Mathematics Limits
Mathematics Limits
Mathematics Limits
The concept of limits forms the basis of calculus and is a very powerful one. Both differential and integral calculus
are based on this concept and as such, limits need to be studied in good detail.
This section contains a general, intuitive introduction to limits.
Consider a circle of radius r.
m
Fig - 1
co
We know that the area of this circle is π r . How?
2
g.
The ancient Greeks derived this result using the concept of limits.
To see how, recall the definition of π .
ng
length of circumference
π=
length of diameter
ye
c c
π= =
d 2r
.m
c = 2π r
With this definition in hand, the Greeks divided the circle as follows (like cutting a cake or a pie):
w
w
B C
w
A D
H E
G F
Fig - 2
Now they took the different pieces of this ‘pie’ and placed them as follows:
MATHS / LIMITS
LOCUS 2
m
co
g.
ng
The figure on the right side starts resembling a rectangle as we increase the number of cuts to the circle. The
ye
sequence of curves that joins x to y starts becoming more and more of a straight line with the same total length π r .
What happens as we increase the number of cuts indefinitely, or equivalently, we decrease θ indefinitely? The
.m
figure ‘almost’ becomes a rectangle, though never becoming a rectangle exactly. The area ‘almost’ becomes
π r × r = π r 2.
w
In the language of limits, we say that the figure tends to a rectangle or the area A tends to π r 2 , or the limiting
w
value of area is π r 2 .
w
In standard terminology.
lim A = π r 2
θ →0
Hence, we see that a limit describes the behaviour of some quantity that depends on an independent variable, as
that independent variable ‘approaches’ or ‘comes close to’ a particular value.
1 1
For example, how does behave when x becomes larger and larger? becomes smaller and smaller and
x x
‘tends’ to 0.
MATHS / LIMITS
LOCUS 3
We write this as
1
lim =0
x →∞ x
1 1
How does behave when x becomes smaller and smaller and approaches 0? obviously becomes larger and
x x
larger and ‘tends’ to infinity.
We write this as:
1
lim =∞
x →0 x
The picture is not yet complete. In the example above, x can ‘approach’ 0 in two ways, either from the left hand
side or from the right hand side:
m
x → 0− : approach is from left side of 0
co
x → 0 + : approach is from right side of 0 g.
0
x
– +
ng
x→0 x→0
Fig - 6
ye
1
How do we differentiate between the two possible approaches? Consider the graph of f ( x ) = carefully.
x
f(x)
.m
f(x)→∞ f(x)= 1x
w
w
f(x)→0+
w
x→-∞
x
x→0 - x→0+ x→∞
f(x)→0 -
f(x)→-∞
Fig - 7
As we can see in the graph above, as x increase in value or as x → ∞, f ( x ) decreases in value and approaches
0 (but it remains positive, or in other words, it approaches 0 from the positive side)
MATHS / LIMITS
LOCUS 4
lim f ( x ) = 0 +
x →∞
Similarly, lim f ( x ) = + ∞
x →0 +
What if x approaches 0, but from the left hand side ( x → 0− ) ? From the graph, we see that as x → 0 − ,
1
x
1
increases in magnitude but it also has a negative sign, that is → − ∞.
x
1 1
What if x → − ∞ ? decreases in magnitude (approaches 0) but it still remains negative, that is, approaches
x x
m
1
0 from the negative side or → 0−
x
co
These concepts and results are summarized below: g.
ng
ye
.m
w
w
Lets consider another example now. We analyse the behaviour of f(x) = [x], as x approaches 0.
w
MATHS / LIMITS
LOCUS 5
lim [ x ] = 0
x →0+
lim [ x ] = −1
x →0−
lim [ x ] = I
x→I +
I is any integer
lim− [ x ] = I − 1
m
x→I
co
lim {x} = 0
x →0 +
g.
lim {x} = 1
ng
x →0 −
lim {x} = 0
ye
x→I +
I is any integer
lim− {x} = 1
x→I
.m
Its possible that the value of the function at x = a is undefined, and yet the LHL or RHL (or both) exist.
x2 − 1
For example, consider f ( x ) =
x −1
MATHS / LIMITS
LOCUS 6
f (x) =
( x − 1)( x + 1) = x + 1
( x − 1)
which is a straight line
f(x)
x2-1
2 f(x) =
x-1
m
x
1
co
g.
Fig - 11
This line has a hole at x = 1 because f (x) is undefined there. In the neighbourhood of x = 1( as x → 1+ or x → 1− ) ,
ng
f(x) ‘approaches’ the value 2 (though it never becomes 2, because to become 2, x has to have the value 1, which
ye
x →1 x →1
To emphasize once again, in evaluating a limit at x = a, we are not concerned with what value f(x) assumes
w
at precisely x = a; we are concerned with only how f(x) behaves as x approaches or nearly becomes a, whether
from the left hand or right hand side, giving rise to LHL and RHL respectively.
w
And finally, the limit of f(x) at x = a is said to exist if the function approaches the same value from both sides
w
LHL= RHL at x = a
implies lim f ( x ) exists
x →a
Note:
We have used the concept of ∞ (infinity) in the discussions above. Lets discuss this concept in some what more
detail:
⇒ Infinity does not stand for any particular real number. In fact, it cannot be defined precisely. This is a
deep concept. For any number you can think of, no matter how large, infinity is still larger. When we
say that x → ∞, we mean that x increase in an unbounded fashion, that is, becomes indefinitely large.
