Complex Functions of a Variable

Complex Numbers and Axiomatic Foundations

  • Definition of Complex Numbers: The set of complex numbers, denoted by C\mathbb{C}, is defined as the set of all ordered pairs (x,y)(x, y) of real numbers x,y∈Rx, y \in \mathbb{R}. For any complex number z=(x,y)z = (x, y), the real number xx is termed the real part of zz (Re(z)=x\text{Re}(z) = x), and the real number yy is termed the imaginary part of zz (Im(z)=y\text{Im}(z) = y). The real numbers are embedded naturally as R⊂C\mathbb{R} \subset \mathbb{C} by identifying x∈Rx \in \mathbb{R} with (x,0)∈C(x, 0) \in \mathbb{C}.

  • Field Operations: Let z1=(x1,y1)z_1 = (x_1, y_1) and z2=(x2,y2)z_2 = (x_2, y_2) be arbitrary elements of C\mathbb{C}. The algebraic operations of equality, addition, and multiplication are defined as:

    • Equality: z1=z2  ⟺  x1=x2 and y1=y2z_1 = z_2 \iff x_1 = x_2 \text{ and } y_1 = y_2.

    • Addition: z1+z2=(x1+x2,y1+y2)z_1 + z_2 = (x_1 + x_2, y_1 + y_2).

    • Multiplication: z1z2=(x1x2−y1y2,x1y2+y1x2)z_1 z_2 = (x_1 x_2 - y_1 y_2, x_1 y_2 + y_1 x_2).

  • Field Axioms of C\mathbb{C}: The set C\mathbb{C} equipped with these operations satisfies all field axioms:

    • Closure under addition and multiplication: ∀z1,z2∈C\forall z_1, z_2 \in \mathbb{C}, z1+z2∈Cz_1 + z_2 \in \mathbb{C} and z1z2∈Cz_1 z_2 \in \mathbb{C}.

    • Commutativity: z1+z2=z2+z1z_1 + z_2 = z_2 + z_1 and z1z2=z2z1z_1 z_2 = z_2 z_1.

    • Associativity: (z1+z2)+z3=z1+(z2+z3)(z_1 + z_2) + z_3 = z_1 + (z_2 + z_3) and (z1z2)z3=z1(z2z3)(z_1 z_2) z_3 = z_1 (z_2 z_3).

    • Distributivity: z1(z2+z3)=z1z2+z1z3z_1 (z_2 + z_3) = z_1 z_2 + z_1 z_3.

    • Existence of Identities: The additive identity is unique and given by 0=(0,0)0 = (0, 0), satisfying z+0=zz + 0 = z. The multiplicative identity is unique and given by 1=(1,0)1 = (1, 0), satisfying z⋅1=zz \cdot 1 = z.

    • Existence of Inverses: Every z=(x,y)∈Cz = (x, y) \in \mathbb{C} has a unique additive inverse −z=(−x,−y)-z = (-x, -y) such that z+(−z)=0z + (-z) = 0. Every non-zero z=(x,y)≠(0,0)z = (x, y) \ne (0, 0) has a unique multiplicative inverse z−1z^{-1} given by:     z−1=(xx2+y2,−yx2+y2)z^{-1} = \left( \frac{x}{x^2 + y^2}, -\frac{y}{x^2 + y^2} \right)     such that zz−1=1z z^{-1} = 1.

  • Subtended Operations and Integer Powers:

    • Subtraction: z−w=z+(−w)z - w = z + (-w).

    • Division: z/w=zw−1z / w = z w^{-1} for w≠0w \ne 0.

  • Positive Integer Powers: For n∈Nn \in \mathbb{N}, zn=z⋅z⋯zz^n = z \cdot z \cdots z (nn times).

  • Zero and Negative Integer Powers: For z≠0z \ne 0, z0=1z^0 = 1 and z−n=1/znz^{-n} = 1 / z^n.

  • The Imaginary Unit and Standard Algebraic Form:

    • The imaginary unit is defined as i=(0,1)i = (0, 1). It satisfies i2=(0,1)(0,1)=(−1,0)=−1i^2 = (0, 1)(0, 1) = (-1, 0) = -1.

    • Cyclic recurrence of integer powers of ii: i0=1i^0 = 1, i1=ii^1 = i, i2=−1i^2 = -1, i3=−ii^3 = -i, i4=1i^4 = 1, and generally i4k+r=iri^{4k + r} = i^r for k,r∈Zk, r \in \mathbb{Z}.

    • Standard form decomposition: Every pair (x,y)(x, y) satisfies (x,y)=(x,0)+(0,1)(y,0)=x+iy(x, y) = (x, 0) + (0, 1)(y, 0) = x + iy.

    • Sum and product in standard form: (x1+iy1)+(x2+iy2)=(x1+x2)+i(y1+y2)(x_1 + i y_1) + (x_2 + i y_2) = (x_1 + x_2) + i(y_1 + y_2), and (x1+iy1)(x2+iy2)=(x1x2−y1y2)+i(x1y2+y1x2)(x_1 + i y_1)(x_2 + i y_2) = (x_1 x_2 - y_1 y_2) + i(x_1 y_2 + y_1 x_2).

  • Modulus and Complex Conjugation:

    • Modulus: For z=x+iyz = x + iy, the modulus is ∣z∣=x2+y2|z| = \sqrt{x^2 + y^2}, with ∣z∣≥0|z| \ge 0 and ∣z∣=0  ⟺  z=0|z| = 0 \iff z = 0.

    • Complex Conjugate: For z=x+iyz = x + iy, the conjugate is zˉ=x−iy\bar{z} = x - iy.

    • Fundamental Identities:     zzˉ=∣z∣2z \bar{z} = |z|^2     ∣z∣=zzˉ|z| = \sqrt{z \bar{z}}     z−1=zˉ∣z∣2(z≠0)z^{-1} = \frac{\bar{z}}{|z|^2} \quad (z \ne 0)     Re(z)=z+zˉ2,Im(z)=z−zˉ2i\text{Re}(z) = \frac{z + \bar{z}}{2}, \quad \text{Im}(z) = \frac{z - \bar{z}}{2i}     z±w‾=zˉ±wˉ,zw‾=zˉwˉ,(zw)‾=zˉwˉ(w≠0)\overline{z \pm w} = \bar{z} \pm \bar{w}, \quad \overline{zw} = \bar{z}\bar{w}, \quad \overline{\left(\frac{z}{w}\right)} = \frac{\bar{z}}{\bar{w}} \quad (w \ne 0)     ∣zw∣=∣z∣∣w∣,∣zw∣=∣z∣∣w∣(w≠0)|zw| = |z||w|, \quad \left|\frac{z}{w}\right| = \frac{|z|}{|w|} \quad (w \ne 0)     zˉˉ=z,∣zˉ∣=∣z∣\bar{\bar{z}} = z, \quad |\bar{z}| = |z|     ∣Re(z)∣≤∣z∣,∣Im(z)∣≤∣z∣|\text{Re}(z)| \le |z|, \quad |\text{Im}(z)| \le |z|

  • Triangle Inequality and Bounds:

    • Theorem: For any z,w∈Cz, w \in \mathbb{C}, ∣z+w∣≤∣z∣+∣w∣|z + w| \le |z| + |w|.

    • Derivation: ∣z+w∣2=(z+w)(zˉ+wˉ)=∣z∣2+2Re(zwˉ)+∣w∣2≤∣z∣2+2∣zwˉ∣+∣w∣2=(∣z∣+∣w∣)2|z + w|^2 = (z + w)(\bar{z} + \bar{w}) = |z|^2 + 2\text{Re}(z\bar{w}) + |w|^2 \le |z|^2 + 2|z\bar{w}| + |w|^2 = (|z| + |w|)^2.

    • Equality condition: Holds if and only if Re(zwˉ)=∣zwˉ∣\text{Re}(z\bar{w}) = |z\bar{w}| and Im(zwˉ)=0\text{Im}(z\bar{w}) = 0, meaning z=0z = 0, w=0w = 0, or zz and ww lie on the same ray emanating from the origin (w=czw = c z with c>0c > 0).

    • Reverse Triangle Inequality: ∣z+w∣≥∣∣z∣−∣w∣∣|z + w| \ge ||z| - |w||.

    • Full Bilateral Bound:     ∣∣z∣−∣w∣∣≤∣z±w∣≤∣z∣+∣w∣||z| - |w|| \le |z \pm w| \le |z| + |w|

    • Generalized Triangle Inequality: For nn complex numbers, ∣∑k=1nzk∣≤∑k=1n∣zk∣\left|\sum_{k=1}^n z_k\right| \le \sum_{k=1}^n |z_k|.

    • Direct Inequality: ∣Re(z)∣+∣Im(z)∣≤2∣z∣|\text{Re}(z)| + |\text{Im}(z)| \le \sqrt{2}|z|.

  • Polar Form and Principal Argument:

    • Polar Coordinates: Given z=x+iyz = x + iy, polar coordinates r,θr, \theta are defined via x=rcos⁡(θ)x = r \cos(\theta), y=rsin⁡(θ)y = r \sin(\theta), with r=∣z∣=x2+y2r = |z| = \sqrt{x^2 + y^2}.

    • Argument: The angle θ\theta is denoted by arg⁡(z)\arg(z) and is defined up to integer multiples of 2π2\pi.

    • Principal Argument: The unique angle Arg(z)∈(−π,π]\text{Arg}(z) \in (-\pi, \pi] such that arg⁡(z)=Arg(z)+2πn\arg(z) = \text{Arg}(z) + 2\pi n (n∈Zn \in \mathbb{Z}).

    • Explicit Representation of Arg(z)\text{Arg}(z):     Arg(z)={arctan⁡(yx)−πif x<0,y<0arctan⁡(yx)if x≥0,(x,y)≠(0,0)arctan⁡(yx)+πif x<0,y≥0\text{Arg}(z) = \begin{cases} \arctan\left(\frac{y}{x}\right) - \pi & \text{if } x < 0, y < 0 \\ \arctan\left(\frac{y}{x}\right) & \text{if } x \ge 0, (x, y) \ne (0, 0) \\ \arctan\left(\frac{y}{x}\right) + \pi & \text{if } x < 0, y \ge 0 \end{cases}     When x=0x = 0, Arg(z)=sgn(y)π2\text{Arg}(z) = \text{sgn}(y)\frac{\pi}{2}. The function is undefined at z=0z = 0.

Graph of Arg(x,y) showing discontinuity along negative real semi-axis
  • Multiplication, Division, and Powers in Polar Form:

    • Polar notation: cis(θ)=cos⁡(θ)+isin⁡(θ)=eiθ\text{cis}(\theta) = \cos(\theta) + i \sin(\theta) = e^{i\theta}.

    • Product: If z1=r1eiθ1z_1 = r_1 e^{i\theta_1} and z2=r2eiθ2z_2 = r_2 e^{i\theta_2}, then z1z2=r1r2ei(θ1+θ2)z_1 z_2 = r_1 r_2 e^{i(\theta_1 + \theta_2)}.

    • Quotient: z1/z2=(r1/r2)ei(θ1−θ2)z_1 / z_2 = (r_1 / r_2) e^{i(\theta_1 - \theta_2)}.

    • Multi-factor product: ∣z1z2⋯zn∣=∣z1∣∣z2∣⋯∣zn∣|z_1 z_2 \cdots z_n| = |z_1||z_2|\cdots|z_n|, and arg⁡(z1z2⋯zn)=∑k=1narg⁡(zk)\arg(z_1 z_2 \cdots z_n) = \sum_{k=1}^n \arg(z_k).

    • de Moivre's Formula: For any n∈Zn \in \mathbb{Z},     (cos⁡(θ)+isin⁡(θ))n=cos⁡(nθ)+isin⁡(nθ)(\cos(\theta) + i \sin(\theta))^n = \cos(n\theta) + i \sin(n\theta)

  • Roots of Complex Numbers:

    • For z≠0z \ne 0 and integer n≥2n \ge 2, the equation wn=zw^n = z admits exactly nn distinct roots given by:     wk=∣z∣neiArg(z)+2πkn,k=0,1,…,n−1w_k = \sqrt[n]{|z|} e^{i\frac{\text{Arg}(z) + 2\pi k}{n}}, \quad k = 0, 1, \dots, n - 1

    • Roots of Unity: The nn-th roots of unity are given by ck=ei2πknc_k = e^{i\frac{2\pi k}{n}} (k=0,1,…,n−1k = 0, 1, \dots, n-1), situated uniformly on the unit circle centered at the origin, with angular separation 2π/n2\pi / n.

    • If ww is any single nn-th root of zz, the set of all roots is {wc0,wc1,…,wcn−1}\{w c_0, w c_1, \dots, w c_{n-1}\}.

  • Geometric Regions in the Complex Plane:

    • Line Segment: Joining z1z_1 and z2z_2 is parametrized by λ(t)=z1+(z2−z1)t\lambda(t) = z_1 + (z_2 - z_1)t for 0≤t≤10 \le t \le 1.

    • Infinite Straight Line: μ(t)=z1+(z2−z1)t\mu(t) = z_1 + (z_2 - z_1)t for −∞<t<∞-\infty < t < \infty.

