Boundary value problems for second order

equations II
H. Begehr
Abstract
This is the continuation of an investigation of basic boundary value problems for first order complex model partial differential equations. Model second order equations are the Poisson and the inhomogeneous Bitsadze equations. Different kinds of boundary conditions are posed as combinations of the
Schwarz, the Dirichlet, and the Neumann conditions. Solvability conditions
and the solutions are given in explicit form for the unit disc. Exemplarily the
inhomogeneous polyanalytic equation is investigated as a model equation of
arbitrary order.

1 Boundary value problems for second order equations

There are two basic second order differential operators, the Laplace operator
z z and the Bitsadze operator z2 . The third one, z2 is just the complex conjugate of the Bitsadze operator and all formulas and results for this operator can
be attained by the ones for the Bitsadze operator through complex conjugation
giving dual formulas and results.
For the Laplace and the Poisson, i.e. the inhomogeneous Laplace equation, the
Dirichlet and the Neumann boundary value problems are well studied. Before
investigating them the Schwarz problem will be studied for both operators.
Theorem 1 The Schwarz problem for the Poisson equation in the unit disc
wzz = f in D , Re w = 0 , Re wz = 1 on D , Im w(0) = c0 , Im wz (0) = c1
is uniquely solvable for f L1 (D; C), 0 , 1 C(D; R), c0 , c1 R. The solution is
Z
1
+ z d
w(z) = ic0 + ic1 (z + z)
0 ()
2i
z
||=1
Z
1
d

1 ()[ log(1 z)2 log(1 z)2 + z z]

2i

||=1

1
+

{f ()[log | z |2 log(1 z)] f () log(1 z)}dd

(1)

||<1

io
n
h log(1 z)
i
h log(1 z)
+
log
|

|
dd
f ()
+
log
|

f
()
2
2

Z
||<1

1
+

h f ()

f () i z z
+
dd .
2

||<1

Proof This result follows from combining the solution

Z
Z h
+ z d
1
() + z () 1 + z i
1
0 ()
+

dd
w(z) = ic0 +
2i
2
1 z
z
z
||=1

||<1

of the Schwarz problem

wz = in D , Re w = 0 on D , Im w(0) = c0
with the solution
(z) = ic1 +

1
2i

Z
1 ()

+ z d
1

z
2

||=1

h f () + z
z

f () 1 + z i
dd
1 z

||<1

of the Schwarz problem

z = f in D , Re = 1 on D , Im (0) = c1 .
Here the relations
Z
1
+ + z dd
log | |2 +z ,
= 2 log | z |2 2 log(1 z )

||<1

1
2

Z
||<1

1
2

Z
||<1

1
2

+ 1 + z dd
+ log | |2 z ,
= 2 log(1 z )
1

1 + + z dd
2
+z ,
=
log(1 z )
1 z

1 + 1 + z dd
2
z ,
= log(1 z )

1 1 z

||<1

and
1
2

h1 + z
z

1 1 + z i
dd = z z
1 z

||<1

are needed. More simple than this is to verify that (1) is the solution.
The uniqueness of the solution can easily be seen. In case w1 and w2 are two
solutions then = w1 w2 would be a harmonic function with homogeneous
data,
zz = 0 in D , Re = 0 , Re z = 0 on D , Im (0) = 0 , Im z (0) = 0 .
As z is analytic, say 0 in D, then integrating the equation z = 0 means
= + where is analytic in D.
Then Re z = 0 on D, Im z (0) = 0 means Re 0 = 0, Im 0 (0) = 0. From
Theorem 6 then 0 is seen to be identically zero, i.e. a constant, say a.
Then from Re = 0 on D, Im (0) = 0 it follows Re = Re a and
Im (0) = Im a. Thus again [3], Theorem 6 shows (z) = a identically in D.
This means vanishes identically in D.
There is a dual result to Theorem 1 where the roles of z and z are interchanged. This can be attained by setting W = w and complex conjugating
(1).
Theorem 10 The Schwarz problem for the Poisson equation in the unit disc
wzz = f in D , Re w = 0 , Re wz = 1 on D , Im w(0) = c0 , Im wz (0) = c1 ,
for f L1 (D; C), 0 , 1 C(D; R), c0 , c1 R is uniquely solvable by
Z
+ z d
1
w(z) = ic0 + ic1 (z + z) +
0 ()
2i
z
||=1

1
+
2i

1 ()[ log(1 z)2 log(1 z)2 + z z]

||=1

{f ()[log | z |2 log(1 z)] f () log(1 z)}dd

(10 )

||<1

Z
||<1

n
h log(1 z)
i
h log(1 z)
io
f ()
f
()
+
log
|

|
dd
+
log
|

2
2

h f ()

f () i z z
dd .
2

||<1

Theorem 2 The Schwarz problem for the inhomogeneous Bitsadze equation

in the unit disc
wzz = f in D , Re w = 0 , Re wz = 1 on D , Im w(0) = c0 , Im wz (0) = c1 ,

for f L1 (D; C), 0 , 1 C(D; R), c0 , c1 R is uniquely solvable through

Z
1
+ z d
w(z) = ic0 + i(z + z) +
0 ()
2i
z
||=1

1
2i

1 ()

+z
d
( z + z)
z

(2)

||=1

1
+
2

 f () + z
z

f () 1 + z 
( z + z)dd .
1 z

||<1

Proof Rewriting the problem as the system

wz = in D , Re w = 0 on D , Im w(0) = c0 ,
wz = f in D , Re = 1 on D , Im (0) = c1 ,
and combining its solutions
Z
Z
1
+ z d
1
w(z) = ic0 +
0 ()

2i
z 2
||=1

(z) = ic1 +

1
2i

 () + z

() 1 + z 
dd ,
1 z

f () 1 + z 
dd ,
1 z

||<1

Z
1 ()

+ z d
1

z
2

||=1

 f () + z
z

||<1

formula (2) is obtained. Here the relations

Z 
1
1 + z 1 1 + z 
dd = z z ,

2
z 1 z
||<1

+ 1 + z
dd =
z

1
2

+ z
( z) ,
z

||<1

1
2

1 + 1 1 + z
dd = + z ,
1 1 z

||<1

1
2

Z
||<1

1
2

Z
||<1

+ 1 1 + z
dd =
1 z

1 + z
( z) ,
1 z

1 + 1 + z
dd =
1 z

1 + z
( z)
1 z

are used. The uniqueness of the solution follows from the unique solvability of
the Schwarz problem for analytic functions, Theorem 6 and Theorem 9.

