Chapter 1 1

Numbers, Equations and

Systems of Equations
1
1 Numbers, Equations and Systems of Equations . . . . . . . 3
1.1 Sets ................................................................. 3
1.2 Natural Numbers ................................................. 5
1.2.1 Peano’s Axioms ................................................... 6
1.2.2 Mathematical Induction (Complete Induction) ............. 7
1.2.3 Geometric Sum ................................................... 10
1.2.4 Permutations ...................................................... 10
1.2.5 The Binomial Theorem ......................................... 11
1.3 Real Numbers ..................................................... 13
1.3.1 Number Sets and Operations .................................. 13
1.3.2 Calculation Laws for Real Numbers .......................... 14
1.3.3 Power Calculations ............................................... 15
1.3.4 Logarithm .......................................................... 16
1.3.5 Arrangement of the Real Numbers ........................... 17
1.4 Equations and Inequalities...................................... 19
1.4.1 Equations .......................................................... 19
1.4.2 Inequalities......................................................... 23
1.5 Systems of Linear Equations ................................... 26
1.5.1 Introduction ....................................................... 26
1.5.2 Formulation and Notation ...................................... 28
1.5.3 Solving Systems of Linear Equations ......................... 29
1.6 Proofs .............................................................. 36
1.6.1 Mathematical Induction......................................... 36
1.6.2 Direct Proof ....................................................... 36
1.6.3 Proof by Contradiction .......................................... 37
1.6.4 Counter Example ................................................. 37
1.7 Problems on Numbers, Equations and Systems ............ 38
1 Numbers, Equations and
Systems of Equations

Numbers and Equations form the most important basic concepts of mathematics, on
which all further structures and constructions are based on. In this chapter, the basics
are covered by both sets and natural numbers. To describe the natural numbers Pea-
no’s axioms are introduced and the principle of mathematical (complete) induction is
demonstrated by many examples. The real numbers and elementary calculation laws are
indicated; the basic rules for powers and logarithms are repeated.

Solving equations is one of the elementary tasks of mathematics. In this chapter we will
also discuss simple equations as well as systems of linear equations which are important
for the applications and introduce the Gauss algorithm to solve them. Since only a few
types of equations are explicitly solvable, we will not deal systematically with the solving
of general equations, but we will show exemplarily how to work on basic equations.

1.1
1.1 Sets
”By a set M we mean any combination of certain well-defined objects of Set
our view or our thinking as a whole”; this definition of the set concept Menge

was introduced by G. Cantor (1895). This definition of a set is enough for

Elements
our purposes. In the following, we always use capital letters to describe
Elemente
sets. The objects of a set A are called elements of A and are described by
lowercase letters.
a 2 A means: a is an element of the set A.
a2 / A means: a is not an element of the set A.

Sets are usually indicated by

listing the elements in a curly bracket
{a1 , a2 , a3 , a4 , . . .} ,
in form of a statement
{a 2 A : a has the attribute E} .

The empty set ; or {} does not contain any elements. B means subset of A
(B ⇢ A), if each element of B is also an element of A.
 4 1. Numbers, Equations and Systems of Equations

Examples 1.1 (Sets):

IN = Set of natural numbers = {1, 2, 3, 4, ...}.
IN0 = Set of natural numbers with zero = {0, 1, 2, 3, 4, ...}.
ZZ = Set of whole numbers (integers) = {0, ±1, ±2, ±3, ...}.
QI = Set of rational numbers = { pq : p 2 ZZ, q 2 IN}.
IR = Set of real numbers
It applies: IN ⇢ ZZ ⇢ Q
I ⇢ IR.

Remarks:

(1) The order of the elements of a set is irrelevant. It is therefore

{a, b, c, d} = {d, c, a, b} .
(2) Each element of a set is written only once, which means
{a, a, a, b, d, d} = {a, b, d}.

Set Operations
Set Operations For two sets A and B , the intersection A \ B, the union A [ B and the com-
Mengenoperationen plement A\B are defined by

A \ B := {x : x 2 A and x 2 B} ,
A [ B := {x : x 2 A or x 2 B} ,
Intersection
A\B := {x : x 2 A and x 2
/ B} .
Schnittmenge

Here ”:=” means that the symbol on the left side is defined by the right side
Union of the equation. Similarly, ”:()” is to be read as logical equivalence according
Vereinigungsmenge
to the definition of what is on the sides of the colon.

Complement
Mengenkomplement

Figure 1.1. Venn Diagrams

By means of Venn Diagrams (see Fig. 1.1) sets and set operations can be repre-
Venn Diagrams sented pictorially. To the left the intersection of two sets A \ B , in the middle
Venn-Diagramme the union of the sets A [ B and on the right the complement of the set A and
B are indicated. The Venn diagrams are used to easily illustrate the following
calculation rules for sets:
 1.2 Natural Numbers 5

1. A[B =B[A
2. A[A=A
3. A [ (B [ C) = (A [ B) [ C
4. A [ (B \ C) = (A [ B) \ (A [ C)
5. (A\B)\C = A\(B [ C)
6. A ⇢ B , A \ B = A , A [ B = B , A\B = ;

The Cartesian product of two sets M1 and M2 is the set consisting of all pairs Cartesian
(x, y), where x 2 M1 and y 2 M2 : Product
Kartesisches
Produkt
M1 ⇥ M2 := {(x, y) : x 2 M1 and y 2 M2 } .

Example 1.2. IR ⇥ IR consists of all pairs of real numbers. This is nothing else
but the plane of numbers; (x, y) is a point in this plane. Instead of IR ⇥ IR we
write shortly IR2 .

1.2
1.2 Natural Numbers
Natural numbers are the simplest arithmetic objects. They form the foundation Natural
of our number building. We call the entirety of all natural numbers the ”set of Numbers
natural numbers” IN. The expression ”natural” numbers for IN = {1, 2, 3, . . .} Natürliche Zahlen

is certainly well chosen, because children begin to count in this way and in
all cultures mathematical thinking begins with these numbers. The zero was
invented quite late by the Indians, in 300 B.C. and is nowadays added to the
natural numbers: IN0 .

It was not until the discovery of the zero that Indian mathematicians were able
to create, for the first time, a numerical system which has been adopted all over
the world and which only requires ten digits (including zero). The number sys-
tem is already described by A. Ries (1492-1559) in his second arithmetic book
1522. There is also the appreciation of the number zero! The fundamental prin-
ciple of natural numbers goes back to the mathematician Peano (1858-1939,
1889).
 6 1. Numbers, Equations and Systems of Equations

1.2.1 Peano’s Axioms

Peano’s Axioms
Peano Axiome (1) 1 is a natural number.

(2) For each natural number n there exists exactly one successor n0 , which
also belongs to the natural number set.

(3) There is no natural number whose successor is 1.

(4) The successors of two di↵erent natural numbers are di↵erent from each
other.

(5) A subset of the natural numbers contains all natural numbers, if two
conditions are fulfilled: (1) 1 belongs to the set and (2) the successor
n0 always belongs to the set for any member n.

Thanks to Peano’s axioms we are able to build up the set of natural numbers,
because we immediately get the following conclusions from the axioms:

Conclusions:

(1) The natural set of numbers has infinitely many di↵erent elements: Because
of (A1) there is at least one natural number: 1. Because of (A2) there is a
successor to 1, according to (A3) 6= 1: We call it 2. Because of (A2) there
is a successor to 2, according to (A3) 6= 1 and (A4) 6= 2: We call it 3 etc.

(2) The elements of natural numbers can be arranged in a certain order, whe-
reby all natural numbers are recorded step by step:

1; 2; 3; 4; 5; ..., n; n + 1; ...

Here n + 1 means the successor of n. This order naturally determines the

addition of natural numbers.

(3) Each subset M of the natural numbers M ⇢ IN, which contains 1 and with
n 2 M always have the successor n + 1 2 M , is equal to the set of all
natural numbers.

