Conformal Mappings: Exercises 20.1
Conformal Mappings: Exercises 20.1
20
Conformal Mappings
EXERCISES 20.1
Complex Functions as Mappings
1. For w =
1
z
, u =
x
x
2
+y
2
and v =
y
x
2
+y
2
. If y = x, u =
1
2
1
x
, v =
1
2
1
x
, and so v = u. The image is the
line v = u (with the origin (0, 0) excluded.)
2. If y = 1, u =
x
x
2
+ 1
and v =
1
x
2
+ 1
. It follows that u
2
+v
2
=
1
x
2
+ 1
= v and so u
2
+(v +
1
2
)
2
= (
1
2
)
2
. This
is a circle with radius r =
1
2
and center at (0,
1
2
) =
1
2
i. The circle can also be described by |w +
1
2
i| =
1
2
.
3. For w = z
2
, u = x
2
y
2
and v = 2xy. If xy = 1, v = 2 and so the hyperbola xy = 1 is mapped onto the line
v = 2.
4. If x
2
y
2
= 4, u = 4 and so the hyperbola x
2
y
2
= 4 is mapped onto the vertical line u = 4.
5. For w = Ln z, u = log
e
|z| and v = Arg z. The semi-circle |z| = 1, y > 0 may also be described by r = 1,
0 < < . Therefore u = 0 and 0 < v < . The image is therefore the open line segment from z = 0 to z = i.
6. If = /4, then v = = /4. In addition u = log
e
r will vary from to . The image is therefore the
horizontal line v = /4.
7. For w = z
1/2
= (re
i
)
1/2
= r
1/2
e
i/2
and =
0
, w =
r e
i0/2
. Therefore Arg w =
0
/2 and so the image is
the ray =
0
/2.
8. If r = 2 and 0
2
, w =
2 e
i/2
. Therefore |w| =
w + 1
2
since i = e
i/2
. Therefore the strip 0 y 1 is rotated through
90
2 e
i/4
z, the mapping is the composite of a rotation through 45
and a magnication by
=
2 . The image of the rst quadrant is therefore the angular wedge /4 Arg w 3/4.
19. The power function w = z
3
changes the opening of the wedge 0 Arg z /4 by a factor of 3. Therefore the
image region is 0 Arg w 3/4.
20. The power function w = z
1/2
changes the opening of the wedge 0 Arg z /4 by a factor of 1/2. Therefore
the image region is 0 Arg w /8.
21. We rst let z
1
= z i to map the region 1 y 4 to the region 0 y
1
3. We then let w = e
i/2
z
1
to rotate
this strip through 90
4
z
.
27. By Example 1, Section 20.1, z
1
= e
z
maps the strip 0 y onto the upper half-plane y
1
0, or
0 Arg z
1
. The power function w = z
3/2
1
changes the opening of this wedge by a factor of 3/2 so the wedge
0 Arg w 3/2 results. The composite of these two mappings is w = (e
z
)
3/2
= e
3z/2
.
28. The power function z
1
= z
2/3
maps the wedge 0 Arg z 3/2 to the upper half-plane 0 Arg z
1
. We
then let w = e
i/2
z
1
+2 = iz
1
+2 to rotate the upper half-plane through 90
(z) = 3(z
2
1), f is conformal at all points except z = 1.
2. Since f
(z) = 1 +e
z
and 1 +e
z
= 0 for z = i 2ni. Therefore f is conformal except for z = i 2ni.
4. f(z) = z + Ln z + 1 is analytic for all z except z = y 0. But f
(z) = z/(z
2
1)
1/2
is non-zero outside this interval.
Therefore f is conformal except for z = x, 1 x 1.
6. The function f(z) = i
1
2
[Ln(z +1) +Ln(z 1)] is analytic except on the branch cut x 1 0 or x 1, and
f
(z) =
1
2
_
1
z + 1
+
1
z 1
_
=
z
z
2
1
is non-zero for z = 0, 1. Therefore f is conformal except for z = x, x 1.
921
20.2 Conformal Mappings
7. f(z) = cos z = sin(/2 z) is the composite of z
1
= /2 z and w = sin z. The strip 0 x is mapped
onto the strip /2 x /2 by z
1
= /2 z and w = sin z maps this strip onto the region shown in
Figure 20.11(b). The horizontal segment z(t) = t + ib, 0 < t < is rst mapped to the horizontal segment
z
i
(t) = (/2 t) ib, 0 < t < . This latter segment is mapped onto the lower or upper portion of the ellipse
u
2
cosh
2
b
+
v
2
sinh
2
b
= 1
according to whether b > 0 or b < 0. See Figure 20.11.
8. f(z) = sinh z = i sin(iz) is the composite of z
1
= iz, z
2
= sin z
1
, and w = iz
2
. The strip /2 y /2,
x 0 is rotated through 90
so that the image is the strip /2 x /2, y 0. This latter strip is mapped
to the upper half-plane y
2
0 by z
2
= sin z
1
, and w = iz
2
rotates this upper half-plane through 90
. The
nal image region is the half-plane u 0. A vertical line segment in the original strip is mapped to the right
hand side of the ellipse
u
2
sinh
2
b
+
v
2
cosh
2
b
= 1.
See the gures below.
9. f(z) = (sin z)
1/4
is the composite of z
1
= sin z and w = z
1/4
1
. The region /2 x /2, y 0 is mapped to
the upper half-plane y
2
0 by z
1
= sin z (See Example 2) and the power function w = z
1/4
1
maps this upper
half-plane to the angular wedge 0 Arg w /4. The real interval [/2, /2] is rst mapped to [1, 1] and
then to the union of the line segments from e
i
4
to 0 and 0 to 1. See the gures below.