It is well known that the Dirichlet problem for the Poisson equation
wzz = f in D , w = on D
is well posed, i.e. it is solvable for any f L1 (D; C), C(D; C) and the
solution is unique. That the solution is unique is easily seen.
Lemma 1 The Dirichlet problem for the Laplace equation
wzz = 0 in D , w = 0 on D
is only trivially solvable.
Proof From the differential equation wz is seen to be analytic. Integrating
this quantity w = + is seen where and are both analytic in D. Without
loss of generality (0) = 0 may be assumed. From the boundary condition
= on D follows. This Dirichlet problem is solvable if and only if, see
Theorem 7,
Z
0 = f rac12i
||=1

zd
1
()
=
1 z
2i

d
1
()

z 2i

||=1

Z
()

d
= (z) .

||=1

This also implies = 0 on D so that w = 0 in D .

As Bitsadze [6] has realized such a result is not true for the equation wzz =
0.
Lemma 2 The Dirichlet problem for the Bitsadze equation
wzz = 0 in D , w = 0 on D
has infinitely many linearly independent solutions.
Proof Here wz is an analytic function in D. Integrating gives w(z) = (z)z +
(z) with some analytic functions in D. On the boundary (z) + z(z) = 0.
As this is an analytic function this relation hold in D too. Hence, w(z) =
(1 | z |2 )(z) for arbitrary analytic . In particular wk (z) = (1 | z |2 )z k is a
solution of the Dirichlet problem for any k N and these solutions are linearly
independent over C.
Because of this result the Dirichlet problem as formulated above is ill-posed
for the inhomogeneous Bitsadze equation.
With regard to the Dirichlet problem for the Poisson equation the representation formula [3], (150 ) is improper as is also [3], (15000 ). The middle terms
are improper. They can easily be eliminated by applying the Gauss Theorem,

see [3]. For the respective term in [3], (150 ) in the case D = D
Z
Z
1
1
2
w () log | z | d =
w () log | 1 z |2 d
2i
2i
||=1

||=1

[w () log | 1 z |2 ]dd

||<1

1
=

1
w () log | 1 z | dd +

||<1

w ()

z
dd
1 z

||<1

follows. Applying the Gauss Theorem again shows

Z
Z
h
1
z
1
z i
w ()
dd =
w()
dd

1 z

1 z
||<1

1
=
2i

||<1

z
1
w()
d =
1 z
2i

||=1

Z
w()

z d
.
z

||=1

Thus inserting these in [3], (150 ) leads to the representation

Z

 d 1 Z
1

1 z 2
w()
+
1

w () log |
| dd.
w(z) =
2i
z z

z
||=1

||<1

(3)
The kernel function in the boundary integral is the Poisson kernel, the one in
the area integral is called Green function for the unit disc with respect to the
Laplace operator.
Definition 1 The function G(z, ) = (1/2)G1 (z, ) with
G1 (z, ) = log |

1 z 2
| , z, D , z 6= ,
z

(4)

is called Green function of the Laplace operator for the unit disc.
Remark The Green function has the following properties. For any fixed
D as a function of z
(1) G(z, ) is harmonic in D \ {} ,
(2) G(z, ) + log | z | is harmonic in D ,
(3) lim G(z, ) = 0 for all t D ,
zt

(4) G(z, ) = G(, z) for z, D, z 6= .

They can be checked by direct calculations.
Green functions exist for other domains than just the unit disc. The existence is related to the solvability of the Dirichlet problem for harmonic functions

in the domain. The Riemann mapping theorem can be used to find it e.g. for
regular simply connected domains. Having the Green function [3], (150 ) and
[3], (15000 ) can be altered as above leading to the Green representation formula,
see e.g. [1]. Green functions exist also in higher dimensional spaces and for
other strongly elliptic differential operators.
For the unit disc the following result is shown.
C) can be represented as
Theorem 3 Any w C 2 (D; C) C 1 (D;
1
w(z) =
2i


 d 1 Z

w()
+
1

z z

||=1

w ()G1 (z, )dd , (30 )

||<1

where G1 (z, ) is defined in (4).

Formulas [3], (150 ) and [3], (15000 ) are both unsymmetric. Adding both gives
some symmetric formula which is for the unit disc
w(z) =

1
4i


 d
w()
+
z z

||=1

4i

(w () + w ()) log | z |2

||=1

w () log | z |2 dd .

(5)

||<1

Motivated by the procedure before, the Gauss Theorems are applied in a symmetric way to
Z
1
w () log | 1 z |2 dd

||<1

1
=
2

n
[w () log | 1 z |2 ] + [w () log | 1 z |2 ]

||<1

h
z i
z io
+ w()
+ w()
dd
1 z
1 z
Z
1
d
=
log | 1 z |2 [w () + w ()]
4i

||=1

1
+
4i

h z
z i d
w()
+
1 z
1 z

||=1

1
=
4i

log | z |2 [w () + w ()]

||=1

1
4i

w()

h z
z i d
+
.
z z

||=1

Here are two possibilities. At first the second term in (5) can be eliminated
giving
Z

1
w(z) =
2i


 d 1 Z

w()
1
+

z z

||=1

w () log |

1 z 2
| dd ,
z

||<1

i.e. (3). Next the first term in (5) is simplified so that

Z
Z
1
d
1
d
w(z) =
w()
(w () + w ()) log | z |2
2i

2i

||=1

1
+

||=1

w () log | ( z)(1 z) |2 dd .