From the conclusion (3) we obtain a principle of proof, which is to be applied

to most important evidence methods of analysis, namely the Mathematical
Induction or in German literature Complete Induction.
 1.2 Natural Numbers 7

1.2.2 Mathematical Induction (Complete Induction)

In order to prove a statement A(n) for all natural numbers, it is sufficient to
point to conclusion (3):

Mathematical
Mathematical Induction: Induction
Vollständige
Step 1: Induction start: For n = 1 the statement is correct. Induktion

Step 2: Induction closure from n0 to n0 +1: If the statement is valid for

any natural number n0 , then it must also be correct for its successor
n0 + 1.

If both operations can be performed, the statement applies to all n 2 IN.

Because after Step 1 the statement for n = 1 is correct. According to Step 2
the statement is also correct for the successor 2. Once again, according to Step
2 the statement is also correct for the successor, i.e. 3 etc.

n(n + 1)
Example 1.3. 1 + 2 + ··· + n = (n 2 IN)
2

Proof with complete induction. The beginning of induction consists in

explicitly checking that the formula for n = 1 is correct. We therefore
insert n = 1 into both the left and right side of the equation: For n = 1
the left side of the equation is 1 and the right side is 1·2
2 = 1. Hence, the
formula for n = 1 is correct.

Induction closure from n0 to n0 + 1: Let n0 2 IN be arbitrary and let’s

assume the formula is correct for this n0 , i.e. 1 + 2 + · · · + n0 = n0 (n20 +1) .
Under these conditions, we must show that the formula then applies to
n0 + 1. We start with the left side of the equation; the sum now goes up
to n0 + 1. We use the induction assumption and simplify the term until
the right side for n0 + 1 is identified:

1 + 2 + · · · + n0 + (n0 + 1) = (1 + 2 + · · · + n0 ) + (n0 + 1)
n0 (n0 + 1)
= + (n0 + 1)
2
n0 (n0 + 1) + 2(n0 + 1)
=
2
(n0 + 1)(n0 + 2)
= .
2
This is the formula to be proven for n0 + 1.
 8 1. Numbers, Equations and Systems of Equations

Note: According to an anecdote, this formula goes back to F. Gauss (1777-

1855), who had to calculate the sum of the first 100 numbers. He decided to
form the sum of the first and the last, the second and the second last, the third
and the third last, and so on:
1 + 2 + 3 + · · · + 98 + 99 + 100 = (1 + 100) + (2 + 99) + (3 + 98) + · · · .
Thus, he received from the 100 addends only 100 2 , each with the value 101:
1 + 2 + 3 + · · · + 100 = 100·101
2 . In fact, both the formula and the calculation
method were already known to Adam Ries (1492-1559; 1522).

The following notations are introduced as abbreviations for sums and products:

Definition:
Sum (1) Sum: For the sum of al , al+1, . . . , an 2 IR we write
Summe
n
X
ak := al + al+1 + . . . + an .
k=l

Product
Produkt (2) Product: For the product of al , al+1, . . . , an 2 IR we write
n
Y
ak := al · al+1 · . . . · an .
k=l
Factorial
Fakultät (3) Factorial: For each n 2 IN we define

n! := 1 · 2 · . . . · n (Factorial of n ) and 0! := 1.

Besides, n! is growing very fast. For example 13! ⇡ 6·109 ; to count this number,
it would take 100 years if we would be able to count to 100 in one minute!

Examples 1.4:
10
X
1 i2 = 52 + 62 + 72 + 82 + 92 + 102 = 355.
i=5
X5
1 1 1 1 1 1 137
2 = + + + + = .
i=1
2i 2·1 2·2 2·3 2·4 2·5 120
Y6
3 (2i 1)2 = 52 · 72 · 92 · 112 = 12006225.
i=3
4 5! = 1 · 2 · 3 · 4 · 5 = 120.
 1.2 Natural Numbers 9

P
n
Example 1.5. 1 + 3 + 5 + · · · + (2n 1) = (2k 1) = n2 (n 2 IN)
k=1

Proof with complete induction. We start the induction with n = 1 by

inserting n = 1 into both the left and right sides of the equation: 1 = 12 .
So the formula is correct for n = 1.

Induction closure from n to n + 1: If n is arbitrary and the formula is

correct for this n, then we shown that the formula also applies to n + 1:
1 + 3 + 5 + · · · + (2n 1) + (2n + 1) = [1 + 3 + · · · + (2n 1)] + (2n + 1)
= n + 2n + 1 = (n + 1)2 .
2

This is the formula for n + 1.

Warning: The principle of the complete induction refers to both the

beginning of induction and the induction closure. If one of these two parts
is missing, the proof is incomplete and the statement might not be correct,
as the following two examples show:

4
! Example 1.6. According to L. Euler (1707-1783) the expression

p = n2 n + 41

produces for n = 1, 2, 3, . . . , 40 prime numbers: p = 41, 43, 47, . . . , 1601. They

can be calculated explicitly by inserting the corresponding n. But this is not
enough for a general proof. For n = 41 it follows p = 412 41 + 41 = 412 .
This is indeed not a prime number. The prime calculation is correct for many
individual n, but in general it is not valid !

4
! Example 1.7. We are examining the wrong formula

n(n + 1)
1 + 2 + 3 + ··· + n = +1
2
(see Example 1.3) and show that the induction conclusion is feasible: So we
assume that the formula is correct for n and show that it is subsequently also
valid for n + 1.
n(n+1)
1 + 2 + 3 + · · · + n + (n + 1) = 2 + 1 + (n + 1)
n(n+1)+2(n+1)
= 2 +1
(n+1)(n+2)
= 2 + 1.

This is the formula for n + 1. Although the induction closure is feasible, the-
re exist no natural numbers n for which the formula would be correct. The
induction closure thus loses its meaning if the proof for n = 1 or for another
fixed integer value cannot be provided.
 10 1. Numbers, Equations and Systems of Equations

1.2.3 Geometric Sum

The geometric sum is important for many applications:

Geometric Sum
Geometrische Statement: (Geometric Sum Formula).
Summe
n
X 1 q n+1
For each real number q 6= 1: qi = (n 2 IN0 )
i=0
1 q

Proof by complete induction.

X 0
1 q1
Induction start n = 0: qi = q0 = 1 = .
i=0
1 q
Induction closure from n to n + 1: If n is arbitrary and the formula is
correct for n, then we check if it is valid also for n + 1:
P
n+1 P
n 1 q n+1
qi = q i + q n+1 = + q n+1
i=0 i=0 1 q
1 q n+1 + (1 q)q n+1 1 q n+2
= = .
1 q 1 q

1.2.4 Permutations
Permutation Permutation of a set means all possible arrangements of the elements of the
Permutation set. If A = {a1 , a2 , a3 , . . . , an }, there is exactly one element for each position
in the set. Another arrangement of the set would be for example

{a2 , a1 , a3 , . . . , an }.

The following statement gives an indication of how large the number of all
di↵erent arrangements of a set with n elements is:

Statement: The number of all possible arrangements of a set with n

elements a1 , . . . , an is equal to n! = 1 · 2 · 3 · . . . · n.

Proof by complete induction. The induction start is again at n = 1: The

number of arrangements of the 1-element set {a1 } is 1. Because of 1! = 1
the number of orders of the 1-element set {a1 } is also 1, thus proving the
formula for n = 1.

Induction closure from n to n + 1: We are looking for the number of all

arrangements of a (n + 1)-elemental set {a1 , a2 , a3 , . . . , an+1 }. We will
 1.2 Natural Numbers 11

consider the special element a1 and all its positions in this set. a1 can
be at the 1st position, then there are n! arrangements for the remaining
n elements after the induction assumption. a1 can also be at the 2nd
position, then there are n! arrangements for the remaining n elements after
induction. a1 can also be at 3rd place; again there are n! arrangements
for the remaining n elements etc. a1 can therefore be at n + 1 di↵erent
positions, and the remaining n elements still have n! di↵erent possible
arrangements. So there are n! · (n + 1) = (n + 1)! possibilities altogether.

Conclusion: The number of k -elemental subsets of a n-elemental set {a1 , . . . , an }

is equal to
n!
.
k!(n k)!

