Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Lecture notes on CALCULUS 1

Chapter 1. NUMBER SYSTEMS

1.1. Axioms for R
The set R is a complete ordered field, i.e., R is a set which contains at least 2 elements, say 0 and 1, satisfies
the following 4 groups of axioms:
 Axioms about operations: On R there are two operations + and . such that (R, +, .) is a commutative field,
i.e.:
1. x, y  R , x + y = y + x (commutative)
2. x, y, z  R, x + (y + z) = (x + y) + z (associative)
3. x  R, x + 0 = x (unit element)
4. x  R,  x  R, x + (x) = 0 (symmetric element)
5. x, y  R, x.y = y.x (commutative)
6. x, y, z  R, x.(y.z) = (x.y).z (associative)
7. x  R, x.1 = x (unit element)
8. x  R\{0},  x1  R, x.x1 = 1 (inverse element)
9. x, y, z  R. x.(y + z) = x.y + x.z (distributive)
Remarks.
 The symmetric element is unique. This implies that –(–x) = x x  R.
 The inverse element is unique. This implies (x–1)–1 = x x  R\{0}.
 For a, b, c  R, a + c = b + c  b = c.
 For a, b, c  R, a  0, ab = ac  b = c.
 x  R, x.0 = 0.
 x, y  R, x.y = 0  (x = 0 or y = 0).
 x, y  R, x.(y) = (x).y =  x.y.
 x  R, (–1)x = –x.
We define the substraction and the quotient as follows:
x, y  R, x  y = x + (y)
x
x  R, y  R\{0},  x.(y 1 )
y

 Axioms about ordered relation: There is a relation  on R satisfies:

1. x  R, x  x (reflexive)
2. x, y  R, (x  y  y  x)  x = y (antisymmetric)
3. x, y, z  R, (x  y  y  z)  x  z (transitive)
4. x, y  R, x  y  y  x (linearity)
5. x, y, z  R, x  y  x + z  y + z
6. x, y, z  R, (x  y  0  z)  xz  yz

Calculus 1 Page 1
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Remarks.
 x  R, x  y  –y  –x
 x, y, z  R, (x < y and z > 0)  xz < yz.
 x, y, z  R, (x  y and z  0)  xz  yz.
 x, y, z  R, (x < y and z < 0)  xz > yz.
 x, y  R, x  y  –x  –y.
 x, y  R, x < y  –x > –y.
Now, we define some terminologies: Let A be a nonempty subset of R.
An upper bound (resp. lower bound) of A is a number m  R such that
a  A, a  m (resp. a  A, m  a)
A number   R is called the maximum (resp. minimum) of A, denoted by maxA (resp. minA) if
  A and  is an upper bound of A.
(resp.   A and  is a lower bound of A)
A number   R is called the least upper bound (resp. greastest lower bound) of A, denoted by l.u.b. A or
sup A (resp. g.l.b A or inf A) if  is the minimum of the set of all upper bounds of A (resp.  is the
maximum of the set of all lower bounds of A).
We said that the set A is bounded above (resp. bounded below) if there exists an upper bound (resp. lower
bound) of A.
We said that the set A is bounded if it is both bounded above and bounded below.
We shall write:
x<yxyxy
xyyx
x>yy<x
and call numbers > 0 are positive numbers, and numbers < 0 are negative numbers.
Notations:
R+ = { x  R  x > 0 }
R = { x  R  x < 0 }

 Axioms for the sup: Every nonempty bounded above subset of R has the least upper bound.

It is easy to verify the following result:

Every nonempty bounded below subset of R has the greatest lower bound.

Now, consider the set N of natural numbers. We identify each n  N with the real number 1  1  ...  1 .
n times

Thus, N is a subset of R. We have two evidently properties:

1N
and n  N, n + 1  N.

Calculus 1 Page 2
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

 Axioms of well - ordered: Every nonempty subset of N has the minimum.

We have an important result:

Theorem 1 (Archimede): Let b > 0. Then a  R,  n  N, nb > a.
Proof.
 

Corollary:
x  R,  n  N, n > x
1
 > 0,  n  N,  
n
Proof.
 

For x  R, we define the absolute value of x, denoted by | x |, as follows:

 x if x  0
x  .
 x if x  0
We have the following properties:
x  R, | x |  0, and | x | = 0  x = 0
x, y  R, | x.y | = | x |.| y |
x, y  R, | x + y |  | x | + | y |
x, y  R, | x  y |  | | x |  | y | |

1.2. N, Z, Q
Notations:
N = { 1, 2, ... } : the set of natural numbers.
Z = { ..., 2, 1, 0, 1, 2, ... } : the set of integers
m
Q={  m, n  Z, n  0 } : the set of rational numbers
n
R : the set of real numbers.

We have N  Z  Q  R.

Some important properties of real numbers:

Theorem 2. x  R, ! n  Z, n  x < n + 1.
This number n is called the floor of x and denoted by [x].
The number x  [x] is called the fractional part of x.