(6)

||<1

Here the normal derivative appears in the second term while a new kernel
function arises in the area integral.
Definition 2 The function N (z, ) = (1/2)N1 (z, ) with
N1 (z, ) = log | ( z)(1 z) |2 , z, D , z 6= ,

(7)

is called Neumann function of the Laplace operator for the unit disc.
Remark The Neumann function, sometimes [7] also called Green function of
second kind or second Green function, has the properties
(1) N (z, ) is harmonic in z D \ {} ,
(2) N (z, ) + log | z | is harmonic in z D for any D ,
(3) N (z, ) = 1 for z D, D ,
(4) N (z, ) = N (, z) for z, D, z 6= .
Z
1
d
(5)
N (z, )
=0.
2

||=1

They can be checked by direct calculations.

The last result may therefore be formulated as follows.
Theorem 4 Any w C 2 (D; C) C 1 (D; C) can be represented as
Z
Z
1
d
1
d
w(z) =
w()
w() log | z |2
2i

2i

||=1

||=1

1
+

(60 )

w ()N1 (z, )dd .

||<1

This formula can also be written as

Z
Z
d 1
1
w(z) =
[w() N1 (z, ) w()N1 (z, )] +
w ()N1 (z, )dd .
4i

||=1

||<1

(600 )
Theorem 3 immediately provides the solution to the Dirichlet problem.
Theorem 5 The Dirichlet problem for the Poisson equation in the unit disc
wzz = f in D , w = on D ,
for f L1 (D; C) and C(D; C) is uniquely given by
Z

1
w(z) =
2i

Z
 d


1
+
1

()
z z

||=1

f ()G1 (z, )dd . (8)

||<1

This is at once clear from the properties of the Poisson kernel and the Green
function.
As the Dirichlet problem formulated as for the Poisson equation is not uniquely
solvable for the Bitsadze equation another kind Dirichlet problem is considered
which is motivated from decomposing this Bitsadze equation in a first order
system.
Theorem 6 The Dirichlet problem for the inhomogeneous Bitsadze equation
in the unit disc
wzz = f in D , w = 0 , wz = 1 on D ,
for f L1 (D; C), 0 , 1 C(D; C) is solvable if and only if for | z |< 1
z
2i

Z
 ()
z
1 () 
0

d +
1 z

||=1

and

1
2i

f ()

z
dd = 0
1 z

(9)

||<1

zd
1
1 ()

1 z

||=1

Z
f ()

zdd
=0.
1 z

(10)

||<1

The solution then is

Z
Z
Z
1
d
1
z
1
z
w(z) =
0 ()

1 ()
d +
f ()
dd .
2i
z 2i
z

z
||=1

||=1

||<1

(11)

Proof Decomposing the problem into the system

wz = in D , w = 0 on D ,
z = f in D , = 1 on D ,
and composing its solutions
1
w(z) =
2i

Z
||=1

1
(z) =
2i

d
1
0 ()

dd
,
z

f ()

dd
,
z

||<1

d
1
1 ()

||=1

||<1

and the solvability conditions

Z
Z
1
1
zd
0 ()
=
2i
1 z

||=1

1
2i

()

()

zdd
,
1 z

f ()

zdd
,
1 z

||<1

Z
1 ()

zd
1
=
1 z

||=1

Z
||<1

proves (11) together with (9) and (10). Here

1

Z
dd
1
z
=

( )(1 z)
1 z 2i

||<1

z d
z
=
1 z 1 z

||=1

and
1

dd
1
=

( )( z)

||<1

1  1
1 
z

dd =
z z
z

||<1

are used.
This problem can also be considered for the Poisson equation.
Theorem 7 The boundary value problem for the Poisson equation in the unit
disc
wzz = f in D , w = 0 , wz = 1 on D ,
for f L1 (D; C), 0 , 1 C(D; C) is uniquely solvable if and only if
Z
Z
1
1
zd

0 ()
+
1 () log(1 z)d
2i
2i
1 z
||=1
||=1
Z
1
=
f () log(1 z)dd

||<1

10

(12)

and

z
2i

d
z
1 ()
=
1 z

||=1

f ()

dd
.
1 z

(13)

||<1

The solution then is

Z

1
w(z) =
2i

||=1

d
1
0 ()

z 2i

1 () log(1 z)d
||=1

f ()(log | z |2 log(1 z))dd .

(14)

||<1

Proof The system

wz = , z = f in D , w = 0 , = 1 on D
is uniquely solvable according to Theorem 10 if and only if
Z
Z
d
dd
z
z
0 ()
()

=
,
2i

1 z
1 z
||=1

z
2i

||<1

Z
1 ()

z
d
=
1 z

||=1

Z
f ()

dd
.
1 z

||<1

The solution then is

1
w(z) =
2i

Z
||=1

(z) =

()

dd
,
z

||<1

1
2i

d
1
0 ()

z
1 ()

d
1

||=1

f ()

dd
.
z

||<1

Inserting into the first condition gives (12) for

Z
Z
Z
1
dd
1
1

()
=
1 ()

2i

1 z

||=1

||<1

dd
d

( )(1 z)

||=1

1
f ()

||<1

dd

dd

( )(1 z)

||<1

with
1

zdd
z)
( )(1

1
= log(1 z )
2i

||<1

Z
log(1 z)

||=1

+ 1
= log(1 z )
2i

Z
||=1

= log(1 z ).

