UNIT 4 APPLICATIONS

Structure
'4.1 Introduction
Objectives
4.2 Applications to Geometry
4.2.1 Geometrical Meaning of the Derivative
4.2.2 Equations of the Tangent and Normal at a Point
4.2.3 Angle of Intersections between Two Curves
4.2.4 Differential Coefficient of the Length of an Arc of y - f ( x )
4.3 Maxima and Minima
4.3.1 Necessary Condition for a Maximum and Minimum
4.3.2 Geometrical Interpretation
4.3.3 Points where f ' (x ) Qes not exist
4.3.4 First Sufficient Condition for a Maximum or a Minimum-Change of Sign of
First Derivative
4.3.5 Rule for Finding Maxima and Minima
4.3.6 Second Sufficient Condition for Maximum and Minimum (Use Second
Derivative)
4.3.7 Greatest and Least Values of a Function in Closed Interval
4.3.8 Maxima and Minima : Problems
4.4 Indeterminate Forms
4.4.1 The Indeterminate Form OX)
4.4.2 The Indeterminate Form m / m
4.4.3 Further Indeterminate Forms
4.5 Summary
4.1 INTRODUCTION
In Unit 2, we defined the derivative of a function.We studied several important theorems
like Rolle's theorem, Mean value theorems etc. Differential Calculus is a very useful tool in
studying practical problems. In this unit we shall study some applications of the derivative
of a function. In Section 4.2 we propose to study application of derivatives to geometry. In
4.3 we shall discuss how derivatives can be used to determine the points where a
diffetentiable function has maxima and minima and how to solve practical problems
involving minimisation or maximisation of some functions.
Objectives
After reading this unit, you should be able to
*
find the tangent and normal to the graph of a given function at given points,
JC
find the angle of intersection between two curves,
*
find the curvature, radius, centre and circle of curvature of a curve,
*
locate the points where a function has a maximum or a minimum,
-
JC solve some problems when it is required to minimise or maximise a function, and
*
to evaluate limits of functions which have indeterminate forms 0/0, oo/oo, 0.a
00 -.a, oO, ooe, or lW.
APPLICATIONS TO GEOMETRY
In this section we study some applications of the derivative to geometry.
4.2.1 Geometrical Meaning of the Derivative
In Unit 1, we defined the derivative of a function y = f (x)as
We now examine the geometrical meaning of this definition. Let P (x, y) and
Q ( x + Ax, y + A y ) be two neighbowing points on the curve y - f (x) (Figure 4.1).
Draw PM, QN the ordinates through P and Q respectively. Let the secant PQ make an
angle 8 with the positive x axis. Draw PR perpendicular to QN. Then from Figure 4.1, we see
that
RQ = NQ-NR = NQ-MP = f ( ~ + Ax ) - f ( ~ ) i Ay
Hence
=
slope of the secant PQ.
I
In the limit A x -+ 0, the point Q -*P along the curve y = f (x) and the secant PQ becomes
the tangent at P. If the tangent at P makes an angle rp with the positive direction of the
1
x-axis, we get
=
slope of the tangent at P.
The equation ( 4.1 ) shows that
representi the slope of the tangent to the curve y = f (x) at any point (x, y ). The above
discussion shows that at any point of the graph of a differentiable function y = f (x), the
slope of the tangent is defmed and is equal to the derivative of the function at that point.
This means that if y = f (x) is differentiable, then the graph of y = f (x) has a tangent at
every point.
(i)
If&/& - 0 at a point P, then tan $= 0. Hence v= 0. This means the tangent
to the curve at P is parallel to the x-axis. Conversely if the tangent at a point is
parallel to the x-axis then dy/dr = 0 at that point, (Figure 4. 2 (a))
F i i 4.2(8)
dy/& = 0, tangent parallel to x-axis
(i)
If &/dx is infinite at a point P then tan *= oo or *= n/2, so that the tangent is
parallel to the y-axis at that point, (Figure 4.2 @)).
Figore 4.2(b)
dy/& infinite, tangent parallel to y-axis.
4.22 Equations of the Tangent and Normal at a Point
You know that the equation of the line through (xl, yl ) with slope rn is
Y -Y1 = m (x -XI)
If this line is the tangent toy = f (x) at (xl, y, ) then we must have
where (%)(
is the value of the derivative at (x,, y, ). Hence the equation of the tangent
*lYl)
to the curve y = f (x) at (XI, yl ) is
The normal at P(xl, yl) is perpendicular to the tangent at P. The normal at P (xl, yl)
therefore has the slope
[ Recall that if two straight lines with slopes m and m' are perpendicular to each other then
mm' = - 1 or m' = - l / m] . Hence the equation of the normal at ( xl, yl ) toy = f ( x) is
Example 1 :
Find the equations of the tangent and normal to the parabola y2 = 4ax at the point
~ 1 ) -
Solution :
From y2 = 4ax, we find
This gives
The equation of the tangent is (by equation (3.2)
YY, = - 2 ~ 1 + Y:
Since (xl, yl ) is a point on the parabola, we have y: = 4arl. Using this we can
write the equation of the tangent as
YYl = ~ - - ~ l + ~ l
/
or
YY, = k ( x + x 1 )
Similarly the equation of the normal by ( 4.3 ) is
Example 2 :
Find the points where the tangent to the circle x2+ y 2 - 25 is parallel to the line
2 r - y + 6 = 0.