Calculus 1 Page 3
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Proof. 

Theorem 3. Given two real numbers a, b with a < b. There exists a rational number q  (a,b).
Proof. 

Theorem 4. Given two real numbers a, b with a < b. There exists an irrational number q  (a,b).
Proof.


Remark. By theorems 2 and 3, there are infinitely many rational numbers in (a,b) and there are infinitely many
irrational numbers in (a,b).

In R, we define for a, b  R,
[a, b] = { x  R : a  x  b }
(a, b) = { x  R : a < x < b }
[a, b) = { x  R : a  x < b }
(a, b] = { x  R : a < x  b }
[a, + ) = { x  R : x  a }
(a, + ) = { x  R : x > a }
(, b] = { x  R : x  b }
(, b) = { x  R : x < b }
1.3. Some useful equalities and inequalities
Theorem 5. For a, b  R, n  N, we have
n n n n!
(i) 
a  bn   an  k bk , where  
 k  k!(n  k )!
k  0 k   
n
(ii) a n1  b n1  (a  b)  a nk b k
k 0

Proof.
 

Theorem 6.
(i) For a, b  R, a2 + b2  2ab.
Equality holds iff a = b.
1
n  n n
1
(ii) For ai  0, i = 1, ..., n,
n i1 
ai  
  ai  .

 i 1 
Equality holds iff a1 = ... = an.

Calculus 1 Page 4
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

(iii) For ai , bi  R, i = 1, ..., n,

2
 n   n  n 
    
 ai bi    a2i  b2i 
  
 i 1   i 1  i 1 
a a
Equality holds iff 1  ...  n .
b1 bn
(iv) For a > 1, n  N, (1 + a)n  1 + na.
Equality holds iff a = 0, or n = 1.
Proof. 


1.4. Extended real number

We define R = R  { , + } and x R,
 < x < +
(+) + (+) = (+) – (–) = +
(–) + (–) = (–) – (+) = –
(+)  (+) = (–)  (–) = +
(+)  (–) = (–)  (+) = –
x + (+) = x – (–) = +
x – (+) = x + (–) = –
 if x  0
x  ()  
 if x  0
 if x  0
x  ()  
 if x  0
x 
0 , 
 x
x
If x  0 then 
0
Remark.
We don’t determine:
(+) – (+)
(+) + (–)
(–) + (+)
(–) – (–)
0
0
0
0

Calculus 1 Page 5
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.



Moreover, we set:
sup =  , inf = +
and if A is not bounded above then supA = +,
if A is not bounded below then infA = .

1.5. Neighborhoods
Let x  R. An open interval which contains x is called a neighborhood of x.
The interval (, b) , b  R, is a neighborhood of .
The interval (a, +) , a  R, is a neighborhood of +.
The set of all neighborhoods of x is denoted by Nx.
Evidently, union and intersection of two neighborhoods of x is also a neighborhood of x.

1.6. Complex numbers

Let C = {a + ib : a, b  R}.
Define for z1 = a1 + ib1 , z2 = a2 + ib2  C,
z1 + z2 = a1 + a2 + i(b1 + b2)
and z1z2 = a1a2 – b1b2 + i(a1b2 + a2b1).
Some properties:
RC
For z1, z2, z3  C,
z1 + z2 = z2 + z1
(z1 + z2) + z3 = z1 + (z2 + z3)
z1 + 0 = z1
z1 + (–z1) = 0
z1z2 = z2z1
(z1z2)z3 = z1(z2z3)
1.z1 = z1
z1(z2 + z3) = z1z2 + z1z3
1 a b
For z = a + ib (a, b  R), if z  0, define z–1 =  2 i 2 then z.z–1 = 1.
z a b 2
a b 2

z1
We define z1 – z2 = z1 + (–z2) and if z2  0,  z1 .z 21 .
z2
For z = a + ib (a, b  R), define:
|z| = a 2  b2 .
z  a  ib ( z is called the conjugate of z)
then z. z = |z|2.

Calculus 1 Page 6
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

z1  z2  z1  z2
z1  z2  z1  z2
z1.z2  z1.z2
 z1  z1
 
 z2  z2

For z = a + ib  0 (a, b  R), define:

z0 = 1
zn = zn–1z and z–n = (zn)–1 (n  N)
then zn+m = zn.zm (n, m  Z)

Representation of complex numbers in the plane:

For each complex number z = a + ib (a, b  R), we associate it with the point (a,b) in a xy-plane.
Put r = |z|,  = (Ox,OM) where M(a,b). Then we have z = r(cos + i.sin).
Theorem 7.
(i) Let z1 = r1(cos1 + i.sin1), z2 = r2(cos2 + i.sin2), then
z1z2 = r1r2[cos(1 + 2) + i.sin(1 + 2)].
z1 r
= 1 [cos(1 – 2) + i.sin(1 – 2)].
z2 r2

(ii) Let z = r(cos + i.sin), n  Z. Then zn = rn[cos(n) + i.sin(n)] (De Moivre’s formula).
Proof.
 