Proof: All k -elemental subsets of the set M = {a1 , . . . , an } can be found

by taking only the first k elements from all n! arrangements. Each k -
elemental subset occurs k! times and the remaining (n k)! times. Thus
the k -elemental subsets are equal to k!(nn! k)! .

Application: The chance to guess the right combination in the Lotto game ”6
out of 49” is about 1:14 million. Because the number of the 6-elemental subsets
of a 49-element set is equal to 6!49!
43! =
44·45·46·47·48·49
1·2·3·4·5·6 = 13.983.816.

1.2.5 The Binomial Theorem

For two natural numbers n and k with 0  k  n the following number is
introduced
Binomial
! Coefficient
n n! n(n 1) . . . (n k + 1) Binomialkoeffizient
:= = .
k k!(n k)! k!

We speak n over k , and call them binomial coefficients. The binomial coeffi-
cients are determined either by the above formula, or by the scheme named
after Pascal, the so-called Pascal’s triangle: Starting with 1: the pyramid
given below is extended by 1 on the right and left in each step. The numbers
in the scheme result from the sum of the two numbers above.
 12 1. Numbers, Equations and Systems of Equations

0
k : 1

1
k : 1 1

2
k : 1 2 1

3
k : 1 3 3 1

4
k : 1 4 6 4 1

5
k : 1 5 10 10 5 1

6
k : 1 6 15 20 15 6 1
..
.

Remark: With the binomial

! coefficients we can briefly formulate the last con-
n
clusion: There are possibilities to choose k from n objects. From this
k
statement we conclude the binomial theorem:

Binomial
Theorem Statement: (Binomial Theorem). For any real numbers a, b 2 IR and
Binomischer Satz
any natural number n 0 the following applies:

n
!
X n
n
(a + b) = an k
bk
k
k=0

Proof: If we expand the right side, the term

! bk occurs as often as we can
n
choose k factors from n factors, i.e. -times (see remark above). The
k
remaining (n k) factors contribute to an k .

Examples 1.8:
1 (x + y)0 = 1
(x + y)1 = x + y
(x + y)2 = x2 + 2xy + y 2
(x + y)3 = x3 + 3x2 y + 3xy 2 + y 3
2 We calculate the value of the power (104)3 with this theorem:
(104)3 = (100 + 4)3 = 1003 + 3 · 1002 · 4 + 3 · 100 · 42 + 43
= 1 000 000 + 120 000 + 4 800 + 64 = 1.124. 864.
 1.3 Real Numbers 13

1.3
1.3 Real Numbers
Let’s assume that the real numbers are available and let’s not deal with its
axiomatic structure. For physical measurements the rational numbers would
be sufficient but for the higher analysis the rational numbers show ”too many
holes”. Only their extension to real numbers makes calculus possible.

1.3.1 Number Sets and Operations

On the natural numbers IN there are + and · as basic arithmetic operations.
The equation x + 1 = 0 can be formulated within IN, but cannot be solved.
Therefore, the number range is extended by all the solutions of the equations,

x + n = 0,

if n 2 IN0 . The solutions are 0, 1, 2, 3, . . . and the extended number range

is called ZZ, the integers. In ZZ the equation x + n = 0 can be solved for every
n 2 ZZ. Not solvable is the equation 2x = 1. To solve these equations we have
to extend ZZ by all solutions of equations of the form

q·x=p

with p, q 2 ZZ and q 6= 0. Thus, the set of rational numbers Q

I is obtained.
In this number range all equations of the above form are solvable. But the
equation
x2 = 2

doesn’t have a solution in Q.

I Thus, the rational numbers are extended by all
the solutions of equations of the above construction and the real numbers IR
are obtained. Within the real numbers there are also the so-called transcendent
numbers, such as e and ⇡ , which we examine in more detail in the chapter on
Sequences. Table 1 shows the number ranges with the corresponding arithmetic
operations in an overview.

Table 1: Number Sets and Basic Arithmetic Operations

Sets Basic Operations Not Solvable
IN0 natural numbers + · x+1=0
ZZ whole numbers + · 2·x=1
QI rational numbers + · \ x2 = 2
2
IR real numbers + · \ x +1=0

Representation of Real Numbers. The known number line directed from

1 to +1 serves to illustrate the real numbers. Each point on the number
line corresponds exactly to a real number.
14 1. Numbers, Equations and Systems of Equations

Figure 1.2. Real number line

1.3.2 Calculation Laws for Real Numbers

In IR there are two operators, + and ·. Addition and multiplication of two
real numbers return real numbers. Formally, two mappings + and · have been
defined:

+ : IR ⇥ IR ! IR with (x, y) 7 ! x + y

· : IR ⇥ IR ! IR with (x, y) 7 ! x · y.

The calculation laws of addition are applicable

(A1) x + (y + z) = (x + y) + z Associative Law

(A2) x+y =y+x Commutative Law
(A3) x+0=x Existence of Zero
(A4) To each x there is a ( x) 2 IR with
x + ( x) = 0 Inverse Element

The calculation laws of multiplication are applicable

(M 1) x · (y · z) = (x · y) · z Associative Law
(M 2) x·y =y·x Commutative Law
(M 3) There is a number 1 2 IR with 1 6= 0, so that
1·x=x Existence of Zero
(M 4) For every x 2 IR\{0} there is a x 1 2 IR with
x·x 1 =1 Inverse Element

The Distribution law is applicable

(D) x · (y + z) = x · y + x · z

All further calculation laws of the real numbers can be traced back to these
elementary laws. Since these calculation laws are not only valid for the set
of real numbers, but also for other constructions, the concept of a field is
 1.3 Real Numbers 15

introduced and generalized: A set K together with two operations

+:K ⇥K !K with (x, y) 7 ! x + y

Field
Körper
· : K ⇥K !K with (x, y) 7 ! x · y,

that satisfy the axioms (A1)-(A4), (M 1)-(M 4), (D), is called field.

Examples 1.9:
1 Both (IR, +, ·) and (Q
I , +, ·) form fields.

2 (ZZ, +, ·) is not a field because (M 4) is violated: For example 2 2 ZZ has

no inverse element with respect to multiplication, such that 2 · x = 1.

3 (IN, +, ·) is not a field, since (A4) is violated.

4 (F2 , +, ·) with F2 = {0, 1} and its operations is a field.

+ 0 1 · 0 1
0 0 1 0 0 0
1 1 0 1 0 1
The calculation laws are checked directly. F2 is the smallest field; because
each field must contain at least two elements: 0 and 1.

1.3.3 Power Calculations

For each real number a 2 IR we define the power of a by means of
Power
0 1 n Potenz
| · .{z
a := 1, a := a, a := a . . · a} (n 2 IN).
n times

n-th Root
Definition: The n-th root of a number a 0 Wurzel
p 1
b := n a := a n (n 2 IN)

is defined as the positive real number b with property bn = a.

p 1
With the notation n a = a n the n-th roots of a number can be interpreted
as powers with rational exponents. For the powers of products or quotients of
real numbers, the general power calculation rules apply, which are summarized
 16 1. Numbers, Equations and Systems of Equations

in the following overview:

The power calculation rules that are applicable

an ⇣ a ⌘n
(1) an · bn = (a · b)n (2) n = for (b 6= 0)
b b
an
(3) = an m for (a 6= 0) (4) (am )n = an·m
am
p
(5) n am = am/n for (a 0) (n, m 2 IN)

Examples 1.10:

a5x 2y a4x+y a5x 2y 4x y ax 3y

1 : = = .
b6m 1 bm 2 b6m 1 m+2 b5m+1
(a2 b)2 1 1 5 3
2 p = a4 b2 12 a 1 a 2 b 2 = 12 a 2 b 2 .
2a ab
(8a3 b 3 ) 2 8 2 a 6 b6 33
3 = = .
(12a 2 b 4 ) 3 12 3 a6 b12 a12 b6

1.3.4 Logarithm

Logarithm
Definition: Given is the equation a = bx (a, b > 0). For given a and b
Logarithmus
we are looking for the exponent x. We call

x = logb (a)

the logarithm of a to the base b.