MATHS / LIMITS
LOCUS 7
⇒ We cannot apply the normal rules of arithmetic to infinity. For example, saying that
∞
∞−∞ =0 or =1
∞
is absurd because such quantities are not defined.
⇒ It should be clear now that an expression like x → ∞ defines a tendency (of unbounded increase)
Num
Consider a fraction f =
Den
As x → a, if (Num) → (a finite number) and (Den) → ∞ , then f tends to become infinitesmally small
or f → 0.
Such cases are listed out below. Observe the details carefully
Num
f=
m
Den
(i) lim ( Num ) = finite
co
x→ a
⇒ lim f = 0
lim ( Den ) = infinity x →a
x→ a
g.
(ii) lim ( Num ) = infinity
ng
x→ a
⇒ lim f = ∞
lim ( Den ) = finite x →a
x→ a
ye
x→ a
⇒ lim f = 0
lim ( Den ) = non-zero x →a
x→ a
w
x→ a p
⇒ lim f =
lim ( Den ) = q ≠ 0 x →a q
w
x→ a
MATHS / LIMITS
LOCUS 8
0
that f is of the form . But as we have seen earlier, the limit of f exists, as x → 1
0
x2 − 1
lim f = lim = lim ( x + 1) = 2
x →1 x →1 x − 1 x →1
0 ∞
Limits of the form , , ∞ − ∞ , ∞ × 0, 1∞ , 0º are called indeterminate forms. The limits of such
0 ∞
forms may exist but it cannot be determined by simple observation (hence the name indeterminate).
Such forms need to be reduced into determinate forms for which the limit can be determined. In the
x2 − 1
above case, for example, an indeterminate form lim was reduced into a determinate form
x →1 x − 1
m
x →1
co
g.
Section - 2 SOME GENERAL RULES PERTAINING TO LIMITS
ng
Now that we’ve had an intuitive introduction to limits, how do we go about evaluating limits for arbitrary functions?
ye
sin x
For example, consider f ( x ) = . How does this function behave as x → 0 ? (This, as we have seen in the
x
.m
implying that the graph of f(x) has no ‘break’ at ‘x = a’ (and in it’s neighbourhood), the limit of f(x) at x = a can be
evaluated by direct substitution. That is,
w
lim f ( x ) = f ( a )
w
x→a
This will become obvious once we consider an example take f (x) = x2.
y
2
f (x) = x
x
x 0– x +
Fig - 12
MATHS / LIMITS
LOCUS 9
From the graph, we see that since f(x) = x2 is ‘continuous’ at x = 0 (it has no break), f(x) will approach the same
value from both sides, equal to f(0)
Hence, lim x 2 = lim− x 2 = LHL = lim+ x 2 = RHL = f (0 ) = 0
x →0 x →0 x →0
Thus, any determinate limit can be evaluated by direct substitution of x = a. (The function needs to have no breaks
at x = a)
For example:
lim ( x 2 + 2 ) = 1 + 2 = 3
x →1
lim ( x 3 + 4x + 5 ) = 0 + 0 + 5 = 5
x →0
m
lim sin x = sin π = 0
x →π
co
What about lim [ x ]?
x →1 g.
LHL = xlim
→1−
[ x] = 0
ng
RHL = xlim
→1+
[x] = 1
LHL ≠ RHL
ye
The issue now is how to evaluate limits for indeterminate forms. This is the subject of the next section (where some
standard indeterminate limits will be evaluated).
For now, observe some general rules pertaining to limits. These are more or less self explanatory.
If l, m exist, then:
(i)
x→ a
{
x→ a
}{ x→ a
}
lim { f ( x ) ± g ( x )} = lim f ( x ) ± lim g ( x ) = l ± m
MATHS / LIMITS
LOCUS 10
f ( x ) lim
lim = x→a
{
f (x)
=
l } if m ≠ 0
{ }
(iii)
x→ a g ( x )
lim g ( x ) m
x→a
(iv)
x →a
{
lim {K f ( x )} = K lim f ( x ) = Kl
x →a
}
(v) lim f ( x ) = lim f ( x ) = l
x→ a x→ a
( )( ) = l
g(x) lim g ( x )
lim ( f ( x )) = lim f ( x )
x→ a
(vi) m
, if the latter is defined.
x→ a x→ a
For example,
m
lim (log f ( x )) = log lim f ( x ) ( )
co
x→ a x→ a
f (x) lim f ( x )
lim e = e x→ a
x→a
g.
There needs to be a slight addition to this rule. Can you see what?
Consider f (x) = [x] and g (x) = x3
ng
describes the behaviour of a function f(x) ‘sandwiched’ between two different functions g(x) and
w
h(x). At x = a, if g(x) and h(x), tend to the same value l, and f(x) is sandwiched between g(x) and
h(x), it will obviously also tend to the same value l.
w
h(x)
f(x)
g(x)
( ) x
x=a
Fig - 13
MATHS / LIMITS
LOCUS 11
For the graph above, in the neighbourghood of x = a, we see that the following always holds:
g ( x) ≤ f ( x) ≤ h ( x)
or f(x) is sandwiched between g(x) and h(x).
