Professional Documents
Culture Documents
Krishnas - Calculus (Etc.) (Pdfnotes - Co)
Krishnas - Calculus (Etc.) (Pdfnotes - Co)
Krishnas - Calculus (Etc.) (Pdfnotes - Co)
co
hna 's
Kris TEXT BOOK on
C alculus
(For B.A. and B.Sc. Ist year students of All Colleges affiliated to universities in Uttar Pradesh)
By
A. R. Vasishtha
Retired Head, Dep’t. of Mathematics
Meerut College, Meerut (U.P.)
Dedicated
to
Lord
Krishna
Authors & Publishers
www.pdfnotes.co
Preface
This book on CALCULUS has been specially written according to the latest
Unified Syllabus to meet the requirements of the B.A. and B.Sc. Part-I Students
of all Universities in Uttar Pradesh.
The subject matter has been discussed in such a simple way that the students will
find no difficulty to understand it. The proofs of various theorems and examples have
been given with minute details. Each chapter of this book contains complete theory
and a fairly large number of solved examples. Sufficient problems have also been
selected from various university examination papers. At the end of each chapter an
exercise containing objective questions has been given.
We have tried our best to keep the book free from misprints. The authors shall be
grateful to the readers who point out errors and omissions which, inspite of all care,
might have been there.
The authors, in general, hope that the present book will be warmly received by
the students and teachers. We shall indeed be very thankful to our colleagues for
their recommending this book to their students.
The authors wish to express their thanks to Mr. S.K. Rastogi, Managing Director,
Mr. Sugam Rastogi, Executive Director, Mrs. Kanupriya Rastogi Director and entire
team of KRISHNA Prakashan Media (P) Ltd., Meerut for bringing out this book
in the present nice form.
The authors will feel amply rewarded if the book serves the purpose for which it is
meant. Suggestions for the improvement of the book are always welcome.
— Authors
www.pdfnotes.co
Syllabus
Calculus
U.P. UNIFIED (w.e.f. 2011-12)
Differential Calculus
Unit-1: ε
-δ
definition of the limit of a function, Continuous functions and
classification of discontinuities, Differentiability, Chain rule of Differentiability,
Rolle’s theorem, First and second mean value theorems, Taylor’s theorems with
Lagrange’s and Cauchy’s forms of remainder, Successive differentiation and
Leibnitz’s theorem.
Unit-3: Maxima and Minima (for functions of two variables), Tangents and
normals (polar form only), Curvature, Envelopes and evolutes.
Integral Calculus
Unit- 4(b): Reduction formulae, Beta and Gamma functions.
B rief C ontents
SECTION
A
DIFFERENTIAL CALCULUS
C hapters
1. Differentiability
2.
1. Differentiation
3.
1. Successive Differentiation
4.
1. Expansions of Functions
5.
1. Indeterminate Forms
6.
1. Partial Differentiation
7.
1. Jacobians
8.
1. Curvature
11.
Envelopes, Evolutes
12. and Involutes
1. Asymptotes
13.
1.2 Limits
lim
Note 1 : It is not at all necessary for x → a f (x) to exist that f be defined at
x = a. It is enough that for some δ > 0, f be defined whenever 0 < ⏐ x − a⏐ < δ.
Note 2 : If N be a neigh bourhood of a, t hen N ~ {a} is called a deleted
neighbourhood of a.
Note 3 : If a function f has a finite limit at a point a, then by the definition of the
limit of a function a deleted neighbourhood of a exists on which f is bounded.
Now we shall prove a theorem which is the foundation on which the definition of
limit rests. If this theorem were not true, the definition of limit would have been useless.
lim lim lim
Theorem : If x → a f (x) = l , and x → a f (x) = m , then l = m i.e., if x → a f (x)
exists, then it is unique.
Proof : Suppose, if possible, l ≠ m.
Let us take ε = 12 ⏐ l − m⏐ . Then ε > 0.
lim
Since x → a f (x) = l, for a given ε > 0, there exists δ 1 > 0 such that
⏐ f (x) − l⏐ < ε whenever 0 < ⏐ x − a⏐ < δ 1. …(1)
lim
Again since x → a f (x) = m, for a given ε > 0, there exists δ 2 > 0 such that
⏐ f (x) − m⏐ < ε whenever 0 < ⏐ x − a⏐ < δ 2 . …(2)
If we ch o o se δ = m i n . {δ 1, δ 2}, t h e n 0 < ⏐ x − a⏐ < δ imp lies t h at bo t h
0 < ⏐ x − a⏐ < δ 1 and 0 < ⏐ x − a⏐ < δ 2 hold, and hence, we have
⏐ f (x)⏐ 1
≤ ⏐ m − g (x)⏐ + ⏐ f (x) − l⏐ …(1)
⏐ m⏐ ⏐ g (x)⏐ ⏐ m⏐
lim
S i n c e x → a f (x) = l, t h e r e fo r e t h e r e e xi st s a d e l e t e d n e i gh b o u r h o o d
] a − δ 1, a + δ 1 [ − {a} of the point x = a in which the function f is bounded. Let
K > 0 be such that
⏐ f (x)⏐ ≤ K whenever 0 < ⏐ x − a⏐ < δ 1 .
lim
Again since g (x) ≠ 0 for all x in the domain of g and x → a g (x) = m ≠ 0, therefore
there exist numbers k > 0 and δ 2 > 0 such that
1 1
⏐ g (x)⏐ > k i.e., < whenever 0 < ⏐ x − a⏐ < δ 2 .
⏐ g (x)⏐ k
[See theorem 1 of article 1.3]
Let δ′ = min (δ 1, δ 2).
The inequality (1) can then be written as
⎪ f (x) l⎪ K 1
⎪ − ⎪≤ ⏐ m − g (x)⏐ + ⏐ f (x) − l⏐ , ...(2)
⎪ g (x) m ⎪ k ⏐ m⏐ ⏐ m⏐
for all x such that 0 < ⏐ x − a⏐ < δ′.
Now take any given ε > 0.
lim lim
Since x → a f (x) = l and x → a g (x) = m, we can find positive numbers δ 3 and
δ 4 such that
ε
⏐ f (x) − l⏐ < ⏐ m⏐ ⋅ whenever 0 < ⏐ x − a⏐ < δ 3
2
k ⏐ m⏐ ε
and ⏐ g (x) − m⏐ < ⋅ whenever 0 < ⏐ x − a⏐ < δ 4.
K 2
Take δ = min {δ′, δ 3 , δ 4}. Then from (2), we get
⎪ f (x) l⎪ ε ε
⎪ − ⎪ < + = ε, whenever 0 < ⏐ x − a⏐ < δ.
⎪ g (x) m⎪ 2 2
lim f (x) lim f (x) lim ⎛ f ⎞ l
∴ x → a g (x) exists and x → a g (x) = x → a ⎜⎝ g ⎟⎠ (x) = m , if m ≠ 0.
Alternative Proof :
lim 1 1
Since m ≠ 0, therefore, by theorem 1 of § 3, x → a exists and equals ⋅
g (x) m
lim ⎛ f ⎞ lim ⎧ 1 ⎫
Now x → a ⎜⎝ g ⎟⎠ (x) = x → a ⎨⎩ f (x) ⋅ g (x) ⎬⎭
⎧ lim ⎫ ⎧ lim 1 ⎫
= ⎨ x → a f (x) ⎬ ⎨ x → a ⎬ [By theorem 3 of article 1.3]
⎩ ⎭⎩ g (x) ⎭
1 l
= l⋅ = ⋅
m m
Theorem 5 : L et f be defined on D and let f (x) ≥ 0 for all x ∈ D.
lim lim
If x → a f (x) exists, then x → a f (x) ≥ 0.
LIMITS AND CONTINUITY D-11
lim
Proof : Suppose that x → a f (x) = l and l is negative.
1
Taking ε = − 2
l, we can find a positive number δ > 0 such that
1
⏐ f (x) − l⏐ < − 2
l whenever 0 < ⏐ x − a⏐ < δ.
It gives that
3l l
< f (x) < < 0 whenever 0 < ⏐ x − a⏐ < δ.
2 2
This is a contradiction since we are given that f (x) ≥ 0 for all x ∈ D. H ence l cannot
be negative.
lim
Consequently x → a f (x) ≥ 0.
Corollary : L et f be defined on D and let f (x) > 0 for all x ∈ D.
lim lim
If x → a f (x) exists, then x → a f (x) ≥ 0.
Proof : Since f (x) > 0 ⇒ f (x) ≥ 0, therefore now we can apply theorem 5 of
article 1.3.
Theorem 6 : L et f and g be defined on D and let f (x) ≥ g (x) for all x ∈ D. Then
lim lim
x → a f (x) ≥ x → a g (x), provided these limits exist.
lim lim
Proof : Let x → a f (x) = l, x → a g (x) = m.
Let us define a function h by h (x) = f (x) − g (x) —
V x ∈ D. Then, we have
—
(i) h (x) ≥ 0 V x ∈ D.
lim lim
(ii) x → a h (x) exsits and x → a h (x) = l − m.
lim
(iii) x → a h (x) ≥ 0, by theorem 5 of article 1.3.
Thus, from (ii) and (iii), we get l − m ≥ 0 i.e., l ≥ m,
lim lim
i.e., x → a f (x) ≥ x → a g (x).
lim lim
Corollary : L et f (x) > g (x) for all x ∈ D. Then x → a f (x) ≥ x → a g (x) , provided
these lim its exist.
Theorem 7 : L et f, g and h be defined on D and let f (x) ≥ g (x) ≥ h (x) for all x.
lim lim
L et x → a f (x) = x → a h (x) .
lim lim lim lim
Then x → a g (x) exists, and x → a g (x) = x → a f (x) = x → a h (x).
lim lim
Proof : Let x → a f (x) = x → a h (x) = l.
Then corresponding to any given ε > 0, we can find positive numbers δ 1 and δ 2
such that
⏐ f (x) − l⏐ < ε whenever 0 < ⏐ x − a⏐ < δ 1
i.e., l − ε < f (x) < l + ε whenever 0 < ⏐ x − a⏐ < δ 1 …(1)
and l − ε < h (x) < l + ε whenever 0 < ⏐ x − a⏐ < δ 2. …(2)
D-12 DIFFERENTIAL CALCULUS
Choosing δ to be smaller than δ 1 and δ 2 , we see from (1) and (2) that
l − ε < h (x) ≤ g (x) ≤ f (x) < l + ε whenever 0 < ⏐ x − a⏐ < δ.
Thus l − ε < g (x) < l + ε whenever 0 < ⏐ x − a⏐ < δ
or ⏐ g (x) − l⏐ < ε whenever 0 < ⏐ x − a⏐ < δ.
lim lim
H ence x → a g (x) exists and x → a g (x) = l.
lim lim
Theorem 8 : If x → a f (x) = l, then x → a ⏐ f (x)⏐ = ⏐ l⏐ .
Proof : We have ⏐ f (x) − l⏐ ≥ ⏐ ⏐ f (x)⏐ − ⏐ l⏐ ⏐ , for all x. …(1)
[... ⏐ p − q⏐ ≥ ⏐ ⏐ p⏐ − ⏐ q⏐ ⏐ ]
Let ε > 0 be given.
lim
Since x → a f (x) = l, therefore, given ε > 0, there exists a number δ > 0 such that
⏐ f (x) − l⏐ < ε whenever 0 < ⏐ x − a⏐ < δ. …(2)
From (1) and (2), we get
⏐⏐ f (x)⏐ − ⏐ l⏐⏐< ε whenever 0 < ⏐ x − a⏐ < δ.
lim lim
Consequently x → a ⏐ f (x)⏐ exists and x → a ⏐ f (x)⏐ = ⏐ l⏐ .
Theorem 9 : I f t h e re i s a n u m b e r δ > 0 s u c h t h a t h (x) = 0 wh en ever
lim
0 < ⏐ x − a⏐ < δ, then x → a h (x) = 0.
Proof : For any ε > 0, the number δ > 0, given in the hypothesis of the theorem
is such that h (x) = 0 whenever 0 < ⏐ x − a⏐ < δ
or ⏐ h (x) − 0⏐ = 0 < ε whenever 0 < ⏐ x − a⏐ < δ.
lim
H ence x → a h (x) = 0.
Corollary : I f t h e re i s a n u m b e r δ > 0 s u c h t h a t f (x) = g (x) wh en ever
lim lim
0 < ⏐ x − a⏐ < δ , then x → a f (x) = x → a g (x).
Proof : Let us define a function h by setting
h (x) = f (x) − g (x) for all x.
Then h (x) = 0 whenever 0 < ⏐ x − a⏐ < δ.
Now, apply theorem 9 of article 1.3.
Note : The above corollary has deep implications. It asserts that the concept of
limit is a ‘local’ one. If two functions agree on some neighbourhood of a point a, then
they cannot approach different limits as x approaches a.
lim
Theorem 10 : I f x → a f (x) = 0 a n d g (x) i s b o u n d ed in so m e d e le t ed
lim
neighbourhood of a, then x → a f (x) g (x) = 0.
Proof : Since g (x) is bounded in some deleted neighbourhood of a, therefore
there exist numbers k > 0 and δ 1 > 0 such that
⏐ g (x)⏐ ≤ k whenever 0 < ⏐ x − a⏐ < δ 1. ...(1)
LIMITS AND CONTINUITY D-13
Important Note : If both right hand limit and left hand limit of f as x → a, exist
and are equal in value, their common value, evidently, will be the limit of f as x → a. If
however, either or both of these limits do not exist, the limit of f as x → a does not exist.
E ven if both these limits exist but are not equal in value then also the limit of f as
x → a does not exist.
D-14 DIFFERENTIAL CALCULUS
1.5 Limits as x → + ∞ (− ∞ )
Definition : A function f is said to approach l as x becom es positively infinite, if
corresponding to each ε > 0, there exists δ > 0 such that ⏐ f (x) − l⏐ < ε whenever
x ≥ δ.
lim
Then we write x → ∞ f (x) = l or f (x) → l as x → ∞ .
Definition : A function f is said to approach l as x becom es negatively infinite, if
corresponding to each ε > 0 there exists δ > 0 such that ⏐ f (x) − l⏐ < ε whenever
x ≤ − δ.
lim
Then we write x → − ∞ f (x) = l
or f (x) → l as x → − ∞ .
Note 1 : The results on the limits of sum, product and quotient of functions also
hold good here provided that in these cases l + m, lm, l ⁄ m are defined.
lim lim
Note 2 : If x → ∞ f (x) = l exists, x → ∞ g (x) does not exist (as a finite real
lim
number), even then x → ∞ f (x) g (x) can exist. Similar is the case as x → − ∞ .
x2 − a2
Example 1 : L et f be the function given by f (x) = , x ≠ a.
x− a
lim
Using (ε, δ) definition show that x → a f (x) = 2a.
Solution : Let ε > 0 be given. In order to show that
lim
x → a f (x) = 2a,
we have to show that for any given ε > 0, there exists a number δ > 0 such that
⏐ f (x) − 2a⏐ < ε whenever 0 < ⏐ x − a⏐ < δ.
⎪ x2 − a 2 ⎪
If x ≠ a, then ⏐ f (x) − 2a⏐ = ⎪ − 2a⎪ = ⏐ (x + a) − 2a⏐ [... x ≠ a]
⎪ x− a ⎪
= ⏐ x − a⏐ .
LIMITS AND CONTINUITY D-15
lim ⎛ 1⎞
Example 2 : Using (ε, δ) definition show that x → 0 ⎜ x sin ⎟ = 0⋅
⎝ x⎠
(Meerut 2012, 13; Rohilkhand 13B)
lim ⎛ 1⎞
Solution : Let ε > 0 be given. In order to show that x → 0 ⎜ x sin ⎟ = 0,
⎝ x⎠
we have to show that for any given ε > 0, there exists a number δ > 0 such that
⎪ 1 ⎪
⎪ x sin − 0⎪ < ε whenever 0 < ⏐ x − 0⏐ < δ.
⎪ x ⎪
⎪ 1 ⎪ ⎪ 1⎪ ⎪ 1⎪
Now ⎪ x sin − 0⎪ = ⏐ x ⏐ ⎪ sin ⎪ ≤ ⏐ x⏐ , because ⎪ sin ⎪ ≤ 1.
⎪ x ⎪ ⎪ x⎪ ⎪ x⎪
⎪ 1 ⎪
∴ ⎪ x sin − 0⎪ < ε whenever ⏐ x⏐ < ε.
⎪ x ⎪
Choosing a number δ such that 0 < δ ≤ ε, we have
⎪ 1 ⎪
⎪ x sin − 0⎪ < ε whenever 0 < ⏐ x⏐ < δ.
⎪ x ⎪
lim 1
H ence x → 0 x sin = 0.
x
Example 3 : Show by (ε, δ) m ethod that the function f, defined on R − {0} by
f (x) = sin (1 ⁄ x) whenever x ≠ 0 , does not tend to 0 as x tends to 0. (Meerut 2013B)
Solution : In order to show that sin (1 ⁄ x) does not tend to 0 as x tends to 0, take
ε = 12 . By Archimedean property of real numbers for any δ > 0 there exists a positive
integer n such that
1 1
n> i.e., δ > ⋅
πδ nπ
2 1 1
∴ 0< < < < δ.
(4n + 1) π 2nπ nπ
2
Take x= ⋅ Then 0 < ⏐ x − 0⏐ < δ.
(4n + 1) π
1
Also, ⏐ sin (1 ⁄ x) − 0⏐ = ⏐ sin (2nπ + π)⏐ = 1 > ε.
2
1
Thus we have shown that there exists an ε > 0, namely , such that for every
2
⎡ 2
δ > 0 t h e r e i s a n x ⎢⎣ = , wh e r e n i s a p o sit i ve i n t e ge r su ch t h a t
(4n + 1) π
2 ⎤
< δ ⎥⎦ such that
(4n + 1) π
0 < ⏐ x − 0⏐ < δ and ⏐ sin (1 ⁄ x) − 0⏐ > ε.
H ence sin (1 ⁄ x) does not tend to 0 as x tends to 0.
lim ⏐ x − 2⏐
Example 4 : Show that x → 2 does not exist.
x− 2
D-16 DIFFERENTIAL CALCULUS
1 ⎛1 ⎞ 1 ⎛1 ⎞ ⎛1 ⎞
⎡
⎢ ⎜ − 1⎟ ⎜ − 1⎟ ⎜ − 2⎟ ⎤
⎥
h ⎝h ⎠ h ⎝h ⎠ ⎝h ⎠
lim ⎢⎢⎢ 1 ⎥
⎥
= 1+ ⋅h+ h2 + h3 + …⎥⎥
h → 0 ⎢⎣ h 1.2 1.2.3 ⎦
lim ⎡ 1 1.(1 − h) 1.(1 − h) (1 − 2h) ⎤
= ⎢1 + + + + …⎥
h →0 ⎣ 1! 2! 3! ⎦
1 1 1
= 1+ + + + …∞ = e.
1! 2! 3!
Similarly, the left hand limit i.e.,
lim lim lim
f (0 − 0) = h → 0 f (0 − h) = h → 0 f (− h) = h → 0 (1 − h) − 1 ⁄ h = e.
LIMITS AND CONTINUITY D-17
sin x
Solution : Let f (x) = ⋅
x
lim lim lim sin h
H ere f (0 + 0) = h → 0 f (0 + h) = f (h) =
h →0 h →0 h
h3 h5
h−
+ − …
lim 3! 5! lim ⎛⎜ h2 h4 ⎞
= = 1− + − …⎟ = 1.
h →0 h h →0 ⎝ 3! 5! ⎠
lim lim
Similarly f (0 − 0) = h → 0 f (0 − h) = h → 0 f (− h)
lim lim
= h → 0 f (h) = h → 0 h sin (1 ⁄ h)
= 0 × a finite quantity lying between − 1 and 1
= 0.
lim
Similarly, left hand limit i.e., f (0 − 0) = h → 0 f (0 − h)
lim lim lim
= h → 0 f (− h) = h → 0 (− h) sin (− 1 ⁄ h) = h → 0 h sin (1 ⁄ h) = 0.
lim
Since f (0 + 0) = f (0 − 0) = 0, therefore y → 0 y sin (1 ⁄ y) = 0
lim sin x
i.e., x → ∞ x = 0.
lim 1
(e) x → 0 sin x ⋅
number l to which sin (1 ⁄ h) tends as h tends to zero. Therefore the right hand limit
f (0 + 0) does not exist.
lim
Similarly the left hand limit f (0 − 0) also does not exist. Thus x → 0 sin (1 ⁄ x) does
not exist.
lim a x − 1
(f) x →0 ⋅ (Meerut 2003; Kanpur 10)
x
ax− 1
Solution : Let f (x) =
x
1 + x log a + (x2 ⁄ 2 !) (log a) 2 + … − 1
=
x
x [log a + 12 x (log a) 2 + …]
= ⋅
x
⎡ h ⎤
h ⎢ log a + (log a) 2 + …⎥
⎣ 2 ⎦
lim lim
H ere f (0 + 0) = h → 0 f (h) = h → 0 = log a.
h
lim lim
Also f (0 − 0) = h → 0 f (0 − h) = h → 0 f (− h) = log a.
lim a x − 1
Since f (0 + 0) = f (0 − 0) = log a, therefore x → 0 = log a.
x
lim 1
(g) x → 0 ⋅ e1 ⁄ x.
x
1 1⁄x
Solution : Let f (x) = .e .
x
lim lim lim 1
Then f (0 + 0) = h → 0 f (0 + h) = h → 0 f (h) = h → 0 e1 ⁄ h
h
= ∞ , since both 1 ⁄ h and e1 ⁄ h tend to ∞ as h → 0.
lim lim lim 1
Also f (0 − 0) = h → 0 f (0 − h) = h → 0 f (− h) = h → 0 − e− 1 ⁄ h
h
lim − 1 lim − 1
= h →0 1 ⁄ h = h →0
he ⎛ 1 1 1 ⎞
h ⎜1 + + ⋅ + …⎟
⎝ h 2 ! h2 ⎠
lim − 1
= h →0 = 0.
h + 1 + (1 ⁄ 2h) + …
lim 1
Since f (0 + 0) ≠ f (0 − 0) , therefore x → 0 e1 ⁄ x does not exist.
x
lim (1 + x) n − 1
(h) x →0 ⋅
x
(1 + x) n − 1
Let f (x) = ⋅
x
lim lim (1 + h) n − 1
Then f (0 + 0) = h → 0 f (h) = h → 0
h
LIMITS AND CONTINUITY D-19
n (n − 1) 2
1 + nh + h + …− 1
lim 2!
= h →0
h
⎡ n (n − 1) ⎤
h ⎢n + h + …⎥
⎣ 2! ⎦
lim
= h →0
h
lim ⎡ n (n − 1) ⎤
= h →0 ⎢ n + h + …⎥ = n.
⎣ 2! ⎦
lim lim
Also f (0 − 0) = h → 0 f (0 − h) = h → 0 f (− h)
lim (1 − h) n − 1
= h →0
− h
n (n − 1)
1 + n (− h) + (− h) 2 + … − 1
lim 2!
= h →0 = n.
− h
lim
Since f (0 + 0) = f (0 − 0) = n, therefore x → 0 f (x) = n.
lim x m − a m
(i) x →a x − a ⋅
xm − a m
Solution : Let f (x) = ⋅
x− a
lim lim (a + h) m − am
Then f (a + 0) = h → 0 f (a + h) = h → 0
a+ h− a
⎡⎛ h⎞ m ⎤
am ⎢ ⎜ 1 + ⎟ − 1⎥
lim ⎣⎝ a⎠ ⎦
= h →0
h
lim am ⎡ 2 ⎤
⎢ 1 + m ⋅ h + m (m − 1) h + … − 1⎥
= h →0 ⎢ ⎥
h ⎢⎣ a 2! a 2 ⎥
⎦
lim ⎡m m (m − 1) h ⎤ m
= h → 0 am ⎢ + . 2 + …⎥ = am ⋅ = mam − 1.
⎣ a 2 a ⎦ a
lim lim (a − h) m − am
Also f (a − 0) = h → 0 f (a − h) = h → 0 = mam − 1.
a− h− a
lim
Since f (a + 0) = f (a − 0) = mam − 1, hence x → a f (x) = m am − 1.
Example 6 : Find the right hand and the left hand limits in the following cases and
discuss the existence of the lim it in each case :
lim 2 x2 − 8 lim e1 ⁄ x − 1
(i) x → 2 ; (ii) x → 0 1 ⁄ x ;
x− 2 e + 1
(Meerut 2003; Kanpur 11; Rohilkhand 14)
D-20 DIFFERENTIAL CALCULUS
lim
(iii) x → 0 f (x) where f (x) is defined as
f (x) = x, when x > 0 ; f (x) = 0, when x = 0 ; f (x) = − x, when x < 0.
(Purvanchal 2008)
2x2− 8
Solution : (i) Let f (x) = ⋅
x− 2
lim lim 2 (2 + h) 2 − 8
We have f (2 + 0) = h → 0 f (2 + h) = h → 0
2+ h− 2
lim 2(4 + 4h + h2) − 8 lim 8h + 2h2
= h →0 = h →0
h h
lim h (8+ 2h) lim
= h →0 = h → 0 (8 + 2h) = 8.
h
lim lim 2 (2 − h) 2 − 8
Again f (2 − 0) = h → 0 f (2 − h) = h → 0
2− h− 2
lim 2 (4 − 4h + h2) − 8 lim − 8h + 2h2
= h →0 = h →0
− h − h
lim − h (8 − 2h) lim
= h →0 = h → 0 (8 −. 2h) = 8
− h
lim 2x2 − 8
Since f (2 + 0) = f (2 − 0) = 8, therefore x → 2 exists and is equal to 8.
x− 2
e1 ⁄ x − 1
(ii) Let f (x) = ⋅
e1 ⁄ x + 1
H ere the right hand limit, i.e.,
lim lim lim e1 ⁄ h − 1
f (0 + 0) = h → 0 f (0 + h) = h → 0 f (h) = h → 0 1 ⁄ h
e + 1
1 ⁄ h 1 ⁄ h
lim e [1 − (1 ⁄ e ]
= h →0 1 ⁄ h = 1.
e [1 + (1 ⁄ e1 ⁄ h )]
Again the left hand limit, i.e.,
lim lim lim e− 1 ⁄ h − 1
f (0 − 0) = h → 0 f (0 − h) = h → 0 f (− h) = h → 0 − 1 ⁄ h
e + 1
(1 ⁄e 1 ⁄ h ) − 1 0− 1
lim
= h →0 = = − 1.
(1 ⁄ e1 ⁄ h ) + 1 0 + 1
lim e1 ⁄ x − 1
Since f (0 + 0) ≠ f (0 − 0), hence x → 0 1 ⁄ x does not exist.
e + 1
(iii) We have the right hand limit i.e., f (0 + 0)
lim
= h → 0 f (0 + h), where h is + ive but sufficiently small
lim lim
= h → 0 f (h) = h → 0 h, [... h > 0 and f (x) = x if x > 0]
= 0.
Also, the left hand limit, i.e., f (0 − 0)
lim
= h → 0 f (0 − h), where h is + ive but sufficiently small
LIMITS AND CONTINUITY D-21
lim lim
= h → 0 f (− h) = h → 0 − (− h), [... − h < 0 and f (x) = − x if x < 0]
lim
= h → 0 h = 0.
Thus both the limits f (0 + 0) and f (0 − 0) exist and are equal to zero.
lim
H ence x → 0 f (x) exists and is equal to zero.
⎧ x if x is rational
Example 7 : L et f (x) = ⎨
⎩ − x if is irrational.
lim
Show that x → a f (x) exists only when a = 0.
(Purvanchal 2007)
lim lim
= h → 0 (− h) or h → 0 h, according as − h is rational or irrational
= 0.
lim lim
Again f (0 + 0) = h → 0 f (0 + h) = h → 0 f (h)
lim lim
= h →0 h or h → 0 (− h), according as h is rational or irrational
= 0.
lim
Since f (0 + 0) = f (0 − 0) = 0, hence x → 0 f (x) exists and is equal to zero.
Case III : If a is an irrational number.
lim
In this case f (a − 0) = h → 0 f (a − h)
lim lim
= h → 0 (a − h) or h → 0 − (a − h),
according as (a − h) is rational or irrational
= a or − a i.e., is not unique.
∴ f (a − 0) does not exist.
lim
∴ x → a f (x) does not exist.
lim
Thus we see that x → a f (x) exists only when a = 0.
D-22 DIFFERENTIAL CALCULUS
lim
Since f (2 + 0) ≠ f (2 − 0), hence x → 2 f (x) does not exist.
lim lim 1
Example 9 : If x → a f (x) = ± ∞ , then x → a f (x) = 0.
lim
Solution : Let x → a f (x) = + ∞ .
Let ε > 0 be given. If ε1 = 1 ⁄ ε, then ε1 > 0.
lim
Since x → a f (x) = ∞ , therefore for ε1 > 0, there exists δ > 0 such that
f (x) > ε1 whenever 0 < ⏐ x − a ⏐ < δ
1 1
i.e., < whenever 0 < ⏐ x − a ⏐ < δ
f (x) ε1
1
i.e., 0< < ε whenever 0 < ⏐ x − a ⏐ < δ [... ε = 1 ⁄ ε1]
f (x)
1
i.e., − ε< < ε whenever 0 < ⏐ x − a ⏐ < δ
f (x)
⎪ 1 ⎪
i.e, ⎪ − 0⎪ < ε whenever 0 < ⏐ x − a ⏐ < δ.
⎪ f (x) ⎪
lim 1
∴ x → a f (x) = 0⋅
Similarly it can be proved that
lim 1 lim
x → a f (x) = 0 when x → a f (x) = − ∞ .
sin [x]
Example 10 : If f (x) = , [x] ≠ 0 and f (x) = 0, [x] = 0,
[x]
lim
where [x] denotes the greatest integer less than or equal to x, then find x → 0 f (x).
(Kanpur 2010)
LIMITS AND CONTINUITY D-23
lim lim
Solution : H ere f (0 + 0) = h → 0 f (0 + h) = h → 0 f (h)
lim
= h →0 0 [... [h] = 0]
= 0.
lim lim
Also f (0 − 0) = h → 0 f (0 − h) = h → 0 f (− h)
lim sin [− h] ,
= h →0 [ [ ... [− h] = − 1 ≠ 0 ]
− h]
lim sin (− 1) sin (− 1)
= h →0 = = sin 1 ≠ 0.
(− 1) (− 1)
lim
Since f (0 + 0) ≠ f (0 − 0), therefore x → 0 f (x) does not exist.
lim
1. U sing definition of limit, show that x → 0 f (x) = 1 where
⎧
⎪ 1 + x2 if x ≠ 0
f (x) = ⎨⎪
⎩ 0 if x = 0.
⎧ 2, if x is irrational
2. If f is defined on R as f (x) = ⎨
⎩ 1, if x is rational,
lim
prove that x → a f (x) does not exist for any a ∈ R.
⎧ 0, if x is irrational
3. If f is defined on R as f (x) = ⎨
⎩ 1, if x is rational,
lim
prove that x → a f (x) does not exist for any a ∈ R.
4. If x → 0, then does the limit of the following function f exist or not ?
f (x) = x, when x < 0 ; f (x) = 1, when x = 0 ; f (x) = x2, when x > 0.
lim ax− 1 lim 2x − 1
5. U se the formula x → 0 = log a to find x → 0 ⋅
x (1 + x) 1 ⁄ 2 − 1
6. If f (x) = e− 1 ⁄ x, show that at x = 0, the right hand limit is zero while the left hand
limit is + ∞ , and thus there is no limit of the function at x = 0.
7. lim
G ive an example to show that x → a f (x) may exist even when the function is not
defined for x = a.
⎧ x, 0≤ x< 1
8. Let f (x) = ⎨⎩
3 − x, 1 ≤ x ≤ 2.
lim
Show that x → 1 + f (x) = 2. Does the limit of f (x) at x = 1 exist ?
G ive reasons for your answer.
lim x − ⏐ x⏐
9. E valuate x → 0 ⋅ (Meerut 2001)
x
D-24 DIFFERENTIAL CALCULUS
lim ⏐ sin x⏐
10. E valuate x → 0 ⋅
x
lim e1 ⁄ x
11. E valuate x → 0 ⋅
e1 ⁄ x + 1 (Avadh 2010)
lim
4. Yes; x → 0 f (x) = 0. 5. 2 log 2. 8. Does not exist.
9. R ight hand limit is 0 and left hand limit is 2 and so the limit does not exist.
10. Does not exist because the right hand limit is 1 and the left hand limit is − 1.
11. The limit does not exist because the right hand limit is 1 and the left hand limit
is 0.
1.7 Continuity
(Purvanchal 2010, 11; Avadh 14)
The intuitive concept of continuity is derived from geometrical considerations. If
the graph of the function y = f (x) is a continuous curve, it is natural to call the function
continuous. This requires that there should be no sudden changes in the value of the
function. A small change in x should produce only a small change in y. Moreover for
the graph to be a continuous running curve, it should possess a definite direction at
each point.
But the continuity as defined in pure analysis is quite distinct from the intuitive
or the geometrical concept of the term. Sometimes drawing a graph is difficult. We now
give the arithmetical definition of continuity given by Cauchy.
Cauchy’s definition of continuity : A real valued function f defined on an open
interval I is said to be continuous at a ∈ I iff for any arbitrarily chosen positive number ε ,
however sm all, we can find a corresponding num ber δ > 0 such that
⏐ f (x) − f (a)⏐ < ε whenever ⏐ x − a⏐ < δ. …(1)
(Bundelkhand 2010; Kanpur 11)
We say that f is a continuous function if it is continuous at every x ∈ I.
In other words, f is continuous at a if for any given ε > 0, we can find a δ > 0 such
that
⏐ x − a⏐ < δ ⇒ ⏐ f (x) − f (a)⏐ < ε.
This means that the function f will be continuous at x = a if the difference between
f (a) and the value of f (x) at any point in the interval ]a − δ, a + δ[ can be made less
than a pre-assigned positive number ε. Note that we choose δ after we have chosen ε.
LIMITS AND CONTINUITY D-25
= a0 c n + a1 c n − 1 + … + a n − 1 c + an
⎡
⎢
... lim x = c⎤⎥
⎣ x →c ⎦
= f (c).
D-26 DIFFERENTIAL CALCULUS
lim
Since x → c f (x) = f (c), therefore f (x) is continuous at x = c.
Thus f (x) is continuous at every real number c and so f (x) is continuous for all
x ∈ R.
Thus remember that a polynomial function f (x) is always continuous at each
point of its domain.
1.8 Discontinuity
Definition : If a function is not continuous at a point, then it is said to be
discontinuous at that point and the point is called a point of discontinuity of this function.
Types of discontinuity : (Avadh 2014)
(i) Removable discontinuity : (Meerut 2011)
lim
A function f is said to have a rem ovable discontinuity at a point a if f (x)
x →a
exists but is not equal to f (a) i.e., if
f (a + 0) = f (a − 0) ≠ f (a).
The function can be made continuous by defining it in such a way that
LIMITS AND CONTINUITY D-27
lim
x → a f (x) = f (a).
(ii) Discontinuity of the first kind or ordinary discontinuity : (Meerut 2010B)
A function f is said to have a discontinuity of the first kind or ordinary discontinuity
at a if f (a + 0) and f (a − 0) both exist but are not equal. The point a is said to be a
point of discontinuity from the left or right according as f (a − 0) ≠ f (a) = f (a + 0) or
f (a − 0) = f (a) ≠ f (a + 0).
(iii) Discontinuity of the second kind : A funct ion f i s sa i d t o h a ve a
discontinuity of the second kind, at a if none of the limits f (a + 0) and f (a − 0) exist.
The point a is said to be a point of discontinuity of the second kind from the left or right
according as f (a − 0) or f (a + 0) does not exist. (Meerut 2003, 10B)
(iv) Mixed discontinuity : (Meerut 2012B)
A function f is said to have a m ixed discontinuity at a, if f has a discontinuity of
second kind on one side of a and on the other side a discontinuity of first kind or may
be continuous.
(v) Infinite discontinuity : A function f is said to have an infinite discontinuity
at a if f (a + 0) or f (a − 0) is + ∞ or − ∞ . O bviously, if f has a discontinuity at a and is
unbounded in every neighbourhood of a, then f is said to have an infinite discontinuity
at a.
lim lim 1
= h → 0 f (h) = h → 0 h sin = 0. [See theorem 10 of article 1.3]
h
⎡ . . lim 1
⎢ . h → 0 h = 0 and ⎪⎪ sin ⎪⎪ ≤ 1 for all h ≠ 0 i.e., sin (1 ⁄ h) is bounded
⎣ ⎪ h⎪
⎤
in some deleted neighbourhood of zero ⎥⎦
lim
Similarly f (0 − 0) = h → 0 f (0 − h), h > 0
lim sin (1 ⁄ x) ⎡ 0⎤
= x →∞ ⎢ Form ⎥
1⁄x ⎣ 0⎦
lim sin θ , 1
= θ →0 θ putting x = θ so that θ → 0 as x → ∞
= 1.
Thus y → 1 as x → ∞ and so the straight line y = 1 is an asymptote of the curve.
Although the function is continuous at the origin, yet the graph of the function
in the vicinity of the origin cannot be drawn, since the function oscillates infinitely often
in any interval containing the origin.
F rom t he graph it is clear that the function makes an infinite number of
oscillations in the neighbourhood of x = 0. The oscillations, however, go on diminishing
in length as x → 0.
LIMITS AND CONTINUITY D-29
lim
= h →0 h = 0 ;
lim lim
and f (n− 0) = h → 0 f (n− h) = h → 0 {(n− h)− (n− 1)} [... n − 1 < n − h < n]
lim
= h → 0 (1 − h) = 1;
lim ⎡ ...
= h → 0 ⎡⎢ 32 − ⎛⎜ 12 + h⎞⎟ ⎤⎥ ⎢
1
2
< 1
2
+ h < 1⎤⎥
⎣ ⎝ ⎠⎦ ⎣ ⎦
lim
= h → 0 (1 − h) = 1 ≠ φ ⎛⎜ 12 ⎞⎟ = 1
2
⋅
⎝ ⎠
Thus the function φ (x) is discontinuous from the right also at x = 12 ⋅
In this way φ (x) has discontinuity of the first kind i.e., ordinary discontinuity at
x = 12 a n d t h e j u m p o f t h e fu n c t i o n a t x = 1 ⁄ 2 i s φ ⎛⎜ 12 + 0⎞⎟ − φ ⎛⎜ 12 − 0⎞⎟ i.e.,
⎝ ⎠ ⎝ ⎠
1 − 0 i.e., 1.
(iii) For x = 1, we have φ (1) = 1,
lim
φ (1 − 0) = h → 0 φ (1 − h)
lim ⎡ Note that
= h → 0 [(3 ⁄ 2) − (1 − h)], ⎢
1
2
< 1 − h < 1⎤⎥
⎣ ⎦
lim
= h → 0 ⎛⎜ 12 + h⎞⎟ = 12 ⋅
⎝ ⎠
Sin ce φ (1) ≠ φ (1 − 0), φ (x) is
discon tinu ous at x = 1 and the discont inuity is
ordinary.
H ence the function φ (x) has three points of
discontinuity at x = 0, 12 and 1.
The graph of the function consists of the
point (0, 0); t he segment of the line y = 12 − x,
0 < x < 12 ; the point ⎛⎜ 12 , 12 ⎞⎟ ; the segment of the line
⎝ ⎠
y = 32 − x, 12 < x < 1; and the point (1, 1).
Thus the graph is as shown in the figure.
From the graph we observe that the function is
discontinuous at x = 0, 12 and 1.
Example 6 : Determ ine the values of a, b, c for which the function
⎧ sin (a + 1) x + sin x
⎪ for x < 0
⎪ x
⎪
⎪
f (x) = ⎨ c for x = 0
⎪
⎪
⎪ (x + bx2) 1 ⁄ 2 − x1 ⁄ 2
⎪ for x > 0
⎩ bx3 ⁄ 2
is continuous at x = 0.
lim lim (h + bh2) 1 ⁄ 2 − h1 ⁄ 2
Solution : H ere f (0 + 0) = h → 0 f (0 + h) = h → 0
bh3 ⁄ 2
1⁄2 − 1 1
lim (1 + bh) lim {1 + 2 bh + …} − 1 1 ,
= h →0 = h →0 =
bh bh 2
which is independent of b and so b may have any real value except 0.
lim lim sin (a + 1) (− h) + sin (− h)
Again f (0 − 0) = h → 0 f (0 − h) = h → 0
(− h)
2 sin ⎛⎜ 12 a + 1⎞⎟ h cos (ah ⁄ 2)
⎝ ⎠
lim sin (a + 1) h + sin h lim
= h →0 = h →0
h h
LIMITS AND CONTINUITY D-33
lim ⎛ sin ah ⎞ 2 2
= h →0 ⎜ ⎟ . a = 1 . a2 = a2 ;
⎝ ah ⎠
lim lim lim sin2 (− ah)
and f (0 − 0) = h → 0 f (0 − h) = h → 0 f (− h) = h → 0
(− h) 2
lim sin2 ah
= h →0 = a 2.
h2
Now f (x) is continuous at x = 0 iff
f (0 + 0) = f (0 − 0) = f (0).
H ence f (x) is discontinuous at x = 0 unless a = 1.
Example 10 : A function f (x) is defined as follows :
f (x) = 1 + x if x ≤ 2 and f (x) = 5 − x if x ≥ 2.
Is the function continuous at x = 2 ? (Meerut 2002, 06)
Solution : H ere f (2) = 1 + 2 or 5 − 2 = 3 ;
lim
f (2 + 0) = h → 0 f (2 + h), where h is + ive and sufficiently small
lim
= h → 0 [5 − (2 + h)], [... 2 + h > 2 and f (x) = 5 − x if x > 2]
lim
= h → 0 (3 − h) = 3 ;
lim
and f (2 − 0) = h → 0 f (2 − h), where h is + ive and sufficiently small
lim
= h → 0 [1 + (2 − h)], [... 2 − h < 2 and f (x) = 1 + x if x < 2]
lim
= h → 0 (3 − h) = 3.
Thus f (2 + 0) = f (2 − 0) = f (2). H ence the function f (x) is continuous at x = 2.
Example 11 : Discuss the continuity of the function f (x) defined as follows:
f (x) = x2 for x < − 2, f (x) = 4 for − 2 ≤ x ≤ 2, f (x) = x2 for x > 2.
Solution : We shall test the continuity of f (x) only at the points x = − 2 and 2.
O bviously it is continuous at all other points.
At x = − 2. We have f (− 2) = 4;
lim lim
f (− 2 + 0) = h → 0 f (− 2 + h) = h → 0 4 = 4 ;
lim lim
f (− 2 − 0) = h → 0 f (− 2 − h) = h → 0 (− 2 − h) 2, [... − 2 − h < − 2]
= 4.
Since f (− 2 + 0) = f (− 2 − 0) = f (− 2), the function is continuous at x = − 2.
At x = 2. We have f (2) = 4 ;
lim lim
f (2 + 0) = h → 0 f (2 + h) = h → 0 (2 + h) 2 = 4 ;
lim lim
f (2 − 0) = h → 0 f (2 − h) = h → 0 4 = 4.
Since f (2 + 0) = f (2 − 0) = f (2), the function is continuous at x = 2.
LIMITS AND CONTINUITY D-35
1 1
4. If f (x) = sin for x ≠ 0 and f (0) = 0, show that f (x) is finite for every value of
x x
x in the interval [− 1, 1] but is not bounded. Determine the points of discontinuity
of the function if any.
⎧ x, if x is rational
5. A function f defined on [0, 1] is given by f (x) = ⎨
⎩ 1 − x, if x is irrational.
Show that f takes every value between 0 and 1 (both inclusive), but it is continuous
1
only at the point x = ⋅
2 (Rohilkhand 2012B)
6. Prove that the function f defined by
⎧1 ,
⎪ 2 if x is rational
f (x) = ⎨ 1
⎪ , if x is irrational
⎩3
is discontinuous everywhere.
xe1 ⁄ x ,
7. (i) Show that the function f defined by f (x) = x ≠ 0, f (0) = 1
1 + e1 ⁄ x
is not continuous at x = 0 and also show how the discontinuity can be
removed. (Meerut 2011; Rohilkhand 06)
(ii) Show that the function f (x) = 3x2 + 2x − 1 is continuous for x = 2.
(iii) S h o w t h a t t h e fu n ct i o n f (x) = (1 + 2x) 1 ⁄ x, x ≠ 0, f (x) = e2, x = 0 i s
continuous at x = 0.
8. E xamine the continuity of the function
⎧ − x2 if x ≤ 0
⎪
⎪ 5x − 4 if 0 < x ≤ 1
f (x) = ⎨ 2
⎪ 4 x − 3x if 1 < x < 2
⎪ 3x + 4 if x ≥ 2
⎩
at x = 0, 1 and 2. (Meerut 2004, 06B, 07B; Avadh 06; Purvanchal 06, 10)
9. (i) Show that the function
e1 ⁄ x − 1 ,
f (x) = 1 ⁄ x x ≠ 0 and f (0) = 0
e + 1
is discontinuous at x = 0.
Show that the following function is continuous at x = 0.
sin− 1 x ,
f (x) = x ≠ 0, f (0) = 1.
x
1 , when x ≠ 0 and f (0) = 0
10. Discuss the continuity of the function f (x) =
1 − e1 ⁄ x
for all values of x. (Rohilkhand 2010B)
⏐ x⏐
11. Prove that the function f (x) = for x ≠ 0, f (0) = 0 is continuous at all points
x
except x = 0. (Meerut 2009; Kanpur 08, 09)
e1 ⁄ x sin (1 ⁄ x),x≠ 0
12. Test the continuity of the function f (x) at x = 0 if f (x) =
1 + e1 ⁄ x
and f (0) = 0. (Meerut 2005)
LIMITS AND CONTINUITY D-37
13. E xamine the following function for continuity at x = 0 and x = 1 :
⎧x2 for x ≤ 0
⎪
⎪1 for 0 < x ≤ 1
f (x) = ⎨
⎪1 (Meerut 2001, 03, 04B, 05)
⎪ for x > 1.
⎩x
lim x2− 1
2. = …… .
x →1 x− 1
lim sin (x ⁄ 4)
3. = …… .
x →0 x
lim x 3 − 1
4. = …… .
x →1 x2 − 1
lim
5. (1 + x) 1 ⁄ x = …… .
x →0
lim ax − 1
6. = …… .
x →0 x
7. If f (x) = x − [x], where [x] denotes the greatest integer less than or equal to x , then
f (x) = …… , for 3 < x < 4.
⎧ x , 0≤ x< 1
8. Let f (x) = ⎨⎩
3 − x , 1 ≤ x ≤ 2.
lim
Then f (x) = …… .
x → 1−
⎧1
⎪
, x< 1
9. Let f (x) = ⎨⎪ 2 − x , 1 ≤ x < 2
⎩2 , x ≥ 2.
3 lim lim
Then (i) f ( ) = …… (ii) f (x) = …… and (iii) f (x) = ……
2 x → 1+ x → 2−
(Meerut 2003)
lim | sin x |
10. = …… .
x →0 − x
lim
11. A function f (x) has a removable discontinuity at x = a if f (x) exists but
x →a
is not equal to …… .
LIMITS AND CONTINUITY D-39
sin x
12. The domain of the function f (x) = is …… .
x
⎧ sin x
⎪ , x≠ 0
13. The domain of the function f (x) = ⎨⎪ x
⎩1 , x= 0
is …… .
Multiple Choice Questions:
Indicate the correct answer for each question by writing the corresponding letter from
(a), (b), (c) and (d).
lim | x |
14. is equal to
x →0 x
(a) 1 (b) –1 (c) 2 (d) The limit does not exist
lim ex − 1
15. is equal to
x →0 x
(a) 0 (b) 1 (c) –1 (d) 2
lim | x − 2|
16. is equal to
x → 2+ x − 2
(a) –1 (b) 1 (c) 2 (d) –2
lim | x − 3|
17. is equal to
x → 3− x − 3
(a) –1 (b) 3 (c) –3 (d) 1 (Meerut 2003; Rohilkhand 14)
lim e 1 ⁄ x − 1
18. is equal to
x → 0+ e 1 ⁄ x + 1
(a) –1 (b) 1 (c) 0 (d) 2
True or False:
Write ‘T’ for true and ‘F’ for false statement.
⎧
⎪
x , when x < 0
19. If f (x) = ⎨ 1 , when x = 0
⎪
⎩ x 2, when x > 0,
lim
then f (x) = 0.
x →0
⎧ sin x
⎪ , x≠ 0
20. The function f (x) = ⎨⎪ x
⎩2 , x= 0
is continuous at x = 0.
⎧
sin x , x ≥ 0
21. The function f (x) = ⎨⎩
− sin x , x < 0
is continuous at x = 0.
lim
22. For f (x) to exist, the function f (x) must be defined at x = a.
x →a
D-40 DIFFERENTIAL CALCULUS
⎧ x cos (1 ⁄ x), x ≠ 0
23. The function f (x) = ⎨⎩
0 , x= 0
is discontinuous at x = 0.
24. If a function f is continuous at a, then | f | is also continuous at a.
⎧ sin x
⎪ , x≠ 0
25. The function f (x) = ⎨⎪ x
⎩ 1 , x= 0
is continuous at x = 0.
⎧1
⎪
, x< 1
26. The function f (x) = ⎨⎪ 2 − x , 1 ≤ x < 2
⎩2 , x≥ 2
is discontinuous at x = 1.
lim sin x
27. = 1.
x →∞ x
lim sin 3x
28. = 1.
x →0 x
lim sin 2 x
29. = 2.
x →0 x
lim sin x 1
30. = ⋅
x →0 2 x 2
1 3
1. f (a). 2. 2. 3. ⋅ 4. ⋅ 5. e.
4 2
6. log e a. 7. x − 3. 8. 1.
1
9. (i) (ii) 1 (iii) 0. 10. − 1. 11. f (a).
2
12. R − {0}. 13. R. 14. (d). 15. (b). 16. (b).
17. (a). 18. (b). 19. T. 20. F. 21. T.
22. F. 23. F. 24. T. 25. T. 26. F.
27. F. 28. F. 29. T. 30. T.
2.1 Definitions
Derivative at a point : (Bundelkhand 2010; Purvanchal 11)
L et I denote the open interval ]a, b[ in R and let x0 ∈ I. Then a function f : I → R is
said to be differentiable (or derivable) at x0 iff
lim f (x0 + h) − f (x0) lim f (x) − f (x0)
or equivalently x → x
h →0 h 0 x − x0
exists finitely and this lim it, if it exists finitely, is called the differential coefficient or
derivative of f with respect to x at x = x0 .
It is denoted by f ′ (x0) or by D f (x0).
Progressive and regressive derivatives :
The progressive derivative of f at x = x0 is given by
lim f (x0 + h) − f (x0)
, h > 0.
h →0 h
It is also called the right hand differential coefficient of f at x = x0 and is denoted
by R f ′ (x0) or by f ′ (x0 + 0).
The regressive derivative of f at x = x0 is given by
D-42 DIFFERENTIAL CALCULUS
lim ( fg) (x) − ( fg) (x0) lim f (x) g (x) − f (x0) g (x0)
Now x → x0 = x →x
x − x0 0 x − x0
1 1 ,
= − f ′ (x0) ⋅ ⋅ using (1) and (2)
f (x0) f (x0)
= − f ′ (x0) ⁄ { f (x0)}2.
H ence 1 ⁄ f is differentiable at x0 and
(1 ⁄ f )′ (x0) = − f ′ (x0) ⁄ { f (x0)}2.
lim f (0 + h) − f (0)
At x = 0. We have R f ′ (0) =
h →0 h
lim f (h) − f (0) lim 1 − 1
= = , as f (x) = 1 if 0 ≤ x ≤ 1
h →0 h h →0 h
lim
= 0 = 0,
h →0
lim f (0 − h) − f (0)
and L f ′ (0) =
h →0 − h
lim f (− h) − f (0) lim [1 − 2 (− h)] − 1
= =
h →0 − h h →0 − h
[... f (x) = 1 − 2x, if x < 0]
lim 2h lim
= = − 2 = − 2⋅
h →0 − h h →0
... R f ′ (0) ≠ L f ′ (0), so the given function is not differentiable at x = 0.
At x = 1. We have
lim f (1 + h) − f (1) lim [2 (1 + h) − 1] − 1
R f ′ (1) = =
h →0 h h →0 h
lim 2 + 2h − 1 − 1 lim 2h lim
= = = 2 = 2,
h →0 h h →0 h h →0
lim f (1 − h) − f (1) lim 1 − 1 lim
and L f ′ (1) = = = 0 = 0.
h →0 − h h →0 − h h →0
... R f ′ (1) ≠ L f ′(1), so the given function f (x) is not differentiable at x = 1.
⎧ 1, − 2≤ x≤ 0
and ⏐ f (x)⏐ = ⎪⎨ − x + 1, 0 < x ≤ 1
⎪
⎩ x − 1, 1 < x ≤ 2⋅
⎧ − x, − 2≤ x≤ 0
∴ g (x) = f (⏐ x⏐ ) + ⏐ f (x)⏐ = ⎪⎨ 0, 0< x≤ 1
⎪
⎩ 2x − 2, 1 < x ≤ 2.
—
We see that g (x) is differentiable V x ∈ ] − 2, 2 [, except possibly at x = 0 and 1.
lim g (0 − h) − g (0) lim g (− h) − g (0)
L g′ (0) = =
h →0 − h h →0 − h
lim h − 0
= = − 1,
h →0 − h
⎧ x− 1 , when x ≠ 1
⎪
Example 8 : Find f ′ (1) if f (x) = ⎨ 2x2 − 7x + 5
⎪
⎩ − 1 ⁄ 3, when x = 1.
lim f (1 + h) − f (1)
Solution : We have f ′ (1) =
h →0 h
=
lim ⎡ 1+ h− 1 ⎛ 1⎞ ⎤
− ⎜− ⎟
h → 0 ⎢⎣ 2 (1 + h) 2 − 7 (1 + h) + 5 ⎝ 3 ⎠ ⎥⎦
h/
lim 3h + 2 (1 + h) 2 − 7 (1 + h) + 5
=
h → 0 3h [2 (1 + h) 2 − 7 (1 + h) + 5]
lim 2h2 lim 2 2
= = = − ⋅
h → 0 3h (− 3h + 2h2) h → 0 − 9 + 6h 9
D-52 DIFFERENTIAL CALCULUS
⎛ 1 ⎞2 1
= 2 + ⎜x − π⎟ in π ≤ x < ∞.
⎝ 2 ⎠ 2 (Avadh 2009)
Solution : We shall test f (x) for continuity and differentiability at x = 0 and
π ⁄ 2. It is obviously continuous as well as differentiable at all other points.
(i) Continuity and differentiability of f (x) at x = 0.
We have f (0) = 1 + sin 0 = 1;
lim lim lim
f (0 + 0) = f (0 + h) = f (h) = (1 + sin h) = 1;
h →0 h →0 h →0
lim lim lim
and f (0 − 0) = f (0 − h) = f (− h) = 1 = 1.
h →0 h →0 h →0
Since f (0) = f (0 + 0) = f (0 − 0), f (x) is continuous at x = 0.
lim f (0 + h) − f (0) lim f (h) − f (0)
Now R f ′ (0) = =
h →0 h h →0 h
lim (1 + sin h) − (1 + sin 0) lim sin h
= = = 1;
h →0 h h →0 h
lim f (0 − h) − f(0) lim f (− h) − f (0)
and L f ′ (0) = =
h →0 − h h →0 − h
lim 1 − (1 + sin 0) lim 0 lim
= = = 0 = 0.
h →0 − h h →0 − h h →0
Since R f ′ (0) ≠ L f ′ (0), f (x) is not differentiable at x = 0.
1
(ii) Continuity and differentiability of f (x) at x = π.
2
⎛1 ⎞ ⎛1 1 ⎞2
We have f ⎜ π⎟ = 2 + ⎜ π − π⎟ = 2 ;
⎝2 ⎠ ⎝2 2 ⎠
2
⎛1 ⎞ lim ⎛ 1 ⎞ lim ⎡⎢ ⎧⎛ 1 ⎞ 1 ⎫
⎤
⎥
f ⎜ π + 0⎟ = f ⎜ π + h⎟ = ⎢ 2 + ⎨⎜ π + h⎟ − π ⎬ ⎥
⎝2 ⎠ h →0 ⎝ 2 ⎠ h → 0 ⎢⎣ ⎩⎝ 2 ⎠ 2 ⎭ ⎥
⎦
lim
= (2 + h2) = 2 ;
h →0
⎛1 ⎞ lim ⎛ 1 ⎞ lim ⎡ ⎛1 ⎞⎤
and f ⎜ π − 0⎟ = f ⎜ π − h⎟ = ⎢ 1 + sin ⎜ π − h ⎟ ⎥
⎝2 ⎠ h →0 ⎝ 2 ⎠ h →0 ⎣ ⎝2 ⎠⎦
lim
= (1 + cos h) = 1 + 1 = 2 .
h →0
⎛1 ⎞ ⎛1 ⎞ ⎛1 ⎞ 1
Since f ⎜ π ⎟ = f ⎜ π + 0⎟ = f ⎜ π − 0⎟ , f is continuous at x = π.
⎝2 ⎠ ⎝2 ⎠ ⎝2 ⎠ 2
⎛1 ⎞ ⎛1 ⎞
f ⎜ π + h⎟ − f ⎜ π ⎟
⎝ 2 ⎠ ⎝2 ⎠
⎛1 ⎞ lim
Now R f ′ ⎜ π ⎟ =
⎝2 ⎠ h →0 h
2 ⎛1 2
⎡ ⎧1 1 ⎫ ⎤ ⎡ 1 ⎞ ⎤
⎢ 2 + ⎨ π + h − π⎬ ⎥ − ⎢ 2 + ⎜ π − π⎟ ⎥
lim ⎣ ⎩2 2 ⎭ ⎦ ⎣ ⎝2 2 ⎠ ⎦
=
h →0 h
DIFFERENTIABILITY D-53
lim 2 + h2 − 2 lim
= = h= 0;
h →0 h h →0
⎛1 ⎞ ⎛1 ⎞ ⎛1 ⎞
f ⎜ π − h⎟ − f ⎜ π ⎟ 1 + sin ⎜ π − h⎟ − 2
⎝2 ⎠ ⎝2 ⎠ ⎝2 ⎠
⎛1 ⎞ lim lim
and L f ′ ⎜ π⎟ = =
⎝2 ⎠ h →0 − h h →0 − h
lim − 1 + cos h lim 1 − cos h lim 2 sin2 (h ⁄ 2)
= = =
h →0 − h h →0 h h →0 h
lim ⎡ sin (h ⁄ 2) ⎤
= ⎢ . sin (h ⁄ 2) ⎥ = 1 × 0 = 0.
h →0 ⎣ h ⁄ 2 ⎦
1
Since R f ′ (0) = L f ′ (0), f (x) is differentiable at x = π.
2
Example 10 : If f (x) = x2 sin (1 ⁄ x) , for x ≠ 0 and f (0) = 0 , then show that f (x) is
continuous and differentiable everywhere and that f ′ (0) = 0. A lso show that the function
f ′ (x) has a discontinuity of second kind at the origin. (Meerut 2006B; Kanpur 14)
lim 2 1 lim 2 1
Solution : We have f (0 + 0) = (0 + h) sin = h sin = 0;
h →0 0 + h h →0 h
lim lim lim
f (0 − 0) = f (0 − h) = f (− h) = (− h) 2 sin (− 1 ⁄ h)
h →0 h →0 h →0
lim 2 1
= − h sin = 0.
h →0 h
... f (0 + 0) = f (0 − 0) = f (0), so the function is continuous at x = 0.
lim f (0 + h) − f (0) lim f (h) − f (0)
Now R f ′ (0) = =
h →0 h h →0 h
lim h2 sin (1 ⁄ h) − 0 lim 1
= = h sin = 0;
h →0 h h →0 h
lim f (0 − h) − f (0) lim f (− h) − f (0)
and L f ′ (0) = =
h →0 − h h →0 − h
lim (− h) 2 sin (− 1 ⁄ h) − 0 lim 1
= = h sin = 0.
h →0 − h h →0 h
Thus R f ′ (0) = L f ′ (0) implies that f (x) is differentiable at x = 0 and f ′(0) = 0.
For all other values of x, f (x) is easily seen to be continuous and differentiable.
1 1
Now f ′ (x) = 2 x sin − cos at x ≠ 0 and f ′ (0) = 0.
x x
lim lim
∴ f ′ (0 + 0) = h → 0 f ′ (0 + h) = h → 0 f ′ (h)
lim ⎛ 1 1⎞
= ⎜ 2 h sin − cos ⎟ , which does not exist.
h →0 ⎝ h h⎠
Similarly it can be shown that f ′ (0 − 0) does not exist.
H ence f ′ is discontinuous at the origin. Since both the limits f ′ (0 − 0) and
f ′ (0 + 0) do not exist, therefore the discontinuity is of the second kind.
Example 11 : A function f is defined by f (x) = x p cos (1 ⁄ x), x ≠ 0; f (0) = 0.
What conditions should be im posed on p so that f m ay be
(i) continuous at x = 0 (ii) differentiable at x = 0 ?
Solution : We have
D-54 DIFFERENTIAL CALCULUS
lim lim ⎡
f (0 + 0) = f (0 + h) = (0 + h) p cos {1 ⁄ (0 + h)}⎤⎦
h →0 h →0 ⎣
lim p
= h cos (1 ⁄ h) ...(1)
h →0
lim lim ⎡
and f (0 − 0) = f (0 − h) = (0 − h) p cos {1 ⁄ (0 − h)}⎤⎦
h →0 h →0 ⎣
lim
= (− h) p cos (1 ⁄ h). …(2)
h →0
Now if the function f (x) is to be continuous at x = 0, then
f (0 + 0) = f (0) = 0 = f (0 − 0)
i.e., the limits given in (1) and (2) must both tend to zero.
This is possible only if p > 0, which is the required condition.
lim f (0 + h) − f (0) lim f (h) − f (0)
Now R f ′ (0) = =
h →0 h h →0 h
lim h p cos (1 ⁄ h) − 0 lim 1
= = h p − 1 cos …(3)
h →0 h h →0 h
lim f (0 − h) − f (0) lim (− h) p cos (− 1 ⁄ h) − 0
and L f ′ (0) = =
h →0 − h h →0 − h
lim
= − (− 1) p h p − 1 cos (1 ⁄ h). …(4)
h →0
Now if f ′ (x) exists at x = 0 then we must have R f ′ (0) = L f ′ (0) and this is
possible only if p − 1 > 0 i.e., p > 1 which gives Rf ′ (0) = 0 = L f ′ (0). H ence in order
that f is differentiable at x = 0, p must be greater than 1.
Example 12 : For a real num ber y, let [ y] denote the greatest integer less than or
tan (π [ x − π])
equal to y. Then if f (x) = , show that f ′ (x) exists for all x.
1 + [ x]2
Solution : From the definition of [ y], we see that [x − π] is an integer for all
values of x. Then π (x − π) is an integral multiple of π and so tan (π [x − π]) = 0 — V x.
Since [x] is an integer so 1 + [x]2 ≠ 0 for any x. Thus f (x) = 0 for all x i.e., f (x) is a
constant function and so it is continuous and differentiable i.e., f ′ (x) exists — V x and is
equal to zero.
Exa mple 13 : D e t e rm i n e t h e s e t o f a l l p o i n t s w h e re t h e f u n c t i o n
f (x) = x ⁄ (1 + ⏐ x⏐ ) is differentiable.
Solution : Since ⏐ x⏐ = x, x > 0, ⏐ x⏐ = − x, x < 0, ⏐ x⏐ = 0, x = 0,
x , x ,
∴ f (x) = x < 0; f (x) = 0, x = 0; f (x) = x > 0.
1− x 1+ x
lim lim lim h
We have f (0 + 0) = f (0 + h) = f (h) = = 0;
h →0 h →0 h →0 1 + h
lim lim lim − h
f (0 − 0) = f (0 − h) = f (− h) = = 0.
h →0 h →0 h →0 1 + h
Since f (0 + 0) = f (0) = f (0 − 0) = 0 so the function is continuous at x = 0.
lim f (0 − h) − f (0) lim f (− h) − f (0)
Further L f ′ (0) = =
h →0 − h h →0 − h
DIFFERENTIABILITY D-55
x = 1, 2 because the slopes at these points are different on the left and right hand sides.
To test it analytically, we write y = f (x). Then
f (x) = 3 − 2x when x ≤ 1
= 1 when 1 ≤ x ≤ 2
= 2x − 3 when x ≥ 2 .
This function is obviously continuous and differentiable at all points of the
interval [0, 3] except possibly at x = 1 and at x = 2 .
At x = 1, we have f (1) = 1;
lim lim
f (1 − 0) = [3 − 2 (1 − h)] = 1; f (1 + 0) = (1) = 1.
h →0 h →0
Since f (1 − 0) = f (1 + 0) = f (1), f is continuous at x = 1.
lim f (1 + h) − f (1) lim 1 − 1
Again R f ′ (1) = = = 0
h →0 h h →0 h
lim f (1 − h) − f (1) lim 3 − 2 (1 − h) − 1
and L f ′ (1) = = = − 2.
h →0 − h h →0 − h
Since R f ′ (1) ≠ L f ′ (1), f is not differentiable at x = 1.
At x = 2 , we have f (2) = 1;
lim lim
f (2 − 0) = (1) = 1 ; f (2 + 0) = [2 (2 + h) − 3] = 1.
h →0 h →0
Since f (2 − 0) = f (2 + 0) = f (2), f is continuous at x = 2.
lim f (2 + h) − f (2) lim 2 (2 + h) − 3 − 1
Again Rf ′ (2) = h → 0 = = 2
h h →0 h
lim f (2 − h) − f (2) lim 1 − 1
and L f ′ (2) = = = 0⋅
h →0 − h h →0 − h
Since Rf ′ (2) ≠ L f ′ (2), f is not differentiable at x = 2 .
lim ⎧ 1 ⎫
= ⎨ 1 + sin log h 2⎬ , which does not exist since sin log h 2
h →0 ⎩ 3 ⎭
oscillates between − 1 and 1 as h → 0.
⎧ 1 ⎫
(0 − h) ⎨ 1 + sin log (0 − h) 2⎬ − 0
⎩ 3 ⎭
lim
L f ′ (0) =
h →0 − h
lim ⎡ 1 ⎤
= ⎢ 1 + sin log h 2⎥ , which does not exist as above.
h →0 ⎣ 3 ⎦
H ence f has no differential coefficient at x = 0.
e1 ⁄ x − e− 1 ⁄ x ,
Example 17 : L et f (x) = x 1 ⁄ x x ≠ 0; f (0) = 0.
e + e− 1 ⁄ x
Show that f (x) is continuous but not derivable at x = 0.
(Meerut 2005; Purvanchal 07, 14; Kanpur 08; Bundelkhand 14)
Solution : We have f (0) = 0;
lim lim lim e1 ⁄ h − e− 1 ⁄ h
f (0 + 0) = f (0 + h) = f (h) = h
h →0 h →0 h →0 e1 ⁄ h + e− 1 ⁄ h
lim 1 − e− 2 / h ,
= h dividing the Nr. and Dr. by e1 ⁄ h
h → 0 1 + e− 2 / h
1− 0
= 0× = 0 × 1 = 0;
1+ 0
lim lim
and f (0 − 0) = f (0 − h) = f (− h)
h →0 h →0
lim e1 ⁄ − h − e− 1 ⁄ − h lim e− 1 ⁄ h − e1 ⁄ h
= − h 1⁄− h = − h − 1⁄h
h →0 e + e− 1 ⁄ − h h → 0 e + e1 ⁄ h
lim e− 2 / h − 1 0− 1
= − h − 2/h = 0× = 0.
h →0 e + 1 0+ 1
Since f (0 + 0) = f (0 − 0) = f (0), the function is continuous at x = 0.
lim f (0 + h) − f (0) lim f (h) − f (0)
Now R f ′ (0) = =
h →0 h h →0 h
lim ⎡⎢ e1 ⁄ h − e− 1 ⁄ h
/ h = hlim→ 0 11 −+ ee
⎤ − 2/h 1− 0
= ⎢h − 0⎥⎥ = = 1,
h → 0 ⎢⎣ e1 ⁄ h + e− 1 ⁄ h ⎥
⎦
− 2/h 1+ 0
lim f (0 − h) − f (0) lim f (− h) − f (0)
and L f ′ (0) = = h →0
h →0 − h − h
lim ⎡⎢ e− 1 ⁄ h − e1 ⁄ h
/ (− h)
⎤
⎥
= ⎢ (− h) − 1 ⁄ h − 0 ⎥
h → 0 ⎢⎣ e + e1 ⁄ h ⎥
⎦
lim e− 2 / h − 1 0 − 1
= = = − 1.
h →0 e− 2 / h + 1 0 + 1
Since R f ′ (0) ≠ L f ′ (0), the function is not derivable at x = 0.
2
Example 18 : L et f (x) = e− 1 ⁄ x sin (1 ⁄ x) when x ≠ 0 and f (0) = 0. Show that at
every point f has a differential coefficient and this is continuous at x = 0.
Solution : We test differentiability at x = 0.
D-58 DIFFERENTIAL CALCULUS
2
lim e− 1 ⁄ h sin (1 ⁄ h) − 0 lim sin (1 ⁄ h)
R f ′ (0) = =
h →0 h h → 0 he1 ⁄ h 2
lim sin (1 ⁄ h) lim sin (1 ⁄ h)
= =
h → 0 ⎧⎪ 1 1 ⎫⎪ h →0 1 1 1
h ⎨1 + 2 + 4
+ …⎬ h+ + ⋅ 3+ …
⎪⎩ h 2!h ⎪
⎭ h 2 ! h
a finite quantity lying between − 1 and + 1
= ∞ = 0.
Similarly L f ′ (0) = 0.
Since R f ′ (0) = L f ′ (0) = 0, hence the function f (x) is differentiable at x = 0 and
f ′ (0) = 0.
If x is any point other than zero, then
2 2
f ′ (x) = (2 ⁄ x3) e− 1 ⁄ x sin (1 ⁄ x) − (1 ⁄ x2) e− 1 ⁄ x cos (1 ⁄ x)
2
= {(2 ⁄ x) sin (1 ⁄ x) − cos (1 ⁄ x)} (1 ⁄ x2) (1 ⁄ e1 ⁄ x ) …(1)
lim lim ⎛ 2 1 1⎞ 1
Now f ′ (0 + 0) = f ′ (0 + h) = ⎜ sin − cos ⎟ ⋅
h →0 h →0 ⎝ h h h ⎠ h2 e1 ⁄ h 2
lim ⎛⎜ 2 sin (1 ⁄ h) cos (1 ⁄ h) ⎞⎟
= −
h → 0 ⎜⎜ h3 e1 ⁄ h 2 2 ⎟
h2 e1 ⁄ h ⎟
⎝ ⎠
lim ⎡ 2 sin (1 ⁄ h) cos (1 ⁄ h) ⎤
= ⎢ − ⎥
h →0 ⎢ 3 ⎛ 1 1 ⎞ 2 ⎛ 1 1 ⎞⎥
⎢ h ⎜1 + 2 + + … ⎟ h ⎜ 1 + + + …⎟⎥
⎣ ⎝ h 2 !. h4 ⎠ ⎝ h2 2 !. h4 ⎠⎦
some finite quantity some finite quantity
= ∞ − ∞ = 0⋅
Similarly f ′ (0 − 0) = 0. H ence f ′ is continuous at x = 0.
(c) Prove that if a function f (x) possesses a finite derivative in a closed interval
[a, b], then f (x) is continuous in [a, b].
Similarly, from (1), for all positive real numbers h such that c + h lies in [a, b], we
have
f (c + h) ≤ f (c).
By the same argument as above, we get
Rf ′ (c) ≤ 0. …(4)
Since f ′ (c) exists, hence, L f ′ (c) = f ′ (c) = R f ′ (c). …(5)
From (3), (4) and (5) we conclude that f ′ (c) = 0.
In the same manner we can consider the case M = f (a) ≠ m.
Note 1 : There may be more than one point like c at which f ′ (x) vanishes.
Note 2 : R olle’s theorem will not hold good
(i) if f (x) is discontinuous at some point in the interval a ≤ x ≤ b,
or (ii) if f ′ (x) does not exist at some point in the interval a < x < b,
or (iii) if f (a) ≠ f (b).
Note 3 : It can be seen that the conditions of R olle’s theorem are not necessary
fo r f ′ (x) t o va n ish a t so m e p o i n t in ] a, b [. F o r e xa m p le , f (x) = cos (1 ⁄ x) i s
discontinuous at x = 0 in the interval [ − 1, 2] but f ′ (x) vanishes at an infinite number
of points in the interval.
Geometrical interpretation of Rolle’s
Theorem :
Suppose the function f (x) is not constant
and satisfies the conditions of R olle’s theorem
in t h e in t e r val [a, b]. T hen it s ge omet rical
interpretation is that the curve representing the
graph of the function f must have a tangent
parallel to x-axis, at least at one point between
a and b.
Algebraical interpretation of Rolle’s Theorem :
R olle’s theorem leads to a very important result in the theory of equations, when
f (a) = f (b) = 0 and f : [a, b] → R is a polynomial function f (x). H ere a and b are the
roots of the equation f (x) = 0. Since a polynomial function f (x) is continuous and
differentiable at every point of its domain and we have taken f (a) = f (b), therefore, all
the three conditions of R olle’s theorem are satisfied and consequently there exists a
point c ∈ ] a, b [ such that f ′ (c) = 0 i.e., if a and b are any two roots of the polynom ial
equation f (x) = 0, then there exists at least one root of the equation f ′ (x) = 0 which lies
between a and b.
2
Now f ′ (x) = ⋅ [1 ⁄ (x − 1) 1 ⁄ 3]. We see that for x = 1, f ′ (x) is not finite while
3
x = 1 is a point of the open interval 0 < x < 2. Thus the second condition of R olle’s
theorem is not satisfied.
H e n c e t h e R o l l e ’s t h e o r e m i s n o t a p p l i c a b l e fo r t h e fu n c t i o n
f (x) = 2 + (x − 1) 2 ⁄ 3 in the interval [0, 2].
Example 2 : Discuss the applicability of Rolle’s theorem in the interval [− 1, 1] to
the function f (x) = ⏐ x⏐ .
Solution : G iven f (x) = ⏐ x⏐ . H ere f (− 1) = ⏐ − 1⏐ = 1, f (1) = ⏐ 1⏐ = 1, so
that f (− 1) = f (1).
Further the function f (x) is continuous throughout the closed interval [− 1, 1] but
it is not differentiable at x = 0 which is a point of the open interval ] − 1, 1 [. Thus the
second condition of R olle’s theorem is not satisfied. Hence the R olle’s theorem is not
applicable here.
Example 3 : A re the conditions of Rolle’s theorem satisfied in the case of the
following functions ?
(i) f (x) = x2 in 2 ≤ x ≤ 3, (ii) f (x) = cos (1 ⁄ x) in − 1 ≤ x ≤ 1,
(iii) f (x) = tan x in 0 ≤ x ≤ π.
Solution : (i) The function f (x) = x2 is continuous and differentiable in the
interval [2, 3]. Also f (2) = 4 and f (3) = 9, so that f (2) ≠ f (3).
Thus the first two conditions of R olle’s theorem are satisfied and the third
condition is not satisfied.
(ii) The function f (x) = cos (1 ⁄ x) is discontinuous at x = 0 and consequently is
not differentiable there. Thus the first two conditions of R olle’s theorem are not
satisfied.
H ere f (− 1) = cos (− 1) = cos 1 and f (1) = cos 1. Thus f (− 1) = f (1) i.e., the
third condition is satisfied.
(iii) The function f (x) = tan x is not continuous at x = π ⁄ 2 and consequently is
not differentiable there. Thus the first two conditions of R olle’s theorem are not
satisfied here.
Further f (0) = tan 0 = 0 and f (π) = tan π = 0. Thus f (0) = f (π) i.e., the third
condition is satisfied.
⎡ x2 + ab ⎤
Example 4 : Discuss the applicability of Rolle’s theorem to f (x) = log ⎢⎢⎢ ⎥
⎥,
⎥
⎣ (a + b) x ⎦
in the interval [a, b] , 0 < a < b. (Rohilkhand 2014)
⎡ a 2 + ab ⎤
Solution : H ere f (a) = log ⎢ ⎥ = log 1 = 0,
⎣ (a + b) a ⎦
⎡ b2 + ab ⎤
and f (b) = log ⎢⎢⎢ ⎥
⎥ = log 1 = 0.
⎥
⎣ (a + b) b ⎦
Thus f (a) = f (b) = 0.
lim f (x + h) − f (x)
Also R f ′ (x) =
h →0 h
lim 1 ⎡⎢ ⎧ (x + h) 2 + ab ⎫ ⎧ x2 + ab ⎫ ⎤
⎪ ⎪ ⎪ ⎪⎥
= ⎢ log ⎨ ⎬ − log ⎨ ⎬⎥
h →0 h ⎣⎢ ⎪ ⎪
⎩ (a + b) (x + h) ⎭
⎪ (a + b) x⎪ ⎥
⎩ ⎭⎦
lim 1 ⎡⎢ ⎧ 2 x h + h2 ⎫⎪ ⎧ h ⎫ ⎤⎥
= log ⎪⎨ 1 + 2 ⎬ − log ⎨ 1 + ⎬ ⎥
h → 0 h ⎢⎢⎣ ⎪
⎩ x + ab ⎭ ⎪ ⎩ x ⎭ ⎥⎦
lim 1 ⎡⎢ 2 x h + h2 h ⎤
= − + …⎥⎥ , …(1)
h → 0 h ⎢⎢⎣ x2 + ab x ⎥
⎦
⎡. . 1 ⎤
⎢ . log (1 + y) = y − y2 + …⎥
⎣ 2 ⎦
2x 1
= − ⋅
x2 + ab x
lim ⎡ f (x − h) − f (x) ⎤
Again L f ′ (x) = ⎢ ⎥
h →0 ⎣ − h ⎦
lim 1 ⎡ − 2hx + h 2 (− h) ⎤
= ⎢ − + …⎥⎥ , replacing h by − h in (1)
⎢
h → 0 (− h) ⎢ x + ab
⎣
2 x ⎥
⎦
2x 1
= − ⋅
x2
+ ab x
Since R f ′ (x) = L f ′ (x), f (x) is differentiable for all values of x in [a, b]. This
implies that f (x) is also continuous for all values of x in [a, b]. Thus all the three
conditions of R olle’s theorem are satisfied. Hence f ′ (x) = 0 for at least one value of x
in the open interval ]a, b[.
2x 1
Now f ′ (x) = 0 ⇒ 2 − = 0 or 2 x2 − (x2 + ab) = 0
x + ab x
or x2 = ab or x = √(ab),
which being the geometric mean of a and b lies in the open interval ] a, b[. H ence the
R olle’s theorem is verified.
Remark : In this question to find f ′ (x), we can also proceed as follows :
We have f (x) = log (x2 + ab) − log (a + b) − log x.
2x 1
∴ f ′ (x) = 2 − ⋅
x + ab x
O bviously f ′ (x) exists for all values of x in [a, b] .
Example 5 : Verify Rolle’s theorem in the case of the functions
(i) f (x) = 2x3 + x2 − 4x − 2,
(ii) f (x) = sin x in [0, π],
(iii) f (x) = (x − a) m (x − b) n , where m and n are + ive integers, and x lies in the
interval [a, b].
Solution : (i) Since f (x) is a rational integral function of x, therefore, it is
continuous and differentiable for all real values of x. Thus the first two conditions of
R olle’s theorem are satisfied in any interval.
H ere f (x) = 0 gives 2x3 + x2 − 4x − 2 = 0
1
or (x2 − 2) (2x + 1) = 0 i.e., x = ± √2 , − ⋅
2
⎛ 1⎞
Thus f (√2) = f (− √2) = f ⎜ − ⎟ = 0.
⎝ 2⎠
D-64 DIFFERENTIAL CALCULUS
If we take the interval [− √2 , √2], then all the three conditions of R olle’s theorem
are satisfied in this interval. Consequently there is at least one value of x in the open
interval ] − √2 , √2[ for which f ′ (x) = 0.
Now f ′ (x) = 0 ⇒ 6 x2 + 2 x − 4 = 0 ⇒ 3x2 + x − 2 = 0
or (3x − 2) (x + 1) = 0 or x = − 1, 2 ⁄ 3 i.e., f ′ (− 1) = f ′ (2 ⁄ 3) = 0.
Since both the points x = − 1 and x = 2 ⁄ 3 lie in the open interval ] − √2, √2[,
R olle’s theorem is verified.
(ii) The function f (x) = sin x is continuous and differentiable in [0, π].
Also f (0) = 0 = f (π). Thus all t he three condt ions of R olle’s theorem are
satisfied. H ence f ′ (x) = 0 for at least one value of x in the open interval ]0, π [.
π 3π 5π
Now f ′ (x) = 0 ⇒ cos x = 0 ⇒ x = ± , ± ,± ,….
2 2 2
Since x = π ⁄ 2 lies in the open interval ]0, π[, the R olle’s theorem is verified.
(iii) We have f (x) = (x − a) m (x − b) n .
As m and n are positive integers, (x − a) m and (x − b) n are polynomials in x on
being expanded by binomial theorem. H ence f (x) is also a polynomial in x. Consequently
f (x) is continuous and differentiable in the closed interval [a, b]. Also f (a) = f (b) = 0.
Thus all the three conditions of R olle’s theorem are satisfied so that there is at
least one value of x in the open interval ]a, b[ where f ′ (x) = 0.
Now f ′ (x) = (x − a) m . n (x − b) n − 1 + m (x − a) m − 1 (x − b) n .
Solving the equation f ′ (x) = 0, we get x = a, b, (na + mb) / (m + n).
O ut of these values the value (na + mb) ⁄ (m + n) is a point which lies in the open
interval ]a, b[, since it divides the interval ]a, b[ internally in the ratio m : n. H ence the
R olle’s theorem is verified.
Example 6 : Verify Rolle’s theorem for
f (x) = x (x + 3) e− x ⁄ 2 in [− 3, 0].
Solution : We have f (x) = x (x + 3) e− x ⁄ 2.
⎛ 1⎞
∴ f ′ (x) = (2x + 3) e− x ⁄ 2 + (x2 + 3x) e− x ⁄ 2. ⎜ − ⎟
⎝ 2⎠
⎡ ⎤
1 1⎛ 2
= e− x ⁄ 2 2 x + 3 −
⎢ (x2 + 3x) = −
⎥ x − x − 6⎞⎠ e− x ⁄ 2,
⎣ 2 ⎦ 2⎝
which exists for every value of x in the interval [− 3, 0]. H ence f (x) is differentiable and
so also continuous in the interval [− 3, 0]. Also f (− 3) = f (0) = 0.
Thus all the three conditions of R olle’s theorem are satisfied. So f ′ (x) = 0 for at
least one value of x lying in the open interval ]− 3, 0[.