    • Non-parametric Line Equation:     {z∈C:Im(z−z1z2−z1)=0}\left\{ z \in \mathbb{C} : \text{Im}\left( \frac{z - z_1}{z_2 - z_1} \right) = 0 \right\}

    • Open Half-Planes: The sets {z∈C:Im(z−z1z2−z1)<0}\left\{ z \in \mathbb{C} : \text{Im}\left( \frac{z - z_1}{z_2 - z_1} \right) < 0 \right\} and {z∈C:Im(z−z1z2−z1)>0}\left\{ z \in \mathbb{C} : \text{Im}\left( \frac{z - z_1}{z_2 - z_1} \right) > 0 \right\} define the regions strictly to the right and left of the oriented line passing from z1z_1 to z2z_2.

    • Circles and Disks: Boundary circle of radius RR centered at z0z_0: {z∈C:∣z−z0∣=R}\{z \in \mathbb{C} : |z - z_0| = R\}, parametrized as γ(θ)=z0+Reiθ\gamma(\theta) = z_0 + R e^{i\theta} (0≤θ≤2π0 \le \theta \le 2\pi). Open disk: {z∈C:∣z−z0∣<R}\{z \in \mathbb{C} : |z - z_0| < R\}; exterior region: {z∈C:∣z−z0∣>R}\{z \in \mathbb{C} : |z - z_0| > R\}.

  • The Extended Complex Plane and Stereographic Projection:

    • Extended Plane: C∞=C∪{∞}\mathbb{C}_\infty = \mathbb{C} \cup \{\infty\}.

    • Geometric Setup: Unit sphere S2={(x1,x2,x3)∈R3:x12+x22+x32=1}S^2 = \{(x_1, x_2, x_3) \in \mathbb{R}^3 : x_1^2 + x_2^2 + x_3^2 = 1\}. The plane x3=0x_3 = 0 is identified with C\mathbb{C}. North pole N=(0,0,1)N = (0, 0, 1).

    • Projection Line: Ray connecting NN to Pz=(x,y,0)P_z = (x, y, 0) intersects the sphere at Q=(x1,x2,x3)Q = (x_1, x_2, x_3).

    • Coordinate Transformations (C→S2\mathbb{C} \to S^2):     x1=2Re(z)1+∣z∣2,x2=2Im(z)1+∣z∣2,x3=∣z∣2−11+∣z∣2x_1 = \frac{2\text{Re}(z)}{1 + |z|^2}, \quad x_2 = \frac{2\text{Im}(z)}{1 + |z|^2}, \quad x_3 = \frac{|z|^2 - 1}{1 + |z|^2}

    • Inverse Transformations (S2→CS^2 \to \mathbb{C}):     z=x1+ix21−x3z = \frac{x_1 + i x_2}{1 - x_3}

Stereographic projection of the complex plane on the Riemann sphere

Metric Spaces and Topology of the Complex Plane

  • Metric Spaces and Distances:

    • A metric space is a pair (S,d)(S, d) where SS is a set and d:S×S→Rd: S \times S \to \mathbb{R} satisfies:

    • Positivity: d(x,y)≥0d(x, y) \ge 0, and d(x,y)=0  ⟺  x=yd(x, y) = 0 \iff x = y.

    • Symmetry: d(x,y)=d(y,x)d(x, y) = d(y, x).

    • Triangle Inequality: d(x,y)≤d(x,z)+d(z,y)d(x, y) \le d(x, z) + d(z, y).

    • Diameter: diam(S)=sup⁡x,y∈Sd(x,y)\text{diam}(S) = \sup_{x, y \in S} d(x, y). A space is bounded if diam(S)<∞\text{diam}(S) < \infty, and unbounded if diam(S)=∞\text{diam}(S) = \infty.

  • Metrics on the Complex Plane:

    • Standard Euclidean Metric on C\mathbb{C}:     d(z,z′)=∣z−z′∣=(Re(z)−Re(z′))2+(Im(z)−Im(z′))2d(z, z') = |z - z'| = \sqrt{(\text{Re}(z) - \text{Re}(z'))^2 + (\text{Im}(z) - \text{Im}(z'))^2}     Under this metric, diam(C)=∞\text{diam}(\mathbb{C}) = \infty.

    • Chordal Metric on C\mathbb{C} and C∞\mathbb{C}_\infty: Derived from Euclidean distance in R3\mathbb{R}^3 between projected points on the Riemann sphere:     d(z,z′)=2∣z−z′∣(1+∣z∣2)(1+∣z′∣2)(z,z′∈C)d(z, z') = \frac{2|z - z'|}{\sqrt{(1 + |z|^2)(1 + |z'|^2)}} \quad (z, z' \in \mathbb{C})     d(z,∞)=21+∣z∣2(z∈C)d(z, \infty) = \frac{2}{\sqrt{1 + |z|^2}} \quad (z \in \mathbb{C})     Under the chordal metric, diam(C∞)=2\text{diam}(\mathbb{C}_\infty) = 2.

  • Open and Closed Balls:

    • Open ball: B(x,r)={y∈S:d(x,y)<r}B(x, r) = \{y \in S : d(x, y) < r\}.

    • Closed ball: Bˉ(x,r)={y∈S:d(x,y)≤r}\bar{B}(x, r) = \{y \in S : d(x, y) \le r\}.

  • Topological Structures (Open and Closed Sets):

    • Open Sets: A set G⊂SG \subset S is open if ∀x∈G,∃ε>0\forall x \in G, \exists \varepsilon > 0 such that B(x,ε)⊂GB(x, \varepsilon) \subset G.

    • Every open ball B(x,r)B(x, r) is an open set.

    • The sets SS and ∅\emptyset are open.

    • Finite intersections of open sets are open: ⋂k=1nGk\bigcap_{k=1}^n G_k is open.

    • Arbitrary unions of open sets are open: ⋃α∈IGα\bigcup_{\alpha \in I} G_\alpha is open.

    • Closed Sets: A set F⊂SF \subset S is closed if its complement Fc=S∖FF^c = S \setminus F is open.

    • Every closed ball Bˉ(x,r)\bar{B}(x, r) is a closed set.

    • The sets SS and ∅\emptyset are closed.

    • Finite unions of closed sets are closed: ⋃k=1nFk\bigcup_{k=1}^n F_k is closed.

    • Arbitrary intersections of closed sets are closed: ⋂α∈IFα\bigcap_{\alpha \in I} F_\alpha is closed.

  • Interior, Closure, and Boundary:

    • Interior: A∘=⋃G⊂A,G openGA^\circ = \bigcup_{G \subset A, G \text{ open}} G. Point x∈A∘  ⟺  ∃ε>0,B(x,ε)⊂Ax \in A^\circ \iff \exists \varepsilon > 0, B(x, \varepsilon) \subset A.

    • Closure: Aˉ=⋂F⊃A,F closedF\bar{A} = \bigcap_{F \supset A, F \text{ closed}} F. Point x∈Aˉ  ⟺  ∀ε>0,B(x,ε)∩A≠∅x \in \bar{A} \iff \forall \varepsilon > 0, B(x, \varepsilon) \cap A \ne \emptyset.

  • Boundary: ∂A=Aˉ∖A∘=Aˉ∩(A∘)c\partial A = \bar{A} \setminus A^\circ = \bar{A} \cap (A^\circ)^c. Point x∈∂A  ⟺  ∀ε>0,B(x,ε)∩A≠∅x \in \partial A \iff \forall \varepsilon > 0, B(x, \varepsilon) \cap A \ne \emptyset and B(x,ε)∩Ac≠∅B(x, \varepsilon) \cap A^c \ne \emptyset.

    • Key Identities:     A is open  ⟺  A=A∘A \text{ is open} \iff A = A^\circ     A is closed  ⟺  A=AˉA \text{ is closed} \iff A = \bar{A}     A∘=((Ac)−)c,Aˉ=((Ac)∘)c,∂A=Aˉ∩Ac‾A^\circ = ((A^c)^-)^c, \quad \bar{A} = ((A^c)^\circ)^c, \quad \partial A = \bar{A} \cap \overline{A^c}     (A∩B)∘=A∘∩B∘,A∪B‾=Aˉ∪Bˉ(A \cap B)^\circ = A^\circ \cap B^\circ, \quad \overline{A \cup B} = \bar{A} \cup \bar{B}     diam(A)=diam(Aˉ)\text{diam}(A) = \text{diam}(\bar{A})

  • Connectedness:

    • A metric space (S,d)(S, d) is connected if the only subsets that are simultaneously open and closed are ∅\emptyset and SS.

    • A subset S⊂RS \subset \mathbb{R} is connected if and only if SS is an interval.

    • Polygonal Path: Given points z1,z2,…,zn∈Cz_1, z_2, \dots, z_n \in \mathbb{C}, a polygonal path is P=⋃k=1n−1[zk,zk+1]P = \bigcup_{k=1}^{n-1} [z_k, z_{k+1}], where [zk,zk+1]={zk+t(zk+1−zk):0≤t≤1}[z_k, z_{k+1}] = \{z_k + t(z_{k+1} - z_k) : 0 \le t \le 1\}.

    • Characterization Theorem: An open set G⊂CG \subset \mathbb{C} is connected if and only if every pair of points in GG can be joined by a polygonal path lying entirely in GG.

  • Sequences, Completeness, and Compactness:

    • Convergence: A sequence (xn)(x_n) converges to xx (lim⁡n→∞xn=x\lim_{n\to\infty} x_n = x) iff ∀ε>0,∃N(ε)\forall \varepsilon > 0, \exists N(\varepsilon) such that d(xn,x)<εd(x_n, x) < \varepsilon for all n≥Nn \ge N.

    • Sequential Characterization of Closed Sets: A set FF is closed iff every convergent sequence in FF has its limit in FF.

    • Limit Points (Adherent Points): xx is a limit point of AA iff there exists a sequence in AA converging to xx. The closure Aˉ\bar{A} coincides with the set of all limit points of AA.

    • Dense Sets: AA is dense in BB (A⊂BA \subset B) iff B⊂AˉB \subset \bar{A}.

    • Cauchy Sequence: ∀ε>0,∃N(ε)\forall \varepsilon > 0, \exists N(\varepsilon) such that d(xn,xm)<εd(x_n, x_m) < \varepsilon for all n,m≥Nn, m \ge N.

    • Every convergent sequence is a Cauchy sequence. A Cauchy sequence with a convergent subsequence converges entirely to the same limit.

    • Complete Metric Space: A space in which every Cauchy sequence converges in the space. The space (C,d)(\mathbb{C}, d) with Euclidean metric is complete. A subspace of a complete space is complete iff it is closed.

    • Compactness: A metric space is sequentially compact if every sequence has a convergent subsequence. Every compact metric space is complete.

    • Total Boundedness: A space is totally bounded if ∀ε>0\forall \varepsilon > 0, it can be covered by a finite number of open balls of radius ε\varepsilon. Total boundedness implies boundedness.

    • Characterization of Compactness: A metric space is compact if and only if it is complete and totally bounded.

    • Heine-Borel Theorem: A subset K⊂RnK \subset \mathbb{R}^n (and specifically K⊂CK \subset \mathbb{C}) is compact if and only if it is closed and bounded.

Limits and Continuity of Complex Functions

  • Definition of Limit: Let (S,d)(S, d) and (Ω,ρ)(\Omega, \rho) be metric spaces and f:S→Ωf: S \to \Omega. The limit lim⁡x→af(x)=ω\lim_{x \to a} f(x) = \omega exists iff:   ∀ε>0,∃δ(ε)>0 such that ρ(f(x),ω)<ε∀x∈B(a,δ)∖{a}\forall \varepsilon > 0, \exists \delta(\varepsilon) > 0 \text{ such that } \rho(f(x), \omega) < \varepsilon \quad \forall x \in B(a, \delta) \setminus \{a\}   The limit point ω\omega, if it exists, is unique.

  • Sequential Criterion for Limits: lim⁡x→af(x)=ω\lim_{x \to a} f(x) = \omega if and only if for every sequence (xn)(x_n) in S∖{a}S \setminus \{a\} such that lim⁡n→∞xn=a\lim_{n\to\infty} x_n = a, it holds that lim⁡n→∞f(xn)=ω\lim_{n\to\infty} f(x_n) = \omega.

  • Composition of Limits: Let f:S1→S2f: S_1 \to S_2 and g:S2→S3g: S_2 \to S_3, with lim⁡x→af(x)=b\lim_{x \to a} f(x) = b and lim⁡y→bg(y)=c\lim_{y \to b} g(y) = c. Then lim⁡x→a(g∘f)(x)=c\lim_{x \to a} (g \circ f)(x) = c if and only if at least one of the following holds:

    • ∃r>0\exists r > 0 such that f(x)≠bf(x) \ne b for all x∈B(a,r)∖{a}x \in B(a, r) \setminus \{a\}.

    • g(b)=cg(b) = c.