Also, since lim g ( x ) = lim h ( x ) = l,
x→a x→a
we have lim f ( x ) = l
x →a
m
(A) sin x tan x
lim = 1; lim =1
co
x →0 x x →0 x
0
g.
Both the limits above are indeterminate, of the form . We are discussing here a geometric interpretation
0
of these limits. Consider a sector OAP of a unit circle as shown in the figure below.
ng
ye
.m
w
w
w
We see that
area ( ∆OAP ) < area (sec tor OAP ) < area ( ∆OAB )
or
1 1 1
sin x < x < tan x
2 2 2
or
sin x tan x
< 1 <
x x
MATHS / LIMITS
LOCUS 12
sin x
⇒ → 1 (but remains less than 1)
x
tan x
→ 1 (remains greater than 1)
x
sin x tan x
= 1− ; lim = 1+
m
⇒ lim
x →0 x x →0 x
co
It is important to observe how any function approaches a limit.
g.
sin x
For example, in the case above, as x → 0 , approaches 1 from the left side while tan x
x x
ng
Why?
ye
sin x tan x
lim = 0 ; lim =1
x →0
x x →0
x
w
w
w
(B) x
1
lim 1 + = e
x →∞
x
x
Consider the expression for f ( x ) = 1 + . As x gets larger and larger or as x → ∞ , the base
1
x
1
1 + gets closer to 1 while the exponent (x), tends to infinity. Hence, this limit is of the indeterminate
x
x
form 1 . Its very important to get a ‘feel’ that the value 1 + will converge to a fixed, definite
∞ 1
x
MATHS / LIMITS
LOCUS 13
value, as x increases. You should get this feel by looking at the table below.
x
1 1
x 1+ 1+
x x
1 2 2
10 1.1 2.5937...
100 1.01 2.7048...
1000 1.001 2.7169...
10000 1.0001 2.7181...
100000 1.00001 2.7182...
Fig - 15
m
x
1
We see that as x becomes larger, the term 1 + converges to some value (This can be proved)
x
co
This limiting value is denoted by e. e is an irrational number and its value is e = 2.7182...
g.
x
1
Try to show that the limit of 1 + is bounded and lies between and 2 and 3 that is, 2 < e < 3
ng
x
(you can use the binomial theorem in conjunction with the sandwich theorem to prove this, by first
ye
If f ( x ) → 0 x → a, then
w
as
w
lim (1 + f ( x )) = lim (1 + f ( x ))
1 1
f ( x) f ( x)
x→a f ( x )→0
w
lim (1 + y ) ( y = f ( x ))
1
y
=e
y →0
(C) ln (1 + x )
lim =1 (ln y ≡ log e y )
x →0 x
0
This limit is of the indeterminate form .
0
We can easily evaluate this limit based on the previous limit.
MATHS / LIMITS
LOCUS 14
ln (1 + x )
lim
x →0 x
1
= lim ln (1 + x )
x →0 x
1
= limln (1 + x ) x (Property of log)
x →0
{
= ln lim (1 + x )
x →0
1
x
}
= ln e = log e e = 1
This limit can alternatively be evaluated by using the expansion series for ln (1 + x)
x2 x3
ln (1 + x ) = x − + − ...∞ when x < 1
m
2 3
ln (1 + x )
co
x x2
⇒ = 1 − + − ...∞
x 2 3 g.
ln (1 + x )
⇒ lim = 1 (All other terms involving x tend to 0)
x →0 x
ng
log a (1 + x )
ye
(D) 1
lim =
x →0 x ln a
.m
lim
x →0 x x →0 ( ln a ) ⋅ x
w
1 ln (1 + x )
lim
w
=
ln a x →0 x
1
=
ln a
We have used the following property of logarithms
above:
log b y
log a y =
log b a
(E)
ax − 1
lim = ln a
x →0 x
This is again an extension of the limits seen previously.
MATHS / LIMITS
LOCUS 15
1 1
= lim =
t →0 log a (1 + t ) log a (1 + t )
lim
t t →0 t
1
= = ln a.
1
ln a
m
Note that for a = e, this limit is 1.
co
(F) xm − 1
lim =m
x →1 x − 1
g.
When m is an integer, it is easy to see that the above relation holds because xm – 1 can be expanded
ng
(1 + y ) − 1
m
xm − 1
Now we have lim = lim
x →1 x − 1 y →0 y
w
Now expand (1 + y)m using the Binomial theorem for a general index.
w
m ( m − 1) y 2
(1 + y ) = 1 + my + + ...
m
w
2!
(1 + y ) −1
m
Hence, lim
y →0 y
m ( m − 1) 2
my + y + ...
= lim 2!
y →0 y
m ( m − 1)
= lim m + y + ...
y →0
2!
MATHS / LIMITS
LOCUS 16
(G) xm − am
lim = m a m −1
x →a x − a
m x
m
a − 1
xm − am
m
= lim
a
lim
x→a x − a x→a x
a − 1
a
m
x
−1 tm − 1 x
a
where t =
m −1 m −1
= a lim = a lim
t →1 t − 1 a
→1
x x
m
a −1
a
co
= m a m–1
g.