1
Now f ′ (x) = 0 ⇒ − (x2 − x − 6) e− x ⁄ 2 = 0 or x2 − x − 6 = 0
2
or (x − 3) (x + 2) = 0 or x = 3, − 2 .
Since the value x = − 2 lies in the open interval ] − 3, 0[, the R olle’s theorem is
verified.
t h eo rem o n [a, c]. Consequently t here exists a real number d bet ween a a n d c
i.e., a < d < c such that
f (c) − f (a) = (c − a) f ′ (d).
But by hypothesis f ′ (x) = 0 throughout the interval ]a, b[, therefore, in particular
f ′ (d) = 0 and hence f (c) − f (a) = 0 or f (c) = f (a). Since c is any point of ]a, b],
t herefo re, it gives t hat f (x) = f (a) — V x in ]a, b]. Th u s f (x) has a const ant value
throughout [a, b].
Theorem 2 : If f (x) and φ (x) are functions continuous on [a, b] and differentiable
on ]a, b[ and if f ′ (x) = φ ′ (x) throughout the interval ]a, b[, then f (x) and φ (x) differ only
by a constant.
Proof : Consider t he fu nction F (x) = f (x) − φ (x). Throughout t he interval
]a, b[, we have
F ′ (x) = f ′ (x) − φ ′ (x) = 0, because f ′ (x) = φ ′ (x).
Consequently, from theorem 1, we get
F (x) = constant or f (x) − φ (x) = constant.
Theorem 3 : If f ′ (x) = k for each point x of [a, b], k being a constant, then
f (x) = k x + C —V x ∈ [a, b] , where C is a constant.
Proof : Consider the interval [a, x] such that [a, x] lies in the interval [a, b] i.e.,
[a, x] ⊂ [a, b]. Since f ′ (x) exists —
V x ∈ [a, b], f is differentiable on [a, b] and hence on
[a, x] and consequently continuous on [a, x]. Thus f satisfies all the conditions of
Lagrange’s mean value theorem on [a, x] and hence there is a point c ∈ ]a, x[ such that
f (x) − f (a) = (x − a) f ′ (c).
But by hypothesis f ′ (x) = k — V x ∈ [a, b], therefore, in particular f ′ (c) = k as
a < c < x < b i.e., a < c < b.
H ence f (x) − f (a) = (x − a) k or f (x) = k x + f (a) − ak
or f (x) = k x + C where C = f (a) − ak is a constant.
Theorem 4 : If f is continuous on [a, b] and f ′ (x) ≥ 0 in ]a, b[ , then f is increasing
in [a, b].
Proof : Let x1 and x2 be any two distinct points of [a, b] such that x1 < x2 . Then
f satisfies the conditions of the Lagrange’s mean value theorem in [x1, x2]. Consequently
there exists a number c such that x1 < c < x2 , and f (x2) − f (x1) = (x2 − x1) f ′ (c).
Now x2 − x1 > 0 and f ′ (c) ≥ 0 (as f ′ (x) ≥ 0 — V x ∈ ]a, b[ and c is a point of
]a, b[), therefore
f (x2) − f (x1) ≥ 0 i.e., f (x1) ≤ f (x2).
Thus x1 < x2 ⇒ f (x1) ≤ f (x2) —
V x1, x2 ∈ [a, b].
H ence f is an increasing function in the interval [a, b].
Similarly, we can prove that if f ′ (x) ≤ 0 in ]a, b[, then f is decreasing in [a, b].
Corollary : If f is continuous on [a, b] , then f is strictly increasing or strictly
decreasing on [a, b] according as
f ′ (x) > 0 or < 0 in ]a, b[⋅
D-68 DIFFERENTIAL CALCULUS
Example 3 : Com pute the value of θ in the first m ean value theorem
f (x + h) = f (x) + hf ′ (x + θh), if f (x) = ax2 + bx + c.
Solution : H ere f (x) = ax2 + bx + c.
∴ f (x + h) = a (x + h) 2 + b (x+ h) + c,
f ′ (x) = 2ax + b, f ′ (x + θh) = 2a (x + θh) + b.
Substituting all these values in the Lagrange’s mean value theorem, we get
a (x + h) 2 + b (x + h) + c = ax2 + bx + c + h [2a (x + θh) + b] …(1)
The relation (1) being identically true for all values of x, hence when x → 0, we have
ah2 + bh + c = c + h [2aθ h + b]
or ah2 = 2aθh2 or θ = 1 ⁄ 2.
Example 4 : A function f (x) is continuous in the closed interval [0, 1] and
differentiable in the open interval ]0, 1[ , prove that
f ′ (x1) = f (1) − f (0), where 0 < x1 < 1.
Solution : H ere a = 0, b = 1 so that
f (b) − f (a) f (1) − f (0)
= = f (1) − f (0).
b− a 1− 0
If we take c = x1, and substitute these values in the result of Lagrange’s mean value
theorem, we get
f (1) − f (0) = f ′ (x1) where 0 < x1 < 1.
This is a particular case of Lagrange’s mean value theorem. Students can give an
independent proof of this.
Example 5 : Separate the intervals in which the polynom ial
2x3 − 15x2 + 36x + 1 is increasing or decreasing.
Solution : We have f (x) = 2x3 − 15x2 + 36x + 1.
∴ f ′ (x) = 6x2 − 30x + 36 = 6 (x − 2) (x − 3).
Now f ′ (x) > 0 for x < 2 or for x > 3,
f ′ (x) < 0 for 2 < x < 3, and f ′ (x) = 0 for x = 2, 3.
Thus f ′ (x) is + ive in the intervals ] − ∞ , 2[ and ]3, ∞ [ and negative in the interval
]2, 3[.
H ence f is mo no to nically increasing in the intervals ] − ∞ , 2 ], [ 3, ∞ [ an d
monotonically decreasing in the interval [2, 3].
x
Example 6 : Show that < log (1 + x) < x for x > 0. (Bundelkhand 2011)
1+ x
x
Solution : Let f (x) = log (1 + x) − ⋅ ∴ f (0) = 0.
1+ x
1 1 . (1 + x) − x . 1 1 1 x
Then f ′ (x) = − = − = ⋅
1+ x (1 + x) 2 1 + x (1 + x) 2 (1 + x) 2
We observe that f ′ (x) > 0 for x > 0. H ence f (x) is monotonically increasing in the
interval [0, ∞ [. Therefore
⎡ x ⎤
f (x) > f (0) for x > 0 i.e., ⎢ log (1 + x) − ⎥ > 0 for x > 0
⎣ 1 + x⎦
x
i.e., log (1 + x) > for x > 0. …(1)
1+ x
DIFFERENTIABILITY D-71
1. State Lagrange’s mean value theorem. Test if Lagrange’s mean value theorem
holds for the function f (x) = ⏐ x⏐ in the interval [− 1, 1].
(Kanpur 2010; Rohilkhand 13B)
2. If f (x) = 1 ⁄ x in [− 1, 1], will the Lagrange’s mean value theorem be applicable to
f (x) ? (Meerut 2012B)
3. Verify Lagrange’s mean value theorem for the function
f : [− 1, 1] → R given by f (x) = x3.
1
4. Find ‘c’ of the mean value theorem, if f (x) = x (x − 1) (x − 2) ; a = 0, b = ⋅
2
5. Find ‘c’ of mean value theorem when
(i) f (x) = x3 − 3x − 2 in [− 2 , 3] (ii) f (x) = 2 x2 + 3 x + 4 in [1, 2]
(iii) f (x) = x (x − 1) in [1, 2] (Meerut 2013B)
⎛ 11 13 ⎞
2
(iv) f (x) = x − 3x − 1 in ⎜⎝ − , ⎟⋅
7 7⎠
6. State the conditions for the validity of the formula
f (x + h) = f (x) + h f ′ (x + θh)
and investigate how far these conditions are satisfied and whether the result is
true, when
f (x) = x sin (1 ⁄ x) (being defined to be zero at x = 0) and x < 0 < x + h.
7. (a) Show that x3 − 3x2 + 3x + 2 is monotonically increasing in every interval.
2x
(b) Show that log (1 + x) − is increasing when x > 0.
2+ x
8. Determine the intervals in which the function (x4 + 6 x3 + 17x2 + 32 x + 32) e− x
is increasing or decreasing.
9. U se the function f (x) = x1 ⁄ x, x > 0 to determine the bigger of the two numbers
e π and π e.
10. Show that the set of all x for which log (1 + x) ≤ x is equal to [0, ∞ [.
11. U se Lagrange’s mean value theorem to prove that
1 + x < e x < 1 + xe x — V x > 0.
DIFFERENTIABILITY D-73
1. Th e mea n value t h eore m doe s not hold since t he given funct ion is not
differentiable at x = 0.
√21
2. not applicable. 4. 1 − ⋅
6
5. (i) ± √(7 ⁄ 3). (ii) 3 ⁄ 2. (iii) 3 ⁄ 2. (iv) 1 ⁄ 7.
6. Con dition of differen tiabilit y is n ot satisfied in x < 0 < x + h sin ce f (x) i s
non-differentiable at x = 0.
8. Increasing in the intervals [− 2, − 1] and [0, 1] and decreasing in the intervals
] − ∞ , − 2], [− 1, 0] and [1, ∞ [ .
9. e π is bigger than π e.
14. R oot of the equation sin c − (8c ⁄ π 2) = 0 in the open interval ] π ⁄ 6, π ⁄ 2 [.
h2 hn − 1 (n − 1) A n
Now φ (a) = f (a) + h f ′ (a) + f ′′ (a) + … + f (a) + h ,
2! (n − 1) ! n!
and φ (a + h) = f (a + h).
H ence A is given by
h2
f (a + h) = f (a) + hf ′ (a) + f ′′ (a) + …
2!
hn − 1 (n − 1) hn
+ f (a) + A. …(1)
(n − 1) ! n!
Now, by hypothesis, all the functions
f (x), f ′ (x), f ′′ (x),……, f (n − 1) (x)
are continuous in the closed interval [a, a + h] and differentiable in the open interval
]a, a + h[.
Further (a + h − x), (a + h − x) 2 ⁄ 2 !, …, (a + h − x) n ⁄ n !, all being polynomials,
are continuous in the closed interval [a, a + h] and differentiable in the open interval
]a, a + h[. Also A is a constant.
∴ φ (x) is continuous in the closed interval [a, a + h] and differentiable in the
open interval ]a, a+ h[.
By our choice of A , φ (a) = φ (a + h). Hence φ (x) satisfies all the conditions of
R olle’s theorem.
Consequently φ ′ (a + θh) = 0, where 0 < θ < 1.
Now φ ′ (x) = f ′ (x) − f ′ (x) + (a + h − x) f ′′ (x) − (a + h − x) f ′′ (x)
(a + h − x) n − 1 (n) A
+ …+ f (x) − (a + h − x) n − 1
(n − 1) ! (n − 1) !
(a + h − x) n − 1 (n)
= [ f (x) − A] ,
(n − 1) !
since other terms cancel in pairs.
∴ φ ′ (a + θh) = 0 gives
[a + h − (a + θh)]n − 1
[ f (n) (a + θh) − A] = 0
(n − 1) !
hn − 1 (1 − θ) n − 1 (n)
or [ f (a + θh) − A] = 0
(n − 1) !
h2 hn − 1
f (a + h) = f (a) + h f ′ (a) + f ′′ (a) + …+ f (n − 1) (a)
2! (n − 1) !
hn
+ (1 − θ) n − 1 f (n) (a + θh).
(n − 1) !
hn
The (n + 1)th term (1 − θ) n − 1 f (n) (a + θh) is called Cauchy’s form of
(n − 1) !
remainder after n terms in the Taylor’s expansion of f (a + h) in ascending integral
powers of h.
Corollary : (Maclaurin’s development with Cauchy’s form of remainder) :
If we change a to 0 and h to x in the above result, we get
x2 xn − 1
f (x) = f (0) + x f ′(0) + f ′′ (0) + … + f (n − 1) (0)
2! (n − 1) !
xn
+ (1 − θ) n − 1 f (n) (θx),
(n − 1) !
which is Maclaurin’s theorem with Cauchy’s form of remainder. The (n + 1)th term
xn
(1 − θ) n − 1 f (n) (θ x) is known as Cauchy’s form of remainder after n terms in
(n − 1) !
Maclaurin’s development of f (x) in the interval [0, x].
(i) Expansion of e x :
Let f (x) = e x.
Then f (n) (x) = e x — V n ∈ N,—
V x ∈ R so that
f (n) (0) = e0 = 1 —
V n ∈ N.
Now Maclaurin’s expansion of f (x) with Lagrange’s form of remainder is
x2 x n − 1 (n − 1)
f (x) = f (0) + x f ′ (0) + f ′′(0) + …+ f (0) + R n
2! (n − 1) !
x n (n)
where Rn = f (θx), 0 < θ < 1.
n!
lim lim x n (n) lim x n θ x
Now R = f (θx) = e
n →∞ n n →∞ n ! n →∞ n !
lim x n ⎡⎢ . . lim x n ⎤
= eθx = e θ x × 0 = 0. ⎢⎢ . = 0⎥⎥⎥
n →∞ n ! ⎣ n → ∞ n! ⎦
Thus f (n) (x) exists in [0, x] for each n ∈ N and R n → 0 as n → ∞ i.e., all the
conditions of Maclaurin’s series expansion are satisfied.
H ence —V x ∈ R the expansion of e x is
x2 x3 xn
e x = 1+ x + + + …+ + …
2! 3! n!
Case I : 0 ≤ x ≤ 1 .
Since 0 ≤ x ≤ 1 and 0 < θ < 1, therefore x < 1 + θx
lim ⎛ x ⎞ n lim (− 1) n − 1
and hence ⎜ ⎟ = 0. Also = 0.
n → ∞ ⎝ 1 + θ x⎠ n →∞ n
D-78 DIFFERENTIAL CALCULUS
lim
Thus, in this case, R n = 0.
n →∞
Case II : − 1 < x < 0 .
In this case it will not be convenient to show that Lagrange’s form of remainder
R n → 0 as n → ∞ because x ⁄ (1 + θx) may not be numerically less than unity. Therefore
we use the Cauchy’s form of remainder. We have
xn
Rn = (1 − θ) n − 1f (n) (θx)
(n − 1) !
xn
= (1 − θ) n − 1(− 1) n − 1(n − 1) ! (1 + θx) − n
(n − 1) !
⎛ 1− θ⎞n− 1 1 ,
= (− 1) n − 1 xn ⎜ ⎟ ⋅ (0 < θ < 1).
⎝ 1 + θx ⎠ 1 + θx
Now as 0 < θ < 1 and − 1 < x < 0 , we have
1− θ lim ⎛ 1 − θ ⎞ n − 1
0< < 1, so that ⎜ ⎟ = 0.
1 + θx n → ∞ ⎝ 1 + θx ⎠
lim
Also, x n = 0 as − 1 < x < 0.
n →∞
Thus, in this case also R n → 0 as n → ∞ .
H ence, f (x) satisfies all the conditions of Maclaurin’s series expansion for
− 1 < x ≤ 1.
Therefore, for − 1< x ≤ 1, we get
x2 x3 x4 xn
log e (1 + x) = x − + − + … + (− 1) n − 1 + …
2 3 4 n
(v) Expansion of (1 + x)m :
Let f (x) = (1 + x) m , —
V x∈ R.
Then f (n) (x) = m(m − 1) (m − 2) … (m − n + 1) (1 + x) m − n — V n∈ N.
Now we consider two cases :
Case I : If m is a positive integer.
In this case, we notice that for n > m , f (n) (x) = 0. So all the terms after the
(m + 1)th term vanish and so the expansion consists of finite number of terms in the
form
x2 x m (m)
f (x) = f (0) + x f ′(0) + f ′′(0) + … + f (0).
2! m!
Case II : If m is a fraction or a negative integer.
In this case, let ⏐ x ⏐ < 1.
We have
f (n) (x) = m (m − 1) (m − 2) … (m − n + 1) (1 + x) m − n , x ≠ − 1.
H ere, we use Maclaurin’s expansion with Cauchy’s form of remainder. Thus, we
have
x2 x n − 1 (n − 1)
f (x) = f (0) + x f ′(0) + f ′′(0) + … + f (0) + R n
2! (n − 1) !
DIFFERENTIABILITY D-79
xn
where Rn = (1 − θ) n − 1f (n) (θ x), 0 < θ < 1
(n − 1) !
xn
= (1 − θ) n − 1. m(m − 1) … (m − n + 1) (1 + θ x) m − n
(n − 1) !
⎛ 1− θ ⎞n− 1 m(m − 1) … (m − n + 1)
= ⎜ ⎟ . (1 + θ x) m − 1 ⋅ ⋅ x n.
⎝ 1 + θ x⎠ (n − 1)!
m(m − 1) … (m − n + 1) an + 1 m − n
If an = ⋅ x n , then = ⋅ x
(n − 1)! an n
(on simplification)
a
lim n + 1 lim m⎛ ⎞
∴ = ⎜ − 1⎟ x = (0 − 1) x = − x.
n → ∞ an n →∞ ⎝ n ⎠
lim
This gives a = 0, since ⏐ − x⏐ = ⏐ x⏐ < 1.
n →∞ n
Further 0 < θ < 1 ⇒ 0 < 1 − θ < 1 + θ x
n− 1
1− θ lim ⎛ 1 − θ ⎞
so that < 1 which gives ⎜ ⎟ = 0.
1 + θx n → ∞ ⎝ 1 + θ x⎠
lim
H ence in this case R = 0. Thus f (x) satisfies the conditions of Maclaurin’s
n →∞ n
series expansion.
Therefore for − 1 < x < 1 , we have
m(m − 1) m(m − 1)(m − 2)
(1 + x) m = 1 + mx + ⋅ x2 + ⋅ x3 + …
2! 3!
x2
Example 1 : If f (x) = f (0) + x f ′ (0) + f ′′ (θx), …(1)
2!
find the value of θ as x → 1, f (x) being (1 − x) 5 ⁄ 2.
Solution : H ere f (x) = (1 − x) 5 ⁄ 2.
5 15
∴ f ′ (x) = − (1 − x) 3 ⁄ 2 and f ′′ (x) = (1 − x) 1 ⁄ 2.
2 4
Example 2 : Show that the num ber θ which occurs in the Taylor’s theorem with
L agrange’s form of rem ainder after n terms approaches the lim it 1 ⁄ (n + 1) as h → 0
provided that f (n + 1) (x) is continuous and different from zero at x = a.
Solution : Applying Taylor’s theorem with Lagrange’s form of remainder after
n terms and (n + 1) terms successively, we get for θ, θ′ ∈ ]0, 1[,
hn − 1 (n − 1) hn (n)
f (a + h) = f (a) + h f ′ (a) + … + f (a) + f (a + θ h),
(n − 1) ! n!
hn (n) hn + 1 (n + 1)
and f (a + h) = f (a) + h f ′ (a) + … + f (a) + f (a + θ′ h).
n! (n + 1) !
Subtracting these, we have
hn f (n) (a) hn + 1 (n + 1) hn (n)
+ f (a + θ′h) = f (a + θh)
n! (n + 1) ! n!
h
or f (n) (a + θh) − f (n) (a) = f (n + 1) (a + θ′h) …(1)
n+ 1
Applying Lagrange’s mean value theorem to the function f (n) (x) in the interval
[a, a + θh], we get
f (n) (a + θh) − f (n) (a) = θhf (n + 1) (a + θθ′′h), 0 < θ′′ < 1. …(2)
From (1) and (2), we have
h
θhf (n + 1) (a + θθ′′h) = f (n + 1) (a + θ′h)
n+ 1
1 f (n + 1) (a + θ′h)
or θ= ⋅
n + 1 f (n + 1) (a + θθ′′h)
lim 1 f (n + 1) (a) 1
∴ θ= = , provided f (n + 1) (a) ≠ 0.
h →0 n + 1 f (n + 1) (a) n + 1
Example 3 : Assuming the derivatives which occur are continuous, apply the m ean
value theorem to prove that
φ ′ (x) = F ′ { f (x)} f ′ (x) where φ (x) = F { f (x)}.
Solution : Let f (x) = t so that φ (x) = F (t).
lim φ (x + h) − φ (x) lim F { f (x + h)} − F { f (x)}
Now φ ′ (x) = =
h →0 h h →0 h
lim F { f (x) + h f ′ (x + θ1 h)} − F{ f (x)}
= , (0 < θ1 < 1)
h →0 h
[... f (x + h) = f (x) + h f ′ (x + θ1 h), by mean value theorem]
lim F (t + H) − F (t)
= , where H = h f ′ (x + θ1h)
h →0 h
lim HF ′ (t + θ2 H)
=
h →0 h
. .
[ . F(t + H) = F (t) + HF ′ (t + θ2 H), by mean value theorem]
lim h f ′ (x + θ1 h) F ′ [t + θ2 h f ′(x + θ1 h)]
=
h →0 h
= f ′ (x) F ′ (t) = F ′ { f (x)} f ′ (x).
Note : This example gives an alternative proof of the chain rule.
DIFFERENTIABILITY D-81
h2
1. If f (x + h) = f (x) + h f ′ (x) + f ′′ (x + θh),
2!
find the value of θ as x → a, f (x) being (x − a) 5 ⁄ 2.
h2
2. Find θ, if f (x + h) = f (x) + hf ′ (x) +
f ′′ (x + θh), 0 < θ < 1, and
2!
(i) f (x) = ax3 + bx2 + c x + d (ii) f (x) = x3.
3. Show that ‘θ’ (which occurs in the Lagrange’s mean value theorem) approaches
1
the limit as ‘h ’ approaches zero provided that f ′′ (a) is not zero. It is assumed
2
that f ′′ (x) is continuous.
4. Show that the number θ which occurs in the Taylor’s theorem with Lagrange’s
form of remainder after n terms approaches the limit 1 ⁄ (n + 1) as h → 0 provided
that f (n + 1) (x) is continuous and different from zero at x = a.
64 1 1
1. θ= ⋅ 2. (i) θ = ⋅ (ii) θ= ⋅
225 3 3
(iii) g′(x) ≠ 0 for any point of the open interval ]a, b[, then there exists at least one
value c of x in the open interval ]a, b[, such that
f (b) − f (a) f ′(c)
= ⋅
… g′(c)
11. If f is continuous in [a, b] and f ′(x) ≥ 0 in ]a, b[, then f is … in [a, b].
12. If f (x) = sin x, then
lim f (x + h) − f (x)
= ….
h →0 h
Multiple Choice Questions:
Indicate the correct answer for each question by writing the corresponding letter from
(a), (b), (c) and (d).
13. The function f (x) = | x − 1| is not differentiable at
(a) x = 0 (b) x = − 1
(c) x = 1 (d) x = 2
14. The function f (x) = | x + 3| is not differentiable at
(a) x = 3 (b) x = − 3
(c) x = 0 (d) x = 1
15. A function f (x) is differentiable at x = a if
(a) R f ′(a) = L f ′(a)
(b) R f ′(a) = 0
(c) L f ′(a) = 0
(d) R f ′(a) ≠ L f ′(a)
16. A function φ (x) is defined as follows :
φ (x) = 1 + x if x ≤ 2
φ (x) = 5 − x if x > 2 .
Then
(a) φ (x) is continuous but not differentiable at x = 2
(b) φ (x) is differentiable at every point of R
(c) φ (x) is neither continuous nor differentiable at x = 2
(d) φ (x) is differentiable at x = 2 but is not continuous at x = 2 .
DIFFERENTIABILITY D-83
17. O ut of the following four functions tell the function for which the conditions of
R olle’s theorem are satisfied.
(a) f (x) = | x | in [− 1, 1]
(b) f (x) = x2 in 2 ≤ x ≤ 3
(c) f (x) = sin x in [0, π]
(d) f (x) = tan x in 0 ≤ x ≤ π.
18. The function f (x) = sin x is increasing in the interval
⎡ π⎤
(a) [0, π] (b) ⎢ 0, ⎥
⎣ 2⎦
⎡ π 3π ⎤ ⎡π ⎤
(c) ⎢⎣ , ⎥⎦ (d) ⎢ , π ⎥
4 4 ⎣ 2 ⎦
19. The value of ‘c ’ of Lagrange’s mean value theorem for f (x) = x (x − 1) in [1, 2] is
given by
5 3
(a) c = (b) c=
4 2
7 11
(c) c = (d) c=
4 6
20. The value of ‘c ’ of R olle’s theorem for the function f (x) = e x sin x in [0, π] is given
by
3π
(a) c =
4
π
(b) c =
4
π
(c) c =
2
5π
(d) c =
6
True or False:
Write ‘T ’ for true and ‘F ’ for false statem ent.
21. If a function f (x) is continuous at x = a, it must also be differentiable at x = a.
22. If a function f (x) is differentiable at x = a, it must be continuous at x = a.
23. If a function f (x) is differentiable at x = a, it may or may not be continuous at
x = a.
24. The function f (x) = | x | is differentiable at every point of R.
25. R olle’s theorem is applicable for f (x) = sin x in [0, 2 π].
26. R olle’s theorem is applicable for f (x) = | x| in [− 1, 1].
27. Lagrange’s mean value theorem is applicable for f (x) = | x| in [− 1, 1].
⎡ π π⎤
28. The function f (x) = sin x is increasing in ⎢ − , ⎥ ⋅
⎣ 2 2⎦
29. If a + b + c = 0, then the quadratic equation 3ax2 + 2 bx + c = 0 has no root in
]0, 1[.
30. If f is continuous on [a, b] and f ′(x) ≤ 0 in ]a, b[, then f is increasing in [a, b].
31. The function f (x) = 2 x3 − 15 x2 + 36 x + 1 is decreasing in the interval [2 , 3].
32. Let f (x) = | x| + | x − 1| . Then Rf ′(0) = 0.
33. R olle’s theorem is not applicable for the function f (x) = x (x + 2) e − x ⁄ 2 in
[− 2 , 0].
D-84 DIFFERENTIAL CALCULUS
34. The value of ‘c ’ of Lagrange’s mean value theorem for the function
5
f (x) = 2 x2 + 3 x + 4 in [1, 2] is given by c = ⋅
4
lim f (x + h) − f (x)
35. If f (x) = x n , then = n x n− 1.
h →0 h
d n
Thus x = nx n − 1.
dx
d 4
Illustration 1 : (x ) = 4x4 − 1 = 4x3.
dx
d ⎛⎜ 1 ⎞⎟ d − 1⁄3 1 1
Illustration 2 : = (x ) = − x− 4⁄3 = − ⋅
dx ⎜⎜ x1 ⁄ 3 ⎟⎟ dx 3 3x 4 ⁄ 3
⎝ ⎠
d x lim e x + δx − e x
We have e = δx → 0 , by definition
dx δx
lim e x eδx − e x lim e x (e δx − 1)
= δx → 0 = δx → 0
δx δx
⎡ δx (δx) 2 (δx) 3 ⎤
ex ⎢ 1 + + + + ..... − 1⎥
⎣ 1! 2! 3! ⎦
lim
= δx → 0
δx
⎡ δx (δx) 2 (δx) 3 ⎤
ex ⎢ + + + ....⎥
⎣1 ! 2! 3! ⎦
lim
= δx → 0
δx
lim ⎡ δx (δx) 2 ⎤
= δx → 0 e x ⎢ 1 + + + …⎥ = e x.
⎣ 2! 3! ⎦
d x
Thus (e ) = e x.
dx
D-88 DIFFERENTIAL CALCULUS
lim c − c
= δx → 0 , since f (x) = c for every value of x
δx
lim 0
= δx → 0 = 0.
δx
Thus, the differential coefficient of a constant is zero.
1 ⎛ 4⎞ − 7 ⁄ 3 2
= ⎜− ⎟ x = − ⋅
2 ⎝ 3⎠ 3x7 ⁄ 3
d x d d
Illustration 1 : (e cos x) = e x cos x + cos x e x
dx dx dx
= − e x . sin x + cos x . e x = e x (cos x − sin x).
d 3 d d
Illustration 2 : (x log x) = x3 (log x) + log x x3
dx dx dx
= x3 . (1 ⁄ x) + (log x) . 3x2 = x2 (1 + 3 log x).
f1 (x)
Let f (x) = ⋅
f2 (x)
f1 (x + δx)
Then f (x + δx) = ⋅
f2 (x + δx)
Therefore, by definition
DIFFERENTIATION D-91
f1 (x + δx) f1 (x)
−
d lim f (x + δx) − f (x) lim f2 (x + δx) f2 (x)
f (x) = δx → 0 = δx → 0
dx δx δx
f1 (x + δx) f1 (x) f1 (x) f1 (x)
− − +
lim f2 (x + δx) f2 (x + δx) f2 (x) f2 (x + δx)
= δx → 0
δx
⎛ f1 (x) ⎞
⎜ by adding and subtracting the term in the numerator⎟⎟
⎜
⎜
⎝ f2 (x + δx) ⎟
⎠
⎧ f (x + δx) − f (x) ⎫
1 ⎪ 2 2 ⎪
{ f (x + δx) − f1 (x)} − f1 (x) ⎨ ⎬
f2 (x + δx) 1 ⎪ f2 (x) f2 (x + δx) ⎪
⎩ ⎭
lim
= δx → 0
δx
⎧ f1 (x + δx) − f1 (x) ⎫⎪
lim ⎪ 1
= δx → 0 ⎨ ⋅ ⎬
⎪ f2 (x + δx) δx ⎪
⎩ ⎭
⎧ f1 (x) f2 (x + δx) − f2 (x) ⎫⎪
lim ⎪
− δx → 0 ⎨ ⋅ ⎬
⎪ f2 (x) f2 (x + δx) δx ⎪
⎩ ⎭
1 d f1 (x) d
= ⋅ f1 (x) − ⋅ f (x)
f2 (x) dx f2 (x) f2 (x) dx 2
d d
f2 (x) . f (x) − f1 (x) . f (x)
dx 1 dx 2
= ⋅
[ f2 (x)]2
Let sin− 1 x = y.
Then x = sin y.
d
Differentiating both sides with respect to x, we get 1 = (sin y)
dx
d sin y . dy dy
or 1= or 1 = cos y ⋅ ⋅
dy dx dx
dy 1 1 1
∴ = = = ⋅
dx cos y √(1 − sin2 y) √(1 − x2)
d 1
Thus, sin− 1 x = ⋅
dx √(1 − x2)
Similarly, we can prove the following other results for inverse circular functions :
d 1 d 1 d 1
cos− 1 x = − 2
; tan− 1 x = 2
; cot− 1 x = − ;
dx √(1 − x ) dx 1+ x dx 1 + x2
d 1 d 1
sec− 1 x = 2
; cosec− 1 x = − ⋅
dx x√(x − 1) dx x √ (x2 − 1)
dy ⎡ 1 ⎤ dy ( y2 tan x)
or ⎢ − log cos x⎥ = − y tan x or = − ⋅
dx ⎣ y ⎦ dx 1 − y log cos x
⎧ 1 ⎫ ⎛ 1 − sin2 t ⎞ cos2 t
= a ⎪ − sin t + ⎪= a⎜ ⎟ = a ⋅
⎨ 2 sin t cos t ⎬⎪
1 1 ⎝ sin t ⎠ sin t
⎪ 2 2 ⎭
⎩
Also dy ⁄ dt = a cos t.
dy dy ⁄ dt a cos t
Now = = = tan t.
dx dx ⁄ dt (a cos2 t) ⁄ sin t
− 1 ⎡ log x
x⎢ sin − 1 x ⎤⎥
x sin ⎢ + ⎥
⎢ √(1 − x2) x ⎥⎦
du du ⁄ dx ⎣
Now = =
dv dv ⁄ dx 1
√(1 − x2)
− 1 ⎡ x log x + √(1 − x2) sin − 1 x⎤
= x sin x⎢ ⎥ ⋅
⎣ x ⎦
a− x
Example 1 : Differentiate tan− 1 with respect to x.
1 + ax
a− x
Solution : Let y = tan− 1 ⋅
1 + ax
Then y = tan− 1 a − tan− 1 x.
dy 1 ,
∴ = 0− since tan − 1 a is constant
dx 1 + x2
= − 1 ⁄ (1 + x2).
Example 2 : Differentiate tan − 1 [{√(1 + x2) − 1} ⁄ x] with respect to tan − 1 x.
√(1 + x2) − 1
Solution : Let u = tan − 1 and v = tan − 1 x.
x
Then to find du ⁄ dv.
Since v = tan − 1 x, therefore x = tan v.
√(1 + tan2 v) − 1 sec v − 1
∴ u = tan − 1 = tan − 1
tan v tan v
D-98 DIFFERENTIAL CALCULUS
1
1 − cos v 2 sin2 v
2 1 1
= tan − 1 = tan − 1 1 1
= tan − 1 tan v = v.
sin v 2 sin v cos v 2 2
2 2
H ence du ⁄ dv = 1 ⁄ 2.
√(1 + x2) + √(1 − x2)
Example 3 : Differentiate tan− 1 ⋅
√(1 + x2) − √(1 − x2)
√(1 + x2) + √(1 − x2)
Solution : Let y = tan− 1 ⋅
√(1 + x2) − √(1 − x2)
Put x2 = cos 2 θ.
√(1 + cos 2 θ) + √(1 − cos 2θ)
Then y = tan − 1
√(1 + cos 2θ) − √(1 − cos 2 θ)
cos θ + sin θ 1 + tan θ
= tan − 1 = tan − 1
cos θ − sin θ 1 − tan θ
⎛ π ⎞ π
= tan − 1 tan ⎜ + θ⎟ = + .θ
⎝4 ⎠ 4
1 1
∴ y= π+ cos − 1 x2.
4 2
dy 1 1 x
H ence = − ⋅ 2x = − ⋅
dx 2 √(1 − x4) √(1 − x4)
d2 y
Example 3 : If x = a (cos θ + θ sin θ) , y = a (sin θ − θ cos θ) , find ⋅
dx2
Solution : We have x = a (cos θ + θ sin θ).
dx
∴ = a (− sin θ + sin θ + θ cos θ) = a θ cos θ.
dθ
Also y = a (sin θ − θ cos θ).
dy
∴ = a (cos θ − cos θ + θ sin θ) = a θ sin θ.
dθ
dy d y ⁄ d θ aθ sin θ
Now = = = tan θ.
dx d x ⁄ d θ aθ cos θ
SUCCESSIVE DIFFERENTIATION D-103
d2 y d ⎛ dy⎞ d ⎡ d ⎤ dθ
∴ 2
= ⎜ ⎟ = (tan θ) = ⎢ (tan θ) ⎥
dx dx dx
⎝ ⎠ dx ⎣ dθ ⎦ dx
1 1 sec3 θ
= sec2 θ ⋅ = ⋅
aθ cos θ a θ
d2 p a 2 b2
Example 4 : If p2 = a2 cos2 θ + b2 sin2 θ, prove that p + = ⋅
dθ2 p3
Solution : We have p2 = a2 cos2 θ + b2 sin2 θ. …(1)
Differentiating both sides of (1) w.r.t. ‘θ’, we have
dp
2p = − 2a2 cos θ sin θ + 2b2 sin θ cos θ
dθ
dp
or p = (b2 − a2) sin θ cos θ.
. …(2)
dθ
Now differentiating both sides of (2) w.r.t. ‘θ’, we have
d2 p ⎛ dp⎞ 2
p 2 + ⎜ ⎟ = (b2 − a2) (cos2 θ − sin2 θ)
dθ ⎝ dθ ⎠
= (b2 cos2 θ + a2 sin2 θ) − (a2 cos2 θ + b2 sin2 θ)
= (b2 cos2 θ + a2 sin2 θ) − p2. [From (1)]
d2p ⎛ dp ⎞ 2
∴ p + p2 = (b2 cos2 θ + a2 sin2 θ) − ⎜ ⎟
dθ2 ⎝ dθ ⎠
(b2 − a2) 2 sin2 θ cos2 θ ,
= (b2 cos2 θ + a2 sin2 θ) −
p2
substituting for dp ⁄ dθ from (2)
1 2 2
= [ p (b cos2 θ + a2 sin2 θ) − (b2 − a2) 2 sin2 θ cos2 θ]
p2
1
= 2 [(a2 cos2 θ + b2 sin2 θ) (b2 cos2 θ + a2 sin2 θ)
p
− (b2 − a2) 2 sin2 θ cos2 θ]
1 2 2
= [a b cos4 θ + a2 b2 sin4 θ + 2a2 b2 sin2 θ cos2 θ]
p2
a 2 b2 2 2
2 θ + sin2 θ) 2 = a b ⋅
= (cos
p2 p2
2
d p 2
a b 2
Thus p 2 + p2 = ⋅
dθ p2
d2 p a 2 b2
Dividing both sides by p, we have + p = ⋅
dθ2 p3
d2 y 1 ⎛ t⎞
1. If x = a (t − sin t ) and y = a (1 + cos t ), prove that = cosec 4 ⎜ ⎟ ⋅
dx2 4a ⎝ 2⎠
D-104 DIFFERENTIAL CALCULUS
a
(v) If y = log (ax + b), then y1 = = a (ax + b) − 1,
ax + b
y2 = a2 (− 1) (ax + b) − 2, y3 = a3 (− 1) (− 2) (ax + b) − 3, etc.
In general, yn = an (− 1) (− 2)…{− (n − 1)} (ax + b) − n .
(− 1) n − 1 (n − 1) ! a n
Thus, D n log (ax + b) = ⋅ (Meerut 2003)
(ax + b) n
(vi) If y = cos (ax + b), then
⎛ π⎞
y1 = − a sin (ax + b) = a cos ⎜ ax + b + ⎟
⎝ 2⎠
π⎞ π⎞
y2 = a2 cos ⎛⎜ ax + b + 2 ⋅ ⎛
⎟ , y = a 3 cos ⎜ ax + b + 3 . ⎟ , etc.
⎝ 2⎠ 3 ⎝ 2⎠
⎛ nπ ⎞
In general, yn = an cos ⎜ ax + b + ⎟ ⋅
⎝ 2⎠
⎛ n π⎞
Thus, D n cos (ax + b) = a n cos ⎜ ax + b + ⎟ .
⎝ 2 ⎠
⎛ n π⎞
(vii) Similarly, D n sin (ax + b) = a n sin ⎜ ax + b + ⎟ .
⎝ 2 ⎠
(Gorakhpur 2005; Bundelkhand 07; Kumaun 08)
x2
Example 1 : Find the nth differential coefficient of ⋅
(x − a) (x − b)
(Rohilkhand 2014)
x2
Solution : Let y= ⋅
(x − a) (x − b)
Since the given fraction is not a proper one, therefore we should first divide the
numerator by the denominator before resolving it into partial fractions. H ere we
observe orally that the quotient will be 1. So let
x2 A B
≡ 1+ + ⋅
(x − a) (x − b) x− a x− b
Clearing the fractions, we get
x2 ≡ (x − a) (x − b) + A (x − b) + B (x − a).
Putting x = a, we get A = a2 ⁄ (a − b) and putting x = b, we get B = b2 ⁄ (b − a).
a2 b2
H ence y= 1 + +
(a − b) (x − a) (b − a) (x − b)
a2 b2
= 1+ (x − a) − 1 − (x − b) − 1.
(a − b) (a − b)
Therefore differentiating both sides n times, we get
a2 b2
yn = (− 1) n n ! (x − a) − n − 1 − (− 1) n n ! (x − b) − n − 1
(a − b) (a − b)
(− 1) n n ! ⎡⎢ a2 b2 ⎤
⎥⋅
= ⎢ +
− + ⎥
(a − b) ⎢⎣ (x − a) n 1 (x − b) n 1 ⎥
⎦
Example 2 : Find the nth differential coefficient of
(i) tan − 1 (x ⁄ a). (Meerut 2001, 05B, 09; Purvanchal 10, 14; Avadh 13)
(ii) tan − 1 x.
x a a
Solution:(i) If y = tan− 1 , then y1 = 2 = ⋅
a x + a2 (x + ia) (x − ia)
a A B
Now let ≡ + ⋅
(x + ia) (x − ia) x + ia x − ia
Clearing the fractions, we get a ≡ A (x − ia) + B (x + ia).
Putting x = ia, we get B = 1 ⁄ 2i
and putting x = − ia, we get A = − 1 ⁄ 2i.
1 ⎡ 1 1 ⎤ 1 ⎡
∴ y1 = ⎢ − ⎥ = (x − ia) − 1 − (x + ia) − 1⎤⎦ ⋅
2 i ⎣ x − ia x + ia ⎦ 2i⎣
SUCCESSIVE DIFFERENTIATION D-107
(− 1) n − 1 (n − 1) ! ⎡ − n − (x + ia) − n ⎤ .
= ⎣ (x − ia) ⎦
2i
Put x = r cos φ and a = r sin φ . Then
(− 1) n − 1 (n − 1) ! ⎡ − n − n − r− n (cos φ + i sin φ ) − n ⎤
yn = ⎣ r (cos φ − i sin φ ) ⎦
2i
(− 1) n − 1 (n − 1) ! − n ⎡
= r ⋅ ⎣ (cos nφ + i sin n φ ) − (cos nφ − i sin n φ ) ⎤⎦
2i
(− 1) n − 1 (n − 1) ! − n
= r . 2i sin nφ = (− 1) n − 1 (n − 1) ! r− n sin nφ
2i
⎛ a ⎞− n a
= (− 1) n − 1 (n − 1) ! ⎜ ⎟ sin nφ , since r =
⎝ sin φ ⎠ sin φ
= (− 1) n − 1 (n − 1) ! a− n sin n φ sin n φ , where φ = tan− 1 (a ⁄ x).
(ii) Proceeding as in part (i), we get
Dn (tan− 1 x) = (− 1) n − 1 (n − 1) ! sinn φ sin n φ , where φ = tan− 1 (1/x).
1 ax 1 ax
= e sin x+ e [sin 3x − sin x]
2 2
as 2 cos A sin B = sin (A + B) − sin (A − B)
1 ax 1 1
= e sin x + eax sin 3x − eax sin x
2 4 4
1 ax ax
= [e sin x + e sin 3x].
4
Now differentiating both sides n times, we have
1⎡ ⎛ 1⎞
yn = ⎢ (1 + a2) n ⁄ 2 eax sin ⎜ x + n tan− 1 ⎟
4⎣ ⎝ a⎠
⎛ 3⎞ ⎤
+ (9 + a2) n ⁄ 2 eax sin ⎜ 3x + n tan− 1 ⎟ ⎥
⎝ a⎠ ⎦
1 ax⎡ ⎛ 1 ⎞
= e ⎢ (1 + a2) n ⁄ 2 sin ⎜ x + n tan− 1 ⎟
4 ⎣ ⎝ a⎠
⎛ 3⎞ ⎤
+ (9 + a2) n ⁄ 2 sin ⎜ 3x + n tan− 1 ⎟ ⎥ ⋅
⎝ a⎠ ⎦
th 5
Example 3 : Find the n differential coefficient of sin x cos x. 3
Solution : Let z = cos x + i sin x, then
z− 1 = (cos x + i sin x) − 1 = cos x − i sin x.
Therefore z + z− 1 = 2 cos x and z − z− 1 = 2i sin x.
Also by De-Moivre’s theorem, z m = cos mx + i sin mx, z− m = cos m x − i sin mx.
Therefore z m + z− m = 2 cos mx and z m − z− m = 2i sin mx.
Now (2i sin x) 5 (2 cos x) 3 = (z − z− 1) 5 (z + z− 1) 3
= (z8 − z− 8) − 2 (z6 − z− 6) − 2 (z4 − z− 4) + 6 (z2 − z− 2)
= 2i sin 8x − 2 (2i sin 6x) − 2 (2i sin 4x) + 6 (2i sin 2x).
Therefore sin5 x cos3 x = 2− 7 [sin 8x − 2 sin 6x − 2 sin 4x + 6 sin 2x].
H ence D n (sin5 x cos3 x)
⎡ ⎛ nπ ⎞ ⎛ n π⎞
= 2− 7 ⎢ 8n sin ⎜ 8x + ⎟ − 2. 6n sin ⎜ 6x + ⎟
⎣ ⎝ 2 ⎠ ⎝ 2 ⎠
⎛ nπ ⎞ ⎛ n π⎞⎤
− 2 . 4n sin ⎜ 4x + ⎟ + 6.2n sin ⎜ 2x + ⎟⎥ ⋅
⎝ 2⎠ ⎝ 2 ⎠⎦
x2 x
(iii) ⋅ (iv) ⋅
(x + 2) (2x + 3) (x − a) (x − b) (x − c)
x4 , n ≥ 3.
4.
(x − 1) (x − 2) (Agra 2014)
⎧1 + x⎫ ⎧ 2x ⎫
5. (i) tan− 1 ⎨ ⎬. (Purvanchal 2011) (ii) tan− 1 ⎪⎨ ⎪.
⎬
⎩1 − x⎭ ⎪ 1 − x2 ⎪
⎩ ⎭
⎧ ⎫
⎪ √(1 + x2) − 1 ⎪
6. If y = tan− 1 ⎨⎪ ⎬ , show that yn = 1 (− 1) n − 1 (n − 1) ! sin n θ sin n θ,
x ⎪ 2
⎩ ⎭
where θ = cot− 1 x.
x , prove that y = (− 1) n n ! a− n − 1 sinn + 1 φ cos (n + 1) φ ,
7. If y = 2 n
(x + a2)
where φ = tan− 1 (a ⁄ x). (Kanpur 2009)
This theorem is useful for finding the nth differential coefficient of the product of
two functions. The statement of this theorem is as follows :
If u and v are any two functions of x such that all their desired differential coefficients
exist, then the nth differential coefficient of their product is given by
D n (uv) = (D n u). v + n C1 D n − 1 u. Dv + n C2 D n − 2 u. D2 v + …
… + n Cr D n − r u. D rv + … + u D n v.
(Meerut 2004, 05BP, 08, 09; Bundelkhand 05; Agra 07;
Kumaun 08; Purvanchal 11; Kashi 13)
Proof : We shall prove this theorem by mathematical induction. By actual
differentiation, we have
D (uv) = (Du) . v + u. Dv. …(1)
From (1) we see that the theorem is true for n = 1.
Now suppose that the theorem is true for a particular value of n. Then we have
D n (uv) = (D n u) v + n C1 D n − 1 u Dv + n C2 D n − 2 u D2 v + …
… + n Cr D n − r u D r v + n Cr + 1 D n − r − 1 u D r + 1 v + … + u. D n v. …(2)
Differentiating both sides of (2) with respect to x, we have
D n + 1 (uv) = {(D n + 1 u). v + D n u Dv} + {n C1 D n u. Dv + n C1 D n − 1 u D2 v}
+ {n C2 D n − 1 u. D2 v + n C2 D n − 2 u. D 3 v} + …
… + {n Cr D n − r + 1 u D r v + n Cr D n − r u D r + 1v}
+ {n Cr + 1 D n − r u. D r + 1 v + n Cr + 1 D n − r − 1 u D r + 2 v}+ …
… + {Du D n v + u D n + 1 v}.
R earranging the terms, we have
D n + 1 (uv) = (D n + 1 u). v + (1 + n C1) (D n u Dv) + ( n C1 + n C2) (D n − 1 u D2 v)
+ … + ( n Cr + n Cr + 1) (D n − r u. D r + 1 v) + … + u D n + 1v. …(3)
But we know that n Cr + n Cr + 1 = n + 1Cr + 1.
Therefore n C0 + n C1 = n + 1 C1 where n C0 = 1, n C1 + n C2 = n + 1C2 , etc.
H ence (3) becomes
SUCCESSIVE DIFFERENTIATION D-111
or x yn + (n − 1) yn − 1 = (n − 1) !+ (n − 1) yn − 1
D-112 DIFFERENTIAL CALCULUS
or x yn = (n − 1) ! or yn = (n − 1) ! ⁄ x.
H ence D n (x n − 1 log x) = (n − 1) ! ⁄ x.
Example 3 : If y = a cos (log x) + b sin (log x) , show that
x2 y2 + xy1 + y = 0, (Bundelkhand 2006, 11, 12; Avadh 08;
Kashi 12; Meerut 13)
and x2 yn + 2 + (2n + 1) xyn + 1 + (n2 + 1) yn = 0.
Solution : We have y = a cos (log x) + b sin (log x).
Differentiating both sides with respect to x, we have
a b
y1 = − sin (log x) + cos (log x)
x x
or xy1 = − a sin (log x) + b cos (log x).
Differentiating both sides again with respect to x, we have
a b
xy2 + y1 = − cos (log x) − sin (log x)
x x
or 2
x y2 + xy1 = − [a cos (log x) + b sin (log x)]
or x2 y2 + xy1 = − y or x2 y2 + xy1 + y = 0.
Differentiating both sides of this equation n times by Leibnitz’s theorem, we get
D n (x2 y2) + D n (xy1) + D n ( y) = 0
or (D n y2) x2 + n C1 (D n − 1 y2) . (Dx2) + n C2 . (D n − 2 y2) . (D2 x2)
+ (D n y1) . x + n C1 (D n − 1 y1) . (Dx) + D n y = 0
n (n − 1)
or x2 yn + 2 + 2 xn yn + 1 + . 2yn + xyn + 1 + nyn + yn = 0
2
or x2 yn + 2 + (2n + 1) xyn + 1 + (n2 + 1) yn = 0.
− 1
Example 4 : If y = ea sin x, show that
(1 − x2) yn + 2 − (2n + 1) x yn + 1 − (n2 + a2) yn = 0.
(Garhwal 2000, 01; Gorakhpur 05; Rohilkhand 05, 08;
Agra 06, 08; Purvanchal 07)
− 1
Solution : We have y= ea sin x .
− 1
Therefore y1 = ea sin x . a ⁄ √(1 − x2)
− 1 − 1
or y1 . √(1 − x2) = aea sin x = ay, [replacing ea sin x by y]
⎡ n (n − 1) ⎤
or ⎢ yn + 2 . (1 − x2) + n yn + 1 (− 2x) + yn (− 2) ⎥
⎣ 2! ⎦
− [ yn + 1 x + n yn . 1] − a2 yn = 0
or (1 − x2) yn + 2 − (2n + 1) x yn + 1 − (n2 + a2) yn = 0.
Example 5 : If y 1 ⁄ m + y − 1⁄m = 2 x, prove that
(x2 − 1) yn + 2 + (2n + 1) x yn + 1 + (n2 − m 2) yn = 0.
(Agra 2002; Meerut 04, 12B; Rohilkhand 06, 09B, 10B, 11; Purvanchal 06)
Solution : We have y1 ⁄ m + y− 1 ⁄ m = 2 x.
Multiplying both sides by y1 ⁄ m , we get
y2 ⁄ m + 1 = 2 x y1 ⁄ m or y2 ⁄ m − 2 x y1 ⁄ m + 1 = 0.
2 x ± √(4x2 − 4)
∴ y1 ⁄ m = = x ± √(x2 − 1) or y = [x ± √(x2 − 1)]m . ...(1)
2
⎧ x ⎫⎪
∴ y1 = m[x ± √(x2 − 1)]m − 1 ⎪⎨ 1 ± 2 ⎬
⎪⎩ √(x − 1) ⎪⎭
my my , from (1).
= ± 2
[x ± √(x2 − 1)]m = ±
√(x − 1) √(x2 − 1)
Squaring both sides, we get
y12 (x2 − 1) = m 2 y2. Differentiating again, we get
2 y1 y2 (x2 − 1) + 2 x y12 = 2 m 2 y y1 or 2 y1 [ y2 (x2 − 1) + x y1 − m 2 y] = 0
or y2 (x2 − 1) + x y1 − m 2 y = 0, since 2 y1 ≠ 0. ...(2)
Differentiating (2) n times by Leibnitz’s theorem, we get
D n { y2 (x2 − 1)} + D n ( y1 x) − m 2 D n y = 0
or yn + 2 . (x2 − 1) + n C1 yn + 1 2 x + n C2 yn . 2 + yn + 1 . x
+ n C1 yn . 1 − m 2 y n = 0
or (x2 − 1) yn + 2 + (2n + 1) x yn + 1 + (n2 − m 2) yn = 0.
dn
Example 6 : If In = (x n log x) , prove that In = n In − 1 + (n − 1) !;
dx n
(Meerut 2004B; Agra 06; Gorakhpur 06; Rohilkhand 08; Avadh 11)
⎛ 1 1 1⎞
hence show that In = n ! ⎜ log x + 1 + + + … + ⎟⎠ ⋅
⎝ 2 3 n
dn n dn − 1 ⎡ d n ⎤
Solution : We have, In = n
[x log x] = n − 1
⎢ (x log x) ⎥
dx dx ⎣ dx ⎦
dn − 1 ⎡ n − 1 1⎤
= ⎢ nx log x + x n ⋅ ⎥
dx n − 1 ⎣ x⎦
dn − 1 n − 1 log x)+ d
n− 1
= n (x (x n − 1)
dx n − 1 dx n − 1
= n In − 1 + (n − 1) Proved.
! ...(1)
We have just proved that In = n In − 1 + (n − 1) !.
D-114 DIFFERENTIAL CALCULUS
In In − 1 1
Dividing both sides by n !, we have = + ⋅ ...(2)
n ! (n − 1) ! n
Changing n to n − 1 in the above relation (2), we have
In − 1 In − 2 1
= + ⋅
(n − 1) ! (n − 2) ! n − 1
In − 1
Putting this value of in (2), we have
(n − 1) !
In In − 2 1 1
= + + ⋅
n ! (n − 2) ! n − 1 n
Thus making repeated use of the reduction formula (2), we ultimately have
In I1 1 1 1
= + + + …+ ⋅
n! 1! 2 3 n
d 1
But I1 = (x log x) = x ⋅ + log x = log x+ 1.
dx x
In 1 1 1
∴ = log x + 1 + + + … +
n! 2 3 n
⎛ 1 1 1⎞
or In = n ! ⎜ log x + 1 + + + … + ⎟ ⋅
⎝ 2 3 n⎠
1. State Leibnitz’s theorem. (Meerut 2005B, 08, 11; Bundelkhand 08; Agra 08)
d n ⎛ log x ⎞ (− 1) n (n !) ⎡ 1 1 1⎤
6. Prove that n
⎜ ⎟ = n + 1
⎢ log x − 1 − − − … − ⎥ .
dx ⎝ x ⎠ x ⎣ 2 3 n ⎦
7. If y = x n log x , prove that x yn+ 1 = n ! .
(Meerut 2001; Bundelkhand 09; Rohilkhand 11B)
⎛ 1 1 1⎞
H ence show that In = n ! ⎜ log x + 1 + + + … + ⎟ .
⎝ 2 3 n⎠
8. By forming in two different ways the nth derivative of x2n , prove that
SUCCESSIVE DIFFERENTIATION D-115
n2 n2 (n − 1) 2 n2 (n − 1) 2 (n − 2) 2 (2 n) !
1+ 2
+ 2 2
+ 2 2 2
+ …= .
1 1 ⋅ 2 1 ⋅ 2 ⋅ 3 (n !) 2
[Hint. Find the nth derivative of x n . x n and of x2n and equate].
9.
⎛ sin x ⎞
Prove that D n ⎜
⎝ x ⎠
⎧ ⎛ 1 ⎞ ⎛ 1 ⎞⎫
⎟ = ⎨ P sin ⎜ x + nπ ⎟ + Q cos ⎜ x + nπ ⎟ ⎬
⎩ ⎝ 2 ⎠ ⎝ 2 ⎠⎭ /x n + 1,
where P = x n − n (n − 1) x n − 2 + n (n − 1) (n − 2) (n − 3) x n − 4 − …,
and Q = nx n − 1 − n (n − 1) (n − 2) x n − 3 + …
− 1
10. If y = etan x, prove that
(1 + x2) yn + 2 + [2 (n + 1) x − 1] yn + 1 + n (n + 1) yn = 0.
(Avadh 2010; Kanpur 14)
dn ⎧ dPn ⎫⎪
H ence if Pn = (x2 − 1) n , show that d ⎪⎨⎪ (1 − x2) ⎬ + n (n + 1) Pn = 0.
dx n dx ⎩ dx ⎪⎭
⎛ y⎞ ⎛ x⎞ n
14. If cos− 1 ⎜ ⎟ = log ⎜ ⎟ , prove that x2 yn + 2 + (2n + 1) xyn + 1 + 2n2 yn = 0.
⎝ b⎠ ⎝ n⎠
(Meerut 2006B; Rohilkhand 13; Purvanchal 14)
15. If y = [ x + √(1 + x2)]m ,
prove that (1 + x2) y2 + x y1 − m 2 y = 0
and (1 + x2) yn + 2 + (2n + 1) x yn + 1 + (n2 − m 2) yn = 0.
(Kanpur 2006; Avadh 09; Bundelkhand 14)
Sometimes we are required to find the nth differential coefficient of y for x = 0 i.e.
( yn ) 0. This may be done even though we may not be able to find the nth differential
coefficient in a compact form for the general value of x. The method will be clear from
the following example :
d 2y
8. If x = a (cos θ + θ sin θ), y = a (sin θ − θ cos θ), then is
d x2
1 sec 3 θ
(a)
a θ
a sec 3 θ
(b)
θ
θ sec 3 θ
(c)
a
(d) a θ sec 3 θ
− 1
10. If y = e tan x, then (1 + x 2) yn + 2 + [2 (n + 1) x − 1] yn+ 1 = …… .
(a) − n (n − 1) yn
n
(b) (n + 1) yn
2
n
(c) − (n − 1) yn
2
(d) − n (n + 1) yn
True or False.
Write ‘T’ for true and ‘F’ for false statement.
13. While applying Leibnitz’s theorem if we observe that one of the two functions is
such that all its differential coefficients after a certain stage become zero, then we
should take that function as second function.
D-120 DIFFERENTIAL CALCULUS
⎛ n π ⎞⎟
1. an sin ⎜ a x + b + ⋅
⎝ 2 ⎠
2. (− 1) n n ! an (a x + b) − n − 1.
1
3. [e x − (5) n ⁄ 2 e x cos (2x + n tan − 1 2)]. 4. − y. 5. n2 yn − 1.
2
6. (− 1) n e− x. 7. (b). 8. (a). 9. (c).
10. (d). 11. T. 12. F. 13. T.
5.1 Accurate Statement of Taylor’s Theorem
If f (x) is a single-valued function of x such that
(i) all the derivatives of f (x) upto (n − 1) th are continuous in a ≤ x ≤ a + h ,
and (ii) f (n) (x) exists in a < x < a + h, then
h2
f (a + h) = f (a) + hf ′ (a) + f ′′ (a) + …
2!
hn − 1 (n − 1) hn (n)
+ f (a) + f (a + θh), where 0 < θ < 1.
(n − 1) ! n!
Taylor’s Series : (Meerut 2009B, 10B; Kashi 11, 13)
Su ppose f (x) possesses continuous derivat ives of all orders in the interval
[a, a + h]. Then for every positive integral value of n, we have
h2
f (a + h) = f (a) + h f ′ (a) + f ′′ (a) + …
2!
hn − 1 (n − 1)
+ f (a) + R n , …(1)
(n − 1) !
hn (n)
where Rn = f (a + θ h), (0 < θ < 1).
n!
Suppose R n → 0, as n → ∞ . Then taking limits of both sides of (1) when n → ∞ ,
we get
D-122 DIFFERENTIAL CALCULUS
h2 hn (n)
f (a + h) = f (a) + h f ′ (a) +f ′′(a) + … + f (a) + … …(2)
2! n!
The series given in (2) is known as Taylor’s infinite series for the expansion of
f (a + h) as a power series in h .
Suppose f (x) possesses continuous derivatives of all orders in the interval [0, x].
Then for every positive integral value of n, we have
x2 x n − 1 (n − 1)
f (x) = f (0) + x f ′ (0) + f ′′ (0) + … + f (0) + R n , … ...(1)
2! (n − 1) !
x n (n)
where Rn = f (θ x), (0 < θ < 1).
n!
Suppose R n → 0, as n → ∞ . Then taking limits of both sides of (1) when n → ∞ ,
we get
x2 x n (n)
f (x) = f (0) + x f ′ (0) + f ′′ (0) + … + f (0) + … …(2)
2! n!
The series given in (2) is known as Maclaurin’s infinite series for the expansion
of f (x) as a power series in x. Maclaurin’s series is a particular case of Taylor’s series. If
in Taylor’s series we put a = 0 and h = x, we get Maclaurin’s series.
Maclaurin’s expansion of f (x) fails if any of the functions f (x), f ′ (x), f ′′ (x), …,
becomes infinite or discontinuous at any point of the interval [0, x] or if R n does not
tend to zero as n → ∞ .
Let the expansion (1) be differentiable term by term any number of times. Then
by successive differentiation, we have
f ′ (x) = A 1 + 2A 2 x + 3 A 3 x2 + 4A 4 x3 + …,
f ′′ (x) = 2 . 1 A 2 + 3 . 2 A 3 x + 4 . 3 A 4 x2 + …,
f ′′′ (x) = 3 . 2 . 1 A 3 + 4 . 3 . 2 A 4 x + …, and so on.
Putting x = 0 in each of these relations, we get
f (0) = A 0 , f ′ (0) = A 1, f ′′ (0) = 2! A 2 , f ′′′ (0) = 3! A 3 , … .
Substituting these values of A 0 , A 1 , A 2 , … in (1), we get
x2 x n (n)
f (x) = f (0) + x f ′ (0) +f ′′ (0) + … + f (0) + …
2! n!
This is Maclaurin’s Theorem. If we denote f (x) by y, then Maclaurin’s theorem
can also be written in the following way :
x x2 x3 xn
y = ( y) 0 + ( y1) 0 + ( y2) 0 + ( y3) 0 + … + (y ) + ….
1! 2! 3! n! n 0
Taylor’s Theorem : (Bundelkhand 2005; Avadh 09, 10, 14; Kashi 11, 13, 14)
L et f (x) be a function of x which possesses, continuous derivatives of all orders in the
interval [a, a + h]. A ssum ing that f (a + h) can be expanded as an infinite power series in
h, we have
h2 hn (n)
f (a + h) = f (a) + hf ′ (a) + f ′′(a) + … + f (a) + …
2! n!
Proof : Suppose f (a + h) = A 0 + A 1 h + A 2 h2 + A 3 h3 + … …(1)
Let the expansion (1) be differentiable term by term any number of times w.r.t.
‘h’. Then by successive differentiation w.r.t. ‘h’, we have
f ′ (a + h) = A 1 + 2 A 2 h + 3A 3 h2 + …,
f ′′ (a + h) = 2 . 1 A 2 + 3 . 2 A 3 h + …,
f ′′′ (a + h) = 3 . 2 . 1 A 3 + …, and so on.
Putting h = 0 in each of the above relations, we get
f (a) = A 0 , f ′ (a) = A 1, f ′′ (a) = 2! A 2 , f ′′′ (a) = 3! A 3 , and so on.
1 1
∴ A 0 = f (a), A 1 = f ′ (a), A 2 = f ′′ (a), A 3 = f ′′′ (a), and so on.
2! 3!
Substituting these values of A 0 , A 1 , A 2 , A 3 , … in (1), we get
h2 hn (n)
f (a + h) = f (a) + h f ′ (a) +
f ′′ (a) + … + f (a) + …
2! n!
This is Taylor’s theorem. Another useful form is obtained on replacing h by
(x − a). Thus
(x − a) 2 (x − a) n (n)
f (x) = f (a) + (x − a) f ′ (a) + f ′′ (a) + … … + f (a) + …,
2! n!
which is an expansion of f (x) as a power series in (x − a).
Note : If we expand f (x + h), by Taylor’s theorem, as a power series in h, then
the result is as follows :
h2 hn (n)
f (x + h) = f (x) + hf ′ (x) + f ′′ (x) + … + f (x) + …
2! n!
D-124 DIFFERENTIAL CALCULUS
(− 1) n − 1 (n − 1)!
Then f (0) = log 1 = 0, f (n) (x) =
(x + 1) n
so that f (n) (0) = (− 1) n − 1 (n − 1)!, where n = 1, 2, 3, 4, … .
Now by Maclaurin’s theorem,
x2
f (x) = f (0) + x f ′ (0) + f ′′ (0) + … .
2!
Substituting the values of f (0), f ′ (0), f ′′ (0), etc., we get
x2 x3 x4
log (1 + x) = 0 + x − ⋅ 1! + ⋅ 2! − ⋅ 3! + …
2! 3! 4!
xn
…+ (− 1) n − 1 (n − 1)! + …
n!
x2 x3 x4 xn
= x− + − + … + (− 1) n −. 1 + …
2 3 4 n
Example 5 : A pply Maclaurin’s theorem to find the expansion in ascending powers
of x of loge (1 + e x) to the term s containing x4. (Kanpur 11; Rohilkhand 12)
Solution : Let y = log e (1 + e x). Then ( y) 0 = log e (1 + e0) = log e 2.
ex (1 + e x) − 1 1 1 1
Now y1 = = = 1− so that ( y1) 0 = 1 − = ,
1 + ex 1 + ex 1 + ex 2 2
ex ex 1
y2 = 0 + = . = y1 (1 − y1) = y1 − y12,
(1 + ex) 2 1 + e x 1 + e x
1 ⎛ ⎞2
1 1
so that ( y2) 0 = ( y1) 0 − [( y1) 0]2 = − ⎜⎝ ⎟⎠ = ,
2 2 4
1 1 1
y3 = y2 − 2y1 y2 so that ( y3) 0 = − 2 ⋅ ⋅ = 0,
4 2 4
⎛ 1⎞ 2 1
y4 = y3 − 2y22 − 2y1 y3 so that ( y4) 0 = 0 − 2 ⋅ ⎜ ⎟ − 0 = − , and so on.
⎝ 4⎠ 8
Now by Maclaurin’s theorem, we have
x2 x3 x4
y = ( y) 0 + x ( y1) 0 + ( y2) 0 + ( y3) 0 + (y ) + ….
2! 3! 4! 4 0
1 x2 1 x3 x4 ⎛ 1 ⎞
∴ log (1 + e x) = log 2 + x ⋅
+ ⋅ + ⋅ 0+ ⋅ ⎜− ⎟ + …
2 2! 4 3! 4! ⎝ 8 ⎠
x x2 x4
= log 2 + + − + ….
2 8 192
Example 6 : Expand log {x + √(1 + x2)} in ascending powers of x and find the
general term .
Solution : Let y = log {x + √(1 + x2)}. …(1)
1 ⎧ 2x ⎫ 1
Then y1 = 2
⋅ ⎪⎨ 1 + ⎪=
⎬ …(2)
x + √(1 + x ) ⎪⎩ 2 √(1 + x ) ⎪⎭ √(1 + x2)
2
∴ y12 (1 + x2) − 1 = 0.
Differentiating again, we get (1 + x2) 2 y1 y2 + 2x y12 = 0
or 2y1 [(1 + x2) y2 + xy1] = 0
D-126 DIFFERENTIAL CALCULUS
⎡1 ⎛
1 ⎞ ⎤ ⎛ 1 ⎞
Now expanding f ⎢ π + ⎜ x− π ⎟⎠ ⎥ by Taylor’s theorem in powers of ⎜ x − π⎟ ,
⎣2 ⎝ 2 ⎦ ⎝ 2 ⎠
we get
⎡1 ⎛ 1 ⎞⎤ ⎛1 ⎞ ⎛ 1 ⎞ ⎛1 ⎞
f (x) = f ⎢⎣ π + ⎜ x − π ⎟ ⎥⎦ = f ⎜ π ⎟ + ⎜ x − π ⎟ f ′ ⎜ π ⎟
2 ⎝ 2 ⎠ ⎝2 ⎠ ⎝ 2 ⎠ ⎝2 ⎠
1 ⎛ 2 3
1 ⎞ ⎛1 ⎞ 1 ⎛ 1 ⎞ ⎛1 ⎞
+ ⎜ x − π ⎟ f ′′ ⎜ π ⎟ + ⎜ x − π ⎟ f ′′′ ⎜ π ⎟ + … …(1)
2!⎝ 2 ⎠ ⎝2 ⎠ 3!⎝ 2 ⎠ ⎝2 ⎠
⎛1 ⎞ 1
Now f (x) = sin x. Therefore f ⎜ π ⎟ = sin π = 1,
⎝2 ⎠ 2
⎛1 ⎞ 1
f ′ (x) = cos x giving f ′ ⎜ π ⎟ = cos π = 0,
⎝2 ⎠ 2
⎛1 ⎞ 1
f ′′ (x) = − sin x so that f ′′ ⎜ π ⎟ = − sin π = − 1,
⎝2 ⎠ 2
⎛1 ⎞ 1
f ′′′ (x) = − cos x so that f ′′′ ⎜ π ⎟ = − cos π = 0,
⎝2 ⎠ 2
⎛ 1 ⎞ 1
f i v (x) = sin x so that f i v ⎜ π ⎟ = sin π = 1, etc.
⎝2 ⎠ 2
1 ⎛ 4
1 ⎞
+ ⎜ x − π⎟ . 1 + …
4!⎝ 2 ⎠
1 ⎛ 2 1 ⎛ 4
1 ⎞ 1 ⎞
= 1− ⎜ x − 2 π⎟ + ⎜ x − 2 π⎟ − … .
2!⎝ ⎠ 4!⎝ ⎠
Example 9 : Expand log sin (x + h) in powers of h by Taylor’s theorem .
(Purvanchal 2006; Meerut 10; Bundelkhand 09; Kashi 14)
Solution : Let f (x + h) = log sin (x + h).
Then by Taylor’s theorem, we have
log sin (x + h) = f (x + h)
h h2 h3
= f (x) + f ′ (x) + f ′′ (x) + f ′′′ (x) + … .
1! 2! 3!
Now f (x + h) = log sin (x + h).
∴ f (x) = log sin x ,
f ′ (x) = (1 ⁄ sin x) . cos x = cot x ,
f ′′ (x) = − cosec 2 x ,
f ′′′ (x) = 2 cosec x cosec x cot x = 2 cosec 2 x cot x .
… … … … … … … … …
H ence,
1 1
log sin (x + h) = log sin x + h cot x − h2 cosec 2 x + h3 cosec 2 x cot x + …
2 3
EXPANSIONS OF FUNCTIONS D-129
1
or y1 = [(x − i) − 1 − (x + i) − 1]. …(1)
2i
Differentiating (1), (n − 1) times, we get
1
yn = [( − 1) n − 1 (n − 1)! (x − i) − n − (− 1) n − 1 (n − 1)! (x + i) − n ]
2i
(− 1) n − 1 (n − 1)!
or yn = [(x − i ) − n − (x + i) − n ]. …(2)
2!
Now put x = r cos θ, 1 = r sin θ in (2). Then
(− 1) n − 1 (n − 1)! − n
yn = .r [(cos θ − i sin θ) − n − (cos θ + i sin θ) − n ]
2i
(− 1) n − 1 (n − 1)! − n
= r [(cos n θ + i sin n θ) − (cos n θ − i sin n θ)],
2i
by De Moivre’s theorem
(− 1) n − 1 (n − 1)! − n
= r . 2i sin n θ
2i
= (− 1) n − 1 (n − 1)! sinn θ sin n θ. [... r− 1 = 1 ⁄ r = sin θ]
H ence f (n) (x) = (− 1) n − 1 (n − 1) ! sinn θ sin n θ,
where cot θ = x, i.e., θ = cot− 1 x.
Putting n = 1, 2, 3,…, we get
f ′ (x) = sin θ . sin θ, f ′′ (x) = − sin2 θ sin 2θ,
f ′′′ (x) = 2! sin3 θ sin 3 θ, and so on.
Substituting these values in Taylor’s series
h2 h3 hn (n)
f (x + h) = f (x) + hf ′ (x) + f ′′ (x) + f ′′′ (x) + … + f (x) + …, we get
2! 3! n!
2
tan− 1 (x + h) = tan− 1 x + h sin θ . sin θ − h sin2 θ sin 2 θ
2!
h3 2! hn
+ sin3 θ sin 3 θ − … + (− 1) n − 1 (n − 1)! sin n θ sin n θ + …
3! n!
or tan− 1 (x + h) = tan− 1 x + h sin θ. sin θ − (h sin θ) 2 sin 2 θ
1 2
sin 3 θ sin n θ
+ (h sin θ) .3 + … + (− 1) − 1 (h sin θ) n .
n + …
3 n
D-130 DIFFERENTIAL CALCULUS
1. (i) State Maclaurin’s theorem. (Meerut 2000; Bundelkhand 01, 08, 11; Agra 07)
(i) State Taylor’s theorem. (Bundelkhand 2006, 08, 11)
(iii) e x cos x
(iv) tan− 1 x (Bundelkhand 2001)
(v) sec x.
3. (i) O btain by Maclaurin’s theorem the first five terms in the expansion of esin x.
(Bundelkhand 2007)
ex
(ii) E xpand by Maclaurin’s theorem as far as the term x3. (Meerut 2006B)
1 + ex
(iii) O btain by Maclaurin’s theorem the first five terms in the expansion of
log (1 + sin x). (Meerut 2007)
(iv) Find the first three terms in the expansion in the powers of x of log (1 + tan x) .
(Rohilkhand 2011B)
1 2 1 4 1 6
4. (i) Apply Maclaurin’s theorem to prove that log sec x = x + x + x + ….
2 12 45
(Bundelkhand 2011; Rohilkhand 13)
2x2 4x3
(ii) U se Maclaurin’s formula to show that e x sec x = 1 + x + + + …
2! 3!
(Meerut 2004; Rohilkhand 08B)
(iii) Expand sinh x cos x to fifth powers of x.
5. Show that
2 x3 22 x4 22 x5 23 x7
(i) e x cos x = 1 + x − − − + + …
3! 4! 5! 7!
⎛ 1 ⎞ 2n ⁄ 2 n
+ cos ⎜ nπ ⎟ ⋅ x + ….
⎝4 ⎠ n!
(Bundelkhand 2014; Agra 14)
2 3 22 5 ⎛ 1 ⎞ 2n ⁄ 2 n
(ii) e x sin x = x + x2 + x − x − …+ sin ⎜ nπ ⎟ x + …
3! 5! ⎝4 ⎠ n!
(Meerut 2003; Gorakhpur 06)
a2 − b2 2 a (a2 − 3b2) 3
(ii) e ax cos bx = 1 + ax + x + x + …
2 3!
(a2 + b2) n ⁄ 2 n ⎛ b⎞
+ x cos ⎜ n tan− 1 ⎟ + … .
n! ⎝ a⎠
x2 x3
7. Show that e x cos α cos (x sin α) = 1 + x cos α + cos 2α + cos 3α + …
2! 3!
(Rohilkhand 2007; Avadh 11)
⎛1 ⎞
(ii) E xpand sin ⎜ π + θ⎟ in powers of θ.
⎝4 ⎠
(iii) E xpand 2x3 + 7x2 + x − 1 in powers of x − 2.
(Meerut 2004B, 05B; Gorakhpur 06; Rohilkhand 09;
Purvanchal 11; Kashi 11)
(iv) Write the value of α , if by Taylor’s theorem
2x3 + 7x2 + x − 1 = α + 53 (x − 2) + 19 (x − 2) 2 + 2 (x − 2) 3.
(Meerut 2001)
(v) E xpand log sin x in powers of (x − a).
(Meerut 2001, 06; Rohilkhand 07B; Avadh 10)
− 1
10. (i) If y = e a sin x,
show that (1 − x2) yn + 2 − (2n + 1) xyn + 1 − (n2 + a2) yn = 0.
H ence by Maclaurin’s theorem, show that
− 1 a2 2 a (12 + a2) 3
e a sin x= 1 + ax +
x + x + … (Kumaun 2008)
2! 3!
1 2
Also deduce that e θ = 1 + sin θ + sin2 θ + sin3 θ + …
2! 3!
(ii) If y = sin (m sin− 1 x) , then show that
d2 y dy
(1 − x2) − x + m 2 y = 0.
dx2 dx
H ence or otherwise expand sin m θ in powers of sin θ.
(iii) If y = sin log (x2 + 2x + 1), prove that
(x + 1) 2 yn + 2 + (2n + 1) (x + 1) yn + 1 + (n2 + 4) yn = 0.
H ence or otherwise expand y in ascending powers of x as far as x6.
D-132 DIFFERENTIAL CALCULUS
x2 x3
(ii) f (x) = f (0) + x f ′ (x) − f ′′ (x) + f ′′′ (x) − …
2! 3!
x2 ⎞ ⎛ ⎛ x ⎞
[ Hint. (i) Write f ⎜ ⎟ = f ⎜x − ⎟.
⎝ 1 + x⎠ ⎝ 1+ x⎠
Now apply Taylor’s theorem.
(ii) We have f (0) = f (x − x).
Apply Taylor’s theorem and transpose the terms to get the result.]
(x log a) 2 (x log a) n
2. (i) 1 + x log a + + …+ + ….
2! n!
x3 2 5
(ii) x + + x + ….
3 15
x2 x3 11 x4 x5
(iii) 1 + x + − − − + ….
2 3 24 5
EXPANSIONS OF FUNCTIONS D-133
x3 x5 x7 x2 n − 1
(iv) x − + − + … + (− 1) n − 1 + ….
3 5 7 (2n − 1)
x2 5x4 61 x6
(v) 1 + + + + ….
2! 4! 6!
x2 x4 1 1 x3 1
3. (i) 1 + x + − + …. (ii) + x − + ….
2 8 2 4 8 3!
x2 x3 x4 x5 1 2
(iii) x − + − + − …. (iv) x − x2 + x3 + … .
2 6 12 24 2 3
2x 3 4x 5
4. (iii) x − − + ….
3! 5!
x2
8. sin− 1 h + x (1 − h2) − 1 ⁄ 2 + h (1 − h2) − 3 ⁄ 2
2!
x3
+ {(1 − h2) − 5 ⁄ 2 (1 + 2h2)}+ … .
3!
⎛ 2
1 ⎞
⎠ /
(1 + π 2 ⁄ 16) − π ⎜ x − π ⎟ / {4 (1 + π 2 ⁄ 16) 2} + … .
⎛ 1 ⎞
9. (i) tan− 1 (π ⁄ 4) + ⎜ x − π ⎟
⎝ 4 ⎝ 4 ⎠
1 ⎛⎜ θ2 θ3 θ4 θ5 ⎞
(ii) 1+ θ− − + + − …⎟ ⋅
√2 ⎝ 2! 3! 4! 5! ⎠
(iii) 45 + 53 (x − 2) + 19 (x − 2) 2 + 2 (x − 2) 3.
(iv) 45.
(x − a) 2 (x − a) 3
(v) log sin a + (x − a) cot a − cosec 2 a + 2 cosec 2 a cot a + … .
2! 3!
m (12 − m 2) m (12 − m 2) (32 − m 2)
10. (ii) sin m θ = m sin θ + sin3 θ + sin5 θ + … .
3! 5!
2 3 5 3
(iii) y = 2x − x2 − x3 + x4 − x5 + x6 + … .
3 2 3 2
x2 5 x4 y2
11. x+ − + … ., y − + …
2! 24 2
x3 7 x3
12. x+ + … ., x − x2 + + …
6 6
2m (2m − 2) (2m − 4)…2
13. a2 m = 0, a2m + 1 = ⋅
(2m + 1) (2m − 1) …3
3. By Maclaurin’s theorem
2 3 2n ⁄ 2 sin (n π ⁄ 4 ) n
y = e x sin x = x + x2 + x + …… + x + …… ,
3! n!
then ( y3) 0 = …… . (Meerut 2001)
4. By Taylor’s theorem the expansion of log (x + h) in ascending powers of x
is …… .
Multiple Choice Questions:
Indicate the correct answer for each question by writing the corresponding letter from
(a), (b), (c) and (d).
5. By Taylor’s theorem the second term in the expansion of log sin x in the powers
of (x − a) is
(a) (x − a) cot a (b) (x − b) cosec a
(c) (x − a) cot a cosec a (d) (x − a) 2 cot a
(Rohilkhand 2005)
6. By Maclaurin’s theorem the second term in the expansion of e ⁄ (1 + e x) is
x
1 x
(a) x (b)
48 4
1 3
(c) 0 (d) − x (Rohilkhand 2007)
48
True or False:
Write ‘T’ for true and ‘F’ for false statement.
7. The function log x does not possess Maclaurin’s series expansion because it is not
defined at x = 0.
8. If in Taylor’s series we put a = 0 and h = x, we get Maclaurin’s series.
x2 x3
9. Is this f (x) = f (0) + x f ′ (0) + f ′′ (0) + f ′′′ (0) + …… Taylor’s Theorem.
2! 3!
(Agra 2005, 06)
12 . x 3 32 . 12 . x 5 52 . 32 . 12 . x 7
1. x+ + + + …… .
3! 5! 7!
x x2 x3
2. 1. 3. 2. 4. log h + − + − …… .
h 2 h2 3 h3
5. (a). 6. (b). 7. T.
8. T. 9. F.
6.1 Indeterminate Forms
The form 0 ⁄ 0 has got no definite value. For if we write 0 ⁄ 0 = y, then the equation
0y = 0 reduces to an identity in y, i.e., it is true for all values of y. We cannot cancel 0
from both sides. Therefore the form 0 ⁄ 0 is meaningless.
lim lim
Now suppose x → a φ (x) = 0 and x → a ψ (x) = 0.
lim
lim φ (x) x → a φ (x)
Then we cannot write x → a =
ψ (x) lim
x → a ψ (x)
lim φ (x)
because in that case x → a takes the form 0 ⁄ 0 which is meaningless. It, however,
ψ (x)
lim φ (x)
does not mean that if x → a takes the form 0 ⁄ 0, then the limit itself does not exist.
ψ (x)
lim 2 2
lim x2 − a2 x → a (x − a )
For example, x → a takes the form 0 ⁄ 0 if we write it as ⋅
x− a lim
x → a (x − a)
lim x2 − a2 lim (x − a) (x + a) lim
But, we have x → a = x →a = x → a (x + a) = 2a
x− a (x − a)
and thus the limit exists.
D-136 DIFFERENTIAL CALCULUS
H ere it should not be confused that we have made an attempt to find the value of
0/0. We have simply evaluated the limit of a function which is the quotient of two
functions such that if we take their limits separately, then the combination takes the
form 0/0.
The form 0/0 is an indeterminate form. It has no definite value. The other
indeterminate forms are ∞ ⁄ ∞ , ∞ − ∞ , 0 × ∞ , 1∞ , 00, ∞ 0. In this chapter we shall
discuss methods which enable us to evaluate the limits of indeterminate forms.
lim e x − e− x − 2 log (1 + x)
Example 1 : Evaluate x → 0 .
x sin x
lim e x − e− x − 2 log (1 + x)
Solution : We have x → 0 [form 0/0]
x sin x
2
e x + e− x −
lim (1 + x)
= x →0 [form 0/0]
sin x + x cos x
2
e x − e− x +
lim (1 + x) 2 1− 1+ 2 2
= x →0 = = = 1.
cos x + cos x − x sin x 1 + 1 − 0 2
1 3
sin x − x + x
lim 6
Example 2 : Evaluate x → 0 ⋅
x5
1 3
sin x − x + x
lim 6
Solution : We have x → 0 [form 0/0]
x5
1 2
cos x − 1 + x
lim 2
= x →0 [form 0/0]
5 x4
D-138 DIFFERENTIAL CALCULUS
lim − sin x + x
= x →0 [form 0/0]
20 x3
lim − cos x + 1
= x →0 [form 0/0]
60 x2
lim sin x
= x →0 [form 0/0]
120 x
lim cos x 1
= x →0 = ⋅
120 120
x2 5 3
− x + …
lim 2 6
= x →0
x2
lim ⎛ 1 5 ⎞ 1
= x → 0 ⎜ − x + terms containing higher powers of x⎟ = .