Theorem 8. Let z = r(cos + i.sin), r > 0 and n  Z. Then there exists n nth-roots of z given by the formula
1
    2k     2k  
cos  n   i sin  n   , k {0, 1, … , n – 1}
r n
    
Proof.
 

***

Calculus 1 Page 7
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Chapter 2. SEQUENCES
2.1. Convergence
A mappping x : N  R is called a sequence of real numbers. The image of n  N by x will be denoted by
xn and is called the n-th term of the sequence. We will write (xn) for this sequence.
The set { xn | n  N } is called the image of the sequence.
We said that the sequence (xn) is bounded (resp. bounded above, bounded below) if its image is bounded
(resp. bounded above, bounded below). Thus,
(xn) is bounded iff there exists C > 0 such that | xn |  C n.
A constant sequence is a sequence such that its image is a single set.
x 
From two sequences (xn), (yn), we can define sequences (xn + yn), (xn  yn), (xnyn); and if yn  0 n,  n 
 yn 
.
The sequence (xn) is convergent if there exists a real number a such that
 > 0, N  N, n  N, n  N  | xn  a | < 
This number a is called the limit of (xn) and we write
lim x n  a or lim x n  a or xn  a as n   or xn  a.
n
Theorem 1. Limit of a convergent sequence is unique.
Proof. 

If (xn) is not convergent, we said that it is divergent.
Theorem 2. Suppose limxn = a, limyn = b and   R. Then:
lim(xn + yn) = a + b
lim(xn  yn) = a  b
lim(xnyn) = ab
lim(xn) = a
and if b  0 and yn  0 n, then
x  a
lim n   .
 yn  b
Moreover, we have:
limxn = a  lim(xn  a) = 0.
limxn = 0  lim| xn | = 0.
Proof. 

Theorem 3. Suppose limxn = a, limyn = b.
If xn  0 n then a  0.
If xn  yn n then a  b

Calculus 1 Page 8
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Proof. 

Theorem 4. Let (xn), (yn), (zn) be three sequences which satisfies xn  yn  zn n. Suppose (xn) and (zn) converge
to a same limit a. Then (yn) is convergent and limyn = a.
Proof. 

Theorem 5. Every convergent sequence is bounded.
This implies that every unbounded sequence is divergent.
Proof. 

Definition. Consider a sequence (xn).
We write xn  + (resp. ) if
M > 0, N  N, n  N, n  N  xn > M
(resp. M > 0, N  N, n  N, n  N  xn < M)
We write xn   if | xn|  +.

Remarks.
Theorem 2 is still true with limits are extended real numbers.

In the cases that can not apply theorem, we said that the limit is in the indetermined forms.
0 
The usual indetermined forms are: , , 0  ,    .
0 
Theorem 6.
1
1/ Let p > 0. Then lim 0
np
2/ Let p > 0. Then lim n p  1
3/ lim n n  1
n
4/ Let p > 0 and   R. Then lim 0
(1  p) n
5/ Let | x| < 1. Then limxn = 0.
Proof. 


2.1.2. Monotonic sequences

The sequence (xn) is said to be increasing if xn  xn + 1 n.
The sequence (xn) is said to be decreasing if xn+1  xn n.
Calculus 1 Page 9
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

The sequence (xn) is said to be strictly increasing if xn < xn + 1 n.

The sequence (xn) is said to be strictly decreasing if xn+1 < xn n.

Increasing sequences and decreasing sequrences are called monotonic sequences.

Theorem 7. Every bounded monotonic sequence is convergent.
Proof. 


2.1.3. Number e
n
 1
The sequence 1   is increasing and bounded. Therefore, it is convergent.
 n
We define
n
 1
e = lim 1   .
 n
2.1.4. Subsequences
Consider a sequence (xn).
Let (nk) be a strictly increasing sequence of natural numbers.
 
The sequence (yk) with yk = x n k is called a subsequence of (xn) and will be symbolized by x nk .
Clearly, every sequence is a subsequence of itself.
Theorem 8. The sequence (xn) converges iff every of its subsequences converge to the same limit.
Proof. 

Corollary. If the sequence (xn) has two subsequences which converge to two different limits then (xn) is divergent.
Proof. 

Lemma. Every bounded sequence has a monotonic subsequence.
Proof. 

Theorem 9 (Bolzano - Weierstrass) Every bounded sequence has a convergent subsequence.
Proof.
 

Calculus 1 Page 10
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

2.1.5. Cauchy sequence

Definition. The sequence (xn) is called a Cauchy sequence if
 > 0, N  N, m, n  N, m, n  N  | xm  xn | < .
Theorem 10. Every Cauchy sequence is bounded.
Proof. 

Theorem 11. Every convergent sequence in R is a Cauchy sequence.

Proof. 



