
By the Cauchy-Riemann equations. ■ \blacksquare ■
A domain D ⊆ C D \subseteq \mathbb{C} D ⊆ C is ** connected** if every simple closed contour in D D D can Be continuously shrunk to a point within D D D .
Cauchy’s theorem may fail on multiply connected domains. For example, ∫ γ 1 z d z = 2 π i \int_\gamma \frac{1}{z}\, dz = 2\pi i ∫ γ z 1 d z = 2 π i where γ \gamma γ is the unit circle (traversing a region that Excludes the singularity at z = 0 z = 0 z = 0 ).
Corollary 5.2. If f f f is analytic on a connected domain D D D Then the integral ∫ z 0 z 1 f ( z ) d z \int_{z_0}^{z_1} f(z)\, dz ∫ z 0 z 1 f ( z ) d z is independent of the path from z 0 z_0 z 0 to z 1 z_1 z 1 in D D D .
Theorem 5.3. If f f f is analytic on a connected domain D D D Then f f f has an antiderivative F F F in D D D (i.e., F ′ ( z ) = f ( z ) F'(z) = f(z) F ′ ( z ) = f ( z ) ), and
∫ γ f ( z ) d z = F ( z 1 ) − F ( z 0 ) \int_\gamma f(z)\, dz = F(z_1) - F(z_0) ∫ γ f ( z ) d z = F ( z 1 ) − F ( z 0 )
Where z 0 z_0 z 0 and z 1 z_1 z 1 are the endpoints of γ \gamma γ .
Theorem 5.4. If f f f is analytic on a domain D D D containing simple closed contours γ , γ 1 , … , γ n \gamma, \gamma_1, \ldots, \gamma_n γ , γ 1 , … , γ n where γ 1 , … , γ n \gamma_1, \ldots, \gamma_n γ 1 , … , γ n Lie in the interior of γ \gamma γ and the region between γ \gamma γ and the γ k \gamma_k γ k is contained in D D D And all contours are positively oriented, then
∫ γ f ( z ) d z = ∑ k = 1 n ∫ γ k f ( z ) d z \int_\gamma f(z)\, dz = \sum_{k=1}^n \int_{\gamma_k} f(z)\, dz ∫ γ f ( z ) d z = ∑ k = 1 n ∫ γ k f ( z ) d z
Theorem 5.5 (Deformation of Contours). If f f f is analytic on a domain containing two simple Closed contours γ 1 \gamma_1 γ 1 and γ 2 \gamma_2 γ 2 where one can be continuously deformed into the other Within the domain of analyticity of f f f Then
∫ γ 1 f ( z ) d z = ∫ γ 2 f ( z ) d z \int_{\gamma_1} f(z)\, dz = \int_{\gamma_2} f(z)\, dz ∫ γ 1 f ( z ) d z = ∫ γ 2 f ( z ) d z
Proof. This follows directly from Theorem 5.4 applied to the region between γ 1 \gamma_1 γ 1 and γ 2 \gamma_2 γ 2 . ■ \blacksquare ■
Remark. This theorem is enormously useful: we can replace a complicated contour with a simpler one (a small circle around each singularity) without changing the value of the integral.
Solution Problem. Evaluate ∫ γ d z z − 2 \int_\gamma \frac{dz}{z - 2} ∫ γ z − 2 d z where γ \gamma γ is the ellipse x 2 4 + y 2 9 = 1 \frac{x^2}{4} + \frac{y^2}{9} = 1 4 x 2 + 9 y 2 = 1 .
Since z = 2 z = 2 z = 2 is inside the ellipse and 1 z − 2 \frac{1}{z - 2} z − 2 1 is analytic everywhere else, By deformation of contours we can replace γ \gamma γ with a small circle around z = 2 z = 2 z = 2 :
∫ γ d z z − 2 = 2 π i \int_\gamma \frac{dz}{z - 2} = 2\pi i ∫ γ z − 2 d z = 2 π i .
Problem. Evaluate ∫ γ e z z d z \int_\gamma \frac{e^z}{z}\, dz ∫ γ z e z d z where γ \gamma γ is the square with vertices ± 2 ± 2 i \pm 2 \pm 2i ± 2 ± 2 i .
e z z \frac{e^z}{z} z e z is analytic on and inside γ \gamma γ except at z = 0 z = 0 z = 0 . By deformation: ∫ γ e z z d z = ∫ ∣ z ∣ = r e z z d z = 2 π i ⋅ e 0 = 2 π i \int_\gamma \frac{e^z}{z}\, dz = \int_{|z|=r} \frac{e^z}{z}\, dz = 2\pi i \cdot e^0 = 2\pi i ∫ γ z e z d z = ∫ ∣ z ∣ = r z e z d z = 2 π i ⋅ e 0 = 2 π i .
Problem. Evaluate ∫ γ d z z 2 − 1 \int_\gamma \frac{dz}{z^2 - 1} ∫ γ z 2 − 1 d z where γ \gamma γ is ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
1 z 2 − 1 = 1 2 ( 1 z − 1 − 1 z + 1 ) \frac{1}{z^2 - 1} = \frac{1}{2}\left(\frac{1}{z-1} - \frac{1}{z+1}\right) z 2 − 1 1 = 2 1 ( z − 1 1 − z + 1 1 ) .
Both z = ± 1 z = \pm 1 z = ± 1 are inside ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
∫ γ d z z 2 − 1 = 1 2 ( 2 π i − 2 π i ) = 0 \int_\gamma \frac{dz}{z^2 - 1} = \frac{1}{2}(2\pi i - 2\pi i) = 0 ∫ γ z 2 − 1 d z = 2 1 ( 2 π i − 2 π i ) = 0 .
Theorem 6.1 (Cauchy’s Integral Formula). If f f f is analytic on a connected domain Containing a simple closed positively oriented contour γ \gamma γ And z 0 z_0 z 0 is inside γ \gamma γ Then
f ( z 0 ) = 1 2 π i ∫ γ f ( z ) z − z 0 d z f(z_0) = \frac{1}{2\pi i}\int_\gamma \frac{f(z)}{z - z_0}\, dz f ( z 0 ) = 2 π i 1 ∫ γ z − z 0 f ( z ) d z
Proof. Let γ ε \gamma_\varepsilon γ ε be a small circle of radius ε \varepsilon ε around z 0 z_0 z 0 . Since f ( z ) z − z 0 \frac{f(z)}{z - z_0} z − z 0 f ( z ) is analytic on the region between γ \gamma γ and γ ε \gamma_\varepsilon γ ε
∫ γ f ( z ) z − z 0 d z = ∫ γ ε f ( z ) z − z 0 d z \int_\gamma \frac{f(z)}{z - z_0}\, dz = \int_{\gamma_\varepsilon} \frac{f(z)}{z - z_0}\, dz ∫ γ z − z 0 f ( z ) d z = ∫ γ ε z − z 0 f ( z ) d z
On γ ε \gamma_\varepsilon γ ε : f ( z ) = f ( z 0 ) + ( z − z 0 ) f ′ ( ζ ) f(z) = f(z_0) + (z - z_0)f'(\zeta) f ( z ) = f ( z 0 ) + ( z − z 0 ) f ′ ( ζ ) for some ζ \zeta ζ between z z z and z 0 z_0 z 0 .
= ∫ γ ε f ( z 0 ) z − z 0 d z + ∫ γ ε f ′ ( ζ ) d z = f ( z 0 ) ⋅ 2 π i + 0 = \int_{\gamma_\varepsilon} \frac{f(z_0)}{z - z_0}\, dz + \int_{\gamma_\varepsilon} f'(\zeta)\, dz = f(z_0) \cdot 2\pi i + 0 = ∫ γ ε z − z 0 f ( z 0 ) d z + ∫ γ ε f ′ ( ζ ) d z = f ( z 0 ) ⋅ 2 π i + 0
Since ∫ γ ε d z z − z 0 = 2 π i \int_{\gamma_\varepsilon} \frac{dz}{z - z_0} = 2\pi i ∫ γ ε z − z 0 d z = 2 π i (parameterize z = z 0 + ε e i θ z = z_0 + \varepsilon e^{i\theta} z = z 0 + ε e i θ ) and ∫ γ ε f ′ ( ζ ) d z → 0 \int_{\gamma_\varepsilon} f'(\zeta)\, dz \to 0 ∫ γ ε f ′ ( ζ ) d z → 0 as ε → 0 \varepsilon \to 0 ε → 0 by the ML inequality. ■ \blacksquare ■
Theorem 6.2 (Cauchy’s Integral Formula for Derivatives). Under the same conditions,
f ( n ) ( z 0 ) = n ! 2 π i ∫ γ f ( z ) ( z − z 0 ) n + 1 d z f^{(n)}(z_0) = \frac{n!}{2\pi i}\int_\gamma \frac{f(z)}{(z - z_0)^{n+1}}\, dz f ( n ) ( z 0 ) = 2 π i n ! ∫ γ ( z − z 0 ) n + 1 f ( z ) d z
Proof. We proceed by induction. The base case n = 0 n = 0 n = 0 is Theorem 6.1. For the inductive step, Assume the formula holds for n n n . Using the difference quotient:
f ( n + 1 ) ( z 0 ) = lim h → 0 f ( n ) ( z 0 + h ) − f ( n ) ( z 0 ) h = lim h → 0 n ! 2 π i ∫ γ 1 h [ f ( z ) ( z − z 0 − h ) n + 1 − f ( z ) ( z − z 0 ) n + 1 ] d z f^{(n+1)}(z_0) = \lim_{h \to 0} \frac{f^{(n)}(z_0 + h) - f^{(n)}(z_0)}{h} = \lim_{h \to 0} \frac{n!}{2\pi i}\int_\gamma \frac{1}{h}\left[\frac{f(z)}{(z - z_0 - h)^{n+1}} - \frac{f(z)}{(z - z_0)^{n+1}}\right] dz f ( n + 1 ) ( z 0 ) = lim h → 0 h f ( n ) ( z 0 + h ) − f ( n ) ( z 0 ) = lim h → 0 2 π i n ! ∫ γ h 1 [ ( z − z 0 − h ) n + 1 f ( z ) − ( z − z 0 ) n + 1 f ( z ) ] d z
= ( n + 1 ) ! 2 π i ∫ γ f ( z ) ( z − z 0 ) n + 2 d z = \frac{(n+1)!}{2\pi i}\int_\gamma \frac{f(z)}{(z - z_0)^{n+2}}\, dz = 2 π i ( n + 1 )! ∫ γ ( z − z 0 ) n + 2 f ( z ) d z
Where we justified passing the limit inside the integral by uniform convergence of the integrand On compact subsets. ■ \blacksquare ■
Corollary 6.3. If f f f is analytic, then f f f is infinitely differentiable.
This is remarkable: a single complex derivative implies the existence of all derivatives.
Corollary 6.4 (Cauchy’s Estimates). If f f f is analytic on and inside a circle ∣ z − z 0 ∣ = R |z - z_0| = R ∣ z − z 0 ∣ = R And ∣ f ( z ) ∣ ≤ M |f(z)| \leq M ∣ f ( z ) ∣ ≤ M on the circle, then
∣ f ( n ) ( z 0 ) ∣ ≤ n ! M R n |f^{(n)}(z_0)| \leq \frac{n!M}{R^n} ∣ f ( n ) ( z 0 ) ∣ ≤ R n n ! M
Proof. From the integral formula: ∣ f ( n ) ( z 0 ) ∣ = n ! 2 π ∣ ∫ ∣ z − z 0 ∣ = R f ( z ) ( z − z 0 ) n + 1 d z ∣ ≤ n ! 2 π ⋅ M R n + 1 ⋅ 2 π R = n ! M R n |f^{(n)}(z_0)| = \frac{n!}{2\pi}\left|\int_{|z-z_0|=R} \frac{f(z)}{(z-z_0)^{n+1}}\, dz\right| \leq \frac{n!}{2\pi} \cdot \frac{M}{R^{n+1}} \cdot 2\pi R = \frac{n!M}{R^n} ∣ f ( n ) ( z 0 ) ∣ = 2 π n ! ∫ ∣ z − z 0 ∣ = R ( z − z 0 ) n + 1 f ( z ) d z ≤ 2 π n ! ⋅ R n + 1 M ⋅ 2 π R = R n n ! M . ■ \blacksquare ■
Theorem 6.5 (Liouville’s Theorem). Every bounded entire function is constant.
Proof. If ∣ f ( z ) ∣ ≤ M |f(z)| \leq M ∣ f ( z ) ∣ ≤ M for all z z z Then by Cauchy’s estimates with R R R arbitrarily large: ∣ f ′ ( z 0 ) ∣ ≤ M R → 0 |f'(z_0)| \leq \frac{M}{R} \to 0 ∣ f ′ ( z 0 ) ∣ ≤ R M → 0 as R → ∞ R \to \infty R → ∞ . So f ′ ( z ) = 0 f'(z) = 0 f ′ ( z ) = 0 for all z z z Meaning f f f is Constant. ■ \blacksquare ■
Corollary 6.6. If f f f is entire and ∣ f ( z ) ∣ ≥ M |f(z)| \geq M ∣ f ( z ) ∣ ≥ M for all z z z (bounded away from zero), then f f f is constant.
Proof. 1 / f 1/f 1/ f is entire and bounded by 1 / M 1/M 1/ M So constant by Liouville. ■ \blacksquare ■
Theorem 6.7 (Fundamental Theorem of Algebra). Every non-constant polynomial p ( z ) ∈ C [ z ] p(z) \in \mathbb{C}[z] p ( z ) ∈ C [ z ] has a root in C \mathbb{C} C .
Proof. Suppose p ( z ) p(z) p ( z ) has no root. Then f ( z ) = 1 / p ( z ) f(z) = 1/p(z) f ( z ) = 1/ p ( z ) is entire. Since ∣ p ( z ) ∣ → ∞ |p(z)| \to \infty ∣ p ( z ) ∣ → ∞ as ∣ z ∣ → ∞ |z| \to \infty ∣ z ∣ → ∞ , f ( z ) → 0 f(z) \to 0 f ( z ) → 0 So f f f is bounded. By Liouville’s theorem, f f f is constant, so p p p Is constant, a contradiction. ■ \blacksquare ■
Corollary 6.8. Every polynomial of degree n ≥ 1 n \geq 1 n ≥ 1 has exactly n n n roots in C \mathbb{C} C Counting multiplicities.
Solution Problem. Evaluate ∫ γ e z z − 1 d z \int_\gamma \frac{e^z}{z - 1}\, dz ∫ γ z − 1 e z d z where γ \gamma γ is ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
Solution. The function e z z − 1 \frac{e^z}{z - 1} z − 1 e z has a singularity at z = 1 z = 1 z = 1 Which lies inside γ \gamma γ . By Cauchy’s integral formula with f ( z ) = e z f(z) = e^z f ( z ) = e z and z 0 = 1 z_0 = 1 z 0 = 1 :
∫ γ e z z − 1 d z = 2 π i ⋅ f ( 1 ) = 2 π i ⋅ e 1 = 2 π i e \int_\gamma \frac{e^z}{z - 1}\, dz = 2\pi i \cdot f(1) = 2\pi i \cdot e^1 = 2\pi i e ∫ γ z − 1 e z d z = 2 π i ⋅ f ( 1 ) = 2 π i ⋅ e 1 = 2 π i e . ■ \blacksquare ■
Problem. Evaluate ∫ γ z 2 + 1 ( z − i ) 3 d z \int_\gamma \frac{z^2 + 1}{(z - i)^3}\, dz ∫ γ ( z − i ) 3 z 2 + 1 d z where γ \gamma γ is ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
By Cauchy’s formula for derivatives with f ( z ) = z 2 + 1 f(z) = z^2 + 1 f ( z ) = z 2 + 1 and z 0 = i z_0 = i z 0 = i :
∫ γ f ( z ) ( z − i ) 3 d z = 2 π i 2 ! f ′ ′ ( i ) \int_\gamma \frac{f(z)}{(z - i)^3}\, dz = \frac{2\pi i}{2!}\,f''(i) ∫ γ ( z − i ) 3 f ( z ) d z = 2 ! 2 π i f ′′ ( i ) .
f ′ ( z ) = 2 z f'(z) = 2z f ′ ( z ) = 2 z , f ′ ′ ( z ) = 2 f''(z) = 2 f ′′ ( z ) = 2 . So f ′ ′ ( i ) = 2 f''(i) = 2 f ′′ ( i ) = 2 .
∫ γ z 2 + 1 ( z − i ) 3 d z = 2 π i 2 ⋅ 2 = 2 π i \int_\gamma \frac{z^2 + 1}{(z - i)^3}\, dz = \frac{2\pi i}{2} \cdot 2 = 2\pi i ∫ γ ( z − i ) 3 z 2 + 1 d z = 2 2 π i ⋅ 2 = 2 π i . ■ \blacksquare ■
Problem. Evaluate ∫ γ sin z z ( z − π ) d z \int_\gamma \frac{\sin z}{z(z - \pi)}\, dz ∫ γ z ( z − π ) s i n z d z where γ \gamma γ is ∣ z ∣ = 4 |z| = 4 ∣ z ∣ = 4 .