D i l l c ~ c ~ t W ~ Solution :
From
x2+ y2 = 25, we get
So the slope of the tangent at any point on the circle is - - . The slope of the
( 3
given line is 2. So the tangent to the circle is parallel to ibe given line if
X
- - - 2 or x = - 2 y
Y
Solving this and x2 + y2 = 25 we get the required points as (- fi, a) and
(m, -n).
/ El
Find the equation of the tangent and the normal at any point of the ellipse
n n
E2
Find where the tangent.is parallel to the x-axis for tbe curve
y3 = x2(2-X)
4.23 Angle of Intersections between Two Curves
Two cutves may intersect at one or more points. The x-coordinates of the points of
intersection of two cvfves given by the equations y - fl (x) and y - fi (x) can be obtained
by salving the equation f, (x) - fi @). They -coordinates of the points of intersection can
be obtained by putting thmvalues of x m any of the equations y = fl (x) or y - f2 (x).
7
If A Is a point of iutemaction of two curves, then the angle betwan the two curves at A is
de6ned to be the acute angle between the tangents to the two a w e s at A. Let C1 and C2 be
two cutves ink- at A and let a be the acute angle (Figure 4.3) between the tangents
to Cl and C2 at A.
Let ml = slope of the tangent ATl = @/dr for the first curve C1 : y = fi (x) at A and
t
m, - slope ofAT2 - dy/& for the second curve C2 : y - f2 ( x ) at A.
I
Hence
I
where the derivatives on the right hand side are to be evaluated at the mint of intersection.
Remark 2 :
If a = 0, the two cutves have a common tangent. The two curves are said to touch
C
each other if they have a common tangent.
Remark 3 :
X
If a = - the two curves intersect orthogoually. In this case tan a = 00 or
2'
m1m2 = -1.
3
Show that the ellipse - +
- 1 and the hyperbola 3 -3 = 5 cut orthogonally.
18 8
Solution :
We first find the points of intersection of the two cutves. To do this we solve
simultaneously
I
I and
Substituting this in (2) we get
X ' *3
Putting these values of x in (2) we get,
9 - y 2 = 5 0 r y 2 - 4 0 r y ' * 2
Hence the points of intersections are
A (3, 2), B(3, -2), C(-3, 2)andD(-3, -2)
Daerentiating (1) we get
Let ml and m2 denote the slope of the tangents to the curves (1) and (2)
respectivd y.
Similarly
Differentiating (2), we get
and
3
q m - a t D
2
Hence the Wo curves cut orthogonally at A.
Similarly you can verify that the two curves cut orthogonally at all the other points
of intersection.
Show that the curves
x2 "' + L = 1
- +L= 1 and -
a2 b2 a 2 + A b 2 + A
intersect orthogonally.
t
Find the angles of intersection of xy = 10 and 3 + 9 = 29.
I
4.2.4 Differentia1 Coeflicient of the Length of an Arc ofy = f ( x)
When we talk about the distance between two points P and Q, we usually mean the distance
measured along the straight line joining P and Q. However, it is sometimes necessary to
I
know the distance between two points along a curve joining the points P and Q, Figures 4.4 ( a)
and 4.4 @).
I
Figure. 4 4 (a) F i 4.4(b)
1
Distance measured along a st. line joining P (x,. y, ) and Q (x,, v, 1
Distance 's' between P and 0 measured alone the arc v = f (XI is eiven bv
To see this, let P ( x , y ) and Q ( x + Ax, y + Ay ) be two neighbouring points on the curve
y = f (x). LetA be some fured point on the curve y = f (x) from which we measure the
distance of any other point on the curve. Let s denote the length ofAP measured along the
Applications
arc of y = f (x). We call 's' the arc length of the arcAP(Figure 45). Then the arc length
AQiss+Asandthean:lcngthPQish.
As Ax + 0, the point Q tends to P along the arc. The chord PQ and the arc PQ become
almost equal in the limit. Hence
Now &om the triangle PQR
Hence
2
Therefore
k - 0
On using (4.3) we get
In the same way, we can show that
4 s .
We can write equation ( 4.4 ) or ( 4.6 ) in terms of differentials as
d2 = dr 2 + dy 2
'
If y, is ise angle made by the tangent at P with the positive direction of the x-axis,
(Figure 4.5)' it follows that
PR dx
cosy, = lim - = -
Q-PK? 'Ik
and similarly
I
sin * = 4 ds
Example 4 :
Given the cycloid x = a (1 - cos 0 ), y = a (0 + sin 0 ), find d s / L and ds/dy.
Solution :
x - a ( l - me ) , y = a(8+sin0)
4
4-e = l + W 0 0
dr dr
= cot-
-
sin 0 2
Therefore
E5
Find cis/& and ds/dy, given that
1 1
(i) .x - -(2t+3)3n,y = 2 * + I
3
4.3 MAXIMA AND MINIMA
In this sitdion we shall study how to locate the points where a difZerentiable function has its
maximum or minimum.
We shall also wnsidtr some practical problems involving the theory of maximum and
minimum of a function.
I
We say that a function f has a muximum at a point x I. c if, for all x in a
sufficiently small neighbourhood of c,
I
We say that a function f has a minimum at a pointx = d iffor all x in a
sufficiently small neighbourhood of d,
f ( 4 r f (4 (4.8)
L'
In Figure 4.6 the graph of some function f in the interval [ a, b ] is shown.
LetA be the point x = c. As you see, at this point f has a maximum because at
x = c the value f (c) of the function (the 0rdinateA.P) is greater than the value
of the function f (x) for all values of x in a small neighbourhood ofA.
