Mathematics B3200: Complex Analysis II Spring Semester 2023
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
Mathematics B3200: Complex Analysis II
Spring Semester 2023
Questions: 1 May 2023
Directions. These questions are motivated by the class discussion. Throughout the statements of he questions we will use the notation and assumptions established in the class meetings .
1. Let f (x, y) = u(x, y) + iv(x, y) be a complex-valued function of two real-
variables x, y, and assume that f has a domain of definition which includes an open set D 仁 R2 . Let z = x + iy and = x iy . The Cauchy-Riemann equations can be written in two ways.
a)
∂u ∂v ∂u ∂v
∂x ∂y ∂y ∂x .
b)
∂f
= 0.
∂
Prove that the two formulations of the Cauchy-Riemann equations are equivalent.
2. Let u = u(x, y) and v = v(x, y) be two continuous functions which are defined fo4 (x, y) in an open domain D 仁 R2 such that (x, y) !f (u, v) is a bijection. Recall that the Jacobian matrix J (u, v; x, y) is defined by
J (u, v; x, y) := = ( ) .
a) Compute the J (r, θ; x, y) when transforming from Cartesian coordi- nates (x, y) to polar coordinates (r, θ).
b) Compute the J (x, y; r, θ) when transforming from polar coordinates (r, θ) Cartesian coordinates (x, y).
c) Compute the matrix product J (x, y; r, θ)J (r, θ; x, y).
3. Let f (z) = u(x, y) + iv(x, y) be a holomorphic function which is defined for z in an open domain in R2 . Use the Cauchy-Riemann equations to prove the following. At every point z 2 D where ∂f/∂z 0, one can write
J (u, v; x, y) = A · ( s(o)i(s)nθθ(z)z c(s)os(in) θ(θ)z(z) )
for some real number θz and positive real number A. What happens if ∂f ∂z tends to zero?
4. Let M := (c(a) d(b)) be any matrix entries a, b, c, d 2 C such that det(M)
0. Define the function fM (z) by
az + b
fM (z) =
1
Similarly, define fN (z) for N := (γ(α) δ(β) ). Prove that
fN (fM (z)) = fNM (z).
In other words, the composition of the functions fN and fM is equal to the function associated to the matrix product NM .
5. Assume that M := (c(a) d(b) ) has real entries a, b, c, d and det(M) 0. Let H := {z = x + iy 2 C : y > 0}. Prove that if det(M) > 0 then fM (z) is a bijective holomorphic map from H to itself.
6. Consider the differential form
dz √dx2 + dy2
=
y y
on H. Let fM (z) be as defined in question 5. Show that
df dz √dx2 + dy2
= =
Im(f (z)) y y
where Im(f (z)) denotes the imaginary part of f (z). In other words, the differential form |dz|/Im(z) is invariant under transformations of the form z '! fM (z).
7. Let x0 and x1 be real numbers with x0 < x1 . Consider the function
z x1
z x0 .
Prove that g maps H to itself. Let Cx0 ,x1 be the Euclidean circle whose diameter connects x0 to x1 on R. Prove that g maps Cx0 ,x1 to the vertical line Re(z) = x = 0.
8. Let fM (z) be as in question 6. Assume that fM (i) = i. Assuming that det(M) = 1, prove there is a θ such that
M = ( s(o)i(s)nθθ c(s)os(in)θ(θ) ) .
9. Let D be the unit disc, meaning D := {z 2 Z||z| < 1}. For any a 2 D and θ 2 [0, 2π), let
gθ,a (z) = eiθ z a
z 1 .
Prove that gθ,a is a holomorphic bijection from D to itself with gθ,a (a) = 0.
10. Using the Schwarz lemma (page 94 of the textbook) to prove the following. If h is any bijective holomorphic map of D to itself, then h = gθ,a for some θ and a.
11. With the notation as in problem 9, show that any bijective holomorphic map h of D to itself for which h(0) = 0, then h = gθ,0 for some θ 2 [0, 2π).
12. Prove that the function
z i
is a holomorphic bijection from the upper half plane H to the unit disc D.
13. With the notation as above, prove the following. Any holomorphic bijection from the upper half plane H to itself is of the form F −1 。gθ,a 。F . If we write fM = F −1 。gθ,a 。F for some matrix M , can you parameterize M?
