Contents

1 Review of Differentiation and Integration with Applications 1

1.1 Limits and Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2 Tangents and Normals . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.3 Series and Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.3.1 Power Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

1.3.2 Taylor’s Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2 Derivatives of Real-Valued Functions of Several Variables 17

2.1 Limits and Continuity of functions of several

variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

2.2 Partial Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

2.2.1 Higher Partial Derivatives . . . . . . . . . . . . . . . . . . . . . 22

2.2.2 Harmonic functions . . . . . . . . . . . . . . . . . . . . . . . . 23

2.3 Derivatives of Composite Functions . . . . . . . . . . . . . . . . . . . 24

3 Applications of Partial Derivatives 33

i
 3.1 Differentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3.2 Extreme Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

3.2.1 Necessary Conditions For Extreme Values . . . . . . . . . . . 36

3.2.2 Sufficient Conditions for Extreme Values . . . . . . . . . . . . 36

3.3 Second Derivative Test for Critical Points . . . . . . . . . . . . . . . . 37

3.4 QUADRATIC FORMS . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

4 Line and Multiple Integrals 49

4.1 Vector Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

4.2 LINE INTEGRALS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

4.3 SURFACE INTEGRAL . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

4.4 VOLUME INTEGRAL . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

References 61

ii
Chapter 1

Review of Differentiation and

Integration with Applications

1.1 Limits and Continuity

A function f is said to be continuous at the point x = a if

1. f ( a) exists

2. If the lim f ( x ) exists

x→a

3. lim f ( x ) = f ( a)
x→a

Example 1.1.1. Investigate the continuity of a function defined by


x2 −4x +3
, x 6= 3


x 2 −9
f (x) =
1


3 , x=3

Solution

1
1. f (3) = 3 and is defined.

1
 x2 −4x +3 ( x −1)( x −3) x −1 1
2. lim f ( x ) = lim 2 = lim = lim =
x →3 x →3 x −9 x →3 ( x −3)( x +3) x →3 x +3 3

3. lim f ( x ) = f (3).
x →3

∴ f ( x ) is continuous at x = 3.

Theorem 1.1.1. If a function f is defined on an interval I has an extremum at an interior

point c and f is differentiable at c then f 0 (c) = 0

Theorem 1.1.2 (Intermediate Value Theorem). Suppose

1. f is continuous on [a, b]

2. f ( a) 6= f (b)

Then for any number y1 lying between f ( a) and f (b) there exists at least one x1 ε( a, b) 3
f ( x1 ) = y1 . (i.e f takes on every intermediate value between f ( a) and f (b).)

Theorem 1.1.3 (Rolle’s Theorem). Let the function f be continuous on a closed interval
[a, b], differentiable in the open interval ( a, b) and f ( a) = f (b). Then there exists at least
one point c in ( a, b) 3 f 0 (c) = 0.

Proof: Since f is continuous on [a, b], it must have a maximum and minimum point
on [a, b]. If f is not constant on [a, b], then the extremum is an interior point c where
f 0 (c) = 0.

N:B If the extremum values were end points and since f ( a) = f (b), it means that f
is constant which is a contradiction.

Example 1.1.2. Determine if each of the following satisfies the conditions of Rolle’s
theorem, where it does satisfy, find all the possible extremum value. If not, give the
condition that is not satisfied.

2
 
 x , 0<x<1

1. f ( x ) =
 0 , x=1


2. f ( x ) = x2 + 2x, −2 ≤ x ≤ 0

3. f ( x ) = x2 , 1≤x≤2

Solution

1. f is differentiable on (0,1)
f (1) = 0, f (0) = 0
∴ f (0) = f (1).
For continuity,

f (1) 6= lim f ( x )
x →1

.∴ f is not continuous on [0,1].

2. f ( x ) = x2 + 2x, −2 ≤ x ≤ 0
f is differentiable on (-2,0)
f (−2) = 0, f (0) = 0
∴ f (−2) = f (0).
f is not continuous on [-2, 0].
∃ a c ∈ [−2, 0] such that f 0 (c) = 0
f 0 ( x ) = 2x + 2
f 0 (c) = 2c + 2 = 0
c = −1.
extreme value ⇒ f (−1) = (−1)2 + 2(−1) = −1.
Hence f ( x ) has a maximum value -1 at x = −1.

Mean Value Theorem

3
Let f be continuous on [a,b] and differentiable in (a,b). Then there exists at least a
point c ∈ ( a, b) such that
f (b) − f ( a)
= f 0 (c)
b−a

Proof: Construct a function F ( x ) which is related to f ( x ) and satisfies the condition

of Rolle’s theorem. One such function is given by

[ f (b) − f ( a)]
F(x) = f (x) − ( x − a ).
b−a

Observe that F ( a) = F (b) = f ( a). Since f is continuous on [a,b] and differentiable in

(a,b), so is F. Therefore by the Rolle’s theorem, there exists at least a point c ∈ ( a, b)
such that F 0 (c) = 0.
f (b) − f ( a)
F0 (x) = f 0 (x) − = 0.
b−a
f (b) − f ( a)
∴ F 0 (c) = f 0 (c) − = 0.
b−a
f (b) − f ( a)
f 0 (c) =
b−a

Example 1.1.3. Determine if each of the following functions satisfies the condition
of the mean value theorem. Where it does, find all the possible points, if not state
the condition that is not satisfied.

1. f ( x ) = 2 − 3x2 , 0≤x≤1

2. f ( x ) = x3 − 3x2 − x + 3, −2 ≤ x ≤ 3

2
3. f ( x ) = 4x 3 , −1 ≤ x ≤ 27

Solution

4
 1. f is continuous on [0,1], since f is a polynomial function.
f is differentiable on (0,1), there exists,

f (1) − f (0)] −1 − 2
f 0 (c) = = = −3
1−0 1

and f 0 (c) = −6c

1
then −6c = −3 ⇒c= 2

1.2 Tangents and Normals

Recall the equation of a straight line

y = mx + c

If this line passes through the point (a,b), then

b−c b−c
 
b = ma + c ⇒m= ⇒ y−c = x
a a

The tangent of a curve is determined by the derivative at a point.

m= gradient of tangent
−1
m = gradient of normal.

Example 1.2.1. Determine the equation of the tangent and normal to the curve at
the given point

1. x2 y3 − x3 y2 = 12, at (−1, 2)
√
2. 2x + y − 2 sin xy = π
2, at ( π4 , 1)

3. x sin xy − y2 = x2 − 1, at (1, 1)

5
Solution

1. For tangent,
d 2 3 d
( x y − x 3 y2 ) = (12) = 0
dx dx
d 2 3 d
( x y ) − ( x 3 y2 ) = 0
dx dx
 
3 2 2 dy 2 2 3 dy
2xy + 3x y − 3x y + 2x y =0
dx dx
dy dy
2xy3 + 3x2 y2 − 3x2 y2 − 2x3 y =0
dx dx
dy dy
3x2 y2 − 2x3 y = 3x2 y2 − 2xy3
dx dx
dy
(3x2 y2 − 2x3 y) = 3x2 y2 − 2xy3
dx
dy 3x2 y2 − 2xy3
= 2 2
dx 3x y − 2x3 y

dy 3(−1)2 (2)2 − 2(−1)(2)3 3 · 4 + 16 28

= 2 2 3
= =
dx 3(−1) (2) − 2(−1) (2) 12 + 4 16
dy 7
dx |(−1,2) = 4
y −2 7
Equation for tangent := x +1 = 4

4( y − 2) = 7( x + 1)
4y − 8 = 7x + 7
7x − 4y + 15 = 0.

−4
For normal, gradient is 7
y −2 −4
x +1 = 7

7( y − 2) = −4( x + 1)
7y − 14 = −4x − 4
7y + 4x − 10 = 0 is the equation of the normal.

6
2. For tangent,
d √ d π
(2x + y − 2 sin xy) = ( )=0
dx dx 2
d √ d
(2x + y) − 2 (sin xy) = 0
dx dx
dy √
 
dy
2+ − 2 cos( xy) · ( x + y) = 0
dx dx

dy √
 
dy
2+ − 2 x cos( xy) + y cos( xy) = 0
dx dx
dy √ dy √
2+ − x 2 cos( xy) − y 2 cos( xy) = 0
dx dx
dy √ dy √
− 2x cos( xy) = 2y cos( xy) − 2
dx dx
dy √ √
(1 − 2x cos( xy)) = 2y cos( xy) − 2
dx
√
dy 2y cos( xy) − 2
= √
dx (1 − 2x cos( xy))
√
dy 2 cos( π4 ) − 2
| π = √
dx ( 4 ,1) (1 − 2 · π4 cos( π4 )
√ 1
2( √ ) − 2
= √ 2 1 π
1− 2· √ · 4
2
1−2
=
1 − π4
−4
=
4−π
Equation of tangent is
y −1 −4
x − π4
= 4− π

(4 − π ) = −4( x − π4 )
4y − 4 − πy + π = −4x + π
4y + 4x − 4 + πy = 0
y(4 + π ) + 4x − 4 = 0.
4− π
For the normal, gradient = 4

7
 y −1 4− π
x − π4
= 4

4(y − 1) = (4 − π )( x − π4 )
π2
4y − 4 = 4x − π − πx + 4
π2
4y − 4 − 4x + π + πx − 4 =0
π2
4y + x (π − 4) + π − 4 − 4 = 0.

1.3 Series and Sequences

A succession of terms such as ao , a1 , a2 , . . . , an is called a sequence. A summation of

the terms of a sequence is called a series. e.g

n
a o + a1 + a2 + · · · + a n = ∑ ai .
i =0

The sum of the 1st n−terms of a series can be denoted by Sn .

Convergence of a Series

If there exists a number S, such that

lim Sn = S,
n→∞

we say that the series converges, otherwise,it diverges.