11

log(1 z) d

Combining the two integral representations for w and leads to (13) as

Z
Z
Z
dd
1
dd
1
1

d
()
=

1 ()

2i

z
( )( z)

||=1

||<1

||<1

f ()

||<1

dd

dd
( )( z)

||<1

where
1

dd

( z)( )

1
= log | z |2
2i

||<1

log | z |2

||=1

1
= log | z |2
2i

Z
log(1 z)

||=1

1
2i

log(1 z)

(1 )

||=1

.
= log | z |2 log(1 z )
Remark In a similar way the problem
wzz = f in D , w = 0 , wz = 1 on D
with f L1 (D; C), 0 , 1 C(D; C) can be solved.
That integral representations may not always be used to solve related
boundary value problems as was done in the case of the Dirichlet problem
with formula (3), can be seen from (60 ). If w is a solution to the Poisson
equation wzz = f in D satisfying w = on D and being normalized by
Z
d
1
w()
=c
2i

||=1

for proper f and then on the basis of Theorem 16 it may be presented as

Z
1
d
w(z) = c
() log | z |2
2i

||=1

f () log | ( z)(1 z) |2 dd.

(15)

||<1

But this formula although providing always a solution to wzz = f does not
for all satisfy the respective boundary behaviour. Such a behaviour is also
known from the Cauchy integral.

12

Theorem 8 The Neumann problem for the Poisson equation in the unit disc
Z
1
d
wzz = f in D , w = on D ,
w()
=c,
2i

||=1

for f L1 (D; C), C(D; C), c C is solvable if and only if

Z
Z
d
2
1
()
f ()dd .
=
2i

||=1

(16)

||<1

The unique solution is then given by (15).

Proof As the Neumann function is a fundamental solution to the Laplace
operator and the boundary integral is a harmonic function, (15) provides a
solution to the Poisson equation. For checking the boundary behaviour the
first order derivatives have to be considered. They are
Z
Z
 1
1

d
1
wz (z) =
()

f ()
+
dd ,
2i
( z)

z 1 z
||=1

1
wz (z) =
2i

||<1

d
1
()

( z)

||=1

 1

f ()
+
dd ,
z 1 z

||<1

so that
1
w(z) =
2i

Z
()


 d
z
+
1
z z

||=1

h z
z
z i
z
f ()
+
+
+
dd
z z 1 z
1 z

||<1

1
=
2i


 d

()
+
2
z z

||=1

h
f () 2

z
z
1
1 i

dd .
z z 1 z
1 z

||=1

For | z |= 1 this is using the property of the Poisson kernel

Z
Z
d
2
1
w(z) = (z)
() +
f ()dd .
2i

||=1

Therefore w = on | z |= 1 if and only if condition (16) holds. At last the

normalization condition has to be verified. It follows from | z |=| 1 z | for

13

| z |= 1 and
Z
Z
Z
1
1
1
dz
dz
2 dz
log | 1 z |
=
log(1 z)
log(1 z)
=0.
2i
z
2i
z
2i
z
|z|=1

|z|=1

|z|=1

Theorem 9 The Dirichlet-Neumann problem for the inhomogeneous Bitsadze

equation in the unit disc
wzz = f in D , w = 0 , wz = 1 on D , wz (0) = c ,
for f L1 (D; C) C(D; C), 0 , 1 C(D; C), c C is solvable if and only if
for z D
Z
Z
1
d
1 | |2
1
c
0 ()
f ()
+
dd = 0
(17)
2i
1 z

(1 z)
||=1

||<1

and
1
2i

Z
(1 () f ())

d
1
+
(1 z)

||=1

zf (z)
dd = 0 .
(1 z)2

(18)

||<1

The solution then is

Z

1
w(z) = c
z+
2i

0 ()

d
z

||=1

1
2i

(1 () f ())

1 | z |2
d
log(1 z)
z

||=1

Z
f ()

| |2 | z |2
dd .
( z)

(19)

||<1

Proof The problem is equivalent to the system

wz = in D , w = 0 on D ,
z = f in D , = 1 on D , (0) = c .
The solvability conditions are
Z
Z
1
d
1
0 ()
=
2i
1 z

||=1

and

1
2i

()

dd
1 z

||<1

d
1
(1 () f ())
+
1 z)

||=1

Z
||<1

14

zf ()
dd = 0
(1 z)2

and the unique solutions

w(z) =

1
2i

0 ()

d
1

||=1

()

dd
z

||<1

and
(z) = c

1
2i

(() f ()) log(1 z)

||=1

zf ()
dd
( z)

||<1

according to [3], Theorems 10 and 11.

From
Z
Z
Z
1
dd
dd
1
1

=1,
log(1 )
=

1 z

1 z
2i
||<1

1
d

||<1

d
=0,
(1 z)

||=1

and

dd

| |2 1
=
z )

1 z
(1

log(1 )

||<1

condition (17) follows. Similarly (19) follows from

Z
Z
2
dd
1
1
dd = 1 | z | log(1 z )

=z ,
log(1 )

z
z
||<1

||<1

and
1

| |2 | z |2
dd
=
.
( ) z
( z)

Z
||<1

Theorem 10 The boundary value problem for the inhomogeneous Bitsadze

equation in the unit disc
wzz = f in D , w = 0 , zwzz = 1 on D , wz (0) = c ,
is solvable for f L1 (D; C), 0 , 1 C(D; C), c C if and only if for z D
condition (17) together with
Z
Z
1
d
z
dd
1 ()
+
f ()
=0
(20)
2i
(1 z)
(1 z)2
||=1

||<1

holds. The solution then is uniquely given by

Z
Z
1
d
1
1 | z |2
d
w(z) = cz +
0 ()
+
1 ()
log(1 z)
2i
z 2i
z

||=1

Z
f ()

||=1

| |2

| z |2

( z)

dd .