For a fixed base b the logarithm calculation rules are applicable

(1) log(u · v) = log(u) + log(v) (u, v > 0)

(2) log( uv ) = log(u) log(v) (u, v > 0)

(3) log(un ) = n · log(u) (u > 0)

Special logarithms are the logarithm to the base 10 log a := log10 a,

the logarithm to the base 2 (logarithm dualism) ld a := log2 a,
and the logarithm to the base e (natural logarithm) ln a := loge a.
 1.3 Real Numbers 17

There is a relation between logarithms with respect to di↵erent bases

logc (y)
logb (y) = (b , c, y > 0).
logc (b)

Thus, it is sufficient to be able to calculate the logarithm with respect to only

one base (e.g. the natural logarithm). The logarithm to another basis is then
calculated by the formula above.

Proof of the logarithmic formula: From bx = y it follows by definition of

the logarithm to the base b that x = logb y . On the other hand, for the
logarithm to the base c according to the logarithm rule (3) it is valid:
logc y
logc y = logc bx = x · logc b ) x= .
logc b
This proves the above formula.

Examples 1.11:
1 1
1 2x = 8 ) x = log2 8 = log2 8 = 3.
2 10x = 0.0001 ) x = log10 10 4 = 4 log10 10 = 4.
p p
ab 2 1
3 ln p3 cd 3 = ln a + ln b 2 ln c 3 ln d 3 = 12 ln a 2 ln b 1
3 ln c + 3 ln d.
qp
3
p4 1 1 1 1 1 1 1 1
4 log a2 b a c2 = log((a2 b a 4 c 2 ) 3 ) 2 = log a 3 b 6 a 24 c 12
9 1 1
= 24 log a + 6 log b + 12 log c.

1.3.5 Arrangement of the Real Numbers

Among the real numbers there is a certain order: Two real numbers a, b 2 IR
are always in one of the following three relationships to each other:

a<b (a is to the left of b),

a=b (a is identical to b),

a>b (a is to the right of b).

 18 1. Numbers, Equations and Systems of Equations

The absolute value of a real number a is the distance from a to the zero
point. It is indicated by the symbol |a|:
Absolute Value 8
Betrag >
< a for a > 0
|a| := 0 for a = 0
>
: a for a < 0

1 1
p p
Examples 1.12: |3| = 3 ; | 5| = 5 ; 2 = 2 ; 2 = 2.

The distance between two numbers x and a on the number line is |x a| .

Figure 1.3. Distance between two numbers

According to the definition of the absolute value, we have to distinguish bet-

ween two di↵erent cases

|x a| = (x a) if x a 0 or x a,
|x a| = (x a) if x a < 0 or x < a.
This clause will play an important role when solving equations and inequalities
that contain absolute values.

The following rules are applicable:

(1) x > 0 , y > 0 ) x + y > 0.

(2) x > 0 , y > 0 ) x · y > 0.
(3) If x > 0 , y > 0, then there is always a natural number n 2 IN with

nx>y (Archimedes0 Axiom).

Conclusion: (Bernoulli’s Inequality)

(1 + x)n 1 + nx (x 1 and n 2 IN)

Proof by complete induction. For n = 1 the equality applies. Induction

closure from n to n + 1: Due to 1 + x > 0, multiplying the induction
requirement by (1 + x):
(1 + x)n+1 (1 + nx)(1 + x) = 1 + (n + 1)x + nx2 1 + (n + 1)x.
 1.4 Equations and Inequalities 19

Intervals: Subsets of real numbers are called real range or interval. A distinc-
tion is made between finite and infinite intervals. To describe these subsets of
IR we introduce the following notations: Intervals
Intervalle

(1) Finite intervals (a < b)

[a, b] := {x : a  x  b} ) closed interval

[a, b) := {x : a  x < b}
half open intervals
(a, b] := {x : a < x  b}
(a, b) := {x : a < x < b} open interval

(2) Infinite intervals

IR a := [a, 1) := {x : a  x < 1}
IR>a := (a, 1) := {x : a < x < 1}
IRa := ( 1, a] := {x : 1 < x  a}
IR<a := ( 1, a) := {x : 1 < x < a}

1.4
1.4 Equations and Inequalities
The methods of solving equations are as diverse as there are types of equations. We show
exemplarily how basic equations and inequalities can be solved. However, we will not deal
systematically with solving equations or inequalities, because in many cases they cannot
be solved exactly and therefore we depend on numerical methods (e.g. see the bisection
method §6.4 or Newton’s method §7.8 ).

1.4.1 Equations
Any relationship between (real) quantities in which an equal sign occurs is Degree of
called equation. If the quantities of the equation occur only as sums and pro- Equation
ducts of powers, they are called algebraic equations. The largest exponent Grad der Gleichung

indicates the degree of the equation.

 20 1. Numbers, Equations and Systems of Equations

1. Quadratic Equations
The equation
x2 + px + q = 0

is called quadratic equation or equation of second degree. If the equation is in

form of ax2 + bx + c = 0 with a 6= 0, we divide by a to get it in the form p/q .
For the quadratic equation x2 + px + q = 0 there is the p/q solution formula
p/q Formula
p/q-Formel
r
p p2
x1/2 = ± q.
2 4

2
We introduce D := p4 q the (discriminant). For D > 0 the equation has two
di↵erent real solutions, for D = 0 a double real solution, and for D < 0 no real
(but two di↵erent complex) solutions.

Examples 1.13:
1 x2 + 2x 3 p
= 0 has two real solutions:
x1/2 = 1± ( 1)2 + 3 = 1±2. This means x1 = 1, x2 = 3.
2 x2 + 4x + 4 p
= 0 has a double real solution:
x1/2 = 2± 22 4 = 2. So x1 = 2 is a double solution.
3 x2 4x + 13 = 0 has no real (but two complex) solutions:
p p
x1/2 = 2 ± 4 13 = 2 ± 9 = 2 ± 3i. There are no real solutions. Here
i means the imaginary unit (see Chapter 5, Complex Numbers).

2. Higher Degree Equations

Equations of higher degree are analytically only partly solvable, because in
the rarest cases a closed solution exists. Occasionally a substitution leads to a
quadratic equation, as the following example shows:

Example 1.14. We are looking for solutions of the equation

x4 5x2 + 4 = 0.

We substitute z = x2 and get the quadratic equation

z2 5z + 4 = 0.

If we apply the p/q formula

r r
5 25 5 25 16 5 3
z1/2 = ± 4= ± = ± .
2 4 2 4 2 2
 1.4 Equations and Inequalities 21

p
Therefore, z1 = 4, z2 = 1. Because of z = x2 and x = ± z , the following
applies
p
x1/2 = ± 4 = ±2
p
x3/4 = ± 1 = ±1.

The solutions of the equation are x = 2, 1, 1, 2.

Sometimes it is possible to guess a zero, and then reduce the problem by

polynomial division or, by applying Horner’s scheme. We will describe these
techniques in detail in the section on polynomials §4.2.

3. Root Equations
Simple root equations are solved by isolating the root, e.g. by placing it on the
left side of the equation, the remaining terms on the right side. Then, square
root equations are squared and resolved according to the variable being sear-
ched for.

4
! Caution: By squaring the equations we might change the solution set: The
equation x = 1 has as solution only the number 1. If we square the equation,
x2 = 1 is obtained. This equation has both x = 1 and x = 1 as solutions. The
squared equations can have more solutions than the original ones. Therefore,
it is necessary to check whether the resulted numbers really represent solutions
of the original root equation.

Example 1.15. We are looking for solutions of the root equation

p
5 x + 3 x = 0.

We isolate the root

p
5 x=x 3,

square it
5 x = x2 6x + 9

and reformulate it as
x2 5x + 4 = 0.

This is a quadratic equation that can be solved with the p/q formula.
Example 1.14 returns the solutions

x1 = 4, x2 = 1.