10. From Example 3, u = (r + 1/r) cos and v = (r 1/r) sin . If r = 1, u = 2 cos and v = 0. Therefore the
image of the circle |z| = 1 is the real interval [2, 2]. If 0 < r < 1, u
2
/a
2
+v
2
/b
2
= 1 where a = r +1/r > 2 and
b = 1/r r. The resulting ellipse together with [2, 2] ll up the entire w-plane. Therefore the image of the
region |z| 1 is the entire w-plane. If z = x and 1 x 1, w = x + 1/x. Therefore w is real and, from an
analysis of the graph of w = x + 1/x for 1 x 1, |w| 2. Therefore the segment [1, 1] (with 0 excluded)
is mapped on those points w = u on the u-axis for which |u| 2.
922
20.2 Conformal Mappings
11. Using H-4 with a = 2, w = cos(z/2) maps R onto the target region R
. The image of AB
is shown in the gure.
12. Using C-3 w = e
z
maps R onto the target region R
.
Therefore
w =
_
_
1 z
1 +z
_
2
1
_
1/2
and the image of AB is shown in the gure.
15. Using H-6, z
1
=
e
/z
+e
/z
e
/z
e
/z
maps R onto the upper half-plane y
1
0, and w = z
1/2
1
maps
this half-plane onto the target region R
. Therefore
w =
_
e
/z
+e
/z
e
/z
e
/z
_1/2
and the image of AB is shown in the gure.
16. We can translate R to the origin, magnify by 2, and then use H-1 to reach the target region
R
. Therefore z
1
= 2(z
1
2
), w = i
1 z
1 +z
and so
w = i
1 (2z 1)
1 + (2z 1)
= i
1 z
z
.
The image of AB is shown in the gure.
17. z
1
= Ln z maps R onto the horizontal strip 0 y
1
, and, to prepare this strip for mapping by the sine
function, we let z
2
= iz
1
/2 to obtain the vertical strip /2 x /2. Finally w = sin z
2
maps this
vertical strip onto the target region R
. Therefore
w = sin
_
iLn z
2
_
.
The image of AB is the real interval (, 1].
923
20.2 Conformal Mappings
18. Using E-9, z
1
= cosh z maps R onto the upper half-plane y
1
0. Using M-7,
w = z
1
+ Ln z
1
+ 1 maps this half-plane onto the target region R
. Therefore
w = cosh z + Ln(cosh z) + 1 and the image of AB is shown in the gure.
In Exercises 19-22, we nd a conformal mapping w = f(z) that maps the given
region R onto the upper half-plane v 0 and transfers the boundary conditions
so that the resulting Dirichlet problem is as shown in the gure.
19. f(z) = z
4
and so u = U(f(z)) =
1
Arg z
4
. The solution may also be written as u(r, ) = 4/.
20. f(z) =
_
1 +z
1 z
_
2
, using H-5, and so u = U(f(z)) =
1
Arg
_
1 +z
1 z
_
2
.
21. f(z) = i
1 z
1 +z
, using H-1, and so u = U(f(z)) =
1
Arg
_
1 z
1 +z
_
. The solution may also be written as
u(x, y) =
1
tan
1
_
1 x
2
y
2
2y
_
.
22. f(z) =
1
2
_
z +
1
z
_
, using H-3 with a = 1, and so u = U(f(z)) =
1
Arg
1
2
_
z +
1
z
_
. The solution may also be
written as u(x, y) =
1
tan
1
_
y
x
x
2
+y
2
1
x
2
+y
2
+ 1
_
.
In Exercises 23-26, we nd a conformal mapping w = f(z) that maps the given
region R onto the upper half-plane v 0 and transfers the boundary conditions
so that the resulting Dirichlet problem is as shown in the gure.
23. f(z) = z
2
and c
0
= 1. Therefore u =
1
[Arg(z
2
1) Arg(z
2
+ 1)].
24. The mapping z
1
= z
2
maps R onto the region R
1
dened by y
1
0, |z
1
| 1 and shown in H-3, and
w =
1
2
_
z
1
+
1
z
1
_
maps R
1
onto the upper half-plane v 0. Letting c
0
= 5,
u =
5
_
Arg
_
1
2
_
z
2
+
1
z
2
_
1
_
Arg
_
1
2
_
z
2
+
1
z
2
_
+ 1
__
.
25. f(z) = e
z
, using H-2 with a = 1. Letting c
0
= 10,
u =
10
[Arg(e
z
1) Arg(e
z
+ 1)].
26. f(z) = cos(z/2) using H-4 with a = 2. Letting c
0
= 4,
u =
4
x
2
+
2
y
2
,
2
u
x
2
+
2
u
y
2
=
4
x
4
+ 2
4
x
2
y
2
+
4
y
4
= 0
924
20.3 Linear Fractional Transformations
since is assumed to be biharmonic.
(b) If g = u +iv, then = Re( zg(z)) = xu +yv.
x
2
= 3
u
x
+x
2
u
x
2
+y
2
v
x
2
y
2
= 2
v
y
+x
2
u
y
2
+y
2
v
y
2
.
Since u and v are harmonic and
u
x
=
v
y
,
x
2
+
2
y
2
= 2
u
x
+ 2
v
y
= 4
u
x
.
Now u
1
=
u
x
is also harmonic and so
2
u
1
x
2
+
2
u
1
y
2
= 0. But
2
u
1
x
2
+
2
u
1
y
2
=
1
4
_
x
4
+ 2
4
x
2
y
2
+
4
y
4
_
and so is biharmonic.
EXERCISES 20.3
Linear Fractional Transformations
1. (a) For T(z) = i/z, T(0) = , T(1) = i, and T() = 0.
(b) If |z| = 1, |w| = |i/z| = 1/|z| = 1. Therefore, the image of the circle |z| = 1 is the circle |w| = 1 in the
w-plane. The circle |z 1| = 1 passes through the pole at z = 0 and so the image is a line. Since T(2) =
1
2
i
and T(1 +i) =
1
2
+
1
2
i, the image is the line v =
1
2
.