Example 1.3.1. Find the sum to infinity S, of the series

1
1. 1·2 + 21·3 + 31·4 + · · ·

1
2. 2·3·4 + 3·42·5 + 4·35·6 + · · ·

2
3. 1·3·5 + 3·54·7 + 5·67·9 + · · ·

8
Solution
First find Sn i.e sum of the first n−terms then find the lim Sn .
n→∞

1
1. 1·2 + 21·3 + 31·4 + · · · + 1
n ( n +1)

∞ ∞  
1 1 1
= ∑ = ∑ −
n =1
n ( n + 1) n =1
n n+1

       
1 1 1 1 1 1 1
Sn = 1− + − + − +···+ −
2 2 3 3 4 n n+1
1
Sn = 1 −
n+1
 
1 1
S = lim 1 − = 1− = 1−0
n→∞ n+1 ∞

S = 1.

Definition 1.3.1. A series is convergent with a sum to infinity S if given e > 0, ∃ a

finite number N such that |Sn − S| < e ∀ n ≥ N.
A series therefore converges or diverges to −∞ or +∞ or it oscillates finitely. For
example,

1.
∞
∑ n = 1 + 2 + 3 + 4 + · · · diverges to + ∞
n =1

2.
∞
∑ (−n) = −1 − 2 − 3 − · · · diverges to − ∞
n =1

3.
∞
∑ (−1)n+1 = 1 − 1 + 1 − 1 + 1 − 1 + · · · oscillates finitely.
n =1

4.
∞
∑ (−2)n = 1 − 2 + 4 − 8 + 16 − · · · oscillates infinitely.
n =0

9
Example 1.3.2. Find the sum to infinity S of the series

1 2 3
+ + +···
2·3·4 3·4·5 4·5·6

1
Hence determine the smallest value of n for which |Sn − S| < 100 .

Solution
n
The general term is given by (n+1)(n+2)(n+3)
. Resolving this into partial fractions,
one obtains

n 1 2 3
=− + − .
(n + 1)(n + 2)(n + 3) 2( n + 1) n + 2 2( n + 3)

∞
1 2 n 1 2 3
∴ + +···+ = ∑− + − .
2.3.4 3.4.5 (n + 1)(n + 2)(n + 3) n=1 2(n + 1) n + 2 2(n + 3)

The sum can then be expressed as

1 2 3 n−2 n−1 n
+ + +···+ + +
2.3.4 3.4.5 4.5.6 (n − 1)n(n + 1) n(n + 1)(n + 2) (n + 1)(n + 2)(n + 3)
       
1 2 3 1 2 3 1 2 3 1 2 3
= − + − + − + − + − + − +···+ − + −
4 3 8 6 4 10 8 5 12 2( n − 1) n 2( n + 1)
  
1 2 3 1 2 3
+ − + − − + − .
2n n + 2 2(n + 2) 2( n + 1) n + 2 2( n + 3)
1 2 n 1 2 1 3 2 3
∴ Sn = + +···+ =− + − − + −
2.3.4 3.4.5 (n + 1)(n + 2)(n + 3) 4 3 6 2( n + 2) n + 2 2( n + 3)
1 2n + 3
= − .
4 2(n + 2)(n + 3)

Hence we obtain the sum to infinity

 
1 2n + 3 1
S = lim Sn = lim − = .
n→∞ n→∞ 4 2(n + 2)(n + 3) 4

1 1 2n + 3 1 1
∴ | Sn − S | < ⇒ − − <
100 4 2(n + 2)(n + 3) 4 100

10
 2n + 3 1
⇒ < ⇒ 100(2n + 3) < 2(n + 2)(n + 3)
2(n + 2)(n + 3) 100

. Thus we obtain the quadratic inequality

n2 − 95n − 144 > 0

. Now consider the equation

n2 − 95n − 144 = 0

which has the solutions

√
95 ± 9025 + 576 95 ± 97.98
n= =
2 2

, that is n = −1.49 or n = 96.49. This gives n > 96.49, hence the smallest value of n
for which |Sn − S| < 10−2 is 97, since n takes positive integral values only.

Test for Convergence

1. Ratio test

2. nth term test

3. Comparison test

4. Root test

a n +1
Ratio test: If the lim an = l then
n→∞
if l < 1 the series converges
if l > 1 the series diverges
if l = 1, the test fails.

11
1.3.1 Power Series

A series of the form

∞
∑ a n ( x − c ) n = a o + a1 ( x − c ) + a2 ( x − c )2 + · · · a n ( x − c ) n
n =0

is called a power series in powers of ( x − c) or a power series about x = c. The point

x = c is called the center of convergence of the power series.
Examples of Power series
Geometric series:
1
1 + x + x2 + x3 + · · · =
1−x

| x | < 1 with center x = 0.

∞
1
∑ n2 n
( x − 5) n ,
n =1

with center x = 5.

Theorem 1.3.1. For any power series ∑∞ n

n=0 an ( x − c ) one of the following alterna-

tives must hold: (I) The sum may converge only at x = c or

(II) The sum may converge at every real number x, or
(III) There may exist a positive real number R 3 the series converges at every x
satisfying | x − c| < R and diverges at every x satisfying | x − c| > R. In this case the
series may not converge at any of the end points: x = c − R and x = c + R.

In each of these cases, the convergence is absolute, except possibly at the end points
x = c − R and x = c + R in case ( I I I ). By this theorem the set of values of x for
which the power series ∑∞ n
n=0 an ( x − c ) converges is an interval centered at x = c

and we call this interval the the interval of convergence of the power series. The
number R is called the radius of convergence of the power series.

Example 1.3.3. Determine the radius and interval of convergence of the series

12
 (2x +5)n
1. ∑∞
n =0 ( n2 +1)3n

( x + 52 )n
2. ∑∞ 2 n
n =0 ( 3 ) n2 +1

Solution

1. Use the ratio test

(2x + 5)n (2x + 5)n+1

an = a n +1 =
( n 2 + 1 )3n ((n + 1)2 + 1)3n+1

a n +1 (2x + 5)n+1 ( n 2 + 1 )3n

lim = lim ×
n→∞ an n→∞ (( n + 1)2 + 1)3n+1 (2x + 5)n
2x + 5 n2 + 1
= lim ×
n→∞ (( n + 1)2 + 1)3 1
(2x + 5)(n2 + 1)
= lim
n→∞ ( n2 + 2n + 2)3

n2 + 1
 
1
= lim |2x + 5| 2
3 n→∞ n + 2n + 2
1 + n12
" #
1
= lim |2x + 5|
3 n→∞ 1 + n2 + n22
" #
1
1 1+ ∞
= |2x + 5| 2
3 1+ ∞ + ∞2
1
= [|2x + 5| · 1]
3
|2x + 5|
= .
3
The above series will converge if
|2x +5|
3 <1 ⇒ |2x + 5| < 3 ⇒ | x + 52 | < 23 .
The radius of convergence is 32 .
−3 < 2x + 5 < 3
−8 < 2x < −2
−4 < x < −1 is the interval of convergence.

13
1.3.2 Taylor’s Series

Let f ( x ) be any function and c be a constant, then if f (c) exists and f ( x ) has deriva-
tives of all orders at x = c,

0 ( x − c)2 00
f ( x ) = f (c) + ( x − c) f (c) + f (c) + · · ·
2!

This is called the Taylor’s series expansion of f ( x ) about the point x = c.

Proof: Suppose x = c + h, then

h2 00
f (c + h) = f (c) + h f 0 (c) + f (c) + · · ·
2!

when c = 0
x2 00 x3 000
f ( x ) = f (0) + x f 0 (0) + f (0) + f (0) + · · ·
2! 3!

This is called Maclaurin’s series.

Example 1.3.4. Find the Taylor’s series expansion of the function f ( x ) = sin x about
the point x = 0.

Solution
f ( x ) = sin x, f (0) = 0
f 0 ( x ) = cos x, f 0 (0) = 1
f 00 ( x ) = − sin x, f 00 (0) = 0
f 000 ( x ) = − cos x, f 000 (0) = −1
f (iv) ( x ) = sin x, f f (iv) (0) = 0
f (v) ( x ) = cos x, f f ( v ) (0) = 1

14
∴ The Taylor’s series becomes

x2 00 x3 000 x4 (iv) x5 (v)

f ( x ) = sin x = f (0) + x f 0 (0) +f (0) + f (0) + f (0) + f (0) + · · ·
2! 3! 4! 5!
x2 x3 x4 x5
= 0+x·1+ ·0+ · (−1) + ·0+ ·1+···
2! 3! 4! 5!
x3 x5
= x− + −···
3! 5!
∞
(−1)n x2n+1
= ∑
n=0 (2n + 1) !

Example 1.3.5. Use the Taylor series expansion above to estimate the value of sin 31◦ .

Solution
sin 31◦ = sin(30◦ + 1◦ ) = sin( π6 + 0.01745)
f (c + h) = sin(c + h) = sin( π6 + 0.01745)
π
where h = 0.01745 and c = 6

π π (0.01745)2 π
sin(30◦ + 1◦ ) = sin + 0.01745 cos + (− sin ) + · · ·
6 6 2! 6
(0.01745)2
= 0.5 + 0.01745 · 0.8660 + (−0.5)
2!
= 0.51504

If

∞
f (x) = ∑ cn ( x − a)n
n =0

≡ c o + c1 ( x − a ) + c2 ( x − a )2 + · · ·

Let x = a then f ( a) = co
f 0 ( x ) = c1 + 2c2 ( x − a) + 3c3 ( x − a)2 + · · ·
f 0 ( a ) = c1
f 00 ( x ) = 2c2 + 6c3 ( x − a) + · · ·

15
f 00 ( a) = 2c2 = 2!c2
f 000 ( a) = 6c3 = 3!c3

( x − a)2 00 ( x − a)3 000

∴ f ( x ) = f ( a) + ( x − a) f 0 ( a) + f ( a) + f ( a) + · · ·
2! 3!