The dotted line near the ordinate AP indicates that all other ordinates at points near
toA are maller than AP.
Simlarly the point B, where x = d, the function f bas a minimum because in the
nei@bourhood of x = d the value of the fundon is greater than f (d). The ordinate
BQ is smaller than any other ordinates at points in a small neighbourhood of B.
Tbis is indicated by the doned lines.
From the f@ue it is clear thatA is not the only point when? there is a maximum
and B is not the only point where there is a minimum. The function has maximum
atAbAz and it has minimum at B1, B2 Also note that the value of the function at
Al (which is a maximum) is less than the value of the function at B.(which is a
minimum).
Thus there can be several maxima and minima in an interval.
In ordinary language, the expression maximum and minimum are used to mean the
greatest value and the least value respectively. For us the meaning of maximum (or
minimum) is not the same, All we need for a function f to have a maximum at
x - c is that f at x - c should be greater than all other values off in some small
neighbourhood of x = c.
Note that the greatest value is automatically a maximum but a maximum need not
be the greatest value of the t i kt i qn in its domain. Similarly a minimum is not
necessarily the smallest value ofthe function in its domain.
In order to distinguish from the greatest value the maximum is sometimes called a
relative or local marimurn and the greatest value is called absolute or global
maximum. Similarly a minimum is called relative or local minimum and the least
value is called the absolute or global minimum. The term extremum (plural
extrema) or extreme value is used to mean both maximum and minimum.
Example 6 :
2
Let f ( x) = x , - 1 s x s 2. The graph off is a parabola, (Figure 4.7). Since f ( x) r 0
for all x in -1 s x s 2 and f (0) = 0, we conclude that f has a minimum at x = 0.
Since this is the only minimum, x = 0 is the absolute minimum. The function has
maxima at the two end points x = -1 and x = 2. At the point x = 2 there is an
absolute maximum.
4.3.1 Necessary Condition for a Maximum and Minimum
In this section we give a method for finding possible points of the maxima and minima of a
L
function.
Theorem 1 :
1
Iff is differentiable at x = xl and iff has a maximum or a minimum at x = xl then
I f ' ( x1) - 0.
Let us assume that f has a maximum at the point x = xl. Then for small h (both
positive and negative), we get
f ( XI + h ) s f ( ~ 1 )
that is,
f + h - f ( xl ) 5 0.
Hence
and
Now
If h +O through positive values then from(1) and (3), we see that
f (XI) 8 0 (4)
If h -0 through negative values then from (2) and (3), we see that
Since f (x,) is a definite number, it is independent of the way h -, 0. From the
inequalities (4) and (5) we see that only possibility is
f (XI) = 0
Similarly we can show that when f has a minimum at x,,
f ( ~1) a 0
Note :
The theorem does not say that iff (xl) = 0 then f has a maximum or a minimum at
x - x,. In fact, even iff (x,) = 0, there may not be any maximum or minimum as
the following example will show.
Example 7 :
Thenf(x) = 32and f (0) = 0
Butf(x) <Oifx<O and f(x)zOi fx > 0.
Therefore we cannot fmd any neighbourhood of x = 0 where
f (0) a f ( 4 or f (0) f (4
Hence x - 0 is neither a maximum nor a minimum even though f (0) = 0. Thus
f(x,) = 0 is a necessary condition for f to have a maximum or a minimum at
x = x, but it is not sufficient
4.3.2 Geometrical Interpretation
We know from Section 3.2 that if @/& = 0 at a point x = x, then the tangent to the curve
y - f (x) is parallel to the x-axis. From theorem 3.1, we know that iff has a maximum or a
minimum at a pointx - x, and iff (xl) exists, thenf (x,) = 0. Combining these two
results we get the following obvious geometric fact :
Iff is diffementiable at a point x = xl and iff has a maximum or a minimum at x = x,, then
the tangent to the curve y = f (x) at x - x, is parallel to the x-axis.
4.33 Points where f ' (x) does not exist
From theorem 3.1 we cannot say anything about points where f is not differentiable. There
may or may not be any extremum at points where the fundion is not differentiable.
Example 8 :
(i) YOU know the function y = 1 x 1 is not differentiable at x - 0. Its graph is
shown in Figure 4.8 (a). The fundion has a minimum at x = 0 but f ' ( 0 )
does not exist.
i
(ii) Let y = (1 -3'4 312. The graph of the function is shown in Figure 4.8 0). For
this function
At x = 0, dy/& becomes infinite. So the function does not have a derivative at
x = 0. ~ o k e v e r at x - 0 there is a maximum.
(iii) Let f (x) = xV3. Its graph is shown in Figure 4.8 (c). We have
Hence &does not exist at x = 0.
dx
Nowf(x) < 0 if x < 0
andf(x) > 0 if x > 0
Hence in any neighbourhood of the point x = 0, there are points x such that
f (x) < 0 and f (x) > 0. Therefore x = 0 is neither a maximum nor a minimum.
4.3.4 First Suflicient Condition for a Maximum or a Minimum -
Change of Sign of First Derivative
We say that a point x = x, is a critical point off if either
(i) f' (x,) = 0 or
(ii)
f' (x,) does not exist.
In Section 4.3.1 we saw that iff is differentiable at xl and iff has a maximum or a minimum
at x = X, then f (x,) = 0 (Theorem 1 ).
In Section 4.3.3 we saw that f can have mqimum or minimum at points wheref' does not exist.