⎝2 6 ⎠ 2
⎡ 1 11 2 ⎤
e ⎢− x+ x + …⎥
⎣ 2 24 ⎦
lim lim ⎡ 1 11 ⎤ 1
= x →0 = x →0 e ⎢ − + x + …⎥ = − e.
x ⎣ 2 24 ⎦ 2
lim x5 − 2x3 − 4 x2 + 9x − 4
3. (i) x → 1 ⋅
x4 − 2x3 + 2x − 1
lim a x − b x
(ii) x → 0 ⋅ (Agra 2003)
x
lim log (1 − x2)
(iii) x → 0 ⋅
log cos x
lim xe x − log (1 + x)
(iv) x → 0 ⋅
x2
lim x − sin x
4. (i) x → 0 ⋅
tan3 x
lim sin 2x + 2 sin2 x − 2 sin x
(ii) x → 0 ⋅
cos x − cos2 x
lim (1 + x) n − 1
(iii) x → 0 .
x
lim log x
(iv) x → 1 ⋅ (Garhwal 2001)
x− 1
lim e x − esin x
5. (i) x → 0 ⋅
x − sin x
lim {1 − √(1 − x2)}
(ii) x → 0 ⋅
x2
lim tan x − x
(iii) x → 0 ⋅
x2 tan x (Avadh 2014)
lim tan x − x
(iv) x → 0 ⋅
x − sin x
INDETERMINATE FORMS D-141
lim φ ′ (x)
= x →a + 1.
ψ ′ (x)
lim φ ′ (x)
Therefore λ = x →a ⋅
ψ ′ (x)
Case III : λ = ∞ . In this case, we have
1 lim ψ (x) lim ψ ′ (x)
= = ⋅ [by case II]
lim φ (x) x → a φ (x) x → a φ ′ (x)
x → a ψ (x)
lim φ (x) lim φ ′ (x)
Therefore x → a = ⋅
ψ (x) x → a ψ ′ (x)
lim lim
H ence in every case in which x → a φ (x) = ∞ and x → a ψ (x) = ∞ , we get
lim log x
Example 1 : Evaluate x → 0 ⋅
cot x (Agra 2014; Purvanchal 14)
lim log x
Solution : We have, x → 0 [form ∞ ⁄ ∞ ]
cot x
lim 1 ⁄ x
= x →0 [form ∞ ⁄ ∞ ]
− cosec 2 x
lim − sin2 x
= x →0 [form 0/0]
x
lim − 2 sin x cos x − 2 × 0 × 1
= x →0 = = 0.
1 1
lim log sin 2x
Example 2 : Evaluate x → 0 ⋅ (Agra 2001)
log sin x
lim log sin 2x
Solution : We have x → 0 [form ∞ ⁄ ∞ ]
log sin x
lim (2 ⁄ sin 2x) . cos 2x lim 2 cot 2x
= x →0 ⋅ = x →0 [form ∞ ⁄ ∞ ]
(1 ⁄ sin x) . cos x cot x
D-144 DIFFERENTIAL CALCULUS
lim − 4 cosec2 2x
= x →0 [form ∞ ⁄ ∞ ]
− cosec2 x
lim 4 sin2 x
= x →0 [form 0/0]
sin2 2x
lim 4 sin2 x
= x →0
(2 sin x cos x) 2
lim 1
= x →0 = 1.
cos2 x
lim log log (1 − x2)
Example 3 : Evaluate x → 0 ⋅ (Kumaun 2000; Avadh 13)
log log cos x
lim log log (1 − x2)
Solution : We have, x → 0 [form ∞ ⁄ ∞ ]
log log cos x
1 1
⋅ ⋅ (− 2 x)
lim log (1 − x2) 1 − x2
= x →0 ⋅
1 1
⋅ ⋅ (− sin x)
log cos x cos x
lim x cos x log cos x
= 2 x →0
sin x . (1 − x2) log (1 − x2)
lim x lim cos x lim log cos x
= 2 x →0 ⋅ ⋅
sin x x → 0 1 − x2 x → 0 log (1 − x2)
lim log cos x
= 2 × 1 × 1 × x →0 [form 0/0]
log (1 − x2)
1
⋅ (− sin x)
lim cos x
= 2 x →0
1
⋅ (− 2x)
1 − x2
1 lim ⎛ sin x 1 − x2 ⎞
= 2× ⋅ ⎜ ⋅ ⎟
2 x →0 ⎝ x cos x ⎠
= 1.
lim log (x − a)
3. (i) x → a ⋅
log (e x − ea)
lim sec π x
(ii) x → 1 ⋅
2 tan 3 π x
lim ⎧⎪ (log x) 3 ⎫⎪
4. (i) x → ∞ ⎨⎪ ⎬⋅
⎩ x ⎪⎭
lim x (log x) 3
(ii) x → ∞ ⋅
1 + x + x2
lim 1
5. (i) x → ∞ x tan ⋅
x (Bundelkhand 2001)
lim log tan 2x
(ii) x → 0 ⋅
log tan 3x
⎛ 1 ⎞
log ⎜ x − π⎟
lim ⎝ 2 ⎠
6. (i) x → π ⁄ 2 ⋅
tan x
lim c {e1 ⁄ (x − a) − 1}
(ii) x → a ⋅
{e1 ⁄ (x − a) + 1}
lim ⎧ 2 1 ⎫⎪
7. (i) x → 1 ⎪⎨ 2 − ⋅
⎪⎩ x − 1 x − 1 ⎬⎪⎭ (Garhwal 2002)
lim
(ii) x → π ⁄ 2 (sec x − tan x).
lim ⎛ 1 ⎞
Example 1 : Evaluate x → π ⁄ 2 ⎜ sec x − ⎟.
⎝ 1 − sin x⎠
lim ⎛ 1 ⎞
Solution : We have, x → π ⁄ 2 ⎜ sec x − ⎟ [form ∞ − ∞ ]
⎝ 1 − sin x ⎠
lim ⎛ 1 1 ⎞
= x →π ⁄ 2 ⎜ − ⎟ [form ∞ − ∞ ]
⎝ cos x 1 − sin x⎠
lim 1 − sin x − cos x
= x →π ⁄ 2 [form 0/0]
cos x − cos x sin x
lim − cos x + sin x − 0+ 1
= x →π ⁄ 2 = = ∞.
2
− sin x + sin x − cos x 2 − 1+ 1− 0
lim ⎛ 1 1 ⎞
Example 2 : Evaluate x → 0 ⎜ 2 − ⎟ . (Kanpur 2011)
⎝x sin2 x⎠
lim ⎛ 1 1 ⎞
Solution : We have, x → 0 ⎜ 2 − ⎟ [form ∞ − ∞ ]
⎝x sin2 x ⎠
lim sin2 x − x2
= x →0 [form 0/0]
x2 sin2 x
⎛ x3 ⎞2
⎜x − + …⎟ − x2
lim ⎝ 3! ⎠
= x →0
⎛ x3 ⎞2
x2 ⎜ x − + …⎟
⎝ 3! ⎠
2 x4
− + terms containing higher powers of x
lim 3!
= x →0
x4 + terms containing higher powers of x
2
− + terms containing powers of x only in the numerator
lim 3!
= x →0
1 + terms containing powers of x only in the numerator
− 2 1
= = − ⋅
3! 3
lim φ (x)
= x →a [form 0/0]
1
ψ (x)
INDETERMINATE FORMS D-147
ψ (x)
lim
= x →a [form ∞ ⁄ ∞ ]
1
φ (x)
We shall reduce the form 0 × ∞ to the form 0 ⁄ 0 or ∞ ⁄ ∞ according to our
convenience.
lim
Example 1 : Evaluate x → 0 x log sin x.
lim
Solution : We have, x → 0 x log sin x [form 0 × ∞ ]
lim log sin x
= x →0 [form ∞ ⁄ ∞ ]
1⁄x
lim (1 ⁄ sin x) . cos x
= x →0 [form ∞ ⁄ ∞ ]
− 1 ⁄ x2
lim − x2 cos x
= x →0 [form 0/0]
sin x
lim x2 sin x − 2x cos x
= x →0 = 0.
cos x
lim ⎛ 1 x ⎞
1. (i) x → 1 ⎜ − ⎟ ⋅
⎝ log x log x⎠
⎛ cot x − 1 ⎞
lim ⎜⎜ x ⎟⎟
(ii) x → 0 ⎜ ⎟ ⋅
⎝ x ⎠
lim ⎛ cosec x − cot x⎞
2. (i) x → 0 ⎜ ⎟ ⋅
⎝ x ⎠
lim ⎛⎜ 1 1⎞
(ii) x → 0 ⎜⎜ x − ⎟⎟ ⋅
⎝ e − 1 x ⎟⎠
lim ⎛ π ⎞
3. (i) x → π ⁄ 2 ⎜ x tan x − sec x⎟ ⋅
⎝ 2 ⎠
lim
(ii) x → 0 x log x.
lim
4. (i) x → 0 sin x. log x.
lim
(ii) x → ∞ x (a1 ⁄ x − 1).
lim a
5. (i) x → ∞ 2 x sin x ⋅
2 (Agra 2003)
lim
(ii) x → π ⁄ 2 (1 − sin x) tan x.
D-148 DIFFERENTIAL CALCULUS
lim m n
6. x → 0 x (log x) , where m and n are positive integers.
1 1 1
1. (i) − 1. (ii) − ⋅ 2. (i) ⋅ (ii) − ⋅
3 2 2
3. (i) − 1. (ii) 0. 4. (i) 0. (ii) log a.
5. (i) a. (ii) 0. 6. 0.
lim
Suppose x → a {φ (x)}ψ (x) takes any one of these three form s.
lim
Then let y = x → a {φ (x)}ψ (x) .
lim
Taking logarithm of both sides, we get log y = x → a ψ (x). log φ (x).
Now in any of the above three cases, log y takes the form 0 × ∞ which can be
evaluated by the process of aricle 6.6.
lim 2
Example 1 : Evaluate x → 0 (cos x) cot x.; (Kumaun 2001; Bundelkhand 14)
lim 2
Solution : Let y = x → 0 (cos x) cot x. [form 1∞ ]
lim
∴ log y = x → 0 cot2 x. log cos x [form ∞ × 0]
lim log cos x
= x →0 [form 0/0]
tan2 x
lim (1 ⁄ cos x) . (− sin x)
= x →0 (by L’H ospital’s rule)
2 tan x sec 2 x
lim − tan x lim 1 1
= x →0 = x →0 = − ⋅
2 tan x sec 2 x − 2 sec 2 x 2
∴ y = e− 1 ⁄ 2.
lim ⎛ a⎞ x
Example 2 : Evaluate x → ∞ ⎜ 1 + ⎟ ⋅
⎝ x⎠
lim ⎛ a⎞ x
Solution : Let y = x → ∞ ⎜ 1 + ⎟ ⋅ [form 1∞ ]
⎝ x⎠
lim ⎧ ⎛ a⎞ ⎫
∴ log y = x → ∞ ⎨ x log ⎜ 1 + ⎟ ⎬ [form ∞ × 0]
⎩ ⎝ x⎠ ⎭
lim log (1 + a ⁄ x)
= x →∞ [form 0/0]
1⁄x
INDETERMINATE FORMS D-149
1
⋅ (− a ⁄ x2)
lim 1 + a ⁄ x lim a
= x →∞ = x →∞ = a.
− 1⁄x 2 1 + a⁄x
Therefore, y = e a.
lim ⎡ 1 1 ⎤
Example 1 : Evaluate x → 0 ⎢ − 2 log (1 + x) ⎥ .
⎣x x ⎦
lim ⎡ 1 1 ⎤ lim x − log (1 + x)
Solution : We have, x → 0 ⎢ − 2 log (1 + x) ⎥ = x → 0 [form 0/0]
⎣ x x ⎦ x2
1
1−
lim 1+ x ,
= x →0 by L’Hospital’s rule [The form is again 0/0]
2x
1
lim (1 + x) 2 1
= x →0 = ⋅
2 2
2
lim ⎛ tan x⎞ 1 ⁄ x
Example 2 : Evaluate x → 0 ⎜ ⎟ .
⎝ x ⎠ (Meerut 2000; Garhwal 01; Kumaun 02;
Agra 03; Kashi 14; Purvanchal 14)
2
lim ⎛ tan x ⎞ 1 ⁄ x
Solution : Let y = x → 0 ⎜ ⎟ ⋅
⎝ x ⎠
lim 1 ⎛ tan x⎞ lim 1 ⎡1 ⎛ x3 2 5 ⎞⎤
∴ log y = x → 0 2 log ⎜ ⎟ = x → 0 2 log ⎢ ⎜ x + + x + …⎟ ⎥
x ⎝ x ⎠ x ⎣x ⎝ 3 15 ⎠⎦
lim 1 ⎡ ⎛ x2 2 4 ⎞⎤
= x → 0 2 log ⎢ 1 + ⎜ + x + …⎟ ⎥
x ⎣ ⎝ 3 15 ⎠⎦
⎛ x2 2 4 ⎞ 1 ⎛ x2 2 4 ⎞2
⎜ + x + …⎟ − ⎜ + x + …⎟ + …
lim ⎝3 15 ⎠ 2⎝3 15 ⎠
= x →0
x2
x2
+ terms containing higher powers of x
lim 3
= x →0
x2
lim ⎡1 ⎤
= x →0 ⎢ + terms containing powers of x only in the numerator⎥
⎣3 ⎦
1
= ⋅
3
∴ y= e1 ⁄ 3.
D-150 DIFFERENTIAL CALCULUS
lim
1. (i) x → 0 x x . (Agra 2002; Kanpur 04)
lim
(ii) x → 0 (cos x) 1 ⁄ x.
lim 2
2. (i) x → 0 (cos x) 1 ⁄ x . (Garhwal 2003)
lim
(ii) x → π ⁄ 2 (sin x) tan x.
lim
3. (i) x → π ⁄ 2 (sec x) cot x.
lim
(ii) x → π ⁄ 4 (tan x) tan 2x.
lim ⎛ x ⎞ tan (π x ⁄ 2a)
4. (i) x → a ⎜ 2 − ⎟
⎝ a⎠ (Rohilkhand 2012)
lim
(ii) x → 1 (1 − x2) 1 ⁄ log (1 − x) .
lim
5. (i) x → 1 x1 ⁄ (1 − x) .
lim
(ii) x → ∞ (a0 x m + a1 x m − 1 + … + am ) 1 ⁄ x.
lim ⎛ tan x⎞ 1 ⁄ x
6. (i) x → 0 ⎜ ⎟ .
⎝ x ⎠ (Garhwal 2001, 03)
3
lim ⎛ tan x⎞ 1 ⁄ x
(ii) x → 0 ⎜ ⎟ .
⎝ x ⎠
2
lim ⎛ sinh x⎞ 1 ⁄ x
7. (i) x → 0 ⎜ ⎟ .
⎝ x ⎠ (Kumaun 2008)
2
lim ⎛ sin x⎞ 1 ⁄ x
(ii) x → 0 ⎜ ⎟ . (Kumaun 2003)
⎝ x ⎠
2
lim ⎧⎪ 2 (cosh x − 1) ⎫⎪ 1 ⁄ x
8. (i) x → 0 ⎨⎪ ⎬⎪ .
⎩ x2 ⎭
lim ⎧ log x⎫ 1 ⁄ x
(ii) x → ∞ ⎨ ⎬ .
⎩ x ⎭
lim π 1⁄x
9. (i) x → ∞ ⎛⎜ − tan− 1 x⎞⎟ .
⎝2 ⎠
lim
(ii) x → 0 (cosec x) 1 ⁄ log x.
lim πx
10. x → 1 (1 − x) tan 2 ⋅
INDETERMINATE FORMS D-151
a 1
1. ⋅ 2. 2 k . 3. − ⋅
b 3
2 1 a
4. π⋅ 5. −
3
⋅ 6. log
b
⋅
7. (c). 8. (c). 9. (b).
10. (c). 11. (d). 12. (c).
13. (d) 14. F. 15. T.
16. T. 17. F.
7.1 Functions of Several Independent Variables
So far we have considered functions of one independent variable only. H owever,
in practice, we often come across functions of more than one independent variable. For
example, the area of a rectangle depends upon two independent variables, namely the
length and the breadth. Similarly, the volume of a rectangular parallelopiped depends
upon three independent variables, namely the length, the breadth and the height.
There are a number of differences between the calculus of one and of two
variables. Fortunately the calculus of functions of three or more variables differs only
slightly from that of functions of two variables. The study here will be limited largely
to functions of two variables.
Definition : L et z be a sym bol which has one definite value for every pair of values
of x and y. Then z is called a function of the two independent variables x and y, and is usually
written as z = f (x, y). A function of x and y is also written as φ (x, y) or ψ(x, y) etc.
(Kanpur 2014)
A similar definition can be given for functions of more than two independent
variables.
If to each point (x, y), of a part of the xy-plane is assigned a unique real number z,
even then z is said to be given as a function, z = f (x, y), of the independent variables
x and y. The locus of all points (x, y, z) satisfying z = f (x, y) is a surface in ordinary
space.
D-154 DIFFERENTIAL CALCULUS
∂2 z ∂ ⎛ ∂z ⎞ ∂2 z ∂ ⎛ ∂z ⎞
= fx x = ⎜ ⎟ and = fy x = ⎜ ⎟ ⋅
∂x 2 ∂x ⎝ ∂x⎠ ∂y ∂x ∂y ⎝ ∂x⎠
∂z
Similarly, from may be obtained
∂y
∂2 z ∂ ⎛ ∂z ⎞ ∂2 z ∂ ⎛ ∂z ⎞
= fx y = ⎜ ⎟ and = fy y = ⎜ ⎟ ⋅
∂x ∂y ∂x ⎝ ∂y⎠ ∂y2 ∂y ⎝ ∂y⎠
If z = f (x, y) and its partial derivatives are continuous (as is true in all ordinary
cases), the order of differentiation is immaterial, that is,
∂2 z ∂2 z
= ⋅
∂x ∂y ∂y ∂x
∂2 u ∂2 u
Example 1 : If u = ax2 + 2hxy + by2, find and ⋅
∂y ∂x ∂x ∂y
Solution : We have, u = ax2 + 2hxy + by2.
∂u
∴ = 2ax + 2hy. (treating y as constant)
∂x
∂2 u ∂ ⎛ ∂u ⎞ ∂
H ence = ⎜ ⎟ = (2ax + 2hy)
∂y ∂x ∂y ⎝ ∂x ⎠ ∂y
= 2h. (treating x as constant)
∂u
Again = 2hx + 2by
. (treating x as constant)
∂y
∂2 u ∂ ⎛ ∂u ⎞ ∂
∴ = ⎜ ⎟ = (2hx + 2by)
∂x ∂y ∂x ⎝ ∂y ⎠ ∂x
= 2h. (treating y as constant)
∂2
u ∂2
u
H ere we note that = ⋅
∂x ∂y ∂y ∂x
⎛ y⎞ ∂u ∂u
Example 2 : If u = f ⎜ ⎟ , show that x + y = 0.
⎝ x⎠ ∂x ∂y (Agra 2003)
⎛ y⎞
Solution : We have, u = f ⎜ ⎟ ⋅
⎝ x⎠
∂u ⎧ ⎛ y⎞ ⎫ ⎛ y⎞
∴ = ⎨f ′ ⎜ ⎟ ⎬ ⎜ − ⎟ ⋅ (treating y as constant)
∂x ⎩ ⎝ x⎠ ⎭ ⎝ x2 ⎠
∂u y ⎛ y⎞
∴ x = − f′ ⎜ ⎟ ⋅ …(1)
∂x x ⎝ x⎠
∂u ⎧ ⎛ y⎞ ⎫ ⎛ 1 ⎞
Again = ⎨f′ ⎜ ⎟⎬ ⎜ ⎟ ⋅ (treating x as constant)
∂y ⎩ ⎝ x⎠ ⎭ ⎝ x ⎠
∂u y ⎛ y⎞
∴ y = f′ ⎜ ⎟ ⋅ …(2)
∂y x ⎝ x⎠
∂u ∂u
Adding (1) and (2), we have x + y = 0.
∂x ∂y
D-156 DIFFERENTIAL CALCULUS
∂r x r cos θ
or = = = cos θ.
∂x r r
∂r ∂x
Thus = ⋅
∂x ∂r
∂x
Again, ∂θ = − r sin θ. (regarding r as constant)
1 ∂x
∴ = − sin θ.
r ∂θ
Also we have, θ = tan− 1 ( y ⁄ x).
∂θ 1 ⎛ y⎞ y r sin θ sin θ
∴ = ⎜− ⎟ = − 2 = − = − ⋅
∂x y2 ⎝ x2 ⎠ x + y2 r2 r
1+
x2
∂θ
∴ r = − sin θ.
∂x
∂x ∂θ
H ence = r ⋅
r∂θ ∂x
Finally, we have θ = tan− 1 ( y ⁄ x).
∂θ y ∂2 θ 2xy
∴ = − 2 ⋅ ∴ = 2 ⋅ …(1)
∂x (x + y2) ∂x2 (x + y2) 2
∂θ 1 1 x ∂2 θ − 2 xy
Also = ⋅ = ⋅ ∴ = 2 ⋅ …(2)
∂y ⎛⎜ y2 ⎞
⎟ x x2 + y2 ∂y2 (x + y2) 2
⎜ 1 + 2⎟
⎜ x ⎟⎠
⎝
∂2 θ ∂2 θ
Adding (1) and (2), we get + = 0.
∂x2 ∂y2
Example 5 : If u = (1 − 2 xy + y2) − 1 ⁄ 2, prove that
∂ ⎧ ∂u ⎫ ∂ ⎧ 2 ∂u ⎫
⎨ (1 − x2) ⎬+ ⎨y ⎬ = 0.
∂x ⎩ ∂x ⎭ ∂y ⎩ ∂y ⎭
Solution : H ere u = (1 − 2xy + y2) − 1 ⁄ 2.
∂u 1
∴ = − (1 − 2xy + y2) − 3 ⁄ 2 (− 2y) = yu3,
∂x 2
∂u 1
and = − (1 − 2xy + y2) − 3 ⁄ 2 . (− 2x + 2y) = (x − y) u3.
∂y 2
∂ ⎧ ∂u ⎫ ∂ ⎧
Now ⎨ (1 − x2) ⎬= ⎨ (1 − x2) . yu 3⎫⎬⎭
∂x ⎩ ∂x ⎭ ∂x ⎩
∂u
= y (− 2x) u3 + y (1 − x2) . 3u2
∂x
= − 2xy u3 + 3y (1 − x2) u2 yu3
= − 2 x y u3 + 3y2 u5 (1 − x2). …(1)
∂ ⎧ 2 ∂u ⎫ ∂ ⎧ 2 ∂
Also ⎨y ⎬= ⎨⎩ y (x − y) u 3⎫⎬⎭ = {(y2x − y3) u3}
∂y ⎩ ∂y ⎭ ∂y ∂y
∂u
= (2xy − 3y2) u3 + ( y2 x − y3) . 3u2
∂y
= (2 x y − 3y2) u3 + y2 (x − y) . 3u2 . (x − y) u3
D-158 DIFFERENTIAL CALCULUS
= (2 x y − 3y2) u3 + y2 (x − y) 2 . 3u5
= 2 x y u3 + 3y2 u5 [(x − y) 2 − u− 2]
= 2 x y u3 + 3y2 u5 [(x − y) 2 − (1 − 2xy + y2)], [... u− 2 = 1 − 2xy + y2]
= 2 x y u3 + 3y2 u5 [x2 − 1] = 2 x y u3 − 3y2 u5 (1 − x2). …(2)
Adding (1) and (2), we have
∂ ⎧ ∂u ⎫ ∂ ⎧ 2 ∂u ⎫
⎨ (1 − x2) ⎬+ ⎨y ⎬ = 0.
∂x ⎩ ∂x ⎭ ∂y ⎩ ∂y ⎭
2 1 ∂ ⎛ 2 ∂θ ⎞ ∂θ
Example 6 : If θ = t n e− r ⁄ 4t, what value of n will m ake 2 ⎜r ⎟ = ?
r ∂r ⎝ ∂r ⎠ ∂t
(Garhwal 2003; Lucknow 11)
∂θ 2 ⎛ 2 r⎞ r 2
Solution : We have = t n . e− r ⁄ 4t . ⎜ − ⎟ = − t n − 1 e− r ⁄ 4t.
∂r ⎝ 4t ⎠ 2
∂θ 1 2
∴ r2 = − r3 t n − 1 e− r ⁄ 4t.
∂r 2
∂ ⎛ 2 ∂θ⎞ 3 r 2 n − 1 − r 2 ⁄ 4t 1 3 n − 1 e− r 2 ⁄ 4t . ⎛⎜ − 2 r⎞
∴ ⎜r ⎟ = − t e − r t ⎟
∂r ⎝ ∂r ⎠ 2 2 ⎝ 4t ⎠
3 2 n − 1 − r 2 ⁄ 4t 1 2
= − r t e + r 4 t n −. 2 e− r ⁄ 4t .
2 4
1 ∂ ⎛ 2 ∂θ⎞ 3 2 1 2
∴ ⎜r ⎟ = − 2 t n − 1 e− r ⁄ 4t + 4 r 2 t n − 2 e− r ⁄ 4t.
r 2 ∂r ⎝ ∂r ⎠
∂θ 2 2 r2
Also = nt n − 1 e− r ⁄ 4t + t n e− r ⁄ 4t . 2
∂t 4t
2 1 2
= nt n− 1 e− r ⁄ 4t + r 2 t n − 2 e− r ⁄ 4t .
4
1 ∂ ⎛ 2 ∂θ⎞ ∂θ
Now ⎜r ⎟ =
r 2 ∂r ⎝ ∂r ⎠ ∂t
3 n − 1 − r 2 ⁄ 4t 1 2 2 1 2
⇒ − t e + r 2 t n − 2 e− r ⁄ 4t = nt n − 1 e− r ⁄ 4t + r 2 t n − 2 e− r ⁄ 4t
2 4 4
3 2 2
⇒ − t − n − 1 e− r ⁄ 4t = nt n − 1 e− r ⁄ 4t, for all possible values of r and t
2
3
⇒ n= − ⋅
2
∂u ∂u x2 y2
1. Find and when u = 2 + 2 − 1.
∂x ∂y a b
∂2 u ∂2 u
2. Prove that = in each of the following cases :
∂x ∂y ∂y ∂x
(i) u = x4 + x2 y2 + y4. (ii) u = log tan (y ⁄ x).
⎧ x2 + y2 ⎫⎪
(iii) u = log ⎪⎨⎪ ⎬⋅ (iv) u = x y.
⎩ x + y ⎪⎭
PARTIAL DIFFERENTIATION D-159
∂u ∂u
3. If u = sin − 1 (x ⁄ y) + tan− 1 ( y ⁄ x), show that x + y = 0.
∂x ∂y
⎛⎜ y ⎞⎟ ∂u ∂u
4. If u = x y f ⎜⎜ ⎟⎟ , then write the value of the expression x + y ⋅
⎝ x⎠ ∂x ∂y
(Meerut 2001; Kanpur 07)
∂2 z ∂2 z
5. If z = f (x + ay) + φ (x − ay), prove that 2
= a2 2 ⋅
∂y ∂x
(Bundelkhand 2001; Kanpur 05)
∂2 u ∂2 u 1
6. If u = f (r), where r = √(x 2 + y 2), show that + = f ′′(r) + f ′(r).
∂x 2 ∂y 2 r
(Meerut 2001; Avadh 04)
∂2 z x2
− y2
7. If z = x2 tan− 1 ( y ⁄ x) − y2 tan− 1 (x ⁄ y), prove that = 2 ⋅
∂x∂y x + y2
∂2 u ∂2 u
8. (i) If u = 2 (ax + by) 2 − (x2 + y2) and a2 + b2 = 1, prove that 2 + = 0.
∂x ∂y2
∂2u ∂2u
(ii) If u = e x(x cos y − y sin y), then show that + = 0. (Garhwal 2003)
∂x2 ∂y2
⎛ ∂z ∂z ⎞ 2 ⎛ ∂z ∂z ⎞
9. If z (x + y) = x2 + y2, show that ⎜ − ⎟ = 4 ⎜1 − − ⎟ .
⎝ ∂x ∂y⎠ ⎝ ∂x ∂y⎠
(Bundelkhand 2011; Kashi 13; Avadh 14)
∂3 u
10. If u = e x y z, show that = (1 + 3xyz + x2 y2 z2) e x y z.
∂x ∂y ∂z
(Kanpur 2011; Kashi 12; Rohilkhand 13)
∂2 z
11. If x x y y z z = c, show that at x = y = z, = − (x log e x) − 1.
∂x ∂y
(Rohilkhand 2013; Bundelkhand 14; Purvanchal 14)
∂2u u ∂2
12. Show that + = 0 when
∂x2 ∂y2
(i) u = e m y cos m x. (Agra 2014)
13. If V = (x2
+ + y2 z2) − 1 ⁄ 2,
show that
∂V ∂V ∂V
(i) x + y + z = − V.
∂x ∂y ∂z
∂2 V ∂2 V ∂2 V
(ii) + + = 0. (Kumaun 2008)
∂x2 ∂y2 ∂z2
2
14. If u = tan− 1
xy , show that ∂ u = 1
⋅
2 2
√(1 + x + y ) ∂x ∂y (1 + x2 + y2) 3 ⁄ 2
D-160 DIFFERENTIAL CALCULUS
x2 y2 z2
15. If 2 + 2 + 2 = 1, prove that
a + u b + u c + u (Rohilkhand 2011B, 12B)
⎛ ∂u ⎞ 2 ⎛ ∂u ⎞ 2 ⎛ ∂u ⎞ 2
⎛ ∂u ∂u ∂u ⎞
⎜ ⎟ + ⎜ ⎟ + ⎜ ⎟ = 2 ⎜x + y + z ⎟.
⎝ ∂x ⎠ ⎝ ∂y ⎠ ⎝ ∂z ⎠ ⎝ ∂x ∂y ∂z ⎠
16. If x = r cos θ, y = r sin θ, prove that
(i) (∂r ⁄ ∂ x) 2 + (∂ r ⁄ ∂ y) 2 = 1.
∂2 r ∂2 r 1 ⎧ ⎛ ∂ r⎞ 2 ⎛ ∂r⎞ 2⎫
(ii) + = ⎨⎜ ⎟ + ⎜ ⎟ ⎬ ⋅
∂x2 ∂y2 r ⎩ ⎝ ∂x ⎠ ⎝ ∂y⎠ ⎭ (Lucknow 2007, 11)
∂u ∂u ∂u
17. (i) If u = x2 y + y2 z + z2 x, show that + + = (x + y + z) 2.
∂x ∂y ∂z
⎪ 1 1 1⎪
⎪ ⎪ ∂u ∂u ∂y
(ii) If u = ⎪ x y z ⎪ , show that + + = 0.
⎪ 2 2 2⎪ ∂x ∂y ∂z
⎪x y z ⎪ (Rohilkhand 2013B)
2x , 2y
1. ⋅ 4. 2 u.
a 2 b2
(x1 ⁄ 4 + y1 ⁄ 4)
Example 1 : Verify Euler’s theorem for the function u = ⋅
(x1 ⁄ 5 + y1 ⁄ 5)
(Rohilkhand 2014)
Solution : H ere we see that u is a homogeneous function of x and y of degree
1 ⁄ 4 − 1 ⁄ 5 i.e., 1 ⁄ 20. Therefore in order to verify Euler’s theorem we are to show that
∂u ∂u 1
x + y = u.
∂x ∂y 20
We have, log u = log (x1 ⁄ 4 + y1 ⁄ 4) − log (x1 ⁄ 5 + y1 ⁄ 5). …(1)
Differentiating (1) partially with respect to x, we have
1 ∂u 1 ⎛
1 ⎞ 1 ⎛
1 − 4 ⁄ 5⎞
= ⋅ ⎜ x− 3 ⁄ 4⎟⎠ − 1 ⁄ 5 ⎜ x ⎟ .
u ∂x x1 ⁄ 4 + y1 ⁄ 4 ⎝ 4 x + y 1 ⁄ 5 ⎝ 5 ⎠
∂u ⎡1 x− 3 ⁄ 4 1 x− 4 ⁄ 5 ⎤⎥
∴ = u ⎢⎢ 1 ⁄ 4 − ⎥.
∂x ⎢4 x
⎣ + y1 ⁄ 4 5 x1 ⁄ 5 + y1 ⁄ 5 ⎥⎦
D-162 DIFFERENTIAL CALCULUS
∂u ⎡1 x1 ⁄ 4 1 x1 ⁄ 5 ⎤
∴ x = u ⎢⎢ 1 ⁄ 4 ⁄
− ⁄ ⁄
⎥.
⎥ …(2)
∂x ⎢4 x
⎣ + y1 4 5 x1 5 + y1 5 ⎥
⎦
Again differentiating (1) partially with respect to y, we get
1 ∂u ⎡⎢ 1 y− 3 ⁄ 4 1 y− 4 ⁄ 5 ⎤⎥
= ⎢ 1⁄4 ⁄
− ⎥.
u ∂y ⎢⎣ 4 x + y 1 4 5 x1 ⁄ 5 + y1 ⁄ 5 ⎥⎦
∂u ⎡1 y1 ⁄ 4 1 y1 ⁄ 5 ⎤
∴ y = u ⎢⎢ 1 ⁄ 4 ⁄
− ⁄ ⁄
⎥.
⎥ …(3)
∂y ⎢4 x
⎣ + y1 4 5 x1 5 + y1 5 ⎥
⎦
Adding (2) and (3), we get
∂u ∂u ⎡1 x1 ⁄ 4 + y1 ⁄ 4 1 x1 ⁄ 5 + y1 ⁄ 5 ⎤⎥ ⎡1 1⎤ 1
x + y = u ⎢⎢ − ⎥ = u ⎢ 4 − 5⎥ = u.
∂x ∂y ⎢4
⎣ x1 ⁄ 4 + y1 ⁄ 4 5 x1 ⁄ 5 + 1
y ⎦⁄ 5 ⎥ ⎣ ⎦ 20
This verifies Euler’s theorem.
⎧ x2 + y2 ⎫⎪ ∂u ∂u
Example 2 : If u = sin− 1 ⎪⎨⎪ ⎬⎪ , show that x + y = tan u.
⎩ x + y ⎭ ∂x ∂y
(Garhwal 2002; Kumaun 08; Avadh 12)
x2 + y2
Solution : We have, sin u = .
x+ y
x2 + y2
Let v = ⋅ Then v is a homogeneous function of x and y of degree 1.
x+ y
∂v ∂v
Therefore by E uler’s theorem, we have x + y = v. …(1)
∂x ∂y
Now v = sin u.
∂v ∂u ∂v ∂u
∴ = cos u and = cos u ⋅
∂x ∂x ∂y ∂y
Putting these values in (1), we get
∂u ∂u
x cos u + y cos u = v
∂x ∂y
⎛ ∂u ∂u ⎞ v sin u ⎡. . x2 + y2 ⎤
or ⎜x + y ⎟= = ⎢ . v = sin u = ⎥
⎝ ∂x ∂y ⎠ cos u cos u ⎣ x+ y ⎦
= tan u .
This proves the result.
⎛ x3 + y3 ⎞ ∂u ∂u
3. (i) If u = tan− 1 ⎜ ⎟ , prove that x + y = sin 2u.
⎝ x− y ⎠ ∂x ∂y
(Rohilkhand 2012B)
√x − √y , ∂u ∂u
(ii) If u = sin− 1 show that x + y = 0.
√x + √y ∂x ∂y
(Gorakhpur 2005; Garhwal 02)
x+ y , ∂u ∂u 1
4. (i) If u = sin − 1 show that x + y = tan u.
√x + √y ∂x ∂y 2
⎛ x+ y ⎞ ∂u ∂u 1
(ii) If u = cos− 1 ⎜ ⎟ , show that x + y + cot u = 0.
⎝ ⎯
√⎯x + ⎯
√ y ⎠ ∂x ∂y 2
(Rohilkhand 2013B)
x3 + y3 ∂u ∂u
5. (i) If u = log , show that x + y = 2.
x+ y ∂x ∂y
x4 + y4 ∂u ∂u
(ii) If u = log , show that x + y = 3.
x+ y ∂x ∂y
x2 + y2 , ∂u ∂u
(iii) If u = log prove that x + y = 1.
x+ y ∂x ∂y (Kanpur 2006; Avadh 13)
6. U se E uler’s theorem on homogeneous functions to show that if u = tan− 1 ( y ⁄ x),
∂u ∂u
then x + y = 0.
∂x ∂y
7. If u be a homogeneous function of x and y of degree n, show that
∂2 u ∂2 u ∂u
(i) x 2 + y = (n − 1) ⋅
∂x ∂x ∂y ∂x
∂2 u ∂2 u ∂u
(ii) x + y 2 = (n − 1) ⋅
∂x ∂y ∂y ∂y
∂2 u ∂2 u ∂2 u
(iii) x2 2 + 2xy + y2 2 = n (n − 1) u.
∂x ∂x ∂y ∂y
xy , ∂2 u ∂2 u ∂2 u
8. If u = show that x2 2 + 2 x y + y2 2 = 0.
x+ y ∂x ∂x ∂y ∂y
∂2 u ∂2 u ∂2 u
9. If u = x φ ( y ⁄ x) + ψ ( y ⁄ x), prove that x2 2 + 2xy + y2 2 = 0.
∂x ∂x ∂y ∂y
(Kanpur 2008)
∂u ∂u ∂u
Adding (1), (2) and (3), we have + + = 0.
∂x ∂y ∂z
dy √(1 − y2)
2. If √(1 − x2) + √(1 − y2) = a (x − y), prove that = ⋅
dx √(1 − x2)
3. Find du ⁄ dx if u = sin (x2 + y2), where a2 x2 + b2 y2 = c2.
4. If u = x4 y5, where x = t 2 and y = t 3, find du ⁄ dt.
∂f ∂φ dz ∂f ∂φ
5. If f (x, y) = 0, φ ( y, z) = 0, show that ⋅ ⋅ = ⋅ ⋅
∂y ∂z dx ∂x ∂y (Lucknow 2009)
6. I f u = √(x2 + y2) a n d x3 + y3 + 3axy = 5a2, fi n d t h e va l u e o f du ⁄ dx wh e n
x = a, y = a.
∂2u a 2∂2u
6. If u = f( y + ax) + φ ( y − ax), then − = ...... . (Agra 2002)
∂x2 ∂y 2
∂u ∂u
7. If u = f( y / x), then, x + y = ...... .
∂x ∂y (Agra 2003)
(x 2 + 2 xy − y 2)
1. 0. 2. ⋅
{(x + y) 2 + (x 2 + y 2) 2}
∂u dx ∂u dy
3. homogeneous. 4. ⋅ + ⋅ ⋅
∂x dt ∂y dt
5. (n − 1) ux . 6. 0.
7. 0. 8. (a).
9. (b). 10. (b).
11. (d). 12. (a).
13. (d). 14. T.
15. F. 16. T.
8.1 Jacobian (Kanpur 2014)
⎪ ∂u 1 ⎪
⎪
⎪ ∂x
0 0 … 0 ⎪⎪
⎪ 1 ⎪
⎪ ⎪
⎪ ∂u ∂u ⎪
∂ (u1, u2 , … , un ) ⎪⎪ 2 2
0 … 0
⎪
⎪ ∂u1 ∂u2 ∂u
= ⎪⎪ ∂x1 ∂x2 ⎪ = …… n ,
∂ (x1, x2 , … , xn ) ⎪ ∂x1 ∂x2 ∂xn
⎪ ⎪
⎪… … … … … ⎪⎪
⎪
⎪ ∂u ∂un ∂un ∂un ⎪⎪
⎪ n
⎪ … ⎪
⎪ ∂x ∂x2 ∂x3 ∂xn ⎪
1
i.e., in such cases the Jacobian reduces to the principal diagonal term of the
determinant.
∂ (x, y, z)
Example 2 : Find the Jacobian being given
∂ (r, θ, φ )
x = r cos θ cos φ , y = r sin θ √(1 − m 2 sin2 φ ) ,
z = r sin φ √(1 − n2 sin2 θ) , where m 2 + n2 = 1.
JACOBIANS D-171
Solution : H ere x2 + y2 + z2
= r2 cos2 θ cos2 φ + r2 sin2 θ − r2 m 2 sin2 θ sin2 φ
+ r2 sin2 φ − r2 n2 sin2 φ sin2 θ
= r2 (cos2 θ cos2 φ + sin2 θ + sin2 φ − sin2 θ sin2 φ ) [ ... m 2 + n2 = 1]
= r2 (cos2 θ cos2 φ + sin2 θ + sin2 φ cos2 θ)
= r2 (sin2 θ + cos2 θ) = r2.
∂x ∂y ∂z ∂x ∂y ∂z ⎫
∴ x + y + z = r;x
∂r ∂r ∂r ∂θ + y ∂θ + z ∂θ = 0 ⎪⎪
⎬ ...(1)
and ∂x ∂y ∂z ⎪
x ∂φ + y ∂φ + z ∂φ = 0. ⎪
⎭
∂x ∂x ∂x
⎪ ⎪ x ∂x ∂x ∂x
⎪ ∂r ∂θ ∂φ ⎪⎪ ⎪ ∂r x ∂θ x ∂φ ⎪
⎪
⎪ ⎪ ⎪ ⎪
⎪ ∂y ∂y ∂y ⎪⎪ 1 ⎪⎪ ∂y ∂y ∂y ⎪
Now J (x, y, z) = ⎪⎪ ∂θ ∂φ ⎪⎪ = x ⎪⎪ ∂r
⎪
⎪
⎪ ∂r ∂θ ∂φ ⎪
⎪ ⎪ ⎪ ⎪
⎪ ∂z ∂z ∂z ⎪⎪ ⎪ ∂z ∂z ∂z ⎪
⎪ ⎪ ⎪
⎪ ∂r ∂θ ∂φ ⎪ ⎪ ∂r ∂θ ∂φ ⎪
⎪
r 0 0 ⎪
⎪ ∂y ∂y ∂y ⎪
1 ⎪⎪ ⎪ by adding y R + zR to R
= ⎪ ∂r ∂θ ∂φ ⎪⎪ , 2 3 1
x ⎪⎪ ⎪ and using the relations (1)
⎪ ∂z ∂z ∂z ⎪⎪
⎪ ∂r ∂θ ∂φ ⎪
⎪ ∂y ∂y ⎪
r ⎪⎪ ∂θ ∂φ ⎪⎪ r ⎧ ∂y ∂z ∂z ∂y ⎫
= ⎪ ⎪= ⎨ − ∂θ ∂φ ⎬
x ⎪⎪ ∂z ∂z ⎪ x ⎩ ∂θ ∂φ ⎭
⎪
⎪ ∂θ ∂φ ⎪
r ⎧⎪ 2 2 2 2
= ⎨ r cos θ √(1 − m sin φ ) . r cos φ √(1 − n sin θ)
x ⎪⎩
r sin φ . n2 sin θ cos θ r sin θ . m 2 sin φ cos φ ⎫⎪
− ⋅ ⎬
√(1 − n2 sin2 θ) √(1 − m 2 sin2 φ ) ⎪⎭
r3 cos θ cos φ ⎡ (1 − m 2 sin 2 φ ) (1 − n 2 sin 2 θ) − n 2 m 2 sin 2 θ sin 2 φ ⎤
⎢ ⎥
= ⎢ ⎥
x ⎢ √[(1 − n2 sin2 θ) (1 − m 2 sin2 φ )] ⎥
⎣ ⎦
r3 cos θ cos φ
= ⋅
r cos θ cos φ
⎡ 1 − m 2 sin2 φ − n 2 sin2 θ + m 2 n 2 sin2 φ sin2 θ − m 2 n 2 sin2 θ sin2 φ ⎤
⎢ ⎥
⎢ ⎥
⎢ √[(1 − n2 sin2 θ) (1 − m 2 sin2 φ )] ⎥
⎣ ⎦
r2 (m 2 cos2 φ + n2 cos2 θ)
= ⋅ [ ... m 2 + n2 = 1]
√[(1 − n2 sin2 θ) (1 − m 2 sin2 φ )]
Example 3 : If y1 = r sin θ1 sin θ2 , y2 = r sin θ1 cos θ2 , y3 = r cos θ1 sin θ3 ,
∂ ( y1, y2 , y3 , y4)
y4 = r cos θ1 cos θ3 , find the value of the Jacobian ⋅
∂ (r, θ1, θ2 , θ3)
Solution : Squaring and adding the given relations, we have
y12 + y22 + y32 + y42 = r2 .
D-172 DIFFERENTIAL CALCULUS
⎪ r 0 0 0 ⎪⎪
⎪ ∂y ∂y2 ∂y2
⎪ ⎪ ⎪ 2 ⎪
⎪ ⎪ ⎪ ∂θ ∂θ2 ∂θ3 ⎪
⎪ ∂y2 ∂y2 ∂y2 ∂y2 ⎪ ⎪ 1 ⎪
⎪ ⎪ ⎪ ⎪
1 ⎪⎪ ∂r ∂θ1 ∂θ2 ∂θ3 ⎪
⎪ r ⎪⎪ ∂y3 ∂y3 ∂y3 ⎪
⎪
J= ⎪ ⎪ = ⎪ ⎪
y1 ⎪ ∂y3 ∂y3 ∂y3 ∂y3 ⎪ y1 ⎪ ∂θ1 ∂θ2 ∂θ3 ⎪
⎪ ⎪ ⎪ ⎪
⎪ ∂θ1 ∂θ2 ∂θ3 ⎪ ⎪ ∂y ∂y4 ∂y4 ⎪
⎪ ∂r ⎪ ⎪ 4 ⎪
⎪ ⎪ ⎪ ⎪
⎪ ∂y4 ∂y4 ∂y4 ∂y4 ⎪ ⎪ ∂θ1 ∂θ2 ∂θ3 ⎪
⎪ ⎪
⎪ ∂θ1 ∂θ2 ∂θ3 ⎪
∂r
⎪ ∂y2 ∂y2 ∂y2 ⎪
⎪ ⎪
⎪ ∂θ1 ∂θ2 ∂θ3 ⎪⎪
r ⎪⎪ 2 ⎪
= ⎪ − r cos θ1 sin θ1 0 0 ⎪,
y1 y3 ⎪ ⎪
⎪ ∂y ∂y4 ∂y4 ⎪
⎪ 4 ⎪
⎪ ∂θ2 ∂θ3 ⎪⎪
⎪ ∂θ1
adding y4 R 3 to y3 R 2 and using the results (2)
r ⎡ ∂y2 ∂y4 ∂y4 ∂y2 ⎤
= ⋅ r2 cos θ1 sin θ1 ⎢⎢ ∂θ ⋅ ∂θ − ∂θ ⋅ ∂θ ⎥⎥
y1 y3 ⎢ 2 3 2 3 ⎥⎦
⎣
r3 cos θ1 sin θ1
= [(− r sin θ1 sin θ2) (− r cos θ1 sin θ3) − 0]
y1 y3
r5 sin2 θ1 cos2 θ1 sin θ2 sin θ3
= = r3 sin θ1 cos θ1 .
r2 sin θ1 cos θ1 sin θ2 sin θ3
JACOBIANS D-173
2. 1 + u + v. 4. − y ⁄ 2 x .
8. (− 1) n sin n x1 sin n − 1 x2 … sin xn .
Proof : We have
∂u1 ∂u1 ∂y1 ∂u1 ∂y2 ∂u1 ∂u1 ∂y1 ∂u1 ∂y2 ⎫
= + , = + , ⎪
∂x1 ∂y1 ∂x1 ∂y2 ∂x1 ∂x2 ∂y1 ∂x2 ∂y2 ∂x2 ⎪
⎬ ...(1)
∂u2 ∂u2 ∂y1 ∂u2 ∂y2 ∂u2 ∂u2 ∂y1 ∂u2 ∂y2 ⎪
= + , = + ⋅ ⎪
∂x1 ∂y1 ∂x1 ∂y2 ∂x1 ∂x2 ∂y1 ∂x2 ∂y2 ∂x2 ⎭
⎪ ∂u1 ∂u1 ⎪ ⎪ ∂y1 ∂y1 ⎪
⎪ ⎪ ⎪ ⎪
∂ (u1, u2) ∂ ( y1, y2) ⎪⎪ ∂y1 ∂y2 ⎪⎪ ⎪
⎪ ∂x1 ∂x2 ⎪⎪
Now ⋅ = ⎪ ⎪× ⎪ ⎪
∂ ( y1, y2) ∂ (x1, x2) ⎪ ∂u2 ∂u2 ⎪ ⎪ ∂y2 ∂y2 ⎪
⎪ ⎪ ⎪ ⎪
⎪
⎪ ∂y1 ∂y2 ⎪⎪ ⎪
⎪ ∂x1 ∂x2 ⎪⎪
⎪ ∂u1 ∂y1 ∂u1 ∂y2 ∂u1 ∂y1 ∂u1 ∂y2 ⎪
⎪ + + ⎪
⎪ ∂y1 ∂x1 ∂y2 ∂x1 ∂y1 ∂x2 ∂y2 ∂x2 ⎪⎪
= ⎪⎪ ⎪,
⎪ ∂u2 ∂y1 ∂u2 ∂y2 ∂u2 ∂y1 ∂u2 ∂y2 ⎪
⎪ + + ⎪
⎪
⎪ ∂y1 ∂x1 ∂y2 ∂x1 ∂y1 ∂x2 ∂y2 ∂x2 ⎪⎪
applying row-by-column multiplication rule
⎪ ∂u1 ∂u1 ⎪
⎪ ⎪
⎪ ∂x1 ∂x2 ⎪⎪
= ⎪⎪ ⎪ , using the relations (1)
⎪ ∂u2 ∂u2 ⎪
⎪ ⎪
⎪
⎪ ∂x1 ∂x2 ⎪⎪
∂ (u1, u2)
= ⋅
∂ (x1, x2)
Note : T h e a b o ve fo r m u l a r e s e m b l e s ve r y m u c h wi t h t h e fo r m u l a
d f d f dt ,
= ⋅ for the derivative of the function of a function.
dx dt dx
Generalization of the above formula : I f u1, u2 , ..., un a re f u n ct io n s o f
y1, y2 , … , yn and y1, y2 , ..., yn are functions of x1, x2 , ..., xn , then
∂ (u1, u2 , …, un ) ∂ (u1, u2 , ..., un ) ∂ ( y1, y2 , ..., yn )
= ⋅ ⋅
∂ (x1, x2 , ..., xn ) ∂ ( y1, y2 , ..., yn ) ∂ (x1, x2 , ..., xn )
The proof may be easily extended as in the case of two variables and has been left
as an exercise for the students.
∂ ( F1, F2 , ..., Fn )
∂ (u1, u2 , ..., un ) ∂ (x1, x2 , ..., xn )
Then, we have = (− 1) n ⋅
∂ (x1, x2 , ..., xn ) ∂ ( F1, F2 , ..., Fn )
∂ (u1, u2 , ..., un )
Proof : Here also we shall establish the result for two variables and the proof
can be extended easily for n variables. The students should themselves write the proof
for n variables on the basis of the proof given below for two variables.
In the case of two variables, the connecting relations are
F1 (u1, u2 , x1, x2) = 0, ⎫
⎪ ...(1)
F2 (u1, u2, x1, x2) = 0. ⎬⎪⎭
From relations (1), we have by differentiation
∂F1 ∂F1 ∂u1 ∂F1 ∂u2 ⎫
+ + = 0, ⎪
∂x1 ∂u1 ∂x1 ∂u2 ∂x1 ⎪
⎪
∂F1 ∂F1 ∂u1 ∂F1 ∂u2 ⎪
+ + = 0, ⎪
∂x2 ∂u1 ∂x2 ∂u2 ∂x2 ⎪
⎬ ...(2)
∂F2 ∂F2 ∂u1 ∂F2 ∂u2 ⎪
+ + = 0, ⎪
∂x1 ∂u1 ∂x1 ∂u2 ∂x1 ⎪
∂F2 ∂F2 ∂u1 ∂F2 ∂u2 ⎪
⎪
+ + = 0. ⎪
∂x2 ∂u1 ∂x2 ∂u2 ∂x2 ⎭
⎪ ∂F 1 ∂F1⎪ ⎪ ∂u 1 ∂u1 ⎪
⎪ ⎪ ⎪ ⎪
∂ ( F1, F2 ) ∂ (u1, u2) ⎪⎪ ∂u1 ∂u2 ⎪ ⎪⎪ ∂x1
⎪ ∂x2 ⎪⎪
Now ⋅ = ⎪ ⎪× ⎪ ⎪
∂ (u1, u2) ∂ (x1, x2) ⎪ ∂F2 ∂F2 ⎪ ⎪ ∂u2 ∂u2 ⎪
⎪ ⎪ ⎪ ⎪
⎪
⎪ ∂u 1 ∂u2 ⎪⎪ ⎪
⎪ ∂x1 ∂x2 ⎪⎪
⎪ ∂F1 ∂u1 ∂F1 ∂u2 ∂F1 ∂u1 ∂F1 ∂u2 ⎪
⎪ + + ⎪
⎪ ∂u1 ∂x1 ∂u2 ∂x1 ∂u1 ∂x2 ∂u2 ∂x2 ⎪⎪
= ⎪⎪ ⎪,
⎪ ∂F2 ∂u1 ∂F2 ∂u2 ∂F2 ∂u1 ∂F2 ∂u2 ⎪
⎪ + + ⎪
⎪
⎪ ∂u1 ∂x1 ∂u2 ∂x1 ∂u1 ∂x2 ∂u2 ∂x2 ⎪⎪
⎪ − ∂F1 − ∂F1 ⎪
⎪ ⎪
⎪ ∂x ∂x2 ⎪⎪
= ⎪⎪
1
⎪ , using the relations (2)
⎪ − ∂F2 − ∂F2 ⎪
⎪ ⎪
⎪
⎪ ∂x1 ∂x2 ⎪⎪
∂ ( F1, F2)
= (− 1) 2 ⋅
∂ (x1, x2)
Accordingly, we have
∂ ( F1, F2)
∂ (u1, u2) ∂ (x1, x2)
= (− 1) 2 ⋅
∂ (x1, x2) ∂ ( F1, F2)
∂ (u1, u2)
D-176 DIFFERENTIAL CALCULUS
∂ (u, v) ∂ (x, y)
Example 1 : Prove that × = 1.
∂ (x, y) ∂ (u, v)
(Bundelkhand 2011; Kanpur 08; Meerut 12; Avadh 13)
Solution : Let u = f1 (x, y), v = f2 (x, y) . ...(1)
O bviously x and y can also be expressed as functions of u and v . Differentiating
relations (1) partially with respect to u and v, we get
∂u ∂x ∂u ∂y , ∂u ∂x ∂u ∂y , ⎫
1= + 0= + ⎪
∂x ∂u ∂y ∂u ∂x ∂v ∂y ∂v ⎪
⎬ ...(2)
∂v ∂x ∂v ∂y , ∂v ∂x ∂v ∂y
0= + 1= + ⋅ ⎪⎪
∂x ∂u ∂y ∂u ∂x ∂v ∂y ∂v ⎭
⎪ ∂u ∂u ⎪ ⎪ ∂x ∂x ⎪
∂ (u, v) ∂ (x, y) ⎪ ∂x ∂y ⎪⎪ ⎪⎪ ∂u ∂v ⎪⎪
⎪
Now × = ⎪ ⎪× ⎪ ⎪
∂ (x, y) ∂ (u, v) ⎪ ∂v ∂v ⎪ ⎪ ∂y ∂y ⎪
⎪ ⎪ ⎪ ⎪
⎪ ∂x ∂y ⎪ ⎪ ∂u ∂v ⎪
⎪ ∂u ∂x + ∂u ∂y ∂u ∂x ∂u ∂y ⎪
+
⎪ ∂x ∂u ∂y ∂u ∂x ∂v ∂y ∂v ⎪⎪ ,
⎪
= ⎪ ⎪
⎪ ∂v ∂x ∂v ∂y ∂v ∂x ∂v ∂y ⎪
⎪
⎪ ∂x ∂u
+ + ⎪
∂y ∂u ∂x ∂v ∂y ∂v ⎪
applying row-by-column multiplication
⎪1 0⎪
= ⎪ ⎪ , using the relations (2)
⎪0 1⎪
= 1.
Example 2 : If u3 + v + w = x + y2 + z2, u + v3 + w = x2 + y + z2,
u + v + w3 = x2 + y2 + z, prove that
∂ (u, v, w) 1 − 4 (x y + yz + z x) + 16 xyz
= ⋅
∂ (x, y, z) 2 − 3 (u2 + v2 + w2) + 27 u2 v2 w2
Solution : The given relations can be written as
F1 ≡ u3 + v + w − x − y2 − z2 = 0,
F2 ≡ u + v3 + w − x2 − y − z2 = 0,
F3 ≡ u + v + w3 − x2 − y2 − z = 0.
Now
∂ (u, v, w)
∂ (x, y, z)
= (− 1) 3
∂ (x, y, z) /
∂ ( F1, F2 , F3) ∂ ( F1, F2 , F3)
∂ (u, v, w)
⋅ ...(1)
∂ ( F1, F2 , F3) ⎪⎪ − 1 − 2y − 2z ⎪
H ere = ⎪ − 2x − 1 − 2z ⎪⎪
∂ (x, y, z) ⎪
⎪ − 2x
⎪
− 2y − 1 ⎪
= − 1 (1 − 4yz) + 2x (2y − 4yz) − 2x (4yz − 2z)
= − 1 + 4 ( yz + zx + xy) − 16xyz .
∂ ( F1, F2 , F3) ⎪ 3u 2 1 1 ⎪⎪
⎪
And = ⎪⎪ 1 3v2 1 ⎪⎪⎪
∂ (u, v, w) ⎪
⎪ 1
⎪ 1 3w2 ⎪⎪
JACOBIANS D-177
4. If u3 + v3 + w3
= x + y + z, + u2
+ v2
= w2
+ x3
+ y3 z3, u + v + w = x2 + y2 + z2,
∂ (u, v, w) ( y − z) (z − x) (x − y)
then prove that = ⋅
∂ (x, y, z) (u − v) (v − w) (w − u)
(Purvanchal 2007; Kanpur 12)
∂ (u, v)
5. Compute the Jacobian where
∂ (r, θ)
u = 2 x y, v = x2 − y2, x = r cos θ, y = r sin θ.
6. I f u1 = x1 + x2 + x3 + x4 , u1u2 = x2 + x3 + x4 , u1 u2 u3 = x3 + x4 , u1 u2 u3 u4 = x4 ,
show that
∂ (x1, x2 , x3 , x4)
= u 13 u 22 u 3 .
∂ (u1, u2 , u3, u4)
7. G iven y1 (x1 − x2) = 0, y2 (x12 + x1 x2 + x22) = 0, show that
∂ ( y1, y2) x1 + x2
= 3y1 y2 3 ⋅
∂ (x1, x2) x1 − x23
8. If u = x (1 − r2) − 1 ⁄ 2, v = y (1 − r2) − 1 ⁄ 2, w = z (1 − r2) − 1 ⁄ 2,
∂ (u, v, w)
where r2 = x2 + y2 + z2, show that = (1 − r2) − 5 ⁄ 2 .
∂ (x, y, z)
9. (a) Find the Jacobian of y1 , y2 , y3 , ..., yn , being given
y1 = x1 (1 − x2 ), y2 = x1 x2 (1 − x3), … ,
yn − 1 = x1 x2 … xn − 1 (1 − xn ) , yn = x1 x2 x3 … xn .
(b) If y1 = r cos θ1, y2 = r sin θ1 cos θ2 , y3 = r sin θ1 sin θ2 cos θ3 , ....
yn − 1 = r sin θ1 sin θ2 … sin θn − 2 cos θn − 1 and yn = r sin θ1 sin θ2 … sin θn − 1 ,
D-178 DIFFERENTIAL CALCULUS
prove that
∂ ( y1 , y2 , … , yn )
= r n − 1 sin n − 2 θ1 sin n − 3 θ2 … sin θn − 2 .
∂ (r, θ1 , ..., θn − 1)
10. If λ , μ , ν are the roots of the equation in k ,
x y z
+ + = 1,
a+ k b+ k c+ k
prove that
∂ (x, y, z) ( μ − ν) (ν − λ) (λ − μ)
= − ⋅
∂ (λ , μ , ν) (b − c) (c − a) (a − b)
11. The roots of the equation in λ ,
( λ − x) 3 + ( λ − y) 3 + ( λ − z) 3 = 0
are u, v, w. Prove that
∂ (u, v, w) ( y − z) (z − x) (x − y)
= − 2 ⋅ (Kumaun 2008; Kanpur 10)
∂ (x, y, z) (v − w) (w − u) (u − v)
12. If x, y, z are connected by a functional relation f (x, y, z) = 0, show that
∂ ( y, z) ⎛ ∂y⎞
= ⎜ ⎟
∂ (x, z) ⎝ ∂x⎠ z = const.
∂ (u, v, w) ∂ (x, y, z)
13. (i) Prove that × = 1. (Kanpur 2009, 11)
∂ (x, y, z) ∂ (u, v, w)
∂ ( y1, y2 , ..., yn ) ∂ (x1, x2 , ..., xn )
(ii) Prove that ⋅ = 1.
∂ (x1, x2 , ..., xn ) ∂ ( y1, y2 , ..., yn )
∂φ r
i.e., = 0 for some value of r between 1 and n .
∂xr
H ence, for that particular value of r the function φ r must not contain xr ; and
accordingly the corresponding equation is of the form
φ r (xr + 1, ..., xn , u1, u2 , ..., ur) = 0.
Consequently between this and the remaining equations φ r + 1 = 0, φ r + 2 = 0, ...,
φ n = 0, the variables xr + 1, xr + 2 , … , xn can be eliminated so as to give a final equation
between u1, u2 , ..., un alone.
H ence the theorem is established.
1. If u = x2 + y2 + z2, v = x + y + z , w = x y + y z + z x ,
∂ (u, v, w)
show that the Jacobian vanishes identically. Also find the relation
∂ (x, y, z)
between u, v and w. (Avadh 2014)
∂ (u, v)
2. If u = (x + y) ⁄ (1 − xy) and v = tan− 1 x + tan− 1 y, find ⋅
∂ (x, y)
Are u and v functionally related ? If so, find the relationship.
3. If the functions u, v, w of three independent variables x, y, z are not independent,
prove that the Jacobian of u, v, w with respect to x, y, z vanishes.
4. Show that the functions u = 3x + 2y − z, v = x − 2y + z and w = x (x + 2y − z) are
not independent and find the relation between them.
5. Show that the functions
u = x + y+ z,
v = xy + yz + zx ,
w = x3 + y3 + z3 − 3xyz
are not independent. Find the relation between them. (Meerut 2013B)
1. v2 = u + 2w. 2. u = tan v . 4. u2 − v2 = 8w .
5. u3 = 3uv + w . 6. u2 − v2 = 4w . 7. uv = w + 1 .
8. uv = r + 2w .
y2 , x2 + y2 ,
6. If u = v= then
2x 2x
∂ (u, v) y
(a) =
∂ (x, y) 2x
∂ (u, v) y
(b) = −
∂ (x, y) 2x
∂ (u, v) 2x
(c) =
∂ (x, y) y
∂ (u, v) 2x
(d) = −
∂ (x, y) y
True or False:
Write ‘T ’ for true and ‘F ’ for false statem ent.
x2 x3 x x x x ∂ (u1, u2 , u3)
7. If u1 = , u = 3 1 , u = 1 2 , then = − 4.
x1 2 x2 3 x3 ∂ (x1, x2 , x3)
8. If u1, u2 are functions of y1, y2 and y1, y2 are functions of x1, x2 , then
∂ (u1, u2) ∂ (u1, u2) ∂ ( y1, y2)
= ⋅ ⋅
∂ (x1, x2) ∂ ( y1, y2) ∂ (x1, x2)
∂ (x, y, z)
9. If x = r sin θ cos φ , y = r sin θ sin φ , z = r cos θ, then = r 2 sin θ.
∂ (r, θ, φ )
10. If u, v and w are functions of three independent variables x, y and z , then
∂ (u, v, w) ∂ (x, y, z)
⋅ = 0.
∂ (x, y, z) ∂ (u, v, w)
11. If u1, u2 and u3 are functions of three independent variables x1, x2 and x3 connected
by the equations
F1 (u1, u2 , u3 , x1, x2 , x3) = 0,
F2 (u1, u2 , u3 , x1, x2 , x3) = 0,
and F3 (u1, u2 , u3 , x1, x2 , x3) = 0,
∂ (F1, F2 , F3)
∂ (u1, u2 , u3) ∂ (x1, x2 , x3)
then = ⋅
∂ (x1, x2 , x3) ∂ (F1, F2 , F3)
∂ (u1, u2 , u3)
12. T h e fu n c t i o n s u = x2 + y2 + z 2, v = x + y+ z, w = xy + yz + zx a r e n o t
independent of each other.
13. If x + y + z = u, y + z = v, z = uvw,
∂ (x, y, z)
then = u2 v.
∂ (u, v, w)
D-184 DIFFERENTIAL CALCULUS
∂u 1
1. ⋅ 2. ⋅ 3. 1. 4. 1 + u + v.
∂y r
5. (a). 6. (b). 7. F. 8. T.
9. T. 10. F. 11. F. 12. T.
13. T. 14. F.
9
M axima and M inima of F unctions
of T wo I ndependent V ariables
D-186 DIFFERENTIAL CALCULUS
1 ⎛⎜ 2 ∂2 f ∂2 f ∂2 f ⎞
+ ⎜h + 2hk + k 2 2 ⎟⎟ + …
⎜
2 ! ⎝ ∂x 2 ∂x ∂y ∂y ⎟⎠ x = a,
y= b
⎛ ∂f ⎞ ⎛ ∂f ⎞
∴ f (a + h, b + k ) − f (a, b) = h ⎜ ⎟ x = a + k ⎜ ⎟ x = a
⎝ ∂x ⎠ ⎝ ∂y ⎠
y= b y= b
+ terms of the second and higher orders in h and k . ...(1)
By taking h and k , sufficiently small, the first degree terms in h and k can be made
to govern the sign of the right hand side and therefore of the left hand side of (1).
Thus the sign of [ f (a + h , b + k) − f (a, b)]
⎡ ⎛ ∂f ⎞ ⎛ ∂f ⎞ ⎤
= the sign of ⎢⎢ h ⎜ ⎟ + k⎜ ⎟ ⎥ ...(2)
⎢ ⎝ ∂x ⎠ x = a ⎝ ∂y ⎠ x = a ⎥⎥
⎣ ⎦
y= b y= b
⎛ ∂f ⎞
Taking k = 0, we find that if ⎜ ⎟ ≠ 0, the right hand side of (2) changes sign
⎝ ∂x ⎠ x = a
y= b
when h changes sign. Therefore f (x, y) cannot have a maximum or minimum at x = a,
⎛ ∂f ⎞
y = b if ⎜ ⎟ ≠ 0.
⎝ ∂x ⎠ x = a
y= b
Similarly taking h = 0, we can see that f (x, y) cannot have a maximum or a
⎛ ∂f ⎞
minimum at x = a, y = b if ⎜ ⎟ ≠ 0.
⎝ ∂y ⎠ x = a
y= b
Thus a set of necessary conditions that f (x, y) should have a m axim um or a m inim um
at x = a, y = b is that
⎛ ∂f ⎞ ⎛ ∂f ⎞
⎜ ⎟ = 0 and ⎜ ⎟ = 0.
⎝ ∂x ⎠ x = a ⎝ ∂y ⎠ x = a
y= b y= b
The above conditions are necessary but not sufficient for the existence of maxima
or minima.
A point (a, b) is called a stationary point, if both the first order partial derivatives
of the function f (x, y) vanish at that point. A stationary point which is either a maximum
or a minimum is called an extrem e point and the value of the function at the point is
called an extrem e value. A stationary point is not necessarily an extreme point. Thus a
stationary point may be a maximum or a minimum or neither of these two. To decide
whether a point is really an exreme point, a further investigation is necessary.
For (3a ⁄ 2 , − a ⁄ 2) ,
3 1 3
r= − a2 , s = − a2 , t = − a2 so that r t − s2 is positive. Since r is negative,
2 2 2
therefore u has a maximum at this point.
Similarly, we can show that u has a maximum at (a ⁄ 2, a ⁄ 2) .
Example 2 (a) : Find the extrem e values of xy (a − x − y) .
Solution : Let u = xy (a − x − y) .
∂u
Then = ay − 2 xy − y2
∂x
∂u
and = ax − x2 − 2 x y .
∂y
∂2 u ∂2 u
Also r= = − 2 y, s = = a − 2 x − 2 y,
∂x2 ∂y ∂x
∂2 u
and t= = − 2 x.
∂y2
For a maximum or minimum of u , we have
∂u ⁄ ∂x = 0 and ∂u ⁄ ∂y = 0.
Thus, we have
ay − 2 x y − y2 = 0 ⎫⎪
⎬⋅
ax − x2 − 2 x y = 0 ⎪⎭
These equations can be written as
y (a − 2 x − y) = 0, x (a − x − 2 y) = 0,
so that we have to consider the four pairs of equations, viz.,
y = 0, x = 0; a − 2 x − y = 0, x = 0; y = 0, a − x − 2 y = 0;
a − 2 x − y = 0, a − x − 2 y = 0 .
Solving these, we get the following pairs of values of x and y which make the
function stationary :
⎛ ⎞
1 1
(0, 0), (0, a), (a, 0), ⎜⎝ a, a⎟⎠ ⋅
3 3
For (0, 0),
r = 0, s = a, t = 0 so that r t − s2 is negative.
Therefore we have neither a maximum nor a minimum of u at (0, 0) .
For (0, a),
r = − 2 a, s = − a, t = 0 so that r t − s2 is negative.
Therefore u has not an extreme value at (0, a) .
Similarly, we may show that u has not an extreme value at (a, 0) .
⎛ ⎞
1 1
For ⎜ a, a⎟⎠ ,
⎝ 3 3
2 1 2
r= − a, s = − a, t = − a so that r t − s2 is positive.
3 3 3
⎛ ⎞
1 1
Therefore u has an extreme value at ⎜ a, a⎟⎠ and it will be a maximum or
⎝ 3 3
minimum according as, r is negative or positive i.e., according as, a is positive or
negative.
∴ the extreme value of u is (1 ⁄ 27) a3.
MAXIMA AND MINIMA of Functions of Two Independent Variables D-191
∂2 u 2 a2 ∂2 u 2 ab
Again, we get r = = 2+ 2 ,s= = 2 ,
∂x 2 c ∂x ∂y c
∂2 u 2 b2
and t= 2
= 2+ 2 ⋅
∂y c
⎛ a2 ⎞ ⎛ b2 ⎞ 4 a 2 b2 ⎛ a 2 b2 ⎞
∴ r t − s2 = 4 ⎜⎜ 1 + 2 ⎟⎟ ⎜⎜ 1 + 2 ⎟⎟ − = 4 ⎜⎜ 1 + 2 + 2 ⎟⎟ ⋅
⎜ c ⎟⎠ ⎜⎝ c ⎟⎠ c 4 ⎜ c c ⎟⎠
⎝ ⎝
Since r t − s2 is positive and r is also positive, therefore u is minimum for the values
of x and y found above.
p2
The minimum value of u , therefore, is 2 ⋅
(a + b2 + c2)
(ii) u = 2 x2 y + x2 − y2 + 2 y.
1 1
(iii) u = 2 sin (x + y) cos (x − y) + cos (x + y) .
2 2
(iv) u = sin x sin y sin (x + y) . (Bundelkhand 2011; Meerut 12B)
[Hint : It is sufficient to consider the values of x and y between 0 and π since the
function u is periodic with period π both for x and y ]
(v) u = x2 y2 − 5x2 − 8 x y − 5y2. (Avadh 2006)
D-194 DIFFERENTIAL CALCULUS
(vi) u = x4 + 2 x2 y − x2 + 3y2.
3. If u = a x3 y2 − x4 y2 − x3 y3, prove that x = a ⁄ 2 , y = a ⁄ 3 make u a maximum.
4. Investigate the maxima and minima of
2 (x − y) 2 − x4 − y4 ,
leaving aside any doubtful case that may arise. (Kanpur 2009)
5. E xamine the following surface for high and low points :
z = x2 + xy + 3x + 2y + 5 .
Find the saddle points of the surface if there are any.
6. Find all the stationary points of the function
x3 + 3x y2− 15x2 − 15y2 + 72 x ,
examining whether they are maxima or minima.
7. E xamine for maximum and minimum values of the function
z = x2 − 3x y + y2 + 2 x .
8. Find the points (x, y) where the function xy (1 − x − y) is maximum or minimum.
What is the maximum value of the function ?
9. E xamine the function z = x2 y − y2 x − x + y for maxima and minima. Find the
saddle points of the given surface.
10. L e t f (x, y) = x2 − 2 x y + y2 + x3 − y3 + x5. S h o w t h a t f (x, y) h a s n e i t h e r a
maximum nor a minimum at (0, 0) .
11. Find a point within a triangle such that the sum of the squares of its distances
from the three vertices is a minimum. (Kanpur 2010; Kumaun 08)
1 ,y= 1⋅
1. (i) Minimum at x = y = a . (ii) Maximum at x =
2 3
(iii) x = y = a gives a maximum or a minimum according as a is negative or positive.
(iv) Minimum at (− 3, 0). (v) Minimum at (1, 1).
2, 4
(vi) Minimum at x = y= − ⋅
3 3
2. (i) Maximum at (− 2, 0) . (ii) No exreme value.
(iii) Maximum when x = y = n π + (− 1) n π ⁄ 6
and minimum when x = y = 2 n π − π ⁄ 2 .
(iv) Maximum at x = y = π ⁄ 3 and minimum at x = y = 2 π ⁄ 3 .
(v) Maximum at x = y = 0.
(vi) Minimum when x = ± √3 ⁄ 2__, y = − 1 ⁄ 4 .
4. Maximum when x = ± √2 , y = + √2 .
5. No high and low points. The point (2, 1, 3) is a saddle point.
6. Maximum at (4, 0) and minimum at (6, 0).
7. The function z is stationary at the point (4 ⁄ 5, 6 ⁄ 5). But it is neither maximum nor
minimum at this point.
⎛ ⎞
1 , 1 ⎟ ⋅ Max. value = 1 ⁄ 27 .
8. Maximum at the point ⎜
⎝ 3 3 ⎠
MAXIMA AND MINIMA of Functions of Two Independent Variables D-195
9. The function is stationary at the points (1, 1) and (− 1, − 1), but it has no exreme
value. The points (1, 1, 0) and (− 1, − 1, 0) are the two saddle points of the surface
z = x2 y − y2 x − x + y.
11. Centroid of the triangle.
True or False:
Write ‘T ’ for true and ‘F ’ for false statem ent.
7. Let f (x, y) be a function of two independent variables x and y. Let
∂2 f , ∂2 f ∂2 f
r= s = and t = ⋅
∂x2 ∂x ∂y ∂y2
∂f ∂f
If at the point (a, b), we have = 0, = 0 and rt − s 2 < 0, then f (x, y) has an
∂x ∂y
extreme value at (a, b).
8. Let f (x, y) be a function of two independent variables x and y. If at (a, b), we have
∂f ∂f
= 0 and = 0, then f (x, y) must have a maximum or a minimum at (a, b).
∂x ∂y
p2
9. The minimum value of x2 + y2 + z 2 when ax + by + c z = p is 2 ⋅
(a + b2 + c 2)
⎛ ⎞
1 , 1⎟ ⋅
10. The function xy (1 − x − y) has a maximum value at the point ⎜
⎝ 2 2⎠
11. The function x3 + y3 + 3xy has a maximum value at the point (− 1, − 1).
12. The minimum value of the function x3 + y3 − 6 x y is − 8.
1. 0. 2. minimum. 3. maximum.
4. (b). 5. (c). 6. (a).
7. F. 8. F. 9. T.
10. F. 11. T. 12. T.
10.1 Tangent and Normal to a Curve
Tangent to a Curve : Let P be any given point on a
curve and Q any other point on it in the neighbourhood
of P. The point Q may be taken on either side of P. As Q
tends to P, the straight line PQ, in general, tends to a
definite straight line TP passing through P. This straight
line is called the tangent to the curve at the point P.
(Kanpur 2014)
Normal to a Curve : The normal to a curve at any
point P of it is the straight line which passes through that
point and is perpendicular to the tangent to the curve at
that point.
If (X, Y ) are current co-ordinates of any point on the chord PQ, then the equation of
the chord PQ is
(y + δy) − y δy
Y − y= (X − x) or Y − y= (X − x). …(1)
(x + δx) − x δx
Now, as Q tends to P, δx → 0 and chord PQ tends to the tangent at P. Therefore
the equation (1) tends to the equation
dy ⎡. . lim δy dy⎤
Y − y= (X − x). ⎢ . δx → 0 = ⎥.
dx ⎣ δx dx ⎦
Hence the equation of the tangent to the curve y = f (x) at the point (x, y) is
dy
Y− y= (X − x).
dx
Note 1 : If we are to find the equation of the tangent to the curve y = f (x) at the
point (x1, y1) on it, we should first find the value of dy ⁄ dx of the curve at the point
(x1, y1). The equation of the tangent at the point (x1, y1) will then be
⎛ dy⎞
y − y1 = ⎜ ⎟ (x − x1),
⎝ dx⎠ (x , y )
1 1
where (x, y) are the current co-ordinates of any point on the tangent.
Note 2 : If the equations of the curve be given in parametric cartesian form say
x = f ( t ) and y = φ ( t ), then
dy dy ⁄ dt φ ′ ( t )
= = ⋅
dx dx ⁄ dt f′ (t)
H ence the equation of the tangent at any point ‘t’ on the curve is given by
φ′ ( t )
Y− φ (t)= [X − f ( t )].
f ′ (t )
Example 1 : Find the equations of the tangent and the norm al at any point (x, y) of
xm ym
the curve m + m = 1.
a b
xm ym
Solution : Let f (x, y) ≡ m + m − 1 = 0.
a b
∂f mx m − 1 ∂f my m − 1
Then = and = ⋅
∂x am ∂y bm
dy ∂f ⁄ ∂x b m xm − 1
∴ = − = − m m − 1.
dx ∂f ⁄ ∂y a y
Hence the equation of the tangent at (x, y) is
b m xm − 1
Y − y= − (X − x)
am y m − 1
ym − 1 xm − 1
i.e., m
(Y − y) = − (X − x)
b am
Y ym − 1 X xm − 1 xm ym
i.e., m
+ m
= m+ m⋅
b a a b
D-200 DIFFERENTIAL CALCULUS
√ 1+⎧ ⎛ dx⎞⎟ 2 ⎫⎬
= PT = y cosec ψ = y√(1 + cot2 ψ) = y ⎨
⎩
⎜
⎝
.
dy⎠ ⎭
Length of sub-tangent
dx y
= TS = y cot ψ = y = ⋅
dy dy ⁄ dx
Length of normal
√ 1+
⎧ ⎛ dy⎞ 2⎫⎬
= PN = y sec ψ = y √ (1 + tan2 ψ) = y ⎨
⎩
⎜ ⎟ ⎭.
⎝ dx⎠
Length of subnormal
dy
= SN = y tan ψ = y ⋅
dx
Intercepts made by the tangent on the coordinate axes.
dy
The equation of the tangent at P (x, y) is Y − y = (X − x).
dx
D-202 DIFFERENTIAL CALCULUS
1
But sin 2 θ = − is inadmissible because it gives imaginary values of r from (1)
2
1 1
and (2). Now sin 2 θ = gives 2 θ = π or θ = π ⁄ 12.
2 6
H en ce, t h e an gle o f in t ersect io n o f (1) an d (2) at t he p o in t θ = π ⁄ 12 i s
π − 4 (π ⁄ 12) i.e., 2π ⁄ 3 .
1 dr
equation cot φ = ⋅
r dθ
1 1 1
Thus from (1), we have 2 = 2 cosec 2 φ = 2 (1 + cot2 φ )
p r r
2
1 ⎡ 1 ⎛ dr ⎞ 2⎤ 1 1 ⎛ dr ⎞
= 2 ⎢1+ 2 ⎜ ⎟ ⎥ = 2 + 4 ⎜ ⎟ ⋅
r ⎣ r ⎝ dθ⎠ ⎦ r r ⎝ dθ⎠
D-208 DIFFERENTIAL CALCULUS
2
1 1 1 ⎛ dr ⎞
H ence = + ⎜ ⎟ ⋅ …(2)
p2 r2 r4 ⎝ dθ⎠
(Meerut 2003)
1
Sometimes, we write u = ⋅
r
2
du 1 dr 1 1 ⎛ du ⎞
Then = − 2 ⋅ ∴ = + ⎜ ⎟
dθ r dθ p2 r2 ⎝ dθ ⎠
1 ⎛ ⎞2
2 + ⎜ du⎟ ⋅
or = u …(3)
p2 ⎝ dθ ⎠
Note : The results (1), (2) and (3) are very important and should be committed
to memory.
The relation between p and r for a given curve is called its pedal equation where r is
the radius vector of any point on the curve and p is the length of the perpendicular from
pole to the tangent at that point.
Case I : To form the pedal equation of a curve whose cartesian equation is given.
Let f (x, y) = 0, …(1)
be the cartesian equation of the given curve.
The equation of the tangent at (x, y) to the curve (1) is
dy dy ⎛ dy ⎞
Y − y= (X − x), or Y − X + ⎜ x − y⎟ = 0.
dx dx ⎝ dx ⎠
If p be the length of perpendicular from (0, 0) to this tangent, we have
dy
x − y
dx
p= ⋅ …(2)
√
⎧ ⎛ dy⎞ 2⎫
⎨1 + ⎜ ⎟ ⎬
⎩ ⎝ dx⎠ ⎭
Also r2 = x2 + y2. …(3)
E liminating x and y between (1), (2) and (3), we get a relation between p and r i.e.,
the required pedal equation of the given curve.
Case II : To form the pedal equation of a curve whose polar equation is given.
Let f (r, θ) = 0 …(1)
be the polar equation of the given curve.
We have p = r sin φ , …(2)
1 dr
and cot φ = ⋅ …(3)
r dθ
E liminating θ and φ between (1), (2) and (3), we obtain the required pedal
equation of the given curve.