Now, we will check whether the two values are solutions of the original
root equation. For this, we insert the values into the root equation:
22 1. Numbers, Equations and Systems of Equations

p
x1 = 4: p1 1 = 0. That means x1 = 4 satisfies the root equation.
x2 = 1: 4 + 2 = 4 6= 0. That means x1 = 1 does not satisfy the root
equation.
Thus the solution set of the root equation is IL = {4}.

4. Absolute Value Equations

The absolute value |a| of a real number a is defined by the distance to the zero
point
(
a if a 0
|a| := .
a if a < 0

This definition always leads to a clause distinction in an absolute value equati-

on: Depending on whether there is a positive term in the absolute value, then
the absolute value is replaced by a parenthesis (), or whether there is a negative
term in the absolute value, then it is replaced by -( ).

Example 1.16 (With Maple-Worksheet). We are looking for the solutions

of the absolute value equation

|4x 1| = 2x + 4.

To get an overview of the two functions, we draw the left and right sides of the
equation:

It can be seen that there are two intersections of the graphs that need to be
determined.

Case 1: 4x 1 0, that is x 14 :
The absolute value for x 14 are non-negative numbers. Under these con-
ditions, the absolute value can be replaced by a single parenthesis. In this
case
4x 1 = 2x + 4.
 1.4 Equations and Inequalities 23

Solving for x we get

5
6x = 5 respectively x = .
6
5 1
x= 6 satisfies the x 4 condition under which we calculated the result.

Case 2: 4x 1 < 0, that is x < 14 :

For x < 14 there are only negative numbers. Under these conditions, the
absolute value is replaced by -( ). In this case

(4x 1) = 2x + 4 ) 4x + 1 = 2x + 4.

Solving for x we get

2
2x = 3 respectively x = .
3
2 1
x= 3 satisfies the x < 4 condition under which we calculated the result.

Thus, the solution set is IL = { 56 , 3

2 }.

Reference: Due to the widespread use of computers to solve mathematical

problems, in particular to solve equations, numerical methods are becoming
more and more important. Therefore, in this textbook two sections are dedi-
cated to numerical methods.

1.4.2 Inequalities
Equivalent operations acting on inequalities are:

Addition (or subtraction) of any term on both sides of the inequality.

Multiplication (or division) of both sides with a positive number K>0.
Multiplication (or division) of both sides with a negative number K<0;
this changes the inequality sign from
(< to >), ( to ), (> to <), ( to ).

Example 1.17 (With Maple-Worksheet). We are looking for solutions of

the absolute value inequality

|2x + 2| > 3.

As in the case of equations with absolute values, when inequalities are resolved,
the absolute value must first be resolved by a case distinction.

Case 1: 2x + 2 0, that is x 1:
The absolute value for x 1 does not contain negative numbers. Then
the absolute value characters can be replaced by a single parenthesis (...).
24 1. Numbers, Equations and Systems of Equations

In this example
2x + 2 > 3.

Solving for x we obtain

1
x > .
2
x > 12 satisfies the x 1 condition under which we performed the calcu-
lation.
1
) IL1 = ( , 1).
2

Case 2: 2x + 2 < 0, that is x < 1:

The argument of the absolute value for x < 1 is negative. Under this
condition, the absolute value are replaced by -(...). In this case

(2x + 2) > 3.

We multiply with ( 1). Hence, the inequality sign changes

2x + 2 < 3

and solving for x we get

5
x< .
2
5
x = 2 satisfies the x < 1 condition under which we calculated the
result.
5
) IL2 = ( 1, ).
2
5
Hence, the solution set consists of two sub-intervals, the open interval ( 1, 2)
combined with the open interval ( 12 , 1):
5 1
IL = IL1 [ IL2 = ( 1, ) [ ( , 1).
2 2

Example 1.18 (With Maple-Worksheet). We are looking for the solutions

of the absolute value inequality

(x 2)2  |x| .

To get an overview of the two functions, we draw the left and right sides
of the inequality:
 1.4 Equations and Inequalities 25

It can be seen that there are two intersections of the graphs that need
to be determined. The solution set then consists of the closed interval in
which the square function (x 1)2 is less than or equal to the absolute
value function |x|.

Determination of the intersection points: The graph shows that the

intersection points are in the positive x range. In this area, we can replace
the absolute value characters |x| with a simple parenthesis, so that we
have to solve the equation

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

The p/q formula returns

r r
3 9 3 5
x1/2 = ± 1= ± .
2 4 2 4

Thus x1 = 0.38 and x2 = 2.62 are the intersections of the curves. The
solution set consists of the closed interval.

3 1p 3 1p
IL = 5, + 5 = [0.38, 2.62] .
2 2 2 2
 26 1. Numbers, Equations and Systems of Equations

1.5
1.5 Systems of Linear Equations
Gauss Systems of Linear Equations (LEq) play a very important role in theory and ap-
Algorithm plications. In this section we introduce a method to solve any LEq: the Gauss
Gauss Algorithmus algorithm. For general correlations and statements about LEq, we refer to Chapter
3, Matrices and Determinants.

1.5.1 Introduction

Application Example 1.19 (Description of DC Circuits).

Given is the electrical network with the
resistors R1 = 1⌦ , R2 = 5⌦ , R3 = 3⌦.
Two DC currents IA = 1A and IB = 2A
are fed into this network. The individual
currents I1 , I2 , I3 are to be found.

To obtain the model equations, we ap-

ply Kirchho↵ ’s laws: The node rule
Kirchho↵’s Figure 1.4. Electrical network states that the sum of the incoming and
Laws outgoing currents in a node is zero. The
Kirchho↵s Gesetze
mesh rule states that in a mesh the sum of all voltages results in zero.

Node Rule
In this example, for the node KA , I3 flows into and IA , I1 flow out of
Knotensatz

(KA ) : I3 = IA + I1 ;
Mesh Rule
Maschensatz for the node KB , IB flows into and I1 , I2 flow out of

(KB ) : IB = I1 + I2 .

For the mesh with specified current directions, the voltage drop over R2 is
equal to the sum of the voltage drops over R1 and R3 :

(M ) : R1 I 1 + R3 I 3 = R2 I 2 .

This results in a system of 3 equations for the single currents I1 , I2 , I3 .

 1.5 Systems of Linear Equations 27

Method for solving LEq: We substitute the given values into the equations
and introduce a method to systematically solve the system:
I1 I2 I3 r.S.
G1 : 1I1 5I2 + 3I3 = 0 1 5 3 0
G2 : 1I1 + 1I3 = 1 1 0 1 1
G3 : 1I1 + 1I2 =2 1 1 0 2

This system is solved by eliminating I1 from equations G2 and G3 . The sum

of equations G1 and G2 and the di↵erence between equations G1 and G3 is
calculated:
I1 I2 I3 r.S .
0
G1 = G1 : 1I1 5I2 + 3I3 = 0 1 5 3 0
G02 = G1 + G2 : 5I2 + 4I3 = 1 0 5 4 1
G03 = G1 G3 : 6I2 + 3I3 = 2 0 6 3 2

0 0 0
Then we take the equation G2 and G3 and eliminate I2 from G3 . To do this
0 0
we add 6 times equation G2 to ( 5) times of equation G3 :

30I2 + 24I3 = 6
30I2 15I3 = 10
9I3 = 16

This will finally result in

I1 I2 I3 r.S .
00
G1 = G 1 : 1I1 5I2 + 3I3 = 0 1 5 3 0
00
G2 = G02 : 5I2 + 4I3 = 1 0 5 4 1
00 0 0
G3 = 6G2 5G3 : 9I3 = 16 0 0 9 16

00
From equation G3 we get
16
9I3 = 16 ) I3 =
9
00
and from equation G2
16 11
5I2 + 4 · = 1 ) I2 = .
9 9
00
Both results substituted into equation G1 gives
11 16 7
I1 5· +3· = 0 ) I1 = .
9 9 9
Hence, the currents I1 , I2 , I3 are calculated.
 28 1. Numbers, Equations and Systems of Equations

In the last column, the variables are not specified; only the coefficient of the
variables and the constants of the right side of the equation are listed. The
coefficient of I1 always appears first, the coefficient of I2 second, and the coef-
ficient of I3 finally third. In principle, this short version of the equation system
is sufficient to solve it. This procedure to solve this system of equations can be
generalized (!Gauss algorithm), if the searched variables occur only linearly
(!LEq).