Combining these two observations we conclude that if a point x = xl is a point where f has
a maximum or a minimum, then x = xl must be a critical point. However, at all critical
points the function need not have a maximum or a minimum. This was illustrated in
Example if.
The following theorem tells us when a critical point is a maximum, a minimum or neither of
the two.
Theorem 2 :
(Sufficient conditions for existence of extremum.)
Let
(i) x, be a critical point of the function f (i.e f(xl) = 0 or f(xl) does not exists).
(ii) f be continuous in an interval containing the critical point x,.
(iii) f(x) exist at all points of the interval except possibly at x = xl.
(1) Iff (x) > 0 for x < xl and f' (x) < 0 for x > xl in the interval, then f has a
maximum at x - x,.
(2) If f(x) < 0 for x < xl and f' (x) > 0 for x > xl in the interval, then f has a
minimum at x = x,.
Remark :
(1) The function may or may not have a derivative at x = x,.
(2) Aceording to theorem 2, the test of maximum or minimum is the change of
sign off(x) when x crosses xl from left to right. There is a maximum at x = xl
iff(x) changes its sign fiom positive to negative and there is a minimum if
f(x) changes its sign from negative to positive.
Iff'Cr) does not change sign while crossing xl from left to right, then the function
has neither a maximum nor a minimum at x = x,.
Froof of Theorem 2 :
Let f (x) > 0 for x < xl and f (x) < 0 for x > xl for all x sufficiently close to x,.
By Lagrange's mean value theorem (Unit 2, Theorem 2) we have
f (x) - f (xl) 5 (X -xl) f (c) where c E (x, x,) or c E (x,, x), depending upon
whetherx<xl, orx>x,.
Case (i) Let x < xl. Then c < x,, f (c) > 0, f (c) (x -x,) < 0 and co~~sequel~tly
Case (ii) Let x > x,. Then
and, consequently,
f (x) -f ( 4 < 0
From (1) and (2) we see that f has a maximum at x = xl.
In a similar way it can be shown that iff (x) changes its sign from negative to
positive whenx crossesxl 6rom left to right, then f has a minimum at x = xl.
4.3.5 Rule for Finding Maxima and Minima
Using the results of the last section, we have the following steps to test a differentiable
function for maximum and minimum:
(1)
Solve f ' (x) = 0.
(2)
FiOd the points where f (x) does not exist.
(3)
The steps (1) and (2) give the set of critical points off The points where f has ~~l a xi ~na
or hinima belong to this set.
(4)
Inyestigate for every critical point x = x, to see whether the derivative/' (x) changes
sign when x moves from left to right ofxl. Iff (x) changes from positive values to
neglative values then f has a maximum at xl. Iff (x) changes from negative to positive
values then f has a minimum atx,. Iff (x) does not change sign then f does not have a
maximum or a minimum at x 1- x,.
Iff (x) remains positive (i.e., f' (x) > 0) when x crosses xl from left to right then f increases
I at x,.
Similarly iff' (x) remains negative (i.e. f' (x) < 0 ) when x crosses xl from left to right, then f
decreases at xl.
We summarise the above discussions in the following table:
Example 9 :
Find the maximum and minimum off ( x) = x2-x.
Solution :
To solve this problem we follow the steps discussed above.
Step 1
We observe that f is differentiable at all points.
Step 2
f'(x) = k- l,f'(x) = 0 givesx = In.
Step 3
So the on1 y critical point is x, = 1/2.
Hence iff has any maximum or minimum, it must be atx = 1/2.
Step 4
1
We test the change of sign off (x) while crossingx = - tiom left to right. We
2
havef (x) = 2r - 1 = 2(x-1/2)
(i) f (x) < 0 if x < 1/2
and
(ii) f' (x) > 0 if x > 1/2
Therefore f has a minimum at x = 1/2.
1
The mlnimum value is f (1/2) = (1/2)' - = -1/4.
Example 10 :
Investigate the maximum and minimum of the function f ( x) - x5 - 5x4 + 5x3 - 1.
Solution :
(i) The function is differentiable at all points.
(ii) f' (x) = 5x4 - 20.r3 + 15x2
I = 5x2( 2- &+3)
I , = 5xZ(x-3)(x-1)
(iii) The critical points are given by f'(x) = 0.
or 5 x 2 ( x - 1 ) ( x - 3 ) = 0
or X = 0,x = 1,x = 3
(iv) We now test the critical points,
The critical point x - 0.
When x < 0,'we have
2 > 0 , x - 1 < 0 and x- 3<0.
Hence
When x crosses over x = 0 but remains very near x = 0, we have x > 0, x < 1, and
3 > 0 , x - 1 < 0 and x-3<O. Sof' (x) = 5?( x - 1) ( x - 3) >0 (2)
From (1) and (2) we see thatf' (x) remains positive where x goes from left to the
right of the critical point x = 0 and does not change its sign. Hence at x = 0, the
function has neither a maximum nor a minimum. The function increases at x = 0.
Critical point x = 1
Hencef (x) = 5 2 ( x - l ) ( x - 3 ) >O (3)
( +) ( - ) ( - )
When x crosses over x = 1 but remains very near x = 1, when have
2 >o , x - 1 >0 , x - 3 <0
Thus from (3) and (4), f ' (x) changes its sign ?om positive to negative. Hence at
x 1 there is a maximum.
Similarly it can be seen that at x = 3, the function has a minimum.
Example 11 :
Find all the maxima and minima off (x) = (x - 1) x?'~.