  • Complex Limits via Real and Imaginary Parts: For f:C→Cf: \mathbb{C} \to \mathbb{C},   lim⁡z→af(z)=w  ⟺  lim⁡z→aRe(f(z))=Re(w)andlim⁡z→aIm(f(z))=Im(w)\lim_{z \to a} f(z) = w \iff \lim_{z \to a} \text{Re}(f(z)) = \text{Re}(w) \quad \text{and} \quad \lim_{z \to a} \text{Im}(f(z)) = \text{Im}(w)

  • Limit Algebra: If lim⁡x→af(x)=wf\lim_{x \to a} f(x) = w_f and lim⁡x→ag(x)=wg\lim_{x \to a} g(x) = w_g, then:

    • lim⁡x→a(f±g)(x)=wf±wg\lim_{x \to a} (f \pm g)(x) = w_f \pm w_g

    • lim⁡x→a(fg)(x)=wfwg\lim_{x \to a} (fg)(x) = w_f w_g

    • lim⁡x→a(f/g)(x)=wf/wg\lim_{x \to a} (f/g)(x) = w_f / w_g (provided wg≠0w_g \ne 0)

  • Limits Involving the Point at Infinity:   lim⁡z→af(z)=∞  ⟺  lim⁡z→a1f(z)=0\lim_{z \to a} f(z) = \infty \iff \lim_{z \to a} \frac{1}{f(z)} = 0   lim⁡z→∞f(z)=w  ⟺  lim⁡z→0f(1z)=w\lim_{z \to \infty} f(z) = w \iff \lim_{z \to 0} f\left(\frac{1}{z}\right) = w   lim⁡z→∞f(z)=∞  ⟺  lim⁡z→01f(1/z)=0\lim_{z \to \infty} f(z) = \infty \iff \lim_{z \to 0} \frac{1}{f(1/z)} = 0

  • Continuity:

    • A function f:S→Ωf: S \to \Omega is continuous at a∈Sa \in S if lim⁡x→af(x)=f(a)\lim_{x \to a} f(x) = f(a).

    • Global Continuity: ff is continuous on SS iff ∀ε>0,∀x∈S,∃δ(ε,x)>0\forall \varepsilon > 0, \forall x \in S, \exists \delta(\varepsilon, x) > 0 such that ρ(f(x),f(y))<ε\rho(f(x), f(y)) < \varepsilon for all y∈B(x,δ)y \in B(x, \delta).

    • Topological Characterization: ff is continuous on SS if and only if the preimage of every open set in Ω\Omega is open in SS, or equivalently, the preimage of every closed set in Ω\Omega is closed in SS.

    • For f:C→Cf: \mathbb{C} \to \mathbb{C}, ff is continuous at aa iff Re(f)\text{Re}(f) and Im(f)\text{Im}(f) are continuous at aa.

  • Uniform and Lipschitz Continuity:

    • Uniform Continuity: ∀ε>0,∃δ(ε)>0\forall \varepsilon > 0, \exists \delta(\varepsilon) > 0 such that ρ(f(x),f(y))<ε\rho(f(x), f(y)) < \varepsilon whenever d(x,y)<δd(x, y) < \delta, for all x,y∈Sx, y \in S.

    • Lipschitz Continuity: ∃M>0\exists M > 0 such that ρ(f(x),f(y))≤Md(x,y)\rho(f(x), f(y)) \le M d(x, y) for all x,y∈Sx, y \in S.

    • Hierarchy: Lipschitz Continuous  ⟹  Uniformly Continuous  ⟹  Continuous\text{Lipschitz Continuous} \implies \text{Uniformly Continuous} \implies \text{Continuous}.

  • Theorems on Compact and Connected Sets:

    • Preservation: If ff is continuous and KK is compact (connected), then f(K)f(K) is compact (connected).

    • Extreme Value Theorem: If f:K→Rf: K \to \mathbb{R} is continuous and KK is compact, ff attains its absolute maximum and minimum on KK.

    • Modulus Extreme Value Theorem: If f:K→Cf: K \to \mathbb{C} is continuous and KK is compact, ∣f∣|f| attains its absolute maximum and minimum on KK.

    • Local Non-Vanishing Theorem: If f(a)≠0f(a) \ne 0 and ff is continuous at aa, then ∃δ>0\exists \delta > 0 such that f(x)≠0f(x) \ne 0 for all x∈B(a,δ)x \in B(a, \delta).

  • Heine-Cantor Theorem: Every continuous function on a compact metric space is uniformly continuous.

Sequences and Series of Functions and Power Series

  • Convergence of Function Sequences:

    • Pointwise Convergence: (fn)(f_n) converges pointwise to ff on SS if ∀x∈S\forall x \in S, lim⁡n→∞fn(x)=f(x)\lim_{n\to\infty} f_n(x) = f(x).

    • Uniform Convergence: (fn)(f_n) converges uniformly to ff on SS if ∀ε>0,∃N(ε)\forall \varepsilon > 0, \exists N(\varepsilon) such that ρ(fn(x),f(x))<ε\rho(f_n(x), f(x)) < \varepsilon for all n≥Nn \ge N and all x∈Sx \in S.

    • Continuity Preservation Theorem: If a sequence of continuous functions (fn)(f_n) converges uniformly to ff on SS, then ff is continuous on SS.

  • Series of Functions:

    • Partial sums: sn(x)=∑k=1nfk(x)s_n(x) = \sum_{k=1}^n f_k(x).

    • A series converges (uniformly) if the sequence of partial sums sn(x)s_n(x) converges (uniformly).

    • Absolute Convergence: ∑k=1∞fk(x)\sum_{k=1}^\infty f_k(x) converges absolutely if ∑k=1∞∣fk(x)∣\sum_{k=1}^\infty |f_k(x)| converges. In C\mathbb{C}, absolute convergence implies convergence.

    • Weierstrass M-Test: Let (fn)(f_n) be functions on SS. If there exist positive constants MnM_n such that ∣fn(x)∣≤Mn|f_n(x)| \le M_n for all x∈Sx \in S and ∑n=1∞Mn<∞\sum_{n=1}^\infty M_n < \infty, then ∑n=1∞fn(x)\sum_{n=1}^\infty f_n(x) converges uniformly and absolutely on SS.

  • Limit Superior and Limit Inferior of Real Sequences:

    • Definitions:     lim sup⁡n→∞an=lim⁡n→∞(sup⁡k≥nak)\limsup_{n\to\infty} a_n = \lim_{n\to\infty} \left( \sup_{k \ge n} a_k \right)     lim inf⁡n→∞an=lim⁡n→∞(inf⁡k≥nak)\liminf_{n\to\infty} a_n = \lim_{n\to\infty} \left( \inf_{k \ge n} a_k \right)

    • Relation: lim inf⁡n→∞an≤lim sup⁡n→∞an\liminf_{n\to\infty} a_n \le \limsup_{n\to\infty} a_n. Equality holds iff lim⁡n→∞an\lim_{n\to\infty} a_n exists.

    • Summation Properties for Bounded Sequences:     lim inf⁡n→∞an+lim inf⁡n→∞bn≤lim inf⁡n→∞(an+bn)≤lim inf⁡n→∞an+lim sup⁡n→∞bn≤lim sup⁡n→∞(an+bn)≤lim sup⁡n→∞an+lim sup⁡n→∞bn\liminf_{n\to\infty} a_n + \liminf_{n\to\infty} b_n \le \liminf_{n\to\infty} (a_n + b_n) \le \liminf_{n\to\infty} a_n + \limsup_{n\to\infty} b_n \le \limsup_{n\to\infty} (a_n + b_n) \le \limsup_{n\to\infty} a_n + \limsup_{n\to\infty} b_n     If lim⁡n→∞an=a\lim_{n\to\infty} a_n = a, then:     lim sup⁡n→∞(an+bn)=a+lim sup⁡n→∞bn\limsup_{n\to\infty} (a_n + b_n) = a + \limsup_{n\to\infty} b_n     lim inf⁡n→∞(an+bn)=a+lim inf⁡n→∞bn\liminf_{n\to\infty} (a_n + b_n) = a + \liminf_{n\to\infty} b_n

  • Power Series and Abel's Theorem:

    • General form: ∑n=0∞an(z−z0)n\sum_{n=0}^\infty a_n (z - z_0)^n.

    • Geometric Series Archetype: ∑n=0∞zn=11−z\sum_{n=0}^\infty z^n = \frac{1}{1 - z}, converging absolutely for ∣z∣<1|z| < 1, uniformly on B(0,r)B(0, r) for any r<1r < 1, and diverging for ∣z∣>1|z| > 1.

    • Hadamard's Formula: The radius of convergence R∈[0,∞]R \in [0, \infty] is given by:     1R=lim sup⁡n→∞∣an∣1/n\frac{1}{R} = \limsup_{n\to\infty} |a_n|^{1/n}

    • Abel's Theorem on Convergence:

    • If ∣z−z0∣<R|z - z_0| < R, the power series converges absolutely.

    • If ∣z−z0∣>R|z - z_0| > R, the terms are unbounded and the series diverges.

    • For any 0<r<R0 < r < R, the series converges uniformly on Bˉ(z0,r)\bar{B}(z_0, r).

    • Ratio Criterion: If the limit exists,     R=lim⁡n→∞∣anan+1∣R = \lim_{n\to\infty} \left| \frac{a_n}{a_{n+1}} \right|

Derivatives and Analytic Functions

  • Complex Derivative:

    • Let G⊂CG \subset \mathbb{C} be open and a∈Ga \in G. The function f:G→Cf: G \to \mathbb{C} is differentiable at aa if the following limit exists in C\mathbb{C}:     f′(a)=lim⁡z→af(z)−f(a)z−a=lim⁡Δz→0f(a+Δz)−f(a)Δzf'(a) = \lim_{z \to a} \frac{f(z) - f(a)}{z - a} = \lim_{\Delta z \to 0} \frac{f(a + \Delta z) - f(a)}{\Delta z}

    • Differentiability implies continuity at aa.

  • Differentiation Arithmetic:

    • (f±g)′(z)=f′(z)±g′(z)(f \pm g)'(z) = f'(z) \pm g'(z)

    • (fg)′(z)=f′(z)g(z)+f(z)g′(z)(fg)'(z) = f'(z)g(z) + f(z)g'(z)

    • (f/g)′(z)=f′(z)g(z)−f(z)g′(z)(g(z))2(f/g)'(z) = \frac{f'(z)g(z) - f(z)g'(z)}{(g(z))^2}

    • Chain Rule: If ff is differentiable at aa and gg is differentiable at f(a)f(a), then (g∘f)′(a)=g′(f(a))f′(a)(g \circ f)'(a) = g'(f(a)) f'(a).

    • Inverse Function Derivative: If ff is bijective, f−1f^{-1} is continuous, and f′(a)≠0f'(a) \ne 0, then (f−1)′(f(a))=1f′(a)(f^{-1})'(f(a)) = \frac{1}{f'(a)}.

  • Cauchy-Riemann Equations (Cartesian Form):

    • Let f(x+iy)=u(x,y)+iv(x,y)f(x + iy) = u(x, y) + i v(x, y). If ff is differentiable at z0=x0+iy0z_0 = x_0 + i y_0, then the first-order partial derivatives of uu and vv exist and satisfy:     ux(x0,y0)=vy(x0,y0)u_x(x_0, y_0) = v_y(x_0, y_0)     uy(x0,y0)=−vx(x0,y0)u_y(x_0, y_0) = -v_x(x_0, y_0)

    • Derivative representation: f′(z0)=ux(x0,y0)+ivx(x0,y0)=vy(x0,y0)−iuy(x0,y0)f'(z_0) = u_x(x_0, y_0) + i v_x(x_0, y_0) = v_y(x_0, y_0) - i u_y(x_0, y_0).

    • Sufficient Condition Theorem: If ux,uy,vx,vyu_x, u_y, v_x, v_y exist in an open neighborhood of z0z_0, are continuous at (x0,y0)(x_0, y_0), and satisfy the Cauchy-Riemann equations at (x0,y0)(x_0, y_0), then ff is differentiable at z0z_0.

  • Cauchy-Riemann Equations (Polar Form):

    • Let z=reiθz = r e^{i\theta}, and write f(z)=u(r,θ)+iv(r,θ)f(z) = u(r, \theta) + i v(r, \theta). The equations become:     rur(r0,θ0)=vθ(r0,θ0)r u_r(r_0, \theta_0) = v_\theta(r_0, \theta_0)     uθ(r0,θ0)=−rvr(r0,θ0)u_\theta(r_0, \theta_0) = -r v_r(r_0, \theta_0)

    • Derivative in polar form:     f′(z0)=(ur(r0,θ0)+ivr(r0,θ0))e−iθ0f'(z_0) = (u_r(r_0, \theta_0) + i v_r(r_0, \theta_0))e^{-i\theta_0}

  • Analytic and Entire Functions:

    • Analyticity: A function ff is analytic (holomorphic) on an open set GG if it is differentiable at every point of GG. If GG is not open, analyticity on GG means analyticity on some open set containing GG.

    • Entire Function: A function analytic on all of C\mathbb{C}.

    • Singular Point: A point z0z_0 where ff fails to be analytic, but every neighborhood of z0z_0 contains points where ff is analytic.

  • Constancy Theorems on Connected Domains: Let D⊂CD \subset \mathbb{C} be open and connected, and ff analytic in DD:

    • If f′(z)=0f'(z) = 0 for all z∈Dz \in D, then ff is constant.

    • If both ff and fˉ\bar{f} are analytic in DD, then ff is constant.

    • If ∣f(z)∣|f(z)| is constant in DD, then ff is constant.

    • If Re(f)\text{Re}(f) or Im(f)\text{Im}(f) is constant in DD, then ff is constant.