***

Calculus 1 Page 11
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Chapter 3. LIMIT AND CONTINUITY OF FUNCTIONS

3.1. Limits
In this chapter, we let D = (a,b),   [a,b].

Definition. Let f : D\{}  R be a function. We said that f(x) tends to  R as x approaches , if

  0,   0, x  D \ {}, x      f (x)    

Notation: f(x)   as x   or lim f (x)  

x 

Theorem 1. lim f (x)    (x n )  D \ {}, if x n   then f (x n )  

x 

Proof. 

Theorem 2. Let f, g : D\{}  R. Suppose lim f (x)   and lim g(x)   . We have:
x  x 

lim[f (x)  g(x)]     and lim[f (x)  g(x)]    

x  x 

lim[f (x).g(x)]  
x 

Moreover, if g(x)  0 x  D\{}, we have:

f (x) 
lim 
x  g(x) 
Proof. 

Remark. Let f(x) = C x  D. We call f be a constant function. Then for any , lim f (x)  C .
x 

Theorem 3.
(i) Suppose limf (x) and lim g(x) exist.
x  x 

If on a neighborhod V of , we have f(x)  g(x). Then lim f (x)  lim g(x) .

x  x 

(ii) Consider 3 functions f, ,  defined on D\{}.

Suppose on a neighborhood V of , we have (x)  f(x)  (x), and lim (x)  lim (x) = .
x  x 

Then lim f (x)   .

x 
Proof.

3.2. Continuity
Definition. Let f be a function defined on D and x0  D. We said that f is continuous at x0 if
 > 0,  > 0, x  D, | x  x0 | <   | f(x)  f(x0) | < 

Calculus 1 Page 12
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

We said that f is continuous on D if f is continuous at every point of D.

Theorem 4. Let f: D  R, x0  D. Then f is continuous at x0 iff lim f (x)  f (x0 ) .
xx 0

Proof.


Theorem 5. Let f: D  R, x0  D.
Then f is continuous at x0  (x n )  D, if xn  x0 then f (x n )  f (x 0 ) .

Proof.


Theorem 6. Suppose f, g are continuous at x0. Then f + g, f – g, fg are continuous at x0.

f
Moreover, if g(x0)  0 then is continuous at x0.
g
Proof.


Theorem 7. Suppose f is continuous at x0, g is continuous at f(x0). Then gof is continuous at x0.
Proof.


Theorem 8. Every elementary function is continuous on its domain of definition.

Proof.

Theorem 9. Let f: [a,b]  R be continuous Then:
(i) f is bounded on [a,b], i.e., the set f([a,b]) is bounded.
(ii) f attains its maximum and its minimum on [a,b], i.e., max f (x) and min f (x) exist.
x[a, b ] x[a,b ]

(iii) f([a,b]) = [m,M] where, M = max f (x) , m = min f (x) .

x[a, b ] x[a,b ]

Proof.


Calculus 1 Page 13
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Corollary. Let f be continuous on [a,b]. If f(a).f(b) < 0 then the equation f(x) = 0 has a solution in (a,b).
Proof.

Definition. We said that the function f: D  R is uniformly continuous on D if
 > 0,  > 0, x1, x2  D, x1  x2 <   f (x1 )  f (x 2 ) < .
Evidently, if f is uniformly continuous on D then f is continuous on D (but the converse is not true).
Theorem 10. If f is continuous on [a,b] then f is uniformly continuous on [a,b].
Proof.


***

Calculus 1 Page 14
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Chapter 4. DIFFERENTIATION
4.1. Derivatives
Definition. Let f : D  R, x0  D.
We said that f is differentiable at x0 if the following limit exist:
f (x )  f (x 0 )
lim
xx 0 x  x0
df
This limit is called the derivative of f at x0 and is denoted by f ' (x0) or Df(x0) or (x 0 ) .
dx
We said that f is differentiable on D if f is differentiable at every point of D.
Note that we have also
f (x 0  h )  f (x 0 )
f ' (x 0 )  lim .
h0 h

Definition. Let y = f(x) be differentiable at x0. The differential of f at x0 , denoted by df(x0) or dy(x0) , or simply
df or dy, is the quantity f ' (x0).h (depends on a variable h)

By consider the function (x) = x, we have h = dx. Hence, we can write df (x) = f ' (x)dx.
We have also: If h(x) = g(t), then h’(x)dx = g’(t)dt.

4.2. The derivative function and higher derivatives

Let f be differentiable on D. Then we have the derivative function f’ or Df of f:
f': D  R
x  f ' (x)
The derivative of f’ at x (if exist) is called the second derivative of f at x, denoted by f"(x) or f(2)(x).
In general, by induction, we define the nth derivative f(n) (n : integer  0) as follows:
f(0) = f
f(n) = [f(n1)]’ (n  1)
If f has n-th derivative on D and f(n) is continuous on D (n  0), we said that f is of class Cn on D and we write
f  Cn(D).
If f  Cn(D) for all n, we write f  C(D).