Important : Sometimes we do not get the value of φ from equation (3) in a
convenient form. In that case instead of using the relations (2) and (3), we can use the
single relation
TANGENTS AND NORMALS D-209
2
1 1 1 ⎛ dr ⎞
= + ⎜ ⎟ . …(4)
p2 r2 r4 ⎝ dθ⎠
E liminating θ between (1) and (4), we obtain the required pedal equation of the
curve.
x2 y2
Example 1 : Show that the pedal equation of the ellipse 2 + 2 = 1, is
a b
1 1 1 r2
= + − ⋅ (Meerut 2010B; Avadh 13)
p2 a 2 b2 a 2 b2
x2 y2
Solution : The equation of the curve is + = 1. …(1)
a 2 b2
Th e co-ordinates (x, y) o f an y p o in t P on ( 1) ma y be t aken as x = a cos t,
y = b sin t.
dx dy dy b cos t
∴ = − a sin t, = b cos t. ∴ = − .
dt dt dx a sin t
H ence the equation of the tangent to the ellipse at the point ‘t’ is
b cos t
Y − b sin t = − (X − a cos t)
a sin t
or ab − b cos t . X − a sin t . Y = 0. …(2)
Therefore p = the length of perpendicular from (0, 0) to (2)
ab
=
√(a2 sin2 t + b2 cos2 t )
1 a2 sin2 t + b2 cos2 t
i.e., = ⋅ …(3)
p2 a 2 b2
Also r2 = x2 + y2 = a2 cos2 t + b2 sin2 t = a2 (1 − sin2 t ) + b2 (1 − cos2 t )
= a2 + b2 − a2 sin2 t − b2 cos2 t . …(4)
E liminating t between (3) and (4), we obtain the pedal equation of the given curve.
From (4), we get a2 sin2 t + b2 cos2 t = (a2 + b2) − r2.
1 (a2 + b2) − r2 1 1 1 r2
H ence (3) gives 2
= 2 2
or 2
= 2+ 2− 2 2,
p a b p a b a b
which is the required pedal equation of the ellipse.
Example 2 : Show that the pedal equation of the parabola y2 = 4a (x + a) is
2
p = ar. (Meerut 2010)
Solution : Differentiating the equation of the curve, we get
dy dy 2a
2y = 4a, or = ⋅
dx dx y
Therefore the equation of the tangent at (x, y) is
Y − y = (2a ⁄ y) (X − x) or (2a ⁄ y) X − Y + y − (2a ⁄ y) x = 0.
y − (2ax ⁄ y) y2 − 2ax
∴ p= =
√(1 + 4a ⁄ y ) √(y2 + 4a2)
2 2
D-210 DIFFERENTIAL CALCULUS
2a (x + 2a)
= = √[a (x + 2a)]
.
√[4a (x + 2a)]
Also r2 = x2 + y2 = x2 + 4a (x + a) = (x + 2a) 2.
∴ r = (x + 2a). ∴ p2 = a (x + 2a) or p2 = ar,
which is the required pedal equation of the given curve.
Example 3 : Form the pedal equation of the sine spiral r n = an sin n θ.
Solution : The curve is r n = an sin n θ. …(1)
Taking logarithm of both sides, we get n log r = n log a + log sin n θ.
Differentiating with respect to θ, we obtain
n dr cos nθ 1 dr
= n = n cot nθ. ∴ cot φ = = cot nθ.
r dθ sin nθ r dθ
Therefore φ = nθ.
Now p = r sin φ . ∴ p = r sin nθ. …(2)
E liminating θ between (1) and (2), we obtain the required pedal equation of the
given curve.
From (2), we have sin n θ = p ⁄ r.
Putting this value in (i), we obtain
r n = an ( p ⁄ r) or pan = r n + 1,
which is the required pedal equation.
l
5. Show that the pedal equation of the conic = 1 + e cos θ is
r
1 1 ⎛⎜ 2l ⎞
= ⎝
− 1 + e2⎟⎠ ⋅
p2 l2 r
6. Show that the pedal equation of the spiral r = a sech n θ is of the form
1 A
= + B.
p2 r2
7. Show that the pedal equation of the cardioid r = a (1 + cos θ) is r3 = 2ap2.
8. Show that the pedal equation of the astroid x2 ⁄ 3 + y2 ⁄ 3 = a2 ⁄ 3 is r2 = a2 − 3p2.
9. Show that the locus of the extremity of the polar subnormal of the curve
⎛ 1 ⎞
r = f (θ) is r = f ′ ⎜ θ − π⎟ ⋅ (Gorakhpur 2006)
⎝ 2 ⎠
H ence show that the locus of the extremity of the polar subnormal of the
equiangular spiral r = aemθ is another equiangular spiral.
10. Prove that the normal at any point (r, θ) to the curve r n = an cos n θ makes an
angle (n + 1) θ with the initial line.
1. l ⁄ e sin θ.
⎧ ⎛ dy⎞ 2⎫ ,
√ ⎨⎩ 1 +
ds
Thus = ± ⎜ ⎟ ⎬
dx ⎝ dx⎠ ⎭
where plus or minus sign is to be taken before the radical sign according as s increases
as x increases or decreases.
H ence if s increases as x increases, we have
√ ⎛ dy⎞ 2⎫
ds ⎧
= ⎨1 + ⎜ ⎟ ⎬⋅
dx ⎩ ⎝ dx⎠ ⎭
Corollary 1 : If x = f (y) be the equation of the curve, then ‘s’ is obviously a
function of y. In this case dividing (1) throughout by (δy) 2 and proceeding to limits, we
get
√ ⎛ dx⎞ 2⎫
ds ⎧
= ± ⎨1 + ⎜ ⎟ ⎬ ,
dy ⎩ ⎝ dy⎠ ⎭
where plus or minus sign is to be taken before the radical sign according as s increases
as y increases or decreases.
Corollary 2 : If the equations of the curve be given in the parametric form
x = f1 ( t ) and y = f2 ( t ), then ‘s’ is evidently a function of t. In this case dividing (1)
throughout by (δt ) 2 and proceeding to limits, we have
√
ds ⎧ ⎛ dx⎞ 2 ⎛ dy⎞ 2⎫
= ± ⎨⎜ ⎟ + ⎜ ⎟ ⎬, (Param etric formula)
dt ⎩ ⎝ dt ⎠ ⎝ dt ⎠ ⎭
where plus or minus sign is to be taken before the radical sign according as s increases
as t increases or decreases.
10.15 dx dy
cos ψ = and sin ψ =
ds ds
We know that tan ψ = dy ⁄ dx.
If s increases as x increases, we have
⎧ ⎛ dy⎞ 2⎫
√
ds
= ⎨ 1 + ⎜ ⎟ ⎬ = √(1 + tan2 ψ) = sec ψ.
dx ⎩ ⎝ dx⎠ ⎭
dx
∴ = cos ψ.
ds
Again if s increases as y increases, we have
√
ds ⎧ ⎛ dx⎞ 2⎫
= ⎨ 1 + ⎜ ⎟ ⎬ = √(1 + cot2 ψ) = cosec ψ.
dy ⎩ ⎝ dy⎠ ⎭
dy
∴ = sin ψ.
ds
Important : We can remember these results very easily
with the help of the adjoining hypothetical figure.
TANGENTS AND NORMALS D-213
⎧ 2 ⎛ dr ⎞ 2⎫
√ ⎨⎩ r +
ds
To prove that = ⎜ ⎟ ⎬ for the curve r = f (θ).
dθ ⎝ dθ ⎠ ⎭
√ dr ⎞⎟ 2⎫
ds ⎧
= ⎨ r2 + ⎛⎜ ⎬.
dθ ⎩ ⎝ dθ⎠ ⎭
Corollary : If θ = f (r) be the equation of the curve, then ‘s’ is a function of r and
we have
⎧ 2 ⎛ dr ⎞ 2⎫
ds ds dθ dθ
=
dr dθ dr
=
dr √
⎨r + ⎜ ⎟ ⎬
⎩ ⎝ dθ⎠ ⎭
√ ⎛ dθ⎞
ds ⎧ 2⎫
or = ⎨1
⎩
+ r2 ⎜⎝ ⎟⎠ ⎬⎭ .
dr dr
D-214 DIFFERENTIAL CALCULUS
10.17 dr dθ
cos φ = and sin φ = r
ds ds
We know that tan φ = r (dθ ⁄ dr).
1 1 1
∴ cos φ = = =
√(sec φ ) √(1 + tan2 φ )
√
2 ⎧ ⎛ dθ ⎞ 2⎫
⎨ 1 + r2 ⎜ ⎟ ⎬
⎩ ⎝ dr ⎠ ⎭
1 dr ⁄ dθ dr ,
= = =
ds ⁄ dθ ds
√
⎛ dθ ⎞ ⎧ 2 ⎛ dr ⎞ 2⎫
⎜ ⎟ ⎨r + ⎜ ⎟ ⎬
⎝ dr ⎠ ⎩ ⎝ dθ ⎠ ⎭
s being measured in such a way that s increases as θ increases.
dr
Thus cos φ = ⋅
ds
dθ dr dθ
Also sin φ = tan φ cos φ = r ⋅ = r ⋅
dr ds ds
dθ
H ence sin φ = r ⋅
ds
Important : We can remember these results very easily
with the help of the adjoining hypothetical figure.
ds ⎛ x⎞
Example 1 : For the curve y = a log sec (x ⁄ a) , prove that = sec ⎜ ⎟ .
dx ⎝ a⎠
Solution : We have y = a log sec (x ⁄ a).
Differentiating with respect to x, we get
dy 1 ⎛ x⎞ ⎛ x⎞ 1
= a⋅ sec ⎜ ⎟ tan ⎜ ⎟ ⋅
dx sec (x ⁄ a) ⎝ a⎠ ⎝ a⎠ a
⎛ x⎞
= tan ⎜ ⎟ .
⎝ a⎠
⎧ ⎛ dy⎞ 2⎫
√ ⎨⎩ 1 + ⎜⎝ dx⎟⎠ ⎬⎭
ds
Now =
dx
⎛ x⎞
√ 1 + tan
⎧ 2 ⎛⎜ x ⎞⎟ ⎫
= ⎨ ⎬ = sec ⎜ ⎟ ⋅
⎩ ⎝ a⎠ ⎭ ⎝ a⎠
Example 2 : For the ellipse x = a cos t, y = b sin t, prove that
ds ⁄ dt = a (1 − e2 cos2 t) 1 ⁄ 2.
dx dy
Solution : H ere = − a sin t and = b cos t.
dt dt
√
ds ⎧ ⎛ dx⎞ 2 ⎛ dy⎞ 2⎫
Now = ⎨⎜ ⎟ + ⎜ ⎟ ⎬
dt ⎩ ⎝ dt ⎠ ⎝ dt ⎠ ⎭
√
ds ⎧ 2 ⎛ dr ⎞ 2⎫
Now = ⎨r + ⎜ ⎟ ⎬
dθ ⎩ ⎝ dθ⎠ ⎭
d2 r
6. For the curve r n = an cos nθ, prove that a2 n 2 + nr 2 n − 1 = 0.
ds
D-216 DIFFERENTIAL CALCULUS
1. (i) (1 + a ⁄ x) 1 ⁄ 2,
(ii) cosh (x ⁄ a),
(iii) (a ⁄ x) 1 ⁄ 3.
2. (i) 2a cos (t ⁄ 2),
(ii) 3a cos t sin t,
(iii) 2 cos t √ (1 + 4 sin2 t ).
dθ 7 ,
6. If = at a point on the curve r = f (θ), then at that point polar subnormal
dr 3
is …… . (Meerut 2001)
True or False:
Write ‘T’ for true and ‘F’ for false statement.
11. If p be the length of perpendicular drawn from the pole O to tangent at any point
P (r , θ) on the curve r = f (θ), then
2
1 1 1 ⎛ dr ⎞
= 2 + 4 ⎜⎝ ⎟⎠ ⋅
p 2 r r dθ
12. The relation between p and r for a given curve is called its polar equation.
⎛ dr ⎞ 2 ⎛ d θ ⎞ 2
13. For the curve r = f (θ), we have ⎜⎝ ⎟⎠ + ⎜⎝ r ⎟⎠ = 1.
ds ds
14. p = r sin θ is the pedal equation of some curve.
D-218 DIFFERENTIAL CALCULUS
⎧ 2⎫
1. r
dθ
dr
⋅ 2. √ ⎪ 2
⎨r
⎪
⎩
⎛ dr ⎞ ⎪
+ ⎜⎝ ⎟⎠ ⎬⎪ ⋅
dθ ⎭
3. π −
1
2
θ.
1 a2 3
4. 2 a cos t. 5. ⋅ 6. ⋅
2 r 7
7. (b). 8. (c). 9. (d).
10. (a). 11. T. 12. F.
13. T. 14. F.
11.1 Meaning of Curvature
In the adjoining figure we see that the curve bends more
sharply at the point P than at the point Q. We express this feeling
by saying that the curve has a greater curvature at P than at Q.
H owever, in order to get a quantitative estimate of curvature, we
should give a mathematical definition of curvature which should be
in agreement with our intuitive notion of curvature.
Let P and Q be two neighbouring points on a curve, ψ and ψ + δψ the angles which
the tangents at P and Q make with the x-axis.
Let A be any fixed point on the curve.
Let arc AP = s, arc AQ = s + δs, so that arc PQ = δs.
The symbol, δψ denotes the angle through which the
tangent turns as a point moves along the curve from P to Q
through a distance δs. The angle δψ is called the contingence
of the arc PQ. O bviously, δψ will be large or small, as
compared with δs, depending on the degree of sharpness of
D-220 DIFFERENTIAL CALCULUS
the bend of the arc PQ. This suggests us to make the following definitions :
(i) δψ is defined to be total curvature of the arc PQ ;
δψ
(ii) the ratio is defined to be the average curvature of the arc PQ ;
δs
lim δψ dψ
(iii) the curvature of the curve at P is defined to be Q → P i.e., ⋅
δs ds
dψ
Thus is a m athem atical m easure for the curvature of curve at any point P.
ds
Lim
Therefore ρ = Q → P PN
⎛ Lim chord PQ ⎞ ⎛ Lim δs ⎞ ⎛ Lim δψ ⎞
= ⎜ Q →P ⎟ ⋅ ⎜ δψ → 0 δψ⎟ ⋅ ⎜ δψ → 0 ⎟
⎝ arc PQ ⎠ ⎝ ⎠ ⎝ sin δψ⎠
⎛ Lim ⎞
⋅ ⎜ Q → P sin NQP⎟
⎝ ⎠
⎛ ds ⎞ π ds
= 1 ⋅ ⎜ ⎟ ⋅ 1. sin = ⋅
⎝ dψ⎠ 2 dψ
ds
H ence, ρ = ⋅
dψ
Corollary : The curvature of the curve at any point P is by definition, equal to
dψ
. H ence the curvature of the curve at any point is equal to the reciprocal of the radius
ds
1
of curvature at that point i.e., curvature = ρ⋅
Example : Find the radius of curvature for the curve whose intrinsic equation is
⎛ π ψ⎞
s = a log tan ⎜ + ⎟ ⋅
⎝4 2⎠
Solution : We have ρ =
ds
= a
1 2⎛π ψ⎞ 1
dψ ⎛ π ψ ⎞ sec ⎜⎝ 4 + 2 ⎟⎠ ⋅ 2
tan ⎜ + ⎟
⎝4 2⎠
a a a
= ⎛ π ψ⎞ ⎛ π ψ⎞ = ⎛π ⎞ = cos ψ = a sec ψ.
2 sin ⎜ + ⎟ cos ⎜ + ⎟ sin ⎜ + ψ⎟
⎝4 2⎠ ⎝4 2⎠ ⎝2 ⎠
⎧ ⎛ dy⎞ 2⎫
√ ⎨⎩ 1 +
ds
= ⎜ ⎟ ⎬,
dx ⎝ dx⎠ ⎭
we take the positive sign before the radical, the value of ρ is positive or negative
d2 y
according as 2 is positive or negative. H owever, if we define the radius of curvature
dx
in such a way that it is to be always positive, then we should ignore the sign whenever
we get a negative value for ρ.
Note 2 : Sin ce t h e radiu s o f cu rvat u re is a len gt h t herefo re it s valu e is
independent of the choice of x-axis and y-axis. H ence interchanging x and y, we obtain
3⁄2
⎧ ⎛ dx⎞ 2⎫
⎨1 + ⎜ ⎟ ⎬
⎩ ⎝ dy ⎠ ⎭
ρ= ⋅
d2 x
dy2
This formula is specially useful when (dy ⁄ dx) is infinite i.e., when the tangent is
perpendicular to x-axis.
⎡ ⎧⎨ 1 + ⎛ dy⎞ 2⎫ 3 ⁄ 2 ⎤
⎜ ⎟ ⎬
⎢ ⎝ dx⎠ ⎭ ⎥
⎢⎩ ⎥
⎢ ⎥ (1 + 1) 3 ⁄ 2 3a
= ⎢ ⎥ = = − ⋅
⎢ d2 y ⎥ 32 1 8√2
⎢ ⎥ − .
⎢ dx2 ⎥ 3 a
⎣ ⎦ (3a ⁄ 2 , 3a ⁄ 2)
⎛ 3a 3a ⎞ 1 8 √2
∴ Curvature at ⎜ , ⎟ = ρ = − ⋅
⎝ 2 2 ⎠ 3a
⎛ 3a , 3a ⎞ 8 √2
If we ignore the negative sign, the value of curvature at ⎜ ⎟ = ⋅
⎝ 2 2⎠ 3a
Example 2 : If a curve is defined by the equations x = f ( t ) and y = φ ( t ) , prove
(x ′ 2 + y ′ 2) 3 ⁄ 2 ,
that the radius of curvature ρ is equal to where accents (i.e., dashes) denote
x ′ y ′′ − y ′ x ′′
differentiation with respect to t.
dy dy ⁄ dt φ ′ ( t ) y′
Solution : We have = = = ⋅
dx dx ⁄ dt f ′ ( t ) x′
d2 y d ⎛ y′⎞ ⎧ d ⎛ y ′ ⎞ ⎫ dt
Also = ⎜ ⎟ = ⎨ ⎜ ⎟⎬ ⋅ ⋅
dx2 dx x
⎝ ⎠′ ⎩ dt ⎝ x ′ ⎠ ⎭ dx
y ′′ x ′ − x ′′ y ′ 1 ⎡ . . dx dt 1⎤
= ⋅ ⎢ . = x ′ and = ⎥
x ′2 x′ ⎣ dt dx x ′⎦
y ′′ x ′ − x ′′ y ′
= ⋅
x ′3
⎧ ⎛ dy⎞ 2⎫ 3 ⁄ 2 ⎧ 2 ⎫3 ⁄ 2
⎪1 + y ′ ⎪
⎨1 + ⎜ ⎟ ⎬ ⎨ ⎬
⎩ ⎝ dx ⎠ ⎭ ⎪ x ′ 2 ⎪⎭
⎩ (x ′ 2 + y ′ 2) 3 ⁄ 2
H ence ρ= = = ⋅
d2 y y ′′ x ′ − x ′′ y ′ y ′′ x ′ − x ′′ y ′
dx2 x ′3
Example 3 : In th e cyclo id x = a (t + sin t ), y = a (1 − cos t ), prove that
1
ρ = 4a cos t. (Bundelkhand 2007; 12, 14)
2
dx dy
Solution : H ere = a (1 + cos t ) and = a sin t.
dt dt
dy dy ⁄ dt a sin t 2 sin t ⁄ 2 cos t ⁄ 2 t
∴ = = = = tan ⋅
dx dx ⁄ dt a (1 + cos t ) 2
2 cos t ⁄ 2 2
d2 y d ⎛ dy⎞ d ⎛ t⎞ 1 t dt
Also 2
= ⎜ ⎟ = ⎜ tan ⎟ = ⋅ sec2 ⋅
dx dx ⎝ dx⎠ dx ⎝ 2⎠ 2 2 dx
1 t 1 1 t 1 1 t
= sec 2 ⋅ = sec 2 ⋅ = sec4 ⋅
2 2 a (1 + cos t) 2 2 2a cos2 t ⁄ 2 4a 2
3⁄2
⎧ t⎫
⎨ 1 + tan2 ⎬ 4a sec3
t
⎩ 2⎭ 2 1
H ence ρ= = = 4a cos t ⋅
1 t t 2
sec4 sec4
4a 2 2
D-224 DIFFERENTIAL CALCULUS
q 2 ( y3) 0 3
y = 0 + 0. x + .x + x + … …(1)
2 3!
where ( y2) 0 = q.
2 2y 2
Multiplying (1) by 2 , we get 2 = q + ( y ) x+ … ...(2)
x x 3! 3 0
Lim 2y
Taking limit as x → 0 of both sides of (2), we get x → 0 2 = q.
x
(1 + 0) 3 ⁄ 2 1 Lim x2
Also in this case ρ at origin = = = x →0 ⋅
q q 2y
Therefore when x-axis is tangent to the curve at the origin,
Lim ⎛ x2 ⎞
ρ (at origin) = x → 0 ⎜ ⎟ ⋅
⎝ 2y⎠
Similarly it can be shown that if y-axis is tangent to the curve at the origin, then
Lim y2
ρ (at origin) = x → 0 ⋅
2x
These two formulae are known as Newton’s formulae.
(1 + 16) 3 ⁄ 2 85√17
∴ When p = 4, q = 2 ⁄ 5, ρ at origin = = −
− 2⁄5 2
(1 + 1) 3 ⁄ 2
and when p = − 1, q = 2 ⁄ 5, ρ at origin = = 5√2.
2⁄5
1. Find the radius of curvature at the point (s, ψ) on the following curves :
(i) s = c tan ψ (Catenary)
1
(ii) s = 8a sin2 ψ (Cardioid)
6
(iii) s = 4a sin ψ (Cycloid) (Bundelkhand 2001; Rohilkhand 08; Kashi 11)
D-228 DIFFERENTIAL CALCULUS
14. Find the radius of curvature at the origin of the following curves :
(i) y = x4 − 4x3 − 18x2. (ii) y = x3 + 5x2 + 6x.
15. Show that the radii of curvature of the curve a (y2 − x2) = x3 at the origin are
± 2a√2.
4 1
1. (i) c sec 2 ψ (ii) a sin ψ (iii) 4a cos ψ (iv) c tan ψ.
3 3
(a4 + 9x4) 3 ⁄ 2 ⎛ 2 ⎞
⎜ ⎟ (x + a) 3 ⁄ 2
(x2 + y2) 3 ⁄ 2
2. (i) (ii) ⎝ √a ⎠
(iii)
6a4x 2c2
1
(iv) (4a + 9x) 3 ⁄ 2 x1 ⁄ 2 (v) y2 ⁄ a (vi) c sec (x ⁄ c)
6a
2 (x + y) 3 ⁄ 2 (x 2 m − 2 + y 2 m − 2) 3 ⁄ 2
(vii) (viii) 3a1 ⁄ 3 x1 ⁄ 3 y1 ⁄ 3 (ix) ⋅
√a (1 − m) x m − 2 y m − 2
1
3. (i) √8. (ii) ⋅ 14. (i) 1 ⁄ 36, (ii) 37 √(37) ⁄ 10.
√2
1 1⎛ dφ ⎞ 1 d 1 dp
or ρ = r ⎜⎝ sin φ + r cos φ dr ⎟⎠ = r dr (r sin φ ) = r dr ⋅ [... p = r sin φ ]
dr
H ence ρ= r ⋅ (Meerut 2003, 07B)
dp
3⁄2
1 ⎧ ⎛ dr ⎞ 2⎫
r6 ⋅ 6 ⎨ r2 + ⎜ ⎟ ⎬
r ⎩ ⎝ dθ ⎠ ⎭
H ence ρ= ⋅
⎛ dr ⎞ 2 d2 r
r2
+ 2⎜ ⎟ − r 2
⎝ dθ⎠ dθ
⎧ 2 ⎛ dr ⎞ 2⎫ 3 ⁄ 2
⎨r + ⎜ ⎟ ⎬
⎩ ⎝ dθ⎠ ⎭
Therefore, ρ= 2
⋅
⎛ dr ⎞ d 2r
r2 + 2 ⎜ ⎟ − r 2
⎝ dθ ⎠ dθ
1 1
Corollary : If we put u = or r = , then
r u
2
dr 1 du d2 r 2 ⎛ du ⎞ 1 d2 u
= − 2 and = ⎜ ⎟ − ⋅
dθ u dθ dθ2 u3 ⎝ dθ ⎠ u2 dθ2
Putting these values in the polar formula for ρ, we get
⎧1 2 ⎫3 ⁄ 2
⎪ + u′ ⎪
⎨ 2 ⎬
⎪u
⎩ u4 ⎪⎭ (u2 + u′ 2) 3 ⁄ 2
ρ= = ,
1 2u′ 2 2u′ 2 u′′ u3 (u + u′′)
2
+ − + 3
u u4 u4 u
where dashes denote differentiation with respect to θ.
Note : We see that the pedal formula for ρ is simpler than the polar formula.
Therefore in case the equation of the curve is given in polar form, it is often convenient
to change it first to pedal equation and then to find ρ with the help of the pedal formula.
CURVATURE D-231
2
Example 1 : Show that for the cardioid r = a (1 + cos θ), ρ = √(2ar).
3
(Purvanchal 2006, 11; Rohilkhand 09B, 10; Agra 14)
Solution : The curve is r = a (1 + cos θ).
dr d2 r
∴ = − a sin θ and = − a cos θ.
dθ dθ2
{r2 + (dr ⁄ dθ) 2}3 ⁄ 2
Now ρ= 2
r + 2 (dr ⁄ dθ) 2 − r (d2 r ⁄ dθ2)
{a2 (1 + cos θ ) 2 + a2 sin2 θ}3 ⁄ 2
= 2
a (1 + cos θ) 2 + 2 (− a sin θ) 2 − a (1 + cos θ) (− a cos θ)
3⁄2
⎛ 2 41 2 21 21 ⎞
⎜ 4a cos 2 θ + 4a cos 2 θ sin 2 θ⎟
⎝ ⎠
=
a2 + 2a2 (cos2 θ + sin2 θ) + 3a2 cos θ
1 1 1 1
(4a2 cos2 θ) 3 ⁄ 2 [cos2 θ + sin2 θ]3 ⁄ 2 8a3 cos3 θ ⎡ 4a ⎤
2 2 2 2 1
= = = ⎢ ⎥ cos θ.
3a2 (1 + cos θ) 2
6a cos2 1
θ ⎣ ⎦ 3 2
2
1
But r = a (1 + cos θ) = 2a cos2 θ .
2
1
∴ cos θ = √(r ⁄ 2a).
2
√
4a ⎡ r ⎤⎥
H ence ρ= ⎢ = (2 ⁄ 3) √(2ar).
3 ⎣ 2a ⎦
Note : We could have solved this problem more easily by changing the equation
of the curve to pedal form.
Example 2 : Show that in the rectangular hyperbola r2 cos 2 θ = a2, ρ = r3 ⁄ a2.
Solution : The curve is r2 cos 2 θ = a2. …(1)
Taking logarithm of both sides of (1), we get 2 log r + log cos 2θ = 2 log a.
Differentiating with respect to θ, we get
2 dr 1
+ (− 2 sin 2θ) = 0
r dθ cos 2θ
1 dr ⎡π ⎤
or = cot φ = tan 2θ = cot ⎢ − 2θ⎥ .
r dθ ⎣2 ⎦
π
∴ φ = − 2 θ.
2
⎡π ⎤ a2
Now p = r sin φ = r sin ⎢ − 2θ⎥ = r cos 2θ. But cos 2θ = ⋅
⎣2 ⎦ r2
H ence the pedal equation of the curve is
a2 a2
p = r. 2 or p= ⋅
r r
dp a2
∴ = − 2⋅
dr r
D-232 DIFFERENTIAL CALCULUS
dr r3
H ence ρ= r = − 2⋅
dp a
r3
Neglecting the negative sign, we have ρ = ⋅
a2
Example 3 : Find the radius of curvature at the point ( p, r) on the ellipse
1 1 1 r2
2
= 2 + 2 − 2 2⋅
p a b a b
Solution : Differentiating the given equation with respect to r, we get
2 dp 2r dp r p3
− = − 2 2⋅ ∴ = 2 2⋅
p3 dr a b dr a b
dr a 2 b2 a 2 b2
H ence ρ= r = r⋅ = ⋅
dp r p3 p3
2p3 2
1. (i) (ii) √(2ar) (iii) a2 ⁄ 3r
a2 3
a 2 b2
(iv) (r2 + a2) 3 ⁄ 2 / (r2 + 2a2). 2. ⋅
p3
an
4. (i) a ⁄ 2 (ii) 2√(r3 ⁄ a) (iii)
(n + 1) r n − 1
D-234 DIFFERENTIAL CALCULUS
an 2 a2
(iv) (v) √(2ar) (vi) ⋅
(n + 1) r n − 1 3 3r
10. 0.
d2 y ⎧ ⎛ dy⎞ 2 ⎫
or ( β − y) 2
− ⎨ ⎜ ⎟ + 1⎬ = 0.
dx ⎩ ⎝ dx⎠ ⎭
⎛dy ⎞2
1 + ⎜⎝ ⎟⎠
dx
∴ β= y+ ⋅
d2 y
dx2
CURVATURE D-235
Example 1 : Find the co-ordinates of the centre of curvature for the point (x, y) on
the parabola y2 = 4ax.
A lso find the equation of the evolute of the parabola.
dy
Solution : H ere 2y = 4a, i.e.,
dx
dy 2a
dx
=
y
= a1 ⁄ 2 x− 1 ⁄ 2 = √
⎛ a⎞
⎜ ⎟⋅
⎝ x⎠
d2 y1 1⁄2 − 3⁄2
∴ a= − x .
dx2 2
If (α, β) be the centre of curvature for the point (x, y), then
dy ⎧ ⎛ dy⎞ 2⎫
√
⎛ a⎞ ⎧ a⎫
⎨1 + ⎜ ⎟ ⎬ ⎜ ⎟ ⎨1 + ⎬
dx ⎩ ⎝ dx ⎠ ⎭ ⎝ ⎠
x ⎩ x⎭
α= x− = x− = x + 2x (1 + a ⁄ x)⋅
√
2
d y ⎛ a⎞
11 ⎜ ⎟
−
dx2 2 x ⎝ x⎠
∴ α = 3x + 2a, ...(1)
⎛ dy⎞ 2 ⎛ a⎞
1+ ⎜ ⎟ 1+ ⎜ ⎟
⎝ dx⎠ ⎝ x⎠
and β= y+ = y+
d2 y
dx2
−
11
2 x √⎛ a⎞
⎜ ⎟
⎝ x⎠
D-236 DIFFERENTIAL CALCULUS
= 2a1 ⁄ 2 x1 ⁄ 2 − 2a− 1 ⁄ 2 x3 ⁄ 2 (1 + a ⁄ x)
= 2a1 ⁄ 2 x1 ⁄ 2 − 2a− 1 ⁄ 2 x3 ⁄ 2 − 2a1 ⁄ 2 x1 ⁄.2
⎛ x⎞
∴ β = − 2x
√ ⎜ ⎟ ⋅
⎝ a⎠
…(2)
⎛ ⎞
√
⎜ ⎛ x⎞ ⎟
Therefore the required centre of curvature is ⎜⎜ (3x + 2a), − 2x ⎜ ⎟ ⎟⎟
⎝ a⎠ ⎠
⋅
⎝
Evolute of the parabola : Let us eliminate x between (1) and (2). From (2),
we get
4 x3 aβ2
β2 = or x3 = ⋅
a 4
α − 2a
From (i), we get x= ⋅
3
⎛ α − 2a ⎞ 3 aβ2
∴ ⎜ ⎟ = or 27 aβ2 = 4 (α − 2a) 3.
⎝ 3 ⎠ 4
H ence the locus of (α, β) is 27 ay2 = 4 (x − 2a) 3, which is the evolute of parabola.
Example 2 : Prove that the co-ordinates (α, β) of the centre of curvature at any point
(x, y) can be expressed in the form
dy dx
α= x− and β = y + ⋅
dψ dψ
Solution : L e t C (α, β) b e t h e ce n t r e o f
curvature of the point P (x, y) on the curve y = f (x).
The line TP is tangent to the curve at P and obviously
PC is normal at P to the curve, since the centre of
curvature is a point on the normal.
ρ = radius of curvature at P = PC and CN is the
o r d in a t e o f C. D r a w PM p e rp e n dicu la r t o CN .
O bviously
1
∠ CPM = π − ψ.
2
Then α = x − PM = x − ρ sin ψ
dy ds ⎡.. ds dy ⎤
= x− ⋅ ⎢ . ρ= and = sin ψ⎥
ds dψ ⎣ dψ ds ⎦
dy
= x− ⋅
dψ
Also β= y+ CM = y + ρ cos ψ
dx ds ⎡.. dx⎤
= y+ . ⎢ . cos ψ = ⎥
ds dψ ⎣ ds ⎦
dx
= y+ ⋅
dψ
Example 1 : Show that the chord of curvature through the pole of the curve
rn = an cos n θ is 2r ⁄ (n + 1). (Purvanchal 2014)
Solution : The curve is r n = an cos n θ. …(1)
Taking logarithm of both sides, we get n log r = n log a + log cos n θ.
D-238 DIFFERENTIAL CALCULUS
n dr n
Differentiating with respect to θ, we get = − sin n θ
r dθ cos nθ
⎛ ⎞
1
i.e., cot φ = − tan n θ = cot ⎜⎝ π + n θ⎟⎠ ⋅
2
1
∴ φ= π + n θ.
2
⎛ ⎞
1
Now p = r sin φ = r sin ⎜⎝ π + n θ⎟⎠ = r cos n θ.
2
1. In the parabola x2 = 4ay, prove that the co-ordinates of the centre of curvature
⎛ x3 , 3x2 ⎞⎟
are ⎜⎜ − 2a + ⎟ ⋅
⎜ 4a 2 4a ⎟
⎝ ⎠
2. In the catenary y = c cosh (x ⁄ c), show that the centre of curvature (α, β) is given
by α = x − y {( y2 ⁄ c2) − 1)}1 ⁄ 2, β = 2y.
3. For the curve a2 y = x3, show that the centre of curvature (α, β) is given by
x ⎧⎪ 9x4 ⎫⎪ 5x3 a2
α= ⎨1 − 4 ⎬ , β = + ⋅
2 ⎪⎩ a ⎪⎭ 2a2 6 x
4. Show that the centre of curvature (α, β) at the point determined by t on the ellipse
a 2 − b2 a 2 − b2
x = a cos t, y = b sin t, is given by α = cos3 t, β = − sin3 t.
a b
Also show that the evolute of the ellipse is (ax) 2 ⁄ 3 + (by) 2 ⁄ 3 = (a2 − b2) 2 ⁄ 3.
5. Prove that the centre of curvature (α, β) for the curve x = 3t, y = t 2 − 6
4 3 3
is α = − t , β = 3t 2 − ⋅
3 2
6. Show that in any curve the chord of curvature perpendicular to the radius vector
is 2ρ √(r2 − p2) ⁄ r.
7. Show that the chord of curvature through the pole of the equiangular spiral
r = ae m θ is 2r.
8. Show that the chord of curvature, through the pole, for the cardioid
4
r = a (1 + cos θ) is r.
3
9. Show t hat t he circle of curvature at the point (am 2, 2am) of the parabola
y2 = 4ax has for its equation
x2 + y2 − 6am 2 x − 4ax + 4am 3 y − 3a2 m 4 = 0.
10. If Cx and Cy be the chords of curvature parallel to the axes at any point of the curve
1 1 1
y = ae x ⁄ a, prove that 2 + 2
= ⋅
Cx Cy 2aC x
(Agra 2007; Rohilkhand 07; Purvanchal 07)
6. For a curve defined by the equations x = f (t) and y = φ (t) the radius of curvature
is …… .
Multiple Choice Questions:
Indicate the correct answer for each question by writing the corresponding letter from
(a), (b), (c) and (d).
7. If y-axis is the tangent to the given curve at the origin, then radius of curvature at
the origin is equal to
x2 y2 x2 y2
(a) Lim (b) Lim (c) Lim (d) Lim
x →0 2 y x →0 2 x x →0 y x →0 x
8. Pedal formula for radius of curvature is
1 dr dr 1 dp dp
(a) (b) r (c) (d) r
r dp dp r dr dr
9. Chord of curvature parallel to y-axis is
(a) 2 ρ sin φ (b) 2 ρ cos φ (c) 2 ρ sin ψ (d) 2 ρ cos ψ
True or False:
Write ‘T’ for true and ‘F’ for false statement.
10. There is no difference between curvature of the circle and circle of curvature.
(Meerut 2003)
11. The curvature of the curve at any point is equal to the reciprocal of the radius of
curvature at that point.
⎧ 2 ⎫3 ⁄ 2
dr ⎪ ⎛ ⎞ ⎪
⎨r2
+ ⎜⎝ ⎟⎠ ⎬⎪
⎪
dθ ⎭ ⎩
12. The polar formula for radius of curvature is : ρ = ⋅
2
⎛ d r⎞ 2
⎜ ⎟
d2r
r + 2⎝ ⎠ − r 2
dθ dθ
d2p
13. The tangential polar formula for radius of curvature is ρ = p + ⋅
d ψ2
dr
14. Pedal formula for radius of curvature is ρ = r ⋅ (Agra 2006)
dp
⎡ 2 ⎤3⁄2
⎢ ⎛ dy ⎞ ⎥
⎢ ⎜ ⎟ ⎥
⎢1 + ⎜ ⎟ ⎥
⎜ ⎟
dψ ⎣ ⎝ dx ⎠ ⎦ ds
1. intrinsic. 2. ⋅ 3. ⋅ 4. ⋅
ds d2y dψ
dx 2
1 (x ′ 2 + y ′ 2) 3 ⁄ 2 ,
5. 4a cos t. 6. where x ′ y ′′ − y ′ x ′′ ≠ 0.
2 x ′ y ′′ − y ′ x ′′
7. (b). 8. (b). 9. (d). 10. F.
11. T. 12. T. 13. T. 14. T.
12.1 One Parameter Family of Curves
An equation of the form
F (x, y, α) = 0 ...(1)
in which α is a constant, represents a curve. If α is a parameter i.e., if α can take all real
values, then (1) is the equation of a one parameter family of curves with param eter α . If
we give different values to α we get different members of this family. O n any particular
curve belonging to this family the value of α is constant but it changes from one curve
to another.
An equation of the type F (x, y, α, β) = 0 also defines a family of curves but in this
case we have two parameters α and β .
Illustrations :
(i) The equation x cos α + y sin α = a determines a family of straight lines, and
α is the parameter of this family.
(ii) The equation y = m x − 2 am − am 3 determines a family of straight lines
which are normals to the parabola y2 = 4ax . H ere m is the parameter.
Su p p o se P i s a p o i n t o f i n t e r s e c t i o n o f t wo m e m b e r s F (x, y, α) = 0 a n d
F (x, y, α + δα) = 0 of t his family corresponding to t he paramet er values α a n d
α + δα . As δα → 0 , let P tend to a definite point Q on the member α . The locus of Q
(for varying values of α ) is called the envelope of the family.
The above definition can be given in concise form as below :
The envelope of a one param eter fam ily of curves is the locus of the limiting positions
of the points of intersection of any two m em bers of the fam ily when one of them tends to
coincide with the other which is kept fixed.
Thus the envelope of a family of curves is the locus of the points of intersection
of consecutive members of the family.
Example 1 : Find the envelope of the fam ily of straight lines y = mx + (a ⁄ m) , the
param eter being m . (Kanpur 2005)
ENVELOPES, EVOLUTES AND INVOLUTES D-243
Example 2 : Find the envelope of the fam ily of straight lines x ⁄ a + y ⁄ b = 1 , where
the two param eters a , b are connected by the relation ab = c2, c being a constant.
(Kumaun 2002; Meerut 12)
Solution : The equation of the given family of straight lines is
x⁄ a + y⁄ b = 1, ...(1)
where the parameters a , b are connected by the relation
ab = c2. ...(2)
We shall eliminate one parameter, say b .
From (2), we have b = c2 ⁄ a . Putting the value of b in (1), we get
x y x ay
+ = 1 or + 2 = 1 , ...(3)
a c2 ⁄ a a c
which is the equation of the given family and it contains only one parameter a .
We can arrange (3) as a quadratic in a . Thus (3) can be written as
c2 x + a2 y = ac2 or a2 y − ac2 + c2 x = 0 . ...(4)
The equation (4) is a quadratic in the parameter a . So the required envelope is
ENVELOPES, EVOLUTES AND INVOLUTES D-245
9. Find the envelope of the straight lines x cos α + y sin α = l sin α cos α , where α
is the parameter. Give the geometrical interpretation.
(Rohilkhand 2012; Avadh 07, 12)
10. Find the envelope of the family of straight lines
x cos mα + y sin mα = a (cos nα) m ⁄ n ,
where α is the parameter. [Hint. Change the equation to polar co-ordinates by
substituting x = r cos θ, y = r sin θ.]
11. Show that the radius of curvature of the envelope of the line
x cos α + y sin α = f (α) is f (α) + f ′′ (α) .
x y
12. If x2 ⁄ 3 + y2 ⁄ 3 = k 2 ⁄ 3 is the envelope of the lines + = 1, then find the
a b
necessary relation between a, b and k .
13. Find the envelope of the family of curves x2 sin α + y2 cos α = a2,
where α is the parameter.
2
14. Find the envelope of the family of curves ( y − c) 2 − (x − c) 3 = 0, where c is
3
the parameter.
1. x2 ⁄ a2 + y2 ⁄ b2 = 1.
E ach line of the given family is a tangent to the ellipse x2 ⁄ a2 + y2 ⁄ b2 = 1.
2. (i) y2 = 4 x . (ii) x2 + y2 = a2 .
(iii) 4 x3 + 27ay2 = 0. (iv) ( p − 1) p − 1 x p + p p ay p − 1 = 0.
2 2
(v) x − y = c . 2 (vi) a2 (x2 + y2) = x2 y2.
3. 2 2 2 2 2 2
(x + y + c ) = 4a (x + y ). 2
of any member of the family can be obtained. In such cases we should first find the
equation of the family of curves in a proper form and then we should find the envelope.
Example 1 : Find the envelope of the circles which pass through the origin and
whose centres lie on the ellipse x2 ⁄ a2 + y2 ⁄ b2 = 1 .
Solution : Any point on the ellipse x2 ⁄ a2 + y2 ⁄ b2 = 1 is (a cos θ, b sin θ) .
Its distance from the origin is √(a2 cos2 θ + b2 sin2 θ) .
Therefore the equation of the given family of circles is
(x − a cos θ) 2 + ( y − b sin θ) 2 = a2 cos2 θ + b2 sin2 θ
or x2 + y2 − 2ax cos θ − 2by sin θ = 0 . ...(1)
We are to find the envelope of the family of circles (1), where θ is the parameter.
The equation (1) may be written as
2ax cos θ + 2by sin θ = x2 + y2. ...(2)
Differentiating (2) partially with respect to θ , we get
− 2ax sin θ + 2by cos θ = 0 . ...(3)
Squaring and adding (2) and (3), we get
4a2 x2 + 4b2 y2 = (x2 + y2) 2
or (x2 + y2) 2 = 4 (a2 x2 + b2 y2) ,
which is the required envelope.
Example 2 : Find the envelope of the circles drawn on the radii vectors of the
parabola y2 = 4ax as diam eter.
Solution : Any point on the parabola y2 = 4ax is (at2, 2at) . E quation of the circle
drawn on the line joining the origin (0, 0) to the point (at2, 2at) as diameter is
(x − 0) (x − at2) + ( y − 0) ( y − 2at) = 0
or x2 + y2 − axt2 − 2aty = 0 . ...(1)
We are to find the envelope of the family of circles (1), where t is the parameter.
Differentiating (1) partially with respect to t , we get
0 − 2axt − 2ay = 0 . ...(2)
E liminating t between (1) and (2), we get the required envelope. From (2), we get
t = − y⁄ x.
Putting this value of t in (1), we get
x2 + y2 − ax .( y2 ⁄ x2) + 2ay .( y ⁄ x) = 0
or x2 + y2 − (ay2 ⁄ x) + (2ay2 ⁄ x) = 0 or x2 + y2 + (ay2 ⁄ x) = 0
or ay2 + x (x2 + y2) = 0 , which is the required envelope.
Example 3 : Find the envelope of the circles drawn on the radii-vectors of the
curve r n = an cos nθ as diam eter.
Solution : Let P be any point on the curve r n = an cos nθ . If α is the vectorial
angle of P , then the radius vector OP is given by
(OP) n = an cos nα .
∴ OP = a (cos nα) 1 ⁄ n .
D-250 DIFFERENTIAL CALCULUS
xm ym
Example 1 : Find the envelope of the fam ily of curves m + m = 1 , where the
a b
param eters a and b are connected by the relation a p + b p = c p.
Solution : The equation of the given family of curves is
xm ym
+ = 1, ...(1)
am bm
where the parameters a and b are connected by the relation
a p + b p = c p. ...(2)
Since there is a relation between a and b , therefore we shall regard b as a function
of a . Now we shall differentiate (1) and (2) with respect to a regarding x and y as
constants and b as a function of a .
From (1), we get
m xm m y m db db x m ⁄ am + 1
− m+ 1− m+ 1 = 0 i.e, = − m m+ 1⋅ ...(3)
a b da da y ⁄b
Again from (2), we get
pa p − 1 + pb p − 1 (db ⁄ da) = 0 i.e., db ⁄ da = − a p − 1 ⁄ b p − 1. ...(4)
E quating the two values of (db ⁄ da) , we get
x m ⁄ am + 1 a p − 1 x m ⁄ am ap ...(5)
m m + 1
= p − 1
or m m
= p⋅
y ⁄b b y ⁄b b
E liminating a and b between (1), (2) and (5), we get the required envelope. From
(5), we have
x m ⁄ am y m ⁄ bm x m ⁄ a m + y m ⁄ bm 1
p
= p
= = p⋅ [ Note]
a b ap + bp c
D-252 DIFFERENTIAL CALCULUS
∴ xm ⁄ a p + m = 1 ⁄ c p
or a p + m = xm c p
or a = (x m c p) 1 ⁄ ( p + m)
2
or a p = (x m c p) p ⁄ ( p + m) = x m p ⁄ ( p + m) c p ⁄ ( p + m) .
2
Similarly b p = y mp ⁄ ( p + m) c p ⁄ ( p + m) .
Substituting these values of a p and b p in (2), we get
2
c p ⁄ ( p + m) {x m p ⁄ ( p + m) + y m p ⁄ ( p + m) } = c p
2
or x m p ⁄ ( p + m) + y m p ⁄ ( p + m) = c p − p ⁄ ( p + m)
or x m p ⁄ ( p + m) + y m p ⁄ ( p + m) = c m p ⁄ ( p + m) ,
which is the required envelope.
1. Find the envelope of the circles drawn upon the radii vectors of the ellipse
x2 ⁄ a2 + y2 ⁄ b2 = 1 as diameter.
2. Show that the envelope of the circles whose centres lie on the parabola y2 = 4ax
and which pass through its vertex is the cissoid y2 (2 a + x) + x3 = 0.
3. Show that the envelope of the circles whose centres lie on the rectangular
hyperbola x2 − y2 = a2 and wh ich p ass through the origin is the lemniscate
r2 = 4a2 cos 2θ.
4. Show t hat t he en velo pe of th e circles described on t he cent ral radii of a
rectangular hyperbola as diameters is a lemniscate
r2 = a2 cos 2θ.
x2 y2
5. Show that the envelope of the polars of points on the ellipse 2 + 2 = 1 with
h k
x2 y2 h2 x2 k 2 y2
respect to the ellipse 2 + 2 = 1 is + = 1. (Bundelkhand 2011)
a b a4 b4
6. Show that the envelope of the straight line joining the extremities of a pair of
conjugate diameters of the ellipse
1
x2 ⁄ a2 + y2 ⁄ b2 = 1 is the ellipse x2 ⁄ a2 + y2 ⁄ b2 = ⋅
2
7. Find the envelopes of circles described on the radii vectors of the following curves
as diameters
(i) l ⁄ r = 1 + e cos θ, (ii) r3 = a3 cos 3θ, (iii) r cos n (θ ⁄ n) = a .
8. Find the envelopes of the straight lines drawn at right angles to the radii vectors
of the following curves through their extremities :
(i) r = ae θ cot α,
(ii) r n = a n cos nθ,
(iii) r = a + b cos θ.
9. Find the envelope of the straight line x ⁄ a + y ⁄ b = 1, where the parameters
a and b are connected by the following relations
ENVELOPES, EVOLUTES AND INVOLUTES D-253
We define the evolute of a curve as the locus of the centre of curvature for that
curve.
Evolute as the envelope of the normals : The centre of curvature of a curve for
a given point P on it is the limiting position of the intersection of the normal at P with
the normal at any other consecutive point Q as Q → P . Therefore by the definition of
D-254 DIFFERENTIAL CALCULUS
envelope, the envelope of the normals to a curve is the evolute of that curve. Hence,
the evolute of a curve is the envelope of the normals to that curve.
Theorem : The norm al at any point of a curve is a tangent to its evolute touching
at the corresponding centre of curvature.
Proof : The co-ordinates (α, β) of the centre of curvature for any point P (x, y)
on the given curve are given by α = x − ρ sin ψ ; β = y + ρ cos ψ .
Differentiating these with respect to x, we get
dα dψ dρ
= 1 − ρ cos ψ − sin ψ
dx dx dx
ds dx d ψ dρ dρ
= 1− ⋅ ⋅ − sin ψ = − sin ψ …(1)
d ψ ds dx dx dx
d β dy dψ dρ
and = − ρ sin ψ + cos ψ
dx dx dx dx
dy ds dy d ψ dρ dρ
= − ⋅ ⋅ + cos ψ = cos ψ ⋅ …(2)
dx d ψ ds dx dx dx
From (1) and (2), we have
dβ
= − cot ψ …(3)
dα
dβ
But is the slope of the tangent at
dα
Q to the evolute and − cot ψ is the slope of
the normal PQ at P to the given curve. These
two slopes are equal, and Q is a common
point on both the lines. H ence the tangent
at Q to the evolute and the normal at P to the
given curve coincide i.e., the normal at P to
the given curve touches its evolute at the
corresponding point.
dρ 1 ⎡ . . dx dψ 1 ⎤
= cos ψ − sin ψ − ρ cos ψ ⋅ ρ ⎢ . = cos ψ , = ρ⎥
ds ⎣ ds ds ⎦
= − (dρ ⁄ ds) sin ψ . ...(3)
Differentiating both sides of (2), w.r.t. ‘s’, we have
dβ dy dρ dψ
= + cos ψ − ρ sin ψ
ds ds ds ds
dρ 1 ⎡ . . dy ⎤
= sin ψ + cos ψ − ρ sin ψ ⋅ ρ ⎢ . = sin ψ⎥
ds ⎣ ds ⎦
dρ
= cos ψ . ...(4)
ds
Squaring and adding (3) and (4), we get
⎛ dα⎞ 2 ⎛ dβ⎞ 2 ⎛ dρ⎞ 2
⎜ ⎟ + ⎜ ⎟ = ⎜ ⎟ ⋅
⎝ ds ⎠ ⎝ ds ⎠ ⎝ ds ⎠
⎡⎢ ⎛ dα⎞ 2 2 ⎤⎥
√
⎛ dβ⎞ dρ ...(5)
∴ ⎢⎢ ⎜ ⎟ + ⎜ ⎟ ⎥⎥ = ⋅
⎣ ⎝ ds ⎠ ⎝ ds ⎠ ⎦ ds
Now s′ denotes the arc length of the locus of the point (α , β). R egarding α , β as
the functions of the parameter s, we have
⎡⎢ ⎛ dα⎞ 2 2 ⎤⎥
√
ds ′ ⎛ dβ⎞
= ⎢⎜ ⎟
⎢⎣ ⎝ ds ⎠ + ⎜ ⎟ ⎥
⎥⎦ ...(6)
ds ⎝ ds ⎠
From (5) and (6), we have ds ′ ⁄ ds = dρ ⁄ ds .
∴ ds ′ = dρ ...(7)
L e t Q 1 and Q 2 be t he point s on t he evolut e corresponding t o t he point s
P1 and P2 on the given curve. Then integrating (7) between these points, we get
⎡ ⎤ Q 2 ⎡ ⎤ P2
⎢ s′ ⎥ = ⎢ ρ⎥
⎣ ⎦Q ⎣ ⎦P
1 1
i.e., (s ′ at Q 2 ) − (s ′ at Q 1 ) = ( ρ at P2 ) − ( ρ at P1 )
i.e., the arc length of the evolute from Q 1 to Q 2 = the difference between the radii of
curvature of the given curve at the points P1 and P2 .
i.e., r ′ 2 = r2 + ρ2 − 2r ρ sin φ
i.e., r ′ 2 = r2 + ρ2 − 2ρp . [... p = r sin φ ] . ...(2)
Now OTPM is a rectangle and so MP = OT = p . From the right angled triangle
OMP , we have
OM 2 = OP2 − MP2
i.e., p′ 2 = r 2 − p2. ...(3)
dr
Also ρ= r ⋅ ...(4)
dp
E liminating r , p and ρ between the equations (1) , (2) , (3) and (4) , we get the pedal
equation of the evolute.
Example 2 : Prove that the evolute of the cardioid r = a (1 + cos θ) is the cardioid
1 2
r= a (1 − cos θ) , the pole of the latter equation being at the point ( a, 0) .
3 3
Solution : The given cardioid is
r = a (1 + cos θ) . ...(1)
Let O be the pole and OX the initial line. The cardioid (1) has been drawn in the
figure.
Take any point P (r, θ) on the given cardioid (1). Also let O′ be the given point
2
( a, 0) [this will be on the initial line because θ = 0].
3
ENVELOPES, EVOLUTES AND INVOLUTES D-259
Let C be the centre of curvature of the curve (1) corresponding to the point P .
Then CP = ρ = radius of carvature of (1) at the point P .
We have to find the locus of the point C w.r.t. O′ as pole. Let O′C = r1 and
∠ CO′X = θ1 . Draw CM and O′N perpendiculars from C and O′ respectively to OP .
Let us first find the value of ρ . Taking logarithm of both sides of (1), we get
log r = log a + log (1 + cos θ) .
Differentiating w.r.t. ‘θ’, we get
1 1
− sin θ − 2 sin θ cos θ
1 dr 2 2
cot φ = = =
r dθ 1 + cos θ 2
2 cos θ
1
2
1 1 1
= − tan θ = cot ( π+ θ) .
2 2 2
1 1
∴ φ= π+ θ.
2 2
1 1 1
Now p = r sin φ = r sin ( π + θ) = r cos θ .
2 2 2
1
From (1), r = 2 a cos2 θ = 2 a ( p ⁄ r) 2.
2
∴ the pedal equation of the curve (1) is
r3 = 2ap2. ...(2)
Differentiating (2) w.r.t. ‘p’, we get
3r2 (dr ⁄ dp) = 4ap .
1⁄2
dr 4ap 4a ⎛ r3 ⎞ ⎡ .. r3 ⎤
∴ ρ= r = = ⎜ ⎟ , ⎢ . from (2), p2 = ⎥
dp 3r 3r ⎝ 2a ⎠ ⎣ 2a ⎦
4a ⎛ r ⎞ 1 ⁄ 2 4a ⎧ a (1 + cos θ) ⎫ 1 ⁄ 2
= ⎜ ⎟ = ⎨ ⎬
3 ⎝ 2a ⎠ 3 ⎩ 2a ⎭
⎛ 1 ⎞1⁄2
⎜ 2a cos2 θ ⎟
4a ⎜ 2 ⎟
= ⎜ ⎟
3 ⎝ 2a ⎠
4a 1
= cos θ. …(3)
3 2
D-260 DIFFERENTIAL CALCULUS
1 1
Since φ = π+ θ,
2 2
1
∴ ∠ CPM = θ.
2
1 1 4 1 1 2
∴ CM = PC sin θ = ρ sin θ = a cos θ sin θ = a sin θ .
2 2 3 2 2 3
2
∴ Also O′N = OO′ sin θ = a sin θ .
3
Thus CM = O′N and consequently O′N MC is a rectangle. Therefore O′C is
parallel to OP ,
i.e., θ1 = θ. ...(4)
Now r1 = O′C = N M = OP − ON − PM
1
= OP − OO′ cos θ − PC cos θ
2
2 4 1 1
= r− a cos θ − a cos θ cos θ
3 3 2 2
2 2
= a (1 + cos θ) − a cos θ − a (1 + cos θ)
3 3
1
= a (3 + 3 cos θ − 2 cos θ − 2 − 2 cos θ)
3
1
= a (1 − cos θ)
3
[... θ1 = θ]
1
= a (1 − cos θ1).
3
E vo l u t e i s t h e l o cu s o f t h e ce n t r e o f cu r va t u r e . H e n ce ge n e r a li si n g
1 1
r1 = a (1 − cos θ1) , the required equation of the evolute is r = a (1 − cos θ) referred
3 3
to O′ as pole.
12.11 Involutes
Definition : If one curve is the evolute of another, then the latter is called an involute
of the form er. Thus a curve C1 is an involute of a given curve C , if the curve C is the
evolute of C1 .
Theorem : Every curve has an infinite number of involutes.
Let C be the given curve. Take an inextensible thin string and tie one end of this
string to a fixed point, say A , of the curve C and then wrap the string round the convex
side of C, keeping it taut all the while. Then any point on the string will describe an
involute of C, since at each instant the free part of the string is a tangent to C whereas
the direction of motion of the point on the
string is at right angles to this tangent i.e., this
tangent is along the normal to the locus of that
point on the string. In the figure the locus of
the point P1 on the string is the curve C1 which
is an involute of C and the locus of the point
P1′ on the string is the curve C2 which is also an
involute of C . Similarly by taking some other
point on the string we may obtain some other
involute of C .
S i n ce P1P1′ = P2 P2′ = P3 P3′ = P4 P4′
ENVELOPES, EVOLUTES AND INVOLUTES D-261
etc., therefore the curves C1 and C2 are at a constant distance along their common
normal at their corresponding points. We have thus shown that there are an infinite
number of involutes to the given curve C and all of them form a system of parallel curves.
1. Show that the equation of the normal to the ellipse x2 ⁄ a2 + y2 ⁄ b2 = 1 at the point
(a cos θ, b sin θ) is
ax by
− = a 2 − b2 .
cos θ sin θ
H ence find the evolute of the above ellipse. (Avadh 2006, 10)
True or False:
Write ‘T ’ for true and ‘F ’ for false statem ent.
9. In general, the envelope of a family of curves touches each curve of the family and
at each point is touched by some member of the family.
10. The envelope of x2 sin α + y2 cos α = a2, the parameter being α is x2 + y2 = a2.
11. The evolute of a curve is the envelope of the tangents of that curve.
12. If one curve is the evolute of another, then the latter is called an involute of the
former.
13. The evolute of an equiangular spiral is not an equiangular spiral.
1. 0. 2. B 2 = 4 A C. 3. x4 + 4 ay = 0.
4. (b). 5. (a). 6. (c).
7. (b). 8. (c). 9. T.
10. F. 11. F. 12. T.
13. F.
13.1 Asymptote
Suppose a curve is not limited in extent i.e. it has some branch or branches which
extend to infinity. Parabola and H yperbola are familiar curves of this type. Take a point
on an infinite branch of such a curve and draw a tangent to the curve at this point. If
the distance of the point of contact from the origin tends to infinity, the tangent itself,
may or may not tend to a definite straight line. In case the tangent tends to a definite
straight line, at a finite distance from the origin, it is called an asymptote of the curve.
Thus we can define the asymptote of a curve as follows.
Definition : A straight line at a finite distance from the origin to which a tangent
to a curve tends, as the distance from the origin of the point of contact tends to infinity, is
called an asym ptote of the curve. (Kanpur 2005, 07, 14; Kashi 11)
The curve approaches the asymptote :
R oughly speaking an asymptote is a tangent with its point
of contact at a great distance from the origin. Therefore, when
a point P on a curve tends to infinity, its perpendicular distance
from the corresponding asymptote tends to zero. This property
of an asymptote enables us to draw more accurately those
curves which have asymptotes. We draw the asymptotes first,
a n d t h e n t h e cu r ve, m a kin g it s bra n che s a pp r o ach t he
corresponding asymptotes.
D-264 DIFFERENTIAL CALCULUS
Branch of a Curve : Suppose the equation of the curve is such that y has two or
more values for every value of x. Corresponding to these distinct values of y we shall get
different branches of the curve. It is just possible, that each branch may have its own
separate asymptote. Therefore a curve may have m ore than one asym ptote.
or x n φ n ( y ⁄ x) + x n − 1 φ n − 1 ( y ⁄ x) + x n − 2 φ n − 2 ( y ⁄ x) + …
⎛ y⎞ ⎛ y⎞
… + x φ 1 ⎜⎝ ⎟⎠ + φ 0 ⎜⎝ ⎟⎠ = 0, …(2)
x x
⎛ y⎞ y
where φ r ⎜ ⎟ is a polynomial in of degree r.
⎝ x⎠ x
Dividing (2) by x n , we get
1 1 ⎛ y⎞
φ n ( y ⁄ x) + φ n − 1 (y ⁄ x) + 2 φ n − 2 ⎜ ⎟ + ……
x x ⎝ x⎠
1 ⎛ y⎞ 1 ⎛ y⎞
………+ n − 1 φ 1 ⎜ ⎟ + n φ 0 ⎜ ⎟ = 0. …(3)
x ⎝ x ⎠ x ⎝ x⎠
E xcluding at present the case of asymptotes parallel to the y-axis (i.e. excluding
Lim
the case in which x → ∞ ( y ⁄ x) is equal to ∞ ), (3) gives, on taking limits as x → ∞ , the
equation
φ n (m) = 0, ...(4)
Lim ⎛ y⎞
where m = x → ∞ ⎜ ⎟ = slope of an asymptote.
⎝ x⎠
The equation (4) is, in general, of degree n in m . Solving this equation, we shall
get the slopes of the asymptotes. This equation will give us n values of m, corresponding
to the n branches of the curve (1). H owever, some of the values of m may be equal, and
this will be the case of parallel asymptotes. Since we are concerned only with real
asymptotes, therefore we shall reject the imaginary roots of (4) if there are any.
Now if y = m x + c is an asymptote of (1), then we know that corresponding to a
specified value of m , we have
Lim
c = x → ∞ ( y − mx).
Therefore to determine the value of c corresponding to the value of m , we put
y − m x = p in the equation of the curve, where p is a variable which → c as x → ∞ .
So putting y = m x + p i.e.,
y p
= m + in (2), we get
x x
⎛ p⎞ ⎛ p⎞ ⎛ p⎞
xn φn ⎜ m + ⎟ + xn − 1 φn − 1 ⎜ m + ⎟ + xn − 2 φn − 2 ⎜ m + ⎟ + …
⎝ x⎠ ⎝ x⎠ ⎝ x⎠
⎛ p⎞ ⎛ p⎞
… + x φ 1 ⎜ m + ⎟ + φ 0 ⎜ m + ⎟ = 0⋅ …(5)
⎝ x⎠ ⎝ x⎠
E xpanding each term of (5) by Taylor’s theorem, we get
⎡ p p2 ⎤
x n ⎢⎢ φ n (m) + φ n ′ (m) + 2 φ n ′′ (m) + …⎥⎥
⎢ x x 2! ⎥
⎣ ⎦
⎡ p ⎤
+ x n − 1 ⎢ φ n − 1 (m) + φ ′ n − 1 (m) + …⎥
⎣ x ⎦
⎡ p ⎤
+ x n − 2 ⎢ φ n − 2 (m) + φ ′ n − 2 (m) + …⎥ + ……= 0. …(6)
⎣ x ⎦
D-266 DIFFERENTIAL CALCULUS
a 2 b2
Example 1 : Find the asymptotes of the curve 2 − 2 = 1⋅
x y
(Bundelkhand 2006; Agra 07, 08; Kashi 14)
Solution : The equation of the curve can be written as
a2 y2 − b2 x2 = x2 y2 or x2 y2 − a2 y2 + b2 x2 = 0.
Since the curve is of degree 4, therefore it cannot have more than four asymptotes.
E quating to zero the coefficient of the highest power of y (i.e. o f y2), the
asymptotes parallel to y-axis are given by x2 − a2 = 0 i.e. x = ± a.
Also equating to zero the coefficient of the highest power of x (i.e. of x2), the
asymptotes parallel to x-axis are given by y2 + b2 = 0, which gives two imaginary
asymptotes.
Thus all the four possible asymptotes of the curve have been found and the only
real asymptotes are x = ± a.
Example 2 : Find the asymptotes of the curve y2 (a2 − x2) = x4. (Meerut 2010)
2 2 4
Solution : The equation of the curve is y x + x − a y = 0.2 2
Since the curve is of degree 4, therefore it cannot have more than four asymptotes.
E quating to zero the coefficient of the highest power of y (i.e., o f y2) the
asymptotes parallel to y-axis are given by x2 − a2 = 0 i.e. x = ± a.
The coefficient of the highest power x4 of x is merely a constant. H ence there is
no asymptote parallel to x-axis.
ASYMPTOTES D-267
To find the remaining oblique asymptotes, we put y = m and x = 1 in the highest
i.e. four degree terms and we get φ 4 (m) = m 2 + 1.
The roots of the equat ion φ 4 (m) = 0 are imaginary and consequently the
corresponding asymptotes are imaginary.
H ence the only real asymptotes of the curve are x = ± a.
5. y2 (x2 − a 2) = x2 (x2 − 4a2). (Meerut 2003, 06; Agra 05; Rohilkhand 05, 06)
1. x = ± a, y = ± b.
2. x = ± a, y = 0.
3. x = 0.
4. x = ± a, y = ± a .
5. x = ± a, y = ± x.
y x
6. = ± ⋅
b a
7. y = 0, y = 1, x = 0, x = 1.
8. x = ± 1, y = ± 1. 9. x = ± a , y = ± a.
In a similar way we can discuss the case of more than three parallel asymptotes.
ASYMPTOTES D-269
The num ber of asym ptotes, real or im aginary, of an algebraic curve of the nth degree
cannot exceed n.
The slopes of the asymptotes which are not parallel to y-axis are given as the roots
of the equation φ n (m) = 0 which is of degree n at the most. If the equation of the curve
D-270 DIFFERENTIAL CALCULUS
possesses some asymptotes parallel to y-axis, then we can easily see that the degree of
φ n (m) = 0 will be smaller than n by at least the same number. In general one value of
m gives only one value of c. In case the equation for determining c is a quadratic, the
equation φ n (m) = 0 has two equal roots. Similarly if the equation for determining c is
cubic, the equation φ n (m) = 0 has three equal roots.
H ence a curve of degree n cannot have m ore than n asymptotes. But the number of
real asymptotes can be less than n. Some roots of the equation φ n (m) = 0 may come
out to be imaginary or even corresponding to a real value of m the value of c may come
out to be infinite.
4. x2 y + xy2 + xy + y2 + 3x = 0.
5. y3 − 5xy2 + 8 x2 y − 4 x3 − 3y2 + 9xy − 6 x2 + 2y − 2x + 1 = 0.
6. 2 x (y − 3) 2 = 3y (x − 1) 2.
7. y2 (x − 2a) = x3 − a3. (Bundelkhand 2008)
1. y = x + 1, y = − x + 1 and x + 2y = 0.
2. y = − x + 2, y = x + 2, y = 2x − 4.
3. 2y + x = 1, y = x, y + x + 1 = 0.
4. x = − 1, y = 0, x + y = 0.
5. y = x, y = 2x + 2, y = 2x + 1.
6. x = 0, y = 0, 2y = 3x + 6.
7. x = 2a, y = x + a, y = − x − a.
8. y = x, y = 2x + 1, y = − x − 2.
9. x + y = ± 1 and x − y = ± 1.
10. y = x, y = − x + 1, y = − x − 1.
11. y = ± 1, x = ± 1, y = − x.
12. x = 0, y = x, y = x + 1.
13. y − x = 0, 2y − x = 0, 2y − x − 1 = 0.
14. y + x = 0, 3y − 2x − 3 = 0, 3y − 2x + 1 = 0.
Lim ( y2 − x2) − x − y − 1
x + y = x →∞ , y ⁄ x →− 1
(x − y) (x + 2y)
⎡ y2 ⎤
⎢ − 1⎥ − 1 − y ⋅ 1 − 1
⎢ 2 ⎥
⎢x ⎥ x x x x2
Lim ⎣ ⎦
= x →∞ , y ⁄ x →− 1
⎡ y⎤ ⎡ y⎤
⎢1 − ⎥ ⎢1 + 2 ⎥
⎣ x⎦ ⎣ x⎦
(1 − 1)
= = 0,
(1 + 1) (1 − 2)
i.e. x + y = 0 is another asymptote of the curve.
The third asymptote of the curve is
Lim ( y2 − x2) + terms of degree lower than 2
x + 2y = x → ∞ , y ⁄ x → − 1
2 (x − y) (x + y)
Lim ⎧ Qn − 1⎫
(ax + by) + x → ∞ , y ⁄ x → − a ⁄ b ⎪⎨ c + ⎪
⎬= 0
⎪ Pn − 1⎪
⎩ ⎭
Lim ⎛ Qn − 1 ⎞⎟
or (ax + by + c) + x → ∞ , y ⁄ x → − a ⁄ b ⎜⎜ ⎟ = 0⋅
⎜P ⎟
⎝ n − 1⎠
Thus if the equation of a curve is given in the form (1), then there is no necessity
of collecting separately the nth degree terms. A similar modification can be made in
case III.
Case V : Asymptotes by Inspection : If the equation of a curve of the nth degree
can be put in the form Fn + P = 0, where Fn is of degree n (i.e., contains term s of degree n
and m ay also contain term s of lower degrees) , and P is of degree n − 2 , or lower, and if
Fn = 0 can be broken up into n linear factors which represent n straight lines no two of
which are parallel or coincident then all the asym ptotes of the curve are given by equating
to zero the linear factors of Fn .
Let ax + by + c = 0 be a non-repeated factor of Fn . Then the equation of the curve
can be written as (ax + by + c) Fn − 1 + P = 0, where Fn − 1 is of degree n − 1.
The asymptote of the given curve parallel to the line ax + by + c = 0 is
lim ⎛ P ⎞⎟
ax + by + c + x → ∞ , y ⁄ x → − a ⁄ b ⎜⎜ ⎟ = 0⋅ …(1)
⎜ Fn −
⎝ 1 ⎟⎠
Since Fn − 1 contains at least one term of degree n − 1 and P is of degree n − 2,
or lower, therefore we shall have
lim ⎛ P ⎞⎟
x → ∞ , y ⁄ x → − a ⁄ b ⎜⎜⎜ F ⎟⎟ = 0⋅
⎝ n − 1⎠
Thus ax + by + c = 0 is an asymptote of the given curve.
or (x − y − 1) 2 + 3 (x − y − 1) − 4 = 0,
− 3 ± √(9 + 16)
or (x − y − 1) = = 1, − 4.
2
Thus x − y − 2 = 0 and x − y + 3 = 0 are two parallel asymptotes of the given
curve. Since the remaining linear factors of the fourth degree terms in the equation of
the curve are imaginary, therefore the other two asymptotes are imaginary.
Example 2 : Find all the asymptotes of the curve
(x2 − y2) (x + 2y + 1) + x + y + 1 = 0.
Solution : The equation of the curve can be written as
(x − y) (x + y) (x + 2y + 1) + x + y + 1 = 0.
Since no two of the straight lines x − y = 0, x + y = 0 and x + 2y + 1 = 0 are
parallel and x + y + 1 is of degree 1, therefore all the asymptotes of the curve are given
by (x − y) (x + y) (x + 2y + 1) = 0.
1. y = x, y = 2x, y + x + 1 = 0, y + 2x + 1 = 0. 2. x = 3, y − x = 1, y − x = 2.
3. y = 2x − 2, y = 2x − 3, y − x = 4. 4. y = x + 4, x − 2y = 2 ± 3 √3.
5. x = ± a, y = x + a, y = − x + a.
3
6. x = 0, y = 0, y = 2x + , 2y + 4x = 15.
2
7. x − y − 2 = 0, x − y − 3 = 0.
1
8. x = ± a, y = x + a, y = ± x − a.
2
9. y + x = 0, y = x − 3 and y = x + 2.
10. x = 0, y = 0, x − y = 0, x + y = 0, x − 2y = 0 and x + 2y = 0.
ASYMPTOTES D-281
Particular Cases :
(i) If the given curve is of degree 3, then the 3 (3 − 2) i.e., 3 points of intersection
of the curve and its asymptotes lie on a curve of degree 3 − 2 = 1 i.e., on a straight line.
(ii) If the curve is of degree 4 then the 4 (4 − 2) i.e., 8 points of intersection of
the curve and its asymptotes lie on a curve of degree 4 − 2 = 2 i.e. on a conic.
D-282 DIFFERENTIAL CALCULUS
Since the points of intersection of a cubic curve with its asymptotes lie on a
straight line, therefore the most general equation of the curve having F3 = 0 as its
asymptotes is
(x − y) (x − 2y) (x − 3y) + ax + by + c = 0
or x3 − 6x2 y + 11xy2 − 6y3 + ax + by + c = 0. …(2)
If (2) passes through the points (0, 0), (1, 0) and (0, 1), then
c = 0, 1 + a = 0 i.e., a = − 1 and − 6 + b = 0 i.e. b = 6.
H ence the required curve is x3 − 6x2 y + 11xy2 − 6y3 − x + 6y = 0.
1. y = 0, x = 1, x − y + 2 = 0. 3. a2 ( y + b) = b2 (x + a).
7. 2 2 2 2 2 2 2
bxy (x − y ) + a (b − a ) (x + y − a ) = 0.
8. x3 − 6x2 y + 11xy2 − 6y3 − x = 0.
D-284 DIFFERENTIAL CALCULUS
1
i.e. OT → − , if f ′ (α) ≠ 0.
f ′ (α)
Also OP and PT will tend to become parallel as is obvious from the dotted lines
in the figure. Therefore the angle OTP will tend to a right angle and OT will tend to
1
OM where OM is perpendicular to the asymptote. H ence OM = − ⋅
f ′ (α)
When θ = α, suppose OP tends to OP ′.
Then ∠ XOP ′ = α.
⎛π ⎞
∴ ∠MOX = − ⎜ − α⎟ , negative sign indicating that it has been measured
⎝2 ⎠
clockwise.
∴ The equation of the asymptote is
π 1
r cos ⎡⎢ θ − ⎧⎨ − ⎛⎜ − α⎞⎟ ⎫⎬ ⎤⎥ = −
⎣ ⎩ ⎝2 ⎠ ⎭⎦ f ′ (α)
π 1
i.e. r cos ⎛⎜ θ + − α⎞⎟ = −
⎝ 2 ⎠ f ′ (α)
1 1
i.e. − r sin (θ − α) = − i.e. r sin (θ − α) = ⋅
f ′ (α) f ′ (α)
1. If α is a root of the equation f (θ) = 0, then write the equation of asymptote of the
1
polar curve = f (θ) corresponding to the root α.
r (Meerut 2001)
Find the asymptotes of the following curves :
2. y = tan x.
3. r sin mθ = a. (Meerut 2000)
4. r θ = a. (Meerut 2008, 12B)
5. 2 ⁄ r = 1 + 2 sin θ.
6. (i) r sin θ = 2 cos 2θ. (ii) r sin θ = a cos 2θ.
7. (i) r cos θ = a sin θ. (Meerut 2004B)
(ii) r sin θ = 2 cos θ.
8. r = 4 (sec θ + tan θ).
9. r sin 2θ = a.
10. r cos θ = 4 sin2 θ. (Meerut 2005)
11. r θ cos θ = a cos 2θ.
12. r (eθ − 1) = a (eθ + 1).
2θ
13. r= ⋅
sin θ
Find the circular asymptotes of the following curves :
14. r (eθ − 1) = a (eθ + 1).
3θ2 + 2θ + 1
15. r= ⋅
2θ2 + θ + 1
1
1. r sin (θ − α) = ⋅ 2. x = ± π ⁄ 2, ± 3π ⁄ 2, …… .
[ f ′(θ)]θ= α
⎛ kπ ⎞ a , where k is any integer.
3. r sin ⎜ θ − ⎟ =
⎝ m⎠ m cos kπ
⎡ 1 ⎤
4. r sin θ = a. 5. r sin ⎢ θ ± π ⎥ = 2 ⁄ √3.
⎣ 6 ⎦
6. (i) r sin θ = 2. (ii) r sin θ = a.
7. (i) r cos θ = ± a. (ii) r sin θ = ± 2.
1 1
8. r cos θ = 8. 9. r sin θ = ± a, r cos θ = ± a.
2 2
a , k is any integer.
10. r cos θ = 4. 11. r sin θ = a, r cos θ =
⎛ 1⎞
⎜k + ⎟ π
⎝ 2⎠
12. r sin θ = 2a. 13. r sin θ = 2kπ, k = ± 1, ± 2,…. .
3
14. r = a. 15. r= ⋅
2
D-288 DIFFERENTIAL CALCULUS
1. highest. 2. n.
1 ,
3. x + y + a = 0. 4. r sin (θ − α) = f ′(α) ≠ 0.
f ′(α)
5. r = a. 6. (c). 7. (c). 8. (c).
9. T. 10. F. 11. T. 12. T. 13. F.
14.1 Concavity and Convexity (Meerut 2009)
Let P be a given point on a curve and A B a given straight line which does not pass
through P . Then the curve is said to be concave or convex at P with respect to A B ,
according as a sufficiently small arc of the curve containing P lies entirely within or
without the acute angle formed by the tangent at P to the curve with the line A B . Thus
in the figure 1 the curve at P is convex to A B , and in figure 2 it is concave to A B .
d2 y
A curve is convex or concave at P to the axis of x according as y is positive or
dx2
negative at P.
Test for concavity upwards or Concavity downwards :
T h e cu rv e y = f (x) i s concave upwards i n [a, b] i . e. , w h e n a ≤ x ≤ b, if
f ′′(x) > 0 — V x ∈ [a, b] and is concave downwards in [a, b] if f ′′(x) < 0 —
V x ∈ [a, b].
Concavity and Convexity with respect to the axis of y :
By considering y as the independent variable, we can easily show that a curve
x = f ( y), at a given point P on it, is convex or concave to the axis of y according as
d2 x
x 2 is positive or negative at P.
dy
Example 1 : Show that the curve y = e x is everywhere concave upwards and the
curve y = log x is everywhere concave downwards.
Solution : First consider the curve y = e x.
dy d2 y
We have = e x and = e x.
dx dx2
d2 y
O bviously > 0, —
V x ∈ R , where R is the set of real numbers.
dx2
H ence , the curve y = e x is everywhere concave upwards.
Now consider the curve y = log x, 0 < x < ∞ .
dy 1 d2 y 1
We have = and = − 2⋅
dx x dx2 x
d2 y
O bviously, < 0, —
V x ∈ ]0, ∞ [.
dx2
H ence, the curve y = log x is everywhere concave downwards.
Example 2 : Find the intervals in which the curve y = e x (cos x + sin x) is concave
upwards or downwards; x varying in the interval ]0, 2 π[.
Solution : The given curve is y = e x (cos x + sin x).
dy
We have = e x (cos x + sin x) + e x (− sin x + cos x) = 2 e x cos x.
dx
d2 y
∴ = 2 e x cos x − 2 e x sin x = 2 e x (cos x − sin x)
dx2
⎛ π π⎞ ⎛ π⎞
= 2√⎯⎯2 . e x ⎜⎝ cos x cos − sin x sin ⎟⎠ = 2 √ ⎯⎯2 . e x cos ⎜⎝ x + ⎟⎠ ⋅
4 4 4
⎤ π⎡ ⎤ 5π ⎡
, 2 π ⎢ , we have d2 y
When x ∈ ⎥ 0, ⎢ or when x ∈ ⎥⎦ ⎣
> 0.
⎦ 4 ⎣ 4 dx2
⎤ π⎡ ⎤ 5π ⎡
H ence, the given curve is concave upwards in ⎥ 0, ⎢ and ⎥⎦ , 2 π⎢ ⋅
⎦ 4 ⎣ 4 ⎣
⎤π 2
Again when x ∈ ⎥ , 5π ⎡⎢ , we have d y < 0.
⎦ 4 4 ⎣ dx2
SINGULAR POINTS : CURVE TRACING D-293
⎤π
H ence, the given curve is concave downwards in ⎥ , 5π ⎡⎢ ⋅
⎦
4 4⎣
Example 3 : Show that the sine curve y = sin x is everywhere concave with respect
to the axis of x excluding the points where it meets the axis of x.
Solution : The given curve is y = sin x.
dy d2 y
We have = cos x and = − sin x.
dx dx2
The function sin x is a periodic function with period 2 π. Hence, it is sufficient to
consider the given curve in the interval [0, 2 π].
In the interval [0, 2 π], we have y = 0 when x = 0 or x = π or x = 2 π.
d2 y
When x ∈ ]0, π[, we have y > 0 and < 0.
dx2
d2 y
So y 2 < 0 when x ∈ ]0, π[.
dx
H ence, the curve y = sin x is concave to the axis of x in the interval ]0, π[.
d2 y
When x ∈ ]π, 2 π[, we have y < 0 and 2 > 0.
dx
d2 y
So y 2 < 0 when x ∈ ]π, 2 π[.
dx
H ence, the curve y = sin x is concave to the axis of x in the interval ]π, 2 π[.
Thus the curve y = sin x is everywhere concave with respect to the axis of x
excluding the points where it meets the axis of x.
dr
But r = radius of curvature at P
dp
⎧ 2 3⁄2
⎪ 2 dr ⎛ ⎞ ⎫⎪
+ ⎜⎝ ⎟⎠ ⎬⎪
⎨r
⎪
⎩ dθ ⎭
=
⎛ dr ⎞ 2 d2 r
r2 + 2 ⎜⎝ ⎟⎠ − r 2
dθ dθ
⎧ 2 ⎫
⎪ ⎛ dr ⎞ d2 r⎪⎬
r ⎨⎪ r2 + 2 ⎜⎝ ⎟⎠ − r
dp ⎩ dθ dθ2 ⎪⎭
∴ =
dr ⎧ ⎛ ⎞2 ⎫3 ⁄ 2
⎨ r + ⎜ dr ⎟
⎪ 2 ⎪
⎬
⎪ ⎝ dθ⎠ ⎪
⎩ ⎭
2
⎛ dr ⎞ d2 r
Hence, if r2 + 2 ⎜⎝ ⎟⎠ − r 2 = 0 at P, there is, in general, a point of inflexion
dθ dθ
at P.
Now d3 y ⁄ dx3 = 72 x − 24 .
When x = 0, d3 y ⁄ dx3 ≠ 0, therefore x = 0 gives a point of inflexion.
2 2
Similarly, when x = , d3 y ⁄ dx3 ≠ 0 ; therefore x = also gives a point of inflexion.
3 3
From the equation of the curve, we have
11 , 2
y = 1 , when x = 0 and y = when x = ⋅
27 3
⎛ 2 11 ⎞
H ence (0, 1) and ⎜⎝ , ⎟⎠ are the required points of inflexion.
3 27
Important : Instead of finding d3 y ⁄ dx3 , we can use another criterion for points
of inflexion. If d2 y ⁄ dx2 = 0 at x = a and the sign of d2 y ⁄ dx2 changes while x passes
through a , then there will be a point of inflexion at x = a.
Example 2 : Find the points of inflexion of the curve x = log ( y ⁄ x) .
(Purvanchal 2011; Rohilkhand 14)
Solution : The given curve is x = log ( y ⁄ x) or y ⁄ x = e x
or y = x e x. ...(1)
Differentiating (1), we get
dy ⁄ dx = x e x + e x = (x + 1) e x
and d2 y ⁄ dx2 = e x + (x + 1) e x = e x (x + 2) .