(H) sin −1 x tan −1 x
= lim =1
ng
lim
x →0 x x →0 x
sin x tan x
ye
sin −1 x
Consider lim .
x →0 x
w
Also, as x → 0,sin −1 x → 0 or t → 0.
w
t 1
Hence, our limit reduces to lim = =1
t →0 sin t sin t
lim
t →0
t
tan −1 x
Similarly, lim =1
x →0 x
sin −1 x tan −1 x
Evaluate lim and lim x
x →0
x x →0
MATHS / LIMITS
LOCUS 17
Having seen some important indeterminate limits, notice the following expansions that can be used to evaluate
limits.
x2 x3
(a) ln (1 + x ) = x − + − ...∞ x <1
2 3
x2 x3
(b) ln (1 − x ) = − x − − − ...∞ x <1
2 3
x x2 x3
(c) ex = 1 + + + + ...∞ xε
1! 2! 3!
x x2 x3
(d) e− x = 1 − + − + ...∞ xε
1! 2! 3!
x3 x5
sin x = x − + − ...∞
m
(e)
3! 5!
co
x2 x4
(f) cos x = 1 − + − ...∞
2! 4! g.
x 3 2x 5 11x7
(g) tan x = x + + + + ...∞
ng
3 15 315
n ( n − 1) x 2 x < 1
(1 + x ) = 1 + nx + + ...∞
n
(h)
ye
2! and n ε Q
.m
g(x)
Note: We finally need to consider one last important form of limits, namely lim ( f ( x ))
x→ a
(i)
x →a
w
g(x) lim g ( x ) ln f ( x )
lim ( f ( x ))
g ( x ) ln f ( x )
= lim e = e x→ a
x→a x→a
w
where lim h ( x ) = 0
x →a
g(x) g(x)
Now lim ( f ( x )) = lim (1 + h ( x ))
x →a x →a
lim g ( x )⋅h ( x )
1
⋅ g ( x )⋅h ( x )
= lim (1 + h ( x )) h( x ) = e x→a
x→a
MATHS / LIMITS
LOCUS 18
Now we discuss the various methods used in obtaining limits. Each method will be accompanied by some examples
illustrating that method.
m
x →1
(ii) lim 5x 2 + 3x + 1 = 27
co
x →2
(iii) lim 4x 3 + 4 = 0
x →−1
g.
(iv) lim | y | +1 = 2
y →1
ng
5x 2 + 4 129
(v) lim =
x →5 2x + 7
ye
17
x 3 + 1 513
(vi) lim = = 57
.m
x →8 x + 1 9
and so on
w
(B) FACTORIZATION
w
x2 − 1
We saw an example of this method in evaluating lim .
x →1 x − 1
w
In such forms, the limit is indeterminate due to a certain factor occuring in the expression (For example,
in the limit above, (x – 1) occurs in both the numerator and denominator and makes the limit
0
indeterminate, of the form ) . Factorization leads to cancellation of that common factor and reduction
0
of the limit to a determinate form.
x3 − 1 ( x − 1 )( x 2 + x + 1 )
(i) lim = lim = lim ( x 2 + x + 1) = 3
x →1 x − 1 x →1 x −1 x →1
xm − 1
Note that this limit is also of the form lim (whose limit is m)
x →1 x − 1
x 2 − 3x + 2 ( x − 1 )( x − 2 ) x −1 1
(ii) lim = lim = lim =
x→2 x −4
2 x → 2 ( x − 1 )( x + 2 ) x→2 x + 2 4
MATHS / LIMITS
LOCUS 19
( 1 + x )( 1 + 2x )( 1 + 3x ) − 1
(iii) lim
x →0 x
1 + 6 x + 11x 2 + 6 x 3 − 1 6 x + 11x 2 + 6 x 3
= lim = lim
x →0 x x →0 x
= lim ( 6 + 11x + 6 x 2 ) = 6
x →0
x 4 − 3x + 2
(iv) lim 5
x →1 x − 4x + 3
We see that upon substitution of x = 1, both the numerator and denominator become 0.
Hence, (x – 1) is a factor of both the numerator and denominator (Factor theorem)
Factorization leads to
x 4 − 3x + 2 ( x − 1 ) ( x3 + x2 + x − 2 ) x3 + x2 + x − 2
lim 5 = lim = lim 4
m
x →1 x − 4x + 3 x →1 ( x − 1 ) ( x 4 + x 3 + x 2 + x − 3 ) x →1 x + x 3 + x 2 + x − 3
1
= =1
co
1 g.
(C) RATIONALIZATION
ng
In this method, the rationalization of an indeterminate expression leads to determinate one. The following
examples elaborate this method.
ye
x2 + 1 − 1 0
(i) = lim of the indeterminate form
x →0
x 2 + 16 − 4 0
.m
w
=
w
w
8
= =4
2
(ii)
= lim
x →∞
( x2 + x + 1 − x2 + 1 × ) x2 + x + 1 + x2 + 1
x2 + x + 1 + x2 + 1
( x2 + x + 1 ) − ( x2 + 1 )
= lim
x →∞
x2 + x + 1 + x2 + 1
MATHS / LIMITS
LOCUS 20
x
= lim
x →∞
x + x + 1 + x2 + 1
2
∞
Rationalization has led us to another indeterminate form of . However, it can easily be made
∞
determinate in the following manner:
x
lim
x →∞
x2 + x + 1 + x2 + 1
1 Divide the numerator
lim and denominator by
= x→∞ 1 1 1
1+ + 2 + 1+ 2
x x x
1 1
Now, as x → ∞ , → 0 and 2 → 0
m
x x
Hence, the limit above reduces to
co
1 1
=
1+0 +0 + 1+0 2
g.