4.3. Properties of derivatives

Theorem 1. If f is differentiable at x then f is continuous at x. Thus, if f is differentiable on D then f is continuous
on D.
Proof.

Theorem 2. Let f, g be two differentiable functions at x. Then f + g, f  g, f . g are differentiable at x and
(f + g)'(x) = f ' (x) + g'(x)

Calculus 1 Page 15
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

(f  g)'(x) = f ' (x)  g'(x)

(f . g)'(x) = f ' (x).g(x) + f(x).g'(x)
f
Moreover, if g(x)  0 then is differentiable at x and
g
'
 f  f ' (x).g(x)  f (x).g' (x)
  
g g(x)2
Thus, if the functions u and v are differentiable on D then u + v, u  v, uv are differentiable on D and
(u + v)' = u' + v'
(u  v)' = u'  v'
(uv)' = u'v + uv'
u
Moreover, is differentiable on the set { x  D : g(x)  0 } and
v
'
u u' v  uv'
   .
v v2
Proof.

Theorem 3. Let f be differentiable at x, g be differentiable at f(x). Then gof is differentiable at x and
(gof)'(x) = g'(f(x)). f ' (x).

Proof.

Theorem 4. Let f be differentiable at x. Suppose the inverse function f 1 exists and f '(x)  0. Then f 1 is
differentiable at y = f(x) and
 
f 1 ' (y) 
1

1
f ' (x) f ' (f 1 (y))

.

Proof.

4.4. Table of derivatives

f(x) f ' (x) f(x) f ' (x)

x (  R) x 1 u (  R) u1.u’
sinx cosx sin u u’.cosu
cosx sinx cos u u’.sinu
1 u'
tanx 1 + tan2x = 2 tan u (1 + tan2u)u’ =
cos x cos 2 u

Calculus 1 Page 16
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

1  u'
cotx (1 + cot2x) = 2 cot u (1 + cot2u).u’=
sin x sin2 u
1 u'
sin–1x 2
sin–1u
1 x 1  u2
1 u'
cos–1x  cos–1 u 
1  x2 1  u2
1 u'
tan–1x tan–1 u
1 x 2
1  u2
1 u'
cot–1x  cot–1 u 
1  x2 1  u2
ex ex eu eu.u’
1 u'
nx n u
x u
1 u'
n x n u
x u
ax (a > 0, a  1) ax na au (a > 0, a  1) au na .u’
1 u'
og a x (a > 0, a  1) og a u (a > 0, a  1)
xna u na

4.5. Properties of differentiable functions

Definition. Let f : D  R, x0  D. We said that:
f has a local maximum at x0 if  V  Nx , x  V, f(x)  f(x0).
f has a local minimum at x0 if  V  Nx , x  V, f(x)  f(x0).
Local maximum and local minimum are called local extremum.
Definition. Let f: D → R, x  D. If f ' (x) = 0 or f ' (x) does not exist then we said that x is a critical point of f.

Theorem 5 (Fermat). Suppose f is differentiable at x and f has a local extremum at x, then f ' (x) = 0.

Proof.

Theorem 6 (Rolle). Suppose f is differentiable on D and [a,b]  D. If f(a) = f(b) then there exists x0  (a,b) such
that f ' (x0) = 0.

Proof.

Theorem 7 (Lagrange). Suppose f is differentiable on D and [a,b]  D. Then there exists c  (a,b) such that

Calculus 1 Page 17
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

f(b)  f(a) = (b  a) f ' (c).

Proof.


An equivalent form of theorem 7 is:

Theorem 8 (Mean value theorem). Suppose f is differentiable on D and [a,b]  D. Then there exists   (0,1) such
that f(b)  f(a) = (b  a). f ' [(1  )a + b] = (b  a). f ' [a + (b  a)].

Proof.

Definition. Let f: (a,b) → R. We said that f is:
increasing on (a,b) if x1, x2  (a,b), x1 < x2  f(x1)  f(x2)
decreasing on (a,b) if x1, x2  (a,b), x1 < x2  f(x1)  f(x2)
strictly increasing on (a,b) if x1, x2  (a,b), x1 < x2  f(x1) < f(x2)
strictly decreasing on (a,b) if x1, x2  (a,b), x1 < x2  f(x1) > f(x2)
Increasing functions and decreasing functions are called monotonic functions.
Strictly increasing functions and strictly decreasing functions are called strictly monotonic functions.

Theorem 9. Let f be differentiable on (a,b).

If f ' (x)  0 x  (a,b) then f is increasing on (a,b).

If f ' (x)  0 x  (a,b) then f is decreasing on (a,b).

If f ' (x) > 0 x  (a,b) then f is strictly increasing on (a,b).

If f ' (x) < 0 x  (a,b) then f is strictly decreasing on (a,b).

If f ' (x) = 0 x  (a,b) then f is constant on (a,b).

Proof.

Theorem 10. Let f  C2 on (a,b) and x  (a,b). Suppose f ' (x) = 0.