Solution :
We have
Hence f is differentiable at all points except at x = 0. We also see that
f' (x) = 0$t x = 2/ 5
Hence the critical points are x = 0 (where f is not differentiable), and x = 2/ 5
(where f (x) = 0).
2
Forx<O,x--<O and xm<O.
5
Hence for x < 0
5 (x - 2/5) (negative) >
= -j xVl (negative)
Farx >O but x < 2/5,lhat&fnO<x <2/5,webavex-2/5 <Oandxln >O,-
f ( x) <O
For x w 2/5, we have x - 2/5 > 0 and xln > 0 and so
f (x)>Oforxw2/5
Hence f (x) changes from positive to negative at x = 0 and Erom negative to
positive at x = 2/5. We conclude that at x = 0 the function has a maximum and
at x = 2/5, a minimum.
43.6 Second Sumcient Condit on for Maximum and Minimum (Use of
Second Derivative) 1 I
In Sedion 4.3.3, we saw how a txitical point can be tested by observing the change
'
of sign of f'(x) when x passes from the left to the right of a critical point. If the
function is complicated, it is sometimes difficult to distinguish the sign of the
derivative. We can use the following theorem to test a critical point.
Theorem 3 :
Let f bt a function such that f ' (x) - 0.
Then at x = xl, f has a maximum iff " (xl) < 0 , a minimum iff " ( xl ) > 0. If
f " ( xl ) = 0 no conclusion can be made.
Proof :
Let f '( q ) < 0.
Since f " (x) is the derivative off'@), it follows that f ' ( x ) is a decreasing function
at the pointx = XI (By theorem 5 of 3.3). And since f (xl) = 0 we must have
f' (xl) > 0 for x slightly less than x ; and < 0 for x slightly greater than xl. Hence
by our first sufficient condition (Theorem 3.2), f (x) has a maximum at x = xl.
It can be similarly seen that whenf'(xl) '= 0 and f " ( xl ) > 0 , the function has a
minimum at x = xl.
When f " (xl) = 0, the function may or may not have maximum or minimum at
x = xl. In such a situation we can use the fmt method (Theorem 3.2).
We summarise the results in the following table.
Example 12 :
\ %
f ( 4
0
0
0
Find the points of maximum and minimum fofthe function of Example 10 by
using second derivative test. c
Solution:
We have (Example 10)
f(x) = x5-5x4+5x3-1
f ' (x) = 5x4 - 202 + 15x2
f" (XI)
< 0
w 0
0
Conclusion
Maximum
Minimum
Unknown
The critical points are x = 0, x = 1, x = 3.
f"(x) = 2&--&+30x
x = 1, ff' (1) = 20-60+30 = -10<O
Therefore by theorem 3, at x = 1, the function f has a maximum.
At x = 3, f " (3) = 20.3~ - 60.3~ + 30.3 - 540 - 540 + 90 = 90 > 0
Therefore at x = 3, the function f bas a minimum. At x = 0, f'(0) = 0 . Hence
theorem 3.3 tells us nothing about the nature of the critical poinf x = O.(However,
by observing the chame of sign off' (x) at x = 0, we have seen in the example 10
that at x = 0 there is neither a maximum nor a minimum.)
Example 13 :
1 Find the maximum value off' (x) = * for x r 0.
X
Solution :
Since f is differentiable whenx > 0, the only critical point is given by
1-10 x
the roots of f ,(x) = --A- I 0
3
Now f' (x) = 0 when 1 -1ogx = Oor whenx = e.
Hence x = e is the only critical point.
Now
Therefore the. function has a maximum at x = e. The maximum value is
4.3.7 Greatest and Least Values of a Function in a Closed Interval
Let f be cofitinuous in a closed interval [ a, b 1. Then we know (see Unit 2 ) that f always
has a greatat value and a least value in this interval [ a, b 1. The greatest ( or least ) value
may be attained either at an interior point of the interval or at the end points x = a (or
x = b) of the interval.
It is easy to verify that
i. f has a maximum at x = a if f'(a +) < 0.
ii. f has a minimum at x = a if f'(a +) > 0.
iii. f has a minimum at x = b if f' (b -) < 0.
iv. f has a maximum at x = b if f' (b -) > 0.
Therefore if we are required to find the greatest (or least) value of a continuous function in a
closed interval [a, b 1, we should consider the following steps :
(1)
F i n d l the maxima (or minima) in the open interval a c x < b. To do this we follow
the procedure descriid in Section 3.3.4. Find the values of the function at the
maximulm (or minimum) points so obtained.
( 2)
Find the value f ( a) and f (b) of the function at the end pointsx = a and x = b of the
interval [ a, b 1.
(3)
Now pick the greatest (or least) of all the values of the function so obtained. This
value will be the greatest value (or the least value) of the function in the closed
interval [ a, b 1.
Example 14 :
Determine the greatest and the least value of the function
f (x) = xS - 5x4 + 5 2 - 1
in the interval [O, 2 1.
In Example 12 we saw that the critical points of this function are
L
x = 0, x = 1 a n d x = 3
Since x 3 is outside the given interval [O, 2 ] an'd x = 0 is an end point, thus the only
critical point in the open interval 0 < x < 2 is at x = 1.
The values of the function at the end poinis are
f(O> = - 1
f ( 2 ) - 25 - 5.24 + 5.23 - 1
= 32- 80+40- 1
= - 9
Thus the greatest value off (x) in [O, 2 ] is 0 and the least value is - 9.