Singularities inside γ \gamma γ : z = 0 z = 0 z = 0 and z = π z = \pi z = π .
sin z z ( z − π ) = 1 π ( sin z z − π − sin z z ) \frac{\sin z}{z(z - \pi)} = \frac{1}{\pi}\left(\frac{\sin z}{z - \pi} - \frac{\sin z}{z}\right) z ( z − π ) s i n z = π 1 ( z − π s i n z − z s i n z ) .
At z = 0 z = 0 z = 0 : by CIF, ∫ γ sin z z d z = 2 π i ⋅ sin ( 0 ) = 0 \int_\gamma \frac{\sin z}{z}\, dz = 2\pi i \cdot \sin(0) = 0 ∫ γ z s i n z d z = 2 π i ⋅ sin ( 0 ) = 0 . At z = π z = \pi z = π : by CIF, ∫ γ sin z z − π d z = 2 π i ⋅ sin ( π ) = 0 \int_\gamma \frac{\sin z}{z - \pi}\, dz = 2\pi i \cdot \sin(\pi) = 0 ∫ γ z − π s i n z d z = 2 π i ⋅ sin ( π ) = 0 .
∫ γ sin z z ( z − π ) d z = 1 π ( 0 − 0 ) = 0 \int_\gamma \frac{\sin z}{z(z - \pi)}\, dz = \frac{1}{\pi}(0 - 0) = 0 ∫ γ z ( z − π ) s i n z d z = π 1 ( 0 − 0 ) = 0 .
Problem. Evaluate ∫ γ e 2 z ( z − 1 ) 2 ( z + 1 ) d z \int_\gamma \frac{e^{2z}}{(z - 1)^2(z + 1)}\, dz ∫ γ ( z − 1 ) 2 ( z + 1 ) e 2 z d z where γ \gamma γ is ∣ z ∣ = 3 |z| = 3 ∣ z ∣ = 3 .
By partial fractions: 1 ( z − 1 ) 2 ( z + 1 ) = 1 / 4 z + 1 − 1 / 4 z − 1 + 1 / 2 ( z − 1 ) 2 \frac{1}{(z-1)^2(z+1)} = \frac{1/4}{z+1} - \frac{1/4}{z-1} + \frac{1/2}{(z-1)^2} ( z − 1 ) 2 ( z + 1 ) 1 = z + 1 1/4 − z − 1 1/4 + ( z − 1 ) 2 1/2 .
∫ γ e 2 z ( z − 1 ) 2 ( z + 1 ) d z = 1 4 ⋅ 2 π i ⋅ e − 2 − 1 4 ⋅ 2 π i ⋅ e 2 + 1 2 ⋅ 2 π i 1 ! ⋅ 2 e 2 \int_\gamma \frac{e^{2z}}{(z-1)^2(z+1)}\, dz = \frac{1}{4} \cdot 2\pi i \cdot e^{-2} - \frac{1}{4} \cdot 2\pi i \cdot e^2 + \frac{1}{2} \cdot \frac{2\pi i}{1!} \cdot 2e^2 ∫ γ ( z − 1 ) 2 ( z + 1 ) e 2 z d z = 4 1 ⋅ 2 π i ⋅ e − 2 − 4 1 ⋅ 2 π i ⋅ e 2 + 2 1 ⋅ 1 ! 2 π i ⋅ 2 e 2
= π i e − 2 2 − π i e 2 2 + 2 π i e 2 = π i e − 2 2 + 3 π i e 2 2 = \frac{\pi i e^{-2}}{2} - \frac{\pi i e^2}{2} + 2\pi i e^2 = \frac{\pi i e^{-2}}{2} + \frac{3\pi i e^2}{2} = 2 π i e − 2 − 2 π i e 2 + 2 π i e 2 = 2 π i e − 2 + 2 3 π i e 2 .
Theorem 7.1. If f f f is analytic on ∣ z − z 0 ∣ < R |z - z_0| \lt R ∣ z − z 0 ∣ < R Then
f ( z ) = ∑ n = 0 ∞ f ( n ) ( z 0 ) n ! ( z − z 0 ) n f(z) = \sum_{n=0}^{\infty} \frac{f^{(n)}(z_0)}{n!}(z - z_0)^n f ( z ) = ∑ n = 0 ∞ n ! f ( n ) ( z 0 ) ( z − z 0 ) n
And the series converges uniformly on compact subsets of ∣ z − z 0 ∣ < R |z - z_0| \lt R ∣ z − z 0 ∣ < R .
Proof. For ∣ z − z 0 ∣ < r < R |z - z_0| \lt r \lt R ∣ z − z 0 ∣ < r < R Apply Cauchy’s integral formula on ∣ ζ − z 0 ∣ = r |\zeta - z_0| = r ∣ ζ − z 0 ∣ = r :
f ( z ) = 1 2 π i ∫ ∣ ζ − z 0 ∣ = r f ( ζ ) ζ − z d ζ f(z) = \frac{1}{2\pi i}\int_{|\zeta - z_0| = r} \frac{f(\zeta)}{\zeta - z}\, d\zeta f ( z ) = 2 π i 1 ∫ ∣ ζ − z 0 ∣ = r ζ − z f ( ζ ) d ζ
Write 1 ζ − z = 1 ( ζ − z 0 ) − ( z − z 0 ) = 1 ζ − z 0 ⋅ 1 1 − ( z − z 0 ) / ( ζ − z 0 ) \frac{1}{\zeta - z} = \frac{1}{(\zeta - z_0) - (z - z_0)} = \frac{1}{\zeta - z_0} \cdot \frac{1}{1 - (z - z_0)/(\zeta - z_0)} ζ − z 1 = ( ζ − z 0 ) − ( z − z 0 ) 1 = ζ − z 0 1 ⋅ 1 − ( z − z 0 ) / ( ζ − z 0 ) 1 = ∑ n = 0 ∞ ( z − z 0 ) n ( ζ − z 0 ) n + 1 = \sum_{n=0}^{\infty} \frac{(z - z_0)^n}{(\zeta - z_0)^{n+1}} = ∑ n = 0 ∞ ( ζ − z 0 ) n + 1 ( z − z 0 ) n (geometric series, convergent since ∣ z − z 0 ∣ / ∣ ζ − z 0 ∣ < 1 |z - z_0|/|\zeta - z_0| \lt 1 ∣ z − z 0 ∣/∣ ζ − z 0 ∣ < 1 ).
Substituting and integrating term by term gives the Taylor series. ■ \blacksquare ■
Remark. The radius of convergence R R R is the distance from z 0 z_0 z 0 to the nearest singularity of f f f .
e z = ∑ n = 0 ∞ z n n ! = 1 + z + z 2 2 ! + ⋯ e^z = \sum_{n=0}^{\infty} \frac{z^n}{n!} = 1 + z + \frac{z^2}{2!} + \cdots e z = ∑ n = 0 ∞ n ! z n = 1 + z + 2 ! z 2 + ⋯
sin z = ∑ n = 0 ∞ ( − 1 ) n z 2 n + 1 ( 2 n + 1 ) ! \sin z = \sum_{n=0}^{\infty} \frac{(-1)^n z^{2n+1}}{(2n+1)!} sin z = ∑ n = 0 ∞ ( 2 n + 1 )! ( − 1 ) n z 2 n + 1
cos z = ∑ n = 0 ∞ ( − 1 ) n z 2 n ( 2 n ) ! \cos z = \sum_{n=0}^{\infty} \frac{(-1)^n z^{2n}}{(2n)!} cos z = ∑ n = 0 ∞ ( 2 n )! ( − 1 ) n z 2 n
1 1 − z = ∑ n = 0 ∞ z n , ∣ z ∣ < 1 \frac{1}{1 - z} = \sum_{n=0}^{\infty} z^n, \quad |z| \lt 1 1 − z 1 = ∑ n = 0 ∞ z n , ∣ z ∣ < 1
ln ( 1 + z ) = ∑ n = 1 ∞ ( − 1 ) n + 1 z n n , ∣ z ∣ < 1 \ln(1 + z) = \sum_{n=1}^{\infty} \frac{(-1)^{n+1} z^n}{n}, \quad |z| \lt 1 ln ( 1 + z ) = ∑ n = 1 ∞ n ( − 1 ) n + 1 z n , ∣ z ∣ < 1
Solution Problem. Find the Taylor series of f ( z ) = 1 z f(z) = \frac{1}{z} f ( z ) = z 1 centered at z 0 = 1 z_0 = 1 z 0 = 1 .
1 z = 1 1 + ( z − 1 ) = ∑ n = 0 ∞ ( − 1 ) n ( z − 1 ) n \frac{1}{z} = \frac{1}{1 + (z - 1)} = \sum_{n=0}^{\infty} (-1)^n (z - 1)^n z 1 = 1 + ( z − 1 ) 1 = ∑ n = 0 ∞ ( − 1 ) n ( z − 1 ) n for ∣ z − 1 ∣ < 1 |z - 1| \lt 1 ∣ z − 1∣ < 1 .
Radius of convergence: distance from z 0 = 1 z_0 = 1 z 0 = 1 to the singularity at z = 0 z = 0 z = 0 Which is 1 1 1 .
Problem. Find the Taylor series of f ( z ) = 1 ( 1 − z ) 2 f(z) = \frac{1}{(1 - z)^2} f ( z ) = ( 1 − z ) 2 1 centered at z 0 = 0 z_0 = 0 z 0 = 0 .
1 ( 1 − z ) 2 = d d z [ 1 1 − z ] = d d z ∑ n = 0 ∞ z n = ∑ n = 1 ∞ n z n − 1 = ∑ n = 0 ∞ ( n + 1 ) z n \frac{1}{(1-z)^2} = \frac{d}{dz}\left[\frac{1}{1 - z}\right] = \frac{d}{dz}\sum_{n=0}^{\infty} z^n = \sum_{n=1}^{\infty} nz^{n-1} = \sum_{n=0}^{\infty} (n+1)z^n ( 1 − z ) 2 1 = d z d [ 1 − z 1 ] = d z d ∑ n = 0 ∞ z n = ∑ n = 1 ∞ n z n − 1 = ∑ n = 0 ∞ ( n + 1 ) z n for ∣ z ∣ < 1 |z| \lt 1 ∣ z ∣ < 1 .
Problem. Find the Taylor series of f ( z ) = e z sin z f(z) = e^z \sin z f ( z ) = e z sin z up to the z 4 z^4 z 4 term.
e z = 1 + z + z 2 / 2 + z 3 / 6 + z 4 / 24 + ⋯ e^z = 1 + z + z^2/2 + z^3/6 + z^4/24 + \cdots e z = 1 + z + z 2 /2 + z 3 /6 + z 4 /24 + ⋯ sin z = z − z 3 / 6 + z 5 / 120 − ⋯ \sin z = z - z^3/6 + z^5/120 - \cdots sin z = z − z 3 /6 + z 5 /120 − ⋯
e z sin z = ( 1 + z + z 2 / 2 + z 3 / 6 + z 4 / 24 + ⋯ ) ( z − z 3 / 6 + ⋯ ) e^z \sin z = (1 + z + z^2/2 + z^3/6 + z^4/24 + \cdots)(z - z^3/6 + \cdots) e z sin z = ( 1 + z + z 2 /2 + z 3 /6 + z 4 /24 + ⋯ ) ( z − z 3 /6 + ⋯ )
= z + z 2 + z 3 / 2 + z 4 / 6 + ⋯ − z 3 / 6 − z 4 / 6 + ⋯ = z + z^2 + z^3/2 + z^4/6 + \cdots - z^3/6 - z^4/6 + \cdots = z + z 2 + z 3 /2 + z 4 /6 + ⋯ − z 3 /6 − z 4 /6 + ⋯ = z + z 2 + z 3 / 3 − z 4 / 30 + ⋯ = z + z^2 + z^3/3 - z^4/30 + \cdots = z + z 2 + z 3 /3 − z 4 /30 + ⋯
Theorem 7.2 (Laurent Series). If f f f is analytic on the annulus r < ∣ z − z 0 ∣ < R r \lt |z - z_0| \lt R r < ∣ z − z 0 ∣ < R Then
f ( z ) = ∑ n = − ∞ ∞ a n ( z − z 0 ) n = ⋯ + a − 2 ( z − z 0 ) 2 + a − 1 z − z 0 + a 0 + a 1 ( z − z 0 ) + ⋯ f(z) = \sum_{n=-\infty}^{\infty} a_n(z - z_0)^n = \cdots + \frac{a_{-2}}{(z - z_0)^2} + \frac{a_{-1}}{z - z_0} + a_0 + a_1(z - z_0) + \cdots f ( z ) = ∑ n = − ∞ ∞ a n ( z − z 0 ) n = ⋯ + ( z − z 0 ) 2 a − 2 + z − z 0 a − 1 + a 0 + a 1 ( z − z 0 ) + ⋯
Where
a n = 1 2 π i ∫ γ f ( z ) ( z − z 0 ) n + 1 d z a_n = \frac{1}{2\pi i}\int_\gamma \frac{f(z)}{(z - z_0)^{n+1}}\, dz a n = 2 π i 1 ∫ γ ( z − z 0 ) n + 1 f ( z ) d z
For any simple closed contour γ \gamma γ in the annulus encircling z 0 z_0 z 0 .
The principal part is ∑ n = − ∞ − 1 a n ( z − z 0 ) n \sum_{n=-\infty}^{-1} a_n(z - z_0)^n ∑ n = − ∞ − 1 a n ( z − z 0 ) n (negative powers). The analytic Part is ∑ n = 0 ∞ a n ( z − z 0 ) n \sum_{n=0}^{\infty} a_n(z - z_0)^n ∑ n = 0 ∞ a n ( z − z 0 ) n (non-negative powers).
The Laurent series expansion depends on the annulus of convergence. A function may have different Laurent expansions in different annuli.
Proposition 7.3. The Laurent series expansion of f f f in an annulus is unique.
Solution Problem. Find the Laurent series of f ( z ) = 1 z ( z − 1 ) f(z) = \frac{1}{z(z-1)} f ( z ) = z ( z − 1 ) 1 in 0 < ∣ z ∣ < 1 0 \lt |z| \lt 1 0 < ∣ z ∣ < 1 .
Solution. Using partial fractions: 1 z ( z − 1 ) = 1 z − 1 − 1 z \frac{1}{z(z-1)} = \frac{1}{z-1} - \frac{1}{z} z ( z − 1 ) 1 = z − 1 1 − z 1 .
In ∣ z ∣ < 1 |z| \lt 1 ∣ z ∣ < 1 : 1 z − 1 = − 1 1 − z = − ∑ n = 0 ∞ z n \frac{1}{z - 1} = -\frac{1}{1 - z} = -\sum_{n=0}^{\infty} z^n z − 1 1 = − 1 − z 1 = − ∑ n = 0 ∞ z n .
So f ( z ) = − ∑ n = 0 ∞ z n − 1 z = ⋯ − z 2 − z − 1 − 1 z f(z) = -\sum_{n=0}^{\infty} z^n - \frac{1}{z} = \cdots - z^2 - z - 1 - \frac{1}{z} f ( z ) = − ∑ n = 0 ∞ z n − z 1 = ⋯ − z 2 − z − 1 − z 1 .
The principal part is − 1 / z -1/z − 1/ z So z = 0 z = 0 z = 0 is a simple pole. ■ \blacksquare ■
Problem. Find the Laurent series of f ( z ) = 1 z ( z − 1 ) f(z) = \frac{1}{z(z-1)} f ( z ) = z ( z − 1 ) 1 in 1 < ∣ z ∣ < ∞ 1 \lt |z| \lt \infty 1 < ∣ z ∣ < ∞ .
In ∣ z ∣ > 1 |z| \gt 1 ∣ z ∣ > 1 : 1 z − 1 = 1 z ⋅ 1 1 − 1 / z = ∑ n = 2 ∞ z − n \frac{1}{z - 1} = \frac{1}{z} \cdot \frac{1}{1 - 1/z} = \sum_{n=2}^{\infty} z^{-n} z − 1 1 = z 1 ⋅ 1 − 1/ z 1 = ∑ n = 2 ∞ z − n .
f ( z ) = ∑ n = 2 ∞ z − n − 1 z = 1 z 2 + 1 z 3 + ⋯ f(z) = \sum_{n=2}^{\infty} z^{-n} - \frac{1}{z} = \frac{1}{z^2} + \frac{1}{z^3} + \cdots f ( z ) = ∑ n = 2 ∞ z − n − z 1 = z 2 1 + z 3 1 + ⋯
Problem. Find the Laurent series of f ( z ) = e z z 2 f(z) = \frac{e^z}{z^2} f ( z ) = z 2 e z in 0 < ∣ z ∣ < ∞ 0 \lt |z| \lt \infty 0 < ∣ z ∣ < ∞ .
e z = ∑ n = 0 ∞ z n n ! e^z = \sum_{n=0}^{\infty} \frac{z^n}{n!} e z = ∑ n = 0 ∞ n ! z n So f ( z ) = ∑ n = 0 ∞ z n − 2 n ! = 1 z 2 + 1 z + 1 2 + z 6 + ⋯ f(z) = \sum_{n=0}^{\infty} \frac{z^{n-2}}{n!} = \frac{1}{z^2} + \frac{1}{z} + \frac{1}{2} + \frac{z}{6} + \cdots f ( z ) = ∑ n = 0 ∞ n ! z n − 2 = z 2 1 + z 1 + 2 1 + 6 z + ⋯
Residue at z = 0 z = 0 z = 0 : a − 1 = 1 a_{-1} = 1 a − 1 = 1 .
