0% found this document useful (0 votes)
55 views9 pages

III-2 Analytic Functions

1) An analytic function is one that is continuously differentiable in an open set. 2) A function is analytic if and only if it can be represented by a power series with a positive radius of convergence. 3) The exponential, trigonometric, and logarithmic functions are all examples of analytic functions. Their derivatives and other properties are discussed.

Uploaded by

TOM DAVIS
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
0% found this document useful (0 votes)
55 views9 pages

III-2 Analytic Functions

1) An analytic function is one that is continuously differentiable in an open set. 2) A function is analytic if and only if it can be represented by a power series with a positive radius of convergence. 3) The exponential, trigonometric, and logarithmic functions are all examples of analytic functions. Their derivatives and other properties are discussed.

Uploaded by

TOM DAVIS
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 9

III.2.

Analytic Functions 1

III.2. Analytic Functions

Recall. When you hear “analytic function,” think power series representation!

Definition. If G is an open set in C and f : G → C, then f is differentiable at


f (a + h) − f (a)
point a ∈ G if f 0 (a) = lim exists.
h→0 h

Proposition III.2.2. If F : G → C is differentiable at a ∈ G, then f is continuous


at a.

Proof. We have

|f (z) − f (a)|
lim |f (z) − f (a)| = lim |z − a|
z→a z→a |z − a|

f (z) − f (a)
= lim lim |z − a|
z→a z − a z→a
= |f 0 (a)| · 0 = 0.

Note. The reason for the following definition will become apparent in Theorem
IV.2.8.

Definition. A function f : G → C is analytic if f is continuously differentiable on


G.
III.2. Analytic Functions 2

Note. All the usual properties of differentiability and continuity hold, so sums,
products, and quotients of analytic functions are analytic (with the obvious caveats
on quotients). In addition:
Chain Rule. Let f and g be analytic on G and Ω respectively and suppose
f (G) ⊂ Ω. Then g ◦ f is analytic on G and (g ◦ f )0(z) = g 0 (f (z))f 0(z) for all z ∈ G.

Note. We will eventually see that a function is analytic if and only if it has a
power series representation. The following is the easy half of this result.


X
Proposition III.2.5. Let f (z) = an (z − a)n have radius of convergence R > 0.
n=0
Then:

(a) for k ≥ 1 the series



X
n(n − 1) · · · (n − k + 1)an (z − a)n−k
n=k

has radius of convergence R,

(b) The function f is infinitely differentiable on B(a; R) and the series of (a) equals
f (k) (z) for all k ≥ 1 and |z − a| < R, and
1 (n)
(c) for n ≥ 0, an = f (a).
n!

d z
Note. We can use Proposition III.2.5 to show that [e ] = ez .
dz
III.2. Analytic Functions 3

Note. Proposition III.2.5 says that power series are infinitely differentiable. So
the following is not surprising.

an(z − a)n has radius of convergence R > 0


P
Corollary III.2.9. If the series
then f (z) = an(z − a)n is analytic (i.e., continuously differentiable) in B(a; R).
P

Proposition III.2.10. If G is open and connected and f : G → C is differentiable


with f 0 (z) = 0 for all z ∈ G, then f is constant.

Note. For the proof of Proposition III.2.10, we need two results given in Appendix
A. Notice that each deals with functions of a real variable with C as the codomain.
Proposition A.3. If a function f : [a, b] → C is differentiable and f 0 (x) = 0 for

all x ∈ [a, b], then f is a constant.


Proposition A.4. Let f : [a, b] → C be a differentiable function. (a) If g : [c, d] →

[a, b] is differentiable then f ◦ g is differentiable and (f ◦ g)0(t) = f 0(g(t))g 0(t). (b) If


G is an open subset of C containing f ([a, b]) and h : G → C is an analytic function
then h ◦ f is differentiable and (h ◦ f )0(x) = h0(f (x))f 0(x).
III.2. Analytic Functions 4

Lemma III.2.A. Properties of ez include:

(a) ea+b = ea eb ,

(b) ez 6= 0 for all z ∈ C,

(c) ez = ez , and

(d) |ez | = eRe(z) .

Proof.
(a) Define g(z) = ez ea+b−z for given a, b ∈ C. Then g 0 (z) = ez ea+b−z +ez (−ea+b−z ) =
0. So by Proposition 2.10, g(z) is constant for all z ∈ C. With z = 0, we have
g(0) = e0 ea+b = ea+b , so ez ea+b−z = ea+b for all z ∈ C. With z = b we have
eb ea = ea+b .
(b) By part (a) we have 1 = e0 = ez e−z for all z ∈ C, and so ez 6= 0 for all z ∈ C.
(c) Since ez = ∞ zn
P
n=0 n! , then

∞ ∞  n ∞
!
X z n X z X (z)n
z
e = = = = ez .
n=0
n! n=0
n! n=0
n!

(d) By (c) we have


|ez |2 = ez ez = ez ez = ez+z by (a)

and so |ez |2 = e2Re(z) and |ez | = eRe(z) .

Definition. Define

X (−1)n
cos z = z 2n , and
n=0
(2n)!

X (−1)n−1 2n−1
sin z = z .
n=1
(2n − 1)!
III.2. Analytic Functions 5

Note. For cos z and sin z the radius of convergence is R = ∞. We also have:
d
[cos z] = − sin z
dz
d
[sin z] = cos z
dz
eiz + e−iz
cos z =
2
eiz − e−iz
sin z =
2i
cos2 z + sin2 z = 1 for all z ∈ C.