Exercises
Obtain the Maclaurin’s series expansion of the following
( a)e x (b) cos x (c) ln(1 + x ).

16
Chapter 2

Derivatives of Real-Valued Functions

of Several Variables

Definition 2.0.1. A real valued function of n variables is a rule which assigns to a

point ( x1 , x2 , . . . , xn ) in some subset D ( f ) of Rn a unique real number f ( x1 , x2 , . . . , xn ).
D ( f ) is called the domain of f while the set of real numbers f ( x1 , x2 , . . . , xn ) is called
the range of f .

Example 2.0.1. Determine the domain of the function

( a) f ( x, y) = x + y (b) f ( x, y) = √ x−y (c) f ( x, y) = √ 1
9− x − y x 2 − y2
√
(d) f ( x, y) = xy

Solution
( a) f is defined on R2 , hence the domain of f is R2 .
(b) f is defined on R2 − {( x, y)/x + y ≥ 9}
(c) D ( f ) = R2 − {( x, y) : |y| ≥ | x |}
(d) D ( f ) = R2 − {( x, y) : xy < 0}

Definition 2.0.2. The function f ( x1 , x2 , . . . , xn ) = c where c is a constant defines the

level curves of f for various values of c (equally spaced)

17
Example 2.0.2. : Obtain the level curves for the following functions:

x y

1. f ( x, y) = 3 1 − 3 − 4 , 0 ≤ x ≤ 2, 0 ≤ y ≤ 4 − 2x
Solution
x y

Let f ( x, y) = 3 1 − 3 − 4 =c
x y c
1− 3 − 4 = 3
y
− 3x − 4 = c
3 −1
x y
3 + 4 = 1 − 3c , 0≤c≤3
The equation above represents segments of the straight lines which lie in the
1st quadrant of the circle.
For c = 0,
x y y x
3 + 4 =1 ⇒ 4 = 1− 3

∴ y = 4(1 − 3x )
For c = 1,
x y 2 4(2− x )
3 + 4 = 3 ⇒y= 3

For c = 2,
x y 1 4(1− x )
3 + 4 = 3 ⇒y= 3

For c = 3,
x y −4x
3 + 4 =0 ⇒y= 3

p
2. f ( x, y) = 9 − x 2 − y2
Solution
Level curves are defined by f ( x, y) = c
p
f ( x, y) = c = 9 − x2 − y2 where c is a constant.
c2 = 9 − x 2 − y2
x 2 + y2 = 9 − c2
√
It is a circle centred at the origin with radius 9 − c2 . In this case, the level
curves are concentric circles centred at (0,0).

18
 3. f ( x, y) = x2 − y2
Solution
Level curves are defined by f ( x, y) = c = x2 − y2 where c is a constant. These
are rectangular hyperbolas
when c = 0 ⇒ x 2 − y2 = 0 ⇒ x 2 = y2
∴ x = ±y
The lines x = y and x = −y are asymptotes.

2.1 Limits and Continuity of functions of several

variables

Just as in the case of single variable functions, we say that f ( x, y) approaches the
limit L as the point ( x, y) approaches the point ( a, b), and we write

lim f ( x, y) = L
( x,y)→( a,b)

we can also write it as

 
lim f ( x, y) = L or lim lim f ( x, y) =L
x → ay→b x→a y→b

If for all points of a neighbourhood ( a, b) except possibly at the point ( a, b) itself.

More formally, we say that lim f ( x, y) = L, if given e > 0, ∃ a δ(e) > 0 3
( x,y)→( a,b)
p
| f ( x, y) − L| < e whenever 0 > ( x − a)2 + (y − b)2 < δ. The properties of
limits in the single variable function apply in this case.

Example 2.1.1. Evaluate the following limits:

1. lim (2x − y2 )
( x,y)→(2,3)

19
 2xy
2. lim x 2 + y2
( x,y)→(0,0)
 
x
3. limπ y sin y
( x,y)→( 3 ,2)

Solution
 
1. lim (2x − y2 ) = lim lim (2x − y2 ) = lim 2x − 9 = 2(2) − 9 = −5
( x,y)→(2,3) x →2 y →3 x →2
 
2xy 2xy
2. lim x 2 + y2 = lim lim x2 +y2
( x,y)→(0,0) x →0 y →0
Lets consider the limit along the line y = kx

2x (kx ) 2kx2 2k 2k
∴ lim = lim = lim = .
( x,kx )→(0,0) x2 + (kx )2 ( x,kx )→(0,0) x2 (1 + k2 ) ( x,kx )→(0,0) 1 + k2 1 + k2

Since the limit depends on k, there are infinitely many of them for each value
of k, hence the limit is not unique. Therefore it does not exist.
    
limπ y sin y = limπ lim y sin yx
x
= limπ 2 sin 2x = 2 sin

π


3. 6 =1
( x,y)→( 3 ,2) x→ 3 y →2 x→ 3

2x2 y
Investigate the limiting behaviour of the function f ( x, y) = x 4 + y2
as ( x, y) → (0, 0)
Consider the limit along the curve y = kx2

2x2 y 2x2 y 2kx4

 
2k 2k
lim = lim lim 4 = lim = lim = .
( x,kx2 )→(0,0) x 4 + y2 x →0 kx2 →0 x + y2 x →0 x 4 (1 + k 2 ) x →0 1 + k 2 1 + k2

Since the limit depends on k and k has many values , hence the limit is not unique.
Therefore it does not exist.

Determine Investigate the limiting behaviour of the function

x 2 −2
1. f ( x, y) = 3+ xy as ( x, y) → (0, 0)
cos( xy)
2. f ( x, y) = 1− x cos y as ( x, y) → (1, π )

The continuity of a function of several variables is defined in a similar fashion as in

the case of single variable i.e, f ( x, y) is continuous at a point ( a, b) if

20
 1. f ( a, b) is defined.

2. lim f ( x, y) exists.
( x,y)→( a,b)

3. lim f ( x, y) = f ( a, b).
( x,y)→( a,b)

2.2 Partial Derivatives

Given y = f ( x ), then

dy f ( x + δx ) − f ( x )
= f 0 ( x ) = lim
dx δx →0 δx

For a function of two variables x and y, we define the first derivative with respect to
x as

f 1 ( x, y) = f x ( x, y)
f ( x + h, y) − f ( x, y) ∂ f ( x, y)
= lim =
h →0 h ∂x

Similarly,

∂ f ( x, y) f ( x, y + k ) − f ( x, y)
f 2 ( x, y) = f y ( x, y) = = lim
∂y k →0 k

Example 2.2.1. 1. If f ( x, y) = x2 sin y. Find

i) f 1 ( x, y) ii) f 2 ( x, y)
Solution
i) f 1 ( x, y) = 2x sin y
ii) f 2 ( x, y) = x2 cos y

2. Find u x and uy if u( x, y) = x3 y2 + x4 y + y4
Solution

21
 u( x, y) = x3 y2 + x4 y + y4
u x = 3x3 y2 + 4x3 y
uy = 2x3 y + x4 + 4y3

xz
3. Find all the first partial derivatives of f ( x, y, z) = y+z at (1, 1, 1).
Solution
z 1
f x ( x, y, z) = y+z ∴ f x | x=1,y=1,z=1 = 2
− xz
f y ( x, y, z) = ( y + z )2
∴ f y | x=1,y=1,z=1 = − 14
xy 1
f z ( x, y, z) = ( y + z )2
∴ f z | x=1,y=1,z=1 = 4

2.2.1 Higher Partial Derivatives

The higher partial derivatives of a function z = f ( x, y) are given by

∂2 z
= f 11 ( x, y) = f xx ( x, y)
∂x2

∂2 z
= f 22 ( x, y) = f yy ( x, y)
∂y2

∂2 z
= f 12 ( x, y) = f xy ( x, y)
∂y∂x

∂2 z
= f 21 ( x, y) = f yx ( x, y)
∂x∂y

If w = f ( x, y, z), then

∂5 w
 
∂ ∂ ∂ ∂ ∂w
2
= = f 32212 ( x, y, z) = f zyyxy ( x, y, z)
∂y∂x∂y ∂z ∂y ∂x ∂y ∂y ∂z

Example 2.2.2. Find the five partial derivatives of the function f ( x, y) = x2 (1 + y2 )

up to second derivatives.

22
Solution
f x ( x, y) = 2x (1 + y2 )
f y ( x, y) = 2x2 y
f xx ( x, y) = 2(1 + y2 )
f yy ( x, y) = 2x2
f xy ( x, y) = 4xy
f yx ( x, y) = 4xy ∴ f xy = f yx .
N:B f xy = f yx , except for discontinuous functions.

Exercise
Find all the second partial derivatives of the following functions;
a) f ( x, y) = x3 y4 b) f ( x, y) = xey − ye x
c) f ( x, y, z) = x3 y3 z3 .

2.2.2 Harmonic functions

Consider the function u( x, y) = ekx cos(ky). We show that for any real number k the
function u( x, y) satisfies the partial differential equation

u xx + uyy = 0

u x = kekx cos(ky)
u xx = k2 ekx cos(ky)
uy = −kekx sin(ky)
uyy = −k2 ekx cos(ky)
∴ u xx + uyy = k2 ekx cos(ky) − k2 ekx cos(ky) = 0

At any point in the xy−plane the result always holds. The equation u xx + uyy = 0 is
called Laplace equation.

23
A function of two variables which has continuous second partial derivatives in the
region of the xy−plane is said to be harmonic if it satisfies Laplace equation.