If f " (x) < 0 then f has a local maximum at x.

If f " (x) > 0 then f has a local minimum at x.
Proof.

Calculus 1 Page 18
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.



Theorem 11 (L'Hospital rule). Let f, g : (a,b)  R be differentiable functions (a, b  R ). Suppose

f ' (x )
lim   (   R ).
x  b g' (x)

If lim f (x)  lim g(x)  0 or lim g(x)   thì

xb xb xb

f (x )
lim 
xb g(x)

We have similar result for the case x  a.

Proof.


***

Calculus 1 Page 19
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Chapter 5. INTEGRATION
5.1. Antiderivative
Definition. Let f(x) be determined on D. A function F(x) is called an antiderivative of f(x) on D if
x  D, F ' (x) = f(x).

Theorem 1.
(i) If F(x) and G(x) be two antiderivatives of f(x) on (a,b) then G(x)  F(x) = constant on (a,b).
(ii) Let F(x) be an antiderivative of f(x) on (a,b). Then for every constant C, F(x) + C is an antiderivative of
f(x) on (a,b). We write:

 f (x)dx  F(x)  C .
Proof.

Theorem 2.
(i)  f (x)  g(x)dx   f (x)dx   g(x)dx
(ii)  f (x)dx   f (x)dx ( is constant)

Proof.


Table of antiderivatives.
The antiderivatives F(x) of f(x) are given in the following table:

f(x) F(x) f(x) F(x)

x  1 u  1
x (  R \ {1}) uu' (  R \ {1})
 1  1
1 u'
n x n u
x u
sinx  cosx u'.sinu  cosu
cosx sinx u'.cosu sinu
1 + tan2x tanx (1 + tg2u)u' tgu
1 + cot2x  cotx (1 + cotg2u)u'  cotgu
1 u'
sin–1x sin–1u
1  x2 1  u2

Calculus 1 Page 20
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

1 x u' u

2 2
(a > 0) sin 1   2 2
(a > 0) sin 1  
a x a a u a

1 u'
tan–1x tan–1u
1 x 2
1  u2
1 1 x u' 1 u
2 2
(a > 0) tan 1   2 2
(a > 0) tan 1  
a x a a a u a a
1 1 1 x u' 1 1 u
2
n 2
n
1 x 2 1 x 1 u 2 1 u

1 1 ax u' 1 au

2 2
(a > 0) n 2 2
(a > 0) n
a x 2a a  x a u 2a a  u
1 u'
2
(k  R) n x  x 2  k (k  R) n u  u 2  k
2
x k u k
ex ex eu.u' eu

ax au
ax au.u'
na na

5.2. Riemannian integrals

Definition. A partition of the closed interval [a,b]  R is a finite sequence P = (a0, a1, ..., an) such that
a = a0 < a1 < ... < an = b. The mesh of this partition is | P | = max { aj  aj -1 : j = 1, ..., n }.
Let P = (a0, a1, ..., an) and Q = (b0, b1, ..., bm) be two partitions of [a,b].
We said that Q is finer than P if {a0, ..., an}  {b0, ..., bm}.
Clearly, if P, Q are two arbitrary patitions of [a,b], then the partition PQ is finer than both P and Q.
Let f : [a,b]  R be a bounded function. Let P = (a0, ..., an) be a partition of [a,b].
Put
m j  inf f (x ) , M j  sup f (x)
x[a j1 ,a j ] x[a j1 ,a j ]

n n
L( f , P )   (a j  a j1 )m j , U( f , P )   (a j  a j1 )M j
j1 j1

L(f,P) and U(f,P) are called the upper and lower Darboux sum of f corresponds to the partition P.
We have
L(f,P)  U(f,P)
And therefore
sup L(f , P)  inf U(f , P)
P P
We said that f is integrable on [a,b] if

Calculus 1 Page 21
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

sup L(f , P)  inf U(f , P)

P P
and the above value is called the integral of f from a to b, denoted by
b

 f (x)dx
a
Note that in the above notation, the letter x can be replaced by another letter, i.e., we have
b b b

 f (x)dx   f (u)du   f (t)dt  

a a a
Suppose f is integrable on [a,b]. We define
a b

 
f (x)dx   f (x)dx .
b a

Let f be integrable on [a,b] and suppose f  0 on [a,b]. We define the area S of of the region limited by Ox, the
b
graph of f, and two lines x = a, x = b; is S  f (x)dx . 
a
Theorem 3. Let f : [a,b]  R be bounded. Then f is integrable on [a,b] if and only if for every  > 0, there exists a
partition P of [a,b] such that U(f, P)  L(f, P) < .
Proof.

Theorem 4. Every continuous funtion on [a,b] is integrable on [a,b].
Proof.