14. With the notation as above, show that the area differential
dxdy
y2
on H is invariant under z '! fM (z).
15. From pre-calculus (or earlier), one learns that one can write a circle in the plane in the form
(x x0 )2 + (y y0 )2 = r2 .
a. Show that one can write the circle using the complex variable z as z + c0 z + c1 + d = 0, for certain constants c0 , c1 and d.
b. Using (a), show that the inversion w '! 1/z maps “circles” to “cir- cles” includes Euclidean lines.
c. Write
az + b B
cz + d cz + d
for constants A and B . From this, discuss why any fractional linear transformation is the composition of a translation, an inversion, and then another translation.
d. Combine the above discussion to show that any fractional linear trans- formation maps “circles” to “circles” .
16. Let z0 , z1 , z2 and w0 , w1 , w2 be two triples of distinct points. Consider the function w = w(z) defined by
(w w0 )(w1 w2 ) (z z0 )(z1 z2 )
(w w2 )(w1 w0 ) (z z2 )(z1 z0 ) .
Argue that w = w(z) can be written as a fractional linear transformation which maps zj to wj for j = 1, 2, 3.
17. Let w = w(z) be a fractional linear transformation which fixes 3 distinct points. Then prove that w is the identity transformation.
18. Let z, z0 , z1 , z2 be four points in the plane. Define
(z — z0 )(z1 — z2 )
(z : z0 , z1 , z2 ) :=
Show that for any M , as above, one has that
(fM (z); fM (z0 ), fM (z1 ), fM (z2 )) = (z : z0 , z1 , z2 ).
19. Show that four distinct points z, z0 , z1 , z2 in the plane lie on a circle, or straight line, if and only if (z : z0 , z1 , z2 ) is real.
20. Let M := (c(a) d(b) ) have real entries a, b, c, d and det(M) 0. Show that if Tr(M) > 2, where Tr(M) is the trace of M , then fM (z) has two fixed points on the real axis when considering fM (z) as a holomorphic bijection from H to itself.
21. This question is a continuation of question 18. Let x0 and x1 , with x0 < x1 be the fixed points of fM (z). As in question 7, consider the function
z — x1
g(z) =
What can you say about the function g −1 o fM o g? (Hint: Using question 7.)
22. Let f be a bijective holomorphic map which sends the top-half of the unit disc to the right-half of the upper half plane with the image of the x axis mapping onto the y axis. Discuss how you can extend f to a map of the unit disc to the upper half plane by using the Schwarz reflection principle.
23. Describe, in as much detail as possible, the rationale behind the following statement: The set of bijective holomorphic maps of the unit disc D to itself can be parameterized by a value of θ 2 [0, 2π) and a value of a 2 D .
24. Confirm the following: The function
w = ( )2
defines a conformal map of the upper half of the unit disc to the upper half plane . Describe why this function does not preserve angles at z = 1 and why this does not contradict the above statement.
25. Confirm the following: For m 2, the function w = zm defines a conformal map of the sector 0 < Arg(z) < π/m onto the upper half plane .
26. Confirm the following: The function w = log(z) (natural log, of course) is a conformal map of the annular region a < |z| < b with the segment along the negative real axis removed onto a domain D . Determine D . (Note: By removing the segment along contained in the negative real axis, the domain is simply connected.)
27.
a. Prove that
sinh(x) sinh(x/2)
x x/2 .
b. Prove that for any integer n 1, we have that
x sinh(x/2n ) n 1
k=1
c. Cite the appropriate convergence theorems to prove that
= .
d. In part (c), introduce the change of variables x = log(θ) = .
In particular, if θ = 2, then
log(2) = · · · · · .
28. Determine the values of the real variable p > 0 for which the product
∞
converges
29. Determine the values of the real variable x for which the products (1 + ( )n ) and ( )
converge.
30. Prove the following theorem. An infinite product
∞
∏ (1 + an )
n=1
converges if and only if an ! 0 and for some sufficiently large m the series
log(1 + an )
n=m+1
converges . Furthermore, if
L = log(1 + an )
n=m+1
then ∞ m
∏ (1 + an ) =∏ (1 + an ) · eL .
n=1 n=1
31. Prove the following theorem. For any complex number z with |z| ≤ 1/2, we
have that
|z| ≤ |Log(1 + z)| ≤ |z|,
where Log(z) denotes the principal value of the logarithm.