Harmonic functions play very important role in science and engineering as they are
used to model various physical quantities such as steady-state temperature distri-
butions, fluid flows and electric and magnetic potential fields.

Exercises

1. Show that the function f ( x, y) = ekx sin(ky) is harmonic in R2 .

2. Show that if f and g are twice differentiable functions of one variable, then
∂2 w
w = f ( x − ct) + g( x + ct) satisfies the one-dimensional wave equation ∂t2
=
2
c2 ∂∂xw2 .
Solution
∂w
∂t = −c f 0 ( x − ct) + c ġ( x + ct)
∂2 w
∂t2
= c2 f 00 ( x − ct) + c2 g̈( x + ct)
∂w
∂x = f 0 ( x − ct) + ġ( x + ct)
∂2 w
∂x2
= f 00 ( x − ct) + g̈( x + ct)
∂2 w 2
∂t2
= c2 f 00 ( x − ct) + c2 g̈( x + ct) = c2 ( f 00 ( x − ct) + g̈( x + ct)) = c2 ∂∂xw2 for one
dimension
 
∂2 w ∂2 w ∂2 w
N:B ∂t2
= c2 ∂x2
+ ∂y2
is the wave equation in two dimensions.

2.3 Derivatives of Composite Functions

Given z = f ( x, y) where x = φ(t), y = ψ(t), then

dz ∂z ∂x ∂z ∂y
= · + ·
dt ∂x ∂t ∂y ∂t

24
Now, for u = f ( x, y, z), where x = φ(t), y = ψ(t), z = χ(t)

du ∂u ∂x ∂u ∂y ∂u ∂z
= · + · + ·
dt ∂x ∂t ∂y ∂t ∂z ∂t

Example 2.3.1. If u( x, y, z) = 2x2 − yz + xz, where x = 2 sin t, y = t2 − t + 1, z=

3e−t find du
dt at t = 0.

Solution
du ∂u ∂x ∂u ∂y ∂u ∂z
= · + · + ·
dt ∂x ∂t ∂y ∂t ∂z ∂t
∂x
∂t = 2 cos 2t, ∂y
∂t = 2t, ∂z
∂t = −3e−t
∂u
∂x = 4x + z, ∂u
∂y = −z, ∂u
∂z = −y + x

du
= (4x + z)2 cos t + (−z)(2t − 1) + (−y + x )(−3e−t )
dt

At t = 0

du
= (4x + z)2 cos 0 + (−z)(2.0 − 1) + (−y + x )(−3e−t )
dt
= 2(4x + z) + z + 3y − 3x

= 8x + 2z + z + 3y − 3x

= 8x + 2z + z + 3y − 3x

= 5x + 3z + 3y

= 5(2 sin t) + 3(t2 − t + 1) + 3(3e−t )

= 0+3+9

= 12

25
Given z = ( x, y) where x = φ(u, v) and y = ψ(u, v) then we have

∂z ∂z ∂x ∂z ∂y
= · + ·
∂u ∂x ∂u ∂y ∂u

and
∂z ∂z ∂x ∂z ∂y
= · + · .
∂v ∂x ∂v ∂y ∂v

In matrix form,     
∂z ∂x ∂y ∂z
 ∂u   ∂u ∂u   ∂x 
 =  
∂z ∂x ∂x ∂z
∂v ∂v ∂v ∂y

The 2 × 2-matrix is called the ”Jacobian matrix”, and its determinant is called the
”Jacobian”.

For a function of three independent variables, f = f ( x, y, z) where x = φ(s, t, u), y=

ψ(s, t, u), z = χ(s, t, u).
Then
∂f ∂ f ∂x ∂ f ∂y ∂ f ∂z
= · + · + ·
∂s ∂x ∂s ∂y ∂s ∂z ∂s
∂f ∂ f ∂x ∂ f ∂y ∂ f ∂z
= · + · + ·
∂t ∂x ∂t ∂y ∂t ∂z ∂t
∂f ∂ f ∂x ∂ f ∂y ∂ f ∂z
= · + · + ·
∂u ∂x ∂u ∂y ∂u ∂z ∂u

In matrix form is     
∂f ∂x ∂y ∂z ∂f
 ∂s   ∂s ∂s ∂s   ∂x 
∂f  
∂z   ∂ f 
∂y
 
  =  ∂x
 ∂t   ∂t ∂t ∂t   ∂y 
    
∂f ∂x ∂y ∂z ∂f
∂u ∂u ∂u ∂u ∂z

The 3 × 3-matrix is the Jacobian matrix.

∂x ∂y
∂u ∂u ∂( x, y)
The Jacobian = .
∂x ∂x ∂(u, v)
∂v ∂v

26
 2x +y
Example 2.3.2. Given that f ( x, y) = y−2x , where x = 2u − 3v and y = u + 2v
∂f ∂f ∂2 f
Find (a) ∂u (b) ∂v (c) ∂u2 .

Solution
∂y ∂y
∂x
(a) ∂u = 2, ∂x
∂v = −3, ∂u = 1, ∂v = 2.

∂f ∂ f ∂x ∂ f ∂y
= · + ·
∂u ∂x ∂u ∂y ∂u
∂f ∂f
= ·2+ ·1
∂x ∂y
∂f (y − 2x )2 − (2x + y)(−2)
=
∂x (y − 2x )2
2y − 4x + 4x + 2y
=
(y − 2x )2
∂f 4y
=
∂x (y − 2x )2

∂f (y − 2x )1 − (2x + y)1
=
∂y (y − 2x )2
y − 2x − 2x − y
=
(y − 2x )2
∂f −4x
=
∂y (y − 2x )2
∂f 4y × 2 (−4x )
= 2
+
∂u (y − 2x ) (y − 2x )2
8y − 4x 4(2y − x )
= 2
=
(y − 2x ) (y − 2x )2
∂f 4(2y − x )
=
∂u (y − 2x )2

27
(b)

∂f ∂ f ∂x ∂ f ∂y
= · + ·
∂v ∂x ∂v ∂y ∂v
4y(−3) −4x · 2
= 2
+
(y − 2x ) (y − 2x )2
−12y − 8x
=
(y − 2x )2
∂f −4(3y + 2x )
=
∂v (y − 2x )2

(c)

∂2 f
 
∂ ∂f
=
∂u2 ∂u ∂u
   
∂ ∂ f ∂x ∂ ∂ f ∂y
= +
∂x ∂u ∂u ∂y ∂u ∂u
∂ 4(2y − x ) ∂ 4(2y − x )
 
= 2
·2+ ( )·1
∂x (y − 2x ) ∂y (y − 2x )2
(y − 2x )2 (−4) − 4(2y − x )(−4(y − 2x )) 8(y − 2x )2 − 4(2y − x ) · 2(y − 2x )
 
=2 +
(y − 2x )4 (y − 2x )4
−8(y − 2x )2 − 32(2y − x )(y − 2x ) 8(y − 2x )2 − 8(2y − x )(y − 2x )
= +
(y − 2x )4 (y − 2x )4
−8(y − 2x ) + 32(2y − x ) 8(y − 2x ) − 8(2y − x )
= +
(y − 2x )3 (y − 2x )3
−8y + 16x + 64y − 32x + 8y − 16x − 16y + 8x
=
(y − 2x )3
56y − 16x − 8y − 8x
=
(y − 2x )3
48y − 24x
=
(y − 2x )3
24(2y − x )
=
(y − 2x )3

Example 2.3.3. Given that x = f (u, v) and y = g(u, v), where u = φ(r, s) and
v = ψ(r, s), show that
∂( x, y) ∂( x, y) ∂(u, v)
= · .
∂(r, s) ∂(u, v) ∂(r, s)

28
Here,
∂x ∂x ∂u ∂x ∂v
= · + ·
∂r ∂u ∂r ∂v ∂r
∂x ∂x ∂u ∂x ∂v
= · + ·
∂s ∂u ∂s ∂v ∂s
∂y ∂y ∂u ∂y ∂v
= · + ·
∂r ∂u ∂r ∂v ∂r
∂y ∂y ∂u ∂y ∂v
= · + ·
∂s ∂u ∂s ∂v ∂s

Hence one obtains the Jacobian as

∂( x, y)
∂x
∂r
∂x
∂s
∂x
∂u · ∂u
∂r + ∂x
∂v · ∂v
∂r
∂x
∂u · ∂u
∂s + ∂x
∂v · ∂v
∂s
= =
∂(r, s) ∂y ∂y ∂y
· ∂u
+ ∂y
· ∂v ∂y
· ∂u
+ ∂y
· ∂v
∂r ∂s ∂u ∂r ∂v ∂r ∂u ∂s ∂v ∂s

        
∂x ∂x ∂u ∂x ∂x ∂u ∂x ∂x ∂u ∂u
 ∂u ∂v   ∂r   ∂u ∂v   ∂s   ∂u ∂v   ∂r ∂s 
=   +    =   
∂y ∂y ∂v ∂y ∂y ∂v ∂y ∂y ∂v ∂v
∂u ∂v ∂r ∂u ∂v ∂s ∂u ∂v ∂r ∂s

=
∂x ∂x ∂u ∂u
∂u ∂v ∂r ∂s ∂( x, y) ∂(u, v)
= ·
∂y ∂y ∂v ∂v ∂(u, v) ∂(r, s)
∂u ∂v ∂r ∂s

EXERCISE
 
Given that f ( x, y) = tan−1 x
y where x = 2t + s, y = 3t − s, find
∂f ∂f
(a) ∂t (b) ∂s

Solution
(a)
∂f ∂ f ∂x ∂ f ∂y
= · + ·
∂t ∂x ∂t ∂y ∂t
∂y ∂y
∂x
∂t = 2, ∂x
∂s = 1, ∂t = 3, ∂s = −1
1
∂f y y
= 2 = x 2 + y2
∂x 1 + x2
y