Problem. Find the Laurent series of f ( z ) = 1 z 2 ( z − 3 ) f(z) = \frac{1}{z^2(z - 3)} f ( z ) = z 2 ( z − 3 ) 1 in 0 < ∣ z ∣ < 3 0 \lt |z| \lt 3 0 < ∣ z ∣ < 3 .
1 z − 3 = − 1 3 ∑ n = 0 ∞ z n 3 n \frac{1}{z - 3} = -\frac{1}{3}\sum_{n=0}^{\infty} \frac{z^n}{3^n} z − 3 1 = − 3 1 ∑ n = 0 ∞ 3 n z n .
f ( z ) = − ∑ n = 0 ∞ z n − 2 3 n + 1 = − 1 3 z 2 − 1 9 z − 1 27 − z 81 − ⋯ f(z) = -\sum_{n=0}^{\infty} \frac{z^{n-2}}{3^{n+1}} = -\frac{1}{3z^2} - \frac{1}{9z} - \frac{1}{27} - \frac{z}{81} - \cdots f ( z ) = − ∑ n = 0 ∞ 3 n + 1 z n − 2 = − 3 z 2 1 − 9 z 1 − 27 1 − 81 z − ⋯
Residue at z = 0 z = 0 z = 0 : a − 1 = − 1 9 a_{-1} = -\frac{1}{9} a − 1 = − 9 1 .
Definition. The residue at infinity of f f f is defined as
R e s ( f , ∞ ) = − 1 2 π i ∫ ∣ z ∣ = R f ( z ) d z \mathrm{Res}(f, \infty) = -\frac{1}{2\pi i}\int_{|z|=R} f(z)\, dz Res ( f , ∞ ) = − 2 π i 1 ∫ ∣ z ∣ = R f ( z ) d z
For sufficiently large R R R (enclosing all finite singularities).
Proposition 7.4. For a function f f f with finitely many singularities in C \mathbb{C} C :
∑ a l l f i n i t e z k R e s ( f , z k ) + R e s ( f , ∞ ) = 0 \sum_{\mathrm{all\ finite\ } z_k} \mathrm{Res}(f, z_k) + \mathrm{Res}(f, \infty) = 0 ∑ all finite z k Res ( f , z k ) + Res ( f , ∞ ) = 0
Proof. By the residue theorem applied to ∣ z ∣ = R |z| = R ∣ z ∣ = R enclosing all finite singularities:
∫ ∣ z ∣ = R f d z = 2 π i ∑ f i n i t e R e s ( f , z k ) \int_{|z|=R} f\, dz = 2\pi i \sum_{\mathrm{finite} \mathrm{Res}(f, z_k)} ∫ ∣ z ∣ = R f d z = 2 π i ∑ finite Res ( f , z k ) .
But R e s ( f , ∞ ) = − 1 2 π i ∫ ∣ z ∣ = R f d z \mathrm{Res}(f, \infty) = -\frac{1}{2\pi i}\int_{|z|=R} f\, dz Res ( f , ∞ ) = − 2 π i 1 ∫ ∣ z ∣ = R f d z So the sum is zero. ■ \blacksquare ■
Let z 0 z_0 z 0 be an isolated singularity of f f f (i.e., f f f is analytic in a punctured neighbourhood of z 0 z_0 z 0 ).
Classification by Laurent series:
Removable singularity : a n = 0 a_n = 0 a n = 0 for all n < 0 n \lt 0 n < 0 . Can be removed by redefining f ( z 0 ) = a 0 f(z_0) = a_0 f ( z 0 ) = a 0 .Pole of order m m m : a − m ≠ 0 a_{-m} \neq 0 a − m = 0 and a n = 0 a_n = 0 a n = 0 for n < − m n \lt -m n < − m . The principal part is finite.Essential singularity : infinitely many non-zero a n a_n a n with n < 0 n \lt 0 n < 0 .Proposition 8.1 (Riemann’s Removable Singularity Theorem). If f f f is bounded near z 0 z_0 z 0 Then z 0 z_0 z 0 is a removable singularity.
Proposition 8.2. z 0 z_0 z 0 is a pole of order m m m if and only if lim z → z 0 ( z − z 0 ) m f ( z ) \lim_{z \to z_0} (z - z_0)^m f(z) lim z → z 0 ( z − z 0 ) m f ( z ) Exists and is non-zero.
Theorem 8.3 (Casorati-Weierstrass). If z 0 z_0 z 0 is an essential singularity of f f f Then f f f takes Values arbitrarily close to any complex number in every neighbourhood of z 0 z_0 z 0 .
Solution Problem. Classify the singularities of f ( z ) = sin z z f(z) = \frac{\sin z}{z} f ( z ) = z s i n z .
z = 0 z = 0 z = 0 : sin z = z − z 3 / 6 + ⋯ \sin z = z - z^3/6 + \cdots sin z = z − z 3 /6 + ⋯ So f ( z ) = 1 − z 2 / 6 + ⋯ f(z) = 1 - z^2/6 + \cdots f ( z ) = 1 − z 2 /6 + ⋯ . No negative powers, so z = 0 z = 0 z = 0 is a removable singularity. f ( 0 ) = 1 f(0) = 1 f ( 0 ) = 1 by continuity.
Problem. Classify the singularities of f ( z ) = e z − 1 z 2 f(z) = \frac{e^z - 1}{z^2} f ( z ) = z 2 e z − 1 .
z = 0 z = 0 z = 0 : e z − 1 = z + z 2 / 2 + ⋯ e^z - 1 = z + z^2/2 + \cdots e z − 1 = z + z 2 /2 + ⋯ So f ( z ) = 1 z + 1 2 + ⋯ f(z) = \frac{1}{z} + \frac{1}{2} + \cdots f ( z ) = z 1 + 2 1 + ⋯ . Principal part is 1 / z 1/z 1/ z So z = 0 z = 0 z = 0 is a simple pole with residue 1 1 1 .
Problem. Classify the singularity of f ( z ) = e 1 / z f(z) = e^{1/z} f ( z ) = e 1/ z at z = 0 z = 0 z = 0 .
e 1 / z = ∑ n = 0 ∞ 1 n ! z n = 1 + 1 z + 1 2 z 2 + ⋯ e^{1/z} = \sum_{n=0}^{\infty} \frac{1}{n!\, z^n} = 1 + \frac{1}{z} + \frac{1}{2z^2} + \cdots e 1/ z = ∑ n = 0 ∞ n ! z n 1 = 1 + z 1 + 2 z 2 1 + ⋯
Infinitely many negative powers ⇒ \Rightarrow ⇒ z = 0 z = 0 z = 0 is an essential singularity.
Problem. Classify the singularities of f ( z ) = z + 1 z 3 ( z 2 + 1 ) f(z) = \frac{z + 1}{z^3(z^2 + 1)} f ( z ) = z 3 ( z 2 + 1 ) z + 1 .
z = 0 z = 0 z = 0 : pole of order 3 3 3 . z = i z = i z = i : simple pole. z = − i z = -i z = − i : simple pole.
Problem. Determine the type of singularity of f ( z ) = z sin z f(z) = \frac{z}{\sin z} f ( z ) = s i n z z at z = 0 z = 0 z = 0 .
sin z = z − z 3 / 6 + ⋯ \sin z = z - z^3/6 + \cdots sin z = z − z 3 /6 + ⋯ So f ( z ) = 1 1 − z 2 / 6 + ⋯ = 1 + z 2 6 + ⋯ f(z) = \frac{1}{1 - z^2/6 + \cdots} = 1 + \frac{z^2}{6} + \cdots f ( z ) = 1 − z 2 /6 + ⋯ 1 = 1 + 6 z 2 + ⋯ .
No negative powers, so z = 0 z = 0 z = 0 is a removable singularity with f ( 0 ) = 1 f(0) = 1 f ( 0 ) = 1 .
Definition. The residue of f f f at an isolated singularity z 0 z_0 z 0 is the coefficient a − 1 a_{-1} a − 1 In the Laurent expansion:
R e s ( f , z 0 ) = a − 1 = 1 2 π i ∫ γ f ( z ) d z \mathrm{Res}(f, z_0) = a_{-1} = \frac{1}{2\pi i}\int_\gamma f(z)\, dz Res ( f , z 0 ) = a − 1 = 2 π i 1 ∫ γ f ( z ) d z
Where γ \gamma γ is a small positively oriented circle around z 0 z_0 z 0 .
For a simple pole at z 0 z_0 z 0 :
R e s ( f , z 0 ) = lim z → z 0 ( z − z 0 ) f ( z ) \mathrm{Res}(f, z_0) = \lim_{z \to z_0} (z - z_0)f(z) Res ( f , z 0 ) = lim z → z 0 ( z − z 0 ) f ( z )
If f = g / h f = g/h f = g / h where g ( z 0 ) ≠ 0 g(z_0) \neq 0 g ( z 0 ) = 0 , h ( z 0 ) = 0 h(z_0) = 0 h ( z 0 ) = 0 , h ′ ( z 0 ) ≠ 0 h'(z_0) \neq 0 h ′ ( z 0 ) = 0 :
R e s ( f , z 0 ) = g ( z 0 ) h ′ ( z 0 ) \mathrm{Res}(f, z_0) = \frac{g(z_0)}{h'(z_0)} Res ( f , z 0 ) = h ′ ( z 0 ) g ( z 0 )
For a pole of order m m m at z 0 z_0 z 0 :
R e s ( f , z 0 ) = 1 ( m − 1 ) ! lim z → z 0 d m − 1 d z m − 1 [ ( z − z 0 ) m f ( z ) ] \mathrm{Res}(f, z_0) = \frac{1}{(m-1)!}\lim_{z \to z_0} \frac{d^{m-1}}{dz^{m-1}}\left[(z - z_0)^m f(z)\right] Res ( f , z 0 ) = ( m − 1 )! 1 lim z → z 0 d z m − 1 d m − 1 [ ( z − z 0 ) m f ( z ) ]
Solution Problem. Find the residue of f ( z ) = z z 2 + 4 z + 3 f(z) = \frac{z}{z^2 + 4z + 3} f ( z ) = z 2 + 4 z + 3 z at each pole.
z 2 + 4 z + 3 = ( z + 1 ) ( z + 3 ) z^2 + 4z + 3 = (z + 1)(z + 3) z 2 + 4 z + 3 = ( z + 1 ) ( z + 3 ) So simple poles at z = − 1 z = -1 z = − 1 and z = − 3 z = -3 z = − 3 .
At z = − 1 z = -1 z = − 1 : R e s = lim z → − 1 z z + 3 = − 1 2 \mathrm{Res} = \lim_{z \to -1} \frac{z}{z + 3} = \frac{-1}{2} Res = lim z → − 1 z + 3 z = 2 − 1 . At z = − 3 z = -3 z = − 3 : R e s = lim z → − 3 z z + 1 = − 3 − 2 = 3 2 \mathrm{Res} = \lim_{z \to -3} \frac{z}{z + 1} = \frac{-3}{-2} = \frac{3}{2} Res = lim z → − 3 z + 1 z = − 2 − 3 = 2 3 .
Problem. Find the residue of f ( z ) = e z ( z − 1 ) 2 ( z − 2 ) f(z) = \frac{e^z}{(z - 1)^2(z - 2)} f ( z ) = ( z − 1 ) 2 ( z − 2 ) e z at each pole.
At z = 1 z = 1 z = 1 (pole of order 2 2 2 ): R e s = d d z [ e z z − 2 ] z = 1 = e z ( z − 2 ) − e z ( z − 2 ) 2 ∣ z = 1 = − e − e 1 = − 2 e \mathrm{Res} = \frac{d}{dz}\left[\frac{e^z}{z - 2}\right]_{z=1} = \frac{e^z(z - 2) - e^z}{(z-2)^2}\Big|_{z=1} = \frac{-e - e}{1} = -2e Res = d z d [ z − 2 e z ] z = 1 = ( z − 2 ) 2 e z ( z − 2 ) − e z z = 1 = 1 − e − e = − 2 e .
At z = 2 z = 2 z = 2 (simple pole): R e s = e 2 ( 2 − 1 ) 2 = e 2 \mathrm{Res} = \frac{e^2}{(2-1)^2} = e^2 Res = ( 2 − 1 ) 2 e 2 = e 2 .
Theorem 8.4 (Residue Theorem). If f f f is analytic inside and on a simple closed positively Oriented contour γ \gamma γ except for isolated singularities z 1 , z 2 , … , z n z_1, z_2, \ldots, z_n z 1 , z 2 , … , z n inside γ \gamma γ Then
∫ γ f ( z ) d z = 2 π i ∑ k = 1 n R e s ( f , z k ) \int_\gamma f(z)\, dz = 2\pi i \sum_{k=1}^{n} \mathrm{Res}(f, z_k) ∫ γ f ( z ) d z = 2 π i ∑ k = 1 n Res ( f , z k )
Proof. For each singularity z k z_k z k Draw a small circle γ k \gamma_k γ k around it. By Cauchy’s theorem Applied to the multiply connected region between γ \gamma γ and the γ k \gamma_k γ k :
∫ γ f d z = ∑ k = 1 n ∫ γ k f d z = ∑ k = 1 n 2 π i ⋅ R e s ( f , z k ) \int_\gamma f\, dz = \sum_{k=1}^n \int_{\gamma_k} f\, dz = \sum_{k=1}^n 2\pi i \cdot \mathrm{Res}(f, z_k) ∫ γ f d z = ∑ k = 1 n ∫ γ k f d z = ∑ k = 1 n 2 π i ⋅ Res ( f , z k ) . ■ \blacksquare ■
Solution Problem 1. Evaluate ∫ γ e z z ( z − 1 ) 2 d z \int_\gamma \frac{e^z}{z(z-1)^2}\, dz ∫ γ z ( z − 1 ) 2 e z d z where γ \gamma γ is ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
Solution. Singularities inside γ \gamma γ : z = 0 z = 0 z = 0 (simple pole) and z = 1 z = 1 z = 1 (pole of order 2 2 2 ).
At z = 0 z = 0 z = 0 : R e s = lim z → 0 e z ( z − 1 ) 2 = 1 ( − 1 ) 2 = 1 \mathrm{Res} = \lim_{z \to 0} \frac{e^z}{(z-1)^2} = \frac{1}{(-1)^2} = 1 Res = lim z → 0 ( z − 1 ) 2 e z = ( − 1 ) 2 1 = 1 .
At z = 1 z = 1 z = 1 : R e s ( f , 1 ) = d d z [ ( z − 1 ) 2 ⋅ e z z ( z − 1 ) 2 ] z = 1 = d d z [ e z z ] z = 1 = e z ⋅ z − e z z 2 ∣ z = 1 = e − e 1 = 0 \mathrm{Res}(f, 1) = \frac{d}{dz}\left[(z-1)^2 \cdot \frac{e^z}{z(z-1)^2}\right]_{z=1} = \frac{d}{dz}\left[\frac{e^z}{z}\right]_{z=1} = \frac{e^z \cdot z - e^z}{z^2}\Big|_{z=1} = \frac{e - e}{1} = 0 Res ( f , 1 ) = d z d [ ( z − 1 ) 2 ⋅ z ( z − 1 ) 2 e z ] z = 1 = d z d [ z e z ] z = 1 = z 2 e z ⋅ z − e z z = 1 = 1 e − e = 0 .