||<1

15

(21)

The proof is as the last one but [3], Theorem 12 is involved rather than [3],
Theorem 11.
Theorem 11 The Neumann problem for the inhomogeneous Bitsadze equation
in the unit disc
wzz = f in D , w = 0 , wz = 1 on D , w(0) = c0 , wz (0) = c1
is uniquely solvable for f C (D; C), 0 < < 1, 0 , 1 C(D; C), c0 , c1 C
if and only if for z D
Z

1
c1 z +
2i

d
1
0 ()

z 2i

||=1

Z
(1 () f ())(1 z log(1 z))d
||=1

f ()  z( z)
1 

dd = 0

(1 z)2
z

(22)

||<1

and
Z

1
2i

d
z
(1 () f ())

||=1

f ()

dd
=0.
(1 z)2

(23)

||<1

The solution then is given as

Z
1
d
w(z) = c0 + c1 z
0 () log(1 z)
2i

||=1

1
2i

Z
(1 () f ())( z) log(1 z)

d
z
+

||=1

f () z
dd. (24)
z

||<1

Proof From applying [3], Theorem 11

Z
Z
1
d
z
(0 () ()) log(1 z)
w(z) = c0
2i

||=1

1
(z) = c1
2i

d
z
(1 () f ()) log(1 z)

f ()

dd
,
( z)

Z
||<1

if and only if
Z
Z
1
z
d
(0 () ())
+
2i
(1 z)
||=1

1
2i

dd
,
( z)

||<1

||=1

()

()
dd = 0 ,
(1 z)2

||<1

d
z
(1 () f ())
+
(1 z)

||=1

Z
||<1

16

f ()
dd = 0 .
(1 z)2

Inserting into the first condition leads to (22) on the basis of

Z
d
1
() 2
2i
(1 z)
||=1

1
2i

= c1 z

f
())

(1 ()

||=1

||=1

1
f ()
2i

||<1

d d
log(1 )
1 z 2

1
2i

d

dd
( )(1 z)

||=1

1
2i

= c1 z +

(1 () f ())

d
z
+

||=1

and
Z
z

()
z
dd =
2
(1 z)

||<1

Z
||<1

( z)()
z

dd
2
(1 z)

||<1

z
f ()dd
(1 z)2

||<1

z
z
()d
2
(1 z)

z
=
2i

f ()
dd
(1 z)

||=1

z
f ()dd,
(1 z)2

Z
||<1

where for | z |= 1
z
z
()d =
2
(1 z)
2i

z
2i

||=1

z
()d
( z)2

||=1

1
2i

() d
z

||=1

1
=
2i

f
())
1
(1 ()
2i

||=1

||<1

1
2i

log(1 )
d
d
( z)

||=1

1
f ()
2i

d

dd

( )( z)

||=1

Z
(1 () f ())

1
d
log(1 z)
z

||=1

1
=
2i

Z
(1 () f ())z log(1 z)
||=1

17

d
.

1
From
2i

1
log(1 z)d =
2i

||=1

1
2i

Z
log(1 z)d = 0 ,
||=1

||=1
+
X

k=1

log(1 z)d
log(1 )

||=1

log(1 z)d = 1
log(1 )
2i
+

1
k 2i

log(1 z)

X
d
k1 log(1 z) |=0
=
2

k!
k=2

||=1
k

+ k1
X
z
=
(k 1)k
k=2

= log(1 z )

1
+ z )
= 1 z log(1 z )
,
(log(1 z )
z
z

and
log(1 z)
1
d =
2i

1
2i

||=1

log(1 z)
d

Z
||=1

1
2i

Z
||=1

log 1 z
d = 0

the relation
1
2i

Z
() log(1 z)d

||=1

1
=
2i

 1 z
 d
(1 () f ())
log(1 z) +
z

||=1

follows. Similarly from

Z
Z
z
dd
1
=

( z)

||<1

dd
1

||<1

dd
= z ,

||<1

log(1 )
dd
z

||<1

1
= ( z) log(1 z )
2i

Z
||=1

18

d
( ) log(1 )
z

(25)

+ 1
= ( z) log(1 z )
2i

log(1 )

(1 )

d
( z)

||=1
2

+ 1 z log(1 z )
= 1 | z | log(1 z )
,
= ( z) log(1 z )
z
z
Z
Z
1
dd

log(1 )
log(1 )dd
=

||<1

||<1

1
2i

d = ,
log(1 )
2

||=1

and
1

dd
1
=

( )( z)
( z)

||<1

 1
1 
z

dd =
z
z

Z
||<1

it follows
Z

()

dd
( z)

||<1

1
= c1 z +
2i

 1 | z |2
 d
(1 () f ())
log(1 z) +
z

Z
||=1

f () z
dd .
z

(26)

||<1

From (25) and (26) the representation (24) follows.

Theorem 12 The boundary value problem for the inhomogeneous Bitsadze
equation in the unit disc
wzz = f in D , zwz = 0 , zwzz = 1 on D , w(0) = c0 , wz (0) = c1 ,
for f L1 (D; C), 0 , 1 C(D; C), c0 , c1 C is uniquely solvable if and only
if for | z |= 1
Z
Z
1
d
1
1
d
0 ()
+
1 () log(1 z)
2i
2i
z

(1 z)
||=1

||=1

Z
f ()

z
dd ,
(1 z)2

(27)

||<1

and

1
2i

d
z
1 ()
+
(1 z)

||=1

Z
f ()
||<1

19

dd
=0.
(1 z)2

(28)