D-296 DIFFERENTIAL CALCULUS
Substituting the value of θ in the given equation of the curve we get the points of
inflexion as
⎡ ⎛ 2π __ √3 ⎞ , 3a ⎤ ⎡ . . ⎛ π⎞ ⎛ π⎞ √3 ⎤
⎢ a ⎜ 4n π ± + ⎟ ⎥ ⋅ ⎢ . sin ⎜ 2n π ± ⎟ = (− 1) 2n sin ⎜ ± ⎟ = ± ⎥⋅
⎣ ⎝ 3 2⎠ 2⎦ ⎣ ⎝ 3⎠ ⎝ 3⎠ 2⎦
Example 4 : Find the ranges of values of x for which the curve
y = x4 − 6 x3 + 12 x2 + 5x + 7
is concave upwards or downwards.
A lso determine the points of inflexion. (Purvanchal 2008)
dy
Solution : We have = 4 x3 − 18 x2 + 24 x + 5,
dx
d2 y
= 12 x2 − 36 x + 24 = 12 (x − 1) (x − 2)
dx2
d3 y
and = 24 x − 36.
dx3
d2 y
We have > 0, —
V x ∈ ]− ∞ , 1[,
dx2
d2 y d2 y
2
< 0, —
V x ∈ ]1, 2[ and 2 > 0, —
V x ∈ ]2 , ∞ [.
dx dx
H ence, the curve is concave upwards in the intervals ]− ∞ , 1[ and ]2 , ∞ [ and
concave downwards in the interval ]1, 2[.
SINGULAR POINTS : CURVE TRACING D-297
d2 y
Now = 0 ⇒ (x − 1) (x − 2) = 0 ⇒ x = 1 or x = 2 .
dx2
d3 y d2 y
At x = 1, 3
= − 12 ≠ 0 and at x = 2 , 2 = 12 ≠ 0. Thus, there are points of
dx dx
inflexion at x = 1 and at x = 2 .
When x = 1, we have y = 19 and when x = 2 , we have y = 33.
H ence, (1, 19) and (2, 33) are the two points of inflexion on the curve.
Example 5 : E xa m in e th e curve y = sin x for concavity upwards, concavity
downwards and for points of inflexion in the interval [− 2 π, 2 π].
Solution : The given curve is y = sin x.
dy d2 y d3 y
We have = cos x, 2
= − sin x and = − cos x.
dx dx dx3
d2 y
We have, < 0, —
V x ∈ ]− 2 π, − π[,
dx2
d2 y d2 y
2
> 0, —
V x ∈ ]− π, 0[, < 0, —
V x ∈ ]0, π[
dx dx2
d2 y
and > 0, —
V x ∈ ]π, 2 π[.
dx2
H ence, the curve is concave downwards in the intervals ]− 2 π, − π[ and ]0, π[ and
concave upwards in the intervals ]− π, 0[ and ]π, 2 π[.
d2 y
Now = 0 ⇒ sin x = 0 ⇒ x = − 2 π or x = − π or x = 0
dx2
or x = π or x = 2 π.
A t e ach o f t h e p o in t s x = − 2 π, x = − π, x = 0, x = π a n d x = 2 π, we h a ve
d3 y
≠ 0.
dx3
Thus there are points of inflexion at each of these points.
Also, y = 0 at each of these points.
H ence, the curve has points of inflexion at (− 2 π, 0), (− π, 0), (0, 0), (π, 0) and
(2 π, 0).
a2 θ2 (θ2 − 3) (θ2 + 2)
or = 0
(θ2 − 1) 3
or θ2 (θ2 − 3) (θ2 + 2) = 0
∴ θ2 = 0 , 3 , − 2 .
R ejecting the values θ2 = − 2 and 0 we see that the points of inflexion are given
by θ2 = 3 i.e., θ = ± √3 .
1. Show that the points of inflexion upon the curve x2 y = a2 (x − y) are given by
x = 0, x = ± a √3 . (Meerut 2013B)
2. Find the points of inflexion of the curve y (a2 + x2) = x3. (Purvanchal 2010)
8. Find the points of inflexion on the curve y2 = x (x + 1) 2 and also obtain the
equations of the inflexional tangents.
9. Show that origin is a point of inflexion of the curve am − 1. y = x m if m is odd and
greater than 2.
10. Show that the abscissae of the points of inflexion on the curve y2 = f (x) satisfy the
equation [ f ′ (x)]2 = 2 f (x) . f ′′ (x).
11. S h o w t h a t t h e l i n e j o i n i n g t h e p o i n t s o f i n fl e xi o n o f t h e c u r ve
y2 (x − a) = x2 (x + a) subtends an angle of π ⁄ 3 at the origin.
1− x ,
12. Prove that the curve y = has three points of inflexion which lie in a
1 + x2
straight line.
− (x ⁄ a) 2
13. Show that the points of inflexion on the curve y = be are given by
x = ± a ⁄ √2 . (Agra 2005)
14. Show that the points of inflexion of the curve r = bθn are given by
r = b {− n (n + 1)}n ⁄ 2 .
⎛ 3 √3 ⎞ , ⎛ − 3 √3a ⎞
2. (0, 0), ⎜ √3a , a⎟ ⎜ − √3a , ⎟⋅
⎝ 4 ⎠ ⎝ 4 ⎠
SINGULAR POINTS : CURVE TRACING D-299
⎛ ⎞
3
3. ⎜ ae− 3 ⁄ 2, ae3 ⁄ 2⎟⎠ ⋅
⎝ 2
4. (0, 1), (8, e2) .
5. Points of inflexion at x = 2 , (11 ± √2) ⁄ 7; point of undulation at x = 1.
8. Points of inflexion are (1 ⁄ 3, ± 4 ⁄ 3√3) . Inflexional tangents are
9x ± 3√3y + 1 = 0.
A point through which more than one branches of a curve pass is called a m ultiple
point on the curve. A point on the curve is called a double point if two branches of the
curve pass through it, a triple point if three branches pass through it. In general, if r
branches pass through a point, it is called a multiple point of the r th order.
Since through a double point two branches of the curve pass, therefore in the
process of finding tangents at a double point we must get two tangents there, one for
each branch. If the two tangents are real and distinct, the double point will be a node.
If the two tangents are imaginary, the double point will be a conjugate point. If the two
tangents are real and coincident, the double point may be a cusp or a conjugate point.
The possibility of the point being a conjugate point in this case arises on account of
the fact that sometimes imaginary expressions A ± iB become real by chance when
B = 0 . In such cases the double point will be a cusp if there are other real points of the
curve in its neighbourhood, otherwise it will be a conjugate point.
If a curve passes through the origin and is given by a rational integral, algebraic
equation, the equation to the tangent or tangents at the origin is obtained by equating to
zero the lowest degree terms in the equation of the curve.
Let the equation of the curve when arranged according to ascending powers of x
and y be
(a1 x + a2 y) + (b1 x2 + b2 xy + b3 y2)
+ (c1 x3 + c2 x2 y + …) + … = 0 , ...(1)
where the constant term is absent since the curve
passes through the origin.
Let P (x, y) be any point on the curve. The slope
of the chord OP is y ⁄ x . Therefore the equation to OP
is Y = ( y ⁄ x) X , where (X, Y ) are current coordinates.
As P → O i.e., as x → 0 and y → 0 , the chord OP
tends to the tangent at O .
E xcluding for the present the case when the tangent is the y-axis i.e., when
⎛ y⎞
lim ⎜ ⎟⎟
⎜
x → 0 ⎜⎝ x⎟⎠ = ± ∞ , we have the equation of the tangent at O as
⎧ ⎛ ⎞ ⎫
⎪ lim ⎜ y⎟ ⎪
Y = ⎨⎪ x → 0 ⎜⎜ ⎟⎟ ⎬⎪ X . ...(2)
⎩ ⎝ x⎠ ⎭
a1 = a2 = 0 .
If b2 = b3 = 0 , then by (v), b1 = 0 .
Case III : If a1 = a2 = b1 = b2 = b3 = 0 , we can show by the same process that
the rule still holds; and so on.
If tangent at the origin is the y-axis, we can easily show by supposing the axes of
x and y to be interchanged for a moment, that the rule is still true.
Hence the equation of the tangent or tangents at the origin is obtained by equating to
zero the lowest degree terms in the equation of the curve.
Corollary : If the origin is a double point on a curve, then the curve has two
tangents at the origin. Therefore the equation of the curve should not contain the
constant and the first degree terms and the second degree terms should be the lowest
degree terms in the equation of the curve.
Example 1 : Show that the origin is a node on the curve x3 + y3 − 3axy = 0.
(Meerut 2003; Purvanchal 14)
Solution : The curve passes through the origin as its equation does not contain
the constant term. Also equating to zero the lowest degree terms in the equation of the
curve, we get the equation to the tangents at origin as − 3axy = 0, i.e. x y = 0, i.e.
x = 0, y = 0 are two real and distinct tangents at the origin. Therefore origin is a node.
Example 2 : Show that the origin is a conjugate point on the curve
a2 x2 + b2 y2 = (x2 + y2) 2 .
Solution : O bviously the curve passes through the origin. The equation to the
tangent at origin is
a2 x2 + b2 y2 = 0, i.e., a x ± iby = 0.
Thus there are two imaginary tangents at the origin. Therefore origin is a
conjugate point.
⎡ ∂ ⎛ ∂f ⎞ ⎧ ∂ ⎛ ∂f ⎞ ⎫ dy⎤
or ⎢ ⎜ ⎟ + ⎨ ⎜ ⎟⎬ ⎥
⎣ ∂x ⎝ ∂x⎠ ⎩ ∂y ⎝ ∂x⎠ ⎭ dx⎦
⎡ ∂ ⎛ ∂f ⎞ ⎧ ∂ ⎛ ∂f ⎞ ⎫ dy⎤ dy ∂f d2 y
+ ⎢ ⎜ ⎟ + ⎨ ⎜ ⎟⎬ ⋅ ⎥ ⋅ + ⋅ = 0
⎣ ∂x ⎝ ∂y⎠ ⎩ ∂y ⎝ ∂y⎠ ⎭ dx⎦ dx ∂y dx2
∂2f ∂2 f dy ∂2 f dy ∂2 f ⎛ dy⎞ 2 ∂f d2 y
or + ⋅ + ⋅ + ⎜ ⎟ + ⋅ = 0
∂x2 ∂x ∂y dx ∂x ∂y dx ∂y2 ⎝ dx⎠ ∂y dx2
∂2 f ∂2 f dy ∂2 f ⎛ dy⎞ 2
or + 2⋅ ⋅ + ⋅ ⎜ ⎟ = 0,
∂x2 ∂x ∂y dx ∂y2 ⎝ dx⎠
since at a multiple point ∂f ⁄ ∂y = 0 .
dy
Therefore at the multiple point, the values of are given by the quadratic
dx
dy ,
in
dx
∂2 f ⎛ dy⎞ 2 ∂2 f ⎛ dy⎞ ∂2 f
⎜ ⎟ + 2 ⎜ ⎟ + = 0. ...(2)
∂y2 ⎝ dx⎠ ∂x ∂y ⎝ dx⎠ ∂x2
∂2 f , ∂2 f , ∂2 f
If are not all zero, the equation (2) will be a quadratic in dy ⁄ dx
∂x2 ∂x ∂y ∂y2
and the multiple point will be a double point.
The two tangents will be real and distinct, coincident, or imaginary according as
⎛ ∂2f ⎞ 2 ∂2 f ∂2 f
4⎜ ⎟ − 4 > , = or < 0
⎝ ∂x ∂y⎠ ∂y2 ∂x2
i.e., in general, the double point will be a node, cusp or conjugate point according as
⎛ ∂2 f ⎞ 2 ⎛ ∂ 2 f⎞ ⎛ ∂ 2 f⎞
⎜ ⎟ > , = or < ⎜⎜ ⎟ ⎜
⎟ ⎜
⎟⋅
⎟
⎝ ∂x ∂y⎠ ⎜ ∂x2 ⎟ ⎜ ∂y2 ⎟ (Meerut 2003)
⎝ ⎠ ⎝ ⎠
∂2 f ∂2 f ∂2 f
If = = = 0 , then the point (x, y) will be a multiple point of order
∂x2 ∂x ∂y ∂y2
higher than the second.
are only two branches of the curve (2) through the origin, therefore terms involving
powers of p above the second will be neglected. Thus we shall get a quadratic in p of the
form
A p2 + Bp + C = 0 , ...(4)
where A , B and C are some functions of x . Solving (4), we get
p = {− B ± √(B 2 − 4 A C )} ⁄ 2A . ...(5)
Also if p1, p2 are the roots of (4), we get
p1 p2 = C ⁄ A . ...(6)
The following different cases arise :
(i) If for all values of x , positive or negative, provided they are numerically small,
the values of p given by (5) are imaginary, the origin will be a conjugate point.
(ii) If for all numerically small values of x , positive or negative, the values of p
given by (5) are real, there will be a double cusp at the origin.
(iii) If the reality of the values of p given by (5) depends on the sign of x , there
will be a single cusp at the origin.
(iv) If for numerically small values of x for which p is real, the sign of p1 p2 is
positive, then p1 and p2 will be of the same sign. Therefore the two perpendiculars lie
on the same side of the common tangent and there will be a cusp of the second species.
If, on the other hand, the sign of p1 p2 is negative, then p1and p2 are of opposite signs.
Therefore the two perpendiculars lie on opposite sides of the common tangent and
there will be a cusp of the first species.
Note : While investigating the sign of an expression for sufficiently small values
of x , we should keep in mind only those terms which involve the lowest power of x .
∂f
and = 0 gives 6y2 − 6y = 0 i.e., y (y − 1) = 0 i.e., y = 0 , 1 .
∂y
∴ the possible double points are (1, 0) , (1, 1), (2, 0) and (2, 1) .
O ut of these only (1, 1) and (2, 0) satisfy the equation of the curve. Therefore
(1, 1) and (2, 0) are the only double points on the given curve.
∂2 f ∂2 f ∂2 f
Now = 12x − 18 , = 0 , = 12y − 6 .
∂x2 ∂x ∂y ∂y2
∂2 f ∂2 f ∂2 f
At the point (1, 1) , = − 6 , = 0 , = 6.
∂x2 ∂x ∂y ∂y2
⎛ ∂2 f ⎞ 2 ∂2 f ∂2 f
∴ at the point (1, 1) , ⎜ ⎟ = 0 and ⋅ = − 36 .
⎝ ∂x ∂y⎠ ∂x2 ∂y2
⎛ ∂2 f ⎞ 2 ∂2 f ∂2 f
Thus at the point (1, 1) , ⎜ ⎟ > ⋅ ⋅
⎝ ∂x ∂y⎠ ∂x2 ∂y2
Therefore there is a node at the point (1, 1) .
∂2 f ∂2 f ∂2 f
At the point (2, 0) , 2 = 6 , = 0, 2 = − 6.
∂x ∂x ∂y ∂y
⎛ ∂2 f ⎞ 2 ∂2 f ∂2 f
∴ at the point (2, 0) , ⎜ ⎟ > ⋅ ⋅
⎝ ∂x ∂y⎠ ∂x2 ∂y2
Thus there is a node at the point (2, 0) .
Example 4 : Find the nature of the origin on the curve a4 y2 = x4 (x2 − a2) .
(Meerut 2006B)
Solution : The given curve is a4 y2 = x4 (x2 − a2) . ...(1)
E quating to zero the lowest degree terms in the equation of the curve, we get the
tangents at the origin as a4 y2 = 0 i.e., y = 0 , y = 0 are two real and coincident tangents
at the origin.
Thus the origin may be a cusp or a conjugate point.
From (1), y = ± (x2 ⁄ a2) √(x2 − a2) .
For small values of x ≠ 0 , + ive or –ive, (x2 − a2) is –ive i.e., y is imaginary. H ence
no portion of the curve lies in the neighbourhood of the origin. H ence origin is a
conjugate point and not a cusp.
Example 5 : Show that the origin is a conjugate point on the curve
x4 − a x2 y + a x y2 + a2 y2 = 0.
Solution : E quating to zero, the lowest degree terms in the given curve, the
tangents at the origin are given by
a2 y2 = 0 i.e., y2 = 0 i.e., y = 0, y = 0.
Thus there are two real and coincident tangents at the origin
∴ origin is either a cusp or a conjugate point.
Now the equation of the given curve is
ay2 (x + a) − ax2 y + x4 = 0.
Solving it for y, we have
a x2 ± √{a2 x4 − 4ax4 (x + a)} a x2 ± x2 √(− 4a x − 3a2)
y= = ⋅
2 a (x + a) 2a (x + a)
D-308 DIFFERENTIAL CALCULUS
Now for small values of x ≠ 0, (− 4ax − 3a2) is − ive. Thus y is imaginary in the
neighbourhood of origin.
H ence origin is a conjugate point.
1. Write down the equations to the tangents at the origin for the following curves :
(i) y2 (a − x) = x2 (a + x), (ii) x4 + 3x3 y + 2 x y − y2 = 0,
(iii) (x2 + y2) (2a − x) = b2 x .
2. For the curve y2 (a2 + x2) = x2 (a2 − x2) , show that the origin is a node.
3. Show that the origin is a conjugate point on the curve y2 = 2 x2 y + x4 y − 2 x4.
4. Show that the curve x3 + x2 y = ay2 has a cusp at the origin.
5. Find the position and nature of double points of the following curves :
(i) y3 = x3 + a x2. (Meerut 2013B)
(ii) y2 + 3a x2 + x3 = 0
(iii) x3 + y3 = 3ax y . (Agra 2006; Rohilkhand 07; Kumaun 08; Meerut 12B, 13)
2. Nature of the Origin on the Curve : We should see whether the curve passes
through the origin or not. If the point (0, 0) satisfies the equation of the curve, it passes
through the origin. In order to know the shape of a curve at any point, we should draw
the tangent or tangents to the curve at that point. Therefore if the curve passes
through the origin, we should find the equation to the tangents at origin by equating to
zero the lowest degree terms in the equation of the curve. If there are two tangents at
the origin, then the origin will be a double point on the curve. We should also observe
the nature of the double point.
3. Points of intersection of the curve with the co-ordinate axes :
We should find the points where the curve cuts the co-ordinate axes. To find the
points where the curve cuts the x-axis we should put y = 0 in the equation of the curve
and solve the resulting equation for x . Similarly the points of intersection with the y-axis
are obtained by putting x = 0 and solving the resulting equation for y . We should also
obtain the tangents to the curve at the points where it meets the co-ordinate axes. In
order to find the tangent at the point (h, k) , we should shift the origin to (h, k) and then
the tangent or tangents at this new origin will be obtained by equating to zero the lowest
degree terms. The value of dy ⁄ dx at the point (h, k) can also be used to find the slope of
the tangent at that point.
4. We should solve the equation of the curve for y or x whichever is convenient.
Suppose we solve for y . Starting from x = 0 , we should see the nature of y as x increases
and then tends to + ∞ . Similarly we should see the nature of y as x decreases and then
tends to − ∞ . We should pay special attention to those values of x for which y = 0 or
→ infinity.
If we solve the equation of the curve for y and the curve is sym m etrical about y-axis,
then we should consider only positive values of x . The curve for negative values of x can be
drawn from sym metry and there is no necessity of considering them afresh.
H owever, if we solve the equation for y and there is symmetry only about x-axis,
then we are to consider both positive as well as negative values of x . If the curve is
symmetrical in opposite quadrants, or if there is symmetry about the x-axis, then only
positive values of y need be considered.
If y → infinity as x → a , then the line x = a will be an asymptote of the curve.
Similarly if x → infinity as y → b , then the line y = b will be an asymptote of the curve.
5. Regions where the curve does not exist : We should find out if there is any
region of the plane such that no part of the curve lies in it. Such a region is easily
obtained on solving the equation for one variable in terms of the other. The curve will
not exist for those values of one variable which make the other imaginary. For example,
in the curve
a2 y2 = x2 (x − a) (2a − x) ,
we find that for 0 < x < a , y2 is negative, i.e., y is imaginary. Therefore the curve does
not exist in the region bounded by the lines x = 0 and x = a . For a < x < 2a , y2 is positive
i.e., y is real. Therefore the curve exists in the region bounded by the lines x = a and
x = 2a . Thus if y is imaginary when x lies between a and b , the curve does not exist in
the region bounded by the lines x = a and x = b .
6. Asymptotes : We should find all the asymptotes of the curve. If an infinte
branch of the curve has an asymptote, then ultimately it must be drawn parallel to the
asymptote. The asymptotes parallel to the x-axis can be obtained by equating to zero
the coefficient of the highest power of x in the equation of the curve. Similarly the
asymptotes parallel to the y-axis can be obtained by equating to zero the coefficient of
the highest power of y in the equation of the curve.
SINGULAR POINTS : CURVE TRACING D-311
y2 (2a + x) = (x + a) 2 {a − (x + a)}
or y2 (2a + x) = − x (x2 + 2ax + a2) .
E quating to zero the lowest degree terms, we get x = 0 (i.e., the new y-axis) as the
tangent at the new origin. Thus the tangent at (a, 0) is perpendicular to x-axis.
(v) Solving the equation of the curve for y , we get
y2 = x2 (a − x) ⁄ (x + a) .
When x = 0 , y2 = 0 and when x = a , y2 = 0 .
When 0 < x < a , y2 is positive. Therefore the curve exists in this region.
When x > a , y2 is negative. Therefore the curve does not exist in the region
x> a.
When x → − a , y2 → ∞ . Therefore x = − a is an asymptote of the curve.
When − a < x < 0 , y2 is positive. Therefore the curve exists in this region.
When x < − a , y2 is negative. Therefore the curve does not exist in the region
x< − a.
(vi) Putting y = m and x = 1 in the highest i.e., third degree terms in the equation
o f t he cu rve, we get φ 3 (m) = m 2 + 1 . The root s of the equat ion φ 3 (m) = 0 are
imaginary. Therefore x = − a is the only real asymptote of the curve.
(vii) For the portion of the curve lying in the first quadrant, we have
⎧ (a − x) ⎫ (1 − x ⁄ a) 1 ⁄ 2
y= x
√ ⎨ ⎬= x
⎩ (a + x) ⎭ (1 + x ⁄ a) 1 ⁄ 2
⋅
When 0 < x < a , y is less than x . Therefore the curve lies below the line y = x which
is tangent at the origin.
For the portion of the curve lying in the second quadrant, we have
(1 − x ⁄ a) 1 ⁄ 2 ,
y= − x x < 0.
(1 + x ⁄ a) 1 ⁄ 2
When − a < x < 0 , y is greater than the numerical value of x . Therefore the curve
lies above the tangent y = − x .
H ence the shape of the curve is as shown in the figure.
Example 3 : (b) Trace the curve y2 (a + x) = x2 (3a − x) . (Purvanchal 2011)
Solution : Proceed as in part (a).
Example 4 : Trace the curve y2 (x2 + y2) + a2 (x2 − y2) = 0 .
Solution : (i) The curve is symmetrical about both the axes.
(ii) It passes through the origin and a2 (x2 − y2) = 0 i.e., y = ± x are the two
tangents at the origin. Therefore the origin is a node.
(iii) The curve intersects the x-axis only at origin. It intersects the y-axis at
(0, 0) , (0, a) and (0, − a) .
(iv) Shifting the origin to (0, a) , the equation of the curve becomes
( y + a) 2 {x2 + ( y + a) 2} + a2 {x2 − ( y + a) 2} = 0
or ( y2 + 2ay + a2) {x2 + y2 + 2ay + a2} + a2 (x2 − y2 − 2ay − a2) = 0 .
E quating to zero the lowest degree terms, we get
2a3 y + 2a3 y − 2a3y = 0 i.e., y = 0
as the tangent at the new origin. Thus the new x-axis is tangent at the new origin.
We need not find the tangent at (0, − a) as the curve is symmetrical about x-axis.
D-314 DIFFERENTIAL CALCULUS
/
1⁄2 1⁄2
⎛ x2 ⎞⎟ ⎛ 2⎞
⎜1 − x ⎟
or y = 2x ⎜⎜ 1 − ⎟ ⎜ ⎟ .
⎜ 2
4a ⎠ ⎟ ⎜ a2 ⎟⎠
⎝ ⎝
When 0 < x < a , y is greater than 2x . Therefore the curve lies above the line
y = 2x which is tangent at the origin.
Combining all these facts we see that the shape of the curve is as shown in the
above figure.
Example 6 : Trace the curve x3 + y3 = 3axy . (Folium of Descartes)
(Meerut 2007, 08, 10B, 13B; Rohilkhand 08; Purvanchal 07)
Solution : (i) The curve is symmetrical about the line y = x , since its equation
remains unchanged on interchanging x and y .
(ii) Th e cu rve p asses th ro ugh t he o rigin an d t he t angen ts at origin are
3axy = 0 i.e., x = 0 , y = 0 . Since there are two real and distinct tangents at the origin,
therefore the origin is a node on the curve.
(iii) The curve intersects the coordinate axes only at (0, 0) .
(iv) From the equation of the curve we see that x and y cannot be both negative
because then the left hand side of the equation of the curve becomes negative while the
right hand side becomes positive. Therefore the curve does not exist in the third
quadrant.
(v) The curve meets the line y = x at the point (3a ⁄ 2, 3a ⁄ 2) . From the equation
of the curve, we have
dy 3x2 − 3ay
= − 2 ⋅
dx 3y − 3ax
D-316 DIFFERENTIAL CALCULUS
⎛ 3a , 3a ⎞ , dy
At ⎜ ⎟ = − 1 . Therefore the tangent at
⎝ 2 2 ⎠ dx
this point makes an angle of 135o with the positive
direction of x-axis.
(vi) Asymptotes : φ 3 (m) = m 3 + 1 .
The only real root of the equation φ 3 (m) = 0
i.e., m 3 + 1 = 0 , is m = − 1 .
Also φ 2 (m) = − 3am .
For m = − 1 , c is given by c (3m 2) − 3am = 0 .
∴ when m = − 1 , c = − a .
H ence y = − x − a is the only real asymptote of the curve.
Combining all these facts we see that the shape of the curve is as shown in the
figure.
3. y= x (x2 − 1) .
4. y2 = 4ax . (parabola)
5. xy 2 = 4a2 (2a − x) . (Witch of Agnesi)
6. x2 y2 = a2 (x2 + y2) . (Gorakhpur 2006)
13. y2 x = a2 (x − a) .
14. 9ay2 = x (x − 3a) 2.
15. y2 (x + a) = (x − a) 3. (Meerut 2004, 06B)
16. x2 y2 = (1 + y) 2 (4 − y2) .
17. y2 (x + 3a) = x (x − a) (x − 2a) . (Meerut 2005B)
θ 0 π π 2π π
3 2 3
r 2a 3 a a 0
a
2 2
The portion of the curve lying in the region
π < θ < 2π can be drawn by symmetry. Hence the
shape of the curve is as shown in the figure.
Example 2 : Trace the curve r = a cos 2θ .
Solution : (i) The curve is symmetrical about the initial line.
(ii) r = 0 , when cos 2θ = 0 , i.e., 2θ = ± π ⁄ 2 i.e., θ = ± π ⁄ 4 .
Therefore the lines θ = ± π ⁄ 4 are tangents to the curve at the pole.
r is maximum when cos 2θ = 1 . Then θ = 0 and r = a .
dθ 1
(iii) tan φ = r = a cos 2θ ⋅
dr − 2a sin 2θ
1
= − cot 2θ .
2
φ = 90o when 2θ = 0 i.e., θ = 0 . Therefore at the point θ = 0 , the tangent is
perpendicular to the radius vector θ = 0 .
(iv) The following table gives the corresponding values of θ and r :
0 π π π π 2π 3π 5
θ π π
6 4 3 2 3 4 6
r a 1
a 0 −
1
a − a −
1
a 0 1
a a
2 2 2 2
⎛ a⎞ ⎛ a⎞
θ 0 π⁄2 cos− 1 ⎜ − ⎟ cos− 1 ⎜ − ⎟ < θ < π π
⎝ b⎠ ⎝ b⎠
r a+ b a 0 r is negative a− b
⎛ 1⎞
∴ φ = tan− 1 ⎜ ⎟ = constant .
⎝ m⎠
Thus in this curve the angle between the radius vector and the tangent always
remains constant.
H ence the shape of the curve is as shown in the adjoining diagram.
5. r2 = a2 sin 2θ . (Lemniscate)
SINGULAR POINTS : CURVE TRACING D-325
Therefore at the point t = π , whose cartesian coordinates are (aπ, 2a) , the
t a n g e n t i s p e r p e n d i c u l a r t o t h e x- a xi s . W h e n t = − π , x = − aπ , y = 2a ,
(dy ⁄ dx) = − ∞ .
(iii) In this curve y cannot be negative. Therefore the curve lies entirely above
the axis of x . Also no portion of the curve lies in the region y > 2a .
(iv) Corresponding values of x, y and (dy ⁄ dx) for different values of t are given
in the following table :
t − π −
1
π 0 1
π π
2 2
x − aπ 1
− a ( π + 1) 0 1
a ( π + 1) aπ
2 2
y 2a a 0 a 2a
dy ⁄ dx − ∞ − 1 0 1 ∞
True or False:
Write ‘T’ for true and ‘F’ for false statement.
13. If the two branches through a double point on a curve are real and have coincident
tangents there, then the double point is called a node.
D-330 DIFFERENTIAL CALCULUS
14. If the equation of the curve r = f (θ) remains unchanged on changing the signs of
π
r and θ both, the curve is symmetrical about the line θ = ⋅ (Meerut 2001, 09)
2
15. If the equation of a curve remains unchanged even when x and y are interchanged,
the curve is symmetrical about the line y = x.
16. A point of inflexion is a point at which a curve is changing concave upward to
concave downward, or vice-versa.
B
INTEGRAL CALCULUS
C hapters
Reduction Formulae
1.
(For Trigonometric Functions)
Multiple Integrals
4. (Double and Triple Integrals, Change
of Order of Integration)
1. Dirichlet’ ' s and Liouville's Integrals
5.
1. Areas of Curves
6.
Rectification
7. (Lengths of Arcs and Intrinsic
Equations of Plane Curves)
or n In = cos n − 1 x . sin x + (n − 1) In − 2 .
⌠ cos n − 1 x sin x n − 1 ⌠
∴ ⎮ cos n x dx = + ⎮ cos n − 2 x dx .
⌡ n n ⌡
⌠
π ⁄2 ⎡ sin n − 1 x cos x⎤⎥ π ⁄ 2 n− 1⌠
π ⁄2
∴ ⎮ sin n x dx = − ⎢⎢⎢ ⎥⎥ + ⎮ sin n − 2 x dx
⌡0 ⎣ n ⎦0 n ⌡ 0
π ⁄2
n− 1⌠
= 0+ ⎮ sin n − 2 x dx . ...(1)
n ⌡0
Putting (n − 2) in place of n in (1), we have
π ⁄2 π ⁄2
⌠ n− 3⌠
⎮ sin n − 2 x dx = ⎮ sin n − 4 x dx .
⌡0 n − 2 ⌡0
Substituting this value in (1), we have
REDUCTION FORMULAE (For Trigonometric Functions) I-5
π ⁄2 π ⁄2
⌠ n− 1 n− 3 ⌠
⎮ sin n x dx = ⋅ ⋅ ⎮ sin n − 4 x dx
⌡0 n n − 2 ⌡0
π ⁄2
n− 1 n− 3 n− 5 ⌠
= ⋅ ⋅ ⋅ ⎮ sin n − 6 x dx . ...(2)
n n − 2 n − 4 ⌡0
Now two cases arise viz, n is even or odd.
1
In = sin n − 1 (2x) [− cos 2x]
2
− ∫ 1
{(n − 1) sin n − 2 2x . cos 2x .2}. (− cos 2x) dx
2
2x . cos 2x + (n − 1) ∫
1
= − sin n − 1 sin n − 2 2x . cos2 2x dx
2
sin n − 1 2x .cos 2x n − 1
or In = − + In − 2 , is the reduction formula.
2n n
π ⁄2
⌠ (2m) ! π
*Example 2 : Prove that ⎮ sin2m x dx =
⋅ ⋅
⌡0 {2m . m !}2 2
Solution : H ere 2m is even. H ence from article 1.3 (Case II), we get
π ⁄2
⌠ (2m − 1) (2m − 3) … 3 .1 π
⎮ sin2m x dx = ⋅ (Walli’s formula)
⌡0 (2m) (2m − 2) … 4 . 2 2
2m (2m − 1) (2m − 2) … 3 . 2 . 1 π
= ⋅ ⋅
{2m (2m − 2) .... 4 . 2}2 2
[Multiplying Nr. & Dr. by 2m (2m − 2) (2m − 4) … 4 . 2]
(2m) ! π (2m) ! π
= ⋅ = ⋅ ⋅
{2m . m (m − 1) (m − 2) ...... 2 . 1}2 2 {2m . m ! }2 2
2a
⌠ x9 ⁄ 2 dx
Example 3 : Evaluate ⎮ ⋅
⌡0 √(2a − x)
Solution : Put x = 2a sin2 θ , so that dx = 2a .2 sin θ cos θ dθ .
Also when x = 0 , sin2 θ = 0 i.e., θ = 0
and when x = 2a , sin2 θ = 1 i.e., θ = π ⁄ 2 .
2a π ⁄2
⌠ x9 ⁄ 2 dx ⌠ (2a sin2 θ) 9 ⁄ 2. 4a sin θ cos θ dθ
Then ⎮ = ⎮
⌡0 √(2a − x) ⌡0 √(2a − 2a sin2 θ)
π ⁄2
⌠ (2a) 9 ⁄ 2. 4a sin10 θ. cos θ dθ
= ⎮
⌡0 (2a) 1 ⁄ 2. cos θ
π ⁄2
⌠
= (2a) 4. 4a ⎮ sin10 θ dθ
⌡0
9 7 5 3 1 π 63 a5 π
= 64 a5 ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ = ⋅
10 8 6 4 2 2 8
REDUCTION FORMULAE (For Trigonometric Functions) I-7
⌠ tan n − 1 x ⌠
or ⎮ tan n x dx = − ⎮ tan n − 2 x dx ,
⌡ n− 1 ⌡
which is the required reduction formula.
Application : Evaluate ∫ tan4 x dx .
Putting n = 4 in the above reduction formula, we have
∫ tan4 x dx =
1
3
tan3 x − ∫ tan2 x dx
=
1
3
tan3 x − ∫ (sec 2 x − 1) dx
1
= tan3 x − tan x + x .
3
(cot x) n − 1 ⌠
= − − ⎮ cot n − 2 x dx
n− 1 ⌡
cot n − 1 x ⌠
or cot n x dx = − − ⎮ cot n − 2 x dx ,
n− 1 ⌡
= −
1
4
cot4 x − [−
1
2
cot2 x − ∫ cot x dx]
= −
1
4
cot4 x +
1
2
cot2 x + ∫ cot x dx
1 1
= − cot4 x + cot2 x + log sin x .
4 2
sin m + 1 x n− 1 ⌠
= ⋅ cos n − 1 x + ⎮ sin m x cos n − 2 x . (1 − cos2 x) dx
m+ 1 m+ 1⌡
sin m + 1 x cos n − 1 x n− 1 ⌠ n− 1
= + ⎮ sin m x cos n − 2 x dx − I
m+ 1 m+ 1⌡ m + 1 m,n.
Transposing the last term to the left, we have
⎛ n − 1⎞ sin m + 1 x. cos n − 1 x n− 1
Im, n ⎜ 1 + ⎟ = + I
⎝ m + 1⎠ m+ 1 m + 1 m, n − 2
⎛ m + n⎞ sin m + 1 x cos n − 1 x n− 1
or Im, n ⎜ ⎟ = + I .
⎝ m + 1⎠ m+ 1 m + 1 m, n − 2
REDUCTION FORMULAE (For Trigonometric Functions) I-9
⌠ dθ
Example 1 : Evaluate ⎮ ⋅
⌡ sin4 1 θ
2
⌠ dθ ⌠ θ
Solution : We have ⎮ = ⎮ cosec 4 dθ = 2 ∫ cosec 4 x dx , putting θ = 2x .
⌡ sin4 1 θ ⌡ 2
2
⌠ cosec n − 2 x cot x n− 2⌠
But ⎮ cosec n x dx = − + ⎮ cosec n − 2 x dx .
⌡ n− 1 n− 1⌡
= 2 ∫ cosec 4 x dx = −
2 4
cosec 2 x cot x − cot x
3 3
[... x = θ ⁄ 2]
2 1 1 4 1
= − cosec 2 θ cot θ − cot θ .
3 2 2 3 2
∴ ∫ sec 5 θ dθ =
1
4
sec 3 θ tan θ +
3
4
∫ sec 3 θ dθ
=
1
4
sec 3 θ tan θ +
3 1
[
4 2
sec θ tan θ +
1
2
∫ sec θ dθ]
1 3 3
= sec 3 θ tan θ + sec θ tan θ + log (sec θ + tan θ)
4 8 8
1 3 3
= [(1 + x2) 3 ⁄ 2 . x] + x (1 + x2) 1 ⁄ 2 + log {x + √(1 + x2)}.
4 8 8
1 5 5 5 5π
1. (i) − sin5 x cos x − sin3 x cos x − sin x cos x + x. (ii) ⋅
6 24 16 16 32
128 63π
(iii) ⋅ (iv) ⋅
315 512
1 1 1 1
2. (i) [log 2 − ]. (ii) [log 2 − ].
2 2 2 2
1 1 1 1
(iii) sec x tan x + log (sec x + tan x). (iv) √2 + log (√2 + 1).
2 2 2 2
a6 3 1 π
3. (i) [67 √2 + 15 log tan ( π)]. (ii) + ⋅
48 8 48 64
1 1
(iii) tan5 x − tan3 x + tan x − x.
5 3
1 1
5. (log 2 − ).
2 2
⌠π ⁄ 2
1.8 Value of ⎮ sin m x cos n x dx in terms of Γ Function,
⌡0
where m and n are positive integers
We have already derived in article 1.6 that
sin m + 1 x cos n − 1 x n− 1 ⌠
∫ sin m x cos n x dx =
m+ n
+
m+ n⌡
⎮ sin m x cos n − 2 x dx .
I-12 INTEGRAL CALCULUS
π ⁄2
⌠
∴ ⎮ sin m x cos n x dx
⌡0
⎡ sin m+ 1 π ⁄2 π ⁄2
x cos n − 1 x⎤⎥ n− 1 ⌠
= ⎢⎢⎢ ⎥⎥ + ⎮ sin m x cos n − 2 x dx
⎣ m+ n ⎦0 m + n ⌡0
π ⁄2
n− 1 ⌠
= 0+ ⎮ sin m x cos n − 2 x dx
m + n ⌡0
π ⁄2 π ⁄2
⌠ (n − 1) ⌠
i.e., ⎮ sin m x cos n x dx = ⎮ sin m x cos n − 2 x dx ...(1)
⌡0 (m + n) ⌡0
Now four cases arise according as m and n take different types of values, odd
or even.
Case I : When m and n are both even :
Successively applying the formula (1) till the power of cos x becomes zero, we
have
π ⁄2
⌠
⎮ sin m x . cos n x dx
⌡0
π ⁄2
(n − 1) (n − 3) (n − 5) … 1 ⌠
= ⋅ ⋅ ⎮ sin m x dx .
(m + n) (m + n − 2) (m + n − 4) (m + 2) ⌡0
π ⁄2
⌠ m− 1 m− 3 m− 5 1 π
Also ⎮ sin m x dx = ⋅ ⋅ … ⋅ ⋅ [See article 1.3]
⌡0 m m− 2 m− 4 2 2
Therefore,
π ⁄2
⌠ (n − 1) (n − 3) (n − 5) … 1
⎮ sin m x cos n x dx = ×
⌡0 (m + n) (m + n − 2) (m + n − 4) … (m + 2)
(m − 1) (m − 3) (m − 5) … 3 .1 π
⋅
m (m − 2) (m − 4) … 4 .2 2
⎧⎛ n − 1 ⎞ ⎛ n − 3 ⎞ ⎛ n − 5 ⎞ 1⎫ ⎧⎛ m − 1 ⎞ ⎛ m − 3 ⎞ 1⎫
⎨⎜ ⎟ ⎜ ⎟ ⎜ ⎟ … ⎬ ⎨⎜ ⎟ ⎜ ⎟ … ⎬
⎩⎝ 2 ⎠ ⎝ 2 ⎠ ⎝ 2 ⎠ 2⎭ ⎩⎝ 2 ⎠ ⎝ 2 ⎠ 2⎭ π
= ⋅
⎧⎛ m + n ⎞ ⎛ m + n − 2 ⎞ 4 2⎫ 2
⎨⎜ ⎟ ⎜ ⎟ … ⋅ ⎬
⎩⎝ 2 ⎠ ⎝ 2 ⎠ 2 2⎭
⎛ m + 1⎞ ⎛ n + 1⎞
Γ⎜ ⎟ Γ⎜ ⎟
⎝ 2 ⎠ ⎝ 2 ⎠
= ⋅
⎛ m + n + 2⎞
2Γ⎜ ⎟
⎝ 2 ⎠
Similarly, the cases for other values of m and n may be considered and it may
be verified that the result is true in other cases too.
Thus for all positive integral values of m and n, we have
⎛ m + 1⎞ ⎛ n + 1⎞
Γ⎜ ⎟ .Γ⎜ ⎟
π ⁄2 ⎝ 2 ⎠ ⎝ 2 ⎠
⌠
⎮ sin m x cos n x dx = ⋅ (Remember)
⌡0 ⎛ m + n + 2⎞
2Γ⎜ ⎟
⎝ 2 ⎠
REDUCTION FORMULAE (For Trigonometric Functions) I-13
π ⁄2
⌠
Walli’s Formula : [An easy way to evaluate ⎮ sin m x cos n x dx where m and
⌡0
π ⁄2
⌠
n are + ive integers]. We have ⎮ sin m x cos n x dx
⌡
0
(m − 1) (m − 3) (m − 5) … (n − 1) (n − 3) (n− 5) …
= × k,
(m + n) (m + n − 2) (m + n − 4) …
1
where k is π if m and n are both even, otherwise k = 1 . The last factor in each of
2
the three products is either 1 or 2. In case any of m or n is 1, we simply write 1 as
the only factor to replace its product. This formula is equally applicable if any of
m or n is zero provided we put 1 as the only factor in its product and we regard 0
as even.
π ⁄2
⌠
Example 1 : Evaluate ⎮ sin4 x cos2 x dx .
⌡0 (Rohilkhand 2014)
Solution : We know that
⎛ m + 1⎞ ⎛ n + 1⎞
π ⁄2 Γ⎜ ⎟ Γ⎜ ⎟
⌠ ⎝ 2 ⎠ ⎝ 2 ⎠
⎮ sin m x cos n x dx = ⋅
⌡0 ⎛ m + n + 2⎞
2Γ⎜ ⎟
⎝ 2 ⎠
⎛ 4 + 1⎞ ⎛ 2 + 1⎞
Γ⎜ ⎟ ⋅ Γ⎜ ⎟ 5 3
Γ Γ
3
⋅
1
√π ⋅
1
√π
⎝ 2 ⎠ ⎝ 2 ⎠ 2 2 2 2 2 π
∴ the given integral = = = = ⋅
⎛ 4 + 2 + 2⎞ 2Γ4 2 .3 .2. 1 32
2Γ⎜ ⎟
⎝ 2 ⎠
[ ... Γ (n + 1) = n Γ n and Γ = √π ]
1
2
Alternate Solution : U sing Walli’s formula, the given integral
3 .1 .1 π π
= ⋅ = ⋅
6 .4 .2 2 32
π ⁄2
⌠
Example 2 : Evaluate ⎮ sin6 θ dθ .
⌡0
π ⁄2 π ⁄2
⌠ ⌠
Solution : We have ⎮ sin6 θ dθ = ⎮ sin6 θ cos0 θ dθ . (Note)
⌡0 ⌡0
Now m = 6 , n = 0 ; using the G amma function, we have the given integral
⎛ 6 + 1⎞ ⎛ 0 + 1⎞
Γ⎜ ⎟ .Γ⎜ ⎟ 7
Γ ⋅ Γ
1 5 3 1
⋅ ⋅ √π . √π 5π
⎝ 2 ⎠ ⎝ 2 ⎠ 2 2 2 2 2
= = = = ⋅
⎛ 6 + 0 + 2⎞ 2Γ4 2 . (3 .2 . 1) 32
2Γ⎜ ⎟
⎝ 2 ⎠
O therwise, by Walli’s formula, the given integral
5 .3 .1 π 5π
= ⋅ = ⋅
6 .4 .2 2 32
I-14 INTEGRAL CALCULUS
π ⁄6
⌠
Example 3 : Evaluate ⎮ sin2 6θ cos5 3θ dθ .
⌡0
Solution : To bring the given integral into the form of G amma function, put
3θ = x, so that 3 dθ = dx . Also for limits, x = 0 at θ = 0 and x = π ⁄ 2 at θ = π ⁄ 6 .
∴ the given integral
π ⁄2
1 ⌠
= ⎮ sin2 2x cos5 x dx
3 ⌡0
π ⁄2 π ⁄2
1 ⌠ 4 ⌠
= ⎮ (2 sin x cos x) 2 cos5 x dx = ⎮ sin2 x cos7 x dx
3 ⌡0 3 ⌡
0
3 3
Γ .Γ4 Γ . 3 .2. 1
4 24 2 64
= = ⋅ = ⋅
3 2 Γ 11 3 2⋅ 9⋅ 7 5
⋅ ⋅
3
⋅ Γ
3 945
2 2 2 2 2 2
π ⁄4
⌠
Example 4 : Evaluate ⎮ (cos 2θ) 3 ⁄ 2 cos θ dθ .
⌡0
π ⁄4
⌠
Solution : The given integral = ⎮ (1 − 2 sin2 θ) 3 ⁄ 2 cos θ dθ . (Note)
⌡0
Now put √2 sin θ = sin x , so that √2 cos θ dθ = cos x dx .
Also when θ = 0 , sin x = √2 sin 0 = 0 giving x = 0
1 1
and when θ= π , sin x = √2 sin (π ⁄ 4) = 1 giving x = π.
4 2
H ence the given integral
π ⁄2 π ⁄2
⌠ 1 1 ⌠
= ⎮ (1 − sin2 x) 3 ⁄ 2⋅ cos x dx = ⎮ cos3 x . cos x dx
⌡0 √2 √2 ⌡0
π ⁄2 π ⁄2
1 ⌠ 1 ⌠
= ⎮ cos4 x dx = ⎮ sin0 x .cos4 x dx
√2 ⌡0 √2 ⌡0
⎛ 0 + 1⎞ ⎛ 4 + 1⎞
Γ⎜ ⎟ ⋅ Γ⎜ ⎟ Γ ⋅ Γ
5 1
1 ⎝ 2 ⎠ ⎝ 2 ⎠ 1 2 2
= ⋅ =
√2 ⎛ 0 + 4 + 2⎞ √2 2Γ3
2Γ⎜ ⎟
⎝ 2 ⎠
3 1
1 2 ⋅ 2 √π √π 3π
= ⋅ = ⋅
√2 2 .2 .1 16 √2
1
⌠
Example 5 : Evaluate ⎮ x4 (1 − x2) 5 ⁄ 2 dx .
⌡0
Solution : H ere we put x = sin θ , so that dx = cos θ dθ .
And now the new limits are θ = 0 to θ = π ⁄ 2 .
π ⁄2
⌠
Thus the given integral ⎮ sin4 θ (1 − sin2 θ) 5 ⁄ 2 cos θ dθ
⌡0
π ⁄2 π ⁄2
⌠ ⌠
= ⎮ sin4 θ . cos5 θ cos θ dθ = ⎮ sin4 θ cos6 θ dθ
⌡0 ⌡0
REDUCTION FORMULAE (For Trigonometric Functions) I-15
3 .1 .5 .3 .1 π ,
= ⋅ [by Walli’s formula]
10 .8 .6 .4 .2 2
3π
= ⋅
512
a
⌠ π a6
Example 6 : Show that ⎮ x4 (a2 − x2) 1 ⁄ 2 dx = ⋅
⌡0 32 (Bundelkhand 2012)
Solution : Put x = a sin θ so that dx = a cos θ dθ .
1
Also when x = 0 , θ = 0 and when x = a , θ = π.
2
π ⁄2
⌠
∴ the given integral = ⎮ a4 sin4 θ .a cos θ .a cos θ dθ
⌡
0
5 3
π ⁄2 Γ .Γ
⌠ 2 2
= a6 ⎮ sin4 θ cos2 θ dθ = a6
⌡0 2Γ4
3 1 1
⋅ √π √π πa6
6 2 2 2
= a = ⋅
2 .3 .2 .1 32
π ⁄4
⌠
Example 7 : Evaluate ⎮ sin4 x cos2 x dx .
⌡0
Solution : The given integral
π ⁄4
⌠
I= ⎮ (sin2 x cos2 x) sin2 x dx
⌡0
π ⁄4
⌠ 1 1
= ⎮ (4 sin2 x cos2 x) ⋅ (2 sin2 x) dx
⌡0 4 2
π ⁄4
1 ⌠
= ⎮ sin2 2x (1 − cos 2x) dx .
8 ⌡0
Put 2 x = t , so that 2 dx = dt .
Also when x = 0 , t = 0 and when x = π ⁄ 4 , t = π ⁄ 2 .
π ⁄2
1 ⌠ 1
∴ I= ⎮ sin2 t (1 − cos t) ⋅ dt
8 ⌡0 2
π ⁄2 π ⁄2
1 ⎡⎢ ⌠ ⌠ ⎤
= ⎢⎮ sin2 t dt − ⎮ sin2 t cos t dt⎥⎥
16 ⎢⎣ ⌡0 ⌡0 ⎥
⎦
1 ⎡ 1 π 1 .1 ⎤ 1 ⎡ π 1⎤
= ⎢ ⋅ − ⎥ = ⎢ − ⎥⋅
16 ⎣ 2 2 3 .1 ⎦ 16 ⎣ 4 3 ⎦
a
⌠
Example 8 : Evaluate ⎮ x2 √(ax − x2) dx .
⌡0
a a
⌠ ⌠
Solution : We have ⎮ x2 √(ax − x2) dx = ⎮ x5 ⁄ 2 √(a − x) dx .
⌡0 ⌡0
I-16 INTEGRAL CALCULUS
Now put x = a sin2 θ , so that dx = 2a sin θ cos θ dθ , and the new limits are
θ = 0 to θ = π ⁄ 2 .
Thus the given integral
π ⁄2
⌠
= ⎮ (a sin2 θ) 5 ⁄ 2 (a cos2 θ) 1 ⁄ 2. 2a sin θ cos θ dθ
⌡0
π ⁄2
⌠
= 2a4 ⎮ sin6 θ cos2 θ dθ
⌡0
7 3 ⎡5 3 1 1 ⎤
Γ .Γ ⎢ ⋅ ⋅ √π . √π ⎥
5πa4
2 2 ⎢ 2 2 2 2 ⎥
= 2a4 = 2a4 ⎢ ⎥ = ⋅
2Γ5 ⎣ 2 .4 .3 .2 .1 ⎦ 128
∞
⌠ x4 dx
Example 9 : Evaluate ⎮ ⋅
⌡0 (a + x2) 4
2
2a a
⌠ ⌠ 3
5. (i) ⎮ x5 √(2 ax − x2) dx. (ii) ⎮ x (2 ax − x2) 3 ⁄ 2 dx.
⌡0 ⌡0
a a
⌠ ⌠ x4
(iii) ⎮ x2 (2 ax − x2) 5 ⁄ 2 dx. (iv) ⎮ dx.
⌡0 ⌡0 (x + a2) 4
2
√ √
a ⎛ a − x⎞ a ⎛ a 2 − x2 ⎞
⌠ ⌠ ⎜ ⎟
6. (i) ⎮ x2 ⎜ ⎟ dx.
⎝ a + x⎠
(ii) ⎮ x ⎜ ⎟
⎜ a 2 + x2 ⎟ dx.
⌡0 ⌡0 ⎝ ⎠
b
⌠
(iii) ⎮ (x − a) m (b − x) n dx.
⌡a
1
⌠ dx √π Γ (1 ⁄ n)
7. Prove that ⎮ = ⋅
⌡0 √(1 − x n ) n Γ { 1 + (1 ⁄ n)}
2
∞
⌠ dx π x4
8. Show that ⎮ = ⋅
⌡0 (1 + x2) 4 32
9. If m, n are positive integers, then prove that
1 1
⌠ ⌠ 1 . 2 . 3 . …… (m − 1)
⎮ x m− 1 (1 − x) n− 1 dx = ⎮ x n− 1 (1 − x) m− 1 dx = ⋅
⌡0 ⌡0 n (n + 1) …… (n + m − 1)
2 3π 8
1. (i) ⋅ (ii) ⋅ (iii) ⋅
15 512 1287
13 19
Γ( )Γ( ) 5π
2 2 1
(iv) ⋅ 2. (i) ⋅ (ii) ⋅
2 Γ (16) 6 2048
1 2 1
(iii) ⋅ (iv) ⋅ (v) ⋅
3 315 15
π 3π 5π
3. (i) ⋅ (ii) ⋅ (iii) ⋅
32 256 256
πa6 Γ (m + 1) Γ (n + 1) 3π
(iv) ⋅ 4. (i) ⋅ (ii) ⋅
32 Γ (m + n + 2) 128
I-18 INTEGRAL CALCULUS
π (2 m + 1) (2 m − 1) …… 3 . 1
(iii) ⋅ (iv) am+ 2 ⋅
16 (m + 2) (m + 1) m (m − 1) …… 2 . 1
33 7 ⎛ 9π 23 ⎞⎟ ⎡ 45π 2 ⎤⎥
5. (i) πa . (ii) a7 ⎜⎝ − ⎠
⋅ (iii) a8 ⎢⎣ − ⋅
16 32 35 256 7⎦
1 1 ⎡ π 1⎤ 1 2 1
(iv) 3 ⋅ ⎢ − ⎥⋅ 6. (i) a3 ( π − ). (ii) a2 (π − 2).
a 16 ⎣ 4 3 ⎦ 4 3 4
⎡ Γ (m + 1) Γ (n + 1) ⎤⎥
(iii) (b − a) m+ n+ 1 ⎢⎣ ⋅
Γ (m + 2 + 2) ⎦
x n cos m x n ⎡⎢ x n − 1 sin m x n − 1 ⌠ n − 2 ⎤
= − + ⎢ − ⎮x sin m x dx⎥⎥ ,
m m ⎢⎣ m m ⌡ ⎥
⎦
[again integrating by parts regarding cos m x as the 2nd function]
x n cos m x nx n − 1 n (n − 1) ⌠ n − 2
= − + sin m x − ⎮x sin m x dx .
m m 2 m2 ⌡
Above is the required reduction formula. Successively applying this formula we
⌠ ⌠
are left with ⎮ x sin mx dx or ⎮ sin mx dx according as n is odd or even.
⌡ ⌡
(b) ∫ x n cos mx dx . Integrating by parts regarding cos mx as the 2nd function, we
have
⌠ n x n sin m x ⌠ nx n − 1 sin m x dx
⎮ x cos m x dx = − ⎮
⌡ m ⌡ m
x n sin m x n ⌠ n− 1
= − ⎮x sin m x dx
m m⌡
x n sin m x n ⎡⎢ n − 1 ⎛ cos m x⎞ n− 1 ⌠ n− 2 ⎤
= + ⎢x ⋅ ⎜ ⎟ − ⎮x cos m x dx⎥⎥ ,
m m ⎢⎣ ⎝ m ⎠ m ⌡ ⎥
⎦
[again integrating by parts regarding sin m x as 2nd function]
xn sin m x nx n − 1 cos m x n (n − 1) ⌠ n − 2
= + − ⎮x cos m x dx ,
m m2 m2 ⌡
⌠
= (x sin n − 1 x) . (− cos x) + ⎮ cos x [sin n − 1 x + x (n − 1) sin n − 2 x cos x] dx ,
⌡
integrating by parts regarding sin x as 2nd function
= − x sin n − 1 x cos x + ∫ sin n − 1 x cos x dx + (n − 1) ∫ x sin n − 2 x cos2 x dx
1 ⌠
= − x sin n − 1 x cos x + sin n x + (n − 1) ⎮ x sin n − 2 x . (1 − sin2 x) dx
n ⌡
1 ⌠ ⌠
= − x sin n − 1 x cos x + sin n x + (n − 1) ⎮ x sin n − 2 x dx − (n − 1) ⎮ x sin n x dx
n ⌡ ⌡
1
= − x sin n − 1 x cos x + sin n x + (n − 1) In − 2 − (n − 1) In .
n
Transposing the last term to the left, we have
1
In (1 + n − 1) = − x sin n − 1 x cos x + sin n x + (n − 1) In − 2
n
1
or n In = − x sin n − 1 x cos x + sin n x + (n − 1) In − 2
n
x sin n x (n − 1)
or In = −. sin n − 1 x . cos x + + In − 2 ,
n n2 n
which is the required reduction formula.
⌠
Similarly, reduction formula for ⎮ x cos n x dx is
⌡
1.11 ⌠ ax n ⌠ ax n
Reduction Formulae for ⎮
⌡
e sin bx dx and ⎮ e cos bx dx .
⌡
⌠
(a) Let In = ⎮ e a x sin n bx dx
⌡
eax nb ⌠ a x
= sin n bx − ⎮ e sin n − 1 bx cos bx dx , ...(1)
a a ⌡
integrating by parts taking e a x as the 2nd function.
⌠ ax
Now ⎮ e sin n − 1 bx cos bx dx
⌡
eax
= (sin n − 1 bx cos bx)
a
⌠ eax
− ⎮ [(n − 1) b sin n − 2 bx cos2 bx − b sin n bx] dx ,
⌡ a
eax
= (sin n − 1 bx cos bx)
a
b ⌠ ax
− ⎮ e [(n − 1) sin n − 2 bx (1 − sin2 bx) − sin n bx] dx
a⌡
eax b⌠
= (sin n − 1 bx cos bx) − ⎮ e a x [(n − 1) sin n − 2 bx − n sin n bx] dx
a a⌡
eax b⌠ nb
= (sin n − 1 bx cos bx) − (n − 1) ⎮ e a x sin n − 2 bx dx + I .
a a⌡ a n
Substituting this value in (1), we get
eax nb
In = sin n bx − 2 e a x sin n − 1 bx cos bx
a a
b2 ⌠ b2
+ n (n − 1) 2 ⎮ e a x sin n − 2 bx dx − n2 2 In .
a ⌡ a
Transposing the last term to L.H .S., we get
⎛ 2 2⎞ ax 2
⎜ 1 + n b ⎟ I = e (a sin bx − nb cos bx) sin n − 1 bx + n (n − 1) b I .
⎜ ⎟
⎜
⎝ a2 ⎟⎠ n
a2 a2 n − 2
eax n (n − 1)
∴ In = (a sin n bx − nb sin n − 1 bx cos bx) + 2 I ,
2
a + n b2 2 a + n 2 b2 n − 2
which is the required reduction formula.
⌠
Similarly, ⎮ e a x cos n bx dx
⌡
eax n (n − 1)
= (a cos n bx + nb sin bx cos n − 1 bx) + 2 I .
a2 2
+ n b 2 a + n 2 b2 n − 2
Note : The above formulae should not be applied when n is small. In that
case sin n bx and cos n bx are converted in terms of multiples of angles.
1.12 n ax ⌠
⌠ n ax
Reduction Formulae For ⎮
⌡
x e sin bx dx and ⎮ x e cos bx dx .
⌡
eax
We know that ∫ e a x sin bx dx =
r
sin (bx − φ ) ,
1 ⌠
= 2 ⎮ e a x sin (bx − 2φ ) .
r ⌡
1 ⌠ ax 1
Similarly ⎮ e sin (bx − 2φ ) dx = 3 e a x sin (bx − 3φ ) , and so on.
r2 ⌡ r
Now ∫ x n e a x sin bx dx can be easily evaluated by repeatedly integrating by parts
REDUCTION FORMULAE (For Trigonometric Functions) I-21
1.13 ⌠ m
Reduction Formulae for ⎮
⌡
cos x sin nx dx . (Meerut 2013B)
Let Im, n = ∫ cos m x sin nx dx . Integrating by parts regarding sin nx as the 2nd
function, we have
cos m x (− cos nx) ⌠ m cos m − 1 x .(− sin x) (− cos nx) dx
Im, n = − ⎮
n ⌡ n
− cos m x cos nx m ⌠
= − ⎮ cos m − 1 x cos nx .sin x dx . ...(1)
n n ⌡
Now sin {(n − 1) x} = sin (nx − x) = sin nx cos x − cos nx sin x .
∴ cos nx sin x = sin nx cos x − sin (n − 1) x .
cos m x cos nx m ⌠
∴ Im, n = − − ⎮ cos m − 1 x {sin nx cos x − sin (n − 1) x} dx
n n ⌡
cos m x cos nx m ⌠
= − − ⎮ cos m x sin nx dx
n n ⌡
m⌠
+ ⎮ cos m − 1 x sin (n − 1) x dx .
n ⌡
Transposing the middle term to the left and simplifying, we get
cos m x cos nx m ⌠
Im, n = − + ⎮ cos m − 1 x sin (n − 1) x dx ,
m+ n m+ n⌡
which is the required reduction formula.
Deduction : If in the above integral we take the limits of integration as 0 to
1
π , we find that
2
π ⁄2 π ⁄2
⌠ 1 m ⌠
⎮ cos m x sin nx dx = + ⎮ cos m − 1 x sin (n − 1) x dx .
⌡0 m + n m + n ⌡0
1.14 m ⌠
Reduction Formulae for ⎮
⌡
cos x cos nx dx
cos m x sin nx m ⌠
= + ⎮ cos m − 1 x . sin nx sin x dx .
n n ⌡
But cos (n − 1) x = cos nx cos x + sin nx sin x .
∴ sin nx sin x = cos (n − 1) x − cos nx cos x .
cos m x .sin nx m ⌠
H ence Im, n = + ⎮ cos m − 1 x {cos (n − 1) x − cos nx cos x} dx
n n ⌡
cos m x sin nx m ⌠ m⌠
= + ⎮ cos m − 1 x cos (n − 1) x dx − ⎮ cos m x cos nx dx
n n ⌡ n ⌡
cos m x sin nx m m
= + Im − 1, n − 1 − I .
n n n m, n
Transposing the last term to the left, we have
⎛ m⎞ cos m x sin nx m
⎜1 + ⎟ I m, n = + I
⎝ n ⎠ n n m − 1, n − 1
cos m x sin nx m
or Im, n = + I ,
m+ n m + n m − 1, n − 1
which is the required reduction formula.
π ⁄2
⌠
Deduction : ⎮ cos m x cos nx dx
⌡0
⎡ cos x sin nx⎤⎥
m π ⁄2 π ⁄2
m ⌠
= ⎢⎢⎢ ⎥ + ⎮ cos m − 1 x cos (n − 1) x dx
⎣ m + n ⎥⎦ 0 (m + n) ⌡0
π ⁄2
m ⌠
= 0+ ⎮ cos m − 1 x cos (n − 1) x dx
(m + n) ⌡0
π ⁄2
⌠ m
or Im, n = ⎮ cosm x cos nx dx = I .
⌡0 m + n m − 1, n − 1
π ⁄2
⌠
Example 1 : If un = ⎮ x n sin x dx and n > 1 ,
⌡0
1
show that un + n (n − 1) un − 2 = n ( π) n − 1.
2
π ⁄2
⌠
Hence evaluate ⎮ x5 sin x dx .
⌡0
(Kanpur 2005; Gorakhpur 05; Avadh 07; Meerut 12B)
π ⁄2
⌠
Solution : We have un = ⎮ x n sin x dx
⌡0
π ⁄2
π ⁄2 ⌠
= ⎡⎣ x n . (− cos x) ⎤⎦ − ⎮ n . x n − 1 . (− cos x) dx ,
0 ⌡0
[Integrating by parts taking sin x as the 2nd function]
REDUCTION FORMULAE (For Trigonometric Functions) I-23
π ⁄2
⌠
= 0+ n⎮ x n − 1 cos x dx
⌡0
π ⁄2 π ⁄2
⎡ ⌠ ⎤
= n ⎢ ⎧⎨⎩ x n − 1 . sin x ⎫⎬⎭ − ⎮ (n − 1) . x n − 2 . sin x dx⎥ ,
⎣ 0 ⌡0 ⎦
again integrating by parts
π ⁄2
1 ⌠
= n . ( π) n − 1 − n (n − 1) ⎮ x n − 2 sin x dx .
2 ⌡0
1
Thus un = n ( π) n − 1 − n (n − 1) un − 2 . ...(1)
2
1
∴ un + n (n − 1) un − 2 = n ( π) n − 1. Proved.
2
π ⁄2
⌠
Now to evaluate ⎮ x5 sin x dx , put n = 5 in (1).
⌡0
1
Then u5 = 5 ( π) 5 − 1 − 5 (5 − 1) u3
2
1 1
= 5⋅ ( π) 4 − 20 [3 ( π) 3 − 1 − 3 (3 − 1) u1] , putting n = 3 in (1)
2 2
5
= π 4 − 15π 2 + 120 u1.
16
π ⁄2
⌠
Now u1 = ⎮ x sin x dx
⌡0
π ⁄2
π ⁄2 ⌠
= ⎡⎢ x . (− cos x) ⎤⎥ + ⎮ cos x dx
⎣ ⎦0 ⌡0
π ⁄2 ⎡ π ⎤
= ⎡⎢ 0 + sin x⎤⎥ = ⎢ sin − sin 0⎥ = 1 .
⎣ ⎦0 ⎣ 2 ⎦
π ⁄2
⌠ 5π 4
H ence u5 = ⎮ x5 sin x dx = − 15π 2 + 120 .
⌡0 16
Example 2(i) : Integrating by parts twice or otherwise, obtain a reduction
form ula for
∞
⌠
Im = ⎮ e− x sin m x dx , where m ≥ 2
⌡0
= m (m − 1) Im − 2 − m 2 Im
or (1 + m 2) Im = m (m − 1) Im − 2 Proved.
m (m − 1)
or Im = I . ...(1)
1 + m2 m − 2
To evaluate I4 , putting m = 4 in (1), we get
4 (4 − 1) 12
I4 = I = I
1 + 16 2 17 2
12 ⎡ 2 (2 − 1) ⎤ 24
= ⎢ I ⎥= I , [To get I2 , we put m = 2 in (1)]
17 ⎣ 1 + 4 0⎦ 85 0
∞ ∞
24 ⌠ 24 ⌠ 24 ⎡ − x⎤ ∞ 24
= ⎮ e− x sin0 x dx = ⎮ e− x dx = − e ⎦0 = ⋅
85 ⌡0 85 ⌡0 85 ⎣ 85
∞
⌠
Example 2(ii) : If Im = ⎮ e− ax sin m x dx , (a > 0, m ≥ 0); prove that
⌡0
∞
⌠
(m 2 + a2) Im = m (m − 1) Im − 2 and hence evaluate ⎮ e− x sin 4 x dx.
⌡0
(Kanpur 2006)
Solution : Proceed as in E x. 2(i).
∞
⌠
Example 3 : Evaluate ⎮ x e− 2 x cos x dx .
⌡0
(Rohilkhand 2014)
Solution : The given integral
∞
⌠
I = ⎮ x .(e− 2x cos x) dx .
⌡0
1
= − (1 ⁄ 5) [0 − e0 cos (− 2φ )] = cos 2φ ,
5
1
where φ = tan− 1 (b ⁄ a) = tan− 1 (− )
2
1 1
= [{1 − tan2 φ } ⁄ {1 + tan2 φ }] , where tan φ = −
5 2
1 1 1 1 3 3
= [(1 − ) ⁄ (1 + )] = ⋅ = ⋅
5 4 4 5 5 25
1 1⎡ 1 1 ⎤
∴ Im, m = + ⎢ + I 2⎥⎦ ⋅
2m 2 ⎣ 2 (m − 1) 2 m − 2 , m −
⎡. . 1 1 ⎤
⎢ . from (1), I m − 1, m − 1 = + I ⎥
⎣ 2 (m − 1) 2 m − 2 , m − 2⎦
1 1 1
= + 2 + I
2m 2 (m − 1) 22 m − 2 , m − 2
1 1 1 1
= + 2 + 3 + 3 Im − 3, m − 3 , and so on.
2m 2 (m − 1) 2 (m − 2) 2
1 1 1
Finally, Im, m = + 2 + 3
2m 2 (m − 1) 2 (m − 2)
1 1 1
+ …+ m − 2 + + I .
2 . 3 2m − 1 .2 2m − 1 1, 1
π ⁄2
⌠ ⎡1 ⎤π ⁄2 1
But I1, 1 = ⎮ cos x sin x dx = ⎢ sin2 x⎥ = ⋅
⌡0 ⎣2 ⎦0 2
1 ⎡ 22 23 2m ⎤
∴ Im, m = ⎢2 + + + …+ ⎥⋅
2m + 1 ⎣ 2 3 m⎦
n 1
Im, n = I = I . ...(1)
n + n n − 1, n − 1 2 n − 1, n − 1
1
Putting (n − 1) for n in (1), we have In − 1, n − 1 = In − 2 , n − 2 .
2
1 1 1
∴ In, n = ⋅ In − 2 , n − 2 = 2 In − 2 , n − 2 .
2 2 2
Thus by repeated application of (1), we get
1 1
In, n = n In − n, n − n = n I0, 0 .
2 2
π ⁄2 π ⁄2
⌠ ⌠ π ⁄2 π
But I0, 0 = ⎮ cos0 x cos 0 x dx = ⎮ 1 . dx = ⎡⎢ x ⎤⎥ = ⋅
⌡0 ⌡0 ⎣ ⎦0 2
1 1 π
∴ In, n = n
⋅ π= n+ 1 ⋅
2 2 2
⌠ sin nx
Example 6 : Find the reduction form ula for the integral ⎮ dx and show
⌡ sin x
π
⌠ sin nx
that ⎮ dx = π or 0, according as n is odd or even. (Kanpur 2009; Rohilkhand 14)
⌡ sin x
0
Solution : To find the required reduction formula, consider
sin nx − sin (n − 2) x = 2 cos (n − 1) x sin x
sin nx sin (n − 2) x
or − = 2 cos (n − 1) x , dividing both sides by sin x
sin x sin x
sin nx sin (n − 2) x
or = 2 cos (n − 1) x + ⋅
sin x sin x
Integrating both the sides, we have
⌠ sin nx 2 sin (n − 1) x ⌠ sin (n − 2) x
⎮ dx = + ⎮ dx ,
⌡ sin x (n − 1) ⌡ sin x
which is the required reduction formula.
Now let
π π
⌠ sin nx ⎧ 2 sin (n − 1) x⎫ π ⌠ sin (n − 2) x
In = ⎮ dx = ⎨ ⎬ + ⎮ dx = 0 + In − 2 .
⌡0 sin x ⎩ n− 1 ⎭0 ⌡0 sin x
H ence, In = In − 2 = In − 4 = In − 6 = …
i.e., when n is even
π π
⌠ sin 2x ⌠ π
In = I2 = ⎮ dx = 2 ⎮ cos x dx = 2 ⎡⎢ sin x ⎤⎥ = 0
⌡0 sin x ⌡0 ⎣ ⎦0
π ⁄2
⌠ ⎧ m (m − 1) ⎫⎪
8. If I(m, n) = ⎮ cos m x cos nx dx , prove that I(m, n) = ⎪⎨ 2 I .
⌡0 ⎪⎩ m − n 2 ⎬⎪⎭ (m − 2 , n)
(Purvanchal 2014)
π
⌠ ⎛ sin nθ ⎞ 2
9. Prove that ⎮ ⎜ ⎟ dθ = nπ .
⌡0 ⎝ sin θ ⎠
π ⁄2 π ⁄2
⌠ sin (2n − 1) x ⌠ ⎛ sin nx⎞ 2
10. If S n = ⎮ dx , V n = ⎮ ⎜ ⎟ dx, (n is an integer), show
⌡0 sin x ⌡ ⎝ sin x ⎠
0
that S n + 1 − S n = 0, V n + 1 − V n = S n + 1 .
π2 2 4 π 16
1. (i) − + ⋅ (ii) − ⋅ (iii) − ⋅
12 27 9 14 245
πa2 1
(iv) (1 + π 2).
8 6
2
2. (i) π.
3
x sin 3 x cos x sin4 x 3 2 3 3
(ii) − + + x − x sin 2x − cos 2x .
4 16 16 16 32
1 7
(iii) [π − ] .
3 3
1 x
(iv) e [x (cos x + sin x) − cos x] .
2
1 2 1 1
3. (i) x e ax sin (bx − φ ) − xe ax sin (bx − 2φ ) + e ax sin (bx − 3φ ) ,
2 2 4
where a = 2 cos α , b = 2 sin α and φ = α .
1 4 3 1
(ii) π − 3π 2 + 24 . (iii) − π.
16 2 4
1
(iv) (58 − 15π) .
30
⌠ ⌠
2. ⎮ tan n x dx = …… − ⎮ tan n− 2 x dx.
⌡ ⌡
REDUCTION FORMULAE (For Trigonometric Functions) I-29
⌠ n− 2⌠
3. ⎮ sec n x dx = …… + ⎮ sec n− 2 x dx.
⌡ n− 1⌡
π ⁄2
⌠
4. ⎮ sin6 θ dθ = …… .
⌡0
π ⁄2
⌠
5. ⎮ sin4 x cos6 x dx = …… .
⌡0
π ⁄2
⌠ (n − 1)
6. If un = ⎮ θ sin n θ dθ and n > 1, then un = un− 2 + …… .
⌡0 n
True or False:
Write ‘T ’ for true and ‘F ’ for false statem ent.
⌠ cosec n− 2 x cot x n − 2 ⌠
9. ⎮ cosec n x dx = − + ⎮ cosec n− 2 x dx.
⌡ n− 1 n− 1⌡
π ⁄2
⌠ 5π
10. ⎮ cos6 x dx = ⋅
⌡0 16
π ⁄2
⌠ 1
11. If un = ⎮ x n sin x dx and n > 1, then un + n (n − 1) un− 2 = n ( π) n− 1.
⌡0 2
I-30 INTEGRAL CALCULUS
π ⁄2
⌠ 1
12. ⎮ sin2 x cos3 x dx = ⋅
⌡0 15
⌠ m
⎮ x (a + bx n ) p dx can be connected with any one of the following six integrals:
⌡
⌠ m− n ⌠ m
(i) ⎮x (a + bx n ) p dx , (ii) ⎮ x (a + bx n ) p − 1 dx ,
⌡ ⌡
⌠ ⌠
(iii) ⎮ x m + n (a + bx n ) p dx , (iv) ⎮ x m (a + bx n ) p + 1 dx ,
⌡ ⌡
⌠ ⌠
(v) ⎮ x m − n (a + bx n ) p + 1 dx , (vi) ⎮ x m + n (a + bx n ) p − 1 dx .
⌡ ⌡
Rule for Connection : In order to connect the given integral with any one of
the six integrals we use the following rule :
Let P = x λ + 1 (a + bx n ) μ + 1, where λ and μ are the smaller of the indices of
x and (a + bx n ) in the two expressions whose integrals we want to connect.
I-32 INTEGRAL CALCULUS
Find (dP ⁄ dx) and rearrange this as a linear combination of the expressions
whose integrals are to be connected.
Finally integrate both sides and transpose suitably to get the required reduction
formula.
⌠ ⌠
(i) To connect ⎮ x m (a + bx n) p dx with ⎮ x m − n (a + bx n) p dx .
⌡ ⌡
H ere λ = m − n and μ = p (choosing smaller indices for λ and μ).
Let us take P = x λ + 1 (a + bx n ) μ + 1 = x m − n + 1 (a + bx n ) p + 1.
∴ (dP ⁄ dx) = (m − n + 1) x m − n (a + bx n ) p + 1
+ x m − n + 1 ( p + 1) (a + bx n ) p bnx n − 1
= (m − n + 1) x m − n (a + bx n ) p (a + bx n ) + ( p + 1) bn x m (a + bx n ) p
= a (m − n + 1) x m − n (a + bx n ) p + b (m − n + 1) x m (a + bx n ) p
+ ( p + 1) bn x m (a + bx n ) p
= a (m − n + 1) x m − n (a + bx n ) p + bx m (a + bx n ) p (m − n + 1 + pn + n)
i.e., (dP ⁄ dx) = a (m − n + 1) x m − n (a + bx n ) p + b (m + pn + 1) x m (a + bx n ) p .
Thus (dP ⁄ dx) is expressed as a linear combination of the two expressions whose
integrals are to be connected.
Integrating both the sides, we have
⌠ ⌠
P = a (m − n + 1) ⎮ x m − n (a + bx n ) p dx + b (m + np + 1) ⎮ x m (a + bx n ) p dx .
⌡ ⌡
Now putting the value of P and transposing suitably, we get
⌠
b ( pn + m + 1) ⎮ x m (a + bx n ) p dx
⌡
⌠
= x m − n + 1 (a + bx n ) p + 1 − a (m − n + 1) ⎮ x m − n (a + bx n ) p dx
⌡
⌠ m
or ⎮ x (a + bx n) p dx
⌡
x m − n + 1 (a + bx n) p + 1 a (m − n + 1) ⌠ m − n
= − ⎮x (a + bx n) p dx .
b ( pn + m + 1) b ( pn + m + 1) ⌡
⌠ ⌠
(ii) To connect ⎮ x m (a + bx n) p dx with ⎮ x m (a + bx n) p − 1 dx .
⌡ ⌡
H ere λ = m and μ = p − 1 ; (choosing smaller indices for λ and μ).
Now let P = x λ + 1 (a + bx n ) μ + 1 = x m + 1 (a + bx n ) p.
dP
∴ = (m + 1) x m (a + bx n ) p + x m + 1. p (a + bx n ) p − 1 bn x n − 1
dx
= (m + 1) x m (a + bx n ) p + pn . x m (a + bx n ) p − 1 bx n (Note)
= (m + 1) xm (a + bx n ) p + pn .xm (a + bx n ) p − 1 (a + bx n − a)
= ( pn + m + 1) xm (a + bx n ) p − a pn xm (a + bx n ) p − 1.
REDUCTION FORMULAE (For Irrational Algebraic and Transcendental Functions) I-33
Thus (dP ⁄ dx) has been expressed as a linear combination of the two expressions
whose integrals are to be connected.
Integrating both the sides, we have
⌠ ⌠
P = ( pn + m + 1) ⎮ x m (a + bx n ) p dx − apn ⎮ x m (a + bx n ) p − 1 dx .
⌡ ⌡
Now putting the value of P , dividing by ( pn + m + 1) and transposing suitably,
we get
⌠ m x m + 1 (a + bx n) p anp ⌠ m
⎮ x (a + bx n) p dx = + ⎮ x (a + bx n) p − 1 dx .
⌡ np + m + 1 np + m + 1 ⌡
⌠ ⌠
(iii) To connect ⎮ x m (a + bx n) p dx with ⎮ x m + n (a + bx n) p dx .
⌡ ⌡
H ere λ = m and μ = p .
Now let P = x λ + 1 (a + bx n ) μ + 1 = x m + 1 (a + bx n ) p + 1.
dP
∴ = x m + 1 ( p + 1) (a + bx n ) p nb x n − 1 + (m + 1) x m (a + bx n ) p + 1
dx
= bn ( p + 1) x m + n (a + bx n ) p + (m + 1) x m (a + bx n ) p (a + bx n )
(Note)
= bn ( p + 1) x m + n (a + bx n ) p + (m + 1) ax m (a + bx n ) p
+ b (m + 1) x m + n (a + bx n ) p
= b (np + n + m + 1) (a + xm + n bx n ) p
+ (m + 1) ax m (a + bx n ) p,
which is linear combination of the two expressions whose integrals are to be
connected.
Integrating both sides, we have
⌠
P = b (np + n + m + 1) ⎮ x m + n (a + bx n ) p dx
⌡
⌠
+ (m + 1) a ⎮ x m (a + bx n ) p dx .
⌡
Now putting the value of P , dividing by a (m + 1) and transposing suitably, we
get
⌠ m x m + 1 (a + bx n) p + 1
⎮ x (a + bx n) p dx =
⌡ a (m + 1)
b (np + n + m + 1) ⌠ m + 1
− ⎮x (a + bx n) p dx .
a (m + 1) ⌡
⌠ ⌠
(iv) To connect ⎮ x m (a + bx n) p dx with ⎮ x m (a + bx n) p − 1 dx .
⌡ ⌡
H ere λ = m and μ = p , λ being the lesser index of x , and μ being the lesser
index of (a + bx n ) in both the integrals.
Now let P = x λ + 1 (a + bx n ) μ + 1 = x m + 1 (a + bx n ) p + 1.
dP
∴ = x m + 1 ( p + 1) (a + bx n ) p. nb x n − 1 + (m + 1) x m (a + bx n ) p + 1
dx
= n ( p + 1) x m (a + bx n ) p bx n + (m + 1) x m (a + bx n ) p + 1 (Note)
I-34 INTEGRAL CALCULUS
= n ( p + 1) x m (a + bx n ) p (a + bx n − a) + (m + 1) x m (a + bx n ) p + 1
= n ( p + 1) x m (a + bx n ) p + 1 − an ( p + 1) x m (a + bx n ) p
+ (m + 1) x m (a + bx n ) p + 1
= (np + n + m + 1) x m (a + bx n ) p + 1 − an ( p + 1) x m (a + bx n ) p
i.e., (dP ⁄ dx) is a linear combination of the two expressions whose integrals are to be
connected.
Integrating both the sides, we have
⌠
P = {n ( p + 1) + m + 1)} ⎮ x m (a + bx n ) p + 1 dx
⌡
⌠
− an ( p + 1) ⎮ x m (a + bx n ) p + 1 dx .
⌡
Now putting the value of P , dividing by {an ( p + 1)} and transposing suitably,
⌠ m
we get ⎮ x (a + bx n) p dx
⌡
x m + 1 (a + bx n) p + 1 (np + n + m + 1) ⌠ m
= − + ⎮ x (a + bx n) p + 1 dx .
an ( p + 1) an (n + 1) ⌡
⌠ ⌠
(v) To connect ⎮ x m (a + bx n) p dx with ⎮ x m − n (a + bx n) p + 1 dx .
⌡ ⌡
(m − n + 1) ⌠ m − n
− ⎮x (a + bx n) p + 1 dx .
bn ( p + 1) ⌡
⌠ ⌠
(vi) To connect ⎮ x m (a + bx n) p dx with ⎮ x m + n (a + bx n) p − 1 dx .
⌡ ⌡
H ere λ = m and μ = p − 1 .
REDUCTION FORMULAE (For Irrational Algebraic and Transcendental Functions) I-35
∴ P = x λ + 1 (a + bx n ) μ + 1 = x m + 1 (a + bx n ) p.
dP
And = x m + 1. p (a + bx n ) p − 1 bn x n − 1 + (m + 1) x m (a + bx n ) p
dx
= bp nx m + n (a + bx n ) p − 1 + (m + 1) x m (a + bx n ) p.