29
 x
∂f
−
y2 −x
= 2 = x 2 + y2
∂y 1 + x2
y

∂f y x
= 2 2
·2− 2 ·3
∂t x +y x + y2
2y 3x
= 2 −
x + y2 x 2 + y2
2y − 3x
= 2 .
x + y2

(b)

∂f ∂ f ∂x ∂ f ∂y
= · + ·
∂s ∂x ∂s ∂y ∂s
y x
= 2 2
·1+ 2
x +y x + y2
x+y
= 2
x + y2

Show that under the transformation x = r cos θ and y = r sin θ, the Laplace equation

∂2 u ∂2 u
+ 2 =0
∂x2 ∂y

becomes
∂2 u 1 ∂u 1 ∂2 u
+ + =0
∂r2 r ∂θ r2 ∂θ 2

Solution
∂u ∂u ∂r ∂u ∂θ
= · + ·
∂x ∂r ∂x ∂θ ∂x
∂u ∂u ∂r ∂u ∂θ
= · + ·
∂y ∂r ∂y ∂θ ∂y

Differentiate x = r cos θ and y = r sin θ with respect to x noting that r and θ are
functions of x, we have

30
d( x ) d
dx = dx (r cos θ ), which yields

∂r ∂θ
1= cos θ − r sin θ (2.1)
∂x ∂x

and similarly, one obtains

d(y) ∂r ∂θ
= sin θ + r cos θ
dx ∂x ∂x

∂r ∂θ
0 = sin θ + r cos θ (2.2)
∂x ∂x

We now solve (2.1) and (2.2) simultaneously. From (2.2), one obtains

∂θ − tan θ ∂r
=
∂x r ∂x
∂r tan θ ∂r
⇒ 1 = cos θ − r sin θ [− ]
∂x r ∂x
∂r ∂r
1 = cos θ + sin θ tan θ
∂x ∂x
∂r
1= (cos θ + sin θ tan θ )
∂x
∂r cos2 θ + sin2 θ
1= ( )
∂x cos θ
∂r 1
1= ( )
∂x cos θ
∂r ∂x 1
= cos θ ⇒ =
∂x ∂r cos θ
∂θ tan θ sin θ
=− cos θ = − .
∂x r r

Similarly, differentiating with respect to y, we get

∂r ∂θ cos θ
∂y = sin θ, ∂y = r .

31
∂y 1 ∂y r
∂r = sin θ , ∂θ = cos θ

∂u ∂u ∂r ∂u ∂θ
= · + ·
∂x ∂r ∂x ∂θ ∂x
∂u ∂u ∂r ∂u ∂θ
= · + ·
∂y ∂r ∂y ∂θ ∂y
∂u ∂u sin θ ∂u
= cos θ − .
∂x ∂r r ∂θ
∂u ∂u cos θ ∂u
= sin θ + .
∂y ∂r r ∂θ
∂ ∂u ∂2 u
( ) = 2
∂x ∂x ∂x
∂ ∂u sin θ ∂u
= [cos θ − ]
∂x ∂r r ∂θ
∂ ∂u sin θ ∂u ∂r ∂ ∂u sin θ ∂u ∂θ
= (cos θ − ) + (cos θ − )
∂r ∂r r ∂θ ∂x ∂θ ∂r r ∂θ ∂x
∂2 u sin θ ∂u sin θ ∂2 u
= (cos θ 2 + 2 − ) cos θ
∂r r ∂θ r ∂r∂θ
∂2 u ∂u cos θ ∂u sin θ ∂2 u sin θ
+ (cos θ − sin θ − − 2
)(− ).
∂r∂θ ∂r r ∂θ r ∂θ r
∂2 u 2 cos θ sin θ ∂u 2 cos θ sin θ ∂2 u sin2 θ ∂u sin2 θ ∂2 u
= cos2 θ 2 + − + + 2
∂r r2 ∂θ r ∂r∂θ r ∂r r ∂θ 2

Similarly,

∂2 u 2
2 ∂ u 2 sin θ cos θ ∂u 2 sin θ cos θ ∂2 u cos2 θ ∂u cos2 θ ∂2 u
= sin θ − + + + 2 .
∂y2 ∂r2 r2 ∂θ r ∂r∂θ r ∂r r ∂θ 2

Thus,
∂2 u ∂2 u ∂2 u 1 ∂u 1 ∂2 u
+ = + + .
∂x2 ∂y2 ∂r2 r ∂r r2 ∂θ 2

32
Chapter 3

Applications of Partial Derivatives

3.1 Differentials

Let f ( x, y) have continuous first partial derivatives in a region R of the xy-plane.

Then

4 f = f ( x + ∆x, y + ∆y) − f ( x, y)

= f x ∆x + f y ∆y + ε 1 ∆x + ε 2 ∆y

where ε 1 → 0,ε 2 → 0 as ∆x → 0, ∆y → 0.
Putting ∆x = dx, ∆y = dy, we obtain

∆ f = f x dx + f y dy + ε 1 dx + ε 2 dy

∴ d f = f x dx + f y dy.
This is called the principal part of ∆ f .

Example 3.1.1. If F ( x, y) = x3 − xy + 3y2 , compute

1. ∆ f

33
 2. d f , where x = 5, y = 4, ∆x = −0.2, ∆y = 0.1

3. and also find ∆ f and d f when x = 5, y = 4, ∆x = −2, ∆y = 1

SOLUTION

∆ f = ( x + ∆x )3 − ( x + ∆x )(y + ∆y) + 3(y + ∆y)2 − x3 + xy − 3y2

= (5 − 0.2)3 − (5 − 0.2)(4 + 0.1) + 3(4 + 0.1)2 − 53 + 5 × 4 − 3 × 42

= 4.83 − 4.8 × 4.1 + 3(4.1)2 − 125 + 20 − 48

= 110.592 − 19.68 + 50.43 − 153

= −11.658

d f = f x dx + f y dy

= (3x2 − y)dx + (6y − x )dy

= (3.52 − 4)(−0.2) + (6.4 − 5)0.1

= (75 − 4)(−0.2) + (24 − 5)0.1

= 71(−0.2) + 19.0.1

= −14.2 + 1.9

= −12.3

∆ f = (5 − 2)3 − (5 − 2)(4 + 1) + 3(4 + 1)2 − (53 − 5.4 + 3.42 )

= 33 − 3.5 + 3.52 − (125 − 20 + 48)

= 27 − 15 + 75 − 153

= 87 − 153

= −66

34
 d f = f x dx + f y dy

= (3x2 − y)dx + (6y − x )dy

= (3.52 − 4)(−2) + (6.4 − 5).1

= (75 − 4)(−2) + (24 − 5)

= −142 + 19

= −123

A comparison of the results in (1) and (2) shows that the result gets more accurate
when the change is small.

EXERCISE
If z = x3 y − 4xy2 + 8y3 ; find ( a) ∆z (b) dz

Determine whether each of the following is an exact differential, if it is, then find the
function.
( a) (2xy2 + 3y cos 3x )dx + (2x2 y + sin 3x )dy
(b) (6xy − y2 )dx + (2x expy − x2 )dy
SOLUTION

∂f
( a) = 2xy2 + 3y cos 3x
∂x
∂f
= 2x2 y + sin 3x
∂y
Z
f = (2xy2 + 3y cos 3x )dx + C (y)
2x2 y2 3y sin 3x
f = + + C (y)
2 3
= x2 y2 + y sin 3x + C (y)
∂f
= 2x2 y + sin 3x + C 0 (y)
∂y

i.e. 2x2 y + sin 3x + C 0 (y) = 2x2 y + sin 3x

35
C 0 (y) = 0
∴ f = x2 y2 + y sin 3x + K

Hence, it is an exact differential and the function is

f ( x, y) = x2 y2 + y sin 3x + K.

3.2 Extreme Values

Definition 3.2.1. A function of two variables has a local maximum or relative max-
imum value at the point ( a, b) in its domain if f ( x, y) ≤ f ( a, b) for all points in the
domain of f that are sufficiently close to the point ( a, b).

Definition 3.2.2. If f ( x, y) ≥ f (c, d), then f has a local minimum at (c, d) in its
domain.

3.2.1 Necessary Conditions For Extreme Values

A function f ( x, y) can have a local or absolute extreme value at a point ( a, b) in its

domain only if ( a, b) is

1. a critical point of f , that is, a point satisfying O f = 0 at ( a, b) i.e. O f ( a, b) = 0.

2.

2. a singular point of f i.e. O f ( a, b) does not exist.

3. a boundary point of the domain of f .

3.2.2 Sufficient Conditions for Extreme Values

If f is a continuous function of n variables whose domain is a closed and bounded

set in <n , then the range of f is a bounded set of real numbers, and there are

36
points in its domain where f takes on absolute maximum and minimum values.
∂f ∂f
Note:O f = ∂x i + ∂y j, the gradient of f in its domain where f takes absolute maxi-
mum or minimum values.

Example 3.2.1. Consider the function f ( x, y) = x2 + y2 .

To determine the critical point(s) of f we find O f .

∂ 2 ∂
O( x 2 + y2 ) = ( x + y2 ) i + ( x 2 + y2 ) j
∂x ∂x
== 2xi + 2yi

Now for O f = 0 ⇒ 2xi + 2yi = 0

2x = 0, 2y = 0 i.e. x = 0, y = 0.
∴ (0, 0) is a critical point of f .

3.3 Second Derivative Test for Critical Points

Suppose that the point ( a, b) is a critical point of f ( x, y) in the domain of f and

suppose that the second partial derivatives of f are continuous in a neighbourhood
of ( a, b) and have at that point the values
A = f 11 ( a, b), B = f 12 ( a, b) = f 21 ( a, b), C = f 22 ( a, b)

1. if B2 < AC and A > 0, then f has a local minimum value at ( a, b).