(D) REDUCTION TO STANDARD FORMS
ng
In this method, we try to reduce the given limit to one of the standard forms we studied in Section-3
lim (1 + sin x )
ye
2 cot x
(i)
x →0
2 cos x
1
w
↓
(This limit has the value e.)
lim 2 cos x
= e x→0 = e 2 (cos x → 1 as x → 0 )
MATHS / LIMITS
LOCUS 21
− tan h + sin h
= lim
h →0
−8h 3
sin h
− sin h +
1 cos h
= lim
8 h→0 h3
1 sin h 1 − cos h 1
= lim . .
8 h →0 h h2 cos h
1 sin h 2 sin 2 h 2 1
= lim . .
4 (h 2 )
2
8 h →0 h cos h
m
2
1 sin h sin h 2 1
= lim . .
( h 2 ) cos h
16 → h
co
h 0
sin x
g.
This expression now only contains the limits lim = 1 and lim cos x = 1
x →0 x x →0
ng
1
Hence, the final result is
16
ye
We will now see examples based on the methods discussed above. We urge you to first try out all
these examples on your own before viewing the solutions.
.m
w
SOLVED EXAMPLES
w
w
Example – 1
n →∞
x 3 − 7 x 2 + 15x − 9
(c) lim
x →3 x 4 − 5x 2 + 27 x − 27
MATHS / LIMITS
LOCUS 22
Solution: (a) This limit is of the indeterminate form ∞ − ∞ . Combining the two fractions in this limit should lead
to a cancellation of the factor giving rise to this indeterminacy, i.e. (x – 2)
2 1
lim − 2
x → 2 x( x − 2 ) x − 3x + 2
2( x − 1 ) − x ( x−2)
= lim = lim
x→2 x( x − 2 )( x − 1 ) x → 2 x( x − 1 )( x − 2 )
1 1
= lim =
x→2 x( x − 1 ) 2
(b) Before trying to solve this, try to feel that this expression will have a finite limit even though the
number of factors being multiplied tends to infinity. This is because the successive factors become
closer and closer to 1 and their ‘contribution’ to the final product becomes smaller and smaller
1− x
m
Now, to simplify this product, we multiply it by . This is what happens:
1− x
co
n
2 4 2
(1–x). (1+x) (1+x ) (1+x ) ...... (1+x )
lim
n (1–x).
g.
8
ng
n
2 2 4 2
(1–x ) (1+x ) (1+x ) ...... (1+x )
= lim
n (1–x)
8
ye
` ` 4 4 2
n
(1–x ) (1+x ) ...... (1+x )
.m
= lim
n 0 (1–x)
w
n +1
1 − x2
w
= lim
n →∞ 1+ x
w
n+1
Since | x | < 1,lim x 2 =0
n →∞
1
Hence, the value of the limit is
1− x
(c) The numerator and denominator both tend to 0 as x→3 because of the common factor (x – 3).
Hence, factorization leads to :
( x − 3 ) ( x2 − 4 x + 3 ) x2 − 4 x + 3
lim = lim 3
x → 3 ( x − 3 ) ( x 2 − 2x 2 − 6 x + 9 ) x → 3 x − 2x 2 − 6 x + 9
There is still another common (x – 3) left in both the numerator and the denominator.
Factorization again leads to
( x − 1 )( x − 3 ) x −1 2
lim = lim =
x →3 ( x − 3 )( x + x − 3 )
2 x →3 x + x − 3
2
9
MATHS / LIMITS
LOCUS 23
Example – 2
(a) lim 1 + 2x − 3
x →4 x −2
(b) lim x − x
2
x →1 x −1
x2 + 1 − 3 x2 + 1
(c) lim
x →∞ 4
x4 + 1 − 5 x4 + 1
( x + x 2 + ...... + x n ) − n
(d) lim
x →1 x −1
m
Solution: (a) This limit can evidently be solved by rationalising both the numerator and the denominator.
co
g.
ng
2x − 8 x + 2
= lim ×
ye
x − 4 1 + 2x + 3
x →4
x +2
.m
= 2 lim
x →4 1 + 2x + 3
w
4
=
3
w
x 4 − x x + 1
= lim × 2
x →1
x − 1 x + x
x + 1
= lim x ( x 2 + x + 1) × 2
x →1
x + x
=3
MATHS / LIMITS
LOCUS 24
∞−∞
(c) This limit is of the indeterminate form (and look very complicated !)
∞−∞
However, division of both the numerator and denominator by x directly reduces the limit to a
determinate form.
x2 + 1 x2 + 1
− 3
x2 x3
= lim 4
4 x +1 − 5 x +1
x →∞ 4
x4 x 5
1 1 1
1+ − 3 +
x2 x x3
= lim These two terms tend to 0 as x → ∞
m
x →∞
4 1 1 1
1+ − +
x x 5
5
x4
co
=1
g.