Thus (dP ⁄ dx) is expressed as a linear combination of the two expressions whose
integrals are to be connected.
Integrating both sides, we have
⌠ ⌠
P = b pn ⎮ x m + n (a + bx n ) p − 1 dx + (m + 1) ⎮ x m (a + bx n ) p dx
⌡ ⌡
⌠ ⌠
or (m + 1) ⎮ x m (a + bx n ) p dx = P − bn p ⎮ x m + n (a + bx n ) p − 1 dx .
⌡ ⌡
Now putting the value of P and dividing by (m + 1) , we get
⌠ m
⎮ x (a + bx n) p dx
⌡
x m + 1 (a + bx n) p bnp ⌠ m + n
= − ⎮x (a + bx n) p − 1 dx .
(m + 1) (m + 1) ⌡
1
⌠
∴ ( p + 1 + nq) ⎮ x p (1 − x q) n dx
⌡
0
1
1 ⌠
= ⎡⎣ x p + 1 (1 − x q) n ⎤⎦ + nq ⎮ x p (1 − x q) n − 1 dx ,
0 ⌡0
putting the value of P , transposing and also
putting the limits of integration
1
⌠
= 0 + nq ⎮ x p (1 − x q) n − 1 dx .
⌡0
⌠
1
⌠
1 ⎡⎢ x p + 1⎤ 1 1
But I0 = ⎮ x p (1 − x q) 0 dx = ⎮ x p dx = ⎥⎥ = ⋅
⌡0 ⌡ ⎢⎢⎣ p + 1 ⎥⎦ 0 p+ 1
0
nq (n − 1) q … q 1
∴ In = ⋅ ⋅ ⋅
nq + p + 1 (n − 1) q + p + 1 q+ p+ 1 p+ 1
a
⌠
Example 2 : If In denotes ⎮ (a2 − x2) n dx , and n > 0 , prove that
⌡0
2na2
In = I . (Avadh 2005; Kanpur 06)
2n + 1 n − 1
a
⌠
Hence evaluate ⎮ (a2 − x2) 3 dx.
⌡0
a
⌠
Solution : We have In = ⎮ (a2 − x2) n .1 dx (Note)
⌡0
a
a ⌠
= ⎡⎣ (a2 − x2) n . x⎤⎦ 0 − ⎮ n (a2 − x2) n − 1 (− 2x) . x dx ,
⌡0
integrating by parts taking unity as the secnod function
a
⌠
= 0 + 2n ⎮ (a2 − x2) n − 1 x2 dx , [... n > 0]
⌡0
a
⌠
= − 2n ⎮ (a2 − x2) n − 1 . {(a2 − x2) − a2} dx ,
⌡0
[... x2 = − {(a2 − x2) − a2}]
REDUCTION FORMULAE (For Irrational Algebraic and Transcendental Functions) I-37
a a
⌠ ⌠
= − 2n ⎮ (a2 − x2) n dx + 2na2 ⎮ (a2 − x2) n − 1 dx
⌡0 ⌡0
= − 2n In + 2na2 In − 1 .
∴ (1 + 2n) In = 2na2 In − 1
2na2
or In = I ...(1) Proved.
2n + 1 n − 1
6
∴ I3 = a2 I2 , putting n = 3 in (1)
7
6 4
= a2. [ a2 I1] , putting n = 2 in (1) to get I2 in terms of I1
7 5
24 4
= a I1 .
35
a a
⌠ 24a4 ⌠
Thus I3 = ⎮ (a2 − x2) 3 dx = ⎮ (a2 − x2) dx
⌡0 35 ⌡0
a
24a4 ⎡ 2 x3 ⎤ 24a4 ⎡ 3 a3 ⎤ 16a7
= ⎢a x − ⎥ = ⎢a − ⎥ = ⋅
35 ⎣ 3 ⎦0 35 ⎣ 3⎦ 35
⌠ dx , where n is Positive
2.2 Reduction Formulae for ⎮ 2
⌡ (x + a2) n
x 2n − 3
or In = + I ,
2a2 (n − 1) (x2 + a2) n − 1 2a2 (n − 1) n− 1
which is the required reduction formula.
(Bundelkhand 2010)
⌠ ⌠
Let Im = ⎮ x m √(2ax − x2) dx = ⎮ x m + 1 ⁄ 2 √(2a − x) dx .
⌡ ⌡
Integrating by parts taking √(2a − x) as the 2nd function, we have
(2a − x) 3 ⁄ 2 ⌠ ⎛⎜ 1⎞ (2a − x) 3 ⁄ 2
Im = x m + 1 ⁄ 2 3 − ⎮ ⎝ m + ⎟⎠ x m − 1 ⁄ 2 3 dx
( ) . (− 1) ⌡ 2 ( ) . (− 1)
2 2
2 2m + 1 ⌠ m − 1 ⁄ 2
= − x m − 1 x3 ⁄ 2 (2a − x) 3 ⁄ 2 + ⎮x (2a − x) √(2a − x) dx
3 3 ⌡
2 m− 1 2m + 1 ⌠
= − x (2ax − x2) 3 ⁄ 2 + ⎮ 2ax m − 1 ⁄ 2 √(2a − x) dx
3 3 ⌡
2m + 1 ⌠ m − 1 ⁄ 2
− ⎮x . x √(2a − x) dx
3 ⌡
2 m− 1 (2m + 1) 2a ⌠ m − 1 1 ⁄ 2
= − x (2ax − x2) 3 ⁄ 2 + ⎮x x √(2a − x) 1 ⁄ 2 dx
3 3 ⌡
2m + 1 ⌠ m − 1 ⁄ 2 1 ⁄ 2 1 ⁄ 2
− ⎮x . x x (2a − x) 1 ⁄ 2 dx
3 ⌡
2 m− 1 2 (2m + 1) a ⌠ m − 1
= − x (2ax − x2) 3 ⁄ 2 + ⎮x √(2ax − x2) dx
3 3 ⌡
2m + 1 ⌠ m
− ⎮ x √(2ax − x2) dx
3 ⌡
2 m− 1 2 (2m + 1) 2m + 1
= − x (2ax − x2) 3 ⁄ 2 + 2Im − 1 − Im .
3 3 3
Transposing the last term to the left, we have
⎛ 2m + 1 ⎞ 2 m− 1 2(2m + 1) a
⎜1 + ⎟ I = − x (2ax − x2) 3 ⁄ 2 + Im − 1
⎝ 3 ⎠ m 3 3
2 (m + 2) 2 2 (2m + 1) a
or Im = − x m − 1 (2ax − x2) 3 ⁄ 2 + Im − 1
3 3 3
x m − 1 (2ax − x2) 3 ⁄ 2 (2m + 1) a
or Im = − + Im − 1 ,
m+ 2 m+ 2
which is the required reduction formula.
I-40 INTEGRAL CALCULUS
⌠ x n
(b) ⎮ (a ⁄ x ) dx .
⌡
⌠ ⌠
We have ⎮ (a x ⁄ x n ) dx = ⎮ a x x− n dx .
⌡ ⌡
Now integrate by parts taking x− n as the 2nd function.
⌠
2.6 Reduction Formulae for ⎮ xm (log x) n dx .
⌡
∞
⌠
Example 1 : Evaluate ⎮ e− x x n dx , n being a positive integer.
⌡0
(Rohilkhand 2007; Kanpur 10, 12)
Solution : Integrating by parts regarding e− x as the 2nd function, we get
∞ ∞
⌠ ∞ ⌠
⎮ e− x x n dx = ⎡⎣ − x n e− x⎤⎦ 0 + ⎮ ne− x x n − 1 dx .
⌡0 ⌡0
lim lim x n ∞
Now x → ∞ x n e− x = x → ∞ x , which is of the form ∞ ⋅
e
∴ differentiating the numerator and the denominator separately, we get
lim x n lim nx n − 1 lim n (n − 1) … 1
x →∞ x = x →∞ x
= … = x →∞ = 0.
e e ex
H ence ⎡ − x n e− x⎤ ∞ = − lim x n e− x − 0 = 0 − 0 = 0 .
⎣ ⎦0 x →∞
∞ ∞
⌠ ⌠
Therefore ⎮ e− x x n dx = n ⎮ e− x x n − 1 dx . ...(1)
⌡0 ⌡0
Now applying the reduction formula (1) repeatedly, we get
∞ ∞
⌠ ⌠
⎮ e− x x n dx = n (n − 1) (n − 2) … 2 . 1 ⎮ e− x x0 dx
⌡0 ⌡
0
∞
⌠ ∞
= n ! ⎮ e− x dx = n ! ⎡⎣ − e− x⎤⎦ 0
⌡0
⎡ 1 ⎤⎥ ∞
= n ! ⎢⎢ − ⎥ = (n !) .1 = n ! .
⎢⎣ e x⎥⎦ 0
REDUCTION FORMULAE (For Irrational Algebraic and Transcendental Functions) I-41
1
⌠
Example 2 : Evaluate ⎮ x m (log x) n dx , when m ≥ 0 and n is an integer ≥ 0 .
⌡0
1
⌠
Solution : Let Im, n = ⎮ x m (log x) n dx .
⌡0
But log 1 = 0 .
lim m + 1 lim (log x) n
Also x →0 x (log x) n = x → 0 − ( m + 1) , [form ∞ ⁄ ∞ ]
x
lim n (log x) n − 1 . (1 ⁄ x)
= x →0
− (m + 1) x− (m + 2)
lim ⎛ n ⎞ (log x) n − 1
= x →0 ⎜ − ⎟ ⋅
⎝ m + 1 ⎠ x − (m + 1)
lim 1
= (some number) × x → 0 − ( m + 1)
x
lim
= some number × x → 0 x m + 1 = 0 .
n
∴ Im, n = − I ...(1)
(m + 1) m, n − 1
n ⎛ n − 1⎞
= − ⋅ ⎜− ⎟ I , applying (1)
(m + 1) ⎝ m + 1 ⎠ m, n − 2
n (n − 1)
= (− 1) 2 I = …= …
(m + 1) 2 m, n − 1
Proceeding similarly by successive application of (1), we have ultimately
n (n − 1) (n − 2) … 2 .1
Im, n = (− 1) n Im, 0 .
(m + 1) n
1 1
⌠ ⌠
But Im, 0 = ⎮ x m (log x) 0 dx = ⎮ x m dx
⌡0 ⌡0
⎡ x m + 1 ⎤⎥ 1
1
= ⎢⎢⎢ ⎥ = ⋅
⎣ m + 1 ⎥⎦ 0 m+ 1
I-42 INTEGRAL CALCULUS
n! 1
∴ Im, n = (− 1) n ⋅ ⋅
(m + 1) n (m + 1)
n!
= (− 1) n ⋅
(m + 1) n + 1
⌠
Example 3 : Find the reduction form ula for ⎮ {x m ⁄ (log x) n } dx .
⌡
Solution : We have
⌠ xm ⌠ ⎡ 1 1⎤
⎮ dx = ⎮ x m + 1 ⎢ ⋅ ⎥ dx
⌡ (log x) n ⌡ ⎣ (log x) n x⎦
⌠ ⎡ 1⎤
= ⎮ x m + 1 ⋅ ⎢ (log x) − n ⎥ dx . (Note)
⌡ ⎣ x⎦
Now integrating by parts regarding x m + 1 as the first function, we have
⌠ x m dx (log x) − n + 1 ⌠ (log x) − n + 1
⎮ = xm + 1 − ⎮ (m + 1) x m dx
⌡ (log x) n − n+ 1 ⌡ − n+ 1
xm + 1 m+ 1⌠ xm
= − + ⎮ dx ,
(n − 1) (log x) n − 1 n − 1 ⌡ (log x) n − 1
which is the required reduction formula.
⌠ x m dx ,
3. If Im, n = ⎮ prove that
⌡ (1 + x2) n
2 (n − 1) Im, n = − x m − 1 (x2 + 1) 1 − n + (m − 1) Im − 2, n − 1 .
x
⌠ x n dx ,
4. If φ (n) = ⎮ prove that
⌡0 √(x − 1)
(2n + 1) φ (n) = 2 x n √(x − 1) + 2 n φ (n − 1) .
REDUCTION FORMULAE (For Irrational Algebraic and Transcendental Functions) I-43
∞
⌠ 1
5. If In denotes ⎮ dx,
⌡0 (a2 + x2) n
where n is a positive integer ≥ 2 , prove that
2n − 3
In= I .
2 a2 (n − 1) n− 1
∞
⌠ 1
H ence or otherwise evaluate ⎮ dx.
⌡0 (a2 + x2) 4
2a
⌠
6. If Im = ⎮ x m √(2ax − x2) dx , prove that
⌡0
2m m ! . (m + 2) ! Im = am + 2 (2m + 1) ! π .
2a
⌠
H ence or otherwise evaluate ⎮ x3 √(2ax − x2) dx . (Bundelkhand 2007; 10)
⌡0
⌠
7. If In = ⎮ x n (a − x) 1 ⁄ 2 dx , prove that
⌡
(2n + 3) In = 2an In − 1 − 2x n (a − x) 3 ⁄ 2 . (Bundelkhand 2011)
a
⌠
H ence evaluate ⎮ x2 √(ax − x2) dx .
⌡0
⌠
8. If un = ⎮ x n (a2 − x2) 1 ⁄ 2 dx , prove that
⌡
x n − 1 (a2 − x2) 3 ⁄ 2 n − 1 2
un = − + a un − 2 .
n+ 2 n+ 2
a
⌠
H ence evaluate ⎮ x4 √(a2 − x2) dx .
⌡0
∞
⌠ n!
9. Show that ⎮ e− ax x n dx = n + 1 ,
⌡0 a
where a is a positive quantity and n is a positive integer.
1
⌠
10 E valuate ⎮ (log x) 4 x m dx .
⌡0
11. If m and n are positive integers, and
1
⌠
f (m , n) = ⎮ x n − 1 (log x) m dx ,
⌡0
prove that
f (m , n) = − (m ⁄ n) f (m − 1, n) .
Deduce that f (m , n) = (− 1) m . m ! ⁄ nm + 1.
∞
⌠ x
12. E valuate ⎮ dx .
⌡0 (1 + e x)
I-44 INTEGRAL CALCULUS
Beta function is also called the Eulerian integral of the first kind.
I-46 INTEGRAL CALCULUS
⎡ . . ⌠a a ⎤
⎢ . ⎮ f (x) dx = ⌠
⎮ f (a − x) dx⎥⎥
⎢ ⌡0 ⌡0
⎢ ⎥
⎣ ⎦
1 1
⌠ ⌠
= ⎮ (1 − x) m − 1 x n − 1 dx = ⎮ x n − 1 (1 − x) m − 1 dx
⌡0 ⌡0
= B (n, m), by the def. of Beta function.
H ence B (m, n) = B (n, m).
(ii) If m or n is a positive integer, B (m, n) can be evaluated in an explicit form:
Case I : When n is a positive integer : If n = 1, the result is obvious because
⌠
1
⌠
1 ⎡⎢ x m ⎤⎥ 1 1
B (m, 1) = ⎮ x m − 1 (1 − x) 1 − 1 dx = ⎮ x m − 1 dx = ⎢⎢ ⎥ = ⋅
⌡0 ⌡0 ⎣ m ⎥⎦
0 m
So let us take n > 1. We have
1
⌠
B (m, n) = ⎮ x m − 1 (1 − x) n − 1 dx
⌡0
1
⌠
= ⎮ (1 − x) n − 1 x m − 1 dx
⌡0
⎡ 1 1
x m ⎤⎥ ⌠ xm
= ⎢⎢⎢ (1 − x) n − 1 ⋅ ⎥⎥ − ⎮ (n − 1) (1 − x) n − 2 (− 1) ⋅ dx ,
⎣ m ⎦ 0 ⌡0 m
(n − 1) ! ⎡ x m + n − 1 ⎤⎥ 1
= ⋅ ⎢⎢⎢ ⎥ ⋅
m (m + 1) (m + 2) ...... (m + n − 2) ⎣ m + n − 1 ⎥⎦ 0
1
∴ B (m, n) = ⋅
m (m + 1) (m + 2) ..... (m + n − 2) (m + n − 1)
1
⌠ dy ,
= ⎮ y (m − 1) ⁄ 2 (1 − y) n ⋅ putting x2 = y so that 2x dx = dy
⌡0 2
1
1⌠
= ⎮ y (m − 1) ⁄ 2 (1 − y) n dy
2 ⌡0
1
1⌠
= ⎮ y [(m + 1) ⁄ 2] − 1 (1 − y) (n + 1) − 1 dy
2 ⌡0
⎛ ⎞
1 1
= B ⎜⎝ (m + 1), n + 1⎟⎠ ⋅
2 2
1 1
⌠ x2 ⌠
(ii) We have ⎮ dx = ⎮ x2 (1 − x5) − 1 ⁄ 2 dx
⌡0 √(1 − x5) ⌡0
1
⌠ 1
= ⎮ x2 ⋅ 4 (1 − x5) − 1 ⁄ 2 . x4 dx
⌡0 x
1
⌠
= ⎮ x− 2 (1 − x5) − 1 ⁄ 2 x4 dx
⌡0
1
⌠ 1
= ⎮ y − 2 ⁄ 5 (1 − y) − 1 ⁄ 2 ⋅ dy, putting x5 = y so that 5x4 dx = dy
⌡0 5
1
1⌠
= ⎮ y − 2 ⁄ 5 (1 − y) − 1 ⁄ 2 dy
5 ⌡0
1
1⌠
= ⎮ y (3 ⁄ 5) − 1 (1 − y) (1 ⁄ 2) − 1 dy
5 ⌡0
1 ⎛3 , 1⎞
= B⎜ ⎟ ⋅
5 ⎝5 2⎠
(iii) Proceed as in part (i).
= (b − a) m + n − 1 B (m, n)
Γm Γn ⎡ Γ m Γ n ⎤⎥
= (b − a) m+ n− 1 ⋅ ⋅ ⎢ ... B (m, n) =
Γ (m + n) ⎣ Γ (m + n) ⎦
1
⌠ x m − 1 (1 − x) n − 1 1
Example 4 : Show that ⎮ dx = B (m, n) .
⌡0 (a + bx) m + n (a + b) m . a n
Solution : The given integral
1
⌠ x m − 1 (1 − x) n − 1
I= ⎮ dx
⌡0 (a + bx) m + n
1
⌠ ⎛ x ⎞m− 1 ⎛ 1− x⎞n− 1 1
= ⎮ ⎜ ⎟ ⋅ ⎜ ⎟ ⋅ dx . [ Note]
⌡0 ⎝ a + bx⎠ ⎝ a + bx⎠ (a + bx) 2
I-50 INTEGRAL CALCULUS
x y (a + bx) . 1 − x . b dy
Put = so that dx =
a + bx a + b (a + bx) 2 a + b
1 dy
i.e., dx = ⋅
(a + bx) 2 a (a + b)
1− x 1 a − ax 1 ⎡ a + bx − ax − bx⎤ 1⎡ x (a + b) ⎤ 1− y
Further = = ⎢ ⎥ = ⎢1 − ⎥ = ⋅
a + bx a a + bx a ⎣ a + bx ⎦ a⎣ a + bx ⎦ a
Also when x = 0, y = 0 and when x = 1, y = 1.
1
⌠ ⎛ y ⎞ m − 1 ⎛ 1 − y⎞ n − 1 dy
∴ I= ⎮ ⎜ ⎟ ⎜ ⎟ ⋅
⌡0 ⎝ a + b⎠ ⎝ a ⎠ a (a + b)
1
1 ⌠
= ⎮ y m − 1 (1 − y) n − 1 dy
(a + b) m . a n ⌡0
B (m, n)
= ⋅
(a + b) m . an
B (m + 1, n) m
Example 5 : Prove that = ⋅
B (m, n) m+ n
Solution : We have B (m + 1, n) = B (n, m + 1)
[By the symmetry of Beta function]
1
⌠
= ⎮ x n − 1 (1 − x) (m + 1) − 1 dx
⌡0
1
⌠
= ⎮ (1 − x) m x n − 1 dx [ Note]
⌡0
⎡ 1 1
x n ⎤⎥ ⌠ xn
= ⎢⎢⎢ (1 − x) m ⋅ ⎥⎥ − ⎮ m (1 − x) m − 1 (− 1) ⋅ dx ,
⎣ n ⎦0 ⌡0 n
(integrating by parts)
1
m ⌠
= 0+ ⎮ x n − 1. x (1 − x) m − 1 dx
n ⌡0
1
m ⌠
= ⎮ x n − 1 [1 − (1 − x)] (1 − x) m − 1 dx
n ⌡0
⎡⌠1 1 ⎤
=
m ⎢ ⎮ x n − 1 (1 − x) m − 1 dx − ⌠
⎮ x n − 1 (1 − x) m dx⎥⎥
⎢⌡ ⌡0
n ⎢ 0
⎣
⎥
⎦
m ⎡
= B (n, m) − B (n, m + 1) ⎤⎥
n ⎢⎣ ⎦
m m
= B (m, n) − B (m + 1, n)
n n
⎛ m⎞ m
or ⎜1 + ⎟ B (m + 1, n) = B (m, n) [By transposition]
⎝ n⎠ n
or (n + m) B (m + 1, n) = m B (m, n)
BETA AND GAMMA FUNCTIONS I-51
B (m + 1, n) m
or = ⋅
B (m, n) m+ n
∞
⌠ xm − 1 − xn − 1
Example 1 : Prove that ⎮ dx = 0 , m > 0 , n > 0 .
⌡0 (1 + x) m + n
Solution : The given integral is
∞ ∞
⌠ xm − 1 ⌠ xn − 1
= ⎮ dx − ⎮ dx
⌡0 (1 + x) m + n ⌡0 (1 + x) m + n
= B (m, n) − B (n, m) = B (m, n) − B (m, n) = 0 .
∞
⌠ xm − 1
Example 2 : Express ⎮ dx in term s of Beta function where
⌡0 (a + bx) m + n
m > 0, n > 0, a > 0, b > 0.
Solution : In the given integral put bx = ay i.e., x = (a ⁄ b) y so that
dx = (a ⁄ b) dy . When x = 0 , y = 0 and when x → ∞ , y → ∞ .
I-52 INTEGRAL CALCULUS
∞ ∞
⌠ xm − 1 ⌠ ⎛ a ⎞m − 1 1 a
∴ ⎮ dx = ⎮ ⎜ y⎟ ⋅ ⋅ dy
⌡0 (a + bx) m + n ⌡0 ⎝ b ⎠ (a + ay) m + n b
∞ ∞
⌠ am − 1 y m − 1 a 1 ⌠ ym − 1
= ⎮ dy = n m ⎮ dy
⌡0 b m − 1 .a m + n (1 + y) m + n b a b ⌡0 (1 + y) m + n
1
= B (m, n) . [By article 3.3]
a n bm
∞ ∞
⌠ ⌠
∴ Γ (m) = z m ⎮ e− z x. x m − 1 dx = ⎮ z m e− z x x m − 1 dx .
⌡0 ⌡0
Multiplying both sides by e− z zn − 1, we get
∞
⌠
Γ (m) e− z z n − 1 = ⎮ e− z (1 + x) z m + n − 1 x m − 1 dx . ...(1)
⌡0
Now integrating both sides of (1) with respect to z from 0 to ∞ , we get
∞ ∞ ⎡ ∞ ⎤
⌠ ⌠ ⎢ ⌠
Γ (m) ⎮ e− z z n − 1 dz = ⎮ ⎢ ⎮ e− z (1 + x) z m + n − 1 x m − 1 dx⎥⎥ dz
⌡0 ⌡0 ⎢ ⌡0 ⎥
⎣ ⎦
∞ ⎡ ∞ ⎤
⌠ ⌠⎢⎮
or Γ (m) Γ (n) = ⎮ ⎢⌡ e− z (1 + x) z m + n − 1 dz⎥⎥ x m − 1 dx
⌡0 ⎢ 0 ⎥
⎣ ⎦
∞
⌠ Γ (m + n) m − 1
= ⎮ x dx [By article 3.6, part (ii)]
⌡0 (1 + x) m + n
∞
⌠ xm − 1
= Γ (m + n) ⎮ dx
⌡0 (1 + x) m + n
= Γ (m + n) . B (m, n) , by § 3 .
Γ (m) Γ (n)
∴ B (m, n) = ⋅
Γ (m + n)
1
⌠ Γ (m) Γ (n)
Thus remember that ⎮ x m − 1 (1 − x) n − 1 dx = ⋅ (Kanpur 2009)
⌡0 Γ (m + n)
∞
⌠ xn − 1
Proof : We know that B (m, n) = ⎮ dx , [See § 3]
⌡0 (1 + x) m + n
Γ (m) Γ (n) ,
and B (m, n) = where m > 0 and n > 0 .
Γ (m + n)
∞
Γ (m) Γ (n) ⌠ xn − 1
∴ = ⎮ dx .
Γ (m + n) ⌡0 (1 + x) m + n
Putting m + n = 1 or m = 1 − n in the above relation, we get
∞
Γ (1 − n) Γ (n) ⌠ x n − 1
= ⎮ dx , where 0 < n < 1 .
Γ (1) ⌡0 1 + x
[Note that m > 0 ⇒ 1 − n > 0 ⇒ n < 1 .]
But Γ (1) = 1 . Also
∞
⌠ xn − 1 π
⎮ dx = ⋅ [ Remember it]
⌡0 1+ x sin nπ
π ,
∴ Γ (n) Γ (1 − n) = where 0 < n < 1 .
sin nπ
BETA AND GAMMA FUNCTIONS I-55
⎛ ⎞
3.8 The Value of Γ ⎜⎝ 12 ⎟⎠
1
To prove that Γ ( ) = √π . (Agra 2000; Meerut 12)
2
Γ (m) Γ (n)
Proof : We know that B (m, n) = ...(1)
Γ (m + n)
1 , n = 1 , then from (1), we have
If we take m =
2 2
1 1 1
⎛ 1 1⎞ Γ ( ) Γ ( ) [Γ ( )]2 ⎡ 2
⎛ 1⎞ ⎤
[... Γ (1) = 1]
2 2 2
B ⎜⎝ , ⎟⎠ = = = ⎢Γ ⎜ ⎟⎥ ⋅
⎣ ⎝ 2⎠ ⎦
2 2 1
Γ( + )
1 Γ (1)
2 2
1 1 , 1
Thus [ Γ ( )]2 = B ( )
2 2 2
1
⌠
= ⎮ x1 ⁄ 2 − 1 (1 − x) 1 ⁄ 2 − 1 dx ,
⌡0
by the definition of Beta function
1
⌠
= ⎮ x− 1 ⁄ 2 (1 − x) − 1 ⁄ 2 dx .
⌡0
Now put x = sin2 θ so that dx = 2 sin θ cos θ dθ .
1
Also when x = 0 , θ = 0 and when x = 1 , θ = π.
2
π ⁄2
1 ⌠ 1 1
∴ [Γ ( )]2 = ⎮ ⋅ ⋅ 2 sin θ cos θ dθ
2 ⌡0 sin θ cos θ
π ⁄2 π ⁄2
⌠
= 2⎮ dθ = 2 ⎡⎢⎣ θ ⎤⎥⎦
⌡0 0
1
= 2 [ π − 0] = π .
2
Taking square root of both the sides, we get
1
Γ ( ) = √π . (Remember)
2
∞
⌠ 2 √π
Important Deduction : To prove that ⎮ e− x dx = ⋅
⌡0 2
∞
⌠ 2
Proof : Let I = ⎮ e− x dx .
⌡0
Put x2 = z so that 2x dx = dz
1 1 1
or dx = dz = dz = z− 1 ⁄ 2 dz .
2x 2 √z 2
Also when x = 0 , z = 0 and when x → ∞ , z → ∞ .
∞ ∞
⌠ 1 1⌠ 1 ⎛ 1⎞ √π
∴ I = ⎮ e− z z− 1 ⁄ 2 dz = ⎮ e− z z1 ⁄ 2 − 1 dz = Γ ⎜⎝ ⎟⎠ = ⋅
⌡0 2 2 ⌡0 2 2 2
I-56 INTEGRAL CALCULUS
∞
⌠ 2 √π
H ence ⎮ e− x dx = ⋅ (Remember)
⌡0 2
⌠π ⁄ 2
3.9 Value of ⎮ cos m θ sin n θ dθ in terms of Γ,
⌡0
where m > − 1 , n > − 1
⎛ m + 1 ⎞ ⎛⎜ n + 1 ⎞⎟
Γ⎜ ⎟ Γ
π ⁄2 ⎝ 2 ⎠ ⎜⎜ 2 ⎟⎟
⌠ ⎝ ⎠
⎮ cos m θ sin n θ dθ = ⋅
⌡0 ⎛ m + n + 2⎞
2Γ ⎜ ⎟
⎝ 2 ⎠
Proof : Put sin2 θ = x so that 2 sin θ cos θ dθ = dx
or 2 sin θ. √(1 − sin2 θ) dθ = dx or 2 x1 ⁄ 2. √(1 − x) dθ = dx .
dx
∴ dθ = ⋅
2 x1 ⁄ 2 (1 − x) 1 ⁄ 2
1
Also when θ = 0 , x = 0 and when θ = π,x= 1.
2
π ⁄2 π ⁄2
⌠ ⌠
∴ ⎮ cosm θ sinn θ dθ = ⎮ (1 − sin2 θ) m ⁄ 2. sin n θ dθ
⌡0 ⌡0
1
⌠ dx
= ⎮ (1 − x) m ⁄ 2. x n ⁄ 2. 1 ⁄ 2
⌡0 2 x (1 − x) 1 ⁄ 2
1
1 ⌠ (n − 1) ⁄ 2
= ⎮ x (1 − x) (m − 1) ⁄ 2 dx
2 ⌡0
1
1 ⌠ {(n + 1) ⁄ 2} − 1
= ⎮ x (1 − x) {(m + 1) ⁄ 2} − 1 dx
2 ⌡0
1 ⎛ m + 1 , n + 1⎞ ,
= B⎜ ⎟ provided m > − 1 and n > − 1
2 ⎝ 2 2 ⎠
1 1
Γ (m + 1) Γ (n + 1) Γ (m) Γ (n) ⎤
1 2 2 ⎡ ..
= ⎢ . B (m, n) = ⎥
2 Γ 1 (m + 1 + n + 1) ⎣ Γ (m + n) ⎦
2
1 1
Γ (m + 1) Γ (n + 1)
2 2
= 1
⋅
2 Γ (m + n + 2)
2
1
⌠
(iii) We know that ⎮ x m − 1 (1 − x) n − 1 dx = B (m, n) .
⌡0
Putting x = sin2 θ , so that dx = 2 sin θ cos θ dθ , we have
I-58 INTEGRAL CALCULUS
1 π ⁄2
⌠ ⌠
⎮ x m − 1 (1 − x) n− 1 dx = 2 ⎮ sin2 m − 1 θ cos2 n − 1 θ dθ .
⌡0 ⌡0
π ⁄2
⌠ B (m , n) Γ (m) Γ (n)
∴ ⎮ sin2 m − 1 θ cos2 n − 1 θ dθ = = ⋅
⌡0 2 2 Γ (m + n)
This result may also be written in the form
⎛ p + 1⎞ ⎛q+ 1 ⎞⎟
Γ⎜ ⎟ ⋅ Γ ⎜⎜ ⎟⎟
π ⁄2 ⎝ 2 ⎠ ⎜⎝ 2
⌠ ⎠ ,
⎮ sin p θ cos q θ dθ =
⌡0 ⎛ p + q + 2⎞
2Γ ⎜ ⎟
⎝ 2 ⎠
by putting 2m − 1 = p and 2n − 1 = q .
1
⌠
(iv) We know that ⎮ y m − 1 (1 − y) n − 1 dy = B (m, n) .
⌡0
x− b , dx ,
Putting y = so that dy = we have
a− b a− b
1 a
⌠ ⌠ ⎛ x − b⎞m − 1⎛ a − x⎞ n − 1 dx
⎮ y m − 1. (1 − y) n − 1 dy = ⎮ ⎜ ⎟ ⎜ ⎟ ⋅
⌡0 ⌡b ⎝ a − b ⎠ ⎝ a − b⎠ a− b
a
1 ⌠
= ⎮ (x − b) m − 1 (a − x) n − 1 dx .
(a − b) m + n − 1 ⌡b
a 1
⌠ ⌠
∴ ⎮ (x − b) m − 1 (a − x) n − 1 dx = (a − b) m + n − 1 ⎮ y m − 1 (1 − y) n − 1 dy
⌡b ⌡0
a
⌠
or ⎮ (x − b) m − 1 (a − x) n − 1 dx = (a − b) m + n − 1 B ( m, n)
⌡b
Γ (m) Γ (n)
= (a − b) m + n − 1 ⋅
Γ (m + n)
π π ⁄2
1 ⌠ 1 ⌠
= ⎮ sin2m − 1 φ dφ = 2m − 1 ⋅ 2 ⎮ sin2m − 1 φ dφ [ Note]
22m − 1 ⌡0 2 ⌡0
π ⁄2
1 ⌠
= ⎮ sin2m − 1 φ .cos0 φ dφ [ Note]
2 2m − 2 ⌡0
1 1 1
1 Γ (2m − 1 + 1) Γ (0 + 1) 1 Γ (m) Γ ( )
2 2 2
= = ⋅
22m − 2 1
2 Γ (2m − 1 + 0 + 2) 22m − 1 Γ (m + 1
)
2 2
1 Γ (m) √π
= ⋅ ⋅ ...(2)
22m − 1 Γ (m + 1 )
2
[... Γ ( ) = √π]
1
2
Now equating the two values of B (m, m) obtained in (1) and (2), we get
[Γ (m)]2 1 Γ (m) .√π
= 2m − 1 ⋅
Γ (2m) 2 Γ (m + )
1
2
1 √π
or Γ (m) Γ (m + ) = Γ (2 m) . (Remember)
2 22 m − 1
⎛ 1⎞ ⎛ 2⎞ ⎛ 3⎞ ⎛n − 1⎞ ,
3.12 Value of Γ ⎜ ⎟ Γ⎜ ⎟ Γ⎜ ⎟ …Γ⎜ ⎟
⎝ n⎠ ⎝ n⎠ ⎝ n⎠ ⎝ n ⎠
where n is a Positive Integer
⎛ 1⎞ ⎛ 2⎞ ⎛ 3⎞ ⎛ n − 1⎞
Let A = Γ⎜ ⎟ Γ⎜ ⎟ Γ⎜ ⎟ …Γ⎜ ⎟⋅ ...(1)
⎝ n⎠ ⎝ n⎠ ⎝ n⎠ ⎝ n ⎠
Writing the above expression in the reverse order, we have
I-60 INTEGRAL CALCULUS
⎛ 1⎞ ⎛ 2⎞ ⎛ n − 2⎞ ⎛ n − 1⎞
A = Γ ⎜1 − ⎟ Γ ⎜1 − ⎟ … Γ ⎜1 − ⎟ Γ ⎜1 − ⎟⋅ ...(2)
⎝ n⎠ ⎝ n⎠ ⎝ n ⎠ ⎝ n ⎠
Multiplying (1) and (2), we get
⎛ 1⎞ ⎛ 1⎞ ⎛ 2⎞ ⎛ 2⎞ ⎛ n − 1⎞ ⎛ n − 1⎞
A 2 = Γ ⎜ ⎟ Γ ⎜ 1 − ⎟ Γ ⎜ ⎟ Γ ⎜ 1 − ⎟ …… Γ ⎜ ⎟ Γ ⎜1 − ⎟
⎝ n⎠ ⎝ n⎠ ⎝ n⎠ ⎝ n⎠ ⎝ n ⎠ ⎝ n ⎠
πn − 1
= . [See corollary of article 3.7] ...(3)
π 2π n− 1
sin sin … sin π
n n n
To calculate this expression, we factorize 1 − x2n .
Now the roots of the equation x2n − 1 = 0 are given by
x = (1) 1 ⁄ 2n = (cos 2 r π + i sin 2 r π) 1 ⁄ 2n
rπ rπ ,
= cos + i sin where r = 0 , 1, 2 , … , 2n − 1 .
n n
H ence, we have 1 − x2n
⎛ π π⎞ ⎛ π π⎞
= (1 − x) (1 + x) ⎜ x − cos − i sin ⎟ ⎜ x − cos + i sin ⎟ …
⎝ n n⎠ ⎝ n n⎠
⎛ n− 1 n− 1 ⎞ ⎛ n− 1 n− 1 ⎞
… ⎜ x − cos π − i sin π ⎟ ⎜ x − cos π + i sin π⎟
⎝ n n ⎠ ⎝ n n ⎠
⎛ π ⎞ ⎛ 2π ⎞
= (1 − x2) ⎜ 1 − 2x cos + x2⎟ ⎜ 1 − 2x cos + x2⎟ …
⎝ n ⎠ ⎝ n ⎠
⎛ n− 1 ⎞
… ⎜ 1 − 2x cos π + x2⎟ ⋅
⎝ n ⎠
1− x2n ⎛ π ⎞ ⎛ 2π ⎞
∴ = ⎜ 1 − 2x cos + x2⎟ ⎜ 1 − 2x cos + x2⎟ …
1− x2 ⎝ n ⎠ ⎝ n ⎠
⎛ n− 1 ⎞
… ⎜ 1 − 2x cos π + x2⎟ ⋅
⎝ n ⎠
Putting x = 1 and x = − 1 respectively, we have in the limit,
π ⎛ 2π ⎞ ⎛ n− 1 ⎞
n = ⎛⎜ 2 − 2 cos ⎞⎟ ⎜ 2 − 2 cos ⎟ … ⎜ 2 − 2 cos π⎟
⎝ n⎠ ⎝ n⎠ ⎝ n ⎠
π ⎛ 2π ⎞ ⎛ n− 1 ⎞
and n = ⎛⎜ 2 + 2 cos ⎞⎟ ⎜ 2 + 2 cos ⎟ … ⎜ 2 + 2 cos π⎟ ⋅
⎝ n⎠ ⎝ n⎠ ⎝ n ⎠
Multiplying these, we get
π 2π n− 1
n2 = 22n − 2 sin2 ⋅ sin2 … sin2 π
n n n
π 2π n− 1
or n = 2n − 1 .sin ⋅ sin … sin π.
n n n
H ence, from (3), we get
πn − 1 (2π) n − 1 (2π) (n − 1) ⁄ 2
A2 = = or A = ⋅
n ⁄ 2n − 1 n n1 ⁄ 2
BETA AND GAMMA FUNCTIONS I-61
π 2π n− 1
Remark : The value of sin sin … sin π can also be found by using
n n n
the trigonometrical identity
sin nθ ⎛ π⎞ ⎛ 2π ⎞ ⎛ 3θ⎞
= 2n − 1 sin ⎜ θ + ⎟ sin ⎜ θ + ⎟ sin ⎜ θ + ⎟ …
sin θ ⎝ n⎠ ⎝ n⎠ ⎝ n⎠
⎛ n− 1 ⎞
… sin ⎜ θ + π⎟ ⋅
⎝ n ⎠
From the above identity, we have
sin nθ θ ⎛ π⎞ ⎛ 2π ⎞ ⎛ n− 1 ⎞
⋅ ⋅ n = 2n − 1 sin ⎜ θ + ⎟ sin ⎜ θ + ⎟ … sin ⎜ θ + π⎟ ⋅
nθ sin θ ⎝ n⎠ ⎝ n⎠ ⎝ n ⎠
Taking limit as θ → 0 , we get
π 2π 3π n− 1
n = 2n − 1 sin sin sin … sin π.
n n n n
⌠∞
3.13 Values of the Integrals ⎮ e− ax cos bx . x m − 1 dx
⌡0
⌠ ∞ − ax
and ⎮ e sin bx . x m − 1 dx
⌡0
We have
∞ ∞
⌠ ⌠ Γ (m)
⎮ e− ax e ibx x m − 1 dx = ⎮ e− (a − ib) x x m − 1 dx =
⌡0 ⌡0 (a − ib) m
[See article 3.6, part (i)]
= (a − ib) − m Γ (m) . ...(1)
Let us first separate (a − ib) − m into real and imaginary parts.
Put a = k cos α and b = k sin α so that
α = tan− 1 (b ⁄ a) and k = √(a2 + b2) .
Then (a − ib) − m = [k (cos α − i sin α)]− m = k − m (cos α − i sin α) − m
= k − m (cos mα + i sin mα) , by De-Moivre’s theorem.
Now from (1), we have
∞
⌠
⎮ e− ax e ibx x m − 1 dx = k − m (cos mα + i sin mα) Γ (m)
⌡0
∞
⌠ Γ (m)
or ⎮ e− ax (cos bx + i sin bx) x m − 1 dx = (cos mα + i sin mα) ,
⌡0 km
Γ5 Γ10
= 2 B (5, 10) = 2
Γ15
4.3.2.1 1
= 2⋅ = ⋅
14.13.12.11.10 5005
∞
⌠ x dx
(iii) Let I = ⎮ ⋅
⌡0 1 + x6
1
Put x6 = y or x = y1 ⁄ 6 , so that dx = y− 5 ⁄ 6 dy .
6
∞ ∞
1⌠ y1 ⁄ 6. y− 5 ⁄ 6 1⌠ y− 2 ⁄ 3
∴ I= ⎮ dy = ⎮ dy
6 ⌡0 1+ y 6 ⌡0 1+ y
∞
1⌠ y(1 ⁄ 3) − 1 1 ⎛ 1 2⎞
= ⎮ dy = B ⎜⎝ , ⎟⎠ , [By article 3.3]
6 ⌡0 (1 + y) (1 ⁄ 3) + (2 ⁄ 3) 6 3 3
1 2 1 1
Γ Γ Γ Γ (1 − ) π
1 3 3 1 3 3 1
= = = ⋅
1 2
6 Γ( + ) 6 Γ1 6 sin 1 π
3 3 3
⎡. . π ⎤
⎢ . Γn Γ (1 − n) = ⎥
⎣ sin n π ⎦
1 π 1 2π π
= ⋅ = ⋅ = ⋅
6 (√3 ⁄ 2) 6 √3 3 √3
1
⌠ dx √π Γ (1 ⁄ n)
Example 2 : Show that ⎮ = ⋅
⌡0 (1 − x n ) 1 ⁄ 2 n Γ (1 ⁄ n + 1 ⁄ 2) (Lucknow 2010)
Solution : Let xn = sin2 θ i.e., x = sin2 ⁄ n
θ so that
2
dx = sin(2 ⁄ n) − 1 θ cos θ dθ .
n
1 π ⁄2
⌠ dx 2⌠ sin(2 ⁄ n) − 1 θ cos θ dθ
Then ⎮ = ⎮
⌡0 √(1 − x n ) n ⌡0 cos θ
π ⁄2
2 ⌠
= ⎮ sin(2 ⁄ n) − 1 θ cos0 θ dθ
n ⌡0
1
Γ (1 ⁄ n) Γ ( )
2 2
= ⋅
n 2 Γ (1 ⁄ n + 1 ⁄ 2)
√π Γ (1 ⁄ n)
= ⋅ ⋅
n Γ (1 ⁄ n + 1 ⁄ 2)
1
⌠ xm − 1 + xn − 1
Example 3 : Evaluate ⎮ dx . (Kumaun 2002)
⌡0 (1 + x) m + n
Solution : We have
1 1 1
⌠ xm − 1 + xn − 1 ⌠ x m − 1 dx ⌠ x n − 1 dx
⎮ dx = ⎮ + ⎮ ⋅ ...(1)
⌡0 (1 + x) m + n ⌡0 (1 + x) m + n ⌡0 (1 + x) m + n
I-64 INTEGRAL CALCULUS
π ⁄2
⌠ π nπ ,
Example 4 : Show that ⎮ (tan x) n dx = sec where − 1 < n < 1 .
⌡0 2 2
Solution : We have
π ⁄2 π ⁄2 π ⁄2
⌠ ⌠ sin n x ⌠
⎮ tan n x dx = ⎮ dx = ⎮ sin n x cos− n x dx
⌡0 ⌡0 n
cos x ⌡0
1 1
Γ (n + 1) . Γ (− n + 1)
2 2 ,
= 1
2 Γ (n − n + 2)
2
where − n + 1 > 0 i.e., n < 1 and n + 1 > 0 i.e., n > − 1
1 1 1 1 1 1
= Γ (n + 1) Γ (1 − n) = Γ (n + 1) Γ [1 − (n + 1)]
2 2 2 2 2 2
1 π π ,
= , [ ... Γ (n) Γ (1 − n) =
2 sin 1 (n + 1) π sin n π
2
by corollary to article 3.7]
π 1
= ⋅
2 sin ( 1 π + 1 n π)
2 2
π 1
= ⋅
2 cos ( 1 n π)
2
π nπ ,
= sec where − 1 < n < 1 .
2 2
BETA AND GAMMA FUNCTIONS I-65
∞ ∞
⌠ n ⌠ xm n
(ii) Let I = ⎮ x m e− ax dx = ⎮ e− a x x n − 1 dx [ Note]
⌡0 ⌡0 xn − 1
∞
⌠ n
= ⎮ x m − n + 1 e− a x x n − 1 dx .
⌡0
Put ax n = t so that na x n − 1 dx = dt . Also when x = 0 , t = 0 and when
x →∞ , t →∞ .
∞
⌠ ⎛ t ⎞ (m − n + 1) ⁄ n − t 1 ⎡ .. ⎛ t ⎞1⁄n ⎤
∴ I= ⎮ ⎜ ⎟ e ⋅ dt , ⎢ . a xn = t ⇒x = ⎜ ⎟ ⎥
⌡0 ⎝ a ⎠ na ⎣ ⎝ a⎠ ⎦
∞
1 ⌠
= ⎮ t {(m + 1) ⁄ n} − 1 e− t dt
na . a(m − n + 1) ⁄ n ⌡0
1
= Γ {(m + 1) ⁄ n}, by the definition of G amma function.
na(m + 1) ⁄ n
1 1 1
⌠ dx ⌠ dx ⌠ ⎛ 1⎞ − 1 ⁄ 2
(iii) Let I= ⎮ = ⎮ = ⎮ ⎜ log ⎟ dx .
⌡0 √(− log x) ⌡0 √{log (1 ⁄ x)} ⌡0 ⎝ x⎠
Put log (1 ⁄ x) = y i.e., 1 ⁄ x = e y i.e., x = e− y so that dx = − e− y dy .
Also when x → ∞ , y → ∞ and when x = 1 , y = 0 .
⌠0
∴ I = − ⎮ y− 1 ⁄ 2 e− y dy
⌡∞
∞
⌠ 1
= ⎮ e− y y1 ⁄ 2 − 1 dy = Γ ( ) , by the def. of G amma function
⌡0 2
= √π .
I-66 INTEGRAL CALCULUS
⎛ 1⎞ ⎛ 2⎞ ⎛ 3⎞ ⎛ 8⎞
Example 7 : Find the value of Γ ⎜ ⎟ Γ ⎜ ⎟ Γ ⎜ ⎟ … Γ ⎜ ⎟ ⋅
⎝ 9⎠ ⎝ 9⎠ ⎝ 9⎠ ⎝ 9⎠
⎛ 1⎞ ⎛ 2⎞ ⎛ 3⎞ ⎛ n − 1⎞ (2 π) (n − 1) ⁄ 2 ,
Solution : We know that Γ ⎜ ⎟ Γ ⎜ ⎟ Γ ⎜ ⎟ … Γ ⎜ ⎟ =
⎝ n⎠ ⎝ n⎠ ⎝ n⎠ ⎝ n ⎠ n1 ⁄ 2
where n is a positive integer.
BETA AND GAMMA FUNCTIONS I-67
⎛1 ⎞ ⎛ 1 − 2 x⎞ ⎛ 1 + 2 x⎞
= ⎜ − x2⎟ Γ ⎜ ⎟ Γ⎜ ⎟
⎝4 ⎠ ⎝ 2 ⎠ ⎝ 2 ⎠
⎛1 ⎞ ⎛ 1 − 2 x⎞ ⎛ 1 − 2 x⎞
= ⎜ − x2⎟ Γ ⎜ ⎟ Γ ⎜1 − ⎟
⎝4 ⎠ ⎝ 2 ⎠ ⎝ 2 ⎠
⎛1 ⎞ π ⎛1 ⎞ π
= ⎜ − x2⎟ = ⎜ − x2⎟ ⋅
⎝4 ⎠ ⎛1 − 2x ⎞ ⎝4 ⎠ ⎛1 ⎞
sin ⎜ π⎟ sin ⎜ π − xπ ⎟
⎝ 2 ⎠ ⎝2 ⎠
⎛1 ⎞ π ⎛1 ⎞
= ⎜ − x2⎟ ⋅ = ⎜ − x2⎟ ⋅ π sec π x .
⎝4 ⎠ cos x π ⎝4 ⎠
Γ (n) ,
Similarly, I2 = provided b and n are + ive.
2bn
Γm Γn
H ence I= ⋅
4am bn
Example 10 : Show that the sum of the series
1 1 m (m + 1) 1 m (m + 1) (m + 2) 1
+ m + ⋅ + ⋅ + …
n+ 1 n+ 2 2! n+ 3 3! n+ 4
Γ (n + 1) Γ (1 − m) ,
is where − 1 < n < 1 .
Γ (n − m + 2)
Γ (n + 1) Γ (1 − m)
Solution : We have = B (n + 1, 1 − m)
Γ (n − m + 2)
1
⌠
= ⎮ x n (1 − x) − m dx
⌡0
1
⌠ ⎡ m (m + 1) 2 m (m + 1) (m + 2) 3 ⎤
= ⎮ xn ⎢1 + mx + x + x + …⎥ dx
⌡0 ⎣ 2! 3! ⎦
1 ⎤
⌠ ⎡ m (m + 1) n + 2 m (m + 1) (m + 2) n + 3
= ⎮ ⎢ xn + m xn + 1 + x + x + …⎥ dx
⌡0 ⎣ 2! 3! ⎦
⎡ xn + 1 x n + 2 m (m + 1) x n + 3 m (m + 1) (m + 2) x n + 4 ⎤1
= ⎢⎢⎢ + m + + + …⎥
⎣n+ 1 n+ 2 2! n+ 3 3! n+ 4 ⎦0
1 1 m (m + 1) 1 m (m + 1) (m + 2) 1
= + m. + ⋅ + + …
n+ 1 n+ 2 2! n+ 3 3! n+ 4
∞
⌠ sin bz π
Example 11 : Prove that ⎮ dz = ⋅
⌡0 z 2
Solution : We have
∞ ∞
⌠ ⌠
I = ⎮ ⎮ e− x z sin bz dx dz
⌡0 ⌡0
∞ ⎡ − x z⎤ ∞
⌠ e ⎢⎢ ⎥
= ⎮ ⎢⎣ − z ⎥⎥⎦ sin bz dz , on first integrating w.r.t. x
⌡0 0
∞
⌠ sin bz
= ⎮ dz . ...(1)
⌡0 z
Again on first integrating w.r.t. z , we have
∞ ∞
⌠ ⌠
I = ⎮ ⎮ e− x z sin bz dx dz
⌡0 ⌡0
∞ ⎡ ∞ ⎤
⌠ ⌠ ⎢⎮
= ⎮ ⎢⌡ e− x z sin bz dz⎥⎥ dx
⌡0 ⎢ 0 ⎥
⎣ ⎦
∞
⌠ b
= ⎮ dx , [See article 3.13, Deduction (ii)]
⌡0 b2 + x2
BETA AND GAMMA FUNCTIONS I-69
⎡ x⎤ ∞
= ⎢ tan− 1 ⎥
⎣ b⎦ 0
π
=
⋅ ...(2)
2
H ence equating the two values (1) and (2) of I , we have
∞
⌠ sin bz π
⎮ dz = ⋅
⌡0 z 2
Example 12 : Show that
∞
⌠ 1 nπ
⎮ cos (bz1 ⁄ n ) dz = n Γ (n + 1) . cos ⋅
⌡0 b 2 (Kanpur 2014; Agra 14)
Solution : Put z1 ⁄ n = x i.e., z = x n , so that dz = nx n − 1 dx .
∞ ∞
⌠ ⌠
∴ ⎮ cos(bz1 ⁄ n ) dz = ⎮ cos (bx) . nx n − 1 dx
⌡0 ⌡0
∞
⌠
= n ⎮ x n − 1 cos (bx) dx
⌡0
∞
⌠
= real part of n ⎮ e− ibx x n − 1 dx
⌡0
Γ (n)
= real part of n
(ib) n
n Γ (n) 1 1
= real part of ⋅ (cos π + i sin π) − n
bn 2 2
Γ (n + 1) ⎛ nπ n π⎞
= real part of ⎜ cos − i sin ⎟
bn ⎝ 2 2 ⎠
1 ⎛ n π⎞
= ⋅ Γ (n + 1) .cos ⎜ ⎟⋅
bn ⎝ 2 ⎠
∞ xc Γ (c + 1) ,
Example 13 : Show that ⌠
⎮
⌡ dx = c > 1.
0 cx (log c) c + 1
(Gorakhpur 2005; Kanpur 07; Lucknow 10; Rohilkhand 13)
Solution : We have
∞ ∞ ∞ ∞
⌠ xc ⌠ xc xc
I= ⎮
⌡ x dx = ⎮
⌡ x dx = ⌠⎮
⌡ dx = ⌠
⎮
⌡ e− x log c x c dx .
0 c 0 e log c 0 e x log c 0
Put x log c = y so that (log c) dx = dy .
When x = 0, we have y = 0 and when x → ∞ , y → ∞ because c > 1 ⇒ log c > 0 .
∞ y ⎞ c dy 1 ∞
∴ I= ⌠⎮ − y⎛ ⌠
⎮ e− y y(c + 1) − 1 dy
⌡ e ⎜⎝ log c ⎟⎠ log c = ⌡
0 ( log c) c + 1 0
1
= Γ (c + 1) ,
( log c) c + 1
provided c + 1 > 0 which is so because c > 1 .
I-70 INTEGRAL CALCULUS
1. Prove that
a
⌠ dx 1 π
(i) ⎮ = ⋅ ⋅ (Kanpur 2010)
⌡0 (an − x n ) 1 ⁄ n n sin (π ⁄ n)
2
⌠ 2π
(ii) ⎮ (8 − x3) − 1 ⁄ 3 dx = ⋅ (Kumaun 2008)
⌡0 3 √3
1
⌠ x m− 1 (1 − x) n− 1 Γ (m) Γ (n)
2. Show that ⎮ dx = n ⋅
⌡0 (a + x) m+ n a (1 + a) m Γ (m + n)
x (1 + a)
Hint. Put = y.
a+ x
π ⁄2
⌠ 1 ⎛⎜⎜ p + 1 , q + 1 ⎞⎟⎟ ,
3. Show that ⎮ sin p θ cos q θ dθ = B p > − 1, q > − 1.
⌡0 2 ⎜⎝ 2 2 ⎟⎠
2
⌠ 4 16 ⎛⎜ 5 , 2 ⎞⎟
Deduce that ⎮ x (8 − x3) − 1 ⁄ 3 dx = B ⋅
⌡0 3 ⎝ 3 3⎠
4. Prove that B (m, n) = B (m + 1, n) + B (m, n + 1) for m > 0, n > 0.
(Kanpur 2005, 11; Gorakhpur 05; Bundelkhand 11; Avadh 06, 11, 14)
5. Prove that
∞
⌠ Γ (n)
(i) ⎮ e− a x x n− 1 dx = ⋅
⌡0 an
1 n− 1
⌠ ⎛ 1⎞
(ii) ⎮ ⎜⎝ log ⎟⎠ dx = Γ (n).
⌡0 x
6. Show that, if m > − 1, then
∞ ⎛ m + 1⎞
⌠ 2 2 1
⎮ x m e− n x dx = Γ⎜ ⎟⋅
⌡0 2 nm+ 1 ⎝ 2 ⎠
7. Prove that
1
⌠ 1 ⎛m + 1 , ⎞
(i) ⎮ x m (1 − x n ) p dx = B ⎜⎝ p + 1⎟⎠ ⋅
⌡0 n n
1
⌠ 1 [Γ (1 ⁄ n)]2
(ii) ⎮ (1 − x n ) 1 ⁄ n dx = ⋅
⌡0 n 2 Γ (2 ⁄ n)
(2 π) 9 ⁄ 2
8. Show that Γ (0⋅ 1) Γ (0⋅ 2) Γ (0⋅ 3) … Γ (0⋅ 9) = ⋅
√(10)
BETA AND GAMMA FUNCTIONS I-71
9. Show that
π ⁄2 π ⁄2
⌠ ⌠ dθ
(i) ⎮ √(sin θ) dθ × ⎮ = π.
⌡0 ⌡0 √(sin θ)
1 1
⌠ x2 dx ⌠ dx π
(ii) ⎮ × ⎮ = ⋅
⌡0 (1 − x4) 1 ⁄ 2 ⌡0 (1 + x4) 1 ⁄ 2 4 √2
π ⁄2 ∞
⎛ 1⎞ ⎛ 3⎞ ⌠ ⌠ x2 dx
10. Show that Γ ⎜⎝ ⎟⎠ Γ ⎜⎝ ⎟⎠ = 2 ⎮ √(tan θ) dθ = 4 ⎮ = π √2 .
4 4 ⌡0 ⌡0 1 + x4
11. Show that the perimeter of a loop of the curve r n = an cos nθ is
a [Γ (1 ⁄ 2 n)]2
⋅ 2(1 ⁄ n)− 1 ⋅ ⋅
n Γ (1 ⁄ n)
⌠
1
dx √2 ⎡ ⎛ 1⎞ ⎤ 2
12. Prove that ⎮ = ⎢Γ ⎜ ⎟⎥ ⋅
⌡0 √(1 − x4) 8 √π ⎣ ⎝ 4⎠ ⎦
π ⁄2
⌠ 1 ⎛⎜ 1 , p + 1 ⎞⎟
13. Show that ⎮ sin p θ dθ = B ⋅
⌡0 2 ⎝2 2 ⎠
14. Show that
∞
⌠ (α2 − β2)
(i) ⎮ xe− α x cos βx dx =
⌡0 (α2 + β2) 2
∞
⌠ 2 αβ
(ii) ⎮ xe− α x sin βx dx = ⋅
⌡0 (α2 + β2) 2
∞⎛1 ⎞
⌠
15. Prove that ⎮ cos ⎜⎝ π x2⎟⎠ dx = 1.
⌡− ∞ 2
⎛ 1, 1⎞ πm − 1
16. Show that B (m, m) . B ⎜⎝ m + m + ⎟⎠ = 4m− 1 ⋅ (Rohilkhand 2005)
2 2 2
B (m + 1, n) m
4. = ⋅
B (m, n) ……
∞
⌠ x m− 1
5. For m > 0, n > 0, ⎮ dx = …… .
⌡0 (1 + x) m+ n
∞
⌠ x m− 1 − x n− 1
6. For m > 0, n > 0, ⎮ dx = …… .
⌡0 (1 + x) m+ n
7. For n > 0, Γ (n + 1) = …… Γ (n).
8. If n is a positive integer, then Γ (n) = …… .
9. If 0 < n < 1, then Γ (n) Γ (1 − n) = …… .
⎛
1 ⎞⎟
10. Γ ⎜⎝ = …… .
2⎠
11. If m > − 1, n > − 1, then
⎛ m + 1⎞ ⎛ n + 1⎞
π ⁄2 Γ ⎜⎝ ⎟ Γ⎜ ⎟
⌠ 2 ⎠ ⎝ 2 ⎠
⎮ cos m θ sin n θ dθ =
⌡0 ……
∞
⌠ 2
12. ⎮ e− x dx = …… .
⌡0
⎛ 1 ⎞⎟ √π
13. For m > 0, Γ (m) Γ ⎜⎝ m + ⎠
= Γ (2 m).
2 ……
∞
⌠ Γ (n)
14. For a > 0, n > 0, ⎮ e− a x x n− 1 dx = ⋅
⌡0 ……
1 2
15. The value of Γ ( )⋅ Γ ( ) is …… . (Agra 2002)
3 3
∞
⌠
17. The value of the integral ⎮ e− x x− 1 ⁄ 2 dx is
⌡0
√π π
(a) (b)
2 2
(c) √π (d) π (Kumaun 2008)
BETA AND GAMMA FUNCTIONS I-73
True or False:
Write ‘T ’ for true and ‘F ’ for false statem ent.
∞ ⎛ 1⎞
⌠
24. ⎮ e− x x1 ⁄ 2 dx = Γ ⎜⎝ ⎟⎠ ⋅
⌡0 2
⎛
1 ⎞⎟ ⎛
3 ⎞⎟
25. Γ ⎜⎝ Γ ⎜⎝ = π √2 .
4⎠ 4⎠
∞
⌠ x dx π
26. ⎮ = ⋅
⌡0 1 + x6 3 √3
∞
⌠ x8 (1 − x6)
27. ⎮ dx = 1.
⌡0 (1 + x) 24
1
⌠
28. For m > 0, n > 0, B (m, n) = ⎮ x m (1 − x) n dx.
⌡0
∞
⌠ x n− 1
29. For m > 0, n > 0, B (m, n) = ⎮ dx.
⌡0 (1 + x) m+ n
∞ ∞
⌠ x m− 1 ⌠ x n− 1
30. For m > 0, n > 0, ⎮ dx = ⎮ dx.
⌡0 (1 + x) m+ n ⌡0 (1 + x) m+ n
31. Γ (6) = 120.
Γ (m) Γ (n)
32. For m > 0, n > 0, B (m, n) = ⋅
2 Γ (m + n)
∞
⌠ 2 √π
33. ⎮ e− x dx = ⋅
⌡0 4
34. B (m + 1, n) + B (m, n + 1) = B (m + 1, n + 1). (Agra 2003)
II. The double integral of the algebraic sum of a fixed number of functions is
equal to the algebraic sum of the double integrals taken for each term. Thus
∫∫A [ f1 ( x, y ) + f2 ( x, y ) + f3 ( x, y ) + …] dx dy
= ∫ ab ⎡⎢⎣ ∫ cd ⎤
f ( x, y ) dy⎥ dx ,
⎦
...(1)
or ∫∫A f ( x, y ) dx dy = ∫ cd∫ ab f ( x, y ) dy dx
= ∫ cd ⎡⎢⎣ ∫ ab ⎤
f ( x, y ) dx⎥ dy
⎦
...(2)
i.e., in this case the order of integration is immaterial, provided the limits of
integration are changed accordingly.
d
Important Note : In formula (1) the definite integral ∫ c f ( x, y ) dy is calculated
first. During this integration x is regarded as a constant. While in the formula (2)
the definite integral ∫ ab f ( x, y ) dx is calculated first and during this integration y is
regarded as a constant.
(b) If the region A is bounded by the curves
y = f1 (x) , y = f2 (x) , x = a and x = b , then
MULTIPLE INTEGRALS I-77
b f (x)
⌠⌠ ⌠ ⌠2
⎮⎮ f ( x, y ) dx dy = ⎮ ⎮ f ( x, y ) dx dy
⌡⌡A ⌡a ⌡f (x)
1
b ⎡ f2 (x) ⎤
⌠ ⌠⎢⎮
= ⎮ ⎢⌡ f ( x, y ) dy⎥⎥ dx ,
⌡a ⎢ f (x) ⎥
⎣ 1 ⎦
where the integration with respect to y is performed first treating x as a constant.
Similarly, if the region A is bounded by the curves
x = f1 ( y) , x = f2 ( y) , y = c , y = d , we have
d f ( y)
⌠⌠ ⌠ ⌠2
⎮⎮ f ( x, y ) dx dy = ⎮ ⎮ f ( x, y ) dy dx
⌡⌡A ⌡c ⌡f ( y)
1
d f ( y)
⌠ ⎡⌠ 2 ⎤
= ⎮ ⎢⎢ ⎮ f ( x, y ) dx⎥⎥ dy ,
⌡c ⎢ ⌡f ( y) ⎥
⎣ 1 ⎦
where the integration with respect to x is performed first treating y as a constant.
Remember : While evaluating double integrals, first integrate w.r.t. the
variable having variable limits (treating the other variable as constant) and then
integrate w.r.t. the variable with constant limits.
b
Remark : In the double integral ∫ a ∫ cd f ( x, y ) dx dy , it is generally understood
that the limits of integration c to d are those of y and the limits of integration a to
b are those of x . H owever this is not a standard convention. Some authors regard
these limits in the reverse order i.e. they regard the limits c to d as those of x and
the limits a to b as those of y . So it is better to write this double integral as
∫ xb= a ∫ yd= c f ( x, y ) dx dy so that there in no confusion about the limits. H owever in
b f (x)
⌠ ⌠2
the double integral ⎮ ⎮ F ( x, y ) dx dy , there is no confusion about the limits.
⌡a ⌡f (x)
1
O bviously the variable limits are those of y because they are in terms of x and so the
constant limits must be those of x . H ere the first integration must be performed with
respect to y regarding x as constant.
a a
⌠ ⎡ b3 ⎤ ⎡ x3 b3 ⎤ ba3 b3a 1
= ⎮ ⎢ bx2 + ⎥ dx = ⎢ b + x⎥ = + = ab (a2 + b2) .
⌡0 ⎣ 3⎦ ⎣ 3 3 ⎦0 3 3 3
2 x 2 x
⌠ ⌠ dx dy ⌠ ⎡⌠ dy ⎤⎥
(ii) We have ⎮ ⎮ 2 = ⎮ ⎢⎢ ⎮ 2 dx
⌡1 ⌡0 x + y2 ⌡1 ⎢ ⌡0 x + y2 ⎥⎥
⎣ ⎦
2
⌠ ⎡1 y⎤ x
= ⎮ ⎢ tan− 1 ⎥ dx (integrating w.r.t. y treating x as constant)
⌡1 ⎣ x x⎦ y = 0
2 2
⌠ ⎡1 ⎤ π ⌠ dx π ⎡ 2
= ⎮ ⎢ (tan− 1 1 − tan− 1 0) ⎥ dx = ⎮ = ⎢ log x ⎤⎥
⌡1 ⎣ x ⎦ 4 ⌡1 x 4⎣ ⎦1
1 1
= π [log 2 − log 1] = π log 2 .
4 4
Example 2 : Evaluate
3 2 1 √(1 + x2)
⌠ ⌠ ⌠ ⌠ dx dy
(i) ⎮ ⎮ xy (1 + x + y ) dx dy . (ii) ⎮ ⎮ ⋅
⌡0 ⌡1 ⌡0 ⌡0 1 + x2 + y2
(Gorakhpur 2005; Kanpur 12; Avadh 14)
3 2
⌠ ⌠
Solution : (i) We have ⎮ ⎮ xy (1 + x + y) dx dy
⌡0 ⌡1
3 y2 y2 y3 ⎤ 2
⌠ ⎡
= ⎮ ⎢x ⋅ + x2 ⋅ + x⋅ ⎥ dx ,
⌡0 ⎣ 2 2 3 ⎦y= 1
(integrating w.r.t. y treating x as constant)
3
⌠ ⎡x x2 x ⎤
= ⎮ ⎢ (4 − 1) + (4 − 1) + (8 − 1) ⎥ dx
⌡0 ⎣ 2 2 3 ⎦
3 ⎡ 23 x2 3
⌠ ⎡ ⎛ 3 7⎞ 3 ⎤ 3 x3 ⎤
= ⎮ ⎢ ⎜ + ⎟ x + x2⎥ dx = ⎢ ⋅ + ⋅ ⎥
⌡0 ⎣ ⎝ 2 3 ⎠ 2 ⎦ ⎣ 6 2 2 3 ⎦0
23 9 27 123 3
= ⋅ + = = 30 ⋅
6 2 2 4 4
1 √(1 + x2)
⌠ ⌠ dx dy
(ii) We have ⎮ ⎮
⌡0 ⌡0 1 + x2 + y2
1 2
⌠ 1 ⎡ − 1 y ⎤ √(1 + x )
= ⎮ ⎢ tan ⎥ dx ,
⌡0 √(1 + x2) ⎣ √(1 + x2) ⎦ y = 0
(integrating w.r.t. y treating x as constant)
1 1
⌠
= ⎮
1 ⎡ tan− 1 1 − tan− 1 0⎤ dx = π ⌠
⎮
dx
⌡0 √(1 + x2) ⎣ ⎦ 4 ⌡0 √(1 + x2)
π 1 π
= ⎡⎣ log {x + √(1 + x2)}⎤⎦ 0 = log (1 + √2) .
4 4
a √(a2 − y2)
⌠ ⌠
Example 3 : Evaluate ⎮ ⎮ √(a2 − x2 − y2) dy dx .
⌡0 ⌡0
MULTIPLE INTEGRALS I-79
Solution : H ere the variable limits are those of x and so the first integration
must be performed w.r.t. x taking y as constant.
a √(a2 − y2) a √(a2 − y2)
⌠ ⌠ ⌠ ⎡ ⌠ ⎤
∴⎮ ⎮ √(a2 − x2 − y2) dy dx = ⎮ ⎢
⎢ ⎮ √{(a2 − y2) − x2} dx⎥⎥ dy
⌡0 ⌡0 ⌡0 ⎢ ⌡0 ⎥
⎣ ⎦
2 2
a
⌠ ⎡ x √(a2 − y2 − x2) (a2 − y2) x ⎤ √(a − y )
= ⎮ ⎢⎢ + sin− 1 ⎥
⎥ dy,
⌡0 ⎢⎣ 2 2 √(a2 − y2) ⎥⎦ x = 0
(integrating w.r.t. x treating y as constant)
a a
⌠ ⎡ a2 − y2 π ⎤ π⎡ y3 ⎤ π⎡ a3 ⎤ 1
= ⎮ ⎢0 + ⋅ ⎥ dy = ⎢ a2y − ⎥ = ⎢ a3 − ⎥ = π a 3.
⌡0 ⎣ 2 2⎦ 4⎣ 3 ⎦0 4 ⎣ 3⎦ 6
2 y⁄ 2 2 x⁄2
⌠ ⌠ ⌠ ⌠
Example 4 : Show that ⎮ ⎮ y dy dx = ⎮ ⎮ x dx dy .
⌡1 ⌡0 ⌡1 ⌡0
Solution : We have
2 y⁄ 2 2 y⁄ 2 2
⌠ ⌠ ⌠ ⎡ ⌠ ⎤ ⌠ y⁄ 2
⎮ ⎮ y dy dx = ⎮ ⎢⎢ y ⎮ dx⎥⎥ dy = ⎮ y ⎡⎢ x ⎤⎥ dy ,
⌡1 ⌡0 ⌡1 ⎢ ⌡0 ⎥ ⌡1 ⎣ ⎦ 0
⎣ ⎦
(integrating w.r.t. x treating y as a constant)
2 2 2
⌠ ⎡y ⎤ 1⌠ 2 1 ⎡ y3 ⎤ 1 7
= ⎮ y ⎢ − 0⎥ dy = ⎮ y dy = ⎢ ⎥ = [8 − 1] = ...(1)
⌡1 ⎣ 2 ⎦ 2 ⌡1 2 ⎣ 3 ⎦1 6 6
2 x⁄2 ⎡⌠ 2 x⁄2 ⎤ 2 x⁄2
⌠ ⌠ ⌠ ⌠
Again ⎮ ⎮ x dx dy = ⎮ x ⎢⎢ ⎮ dy⎥⎥ dx = ⎮ x ⎡⎢⎣ y ⎤⎥⎦ dx ,
⌡1 ⌡0 ⌡1 ⎢ ⌡0 ⎥ ⌡1 0
⎣ ⎦
(integrating w.r.t. y treating x as a constant)
2 2 2
⌠ ⎡x ⎤ 1⌠ 1 ⎡ x3 ⎤ 1 7
= ⎮ x ⎢ − 0⎥ dx = ⎮ x2 dx = ⎢ ⎥ = (8 − 1) = ...(2)
⌡1 ⎣ 2 ⎦ 2 ⌡1 2 ⎣ 3 ⎦1 6 6
From (1) and (2), we see that
2 y⁄ 2 2 x⁄2
⌠ ⌠ ⌠ ⌠
⎮ ⎮ y dy dx = ⎮ ⎮ x dx dy .
⌡1 ⌡0 ⌡1 ⌡0
⌠⌠
Example 7 : Evaluate ⎮⎮ ( x2 + y2) dx dy over the region in the positive quadrant
⌡⌡
for which x + y ≤ 1 . (Rohilkhand 2012; Avadh 14)
MULTIPLE INTEGRALS I-81
⌠⌠
Example 8 : Evaluate ⎮⎮ xy ( x + y ) dx dy over the area between y = x2 and
⌡⌡
y = x. (Gorakhpur 2005, 06)
Solution : Draw the given curves y = x2 and y = x in the same figure. The two
curves intersect at the points whose abscissae are given by x2 = x or x ( x − 1) = 0
i.e., x = 0 or 1. When 0 < x < 1 , we have x > x2 . So the area of integration can be
considered as lying between the curves y = x2, y = x , x = 0 and x = 1 .
Therefore the required integral
1 x 1 x
⌠ ⌠ ⌠ ⎡⌠ ⎤
= ⎮ ⎮ xy ( x + y ) dx dy = ⎮ ⎢⎢ ⎮ ( x2 y + xy2) dy⎥⎥ dx
⌡x = 0 ⌡y = x2 ⌡0 ⎢ ⌡x2 ⎥
⎣ ⎦
1 x 1
⌠ ⎡ x2 y2 xy3 ⎤ ⌠ ⎡ ⎛ x4 x4 ⎞ ⎛ x6 x7 ⎞ ⎤
= ⎮ ⎢ + ⎥ dx = ⎮ ⎢ ⎜ + ⎟ − ⎜ + ⎟ ⎥ dx
⌡0 ⎣ 2 3 ⎦ x2 ⌡0 ⎣ ⎝ 2 3⎠ ⎝ 2 3⎠⎦
1 1
⌠ ⎡ 5x4 x6 x7 ⎤ ⎡ x5 x7 x8 ⎤
= ⎮ ⎢ − − ⎥ dx = ⎢ − − ⎥
⌡0 ⎣ 6 2 3⎦ ⎣6 14 24⎦ 0
1 1 1 28 − 12 − 7 9 3
= − − = = = ⋅
6 14 24 168 168 56
Example 9 : Prove by the m ethod of double integration that the area lying between
16 2
the parabolas y2 = 4ax and x2 = 4ay is a .
3
Solution : Draw the two parabolas in the same figure. The two parabolas
intersect at the points whose abscissae are given by (x2 ⁄ 4a) 2 = 4ax i.e.,
x (x3 − 64a3) = 0 i.e., x = 0 and x3 = 64a3 . Thus the two parabolas intersect at the
points where x = 0 and x = 4a .
Now the area of a small element situated at any point ( x, y ) = dx dy .
∴ the required area
4a √(4 ax) 4a
⌠ ⌠ ⌠ ⎡ y ⎤ √(4 ax) dx
= ⎮ ⎮ dx dy = ⎮ ⎢ ⎥ 2
⌡x = 0 ⌡y = x2 ⁄ 4 a ⌡0 ⎣ ⎦ x ⁄4a
I-82 INTEGRAL CALCULUS
4a 4a
⌠ ⎡ 1 2⎤ ⎡ 2 1 x3 ⎤
= ⎮ ⎢ 2 √a. x1 ⁄ 2 − . x ⎥ dx = ⎢ 2 √a . x3 ⁄ 2 . − ⋅ ⎥
⌡0 ⎣ 4a ⎦ ⎣ 3 4a 3 ⎦ 0
4 1 32 2 16 2 16 2
= √a. (4a) 3 ⁄ 2 − . 64a3 = a − a = a .
3 12a 3 3 3
a √(a2− x2)
⌠ ⌠
4. (i) E valuate the double integral ⎮ ⎮ x2 y dx dy.
⌡0 ⌡0
Mention the region of integration involved in this double integral.
⌠⌠
(ii) E valuate ⎮ ⎮ x2 y3 dx dy over the circle x2 + y2 = a2.
⌡⌡ (Rohilkhand 2013B)
⌠⌠
5. E valuate ⎮ ⎮ (x + y + a) dx dy over the circular area x2 + y2 ≤ a2.
⌡⌡
⌠⌠
6. E valuate ⎮ ⎮ x2 y2 dx dy over the region bounded by x = 0, y = 0 and x2 + y2 = 1.
⌡⌡
(Avadh 2012)
⌠⌠
7. E valuate ⎮ ⎮ xy dx dy over the region in the positive quadrant for which x + y ≤ 1.
⌡⌡
⌠⌠
8. E valuate ⎮ ⎮ e 2 x+ 3y dx dy over the triangle bounded by x = 0, y = 0 and x + y = 1.
⌡⌡
⌠⌠ xy
9. E valuate ⎮ ⎮ dx dy over the positive quadrant of the circle x2 + y2 = 1.
⌡ ⌡ √(1 − y2)
10. Find the area of the ellipse x2 ⁄ a2 + y2 ⁄ b2 = 1, by double integration.
⌠⌠
11. Compute the value of ⎮⎮ y dx dy, where R is the region in the first quadrant
⌡⌡R
bounded by the ellipse x2 ⁄ a2 + y2 ⁄ b2 = 1.
12. Find the mass of a plate in the form of a quadrant of an ellipse
x2 ⁄ a2 + y2 ⁄ b2 = 1 whose density per unit area is given by ρ = k x y.
1
1. (i) π log (1 + √2). (ii) (log b) (log a). (iii) − 2 .
4
1 1
(iv) ⋅ (v) 7. (vi) π.
2 2
1 4
2. (i) π 2. (ii) ⋅ (iii) 3 ⁄ 35.
4 3
(iv) 1 − log (3 ⁄ 2). (v) πa4 ⁄ 8. (vi) 2 a3 ⁄ 3.
1 1
3. (ii) and − ⋅
2 2
4. (i) 5
a ⁄ 15. The area of the circle x2 + y2 = a2 in the positive quadrant.
(ii) 0. 5. πa3. 6. π ⁄ 96.
1 1 1
7. ⋅ 8. (e − 1) 2 (2 e + 1). 9. ⋅
24 6 6
Let a function f ( r, θ) of
the polar coordinates ( r, θ) be
continuous inside some region
A and on its boundary. Let the
region A be bounded by the
curves r = f1 (θ) , r = f2 (θ) and
the lines θ = θ1 , θ = θ2 .
Divide the area A into
elements by a series of
concentric circular arcs with
centre at origin and successive
radii differing by equal amounts
and a series of straight lines
drawn through the origin at
equal intervals of angles. Let δr
be the distance between two
consecutive circles and δθ be the angle between two consecutive lines. There is thus
a network of elementary areas (say n in number) of which a typical one is PQRS . If
P is the point (r, θ) , the area of the element PQRS situated at the point P is
1 1 2 1
(r + δr) 2 δθ − r δθ = r δθ δr , by neglecting the term (δr) 2 δθ being an
2 2 2
infinitesimal of higher order.
Now by the definition of the double integral of f (r, θ) over the region A , we
have
n
⌠⌠ lim n→∞ ∑ f (r , θ ) r δθ δr ,
⎮⎮ f (r, θ) d A = δr→0 , δθ→0 k k k
⌡⌡A k= 1
where rk δθδr is the area of the element situated at the point (rk , θk) .
U sing the area of integration, this double integral is generally written as
θ f (θ) θ f (θ)
⌠ 2⌠ 2 ⌠ 2 ⌠2
⎮ ⎮ f (r, θ) dθ dr , or ⎮ dθ ⎮ f (r, θ) dr .
⌡θ ⌡f (θ) ⌡θ ⌡f (θ)
1 1 1 1
Remark : The area of the typical element PQRS situated at the point P (r, θ)
can also be found as below :
We have OP = r , OQ = r + δr so that PQ = δr . Also PS is the arc of a circle of
radius r subtending an angle δθ at the centre of the circle and so arc PS = r δθ .
Therefore the area of the element PQRS is δr. r δθ i.e., r δθ δr .
MULTIPLE INTEGRALS I-85
π a (1 + cos θ)
⌠ ⌠
Example 1 : Evaluate ⎮ ⎮ r2 cos θ dθ dr .
⌡0 ⌡0
Solution : We have
π a (1 + cos θ) π
⎡ r3 ⎤ a (1 + cos θ)
⌠ ⌠ ⌠
⎮ ⎮ r2 cos θ dθ dr = ⎮ cos θ ⎢ ⎥ dθ
⌡0 ⌡0 ⌡0 ⎣ 3⎦0
π
1⌠
= ⎮ cos θ . a3 (1 + cos θ) 3 dθ
3 ⌡0
π
a3 ⌠
= ⎮ cos θ (1 + 3 cos θ + 3 cos2 θ + cos3 θ) dθ
3 ⌡0
π
a3 ⌠
= ⎮ [cos θ + 3 cos2 θ + 3 cos3 θ + cos4 θ] dθ
3 ⌡0
π ⁄2 ⎡ . . ⌠π
a3 ⌠ ⎢ . ⎮ cos n θ dθ = 0
= 2. ⎮ [3 cos2 θ + cos4 θ] dθ , ⎢
3 ⌡0 ⎢
⎣
⌡0
π ⁄2 ⎤
⌠
or 2 ⎮ cos n θ dθ according as n is odd or even⎥⎥
⌡0 ⎥
⎦
2a3 ⎡ 1 π 3.1 π ⎤ 2a3 3π ⎡ 1⎤
= ⎢3 ⋅ ⋅ + ⋅ ⎥= ⋅ ⎢1 + ⎥
3 ⎣ 2 2 4.2 2 ⎦ 3 4 ⎣ 4⎦
2a 3 3π 5 5πa 3
= ⋅ ⋅ = ⋅
3 4 4 8
⌠⌠ r dθ dr
Example 2 : Evaluate ⎮⎮ over one loop of the lem niscate
⌡⌡ √(a2 + r2)
r2 = a2 cos 2θ .
Solution : In the equation of the lemniscate r2 = a2 cos 2θ , putting r = 0 , we
get cos 2θ = 0 i.e., 2θ = ± π ⁄ 2 i.e., θ = ± π ⁄ 4 . Therefore for one loop of the given
lemniscate θ varies from − π ⁄ 4 to π ⁄ 4 and r varies from 0 to a √ (cos 2θ) .
Therefore the required integral
π ⁄4 a √(cos 2θ)
⌠ ⌠ r dθ dr
= ⎮ ⎮
⌡θ = − π ⁄ 4 ⌡ r = 0 √(a2 + r2)
π ⁄4 a √(cos 2θ)
⌠ ⌠ 1
= ⎮ ⎮ (a2 + r2) − 1 ⁄ 2 (2 r) dθ dr
⌡− π ⁄ 4 ⌡0 2
π ⁄4
⌠ ⎡⎢ (a 2 + r2) 1 ⁄ 2⎤⎥ a √(cos 2θ) dθ
= ⎮
⌡− π ⁄ 4 ⎣ ⎦0
π ⁄4
⌠
= ⎮ [a (1 + cos 2θ) 1 ⁄ 2 − a] dθ
⌡− π ⁄ 4
I-86 INTEGRAL CALCULUS
π ⁄4 π ⁄4
⌠ ⌠
= 2a ⎮ [(2 cos2 θ) 1 ⁄ 2 − 1] dθ = 2 a ⎮ (√2 cos θ − 1) dθ
⌡0 ⌡0
⎡ 1 π⎤ π a
= 2a ⎡⎣ √2 sin θ − θ⎤⎦ π ⁄ 4 = 2a ⎢ √2 . − ⎥ = 2a ⎡⎢ 1 − ⎤⎥ = (4 − π) .