  • Complex Functions of a Real Variable:

    • Let f:[a,b]→Cf: [a, b] \to \mathbb{C}, with f(t)=u(t)+iv(t)f(t) = u(t) + i v(t). Then f′(t)=u′(t)+iv′(t)f'(t) = u'(t) + i v'(t).

    • Failure of the Mean Value Theorem: The real Mean Value Theorem fails for complex-valued functions. For f(t)=eitf(t) = e^{it} on [0,2π][0, 2\pi], f(2π)−f(0)=1−1=0f(2\pi) - f(0) = 1 - 1 = 0, whereas ∣f′(t)∣=∣ieit∣=1≠0|f'(t)| = |i e^{it}| = 1 \ne 0 for all tt.

  • Conformal Transformations:

    • Definition: A transformation ff is conformal at z0z_0 if it preserves both the magnitude and orientation of angles between regular intersecting paths, and the local scale limit exists and is non-zero:     lim⁡z→z0∣f(z)−f(z0)∣∣z−z0∣=∣f′(z0)∣≠0\lim_{z \to z_0} \frac{|f(z) - f(z_0)|}{|z - z_0|} = |f'(z_0)| \ne 0

  • Theorem: If ff is analytic on an open set GG, then ff is conformal at every point z0∈Gz_0 \in G where f′(z0)≠0f'(z_0) \ne 0.

  • Orthogonal Trajectories: If f=u+ivf = u + iv is analytic with f′(z0)≠0f'(z_0) \ne 0, the level curves u(x,y)=c1u(x, y) = c_1 and v(x,y)=c2v(x, y) = c_2 intersect orthogonally.

Elementary Functions in the Complex Plane

  • The Complex Exponential:

    • Definition: For z=x+iyz = x + iy,     exp⁡(z)=ez=ex(cos⁡(y)+isin⁡(y))\exp(z) = e^z = e^x (\cos(y) + i \sin(y))

    • Properties: exp⁡(z1+z2)=exp⁡(z1)exp⁡(z2)\exp(z_1 + z_2) = \exp(z_1)\exp(z_2), exp⁡(z1−z2)=exp⁡(z1)/exp⁡(z2)\exp(z_1 - z_2) = \exp(z_1)/\exp(z_2), and 1/exp⁡(z)=exp⁡(−z)1/\exp(z) = \exp(-z).

    • Analyticity: exp⁡(z)\exp(z) is an entire function with ddzez=ez\frac{d}{dz} e^z = e^z.

    • Modulus and Argument: ∣ez∣=eRe(z)=ex|e^z| = e^{\text{Re}(z)} = e^x, and arg⁡(ez)=Im(z)+2πn=y+2πn\arg(e^z) = \text{Im}(z) + 2\pi n = y + 2\pi n.

    • Periodicity: Entirely periodic with pure imaginary period 2πi2\pi i:     ez+2πi=eze^{z + 2\pi i} = e^z

Real and imaginary parts of exp(z)
  • The Complex Logarithm:

    • Multi-valued Logarithm: For z≠0z \ne 0,     log⁡(z)=ln⁡∣z∣+iarg⁡(z)=ln⁡∣z∣+i(Arg(z)+2πn),n∈Z\log(z) = \ln|z| + i \arg(z) = \ln|z| + i(\text{Arg}(z) + 2\pi n), \quad n \in \mathbb{Z}     where ln⁡(⋅)\ln(\cdot) denotes the standard real natural logarithm. Defined as the inverse of the exponential: exp⁡(log⁡(z))=z\exp(\log(z)) = z.

  • Composition Inverse Ambiguity: log⁡(exp⁡(z))=z+2πin\log(\exp(z)) = z + 2\pi i n.

  • Principal Value: The single-valued function   Log(z)=ln⁡∣z∣+iArg(z),−π<Arg(z)≤π\text{Log}(z) = \ln|z| + i \text{Arg}(z), \quad -\pi < \text{Arg}(z) \le \pi

  • Branches of the Logarithm: A continuous function f:D→Cf: D \to \mathbb{C} (DD open and connected, 0∉D0 \notin D) such that exp⁡(f(z))=z\exp(f(z)) = z for all z∈Dz \in D.

  • Totality of Branches: If ff is a branch on DD, all branches are of the form f(z)+2πinf(z) + 2\pi i n for fixed n∈Zn \in \mathbb{Z}.

  • Analyticity of Branches: Any branch f(z)f(z) is analytic on DD with:   ddzf(z)=1z\frac{d}{dz} f(z) = \frac{1}{z}

  • Principal Branch: Defined on the cut plane D=C∖{z=−t:t∈[0,∞)}D = \mathbb{C} \setminus \{z = -t : t \in [0, \infty)\}. The ray (−∞,0](-\infty, 0] is the branch cut (branch line), and z=0z = 0 is the branch point.

Real and imaginary parts of the principal branch of log(z)Imaginary parts of three consecutive branches of log(z)
  • Powers with Complex Exponents and Bases:

    • Power Function: For c∈Cc \in \mathbb{C} and z≠0z \ne 0,     zc=exp⁡(clog⁡(z))z^c = \exp(c \log(z))

  • Multi-valuedness:

    • If c=n∈Zc = n \in \mathbb{Z}, znz^n is single-valued and entire (for n≥0n \ge 0).

    • If c=n/m∈Qc = n/m \in \mathbb{Q} (irreducible), zn/mz^{n/m} admits exactly mm distinct branches.

    • If c∉Qc \notin \mathbb{Q}, zcz^c has infinitely many branches.

  • Derivative of Principal Branch: On the cut plane, ddzzc=czc−1\frac{d}{dz} z^c = c z^{c-1}.

  • Exponentials with Complex Base: For c∈C∖{0}c \in \mathbb{C} \setminus \{0\}, cz=exp⁡(zlog⁡(c))c^z = \exp(z \log(c)). For a chosen branch of log⁡(c)\log(c), czc^z is an entire function with derivative ddzcz=czlog⁡(c)\frac{d}{dz} c^z = c^z \log(c).

Real and imaginary parts of the principal branch of sqrt(z)
  • Trigonometric Functions:

    • Definitions via Euler formulas:     sin⁡(z)=eiz−e−iz2i,cos⁡(z)=eiz+e−iz2\sin(z) = \frac{e^{iz} - e^{-iz}}{2i}, \quad \cos(z) = \frac{e^{iz} + e^{-iz}}{2}

    • Entireness and Derivatives: Both functions are entire, with ddzsin⁡(z)=cos⁡(z)\frac{d}{dz} \sin(z) = \cos(z) and ddzcos⁡(z)=−sin⁡(z)\frac{d}{dz} \cos(z) = -\sin(z).

    • Periodicity: Both have period 2π2\pi, and anti-period π\pi (sin⁡(z+π)=−sin⁡(z)\sin(z + \pi) = -\sin(z), cos⁡(z+π)=−cos⁡(z)\cos(z + \pi) = -\cos(z)).

    • Real and Imaginary Decompositions (z=x+iyz = x + iy):     sin⁡(z)=sin⁡(x)cosh⁡(y)+icos⁡(x)sinh⁡(y)\sin(z) = \sin(x)\cosh(y) + i \cos(x)\sinh(y)     cos⁡(z)=cos⁡(x)cosh⁡(y)−isin⁡(x)sinh⁡(y)\cos(z) = \cos(x)\cosh(y) - i \sin(x)\sinh(y)

    • Moduli:     ∣sin⁡(z)∣2=sin⁡2(x)+sinh⁡2(y)|\sin(z)|^2 = \sin^2(x) + \sinh^2(y)     ∣cos⁡(z)∣2=cos⁡2(x)+sinh⁡2(y)|\cos(z)|^2 = \cos^2(x) + \sinh^2(y)

    • Unboundedness: Both sin⁡(z)\sin(z) and cos⁡(z)\cos(z) are unbounded in C\mathbb{C}.

    • Zeros: sin⁡(z)=0  ⟺  z=kπ\sin(z) = 0 \iff z = k\pi (k∈Zk \in \mathbb{Z}); cos⁡(z)=0  ⟺  z=π2+kπ\cos(z) = 0 \iff z = \frac{\pi}{2} + k\pi (k∈Zk \in \mathbb{Z}).

    • Tangent: tan⁡(z)=sin⁡(z)cos⁡(z)\tan(z) = \frac{\sin(z)}{\cos(z)}, analytic everywhere except simple poles at z=π2+kπz = \frac{\pi}{2} + k\pi, with ddztan⁡(z)=1cos⁡2(z)\frac{d}{dz} \tan(z) = \frac{1}{\cos^2(z)}.

Real and imaginary parts of sin(z)
  • Hyperbolic Functions:

    • Definitions:     sinh⁡(z)=ez−e−z2,cosh⁡(z)=ez+e−z2\sinh(z) = \frac{e^z - e^{-z}}{2}, \quad \cosh(z) = \frac{e^z + e^{-z}}{2}

    • Derivatives: ddzsinh⁡(z)=cosh⁡(z)\frac{d}{dz} \sinh(z) = \cosh(z), ddzcosh⁡(z)=sinh⁡(z)\frac{d}{dz} \cosh(z) = \sinh(z).

    • Relations to Trigonometric Functions:     sinh⁡(z)=−isin⁡(iz),cosh⁡(z)=cos⁡(iz)\sinh(z) = -i \sin(iz), \quad \cosh(z) = \cos(iz)     sin⁡(iy)=isinh⁡(y),cos⁡(iy)=cosh⁡(y)\sin(iy) = i \sinh(y), \quad \cos(iy) = \cosh(y)

    • Periodicity: Period 2πi2\pi i, anti-period πi\pi i.

    • Real and Imaginary Parts:     sinh⁡(z)=sinh⁡(x)cos⁡(y)+icosh⁡(x)sin⁡(y)\sinh(z) = \sinh(x)\cos(y) + i \cosh(x)\sin(y)     cosh⁡(z)=cosh⁡(x)cos⁡(y)+isinh⁡(x)sin⁡(y)\cosh(z) = \cosh(x)\cos(y) + i \sinh(x)\sin(y)

    • Moduli:     ∣sinh⁡(z)∣2=sinh⁡2(x)+sin⁡2(y)|\sinh(z)|^2 = \sinh^2(x) + \sin^2(y)     ∣cosh⁡(z)∣2=cosh⁡2(x)+sin⁡2(y)|\cosh(z)|^2 = \cosh^2(x) + \sin^2(y)

    • Zeros: sinh⁡(z)=0  ⟺  z=ikπ\sinh(z) = 0 \iff z = i k\pi (k∈Zk \in \mathbb{Z}); cosh⁡(z)=0  ⟺  z=i(π2+kπ)\cosh(z) = 0 \iff z = i\left(\frac{\pi}{2} + k\pi\right) (k∈Zk \in \mathbb{Z}).

Real and imaginary parts of sinh(z)
  • Inverse Trigonometric and Hyperbolic Functions:

    • Multi-valued logarithmic representations:     arcsin⁡(z)=−ilog⁡(iz+1−z2)\arcsin(z) = -i \log\left(iz + \sqrt{1 - z^2}\right)     arccos⁡(z)=−ilog⁡(z+i1−z2)\arccos(z) = -i \log\left(z + i\sqrt{1 - z^2}\right)     arctan⁡(z)=−i2log⁡(i−zi+z)=i2log⁡(1−iz1+iz)\arctan(z) = -\frac{i}{2} \log\left(\frac{i - z}{i + z}\right) = \frac{i}{2} \log\left(\frac{1 - iz}{1 + iz}\right)     arcsinh(z)=log⁡(z+z2+1)\text{arcsinh}(z) = \log\left(z + \sqrt{z^2 + 1}\right)     arccosh(z)=log⁡(z+z2−1)\text{arccosh}(z) = \log\left(z + \sqrt{z^2 - 1}\right)     arctanh(z)=12log⁡(1+z1−z)\text{arctanh}(z) = \frac{1}{2} \log\left(\frac{1 + z}{1 - z}\right)

    • Branch Cuts and Derivatives:

    • Principal branch of arcsin⁡(z)\arcsin(z): Analytic on C∖((−∞,−1]∪[1,∞))\mathbb{C} \setminus ((-\infty, -1] \cup [1, \infty)), with ddzarcsin⁡(z)=11−z2\frac{d}{dz} \arcsin(z) = \frac{1}{\sqrt{1 - z^2}}.

    • Principal branch of arctan⁡(z)\arctan(z): Analytic on C∖((−i∞,−i]∪[i,i∞))\mathbb{C} \setminus ((-i\infty, -i] \cup [i, i\infty)), with ddzarctan⁡(z)=11+z2\frac{d}{dz} \arctan(z) = \frac{1}{1 + z^2}.

Real and imaginary parts of the principal branch of arcsin(z)Real and imaginary parts of the principal branch of arctan(z)

Complex Integration and the Cauchy-Goursat Framework

  • Contour Integrals of Complex Functions of a Real Variable:

    • For continuous f:[a,b]→Cf: [a, b] \to \mathbb{C}, the integral is:     ∫abf(t) dt=∫abRe(f(t)) dt+i∫abIm(f(t)) dt\int_a^b f(t)\,dt = \int_a^b \text{Re}(f(t))\,dt + i \int_a^b \text{Im}(f(t))\,dt

    • Fundamental Theorem of Calculus: If F′(t)=f(t)F'(t) = f(t), then ∫abf(t) dt=F(b)−F(a)\int_a^b f(t)\,dt = F(b) - F(a).