1.5.2 Formulation and Notation

A linear relationship between two values x and y exists, if x is proportional to
y (x ⇠ y), i.e. a x + b y = const. More generally, an equation of the form

ax1 + bx2 + cx3 = d

is a linear equation in x1 , x2 , x3 since each of the variables x1 , x2 and x3 oc-

cur only in linear form, i.e. to the power of 1. Every triple of real numbers
(x1 , x2 , x3 ) 2 IR3 = IR ⇥ IR ⇥ IR that satisfies the equation is called solution.

Examples 1.20:
1 x1 x2 + x3 = 0 is a linear equation, because the variables x1 , x2 , x3 are
represented proportional in the equation. This equation has for example
(0, 1, 1), (1, 1, 0), (1, 2, 1) as solutions.
2 The equation x2 +2x y = 0 is not a linear equation, because the variable
x is square.
3 x1 x2 + x3 = 0 and 2x1 + 3x2 x3 = 0 form a system of linear equations,
a linear system of equations.

LEq Definition: A system of m linear equations in the n unknowns

LGS x 1 , x2 , . . . , x n

a11 x1 + a12 x2 + ··· + a1n xn = b1

a21 x1 + a22 x2 + ··· + a2n xn = b2
.. .. .. ..
. . . .
Coefficients of am1 x1 + am2 x2 + · · · + amn xn = bm
LEq
Koeffizienten is called a system of linear equation (LEq) . The real numbers aij are
the coefficients and bi the constants of the right side of the LEq.
 1.5 Systems of Linear Equations 29

Abbreviating for the LEq the coefficients and the right side are written in
a compact scheme
0 1
a11 a12 a13 · · · a1n b1
Ba C
B 21 a22 a23 · · · a2n b2 C
B . .. .. .. .. C
B . C
@ . . . . . A
am1 am2 am3 · · · amn bm Coefficient
Matrix
This scheme is called the extended coefficient matrix or in short Matrix. Koeffizientenmatrix

The solid line should remind you that the coefficients are to the left and
the constants to the right of the equal sign. Matrix
Matrix

A system of linear equations means homogeneous if all bi are equal to

zero, otherwise inhomogeneous.

Each row of the matrix represents an equation; each column is assigned to the
corresponding unknown. The solution consists of all n-tuples (x1 , x2 , ..., xn ),
which fulfill all m equations.

As we saw in the introductory example, the elimination method only changes

the coefficients and the constants, but not the variables. Therefore, when sol-
ving LEq, the variables are completely omitted and all calculation steps are
carried out in matrix notation.

1.5.3 Solving Systems of Linear Equations

Operations that do not change the solution set of a system are called equiva-
lence manipulations. The following operations are equivalence manipulati-
ons of a system of linear equations:

Equivalence
Manipulations
(1) The order of the equations can be changed.
Umformungen

(2) An equation can be multiplied by any real number 6= 0.

(3) An equation of the system can be added to another one.

 30 1. Numbers, Equations and Systems of Equations

If these three manipulations are applied systematically - as described below

- the solution of each LEq can be determined. As shown in the introductory
example, in each calculation step a variable is eliminated from the system and
thus reduced by one equation, until only one equation remains for one variable.
The method, which goes back to Gauss (1777-1855), is called the Gaussian
elimination method or the Gaussian algorithm. For simplification, we li-
mit ourselves to quadratic systems with n equations for n unknowns. But the
Gauss algorithm is extendable to arbitrary (n ⇥ m) systems.

Elimination Method (Gaussian Algorithm)

Elimination
Method
Eliminationsverfahren (1) Choose an equation with a non-zero coefficient of x1 as the first
equation.

(2) Eliminate the variable x1 from the remaining (n 1) equations. The

first line is multiplied by aa21
11
and added to the second equation.
aj1
The same applies to the remaining lines: Add the a11 -fold of the
1st line to the j-th line. Thus we get (n 1) equations with (n 1)
Gauss unknown x2 , x3 , . . . , xn .
Algorithm
Gauss Algorithmus
(3) Step (2) is applied to the reduced system by eliminating the unknown
x2 from lines 3 to n. After a total of (n 1) steps, only one equation
for the unknown xn remains.

(4) The eliminated equations form a scaled system of lines from which the
unknowns can be calculated in the order xn , xn 1 , . . . , x2 , x1 .

4
! The above algorithm assumes that none of the coefficients aii equals zero;

otherwise the lines must be swapped. If all remaining coefficients of the variable
xi to be eliminated are equal to zero, this step can be skipped, because the
LEq already has the desired form. However, when the algorithm is executed
numerically, calculation inaccuracies occur even if these coefficients are very
small. In order to keep such errors as small as possible, it is advisable to swap
0
the lines in each step so that the line with the largest coefficient aii is selected
as the uppermost equation. This is called pivoting.
 1.5 Systems of Linear Equations 31

Examples 1.21:
1 A system with exactly one solution: The solution set of the LEq is

searched for LEq

2x1 + x2 x3 = 3
3x1 + 5x2 4x3 = 1
4x1 3x2 + 2x3 = 2
In matrix notation, this LEq is as follows
0 1
G1 : 2 1 1 3
B C
G2 : @3 5 4 1A
G3 : 4 3 2 2

For the solution we use the Gaussian algorithm. To do this, we copy the first
line; multiply G1 by (-3) and add the result to 2 times the second line. We also
multiply the first row by (-2) and add the result to the third row:
0
0 1
G1 : 2 1 1 3 (G1 )
0 B C
G2 : @0 7 5 7A (2G2 3G1 )
0
G3 : 0 5 4 4 (G3 2G1 )

Now, we leave the first two equations unchanged and modify the last equation
such that the coefficient of x2 becomes zero.
00
0 1 0
G1 : 21 1 3 (G1 )
00 B C 0
G2 : @0 7 5 7A (G2 )
00 0 0
G3 : 00 3 63 (7G3 + 5G2 )
From the equivalent system (00 ) now the solutions can be calculated easily. The
last equation returns
3x3 = 63 ) x3 = 21.
00
Used in G2 : 7x2 5 · ( 21) = 7 ) x2 = 16.
00
Both used in G1 : 2x1 + ( 16) ( 21) = 3 ) x1 = 1.
Hence, the system has one unique solution ( 1; 16; 21) and the solution
set is 8 0 1 0 19
>
< x1 1 > =
B C B C
IL = (x1 , x2 , x3 ) 2 IR3 : @ x2 A = @ 16 A .
>
: >
x3 21 ;

The system (00 ) is called a system with upper triangular matrix, because the Upper
00 00 00 Triangular
entries below the main diagonal (a11 , a22 , a33 ) are equal to zero. If the system Matrix
has an upper triangular shape, the elimination procedure is complete. The Obere
unknown x1 , x2 , x3 can then be determined by reverse resolution. Dreiecksmatrix
32 1. Numbers, Equations and Systems of Equations

2 The solution has a free parameter: In order to solve the system

x1 3x2 + 2x3 = 4
2x1 + x2 + 3x3 = 2
2x1 16x2 + 18x3 = 28
we modify the coefficient matrix in two steps so that we obtain triangular form
0 1
G1 1 3 2 4
B C
G2 @ 2 1 3 2A
G3 2 16 18 28
0
0 1
G1 1 3 2 4
0 B C
G2 @ 0 5 7 10 A (G2 + 2G1 )
0
G3 0 15 21 30 (G3 + G2 )
00
0 1
G1 1 3 2 4
00 B C
G2 @ 0 5 7 10 A
00 0 0
G3 0 0 0 0 (G3 3G2 )

From the last line follows 0 · x3 = 0 , which is fulfilled for any x3 . Therefore
00
we set x3 = (arbitrary). Inserted into G2 it follows
7
5x2 + 7 = 10 ) x2 = 2+ .
5
00
Both inserted into G1 gives
7 11
x1 = 4 + 3( 2 + ) 2 = 2+ .
5 5
To get a better representation, we select = 5k, such that the solution is
8 0 1 0 1 0 1 9
>
< x1 2 11 >
=
3 B C B C B C
IL = (x1 , x2 , x3 ) 2 IR : @ x2 A = @ 2 A + k @ 7 A with k 2 IR .
>
: >
;
x3 0 5

3 The system has no solution: We look at the system from 2 by mo-

difying the last equation: The constant 28 is replaced by 27. By elementary
manipulations we obtain
0 1
1 32 4
B C
@ 0 5 7 10 A .
0 00 1

From the last line it follows 0 · x3 = 1 . This equation cannot be fulfilled

because the left side always returns zero. Therefore, IL = {} .