(d) since the denominator is x–1, we can get a hint that the numerator
(x + x2 + ...+xn) – n
ng
can be written as
(x – 1) + (x2 – 1) + ... + (xn – 1)
ye
so that
(x + x 2
+ ... + x n ) − n x − 1 x2 − 1
= lim + + ... +
xn − 1
lim
.m
x →1 x −1 x →1 x − 1
x −1 x −1
= 1 + 2 + 3 + ...+ n
n ( n + 1)
w
=
2
w
w
Example – 3
x2 − 1
Find the values of a and b if lim − ( ax + b ) = 2
x →∞
x+1
Solution: The limit can be rearranged as
lim {( x − 1) − ( ax + b )}
x →∞
= lim
x →∞
{(1 − a ) x − (1 + b )} = 2
Since the limit is finite, the coefficient of a has to be necessarily 0.
Therefore
MATHS / LIMITS
LOCUS 25
Example – 4
Solution: One might say that since the terms in the numerator are all integral, the numerator is not continuous
and hence the limit will not exist. However note first of all that the limit is on n, and the secondly,
addition of a large number of integral terms in the numerator ( n → ∞ ) would tend to ‘overcome’
or ‘make negligible’ the effect of fractional parts that would otherwise have been present had there
been no greatest integer functions on any of the terms. This implies that whether I consider
m
Hence, the limit in question is equivalent to
co
x + 2 2 x + 32 x + ...n 2 x
lim
n →∞ n3 g.
x (1 + 2 2 + 32 + ...n 2 )
= lim
ng
n →∞ n3
x n ( n + 1)( 2n + 1)
ye
= lim
6 n→∞ n3
x 1 1
.m
= lim 1 + ⋅ 2 +
6 n →∞ n n
x
w
=
3
w
For students who like more rigor here is the proof of the above result using Sandwich theorem
(In this proof, it will become clear that the effect of the fractional part’s is negligible as n → ∞
w
MATHS / LIMITS
LOCUS 26
x x
It is easy to that the left and right limits are both , and hence the centre limit is also .
3 3
Example – 5
Evaluate the following limits:
cos x − sin x
(a) lim
π π
− x ( cos x + sin x )
x→
m
4
4
co
cos x − cos a
(b) lim
x→ a x−a g.
x − sin x
(c) lim
x →0 x3
ng
x x x
(d) lim cos ⋅ cos 2 ...cos n
n →∞
2 2 2
ye
Solution: Before proceeding to solve these limits, it should be mentioned that most limits can be evaluated in
more than one manner. (In fact, L’Hospital’s rule is a technique that can be used to solve most
.m
indeterminate limits: we will discuss it in detail in the topic of differentiation). In these examples and
the ones that follow we will be discussing multiple ways for solving limits, wherever they are
w
important.
w
π
4 − x ( cos x + sin x ) − x (1 + tan x )
4 4
π 1 + tan h
Let x = + h, so that tan x = and the limit above becomes
4 1 − tan h
1 + tan h
1 − 1 − tan h 1
lim ⋅ lim
h →0
−h x → 4 1 + tan x
π
−2 tan h 1 tan h 1
= lim ⋅ = lim . =1
h →0 − h (1 − tan h )
2 h →0
h 1 − tan h
MATHS / LIMITS
LOCUS 27
Alternatively, we could proceed by multiplying the numerator and denominator of the original limit
by (cosx + sinx):
cos x − sin x cos x + sin x
lim ⋅
x→ π cos x + sin x
π
4 − x ( cos x + sin x )
4
cos 2 x − sin 2 x
= limπ
x→ π
− x (cos x + sin x )
2
4
4
cos 2x 1
= limπ ⋅ lim
x→ π
− x x → 4 ( cos x + sin x )
π 2
m
4
4
co
= g.
sin 2h
= lim
ng
h →0 2h
=1
ye
x→ a x−a
x + a x − a
w
−2 sin sin
2 2
w
= lim
x→ a
(
x+ a )(
x− a )
w
C+D C−D
{We used cos C – cos D = −2 sin sin
2 2
x − a
sin
sin a 2
= − ⋅ lim
x− a
2 a x → a
2
sin a
= −
2 a
MATHS / LIMITS
LOCUS 28
(c) This limit can be solved very easily by using the expansion series for sin x :
x3 x5
sin x = x − + − ...∞
3! 5!
x3 x5
− + ...∞
x − sin x
Therefore, lim 3
= lim 3! 5!3
x →0 x → x
x 0
1 x 2 x4
= lim − + − ...∞
x →0 3! 5! 7 !
1 1
= =
3! 6
m
Alternatively, we know that sin 3x = 3 sin x – 4 sin3 x
co
Hence, 3x – 3 sin x = 3 x – sin 3x – 4 sin3x
x − sin x 1 3x − 3 sin x
L = lim = lim
x →0 x 3
3 x →0 x3
g.
3x − sin 3x 4sin3 x
ng
1
= x →0
lim −
3 x3 x3
ye
1
⋅
(3x ) − sin (3x ) − 4sin3 x
= 3 x →0
lim 27
(3x )
3
x 3
.m
1 θ − sin θ sin x
3
→0 θ3 x →0
x
w
4
=9L–
w
3
4 1
Hence, L = 9L – or L =
3 6
(d) As in Example - 1 Part - (b), we try to reduce this expression into a closed form by multiplying it
with an appropriate factor as follows:
MATHS / LIMITS
LOCUS 29
combine:sin
2θ = 2 sinθ cos θ
x x x x x
cos ⋅ cos 2 ...cos n −1 ⋅ 2 cos n ⋅ sin n
2 2 2 2 2
= lim
n →∞
x
2 sin n
2
combine
x x x x
cos 2 ⋅ cos 2 2 ...cos 2 n −1 ⋅ sin 2 n −1
lim
m
= n →∞ x
2 sin n
2
co
x x x x
g.