    • Fundamental Modulus Inequality: ∣∫abf(t) dt∣≤∫ab∣f(t)∣ dt\left|\int_a^b f(t)\,dt\right| \le \int_a^b |f(t)|\,dt.

  • Paths, Curves, and Regularity:

    • Path: Continuous mapping γ:[a,b]→C\gamma: [a, b] \to \mathbb{C}.

    • Simple Path: γ\gamma is injective on (a,b)(a, b).

    • Closed Path: γ(a)=γ(b)\gamma(a) = \gamma(b).

    • Simple Closed Path (Jordan Curve): γ(a)=γ(b)\gamma(a) = \gamma(b) and γ\gamma is injective on [a,b)[a, b).

    • Regular Path: Continuous first derivative γ′(t)\gamma'(t). Piecewise regular if [a,b][a, b] can be partitioned into subintervals on each of which γ\gamma is regular.

    • Path Length: Lγ=∫ab∣γ′(t)∣ dtL_\gamma = \int_a^b |\gamma'(t)|\,dt. Length is invariant under reparametrization.

    • Equivalent Paths and Curves: An equivalence class under continuous strictly increasing parameter changes ϕ:[c,d]→[a,b]\phi: [c, d] \to [a, b] with η=γ∘ϕ\eta = \gamma \circ \phi.

    • Jordan Curve Theorem: Every simple closed curve γ\gamma divides C∖{γ}\mathbb{C} \setminus \{\gamma\} into exactly two disjoint open connected sets: a bounded interior Int(γ)\text{Int}(\gamma) and an unbounded exterior Ext(γ)\text{Ext}(\gamma), with ∂Int(γ)=∂Ext(γ)={γ}\partial\text{Int}(\gamma) = \partial\text{Ext}(\gamma) = \{\gamma\}.

    • Positive Orientation: Traversal keeps Int(γ)\text{Int}(\gamma) immediately to the left.

  • Complex Line Integrals and Darboux Inequality:

    • Definition: Along a piecewise regular path γ\gamma,     ∫γf(z) dz=∫abf(γ(t))γ′(t) dt\int_\gamma f(z)\,dz = \int_a^b f(\gamma(t))\gamma'(t)\,dt

    • Path Splitting: If γ=γ1+γ2\gamma = \gamma_1 + \gamma_2, then ∫γ1+γ2f(z) dz=∫γ1f(z) dz+∫γ2f(z) dz\int_{\gamma_1 + \gamma_2} f(z)\,dz = \int_{\gamma_1} f(z)\,dz + \int_{\gamma_2} f(z)\,dz.

    • Path Reversal: ∫−γf(z) dz=−∫γf(z) dz\int_{-\gamma} f(z)\,dz = -\int_\gamma f(z)\,dz.

    • Darboux Inequality (ML-Inequality): If ff is continuous on {γ}\{\gamma\},     ∣∫γf(z) dz∣≤Lγsup⁡z∈{γ}∣f(z)∣\left|\int_\gamma f(z)\,dz\right| \le L_\gamma \sup_{z \in \{\gamma\}} |f(z)|

  • Primitives and Path Independence:

    • Primitive: An analytic function FF in DD such that F′(z)=f(z)F'(z) = f(z).

    • Theorem on Primitives: In an open connected domain DD, the following statements are equivalent:

    • ff admits a primitive FF in DD.

    • For any two points z1,z2∈Dz_1, z_2 \in D, ∫γf(z) dz\int_\gamma f(z)\,dz is independent of the path γ⊂D\gamma \subset D connecting them (∫z1z2f(z) dz=F(z2)−F(z1)\int_{z_1}^{z_2} f(z)\,dz = F(z_2) - F(z_1)).

    • For every closed piecewise regular path γ⊂D\gamma \subset D, ∫γf(z) dz=0\int_\gamma f(z)\,dz = 0.

  • Cauchy-Goursat Theorem:

    • Theorem (Simple Closed Curves): Let ff be analytic on and within a piecewise regular simple closed curve γ\gamma. Then:     ∫γf(z) dz=0\int_\gamma f(z)\,dz = 0

    • Proof Strategy (Goursat Lemma): Meshing Int(γ)\text{Int}(\gamma) into a finite grid of squares QjQ_j; estimating the incremental derivative remainder δj(z)=f(z)−f(zj)z−zj−f′(zj)\delta_j(z) = \frac{f(z) - f(z_j)}{z - z_j} - f'(z_j); cancellation of boundary line integrals over adjacent squares; and bounding the remainder by ε2D(4D+L)\varepsilon \sqrt{2} D (4D + L), vanishing as ε→0\varepsilon \to 0.

  • Homotopy and Simply Connected Domains:

    • Homotopic Curves: Two paths γ0,γ1\gamma_0, \gamma_1 are homotopic in GG (γ0∼γ1\gamma_0 \sim \gamma_1) if there exists a continuous map H(s,t):[a,b]×[0,1]→GH(s, t): [a, b] \times [0, 1] \to G deforming γ0\gamma_0 into γ1\gamma_1.

    • Homotopic to Zero: A closed curve is homotopic to zero (γ∼0\gamma \sim 0) if it can be continuously deformed to a single point in GG.

    • Simply Connected Domain: An open connected domain where every closed curve is homotopic to zero.

    • General Cauchy-Goursat Theorem: If GG is open and simply connected, and ff is analytic in GG, then ∫γf(z) dz=0\int_\gamma f(z)\,dz = 0 for every closed piecewise regular curve γ⊂G\gamma \subset G.

    • Existence of Primitive Corollary: Every analytic function in a simply connected domain admits a primitive.

    • Path Deformation Principle: If γ1,γ2\gamma_1, \gamma_2 are positively oriented simple closed curves with γ1⊂Int(γ2)\gamma_1 \subset \text{Int}(\gamma_2), and ff is analytic in the region between and on them, then:     ∫γ2f(z) dz=∫γ1f(z) dz\int_{\gamma_2} f(z)\,dz = \int_{\gamma_1} f(z)\,dz

    • Multi-Boundary Generalization: For disjoint inner curves γ1,…,γn\gamma_1, \dots, \gamma_n within γ\gamma,     ∫γf(z) dz=∑k=1n∫γkf(z) dz\int_\gamma f(z)\,dz = \sum_{k=1}^n \int_{\gamma_k} f(z)\,dz

  • Cauchy Integral Formula and Higher Derivatives:

    • Fundamental Formula: If ff is analytic on and within a positively oriented simple closed curve γ\gamma, then for all z∈Int(γ)z \in \text{Int}(\gamma),     f(z)=12πi∫γf(w)w−z dwf(z) = \frac{1}{2\pi i} \int_\gamma \frac{f(w)}{w - z}\,dw

    • Higher Derivatives: An analytic function is infinitely differentiable, with derivatives of all orders given by:     f(n)(z)=n!2πi∫γf(w)(w−z)n+1 dw,n=0,1,2,…f^{(n)}(z) = \frac{n!}{2\pi i} \int_\gamma \frac{f(w)}{(w - z)^{n+1}}\,dw, \quad n = 0, 1, 2, \dots

    • Corollary: Every derivative f(n)f^{(n)} of an analytic function is itself analytic.

  • Foundational Corollaries:

    • Morera's Theorem: If ff is continuous in an open connected domain DD and ∫γf(z) dz=0\int_\gamma f(z)\,dz = 0 for all closed piecewise regular curves γ⊂D\gamma \subset D, then ff is analytic in DD.

    • Cauchy Inequalities: If ff is analytic in B(z0,R)B(z_0, R) and bounded by MM (∣f(z)∣≤M|f(z)| \le M), then:     ∣f(n)(z0)∣≤n!MRn,n=0,1,2,…|f^{(n)}(z_0)| \le \frac{n! M}{R^n}, \quad n = 0, 1, 2, \dots

    • Liouville's Theorem: Every bounded entire function on C\mathbb{C} is constant.

    • Fundamental Theorem of Algebra: Every polynomial Pn(z)=a0+a1z+⋯+anznP_n(z) = a_0 + a_1 z + \dots + a_n z^n (an≠0a_n \ne 0, n≥1n \ge 1) has at least one root in C\mathbb{C}. Consequently, it completely factors into linear factors:     Pn(z)=an(z−z1)(z−z2)⋯(z−zn)P_n(z) = a_n (z - z_1)(z - z_2) \cdots (z - z_n)

Taylor and Laurent Series Expansions

  • Taylor Series Expansion:

    • Theorem: Let ff be analytic in the open ball B(z0,R)B(z_0, R). Then for every z∈B(z0,R)z \in B(z_0, R), f(z)f(z) expands uniquely into the convergent Taylor series:     f(z)=∑n=0∞an(z−z0)n,an=f(n)(z0)n!f(z) = \sum_{n=0}^\infty a_n (z - z_0)^n, \quad a_n = \frac{f^{(n)}(z_0)}{n!}

    • Derivation: Uses Cauchy's integral formula along a circular path γ\gamma of radius rr (∣z−z0∣<r<R|z - z_0| < r < R), coupled with the geometric series expansion:     1w−z=∑k=0n−1(z−z0)k(w−z0)k+1+(z−z0)n(w−z)(w−z0)n\frac{1}{w - z} = \sum_{k=0}^{n-1} \frac{(z - z_0)^k}{(w - z_0)^{k+1}} + \frac{(z - z_0)^n}{(w - z)(w - z_0)^n}

    • Maclaurin Series: The special case with expansion center z0=0z_0 = 0.

    • Term-by-Term Integration: If f(z)=∑n=0∞an(z−z0)nf(z) = \sum_{n=0}^\infty a_n (z - z_0)^n in B(z0,R)B(z_0, R) and gg is continuous on a contour γ⊂B(z0,R)\gamma \subset B(z_0, R), then:     ∫γg(z)f(z) dz=∑n=0∞an∫γg(z)(z−z0)n dz\int_\gamma g(z) f(z)\,dz = \sum_{n=0}^\infty a_n \int_\gamma g(z)(z - z_0)^n\,dz

    • Analyticity of Power Series Sum: The sum of a power series is analytic within its disk of convergence.

    • Term-by-Term Differentiation: The derivative of f(z)f(z) within its circle of convergence is given by:     f′(z)=∑n=1∞nan(z−z0)n−1f'(z) = \sum_{n=1}^\infty n a_n (z - z_0)^{n-1}

    • Uniqueness of Taylor Expansion: If ∑n=0∞an(z−z0)n=f(z)\sum_{n=0}^\infty a_n (z - z_0)^n = f(z) within a disk, then necessarily an=f(n)(z0)/n!a_n = f^{(n)}(z_0) / n!.

  • Canonical Maclaurin Series Expansions:

    • ez=∑n=0∞znn!=1+z+z22!+z33!+…(∣z∣<∞)e^z = \sum_{n=0}^\infty \frac{z^n}{n!} = 1 + z + \frac{z^2}{2!} + \frac{z^3}{3!} + \dots \quad (|z| < \infty)

    • sin⁡(z)=∑k=0∞(−1)k(2k+1)!z2k+1=z−z33!+z55!−…(∣z∣<∞)\sin(z) = \sum_{k=0}^\infty \frac{(-1)^k}{(2k+1)!} z^{2k+1} = z - \frac{z^3}{3!} + \frac{z^5}{5!} - \dots \quad (|z| < \infty)

    • cos⁡(z)=∑k=0∞(−1)k(2k)!z2k=1−z22!+z44!−…(∣z∣<∞)\cos(z) = \sum_{k=0}^\infty \frac{(-1)^k}{(2k)!} z^{2k} = 1 - \frac{z^2}{2!} + \frac{z^4}{4!} - \dots \quad (|z| < \infty)

    • sinh⁡(z)=∑k=0∞z2k+1(2k+1)!=z+z33!+z55!+…(∣z∣<∞)\sinh(z) = \sum_{k=0}^\infty \frac{z^{2k+1}}{(2k+1)!} = z + \frac{z^3}{3!} + \frac{z^5}{5!} + \dots \quad (|z| < \infty)

    • cosh⁡(z)=∑k=0∞z2k(2k)!=1+z22!+z44!+…(∣z∣<∞)\cosh(z) = \sum_{k=0}^\infty \frac{z^{2k}}{(2k)!} = 1 + \frac{z^2}{2!} + \frac{z^4}{4!} + \dots \quad (|z| < \infty)

    • 11−z=∑n=0∞zn(∣z∣<1)\frac{1}{1 - z} = \sum_{n=0}^\infty z^n \quad (|z| < 1)

    • 11+z=∑n=0∞(−1)nzn(∣z∣<1)\frac{1}{1 + z} = \sum_{n=0}^\infty (-1)^n z^n \quad (|z| < 1)

    • Log(1+z)=∑n=1∞(−1)n−1nzn(∣z∣<1)\text{Log}(1 + z) = \sum_{n=1}^\infty \frac{(-1)^{n-1}}{n} z^n \quad (|z| < 1)

    • Binomial series: (1+z)c=∑n=0∞(cn)zn(∣z∣<1)(1 + z)^c = \sum_{n=0}^\infty \binom{c}{n} z^n \quad (|z| < 1), where (cn)=c(c−1)⋯(c−n+1)n!\binom{c}{n} = \frac{c(c-1)\cdots(c-n+1)}{n!}.