2. if B2 < AC and A < 0, then f has a local maximum value at ( a, b).

3. if B2 > AC, then f has a saddle point.

4. B2 = AC then the test fails

Example 3.3.1. : Classify the critical points of the function

f ( x, y) = 2x3 − 6xy + 3y2 .

37
Solution

f 1 ( x, y) = 6x2 − 6y = f x

f 2 ( x, y) = −6x + 6y = f y

for critical points, we obtain

f x = 0, fy = 0

6x2 − 6y2 = 0and − 6x2 + 6y = 0

x2 − y = 0, −x + y = 0
y = x2 , y=x
x = x2
x2 − x = 0
x ( x − 1) = 0
x = 0, or x=1
when x = 0, y = 0 when x = 1, y = 1, then (0, 0) and (1, 1) are critical points of
f.
f xx = 12x, f yy = 6, f xy = −6

At (0, 0),

A = f 11 (0, 0) = 0, B = f 12 (0, 0) = −6, C = f 22 (0, 0) = 6

∴ B2 = 36, AC = 0 i.e. B2 > AC i.e. 36 > 0 and A = 0

therefore, (0, 0) is a saddle point.
At(1, 1) :
A = 12, B = −6, C = 6.

38
B2 = 36, AC = 72
∴ B2 < AC i.e. 36 < 72 and A>0
⇒ (1, 1), is a minimum critical point of f . Thus, the minimum value is given by
f (1, 1) = 2 − 6 + 3 = −1

Example 3.3.2. : Find and classify the critical points of the function
2 + y2 )
f ( x, y) = x2 ye−( x .
Solution

2 + y2 ) 2 + y2 )
f x = 2xye−( x − 2x3 ye−(x
2 + y2 )
= 2xye−(x [1 − x 2 ]
2 + y2 ) 2 + y2 )
f y = x2 e−( x − 2x2 y2 e−(x
2 + y2 )
= x2 (1 − 2y2 )e−(x

for critical points,

f x = 0, and fy = 0

2 + y2 )
2xy(1 − x2 )e−( x =0 (3.1)

and
2 + y2 )
x2 (1 − 2y2 )e−( x =0 (3.2)

from (3.1),
2xy(1 − x2 ) = 0 (3.3)

and from (3.2),

x2 (1 − 2y2 ) = 0 (3.4)

from (3.3), 2xy = 0 or 1 − x2 = 0

39
x=0 or y=0 or x = ±1
from (3.4) x2 = 0, 1 − 2y2 = 0
or
q √
x=0 repeated or y = ± 12 = ± 22 .
√ √ √ √
2 2 2 2
Therefore the critical points are (0, 0), (1, 2 ), (1, − 2 ), (−1, 2 ), (−1, − 2 )

2 + y2 ) 2 + y2 ) 2 + y2 ) 2 + y2 )
f xx = 2ye−( x − 4x2 ye−(x − 6x2 ye−(x + 4x4 ye−(x
2 + y2 ) 2 + y2 ) 2 + y2 ) 2 + y2 )
f xy = 2xe−( x − 4xye−(x − 2x3 e−(x + 4x3 y2 e−(x
2 + y2 ) 2 + y2 ) 2 + y2 )
f yy = −2x2 ye−( x − 4x2 ye−(x + 4x2 y3 e−(x

At (0, 0),
A = f xx (0, 0) = 0, B = f xy (0, 0) = 0, C = f yy (0, 0) = 0, B2 = AC the test is
inconclusive.
 √ 
At 1, 22 ,

2 + y2 )
f xx = 2y(1 − 2x2 − 3x2 + 2x4 )e−( x
2 + y2 )
= 2y(1 − 5x2 + 2x4 )e−(x
2 + y2 )
f xy = 2x (1 − 2y2 − x2 + 2x2 y2 )e−( x
2 + y2 )
f yy = 2x2 y(−3 + 2y2 )e−( x

40
 0.15

0.1

0.05
f(x,y)

−0.05

−0.1

−0.15

100
90
80
100
70
60 80
50
60
40
30 40
20
20
10
y
x

Figure 3.1: Surface indicating minimum, maximum and saddle points.

 √ 
At 1, 22 ,

√ ! √
2 2 1 √ 3
A = f xx 1, = 2· (1 − 5 + 2)e−(1+ 2 ) = −2 2e− 2
2 2
√ ! √ √
2 2 2 2 2 −3
B = f xy 1, = 2 · 1(1 − 2( ) − 1 + 2 · (1)2 · ( ) e 2)
2 2 2
1 3
= 2(1 − 1 − 1 + 2 · ) e − 2 = 0
2
√ ! √ √
2 2 2 2 3 √ 3
C = f yy 1, = 2(1)2 (2( ) − 3)e− 2 = −2 2e− 2
2 2 2
√ 3
B2 − AC = 02 − (−2 2e− 2 )2 = −4.2e−3 = −8e−3

B2 < AC, A > 0.

√
2
∴ (1, 2 ) is a maximum point of f .

41
EXERCISE
Find and classify the critical points of the following functions:

1. f ( x, y) = xy − x2 − y2 − 2x − 2y + 4

−( x2 +y2 )
2. f ( x, y) = xye 2 .

3.4 QUADRATIC FORMS

It is not clear how the second derivative test for classifying points can be extended
to function of more than two variables. Such an extension is possible if the con-
ditions for classifying critical points for two variable functions as stated in terms of
quadratic forms. If u = ui + vj is a unit vector then the second directional derivative
of f ( x, y) at the point ( a, b) in the direction of the unit vector is given by:

42
 Du2 f ( a, b) = u · 5(u · 5 f )( a, b)
∂f ∂f
5f = i+ j
∂x ∂y

u = ui + vj
 
∂f ∂f
u · 5 f = (ui + vj) · i+ j
∂x ∂y
∂f ∂f
=u +v =F
∂x ∂y
∂F ∂F
5F = i+ j.
∂x ∂y
   
∂ ∂f ∂f ∂ ∂f ∂f
= u +v i+ u +v j
∂x ∂x ∂y ∂y ∂x ∂y
 2
∂2 f ∂2 f ∂2 f
  
∂ f
= u 2 +v i+ u +v 2 j
∂x ∂x∂y ∂x∂y ∂y
 2 2 ∂2 f ∂2 f
   
∂ f ∂ f
u · 5f = (ui + vj) · u 2 + v i+ u +v 2 j
∂x ∂x∂y ∂x∂y ∂y
 2
∂2 f ∂2 f ∂2 f
  
∂ f
= u u 2 +v +v u +v 2
∂x ∂x∂y ∂y∂x ∂y
2
∂ f 2
∂ f 2
∂ f 2
∂ f
= u2 2 + uv + uv + v2 2 .
∂x ∂x∂y ∂x∂y ∂y
∂2 f ∂2 f ∂2 f
= u2 2 + 2uv + v2 2 .
∂x ∂x∂y ∂y

Q(u, v) = Au2 + 2Buv + Cv2 (3.5)

∂2 f ∂2 f ∂2 f
evaluated at ( a, b) where A = ∂x2
, B= ∂x∂y , C= ∂y2

The expression (3.5) is called a quadratic form in the variables u and v. Such a
quadratic form is said to be ”positive definite” or ”negative definite” if Q(u, v) > 0
or Q(u, v) < 0 for all non-zero vectors u. Otherwise, it is said to be ”indefinite”.
Therefore, f has a local minimum, local maximum or a saddle point at the criti-

43
cal point ( a, b) if Q(u, v) is positive definite, negative definite or vanishes for some
vectors, positive for some and negative for others respectively.

For a function of three variables f ( x, y, z) with continuous second partial derivatives

the second directional derivatives at the critical point ( a, b, c) in the direction of the
unit vector u = ui + vj + wk is a given by u · 5(u · 5 f )( a, b, c)

  
∂f ∂f ∂f
u · 5(u · 5 f )( a, b, c) = (ui + vj + wk) · 5 (ui + vj + wk) · i+ j+ k
∂x ∂y ∂z
 
∂f ∂f ∂f
= (ui + vj + wk) · 5 u + v + w
∂x ∂y ∂z
    
∂ ∂f ∂f ∂f ∂ ∂f ∂f ∂f
= (ui + vj + wk) · u +v +w i+ u +v +w j
∂x ∂x ∂y ∂z ∂y ∂x ∂y ∂z
∂ ∂f ∂f ∂f
+ (u + v + w )k
∂z ∂x ∂y ∂z
   
∂ ∂f ∂f ∂f ∂ ∂f ∂f ∂f
=u u +v +w +v u +v +w
∂x ∂x ∂y ∂z ∂y ∂x ∂y ∂z
 
∂ ∂f ∂f ∂f
+w u +v +w
∂z ∂x ∂y ∂z
 2 2 2 ∂2 f ∂2 f ∂2 f
  
∂ f ∂ f ∂ f
= u u 2 +v +w +v u +v 2 +w
∂x ∂x∂y ∂x∂z ∂x∂y ∂y ∂y∂z
 2
∂2 f ∂2 f

∂ f
+w u +v +w 2
∂z∂x ∂z∂y ∂z
2
∂ f 2
∂ f 2
∂ f ∂2 f ∂2 f
= u2 2 + uv + uw + uv + v2 2
∂x ∂x∂y ∂x∂z ∂y∂x ∂y
∂2 f ∂2 f ∂2 f ∂2 f
+ vw + uw + vw + w2 2
∂y∂z ∂z∂x ∂z∂y ∂z
2
∂ f 2
∂ f 2
∂ f 2
∂ f
= u2 2 + v2 2 + w2 2 + 2uv
∂x ∂y ∂z ∂y∂x
∂2 f ∂2 f
+ 2uw + 2vw .
∂x∂z ∂z∂y

∴ The quadratic form is given by

Q(u, v, w) = Du2 f ( a, b, c) = Au2 + Bv2 + Cw2 + 2Duv + 2Euw + 2Fvw (3.6)

44
where A = f 11 ( a, b, c), B = f 22 ( a, b, c), C = f 33 ( a, b, c), D = f 12 ( a, b, c),
E = f 13 ( a, b, c), F = f 23 ( a, b, c).
Thus f ( x, y, z) has a local minimum at the critical point ( a, b, c) if Q(u, v, w) is pos-
itive definite. But a local maximum if Q(u, v, w) is negative definite and a saddle
point if Q(u, v, w) is positive for some vectors and negative for others. The test fails
if it vanishes.