4 Homogeneous LEq: According to example 2 we can immediately identify

the solution set of the homogeneous LEq
 1.5 Systems of Linear Equations 33

x1 3x2 + 2x3 = 0
2x1 + x2 + 3x3 = 0 .
2x1 16x2 + 18x3 = 0
Because the elementary line manipulations give
0 1
1 32 0
B C
@0 5 7 0A.
0 00 0

By backward resolving we get 0 · x3 = 0 from line 3. Therefore x3 is arbitrary.

We set x3 = 5 k. Inserted into line 2, follows

5x2 + 7 · 5k = 0 ) x2 = 7k

and both are inserted into line 1:

x1 = +3 · 7k 2 · 5k = 11k.

Therefore
8 0 1 0 1 9
>
< x1 11 >
=
B C B C
IL = (x1 , x2 , x3 ) : @ x2 A = k @ 7 A and k 2 IR .
>
: >
;
x3 5

Examples 1 - 4 suggest the following general conclusion, which we examine

in more detail in the chapter: Matrices and Determinants:

Solution Behavior of Systems of Linear Equations

(1) An inhomogeneous LEq has either exactly one solution or infinitely

many solutions or no solution at all.

(2) A homogeneous LEq has either exactly one solution, namely the trivial
0 1
0
B .. C
zero solution x = @ . A, or infinite many solutions.
0
(3) If the inhomogeneous LEq is solvable, the general solution consists of all
homogeneous solutions plus one solution of the inhomogeneous system:

ILi = ILh + xs ,

if ILi = solution set of the inhomogeneous LEq, ILh = solution set of the
associated homogeneous LEq and xs is a special solution of the inhomo-
geneous system.
34 1. Numbers, Equations and Systems of Equations

Application Example 1.22 (Chemical Reaction).

From quartz (SiO2 ) and caustic soda (N a OH) we obtain as reaction sodium
silicate (N a2 SiO3 ) and water (H2 O):

x1 SiO2 + x2 N a OH ! x3 N a2 SiO3 + x4 H2 O

We are looking for the proportions of the substances x1 , x2 , x3 , x4 for which the
reaction takes place. Since only integer multiples are possible, natural numbers
x1 , x2 , x3 , x4 must be determined, so that each of the chemical elements Si,
O, N a, H occurs equally often on both sides of the reaction equation. This
leads to the following homogeneous system of linear equations:
Si : x1 = x3
Na : x2 = 2x3
O: 2x1 + x2 = 3x3 + x4
H: x2 = 2x4 .

In matrix form, the LEq is as follows

0 1 0 1
10 1 0 0 1 0 1 0 0
B0 1 2 0 0C B0 1 2 0 0C
B C B C
B C ,! B C
@2 1 3 1 0A @0 0 1 1 0A
01 0 2 0 0 0 0 0 0
The last line is followed by 0 · x4 = 0. Therefore x4 is arbitrary. We select
x4 = k. Inserted in line 3, follows x3 = k. Both results inserted into line 2 or
line 1 respectively returns x2 = 2k and x1 = k. The solution with the smallest
fractions of the substances is therefore (for k = 1)

SiO2 + 2 N a OH ! N a2 SiO3 + H2 O.

Application Example 1.23 (Mixing of Alloys).

Stainless steel is an alloy of iron, chromium and nickel. For example, V2A steel
consists of 74% iron, 18% chromium and 8% nickel. The table below shows
existing alloys (I - IV) with which 1000 kg of V2A steel is to be mixed.

I II III IV
Iron 70% 72% 80% 85%
Chrome 22% 20% 10% 12%
Nickel 8% 8% 10% 3%
 1.5 Systems of Linear Equations 35

If x1 , x2 , x3 , x4 are the proportions of alloys I - IV in units kg, then the

following applies to the sum of all mixture proportions in kg

x1 + x2 + x3 + x4 = 1000.

The conservation equations for the individual constituents iron, chromium and
nickel are as follows
0.7 x1 + 0.72 x2 + 0.8 x3 + 0.85 x4 = 740
0.22 x1 + 0.2 x2 + 0.1 x3 + 0.12 x4 = 180
0.08 x1 + 0.08 x2 + 0.1 x3 + 0.03 x4 = 80.

Note that for 1000 kg alloy with 74% iron, the iron weight is 740 kg. The same
applies to chromium and nickel. These four equations provide an inhomoge-
neous system of linear equations
0 1 0 1
1 1 1 1 1000 1 1 1 1 1000
B 70 72 80 85 74000 C B 0 2 10 15 4000 C
B C B C
B C ,! B C.
@ 22 20 10 12 18000 A @0 0 2 5 0A
8 8 10 3 8000 00 0 0 0

We choose x4 = k arbitrarily. From line 3, x3 = 52 k follows. Both results in line

2 and line 1, respectively, provide x2 = 2000 20k and x1 = 1000 + 16.5k.
For the solution to be feasible, all shares must be x1 , x2 , x3 , x4 0. Because
of the condition for x2 it follows 100 k . And due to the condition for x1
follows k 60.6. Which means for 100 k 60.6 the problem can be solved
physically.
 36 1. Numbers, Equations and Systems of Equations

1.6
1.6 Proofs

Axioms Mathematics can partly be regarded as a set of statements which are derived
Axiome (=proven) from basic statements in a purely logical way. These statements
are then generally valid and cannot be disproved under the given conditions
Statements (axioms). This is the principle of mathematics that goes back to Euclid (ca.
Sätze 300 BC). For the first time Euclid listed in his ”Elements” not observed laws
of nature, but proved mathematical laws (=statements). Since that time, this
way of proceeding has represented a fundamental di↵erence between mathema-
tics and natural sciences. There, a law of nature is considered to be confirmed
if several independent experiments repeatedly confirm the same statement. A
law of nature is valid until it is refuted by another experiment.

Although in this textbook the focus is more on the application of mathematics

than on rigorous mathematical proofs, the most important proof methods have
to be introduced.

1.6.1 Mathematical Induction

Mathematical induction is one of the most important elementary proof me-
thods in mathematics. A statement A(n) is true for all natural numbers if it is
explicitly checked for n = 1 and the statement A(n + 1) can be shown under
the assumption of A(n). This method is discussed in detail in section 1.2.2.

1.6.2 Direct Proof

A direct proof is the method of directly deriving a statement based on pre-
conditions or valid formulas. Examples are the proof of the binomial theorem
(1.2.5) or the following proof of the geometric sum formula:

Theorem (Geometric Sum:)

n
X 1 q n+1
For each real number q 6= 1 applies: qi = (n 2 IN0 )
i=0
1 q

Proof: We define
n
X
sn := qi = q0 + q1 + . . . + qn (1)
i=0

and multiply this equation by q

q · sn = q 1 + q 2 + . . . + q n+1 . (2)
 1.6 Proofs 37

By subtracting equation (2) from equation (1) we get

sn q · sn = q 0 q n+1 .

Therefore,
n
X 1 q n+1
(1 q) · sn = 1 q n+1 ) sn = qi = .
i=0
1 q

Thus, the geometric sum formula is directly proven.

1.6.3 Proof by Contradiction

Another frequently used method is the proof by contradiction. To prove a state-
ment the opposite is assumed which leads to an contradiction. As an example,
we take the Euclidean proof for the theorem that there are infinitely many
prime numbers.