(c) The disk |z| 1 is mapped onto the disk |w| 1.
2. (a) For T(z) =
1
z 1
, T(0) = 1, T(1) = , and T() = 0.
(b) The circle |z| = 1 passes through the pole at z = 1 and so the image is a line. Since T(1) =
1
2
and
T(i) =
1
2
1
2
i, the image is the line u =
1
2
. If |z 1| = 1, |w| = 1/|z 1| = 1 and the image is the
circle |w| = 1 in the w-plane.
(c) Since T(0) = 1, the image of the disk |z| 1 is the half-plane u =
1
2
.
3. (a) For T(z) =
z + 1
z 1
, T(0) = 1, T(1) = , and T() = 1.
(b) The circle |z| = 1 passes through the pole at z = 1 and so the image is a line. Since T(1) = 0 and
T(i) = i, the image is the line u = 0. If |z 1| = 1,
|w 1| =
z + 1
z 1
1
=
2
|z 1|
= 2
and so the image is the circle |w 1| = 2 in the w-plane.
(c) Since T(0) = 1, the image of the disk |z| 1 is the half-plane u 0.
925
20.3 Linear Fractional Transformations
4. (a) For T(z) =
z i
z
, T(0) = , T(1) = 1 i, and T() = 1.
(b) T(i) = 0, T(1) = 1 i and T(i) = 2. The image of |z| = 1 is therefore a circle which passes through 0,
1 i, and 2. This is the circle |w 1| = 1. The circle |z 1| = 1 passes through the pole at z = 0 and so
the image is a line. Since T(2) = 1
1
2
i and T(1 +i) =
1
2
1
2
i, the image is the line v =
1
2
.
(c) T(
1
2
) = 1 2i which is exterior to the circle |w 1| = 1. Therefore, the image of |z| 1 is |w 1| 1, the
exterior of the circle |w 1| = 1 in the w-plane.
5. S
1
(T(z)) =
az +b
cz +d
where
_
a b
c d
_
= adj
__
i 1
1 1
___
1 0
i 1
_
=
_
1 i 1
2 i
_
.
Therefore,
S
1
(T(z)) =
(1 i)z + 1
2z i
=
(1 +i)z 1
2z +i
and S
1
(w) =
w 1
w +i
=
w + 1
w i
.
6. S
1
(T(z)) =
az +b
cz +d
where
_
a b
c d
_
= adj
__
2 1
1 1
___
i 0
1 2i
_
=
_
1 +i 2i
2 i 4i
_
.
Therefore,
S
1
(T(z)) =
(1 +i)z + 2i
(2 i)z 4i
and S
1
(w) =
w 1
w + 2
.
7. S
1
(T(z)) =
az +b
cz +d
where
_
a b
c d
_
= adj
__
1 2
1 1
___
2 3
1 3
_
=
_
0 3
1 0
_
.
Therefore,
S
1
(T(z)) =
3
z
=
3
z
and S
1
(w) =
w + 2
w + 1
=
w 2
w 1
.
8. S
1
(T(z)) =
az +b
cz +d
where
_
a b
c d
_
= adj
__
2 i 0
1 1 i
___
1 1 +i
i 2
_
=
_
1 i 2
2i 3 +i
_
.
Therefore,
S
1
(T(z)) =
(1 i)z + 2
2iz 3 +i
and S
1
(w) =
(1 i)w
w + 2 i
=
(1 +i)w
w 2 + 1
.
9. T(z) =
(z z
1
)(z
2
z
3
)
(z z
3
)(z
2
z
1
)
maps z
1
, z
2
, z
3
to 0, 1, . Therefore, T(z) =
(z + 1)(2)
(z 2)(1)
= 2
z + 1
z 2
maps 1, 0,
2 to 0, 1, .
10. T(z) =
(z i)i
(z +i)(i)
=
z i
z +i
maps i, 0, i to 0, 1, .
926
20.3 Linear Fractional Transformations
11. S(w) =
(w w
1
)(w
2
w
3
)
(w w
3
)(w
2
w
1
)
maps w
1
, w
2
, w
3
to 0, 1, and so S
maps 0, 1, to w
1
, w
2
, w
3
. Therefore,
z =
(w 0)(i 2)
(w 2)(i 0)
and so w =
2z
z 1 2i
maps 0, 1, to 0, i, 2.
12. As in Exercise 11, z =
(w 1 i)(1 +i)
(w 1 +i)(1 i)
and, solving for w, w =
2z 2
(1 +i)z 1 +i
maps 0, 1, to 1 +i, 0,
1 i.
13. Using the cross-ratio formula (7),
(w i)(1)
w(1)
=
(z + 1)(1)
(z 1)(1)
and so w =
i
2
z 1
z
maps 1, 0, 1 to i, , 0.
14. Using the cross-ratio formula (7),
(1)(i 1)
(w 1)(1)
=
(z + 1)(1)
(z 1)(1)
and so w =
(2 +i)z i
z + 1
maps 1, 0, 1 to , i, 1.
15. Using the cross-ratio formula (7),
S(w) =
(w + 1)(3)
(w 3)(1)
=
(z 1)(2i)
(z +i)(i 1)
= T(z).
We can solve for w to obtain
w = 3
(1 +i)z + (1 i)
(3 + 5i)z 3 5i
.
Alternatively we can apply the matrix method to compute w = S
1
(T(z)).
16. Using the cross-ratio formula (7),
S(w) =
(w i)(i 1)
(w 1)(2i)
=
(z 1)(2i)
(z +i)(i 1)
= T(z).
We can solve for w to obtain
w =
(1 + 2i)z i
iz + 1 2i
.
Alternatively we can apply the matrix method to compute w = S
1
(T(z)).