Example 3.4.1. Find and classify the critical points of the function
f ( x, y, z) = xyz − x2 − y2 − z2 .

Solution
f x = yz − 2x = 0 (3.7)

f y = xz − 2y = 0 (3.8)

f z = xy − 2z = 0 (3.9)

from (3.9),
xy
z= (3.10)
2

Using (3.10) in (3.7) and (3.8) yields

xy2
2 − 2x = 0
x (y2 − 4) = 0, x=0 and y = ±2
x2 y
from (3.8), 2 − 2y = 0
y( x2 − 4) = 0, y = 0, or x = ±2
when x = 0, y = 0, z=0
when x = 2, y = 2, z=2
when x = −2, y = 2, z = −2
when x = 2, y = −2, z = −2
when x = −2, y = −2, z=2
The critical points are (0, 0, 0), (2, 2, 2), (2, −2, −2), (−2, 2, −2), and (−2, −2, −2).

45
f xx = −2, f yy = −2, f zz = −2, f xy = z, f xz = y, f yz = x at (0, 0, 0),
A = −2, B = −2, C = −2, D = 0, E=F=0
Thus, the quadratic form

Q(u, v, w) = Au2 + Bv2 + Cw2 + 2Duv + 2Euw + 2Fvw

= −2u2 − 2v2 − 2w2

= −2( u + v2 + w2 ) < 0

∴ (0, 0, 0) is a local maximum.

At (2, 2, 2), D = 2, E=2 and F=2

Q(u, v, w) = −2u2 − 2v2 − 2w2 + 4uv + 4uw + 4vw

= 2(−u2 + 2uv − v2 + 2vw − w2 + 2uw)

h i
= 2 −(u − v)2 − (u − w)2 − (v − w)2 + 2u2 + 2v2 + 2w2
h i
= −2 ( u − v )2 + ( u − w )2 + ( v − w )2 + 2( u2 + v2 + w2 )

Therefore, (2, 2, 2) is a saddle point, since the quadratic form is negative for some
vectors and positive for others. The quadratic form can easily be verified by the
use of an alternative procedure for determining the nature of critical points of func-
tions. It is an orderly procedure for determining whether or not a quadratic form is
negative or positive definite.

Q(u, v, w) = Au2 + Bv2 + Cw2 + 2Duv + 2Euw + 2Fvw

can be expressed as Q(u, v, w) = u T Qu

46
where    
u  A D E
   
u=
v
 and Q=
 D B F .

   
w E F C

Hence Q is the positive definite if A > 0

A D E
A D
> 0 and D B F >0
D B
E F C

Q is negative definite if A < 0

A D E
A D
> 0 and D B F <0
D B
E F C

If det( Q) 6= 0 but neither Q nor − Q is positive definite, then Q will be positive

for some vectors and negative for others in which case, we have a saddle point,
otherwise, it vanishes.
At (0, 0, 0), A = B = C = −2, 12 = z, f 13 = y, f 23 = x, D = E = F = 0.

A D −2 0
A = −2 < 0, = =4>0
D B 0 −2

−2 0 0
−2 0
det( Q) = 0 −2 0 = −2 = −2(4) = −8 < 0
0 −2
0 0 −2

Q has a maximum point at (0, 0, 0).

47
At (2, 2, 2), A = B = C = −2, D = E = F = 2

A D −2 2
A < 0, = =0
D B 2 −2

−2 2 2
−2 2 2 2 2 −2
det( Q) = 2 −2 2 = −2 −2 +2 = −2 · 0 − 2(−8) + 2(8) = 32 > 0
2 −2 2 −2 2 2
2 2 −2

Hence (2, 2, 2), is a saddle a point.

Similarly, the nature of the remaining critical points can be determined.

48
Chapter 4

Line and Multiple Integrals

4.1 Vector Integration

Vectors can be integrated in a similar way to scalar functions as illustrated in the

following examples:

R2
1. If ~f = u2 − u i + 2u3 j − 3k, find 1 ~f (u)du

2
u3 u2 u4
Z 2 h  i  
2 3
u − u i + 2u j − 3k du = − i + j − 3uk
1 3 2 2
   1  
8 1 1 1
= − 2 i + 8j − 6k − − i + j − 3k
3 3 2 2
2 1 1
= i + 8j − 6k + i − j + 3k
3 6 2
5 15
= i + j − 3k
6 2

2. The acceleration of a particle at any time t is given by

~a = 12 cos 2ti − 8 sin 2tj + 16tk

49
 ~ and displacement ~r are zero at t = 0, find V
If the velocity V ~ and ~r at any time

t.
Solution

Z Z
~ =
V ~adt = [12 cos 2ti − 8 sin 2tj + 16tk] dt

= 6 sin 2ti + 4 cos 2tj + 8t2 k + C

~

~ =0
Now, when t = 0, V
⇒ ~ =0
0i + 4j + 0k + C
~ = −4j
⇒C

∴ ~ = 6 sin 2ti + 4 cos 2tj + 8t2 k − 4j

V

= 6 sin 2ti + 4 (cos 2t − 1) j + 8t2 k

Similarly, we obtain ~r as follows

Z Z h i
~ = 2
~r = Vdt 6 sin 2ti + 4 (cos 2t − 1) j + 8t k dt
8
= −3 cos 2ti + (2 sin 2t − 4t) j + t3 k + ~B
3
when t = 0,~r = 0
⇒ −3i + 0j + 0k + ~B = 0
⇒ ~B = 3i

8
∴ ~r = −3 cos 2ti + (2 sin 2t − 4t) j + t3 + 3i
3
8
= 3 (1 − cos 2t) i + (2 sin 2t − 4t) j + t3 k
3

50
4.2 LINE INTEGRALS

~ = A1 i +
Let C be a space curve and ~r the position vector of any point on C. Let A
~ along C is given by
A2 j + A3 k be a force. The line integral of A

Z
~ ~r,
A.d
c

provided A1 , A2 , A3 are integrable functions of the indicated variable. If P1 , P2 are

~ from P1 to P2 is given by
points on C , then the line integral of A

Z P2
~ · d~r
A
P1

Example
~ = 3x2 + 6y i − 14yzj + 20xz2 k,


If A
~ · d~r from (0, 0, 0) to (1, 1, 1) along the following path C :
R
evaluate c A
(i) x = t, y = t2 , z = t3
(ii) the straight line from (0,0,0) to (1,0,0) then (1,1,0) and then (1,1,1).
(iii) the straight line joining (0,0,0) to (1,1,1).
Solution
(i) x = t, y = t2 , z = t3 at (0,0,0) to (1,1,1)
Let t vary from 0 to 1, then

dx = dt, dy = 2tdt, dz = 3t2 dt

51
 Z Z 1 h  ih i
∴ ~ · d~r =
A 3t2 + 6t2 i − 14t2 · t3 j + 20t(t3 )2 k dti + 2tdtj + 3t2 dtk
C 0
Z 1  
= 9t2 i − 14t5 j + 20t7 k i + 2tj + 3t2 k dt
0
Z 1h i
= 9t2 − 14t5 (2t) + 20t7 (3t2 ) dt
0
Z 1 
= 9t2 − 28t6 + 60t9 k dt
0
1
9t3 28t7 60t10

= − +
3 7 10 0

= 3 − 4 + 6 = 5.

(ii) (0,0,0) to (1,0,0) to (1,1,0) to (1,1,1)

Along (0,0,0) to (1,0,0), x varies from 0 to 1, y = 0, z = 0

Z Z 1  Z 1 
2 2
∴ ~ · d~r =
A 3x i · idx = 3x dx = x3 |10 = 1
C 0 0

Along (1,0,0) to (1,1,0), x = 1, dx = 0, y varies from 0 to 1, z = 0

Z Z 1
⇒ ~ · d~r =
A (3 + y) i · jdy = 0
C 0

Along (1,1,0) to (1,1,1),x = 1, y = 1, z varies from 0 to 1, dx = 0, dy = 0

Z 1 Z 1
20 3 1 20
Z
∴ ~ · d~r = 20
A xz2 dz = 20 z2 dz = z |0 = .
C 0 0 3 3

∴ The whole integral is

~ · d~r = 1 + 0 + 20 = 23
Z
A
C 3 3

52
(iii) From (0,0,0) to (1,1,1)
Let t vary from 0 to 1. The parametric equations are paid out as follows:

t 0 1 grad Eqn

x 0 1 1 x=t
y 0 1 1 y=t
z 0 1 1 z=t

Thus the parametric equation of the straight line joining (0,0,0) and (1,1,1) are x =
t, y = t, z = t

Z Z 1 
~ · d~r = 2 2 3
∴ A 3t + 6t − 14t + 20t dt
C 0
Z 1 
2 3
= −11t + 6t + 20t dt
0
−11 3
= t + 3t2 + 5t4 |10
3
−11
= +3+5
3
13
=
3

Example 4.2.1. Calculate the work done by the force ~F = xyi − z2 j + xyzk when a
particle is moved from the point (1,0,2) to the point (3, -2, -1) along the straight line
joining them.