The solution then is

1
w(z) = c0 + c1 z
2i

Z
0 () log(1 z)

||=1

 1 | z |2
 d
1 ()
log(1 z) +
z

1
2i

(29)

||=1

f () z
dd .
z

||<1

Proof The system

wz = in D , zwz = 0 on D , w(0) = c0 ,
z = f in D , zz = 1 on D , (0) = c1 ,
is uniquely solvable if and only if
Z
Z
1
d
z
0 ()
+
2i
(1 z)

||=1

and

1
2i

()

dd
=0
(1 z)2

||<1

z
d
1 ()
+
(1 z)

||=1

Z
()

dd
=0.
(1 z)2

||=1

The solution then is

1
w(z) = c0
2i

z
d
0 () log(1 z)

||=1

1
(z) = c1
2i

Z
()

dd
,
( z)

f ()

dd
.
( z)

||<1

d
z
1 () log(1 z)

||=1

Z
||<1

Inserting the expression for into the first condition gives (27) because as in
the preceding proof on | z |= 1
Z
z
()
dd

(1 z)2
||<1

1
2i

Z
1 ()

1
d
z
log(1 z)
z

||=1

20

z
f ()dd .
(1 z)2

Also from
Z
dd
z
()

( z)

= c1 z

||<1

 1 | z |2
 d
1 ()
log(1 z) +
z

1
+
2i

||=1

f () z
dd
z

||<1

formula (29) follows.

Boundary value problems as in Theorem 12 can also be solved for the Poisson equation. One case is considered in the next theorem.
Theorem 13 The boundary value problem for the Poisson equation in the unit
disc
wzz = f in D , zwz = 0 , zwzz = 1 on D , w(0) = c0 , wz (0) = c1
are uniquely solvable for f L1 (D; C), 0 , 1 C(D; C), c0 , c1 C if and only
if
Z
Z
 1

1
d
1
1
0 ()
=
f ()
+
1 dd
(30)
2i

1 z
(1 z)
1 z
||=1

and

||<1

1
2i

d
z
1 ()
+
(1 z)

||=1

Z
f ()

dd
=0.
(1 z)2

(31)

||<1

The solution then is

w(z) = c0 + c1 z

1
2i

0 () log(1 z)

||=1

1
2i

1 ()

 1 z

log(1 z) + z

 d

(32)

||=1

Z
f ()

z


+ log | z |2 log(1 z) log | |2 dd .

||<1

Proof The problem is equivalent to the system

wz = in D , zwz = 0 on D , w(0) = c0 ,
z = f in D , z = 1 on D , (0) = c1 .

21

It is solvable if and only if

Z
Z
d
1
z
0 ()
+
2i

(1 z)
||=1

and

1
2i

()

dd
=0
(1 z)2

f ()

dd
=0
(1 z)2

||<1

d
z
1 ()
+
(1 z)

||=1

||<1

and the solutions are according to [3], Theorem 12

Z
Z
1
d
z
w(z) = c0
0 () log(1 z)
2i

||=1

1
(z) = c1
2i

()

dd
,
( z)

f ()

dd
.
( z)

||<1

d
z
1 () log(1 z)

||=1

||<1

For the first problem [3], Theorem 12 is applied to w and the formulas then
complex conjugated. For (30)
z

dd
()
(1 z)2

||<1

h h () 
f () i

dd

1 z
1 z

||<1

1
2i

()
1
d

1 z

||=1

has to be evaluated. For | z |= 1

Z
Z
1
z
()
d =
2i
2i
1 z
||=1

f ()
dd
1 z

||<1

()
d
1 z

||=1

1
2i

1 ()

1
2i

||=1

||<1

log(1 )

z
d
d
1 z

||=1

1
f ()
2i

z

ddd
1 z

||=1

Z
f ()

z
dd
1 z

||<1

so that
z

Z
||<1

dd
1
()
=
2

(1 z)

Z
||<1

22

 z
1 
f ()
+
dd .
1 z
1 z

For (32)

()

dd
= c1 z
z

||<1

1
+
2i

||=1

||<1

1 Z
1 ()

dd d
log(1 )
z

||<1

1
f ()

dd
dd

||<1

needs some modification. From

Z
1 z
1
1

1) d
(log(1 z ) 1) =
(log(1 )
2i
z

||=1
Z
1
dd
+
log(1 )

z
||<1

1
2i

Z
||=1

1
+

1
d
1)
(log(1 )
(1
z)

log(1 )

dd
z

||<1

1 1
= +

from which also

dd
log(1 )
z

||<1

dd = 0
log(1 )

||<1

follows, and
Z
1
1 dd

= log | z |2

1
2i

||<1

log | z |2

||=1

1
= log | z |2 +
2i

Z
log(1 z)
||=1

1
2i

Z
log(1 z)

||=1

= log | z |2 log(1 z )

23

(1 )

from what
Z

1
dd
=

||<1

1+

 dd
z

||<1

= z + (log
| z | log(1 z ))
2

and

dd
= log | |2

||<1

is seen,
z

dd
1
()
= c1 z +
2i
( z)

||<1

 1 z
 d
1 ()
log(1 z) + z

||=1

Z
f ()

z


+ log | z |2 log(1 z) log | |2 dd

||<1

follows.
Remark Instead of this constructive way the proof can be given by verification.
From (32)
Z
Z
 1
1
z d
z

0 ()

f ()
dd .
zwz (z) =

2i
1 z

z 1 z
||=1

||<1

This obviously coincides with 0 on D if and only if (31) is satisfied. Similarly

from
Z
Z
 1
1
1
1
d
wz (z) = c1
1 () log(1 z)
f ()

dd ,
2i

z
||=1

1
zwzz (z) =
2i

||<1

z d
1
1 ()

1 z

||=1

Z
f ()

z
dd
( z)2

||<1

it is seen that zwzz coinsides with 1 on D if and only if (31) holds. The other
two conditions are obviously satisfied.

