Escolar Documentos
Profissional Documentos
Cultura Documentos
APPENDIX
In Sections 3.1, 3.4, and 6.1, we introduced the curl, divergence, and gradient, respec-
tively, and derived the expressions for them in the Cartesian coordinate system. In this
appendix, we shall derive the corresponding expressions in the cylindrical and spheri-
cal coordinate systems. Considering first the cylindrical coordinate system, we recall
from Appendix A that the infinitesimal box defined by the three orthogonal surfaces
intersecting at point P(r, u, f) and the three orthogonal surfaces intersecting at point
Q(r + dr, f + df, z + dz) is as shown in Figure B.1.
dz e
b
r df
g
a
P(r, f, z)
dr (r ! dr) df
FIGURE B.1
Infinitesimal box formed by incrementing the coordinates in the
cylindrical coordinate system.
420
Z02_RAO3333_1_SE_APPB.QXD 4/9/08 2:42 PM Page 421
Appendix B 421
From the basic definition of the curl of a vector introduced in Section 3.3 and
given by
AC A dl
#
§ : A = Lim c d an (B.1)
¢S : 0 ¢S max
Aabcda A dl
#
(§ : A)r = Lim
df : 0 area abcd
dz:0
Aadefa A dl
#
(§ : A)f = Lim
dz : 0 area adef
dr : 0
Aafgba A dl
#
(§ : A)z = Lim
dr : 0 area afgb
df : 0
Combining (B.2a), (B.2b), and (B.2c), we obtain the expression for the curl of a vector
in cylindrical coordinates as
To find the expression for the divergence, we make use of the basic definition of
the divergence of a vector, introduced in Section 3.6 and given by
AS A dS
# #
§ A = Lim (B.4)
¢v : 0 ¢v
Evaluating the right side of (B.4) for the box of Figure B.1, we obtain
To obtain the expression for the gradient of a scalar, we recall from Appendix A
that in cylindrical coordinates,
dl = dr ar + r df af + dz az (B.6)
and hence
0£ 0£ 0£
d£ = dr + df + dz
0r 0f 0z
= a
0£
0r
ar +
1 0£
r 0f
af +
0£ #
a b (dr ar + r df af + dz az)
0z z
= §£ dl # (B.7)
Z02_RAO3333_1_SE_APPB.QXD 4/9/08 2:42 PM Page 423
Appendix B 423
Thus,
0£ 1 0£ 0£
§£ = a + a + a (B.8)
0r r r 0f f 0z z
Turning now to the spherical coordinate system, we recall from Appendix A that
the infinitesimal box defined by the three orthogonal surfaces intersecting at P(r, u, f)
and the three orthogonal surfaces intersecting at Q(r + dr, u + du, f + df) is as
shown in Figure B.2. From the basic definition of the curl of a vector given by (B.1), we
then find the components of § : A as follows with the aid of Figure B.2:
Aabcda A dl
#
(§ : A)r = Lim
du : 0 area abcd
df : 0
d (r ! dr) sin u df
f
r sin u df
h
dr Q(r ! dr, u ! du, f ! df)
a
P(r, u, f) c
(r ! dr) du
r du r sin (u ! du) df
g
b
FIGURE B.2
Infinitesimal box formed by incrementing the coordinates in the spherical
coordinate system.
Z02_RAO3333_1_SE_APPB.QXD 4/9/08 2:42 PM Page 424
Aadefa A dl
#
(§ : A)u = Lim
df : 0 area adef
dr : 0
[Af](r, u)r sin u df + [Ar](u, f + df) dr
e f
- [Af](r + dr, u)(r + dr) sin u df - [Ar](u, f) dr
= Lim
df : 0 r sin u dr df
dr : 0
[Ar](u, f + df) - [Ar](u, f)
= Lim
df : 0 r sin u df
[rAf](r, u) - [rAf](r + dr, u)
+ Lim
dr : 0 r dr
1 0Ar 1 0
= - (rAf) (B.9b)
r sin u 0f r 0r
Aafgba A dl
#
(§ : A)f = Lim
dr : 0 area afgb
du : 0
Combining (B.9a), (B.9b), and (B.9c), we obtain the expression for the curl of a vector
in spherical coordinates as
1 0 0Au
§ : A = c (Af sin u) - da
r sin u 0u 0f r
1 1 0Ar 0 1 0 0Ar
+ c - (rAf) dau + c (rAu) - da
r sin u 0f 0r r 0r 0u f
ar au af
r2 sin u r sin u r
5 0 0 0 5
= (B.10)
0r 0u 0f
Ar rAu r sin uAf
Z02_RAO3333_1_SE_APPB.QXD 4/9/08 2:42 PM Page 425
Appendix B 425
To find the expression for the divergence, we make use of the basic definition of
the divergence of a vector given by (B.4) and by evaluating its right side for the box of
Figure B.2, we obtain
#
§ A = Lim
dr : 0
+ [Af]f + dfr dr du - [Af]fr dr du
r2 sin u dr du df
du : 0
df : 0
1 0 2 1 0 1 0Af
= 2 0r
(r Ar) + (Au sin u) + (B.11)
r r sin u 0u r sin u 0f
To obtain the expression for the gradient of a scalar, we recall from Appendix A
that in spherical coordinates,
dl = dr ar + r du au + r sin u df af (B.12)
and hence
0£ 0£ 0£
d£ = dr + du + df
0r 0u 0f
= a
0£
0r
ar +
1 0£
r 0u
au +
1 0£ #
a b (dr ar + r du au + r sin u df af)
r sin u 0f f
= §£ dl# (B.13)
Thus,
0£ 1 0£ 1 0£
§£ = a + a + a (B.14)
0r r r 0u u r sin u 0f f
REVIEW QUESTIONS
B.1. Briefly discuss the basic definition of the curl of a vector.
B.2. Justify the application of the basic definition of the curl of a vector to determine sepa-
rately the individual components of the curl.
B.3. How would you generalize the interpretations for the components of the curl of a vec-
tor in terms of the lateral derivatives involving the components of the vector to hold in
cylindrical and spherical coordinate systems?
Z02_RAO3333_1_SE_APPB.QXD 4/9/08 2:42 PM Page 426
PROBLEMS
B.1. Find the curl and the divergence for each of the following vectors in cylindrical coordi-
1 1
nates: (a) r cos f ar - r sin f af; (b) ar; (c) af.
r r
B.2. Find the gradient for each of the following scalar functions in cylindrical coordinates:
1
(a) cos f; (b) r sin f.
r
B.3. Find the expansion for the Laplacian, that is, the divergence of the gradient, of a scalar
in cylindrical coordinates.
B.4. Find the curl and the divergence for each of the following vectors in spherical coordi-
e- r 1
nates: (a) r2 ar + r sin u au; (b) a ; (c) ar.
r u r2
B.5. Find the gradient for each of the following scalar functions in spherical coordinates:
sin u
(a) ; (b) r cos u.
r
B.6. Find the expansion for the Laplacian, that is, the divergence of the gradient, of a scalar
in spherical coordinates.