∫ γ f d z = 2 π i ( 1 + 0 ) = 2 π i \int_\gamma f\, dz = 2\pi i(1 + 0) = 2\pi i ∫ γ f d z = 2 π i ( 1 + 0 ) = 2 π i . ■ \blacksquare ■
Problem 2. Evaluate ∫ γ 1 z 4 + 1 d z \int_\gamma \frac{1}{z^4 + 1}\, dz ∫ γ z 4 + 1 1 d z where γ \gamma γ is ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
Solution. The poles are the fourth roots of − 1 -1 − 1 : z k = e i π / 4 + i k π / 2 z_k = e^{i\pi/4 + ik\pi/2} z k = e iπ /4 + ik π /2 for k = 0 , 1 , 2 , 3 k = 0, 1, 2, 3 k = 0 , 1 , 2 , 3 . All four lie inside ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
Each is a simple pole with R e s ( f , z k ) = 1 4 z k 3 \mathrm{Res}(f, z_k) = \frac{1}{4z_k^3} Res ( f , z k ) = 4 z k 3 1 . Since z k 4 = − 1 z_k^4 = -1 z k 4 = − 1 : z k − 3 = − z k z_k^{-3} = -z_k z k − 3 = − z k So the sum equals − 1 4 ∑ z k = 0 -\frac{1}{4}\sum z_k = 0 − 4 1 ∑ z k = 0 .
∫ γ d z z 4 + 1 = 2 π i ⋅ 0 = 0 \int_\gamma \frac{dz}{z^4 + 1} = 2\pi i \cdot 0 = 0 ∫ γ z 4 + 1 d z = 2 π i ⋅ 0 = 0 . ■ \blacksquare ■
Contour integration is a powerful tool for evaluating definite integrals.
Theorem 9.1. If f ( x ) = P ( x ) / Q ( x ) f(x) = P(x)/Q(x) f ( x ) = P ( x ) / Q ( x ) where deg ( Q ) ≥ deg ( P ) + 2 \deg(Q) \geq \deg(P) + 2 deg ( Q ) ≥ deg ( P ) + 2 and Q Q Q has no real roots, Then
∫ − ∞ ∞ f ( x ) d x = 2 π i ∑ I m ( z k ) > 0 R e s ( f , z k ) \int_{-\infty}^{\infty} f(x)\, dx = 2\pi i \sum_{\mathrm{Im}(z_k) > 0} \mathrm{Res}(f, z_k) ∫ − ∞ ∞ f ( x ) d x = 2 π i ∑ Im ( z k ) > 0 Res ( f , z k )
Where the sum is over poles in the upper half-plane.
Proof. Integrate f ( z ) f(z) f ( z ) over the semicircular contour γ R \gamma_R γ R consisting of [ − R , R ] [-R, R] [ − R , R ] on the Real axis and the semicircle ∣ z ∣ = R |z| = R ∣ z ∣ = R in the upper half-plane. As R → ∞ R \to \infty R → ∞ The integral over The semicircle vanishes (since ∣ f ( z ) ∣ ≤ M / R 2 |f(z)| \leq M/R^2 ∣ f ( z ) ∣ ≤ M / R 2 and the length is π R \pi R π R ). ■ \blacksquare ■
Problem. Evaluate ∫ − ∞ ∞ d x x 2 + 1 \int_{-\infty}^{\infty} \frac{dx}{x^2 + 1} ∫ − ∞ ∞ x 2 + 1 d x .
Solution. f ( z ) = 1 z 2 + 1 f(z) = \frac{1}{z^2 + 1} f ( z ) = z 2 + 1 1 has simple poles at z = ± i z = \pm i z = ± i .
Only z = i z = i z = i is in the upper half-plane.
R e s ( 1 z 2 + 1 , i ) = 1 2 z ∣ z = i = 1 2 i \mathrm{Res}\left(\frac{1}{z^2 + 1}, i\right) = \frac{1}{2z}\Big|_{z = i} = \frac{1}{2i} Res ( z 2 + 1 1 , i ) = 2 z 1 z = i = 2 i 1 .
∫ − ∞ ∞ d x x 2 + 1 = 2 π i ⋅ 1 2 i = π \int_{-\infty}^{\infty} \frac{dx}{x^2 + 1} = 2\pi i \cdot \frac{1}{2i} = \pi ∫ − ∞ ∞ x 2 + 1 d x = 2 π i ⋅ 2 i 1 = π . ■ \blacksquare ■
For integrals of the form ∫ 0 2 π R ( cos θ , sin θ ) d θ \int_0^{2\pi} R(\cos\theta, \sin\theta)\, d\theta ∫ 0 2 π R ( cos θ , sin θ ) d θ Substitute z = e i θ z = e^{i\theta} z = e i θ So d z = i z d θ dz = iz\, d\theta d z = i z d θ , cos θ = z + z − 1 2 \cos\theta = \frac{z + z^{-1}}{2} cos θ = 2 z + z − 1 sin θ = z − z − 1 2 i \sin\theta = \frac{z - z^{-1}}{2i} sin θ = 2 i z − z − 1 .
The integral becomes ∫ ∣ z ∣ = 1 f ( z ) d z \int_{|z|=1} f(z)\, dz ∫ ∣ z ∣ = 1 f ( z ) d z where f ( z ) f(z) f ( z ) is a rational function.
Problem. Evaluate ∫ 0 2 π d θ 2 + cos θ \int_0^{2\pi} \frac{d\theta}{2 + \cos\theta} ∫ 0 2 π 2 + c o s θ d θ .
Solution. Substitute z = e i θ z = e^{i\theta} z = e i θ : d θ = d z i z d\theta = \frac{dz}{iz} d θ = i z d z cos θ = z + 1 / z 2 \cos\theta = \frac{z + 1/z}{2} cos θ = 2 z + 1/ z .
∫ ∣ z ∣ = 1 d z i z ( 2 + z + 1 / z 2 ) = ∫ ∣ z ∣ = 1 2 d z i ( z 2 + 4 z + 1 ) \int_{|z|=1} \frac{dz}{iz\left(2 + \frac{z + 1/z}{2}\right)} = \int_{|z|=1} \frac{2\, dz}{i(z^2 + 4z + 1)} ∫ ∣ z ∣ = 1 i z ( 2 + 2 z + 1/ z ) d z = ∫ ∣ z ∣ = 1 i ( z 2 + 4 z + 1 ) 2 d z
Poles: z 2 + 4 z + 1 = 0 ⇒ z = − 2 ± 3 z^2 + 4z + 1 = 0 \Rightarrow z = -2 \pm \sqrt{3} z 2 + 4 z + 1 = 0 ⇒ z = − 2 ± 3 .
∣ z 1 ∣ = ∣ − 2 + 3 ∣ = 2 − 3 < 1 |z_1| = |-2 + \sqrt{3}| = 2 - \sqrt{3} \lt 1 ∣ z 1 ∣ = ∣ − 2 + 3 ∣ = 2 − 3 < 1 (inside). ∣ z 2 ∣ = ∣ − 2 − 3 ∣ = 2 + 3 > 1 |z_2| = |-2 - \sqrt{3}| = 2 + \sqrt{3} \gt 1 ∣ z 2 ∣ = ∣ − 2 − 3 ∣ = 2 + 3 > 1 (outside).
R e s ( 1 z 2 + 4 z + 1 , z 1 ) = 1 2 3 \mathrm{Res}\left(\frac{1}{z^2 + 4z + 1}, z_1\right) = \frac{1}{2\sqrt{3}} Res ( z 2 + 4 z + 1 1 , z 1 ) = 2 3 1 .
∫ 0 2 π d θ 2 + cos θ = 2 i ⋅ 2 π i ⋅ 1 2 3 = 2 π 3 \int_0^{2\pi} \frac{d\theta}{2 + \cos\theta} = \frac{2}{i} \cdot 2\pi i \cdot \frac{1}{2\sqrt{3}} = \frac{2\pi}{\sqrt{3}} ∫ 0 2 π 2 + c o s θ d θ = i 2 ⋅ 2 π i ⋅ 2 3 1 = 3 2 π . ■ \blacksquare ■
Theorem 9.2 (Jordan’s Lemma). If f ( z ) → 0 f(z) \to 0 f ( z ) → 0 uniformly as ∣ z ∣ → ∞ |z| \to \infty ∣ z ∣ → ∞ in the upper Half-plane and a > 0 a \gt 0 a > 0 Then
lim R → ∞ ∫ C R e i a z f ( z ) d z = 0 \lim_{R \to \infty} \int_{C_R} e^{iaz}f(z)\, dz = 0 lim R → ∞ ∫ C R e ia z f ( z ) d z = 0
Where C R C_R C R is the upper semicircle ∣ z ∣ = R |z| = R ∣ z ∣ = R , I m ( z ) ≥ 0 \mathrm{Im}(z) \geq 0 Im ( z ) ≥ 0 .
This allows evaluation of integrals of the form ∫ − ∞ ∞ f ( x ) cos ( a x ) d x \int_{-\infty}^{\infty} f(x)\cos(ax)\, dx ∫ − ∞ ∞ f ( x ) cos ( a x ) d x and ∫ − ∞ ∞ f ( x ) sin ( a x ) d x \int_{-\infty}^{\infty} f(x)\sin(ax)\, dx ∫ − ∞ ∞ f ( x ) sin ( a x ) d x .
Solution Problem. Evaluate ∫ − ∞ ∞ cos x x 2 + 1 d x \int_{-\infty}^{\infty} \frac{\cos x}{x^2 + 1}\, dx ∫ − ∞ ∞ x 2 + 1 c o s x d x .
Consider ∫ − ∞ ∞ e i x x 2 + 1 d x = 2 π i ⋅ R e s ( e i z z 2 + 1 , i ) \int_{-\infty}^{\infty} \frac{e^{ix}}{x^2 + 1}\, dx = 2\pi i \cdot \mathrm{Res}\!\left(\frac{e^{iz}}{z^2+1}, i\right) ∫ − ∞ ∞ x 2 + 1 e i x d x = 2 π i ⋅ Res ( z 2 + 1 e i z , i ) .
R e s ( e i z z 2 + 1 , i ) = e i ⋅ i 2 i = e − 1 2 i \mathrm{Res}\!\left(\frac{e^{iz}}{z^2+1}, i\right) = \frac{e^{i \cdot i}}{2i} = \frac{e^{-1}}{2i} Res ( z 2 + 1 e i z , i ) = 2 i e i ⋅ i = 2 i e − 1 .
∫ − ∞ ∞ e i x x 2 + 1 d x = 2 π i ⋅ e − 1 2 i = π e \int_{-\infty}^{\infty} \frac{e^{ix}}{x^2 + 1}\, dx = 2\pi i \cdot \frac{e^{-1}}{2i} = \frac{\pi}{e} ∫ − ∞ ∞ x 2 + 1 e i x d x = 2 π i ⋅ 2 i e − 1 = e π .
Taking real parts: ∫ − ∞ ∞ cos x x 2 + 1 d x = π e \int_{-\infty}^{\infty} \frac{\cos x}{x^2 + 1}\, dx = \frac{\pi}{e} ∫ − ∞ ∞ x 2 + 1 c o s x d x = e π .
Problem. Evaluate ∫ − ∞ ∞ x sin x x 2 + a 2 d x \int_{-\infty}^{\infty} \frac{x \sin x}{x^2 + a^2}\, dx ∫ − ∞ ∞ x 2 + a 2 x s i n x d x for a > 0 a \gt 0 a > 0 .
Consider ∫ − ∞ ∞ z e i z z 2 + a 2 d z \int_{-\infty}^{\infty} \frac{z\, e^{iz}}{z^2 + a^2}\, dz ∫ − ∞ ∞ z 2 + a 2 z e i z d z . Only z = i a z = ia z = ia is in the upper half-plane.
R e s ( z e i z z 2 + a 2 , i a ) = i a ⋅ e i ⋅ i a 2 i a = e − a 2 \mathrm{Res}\!\left(\frac{ze^{iz}}{z^2 + a^2}, ia\right) = \frac{ia \cdot e^{i \cdot ia}}{2ia} = \frac{e^{-a}}{2} Res ( z 2 + a 2 z e i z , ia ) = 2 ia ia ⋅ e i ⋅ ia = 2 e − a .
∫ − ∞ ∞ x e i x x 2 + a 2 d x = 2 π i ⋅ e − a 2 = π i e − a \int_{-\infty}^{\infty} \frac{x\, e^{ix}}{x^2 + a^2}\, dx = 2\pi i \cdot \frac{e^{-a}}{2} = \pi i\, e^{-a} ∫ − ∞ ∞ x 2 + a 2 x e i x d x = 2 π i ⋅ 2 e − a = π i e − a .
Taking imaginary parts: ∫ − ∞ ∞ x sin x x 2 + a 2 d x = π e − a \int_{-\infty}^{\infty} \frac{x \sin x}{x^2 + a^2}\, dx = \pi\, e^{-a} ∫ − ∞ ∞ x 2 + a 2 x s i n x d x = π e − a .
Problem. Evaluate ∫ 0 2 π cos 2 θ 5 + 4 cos θ d θ \int_0^{2\pi} \frac{\cos 2\theta}{5 + 4\cos\theta}\, d\theta ∫ 0 2 π 5 + 4 c o s θ c o s 2 θ d θ .
Substitute z = e i θ z = e^{i\theta} z = e i θ : cos θ = ( z + z − 1 ) / 2 \cos\theta = (z + z^{-1})/2 cos θ = ( z + z − 1 ) /2 , cos 2 θ = ( z 2 + z − 2 ) / 2 \cos 2\theta = (z^2 + z^{-2})/2 cos 2 θ = ( z 2 + z − 2 ) /2 .
I = 1 2 i ∫ ∣ z ∣ = 1 z 4 + 1 z 2 ( 2 z + 1 ) ( z + 2 ) d z I = \frac{1}{2i}\int_{|z|=1} \frac{z^4 + 1}{z^2(2z + 1)(z + 2)}\, dz I = 2 i 1 ∫ ∣ z ∣ = 1 z 2 ( 2 z + 1 ) ( z + 2 ) z 4 + 1 d z .
Poles inside ∣ z ∣ = 1 |z| = 1 ∣ z ∣ = 1 : z = 0 z = 0 z = 0 (order 2 2 2 ) and z = − 1 / 2 z = -1/2 z = − 1/2 (simple).
At z = 0 z = 0 z = 0 : R e s = d d z [ z 4 + 1 ( 2 z + 1 ) ( z + 2 ) ] z = 0 = − 5 4 \mathrm{Res} = \frac{d}{dz}\left[\frac{z^4 + 1}{(2z+1)(z+2)}\right]_{z=0} = -\frac{5}{4} Res = d z d [ ( 2 z + 1 ) ( z + 2 ) z 4 + 1 ] z = 0 = − 4 5 .
At z = − 1 / 2 z = -1/2 z = − 1/2 : R e s = 17 / 16 3 / 4 = 17 12 \mathrm{Res} = \frac{17/16}{3/4} = \frac{17}{12} Res = 3/4 17/16 = 12 17 .
I = 1 2 i ⋅ 2 π i ( − 5 4 + 17 12 ) = π 6 I = \frac{1}{2i} \cdot 2\pi i \left(-\frac{5}{4} + \frac{17}{12}\right) = \frac{\pi}{6} I = 2 i 1 ⋅ 2 π i ( − 4 5 + 12 17 ) = 6 π .
For integrals where the integrand has poles on the real axis, we use the Cauchy principal value :
P V ∫ − ∞ ∞ f ( x ) d x = lim ε → 0 + ( ∫ − ∞ a − ε f ( x ) d x + ∫ a + ε ∞ f ( x ) d x ) \mathrm{PV}\!\int_{-\infty}^{\infty} f(x)\, dx = \lim_{\varepsilon \to 0^+} \left(\int_{-\infty}^{a-\varepsilon} f(x)\, dx + \int_{a+\varepsilon}^{\infty} f(x)\, dx\right) PV ∫ − ∞ ∞ f ( x ) d x = lim ε → 0 + ( ∫ − ∞ a − ε f ( x ) d x + ∫ a + ε ∞ f ( x ) d x )
Solution Problem. Evaluate P V ∫ − ∞ ∞ sin x x d x \mathrm{PV}\!\int_{-\infty}^{\infty} \frac{\sin x}{x}\, dx PV ∫ − ∞ ∞ x s i n x d x .
Consider ∮ γ e i z z d z \oint_\gamma \frac{e^{iz}}{z}\, dz ∮ γ z e i z d z where γ \gamma γ consists of [ − R , − ε ] [-R, -\varepsilon] [ − R , − ε ] [ ε , R ] [\varepsilon, R] [ ε , R ] on the real axis, small upper semicircle C ε C_\varepsilon C ε around 0 0 0 And large Upper semicircle C R C_R C R .
No poles inside the contour, so the integral is 0 0 0 .
On C R C_R C R : vanishes as R → ∞ R \to \infty R → ∞ by Jordan’s lemma. On C ε C_\varepsilon C ε (indenting above): ∫ C ε e i z z d z → − i π \int_{C_\varepsilon} \frac{e^{iz}}{z}\, dz \to -i\pi ∫ C ε z e i z d z → − iπ as ε → 0 \varepsilon \to 0 ε → 0 (half residue contribution).
0 = P V ∫ − ∞ ∞ e i x x d x + ( − i π ) 0 = \mathrm{PV}\!\int_{-\infty}^{\infty} \frac{e^{ix}}{x}\, dx + (-i\pi) 0 = PV ∫ − ∞ ∞ x e i x d x + ( − iπ ) .