2 The inhomogeneous polyanalytic

equation
As second order equations of special type were treated in the preceding section
model equations of third, forth, fifth etc. order can be investigated. From the

24

material presented it is clear how to proceed and what kind of boundary conditions can be posed. However, there is a variety of boundary conditions possible.
All kind of combinations of the three kinds, Schwarz, Dirichlet, Neumann conditions can be posed. And there are even others e.g. boundary conditions of
mixed type which are not investigated here.
As simple examples the Schwarz problem will be studied for the inhomogeneous
polyanalytic equation. Another possibility is the Neumann problem for the inhomogeneous polyharmonic equation, see [4, 5], and the Dirichlet problem, see
[2].
Lemma 3 For | z |< 1, | |< 1 and k N0

1
2

1

1
(1)k+1
( z + z)k+1 =
(z + z)k+1
k+1
k+1

+ 1 1 +

( z + z)k dd .

(33)

||<1

= i( z + z)k+1 /(k + 1) satisfies the Schwarz

Proof The function w()
condition
k+1
= i(z+

= 0 on D , Im w(0) = (1)
w ()
z)k in D , Re w()
(z+z)k+1 ,

k+1

so that according to [3], (33)

k+1
i
= i (1)
(z +z)k+1
w()
k+1
2

 1 +


1 1 +
( z + z)k dd .

||<1

This is (33).
Corollary 1 For | z |< 1 and k N0
Z 
1
1 1

( z + z)k dd = 0
2

(34)

||<1

and
1
2

1 + z
z

1 1 + z 
(1)k+1
( z + z)k dd =
(z + z)k+1 . (35)
k+1
1 z

||<1

Proof (34) and (35) are particular cases of (33) for = 0 and = z, respectively.
Theorem 14 The Schwarz problem for the inhomogeneous polyanalytic equation in the unit disc
zn w = f in D , Re z w = on D , Im z w(0) = 0 , 0 n 1 ,

25

is uniquely solvable for f L1 (D; C), C(D; R), c R, 0 n 1.

The solution is
n1
n1
X c
X (1) Z
+z
d

w(z) = i
()
(z + z) +
( z + z)
!
2i!
z

=0

=0

(1)n

 f () + z

2(n 1)!

||=1

f () 1 + z 
( z + z)n1 dd . (36)
1 z

||<1

Proof For n = 1 formula (36) is just [3], (33). Assuming it holds for n 1
rather than for n the Schwarz problem is rewritten as the system
zn1 w = in D , Re z w = on D , Im z w(0) = c , 0 n 2 ,
z = f in D , Re = n1 on D , Im (0) = cn1 ,
having the solution
w(z) = i

n2
X

n2

X (1)
c
(z + z) +
!
2i!

=0

=0

(1)n1
2(n 2)!

 () + z
z

Z
()

+z
d
(z z + z)
z

||=1

() 1 + z 
( z + z)n2 dd ,
1 z

||<1

(z) = icn2 +

1
2i

n1 ()

+ z d
z

||=1

1
2

 f () + z
z

f () 1 + z 
dd .
1 z

||<1

Using (35)
(1)n1
2(n 2)!

 () + z

() 1 + z 
( z + z)n2 dd
1 z

||<1

=i

cn1
(1)n1
(z + z)n1 +
(n 1)!
2i(n 2)!

n1 ()

||=1

1
2

 + 1 + z
z

||<1

+ 1 1 + z 
d
( z + z)n2 dd

1 z

(1)n
2(n 2)!

||<1

26

f ()

Z 
+ 1 + z 1 + 1 1 + z 

( z + z)n2 dddd
+

z
1

1
2

||<1

(1)n
+
2(n 2)!

||<1

1
2

f ()

Z 
1 + 1 + z + 1 1 + z 

( z + z)n2 dddd
+

z
1
1 z

||<1

follows. Because
+ 1 + z 1 + 1 1 + z
+
z 1 1 z
 2
 2
 2z
1  2
1
+
=

+
+1
1
z
1 z

1



4
1
2
2
1
2
1
=

+

+

z

z


z
2
2
2z
1
4z

+
+

1 z

1 z

1
z 1



+ z
1
+ z
1

z  1 1
= 2
+2

1 z

z z
z 1


1
1 1
z
+ z

1
+
+
= 2

1 z

z
1

+ z  2
1
2
1
2
1
2z
1 1 1


+ +
+
+

1 z

z
1



+z 1 +
1 1 +
1 + z 1 1 + z
1 1

+
=
+

z 1 z

27

and similarly
1 + 1 + z + 1 1 + z
+
1 z 1 z
 2
 2

1  2
1  2
=
1

1
z
1
1 z
 1
4
  2
2
=
+

+
1 z z 1
1

z 
1
2
1
4
1
2z
2
2
+

+
+
+ +
z

1 z
1 z

1 z

1
z  1 1

1 + z  1

+
+
=2

1 z z 1 1 z
1 + z  2
1
2
1
2
1
2z
1 1 1
=


+ +
+
+
z

1 z

1 z
1
1 + z  1 + 1 1 + 1 + z 1 1 + z  1 1
=

+
+

1 z 1 z 1 z
and applying (33), (34), and (35)
1
2

 + 1 + z

1 + 1 1 + z 
( z + z)n2 dd

z
1

||<1

Z
+ z 1
=
z 2

 1 + 1 1 +
1 + z 1 1 + z 
(z+ z)n2 dd

+
z 1 z
1

||<1

i
+ z h 1
(1)n1
(1)n1
( z + z)n1
(z + z)n1 +
(z + z)n1
n1
n1
z n 1
+ z 1
( z + z)n1 ,
z n 1
 + 1 + z 1 + 1 1 + z 
+
( z + z)n2 dd
z 1 1 z
=

1
2

Z
||<1

Z
1 + z 1
=
1 z 2

||<1

 1 + 1 1
1 + z 1 1 z 

+
(z+ z)n2 dd
z 1 z
1

i
1 + z h 1
(1)n1
(1)n1
(z + z)n1
(z +z)n1 +
(z +z)n1
n1
n1
1 z n 1

28

1 + z 1
( z + z)n1 ,
n

1 z

then
(1)n1
2(n 2)!