  • Annulus and Laurent Series Expansion:

    • Annulus: Defined for 0≤R1<R2≤∞0 \le R_1 < R_2 \le \infty by A(z0,R1,R2)={z∈C:R1<∣z−z0∣<R2}A(z_0, R_1, R_2) = \{z \in \mathbb{C} : R_1 < |z - z_0| < R_2\}. When R1=0R_1 = 0, this is the punctured disk B˙(z0,R2)=A(z0,0,R2)\dot{B}(z_0, R_2) = A(z_0, 0, R_2).

    • Theorem: Let ff be analytic in the annulus A(z0,R1,R2)A(z_0, R_1, R_2). Then f(z)f(z) expands into a generalized power series:     f(z)=∑n=0∞an(z−z0)n+∑n=1∞bn(z−z0)n=∑n=−∞∞cn(z−z0)nf(z) = \sum_{n=0}^\infty a_n (z - z_0)^n + \sum_{n=1}^\infty \frac{b_n}{(z - z_0)^n} = \sum_{n=-\infty}^\infty c_n (z - z_0)^n

    • Integral Formulas for Coefficients: For any positively oriented simple closed curve γ⊂A(z0,R1,R2)\gamma \subset A(z_0, R_1, R_2) enclosing z0z_0:     an=12πi∫γf(z)(z−z0)n+1 dz,n=0,1,2,…a_n = \frac{1}{2\pi i} \int_\gamma \frac{f(z)}{(z - z_0)^{n+1}}\,dz, \quad n = 0, 1, 2, \dots     bn=12πi∫γf(z)(z−z0)n−1 dz,n=1,2,3,…b_n = \frac{1}{2\pi i} \int_\gamma f(z)(z - z_0)^{n-1}\,dz, \quad n = 1, 2, 3, \dots     cn=12πi∫γf(z)(z−z0)n+1 dz,n∈Zc_n = \frac{1}{2\pi i} \int_\gamma \frac{f(z)}{(z - z_0)^{n+1}}\,dz, \quad n \in \mathbb{Z}

    • Decomposition: f(z)=fa(z)+fb(z)f(z) = f_a(z) + f_b(z), where the regular part fa(z)=∑n=0∞an(z−z0)nf_a(z) = \sum_{n=0}^\infty a_n (z - z_0)^n converges for ∣z−z0∣<R2|z - z_0| < R_2 and the principal (singular) part fb(z)=∑n=1∞bn(z−z0)−nf_b(z) = \sum_{n=1}^\infty b_n (z - z_0)^{-n} converges for ∣z−z0∣>R1|z - z_0| > R_1.

    • Term-by-term integration and differentiation apply throughout the convergence annulus.

    • Uniqueness of Laurent Series: The Laurent expansion of an analytic function in a given annulus is unique.

  • Algebra of Power Series:

    • Cauchy Product of Two Power Series: If f(z)=∑n=0∞an(z−z0)nf(z) = \sum_{n=0}^\infty a_n (z - z_0)^n on B(z0,R1)B(z_0, R_1) and g(z)=∑n=0∞bn(z−z0)ng(z) = \sum_{n=0}^\infty b_n (z - z_0)^n on B(z0,R2)B(z_0, R_2), then on B(z0,min⁡(R1,R2))B(z_0, \min(R_1, R_2)):     f(z)g(z)=∑n=0∞cn(z−z0)n,cn=∑k=0nakbn−kf(z)g(z) = \sum_{n=0}^\infty c_n (z - z_0)^n, \quad c_n = \sum_{k=0}^n a_k b_{n-k}

    • Ratio of Two Power Series: If g(z)≠0g(z) \ne 0 on B(z0,R)B(z_0, R), then f(z)/g(z)=∑n=0∞dn(z−z0)nf(z)/g(z) = \sum_{n=0}^\infty d_n (z - z_0)^n, where the coefficients are determined recursively by:     dn=1b0(an−∑k=0n−1dkbn−k)d_n = \frac{1}{b_0} \left( a_n - \sum_{k=0}^{n-1} d_k b_{n-k} \right)     with base case d0=a0/b0d_0 = a_0 / b_0.

Singularities, Zeros, and Residue Theory

  • Singularities:

    • Singular point: A point z0z_0 where ff is not analytic, but is analytic at some points in every open ball B(z0,ε)B(z_0, \varepsilon).

    • Isolated Singular Point: A singularity z0z_0 such that ff is analytic in a punctured disk A(z0,0,r)A(z_0, 0, r).

  • The Residue:

    • Definition: In the Laurent expansion of ff about an isolated singularity z0z_0 in A(z0,0,r)A(z_0, 0, r),     f(z)=∑n=0∞an(z−z0)n+∑n=1∞bn(z−z0)nf(z) = \sum_{n=0}^\infty a_n (z - z_0)^n + \sum_{n=1}^\infty \frac{b_n}{(z - z_0)^n}     the coefficient of (z−z0)−1(z - z_0)^{-1} is called the residue of ff at z0z_0:     Resz=z0f(z)=b1=c−1\text{Res}_{z=z_0} f(z) = b_1 = c_{-1}

    • Residue Theorem: Let γ\gamma be a positively oriented simple closed piecewise regular curve. If ff is analytic on and within γ\gamma except for isolated singularities z1,z2,…,zn∈Int(γ)z_1, z_2, \dots, z_n \in \text{Int}(\gamma), then:     ∫γf(z) dz=2πi∑k=1nResz=zkf(z)\int_\gamma f(z)\,dz = 2\pi i \sum_{k=1}^n \text{Res}_{z=z_k} f(z)

    • Residue at Infinity: Let ff be analytic on and outside a closed curve γ\gamma. The residue at infinity is defined such that:     ∫γf(z) dz=2πi Resz=0[1z2f(1z)]\int_\gamma f(z)\,dz = 2\pi i \, \text{Res}_{z=0} \left[ \frac{1}{z^2} f\left(\frac{1}{z}\right) \right]     where Resz=∞f(z)=−Resw=0[1w2f(1w)]=−c−1\text{Res}_{z=\infty} f(z) = -\text{Res}_{w=0} \left[ \frac{1}{w^2} f\left(\frac{1}{w}\right) \right] = -c_{-1}.

  • Classification of Isolated Singularities: Based on the principal part ∑k=1∞bk(z−z0)k\sum_{k=1}^\infty \frac{b_k}{(z - z_0)^k} in the Laurent expansion:

    • Removable Singularity: bk=0b_k = 0 for all k≥1k \ge 1. The principal part vanishes entirely.

    • Pole of Order mm (m∈Nm \in \mathbb{N}): bm≠0b_m \ne 0 and bk=0b_k = 0 for all k>mk > m. If m=1m = 1, it is a simple pole.

  • Essential Singularity: Infinitely many bk≠0b_k \ne 0.

  • Criteria and Computational Formulas for Poles:

    • Analytic Factorization Criterion: An isolated singularity z0z_0 is a pole of order mm if and only if f(z)f(z) can be factored as:     f(z)=g(z)(z−z0)mf(z) = \frac{g(z)}{(z - z_0)^m}     where g(z)g(z) is analytic at z0z_0 and g(z0)≠0g(z_0) \ne 0.

  • General Residue Formula for a Pole of Order mm:   Resz=z0f(z)=g(m−1)(z0)(m−1)!=lim⁡z→z01(m−1)!dm−1dzm−1[(z−z0)mf(z)]\text{Res}_{z=z_0} f(z) = \frac{g^{(m-1)}(z_0)}{(m - 1)!} = \lim_{z \to z_0} \frac{1}{(m - 1)!} \frac{d^{m-1}}{dz^{m-1}} \left[ (z - z_0)^m f(z) \right]

  • Simple Pole Quotient Formula: If f(z)=p(z)q(z)f(z) = \frac{p(z)}{q(z)} with p,qp, q analytic at z0z_0, p(z0)≠0p(z_0) \ne 0, q(z0)=0q(z_0) = 0, and q′(z0)≠0q'(z_0) \ne 0 (a simple zero of qq), then z0z_0 is a simple pole of ff and:     Resz=z0p(z)q(z)=p(z0)q′(z0)\text{Res}_{z=z_0} \frac{p(z)}{q(z)} = \frac{p(z_0)}{q'(z_0)}

  • Double Pole Quotient Formula: If p(z0)≠0p(z_0) \ne 0, q(z0)=q′(z0)=0q(z_0) = q'(z_0) = 0, and q′′(z0)≠0q''(z_0) \ne 0, then z0z_0 is a double pole with residue:     Resz=z0p(z)q(z)=p′(z0)q′′(z0)2−p(z0)q′′′(z0)6(q′′(z0)2)2\text{Res}_{z=z_0} \frac{p(z)}{q(z)} = \frac{p'(z_0)\frac{q''(z_0)}{2} - p(z_0)\frac{q'''(z_0)}{6}}{\left(\frac{q''(z_0)}{2}\right)^2}

  • Zeros of Analytic Functions:

    • Zero of Order mm: Point z0z_0 where f(z0)=f′(z0)=⋯=f(m−1)(z0)=0f(z_0) = f'(z_0) = \dots = f^{(m-1)}(z_0) = 0 and f(m)(z0)≠0f^{(m)}(z_0) \ne 0.

  • Factorization Criterion: z0z_0 is a zero of order mm iff f(z)=(z−z0)mg(z)f(z) = (z - z_0)^m g(z) with g(z)g(z) analytic at z0z_0 and g(z0)≠0g(z_0) \ne 0.

    • Zero Isolation Theorem: Every zero of a non-constant analytic function on an open connected domain is isolated and of finite order.

    • Identity Theorem: Let f,gf, g be analytic in an open connected domain DD. If the set of coincidences {z∈D:f(z)=g(z)}\{z \in D : f(z) = g(z)\} has an accumulation point (limit point) in DD, then f(z)=g(z)f(z) = g(z) identically throughout DD.

  • Singularity Behavior Theorems:

    • Limit Characterization of a Pole: A point z0z_0 is a pole of ff if and only if lim⁡z→z0∣f(z)∣=∞\lim_{z \to z_0} |f(z)| = \infty.

    • Riemann's Theorem on Removable Singularities: If ff has an isolated singularity at z0z_0 and is bounded in some punctured neighborhood A(z0,0,ε)A(z_0, 0, \varepsilon), then z0z_0 is a removable singularity. Setting f(z0)=lim⁡z→z0f(z)f(z_0) = \lim_{z \to z_0} f(z) extends ff analytically to z0z_0.

    • Casorati-Weierstrass Theorem: If z0z_0 is an essential singularity of ff, then for any δ>0\delta > 0, the image of the punctured disk f(A(z0,0,δ))f(A(z_0, 0, \delta)) is dense in C\mathbb{C}. That is, for every w∈Cw \in \mathbb{C} and every ε>0\varepsilon > 0, there exists z∈A(z0,0,δ)z \in A(z_0, 0, \delta) such that ∣f(z)−w∣<ε|f(z) - w| < \varepsilon.

Applications of the Residue Calculus

  • Improper Integrals and Principal Value:

    • Cauchy Principal Value on R\mathbb{R}:     PV∫−∞+∞f(x) dx=lim⁡R→∞∫−RRf(x) dx\text{PV} \int_{-\infty}^{+\infty} f(x)\,dx = \lim_{R \to \infty} \int_{-R}^R f(x)\,dx

    • Cauchy Principal Value around an Interior Singularity c∈(a,b)c \in (a, b):     PV∫abf(x) dx=lim⁡r→0+(∫ac−rf(x) dx+∫c+rbf(x) dx)\text{PV} \int_a^b f(x)\,dx = \lim_{r \to 0^+} \left( \int_a^{c-r} f(x)\,dx + \int_{c+r}^b f(x)\,dx \right)

  • Integrals of Trigonometric Functions over [0,2π][0, 2\pi]:

    • Form: ∫02πR(cos⁡(θ),sin⁡(θ)) dθ\int_0^{2\pi} R(\cos(\theta), \sin(\theta))\,d\theta.

    • Parametric Transformation: Substitute z=eiθz = e^{i\theta}, dθ=dzizd\theta = \frac{dz}{iz}, cos⁡(θ)=z+z−12\cos(\theta) = \frac{z + z^{-1}}{2}, sin⁡(θ)=z−z−12i\sin(\theta) = \frac{z - z^{-1}}{2i}.

    • Contour: Transformed into a contour integral along the positively oriented unit circle γ:∣z∣=1\gamma: |z| = 1:     ∫02πR(cos⁡(θ),sin⁡(θ)) dθ=∫∣z∣=1R(z+z−12,z−z−12i)dziz=2πi∑∣zk∣<1Resz=zkR~(z)\int_0^{2\pi} R(\cos(\theta), \sin(\theta))\,d\theta = \int_{|z|=1} R\left( \frac{z + z^{-1}}{2}, \frac{z - z^{-1}}{2i} \right) \frac{dz}{iz} = 2\pi i \sum_{|z_k| < 1} \text{Res}_{z=z_k} \tilde{R}(z)

  • Integrals of Rational Functions on (−∞,+∞)(-\infty, +\infty):

    • Form: ∫−∞+∞Pn(x)Qm(x) dx\int_{-\infty}^{+\infty} \frac{P_n(x)}{Q_m(x)}\,dx, with Qm(x)≠0Q_m(x) \ne 0 on R\mathbb{R} and m≥n+2m \ge n + 2.