Note. Since the series for ez converges absolutely, we have eiz = cos z + i sin z.

Definition. A function f is periodic if for some c ∈ C (c 6= 0) we have f (z + c) =


f (z) for all z ∈ C.

Note. From above, we see that ez has period 2πi:

ez+2πi = ez e2πi = ez (1) = ez .

So ez is NOT a one to one function and does not have an inverse.

Note. We have that ez = w if and only if z = log |w| + i(arg(w) + 2πk) where
k ∈ Z:

elog |w|+i(arg(w)+2πk) = elog |w|ei(arg(w)+2πk)

= |w| (cos(arg(w) + 2πk) + i sin (arg(w) + 2πk))

= |w| (cos(arg(w)) + i sin(arg(w))) = |w|cis(arg(w)) = w.


III.2. Analytic Functions 6

Definition. Let G be an open connected set in C and f : G → C be a continuous


function such that z = ef(z) on set G. Then f is a branch of the logarithm on G.

Note. Since there is no z ∈ C such that ez = 0, then 0 ∈


/ G.

Note. We cannot define a branch of the log on C \ {0} because of continuity (this
is an informal solution to Exercise III.2.21):

But log |z1 | and log |z2 | are “close together,” however arg(z1 ) and arg(z2 ) must
be about 2π apart (to keep continuity on the path) and then we cannot have
continuity on a circle around the origin. Therefore, we can only define a branch of
log on C \ {0} \ { a path from 0 to ∞}.

Proposition III.2.19. If G ⊆ C is open and connected and f is a branch of log z


on G, then the totality of branches of log z are the functions {f (z) + 2kπi | k ∈ Z}.
III.2. Analytic Functions 7

Note. One branch of the logarithm is f (z) = f (reiθ ) = log r+iθ where −π < θ < π
and G = C \ {z ∈ R | z ≤ 0}. In a complex variables class, you might use the
“principal argument” to define a “principal branch” of the logarithm. To see that
a branch of log is analytic, we need the following.

Proposition III.2.20. Let G and Ω be open subsets of C. Let f : G → C and


g : Ω → C be continuous where f (G) ⊆ Ω and g(f (z)) = z for all z ∈ G. If g is
1
differentiable and g 0 (z) = 0, then f is differentiable and f 0 (z) = 0 . If g is
g (f (z))
analytic, then f is analytic.

Corollary III.2.21. A branch of the log is analytic and its derivative is 1/z.

Proof. Let f (z) be a branch of log, let g be ez , and apply Proposition III.2.20.

Definition. Let G = C \ {z ∈ R | z ≤ 0} and define the principal branch of log of


G as given above:

f (z) = f (reiθ ) = log(r) + iθ where − π < θ < π.

If f is a branch of log on some open connected set H, then g : H → C defined as


g(z) = exp(bf (z)) is a branch of z b (where b ∈ C). If we write z b we will mean the
principal branch of z b (based on the principal branch of log z).

Notice. z b is analytic since ez and the principal branch of log are analytic.
III.2. Analytic Functions 8

Definition. A region is an an open connected set in C.

Note. We now derive a necessary and sufficient condition for function f to be


analytic on a region.

Definition. Let u : R2 → R and v : R2 → R. The Cauchy-Riemann equations on


u and v are
∂u ∂v ∂u ∂v
= and =− .
∂x ∂y ∂y ∂x

Note. If u and v satisfy the Cauchy-Riemann equations, then

∂u ∂v ∂ 2u ∂ 2v
= implies = , and
∂x ∂y ∂x2 ∂x∂y
∂u ∂v ∂ 2u ∂ 2v
=− implies =− .
∂y ∂x ∂y 2 ∂y∂x
Then, if u has continuous second partial derivatives (and second mixed partials are
equal), then
∂ 2u ∂ 2u
+ = 0.
∂x2 ∂y 2
Similarly,
∂ 2v ∂ 2v
+ = 0.
∂x2 ∂y 2

Definition. A function u : R2 → R satisfying

∂ 2u ∂ 2u
+ =0
∂x2 ∂y 2

is a harmonic function.
III.2. Analytic Functions 9

Theorem III.2.29. Let u and v be real-valued functions defined on a region G and


suppose u and v have continuous partial derivatives (so we view u and v as functions
of x and y where z = x + iy). Then f : G → C defined by f (z) = u(z) + iv(z) is
analytic if and only if the Cauchy-Riemann equations are satisfied.

Note. There are certainly differences between real and complex functions. For
example, f (x) = |x|2 = x2 is differentiable for all x ∈ R. However, f (z) = |z|2 =
x2 + y 2 is only differentiable at z = 0. (This is Exercise III.2.1.)

Definition. If u and v are harmonic on region G, and if f = u + iv satisfies the


Cauchy-Riemann equations, then u and v are harmonic conjugates.

Note. The following gives conditions under which u has a harmonic conjugate on
G.

Theorem III.2.30. Let G be either the whole plane C or some open disk. If
u : G → R is harmonic, then u has a harmonic conjugate on G.

Idea of the Proof. The harmonic conjugate is found as follows:


Z y Z x
v(x, y) = ux (x, t) dt − uy (s, 0) ds.
0 0

u and v then satisfy the Cauchy-Riemann equations. 

Revised: 9/4/2017

You might also like