P V ∫ − ∞ ∞ e i x x d x = i π \mathrm{PV}\!\int_{-\infty}^{\infty} \frac{e^{ix}}{x}\, dx = i\pi PV ∫ − ∞ ∞ x e i x d x = iπ .
Taking imaginary parts: P V ∫ − ∞ ∞ sin x x d x = π \mathrm{PV}\!\int_{-\infty}^{\infty} \frac{\sin x}{x}\, dx = \pi PV ∫ − ∞ ∞ x s i n x d x = π .
Definition. An analytic function f f f is conformal at z 0 z_0 z 0 if f ′ ( z 0 ) ≠ 0 f'(z_0) \neq 0 f ′ ( z 0 ) = 0 . A conformal Mapping preserves angles (both magnitude and orientation) between curves.
If f ′ ( z 0 ) = r e i θ f'(z_0) = re^{i\theta} f ′ ( z 0 ) = r e i θ Then near z 0 z_0 z 0 the mapping f f f acts as a rotation by θ \theta θ followed By a scaling by r r r . The Jacobian determinant is ∣ f ′ ( z 0 ) ∣ 2 > 0 |f'(z_0)|^2 \gt 0 ∣ f ′ ( z 0 ) ∣ 2 > 0 So orientation is preserved.
Mapping Effect w = a z + b w = az + b w = a z + b (a ≠ 0 a \neq 0 a = 0 )Translation, rotation, scaling w = 1 / z w = 1/z w = 1/ z Inversion in the unit circle w = z 2 w = z^2 w = z 2 Squaring (doubles angles) w = e z w = e^z w = e z Exponential (maps strips to sectors) w = z − a 1 − a ˉ z w = \frac{z - a}{1 - \bar{a}z} w = 1 − a ˉ z z − a Möbius (maps disk to disk)
A Möbius transformation (or linear fractional transformation) is
T ( z ) = a z + b c z + d , a d − b c ≠ 0 T(z) = \frac{az + b}{cz + d}, \quad ad - bc \neq 0 T ( z ) = cz + d a z + b , a d − b c = 0
Proposition 10.1. Möbius transformations are conformal (where defined) and map circles and lines To circles and lines.
Proposition 10.2. Three points determine a unique Möbius transformation: T ( z 1 ) = w 1 T(z_1) = w_1 T ( z 1 ) = w 1 T ( z 2 ) = w 2 T(z_2) = w_2 T ( z 2 ) = w 2 , T ( z 3 ) = w 3 T(z_3) = w_3 T ( z 3 ) = w 3 .
Definition. The cross-ratio of four distinct points z 1 , z 2 , z 3 , z 4 z_1, z_2, z_3, z_4 z 1 , z 2 , z 3 , z 4 is
( z 1 , z 2 , z 3 , z 4 ) = ( z 1 − z 3 ) ( z 2 − z 4 ) ( z 1 − z 4 ) ( z 2 − z 3 ) (z_1, z_2, z_3, z_4) = \frac{(z_1 - z_3)(z_2 - z_4)}{(z_1 - z_4)(z_2 - z_3)} ( z 1 , z 2 , z 3 , z 4 ) = ( z 1 − z 4 ) ( z 2 − z 3 ) ( z 1 − z 3 ) ( z 2 − z 4 )
Proposition 10.3. The cross-ratio is invariant under Möbius transformations: ( T z 1 , T z 2 , T z 3 , T z 4 ) = ( z 1 , z 2 , z 3 , z 4 ) (Tz_1, Tz_2, Tz_3, Tz_4) = (z_1, z_2, z_3, z_4) ( T z 1 , T z 2 , T z 3 , T z 4 ) = ( z 1 , z 2 , z 3 , z 4 ) .
Proposition 10.4. The unique Möbius transformation sending z 1 ↦ 0 z_1 \mapsto 0 z 1 ↦ 0 , z 2 ↦ 1 z_2 \mapsto 1 z 2 ↦ 1 z 3 ↦ ∞ z_3 \mapsto \infty z 3 ↦ ∞ is
T ( z ) = ( z − z 1 ) ( z 2 − z 3 ) ( z − z 3 ) ( z 2 − z 1 ) T(z) = \frac{(z - z_1)(z_2 - z_3)}{(z - z_3)(z_2 - z_1)} T ( z ) = ( z − z 3 ) ( z 2 − z 1 ) ( z − z 1 ) ( z 2 − z 3 )
A Möbius transformation T ( z ) = a z + b c z + d T(z) = \frac{az + b}{cz + d} T ( z ) = cz + d a z + b is classified by its fixed points (solutions of T ( z ) = z T(z) = z T ( z ) = z ).
Parabolic: Exactly one fixed point. Conjugate to w = z + k w = z + k w = z + k .Elliptic: Two fixed points, ∣ T ′ ( z 0 ) ∣ = 1 |T'(z_0)| = 1 ∣ T ′ ( z 0 ) ∣ = 1 . Conjugate to a rotation w = e i θ z w = e^{i\theta} z w = e i θ z .Hyperbolic: Two fixed points, T ′ ( z 0 ) ∈ R + T'(z_0) \in \mathbb{R}^+ T ′ ( z 0 ) ∈ R + , T ′ ( z 0 ) ≠ 1 T'(z_0) \neq 1 T ′ ( z 0 ) = 1 . Conjugate to w = k z w = kz w = k z .Loxodromic: Two fixed points, T ′ ( z 0 ) ∉ R ∪ { z : ∣ z ∣ = 1 } T'(z_0) \notin \mathbb{R} \cup \{z : |z| = 1\} T ′ ( z 0 ) ∈ / R ∪ { z : ∣ z ∣ = 1 } . Conjugate to w = k e i θ z w = ke^{i\theta}z w = k e i θ z .Solution Problem. Find the Möbius transformation mapping 0 ↦ i 0 \mapsto i 0 ↦ i , 1 ↦ 0 1 \mapsto 0 1 ↦ 0 , ∞ ↦ − i \infty \mapsto -i ∞ ↦ − i .
T ( z ) = a z + b c z + d T(z) = \frac{az + b}{cz + d} T ( z ) = cz + d a z + b with T ( 0 ) = i ⇒ b / d = i ⇒ b = i d T(0) = i \Rightarrow b/d = i \Rightarrow b = id T ( 0 ) = i ⇒ b / d = i ⇒ b = i d . T ( 1 ) = 0 ⇒ a = − b = − i d T(1) = 0 \Rightarrow a = -b = -id T ( 1 ) = 0 ⇒ a = − b = − i d . T ( ∞ ) = − i ⇒ a / c = − i ⇒ c = d T(\infty) = -i \Rightarrow a/c = -i \Rightarrow c = d T ( ∞ ) = − i ⇒ a / c = − i ⇒ c = d .
T ( z ) = − i d z + i d d z + d = i ( 1 − z ) z + 1 T(z) = \frac{-idz + id}{dz + d} = \frac{i(1 - z)}{z + 1} T ( z ) = d z + d − i d z + i d = z + 1 i ( 1 − z ) .
Problem. Show that T ( z ) = z − 1 z + 1 T(z) = \frac{z - 1}{z + 1} T ( z ) = z + 1 z − 1 maps the right half-plane to the unit disk.
If R e ( z ) > 0 \mathrm{Re}(z) \gt 0 Re ( z ) > 0 Then ∣ z − 1 ∣ < ∣ z + 1 ∣ |z - 1| \lt |z + 1| ∣ z − 1∣ < ∣ z + 1∣ So ∣ T ( z ) ∣ < 1 |T(z)| \lt 1 ∣ T ( z ) ∣ < 1 .
Check boundary: T ( i ) = i − 1 i + 1 = ( i − 1 ) ( − i + 1 ) ( i + 1 ) ( − i + 1 ) = 2 2 = 1 T(i) = \frac{i - 1}{i + 1} = \frac{(i-1)(-i+1)}{(i+1)(-i+1)} = \frac{2}{2} = 1 T ( i ) = i + 1 i − 1 = ( i + 1 ) ( − i + 1 ) ( i − 1 ) ( − i + 1 ) = 2 2 = 1 . ∣ T ( i ) ∣ = 1 |T(i)| = 1 ∣ T ( i ) ∣ = 1 . ✓ \checkmark ✓
Problem. Classify T ( z ) = 2 z + 1 z + 2 T(z) = \frac{2z + 1}{z + 2} T ( z ) = z + 2 2 z + 1 .
Fixed points: z = 2 z + 1 z + 2 ⇒ z 2 = 1 ⇒ z = ± 1 z = \frac{2z + 1}{z + 2} \Rightarrow z^2 = 1 \Rightarrow z = \pm 1 z = z + 2 2 z + 1 ⇒ z 2 = 1 ⇒ z = ± 1 .
T ′ ( z ) = 3 ( z + 2 ) 2 T'(z) = \frac{3}{(z + 2)^2} T ′ ( z ) = ( z + 2 ) 2 3 . T ′ ( 1 ) = 1 / 3 T'(1) = 1/3 T ′ ( 1 ) = 1/3 , T ′ ( − 1 ) = 3 T'(-1) = 3 T ′ ( − 1 ) = 3 .
Both multipliers are real and positive (not equal to 1 1 1 ), so T T T is hyperbolic.
Theorem 10.5 (Riemann Mapping Theorem). Let U U U be a connected open proper subset of C \mathbb{C} C . Then there exists a bijective conformal map from U U U onto the unit disk D = { z : ∣ z ∣ < 1 } \mathbb{D} = \{z : |z| \lt 1\} D = { z : ∣ z ∣ < 1 } .
This is one of the most profound results in complex analysis, establishing that all connected Domains (other than C \mathbb{C} C itself) are conformally equivalent.
Remark. The Riemann mapping theorem is an existence theorem; it does not provide an explicit Formula for the conformal map .
Theorem 11.1 (Liouville’s Theorem). Every bounded entire function is constant.
Proof. If ∣ f ( z ) ∣ ≤ M |f(z)| \leq M ∣ f ( z ) ∣ ≤ M for all z z z Then by Cauchy’s estimates with R R R arbitrarily large: ∣ f ′ ( z 0 ) ∣ ≤ M R → 0 |f'(z_0)| \leq \frac{M}{R} \to 0 ∣ f ′ ( z 0 ) ∣ ≤ R M → 0 as R → ∞ R \to \infty R → ∞ . So f ′ ( z ) = 0 f'(z) = 0 f ′ ( z ) = 0 for all z z z Meaning f f f is Constant. ■ \blacksquare ■
Theorem 11.2 (Fundamental Theorem of Algebra). Every non-constant polynomial p ( z ) ∈ C [ z ] p(z) \in \mathbb{C}[z] p ( z ) ∈ C [ z ] has a root in C \mathbb{C} C .
Proof. Suppose p ( z ) p(z) p ( z ) has no root. Then f ( z ) = 1 / p ( z ) f(z) = 1/p(z) f ( z ) = 1/ p ( z ) is entire. Since ∣ p ( z ) ∣ → ∞ |p(z)| \to \infty ∣ p ( z ) ∣ → ∞ as ∣ z ∣ → ∞ |z| \to \infty ∣ z ∣ → ∞ , f ( z ) → 0 f(z) \to 0 f ( z ) → 0 So f f f is bounded. By Liouville’s theorem, f f f is constant, so p p p Is constant, a contradiction. ■ \blacksquare ■
Theorem 11.3 (Maximum Modulus Principle). If f f f is analytic and non-constant on a domain D D D Then ∣ f ∣ |f| ∣ f ∣ has no local maximum in D D D .
Corollary 11.4. If f f f is analytic on a bounded domain D D D and continuous on D ˉ = D ∪ ∂ D \bar{D} = D \cup \partial D D ˉ = D ∪ ∂ D Then ∣ f ∣ |f| ∣ f ∣ attains its maximum on ∂ D \partial D ∂ D .
Theorem 11.5 (Minimum Modulus Principle). If f f f is analytic and non-zero on a bounded domain D D D And continuous on D ˉ \bar{D} D ˉ Then ∣ f ∣ |f| ∣ f ∣ attains its minimum on ∂ D \partial D ∂ D .
Remark. If f f f has zeros in D D D Then ∣ f ∣ |f| ∣ f ∣ attains its minimum of 0 0 0 at those zeros. The minimum modulus principle requires the non-vanishing hypothesis.
Theorem 11.6 (Schwarz Lemma). If f : D → D f : \mathbb{D} \to \mathbb{D} f : D → D is analytic with f ( 0 ) = 0 f(0) = 0 f ( 0 ) = 0 Then
∣ f ( z ) ∣ ≤ ∣ z ∣ f o r a l l z ∈ D |f(z)| \leq |z| \quad \mathrm{for\ all\ } z \in \mathbb{D} ∣ f ( z ) ∣ ≤ ∣ z ∣ for all z ∈ D
And ∣ f ′ ( 0 ) ∣ ≤ 1 |f'(0)| \leq 1 ∣ f ′ ( 0 ) ∣ ≤ 1 . Equality in either case implies f ( z ) = e i θ z f(z) = e^{i\theta} z f ( z ) = e i θ z for some real θ \theta θ .
Proof. Define g ( z ) = f ( z ) / z g(z) = f(z)/z g ( z ) = f ( z ) / z for z ≠ 0 z \neq 0 z = 0 and g ( 0 ) = f ′ ( 0 ) g(0) = f'(0) g ( 0 ) = f ′ ( 0 ) . Then g g g is analytic on D \mathbb{D} D . For ∣ z ∣ = r < 1 |z| = r \lt 1 ∣ z ∣ = r < 1 : ∣ g ( z ) ∣ = ∣ f ( z ) ∣ / ∣ z ∣ ≤ 1 / r |g(z)| = |f(z)|/|z| \leq 1/r ∣ g ( z ) ∣ = ∣ f ( z ) ∣/∣ z ∣ ≤ 1/ r . By the maximum modulus Principle, ∣ g ( z ) ∣ ≤ 1 / r |g(z)| \leq 1/r ∣ g ( z ) ∣ ≤ 1/ r for ∣ z ∣ ≤ r |z| \leq r ∣ z ∣ ≤ r . Letting r → 1 r \to 1 r → 1 : ∣ g ( z ) ∣ ≤ 1 |g(z)| \leq 1 ∣ g ( z ) ∣ ≤ 1 So ∣ f ( z ) ∣ ≤ ∣ z ∣ |f(z)| \leq |z| ∣ f ( z ) ∣ ≤ ∣ z ∣ . Also ∣ f ′ ( 0 ) ∣ = ∣ g ( 0 ) ∣ ≤ 1 |f'(0)| = |g(0)| \leq 1 ∣ f ′ ( 0 ) ∣ = ∣ g ( 0 ) ∣ ≤ 1 . If ∣ f ′ ( 0 ) ∣ = 1 |f'(0)| = 1 ∣ f ′ ( 0 ) ∣ = 1 Then ∣ g ∣ |g| ∣ g ∣ attains its maximum At an interior point, so g g g is constant: g ( z ) = e i θ g(z) = e^{i\theta} g ( z ) = e i θ . ■ \blacksquare ■
Theorem 12.1 (Argument Principle). If f f f is meromorphic inside and on a simple closed contour γ \gamma γ with no zeros or poles on γ \gamma γ Then
1 2 π i ∫ γ f ′ ( z ) f ( z ) d z = N − P \frac{1}{2\pi i}\int_\gamma \frac{f'(z)}{f(z)}\, dz = N - P 2 π i 1 ∫ γ f ( z ) f ′ ( z ) d z = N − P
Where N N N is the number of zeros and P P P is the number of poles of f f f inside γ \gamma γ (counting Multiplicities).
Theorem 12.2 (Rouché’s Theorem). If f f f and g g g are analytic inside and on a simple closed Contour γ \gamma γ And ∣ f ( z ) ∣ > ∣ g ( z ) ∣ |f(z)| \gt |g(z)| ∣ f ( z ) ∣ > ∣ g ( z ) ∣ on γ \gamma γ Then f f f and f + g f + g f + g have the same number of Zeros inside γ \gamma γ .
Proof. On γ \gamma γ : ∣ g ( z ) / f ( z ) ∣ < 1 |g(z)/f(z)| \lt 1 ∣ g ( z ) / f ( z ) ∣ < 1 . The function h ( z ) = 1 + g ( z ) / f ( z ) h(z) = 1 + g(z)/f(z) h ( z ) = 1 + g ( z ) / f ( z ) satisfies ∣ h ( z ) − 1 ∣ < 1 |h(z) - 1| \lt 1 ∣ h ( z ) − 1∣ < 1 on γ \gamma γ So h ( γ ) h(\gamma) h ( γ ) does not wind around 0 0 0 . By the argument principle Applied to h h h : 0 = N h − P h 0 = N_h - P_h 0 = N h − P h Meaning h h h has the same number of zeros and poles inside γ \gamma γ . But h = ( f + g ) / f h = (f + g)/f h = ( f + g ) / f So zeros of h h h are zeros of f + g f + g f + g and poles of h h h are zeros of f f f . Therefore f f f and f + g f + g f + g have the same number of zeros. ■ \blacksquare ■
Problem. Show that z 4 + 6 z + 3 z^4 + 6z + 3 z 4 + 6 z + 3 has exactly one root in ∣ z ∣ < 1 |z| \lt 1 ∣ z ∣ < 1 .