17. From Example 2, z =
w + 2
w 1
maps the annular region 1 < |w| < 2 onto the
region R and the circle |w| = 1 corresponds to the line x = 1/2. Solving for
w, w =
z + 2
z 1
maps R onto the annulus and the transferred boundary conditions
are shown in the gure to the right. The solution to this new Dirichlet problem is
U = log
e
r/ log
e
2 and so
u = U(f(z)) =
1
log
e
2
log
e
z + 2
z 1
is the solution to the Dirichlet problem in Figure 20.37. The level curves are the images of the level curves of
U, |w| = r for 1 < r < 2 under the mapping z =
w + 2
w + 1
. Since these circles do not pass through the pole at
w = 1, the images are circles.
927
20.3 Linear Fractional Transformations
18. The mapping T(z) =
1
2
z + 1
z
maps 1, 1, 0 to 0, 1, and maps each of the
two circles in R to lines since both circles pass through the pole at z = 0. Since
T(
1
2
+
1
2
i) = 1 i and T(1) = 1, the circle |z
1
2
| =
1
2
is mapped onto the line
u = 1. Likewise, the circle |z +
1
2
| =
1
2
is mapped onto the line u = 0. The transferred
boundary conditions are shown in the gure and U(u, v) = u is the solution. The
solution to the Dirichlet problem in Figure 20.38 is
u = U(T(z)) = Re
_
1
2
z + 1
z
_
=
1
2
+
1
2
x
x
2
+y
2
.
The level curves u = c are the circles with centers on the x-axis which pass through the origin. The level curve
u =
1
2
, however, is the vertical line x = 0.
19. The linear fractional transformation that sends 1, i, i to 0, 1, 1 satises the cross-ratio equation
(w 0)(2)
(w + 1)(1)
=
(z 1)(2i)
(z +i)(i 1)
.
Solving for w, w = i
1 z
1 +z
= T(z). Since T(0) = i, the image of the disk |z| 1 is the upper half-plane v 0.
20. The linear fractional transformation that sends 1, i, 1 to , i, 0 is T(z) =
1 +z
1 z
and so f(z) =
_
1 +z
1 z
_
2
maps 1, i, 1 to , 1, 0. The upper semi-circle is mapped by T to the positive imaginary axis, and the real
interval [1, 1] is mapped to the positive real axis. Since T
_
i
2
_
=
3
5
+
4
5
i, the image of R under T is the rst
quadrant. w = z
2
1
doubles the size of the opening so that the image under f is the upper half-plane v 0. See
the gures below.
21. T
2
(T
1
(z)) =
a
2
T
1
(z) +b
2
c
2
T
1
(z) +d
2
=
a
2
a
1
z +b
1
c
1
z +d
1
+b
2
c
2
a
1
z +b
1
c
1
z +d
1
+d
2
=
a
1
a
2
z +a
2
b
1
+b
2
c
1
z +b
2
d
1
a
1
c
2
z +b
1
c
2
+c
1
d
2
z +d
1
d
2
=
(a
1
a
2
+c
1
b
2
)z + (b
1
a
2
+d
1
b
2
)
(a
1
c
2
+c
1
d
2
)z + (b
1
c
2
+d
1
d
2
)
22. |w w
1
| = |w w
2
| =(u u
1
)
2
+ (v v
1
)
2
=
2
[(u u
2
)
2
+ (v v
2
)
2
]
The latter equation may be put in the form
Au
2
+Bv
2
+Cu +Dv +F = 0
where A = B = 1
2
. If = 1, the line Cu + Dv + F = 0 results. If > 0 and = 1, then the equation
denes a circle.
928
20.4 Schwarz-Christoel Transformations
EXERCISES 20.4
Schwarz-Christoffel Transformations
1. arg f
(t) =
1
2
Arg(t 1) =
_
/2, t < 1
0, t > 1
. Since f(1) = 0, the image is the rst quadrant.
2. arg f
(t) =
1
3
Arg(t + 1) =
_
/3, t < 1
0, t > 1
. In (2),
1
= 2/3 and since f(1) = 0, the image is the wedge
0 Arg w 2/3.
3. arg f
(t) =
1
2
Arg(t + 1) +
1
2
Arg(t 1) =
_
_
_
0, t < 1
/2, 1 < t < 1
0, t > 1
and
1
= /2 and
2
= 3/2. Since f(1) = 0, the image of the upper half-plane
is the region shown in the gure.
4. arg f
(t) =
1
2
Arg(t + 1)
3
4
Arg(t 1) =
_
_
_
5/4, t < 1
3/4, 1 < t < 1
0, t > 1
and
1
= /2 and
2
= /4. Since f(1) = 0, the image of the upper half-plane is
the region shown in the gure.
5. Since
1
=
2
=
3
= /2,
i
/ 1 = 1/2 and so f
(z) = A(z + 1)
2/3
z
1/2
for
some constant A.
7. Since
1
=
2
= 2/3,
i
/ 1 = 1/3 and so f
(z) = A(z + 1)
1/3
z
1/3
for some constant A.
8. Since
1
= 3/2 and
2
= /4,
1
/ 1 = 1/2 and
2
/ 1 = 3/4. Therefore, f
(z) = A(z + 1)
1/2
z
3/4
for
some constant A.
9. Since
1
=
2
= /2, f
(z) = A(z + 1)
1/2
(z 1)
1/2
= A(z
2
1)
1/2
. Therefore,
f(z) = A
_
z(z
2
1)
1/2
2
1
2
Ln(z + (z
2
1)
1/2
_
+B.
but f(1) = ai and f(1) = ai. It follows that
ai = f(1) = B, ai = f(1) = A
_
i
2
_
+B
and so B = ai and A = 4a/. Therefore,
f(z) =
4a
_
z(z
2
1)
1/2
2
1
2
Ln(z + (z
2
1)
1/2
)
_
+ai =
4a
_
z(z
2
1)
1/2
2
1
2
cosh
1
z
_
+ai.