Solution
~F · d~r
R
Work done = C

Now, ~F · d~r = xydx − z2 dy + xyzdz

To obtain the parametric equations of the straight line joining (1,0,2) and (3, -2, -1),

53
let t vary from 0 to 1 and then proceed as follows.

t 0 1 grad equation

x 1 3 2 x = 2t + 1
y 0 -2 -2 y = −2t
z 2 -1 -3 z = 2 − 3t

Using the parametric equations, we obtain

x = 2t + 1 ⇒ dx = 2dt
y = −2t ⇒ dy = −2dt
z = 2 − 3t ⇒ dz = −3dt

Z Z 1h i
∴ ~F · d~r = (2t + 1)(−2t) · 2dt − (2 − 3t)2 (−2)dt + (2t + 1)(−2t)(2 − 3t)(−3)dt
C 0
Z 1 Z 1 Z 1
= −4 (2t + t)dt + 2 (4 − 12t + 9t )dt + 6 (2t + t2 − 6t3 )dt
2 2
0 0 0
2 1 1
t3 3 4
 
2 3 t h
2 3
i 1
2
= −4 t + + 2 4t − 6t + 3t + 6 t + − t
3 2 0 0 3 2 0
   
2 1 1 3
= + + 2(4 − 6 + 3) + 6 1 + −
3 2 3 2
−14
= +2−1
3
−11
=
3
Assignment
If ~F = xyi − xzj + 2yz2 k, evaluate ~F · d~r if C is the straight line joining (0,0,1) to
R
C

(1,2,-3).

54
4.3 SURFACE INTEGRAL

Consider a plane area S as shown above. The vector area ~S is defined as ~S = Sn̂,
where n̂ is a unit normal to the plane area. In the case of a surface S in R3 ,

∂~S = ∂Sn̂ (4.1)

where ∂S is an elemental surface area.

Definition 4.3.1. Let V (~r ) be a scalar field and ~F (~r ) be a vector field both with a
prescribed surface S. Let ∂Si and ∂~
Si be elemental part of S and ~S respectively for
i = 1, 2, . . . , N.

1. The surface integral of V over S is given as

Z N

S
V (~r )d~S = lim ∑ V (~r)∂~Si
∂Si →o i =1
(4.2)

and this is a vector.

2. The scalar surface integral of ~F over S is given as

Z N

S
~F (~r ) · d~S = lim ∑ ~F(~r) · ∂~Si
∂Si →o i =1
(4.3)

which is a scalar.

3. The vector surface integral of ~F over S is given as

Z N

S
~F (~r ) ∧ d~S = lim
∂Si →o
∑ ~F(~r) ∧ ∂~Si (4.4)
i =1

and this is a vector.

55
 Evaluation of Surface Integral

From (4.1),
∂~S grad f
n̂ = =
∂S | grad f |

where the Cartesian equation of S is known in the form f ( x, y, z) = 0, since grad f is

a vector normal to the surface f ( x, y, z) = constant. Thus

grad f
d~S = dS (4.5)
| grad f |

∂f ∂f ∂f
But grad f = ∂x i + ∂y j + ∂z k, then from (4.2) and (4.3), we have

1.

Z Z ∂f Z ∂f Z ∂f
∂y
Vd~S = i V ∂x
dS + j V dS + k V ∂z
dS (4.6)
S S | grad f | S | grad f | S | grad f |

2.
     
∂f ∂f ∂f

Z Z  Fx Fy Fz 
∂x ∂y ∂z
~F · d~S = + + dS (4.7)
S S  | 5 f| | 5 f| | 5 f| 

3.
            
∂f ∂f ∂f ∂f ∂f ∂f

Z Z  Fy
∂z − Fz ∂y Fz ∂x − Fx ∂z Fx ∂y − Fy ∂x

~F ∧ d~S = i+ j+ k dS(4.8)
S S  | grad f | | grad f | | grad f | 

That is,
~F = Fx i + Fy j + Fz k

∂f ∂f ∂f
d~S = i+ j+ k
∂x ∂y ∂z

56
 i j k
∴ ~F ∧ d~S = Fx Fy Fz
∂f ∂f ∂f
∂x ∂y ∂z
     
= i Fy ∂∂zf − Fz ∂∂yf − j Fx ∂∂zf − Fz ∂∂xf + k Fx ∂∂yf − Fy ∂∂xf

In this way, the evaluation of the surface integrals is reduced to integration over the
surface S. The coordinate system to be used for performing the double integration
depends on the nature of S, as demonstrated in the following examples.

Given that V = x2 y2 z2 , evaluate ~ where S is the curved surface of the cylin-

R
S Vd S.

der x2 + y2 = 9 lying between the planes z = 0 and z = 2 and included in the first
quadrant.

Solution
~
R R
S Vd S = V n̂dS
grad f
But n̂ = | grad f |
, where f ( x, y, z) = x2 + y2

5( x2 + y2 ) 2xi + 2yj xi + yj xi + yj x y
∴ n̂ = 2 2
=p =p = √ = i+ j
| 5 ( x + y )| 4x2 + 4y2 x 2 + y2 9 3 3

1 1
Z Z Z  
3 2 2 2 3 2
∴ Vd~S = V ( xi + yj)dS = x y z i + x y z j dS
S 3 3

We now change the coordinate from ( x, y, z) to the cylindrical coordinates (ρ, ϕ, z),
x = ρ cos ϕ, y = ρ sin ϕ, z = z.
p
dS = ρdϕdz. But ρ = x2 + y2 = 3
x = 3 cos ϕ, y = 3 sin ϕ, z = z.

57
 Z π h
1 2
Z Z i
2 3 3 2 2 2 2 2 3 3 2
~
∴ VdS = 3 cos ϕ · 3 sin ϕz i + 3 cos ϕ · 3 sin ϕz j 3.dϕdz
S 3 z =0 ϕ =0
Z π Z 2 Z π
1 2 1
Z
2 5 3 2 2 2
= i 3 cos ϕ · sin ϕz dϕdz + j 35 cos2 ϕ · sin3 ϕz2 dϕdz
3 z =0 ϕ =0 3 z =0 ϕ =0
Z 2 Z π Z 2 Z π
4 2 2 3 2 4 2 2
=3 i z dz cos ϕ · sin ϕdϕ + 3 j z dz cos2 ϕ · sin3 ϕdϕ
z =0 ϕ =0 z =0 ϕ =0
8 2 8 2
= 34 · · i + 34 · · j
3 15 3 15
2 2
= 33 · 8 · i + 33 · 8 · j
15 15
9·8·2 9·8·2
= i+ j
5 5
144
= ( i + j ).
5

Exercise
If ~F = xi + xyj + xyzk, evaluate ~F · d~S, where S is the complete surface of the
R
S

sphere x2 + y2 + z2 = 4.

4.4 VOLUME INTEGRAL

The volume integral of a scalar field V (~r ) over a given volume τ is the sum of all
V (~r )∂τi for all elemental volumes ∂τi , 1 ≤ i ≤ n and taking the limit of this ∂τi → 0
(i.e. as n → ∞). This is given by

Z n

τ
V (~r )∂τ = lim ∑ V (~r)∂τi
∂τi →0 i =1
(4.9)

and this is a scalar quantity.

Similarly, for any vector field ~F (~r ), the definition of the volume integral of ~F over τ

58
is

Z n

τ
~F (~r )∂τ = lim ∑ ~F(~r)∂τi
∂τi →0 i =1
(4.10)

and this is a vector quantity.

Note that τ = τ ( x, y, z).
The evaluation of the volume integrals depends on the shape of the volume τ. The
shape of volume helps in choosing coordinates. Note that

1. for Cartesian coordinates ( x, y, z), dτ = dxdydz

2. for cylindrical polar coordinates (ρ, ϕ, z), dτ = ρdρdϕdz

3. for spherical polar coordinates (r, θ, ϕ), dτ = r2 sin θdrdθdϕ

To change from ( x, y) to (r, θ ), use the Jacobian transformation

dx dx
dr dθ
dxdy = drdθ
dy dy
dr dθ

cos θ −r sin θ
= drdθ
sin θ r cos θ

= rdrdθ

Example
R
If V = xy, evaluate τ Vdτ, where τ is the volume bounded by the planes x = y =
z = 0, 3x + 2y + z = 6

Solution

59
 Z ZZZ
Vdτ = ( xy)dxdydz
τ

Since 3x + 2y + z = 6 ⇒ z = 6 − 3x − 2y
∴ z varies from 0 to 6 − 3x − 2y.
6−3x
For the y−component, y varies from 0 to 2 , that is, z is left out.
For the x −component, x varies from 0 to 2, that is, y and z are left out.

Z Z 2 Z 6−3x Z 6−3x−2y
2
∴ Vdτ = xydzdydx
τ x =0 y =0 z =0

But
Z 6−3x−2y
6−3x −2y
xydz = xyz|0
z =0

= xy(6 − 3x − 2y)

= 6xy − 3x2 y − 2xy2 .

Also,

6−3x
3 2xy3 6−23x
Z  
2
6xy − 3x2 y − 2xy2 dy = 3xy2 − x2 y2 − |
y =0 2 3 0
3x 3x2 x
= (6 − 3x )2 − (6 − 3x )2 − (6 − 3x )3
4  8 12

x 2 3x 1
= (6 − 3x ) 3 − − (6 − 3x )
4 2 3
9x
= (2 − x )3
8

60
Hence
Z 2
9x
Z
Vdτ = (2 − x )3 dx
τ x =0 8
1 2
 
9 x
Z
4 2 4
= − (2 − x ) |0 + (2 − x ) dx
8 4 4 0
 2
9 x 4 1 5
= − (2 − x ) − (2 − x )
8 4 20 0
  
9 1
= 0 − − · 32
8 20
 
9 32 9
= = .
8 20 5

61
References

62