Solution. On ∣ z ∣ = 1 |z| = 1 ∣ z ∣ = 1 : ∣ 6 z ∣ = 6 > ∣ z 4 + 3 ∣ ≤ ∣ z ∣ 4 + 3 = 4 |6z| = 6 \gt |z^4 + 3| \leq |z|^4 + 3 = 4 ∣6 z ∣ = 6 > ∣ z 4 + 3∣ ≤ ∣ z ∣ 4 + 3 = 4 . By Rouché’s theorem with f ( z ) = 6 z f(z) = 6z f ( z ) = 6 z and g ( z ) = z 4 + 3 g(z) = z^4 + 3 g ( z ) = z 4 + 3 : f + g = z 4 + 6 z + 3 f + g = z^4 + 6z + 3 f + g = z 4 + 6 z + 3 has the same number of zeros in ∣ z ∣ < 1 |z| \lt 1 ∣ z ∣ < 1 as f ( z ) = 6 z f(z) = 6z f ( z ) = 6 z Which has exactly one zero (at z = 0 z = 0 z = 0 ). ■ \blacksquare ■
Solution Problem. Show that all roots of z 4 + z + 1 = 0 z^4 + z + 1 = 0 z 4 + z + 1 = 0 satisfy ∣ z ∣ < 2 |z| \lt 2 ∣ z ∣ < 2 .
On ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 : ∣ z 4 ∣ = 16 > ∣ z + 1 ∣ ≤ 3 |z^4| = 16 \gt |z + 1| \leq 3 ∣ z 4 ∣ = 16 > ∣ z + 1∣ ≤ 3 . By Rouché with f ( z ) = z 4 f(z) = z^4 f ( z ) = z 4 and g ( z ) = z + 1 g(z) = z + 1 g ( z ) = z + 1 : z 4 + z + 1 z^4 + z + 1 z 4 + z + 1 has 4 4 4 zeros in ∣ z ∣ < 2 |z| \lt 2 ∣ z ∣ < 2 (same as z 4 z^4 z 4 ).
Problem. Show that z 5 + 3 z 2 + 1 z^5 + 3z^2 + 1 z 5 + 3 z 2 + 1 has exactly two roots in ∣ z ∣ < 1 |z| \lt 1 ∣ z ∣ < 1 .
On ∣ z ∣ = 1 |z| = 1 ∣ z ∣ = 1 : ∣ 3 z 2 + 1 ∣ ≥ ∣ 3 z 2 ∣ − ∣ 1 ∣ = 2 > ∣ z 5 ∣ = 1 |3z^2 + 1| \geq |3z^2| - |1| = 2 \gt |z^5| = 1 ∣3 z 2 + 1∣ ≥ ∣3 z 2 ∣ − ∣1∣ = 2 > ∣ z 5 ∣ = 1 . By Rouché with f ( z ) = 3 z 2 + 1 f(z) = 3z^2 + 1 f ( z ) = 3 z 2 + 1 and g ( z ) = z 5 g(z) = z^5 g ( z ) = z 5 : z 5 + 3 z 2 + 1 z^5 + 3z^2 + 1 z 5 + 3 z 2 + 1 has the same number of zeros as 3 z 2 + 1 3z^2 + 1 3 z 2 + 1 in ∣ z ∣ < 1 |z| \lt 1 ∣ z ∣ < 1 . 3 z 2 + 1 = 0 ⇒ z = ± i / 3 3z^2 + 1 = 0 \Rightarrow z = \pm i/\sqrt{3} 3 z 2 + 1 = 0 ⇒ z = ± i / 3 Both in ∣ z ∣ < 1 |z| \lt 1 ∣ z ∣ < 1 . So 2 2 2 zeros.
Definition. If f 1 f_1 f 1 is analytic on D 1 D_1 D 1 and f 2 f_2 f 2 is analytic on D 2 D_2 D 2 with D 1 ∩ D 2 ≠ ∅ D_1 \cap D_2 \neq \emptyset D 1 ∩ D 2 = ∅ and f 1 = f 2 f_1 = f_2 f 1 = f 2 on D 1 ∩ D 2 D_1 \cap D_2 D 1 ∩ D 2 Then f 2 f_2 f 2 is an analytic Continuation of f 1 f_1 f 1 .
Theorem 13.1 (Identity Theorem). If f f f and g g g are analytic on a domain D D D and agree on a set With a limit point in D D D Then f = g f = g f = g on all of D D D .
Proof. Let E = { z ∈ D : f ( n ) ( z ) = g ( n ) ( z ) f o r a l l n ≥ 0 } E = \{z \in D : f^{(n)}(z) = g^{(n)}(z) \mathrm{\ for\ all\ } n \geq 0\} E = { z ∈ D : f ( n ) ( z ) = g ( n ) ( z ) for all n ≥ 0 } . E E E is Non-empty (it contains the limit point by continuity of derivatives). E E E is closed (by continuity). If z 0 ∈ E z_0 \in E z 0 ∈ E The Taylor series of f f f and g g g at z 0 z_0 z 0 coincide, so f = g f = g f = g in a neighbourhood of z 0 z_0 z 0 Giving E E E open. Since D D D is connected, E = D E = D E = D . ■ \blacksquare ■
:::caution Common Pitfall The Cauchy-Riemann equations are necessary but not sufficient for Differentiability. The partial derivatives must also be continuous. For example, f ( z ) = exp ( − 1 / z 4 ) f(z) = \exp(-1/z^4) f ( z ) = exp ( − 1/ z 4 ) extended by f ( 0 ) = 0 f(0) = 0 f ( 0 ) = 0 satisfies the Cauchy-Riemann equations at the origin But is not differentiable there. :::
:::caution Common Pitfall Cauchy’s theorem requires a connected domain. On a multiply Connected domain, the integral of an analytic function around a closed contour may be non-zero. The Classic example is ∮ ∣ z ∣ = 1 d z / z = 2 π i \oint_{|z|=1} dz/z = 2\pi i ∮ ∣ z ∣ = 1 d z / z = 2 π i . :::
:::caution Common Pitfall When computing residues at poles of order m ≥ 2 m \geq 2 m ≥ 2 The formula involves Differentiation. A common error is forgetting the ( m − 1 ) ! (m-1)! ( m − 1 )! in the denominator or differentiating ( z − z 0 ) m f ( z ) (z - z_0)^m f(z) ( z − z 0 ) m f ( z ) the wrong number of times. :::
:::caution Common Pitfall The residue at infinity is R e s ( f , ∞ ) = − R e s ( 1 / z 2 ⋅ f ( 1 / z ) , 0 ) \mathrm{Res}(f, \infty) = -\mathrm{Res}(1/z^2 \cdot f(1/z), 0) Res ( f , ∞ ) = − Res ( 1/ z 2 ⋅ f ( 1/ z ) , 0 ) . It is NOT f ( ∞ ) f(\infty) f ( ∞ ) . For A function that is analytic everywhere in the finite plane except for finitely many singularities, The sum of all residues (including the residue at infinity) is zero. :::
:::caution Common Pitfall A conformal mapping preserves angles but not necessarily distances. The Mapping w = z 2 w = z^2 w = z 2 is conformal at every z ≠ 0 z \neq 0 z = 0 But it doubles the angle between curves at each Point. At z = 0 z = 0 z = 0 It is not conformal because f ′ ( 0 ) = 0 f'(0) = 0 f ′ ( 0 ) = 0 . :::
:::caution Common Pitfall The maximum modulus principle says that ∣ f ∣ |f| ∣ f ∣ has no local maximum in the Interior, but the minimum can occur in the interior (e.g., f ( z ) = z f(z) = z f ( z ) = z on the unit disk has minimum ∣ f ∣ = 0 |f| = 0 ∣ f ∣ = 0 at z = 0 z = 0 z = 0 ). For the minimum principle, one needs the additional hypothesis that f f f has No zeros in the domain. :::
:::caution Common Pitfall The complex logarithm is multi-valued. When a problem asks for “logarithm” without specifying a branch, you must either compute all values or explicitly state which Branch you are using. The principal branch L o g z \mathrm{Log}\, z Log z has a branch cut along ( − ∞ , 0 ] (-\infty, 0] ( − ∞ , 0 ] And is undefined on this cut. :::
:::caution Common Pitfall When applying the ML inequality, make sure M M M is a valid upper bound for ∣ f ( z ) ∣ |f(z)| ∣ f ( z ) ∣ on the entire contour. A common error is bounding ∣ f ∣ |f| ∣ f ∣ on only part of the contour. Also, L L L must be the arc length of the contour, not a diameter or radius. :::
Express z = − 3 + i z = -\sqrt{3} + i z = − 3 + i in polar form and find all values of z 1 / 3 z^{1/3} z 1/3 .
Solution ∣ z ∣ = 3 + 1 = 2 |z| = \sqrt{3 + 1} = 2 ∣ z ∣ = 3 + 1 = 2 . Since R e ( z ) < 0 \mathrm{Re}(z) \lt 0 Re ( z ) < 0 and I m ( z ) > 0 \mathrm{Im}(z) \gt 0 Im ( z ) > 0 : arg ( z ) = π − π / 6 = 5 π / 6 \arg(z) = \pi - \pi/6 = 5\pi/6 arg ( z ) = π − π /6 = 5 π /6 .
z = 2 e 5 π i / 6 z = 2\,e^{5\pi i/6} z = 2 e 5 π i /6 .
z 1 / 3 = 2 1 / 3 e ( 5 π / 6 + 2 π k ) / 3 z^{1/3} = 2^{1/3}\, e^{(5\pi/6 + 2\pi k)/3} z 1/3 = 2 1/3 e ( 5 π /6 + 2 π k ) /3 for k = 0 , 1 , 2 k = 0, 1, 2 k = 0 , 1 , 2 .
z 0 = 2 1 / 3 e 5 π i / 18 z_0 = 2^{1/3}\, e^{5\pi i/18} z 0 = 2 1/3 e 5 π i /18 , z 1 = 2 1 / 3 e 17 π i / 18 z_1 = 2^{1/3}\, e^{17\pi i/18} z 1 = 2 1/3 e 17 π i /18 , z 2 = 2 1 / 3 e 29 π i / 18 z_2 = 2^{1/3}\, e^{29\pi i/18} z 2 = 2 1/3 e 29 π i /18 .
If you get this wrong, revise: Section 1.5 (Roots of Complex Numbers).
Let f ( z ) = z 2 + z ˉ 2 f(z) = z^2 + \bar{z}^2 f ( z ) = z 2 + z ˉ 2 . Find where f f f is differentiable and where it is analytic.
Solution f ( z ) = ( x + i y ) 2 + ( x − i y ) 2 = 2 ( x 2 − y 2 ) f(z) = (x + iy)^2 + (x - iy)^2 = 2(x^2 - y^2) f ( z ) = ( x + i y ) 2 + ( x − i y ) 2 = 2 ( x 2 − y 2 ) . So u = 2 ( x 2 − y 2 ) u = 2(x^2 - y^2) u = 2 ( x 2 − y 2 ) , v = 0 v = 0 v = 0 .
u x = 4 x u_x = 4x u x = 4 x , u y = − 4 y u_y = -4y u y = − 4 y , v x = 0 v_x = 0 v x = 0 , v y = 0 v_y = 0 v y = 0 .
CR: 4 x = 0 ⇒ x = 0 4x = 0 \Rightarrow x = 0 4 x = 0 ⇒ x = 0 , − 4 y = 0 ⇒ y = 0 -4y = 0 \Rightarrow y = 0 − 4 y = 0 ⇒ y = 0 .
f f f is differentiable only at z = 0 z = 0 z = 0 and analytic nowhere.
f ′ ( 0 ) = 0 f'(0) = 0 f ′ ( 0 ) = 0 (verified by direct computation).
If you get this wrong, revise: Sections 2.4 and 3.1 (Analyticity and Cauchy-Riemann).
Verify that f ( z ) = 1 z 2 + 1 f(z) = \frac{1}{z^2 + 1} f ( z ) = z 2 + 1 1 satisfies the Cauchy-Riemann equations on its domain and Find f ′ ( z ) f'(z) f ′ ( z ) .
Solution f ( z ) = 1 / ( z 2 + 1 ) f(z) = 1/(z^2 + 1) f ( z ) = 1/ ( z 2 + 1 ) is a rational function with denominator non-zero away from ± i \pm i ± i So f f f Is analytic on C ∖ { i , − i } \mathbb{C} \setminus \{i, -i\} C ∖ { i , − i } .
By the quotient rule: f ′ ( z ) = − 2 z ( z 2 + 1 ) 2 f'(z) = \frac{-2z}{(z^2 + 1)^2} f ′ ( z ) = ( z 2 + 1 ) 2 − 2 z .
Verify via CR at z = 1 z = 1 z = 1 : u = x 2 − y 2 + 1 ( x 2 − y 2 + 1 ) 2 + 4 x 2 y 2 u = \frac{x^2 - y^2 + 1}{(x^2 - y^2 + 1)^2 + 4x^2y^2} u = ( x 2 − y 2 + 1 ) 2 + 4 x 2 y 2 x 2 − y 2 + 1 v = − 2 x y ( x 2 − y 2 + 1 ) 2 + 4 x 2 y 2 v = \frac{-2xy}{(x^2 - y^2 + 1)^2 + 4x^2y^2} v = ( x 2 − y 2 + 1 ) 2 + 4 x 2 y 2 − 2 x y .
u x ( 1 , 0 ) = − 1 / 2 = f ′ ( 1 ) u_x(1, 0) = -1/2 = f'(1) u x ( 1 , 0 ) = − 1/2 = f ′ ( 1 ) . ✓ \checkmark ✓
If you get this wrong, revise: Sections 3.1 and 3.3 (CR Equations).
Show that u ( x , y ) = x 3 − 3 x y 2 + 3 x 2 − 3 y 2 u(x, y) = x^3 - 3xy^2 + 3x^2 - 3y^2 u ( x , y ) = x 3 − 3 x y 2 + 3 x 2 − 3 y 2 is harmonic and find its harmonic conjugate.
Solution u x x = 6 x + 6 u_{xx} = 6x + 6 u xx = 6 x + 6 , u y y = − 6 x − 6 u_{yy} = -6x - 6 u y y = − 6 x − 6 . Δ u = 0 \Delta u = 0 Δ u = 0 . ✓ \checkmark ✓
By CR: v y = u x = 3 x 2 − 3 y 2 + 6 x v_y = u_x = 3x^2 - 3y^2 + 6x v y = u x = 3 x 2 − 3 y 2 + 6 x . v = 3 x 2 y − y 3 + 6 x y + g ( x ) v = 3x^2 y - y^3 + 6xy + g(x) v = 3 x 2 y − y 3 + 6 x y + g ( x ) .
v x = − u y = 6 x y + 6 y v_x = -u_y = 6xy + 6y v x = − u y = 6 x y + 6 y . 6 x y + 6 y = 6 x y + 6 y + g ′ ( x ) ⇒ g ′ ( x ) = 0 ⇒ g ( x ) = C 6xy + 6y = 6xy + 6y + g'(x) \Rightarrow g'(x) = 0 \Rightarrow g(x) = C 6 x y + 6 y = 6 x y + 6 y + g ′ ( x ) ⇒ g ′ ( x ) = 0 ⇒ g ( x ) = C .
Harmonic conjugate: v ( x , y ) = 3 x 2 y − y 3 + 6 x y + C v(x, y) = 3x^2 y - y^3 + 6xy + C v ( x , y ) = 3 x 2 y − y 3 + 6 x y + C .
f ( z ) = u + i v = z 3 + 3 z 2 f(z) = u + iv = z^3 + 3z^2 f ( z ) = u + i v = z 3 + 3 z 2 .
If you get this wrong, revise: Section 3.4 (Harmonic Functions).
Evaluate ∫ γ ( z 2 + 2 z ) d z \int_\gamma (z^2 + 2z)\, dz ∫ γ ( z 2 + 2 z ) d z where γ \gamma γ is the upper half of the unit circle from z = 1 z = 1 z = 1 to z = − 1 z = -1 z = − 1 .
Solution Since z 2 + 2 z z^2 + 2z z 2 + 2 z is entire, the integral is path-independent. Let F ( z ) = z 3 / 3 + z 2 F(z) = z^3/3 + z^2 F ( z ) = z 3 /3 + z 2 .