11. f
(z) = A(z + 1)
(1/)1
z
(2/)1
(z 1)
(3/)1
from (3). Since f(1) = i,
1
as w
1
in the
horizontal direction. Likewise
2
0 and
3
. This suggests we examine f
(z) = Az
1
= A/z. Therefore,
929
20.4 Schwarz-Christoel Transformations
f(z) = ALn z + B. But f(1) = i and f(1) = 0. It follows that A = 1 and B = 0 so that f(z) = Ln z. We
veried in Example 1, Section 20.1 that f(z) = Ln z maps the upper half-plane y 0 to the horizontal strip
0 v .
12. From (3), f
(z) = Az
3/4
(z 1)
(2/)1
. But
2
as 0. This suggests that we examine f
(z) = Az
3/4
.
Therefore, f(z) = A
1
z
1/4
+B
1
. But f(0) = 0 and f(1) = 1 so that B
1
= 0 and A
1
= 1. Hence f(z) = z
1/4
and
we recognize that this power function maps the upper half-plane onto the wedge 0 Arg w /4.
13. From (3), f
(z) = A(z + 1)
1/2
z(z 1)
1/2
= A
z
(z
2
1)
1/2
.
Therefore, f(z) = A(z
2
1)
1/2
+B. But f(1) = f(1) = 0 and f(0) = ai. This implies that B = 0 and ai = Ai
or A = a. Therefore, f(z) = a(z
2
1)
1/2
. By expressing f(z) as the composite of z
1
= z
2
, z
2
= z
1
1, z
3
= z
1/2
2
and w = az
3
we can show that the image of the upper half-plane is R
.
14. If w(t) = u(t) +iv(t), then w
(t) = u
(t) +iv
(t) and so
tan(arg w
(t)) =
v
(t)
u
(t)
=
dv
du
.
If arg w
Arg
_
z
z + 1
_
+
1
Arg
_
z 1
z
_
.
2. Using (3) with x
0
= 2, x
1
= 0, x
2
= 1 and u
1
= 5 and u
2
= 1,
u =
5
Arg
_
z
z + 2
_
+
1
Arg
_
z 1
z
_
.
3. The harmonic function
u
1
=
5
[ Arg(z 1)] =
_
5, x > 1
0, x < 1
, and u
2
=
1
Arg
_
z + 1
z + 2
_
+
1
Arg
_
z
z + 1
_
from (3) satises all boundary conditions except that u
2
= 0 for x > 1. Therefore u = u
1
+ u
2
is the solution
to the given Dirichlet problem.
4. The harmonic function
u
1
=
1
Arg(z + 2) =
_
1, x < 2
0, x > 2
, and u
2
=
1
Arg
_
z + 1
z + 2
_
+
1
Arg
_
z + 1
z 1
_
satises all boundary conditions except that u
2
= 0 for x < 2. Therefore u = u
1
+ u
2
is the solution to the
given Dirichlet problem.
930
20.5 Poisson Integral Formulas
5. By Theorem 20.5,
u(x, y) =
y
_
1
0
t
2
(x t)
2
+y
2
dt.
If we let s = x t, this integral can be expressed in terms of the natural log and inverse tangent. Using Maple
we obtain
u =
y
_
y
2
x
2
y
_
tan
1
_
x 1
y
_
tan
1
_
x
y
__
+xln
(x 1)
2
+y
2
x
2
+y
2
+ 1
_
.
6. From Theorem 20.5,
u(x, y) =
y
cos t
(x t)
2
+y
2
dt =
y
cos(x s)
s
2
+y
2
ds
letting s = x t. But cos(x s) = cos xcos x + sin xsin s. It follows that
u(x, y) =
y cos x
cos s
s
2
+y
2
ds +
y sin x
sin s
s
2
+y
2
ds =
y cos x
_
e
y
y
_
= e
y
cos x, y > 0.
7. The mapping f(z) = z
2
maps R onto the upper half-plane R
. The corresponding
boundary value problem in R
Arg(w + 1) +
1
Arg
_
w 1
w
_
is the solution in R
. Therefore
u = U(f(z)) =
5
Arg(z
2
+ 1) +
1
Arg
_
z
2
1
z
2
_
is the solution to the original Dirichlet problem in R.
8. Using H-4 with a = 3, f(z) = cos(z/3) maps R onto the upper half-plane R
. The
corresponding Dirichlet problem in R
[ Arg(w 1)] +
1
Arg(w + 1)
is the solution in R
. Therefore
u = U(f(z)) = 1 +
1
. The corresponding
Dirichlet problem in R
Arg
_
w
w + 1
_
+
1
Arg
_
w 1
w
_
=
1
_
Arg
_
w 1
w
_
Arg
_
w
w + 1
__
.
The harmonic function u = U(f(z)) may be simplied to
u =
1
_
Arg
_
(1 i)z (1 +i)
1 z
_
Arg
_
1 z
(1 +i)z + 1 i
__
and is the solution to the original Dirichlet problem in R.
931
20.5 Poisson Integral Formulas
10. Using H-5, f(z) =
_
1 +z
1 z
_
2
maps R onto the upper half-plane R
. The corre-
sponding Dirichlet problem in R
Arg
_
w
w + 1
_
+
1
[ Arg(w 1)] = 1
1
Arg(w 1) +
1
w
Arg
_
w
w + 1
_
.
The harmonic function u = U(f(z)) may be simplied to
u = 1
1
Arg
_
4z
(1 z)
2
_
+
1
Arg
_
(1 +z)
2
2(1 +z
2
)
_
.
11. From Theorem 20.6, u(x, y) =
1
2
_
t
2
2
1 |z|
2
|e
it
z|
2
dt. Therefore,
u(0, 0) =
1
2
_
t
2
2
dt =
1
3
.