 () + z
z

() 1 + z 
( z + z)n2 dd
1 z

||<1

cn1
(1)n1
=i
(z + z)n1 +
(n 1)!
2i(n 1)!

Z
n1 ()

+z
d
( z + z)n1
z

||=1

(1)n
+
2(n 1)!

 f () + z
z

||<1

f () 1 + z 
( z + z)n1 dd .
1 z

This proves formula (36).

Theorem 15 The Dirichlet problem for the inhomogeneous polyanalytic equation in the unit disc
zn w = f in D , z w = on D , 0 n 1 ,
is uniquely solvable for f L1 (D; C), C(D; C), 0 n 1, if and only
if for 0 n 1
n1
X
=

z
2i

(1)

() ( z)
d
1 z ( )!

||=1

(1)n z

f () ( z)n1
dd = 0 .
1 z (n 1 )!

(37)

||<1

The solution then is

w(z) =

n1
X
=0

(1)
2i

(1)n

() ( z)
d
!
z

Z
||=1

f () ( z)n1
dd .
(n 1)! z

(38)

||<1

Proof For n = 1 condition (37) coinsides with [3], (34) and (38) is [3], (35).
Assuming Theorem 27 is proved for n 1 rather than for n the problem is
decomposed into the system
zn1 w = in D , z w = on D , 0 n 2 ,
z = f in D , z = n1 on D ,

29

with the solvability conditions (37) for 0 n 2 and instead of f

together with
Z
Z
d
dd
z
z
n1 ()
f ()

=0
2i
1 z

1 z
||=1

||<1

and the solutions (38) for n 1 instead of n and instead of f where

Z
Z
1
1
dd
d
(z) =
n1 ()

f ()
.
2i
z
z
||=1

||<1

Then for 0 n 2
Z
() ( z)n2
1
dd

1 z (n 2 )!
||<1

1
2i

(,
z)d
n1 ()

||=1

(,
z)dd
,
f ()

||<1

where
Z

z) = 1
(,

( z)n2
dd
(n 2 )!(1 z)

||<1

( z)n1
1

2i
(n 1 )!(1 z )

( z)n1
d
(n 1 !(1 z)

||=1

( z)n1
.

(n 1 )!(1 z )

The last equality holds because

Z

1
2i

( z)n1
1
d =

2i
(1 z)( )

||=1

1
2i

Z
||=1

z)n2

d = 0 .

||=1

30

( z)n1 d

( z)(1 )

Thus for 0 n 2
n2
X
=

z
2i

(1)

() ( z)
d
1 z ( )!

||=1

(1)n1 z

() ( z)n2
dd
1 z (n 2 )!

||<1

n1
X
=

z
2i

(1)

() ( z)
d
1 z ( )!

||=1

(1)n z

f () ( z)n1
dd = 0 .
1 z (n 1 )!

||<1

This is (37). For showing (38) similarly

1

Z
()

( z)n2
dd
(n 2)!( z)

||<1

1
=
2i

n1 (,
z)d 1
n1 ()

||=1

n1 (,
z)dd

f ()

||<1

with
z) =
n1 (,

( z)n2 dd
(n 2)!( z)

Z
||<1

( z)n2  1
1 
( z)n1
dd =

(n 2)!( z) z
(n 1)!( z)

||<1

1
2i(n 1)!( z)

 ( z)n1

( )

( z)n1 
( z)n1
d =
z
(n 1)!( z)

||=1

1
2i(n 1)!( z)

( z)n1

1
1  d
( z)n1

.
=

1 z
1
(n 1)!( z)

||=1

Hence,
w(z) =

n2
X
=0

(1)
2i

Z
||=1

31

() ( z)
d
!
z

(1)n1

() ( z)n2
dd
(n 2)! z

Z
||<1

n1
X
=0

(1)

() ( z)
d
!
z

2i
||=1

(1)n

f () ( z)n1
dd ,
(n 1)! z

||<1

i.e. (38) is valid.

Acknowledgement
The author is very grateful for the hospitality of the Mathematical Department
of the Simon Bolivr University. In particular his host, Prof. Dr. Carmen Judith
Vanegas has made his visit very interesting and enjoyable.

References
[1] Begehr, H.: Complex analytic methods for partial differential equations.
An introductory text. World Scientific, Singapore, 1994.
[2] Begehr, H.: Combined integral representations. Proc. 4th Intern. ISAAC
Congress, Toronto, 2003, World Sci., Singapore, to appear.
[3] Begehr, H.: Boundary value problem in complex analysis, I. Bol. Asoc.
Mat. Venezolana, to appear.
[4] Begehr, H., Vanegas, C. J.: Iterated Neumann problem for higher order
Poisson equation. Preprint, FU Berlin, 2003, Math. Nachr., to appear.
[5] Begehr, H., Vanegas, C. J.: Neumann problem in complex analysis. Proc.
11. Intern. Conf. Finite, Infinite Dimensional Complex Analysis, Appl.,
Chiang Mai, Tahiland, 2003, eds. P. Niamsup, A. Kananthai, 212225.
[6] Bitsadze, A. V.: About the uniqueness of the Dirichlet problem for elliptic
partial differential equations. Uspekhi Mat. Nauk 3 (1948), 6 (28), 211-212
(Russian).
[7] Haack, W., Wendland, W.: Lectures on partial and Pfaffian differential
equations. Pergamon Press, Oxford; Birkhauserr, Basel, 1969 (German).
Heinrich Begehr
I. Math. Inst., FU Berlin
Arnimallee 3
14195 Berlin, Germany
email: begehr@math.fu-berlin.de

32