    • Semicircular Upper Contour: γ=λR+γR\gamma = \lambda_R + \gamma_R, where λR(x)=x\lambda_R(x) = x (−R≤x≤R-R \le x \le R) and γR(θ)=Reiθ\gamma_R(\theta) = R e^{i\theta} (0≤θ≤π0 \le \theta \le \pi).

    • Result: By the ML-inequality, lim⁡R→∞∫γRPn(z)Qm(z) dz=0\lim_{R\to\infty} \int_{\gamma_R} \frac{P_n(z)}{Q_m(z)}\,dz = 0, yielding:     ∫−∞+∞Pn(x)Qm(x) dx=2πi∑Im(zk)>0Resz=zkPn(z)Qm(z)\int_{-\infty}^{+\infty} \frac{P_n(x)}{Q_m(x)}\,dx = 2\pi i \sum_{\text{Im}(z_k) > 0} \text{Res}_{z=z_k} \frac{P_n(z)}{Q_m(z)}

    • Conjugate Evaluation: Equivalently, integrating along the lower half-plane gives −2πi∑Im(zk)<0Resz=zkPn(z)Qm(z)-2\pi i \sum_{\text{Im}(z_k) < 0} \text{Res}_{z=z_k} \frac{P_n(z)}{Q_m(z)}.

  • Integrals of Rational Functions with Trigonometric Factors:

    • Jordan's Lemma: Let ff be analytic in {z∈C:∣z∣>R0,Im(z)≥0}\{z \in \mathbb{C} : |z| > R_0, \text{Im}(z) \ge 0\}. If lim⁡R→∞MR=0\lim_{R\to\infty} M_R = 0, where MR=sup⁡z∈{γR}∣f(z)∣M_R = \sup_{z \in \{\gamma_R\}} |f(z)|, then for all a>0a > 0:     lim⁡R→∞∫γRf(z)eiaz dz=0\lim_{R\to\infty} \int_{\gamma_R} f(z) e^{iaz}\,dz = 0

    • Derivation employs Jordan's inequality: sin⁡(θ)≥2θπ\sin(\theta) \ge \frac{2\theta}{\pi} for θ∈[0,π/2]\theta \in [0, \pi/2].

    • Integration Formula: If m≥n+1m \ge n + 1 and Qm(x)≠0Q_m(x) \ne 0 on R\mathbb{R}, then for a>0a > 0:     PV∫−∞+∞Pn(x)Qm(x)eiax dx=2πi∑Im(zk)>0Resz=zk[Pn(z)Qm(z)eiaz]\text{PV} \int_{-\infty}^{+\infty} \frac{P_n(x)}{Q_m(x)} e^{iax}\,dx = 2\pi i \sum_{\text{Im}(z_k) > 0} \text{Res}_{z=z_k} \left[ \frac{P_n(z)}{Q_m(z)} e^{iaz} \right]

    • Rectangular Path Proof: A rectangular box with corners −R1,R2,R2+i(R1+R2),−R1+i(R1+R2)-R_1, R_2, R_2 + i(R_1+R_2), -R_1 + i(R_1+R_2) proves convergence in the ordinary improper sense.

  • Indented Paths Around Simple Poles on the Real Axis:

    • Fractional Residue Lemma: If ff has a simple pole at x0∈Rx_0 \in \mathbb{R}, and γr±(θ)=x0+re±iθ\gamma_r^\pm(\theta) = x_0 + r e^{\pm i\theta} (−π≤θ≤0-\pi \le \theta \le 0), then:     lim⁡r→0∫γr±f(z) dz=±πi Resz=x0f(z)\lim_{r \to 0} \int_{\gamma_r^\pm} f(z)\,dz = \pm \pi i \, \text{Res}_{z=x_0} f(z)

    • Indented Formula (Rational):     PV∫−∞+∞Pn(x)Qm(x) dx=2πi∑Im(zk)>0Resz=zkPn(z)Qm(z)+πi∑xj∈RResz=xjPn(z)Qm(z)\text{PV} \int_{-\infty}^{+\infty} \frac{P_n(x)}{Q_m(x)}\,dx = 2\pi i \sum_{\text{Im}(z_k) > 0} \text{Res}_{z=z_k} \frac{P_n(z)}{Q_m(z)} + \pi i \sum_{x_j \in \mathbb{R}} \text{Res}_{z=x_j} \frac{P_n(z)}{Q_m(z)}

    • Indented Formula (Fourier / Oscillatory, a>0a > 0):     PV∫−∞+∞Pn(x)Qm(x)eiax dx=2πi∑Im(zk)>0Resz=zk[Pn(z)Qm(z)eiaz]+πi∑xj∈RResz=xj[Pn(z)Qm(z)eiaz]\text{PV} \int_{-\infty}^{+\infty} \frac{P_n(x)}{Q_m(x)} e^{iax}\,dx = 2\pi i \sum_{\text{Im}(z_k) > 0} \text{Res}_{z=z_k} \left[ \frac{P_n(z)}{Q_m(z)} e^{iaz} \right] + \pi i \sum_{x_j \in \mathbb{R}} \text{Res}_{z=x_j} \left[ \frac{P_n(z)}{Q_m(z)} e^{iaz} \right]

  • Branch-Cut Contours and Multi-valued Integrals:

    • Keyhole Contour: Used for integrals of the form ∫0∞xaR(x) dx\int_0^\infty x^a R(x)\,dx (−1<a<0-1 < a < 0 or non-integer) or ∫0∞ln⁡(x)R(x) dx\int_0^\infty \ln(x) R(x)\,dx.

    • The contour surrounds the branch cut along [0,∞)[0, \infty) with an upper line segment x+i0+x + i0^+, a large circle γR\gamma_R, a lower returning line segment x−i0−x - i0^-, and an inner indenting circle γr\gamma_r around the origin.

    • Sector / Wedge Contours: Employed for symmetries such as zn+1z^n + 1, spanning angles [0,2π/n][0, 2\pi / n].

    • Fresnel Integrals Evaluation: Integrating eiz2e^{iz^2} along a sector contour of angle π/4\pi/4 proves:     ∫0∞cos⁡(x2) dx=∫0∞sin⁡(x2) dx=12π2\int_0^\infty \cos(x^2)\,dx = \int_0^\infty \sin(x^2)\,dx = \frac{1}{2} \sqrt{\frac{\pi}{2}}

  • Bromwich Integral (Inverse Laplace Transform):

    • Definition: For g(z)g(z) analytic in C\mathbb{C} except for isolated singularities with max⁡kRe(zk)<x0\max_k \text{Re}(z_k) < x_0:     f(t)=12πilim⁡y0→∞∫x0−iy0x0+iy0eztg(z) dz,t>0f(t) = \frac{1}{2\pi i} \lim_{y_0 \to \infty} \int_{x_0 - i y_0}^{x_0 + i y_0} e^{zt} g(z)\,dz, \quad t > 0

    • Closed Contour Evaluation: By completing the vertical Bromwich path with a leftward semicircle γy0(θ)=x0+y0eiθ\gamma_{y_0}(\theta) = x_0 + y_0 e^{i\theta} (π/2≤θ≤3π/2\pi/2 \le \theta \le 3\pi/2), if lim⁡y0→∞sup⁡z∈{γy0}∣g(z)∣=0\lim_{y_0 \to \infty} \sup_{z \in \{\gamma_{y_0}\}} |g(z)| = 0, then:     f(t)=∑k=1NResz=zk[eztg(z)],t>0f(t) = \sum_{k=1}^N \text{Res}_{z=z_k} \left[ e^{zt} g(z) \right], \quad t > 0

    • Higher-Order Pole Inversion: If g(z)g(z) has a pole of order mm at zkz_k with Laurent principal part ∑j=1mbj(z−zk)j\sum_{j=1}^m \frac{b_j}{(z - z_k)^j}, then:     Resz=zk[eztg(z)]=ezkt(b1+b2t1!+b3t22!+⋯+bmtm−1(m−1)!)\text{Res}_{z=z_k} [e^{zt} g(z)] = e^{z_k t} \left( b_1 + b_2 \frac{t}{1!} + b_3 \frac{t^2}{2!} + \dots + b_m \frac{t^{m-1}}{(m - 1)!} \right)

Further Properties of Analytic Functions and Asymptotic Methods

  • Analytic Continuation:

    • Definition: If f1f_1 is analytic in open connected D1D_1 and f2f_2 is analytic in open connected D2D_2 with D1∩D2≠∅D_1 \cap D_2 \ne \emptyset, such that f1(z)=f2(z)f_1(z) = f_2(z) for all z∈D1∩D2z \in D_1 \cap D_2, then f2f_2 is the unique analytic continuation of f1f_1 to D2D_2.

    • Borel Analytic Continuation: For f1(z)=∑k=0∞akzkf_1(z) = \sum_{k=0}^\infty a_k z^k on B(0,R)B(0, R), construct the entire function h(t)=∑k=0∞akk!tkh(t) = \sum_{k=0}^\infty \frac{a_k}{k!} t^k. The analytic continuation is:     f2(z)=∫0∞e−th(zt) dtf_2(z) = \int_0^\infty e^{-t} h(zt)\,dt     valid on the half-plane where the integral converges.

    • Weierstrass Analytic Continuation: Chain of overlapping disks along a polygonal path PP. Expanding in successive Taylor series centered at points z0,z1,…,zm=zz_0, z_1, \dots, z_m = z.

    • Schwarz Reflection Principle: Let DD be an open connected region symmetric across the real axis (D=D∗={z:zˉ∈D}D = D^* = \{z : \bar{z} \in D\}). Let D+={z∈D:Im(z)>0}D_+ = \{z \in D : \text{Im}(z) > 0\} and D0={z∈D:Im(z)=0}D_0 = \{z \in D : \text{Im}(z) = 0\}. If ff is continuous on D+∪D0D_+ \cup D_0, analytic on D+D_+, and purely real on D0D_0, then the function     g(z)={f(z)z∈D+∪D0f(zˉ)‾z∈D−g(z) = \begin{cases} f(z) & z \in D_+ \cup D_0 \\ \overline{f(\bar{z})} & z \in D_- \end{cases}     is analytic throughout all of DD.

    • Monodromy Theorem: Let GG be a simply connected open set and ff an analytic function element in a neighborhood of z1∈Gz_1 \in G. If ff can be continued analytically along every path in GG starting at z1z_1, then the analytic continuation is single-valued and defines an analytic function on GG.

  • Maximum and Minimum Modulus Principles:

    • Gauss Mean Value Theorem: If ff is analytic in B(z0,r)B(z_0, r),     f(z0)=12π∫02πf(z0+ρeit) dt(0<ρ<r)f(z_0) = \frac{1}{2\pi} \int_0^{2\pi} f(z_0 + \rho e^{it})\,dt \quad (0 < \rho < r)

    • Maximum Modulus Principle: Let ff be analytic and non-constant in an open connected domain DD. Then ∣f(z)∣|f(z)| attains no local maximum in DD.

    • Boundary Maximum Corollary: If DD is bounded and ff is continuous on Dˉ\bar{D} and analytic on DD, the maximum of ∣f(z)∣|f(z)| on Dˉ\bar{D} is attained strictly on the boundary ∂D\partial D.

    • Minimum Modulus Principle: If ff is analytic, non-constant, and has no zeros in DD, then ∣f(z)∣|f(z)| attains no local minimum in DD. The minimum is achieved on ∂D\partial D.

    • Real and Imaginary Parts: The functions Re(f)\text{Re}(f) and Im(f)\text{Im}(f) attain their absolute maxima and minima on ∂D\partial D.

  • Meromorphic Functions and the Argument Principle:

    • Meromorphic Function: A function analytic on an open connected domain DD except for isolated poles of finite order.

    • Finiteness of Zeros and Poles: Within any bounded region bounded by a simple closed curve on which ff is analytic and non-zero, the number of zeros and poles is finite.

    • Generalized Argument Principle: If ff is meromorphic inside and on γ\gamma, and gg is analytic inside and on γ\gamma, then:     12πi∫γg(z)f′(z)f(z) dz=∑k=1mαkg(ak)−∑k=1nβkg(bk)\frac{1}{2\pi i} \int_\gamma g(z) \frac{f'(z)}{f(z)}\,dz = \sum_{k=1}^m \alpha_k g(a_k) - \sum_{k=1}^n \beta_k g(b_k)     where aka_k are zeros of order αk\alpha_k and bkb_k are poles of order βk\beta_k.

    • Index / Winding Number: The index of a closed curve γ\gamma with respect to z0∉{γ}z_0 \notin \{\gamma\} is an integer:     n(γ,z0)=12πi∫γdzz−z0∈Zn(\gamma, z_0) = \frac{1}{2\pi i} \int_\gamma \frac{dz}{z - z_0} \in \mathbb{Z}

    • Standard Argument Principle (g(z)=1g(z) = 1):     n(f∘γ,0)=12πi∫γf′(z)f(z) dz=Z−Pn(f \circ \gamma, 0) = \frac{1}{2\pi i} \int_\gamma \frac{f'(z)}{f(z)}\,dz = Z - P     where ZZ and PP are the total numbers of zeros and poles of ff inside γ\gamma, counted with multiplicities.