Theorem 5.
(i) Let f, g be two integrable functions on [a,b] and   R. Then f + g, f are integrable on [a,b] and
b b b

 f (x)  g(x)dx   f (x)dx   g(x)dx

a a a
b b

 f (x)dx   f (x)dx 
a a
b
(ii) Let f be integrable on [a,b]. If f  0 on [a,b] then  f (x)dx  0 .
a
b b
Let f, g be integrable on [a,b]. If f  g on [a,b] then  f (x)dx   g(x)dx .
a a

Calculus 1 Page 22
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

(iii) Let f be integrable on [a,b]. Then f is integrable on [a,b] and

b b

 f (x)dx   f (x) dx
a a

(iv) Let f be integrable on [a,b], , ,   [a,b]. Then

  

 f (x)dx   f (x)dx   f (x)dx .

  

Proof.


5.3. Fundamental theorem

Theorem 6. Let f be integrable on [a,b], x0  [a,b]. Define
x
F(x )   f (t)dt x  [a,b]
x0

Then F is continuous on [a,b].

Moreover, if f is continuous at x  [a,b] then F is differentiable at x and F ' (x) = f(x).
Thus, if f is continuous on [a,b] then F is the antiderivative of f on [a,b] which vanishes at x0.
Proof.

Theorem 7. Let F be of class C1 on [a,b]. Then
b

 F ' (x)dx  F(b)  F(a) .

a

Proof.


Corollary (Newton - Leibnitz). If F is an antiderivative of f on [a,b] then

b
b
 f (x)dx  F(b)  F(a)  F(x) a .
a

Proof.


Calculus 1 Page 23
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

5.3. Methods to compute integrals

Theorem 8 (Integration by parts) Let u, v be of class C1 on [a,b]. Then
b b
b
 u' (x)v(x)dx  u(x)v(x) a 
 u(x)v' (x)dx
a a

Proof.

Theorem 9 (Change variables)
Let f be of class C1 on [a,b]. Suppose  : [,]  [a,b] is of class C1 on [,]. Then
 ()

 f ((t))' (t)dt   f (x)dx

 ( )

Proof.


5.4. Taylor’s formula

Theorem 10. Let f be of class Cn on a neighborhood V of a. For h small enough, we have
n 1
h k 1 ( k 1)
f (a  h)   ( k  1)!
f (a)  R n
k 1
where
1
hn
Rn  (1  t )n 1 f (n) (a  th)dt
 (*)
(n  1)!
0

Proof.


Remark.
(i) For p  N, p  n, there exists   (0,1) such that
(1  )n  p hn (n)
Rn  f (a  h)
p(n  1)!

Let p = n, we have:

h n (n)
Rn  f (a  h) with   (0,1) (**)
n!
Let p = 1, we haver:

Calculus 1 Page 24
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

(1  ) n1 h n (n )
Rn  f (a  h) with   (0,1) (***)
(n  1)!

(ii) Put M sup f (n) (x)

a  h  x a  h

n
h
then Rn  M.
n!
In particular, we have the Maclaurin’s formula:
n 1
x k 1 ( k 1)
f (x )   (k  1)!
f (0)  R n
k 1
n
x
with Rn  M , M  sup f (n) (t) .
n!  x t x

5.5. Improper integral

We study 2 kinds of improper integrals:
 Type 1: Infinite intervals.
 Type 2: Discontinuous integrands.
Definition. Let f : [a,+)  R be integrable on every interval [a,x], x > a. We define:
 x

 f (t )dt  lim
x   f (t)dt .
a a
b b

 f (t)dt  xlim
 
f (t )dt
 x
 c 

 f (t)dt   f (t)dt   f (t)dt (c  R)

  c
If the limit exists, we said that the corresponding improper integral is convergent.
Theorem 11. For  > 0, we have

dx
 x is convergent   > 1
1

Proof.


Definition.
(i) Let f : [a,b)  R be integrable on every interval [a,x], a  x < b. We define:

Calculus 1 Page 25
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

b x

 f (t)dt  xlim
b 
f (t )dt
a a
(ii) Let f : (a,b]  R be integrable on every interval [x,b], a < x  b. We define:
b b

 f (t)dt  xlim
a 
f (t )dt
a x
(iii) Let f : [a,b]\{c}  R, a < c < b. We define:
b c b

 f (t)dt   f (t)dt   f (t)dt (c  R)

a a c

Theorem 12. For  > 0, we have

1
dx
 x is convergent  0 <  < 1.
0

Proof.

b
Theorem 13 (Comparison test 1) Consider the improper integral  f (t)dt (a, b  [–,+]), where f is a
a
nonnegative function.
b b
(i) If there exists a nonnegative function g such that f  g and  g(t)dt is convergent then  f (t)dt is
a a
convergent.
b b
(ii) If there exists a nonnegative function g such that f  g and  g(t)dt is divergent then  f (t)dt is
a a
divergent.
Proof.


Theorem 14 (Comparison test 2) Let f, g: [a, b)  R be nonnegative, integrable on every interval [a,x], a  x< b.
b b
f (x)
Let   lim
xb g(x)
 [0,] , If 0 <  < + then either both  g(x)dx and  f (x)dx converge or both diverge.
a a

Proof.