∫ γ ( z 2 + 2 z ) d z = F ( − 1 ) − F ( 1 ) = 2 3 − 4 3 = − 2 3 \int_\gamma (z^2 + 2z)\, dz = F(-1) - F(1) = \frac{2}{3} - \frac{4}{3} = -\frac{2}{3} ∫ γ ( z 2 + 2 z ) d z = F ( − 1 ) − F ( 1 ) = 3 2 − 3 4 = − 3 2 .
If you get this wrong, revise: Sections 4.5 and 4.7 (Contour Integrals).
Use the ML inequality to bound ∣ ∫ γ e z z − 2 d z ∣ \left|\int_\gamma \frac{e^z}{z - 2}\, dz\right| ∫ γ z − 2 e z d z where γ \gamma γ Is the circle ∣ z ∣ = 1 |z| = 1 ∣ z ∣ = 1 .
Solution On γ \gamma γ : ∣ z ∣ = 1 |z| = 1 ∣ z ∣ = 1 So ∣ e z ∣ ≤ e |e^z| \leq e ∣ e z ∣ ≤ e and ∣ z − 2 ∣ ≥ 1 |z - 2| \geq 1 ∣ z − 2∣ ≥ 1 .
∣ e z z − 2 ∣ ≤ e \left|\frac{e^z}{z - 2}\right| \leq e z − 2 e z ≤ e . L = 2 π L = 2\pi L = 2 π .
∣ ∫ γ e z z − 2 d z ∣ ≤ 2 π e \left|\int_\gamma \frac{e^z}{z - 2}\, dz\right| \leq 2\pi e ∫ γ z − 2 e z d z ≤ 2 π e .
If you get this wrong, revise: Section 4.6 (ML Inequality).
Evaluate ∮ γ z + 1 z 2 − z d z \oint_\gamma \frac{z + 1}{z^2 - z}\, dz ∮ γ z 2 − z z + 1 d z where γ \gamma γ is ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
Solution z + 1 z 2 − z = z + 1 z ( z − 1 ) \frac{z + 1}{z^2 - z} = \frac{z + 1}{z(z - 1)} z 2 − z z + 1 = z ( z − 1 ) z + 1 . Simple poles at z = 0 z = 0 z = 0 and z = 1 z = 1 z = 1 Both inside ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
At z = 0 z = 0 z = 0 : R e s = lim z → 0 z + 1 z − 1 = − 1 \mathrm{Res} = \lim_{z \to 0} \frac{z + 1}{z - 1} = -1 Res = lim z → 0 z − 1 z + 1 = − 1 . At z = 1 z = 1 z = 1 : R e s = lim z → 1 z + 1 z = 2 \mathrm{Res} = \lim_{z \to 1} \frac{z + 1}{z} = 2 Res = lim z → 1 z z + 1 = 2 .
∮ γ z + 1 z 2 − z d z = 2 π i ( − 1 + 2 ) = 2 π i \oint_\gamma \frac{z + 1}{z^2 - z}\, dz = 2\pi i(-1 + 2) = 2\pi i ∮ γ z 2 − z z + 1 d z = 2 π i ( − 1 + 2 ) = 2 π i .
If you get this wrong, revise: Sections 8.4 and 8.5 (Residues).
Classify all singularities of f ( z ) = e 1 / z z 2 + 1 f(z) = \frac{e^{1/z}}{z^2 + 1} f ( z ) = z 2 + 1 e 1/ z and find all residues.
Solution z = 0 z = 0 z = 0 : e 1 / z e^{1/z} e 1/ z has an essential singularity at 0 0 0 So z = 0 z = 0 z = 0 is an essential singularity of f f f . z = i z = i z = i : simple pole. z = − i z = -i z = − i : simple pole.
At z = i z = i z = i : R e s = e 1 / i 2 i = e − i 2 i \mathrm{Res} = \frac{e^{1/i}}{2i} = \frac{e^{-i}}{2i} Res = 2 i e 1/ i = 2 i e − i . At z = − i z = -i z = − i : R e s = e 1 / ( − i ) − 2 i = e i − 2 i \mathrm{Res} = \frac{e^{1/(-i)}}{-2i} = \frac{e^{i}}{-2i} Res = − 2 i e 1/ ( − i ) = − 2 i e i .
At z = 0 z = 0 z = 0 : find the coefficient of 1 / z 1/z 1/ z in e 1 / z z 2 + 1 \frac{e^{1/z}}{z^2 + 1} z 2 + 1 e 1/ z . 1 z 2 + 1 = 1 − z 2 + z 4 − ⋯ \frac{1}{z^2 + 1} = 1 - z^2 + z^4 - \cdots z 2 + 1 1 = 1 − z 2 + z 4 − ⋯ near z = 0 z = 0 z = 0 . e 1 / z = 1 + 1 / z + 1 / ( 2 z 2 ) + ⋯ e^{1/z} = 1 + 1/z + 1/(2z^2) + \cdots e 1/ z = 1 + 1/ z + 1/ ( 2 z 2 ) + ⋯ . The 1 / z 1/z 1/ z coefficient in the product: from 1 ⋅ 1 / z = 1 / z 1 \cdot 1/z = 1/z 1 ⋅ 1/ z = 1/ z Giving residue 1 1 1 .
If you get this wrong, revise: Sections 8.1 and 8.4 (Singularities and Residues).
Evaluate ∫ 0 2 π cos θ 5 + 4 cos θ d θ \int_0^{2\pi} \frac{\cos\theta}{5 + 4\cos\theta}\, d\theta ∫ 0 2 π 5 + 4 c o s θ c o s θ d θ .
Solution Substitute z = e i θ z = e^{i\theta} z = e i θ :
I = ∫ ∣ z ∣ = 1 ( z + z − 1 ) / 2 5 + 4 ( z + z − 1 ) / 2 ⋅ d z i z = 1 2 i ∫ ∣ z ∣ = 1 z 2 + 1 z ( 2 z 2 + 5 z + 2 ) d z = 1 2 i ∫ ∣ z ∣ = 1 z 2 + 1 z ( 2 z + 1 ) ( z + 2 ) d z I = \int_{|z|=1} \frac{(z + z^{-1})/2}{5 + 4(z + z^{-1})/2} \cdot \frac{dz}{iz} = \frac{1}{2i}\int_{|z|=1} \frac{z^2 + 1}{z(2z^2 + 5z + 2)}\, dz = \frac{1}{2i}\int_{|z|=1} \frac{z^2 + 1}{z(2z + 1)(z + 2)}\, dz I = ∫ ∣ z ∣ = 1 5 + 4 ( z + z − 1 ) /2 ( z + z − 1 ) /2 ⋅ i z d z = 2 i 1 ∫ ∣ z ∣ = 1 z ( 2 z 2 + 5 z + 2 ) z 2 + 1 d z = 2 i 1 ∫ ∣ z ∣ = 1 z ( 2 z + 1 ) ( z + 2 ) z 2 + 1 d z .
Poles inside ∣ z ∣ = 1 |z| = 1 ∣ z ∣ = 1 : z = 0 z = 0 z = 0 (simple) and z = − 1 / 2 z = -1/2 z = − 1/2 (simple).
At z = 0 z = 0 z = 0 : R e s = 1 ( 2 ⋅ 0 + 1 ) ( 0 + 2 ) = 1 2 \mathrm{Res} = \frac{1}{(2 \cdot 0 + 1)(0 + 2)} = \frac{1}{2} Res = ( 2 ⋅ 0 + 1 ) ( 0 + 2 ) 1 = 2 1 . At z = − 1 / 2 z = -1/2 z = − 1/2 : R e s = 1 / 4 + 1 ( − 1 / 2 ) ( − 1 + 2 ) = 5 / 4 − 1 / 2 = − 5 2 \mathrm{Res} = \frac{1/4 + 1}{(-1/2)(-1 + 2)} = \frac{5/4}{-1/2} = -\frac{5}{2} Res = ( − 1/2 ) ( − 1 + 2 ) 1/4 + 1 = − 1/2 5/4 = − 2 5 .
I = 1 2 i ⋅ 2 π i ( 1 2 − 5 2 ) = π ( − 2 ) = − π 3 I = \frac{1}{2i} \cdot 2\pi i\left(\frac{1}{2} - \frac{5}{2}\right) = \pi(-2) = -\frac{\pi}{3} I = 2 i 1 ⋅ 2 π i ( 2 1 − 2 5 ) = π ( − 2 ) = − 3 π .
If you get this wrong, revise: Section 9.4 (Trigonometric Integrals).
Evaluate ∫ − ∞ ∞ d x ( x 2 + 1 ) ( x 2 + 4 ) \int_{-\infty}^{\infty} \frac{dx}{(x^2 + 1)(x^2 + 4)} ∫ − ∞ ∞ ( x 2 + 1 ) ( x 2 + 4 ) d x .
Solution f ( z ) = 1 ( z 2 + 1 ) ( z 2 + 4 ) f(z) = \frac{1}{(z^2 + 1)(z^2 + 4)} f ( z ) = ( z 2 + 1 ) ( z 2 + 4 ) 1 . Poles in upper half-plane: z = i z = i z = i (simple) and z = 2 i z = 2i z = 2 i (simple).
At z = i z = i z = i : R e s = 1 ( 2 i ) ( i 2 + 4 ) = 1 2 i ⋅ 3 = 1 6 i \mathrm{Res} = \frac{1}{(2i)(i^2 + 4)} = \frac{1}{2i \cdot 3} = \frac{1}{6i} Res = ( 2 i ) ( i 2 + 4 ) 1 = 2 i ⋅ 3 1 = 6 i 1 . At z = 2 i z = 2i z = 2 i : R e s = 1 ( 4 i − 1 ) ( 4 i ) = 1 4 i ( − 3 ) = − 1 12 i \mathrm{Res} = \frac{1}{(4i - 1)(4i)} = \frac{1}{4i(-3)} = -\frac{1}{12i} Res = ( 4 i − 1 ) ( 4 i ) 1 = 4 i ( − 3 ) 1 = − 12 i 1 .
∫ − ∞ ∞ f ( x ) d x = 2 π i ( 1 6 i − 1 12 i ) = 2 π i ⋅ 1 12 i = π 6 \int_{-\infty}^{\infty} f(x)\, dx = 2\pi i\left(\frac{1}{6i} - \frac{1}{12i}\right) = 2\pi i \cdot \frac{1}{12i} = \frac{\pi}{6} ∫ − ∞ ∞ f ( x ) d x = 2 π i ( 6 i 1 − 12 i 1 ) = 2 π i ⋅ 12 i 1 = 6 π .
If you get this wrong, revise: Section 9.2 (Rational Function Integrals).
Find the Taylor series of f ( z ) = z z 2 + 4 f(z) = \frac{z}{z^2 + 4} f ( z ) = z 2 + 4 z centered at z 0 = 0 z_0 = 0 z 0 = 0 and state the radius Of convergence.
Solution z z 2 + 4 = z 4 ⋅ 1 1 + z 2 / 4 = z 4 ∑ n = 0 ∞ ( − 1 ) n z 2 n 4 n = ∑ n = 0 ∞ ( − 1 ) n z 2 n + 1 4 n + 1 \frac{z}{z^2 + 4} = \frac{z}{4} \cdot \frac{1}{1 + z^2/4} = \frac{z}{4}\sum_{n=0}^{\infty} (-1)^n \frac{z^{2n}}{4^n} = \sum_{n=0}^{\infty} \frac{(-1)^n z^{2n+1}}{4^{n+1}} z 2 + 4 z = 4 z ⋅ 1 + z 2 /4 1 = 4 z ∑ n = 0 ∞ ( − 1 ) n 4 n z 2 n = ∑ n = 0 ∞ 4 n + 1 ( − 1 ) n z 2 n + 1
For ∣ z ∣ < 2 |z| \lt 2 ∣ z ∣ < 2 . Radius of convergence: distance from 0 0 0 to nearest singularity (± 2 i \pm 2i ± 2 i ), which is 2 2 2 .
If you get this wrong, revise: Section 7.1 (Taylor Series).
Find the Laurent series of f ( z ) = 1 ( z − 1 ) ( z − 2 ) f(z) = \frac{1}{(z - 1)(z - 2)} f ( z ) = ( z − 1 ) ( z − 2 ) 1 in the annulus 1 < ∣ z ∣ < 2 1 \lt |z| \lt 2 1 < ∣ z ∣ < 2 .
Solution 1 ( z − 1 ) ( z − 2 ) = 1 z − 2 − 1 z − 1 \frac{1}{(z-1)(z-2)} = \frac{1}{z - 2} - \frac{1}{z - 1} ( z − 1 ) ( z − 2 ) 1 = z − 2 1 − z − 1 1 .
For ∣ z ∣ > 1 |z| \gt 1 ∣ z ∣ > 1 : 1 z − 1 = 1 z ⋅ 1 1 − 1 / z = ∑ n = 0 ∞ z − n − 1 \frac{1}{z - 1} = \frac{1}{z} \cdot \frac{1}{1 - 1/z} = \sum_{n=0}^{\infty} z^{-n-1} z − 1 1 = z 1 ⋅ 1 − 1/ z 1 = ∑ n = 0 ∞ z − n − 1 .
For ∣ z ∣ < 2 |z| \lt 2 ∣ z ∣ < 2 : 1 z − 2 = − 1 2 ⋅ 1 1 − z / 2 = − ∑ n = 0 ∞ z n 2 n + 1 \frac{1}{z - 2} = -\frac{1}{2} \cdot \frac{1}{1 - z/2} = -\sum_{n=0}^{\infty} \frac{z^n}{2^{n+1}} z − 2 1 = − 2 1 ⋅ 1 − z /2 1 = − ∑ n = 0 ∞ 2 n + 1 z n .
f ( z ) = − ∑ n = 0 ∞ z n 2 n + 1 − ∑ n = 0 ∞ z − n − 1 f(z) = -\sum_{n=0}^{\infty} \frac{z^n}{2^{n+1}} - \sum_{n=0}^{\infty} z^{-n-1} f ( z ) = − ∑ n = 0 ∞ 2 n + 1 z n − ∑ n = 0 ∞ z − n − 1 .
If you get this wrong, revise: Section 7.4 (Laurent Series).
Using Rouché’s theorem, determine the number of roots of z 5 − 5 z + 1 = 0 z^5 - 5z + 1 = 0 z 5 − 5 z + 1 = 0 in ∣ z ∣ < 1 |z| \lt 1 ∣ z ∣ < 1 .
Solution On ∣ z ∣ = 1 |z| = 1 ∣ z ∣ = 1 : ∣ − 5 z ∣ = 5 > ∣ z 5 + 1 ∣ ≤ 2 |-5z| = 5 \gt |z^5 + 1| \leq 2 ∣ − 5 z ∣ = 5 > ∣ z 5 + 1∣ ≤ 2 .
By Rouché with f ( z ) = − 5 z f(z) = -5z f ( z ) = − 5 z and g ( z ) = z 5 + 1 g(z) = z^5 + 1 g ( z ) = z 5 + 1 : z 5 − 5 z + 1 z^5 - 5z + 1 z 5 − 5 z + 1 has the same number of zeros In ∣ z ∣ < 1 |z| \lt 1 ∣ z ∣ < 1 as − 5 z -5z − 5 z Which has exactly one zero (at z = 0 z = 0 z = 0 ).
So exactly one root in ∣ z ∣ < 1 |z| \lt 1 ∣ z ∣ < 1 .
If you get this wrong, revise: Section 12.2 (Rouché’s Theorem).
Find the Möbius transformation that maps 1 ↦ 0 1 \mapsto 0 1 ↦ 0 , i ↦ 1 i \mapsto 1 i ↦ 1 , − 1 ↦ ∞ -1 \mapsto \infty − 1 ↦ ∞ .
Solution T ( z ) = ( z − 1 ) ( i − ( − 1 ) ) ( z − ( − 1 ) ) ( i − 1 ) = ( z − 1 ) ( i + 1 ) ( z + 1 ) ( i − 1 ) T(z) = \frac{(z - 1)(i - (-1))}{(z - (-1))(i - 1)} = \frac{(z - 1)(i + 1)}{(z + 1)(i - 1)} T ( z ) = ( z − ( − 1 )) ( i − 1 ) ( z − 1 ) ( i − ( − 1 )) = ( z + 1 ) ( i − 1 ) ( z − 1 ) ( i + 1 ) .
Simplify: i + 1 i − 1 = ( i + 1 ) ( − i − 1 ) ( i − 1 ) ( − i − 1 ) = − i 2 − 2 i − 1 − i 2 + 1 = − 2 i 2 = − i \frac{i + 1}{i - 1} = \frac{(i+1)(-i-1)}{(i-1)(-i-1)} = \frac{-i^2 - 2i - 1}{-i^2 + 1} = \frac{-2i}{2} = -i i − 1 i + 1 = ( i − 1 ) ( − i − 1 ) ( i + 1 ) ( − i − 1 ) = − i 2 + 1 − i 2 − 2 i − 1 = 2 − 2 i = − i .