4.3.8 Maxima and Minima : Problems
The theory of maxima and minima developed in the previous sections provides a powerful
tool for solving problems that require minimising or maximising' certain functions. In ihis
section, we illustrate how this is done by solving some problems. We shall summarise the
technique at the end.
Example 15 :
Find two positive numbers such that their sum is 10 and their product is as large as
possible.
Solution :
Let one of the numbers be x . Then the other number must be 10 -x. The product
of the two numbers is
2
f ( x) = x( 10- x) = l o x - x
(1)
Since both the numbers are positive, we have
x>O and 10- x>0 or x<10.
We have to choose x in such a way that f (x) is maximum.
From ( I ) , we find
f ' ( x) = 1 0 - 2 = 2( 5- x)
Therefore f ' ( x) = 0 at x = 5
Now
f ' ( x) >O if x < 5 and
f (x)<O if x > 5
Hence f (x) has a maximum at x = 5.
Ifence the two required numbers are x = 5 and 10 -x = 5.
A cyhhical dontainer is to be made with capacity 1000 cubic metre. The
material for the side costs Rs.2001- per square metre and for the ends Rs.1501- per
squut metre. Find the radius of the base so that the cost of material for making the
container is least
Let the cylinder be of radius rand height h,
Its volume is
V - x?h - 1000 (given)
The area of the side is
A, - 2zrh
Cost of material for side = Rs. 200. A1 - 2z rh . UK)
(2)
Amofbottomend- x?
So that sum of the areas of the two ends = 2n r
Cost of material of the two ends = 2z r .I50
Hence total cost [ adding (2) and (3) j
C = 2~?. 150+2z r h. 200
= 300~?+400xr h
(Substituting the value of h from (1))
= 0 (when C is minimum)
Thisgives
So radius of the base is 5% an
(4,)
Note that
for the value of r . Hence the cost C is minimum.
The cost of running an eagiue is proportional to the square of its speed and is
Rs. W- pa hour for a speed of 16 m.p.h. Other expenses amount to Rs. 3001- per
bout, What is the most eumomical speed ?
Let v1km.p.h be t&e speed If G is the cost of nrnning the engine then C = K?
The running cost per hour is=
I If the distance travelled is s km, then number of hours the engine is run is 7 s
Hence the total cost of running is
For the most economical speed we must have
= 0 and $ 0.
dv
gives v = 40
Hence the most economical speed is 40 km.p.h.
Example 18 :
The efficiency of a screw is given by
where t is the tangent of the angle and p is the coefficient of friction. Find for what
value of t efficiency will be maximum.
Solution :
Since p is a constant, E depends only on t.
Now
gives
The maximum eaciency is obtained by the value of t obtained from this equation.
Note that minimum efficiency is zero. Hence the value of t obtained from (1) can
give only maximum efficikncy.
The examples that we have worked out shows that when solving a problem that
requires maximising or minimising a function, the following steps are to be
followed :
(1) If possible draw a figure describing the given data.
(2) Write an equation for the quantity to be maximised or minimised.
(3) NLW find the critical points and test the critical points as usual.
(4) The critical points will lead to the solution of the problem.
Find the maximum and minimum of
2 - 3x + 3 on the interval [-3,3 / 21.
(iv) y = (x - 1) (x - 212.
E7
Show that f ( x) - sin x (1 + cos x) has a maximum at x .I x/3.
Find the extreme values of f ( x ) - sin 2r for 0 r x r 2 R.
E9
Show that the maximum rectangle that can be inscribed in a circle is a square.
Show that of all the rectangles of given area the square has the smallest perimeter,
/
Show that the conical tent of given capacity will require the least amount of canvas
if its height is fl t i mes its base radius.
h open storage bin with square base and vertical sides is to be constructed from a
given amount of material. Determine its dimensions if its volume is to be
maximum neglecting the thickness of material and waste in constructing i t
Find the height of a right cylinder with greatest lateral surface area that may be
inscribed in a given sphere of radius R.
Given a point on the axis of the parabola 4 = 2px at a distance a fiom the vertex,
find the absdssa of the point of the curve closest to it.
l l i
El5
A load w is hoisted by a levee a f m F is applied to one d, the point of support
is at the other cad of the lever. If the load is suspcdd from a point a cmtimtters
6rom the fulcrum, and the lever rod weighs v grams per centimeter of length, what
should the lengtb of the rod be fot the torte (required to raise the load) to be a
minimum ?
4.4 INDETERMINATE FORMS
In this seation we introduce you to a rule that will be highly helpful in evaluating limits.
If we d e f k the function H by
sin x
, x*o, H Q - 7
then
(x) = g (x)
where f (x) - sin x, g (x) = x.
We axmot evaluate
l h HCr)
r--0
by using the quotient limit theofem
lim f (x)
lim H (x) =
r-0
r-0 I h g ( 4
x-0
lim g(x) - 0
X ~ O
since
However, in addition,
and therefore
lim H O
x-0
still might wrist Let us show how Cauchy's M.V.T. can be used to evaluate this limit
for some number z between 0 and x.
Applications
Since f (z) = cosz,gt(z) = 1,
H(x) = cosz = cosBx, where O<8<1.
Hence
sin x
lim - = lim cos8x = 1
2- 0 x x- 0
sin x 0
The function - is said to have the indeterminate form - at x = 0.
X 0
00
The other indeterminate forms are 0 . 00, 00 - 00, 0', a', 1".
4.4.1 The Indeterminate Form 010
The following rule allows us to investigate limits of functions possessing indeterminate
form om.