Calculus 1 Page 26
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

5.6. Applications
1/ Area.
 Let S be the area of the region limited by Ox, the curve (C): y = f(x), where f is continuous on [a,b], and
two lines x = a, x = b. Then
b
S   f (x) dx
a
 Let S be the area of the region limited by two curves (C1): y = f1(x), (C2): y = f2(x), where f1, f2 are
continuous on [a,b], and two lines x = a, x = b. Then
b
S  f1 (x)  f2 (x) dx
a
2/ Volume.
 Let (H) be the region limited by Ox, the curve (C): y = f(x), where f is continuous on [a,b], and two lines
x = a, x = b. Let V be the volume of the solid obtained by rotating about the x-axis the region (H). Then:
b
V   f (x) 2 dx

a
3/ Arc length.
 Suppose f is of class C1 on [a,b]. Then the length of the curve (C): y = f(x), x  [a,b], is
b
  1  f ' (x)2 . dx
a
***

Calculus 1 Page 27
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Chapter 6. SERIES
6.1. Definition
Consider a real sequence (an). We call

 an or  an (1)
n 1
a series of real numbers.
n
Put sn  a1  ...  an   ai (n  N)
i 1
th
We call sn the n partial sum of the series.
If lim sn =   R exists, we said that the series (1) is convergent and write
 an   .
If the sequence (sn) is divergent, we said that the series (1) is divergent.


Theorem 1. The geometric series  x n is convergent  | x| < 1.
n 0

Proof.


Theorem 2. Suppose  an  a and  b n  b . Then

 an  b n   a  b and  ka n  ka (k  R)

Proof.


Lemma. (Cauchy) The series  an is convergent iff

m
 > 0, n0  N, m, n  N, m  n  n0   ai  .
in

Proof.


Theorem 3. If  an is convergent then lim an = 0.

Deduce that if (an) does not converge to 0 then  an is divergent.

Proof.


Calculus 1 Page 28
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Theorem 4. If  an is convergent then  a n is convergent.

Proof.


Definition. If  an is convergent, we said that the series  an is absolutely convergent.

6.2. Positive series

We call  an a positive series if an  0 n.
Note that for positive series, the sequence of its partial sum is increasing. We have

Theorem 5. A positive series is convergent iff the sequence of its partial sum is bounded.
Proof.


Theorem 6 (Comparison test 1). Consider two positive series  an and  b n . Suppose an  bn n.
(i) If  a n is divergent then  b n is divergent.
(ii) If  b n is convergent then  a n is convergent.

Proof.


Theorem 7 (Comparison test 2) Consider two positive series  an and  b n . Put

an
  lim .
bn
If 0 <  < + then either both series converge or both diverge.
Proof.


Theorem 8. Suppose (an) is a decreasing positive sequence. Then either both the series  an and  2 n a2 n

converge or both diverge.

Proof.


Calculus 1 Page 29
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Theorem 9 (Integral test) Let f : [1, +) → [0,+) be a continuous, decreasing function. Then the series


 f (n) is convergent iff the improper integral  f (x)dx is convergent.

1

Proof.


1
Theorem 10 (Harmonic series) The harmonic series  np is convergent iff p > 1.
n 1

Proof.

an 1
Theorem 11 ( D'Alember ratio test) Consider the positive series  a n . Suppose lim
an
  . Then

(i) If  < 1 then  a n is convergent.

(ii) If  > 1 then  a n is divergent.

Proof.

Theorem 12 (Cauchy root test) Consider the positive series  a n . Suppose lim n an   . Then
(i) If  < 1 then  a n is convergent.
(ii) If  > 1 then  a n is divergent.

Proof.


a n 1
Theorem 13. Consider the positive series  a n . If lim n a n exists then lim
an
exists and two limits are the

same.
Proof.

6.3. Arbitrary series
Theorem 14 (D'Alembert ratio test) Consider the series  an .
an 1
Put   lim . We have:
an
i) If  < 1 then  a n is convergent.
ii) If  > 1 then  a n is divergent.

Calculus 1 Page 30
 Faculty of Mathematics & Statistics, TDTU Chu Đức Khánh, PhD.

Proof.


Theorem 15 (Cauchy root test) Consider the series  an .

Put   lim n a n . We have:
i) If  < 1 then  a n is convergent.
ii) If  > 1 then  a n is divergent.

Proof.


Theorem 16 (Leibnitz). Let (an) be a decreasing sequence which converges to 0.

Then the alternating series  (1) n1a n is convergent. Moreover, a1  a2   (1) n1a n  a .
1

Proof.


2.2.4. Power series

Definition. The series  anxn is called a power series.
an 1 1
Theorem 17. Let  anxn be a power series. Put   lim
an
= lim n a n and R 

. Then:

If | x | < R then the series is convergent, if | x | > R then the series is divergent.

We call R the radius of convergence of the power series.

Proof.


***

Calculus 1 Page 31