T ( z ) = − i ⋅ z − 1 z + 1 T(z) = -i \cdot \frac{z - 1}{z + 1} T ( z ) = − i ⋅ z + 1 z − 1 .
Verify: T ( 1 ) = 0 T(1) = 0 T ( 1 ) = 0 ✓ \checkmark ✓ , T ( i ) = − i ⋅ i − 1 i + 1 = − i ⋅ ( − i ) = − 1 T(i) = -i \cdot \frac{i-1}{i+1} = -i \cdot (-i) = -1 T ( i ) = − i ⋅ i + 1 i − 1 = − i ⋅ ( − i ) = − 1 .
That gives − 1 -1 − 1 Not 1 1 1 . Let me recompute.
T ( z ) = ( z − z 1 ) ( z 2 − z 3 ) ( z − z 3 ) ( z 2 − z 1 ) T(z) = \frac{(z - z_1)(z_2 - z_3)}{(z - z_3)(z_2 - z_1)} T ( z ) = ( z − z 3 ) ( z 2 − z 1 ) ( z − z 1 ) ( z 2 − z 3 ) with z 1 = 1 z_1 = 1 z 1 = 1 , z 2 = i z_2 = i z 2 = i , z 3 = − 1 z_3 = -1 z 3 = − 1 .
T ( z ) = ( z − 1 ) ( i + 1 ) ( z + 1 ) ( i − 1 ) T(z) = \frac{(z - 1)(i + 1)}{(z + 1)(i - 1)} T ( z ) = ( z + 1 ) ( i − 1 ) ( z − 1 ) ( i + 1 ) .
T ( i ) = ( i − 1 ) ( i + 1 ) ( i + 1 ) ( i − 1 ) = 1 T(i) = \frac{(i - 1)(i + 1)}{(i + 1)(i - 1)} = 1 T ( i ) = ( i + 1 ) ( i − 1 ) ( i − 1 ) ( i + 1 ) = 1 . ✓ \checkmark ✓
T ( 1 ) = 0 T(1) = 0 T ( 1 ) = 0 . ✓ \checkmark ✓ . T ( − 1 ) = ∞ T(-1) = \infty T ( − 1 ) = ∞ . ✓ \checkmark ✓ .
So T ( z ) = ( z − 1 ) ( i + 1 ) ( z + 1 ) ( i − 1 ) T(z) = \frac{(z - 1)(i + 1)}{(z + 1)(i - 1)} T ( z ) = ( z + 1 ) ( i − 1 ) ( z − 1 ) ( i + 1 ) .
If you get this wrong, revise: Section 10.5 (Cross-Ratio).
Evaluate ∫ γ z 3 z 2 + 1 d z \int_\gamma \frac{z^3}{z^2 + 1}\, dz ∫ γ z 2 + 1 z 3 d z where γ \gamma γ is ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
Solution z 3 z 2 + 1 \frac{z^3}{z^2 + 1} z 2 + 1 z 3 has simple poles at z = ± i z = \pm i z = ± i Both inside ∣ z ∣ = 2 |z| = 2 ∣ z ∣ = 2 .
At z = i z = i z = i : R e s = i 3 2 i = − i 2 i = − 1 2 \mathrm{Res} = \frac{i^3}{2i} = \frac{-i}{2i} = -\frac{1}{2} Res = 2 i i 3 = 2 i − i = − 2 1 . At z = − i z = -i z = − i : R e s = ( − i ) 3 − 2 i = i − 2 i = − 1 2 \mathrm{Res} = \frac{(-i)^3}{-2i} = \frac{i}{-2i} = -\frac{1}{2} Res = − 2 i ( − i ) 3 = − 2 i i = − 2 1 .
∫ γ z 3 z 2 + 1 d z = 2 π i ( − 1 2 − 1 2 ) = − 2 π i \int_\gamma \frac{z^3}{z^2 + 1}\, dz = 2\pi i\left(-\frac{1}{2} - \frac{1}{2}\right) = -2\pi i ∫ γ z 2 + 1 z 3 d z = 2 π i ( − 2 1 − 2 1 ) = − 2 π i .
Alternatively: z 3 z 2 + 1 = z − z z 2 + 1 \frac{z^3}{z^2 + 1} = z - \frac{z}{z^2 + 1} z 2 + 1 z 3 = z − z 2 + 1 z . ∫ γ z d z = 0 \int_\gamma z\, dz = 0 ∫ γ z d z = 0 (entire), and ∫ γ z z 2 + 1 d z = 2 π i ( 1 / 2 + 1 / 2 ) = 2 π i \int_\gamma \frac{z}{z^2 + 1}\, dz = 2\pi i(1/2 + 1/2) = 2\pi i ∫ γ z 2 + 1 z d z = 2 π i ( 1/2 + 1/2 ) = 2 π i . So the integral equals 0 − 2 π i = − 2 π i 0 - 2\pi i = -2\pi i 0 − 2 π i = − 2 π i . ✓ \checkmark ✓
If you get this wrong, revise: Sections 8.4 and 8.5 (Residues).
Show that ∫ − ∞ ∞ cos 2 x x 2 + 1 d x = π e 2 \int_{-\infty}^{\infty} \frac{\cos 2x}{x^2 + 1}\, dx = \frac{\pi}{e^2} ∫ − ∞ ∞ x 2 + 1 c o s 2 x d x = e 2 π .
Solution Consider ∫ − ∞ ∞ e 2 i x x 2 + 1 d x \int_{-\infty}^{\infty} \frac{e^{2ix}}{x^2 + 1}\, dx ∫ − ∞ ∞ x 2 + 1 e 2 i x d x .
f ( z ) = e 2 i z z 2 + 1 f(z) = \frac{e^{2iz}}{z^2 + 1} f ( z ) = z 2 + 1 e 2 i z has a simple pole at z = i z = i z = i in the upper half-plane.
R e s ( e 2 i z z 2 + 1 , i ) = e 2 i ⋅ i 2 i = e − 2 2 i \mathrm{Res}\!\left(\frac{e^{2iz}}{z^2 + 1}, i\right) = \frac{e^{2i \cdot i}}{2i} = \frac{e^{-2}}{2i} Res ( z 2 + 1 e 2 i z , i ) = 2 i e 2 i ⋅ i = 2 i e − 2 .
∫ − ∞ ∞ e 2 i x x 2 + 1 d x = 2 π i ⋅ e − 2 2 i = π e 2 \int_{-\infty}^{\infty} \frac{e^{2ix}}{x^2 + 1}\, dx = 2\pi i \cdot \frac{e^{-2}}{2i} = \frac{\pi}{e^2} ∫ − ∞ ∞ x 2 + 1 e 2 i x d x = 2 π i ⋅ 2 i e − 2 = e 2 π .
Taking real parts: ∫ − ∞ ∞ cos 2 x x 2 + 1 d x = π e 2 \int_{-\infty}^{\infty} \frac{\cos 2x}{x^2 + 1}\, dx = \frac{\pi}{e^2} ∫ − ∞ ∞ x 2 + 1 c o s 2 x d x = e 2 π .
If you get this wrong, revise: Section 9.7 (Fourier-Type Integrals).
Find the residue of f ( z ) = sin z z 4 f(z) = \frac{\sin z}{z^4} f ( z ) = z 4 s i n z at z = 0 z = 0 z = 0 .
Solution sin z = z − z 3 / 6 + z 5 / 120 − ⋯ \sin z = z - z^3/6 + z^5/120 - \cdots sin z = z − z 3 /6 + z 5 /120 − ⋯
f ( z ) = z − z 3 / 6 + z 5 / 120 − ⋯ z 4 = 1 z 3 − 1 6 z + z 120 − ⋯ f(z) = \frac{z - z^3/6 + z^5/120 - \cdots}{z^4} = \frac{1}{z^3} - \frac{1}{6z} + \frac{z}{120} - \cdots f ( z ) = z 4 z − z 3 /6 + z 5 /120 − ⋯ = z 3 1 − 6 z 1 + 120 z − ⋯
The coefficient of 1 / z 1/z 1/ z is − 1 / 6 -1/6 − 1/6 So R e s ( f , 0 ) = − 1 6 \mathrm{Res}(f, 0) = -\frac{1}{6} Res ( f , 0 ) = − 6 1 .
If you get this wrong, revise: Section 8.4 (Computing Residues).
Evaluate ∫ γ d z ( z − 1 ) 2 ( z − 2 ) \int_\gamma \frac{dz}{(z - 1)^2(z - 2)} ∫ γ ( z − 1 ) 2 ( z − 2 ) d z where γ \gamma γ is ∣ z − 1 ∣ = 1 / 2 |z - 1| = 1/2 ∣ z − 1∣ = 1/2 .
Solution Only z = 1 z = 1 z = 1 is inside γ \gamma γ (a pole of order 2 2 2 ). z = 2 z = 2 z = 2 is outside.
R e s ( f , 1 ) = d d z [ 1 z − 2 ] z = 1 = − 1 ( z − 2 ) 2 ∣ z = 1 = − 1 \mathrm{Res}(f, 1) = \frac{d}{dz}\left[\frac{1}{z - 2}\right]_{z=1} = -\frac{1}{(z-2)^2}\Big|_{z=1} = -1 Res ( f , 1 ) = d z d [ z − 2 1 ] z = 1 = − ( z − 2 ) 2 1 z = 1 = − 1 .
∫ γ f d z = 2 π i ⋅ ( − 1 ) = − 2 π i \int_\gamma f\, dz = 2\pi i \cdot (-1) = -2\pi i ∫ γ f d z = 2 π i ⋅ ( − 1 ) = − 2 π i .
If you get this wrong, revise: Section 6.2 (CIF for Derivatives) and 8.4 (Residues).
Use the Cauchy-Riemann equations to show that f ( z ) = ∣ z ∣ 2 + 2 z ˉ f(z) = |z|^2 + 2\bar{z} f ( z ) = ∣ z ∣ 2 + 2 z ˉ is differentiable at Exactly one point and find f ′ ( z ) f'(z) f ′ ( z ) there.
Solution f ( z ) = x 2 + y 2 + 2 x − 2 i y f(z) = x^2 + y^2 + 2x - 2iy f ( z ) = x 2 + y 2 + 2 x − 2 i y . So u = x 2 + y 2 + 2 x u = x^2 + y^2 + 2x u = x 2 + y 2 + 2 x , v = − 2 y v = -2y v = − 2 y .
u x = 2 x + 2 u_x = 2x + 2 u x = 2 x + 2 , u y = 2 y u_y = 2y u y = 2 y , v x = 0 v_x = 0 v x = 0 , v y = − 2 v_y = -2 v y = − 2 .
CR: 2 x + 2 = − 2 ⇒ x = − 2 2x + 2 = -2 \Rightarrow x = -2 2 x + 2 = − 2 ⇒ x = − 2 And 2 y = 0 ⇒ y = 0 2y = 0 \Rightarrow y = 0 2 y = 0 ⇒ y = 0 .
f f f is differentiable only at z = − 2 z = -2 z = − 2 .
f ′ ( − 2 ) = u x ( − 2 , 0 ) + i v x ( − 2 , 0 ) = ( 2 ( − 2 ) + 2 ) + 0 = − 2 f'(-2) = u_x(-2, 0) + iv_x(-2, 0) = (2(-2) + 2) + 0 = -2 f ′ ( − 2 ) = u x ( − 2 , 0 ) + i v x ( − 2 , 0 ) = ( 2 ( − 2 ) + 2 ) + 0 = − 2 .
If you get this wrong, revise: Section 3.1 (Cauchy-Riemann Equations).
Evaluate ∫ γ e z sin z ( z − π ) 3 d z \int_\gamma \frac{e^z \sin z}{(z - \pi)^3}\, dz ∫ γ ( z − π ) 3 e z s i n z d z where γ \gamma γ is ∣ z ∣ = 4 |z| = 4 ∣ z ∣ = 4 .
Solution Only z = π z = \pi z = π is inside γ \gamma γ (a pole of order 3 3 3 ).
By CIF for derivatives: ∫ γ f ( z ) ( z − π ) 3 d z = 2 π i 2 ! f ′ ′ ( π ) \int_\gamma \frac{f(z)}{(z - \pi)^3}\, dz = \frac{2\pi i}{2!}\,f''(\pi) ∫ γ ( z − π ) 3 f ( z ) d z = 2 ! 2 π i f ′′ ( π ) Where f ( z ) = e z sin z f(z) = e^z \sin z f ( z ) = e z sin z .
f ′ ( z ) = e z sin z + e z cos z = e z ( sin z + cos z ) f'(z) = e^z \sin z + e^z \cos z = e^z(\sin z + \cos z) f ′ ( z ) = e z sin z + e z cos z = e z ( sin z + cos z ) . f ′ ′ ( z ) = e z ( sin z + cos z ) + e z ( cos z − sin z ) = 2 e z cos z f''(z) = e^z(\sin z + \cos z) + e^z(\cos z - \sin z) = 2e^z \cos z f ′′ ( z ) = e z ( sin z + cos z ) + e z ( cos z − sin z ) = 2 e z cos z .
f ′ ′ ( π ) = 2 e π cos π = − 2 e π f''(\pi) = 2e^\pi \cos\pi = -2e^\pi f ′′ ( π ) = 2 e π cos π = − 2 e π .
∫ γ e z sin z ( z − π ) 3 d z = π i ⋅ ( − 2 e π ) = − 2 π i e π \int_\gamma \frac{e^z \sin z}{(z - \pi)^3}\, dz = \pi i \cdot (-2e^\pi) = -2\pi i\, e^\pi ∫ γ ( z − π ) 3 e z s i n z d z = π i ⋅ ( − 2 e π ) = − 2 π i e π .
If you get this wrong, revise: Section 6.2 (Cauchy’s Integral Formula for Derivatives).
Example 1: Definite integration
Evaluate ∫ 0 2 ( 3 x 2 + 2 x ) d x \displaystyle\int_0^2 (3x^2 + 2x)\,dx ∫ 0 2 ( 3 x 2 + 2 x ) d x .
Solution:
∫ ( 3 x 2 + 2 x ) d x = x 3 + x 2 + c \int (3x^2 + 2x)\,dx = x^3 + x^2 + c ∫ ( 3 x 2 + 2 x ) d x = x 3 + x 2 + c
[ x 3 + x 2 ] 0 2 = ( 8 + 4 ) − ( 0 ) = 12 \left[x^3 + x^2\right]_0^2 = (8 + 4) - (0) = 12 [ x 3 + x 2 ] 0 2 = ( 8 + 4 ) − ( 0 ) = 12
Example 2: Integration by parts
Find ∫ x e 2 x d x \displaystyle\int x e^{2x}\,dx ∫ x e 2 x d x .
Solution:
Let u = x ⟹ d u d x = 1 u = x \implies \frac{du}{dx} = 1 u = x ⟹ d x d u = 1 and d v d x = e 2 x ⟹ v = 1 2 e 2 x \frac{dv}{dx} = e^{2x} \implies v = \frac{1}{2}e^{2x} d x d v = e 2 x ⟹ v = 2 1 e 2 x .
∫ x e 2 x d x = x ⋅ 1 2 e 2 x − ∫ 1 2 e 2 x d x = x e 2 x 2 − e 2 x 4 + c = e 2 x ( 2 x − 1 ) 4 + c \int x e^{2x}\,dx = x \cdot \frac{1}{2}e^{2x} - \int \frac{1}{2}e^{2x}\,dx = \frac{x e^{2x}}{2} - \frac{e^{2x}}{4} + c = \frac{e^{2x}(2x - 1)}{4} + c ∫ x e 2 x d x = x ⋅ 2 1 e 2 x − ∫ 2 1 e 2 x d x = 2 x e 2 x − 4 e 2 x + c = 4 e 2 x ( 2 x − 1 ) + c
Holomorphic functions: complex differentiable on an open set; Cauchy-Riemann equations u x = v y u_x = v_y u x = v y , u y = − v x u_y = -v_x u y = − v x are necessary conditions. Cauchy’s integral theorem: ∮ γ f ( z ) d z = 0 \oint_\gamma f(z)\,dz = 0 ∮ γ f ( z ) d z = 0 for f f f holomorphic inside and on γ \gamma γ ; Cauchy’s integral formula evaluates f ( a ) f(a) f ( a ) and all derivatives via contour integrals. Residue theorem: ∮ γ f ( z ) d z = 2 π i ∑ Res ( f , z k ) \oint_\gamma f(z)\,dz = 2\pi i \sum \text{Res}(f, z_k) ∮ γ f ( z ) d z = 2 π i ∑ Res ( f , z k ) ; residues computed via Laurent series coefficients. Laurent series: generalises Taylor series to annular regions; classifies singularities as removable, poles, or essential. Conformal mappings: holomorphic functions with non-zero derivative preserve angles; applications in fluid dynamics and electrostatics. Topic Site Link Real Analysis WyattsNotes View Multivariable Calculus WyattsNotes View Differential Equations WyattsNotes View Complex Analysis — MIT 18.04 MIT OCW View