32. For what values of z will the following products converge absolutely:
∞ ∞
∏ (1 + zn ) and ∏ (1 + sin2 (z/n)) .
n=1 n=1
33. In class it was shown that
= (1 — ) .
a. Compute the logarithmic derivative of this product to obtain a series expansion for cot(πz)
b. With this series, study the coefficients of the expansion near z = 0 to
evaluate ∞
.
c. Differentiate the series from (a) with respect to z and, as in (b), study the expansion near z = 0 to evaluate
.
34. Prove that
cos(πz) = (1 — ) .
34. Using the result from the above question, evaluate at special values of z , say z = 1/2, z = 1/6, etc., and numerical product expansions for ^2 and ^3.
35. Define the function Γ(s) by
∞
Γ(s) = ∫ e −tts−1dt.
0
a. Prove that the integral which defines Γ(s) is holomorphic for all s 2 C with Re(s) > 0.
b. For any Re(s) > 0, prove that
Γ(s + 1) = sΓ(s).
c. Using the above results, argue that Γ(s) admits a meromorphic con- tinuation to all s 2 C.
36. Refer to the notation from above for Γ(s).
a. Proof that the function
ζ(s) =
is holomorphic for Re(s) > 1.
b. Show that for any real number a > 0 and Re(s) > 0, we have that
∞
a −s = ∫ e −atts−1dt.
0
c. For any t > 0, show that
e −nt = .
d. Prove that for any Re(s) > 1 one has that
1 ∞
ζ(s) = ∫ ts−1dt + ∫ ts−1dt.
0 1
e. From the information above, prove that the function ζ(s) —
extends to a holomorphic function for all s 2 C.
37. Let S be any finite set of prime numbers, and let P denote the smallest prime number not in S. Assume that Re(s) > 1, and let
F (s) = — (1 — p −s) −1 .
a. If x is a real number and x > 1, prove that F (x) 0.
b. Choose any δ > 0 and assume that Re(s) 1 + δ . Prove there is a constant C6 such that
|F (s)| ≤ C6 P −s .
c. Let P be the set of all prime numbers. By quoting results from the textbook, prove that the product
∏ (1 — p −s) −1
p∈P
converges for all Re(s) > 1.
d. With these results, prove that
= (1 — p −s) −1
for Re(s) > 1.
38. Prove that
∏ (1 — p −s) −1 — ∑
is holomorphic for Re(s) > 1 and extends to a holomorphic function for Re(s) > 1/2. (Hint: Expand the product using the geometric series expan- sion and the fundamental theorem of arithmetic.)
39. Let x be a real number. Using the above results, show that the limit x1 —
exists. From this, conclude that the set P has an infinite number of ele- ments, meaning there are an infinite number of prime numbers.
40. Using that
∞
Γ(s) = ∫ e −tts−1dt,
0
prove that
π
sin(πs) .
From this, prove that Γ(s) = 0 has no solutions.
9
41. Assume the notation as above.
a. Prove that the quotient
Γ(2s)/ (Γ(s)Γ(s + 1/2)])
is holomorphic and non-vanishing.
b. For any integer m 2, prove that the quotient
Γ(ms)/ (Γ(s + k/m))
is holomorphic and non-vanishing.
42. Let m > 1 be an integer. If s any complex number for which is not an integer, then prove that
sin ( k m(一) s π ) = 21−m sin(πs).
(Hint: Show that both sides have the same zeros with the same order of vanishing. Then, show that both sides are period in s with period 1. Then, compute the asymptotics as s ! σ + i1 for any σ 2 [一1/2, 1/2].)
43. Prove the following identities.
a. Γ(n + 1) = n! for any integer n 1.
b. Γ(1/2) = ^π .
c. |Γ(ix)|2 = π
cosh(πx) .
44. Compute all of the poles and residues of Γ(s).
45. Show that ζ(0) = 一1/2.
46. Prove that the following statement is false. Let pm be the m-th integer. Then there is a constant C > 0 such that pm C · m(log(m))2 . Then, show that is also not true that you can replace the lower bound for pm by C · m(log(m))1+ε for any ε > 0.
2023-05-06