To estimate u(0.5, 0) and u(0.5, 0) we must use a numerical integration method. With the aid of Simpsons
Rule, u(0.5, 0) = 0.1516 and u(0.5, 0) = 0.5693.
12. From Theorem 20.6, u(x, y) =
1
2
_
e
|t|
1 |z|
2
|e
it
z|
2
dt. Therefore,
u(0, 0) =
1
2
_
e
|t|
dt =
1
_
0
e
t
dt =
1
(1 e
).
With the aid of Simpsons Rule, u(0.5, 0) = 0.5128 and u(0.5, 0) = 0.1623.
13. u(0.0) =
1
2
_
u(e
it
) dt. The latter integral is just the average value of u(e
it
) for t .
14. For u(e
i
) = cos 2, the Fourier series solution (6) reduces to
u(r, ) = r
2
cos 2 = Re(z
2
) or u(x, y) = x
2
y
2
.
The corresponding system of level curves is shown in the gure.
15. For u(e
i
) = sin + cos , the Fourier series solution (6) reduces to
u(r, ) = r sin +r cos or u(x, y) = y +x.
The corresponding system of level curves is shown in the gure.
932
20.6 Applications
EXERCISES 20.6
Applications
1. g(z) = cos
0
i sin
0
= e
i0
is analytic everywhere and so div F = 0 and curl F = 0
by Theorem 20.7. A complex potential is G(z) = e
i0
z and
(x, y) = Re(G(z)) = xcos
0
+y sin
0
.
The equipotential lines (corresponding to
0
= /6) are shown in the gure.
2. g(z) = y + xi = i(x + iy) = iz is analytic everywhere and so div F = 0 and
curl F = 0 by Theorem 20.7. A complex potential is G(z) =
i
2
z
2
and
(x, y) = Re
_
i
2
z
2
_
= xy.
The equipotential lines xy = c are shown in the gure and are hyperbolas.
3. g(z) =
x
x
2
+y
2
y
x
2
+y
2
i =
1
z
is analytic for z = 0 and so div F = 0 and
curl F = 0 by Theorem 20.7. A complex potential is G(z) = Lnz and
(x, y) = Re(G(z)) =
1
2
log
e
(x
2
+y
2
).
The equipotential lines (x, y) = c are circles x
2
+ y
2
= e
2c
and are shown in the
gure.
4. g(z) =
x
2
y
2
2xyi
(x
2
+y
2
)
2
=
1
z
2
is analytic for z = 0 and so div F = 0 and
curl F = 0 by Theorem 20.7. A complex potential is G(z) =
1
z
and
(x, y) = Re
_
1
z
_
=
x
x
2
+y
2
.
The equipotential lines
x
x
2
+y
2
= c can be writen as
_
x +
1
2c
_
2
+ y
2
=
_
1
2c
_
2
for c = 0. See the gure.
5. The mapping f(z) = z
4
maps the wedge 0 Arg z /4 to the upper half-plane R
and U =
1
Arg w is the
solution to the corresponding Dirichlet problem in R
. Therefore,
(x, y) = U(z
4
) =
1
Arg z
4
=
4
Arg z
is the potential in the wedge. A complex potential is G(z) =
4
, the equipotential
933
20.6 Applications
lines are the rays =
4
c. Finally
F = G
(z) =
4
1
z
=
4
_
x
x
2
+y
2
,
y
x
2
+y
2
_
.
6. The function f(z) =
1
z
maps the original region R to the strip
1
2
v 0 (see
Example 2, Section 20.1). The boundary conditions transfer as shown in the gure.
U = 2v is the solution in the horizontal strip and so
(x, y) = 2Im
_
1
z
_
=
2y
x
2
+y
2
is the potential in the original region R. The equipotential lines
2y
x
2
+y
2
= c may be written as
x
2
+
_
y +
1
c
_
2
=
_
1
c
_
2
for c = 0 and are circles. If c = 0, we obtain the line y = 0. Note that
(x, y) = Re
_
2i
z
_
and so G(z) =
2i
z
is a complex potential. The corresponding vector eld is
F = G
(z) =
2i
z
2
=
_
4xy
(x
2
+y
2
)
2
,
2(x
2
y
2
)
(x
2
+y
2
)
2
_
.
7. Using H-5, w =
_
1 z
1 +z
_
2
=
_
z 1
z + 1
_
2
maps R onto the upper half-plane R
and U =
1
. Therefore,
=
1
Arg
_
z 1
z + 1
_
2
.
In R
5
2
and r
0
=
7 3
5
2
, so T(z) =
z a
az 1
maps R onto
the annular region R
dened by r
0
|w| 1. The solution to the corresponding Dirichlet problem in R
is
U =
log
e
|w|
log
e
r
0
and so
=
1
log
e
r
0
log
e
z a
az 1
.
The level curves of U are circles |w| = r where r
0
< r < 1, and the equipotential lines (x, y) = c are the images
of these circles under the inverse transformation T
1
(w) =
w +a
aw + 1
. T
1
has a pole at w =
1
a
=
2
3 +
5
( 0.38) and r
0
<
1
a
< 1. All circles |w| = r are mapped onto circles in the z-plane with the exception of
|w| =
1
a
which is mapped onto a line.
9. (a) (x, y) = Im(z
4
) = 4xy(x
2
y
2
) and so (x, y) = 0 when y = x and y = 0.
934
20.6 Applications
(b) V = G
(z) = 4z
3
= 4(x
3
3xy
2
, y
3
3x
2
y)
(c) In polar coordinates r
4
sin 4 = c or r = (c csc 4)
1/4
, for 0 < < /4, are the
streamlines. See the gure.
10. (a) Since G(re
i
) = r
2/3
e
i2/3
, (r, ) = Im(G(re
i
)) = r
2/3
sin
2
3
. Note that = 0 on the boundary where
= 0 and = 3/2.