Definition: A natural number p > 1 means Prime number, if it is divisible

only by 1 and by itself .

Theorem: There are infinitely many prime numbers.

Proof by Contradiction: Assume that there are only finite many prime
numbers, namely p1 , p2 , . . . , pn > 1. Then, we consider the natural number

m := p1 · p2 · . . . · pn + 1.

This number m is greater than 1, because the prime number 2 occurs as a fac-
tor. The number m cannot be another prime number, since we have assumed
that p1 , p2 , . . . , pn represent all primes. Therefore, m is divisible by at least
one pi 2 {p1 , p2 , . . . , pn }. pi thus divides both p1 · p2 · . . . · pn and 1. However,
this is a contradiction, since 1 does not have any dividers greater than 1. So
we have made the assumption (there are only a finite number of primes) a
contradiction. If there are not finitely many prime numbers, then there must
be infinitely many.

1.6.4 Counter Example

Giving a counter example to an assertion is also a possible form of evidence.
All prime numbers are odd. (Counter example is the number 2.)
With the formula p = n2 n + 41 prime numbers are obtained. (Counter
example n = 41; see example 1.6.)
 38 1. Numbers, Equations and Systems of Equations

1.7
1.7 Problems on Numbers, Equations and Systems of
Equations

1.1 Specify the following sets by enumerating their elements:

a) {x : x is prime and x < 20}
b) {x : x is real and x2 + 1 = 0}
1.2 The following sets are given A = {x 2 IR : 0 < x < 2} and B = {x 2 IR : 1 
x  3}. Determine graphically as well as computationally
(i) A \ B, (ii) A [ B, (iii) A ⇥ B, (iv) A\B.

1.3 Form the union, intersection and both complements of the following sets
a) M1 = {2, 4, 6, . . .}, M2 = {3, 6, 9, . . .}
b) M1 = {x : x2 + x 2 = 0}, M2 = {x : x2 3x + 2 = 0}

1.4 Given are three sets M1 , M2 , and M3 . Use Venn diagrams to show
a) M1 \ (M2 [ M3 ) = (M1 \ M2 ) [ (M1 \ M3 )
b) M1 [ (M2 \ M3 ) = (M1 [ M2 ) \ (M1 [ M3 )

1.5 Show by mathematical induction that all n 2 IN

Xn
n (n + 1) (2n + 1)
a) 12 + 22 + 32 + . . . + n2 = k2 =
6
k=1
Xn
b) 20 + 21 + 22 + . . . + 2n = 2k = 2n+1 1
k=0

1 1 1 1 n
c) + + + ... + =
1·2 2·3 3·4 n (n + 1) n+1

1.6 Show by mathematical induction

a) 2n  n! for each n 4
b) 2 n + 1  2n for each n 3
c) n2  2n for each n 6= 3

1.7 Calculate
! ! ! ! ! ! ! ! !
n n 3 3 4 4 4 4 4
a) , , , , , , , ,
0 n 1 2 0 1 2 3 4
b) 1024

1.8 Check that

Xn nX1 n
X1 n
X
ak 1 = ak ; ak+1 = ak .
k=1 k=0 k=0 k=1
!
n 1 1
1.9 Show nk
 k!
for each n 2 IN.
k
1.10 Develop the binomials
3
a) (x + 4)5 b) (1 5 y)4 c) a2 2b .

1.13 Simplify the expressions as much as possible

18 xa+4 4 x7 3a
a) : b) (an+1 bx 1 + an bx + an 1 x+1
b ) : (an 2 x 1
b )
2 y 5a+7 9 y 8+5a
 1.7 Problems on Numbers, Equations and Systems of Equations 39

1.14 Simplify formally the root terms as much as possible

x(2r2 4x2 ) p p (2x k)2
a) p 8x r2 x2 b) 2 (x k)2 + x2 p
r 2 2
xr 2x2 2kx + k2
p 3x 3
c) 6x2 6
2x + 2

1.15 Calculate r q
pp p4 p p
3 p 3 p
5 p
3 6 12 3 4
a) a b b) a2 a2 c) a6 b8 d) a3 a2 a8 a3
p6 p p p
3 9
a5 a2 a3 a7
e) p3 p : p9
p
6
a2 a4 a7 a

1.16 Calculate q p
p p
a) ld 24 , log 10, ln e3 b) ln( e)3 , ln p 1
3 2 , ln e3(ln e2 +ln e6 )
n+1
p p
e
m
c) log an b 1

1.17 Show that the two sets together with the + and · arithmetic operations fulfill
the body axioms.
p
a) ({a + b 2 with a, b 2 Q},
I +, ·) with the arithmetic operations in IR.
b) (F2 , +, ·) referring to the Example 1.9 4 for the specified link tables.

1.18 Specify the real solutions of the following quadratic equations:

a) 4 x2 + 8 x 60 = 0 b) x2 4 x + 13 = 0 c) 1 = 9 (x 2)2
d) 5 x2 + 20 x + 20 = 0 e) (x 1) (x + 3) = 4

1.19 Determine the parameter c such that the equation 2 x2 + 4 x = c has exactly
one real solution.

1.20 What real-valued solutions does the equations have?

a) 2 x3 + 8 x2 = 8 x
b) t4 13 t2 + 36 = 0
c) 12 3x2 6 x2 25 (x + 3) = 0

1.21 Solve the following root equations:

p p
a) 3 + 2x = 2 b) x2 + 4 = x 2
p p p
c) x 1 = x + 1 d) 2 x2 1 + x = 0

1.22 Which real solutions do the absolute value equations have?

a) x2 x = 24 b) |2 x + 4| = x2 x 6

1.23 Determine the real solution sets of the following inequalities:

a) 2 x 8 > |x| b) x2 + x + 1 0 c) |x|  x 2 d) |x 4| > x2

1.24 Solve the following systems of equations:

a) 4 x1 + 2 x2 + 4 x3 = 10
x1 + x2 + x3 = 3
2 x1 + 3 x2 + 3 x3 = 8

b) 2 x1 + x2 + x3 = 7
2 x1 + 2 x2 + x3 = 10
3 x1 + x3 = 5
40 1. Numbers, Equations and Systems of Equations

c) 2 x1 + x2 + x3 = 7
2 x1 + x2 + x3 = 0
3 x1 + x3 = 5
1.25 Determine the solution set of the following systems:
a) x1 3 x2 + x3 = 3
3 x1 + x2 + x3 = 5

b) x1 + x2 + x3 = 6
x1 + 2 x2 + x3 = 7
2 x1 + x2 + 2 x3 = 11

c) x1 + x2 + x3 = 7
x1 + 2 x2 + x3 = 7
2 x1 + x2 + 2 x3 = 11
1.26 Determine the solution set of the systems of linear equations:
a) 2 x1 + 3 x2 + 4 x3 = 4

b) x1 x2 + x3 = 1
3 x1 + 3 x2 3 x3 = 3
5 x1 5 x2 + 5 x3 = 5
c) x1 x2 + x3 = 1
3 x1 + 3 x2 3 x3 = 1
5 x1 5 x2 + 5 x3 = 5
1.27 Determine the solution of the corresponding homogeneous systems 1.24 to 1.26.
Which general statements can be identified with respect to these homogeneous
systems?

1.28 In the next chemical reactions the variables x1 , x2 , . . . represent natural num-
bers as small as possible to balance the reactions:
a) x 1 F e + x 2 O2 ! x 3 F e 2 O3
b) x1 F eS2 + x2 O2 ! x3 F e3 O3 + x4 SO4
c) x1 C6 H12 O6 + x2 O2 ! x3 CO2 + x4 H2 O
d) x1 C3 H5 N3 O9 ! x2 CO2 + x3 H2 O + x4 N2 + x5 O2
e) x1 N H3 + x2 CuO2 ! x3 N2 + x4 Cu + x5 H2 O
f) x1 Al + x2 H2 SO4 ! x3 Al2 (SO4 )3 + x4 H2
g) x1 Ca3 (P O4 ) + x2 HCl ! x3 Cacl2 + x4 H3 (P O4 )