Theorem 4 :
(L' Hospitals' Rule 010)
Let the functions f and g be differentiable at every point other than possibly a in
some interval, with gl(x) * 0, if x a.
If
lim f (x) = 0 = lirn g (x),
x- a I - a
Then lim fkl = lim f"
x- a g(x) x- a g' (x)
provided this latter limit exists or is infinite.
(The rule is named after a French mathematician G.FA. L' Hospital
( 1661 - 1704)).
The proof is based on Cauchy's Mean Value Theorem, as in the above example.
However, we leave it as an exercise for you.
Example 19 :
1- cosx
Evaluate lirn -
x- 0 X + X
2
Solution :
Since
lim (1-cosx) - 0 = lirn ( x+x2) ,
x- 0 x - 0
We apply L ' Hospital's rule.
sin x
= lim - = 0.
,-0 1 + 2x
It may happen that in applying L' Hospital's rule we fmd that
lim f ' (x) = lirn g ' (x) = 0
x- 0 x- 0
If this happens, we may apply L ' Hospital's rule again.
Example 20 :
sinx-x
Find lim
x - 0 im
Solution :
lim (sinx-x) = lim? - 0
x - 0 x - 0
Therefore, by L' Hospital's rule,
cosx-1
- lim
x - o 3 2
we use L ' Hospital rule again.
Iim -2-l
[+I
x - 0
Hence
sinx-x 1
lim -
I --
x - 0 ? . 6
It may happen that one of lim f' (x) and lim g' (x) is zero. while the other is not.
x - 0 x - 0
Then
lim rn
x - 0 gl(x)
is either zero or infinity, as in the next example :
Evaluate
Solution :
cotx
1" -
- 5 t-;f
cotx
lim. 7 [ ; ]
x- T t-;I
- cscL x
= lim
2 3x- -
x- E
( ;s
=-.-
El6
Find the limits :
1- sinx
(9 limz1+cos2r
X- T
4.4.2 The Indeterminate Form
The symbol = is used to represent "infinity". However, infinity is not a real number.
The specific meaning of the expression
lim f ( x ) =
X- - a
t (f ( x) 1 r M for x satisfying
If we replace 1 f ( x) ('r M by f ( x) r M in the statement above, we say lim f ( x) = + m,
x- - a
and if we replace ( f ( x) ( < M by f ( x) < -M,
we
f ( x) = - co
if x -. 0 from the left.
While speaking of infinity as the limit of a functionf, we do not mean that the difference
between f ( x) and infmity becomes large, but rather f ( x) is numerically large when x is close
Theorem 5 :
00
(L ' Hospital's Rule 2 )
Let the functions f and g be differentiable' at every point other than a in some
interval, withg ' ( x) + 0 if x t a. If
lim f ( x ) 3 f and lim g' ( x) = * a,
X - a x - a
lim f o=l hf ' o,
x - a g( x ) x - a gl ( x )
provided this latter limit exists or is infinite.
Tbe rule in Tbeorem 5 is the same as in Tbeorem 4. In other words, L ' Hospital's
0 m
rule applies to tbe indeterminate form - as well as to the form 2
0
In x
Findlim -.
cotx
x -- o+
Applications
l a x
lim - = [ ; I .
+ Cotx
x-0
sin x
=-lim (smx)-
x-o+
X
Evaluate lim [x ( ln 1 x 1 )2 1.
x-0
lim [ X ( ~ I ] X ~ ) ~ ] [O.m].
x--0
- lim
x-0 l / x [ : I
1
2 ( l n l x l ) ;
= lim
1-0 1
-2
= lim
x-0 - - [=I
X
2
--
X
= lim -
x-0 1
-2
\
El7
Find the l i mb :
Cot&
(i) lim -
x-.o CotSx
4.43 Further Indeterminate Forms
You may use L ' Hospital's rule in the evaluation of limits of the form
lim fll
x- . g(x)
2+ 3x
Evaluate lim -
x - m 4 2 - 1
Solution :
lim ( @- I ) = lim 2
x-OD / x- .
2 + 3 x a
Therefore. lim - assumes the form -.
x- . 4 2 - 1 a
We apply L'Hospital's rule and have
2 + 3 x m
lim -
x- . 4 2 - 1 [a]
1
@) Note that the limit could have been found by writing x - - and letting t
t
2 + 3 x
lim -
3 ' t
- lim -
1+3t 1
lim -I -.
4
x- . 4 2 - 1 ,-0 t - 0 4 - 2 4
2
(c) Another method is to divide both tbe numerator and denominator by the
highest paws of x that occurs 'in either numerator or denominator.
3
2+3x
1 + -
X
- lim -
% x 4 - Y
1 1
Botb - and approach zero as x increases without bound.
x f
Therefore
Example 25 :
Find lim [L - L] .
x-o sin2x 2
Solution :
0
This is of the form a- w. In using L ' Hospital's rule, we shall reduce this into -
0'
lim [+-$I [=-=I
-. o sin x
2-2ws2.X
= lirn
x-0 12r2
1 -cos2r
- lim
x-o 6 2
2
1 sinx
- lirn - -
x-0 3 ( x ]
1
t -
3
Example 26 :
Show that Jim (2 + l)lAnX e2.
x - m
Solution :
0 =
This is of the f om aO. To reduce this to the fonn - or 3 we take the logarithm.