(b) V = G
(z) =
2
3
z
1/3
. Therefore, letting z = re
i
, V =
2
3
r
1/3
(cos(/3), sin(/3)) for 0 < < 3/2.
(c) r
2/3
sin(2/3) = c implies that r = [c csc(2/3)]
2/3
for 0 < < 3/2. The
streamlines are shown in the gure.
11. (a) (x, y) = Im(sin z) = cos xsinh y and (x, y) = 0 when x = /2 or when y = 0.
(b) V = G
(z) =
i
(1 z
2
)
1/2
=
i
(1 z
2
)
1/2
(c) The streamlines are the images of the lines v = b, /2 < b < /2 under
z = i sin w and are therefore hyperbolas. See Example 2, Section 20.2, and the
gure. Note that at z = 0, v = i and the ow is downward.
13. (a) If z = re
i
, G(re
i
) = r
2
e
2i
+
1
r
2
e
2i
and so
(r, ) = Im(G(re
i
)) =
_
r
2
1
r
2
_
sin 2.
Note that = 0 when r = 1 or when either = 0 or = /2. Therefore = 0 or the boundary of R.
(b) V = G
(z) = (2z 2z
3
) and so in polar coordinates
V = 2re
i
2
r
3
e
3i
= 2(r cos r
3
cos 3, r sin r
3
sin 3).
(c) In rectangular coordinates, the streamlines are
(x, y) = 2xy
_
1
1
(x
2
+y
2
)
2
_
= c.
935
20.6 Applications
14. (a) (x, y) = Im(e
z
) = e
x
sin y and so = 0 when y = 0 or . Therefore = 0 on the boundary of R.
(b) V = G
(z) = e
z
= (e
x
cos y e
x
sin y)
(c) The streamlines are e
x
sin y = c and so x = log
e
(c csc y) for 0 < y < . See
the gure.
15. (a) For f(z) = i
1
2
[Ln(z + 1) + Ln(z 1)]
f(t) = i
1
2
[log
e
|t + 1| + log
e
|t 1| +iArg(t + 1) +iArg(t 1)]
and so Im(f(t)) =
_
_
_
0, t < 1
/2, 1 < t < 1
, t > 1
. Hence Im(G(z)) = (x, y) = 0 on the boundary of R.
(b) x =
1
2
[log
e
|t + 1 +ic| + log
e
|t 1 +ic|], y =
1
2
[Arg(t + 1 +ic) + Arg(t 1 +ic)] for c > 0
(c)
16. (a) For f(z) = (z
2
1)
1/2
,
f(t) = |t
2
1|
1/2
cos
_
1
2
Arg(t
2
1)
_
+i|t
2
1|
1/2
sin
_
1
2
Arg(t
2
1)
_
and so f(t) =
_
|t
2
1|
1/2
, |t| > 1
i|t
2
1|
1/2
, |t| < 1
. Hence Im(G(z)) = 0 on the boundary of R.
(b) x = |(t +ic)
2
1|
1/2
cos
_
1
2
Arg((t +ic)
2
1)
_
, y = |(t +ic)
2
1|
1/2
sin
_
1
2
Arg((t +ic)
2
1)
_
for c > 0
(c)
17. (a) For f(z) =
1
[(z
2
1)
1/2
+ cosh
1
z],
f(t) =
1
_
(t
2
1)
1/2
+ cosh
1
t
_
=
1
_
(t
2
1)
1/2
+ Ln(t + (t
2
1)
1/2
)
_
and so Im(f(t)) =
_
1, t < 1
0, t > 1
and Re(f(t)) = 0, for 1 < t < 1. Hence Im(G(z)) = (x, y) = 0 on the
boundary of R.
(b) x = Re
_
1
_
_
(t +ic)
2
1
_
1/2
+ cosh
1
(t +ic)
_
_
,
y = Im
_
1
_
_
(t +ic)
2
1
_
1/2
+ cosh
1
(t +ic)
_
_
for c > 0
936
20.6 Applications
(c)
18. (a) For f(z) = 2(z + 1)
1/2
+ Ln
_
(z + 1)
1/2
1
(z + 1)
1/2
+ 1
_
,
f(t) = 2(t + 1)
1/2
+ Ln
(t + 1)
1/2
1
(t + 1)
1/2
+ 1
.
If we write (t + 1)
1/2
= |t + 1|
1/2
e
(i/2)Arg(t+1)
, we may conclude that
Im(f(t)) =
_
0, t > 0
, 1 < t < 0
and Re(f(t)) = 0 for t < 1.
Therefore Im(G(z)) = (x, y) = 0 on the boundary of R.
(b) x = Re
_
2(t +ic + 1)
1/2
+ Ln
(t +ic + 1)
1/2
1
(t +ic + 1)
1/2
+ 1
_
y = Im
_
2(t +ic + 1)
1/2
+ Ln
(t +ic + 1)
1/2
1
(t +ic + 1)
1/2
+ 1
_
for c > 0
(c)
19. In Example 5, V = (2x, 2y) and so the only stagnation point occurs at z = 0. In Example 6, V = 1 1/ z
2
and so if V = 0, z
2
= 1. Therefore z = 1, 1 are the only stagnation points.
20. (a) (x, y) = Im(G(z)) = kArg(z x
1
) and so if (x, y) = c, Arg(z x
1
) is constant. This implies that the
streamlines are rays with vertex at z = x
1
.
(b) V = G
(z) =
k
z x
1
=
k
z x
1
=
k
|z x
1
|
2
(z x
1
).
The direction of the ow is determined by the sign of k, and if k < 0 the ow is directed towards z = x
1
.
21. f(z) = z
2
maps the rst quadrant onto the upper half-plane and f(
0
) = f(1) = 1. Therefore G(z) = Ln(z
2
1)
is the complex potential, and so
(x, y) = Arg(z
2
1) = tan
1
_
2xy
x
2
y
2
1
_
is the streamline function where tan
1
is chosen to be between 0 and . If (x, y) = c, then x
2
+Bxyy
2
1 = 0
where B = 2 cot c. Each hyperbola in the family passes through (1, 0) and the boundary of the rst quadrant
satises (x, y) = 0.