0 ⎣ √2 4 ⎦ ⎣ 4⎦ 2
Example 3 : Find by double integration the area lying inside the circle
r = a sin θ and outside the cardioid r = a (1 − cos θ) .
Solution : The given circle is r = a sin θ and the cardioid is r = a (1 − cos θ) .
Note that the given circle passes through the pole and the diameter through the pole
makes an angle π ⁄ 2 with the initial line.
E liminating r between the two equations, we have
a sin θ = a (1 − cos θ)
1 1
sin θ 2 sin θ cos θ θ
2 2
or 1= = = tan
1 − cos θ 1
2 cos2 θ 2
2
1 1
or θ= π i.e., θ = π ⁄ 2 .
2 4
Thus the two curves meet at the point where
θ = π ⁄ 2 . Also for both the curves r = 0 when
θ = 0 and so the two curves also meet at the pole
O where θ = 0 . To cover the required area the limits
of integration for r are a (1 − cos θ) to a sin θ and
for θ are 0 to π ⁄ 2 . Therefore the required area
π ⁄2 a sin θ
⌠ ⌠
= ⎮ ⎮ r dθ dr
⌡0 ⌡a (1 − cos θ)
π ⁄ 2 ⎡ 2 ⎤ a sin θ π ⁄2
⌠ r 1⌠
= ⎮ ⎢ ⎥ dθ = ⎮ [a2 sin2 θ − a2 (1 − cos θ) 2] dθ
⌡0 ⎣ 2 ⎦ a (1 − cos θ) 2 ⌡0
π ⁄2
a2 ⌠
= ⎮ [sin2 θ − 1 + 2 cos θ − cos2 θ] dθ
2 ⌡0
a2 ⎡ 1 π π 1 π⎤ a2 ⎡ π⎤ a2
= ⎢ ⋅ − + 2.1 − ⋅ ⎥ = ⎢2 − ⎥ = (4 − π) .
2 ⎣2 2 2 2 2⎦ 2 ⎣ 2⎦ 4
2 √(2 x − x2)
⌠ ⌠ x dx dy
Example 4 : Transform the integral ⎮ ⎮ by changing to
⌡0 ⌡0 √( x2 + y2)
polar coordinates and hence evaluate it. (Kumaun 2008)
Solution : From the limits of
integration it is obvious that the region of
integration is bounded by y= 0,
2
y = √(2x − x ) and x = 0 , x = 2 i.e., the region
of integration is the area of the circle
x2 + y2 − 2x = 0 between the lines x = 0 ,
x = 2 and lying above the axis of x i.e., the
line y = 0 .
MULTIPLE INTEGRALS I-87
π ⁄2 π ⁄2
⌠ 1 ⌠ 2 4
= ⎮ cos θ .4 cos2 θ dθ = 2 ⎮ cos3 θ dθ = 2 . = ⋅
⌡0 2 ⌡0 3 3
π a sin θ
⌠ ⌠
1. (i) E valuate ⎮ ⎮ r dθ dr.
⌡0 ⌡0 (Kashi 2013)
π ⁄2 a cos θ
⌠ ⌠
(ii) E valuate ⎮ ⎮ r sin θ dθ dr.
⌡0 ⌡0
π a (1 + cos θ)
⌠ ⌠
(iii) E valuate ⎮ ⎮ r 3 sin θ cos θ dθ dr. (Agra 2003)
⌡0 ⌡0
⌠⌠
2. E valuate ⎮⎮ r 2 dθ dr over the area of the circle r = a cos θ. (Kanpur 2010)
⌡⌡
3. Integrate r sin θ over the area of the cardioid r = a (1 + cos θ), lying above the
initial line. (Kanpur 2010)
1 √(2 x− x2)
⌠ ⌠
(ii) ⎮ ⎮ (x2 + y2) dx dy.
⌡0 ⌡x
a √(a2 − x2)
⌠ ⌠
(iii) ⎮ ⎮ y2 √(x2 + y2) dx dy.
⌡0 ⌡0
1 a2 16 4
1. (i) πa2. (ii) ⋅ (iii) a .
4 6 15
4a3 4a3 π ka4
2. ⋅ ⋅ 3. 4. ⋅
9 3 16
1 2 9π + 16
5. a (π + 8). 6. ⋅
4 12
π ⁄2 a π ⁄2 2 cos θ
⌠ ⌠ πa4 ⌠ ⌠ ⎛ 3π ⎞
7. (i) ⎮ ⎮ (a2 − r2) r dθ dr ; ⋅ (ii) ⎮ ⎮ r 3 dθ dr ; ⎜⎝ ⎟ − 1.
⌡0 ⌡0 8 ⌡π ⁄ 4 ⌡0 8⎠
π ⁄2 a
⌠ ⌠ 4 2 πa5
(iii) ⎮ ⎮ r sin θ dθ dr ; ⋅
⌡0 ⌡0 20
= ∫ ab dx ∫
c
d
dy ∫
e
f
f ( x, y, z ) dz .
MULTIPLE INTEGRALS I-89
H ere the order of integration is immaterial and the integration with repsect to
any of x, y and z can be performed first.
(b) If the limits of z are given as functions of x and y , the limits of y as functions
of x while x takes the constant values say from x = a to x = b , then
b y (x) z ( x, y )
⌠⌠⌠ ⌠ ⌠ 2 ⌠ 2
⎮⎮⎮ f ( x, y, z ) dx dy dz = ⎮ dx ⎮ dy ⎮ f ( x, y, z ) dz .
⌡⌡⌡V ⌡a ⌡y (x) ⌡z ( x, y )
1 1
3 2
Example 1 : Evaluate ∫ y = 0 ∫ x = 0 ∫ z1= 0 ( x + y + z ) dz dx dy .
Solution : The given integral
= ∫ y3= 0 ∫ x2= 0 ⎧⎨⎩∫ 01 ⎫
( x + y + z ) dz⎬ dx dy
⎭
3 2 1 3 ⎧⌠ 2
⌠ ⌠ ⎧⎪ z2 ⎫⎪ ⌠ ⎫
⎪ ⎮ ( x + y + 1) dx⎪ dy
= ⎮ ⎮ ⎨⎪ xz + yz + ⎬⎪ dx dy = ⎮ ⎨⌡ ⎬
⌡y = 0 ⌡x = 0 ⎩ 2 ⎭0 ⌡0 ⎪ 0 2 ⎪
⎩ ⎭
3 2 3 3
⌠ ⎧ x2 x⎫ ⌠ ⎡ 2y2 ⎤
= ⎮ ⎪⎨⎪ + xy + ⎪⎬⎪ dy = ⎮ (3 + 2y ) dy = ⎢ 3y + ⎥ = 18 .
⌡0 ⎩ 2 2 ⎭0 ⌡0 ⎣ 2 ⎦0
Example 4 : Find the volume of the tetrahedron bounded by the coordinate planes
and the plane x + y + z = 1 . (Rohilkhand 2013B)
Solution : H ere the region of integration V to cover the volume of the
tetrahedron can be expressed as 0 ≤ x ≤ 1 , 0 ≤ y ≤ 1 − x, 0 ≤ z ≤ 1 − x − y .
Therefore the required volume of the tetrahedron
1 1− x 1− x− y
⌠⌠⌠ ⌠ ⌠ ⌠
= ⎮⎮⎮ dx dy dz = ⎮ ⎮ ⎮ dx dy dz (Note)
⌡⌡⌡V ⌡0 ⌡0 ⌡0
1 1− x 1 1− x
⌠ ⌠ ⎡ z ⎤ 1 − x − y dx dy = ⌠ ⌠
= ⎮ ⎮ ⎢ ⎥ ⎮ ⎮ (1 − x − y ) dx dy
⌡0 ⌡0 ⎣ ⎦0 ⌡0 ⌡0
1 1− x 1
⌠ ⎡ y2 ⎤ ⌠ ⎡ (1 − x ) 2 ⎤
= ⎮ ⎢ (1 − x ) y − ⎥ dx = ⎮ ⎢ (1 − x ) 2 − ⎥ dx
⌡0 ⎣ 2 ⎦0 ⌡0 ⎣ 2 ⎦
1 1
⌠ 1 1 ⎡ (1 − x ) 3 ⎤
= ⎮ (1 − x ) 2 dx = ⎢ ⎥
⌡0 2 2 ⎣ 3 .(− 1) ⎦ 0
1 1
= − [0 − 1] = ⋅
6 6
⌠⌠⌠
Example 5 : Evaluate ⎮⎮⎮ ( x + y + z ) dx dy dz over the tetrahedron x = 0 ,
⌡⌡⌡
y = 0 , z = 0 and x + y + z = 1 .
Solution : The region of integration V for the given tetrahedron can be
expressed as
0 ≤ x ≤ 1 , 0 ≤ y ≤ 1 − x, 0 ≤ z ≤ 1 − x − y.
⌠⌠⌠
H ence the required triple integral = ⎮⎮⎮ ( x + y + z ) dx dy dz
⌡⌡⌡V
1 1− x 1− x− y
⌠ ⌠ ⌠
= ⎮ ⎮ ⎮ ( x + y + z ) dx dy dz
⌡0 ⌡0 ⌡0
1 1− x ⎡ 1− x− y
⌠ ⌠ z2 ⎤
= ⎮ ⎮ ⎢ ( x + y) z + ⎥ dx dy
⌡0 ⌡0 ⎣ 2 ⎦0
1 1− x ⎡
⌠ ⌠ (1 − x − y ) 2 ⎤
= ⎮ ⎮ ⎢ ( x + y ) (1 − x − y ) + ⎥ dx dy
⌡0 ⌡0 ⎣ 2 ⎦
1 1− x
⌠ ⌠ ⎛ 1 − x − y⎞
= ⎮ ⎮ (1 − x − y ) ⎜ x + y + ⎟ dx dy
⌡0 ⌡0 ⎝ 2 ⎠
1 1− x
⌠ ⌠ 1
= ⎮ ⎮ (1 − x − y ) (1 + x + y ) dx dy
⌡0 ⌡0 2
I-92 INTEGRAL CALCULUS
1 1− x 1 1− x
1⌠ ⌠ 1⌠ ⎡ ( x + y )3⎤
= ⎮ ⎮ [1 − ( x + y ) 2] dx dy = ⎮ ⎢y − ⎥ dx
2 ⌡0 ⌡0 2 ⌡0 ⎣ 3 ⎦0
(Note)
1 1
1⌠ ⎛ 1 x3 ⎞ 1 ⌠ ⎛2 x3 ⎞
= ⎮ ⎜ 1 − x − + ⎟ dx = ⎮ ⎜ − x + ⎟ dx
⌡
2 0 ⎝ 3 3⎠ ⌡
2 0 ⎝3 3⎠
1
1 ⎡2 x2 x4 ⎤ 1 ⎡2 1 1⎤ 1 1 1
= ⎢ x− + ⎥ = ⎢ − + ⎥ = ⋅ = ⋅
2 ⎣3 2 3 × 4 ⎦ 0 2 ⎣ 3 2 12⎦ 2 4 8
⌠⌠⌠
Example 6 : Evaluate ⎮⎮⎮ z2 dx dy dz over the sphere x2 + y2 + z 2 = 1 .
⌡⌡⌡
Solution : H ere the region of integration can be expressed as
− 1 ≤ x ≤ 1 , − √(1 − x2) ≤ y ≤ √(1 − x2) , − √(1 − x2 − y2) ≤ z ≤ √(1 − x2 − y2) .
∴ the required triple integral
1 √(1 − x2) √(1 − x2 − y2)
⌠ ⌠ ⌠
= ⎮ ⎮ ⎮ z2 dx dy dz
⌡− 1 ⌡− √(1 − x2) ⌡− √(1 − x2 − y2)
1 √(1 − x2) 2 2
⎡ z3 ⎤ √(1 − x − y )
⌠ ⌠
= ⎮ ⎮ ⎢ ⎥ dx dy
⌡− 1 ⌡− √(1 − x2) ⎣ 3 ⎦ − √(1 − x2 − y2)
1 √(1 − x2)
1 ⌠ ⎡⌠ ⎤
= ⎮ ⎢⎢ ⎮ 2 (1 − x2 − y2) 3 ⁄ 2 dy⎥⎥ dx
3 ⌡− 1 ⎢⎣ ⌡− √(1 − x2) ⎥
⎦
2 1 ⎡ π ⁄2 ⎤
= ∫ ⎢∫
3 − 1 ⎣ − π ⁄2
[(1 − x2) cos2 θ]3 ⁄ 2. √(1 − x2). cos θ dθ⎥ dx
⎦
[putting y = √(1 − x2) sin θ so that dy = √(1 − x2) cos θ dθ ;
also when y = 0 , θ = 0 and when y = √(1 − x2) , θ = π ⁄ 2]
2 1 ⎡
= ∫ ⎢ 2 .∫ 0π ⁄ 2 (1 − x2) 2 cos4 θ dθ⎤⎥⎦ dx
3 − 1 ⎣
4 1 3.1 π π 1
= ∫ (1 − x2) 2⋅ ⋅ dx = ∫ (1 − x2) 2 dx
3 − 1 4.2 2 4 − 1
π 1 π⎡ 2 1 5⎤ 1
= . 2 ∫ (1 − 2x2 + x4) dx = ⎢ x − x3 + x⎥
4 0 2⎣ 3 5 ⎦0
π⎡ 2 1⎤ π 8 4π
= ⎢1 − + ⎥= ⋅ = ⋅
2⎣ 3 5⎦ 2 15 15
π ⁄2 a sin θ (a2− r 2) ⁄ a
⌠ ⌠ ⌠
(iv) ⎮ dθ ⎮ dr ⎮ r dz.
⌡0 ⌡0 ⌡0
a x x+ y
⌠ ⌠ ⌠
3. (i) ⎮ ⎮ ⎮ e x+ y+ z dx dy dz.
⌡0 ⌡0 ⌡0
a a− x a− x− y
⌠ ⌠ ⌠
(ii) ⎮ ⎮ ⎮ x2 dx dy dz.
⌡0 ⌡0 ⌡0
4. E valuate the triple integral of the function f (x, y, z) = x2 over the region V
enclosed by the planes x = 0, y = 0, z = 0 and x + y + z = a.
(Rohilkhand 2012; Avadh 12)
5. Find the volume of the tetrahedron bounded by the plane x ⁄ a + y ⁄ b + z ⁄ c = 1
and the coordinate planes.
⌠⌠⌠ dx dy dz
6. (i) E valuate ⎮⎮⎮ over the region x ≥ 0, y ≥ 0, z ≥ 0, x + y + z ≤ 1.
⌡⌡⌡ (x + y + z + 1) 3
(Avadh 2013)
⌠⌠⌠ x2 y2 z2
(ii) E valuate ⎮⎮⎮ xyz dx dy dz over the ellipsoid 2 + 2 + 2 = 1.
⌡⌡⌡ a b c (Kanpur 2011)
⌠⌠⌠
(iii) E valuate ⎮⎮⎮ (z5 + z) dx dy dz over the sphere x2 + y2 + z2 = 1.
⌡⌡⌡
⌠⌠⌠
(iv) E valuate ⎮⎮⎮ u2 v2 w du dv dw, where R is the region u2 + v2 ≤ 1, 0 ≤ w ≤ 1.
⌡⌡⌡R
1 ⎛⎜ 26 ⎞ 5a3 π 1
(iii) − log 3⎟⎠ ⋅ (iv) ⋅ 3. (i) (e 4 a − 6 e2 a + 8 e a − 3).
6⎝ 3 64 8
a5 a5 abc
(ii) ⋅ 4. ⋅ 5. ⋅
60 60 6
1 ⎛
5 ⎞⎟ π
6. (i) ⎜ log 2 − ⋅ (ii) 0. (iii) 0. (iv) ⋅
2 ⎝ 8⎠ 48
1 y
⌠ ⌠
1. ⎮ ⎮ f (x, y) dy dx.
⌡0 ⌡1− √(1− y)
2 4 − x2
⌠ ⌠
2. ⎮ ⎮ (x + y) dx dy.
⌡1 ⌡0
a sin α y cot α a √(a2− y2)
⌠ ⌠ ⌠ ⌠
3. ⎮ ⎮ f (x, y) dy dx + ⎮ ⎮ f (x, y) dy dx.
⌡0 ⌡0 ⌡a sin α ⌡0
am y⁄ m al a
⌠ ⌠ ⌠ ⌠
4. ⎮ ⎮ f (x, y) dy dx + ⎮ ⎮ f (x, y) dy dx.
⌡0 ⌡y ⁄ l ⌡am ⌡y ⁄ l
MULTIPLE INTEGRALS I-99
a √(4ay) 3a 3a− y
⌠ ⌠ ⌠ ⌠
5. ⎮ ⎮ f (x, y) dy dx + ⎮ ⎮ f (x, y) dy dx.
⌡0 ⌡0 ⌡a ⌡0
b ⁄ (a+ b) a 1 b (1− y) ⁄ y
⌠ ⌠ ⌠ ⌠
6. ⎮ ⎮ f (x, y) dy dx + ⎮ ⎮ f (x, y) dy dx.
⌡0 ⌡0 ⌡b ⁄ (a+ b) ⌡0
a y ∞ a2 ⁄ y
⌠ ⌠ ⌠ ⌠
7. ⎮ ⎮ f (x, y) dy dx + ⎮ ⎮ f (x, y) dy dx.
⌡0 ⌡0 ⌡a ⌡0
b √{1− (c2 ⁄ a2)} a b a
⌠ ⌠ ⌠ ⌠
8. ⎮ ⎮ f (x, y) dy dx + ⎮ ⎮ f (x, y) dy dx.
⌡0 ⌡a √{1− ( y2 ⁄ b2)} ⌡b √{(1− (c2 ⁄ a2)} ⌡c
a⁄4 √(ay)
⌠ ⌠
9. ⎮ ⎮ f (x, y) dy dx.
⌡0 ⌡1 [a − √(a2− 4ay)]
2
a a − √(a2− y2) a 2a 2a 2a
⌠ ⌠ ⌠ ⌠ ⌠ ⌠
10. ⎮ ⎮ f (x, y) dy dx + ⎮ ⎮ f (x, y) dy dx + ⎮ ⎮ f (x, y) dy dx.
⌡0 ⌡y2 ⁄ 2 a ⌡0 ⌡a + √(a2− y2) ⌡a ⌡y2 ⁄ 2 a
ab ⁄ √(a2+ b2) ab ⁄ √(a2+ b2) a (b ⁄ a) √(a2− x2)
⌠ ⌠ ⌠ ⌠
11. ⎮ ⎮ f (x, y) dx dy + ⎮ ⎮ f (x, y) dx dy.
⌡0 ⌡0 ⌡ab ⁄ √(a2+ b2) ⌡0
2a cos− 1
(r ⁄ 2 a)
⌠ ⌠
12. ⎮ ⎮ f (r, θ) dr dθ.
⌡0 ⌡0
a a
⌠ ⌠ φ ′ ( y) dy dx
13. ⎮ ⎮ = π [φ (a) − φ (0)].
⌡0 ⌡y √{(a − x) (x − y)}
To evaluate the integral put x = a sin2 θ + y cos2 θ.
⌠⌠
Example 1 : Transform ⎮⎮ f ( x, y ) dx dy by the substitution x + y = u , y = uv .
⌡⌡
Solution : We have x + y = u and y = uv . ...(1)
From these, we have
x = u − y = u − uv and y = uv . ...(2)
∂x ∂x ∂y ∂y
∴ = 1 − v, = − u, = v and = u.
∂u ∂v ∂u ∂v
⎪ ∂x ∂x ⎪
∂ ( x, y ) ⎪⎪ ∂u ∂v ⎪⎪ ⎪ 1 − v − u⎪
∴ J= = ⎪ ⎪= ⎪ ⎪ = u.
∂ (u, v ) ⎪ ∂y ∂y ⎪ ⎪ v u ⎪
⎪ ⎪
⎪ ∂u ∂v ⎪
∴ dx dy = J du dv = u du dv .
H ence the given integral transforms to
⌠⌠
⎮⎮ F (u, v ) u du dv .
⌡⌡
⌠⌠
Example 2 : Transform ⎮⎮ f ( x, y ) dx dy to polar coordinates.
⌡⌡
Solution : We have x = r cos θ , y = r sin θ .
⎪ ∂x ∂x ⎪
∂ ( x, y ) ⎪⎪ ∂r ∂θ⎪⎪ ⎪ cos θ − r sin θ⎪
Now J= = ⎪ ⎪= ⎪ ⎪ = r.
∂ (r, θ) ⎪ ∂y ∂y ⎪ ⎪ sin θ r cos θ ⎪
⎪ ⎪
⎪ ∂r ∂θ⎪
∴ dx dy = J dθ dr = r dθ dr .
⌠⌠
H ence the given integral transforms to ⎮⎮ F (r, θ) r dθ dr .
⌡⌡
⌠⌠
Example 3 : Evaluate ⎮⎮ √(a2 − x2 − y2) dx dy over the sem i-circle x2 + y2 = ax
⌡⌡
in the positive quadrant.
Solution : H ere the region of integration is a semi-circle. Therefore, for the
sake of convenience, changing to polar coordinates by putting x = r cos θ and
y = r sin θ in x2 + y2 = ax , we have
MULTIPLE INTEGRALS I-101
a a− x
⌠ ⌠
1. Transform ⎮ ⎮ f (x, y) dx dy, by the substitution x + y = u, y = uv.
⌡0 ⌡0
2. By using the transformation x + y = u, y = uv, show that
1 1− x
⌠ ⌠ 1
⎮ ⎮ e y ⁄ (x+ y) dx dy = (e − 1).
⌡0 ⌡0 2
3. By using the transformation x + y = u, y = uv, prove that
⌠⌠
⎮⎮ {xy (1 − x − y)}1 ⁄ 2 dx dy
⌡⌡
taken over the area of the triangle bounded by the lines
x = 0, y = 0, x + y = 1 is 2 π ⁄ 105.
⌠⌠
4. E valuate ⎮⎮ (x2 + y2) 7 ⁄ 2 dx dy over the circle x2 + y2 = 1.
⌡⌡
⌠⌠
5. E valuate ⎮⎮ xy (x2 + y2) 3 ⁄ 2 dx dy over the positive quadrant of the circle x2 + y2 = 1.
⌡⌡
⌠⌠ 2 2
6. E valuate ⎮⎮ e− (x + y ) dx dy over the circle x2 + y2 = a2.
⌡⌡
a 1
⌠ ⌠
1. ⎮ ⎮ F (u, v) u du dv. 4. 2 π ⁄ 9.
⌡0 ⌡0
2
5. 1 ⁄ 14 . 6. π (1 − e− a ).
I-102 INTEGRAL CALCULUS
3 2
⌠ ⌠
1. The value of the double integral ⎮ ⎮ dx dy is …… . (Agra 2002)
⌡0 ⌡1
1 1
⌠ ⌠
2. The value of the double integral ⎮ ⎮ xy dx dy is …… .
⌡0 ⌡0
1 x
⌠ ⌠
3. The value of the double integral ⎮ ⎮ xy dx dy is …… .
⌡0 ⌡0
π ⁄2 2 a cos θ
⌠ ⌠
4. The value of the double integral ⎮ ⎮ r dθ dr is …… .
⌡0 ⌡0
2 2 2
⌠ ⌠ ⌠
5. The value of the triple integral ⎮ ⎮ ⎮ xyz dx dy dz is …… .
⌡0 ⌡0 ⌡0
2 2 3
⌠ ⌠ ⌠
6. The value of the triple integral ⎮ ⎮ ⎮ dx dy dz is …… .
⌡1 ⌡1 ⌡1
a √(a2− x2)
⌠ ⌠
7. The value of the double integral ⎮ ⎮ dx dy is …… .
⌡− a ⌡0
1 1 1
⌠ ⌠ ⌠
9. The value of the triple integral ⎮ ⎮ ⎮ xyz dx dy dz is
⌡0 ⌡0 ⌡0
1 1
(a) (b)
2 8
1
(c) (d) 1.
4
MULTIPLE INTEGRALS I-103
a √(a2− y2)
⌠ ⌠
10. The value of the double integral ⎮ ⎮ dy dx is
⌡0 ⌡0
(a) πa2 (b) 2 πa2
πa2 πa2
(c) (d) ⋅
2 4
π ⁄2 π ⁄2 π ⁄2
⌠ ⌠ ⌠
11. The value of the triple integral ⎮ ⎮ ⎮ cos2 x cos2 y cos2 z dx dy dz is
⌡0 ⌡0 ⌡0
π2 π
(a) (b)
16 64
π3 π3
(c) (d) ⋅
8 64
1 1 1
⌠ ⌠ ⌠
12. The value of the triple integral ⎮ ⎮ ⎮ e x+ y+ z dx dy dz is
⌡0 ⌡0 ⌡0
e3
(a) e 3 (b)
4
(c) (e − 1) 3 (d) (e + 1) 3.
π ⁄2 π ⁄2 π ⁄2
⌠ ⌠ ⌠
13. The value of the triple integral ⎮ ⎮ ⎮ cos x cos y cos z dx dy dz is
⌡0 ⌡0 ⌡0
π
(a) 1 (b)
2
3π
(c) π (d) ⋅ (Rohilkhand 2005)
2
π ⁄2 sin θ
⌠ ⌠
14. The value of ⎮ ⎮ r dθ dr is equal to
⌡0 ⌡0
π ⁄2 sin θ
⌠ ⌠ π
(a) ⎮ sin θ dθ (b) ⎮ r dr
⌡0 ⌡0 2
π ⁄2
⌠ sin2 θ
(c) ⎮ dθ (d) none of these.
⌡0 2
(Garhwal 2003)
True or False:
Write ‘T ’ for true and ‘F ’ for false statem ent.
π ⁄2 a
⌠ ⌠ πa2
15. The value of the double integral ⎮ ⎮ r dθ dr is ⋅
⌡θ= − π ⁄ 2 ⌡r= 0 2
a √(a2− x2)
⌠ ⌠
16. The value of the double integral ⎮ ⎮ dx dy is πa2.
⌡− a ⌡− √(a2− x2)
a √(a2− x2)
⌠ ⌠
17. The value of the double integral ⎮ ⎮ x dx dy is 0.
⌡− a ⌡0
I-104 INTEGRAL CALCULUS
1 1
1. 3. 2. ⋅ 3. ⋅
4 8
πa2
4. ⋅ 5. 8. 6. 2.
2
πa2
7. ⋅ 8. (a). 9. (b).
2
10. (d). 11. (d). 12. (c).
13. (a). 14. (c). 15. T.
16. T. 17. T.
5.1 Dirichlet’s Theorem for Three Variables
If l, m, n are all positive, then the triple integral
⌠⌠⌠ l − 1 m − 1 n − 1 Γ (l) Γ (m) Γ (n) ,
⎮⎮⎮ x y z dx dy dz =
⌡⌡⌡ Γ (l + m + n + 1)
where the integral is extended to all positive values of the variables x, y and z subject to
the condition x + y + z ≤ 1 . (Agra 2001; Kanpur 14; Avadh 14; Kashi 14; Purvanchal 14)
1 ⎡ ym ⎤ 1− x
⌠
= ⎮ x l − 1 ⎢⎢⎢ ⎥⎥⎥ dx
⌡0 ⎣ m ⎦0
1
⌠ 1 l− 1
= ⎮ x (1 − x) m dx
⌡0 m
1
1 ⌠ 1
= ⎮ x l − 1 (1 − x) m + 1 − 1 dx = B (l, m + 1) ,
m ⌡0 m
by the def. of Beta function
1 Γ (l) Γ (m + 1) 1 Γ (l) . m Γ (m) ,
= = [... Γ (n + 1) = n Γ (n)]
m Γ (l + m + 1) m Γ (l + m + 1)
Γ (l) Γ (m)
= ⋅ (Remember) ...(1)
Γ (l + m + 1)
This is Dirichlet’s theorem for two variables.
⌠⌠
Now consider the double integral U2 = ⎮⎮ x l − 1 y m − 1 dx dy ,
⌡⌡
where the integral is extended to all positive values of the variables x and y subject
to the condition x + y ≤ h .
x y
We have x + y ≤ h ⇒ + ≤ 1 .
h h
So putting x ⁄ h = u and y ⁄ h = v so that dx = h du and dy = h dv , the integral
U2 becomes
⌠⌠
U2 = ⎮⎮ (hu) l − 1 (hv) m − 1 h2 du dv
⌡⌡
⌠⌠
= hl + m ⎮⎮ ul − 1 vm − 1 du dv , where u + v ≤ 1
⌡⌡
Γ (L ) Γ (m) ,
= hl+ m by (1). ...(2)
Γ (l + m + 1)
Now we consider the triple integral
⌠⌠⌠
I3 = ⎮⎮⎮ x l − 1 y m − 1 z n − 1 dx dy dz ,
⌡⌡⌡
subject to the condition x + y + z ≤ 1 i.e., y + z ≤ 1 − x and 0 ≤ x ≤ 1 .
We have
1 ⎡ ⌠⌠ ⎤
⌠ ⎢ ⎮⎮ y m − 1 z n − 1 dy dz⎥ x l − 1 dx , where y + z ≤ 1 − x
I3 = ⎮ ⎢ ⌡⌡ ⎥
⌡x= 0 ⎢ ⎥
⎣ ⎦
1 Γ (m) Γ (n)
⌠
= ⎮ (1 − x) m + n x l − 1 dx , by using (2)
⌡0 Γ (m + n + 1)
1
Γ (m) Γ (n) ⌠
= ⎮ x l − 1 (1 − x) m + n + 1 − 1 dx
Γ (m + n + 1) ⌡0
Γ (m) Γ (n)
= B (l, m + n + 1)
Γ (m + n + 1)
DIRICHLET’S AND LIOUVILLE’S INTEGRALS I-107
⌠ ⌠
1 1− x n 1 − x− y
⎡z ⎤
= ⎮ ⎮ x l − 1 y m − 1 ⎢⎢⎢ ⎥⎥⎥ dx dy
⌡0 ⌡0 ⎣ n ⎦0
1 1− x
1 ⌠ ⌠
= ⎮ ⎮ x l − 1 y m − 1 (1 − x − y) n dx dy
n ⌡0 ⌡0
1 ⎡⌠ 1− x ⎤
1 ⌠
= ⎮ x l − 1 ⎢⎢ ⎮ y m − 1 {(1 − x) − y }n dy⎥⎥ dx .
n ⌡0 ⎢ ⌡0 ⎥
⎣ ⎦
To integrate w.r.t. y , put y = (1 − x) t so that dy = (1 − x) dt ; also when
y = 0 , t = 0 and when y = 1 − x , t = 1 .
∴ the required integral
1 ⎡⌠ 1 ⎤
1 ⌠
I3 = ⎮ x l − 1 ⎢⎢ ⎮ (1 − x) m − 1 t m − 1 {(1 − x) n (1 − t) n } (1 − x) dt⎥⎥ dx
n ⌡0 ⎢ ⌡0 ⎥
⎣ ⎦
1 1
1 ⌠ ⌠
= ⎮ ⎮ x l − 1 (1 − x) m + n t m − 1 (1 − t) n dx dt
n ⌡0 ⌡0
1 1
1 ⌠ ⌠
= ⎮ x l − 1 (1 − x) m + n dx × ⎮ t m − 1 .(1 − t) n dt
n ⌡0 ⌡0
1
= B (l, m + n + 1) B (m, n + 1) ,
n
(by the definition of Beta function)
5.1,
DIRICHLET’S AND LIOUVILLE’S INTEGRALS I-109
⌠⌠ ⌠
= h l1 + l2 + … + ln ⎮⎮ … ⎮ u1 l1 − 1 u2 l2 − 1 … un ln − 1 du1 du2 … dun
⌡⌡ ⌡
subject to the condition u1 + u2 + … + un ≤ 1
Γ (l1) Γ (l2) … Γ (ln )
= h l1 + l2 + … + ln , ...(3)
Γ (1 + l1 + l2 + … + ln )
using the assumed result (2).
Now for n + 1 variables the condition is
x1 + x2 + … + xn + xn + 1 ≤ 1
i.e., x2 + x3 + … + xn + xn + 1 ≤ 1 − x1, and 0 ≤ x1 ≤ 1 .
We then have
⌠⌠ ⌠ l − 1 l − 1
⎮⎮ … ⎮ x1 1 x2 2 … xn ln − 1 xn + 1 ln+ 1 − 1 dx1 dx2 … dxn dxn + 1,
⌡⌡ ⌡
where x1 + x2 + … + xn + 1 ≤ 1
1 ⎡ ⌠⌠ ⎤
⌠ ⌠
= ⎮ x l1 − 1 ⎢⎢ ⎮⎮ … ⎮ x2 l2 − 1 … xn + 1 ln + 1 − 1 dx2 … dxn + ⎥ dx
1⎥⎥
⌡x = 0 1 ⎢ ⌡⌡ ⌡ 1
1 ⎣ ⎦
1 Γ (l2) Γ (l3) … Γ (ln + 1)
⌠
= ⎮ x1 l1 − 1 ⋅ ⋅
⌡x = 0 Γ (1 + l2 + l3 + … + ln + ln + 1)
1
(1 − x1) l2 + l3 + … + ln + 1 dx1,
using (3)
Γ (l2) Γ (l3) … Γ (ln + 1)
= ⋅
Γ (1 + l2 + … + ln + ln + 1)
1
⌠
⎮ x1 l1 − 1 (1 − x1) (1 + l2 + l3 + … + ln + 1)− 1 dx1
⌡0
Γ (l2) Γ (l3) … Γ (ln + 1) Γ (l1) Γ (1 + l2 + … + ln + 1)
= ⋅
Γ (1 + l2 + … + ln + ln + 1) Γ (1 + l1 + l2 + … + ln + ln + 1)
Γ (l1) Γ (l2) … Γ (ln + 1)
= ⋅ ...(4)
Γ (1 + l1 + l2 + … + ln + 1)
The result (4) shows that the theorem holds for (n + 1) variables if it holds for
n variables. But we have seen that the theorem is true for two variables. Hence by
mathematical induction the theorem is true for all values of n .
⌠⌠
Exam ple 1 : Evaluate ⎮⎮ x2 l − 1 y2 m − 1 dx dy for all positive values of x and y such
⌡⌡
that x2 + y2 ≤ c2.
Solution : Let us denote the given integral by I . Then we have to find the
value of I extended to all positive values of x and y subject to the condition
I-110 INTEGRAL CALCULUS
⎛ x ⎞ 2 ⎛ y⎞ 2
⎜ ⎟ + ⎜ ⎟ ≤ 1.
⎝ c⎠ ⎝ c⎠
1
Put (x ⁄ c) 2 = u i.e., x = cu1 ⁄ 2, so that dx = cu− 1 ⁄ 2 du ,
2
1 − 1⁄2
and ( y ⁄ c) 2 = v i.e., y = cv1 ⁄ 2, so that dy = cv dv .
2
Then the required integral
⌠⌠ 1 1
I = ⎮⎮ (cu1 ⁄ 2) 2 l − 1 (cv1 ⁄ 2) 2 m − 1 . cu− 1 ⁄ 2 ⋅ cv− 1 ⁄ 2 du dv
⌡⌡ 2 2
1 2l+ 2m ⌠⌠ l − 1 m − 1
= c ⎮⎮ u v du dv , where u, v take all + ive values
4 ⌡⌡
subject to the condition u + v ≤ 1
1 2l+ 2m Γ (l) Γ (m) ,
= c ⋅ by Dirichlet’s theorem.
4 Γ (l + m + 1)
⌠⌠ ⌠
Example 2 : Find the value of ⎮⎮ … ⎮ dx1 dx2 … dxn extended to all positive
⌡⌡ ⌡
values of the variables, subject to the condition x12 + x22 + … + xn 2 < R 2.
Solution : Let us denote the given integral by I . Then we have to find the
value of I extended to all positive values of x1, x2 , … , xn subject to the condition
x12 x22 xn 2
+ + … + < 1.
R2 R2 R2
1
Put (x1 ⁄ R) 2 = u1 i.e., x1 = Ru11 ⁄ 2, so that dx1 = Ru−1 1 ⁄ 2 du1,
2
1
(x2 ⁄ R) 2 = u2 i.e., x2 = Ru21 ⁄ 2, so that dx2 = Ru−2 1 ⁄ 2 du2 , and so on.
2
Then the required integral
⌠⌠ ⌠ ⎛ 1⎞ n n − 1 ⁄ 2 − 1 ⁄ 2
I = ⎮⎮ … ⎮ ⎜ ⎟ R u
⎝ 2⎠ 1 u2 … un − 1 ⁄ 2 du1 du2 … dun
⌡⌡ ⌡
⎛ R ⎞ n ⌠⌠ ⌠
= ⎜ ⎟ ⎮⎮ … ⎮ u1(1 ⁄ 2) − 1 u2(1 ⁄ 2) − 1 .... un (1 ⁄ 2) − 1 du1 du2 … dun ,
⎝ 2 ⎠ ⌡⌡ ⌡
subject to the condition u1 + u2 + … + un < 1
1
{Γ ( )}n
⎛ R⎞n 2 , by Dirichlet’s theorem
= ⎜ ⎟
⎝ 2 ⎠ Γ (1 + n ⋅ 1 )
2
⎛ R⎞n πn⁄2
[... Γ ( ) = √π]
1
= ⎜ ⎟ ⋅ 1
⋅
⎝ 2⎠ Γ (1 + n) 2
2
So the volume of the solid surrounded by this surface = 8 × the volume of the
portion of this solid lying in the positive octant.
Now the volume of a small element situated at any point (x, y, z) = dx dy dz .
∴ the volume of the solid in the positive octant
⌠⌠⌠
= ⎮⎮⎮ dx dy dz ,
⌡⌡⌡
where the integral is extended to all positive values of the variables x, y, z subject to
the condition (x ⁄ a) 2 ⁄ 3 + ( y ⁄ b) 2 ⁄ 3 + (z ⁄ c) 2 ⁄ 3 ≤ 1 .
Now put (x ⁄ a) 2 ⁄ 3 = u , ( y ⁄ b) 2 ⁄ 3 = v , (z ⁄ c) 2 ⁄ 3 = w
i.e., x = au3 ⁄ 2, y = bv3 ⁄ 2, z = cw3 ⁄ 2
3 3 3
so that dx = au1 ⁄ 2 du , dy = bv1 ⁄ 2 dv , dz = cw1 ⁄ 2 dw .
2 2 2
∴ the volume in the positive octant
⌠⌠⌠ 27
= ⎮⎮⎮ abc u(3 ⁄ 2) − 1 v(3 ⁄ 2) − 1 w(3 ⁄ 2) − 1 du dv dw ,
⌡⌡⌡ 8
where u + v + w ≤ 1
1
[Γ (3 ⁄ 2)]3 27 ⋅ ( √π) 3
27 2
= abc 9
= abc ⋅ 9 7 5 3 1
8 Γ ( + 1) 8 ⋅ ⋅ ⋅ ⋅ √π
2 2 2 2 2 2
27 π 32 π abc 4
= abc ⋅ ⋅ = ⋅ ⋅
8 8 27.35 8 35
H ence the required volume
π abc 4 4π abc
= 8⋅ ⋅ = ⋅
8 35 35
Remark : The above theorem holds good even if we take the condition as
h1 < x + y + z < h2 in place of h1 ≤ x + y + z ≤ h2 .
⌠⌠⌠
Example 1 : Find the value of ⎮⎮⎮ log (x + y + z) dx dy dz , the integral extending
⌡⌡⌡
over all positive values of x, y, z subject to the condition x + y + z < 1 .
(Kanpur 2014; Purvanchal 14)
Solution : H ere the integral is to be extended for all positive values of x, y and
z such that 0 < x + y + z < 1 .
DIRICHLET’S AND LIOUVILLE’S INTEGRALS I-113
Γ (α + 1) Γ (β + 1) Γ (γ + 1) Γ (α + β + γ + 3) Γ (λ + 1)
= ⋅
Γ (α + β + γ + 3) Γ (α + β + γ + λ + 4)
Γ (α + 1) Γ (β + 1) Γ (γ + 1) Γ (λ + 1)
= ⋅
Γ (α + β + γ + λ + 4)
Example 5 : Evaluate
⌠⌠⌠
⎮⎮⎮ √(a2 b2 c2 − b2 c2 x2 − c2 a2 y2 − a2 b2 z2) dx dy dz
⌡⌡⌡
taken throughout the ellipsoid x2 ⁄ a2 + y2 ⁄ b2 + z2 ⁄ c2 = 1 .
Solution : The given ellipsoid x2 ⁄ a2 + y2 ⁄ b2 + z2 ⁄ c2 = 1 is symmetrical in all
the eight octants. Let us first evaluate the given integral over the region of the
ellipsoid which lies in the positive octant i.e., where x, y, z are all positive.
Put x2 ⁄ a2 = u , y2 ⁄ b2 = v , z2 ⁄ c2 = w .
1
Then x = au1 ⁄ 2, dx = au− 1 ⁄ 2 du etc.
2
Now the given integral extended over the positive octant of the given ellipsoid is
√ 1 − ax
⌠⌠⌠ ⎛ 2 y2 z2 ⎞
I = abc ⎮⎮⎮ ⎜ − − 2 ⎟⎟ dx dy dz , where 0 < x2 ⁄ a2 + y2 ⁄ b2 + z2 ⁄ c2 ≤ 1
⎜
⌡⌡⌡ ⎜
⎝
2 b 2 c ⎟⎠
⌠⌠⌠ 1
= abc ⎮⎮⎮ √(1 − u − v − w) ⋅ abc u− 1 ⁄ 2 v− 1 ⁄ 2 w− 1 ⁄ 2 du dv dw
⌡⌡⌡ 8
where 0 < u + v + w ≤ 1
a2 b2 c2 ⌠⌠⌠ (1 ⁄ 2) − 1 (1 ⁄ 2) − 1 (1 ⁄ 2) − 1
= ⎮⎮⎮ u v w √{1 − (u + v + w)} du dv dw
8 ⌡⌡⌡
1
[Γ ( )]3 1
a2 b2 c2 2 ⌠
= ⋅ ⎮ √(1 − t) . t1 ⁄ 2 + 1 ⁄ 2 + 1 ⁄ 2 − 1 dt ,
8 Γ (3 ⁄ 2) ⌡0
by Liouville’s extension of Dirichlet’s theorem
1
a2 b2 c2 (√π) 3 ⌠
= ⋅ 1 ⎮ (1 − t) (3 ⁄ 2) − 1 t(3 ⁄ 2) − 1 dt
8 . √π ⌡0
2
a2 b2 c2 Γ (3 ⁄ 2) Γ (3 ⁄ 2) π 2 a2 b2 c2
= ⋅ 2π ⋅ = ⋅
8 Γ (3) 32
H ence if the integration is extended throughout the ellipsoid, the given integral
π 2 a2 b2 c2 π 2 a2 b2 c2
= 8I= 8⋅ = ⋅
32 4
√
⌠⌠ ⎧ 1 − x2 ⁄ a 2 − y2 ⁄ b2 ⎫
Example 6 : Evaluate ⎮⎮ ⎪ ⎪ dx dy
⎨ ⎬
⌡⌡ ⎪ 1 + x2 ⁄ a 2 + y2 ⁄ b2 ⎪
⎩ ⎭
where x2 ⁄ a2 + y2 ⁄ b2 ≤ 1 .
Solution : The ellipse x2 ⁄ a2 + y2 ⁄ b2 = 1 is symmetrical in all the four
quadrants.
Let us first evaluate the given integral over the region of the ellipse
x2 ⁄ a2 + y2 ⁄ b2 = 1 which lies in the first quadrant i.e., where x and y are both positive.
Put x2 ⁄ a2 = u , y2 ⁄ b2 = v .
I-116 INTEGRAL CALCULUS
1
Then x = au1 ⁄ 2, dx = au− 1 ⁄ 2 du ,
2
1 − 1⁄2
y = bv1 ⁄ 2, dy = bv dv .
2
∴ the given integral extended over the region of the ellipse x2 ⁄ a2 + y2 ⁄ b2 = 1
which lies in the first quadrant is given by
⌠⌠
I = ⎮⎮
⌡⌡ √ ⎛ 1 − u − v⎞ 1
⎜ ⎟ ⋅
⎝ 1 + u + v⎠ 2
abu− 1 ⁄ 2 v− 1 ⁄ 2 du dv , where 0< u+ v≤ 1
=
ab ⌠⌠
⎮⎮
4 ⌡⌡ √ ⎧ 1 − (u + v) ⎫ (1 ⁄ 2) − 1 (1 ⁄ 2) − 1
⎨ ⎬u
⎩ 1 + (u + v) ⎭
v du dv
1 1
Γ( )Γ( ) 1
=
ab
4
⋅
2
Γ (1)
2 ⌠
⎮
⌡0 √ ⎛ 1 − t⎞
⎜ ⎟ ⋅ t1 ⁄ 2 + 1 ⁄ 2 − 1 dt
⎝ 1 + t⎠
1 π ⁄2
π ab ⌠ 1− t π ab ⌠ 1 − sin θ
= ⋅ ⎮ dt = ⎮ cos θ dθ ,
4 ⌡0 √(1 − t2) 4 ⌡0 cos θ
putting t = sin θ so that dt = cos θ dθ
π ⁄2
π ab ⌠ π ab ⎡ ⎤π ⁄2
= ⎮ (1 − sin θ) dθ = ⎢ θ + cos θ⎥
4 ⌡0 4 ⎣ ⎦0
π ab ⎡ ⎛ π ⎞ ⎤ π ab ⎛ π ⎞
= ⎢ ⎜ + 0⎟ − ⎛⎜ 0 + 1⎞⎟ ⎥ = ⎜ − 1⎟ ⋅
4 ⎣⎝ 2 ⎠ ⎝ ⎠⎦ 4 ⎝2 ⎠
H ence the given integral extended over the whole region of the ellipse
1
x2 ⁄ a2 + y2 ⁄ b2 = 1 = 4. I = π ab ( π − 1) .
2
⌠⌠⌠⌠
Example 7 : Prove that I = ⎮⎮⎮⎮ dx dy dz dw , for all positive values of the
⌡⌡⌡⌡
variables for which x2 + y2 + z2 + w2 is not less than a2 and not greater than b2 is
π 2 (b4 − a4) ⁄ 32 .
Solution : We have to evaluate I subject to the condition
a2 < x2 + y2 + z2 + w2 < b2 .
1
Putting x2 = u1 i.e., x = u11 ⁄ 2, dx = u1− 1 ⁄ 2 du1 etc., we get
2
⌠⌠⌠⌠ 1 1 1 1
I = ⎮⎮⎮⎮ u−1 1 ⁄ 2 ⋅ u−2 1 ⁄ 2⋅ u−3 1 ⁄ 2 ⋅ u−4 1 ⁄ 2 du1 du2 du3 du4
⌡⌡⌡⌡ 2 2 2 2
subject to the condition a2 < u1 + u2 + u3 + u4 < b2
1 ⌠⌠⌠⌠ (1 ⁄ 2) − 1 (1 ⁄ 2) − 1 (1 ⁄ 2) − 1 (1 ⁄ 2) − 1
or I= ⎮⎮⎮⎮ u u2 u3 u4 du1 du2 du3 du4
16 ⌡⌡⌡⌡ 1
1
[Γ ( )]4 b2
1 2 ⌠
= ⋅ ⎮ t1 ⁄ 2 + 1 ⁄ 2 + 1 ⁄ 2 + 1 ⁄ 2 − 1 dt ,
16 Γ ( 1 + 1 + 1 + 1 ) ⌡a2
2 2 2 2
by Liouville’s theorem
2 2
b
(√π) 4 ⌠ π 2 ⎡ t2 ⎤ b π2 4
= ⎮ t dt = ⋅ ⎢ ⎥ = (b − a4) .
16 Γ (2) ⌡a2 16 ⎣ 2 ⎦ a2 32
DIRICHLET’S AND LIOUVILLE’S INTEGRALS I-117
⌠⌠⌠
1. (i) Show that the integral ⎮⎮⎮ x l− 1 y m− 1 z n− 1 dx dy dz integrated over the region
⌡⌡⌡
in the first octant below the surface (x ⁄ a) p + ( y ⁄ b) q + (z ⁄ c) r = 1 is,
a l bm c n Γ (l ⁄ p) Γ (m ⁄ q) Γ (n ⁄ r)
⋅ ⋅
pqr Γ (l ⁄ p + m ⁄ q + n ⁄ r + 1) (Avadh 2011)
⌠⌠⌠
8. (i) E valuate ⎮⎮⎮ x− 1 ⁄ 2 y− 1 ⁄ 2 z− 1 ⁄ 2 (1 − x − y − z) 1 ⁄ 2 dx dy dz extended to all posi
⌡⌡⌡
tive values of the variables subject to the condition x + y + z < 1. (Kanpur 2007)
⌠⌠⌠ dx dy dz π2
, the integral extended to all positive
(ii) Prove that ⎮⎮⎮ =
⌡⌡⌡ √(1 − x2 − y2 − z2) 8
values of the variables for which the expression is real.
(iii) If S is a unit sphere with its centre at the origin, then prove that
⌠⌠⌠ dx dy dz
⎮⎮⎮ = π 2.
⌡⌡⌡S √(1 − x2 − y2 − z2)
⌠⌠⌠
9. E valuate ⎮⎮⎮ (x + y + z + 1) 2 dx dy dz, where R is the region defined by
⌡⌡⌡R
x ≥ 0, y ≥ 0, z ≥ 0, x + y + z ≤ 1.
⌠⌠ ⎛ 1 − x2 − y2 ⎞ 1 ⁄ 2 π
⎜ ⎟
10. Show that ⎮⎮ ⎜ ⎟
⎜ 1 + x2 + y2 ⎟ dx dy = (π − 2)
⌡⌡ ⎝ ⎠ 8
over the positive quadrant of the circle x2 + y2 = 1.
√
⌠⌠⌠ ⎛ 1 − x2 − y2 − z2 ⎞
⎜ ⎟
11. E valuate ⎮⎮⎮ ⎜ ⎟
⎜ 1 + x2 + y2 + z2 ⎟ dx dy dz integral being taken over all positive
⌡⌡⌡ ⎝ ⎠
values of x, y, z such that x2 + y2 + z2 ≤ 1.
⌠⌠
12. (i) E valuate ⎮⎮ √(x2 + y2) dx dy
⌡⌡R
where R is the region in the xy-plane bounded by x2 + y2 = 4 and x2 + y2 = 9.
⌠⌠
(ii) Evaluate ⎮⎮ √(x2 + y2) dx dy where R is the region x2 + y2 ≤ a2.
⌡⌡R
13. E valuate the integral
⌠⌠⌠
⎮⎮⎮ √(1 − x2 − y2 − z2) dx dy dz
⌡⌡⌡R
where R is the region interior to the sphere x2 + y2 + z2 = 1.
14. Find the mass of the region bounded by the ellipsoid
x2 ⁄ a2 + y2 ⁄ b2 + z2 ⁄ c2 = 1
if the density varies as the square of the distance from its centre.
⌠⌠⌠ dx dy dz 1 ⎡ 5 ⎤⎥
15. Prove that ⎮⎮⎮ = ⎢ log 2 −
⌡⌡⌡ (x + y + z + 1) 3 2 ⎣ 8⎦
throughout the volume bounded by the coordinate planes and the plane x + y + z = 1.
(Rohilkhand 2013)
4 4
3. (i) π abc ⁄ 6. (ii) π abc. (iii) π abc.
3 3
4. a2b2c2 ⁄ 48. 5. k a2b2c2 ⁄ 720. 6. 1 ⁄ 2520.
DIRICHLET’S AND LIOUVILLE’S INTEGRALS I-119
True or False:
Write ‘T ’ for true and ‘F ’ for false statem ent.
1
⌠⌠⌠ dx dy dz 1⌠ u2
5. ⎮⎮⎮ = ⎮ du,
⌡ ⌡ ⌡ (x + y + z + 1) 3 3 ⌡0 (u + 1) 3
where the region of integration is the volume bounded by the coordinate planes
and the plane x + y + z = 1.
1
⌠⌠⌠ 1⌠
6. ⎮ ⎮ ⎮ (x + y + z + 1) 2 dx dy dz = ⎮ u2 (u + 1) 2 du,
⌡⌡⌡ 2 ⌡0
where the region of integration is the volume bounded by the coordinate planes
and the plane x + y + z = 1.
7. If l and m are both positive, then the double integral
⌠ ⌠ l− 1 m− 1 Γ (l) Γ (m) ,
⎮⎮ x y dx dy =
⌡⌡ Γ (l + m)
where the integral is extended to all positve values of the variables x and y
subject to the condition x + y ≤ 1.
x2 y2
Example 1 : Find the area bounded by the ellipse 2
+ 2 = 1, the ordinates
a b
x = c , x = d and the x-axis. (Meerut 2000)
Solution : E quation of the ellipse is
x2 y2 y2 x2
2
+ 2= 1 or 2
= 1− 2
a b b a
AREAS OF CURVES I-123
b
giving y= √(a2 − x2) . ...(1)
a
∴ the required area = the area ABDC
d d
⌠ ⌠ b
= ⎮ y dx = ⎮ √(a2 − x2) dx ,
⌡c ⌡c a
from (1)
b ⎡1 1 ⎛ x⎞⎤d
= ⎢ x √(a 2 − x2) + a 2 sin− 1 ⎜ ⎟ ⎥
a ⎣2 2 ⎝ a⎠⎦c
b ⎡ ⎛ d c⎞ ⎤
= ⎢ d √(a 2 − d2) − c √(a 2 − c2) + a 2 ⎜ sin− 1 − sin− 1 ⎟ ⎥ ⋅
2a ⎣ ⎝ a a⎠ ⎦
Example 2 : Find the area bounded by the parabola y2 = 4ax and its latus rectum.
(Agra 2005; Avadh 05; Bundelkhand 08)
Solution : Latus rectum is a line through the focus S (a, 0) and perpendicular to
x-axis i.e., its equation is x = a . Also the curve is symmetrical about x-axis.
∴ the required area L OL′
a
⌠
= 2 × area OSL = 2 . ⎮ y dx
⌡0
a
⌠
= 2 ⎮ √(4ax ) dx ,
⌡0
Note that the shifting of the origin only changes the equation of the curve and has
no effect on its shape. Now the origin being at the point A , the new limits for the loop
are x = − a to x = 0 .
∴ required area of the loop = 2 × area CPA
0
⌠
= 2 ⎮ y dx , [the value of y to be put from (1)]
⌡− a
0 ⎧ ⎫
√
⌠ ⎪ ⎛ a + x⎞ ⎪⎬
= 2⎮ ⎨− x ⎜ ⎟ ⎪ dx , [Note that in the equation (1), for
⌡− a ⎪ ⎝ a − x⎠ ⎭
⎩
the portion CPA , y = − x √{(a + x ) ⁄ (a − x )}]
0
⌠ − x (a + x )
= 2⎮ dx , multiplying the numerator and the
⌡− a √(a2 − x2)
denominator by √(a + x )
0
⌠ − (− a sin θ) (a − a sin θ)
= 2⎮ ⋅ (− a cos θ) dθ ,
⌡π ⁄ 2 a cos θ
putting x = − a sin θ and dx = − a cos θ dθ
0 π ⁄2
⌠ ⌠
= − 2a2 ⎮ (sin θ − sin2 θ) dθ = 2a2 ⎮ (sin θ − sin2 θ) dθ
⌡π ⁄ 2 ⌡0
1 1
= 2a2 [1 − ⋅ π] , by Walli’s formula
2 2
1
= 2a2 (1 − π) .
4
√
a a ⎛ a 2 x2 ⎞
⌠ ⌠ ⎜ ⎟
= 4 ⎮ y dx = 4 ⎮ ⎜ ⎟
⎜ a 2 − x2 ⎟ dx ,
⌡0 ⌡0 ⎝ ⎠
[ ... from the equation of the given curve, y2 = a2 x2 ⁄ (a2 − x2)]
a a ⎡ (a 2 − x2) 1 ⁄ 2 ⎤ a
⌠ ax dx ⌠ − 2x dx ⎢⎢ ⎥⎥
= 4⎮ = − 2a ⎮ = − 2a ⎢⎣ ⎥⎦
⌡0 √(a2 − x2) ⌡0 √(a2 − x2) 1⁄2 0
= − 4a [0 − a] = 4a2.
1. Find the area bounded by the axis of x , and the following curves and the given
ordinates :
(i) y = log x ; x = a , x = b (b > a > 1) .
(ii) xy = c2 ; x = a , x = b , (a > b > 0) . (Kashi 2012)
2. (i) Find the area bounded by the curve y = x3,
the y-axis and the lines y = 1 and
y= 8.
(ii) Show that the area cut off a parabola by any double ordinate is two third of
the corresponding rectangle contained by that double ordinate and its distance
from the vertex.
3. (i) Find the area of the quadrant of an ellipse ( x2 ⁄ a2) + ( y2 ⁄ b2) = 1 .
(Bundelkhand 2010; Kanpur 11)
4. (i) Trace the curve ay2 = x2 (a − x ) and show that the area of its loop is 8a2 ⁄ 15 .
(Avadh 2008)
I-126 INTEGRAL CALCULUS
(ii) Find the area enclosed by the curve xy2 = a2 (a − x ) and y-axis.
(iii) Trace the curve a2 y2 = a2 x2 − x4 and find the whole area within it.
(Rohilkhand 2012; Avadh 12, Bundelkhand 14)
8. (i) Prove that the area of a loop of the curve a4 y2 = x4 (a2 − x2) is πa2 ⁄ 8 .
(ii) Show that the whole area of the curve a4 y2 = x5 (2a − x ) is to that of the circle
whose radius is a, as 5 to 4. (Kanpur 2010)
9. (i) Find the area between the curve y2 (a − x ) = x3 (cissoid) and its asymptotes.
Also find the ratio in which the ordinate x = a ⁄ 2 divides the area.
(ii) Find the area of the loop of the curve y2 (a − x ) = x2 (a + x ) .
(Purvanchal 2011)
10. Trace the curve y2 (a + x ) = (a − x ) 3. Find the area between the curve and its
asymptote. (Purvanchal 2007)
Example 1 : Find the area included between the curves y2 = 4ax and x2 = 4by .
(Bundelkhand 2011; Rohilkhand 10)
Solution : Solving the equations of the two given curves, we have
y4 = 16a2 (4by ) = 64a2 by .
∴ y ( y3 − 64 a2b) = 0 , giving y = 0 , 4a2 ⁄ 3 b1 ⁄ 3.
W h e n y = 0 , x = 0 a n d wh e n y = 4a2 ⁄ 3 b1 ⁄ 3,
x = 4a1 ⁄ 3 b2 ⁄ 3.
H ence, the points of intersection of the given
curves are O (0, 0) and A (4a1 ⁄ 3 b2 ⁄ 3, 4b1 ⁄ 3 a2 ⁄ 3) .
∴ the required area (i.e., the shaded area)
= area OPA L − area OQA L
4a1 ⁄ 3 b2 ⁄ 3
⌠
= ⎮ y dx , from the curve y2 = 4ax
⌡0
4a1 ⁄ 3 b2 ⁄ 3
⌠
− ⎮ y dx , from the curve x2 = 4by
⌡0
(Note that for the required area
x varies from 0 to 4a1 ⁄ 3 b2 ⁄ 3)
4a1 ⁄ 3 b2 ⁄ 3 4a1 ⁄ 3 b2 ⁄ 3 ⎛ 2 ⎞
⌠ ⌠ x
= ⎮ √(4ax ) dx − ⎮ ⎜ ⎟ dx
⌡0 ⌡0 ⎝ 4b ⎠
1⁄ 3 2⁄ 3
⎡ 2x3 ⁄ 2 ⎤⎥ 4a b 4a1 ⁄ 3 b2 ⁄ 3
1 ⎡ x3 ⎤
= 2 √a ⎢⎢⎢ ⎥ − ⎢ ⎥
⎣ 3 ⎥⎦ 0 4b ⎣ 3 ⎦ 0
4 √a 1 32 16 16
= [8√(a) . b] − (64 ab2) = ab − ab = ab .
3 12b 3 3 3
Example 2 : Find the area of the segm ent cut off from the parabola y2 = 2x by the
straight line y = 4x − 1 .
I-128 INTEGRAL CALCULUS
1 b b⎡x 1 x⎤ a
= x . √(a2 − x2) + ⎢ √(a2 − x2) + a2 sin− 1 ⎥
2 a a ⎣2 2 a⎦ x
bx b⎡ 1 π x 1 x⎤
= √(a2 − x2) + ⎢ 0 + a2. − √(a2 − x2) − a2 sin− 1 ⎥
2a a⎣ 2 2 2 2 a⎦
bx b 1 ⎛π x⎞ bx
= √(a2 − x2) + ⋅ a2 ⎜ − sin− 1 ⎟ − √(a2 − x2)
2a a 2 ⎝2 a⎠ 2a
ab ⎛ π x⎞ ab x
= ⎜ − sin− 1 ⎟ = cos− 1 ⋅
2 ⎝2 a⎠ 2 a
ab x
Thus S= cos− 1 ⋅
2 a
x 2S x 2S 2S
∴ cos− 1 = or = cos or x = a cos ⋅
a ab a ab ab
b b 2S
Also y = √(a2 − x2) = √{a2 − a2 cos2 (2S ⁄ ab)} = b sin ⋅
a a ab
1 ⎡ y3 ⎤ 4a √3 ⎡1 64a2 y⎤
8a
= 32 πa2 + ⎢ ⎥ + 2 ⎢ y √(64a2 − y2) + sin− 1 ⎥
6a ⎣ 3 ⎦0 ⎣2 2 8a ⎦ 4a √3
1 ⎡ 64 × 3 √3a3 ⎤
= 32 πa2 + ⎢ ⎥
6a ⎣ 3 ⎦
+ 2 [{0 − 8a2 √3} + 32a2 {sin− 1 1 − sin− 1 (√3 ⁄ 2)}]
32 √3a2 32 2
= 32 πa2 + − 16a2 √3 + a π
3 3
I-130 INTEGRAL CALCULUS
128 2 16 2 16 2
= a π− a √3 = a (8π − √3) .
3 3 3
Example 5 : Find by double integration the area of the region enclosed by the curves
x2 + y2 = a2, x + y = a (in the first quadrant).
Solution : The given equations of the circle
x2 + y2 = a2
[centre (0, 0) and radius a] and of the
st r a i gh t l i n e x + y = a ( wi t h e q u a l
intercepts a on both the axes) can be
easily traced as shown in the figure.
T h e r e quir e d a r e a is t he ar ea
bounded by the arc AB an d t he line
A B . To find it with the help of double
integration take any point P ( x, y ) in this
portion and consider an elementary area
δx δy at P . The required area can now be
cover ed by fir st mo ving y fr o m t h e
straight line x + y = a to the arc of the
circle x2 + y2 = a2 an d then moving x
from 0 to a .
a √(a2 − x2)
⌠ ⌠
∴ the required area = ⎮ ⎮ dx dy , the first integration to be
⌡x = 0 ⌡y = (a − x )
performed w.r.t. y whose limits are variable
a a
⌠ √(a2 − x2) ⌠
= ⎮ ⎡⎢ y ⎤⎥ dx = ⎮ [√(a2 − x2) − (a − x)] dx
⌡0 ⎣ ⎦ (a − x ) ⌡0
⎡ ⎧1 ⎫ a
1 1 ⎤
= ⎢ ⎨ x √(a2 − x2) + a2 sin− 1 ( x ⁄ a) ⎬ − ax + x2⎥
⎣ ⎩2 2 ⎭ 2 ⎦0
1 1 1 1 1 1
= a2. ( π) − a2 + a2 = a2 ( π − 1) = a2 (π − 2) .
2 2 2 2 2 4
Note : The required area can also the covered by first moving x from the st. line
x + y = a to the arc of the circle x2 + y2 = a2 and then moving y from 0 to a .
1. Find the common area between the curves y2 = 4ax and x2 = 4ay .
(Meerut 2004B, 08; Agra 14)
8. Find the area included between the parabola x2 = 4ay and the curve
y= 8a3 ⁄ ( x2 + 4a2) . (Rohilkhand 2008B)
9. Find by double integration the area bounded by the curves y (x2 + 2) = 3x and
4y = x2.
10. Find by double integration the area lying between the parabola y = 4x − x2 and
the straight line y = x.
π ⁄2
⌠ 1 1
= 8a2 ⎮ sin4 φ 2dφ , putting θ = φ so that dθ = dφ
⌡0 2 2
π ⁄2
⌠ 3.1 π ,
= 16a2 ⎮ sin4 φ dφ = 16a2 ⋅ ⋅ by Walli’s formula
⌡0 4.2 2
= 3π a2.
Example 2 : Find the whole area of the curve (hypocycloid) given by the equations
x = a cos3 t , y = b sin3 t .
(Gorakhpur 2005; Rohilkhand 09; Kashi 11)
Solution : E liminating t from the given equations the cartesian equation of the
curve is obtained as (x ⁄ a) 2 ⁄ 3 + (y ⁄ b) 2 ⁄ 3 = 1 i.e., {(x ⁄ a) 2 }1 ⁄ 3 + {( y ⁄ b) 2}1 ⁄ 3 = 1 .
AREAS OF CURVES I-133
1. Find the area included between the curve x = a (t + sin t) , y = a (1 − cos t) and
its base. (Agra 2005)
2. Find the area of a loop of the curve x = a sin 2t , y = a sin t or
a2 x2 = 4y2 (a2 − y2) .
3. Show that the area bounded by the cissoid x = a sin2 t , y = (a sin3 t) ⁄ cos t
and its asymptote is 3 πa2 ⁄ 4 .
(Purvanchal 2006; Avadh 09; Rohilkhand 11; Purvanchal 14)
4. Find the area of the loop of the curve
x = a (1 − t2) , y = at (1 − t2) , where − 1 ≤ t ≤ 1 .
1. 3πa2. 2. 4a2 ⁄ 3
3. 3πa2 ⁄ 4. 4. 8a2 ⁄ (15).
Now the L.H.S. = the value of A for θ equal to θ2 − the value of A for θ equal to θ1
= (the area A OB) − 0 = area A OB .
θ
1⌠ 2 2
H ence the required area A OB = ⎮ r dθ .
2 ⌡θ
1
Note : In some cases it is more convenient to find the required area by using
double integration. In that case the area is given by
θ f (θ)
⌠ 2 ⌠
⎮ ⎮ r dθ dr , (θ1 < θ2) .
⌡θ = θ ⌡r = 0
1
1. Find the area between the following curves and the given radii vectors :
(i) The spiral r θ1 ⁄ 2 = a ; θ = α , θ = β .
(ii) The parabola l ⁄ r = 1 + cos θ ; θ = 0 , θ = α .
2. Find the area of the loop of the curve r = a θ cos θ between θ = 0 and θ = π ⁄ 2 .
(Kanpur 2009)
3. (i) Find the area of one loop of r = a cos 4θ . (Rohilkhand 2007)
(ii) Find the area of a loop of the curve r = a sin 3θ .
4. (i) Find the whole area of the curve r = a sin 2θ . (Bundelkhand 2009)
(ii) Find the whole area of the curve r = a cos 2θ .
5. (i) Find the whole area of the curve r2 = a2 cos2 θ + b2 sin2 θ .
(ii) Find the area of the cardioid r = a (1 − cos θ) .
AREAS OF CURVES I-137
6. (i) Show that the area of the limacon r = a + b cos θ , (b < a) is equal to
1 2
π (a2 + b ).
2
( ii) P r o ve t h a t t h e su m o f t h e ar e a s o f t h e t wo lo o p s o f t h e lim a co n
r = a + b cos θ , (b > a) is equal to π (2a2 + b2) ⁄ 2 .
7. Calculate the ratio of the area of the larger to the area of the smaller loop of the
1
curve r = + cos 2θ .
2
8. Show that the area of a loop of r = a cos nθ is π a2 ⁄ 4n , n being integral. Also prove
that the whole area is π a2 ⁄ 4 or π a2 ⁄ 2 according as n is odd or even.
9. Trace the curve r = √3 cos 3θ + sin 3θ , and find the area of a loop.
1 1 2 ⎡
1 1 1 ⎤ πa2 2
1. (i) a2 log ( β ⁄ α), (ii) l ⎢ tan α + tan3 α⎥⎦ ⋅ 2. (π − 6).
2 4 ⎣ 2 3 2 96
3. (i) πa2 ⁄ (16). (ii) πa2 ⁄ (12). 4. (i) πa2 ⁄ 2 .
1
(ii) πa2 ⁄ 2 . 5. (i) π (a2 + b 2). (ii) 3πa2 ⁄ 2 .
2
4π + 3 √3
7. ⋅ 9. π ⁄ 3.
2 π − 3 √3
π π ⁄2
⌠ 1 ⌠
= 16a2 ⎮ cos4 θ dθ = 16a2 ⎮ cos4 φ .2 dφ ,
⌡0 2 ⌡0
1
(putting θ = φ so that 12 dθ = dφ ;
2
1
also when θ = 0 , φ = 0 and when θ = π , φ = π)
2
3.1 π
= 32a2 ⋅ = 6π a2. ...(1)
4.2 2
π ⁄2
1⌠ 2a
Area OA CPO = ⎮ r2 dθ , for the parabola r =
2 ⌡0 1 + cos θ
π ⁄2 π ⁄2
1 ⌠ dθ a2 ⌠ 1
= ⋅ 4a2 ⎮ = ⎮ sec4 θ dθ
2 ⌡0 (1 + cos θ) 2 2 ⌡0 2
AREAS OF CURVES I-139
π ⁄2
1 ⌠ 1 1
= a2 ⎮ (1 + tan2 θ) sec2 θ dθ
2 ⌡0 2 2
⎡ π ⁄2 4a2
1 1 1 ⎤ 1
= a2 ⎢ tan θ + tan3 θ⎥ = a2 (1 + ) = ⋅ ...(2)
⎣ 2 3 2 ⎦
0
3 3
π
1 ⌠
Also area OPBO = ⎮ r2 dθ , for the cardioid r = 2a (1 + cos θ)
2 ⌡π ⁄ 2
π
1⌠
= ⎮ 4a2 (1 + cos θ) 2 dθ
2 ⌡π ⁄ 2
π
⌠
= 2a2 ⎮ [1 + 2 cos θ + cos2 θ] dθ
⌡π ⁄ 2
π
⌠ 1 1
= 2a2 ⎮ (1 + 2 cos θ + + cos 2θ) dθ
⌡π ⁄ 2 2 2
⎡3 1 ⎤π 3
= 2a2 ⎢ θ + 2 sin θ + sin 2θ⎥ = π a2 − 4a2. ...(3)
⎣2 4 ⎦π ⁄2 2
Adding (2) and (3) and multiplying by 2, we get the whole area included between
the two curves i.e., the area of the smaller portion of the cardioid
⎡4 ⎛3 ⎞⎤ ⎡ 16⎤
= 2 × ⎢ a2 + ⎜ πa2 − 4a2⎟ ⎥ = a2 ⎢ 3π − ⎥
⎣3 ⎝2 ⎠⎦ ⎣ 3⎦
1
= a2 [9π − 16] . ...(4)
3
Also the shaded area (i.e., the area of the larger portion of the cardioid)
= (Area of the whole cardioid) − (unshaded area) i.e., = (1) − (4)
1 1
= 6 π a2 − a2 (9π − 16) = a2 (9π + 16) . ...(5)
3 3
1
Larger area a2 (9π + 16) 9π + 16
3
∴ R atio of the two parts = = = ⋅
Smaller area 1 a2 (9π − 16) 9π − 16
3
Example 3 : Find the area lying between the cardioid r = a (1 − cos θ) and its
double tangent.
Solution : L e t PQ b e t h e d o u b le t a n ge n t o f t h e ca r d io id . C le ar ly it is
perpendicular to OX i.e., it must be inclined at an angle of 90o to the initial line
i.e., ψ = 90o at P .
Also we know that at any point of a curve,
ψ= θ+ φ. ...(1)
Now tan φ = r (dθ ⁄ dr) = r ⁄ (dr ⁄ dθ) = a (1 − cos θ) ⁄ (a sin θ) ,
[... r = a (1 − cos θ)]
I-140 INTEGRAL CALCULUS
1
2 sin2 θ
2 1
= 1 1
= tan θ .
2
2 sin θ cos θ
2 2
1
∴ φ= θ.
2
1 3
Putting the value of φ in (1), we get ψ = θ + θ= θ.
2 2
1 1 3
Since at P , ψ = π , therefore at P , π = θ or θ = π ⁄ 3 .
2 2 2
∴ the vectorial angle of the point of contact P of the double tangent is π ⁄ 3 i.e.,
60o . Substituting this value of θ in the equation of the curve, we get the radius vector
OP = a (1 − cos 60o ) = a ⁄ 2 .
Thus in the triangle OPM ,
1
OP = a , ∠ POM = 60o , ∠ PMO = 90o .
2
1 1 1 1 1
∴ OM = a cos 60o = a ⋅ and PM = a sin 60o = a (√3 ⁄ 2) .
2 2 2 2 2
1 1 1
∴ area of the triangle OPM = OM . PM = ( a) (√3a ⁄ 4) = (1 ⁄ 32) a2 √3 .
2 2 4
Also the sectorial area OPO of the cardioid r = a (1 − cos θ) i.e., the dotted area
π ⁄3 π ⁄3
1⌠ 1⌠
= ⎮ r2 dθ = ⎮ a2 (1 − cos θ) 2 dθ
2 ⌡0 2 ⌡0
π ⁄3
1 2⌠ 1 1
= a ⎮ (1 − 2 cos θ + + cos 2θ) dθ
2 ⌡0 2 2
1 ⎡3 1 ⎤π ⁄3
= a2 ⎢ θ − 2 sin θ + sin 2θ⎥
2 ⎣2 4 ⎦0
1 1 1 1 2
= a2 ( π − √3 + √3) = a (4π − 7 √3) .
2 2 8 16
H ence the required area (i.e., the area shaded by vertical lines)
= 2 [area of Δ OPM − area of sector OPO]
⎡ ⎤
1 2 1 2 1 2
= 2 ⎢⎣ a √3 − a (4π − 7 √3) ⎥⎦ = a (15 √3 − 8π) .
32 16 16
Example 4 : Find the area of a loop of the curve r = a sin 3θ outside the circle
r = a ⁄ 2 and hence find the whole area of the curve outside the circle r = a ⁄ 2 .
Solution : E liminating r between the two given equations, we get
AREAS OF CURVES I-141
1
i.e., (a ⁄ 2) = a sin 3θ sin 3θ =
2
i.e., 3θ = π ⁄ 6 or 5π ⁄ 6
i.e., θ = π ⁄ 18 or 5π ⁄ 18
i.e., θ = 10o or 50o .
T h u s t h e l o o p o f t h e c u r ve
r = a sin 3θ l yi n g b e t we e n θ = 0 a n d
θ = π ⁄ 3 intersects the circle r = a ⁄ 2 at the
points B and B′ where θ = 10o at B and
θ = 50o at B′ . This loop is symmetrical
about OA and θ = π ⁄ 6 at A .
Now the required area of a loop of
t h e cu rve r = a sin 3θ lying ou tside t he
circle r = a ⁄ 2
= the area BAB′ CB
(i.e., the shaded area)
= 2 × area BA CB , (by symmetry)
= 2 × [(area of the curve r = a sin 3θ between the radii vectors OB and OA i.e.,
θ = π ⁄ 18 and θ = π ⁄ 6) − (area of the circle r = a ⁄ 2 between
the radii vectors OB and OC i.e., θ = π ⁄ 18 and θ = π ⁄ 6 )]
⎡1 ⌠ π ⁄6
= 2 ⎢⎢ ⎮ r2 dθ , for the curve r = a sin 3θ
⎢ 2 ⌡π ⁄ 18
⎣
π ⁄6
1⌠ a⎤
− ⎮ r2 dθ , for the circle r = ⎥
2 ⌡π ⁄18 2⎦
π ⁄6 π ⁄6 π ⁄6
⌠ ⌠ a2 a2 ⌠ a2 ⎡ ⎤ π ⁄ 6
= ⎮ a2 sin2 3θ dθ − ⎮ dθ = ⎮ (1 − cos 6θ) dθ − θ
⌡π ⁄ 18 ⌡π ⁄ 18 4 2 ⌡π ⁄ 18 4 ⎢⎣ ⎥⎦ π ⁄ 18
a2 ⎡ sin 6θ⎤ π ⁄ 6 a2 ⎡ π π ⎤ a2 ⎡ ⎧ π sin π ⎫ ⎧ π 1 π ⎫⎤ a2 π
= ⎢θ − ⎥ − ⎢ − ⎥= ⎢⎨ − ⎬− ⎨ − sin ⎬ ⎥ − ⋅
2 ⎣ 6 ⎦ π ⁄ 18 4 ⎣ 6 18⎦ 2 ⎣⎩6 6 ⎭ ⎩ 18 6 3 ⎭⎦ 4 9
a2 ⎡ π ⎧ π 1 √3 ⎫ ⎤ a2 π a2 ⎡ π √3 ⎤ a2 π a2
= ⎢ − ⎨ − ⋅ ⎬⎥ − = ⎢ + ⎥ − = [2π + 3 √3] .
2 ⎣ 6 ⎩ 18 6 2 ⎭ ⎦ 36 2 ⎣9 12 ⎦ 36 72
Again the curve r = a sin 3θ has 3 equal loops ( ... n = 3 which is odd).
∴ whole area of the curve r = a sin 3θ outside the circle r = a ⁄ 2
= 3 × area BA B′ CB i.e., 3 times the shaded area
1 2 1 2
= 3× a [2π + 3 √3] = a [2π + 3 √3] .
72 24
1. Find the area outside the circle r = 2a cos θ and inside the cardioid
r = a (1 + cos θ) .
2. Find the total area inside r = sin θ and outside r = 1 − cos θ .
I-142 INTEGRAL CALCULUS
3. Find by double integration the area lying inside the circle r = a sin θ and outside
the cardioid r = a (1 − cos θ).
4. Find the area common to the circle r = a and the cardioid r = a (1 + cos θ) .
(Meerut 2007)
5. Find the area of the portion included between the cardioids
r = a (1 + cos θ) and r = a (1 − cos θ) .
6. Show that the area contained between the circle r = a and the curve r = a cos 5θ
is equal to three-fourth of the area of the circle.
7. Find the area between the curve r = a (sec θ + cos θ) and its asymptote.
(Purvanchal 2010)
8. O is the pole of the lemniscate = r2 a2
cos 2θ and PQ is a common tangent to its
two loops. Find the area bounded by the line PQ and the arcs OP and OQ of the
curve.
Example 1 : Find the area of a loop of the folium x3 + y3 = 3axy . (Meerut 2001)
1
1. a2 (π ⁄ 2). 2. πa2 ⁄ 4. 4. a2 log 2 .
4
b
⌠ a2 2 m β
1. quadrature 2. ⎮ f (x) dx 3. (e − e 2 m α)
⌡a 4m
a
4. three 5. c 2 sinh ⋅ 6. (a). 7. (a).
c
8. (b). 9. (b). 10. (a). 11. T. 12. F.
13. T.
7.1 Rectification
The process of finding the length of an arc of a curve between two given points is
called rectification.
If s denotes the arc length of a curve measured from a fixed point to any point on
it, then as proved in Unit I on Differential Calculus, we have
ds
dx
= ±
√ ⎧ ⎛ dy⎞ 2 ⎫
⎨1 + ⎜ ⎟ ⎬ ,
⎩ ⎝ dx⎠ ⎭
where + ive or –ive sign is to be taken before the radical sign according as x increases
or decreases as s increases. H ence if s increases as x increases, we have
ds
dx
=
√ ⎧ ⎛ dy⎞ 2 ⎫
⎨ 1 + ⎜ ⎟ ⎬ or ds =
⎩ ⎝ dx⎠ ⎭ √ ⎧ ⎛ dy⎞ 2 ⎫
⎨ 1 + ⎜ ⎟ ⎬ dx .
⎩ ⎝ dx⎠ ⎭
I-146 INTEGRAL CALCULUS
x
⌠
Integrating, we have s = ⎮
⌡a √ ⎧ ⎛ dy⎞ 2 ⎫
⎨ 1 + ⎜ ⎟ ⎬ dx ,
⎩ ⎝ dx⎠ ⎭
where a is the abscissa of the fixed point from which s is measured.
H ence the arc length of the curve y = f (x) included between two points for which
x = a and x = b is equal to
√ 1+
b
⌠ ⎧
⎨ ⎛ dy⎞ 2 ⎫⎬
⎮ ⎩ ⎜ ⎟ ⎭ dx , (b > a) .
⌡a ⎝ dx⎠
Sometimes it is more convenient to take y as the independent variable. Then the
length of the arc of the curve x = f ( y) between y = a and y = b is equal to
√
b
⌠ ⎧ dx 2 ⎫
⎨ 1 + ⎛⎜ ⎞⎟ ⎬ dy ,
⎮ ⎩
(b > a) .
⌡a ⎝ dy⎠ ⎭
Remark : Suppose we have to find the length of the arc of a curve (whose
cartesian equation is given) lying between the points (x1, y1) and (x2 , y2) . We can use
either of the two formulae
x y
⌠ 2 ⌠ 2
s = ⎮ √{1 + (dy ⁄ dx) 2} dx and s = ⎮ √{1 + (dx ⁄ dy) 2} dy .
⌡x ⌡y
1 1
If we feel any difficulty in integration while using one of these two formulae, we
must try the other formula also.
Example 1 : Show that the length of the curve y = log sec x between the points where
1
x = 0 and x = π is log (2 + √3) . (Kanpur 2005; Rohilkhand 14)
3
Solution : The given curve is y = log sec x . ...(1)
Differentiating (1) w.r.t. x , we get
dy 1
= sec x tan x = tan x .
dx sec x
⎛ ds ⎞ 2 ⎛ dy⎞ 2
Now ⎜ ⎟ = 1 + ⎜ ⎟ = 1 + tan2 x = sec 2 x . ...(2)
⎝ dx⎠ ⎝ dx⎠
If the arc length s of the given curve is measured from x = 0 in the direction of x
ds
increasing, we have = sec x or ds = sec x dx .
dx
1
Therefore if s1 denotes the arc length from x = 0 to x = π , then
3
s π ⁄3
⌠1 ⌠ π ⁄3
⎮ ds = ⎮ sec x dx = ⎡⎢ log (sec x + tan x) ⎤⎥
⌡0 ⌡0 ⎣ ⎦0
1 1
or s1 = [log (sec π + tan π) − log 1] = log (2 + √3) .