    • Inverse Function Formula: If ff is analytic in B(z0,R)B(z_0, R) and bijective onto f(B(z0,R))f(B(z_0, R)), then for any w∈f(B(z0,R))w \in f(B(z_0, R)),     f−1(w)=12πi∫∣z−z0∣=Rzf′(z)f(z)−w dzf^{-1}(w) = \frac{1}{2\pi i} \int_{|z-z_0|=R} z \frac{f'(z)}{f(z) - w}\,dz

    • Abel-Plana Summation Formula:     ∑n=0∞f(n)−∫0∞f(x) dx=12f(0)+i∫0∞f(it)−f(−it)e2πt−1 dt\sum_{n=0}^\infty f(n) - \int_0^\infty f(x)\,dx = \frac{1}{2}f(0) + i \int_0^\infty \frac{f(it) - f(-it)}{e^{2\pi t} - 1}\,dt

    • Rouché's Theorem: Let ff and gg be analytic on and within a simple closed curve γ\gamma. If the strict inequality holds on the boundary:     ∣f(z)∣>∣g(z)∣∀z∈{γ}|f(z)| > |g(z)| \quad \forall z \in \{\gamma\}     then ff and f+gf + g have the exact same number of zeros inside γ\gamma, counting multiplicities.

  • Harmonic Functions and Harmonic Conjugates:

    • Definition: A function u:D→Ru: D \to \mathbb{R} is harmonic if it is C2C^2 and satisfies Laplace's equation:     Δu=uxx+uyy=0\Delta u = u_{xx} + u_{yy} = 0

    • Conjugate Harmonic Function: vv is the harmonic conjugate of uu in DD iff f(x+iy)=u(x,y)+iv(x,y)f(x + iy) = u(x, y) + i v(x, y) is analytic in DD.

    • Theorem: An analytic function has real and imaginary parts that are mutually conjugate harmonic functions.

    • Existence: In a simply connected domain DD, every harmonic function uu admits a harmonic conjugate vv, unique up to an additive constant:     v(x,y)=∫(x0,y0)(x,y)(−ut(s,t) ds+us(s,t) dt)v(x, y) = \int_{(x_0, y_0)}^{(x, y)} \left( -u_t(s, t)\,ds + u_s(s, t)\,dt \right)

    • Simplified Star-Domain Formula (centered at (0,0)(0, 0)):     v(x,y)=−∫0xuy(s,0) ds+∫0yux(x,t) dt+Cv(x, y) = -\int_0^x u_y(s, 0)\,ds + \int_0^y u_x(x, t)\,dt + C

  • Asymptotic Methods for Integrals:

    • Laplace's Method (Real Integrals): For I(λ)=∫abeλf(t)g(t) dtI(\lambda) = \int_a^b e^{\lambda f(t)} g(t)\,dt where f(t)f(t) has a unique maximum at an interior point t0∈(a,b)t_0 \in (a, b) with f′′(t0)<0f''(t_0) < 0 and g(t0)≠0g(t_0) \ne 0:     I(λ)=eλf(t0)(2πλ∣f′′(t0)∣g(t0)+O(λ−3/2))(λ→∞)I(\lambda) = e^{\lambda f(t_0)} \left( \sqrt{\frac{2\pi}{\lambda |f''(t_0)|}} g(t_0) + O(\lambda^{-3/2}) \right) \quad (\lambda \to \infty)

  • Boundary Maximum Variation: If the maximum is attained at an endpoint (e.g. t0=at_0 = a):   I(λ)=eλf(a)(π2λ∣f′′(a)∣g(a)+O(λ−1))I(\lambda) = e^{\lambda f(a)} \left( \sqrt{\frac{\pi}{2\lambda |f''(a)|}} g(a) + O(\lambda^{-1}) \right)

  • Stirling's Factorial Approximation: Applying Laplace's method to n!=nn+1∫0∞en(ln⁡(t)−t) dtn! = n^{n+1} \int_0^\infty e^{n(\ln(t) - t)}\,dt gives:   n!=2πn(ne)n(1+O(n−1))n! = \sqrt{2\pi n} \left(\frac{n}{e}\right)^n \left( 1 + O(n^{-1}) \right)

  • Saddle Point (Steepest Descent) Method (Complex Integrals): For I(λ)=∫γeλf(z)g(z) dzI(\lambda) = \int_\gamma e^{\lambda f(z)} g(z)\,dz where f(z)f(z) has a simple saddle point at z0z_0 (f′(z0)=0f'(z_0) = 0, f′′(z0)≠0f''(z_0) \ne 0):

  • Local geometry: z0z_0 is never an isolated local extremum of Re(f(z))\text{Re}(f(z)), but a saddle point.

  • The paths of steepest descent are orthogonal to the level lines of Re(f(z))\text{Re}(f(z)) and maintain Im(f(z))=constant\text{Im}(f(z)) = \text{constant}.

  • Asymptotic Formula:   I(λ)=eλf(z0)2πλ∣f′′(z0)∣g(z0)ei(π2−12Arg(f′′(z0)))+O(λ−3/2)(λ→∞)I(\lambda) = e^{\lambda f(z_0)} \sqrt{\frac{2\pi}{\lambda |f''(z_0)|}} g(z_0) e^{i\left( \frac{\pi}{2} - \frac{1}{2}\text{Arg}(f''(z_0)) \right)} + O(\lambda^{-3/2}) \quad (\lambda \to \infty)

Graph of Re(z^2) displaying a saddle point at the origin
  • Applications to Statistical Physics:

    • Spin-1/2 System Density of States: Microstates of energy E=−J∑j=1NσjE = -J \sum_{j=1}^N \sigma_j with σj=±1\sigma_j = \pm 1. The density of states in the thermodynamic limit N→∞N \to \infty with ε=E/(NJ)\varepsilon = E / (NJ) constant:     Ω(E)=2N2π∫−∞+∞eNf(k) dk,f(k)=ikEN+log⁡(cos⁡(kJ))\Omega(E) = \frac{2^N}{2\pi} \int_{-\infty}^{+\infty} e^{N f(k)}\,dk, \quad f(k) = i k \frac{E}{N} + \log(\cos(kJ))     Saddle point evaluation yields the entropy formula:     1Nln⁡(Ω(E)J)=ln⁡(2)−12(1+ε)ln⁡(1+ε)−12(1−ε)ln⁡(1−ε)+O(ln⁡(N)N)\frac{1}{N} \ln(\Omega(E) J) = \ln(2) - \frac{1}{2}(1 + \varepsilon)\ln(1 + \varepsilon) - \frac{1}{2}(1 - \varepsilon)\ln(1 - \varepsilon) + O\left(\frac{\ln(N)}{N}\right)     which vanishes at the extremal energy bounds ε=±1\varepsilon = \pm 1.

    • Quantum Harmonic Oscillators: For NN independent oscillators with energy E=∑j=1Nℏω(nj+1/2)E = \sum_{j=1}^N \hbar\omega(n_j + 1/2), saddle point integration evaluates the dominant density of states in terms of ε=2E/(Nℏω)\varepsilon = 2E / (N\hbar\omega), reproducing the thermodynamic entropy.

Detailed Branch Cut Analyses and Solved Case Studies

  • Branch Cut Analysis of Inverse Trigonometric Functions:

    • Principal branch of arccos⁡(z)=−ilog⁡(z+i1−z2)\arccos(z) = -i \log(z + i\sqrt{1 - z^2}): Non-analytic along the real intervals (−∞,−1](-\infty, -1] and [1,∞)[1, \infty), where the argument of the logarithm and root cross real negative thresholds.

Real and imaginary parts of the principal branch of arccos(z)
  • Branch Cuts of Algebraic Roots:

    • The function z2−1=exp⁡(12log⁡(z2−1))\sqrt{z^2 - 1} = \exp\left(\frac{1}{2}\log(z^2 - 1)\right): The principal branch requires z2−1=−uz^2 - 1 = -u (u≥0u \ge 0). For 0≤u<10 \le u < 1, z(u)=±1−uz(u) = \pm \sqrt{1 - u} (the real segment [−1,1][-1, 1]). For u≥1u \ge 1, z(u)=±iu−1z(u) = \pm i\sqrt{u - 1} (the entire imaginary axis).

Real and imaginary parts of the principal branch of sqrt(z^2 - 1)
  • Branch Cut of i1−z2i\sqrt{1 - z^2}:

    • Non-analyticity requires 1−z2=−u1 - z^2 = -u (u≥0u \ge 0), yielding z(u)=±1+uz(u) = \pm \sqrt{1 + u}. This forms branch cuts along the two rays on the real axis: (−∞,−1](-\infty, -1] and [1,∞)[1, \infty).

Real and imaginary parts of the principal branch of i sqrt(1 - z^2)
  • Branch Cuts of Iterated Logarithmic Functions:

    • The function log⁡(log⁡2(z))\log(\log^2(z)): Discontinuities arise when z=−tz = -t (t≥0t \ge 0) and when log⁡2(z)=−u\log^2(z) = -u (u≥0u \ge 0), giving log⁡(z)=±iu\log(z) = \pm i \sqrt{u}, which implies ∣z∣=1|z| = 1. The resulting domain of analyticity is D=C∖({z:∣z∣=1}∪{z=−t:t≥0})D = \mathbb{C} \setminus (\{z : |z| = 1\} \cup \{z = -t : t \ge 0\}).

Real and imaginary parts of the principal branch of log(log^2(z))
  • Branch Cuts of Algebraic Quotients:

    • The function f(z)=1−zi+z2f(z) = \frac{\sqrt{1-z}}{i + z^2}: Branch cut along the real ray [1,∞)[1, \infty) due to 1−z\sqrt{1-z}, and two isolated simple poles at the zeros of the denominator: z2=−i  ⟹  z=e−iπ/4z^2 = -i \implies z = e^{-i\pi/4} and z=ei3π/4z = e^{i3\pi/4}.

Real and imaginary parts of the principal branch of sqrt(1-z)/(i+z^2)
  • Four-Fold Branch Cuts:

    • The function 4+z4=exp⁡(12log⁡(4+z4))\sqrt{4 + z^4} = \exp\left(\frac{1}{2}\log(4 + z^4)\right): Non-analyticity occurs when 4+z4=−t4 + z^4 = -t (t≥0t \ge 0), yielding z(t)=(4+t)1/4ei(π+2kπ)/4z(t) = (4 + t)^{1/4} e^{i(\pi + 2k\pi)/4} for k=0,1,2,3k = 0, 1, 2, 3. These form four infinite rays emanating from points at distance 2\sqrt{2} from the origin bisecting the four quadrants.

Real and imaginary parts of the principal branch of sqrt(4+z^4)
  • Root-Cosine Trigonometric Combinations:

    • The function f(z)=sin⁡(z)cos⁡(z)f(z) = \frac{\sin(z)}{\sqrt{\cos(z)}}: Analyticity fails at the branch cut of log⁡(cos⁡(z))\log(\cos(z)), given by cos⁡(z)=−t\cos(z) = -t (t≥0t \ge 0). For 0≤t≤10 \le t \le 1, this yields real segments [π/2+2kπ,3π/2+2kπ][\pi/2 + 2k\pi, 3\pi/2 + 2k\pi]. For t>1t > 1, it yields vertical lines parallel to the imaginary axis emanating from (2k+1)π(2k + 1)\pi.

Real and imaginary parts of the principal branch of sin(z)/sqrt(cos(z))
  • Non-Analytic Functions with Directional Derivatives:

    • The function g(z)=log⁡((3z2−zˉ2)z/2)g(z) = \log((3z^2 - \bar{z}^2)z/2): Setting z=x+iyz = x + iy, the inner polynomial is x3+3xy2+i(y3+3x2y)x^3 + 3xy^2 + i(y^3 + 3x^2y). Cauchy-Riemann equations hold only on the coordinate axes x=0x = 0 and y=0y = 0. Because these axes contain no open neighborhoods where the Cauchy-Riemann equations hold, the function is differentiable only along the axes and nowhere analytic.

Real and imaginary parts of log((3z^2 - conjugate(z)^2)z/2)
  • Branch Cuts of Modified Root Angles:

    • The branch of cos⁡(z)\sqrt{\cos(z)} chosen with branch cut along the positive imaginary axis satisfies cos⁡(z)=it\cos(z) = it (t≥0t \ge 0). In Cartesian coordinates, cos⁡(x)cosh⁡(y)=0\cos(x)\cosh(y) = 0 and sin⁡(x)sinh⁡(y)=−t\sin(x)\sinh(y) = -t. This generates semi-infinite vertical cuts along x=π/2+2kπx = \pi/2 + 2k\pi (y≤0y \le 0) and x=3π/2+2kπx = 3\pi/2 + 2k\pi (y≥0y \ge 0).

Real and imaginary parts of the branch of sqrt(cos(z)) cut along the positive imaginary axis
  • Branch Lines of log⁡(cos⁡(z))\log(\cos(z)):

    • For the principal branch of log⁡(cos⁡(z))\log(\cos(z)), the cut equation cos⁡(z)=−t\cos(z) = -t (t≥0t \ge 0) decomposes into periodic segments on the real axis [(2k+1/2)π,(2k+3/2)π][(2k + 1/2)\pi, (2k + 3/2)\pi] and semi-infinite vertical lines emerging from (2k+1)π(2k + 1)\pi.

Real and imaginary parts of the principal branch of log(cos(z))