937
20.6 Applications
22. (a) From E-5, f(z) = e
z
maps the horizontal strip 0 < y < onto the upper half-plane and f(
0
) = f(0) = 1.
Therefore G(z) = kLn(e
z
1) is a complex potential. To construct a sink at
0
= 0, we must have k < 0.
(b) (x, y) = Im(kLn(e
z
1)) = kArg(e
z
1) = k tan
1
_
e
x
sin y
e
x
cos y 1
_
where tan
1
is chosen to be between 0 and . If we set k = 1, then the
streamlines (x, y) = c, < c < 0, satisfy e
x
[Bcos y sin y] = B where
B = tan c, and so
x = log
e
_
B
Bcos y sin y
_
.
Note that B will vary over all real values. The streamlines are also the images of rays through w = 1 under
the inverse transformation z = Ln w. See the gure.
23. = Im(G(z)) = kArg(z 1) kArg(z + 1) = kArg
_
z 1
z + 1
_
.
(See (1) and (2) in this section in the text). In rectangular coordinates
(x, y) = k tan
1
_
2y
x
2
+y
2
1
_
where tan
1
is chosen to be between 0 and . Level curves (x, y) = c can be put in the form
x
2
+y
2
2By = 1 or x
2
+ (y B)
2
= 1 +B
2
.
Each member of the family passes through (1, 0) and (1, 0).
24. (a) V =
a +ib
z
=
_
ax by
x
2
+y
2
,
bx +ay
x
2
+y
2
_
and since (x
(t), y
.
3. The wedge 0 Arg w 2/3. See gure 20.6 in the text.
4. f
(z
1
+ Ln z
1
+ 1) to map this half-plane onto the target region R
. The composite
transformation is
w =
1
[z
2
+ Ln(z
2
) + 1]
and the image of AB is the ray extending to the left from w = i along the line v = 1.
13. From H-4, z
1
= cos z maps R onto the upper half-plane y
1
0, and AB onto the real segment (, 1].
Using C-4, w =
i z
1
1 +z
1
maps this upper half-plane onto the target region R
Arg
_
w 1
w + 1
_
and so
u =
1
Arg
_
z
4
1
z
4
+ 1
_
is the solution to the original Dirichlet problem in R.
939
CHAPTER 20 REVIEW EXERCISES
15. The inversion w = 1/z maps R onto the horizontal strip 1 v 1/2 and the
transferred boundary conditions are shown in the gure. The solution in the strip
is U = 2v + 2 and so
u = U
_
1
z
_
= 2Im
_
1
z
_
+ 2 =
2y
x
2
+y
2
+ 2
is the solution to the original Dirichlet problem in R.
16. Using the cross ratio formula,
(w i)(i + 1)
(w + 1)(2i)
=
(z 1) 1
1 (2)
we see that w =
z i
z +i
maps 1, 1, to i, i, 1.
17. (a) We note that as u
1
+,
1
0,
2
2, and
3
0.
(b) Since
f
(z) = A(z + 1)
(1/)1
z
(2/)1
(z 1)
(3/)1
,
this suggests that we examine
f
(z) = A(z + 1)
1
z(z 1)
1
= A
z
z
2
1
.
We may write
f
(z) =
A
2
_
1
z + 1
+
1
z 1
_
and so
f(z) =
1
2
A[Ln(z + 1) + Ln(z 1)] +B.
(c) For t real, we can write
f(t) =
1
2
A[log
e
|t + 1| + log
e
|t 1| +iArg(t + 1) +iArg(t 1)] +B.
Since f(0) =
i
2
,
i
2
=
i
2
A+B. For t real,
f(t) =
_
1
2
A[ln |t
2
1| + 2i] +B, t < 1
1
2
Aln |t
2
1| +B, t > 1
.
If A is real and we require that Im(f(t)) = 0 for t < 1, then 0 = Ai + Im(B). If Im(f(t)) = for t > 1,
then = Im(B). All three equations are satised by letting B = i and A = 1. Therefore
f(z) = i
1
2
[Ln(z + 1) + Ln(z 1)].
18. (a) From Theorem 20.5,
u(x, y) =
y
sin t
(x t)
2
+y
2
dt =
y
sin(x s)
s
2
+y
2
] ds (letting s = x t).
But sin(x s) = sin xcos s cos xsin s. We now proceed as in the solution to Problem 6, Section 20.5 to
show that u(x, y) = e
y
sin x.
(b) For u(e
i
) = sin , the Fourier Series solution (6) in Section 20.5 reduces to u(r, ) = r sin .
940
CHAPTER 20 REVIEW EXERCISES
19. If f(w) = w + e
w
+ 1, G(z) = f
1
(z) maps R onto the strip 0 v and the
transferred boundary conditions are shown in the gure to the right. The solution
for the strip is U = v/ and so the solution in R is
(x, y) = U(G(z)) =
1
Im(G(z)) =
1
(x, y).
Therefore the equipotential lines (x, y) = c are the streamlines (x, y) = c of the ow in Figure 20.72.
20. G(re
i
) = r
1/2
sin(/2) + i[r
1/2
cos(/2) 1] and so (r, ) =
r cos(/2) = 1. If
(r, ) = 0, r cos
2
(/2) = 1 or r cos + r = 2 (using cos
2
(/2) = (1 + cos )/2.) In
rectangular coordinates, x +
_
x
2
+y
2
= 2 or y
2
= 4 4x. Therefore the boundary
of R is a streamline. To sketch the streamlines, note that (r, ) = c implies that
r = (c + 1)
2
sec
2
(/2). See the gure to the right.
941