Department of Mathematics, School of Natural Sciences, Shiv Nadar University

Monsoon Semester 2014-15

MAT 101 Calculus I

Lecture Notes Amber Habib December 1

1 Review of Key Concepts from Previous Lectures

1. Improper Integrals, Type I (Notes of Nov 28)

2. Mean Value Theorem for Integration (Nov 24)

3. Applications of Lagranges Mean Value Theorem (Oct 31)

2 Improper Integrals

In the last lecture we discussed integrals over unbounded intervals theseZare called Type
Z a
I improper integrals and have one of the following forms: f (x) dx, f (x) dx, or
Z a

f (x) dx.

The other kind is when the integrand f (x) is unbounded. Such integrals are called Type
II improper integrals and occur when f (x) has a vertical asymptote at some point.

For the definition and examples see pages 349350 of Stewart.

3 Weighted Mean Value Theorem

Theorem 3.1 Let f, g : [a, b] R be continuous functions such that g does not change
sign. Then c [a, b] such that
Z b Z b
f (x) g(x) dx = f (c) g(x) dx
a a

Proof: Assume g(x) 0.

By the Extreme Value Theorem, f achieves its minimum and maximum values m, M at
points u, v respectively. Then

m f (x) M = m g(x) f (x)g(x) M g(x)

Z b Z b Z b
= m g(x) dx f (x)g(x) dx M g(x) dx
a a a
 Z b
Consider the function H : [a, b] R defined by H(t) = f (t) g(x) dx. We have H(u) =
Z b Z b a

m g(x) dx and H(v) = M g(x) dx. So IVT, applied to the continuous function H,
a a Z
b
gives c [a, b] such that H(c) = f (x)g(x) dx. 2
a

One use of the Weighted MVT is in estimating integrals.

1
x9
Z
Example 3.2 Suppose we wish to get an estimate of dx. A first attempt
0 1 + x2
could be to use the earlier MVT for integration. For that we need to find the minimum
and maximum vaues of the integrand. So we look the first derivative:

x9 9x8 (1 + x2 ) x10 x8 (9 + 8x2 )

= = 0
1 + x2 (1 + x2 )3/2 (1 + x2 )3/2

Therefore the extreme values of the integrand occur at the endpoints and are 0 and 1/ 2.
Since the width of the interval is 1, this gives
Z 1
x9 1
0 dx 0.7
1+x 2 2
0

We can do much better by using the Weighted MVT. Take f (x) = 1/ 1+ x2 and
g(x) = x9 . The function f (x) is monotone and so has extreme values 1/ 2 and 1.
Z 1
1
Also, g(x) dx = . So the Weighted MVT gives:
0 10
Z 1
1 x9 1
0.07 dx = 0.1
10 2 0 1 + x2 10

107 2 128
(For the record, the precise value of this integral is = 0.074.) 2
315

4 Exponential Growth and Decay

A problem of indefinite integration asks us to figure out a function y = F (x) from its
derivative f (x). In other words, we have to solve the equation y = f for y.

This is the simplest case of what are called Ordinary Differential Equations or ODEs.
An ODE is an equation involving various derivatives of y, and the task is to solve for y.

Some ODEs for a function y(x):

1. y = cos x 3. y = 3y + 1

2. y = 5xy 4. (y )3 + y sec(y ) + y /y + tan x = 0

One measure of how difficult it may be to solve an ODE is the order of the ODE. This
is the order of the highest derivative of y that occurs in the ODE. In the list above, the
first two are first-order, the next is second-order, and the last is third-order.
In this section, we consider the following first-order ODE:

y = ky, where k is a constant

It is easy to check that every function of the form y = Aekx is a solution of this ODE.
Further, we proved earlier (as a consequence of the Lagrange MVT) that every function
that satisfies y = ky has to be of the form y = Aekx . Thus we know all the solutions of
this ODE.

We can solve for the constant A if we know even one value of y(x). Such knowledge, of
the form y(a) = b, is called an initial value condition. We can substitute it in the
general form of the solution to get b = Aeka , or A = beka .

Example 4.1 (Population Model) Consider the population P (t) of some species at
time t. The species could be an animal or plant in the wild, bacteria in a petri dish, a
chemical in a reaction. It is often reasonable to model the rate of change of the population
as being proportional to the population itself (100 rabbits should have twice as many
offspring as 50, and twice as many deaths, assuming that food and predators are not an
issue). In such a case we get
dP
= kP
dt
with the general solution
P (t) = Aekt
Suppose we know the starting population P0 at time t = 0. Then we get P0 = Ae0 or
A = P0 , so that the particular solution is

P (t) = P0 ekt

To find k we just need one more measurement of the population at another time. For
example, suppose we know the population P2 at time t = 2. Then we get
1 P2
P2 = P0 e2k = k = ln
2 P0
Note that k could be any real number. If k = 0 we have a constant population. If k > 0
we have exponential growth, with the population expanding without bound. If k < 0 we
have exponential decay, with the population dying down towards zero. 2

Example 4.2 (Radioactive Decay) Consider a mass M (t) of some radioactive mate-
rial at time t. The amount of material decreases as it emits radioactivity and changes to
different elements. A natural model is that the emission is proportional to the amount of
material, and so we have:
dM
= kM
dt
with the general solution
M (t) = M0 ekt
where M0 is the amount of material at time t = 0. In radioactive decay we must have
k < 0 with the mass decaying exponentially towards zero.
 25

20

15

10

10 20 30 40

The dashed lines represent counts of emissions from an amount of radioactive material
over 40 days, from a paper published in 1905. The solid curve is an exponential fit to the
data using the model just discussed. 2

References

1. Stewart, Essential Calculus, Sections 3.4 and 6.6.

2. Apostol, Calculus Vol I, Section 3.19.

Exercises

Stewart, Sec 3.4: 2, 5, 12

Stewart, Sec 6.6: 29, 31, 47

Apostol, Sec 3.20: 2