3 3
Example 2 : Find the length of the arc of the parabola y2 = 4ax extending from the
vertex to an extremity of the latus rectum . (Meerut 2009)
Solution : The given equation of parabola is y2 = 4ax . ...(1)
RECTIFICATION I-147
√ ⎛ dy⎞ 2 ⎫
√
⌠ ⎧ ⌠ ⎧⎪ (4a − 3x) 2 ⎫⎪
= 2⎮ ⎨ 1 + ⎜ ⎟ ⎬ dx = 2 ⎮ ⎨⎪ ⎬ dx
⌡0 ⎩ ⎝ dx⎠ ⎭ ⌡0 ⎩ 12a (a − x) ⎪⎭
I-148 INTEGRAL CALCULUS
a a
1 ⌠ (4a − 3x) 1 ⌠ 3 (a − x) + a
= ⎮ dx = ⎮ dx
√(3a) ⌡0 √(a − x) √(3a) ⌡0 √(a − x)
a
1 ⌠ ⎡ 3 (a − x) a ⎤
= ⎮ ⎢ + ⎥ dx
√(3a) ⌡0 ⎣ √(a − x) √(a − x) ⎦
a
1 ⌠
= ⎮ [3 √(a − x) + a (a − x) − 1 ⁄ 2] dx
√(3a) ⌡0
1 ⎡ 2 ⎤a
= ⎢− 3 ⋅ (a − x) 3 ⁄ 2 − a . 2 (a − x) 1 ⁄ 2⎥
√(3a) ⎣ 3 ⎦0
1 ⎡ 3⁄2 4a
= ⎢ 2a + 2a3 ⁄ 2⎤⎥⎦ = ⋅
√(3a) ⎣ √3
⎛ 2 ⁄ 3⎞
√ 1 + yx
a a
⌠
= 4⎮
⌡0 √ ⎧
⎩
⎛ dy⎞ 2 ⎫
⎝ dx⎠ ⎭
⌠
⎨ 1 + ⎜ ⎟ ⎬ dx = 4 ⎮
⌡0
⎜
⎜
⎜
⎝
⎟
⎟ dx
2 ⁄ 3⎟
⎠
a a
⌠ √(x2 ⁄ 3 + y 2 ⁄ 3) ⌠ √(a2 ⁄ 3)
= 4⎮ dx = 4 ⎮ dx
⌡0 x1 ⁄ 3 ⌡0 x1 ⁄ 3
a
⌠ ⎡3 ⎤a
= 4a1 ⁄ 3 ⎮ x − 1 ⁄ 3 dx = 4a1 ⁄ 3 ⎢ x2 ⁄ 3⎥ = 6a .
⌡0 ⎣2 ⎦0
1 2 1
1. (i) Find the arc length of the curve y = x − log x from x = 1 to x = 2 .
2 4
(Meerut 2012B)
ex − 1
(ii) Find the length of the curve y = log x from x = 1 to x = 2 .
e + 1
(Meerut 2004B; Agra 06; Avadh 08; Kanpur 11;
Rohilkhand 13; Kashi 13)
RECTIFICATION I-149
2. (i) Show that in the catenary y = c cosh (x ⁄ c) , the length of arc from the vertex
to any point is given by s = c sinh (x ⁄ c) .
(ii) If s be the length of the arc of the catenary y = c cosh (x ⁄ c) from the vertex
(0, c) to the point (x, y) , show that s2 = y2 − c2 .
3. (i) Find the length of an arc of the parabola y2 = 4ax measured from the vertex.
(ii) Find the length of the arc of the parabola y2 = 4ax cut off by its latus rectum.
4. (i) Find the length of the arc of the parabola x2 = 4ay from the vertex to an
extremity of the latus rectum. (Kanpur 2008; Purvanchal 09)
(ii) Find the length of the arc of the parabola x2 = 8y from the vertex to an
extremity of the latus rectum.
5. (i) Find the length of the arc of the parabola y2 = 4ax cut off by the line y = 3x .
(ii) Show that the length of the arc of the parabola y2 = 4ax which is intercepted
between the points of intersection of the parabola and the straight line 3y = 8x is
⎛ 15 ⎞⎟
a ⎜⎝ log 2 + ⋅
16 ⎠ (Gorakhpur 2006; Purvanchal 06)
6. (i) Find the perimeter of the curve x2 + y2 = a2. (Avadh 2010; Rohilkhand 13B)
(ii) Find the length of the arc of the semi-cubical parabola ay2 = x3 from the vertex
to the point (a, a). (Bundelkhand 2010)
7. (i) Show that the length of the arc of the curve x2 = a2 (1 − e y ⁄ a) measured from
the origin to the point (x, y) is a log {(a + x) ⁄ (a − x)} − x . (Rohilkhand 2010B)
(ii) Prove that the length of the loop of the curve 3ay2 = x (x − a) 2 is 4a / √3.
(Meerut 2005B, 08, 09B)
8. (i) Find the perimeter of the loop of the curve 9ay2 = (x − 2a) (x − 5a) 2.
(ii) Show that the whole length of the curve x2 (a2 − x2) = 8a2 y2 is π a √2 .
(Bundelkhand 2006; Purvanchal 11)
3 1 ⎛ 1⎞
1. (i) + log 2 . (ii) log ⎜⎝ e + ⎟⎠
2 4 e
⎡ ⎧ 2 2 ⎫⎤
1 ⎢⎢ ⎪ y + √( y + 4a ) ⎪ ⎥
2 2 2 ⎨ ⎬⎥ ⋅
3. (i) ⎢ y √( y + 4a ) + 4a log ⎪ ⎪⎥
4a ⎣ ⎩ 2a ⎭⎦
(ii) 2 a [√2 + log (1 + √2)].
4. (i) a [√2 + log (1 + √2)].
(ii) 2 [√2 + log (1 + √2)].
⎡ 2 √(13) ⎧ 2 + √(13) ⎫ ⎤
5. (i) a ⎢⎣ + log ⎨⎩ ⎬⎥ ⋅
⎭⎦
9 3
1
6. (i) 2 aπ. (ii) a [13 √(13) − 8].
27
8. (i) 4a √3.
I-150 INTEGRAL CALCULUS
√ ⎧ ⎛ dx⎞ 2 ⎛ dy⎞ 2 ⎫
√ ⎧ ⎛ dx⎞ 2 ⎛ dy⎞
ds 2
⎫
= ⎨⎜ ⎟ + ⎜ ⎟ ⎬ or ds = ⎨⎜ ⎟ + ⎜ ⎟ ⎬ dt .
dt ⎩ ⎝ dt ⎠ ⎝ dt ⎠ ⎭ ⎩ ⎝ dt ⎠ ⎝ dt ⎠ ⎭
O n integrating between proper limits, the required length
√
t
⌠2 ⎧ ⎛ dx⎞ 2 ⎛ dy⎞ 2 ⎫
s= ⎮ ⎨ ⎜ ⎟ + ⎜ ⎟ ⎬ dt . (Meerut 2003)
⌡t ⎩ ⎝ dt ⎠ ⎝ dt ⎠ ⎭
1
Example 1 : Show that 8a is the length of an arch of the cycloid whose equations
are x = a (t − sin t ), y = a (1 − cos t ) . (Meerut 2006; Rohilkhand 08; Kashi 11;
Avadh 12; Purvanchal 14)
Solution : The given equations of the cycloid are
x = a (t − sin t ), y = a (1 − cos t ) .
We have dx ⁄ dt = a (1 − cos t ), and dy ⁄ dt = a sin t .
1 1
2 sin t cos t
dy dy ⁄ dt a sin t 2 2 1
∴ = = = = cot t .
dx dx ⁄ dt a (1 − cos t ) 2
2 sin t
1 2
2
N o w y = 0 wh e n cos t = 1 i.e., t = 0. A t
t = 0, x = 0, y = 0 and dy ⁄ dx = ∞ . Thus the curve
passes through the point (0, 0) and the tangent there
is perpendicular to the x-axis.
Again y is maximu m when cos t = − 1 i.e.,
t = π . When t = π , x = aπ , y = 2a , dy ⁄ dx = 0. Thus
at the point (aπ, 2a) the tangent to the curve is
parallel to the x-axis.
Also in this curve y cannot be negative. Thus an arch OBA of the given cycloid is
as shown in the figure. It is symmetrical about the line BM which is the axis of the cycloid.
⎛ ds⎞ 2 ⎛ dx⎞ 2 ⎛ dy⎞ 2
We have ⎜ ⎟ = ⎜ ⎟ + ⎜ ⎟
⎝ dt ⎠ ⎝ dt ⎠ ⎝ dt ⎠
1. (i) Find the whole length of the curve (astroid) x = a cos3 t , y = a sin3 t .
(Rohilkhand 2011)
(ii) Find the whole length of the curve (H ypocycloid) x = a cos3 t , y = b sin3 t .
2. R ectify the curve or find the length of an arch of the curve
x = a (t + sin t ), y = a (1 − cos t ). (Rohilkhand 2009B)
3. Prove that the length of an arc of the cycloid x = a (t + sin t ), y = a (1 − cos t )
from the vertex to the point (x, y) is √(8ay) . (Bundelkhand 2007; Meerut 12)
4. Find the length of the arc of the curve
1
x = e t sin t , y = e t cos t , from t = 0 to t = π.
2
(Kumaun 2008; Kanpur 09)
5. Show that in the epi-cycloid for which
x = (a + b) cos θ − b cos {(a + b) ⁄ b} θ,
y = (a + b) sin θ − b sin {(a + b) ⁄ b} θ,
the length of the arc measured from the point θ = πb ⁄ a is
{4b (a + b) ⁄ a} cos {(a ⁄ 2b) θ} .
6. In the ellipse x = a cos φ , y = b sin φ , show that ds = a √(1 − e2 cos2 φ ) dφ ,
and hence show that the whole length of the ellipse is
⎡ ⎛ 1 ⎞ 2 e2 ⎛ 1 . 3 ⎞ 2 e4 ⎛ 1 . 3 . 5 ⎞ 2 e6 ⎤
2 πa ⎢ 1 − ⎜ ⎟ ⋅ − ⎜ ⎟ ⋅ − ⎜ ⎟ ⋅ − …⎥ ,
⎣ ⎝ 2⎠ 1 ⎝ 2 . 4⎠ 3 ⎝ 2 . 4 . 6⎠ 5 ⎦
√
⌠ 2 ⎧⎪ ⎛ dr ⎞ ⎫⎪
s= ⎮ ⎨⎪ r 2 + ⎜ ⎟ ⎬⎪ dθ . (Meerut 2003)
⌡θ ⎩ ⎝ dθ⎠ ⎭
1
RECTIFICATION I-153
If the equation of the curve be θ = f (r) , then the required length is given by
2
√
r ⎧⎪ ⎫⎪
⌠2 ⎛ dθ⎞
s= ⎮ ⎨1 + ⎜ r ⎟
⎪⎩
⎬ dr .
⎪⎭
⌡r ⎝ dr ⎠
1
⎡ θ⎤ π ⎡ θ⎤ π
= 4a ⎢ − cos ⎥ = − 4a ⎢ cos ⎥ = − 4a (0 − 1) = 4a .
⎣ 2 ⎦0 ⎣ 2⎦ 0
∴ the perimeter of the cardioid = 2 × 4a = 8a .
Example 2 : Find the length of the arc of the equiangular spiral r = ae θ cot α
between the points for which radii vectors are r1 and r2 . (Kanpur 2007)
Solution : The given curve is r = ae θ cot α. ...(1)
Differentiating (1) w.r.t. θ, we get
dr ⁄ dθ = ae θ cot α . cot α = r cot α , from (1).
I-154 INTEGRAL CALCULUS
√ 1+ r
2
ds
=
⎧
⎨ 2 ⎛⎜ dθ⎞⎟ ⎫
⎬ (Note)
dr ⎩ ⎝ dr ⎠ ⎭
Example 3 : Prove that the perim eter of the lim acon r = a + b cos θ, if b ⁄ a be sm all,
1 2 2
is approximately 2 πa (1 + b ⁄a ).
4
Solution : The given curve is
r = a + b cos θ, (a > b) . ...(1)
Note that b ⁄ a is given to be small so
we must have b < a. Th e cu rve (1) is
symmetrical about the initial line and for
the portion of the curve lying above the
initial line θ varies from θ = 0 to θ = π .
By symmetry, the perimeter of the
limacon
= 2 × the arc length of the upper half of the limacon.
Now differentiating (1) w.r.t. θ, we have
dr ⁄ dθ = − b sin θ.
We have
2
⎛ ds ⎞ 2 ⎛ dr ⎞
⎜ ⎟ = r 2 + ⎜ ⎟ = (a + b cos θ) 2 + (− b sin θ) 2
⎝ dθ⎠ ⎝ dθ⎠
= a2 + b2 cos2 θ + 2ab cos θ + b2 sin2 θ = a2 + b2 + 2ab cos θ.
If we measure the arc length s in the direction of θ increasing,
we have ds ⁄ dθ = √(a2 + b2 + 2ab cos θ)
or ds = √(a2 + b2 + 2ab cos θ) dθ .
The arc length of the upper half of the limacon
π π 1⁄2
⌠ ⌠ ⎛ 2b b2 ⎞
= ⎮ √(a2 + b2 + 2ab cos θ) dθ = a ⎮ ⎜⎜ 1 + cos θ + 2 ⎟⎟ dθ
⌡0 ⌡0 ⎜⎝ a a ⎟⎠
⎡ 1 1 ⎤
π⎢ 2 ( − 1) ⎛ 2 ⎞ ⎥
⌠ ⎢ 1 + b cos θ + 1 ⋅ b + 2 2 ⎜ 4 b cos2 θ⎟ ⎥ dθ
= a⎮ ⎢ ⎜ ⎟ ⎥
⌡0 ⎣ a 2 a 2 2! ⎜ a
⎝
2 ⎟
⎠ ⎦
[E xpanding by binomial theorem and neglecting powers of b ⁄ a
higher than two because b ⁄ a is small]
π⎡ 2 ⎤
⌠ ⎢ 1 + b cos θ + 1 b (1 − cos2 θ) ⎥ dθ
= a⎮ ⎢ ⎥
⌡0 ⎢
⎣ a 2 a 2 ⎥
⎦
RECTIFICATION I-155
π⎡ 2 ⎤
⌠ ⎢ 1 + b cos θ + 1 b sin2 θ⎥ dθ
= a⎮ ⎢ ⎥
⌡0 ⎢
⎣ a 2 a 2 ⎥
⎦
π ⁄2
⎡⎧ b ⎫π 1 b2 ⌠ ⎤
= a ⎢⎢ ⎨ θ + sin θ⎬ + 2⎮ sin2 θ dθ⎥⎥
⎢⎩
⎣ a ⎭0 2 a2 ⌡0 ⎥
⎦
⎡ 1 b2 1 π⎤
= a ⎢⎢ π + ⋅ 2 ⋅ ⋅ ⎥⎥
⎢ 2a 2 2 2 ⎥⎦
⎣
⎡ 2
b ⎥ ⎤
= aπ ⎢⎢ 1 + 2 ⎥⎥
⋅
⎢ 4a
⎣ ⎦
∴ the perimeter of the limacon
= 2 × aπ [1 + (b2 ⁄ 4a2)] = 2aπ [1 + (b2 ⁄ 4a2)] .
1
Example 4 : If s be the length of the curve r = a tanh θ between the origin and
2
θ = 2π, and Δ be the area under the curve between the sam e two points, prove that
Δ = a (s − aπ) .
1
Solution : The given curve is r = a tanh θ. ...(1)
2
1 1
Differentiating (1) w.r.t. θ, we get dr ⁄ dθ = a ⋅ sech2 θ.
2 2
⎛ ds ⎞ 2 ⎛ dr ⎞ 2 a 2
1 1
We have ⎜ ⎟ = r 2 + ⎜ ⎟ = a 2 tanh2 θ + sech4 θ
⎝ dθ⎠ ⎝ dθ⎠ 2 4 2
1 1 1 1 1 1
= a2 [4 tanh2 θ + sech4 θ] = a2 [4 (1 − sech2 θ) + sech4 θ]
4 2 2 4 2 2
1 1
= a2 [2 − sech2 θ]2. ...(2)
4 2
If we measure the arc length s in the direction of θ increasing, we have
1 1
ds ⁄ dθ = a (2 − sech2 θ)
2 2
[R etaining + ive sign while taking the square root of (2)]
1 1
or ds = a (2 − sech2 θ) dθ .
2 2
Now at the origin r = 0 and putting r = 0 in (1), we get θ = 0.
∴ the arc length of the given curve between the origin (θ = 0) and θ = 2π is
given by
2π 2π 2π
1 ⌠ 1 1 ⌠ 1 ⌠ 1
s= a⎮ (2 − sech2 θ) dθ = a⎮ 2dθ − a⎮ sech2 θ dθ
2 ⌡0 2 2 ⌡0 2 ⌡0 2
2π ⎡ 2π
1 ⎤
a . 2 ⎡⎢⎣ θ⎤⎥⎦
1 1
= − a ⎢ 2 tanh θ⎥ = 2aπ − a tanh π . ...(3)
2 0 2 ⎣ 2 ⎦
0
Also the area between the radii vectors θ = 0, θ = 2π and the curve is
2π 2π
1⌠ 1 2⌠ 1
Δ= ⎮ r 2 dθ = a ⎮ tanh2 θ dθ
2 ⌡0 2 ⌡0 2
2π 2π
1 ⌠ 1 1 ⎡ 1 ⎤
= a2 ⎮ (1 − sech2 θ) dθ = a2 ⎢ θ − 2 tanh θ⎥
2 ⌡0 2 2 ⎣ 2 ⎦
0
I-156 INTEGRAL CALCULUS
1
= a2 [2π − 2 tanh π] = a2 [π − tanh π]
2
= a [aπ − a tanh π] = a [(2aπ − a tanh π) − aπ] = a (s − aπ), from (3).
The value of p should be put in terms of r from the equation of the curve.
Important Remark : If the curve is symmetrical about one or more lines, then
find out the length of one symmetrical part and then multiply it by the number of
symmetrical parts.
⌠ r dr
Example 1 : Prove the form ula s = ⎮ ⋅
⌡ √(r 2 − p2)
Show that the arc of the curve p2 (a4 + r4) = a4 r 2 between the limits r = b, r = c is
equal in length to the arc of the hyperbola xy = a2 between the lim its x = b, x = c .
Solution : From differential calculus, we know that
tan φ = r
dθ
dr
and
ds
dr
=
√ 1+ ⎡
⎢
⎣
⎛ dθ⎞ 2 ⎤
⎜r ⎟ ⎥ ⋅
⎝ dr ⎠ ⎦
ds
∴ = √(1 + tan2 φ ) = √(sec 2 φ ) = sec φ
dr
1 1 1
= = = [ ... p = r sin φ ]
cos φ √(1 − sin2 φ ) √{1 − ( p2 ⁄ r 2)}
r
= ⋅
√(r − p2)
2
r
Thus ds = dr .
√(r − p2)
2
⌠ r
Integrating between the given limits, we get s = ⎮ dr . ...(1)
⌡ √(r 2 − p2)
Now the given curve is p2 (a4 + r 4) = a4 r 2 or p2 = a4 r 2 ⁄ (a4 + r 4) .
a4 r 2 r6
We have r 2 − p2 = r 2 − = ⋅ ...(2)
(a4
+ r ) (a + r 4)
4 4
Therefore from (1), the arc of the given curve between the limits r = b, r = c is
RECTIFICATION I-157
c c
⌠ r dr ⌠ r dr
= ⎮ = ⎮ [From (2)]
⌡b √(r 2 − p2) ⌡b √{r 6 ⁄ (a4 + r 4)}
c c
⌠ r √(a4 + r 4) ⌠ √(a4 + r 4)
= ⎮ dr = ⎮ dr . ...(3)
⌡b r3 ⌡b r2
Also, for the hyperbola xy = a2 i.e., y = a2 ⁄ x, dy ⁄ dx = − a2 ⁄ x2.
∴ the arc length of the hyperbola xy = a2 between the limits x = b, x = c
c ⎧⎪ 2 ⎫⎪ c
√ √ 1 + ax
⌠ ⎛ dy⎞ ⌠ ⎧ 4⎫
= ⎮ ⎨⎪ 1 + ⎜ ⎟ ⎬⎪ dx = ⎮ ⎪ ⎪ dx
⎨ 4⎬
⌡b ⎩ ⎝ dx⎠ ⎭ b
⌡ ⎪
⎩
⎪
⎭
c c
⌠ √(x4 + a4) 4
⌠ √(r + a 4)
= ⎮ dx = ⎮ dr
⌡b x2 ⌡b r2
[Changing the variable from x to r by a property of definite integrals]
c
⌠ √(a4 + r 4)
= ⎮ dr . ...(4)
⌡b r2
From (3) and (4) we observe that the two lengths are equal.
1. 8a. 5. 4a √3.
6. a sec α [e θ cot α − 1].
1
7. f (θ2) − f (θ1), where f (θ) = a √(sec 2 θ + 3) − a √3 log {cos θ + √(cos2 θ + )}.
3
I-158 INTEGRAL CALCULUS
E liminating θ and φ between (1), (2) and (3), we get a relation between s and ψ,
which is the intrinsic equation of the curve.
(c) Intrinsic equation from Pedal Equation :
Let the pedal equation of the curve be p = f (r) . ...(1)
r
⌠ r dr ,
Then s= ⎮ ...(2)
⌡a √(r2 − p2)
the arc length s being measured from the point r = a .
ds dr
Also the radius of curvature ρ = = r ⋅ …(3)
dψ dp
E liminating p and r between (1), (2) and (3), we obtain the required intrinsic
equation.
√ ⎛ dx⎞ 2 ⎫
√
ds ⎧ 2 ⎫
=
⎧
⎨1 + ⎜ ⎟ ⎬ = ⎪1 + y ⎪ ⎡ . . dx
⎢ . =
y⎤
⎥
⎨ ⎬
dy ⎩ ⎝ dy⎠ ⎭ ⎪ 4a2 ⎪⎭ ⎣ dy 2a ⎦
⎩
√
⎧ 4a 2 + y2 ⎫
= ⎪ ⎪ = 1 √(4a 2 + y2) .
⎨ ⎬
⎪ 4a 2 ⎪ 2a
⎩ ⎭
1
∴ ds = √(4a2 + y2) dy .
2a
s y
⌠ 1 ⌠
Integrating, ⎮ ds = ⎮ √(4a2 + y2) dy
⌡0 2a ⌡0
1 ⎡1 1 ⎤y
or s= ⎢ y √(4a 2 + y2) + ⋅ 4a 2 log { y + √(4a 2 + y2)}⎥
2a ⎣ 2 2 ⎦0
1 1
= (1 ⁄ 2a) [ y √(4a2 + y2) + ⋅ 4a2 log { y + √(4a2 + y2)}
2 2
1
− ⋅ 4a2 log 2a]
2
1 ⎡ y + √(4a2 + y2) ⎤
= ⎢ y √(4a 2 + y2) + 4a 2 log ⎥⋅ ...(3)
4a ⎣ 2a ⎦
Now to obtain the intrinsic equation of the given parabola we eliminate y between
(2) and (3). From (2), we have y = 2a cot ψ . Putting this value of y in (3), we get
I-160 INTEGRAL CALCULUS
1 ⎡
⎢
s= ⎢ 2a cot ψ √(4a 2 + 4a 2 cot2 ψ)
4a ⎢
⎣
2a cot ψ + √(4a2 + 4a2 cot 2 ψ) ⎤
+ 4a2 log ⎥
2a ⎦
1
= [(2a cot ψ) . 2a √(1 + cot2 ψ) + 4a2 log {cot ψ + √(1 + cot 2 ψ)}]
4a
= a cot ψ cosec ψ + a log (cot ψ + cosec ψ),
which is the required intrinsic equation.
Example 2 : Show that the intrinsic equation of the cycloid
x = a (t + sin t ), y = a (1 − cos t )
is s = 4a sin ψ .
Hence or otherwise find the length of the com plete cycloid.
(Meerut 2001, 06B, 07, 10; Kanpur 04; Avadh 04, 09, 10; Rohilkhand 07B)
Solution : The given equations of the cycloid are
x = a (t + sin t ), y = a (1 − cos t ) . ...(1)
We have dx ⁄ dt = a (1 + cos t ), and dy ⁄ dt = a sin t .
1 1
2 sin t cos t
dy dy ⁄ dt a sin t 2 2 1
∴ = = = = tan t .
dx dx ⁄ dt a (1 + cos t ) 2
2 cos t
1 2
2
1 1
H ence tan ψ = dy ⁄ dx = tan t or ψ= t. ...(2)
2 2
If s denotes the arc length of the cycloid measured from the vertex (i.e., the point
t = 0) to any point P (i.e., the point ‘t ’) in the direction of t increasing, then
t
⌠
s= ⎮
⌡0 √ ⎧ ⎛ dx⎞ 2 ⎛ dy⎞ 2 ⎫
⎨ ⎜ ⎟ + ⎜ ⎟ ⎬ dt
⎩ ⎝ dt ⎠ ⎝ dt ⎠ ⎭
t
⌠
= ⎮ √{a2 (1 + cos t ) 2 + a2 sin2 t } dt
⌡0
t
⌠
= ⎮ √{2a2 (1 + cos t )} dt
⌡0
t t
⌠ 1 ⎡ 1 ⎤ 1
= 2a ⎮ cos t dt = 2a ⎢ 2 sin t⎥ = 4a sin t ...(3)
⌡0 2 ⎣ 2 ⎦
0
2
⎡ θ
1 ⎤ 1
= 2a ⎢ 2 sin θ⎥ = 4a sin θ. ...(3)
⎣ 2 ⎦ 2
0
1 1
E liminating θ between (2) and (3), we get s = 4a sin { (ψ − π)} , ...(4)
3 2
which is the required intrinsic equation.
ds 4a 1 1
Also ρ= = cos (ψ − π), from (4)
dψ 3 3 2
1 1
or 3ρ = 4a cos (ψ − π) . ...(5)
3 2
Squaring and adding (4) and (5), we get
1 1 1 1
s2 + 9ρ2 = (4a) 2 {sin2 (ψ − π) + cos2 (ψ − π)}
3 2 3 2
= 16a2 . 1 = 16a2.
Example 4 : Find the intrinsic equation of the equiangular spiral p = r sin α .
(Meerut 2000, 01, 04, 06, 09, 10B)
Solution: The given pedal equation of the curve is p = r sin α . s ...(1)
Differentiating (1) w.r.t. r, we have dp ⁄ dr = sin α .
I-162 INTEGRAL CALCULUS
ds dr r r
∴ ρ= = r = = = r cosec α . ...(2)
dψ dp dp ⁄ dr sin α
If we measure the arc length s from the point r = 0 in the direction of r increasing,
we have
r r r
⌠ r dr ⌠ r dr ⌠
s= ⎮ = ⎮ = ⎮ sec α dr
⌡0 √(r2 − p2) ⌡0 √(r2 − r2 sin2 α) ⌡0
r
⌠ r
= sec α ⎮ dr = sec α ⎡⎢⎣ r ⎤⎥⎦ = r sec α . ...(3)
⌡0 0
1 dψ ⎡ . . dx ⎤
or = 4a cos ψ ⎢ . = cos ψ⎥
cos ψ dx ⎣ ds ⎦
7. Find the intrinsic equation of r = a eθ cot α, where s is measured from the point
(a, 0).
8. Find the intrinsic equation of the spiral r = aθ, the arc being measured from the
pole.
9. Find the intrinsic equation of the curve p2 = r2 − a2.
10. In the four-cusped astroid x2 ⁄ 3 + y2 ⁄ 3 = a2 ⁄ 3, show that
3
(i) s = a cos 2 ψ, s being measured from the vertex;
4
3
(ii) s = a sin2 ψ , s being mesured from the cusp on x-axis;
2 (Purvanchal 2014)
(iii) whole length of the curve is 6a .
11. Find the cartesian equation of the curve whose intrinsic equation is s = c tan ψ
when it is given that at ψ = 0, x = 0 and y = c .
I-164 INTEGRAL CALCULUS
⎛ ds ⎞ 2
9. If x = a cos3 t, y = a sin 3 t, then ⎜ ⎟ is
⎝ dt ⎠
(a) (a sin t cos t) 2 (b) (sin t cos t) 2
(c) (3 a sin t cos t) 2 (d) 3a sin t cos t
10. The entire length of the cardioid r = a (1 + cos θ) is
(a) 8 a (b) 4 a
(c) 6 a (d) 2 a (Rohilkhand 2008)
True or False:
Write ‘T’ for true and ‘F’ for false statement.
11. The length of the arc of the curve x = f ( y) between y = a and y = b, (b > a) is
equal to
b ⎧ ⎛ ⎞2⎫
⌠
⎮
⌡a √ ⎨ 1 + ⎜ dx⎟ ⎬ dy.
⎩ ⎝ dy⎠ ⎭
12. The relation between s and ψ for any curve is called its polar equation.
13. If the equation of the curve be θ = f (r ), then the arc length from r = r1 to r = r2
is given by
r ⎧ ⎛ ⎞2⎫
⌠2
⎮
⌡r
1
√ ⎨ 1 + ⎜ r d θ ⎟ ⎬ d r.
⎩ ⎝ dr⎠ ⎭
14. If the equation of the curve be r = f (θ), then the arc length from θ = θ1 to
θ = θ2 is given by
θ ⎡ ⎛ ⎞2⎤
⌠ 2
⎮
⌡θ
1
√ ⎢ r2 + ⎜ dr ⎟ ⎥ dr.
⎣ ⎝ dθ ⎠ ⎦ (Meerut 2003)
I-166 INTEGRAL CALCULUS
b ⎧ ⎛ ⎞2⎫
1. R ectification. 2.
⌠
⎮
⌡a √ ⎨ 1 + ⎜ dy⎟ ⎬ dx .
⎩ ⎝ dx⎠ ⎭
3.
3 1
+ log 2 .
2 4
⎧ 2⎫
⎪ ⎛ d θ⎞⎟ ⎪
4. r cosec α d θ . 5. ⎨1 + ⎜r ⎬⋅ 6. 8a.
⎪ ⎝ ⎪
⎩ d r⎠ ⎭
7. (a). 8. (b). 9. (c).
10. (a). 11. T. 12. F.
13. T. 14. F.
8.1 Revolution : Definitions
Solid of Revolution : If a plane area is revolved about a fixed line lying in its own
plane, then the body so generated by the revolution of the plane area is called a solid
of revolution.
Surface of Revolution : If a plane curve is revolved about a fixed line lying in its
own plane, then the surface generated by the perimeter of the curve is called a surface
of revolution.
Axis of Revolution : The fixed straight line, say A B, about which the area
revolves is called the axis of revolution or axis of rotation.
Proof : Let A B be the arc of the curve y = f (x) included between the ordinates
x = a and x = b. It is being assumed that the curve does not cut the x-axis and f (x) is a
continuous function of x in the interval (a, b) .
Let P (x, y) and Q (x + δx, y + δy) be any two
neighbouring points on the curve y = f (x). Draw
the ordinates PM and QN. Also draw PP′ and QQ′
perpendiculars to these ordinates.
L e t V d e n o t e t h e vo lu m e o f t h e so l id
generated by the revolution of the area A CMP
about the x-axis and let the volume of revolution
obtained by revolving the area A CNQ about x-axis
be V + δV, so that volume of the solid generated by
the revolution of the strip PMNQ about the x-axis
is δV.
Now PM = y, QN = y + δy and MN = (x + δx) − x = δx . Then the volume of the
solid generated by revolving the area PMN P′ = πy2 δx and the volume of the solid
generated by revolving the area Q′MN Q = π ( y + δy) 2 δx .
Also the volume of the solid generated by the revolution of the area PMNQP
(i.e., the volume δV ) lies between the volumes of the right circular cylinders generated
b y t h e r e vo l u t i o n o f t h e a r e a s PMN P′P a n d MN QQ′ i.e., δV l ie s be t we e n
π y2 δx and π ( y + δy) 2 δx
or (δV ⁄ δx) lies between π y2 and π ( y + δy) 2
i.e., π y2 < (δV ⁄ δx) < π ( y + δy) 2.
In the limiting position as Q → P, δx → 0 (and therefore δy → 0), we have
dV ⁄ dx = π y2 or dV = π y2 dx .
b b
⌠ ⌠ ⎡ ⎤x= b
H ence ⎮ π y2 dx = ⎮ dV = ⎢⎣ V ⎥⎦
⌡a ⌡a x= a
(ii) If the curve is symmetrical about x-axis and it is required to find the volume
generated by the revolution of the area about y-axis, then the volume generated will
be twice the volume generated by half of the symmetrical portion of the curve.
4
Example 1 : Show that the volum e of a sphere of radius a is πa3.
3
(Bundelkhand 2010; Avadh 10)
Example 3 : Find the volume of the solid generated by the revolution of the cissoid
y2 (2a − x) = x3 about its asymptote. (Meerut 2007; Kanpur 14)
Solution : T h e gi ve n c u r ve i s
y2 (2a − x) = x3. Its shape is as shown in the
figure. E quating to zero t he coefficient of
highest power of y, the asymptote parallel to
the axis of y is x = 2a . Take an elementary strip
PMNQ perpendicular to the asymptote x = 2a
where P is the point (x, y) and Q is the point
(x + δx, y + δy) .
We have PM = 2a − x and MN = δy .
N ow vo lume of t h e ele ment ary disc
formed by revolving the strip PMN Q about the
line x = 2a is
= π . PM 2. MN = π (2a − x) 2 δy.
The given curve is symmetrical about
x-axis and for the portion of the curve above x-axis y varies from 0 to ∞ .
∞
⌠
∴ the required volume = 2 ⎮ π (2a − x) 2 dy. ...(1)
⌡y = 0
From the given equation of the curve y2 (2a − x) = x3 we observe that the value
of x cannot be easily found in terms of y. H ence for the sake of integration we change
the independent variable from y to x . (Note)
x3 ;
The curve is y2 =
2a − x
dy (2a − x) . 3x2 − x3 (− 1) 2 (3a − x) x2
∴ 2y = =
dx (2a − x) 2 (2a − x) 2
(3a − x) x2 √(2a − x) (3a − x) √x √(2a − x)
or dy = ⋅ dx = dx .
(2a − x) 2 x √x (2a − x) 2
Also when y = 0, x = 0 and when y → ∞ , x → 2a .
H ence from (1), the required volume
2a
⌠ ⎡ (3a − x) √x √(2a − x) ⎤
= 2π ⎮ (2a − x) 2 ⎢ ⎥ dx
⌡x = 0 ⎣ (2a − x) 2 ⎦
VOLUMES AND SURFACES OF SOLIDS OF REVOLUTION I-171
2a
⌠
= 2π ⎮ (3a − x) √x √(2a − x) dx .
⌡0
Now put x = 2a sin 2 θ so that dx = 4a sin θ cos θ dθ. When x = 0, θ = 0 and when
x = 2a, θ = π ⁄ 2 . Therefore the required volume
π ⁄2
⌠
= 2π ⎮ (3a − 2a sin2 θ) √(2a) sin θ √[2a (1 − sin2 θ)] × 4a sin θ cos θ dθ
⌡0
π ⁄2
⌠
= 16πa3 ⎮ (3 sin2 θ cos2 θ − 2 sin4 θ cos2 θ) dθ
⌡0
⎡ 3 3 5 3 ⎤
⎢ 3Γ ( ) Γ ( ) 2 Γ ( ) Γ ( )⎥
16πa3 ⎢
2 2 2 2 ⎥
= ⎢ − ⎥
⎣ 2 Γ (3) 2 Γ (4) ⎦
⎡ 1 1 3 1 1 ⎤
⎢ 3 ⋅ ⋅ √π ⋅ ⋅ √π 2 ⋅ ⋅ ⋅ √π ⋅ ⋅ √π ⎥
16πa3 ⎢ ⎥
2 2 2 2 2
= ⎢ − ⎥
⎣ 2.2.1 2.3.2.1 ⎦
⎡ 3π π⎤
= 16πa3 ⎢ − ⎥ = 2π 2a 3.
⎣ 16 16⎦
Note : If the given curve is y2 (a − x) = x3, then the required volume can be
1
obtained by putting a for 2a in the above E xercise. The volume so obtained is π 2a3.
4
Important Remark : When we are to revolve an area about a line which is neither
the x-axis nor the y-axis we m ust take an elem entary strip which is perpendicular to the line
of revolution as explained in the above exam ple.
Example 4 : The area between a parabola and its latus rectum revolves about the
directrix. Find the ratio of the volum e of the ring thus obtained to the volum e of the sphere
whose diam eter is the latus rectum .
Solution : Let the parabola be y2 = 4ax . Then
the directrix is the line x = − a . Let L L ′ be the latus
re ct um. The area L OL′SL is revo lved abo u t t h e
directrix. The volume of the ring thus obtained = the
volume V 1 of the cylinder formed by the revolution of
t h e r e ct a n gle L L′R′R a bo u t t h e di r e ct r ix − the
volume V 2 of the reel formed by the revolution of the
arc L OL′ about the directrix.
N o w t h e vo l u m e V 1 o f t h e c yl i n d e r
= π r2 h = π (L R) 2 . L L′ = π (2a) 2. 4a = 16πa3.
To find the volume V 2 of the reel consider an
e l e m e n t a r y s t r i p PMN Q wh e r e P (x, y) a n d
Q (x + δx, y + δy) are two neighbouring points on the
arc OL and PM, QN are perpendiculars from P and Q
on the directrix.
We have PM = a + x and MN = δy.
∴the volume V 2 of the reel
2a
⌠
= 2⎮ π (a + x) 2 dy [By symmetry about x-axis]
⌡0
I-172 INTEGRAL CALCULUS
2a 2a ⎛ 2 4 ⎞
⌠ ⌠ ⎜ a 2 + 2a ⋅ y + y ⎟ dy
= 2⎮ π (a2 + 2ax + x2) dy = 2π ⎮ ⎜ ⎟
⌡0 ⌡0 ⎜
⎝ 4a 16a2 ⎟⎠
[ ... x = y2 ⁄ 4a]
2a
⎡ 1 y3 1 y5 ⎤
= 2π ⎢⎢ a2y + + ⋅ ⎥⎥
⎢ 2 3 16a 2 5 ⎥⎦
⎣ 0
⎡ 3 4 3 2 3⎤ 56 112 πa3
= 2π ⎢ 2a + a + a ⎥ = 2πa3 ⋅ = ⋅
⎣ 3 5 ⎦ 15 15
∴ Volume of the ring = volume of the cylinder − volume of the reel
112 3 128 3
= V 1 − V 2 = 16πa3 − πa = πa .
15 15
Volume of the sphere whose diameter is the latus rectum 4a i.e., the radius is 2 a
4 4 32
= π r3 = π (2a) 3 = πa3.
3 3 3
3
128πa ⁄ 15 4
∴ the required ratio = = ⋅
32πa3 ⁄ 3 5
7. Show that the volume of the solid generated by the revolution of the upper half
4
of the loop of the curve y2 = x2 (2 − x) about x-axis is π . (Meerut 2005)
3
8. The area of the curve x2 ⁄ 3 + y2 ⁄ 3 = a2 ⁄ 3 lying in the first quadrant revolves about
x-axis. Find the volume of the solid generated. (Agra 2014)
9. Find the volume of the solid obtained by revolving the loop of the curve
a2 y2 = x2 (2a − x) (x − a) about x-axis.
10. A basin is formed by the revolution of the curve x3 = 64y, ( y > 0) about the axis
of y. If the depth of the basin is 8 inches, how many cubic inches of water it will
hold ?
11. Show that the volume of the solid generated by the revolution of the curve
1 2 3
(a − x) y2 = a2 x , about its asymptote is π a .
2
(Meerut 2004B, 06B; Kumaun 08; Rohilkhand 12)
12. The figure bounded by a quadrant of a circle of radius a and tangents at its
extremities revolves about one of the tangents. Prove that the volume of the solid
5 1
generated is ( − π) πa3.
3 2
13. The area cut off from the parabola y2 = 4ax by the chord joining the vertex to an
end of the latus rectum is rotated through four right angles about the chord. Find
the volume of the solid generated. (Rohilkhand 2008; Bundelkhand 09)
⎡ ⎤
2 1
1. (i) πa3. (ii) πh2 ⎢⎣ a − h⎥⎦ ⋅
3 3
4 πc 2 ⎡⎢ c 2 x⎤ π 2 a3
2. (ii) πa3. 4. (i) x + sinh ⎥⎦ ⋅ (ii) ⋅
5 2 ⎣ 2 c 2
⎡
2 ⎤⎥ 1 16
6. (i) 2 a3π ⎢⎣ log 2 − ⋅ (ii) πa4. 8. πa3.
3⎦ 12 105
23 3 1536 2
9. πa . 10. π cubic inches. 13. √5 πa3.
60 5 75
b b
⌠ dy ⌠
= ⎮ π x2 dt = π ⎮ { φ (t)} 2 . ψ′ (t) dt .
⌡a dt ⌡a
Example 1 : Find the volum e of the solid form ed by revolving the cycloid
We have PM = y and MN = δx .
Now the volume of the elementary disc formed by revolving the strip PMN Q about
the base (i.e., the x-axis) is π PM 2 . MN = π y2 δx .
Now the cycloid is symmetrical about the line BH .
∴ the required volume
⌠
= 2 ⎮ π y2 dx , the limits of integration being extended from O to B
⌡
π π
⌠ dx ⌠
= 2π ⎮ y2 dθ = 2π ⎮ a2 (1 − cos θ) 2 a (1 − cos θ) dθ [From (1)]
⌡θ = 0 dθ ⌡0
π
⌠
= 2π ⎮ a3 (1 − cos θ) 3 dθ
⌡0
π π
⌠ θ⎞ 3 ⌠ θ
= 2π a3 ⎮ ⎛⎜ 2 sin2 ⎟ dθ = 16π a 3 ⎮ sin6 dθ
⌡0 ⎝ 2⎠ ⌡0 2
π ⁄2
⌠ θ
= 32πa3 ⎮ sin6 φ dφ , putting = φ so that dθ = 2 dφ
⌡0 2
5 3 1 1
= 32π a3 ⋅ ⋅ ⋅ ⋅ π = 5π 2a3.
6 4 2 2
(ii) When the curve revolves about y-axis, the required volume of the solid
generated
VOLUMES AND SURFACES OF SOLIDS OF REVOLUTION I-175
= the volume generated by the revolution of the area OA BDO about y-axis
− the volume generated by the revolution of the area OBDO about the y-axis.
...(2)
Also at A, θ = 2π ; at B, θ = π and at O, θ = 0.
Now the area OA BD is bounded by the arc A B of the cycloid and the axis of y .
Therefore volume of the solid generated by the revolution of the area OA BDO about
y -axis
π π
⌠ ⌠ dy
= ⎮ π x2 dy = ⎮ π x2 dθ
⌡θ = 2π ⌡θ = 2π dθ
π
⌠
= π ⎮ a2 (θ − sin θ) 2 a sin θ dθ [From (1)]
⌡2π
π
⌠
= π ⎮ a2 (θ2 − 2θ sin θ + sin2 θ) a sin θ dθ
⌡2π
π
⌠
= πa3 ⎮ (θ2 sin θ − 2θ sin2 θ + sin3 θ) dθ
⌡2π
π
⌠ 1
= πa3 ⎮ [θ2 sin θ − θ (1 − cos 2θ) + (3 sin θ − sin 3θ)] dθ (Note)
⌡2π 4
⎡ 1 2 1
= πa3 ⎢ θ2 . (− cos θ) − 2θ (− sin θ) + 2 cos θ − θ + θ ( sin 2θ)
⎣ 2 2
1 3 1 ⎤π
− 1 (− cos 2θ) − cos θ + cos 3θ⎥ ,
4 4 12 ⎦ 2π
⌠ ⌠
the values of the integrals ⎮ θ2 sin θ dθ and ⎮ θ cos 2θ dθe
⌡ ⌡
have been written after applying integration by parts
1 1 3 1 1 3 1
= πa3 [(π 2 − 2− π2 + + − ) − (− 4π 2 + 2 − 2π 2 + − + )]
2 4 4 12 4 4 12
13 8
= πa3 [ π2 − ]⋅ ...(3)
2 3
Again volume of the solid generated by the revolution of the area OBDO about
y-axis
π π π
⌠ ⌠ dy ⌠
= ⎮ π x2 dy = ⎮ π x2 dθ= π ⎮ a2 (θ − sin θ) 2 . a sin θ dθ
⌡θ = 0 ⌡θ = 0 dθ ⌡0
π
⌠
= πa3 ⎮ (θ2 − 2θ sin θ + sin2 θ) sin θ dθ
⌡0
π
⌠
= πa3 ⎮ (θ2 sin θ − 2θ sin2 θ + sin3 θ) dθ
⌡0
π
⌠ 1
= πa3 ⎮ [θ2 sin θ − θ (1 − cos 2θ) + (3 sin θ − sin 3θ)] dθ
⌡0 4
⎡ 1 2 1
= πa3 ⎢ θ2 (− cos θ) − 2θ (− sin θ) + 2 cos θ − θ + θ ( sin 2θ)
⎣ 2 2
1 3 1 ⎤π
− 1 (− cos 2θ) − cos θ + cos 3θ⎥
4 4 12 ⎦0
I-176 INTEGRAL CALCULUS
1 1 3 1 1 3 1
= πa3 [(π 2 − 2 − π2 + + − ) − (2 + − + )]
2 4 4 12 4 4 12
1 8
= πa3 ( π 2 − ). ...(4)
2 3
∴ from (2), the required volume = (3) − (4)
13 8 1 8
= πa3 [ π2 − ] − πa3 [ π 2 − ] = π a3 [6π 2] = 6π 3a3.
2 3 2 3
Example 2 : Find the volum e of the solid generated by the revolution of the tractrix
1
x = a cos t + a log tan2 (t ⁄ 2), y = a sin t about its asymptote.
2
(Meerut 2000, 05B; Rohilkhand 06; Avadh 09, 11; Kashi 12; Purvanchal 14)
Solution : The given curve is
1
x = a cos t + a log tan2 (t ⁄ 2), y = a sin t . ...(1)
2
dx 1 1 1
∴ = − a sin t + a ⋅ 2
⋅ 2 tan (t ⁄ 2) sec2 (t ⁄ 2) ⋅
dt 2 tan (t ⁄ 2) 2
a a
= − a sin t + = − a sin t +
2 sin (t ⁄ 2) cos (t ⁄ 2) sin t
(1 − sin2 t) cos2 t
= a = a ...(2)
sin t sin t
N o w t h e g i ve n c u r ve i s
symmetrical about both the axes and the
asymptote is the line y = 0 i.e., x-axis.
For the portion of the curve lying
in the second quadrant y varies from a to
0, t varies from π ⁄ 2 to 0 and x varies from
0 to − ∞ .
∴ the required volume
0
⌠
= 2 ⎮ π y2 dx
⌡− ∞
π ⁄2
⌠ dx
= 2⎮ π y2 ⋅ dt
⌡0 dt
π ⁄2
⌠ a cos2 t
= 2π ⎮ a2 sin2 t . dt [From (1) and (2)]
⌡0 sin t
π ⁄2
⌠ 1 2
= 2πa3 ⎮ cos2 t sin t dt = 2πa3 = πa3.
⌡0 3.1 3
1. Find the volume of the solid generated by the revolution of the cycloid
x = a (θ + sin θ), y = a (1 − cos θ), − π ≤ θ ≤ π ,
(i) about the x-axis, (ii) about the base.
VOLUMES AND SURFACES OF SOLIDS OF REVOLUTION I-177
2. Show that the volume of the solid generated by the revolution of the cycloid
x = a (θ + sin θ), y = a (1 − cos θ), 0 ≤ θ ≤ π ,
3 8
about the y-axis is πa3 ( π 2 − ) ⋅
2 3
3. Prove that the volume of the reel formed by the revolution of the cycloid
x = a (θ + sin θ), y = a (1 − cos θ)
about the tangent at the vertex is π 2a3.
4. Prove that the volume of the solid generated by the revolution about the x-axis
1 3
of the loop of the curve x = t2, y = t − t3 is π.
3 4
5. Find the volume of the spindle shaped solid generated by revolving the astroid
x2 ⁄ 3 + y2 ⁄ 3 = a2 ⁄ 3 about the x-axis.
6. Find the volume of the solid generated by the revolution of the cissoid
x = 2a sin2 t , y = 2 a sin3 t ⁄ cos t about its asymptote.
(Kanpur 2006; Bundelkhand 14)
32
1. (i) π 2 a3, (ii) 5π 2 a3. 5. πa3. 6. 2 π 2 a3.
105
8.5 Volume of the Solid Generated by the Revolution when The Axis
of Rotation being any Line
If, however, the axis of rotation is neither x-axis nor y-axis, but is any other line
CD, then the volum e of the solid generated by the revolution about CD of the area bounded
by the curve A B, the axis CD and the perpendiculars A C, BD on the aixs is
OD
⌠
⎮ π (PM) 2 d (OM),
⌡OC
where PM is the perpendicular drawn from any point P on the curve to the axis of rotation
and O is som e fixed point on the axis of rotation.
Example 1 : The cardioid r = a (1 + cos θ) revolves about the initial line. Find the
volum e of the solid thus generated.
(Meerut 2001, 03, 07B; Agra 06, 07, 08; Rohilkhand 13, 13B)
Solution : The given curve is r = a (1 + cos θ) . ...(1)
It is symmetrical about the initial line. We have r = 0 when cos θ = − 1 i.e.,
θ= π.
Also r is maximum when cos θ = 1 i.e., θ = 0 and then r = 2a . As θ increases from
0 to π, r decreases from 2a to 0. H ence the
sh ape o f the curve is as shown in the
figure. For the upper half of the curve, θ
varies from 0 to π .
∴ the required volume
π
2⌠
= ⎮ π r 3 sin θ dθ
3 ⌡0
π
⌠
2π
= ⎮ a3 (1 + cos θ) 3 sin θ dθ
⌡0
3
[From (1)]
π
2 ⌠
= − πa3 ⎮ (1 + cos θ) 3 (− sin θ) dθ (Note)
3 ⌡0
⎡ (1 + cos θ) 4 ⎤ π
= −
2
πa3 ⎢ ⎥ , using power formula
3 ⎣ 4 ⎦0
⌠ [ f (x)]n+ 1
i.e., ⎮ [ f (x)]n f ′ (x) dx =
⌡ n+ 1
1 8
= − πa3 (0 − 24) = πa3.
6 3
VOLUMES AND SURFACES OF SOLIDS OF REVOLUTION I-179
⌠
π ⎡ r 3 ⎤⎥ a (1 + cos θ) 2π ⌠
π
= ⎮ 2π ⎢⎢⎢ ⎥⎥ sin θ dθ = ⎮ a3 (1 + cos θ) 3 sin θ dθ
⌡0 ⎣ 3 ⎦0 3 ⌡0
π π
2πa3 ⌠ 2πa3 ⎡ (1 + cos θ) 4 ⎤
= − ⎮ (1 + cos θ) 3 (− sin θ) dθ = − ⎢ ⎥
3 ⌡0 3 ⎣ 4 ⎦0
2πa3 1 2 1 8
= − ⋅ [0 − 24] = ⋅ πa3 ⋅ ⋅ 16 = πa3.
3 4 3 4 3
Example 2 : Find the volum e of the solid form ed by revolving one loop of the curve
r 2 = a2 cos 2θ about the initial line. (Rohilkhand 2007B)
Solution : For the upper half of the loop θ varies from 0 to π ⁄ 4 . H ere the curve
is revolving about the initial line (i.e., x-axis).
π ⁄4
2 ⌠
∴ the required volume = π⎮ r 3 sin θ dθ
3 ⌡0
π ⁄4
2π ⌠
= ⎮ {a √(cos 2θ)}3 sin θ dθ [ ... r 2 = a2 cos 2θ]
3 ⌡0
π ⁄4
2πa3 ⌠
= ⎮ (2 cos2 θ − 1) 3 ⁄ 2 sin θ dθ. (Note)
3 ⌡0
Put √2 cos θ = sec φ so that − √2 sin θ dθ = sec φ tan φ dφ .
When θ = 0, φ = π ⁄ 4 and when θ = π ⁄ 4, φ = 0.
∴ the required volume
0
2πa3 ⌠ (− sec φ tan φ )
= ⎮ (sec 2 φ − 1) 3 ⁄ 2 dφ
3 ⌡π ⁄ 4 √2
π ⁄4 π ⁄4
√2πa3 ⌠ √2πa3 ⌠
= ⎮ tan4 φ sec φ dφ = ⎮ (sec 2 φ − 1) 2 sec φ dφ
3 ⌡0 3 ⌡0
π ⁄4
√2πa3 ⌠
= ⎮ (sec 5 φ − 2 sec 3 φ + sec φ ) dφ . ...(1)
3 ⌡0
I-180 INTEGRAL CALCULUS
π⎡ 2 3 x 1
= ⎢− x − a2 x + 2 √2a ⋅ √(x2 + a2)
2⎣ 3 2 8
1 2 1 ⎤a
+ 2√2a ⋅ a log {x + √(x2 + a2)}⎥
16 8 ⎦0
π⎡ 2 3 a 3a
= ⎢− a − a3 + 2 √2a ⋅ ⋅
2⎣ 3 2 2 √2
1 ⎧ ⎛ 3a ⎞ a ⎫⎤
+ √2a3 ⎨ log ⎜ a + ⎟ − log ⎬⎥
8 ⎩ ⎝ 2 √2 ⎠ 2 √2 ⎭ ⎦
π⎡ 5 3 1 ⎧ a (2 √2 + 3) 2 √2 ⎫ ⎤
= ⎢ − a3 + a3 + √2a3 log ⎨ ⋅ ⎬⎥
2⎣ 3 2 8 ⎩ 2 √2 a ⎭⎦
π⎡ 1 1 ⎤
= ⎢ − a3 + √2a3 log (2 √2 + 3) ⎥
2⎣ 6 8 ⎦
π⎡ 1 3 1 ⎤
= ⎢ − a + √2a3 log (√2 + 1) 2⎥
2⎣ 6 8 ⎦
πa3 ⎡ 1 1⎤ πa3 ⎡ 1 1⎤
= ⎢2 ⋅ √2 log (√2 + 1) − ⎥ = ⎢ √2 log (√2 + 1) − ⎥
2 ⎣ 8 6⎦ 2 ⎣4 6⎦
πa3 πa3 √2
= ⎡ 3√2 log (√2 + 1) − 2⎤ =
⎣ ⎦ [3 log (√2 + 1) − √2] .
24 24
Example 3 : Show that if the area lying within the cardioid r = 2a (1 + cos θ) and
without the parabola r (1 + cos θ) = 2a revolves about the initial line, the volum e
generated is 18πa3.
Solution : The equation of the cardioid is
r = 2a (1 + cos θ), ...(1)
and that of the parabola is r = 2a ⁄ (1 + cos θ) . ...(2)
E quating the values of r from (1) and (2), we get
2a (1 + cos θ) = 2a ⁄ (1 + cos θ)
or (1 + cos θ) 2 = 1
or cos θ (cos θ + 2) = 0.
Now cos θ ≠ − 2 .
Therefore cos θ = 0 i.e., θ = π ⁄ 2 , − π ⁄ 2 .
Thus the curves (1) and (2) intersect
where θ = π ⁄ 2 and θ = − π ⁄ 2 .
A l s o b o t h t h e c u r ve s a r e
symmet rical abo ut th e initial line (i.e.,
x-axis). The required volume is generated
by revolving the upper half of the shaded
area about the initial line.
∴ the required volume = Volume
generated by the revolution of the area
OA BO o f t h e c a r d i o i d − volume
generated by the revolution of the area
OPBO of the parabola
I-182 INTEGRAL CALCULUS
π ⁄2 π ⁄2
2π ⌠ 2π ⌠
= ⎮ r 3 sin θ dθ − ⎮ r 3 sin θ dθ
3 ⌡0 3 ⌡0
(for cardioid) (for parabola)
π ⁄2 ⎡
2π ⌠ ⎢ 8a 3 (1 + cos θ) 3 − 8a3 ⎤
⎥ sin θ dθ
= ⎮ ⎢ ⎥
3 ⌡0 ⎢
⎣ (1 + cos θ) ⎥⎦
3
π ⁄2
− 16πa3 ⌠
= ⎮ [(1 + cos θ) 3 − (1 + cos θ) − 3] (− sin θ) dθ (Note)
3 ⌡0
1. Find the volume of the solid generated by the revolution of r = 2a cos θ about the
initial line.
2. F ind t he volume of the solid generated by t he revolution of the cardioid
r = a (1 − cos θ) about the initial line. (Rohilkhand 2010)
3. The arc of the cardioid r = a (1 + cos θ) , specified by − π ⁄ 2 ≤ θ ≤ π ⁄ 2 , is rotated
5
about the line θ = 0, prove that the volume generated is πa3.
2
4. Show that the volume of the solid formed by the revolution of the curve
4
r = a + b cos θ (a > b) about the initial line is πa (a2 + b2) . (Meerut 2008)
3
5. Find the volume of the solid generated by revolving one loop of the lemniscate
1
r2 = a2 cos 2θ about the line θ = π. (Meerut 2006)
2
4 8 4
1. π a 3. 2. π a3. 4. πa (a2 + b 2). 5. (π 2 a3) ⁄ 4 √2 .
3 3 3
VOLUMES AND SURFACES OF SOLIDS OF REVOLUTION I-183
(a) Revolution about the axis of x : To prove that the curved surface of the solid
generated by the revolution, about x-axis, of the area bounded by the curve y = f (x), the
ordinates x = a, x = b and the x-axis is
x= b
⌠
⎮ 2 πy ds ,
⌡x = a
where s is the length of the arc m easured from x = a to any point (x, y).
Or
Show that the area of the surface of the solid obtained by revolving about x-axis the
arc of the curve intercepted between the points whose abscissae are a and b is
b
⌠ ds
⎮ 2πy dx .
⌡a dx
Proof : Let AB be the arc of the curve y = f (x) included between the ordinates
x = a and x = b. It is being assumed that the curve does not cut x-axis and f (x) is a
continuous function of x in the interval (a, b) .
Let P (x, y) and Q (x + δx, y + δy) be any two neighbouring points on the curve
y = f (x) .
Let the length of the arc AP be s and arc A Q = s + δs
so that arc PQ = δs.
D raw t h e o rdinat es PM a n d QN . L e t S
denote the curved surface of the solid generated by
the revolution of the area CMPA about the x-axis.
Then the curved surface of the solid generated by
the revolution of the area MN QP = δS .
We shall take it as an axiom that the curved
surface of the solid generated by the revolution of
the area MN QP about the x-axis lies between the
cu rved su rfaces of th e right circular cylinders
whose radii are PM and N Q and which are of the
same thickness (height) δs. There is no loss in
assuming so because ultimately Q is to tend to P .
Thus δS lies between 2π yδs and 2π ( y + δy) δs
i.e., 2π y δs < δS < 2π ( y + δy) δs or 2π y < (δS ⁄ δs) < 2π ( y + δy) .
Now as Q approaches P i.e., δs → 0, δy will also tend to zero. H ence by taking limits
as δs → 0, we have
dS
= 2π y or dS = 2π y ds.
ds
x= b x= b x= b
⌠ ⌠
∴ ⎮ 2π y ds = ⎮ dS = ⎡⎢⎣ S ⎤⎥⎦
⌡x = a ⌡x = a x= a
x= b
= ⎮
⌠
⌡x = a
2π y
ds
dx
dx , where
ds
dx
=
⎧
√⎛ dy⎞ 2 ⎫
⎨1 + ⎜ ⎟ ⎬ ⋅
⎩ ⎝ dx⎠ ⎭
(b) Axis of revolution as y-axis : Sim ilarly the curved surface of the solid
generated by the revolution about the y-axis, of the area bounded by the curve x = f ( y), the
lines y = a , y = b and the y-axis is
y= b b
⌠
2π ⎮
⌡y = a
⌠
x ds or S = 2π ⎮
⌡y = a
ds
x dy , where
dy
ds
dy
=
√ ⎧ ⎛ dx⎞ 2 ⎫
⎨1 + ⎜ ⎟ ⎬ ⋅
⎩ ⎝ dy ⎠ ⎭
Important Remark : If an arc length revolves about x-axis, the basic formula for
⌠
the surface of revolution in all cases is ⎮ 2π y ds , between the suitable limits. If we want
⌡
to integrate w.r.t. x , we shall change ds as (ds ⁄ dx) dx and adjust the limits accordingly.
A similar transformation can be made if we want to integrate w.r.t. y or with
respect to θ or w.r.t. some parameter, say t .
√ ⎛ dy⎞ 2 ⎫
√
ds ⎧ 2⎫
Therefore =
⎧
⎨1 + ⎜ ⎟ ⎬ = ⎪1 + x ⎪
⎨ ⎬
dx ⎩ ⎝ dx⎠ ⎭ ⎪ y2 ⎪⎭
⎩
√ √
⎧ y2 + x2 ⎫ ⎛ a2 ⎞
= ⎪ ⎪= ⎜ ⎟ [From (1)]
⎨ ⎬ ⎜ 2⎟
⎪ y2 ⎪ ⎜y ⎟
⎩ ⎭ ⎝ ⎠
= a ⁄ y.
For the arc of the circle (1) lying in the first quadrant x varies from 0 to a .
∴ the required surface
a a
⌠ ⌠ ds
= 2π ⎮ y ds = 2π ⎮ y ⋅ dx
⌡x = 0 ⌡0 dx
a a
⌠ a ⌠ a
= 2π ⎮ y ⋅ dx = 2π ⎮ a dx = 2π a ⎡⎢ x ⎤⎥
⌡0 y ⌡0 ⎣ ⎦0
= 2πa . a = 2πa2.
Example 2 : Find the surface generated by the revolution of an arc of the catenary
y = c cosh (x ⁄ c) about the axis of x . (Meerut 2000, 04B, 07, 07B, 10; Rohilkhand 14)
Solution : The given curve is, y = c cosh (x ⁄ c) . ...(1)
Differentiating (1) w.r.t x , we get
VOLUMES AND SURFACES OF SOLIDS OF REVOLUTION I-185
dy x 1 x
= c sinh ⋅ = sinh ⋅
dx c c c
∴
ds
dx
=
√ ⎧ ⎛ dy⎞ 2 ⎫
⎨1 + ⎜ ⎟ ⎬ =
⎩ ⎝ dx⎠ ⎭ √ 1 + sinh xc
⎧
⎨
⎩
2 ⎫
⎬
⎭
= cosh
x
c
...(2)
If the arc be measured from the vertex (x = 0) to any point (x, y), then the required
surface formed by the revolution of this arc about x-axis
x x
⌠ ds ⌠ x x
= ⎮ 2π y dx = 2π ⎮ c cosh ⋅ cosh dx , from (1) and (2)
⌡x = 0 dx ⌡0 c c
x x
⌠ x ⌠ ⎡ 2x⎤
= π c ⎮ 2 cosh2 dx = π c ⎮ ⎢ 1 + cosh ⎥ dx (Note)
⌡0 c ⌡0 ⎣ c⎦
⎡ c 2x⎤ x ⎡ c 2x⎤
= π c ⎢x + sinh ⎥ = π c ⎢ x + sinh ⎥
⎣ 2 c ⎦0 ⎣ 2 c⎦
⎡ x x⎤
= π c ⎢ x + c sinh cosh ⎥ ⋅
⎣ c c⎦
Example 3 : Prove that the surface of the prolate spheroid form ed by the revolution
of the ellipse of eccentricity e about its m ajor axis is equal to 2 × area of the ellipse
× [ √(1 − e2) + (1 ⁄ e) sin− 1 e] .
Solution : [ Note : Prolate spheroid is generated by the revolution of an ellipse
about its major axis]
x2 y2
Let the equation of the ellipse be 2 + 2 = 1, ...(1)
a b
the x-axis being the major axis so that a > b.
The parametric equations of (1) are x = a cos t , y = b sin t .
∴ dx ⁄ dt = − a sin t and dy ⁄ dt = b cos t .
We have
ds
dt
=
√ ⎧ ⎛ dx⎞ 2 ⎛ dy⎞ 2 ⎫
⎨ ⎜ ⎟ + ⎜ ⎟ ⎬ = √(a 2 sin2 t + b2 cos2 t )
⎩ ⎝ dt ⎠ ⎝ dt ⎠ ⎭
π ⁄2
⌠
= 4π ab ⎮ sin t √(1 − e2 cos2 t ) dt .
⌡0
1
Pu t e cos t = z so t h at − e sin t dt = dz . W h e n t = 0, z = e an d wh en t = π,
2
z = 0.
0 e
⌠ 1 4πab ⌠
∴ S = − 4π ab ⎮ √(1 − z2) dz = ⎮ √(1 − z2) dz
⌡e e e ⌡0
4πab ⎡ z 1 ⎤e 4πab ⎡ e 1 ⎤
= ⎢ √(1 − z2) + sin− 1 z⎥ = ⎢ √(1 − e2) + sin− 1 e⎥
e ⎣2 2 ⎦0 e ⎣2 2 ⎦
Example 4 : The part of the parabola y2 = 4ax cut off by the latus rectum revolves
about the tangent at the vertex. Find the curved surface of the reel thus generated.
(Bundelkhand 2011)
Solution : The given parabola is = 4ax . y2 ...(1)
Differentiating (1) w.r.t. x , we get dy ⁄ dx = 2a ⁄ y .
√ ⎛ dy⎞ 2 ⎫
√
ds ⎧ 2⎫
∴ =
⎧
⎨1 + ⎜ ⎟ ⎬ = ⎪ 1 + 4a ⎪
⎨ ⎬
dx ⎩ ⎝ dx⎠ ⎭ ⎪ y ⎪⎭
2
⎩
√ 4a2 ⎫⎪
√
⎧⎪ ⎛ x + a⎞
= ⎨⎪ 1 + ⎬⎪ = ⎜ ⎟ ⋅
⎩ 4ax ⎭ ⎝ x ⎠
The required curved surface is generated
by the revolution of the arc L OL′ (L SL′ is the
latus rectum), about the tangent at the vertex
i.e., y-axis. The curve is symmetrical about x-axis
and for the arc OL, x varies from 0 to a .
∴ the required surface
a
⌠ ds
= 2⎮ 2π x dx
⌡x = 0 dx
a
⌠
= 4π ⎮ x
⌡0
⎜
√
⎛ x + a⎞
⎝ x ⎠
⎟ dx
a
⌠
= 4π ⎮ √(x2 + ax) dx
⌡0
a
√ a⎞ 2 ⎛ a⎞ 2
⌠ ⎧⎛ ⎫
= 4π ⎮ ⎨⎜ x + ⎟ − ⎜ ⎟ ⎬ dx (Note)
⌡0 ⎩⎝ 2⎠ ⎝ 2⎠ ⎭
a
⎡1 ⎛ a⎞ 1 a2 ⎧⎛ a⎞ ⎫⎤
= 4π ⎢ ⎜ x + ⎟ √(x2 + ax) − ⋅ log ⎨ ⎜ x + ⎟ + √(x2 + ax) ⎬ ⎥
⎣2 ⎝ 2⎠ 2 4 ⎩⎝ 2⎠ ⎭⎦ 0
⎡ .. ⌠ 1 1 ⎤
⎢ . ⎮ √(x2 − a2) dx = x √(x2 − a2) − a2 log {x + √(x2 − a2)}⎥
⎣ ⌡ 2 2 ⎦
VOLUMES AND SURFACES OF SOLIDS OF REVOLUTION I-187
1 3 1 3 1 1
= 4π [ ⋅ a a √2 − a2 log { a + a √2} + a2 log ( a)]
2 2 8 2 8 2
3 2 1 2 3 1 1
= 4π [ a √2 − a log {( a + a √2) ⁄ ( a)}] = πa2 [3 √2 − log (3 + 2 √2)]
4 8 2 2 2
1
= πa2 [3 √2 − log (√2 + 1) 2] (Note)
2
= πa2 [3 √2 − log (√2 + 1)] .
8
1. 4πa2. 3. πa2 [2 √2 − 1].
3
⎡ c 2b c 2 a ⎤⎥ ⎡ 4π ⎤⎥
4. πc ⎢⎣ (b − a) + sinh − sinh ⋅ 6. 8π ⎢⎣ 1 + ⋅
2 c 2 c ⎦ 3 √3 ⎦
128
7. πa2 [125 √(10) + 1].
1215
where
ds
dt
=
√ ⎧ ⎛ dx⎞ 2 ⎛ dy⎞ 2 ⎫
⎨⎜ ⎟ + ⎜ ⎟ ⎬ ⋅
⎩ ⎝ dt ⎠ ⎝ dt ⎠ ⎭
I-188 INTEGRAL CALCULUS
Example 1 : Find the surface of the solid generated by revolution of the astroid
x2 ⁄ 3 + y2 ⁄ 3 = a2 ⁄ 3 or x = a cos3 t , y = a sin3 t about the x-axis. (Meerut 2006, 09;
Kanpur 05; Rohilkhand 07, 09, 11B; Avadh 11; Kashi 12)
Solution : The parametric equations of the the
curve are x = a cos3 t , y = a sin3 t .
dx
∴ = − 3a cos2 t sin t
dt
dy
and = 3a sin2 t cos t .
dt
H ence
ds
dt
=
√ ⎡ ⎛ dx⎞ 2 ⎛ dy⎞ 2 ⎤
⎢⎜ ⎟ + ⎜ ⎟ ⎥
⎣ ⎝ dt ⎠ ⎝ dt ⎠ ⎦
⎛ 1 ⎞ (− sin2 t + 1) a cos2 t
= a ⎜ − sin t + ⎟ = a =
⎝ sin t ⎠ sin t sin t
dy
and = a cos t .
dt
√ ⎧ ⎛ dx⎞ 2 ⎛ dy⎞ 2 ⎫
√
ds ⎧ a 2 cos4 t ⎫ a cos t
H ence = ⎨⎜ ⎟ + ⎜ ⎟ ⎬ = ⎪ + a2 cos2 t⎪⎬ = ⋅
⎨
dt ⎩ ⎝ dt ⎠ ⎝ dt ⎠ ⎭ ⎪ sin2 t ⎪ sin t
⎩ ⎭
VOLUMES AND SURFACES OF SOLIDS OF REVOLUTION I-189
The given curve is symmetrical about both the axes and the asymptote is the line
y = 0 i.e., x-axis. For the arc of the curve lying in the second quadrant t varies from 0
1
to π.
2
π ⁄2
⌠ ds
∴ the required surface = 2 . ⎮ 2π y dt (Note)
⌡0 dt
π ⁄2 π ⁄2
⌠ a cos t ⌠
= 4π ⎮ a sin t . dt = 4πa2 ⎮ cos t dt
⌡0 sin t ⌡0
π ⁄2
= 4πa2 ⎡⎢ sin t⎤⎥ = 4πa2
⎣ ⎦ 0
= the surface of a sphere of radius a .
1. Find the surface area of the solid generated by revolving the cycloid
x = a (θ − sin θ), y = a (1 − cos θ) about the x-axis.
2. Find the area of the surface generated by revolving an arch of the cycloid
x = a (θ + sin θ), y = a (1 − cos θ) about the tangent at the vertex.
3. The portion between two consecutive cusps of the cycloid x = a (θ + sin θ),
y = a (1 + cos θ) is revolved about the x-axis. Prove that the area of the surface so
formed is to the area of the cycloid as 64 : 9.
4. Prove that the surface area of the solid generated by the revolution, about the
x-axis of the loop of the curve
1
x = t2, y = t − t3 is 3π .
3
5. Prove that the surface of the oblate spheroid formed by the revolution of the
ellipse of the semi-major axis a and eccentricity e is
⎡ 1 − e2 ⎛ 1 + e⎞ ⎤ .
2πa2 ⎢ 1 + log ⎜ ⎟⎥
⎣ 2e ⎝ 1 − e⎠ ⎦
64 2 32 2
1. πa . 2. πa .
3 3
Example 1 : Find the surface of the solid generated by the revolution of the
lemniscate r 2 = a2 cos 2θ about the initial line. (Meerut 2004, 10B, 11; Rohilkhand 08B;
Agra 14; Purvanchal 14)
Solution : The given curve is r 2 = a2 cos 2θ. ...(1)
Differentiating (1) w.r.t. θ, we get
dr
2r = − 2a2 sin 2θ
dθ
dr − a2 sin 2θ
or = ⋅
dθ r
∴
ds
dθ
=
√ ⎧ 2 ⎛ dr ⎞ 2 ⎫
⎨r + ⎜ ⎟ ⎬
⎩ ⎝ dθ⎠ ⎭
√ a cos 2θ + a sinr 2θ
⎧ 2 4 2 ⎫
= ⎪ ⎪
⎨ 2 ⎬
⎪ ⎪
⎩ ⎭
1
= √{r 2 . a2 cos 2θ + a4 sin2 2θ}
r
1
= √{a4 cos2 2θ + a4 sin2 2θ} , [ ... r 2 = a2 cos 2θ]
r
= a2 ⁄ r . ...(2)
The given curve is symmetrical about the initial line and about the pole.
1 1
Putting r = 0 in (1), we get cos 2θ = 0 giving 2θ = ± π i.e., θ = ± π.
2 4
1 1
Therefore one loop of the curve lies between θ = − π and θ = π.
4 4
There are two loops in the curve and for the upper half of one of these two loops
1
θ varies from 0 to π.
4
∴ the required surface
= 2 × the surface generated by the revolution of one loop
π ⁄4
⌠ ds
= 2.⎮ 2π y dθ, where y = r sin θ
⌡0 dθ
π ⁄4
⌠ a2
= 4π ⎮ r sin θ ⋅ dθ [From (2)]
⌡0 r
π ⁄4
⌠ π ⁄4
= 4πa2 ⎮ sin θ dθ = 4πa2 ⎡⎢ − cos θ⎤⎥
⌡0 ⎣ ⎦ 0
Example 2 : A circular arc revolves about its chord. Find the area of the surface
generated, when 2α is the angle subtended by the arc at the centre.
Solution : Let the parametric equations of the circle be
x = a cos θ, y = a sin θ, ...(1)
θ being the parameter.
Take any point P (a cos θ, a sin θ) on the
circular arc A BC which is symmetrical about the
x-axis and which subtends an angle 2α at the
centre O so that ∠ A OB = α .
W e h a ve OD = OA cos α = a cos α .
Draw PM perpendicular from P to A C , the axis
of rotation. Then
PM = ON − OD = a cos θ − a cos α . ...(2)
For the upper half of the arc to be rotated
i.e., for the arc BA, θ varies from 0 to α .
Also
ds
dθ
=
√ ⎧ ⎛ dx ⎞ 2 ⎛ dy⎞ 2 ⎫
⎨⎜ ⎟ + ⎜ ⎟ ⎬
⎩ ⎝ dθ⎠ ⎝ dθ⎠ ⎭
32 2 48
1. 4πa2. 2. πa . 3. √2 πa2.
5 5
Then the surface generated by the revolution of the arc A B about x-axis is
x= b
⌠
= ⎮ 2π y ds ...(1)
⌡x = a
_
Also we know that (see the chapter on centre of gravity) the ordinate y, of the
centroid of the arc from x = a to x = b, of length l, is given by
b
⌠
⎮ y ds
_ ⌡x = a
y= ...(2)
l _ _
From (1) and (2), we get the required surface = 2π y l = l × 2π y
= (length of the arc) × (length of the path described by the centroid of the arc).
Note 1 : The closed curve or arc in the above theorems must not cross the axis
of revolution but may be terminated by it.
Note 2 : When the volume or surface generated is known, the theorems may be
applied to find the position of the centroid of the generating area or arc.
Example 1 : Find the volum e and surface-area of the anchor-ring generated by the
revolution of a circle of radius a about an axis in its own plane distant b from its centre
(b > a) .
Solution : H ere the given curve (circle) does not
intersect the axis of rotation, so Pappus theorem can be
applied.
In this case A
= area of the region of the closed curve
= area of the circle of radius a= πa2
and l = length of the arc of the curve
= circumference of the circle = 2πa .
As the centroid of the area of a circle
_ and also of
its circumference lies at the centre, so y = b in both the
cases and hence the length of the path described by the
C.G. = 2πb .
Now by Pappus theorem, the required volume of the anchor-ring
= (area of the circle)
× (circumference of the circle generated by the centroid)
= πa2 . 2 πb = 2 π 2 a2 b.
And the surface area of the anchor-ring
= (arc length of the circle)
× (circumference of the circle generated by the centroid )
= 2πa . 2πb = 4π 2ab.
Example 2 : Show that the volum e generated by the revolution of the ellipse
x2 ⁄ a2 + y2 ⁄ b2 = 1 about the line x = 2a is 4π 2 a2 b.
I-194 INTEGRAL CALCULUS
Example 3 : The loop of the curve 2ay2 = x (x − a) 2 revolves about the straight line
y = a . Find the volume of the solid generated.
Solution : The given curve is 2ay2 = x (x − a) 2. ...(1)
The cu rve (1) is symmet rical about the
x-axis and the loop lies between x = 0 and x = a .
Differentiating (1) w.r.t. x , we get
4ay (dy ⁄ dx) = 2x (x − a) + (x − a) 2
= 3x2 − 4ax + a2.
Now (dy ⁄ dx) = 0 when 3x2 − 4ax + a2 = 0
or when x = a ⁄ 3 which gives from (1),
y = (a √2) ⁄ (3 √3) i.e., < a showing that the loop
does not intersect the straight line y = a .
By symmetry the C.G . of the loop lies on
x-axis i.e., the distance of the C.G . from the axis of revolution ( y = a) is a . When the
loop is rotated about y = a, its C.G . will describe a circle of radius a whose perimeter
is 2πa .
Also the area A of the loop
a a
⌠ ⌠ (x − a) √x ⎡ .. (x − a) √x⎤
= 2 ⎮ y dx = 2 ⎮ dx , ⎢ . from (1), y = ⎥
⌡0 ⌡0 √(2a) ⎣ √(2a) ⎦
a
=
√ ⎛ 2⎞ ⌠
⎜ ⎟ ⎮ (x3 ⁄ 2 − ax1 ⁄ 2) dx
⎝ a ⎠ ⌡0
⎡⎢ x5 ⁄ 2 a
ax3 ⁄ 2 ⎤⎥
=
√ ⎛ 2⎞
⎜ ⎟
⎝ a⎠
⎢⎢
⎣ 5 ⁄ 2
−
3 ⁄ 2
⎥⎥ =
⎦0
4
15
√2 a2.
Example 4 : Prove that the volum e of the solid form ed by the rotation about the
line θ = 0 of the area bounded by the curve r = f (θ) and the lines
θ
2 ⌠ 2 3
θ = θ1, θ = θ2 is π ⎮ r sin θ dθ.
3 ⌡θ
1
VOLUMES AND SURFACES OF SOLIDS OF REVOLUTION I-195
Solution : L e t OA B b e t h e a r e a
bounded by the curve r = f (θ) and the radii
vectors θ = θ1 and θ = θ2. We have to find the
vo lu me fo rmed by th e revolut io n o f area
OA B about the initial line OX .
Take an y point (r, θ) in side the area
OA B and take a small element of the area
r δθ δr at the point P. Drop PM perpendicular
from P to the axis of rotation OX . We have
PM = OP sin θ = r sin θ.
Now the volume of the ring formed by
revolving the element of area r δθ δr about
OX
= 2 πr sin θ . rδθ δr = 2 π r 2 sin θ δθ δr.
Therefore the whole volume formed by revolving the area OA B about OX
θ f (θ)
⌠ 2 ⌠
= ⎮ ⎮ 2 π r 2 sin θ d θ dr
⌡θ = θ ⌡r = 0
1
θ ⎡ 3⎤f (θ)
⌠ 2 ⎢r ⎥
= ⎮ 2 π sin θ ⎢⎢ ⎥⎥ dθ
⌡θ = θ ⎣ 3 ⎦0
1
θ θ
2 ⌠ 2 2 ⌠ 2
= π⎮ [ f (θ)]3 sin θ d θ = π ⎮ r 3 sin θ d θ,
3 ⌡θ = θ 3 ⌡θ
1 1
5. A semi-circular bend of lead has a mean radius of 8 inches; the initial diameter of
1
the pipe is 4 inches and the thickness of the lead is inch. Applying the theorem
2
of Pappus and G uldin find the volume of the lead and its weight, given that 1 cubic
inch of lead weighs 0⋅ 4 lb.
[Hint. Internal diameter of pipe = 4 inches.
1
Thickness of metal = inch
2
∴ external diameter of the pipe = 4 + 1 = 5 inches.
1 9
∴ area of lead = π (52 − 42) = π.
4 4
The centroid of this area is at a distance of 8 inches from the axis of rotation.
Therefore the length of path traced out by its centroid in describing a semi-circle
= 8π inches.
9
∴ volume of the lead = 8π × π = 18π 2 cu. inch.
4
∴ weight of the pipe = volume × density = 18π 2 × 0⋅ 4 lb. = 71⋅ 1 lb]
6. State the theorems of Pappus and G uldin. (Meerut 2008)
1. 4a ⁄ 3π. 2. 2 π 2 a2 b or 2 π 2 ab 2.
3. 6 √2 π 2 a3, where a is the semi-major axis.
11 2 2
4. π a .
2
1. The volume of the solid generated by the revolution of the area bounded by the
curve y = f (x), x-axis and the ordinates x = a, x = b about the x-axis is …… .
(Meerut 2003)
2. The volume of the solid generated by the revolution of the area bounded by the
curve r = f (θ) and the radii vectors θ = θ1, θ = θ2 about the initial line is …… .
(Meerut 2001)
3. If the equation of the curve in the polar form is r = f (θ), then the curved surface
generated by the revolution about the initial line of the arc intercepted between
θ= β
⌠ ds ds
the radii vectors θ = α and θ = β is ⎮ 2 π (r sin θ) d θ, where = …… .
⌡ dθ dθ
θ= α
VOLUMES AND SURFACES OF SOLIDS OF REVOLUTION I-197
6. The volume of the paraboloid generated by the revolution about the x-axis of the
parabola y2 = 4 a x from x = 0 to x = h is
(a) 2 π a h 2 (b) 2 π a h
2 2
(c) π a h2 (d) πah
3 3 (Rohilkhand 2005)
7. The curved surface of the solid generated by the revolution about the y-axis of the
area bounded by the curve x = f ( y), the lines y = a, y = b and y-axis is
b b
⌠ ⌠
(a) ⎮ π x d s (b) ⎮ 2 π x ds
⌡a ⌡a
b b
⌠ 2 ⌠
(c) ⎮ π x ds (d) ⎮ π 2 x ds
⌡a 3 ⌡a
True or False:
Write ‘T’ for true and ‘F’ for false statement.
8. The volume of the solid generated by the revolution of the area bounded by the
curve r = f (θ) and the radii vectors θ = θ1, θ = θ2 about the initial line θ = 0 is
θ
⌠2 2
⎮ π r 3 sin θ d θ.
⌡θ 3
1
I-198 INTEGRAL CALCULUS
9. If an arc length revolves about x-axis, the basic formula for the surface of
⌠
revolution in all cases is ⎮ 2 π y d s, between the suitable limits.
⌡
b θ
⌠ ⌠ 2 2
1. ⎮ π y 2 dx . 2. ⎮ π r 3 sin θ d θ .
⌡a ⌡θ 3
1
⎧ ⎛ ⎞2⎫ 2 2
⎛ dy⎞ ⎫
√ √
⎧⎛ dx⎞⎟
⎨ r 2 + ⎜ d r⎟ ⎬ ⋅
⎜ ⎟
3. 4. ⎨⎜ + ⎜ ⎟ ⎬⋅
⎩ ⎜ ⎟
⎝ d θ⎠ ⎭ ⎩⎝ d t⎠ ⎝ dt ⎠ ⎭
5. (b). 6. (a).
7. (b). 8. T.
9. T.