cos ⋅ cos ...cos −
⋅ sin
2 2 2
2 n 2
2 −2
n
= nlim
ng
→∞
x
2 2 ⋅ sin n
2
:
ye
:
:
.m
sin x
= lim
n →∞ x
2 n sin n
2
w
w
sin x sin x 1
= lim = ⋅
n →∞ x x
w
x
x ⋅ sin n sin 2 n
2 lim
x x
n →∞
n 2 n
2
x
Now, as n → ∞ , → 0 and hence
2n
x
sin
2n
lim =1
xn
n →∞
2
sin x
Therefore, our final result is
x
MATHS / LIMITS
LOCUS 30
Example – 6
(1 + x ) −e
1x
(c) lim
x →0 x
27 x − 9 x − 3 x + 1
(d) lim
x →0 2 − 1 + cos x
m
0
Solution : (a) The limit is of the indeterminate form , but can be reduced into a combination of two standard
co
0
limits as follows: g.
e x − cos x e x − 1 + 1 − cos x
2 2
lim = lim
x2 x2
ng
x →0 x →0
e x − 1 1 − cos x
2
= lim 2 +
ye
x →0
x x 2
e x2 − 1 2 sin 2 x 2
.m
= lim 2 +
4 ( x 2)
x →0 2
x
w
1 3
= 1+ =
w
2 2
w
x2 x3
(b) ln (1 + x) can be expanded as x − + − ...
2 3
Hence,
x2 x3
x − ln (1 + x ) 2 − 3 + ...
lim = lim
x →0 x2 x →0
x2
1
=
2
(c) The numerator in this limits tends to 0 as x → 0 because lim (1 + x ) = e.
1x
x →0
MATHS / LIMITS
LOCUS 31
Evaluating this limit will require a little artifice in the following manner:
ln(1+ x )
(1 + x )1 x − e e x − e
lim
x →0
= lim
x →0
x x
ln (1+x )
Now as x → 0, − 1 → 0 so that the numerator in the limit above is of the form e h − 1
x
where h → 0 .
m
ln (1 + x )
What should we do now? Multiply and divide by h h = − 1 .
co
x
We get
g.
ln(1+ x )
ng
−1
e x − 1 ln (1 + x ) 1
= e lim ⋅ − 1 ⋅
x →0 ln (1 + x ) x x
ye
− 1
x
eh − 1 ln (1 + x ) − x
.m
= e lim ⋅ lim
h →0
h x →0 x2
w
From the previous example, it follows that the second limit has the value − 1 .
2
w
e
Hence, the overall value for this limit is −
w
2
(d) Factoring the numerator and rationalising the denominator gives.
x x
(9 –1) (3 –1) 2+ 1+cosx
lim
x 0
2– 1+cosx 2+ 1+cosx
= lim
(9 x
− 1) ( 3 x − 1 )
⋅ ⋅
( 2 + 1 + cos x )⋅x 2
x →0 x x 1 − cos x
(We have multiplied and divided by x2 above)
MATHS / LIMITS
LOCUS 32
= ln 9. ln 3. 2 2 . 2
= 8 2 (ln 3 )
2
Example – 7
Evaluate the following limits:
1
a +b +c x x x x
(a) lim
m
x →0
3
co
1
π x
(b) lim tan + x
x →0
4
g.
1 x2
(c) lim sin x
ng
x →0
x
ye
x →0
.m
1
(e) lim ( cos x + a sin b x ) x
x →0
Notice that all the limits above are of the form (f (x))g(x) where f ( x ) → 1 and g ( x ) → ∞ that is,
w
Solution:
these limits are of the indeterminate form 1∞ .
w
In Section - 3, we saw how to evaluate such limits. Writing f (x) as (1 + h(x)) reduces this limit to
w
lim g ( x )⋅h( x )
e x→a , where h(x) = f(x) – 1.
We will now directly apply this result to evaluate the limits above.
1
ax + bx + cx x
(a) lim
x →0
3
1 a x + b x + c x
lim ⋅ −1
x →0 x 3
=e
x ( ) (
1 a −1 b −1
lim ⋅ +
x
+
c x −1 ) ( )
x →0 3 x x x
=e
1
((ln a )+(lnb )+(ln c ))
=e 3
= abc1 3
MATHS / LIMITS
LOCUS 33
1
π x
(b) lim tan + x
x →0
4
1 π
lim ⋅tan + x −1
x→0 x 4
=e
1 2tan x
lim ⋅ π 1 + tan x
Using tan 4 + x = 1 − tan x
x→0 x 1−tan x
=e
tan x 1
2lim ⋅
x→0 x 1− tan x
=e
2
=e
1
(c) lim
sin x x2
m
x →0
x
co
1 sin x
lim ⋅ −1
x→0 x 2 x
=e g.
sin x − x
lim
x→0 x3
=e
ng
In example -5 Part -C, we considered the limit in the exponent above.
−1 6
The value of this limit is therefore e
ye
1
(d) lim (cos x ) sin x
x →0
.m
1
lim .(cos x −1 )
x→0 sin x
=e
w
2 x
−2 sin 2
lim
x→0 x x
2 sin cos
w
2 2
=e
w
lim ( − tanx 2 )
= e x→0 = e0 = 1 .