0
lirn ln @ + 1)
x - m
- lim
22
= lim 7
x-m x + 1 [ : I
4x
- lim -
.-.a 2x:
It follows that
lim ( x 2 + = e2.
x+m
El8
Find the limits :
(ii) lim 1 sin x
x + o
(iii) Iim I sec x lmx
x
X-T
(v) lim (2 + 1)
x +m
4.5 SUMMARY
The important points wvered in the unit are as under :
(1)
Given the curve y = f (x), represents the slope of the tangent to the curve at the
dr
point ( x , y ) .
(2)
The equation of the tangent to the curve y = f ( x) at the point (xl, y l ) is
(3)
The equation of the normal to the curve y = f (x) at the point (xl , yl)'is
(4) If s is the a d distance of a point (x, y) on the m e y = f @) from a fixed point on
the curve, then
(5)
A pi nt x = xl is a critical point off O if either f (xl) = 0 or f (xl) does not exist.
(6) f (x) has a maximum at x - xl iff (xl) = 0 and f" (xl) < 0.
(7) f (x) has a minimum of x = xl if)'(xl) = 0 a d f" (xl) > 0.
(8) If lim
0 00
mi s of t hehga3t henl i m fll= lim fo
x-a g ( x ) x-a B(X) x-a g ' ( ~) '
4.6 SOLUTIONS/ANSWERS
El.
* ' t L 1
3 b2
(3, x b2
*- = --
& Y ; ; I
I -- -
Xl 62
EquationofTangenty-yl - -- -(x-xl)
YI 3
1 1
Shilarly normal Equation + $ - 2 = xl yl ( 2 - ;;i )
a
E2.
y3 - 2 ( 2 - x )
dy 4r-32
T
32
tangent parallel a x -axis, 4 - O: at x = *): y -
&
E3.
Sbow as per given condition
tan a =
"'1 -m2
1 +mlm2
tan a =
-522 4% 21
= --
5 2 20
1 + - x -
2 5
similarly find angles of intersection at (- 5, - 2), (2 J), ( - 2, - 5).
ES.
ii) ( y + I )' = 4x3
- - y + l 4r3
=m
I
i dy
I iii)
= at2
dx
-
dt
= 2at
y = 2at + * = &
dt
4 = L
dx t
SO, - =
dx t
f ' ( x ) = 0 =>x = 0.-4
f " ( o ) > o Minimum
PI ( - 4) <0 Maximum
Local Minimum Value f ( 0 ) = 2
Local Maximum Value f ( - 4 ) = 4
f ( x ) = 3 2 - 3
f ' ( x ) = O * x = * l
f ( - 3 ) = - 15; f ( - 1) = 5
f ( 1 ) = 1 ; f ( 3 h ) = 1.875
Min = -15; Max. = 5
iii) - 1 - 2
f ( x ) - -32;
f " ( x ) - 6 ~
f l ( x ) = 0 * X = 0
f " ( 0) = 0
'v!, f ( x ) = ( x - 1 ) ( x - 2) 2
f ( x ) = ( x - 2 ) ( 3x - 4)
x = 2 is a minima
x 3 4A is a maxima
Maximum value = 427
Minimum value = 0
sin 2 x
= sinx+-
2
f ( x ) = cosx+cos2x
Since f" ( % ) is negative; f ( x ) has a maximum at x = %
E9
A = ab
a 4r2 = a2 + b2
*b2 = 4 2 - a 2
Substituting the value of b in equation (1) and after maximazation;
a = 6. So, it is a square.
El0
Now; - d~ -
2A
da - 2 - ~
a A = a2 a = 6. So, it is a square.
El l
v = I d h
3
S = ml
i2 = 2+?
* s=mdh2+3
3 v
We know ? = -
dl
Substituting the value of r 2, and after minimizing for S, we get h = fi r.
Therefore, V = - Sa - a
( 3 ,
a S = 3a2. We get h = &,
where h is the vertical side and a is the'side of the square base.
After puting the conditions for greatest laleral surface, you will get h = 2 r
The point of the curvet8 closes to the point on the axis of the parabola y = 2p x at
a distance a 'is ( a,
) or ( a, -* ).
El5
Taking Moment about the point (Ncum),
dF w a v
Now-- = --
dF
dl +? Pu t - - 0 dl (for minimization)
we get
1 -sinx
i) lim
x--
2
= lim
1
= lim -
4 sinx
X-2
= V4
ii) lim
- 0 2
,, -10 sin x [ ; ]
x- 0 6x
8 - e-x
= lim - Lt
ex+e- x
x,o 2sin2r x,o 4ms 2r
ii) lim ( xl nl si nxl )
x-0
In I sinx[' 2 cosx xcosx
= lini
= lim-- - lim-- = 0
x-0 lk x-0 Isinxl .-0 lslnxl
X
In x
iii) lim -
x-0
+cotx
1
-
x sin x
= lim = lim(-sinx-)
-c , , + - cosec x
x,o+
X
x ~ o x + ( l + x ) l n ( l + x )
x ~ o l + l + l n ( l + x )
ii) y = lim I sin x
x-0
lny = lim x1n)sinxI
x-0
- lim
In I sinx 1
x-0 v x
cos X
0
* y = e * y = 1
iii) y = lim 1 sec x lax
X
x--
2
In ( sec x 1 _ lim
tan x
lny = lim cosxlnlsecxl - lim
X
secx secxtanx
X-2 "'5 X-Z
0
* y = e * y - 1
X
l nx+ 1
= lim - -
2 + lnx
- 1/L
1
=s h y = lim - l n ( d + 1 )
x 4 a X
1
-, Iny = lim - = 1
x - a 1 + '/e'