1
(e) lim ( cos x + a sin bx ) x
x →0
1
lim ⋅{cos x + a sinbx −1}
x→0 x
=e
cos x −1 sinbx
lim + ab
x →0 x bx
=e
−2 sin x 2 sinbx
2
lim ⋅ x + ab
x →0 4 ( x / 2 )2 bx
=e
1 sin ( x 2 ) sinbx
2
lim − ⋅ x + ab
x →0 2 ( x 2 )2 bx
=e
ab
=e
MATHS / LIMITS
LOCUS 34
Example – 8
xn
(e) lim
m
x →+∞ n!
1 1
co
Solution: (a) Notice that as x → 0, → ∞, that is , has no particular limit to which it converges. Hence
x x g.
1
sin keeps oscillating between +1 and –1 as x becomes smaller and smaller, i.e., x → 0.
x
ng
Therefore, the limit for this function does not exist.
1
This is also clear from the graph (approximate) of sin sketched below:
ye
x
.m
w
w
w
MATHS / LIMITS
LOCUS 35
1 1
(b) In this limit, in addition to sin , ‘x’ is also present. Thus, although sin remains oscillating and
x x
does not approach any particular limit, it nevertheless remains somewhere between +1 and –1,
and when it gets multiplied by x(where x → 0), the whole product gets infinitesimally small.
That is,
1
lim x sin =0
x →0 x
Again, this is evident from the graph below:
m
co
g.
ng
ye
.m
w
Although ln x and x are both tending to infinity, ln x increases very slowly as compared to x.
For example, when x = e10 , ln x is just 10. When x = e10000 (a very large number indeed !), ln x
w
is just 10000.
ln x
Therefore, decreases and becomes infinitesimally small as x → ∞, i.e.,
x
ln x
lim =0
x →∞ x
(We can also use the LH rule to evaluate the limit above: this rule will be discussed later)
MATHS / LIMITS
LOCUS 36
(d) Consider x ln x.
As x → 0 + , ln x → −∞ , so that this limit is of the indeterminate form 0 × ∞.
But as in parts (b) and (c), try to see that the product becomes infinitesimally small as x → 0 .
−10
For example, at x = e −10 , ln x = –10 and x ln x =
e10
−1000
At x = e −1000 , x ln x = (which is very very small)
e1000
Hence, here again, lim+ x ln x = 0
x →0
m
xn
co
For x > 1 we write as
n! g.
xn x x x x
= ⋅ ⋅ ...
n! 1 2 3 n
ng
Now, since x is finite, let N be the integer just less than or equal to x ; N = [x]
xn x x x x x x
ye
xn x x x
Thus lim = P lim ⋅ ...
n →∞ N + 1 N + 2
n →∞ n!
n
w
The product inside the limit consists of all terms less than 1. Also successive terms become smaller
w
xn
lim =0
n →∞ n!
MATHS / LIMITS
LOCUS 37
EXERCISE
1. Evaluate the following limits
xm − am n!
(i) lim (ii) lim
x→a x n − a n n →∞ ( n + 1) ! − n!
(iii) lim
( n + 1) ! + ( n + 2 ) ! (iv) lim
x
n →∞ ( n + 3 )! x →∞
x+ x+ x
m
1 − cos x cos 2x 1 − cos (1 − cos x )
(vii) lim (viii) lim
x →0 x2 x →0 x4
co
cos ( sin x ) − cos x 1 + x sin x − cos x
(ix) lim (x) lim
g.
x →0 x 4 x →0 x
tan 2
2
ng
x →∞
(xiv) lim
x→
π
3 α →β α2 − β2
6 − cos x
2
w
2 x −1 − 1
(xv) lim x tan 2x − 2x tan x (xvi) lim
w
(1 − cos 2x ) x →1 sin π x
x →0 2
w
tan π x
ln x − 1
(xviii) lim 2 − a
2a
(xvii) lim
x →e x − e x →a
x
x a x
1 1 1x + 2 x + ... + n x
(xix) lim sin + cos (xx) lim
x →∞
x x x →0
n
1
tan x x π
(xxi) lim (xxii) lim x tan x − sec x
x →0
x x→
π
2
2
2x − 1 sin (π cos 2 x )
(xxiii) lim (xxiv) lim
x →0 1+ x −1 x →0 x2
MATHS / LIMITS
LOCUS 38
6 x − 2 x − 3x + 1
(xxv) lim (xxvi) lim x cot π x
x →0 sin 2 x x →1
2x
x+3 cos x
(xxvii) lim 5 (xxviii) lim
(1 − sin x )
x →∞ π 2
x→ 3
2
1 1 a sin x − x sin a
(xxix) lim − (xxx) lim
x →0 x sin x x→ a ax 2 − a 2 x
sin 2x + a sin x
2. Find a and b if lim =b
x →0 x3
3. If Sn = sin θ + sin 2θ + ... + sin nθ , prove that
S1 + S2 + ... + S n 1 θ
lim = cot
n →∞ n 2 2
m
nθ
⋅ sin
( n + 1 )θ
co
sin
You may use this relation: S n = 2 2
θ
sin
2
g.
1
1
sin x
4. +
x →0
x
{(a − n ) nx − tan x} sin ( nx ) = 0 , find a.
ye
5. If lim
x →0 x2
.m
MATHS / LIMITS