02-proof

lOMoARcPSD|27395923
02 Proof
Maths extension 2 (Saint Mary & Saint Mina'S Coptic Orthodox College)
Scan to open on Studocu
Studocu is not sponsored or endorsed by any college or university

Downloaded by Eric Gao (442398780@student.sbhs.nsw.edu.au)
lOMoARcPSD|27395923
2
Proof
Chapter Overview: There are many opportunities for presenting or reading

proofs in the Extension 1 and Extension 2 Mathematics courses. In this chapter,
the basic concepts, terminology and notation are discussed, and a few common
methods of proof are presented. Logic plays an important part in proof, but the
emphasis here is on clear argument rather than fancy notation.
Section 2A introduces the necessary language. The remaining sections each focus
on a common type of proof. Section 2B investigates various simple problems in
number theory, that is, the basic structures of the integers. Proof by contradiction
is presented in Section 2C. Algebraic inequalities are proven in Section 2D, whilst
2E introduces harder types of induction. The final section considers inequalities
in calculus, which some may prefer to postpone until after the Integration chapter.
2A The Language of Proof

This long first section introduces some basic concepts, terminology and notation
which are used in the proofs throughout the remainder of the chapter.
Statements: Statements are the basic building blocks of proof. Despite being so
fundamental, it is rather difficult to define what a statement is. Two of the
definitions given in The Macquarie Dictionary are as follows:
1. something stated
2. a communication or declaration in speech or writing setting forth facts,
particulars, etc.
The latter is perhaps more relevant to mathematical proof. Thus a statement
can be a simple sentence, such as:
n is a multiple of 3.
It may also be a mathematical declaration, such as:
let p be a prime number.
A statement may take the form of an assertion or definition, such as:
even numbers are divisible by 2,
or it may be a claim to be proven or a deduction which has already been proven,
such as:
every multiple of 6 is also a multiple of 3.
Although these four examples do not say precisely what a statement is, they will
enable discussions about statements with some agreement about what is meant.
Mathematics Extension 2 Year 12 ISBN 978-1-108-77105-4 © Sadler and Ward 2020 Cambridge University Press
Cambridge Maths Stage 6 Photocopying is restricted under law
and this
Downloaded bymaterial
Eric Gaomust not be transferred to another party
(442398780@student.sbhs.nsw.edu.au)
.
lOMoARcPSD|27395923
2A The Language of Proof 57
Logical Values: Every statement must take one of two logical values: true or false.
Using the first example of a statement above,
n is a multiple of 3
is true if it is known that n = 6, but is clearly false if it is known that n = 7.
1 LOGICAL VALUES: A statement must take one of two logical values: true or false.
Notice in this example that the logical value of a statement may change with
circumstance, but it cannot be simultaneously both true and false. There is no
integer n which is both a multiple of 3 and not a multiple of 3. Likewise a
statement cannot be neither true nor false. Given an integer n, it must always
fall into one of two categories: a multiple of 3, or not a multiple of 3.
A Proven Statement: Whilst some statements may change logical value due to
circumstance, other statements never change their logical value. The statement
34 is a Fibonacci number
is always true, as is determined by simply writing out the first few terms:
1, 1, 2, 3, 5, 8, 13, 21, 34, . . .
A statement which is shown to be true is said to be proven, and the evidence

used to establish the truth is called the proof. Thus the statement “34 is a
Fibonacci number” is proven, and the proof is the listing of the first few terms
of the sequence.
A PROVEN STATEMENT: A statement which is shown to be true is said to be proven,

2
and the evidence used to establish the truth is called the proof.
Examples and Counterexamples: Whilst it is rarely feasible, a statement can be

shown to be true by example, but care must be taken to examine every possible
case. One statement where this is feasible is:
every prime on a regular die is also a Fibonacci number.
The primes on a die are 2, 3 and 5, which are also Fibonacci numbers, so the
statement is proven. Clearly this is a contrived situation, but it serves as a
reminder that proof by exhaustive examples is sometimes possible.
In contrast, the statement

all primes are odd
is clearly false, since 2 is both prime and even. When a statement is shown to
be false by example in this way, that example is called a counterexample, and
the statement is said to be disproved. It is important to note that only one
counterexample is needed in order to disprove a statement.
A COUNTEREXAMPLE: A statement may be disproved (shown to be false) by a single

3
example, called a counterexample.
and this
.
lOMoARcPSD|27395923
58 CHAPTER 2: Proof 2A
Negation: The negation of a statement changes its logical value. In English, negation
is commonly associated with the word not . Thus, whilst the statement
all Fibonacci numbers are odd
is clearly false, its negation
not all Fibonacci numbers are odd
is equally clearly true. In English, this last statement might be considered clumsy,
and so may be replaced with
some Fibonacci numbers are even.
Thus, whilst the presence of the word “not” clearly indicates a negation, its
presence as an indicator cannot always be relied upon.
It is sometimes convenient to associate negation with
complementary sets or events. By way of example: S
_
let x be a Fibonacci number on a die F F
1 3 4
is true when x = 1, 2, 3 or 5 and false when x = 4 or 6. 5
2 6
In contrast:
let x be a non-Fibonacci number on a die
is false when x = 1, 2, 3 or 5 and true when x = 4 or 6. Notice that, whilst the
“not” has been modified into the prefix “non-”, the logical value of the statement
has been changed as expected. All this is clearly evident in the Venn diagram
showing the set of Fibonacci numbers F and its complementary set F .
NEGATION: The negation of a statement changes its logical value. It is commonly

4
associated with the word not, and with complementary sets.
Notation: It is sometimes convenient to present and manipulate statements in an

algebraic manner. Pronumerals are used to represent statements, as in:
suppose that p is the statement ‘x is a multiple of 2.’
The letter p is often used as it is the first letter of the word proposition, a synonym
for a statement. There are two symbols which are commonly used to negate a
statement. These are ¬ and ∼. Thus ¬p = ∼p, which is said “not-p” and means:
x is not a multiple of 2.
And, Or and Negation: In the Advanced and Extension 1 work on probability

it was established that and corresponds to the intersection of sets, whilst or
corresponds to the union. Thus if F is the set of Fibonacci numbers and T is the
set of triangular numbers on a die, then
S
F and T = F ∩ T
F T
= {1, 3} , 2 1
6
whereas F or T = F ∪ T 5 3
4
= {1, 2, 3, 5, 6} .
It makes sense to extend this correspondence to negation and so write:
¬(F and T ) = F ∩ T ,
with ¬(F or T ) = F ∪ T .
and this
.
lOMoARcPSD|27395923
Here is the Venn diagram of F ∩ T . And here is the Venn diagram of F ∩ T .

S S
F T F T
2 1 2 1
6 6
5 3 5 3
4 4
Analysing the latter diagram carefully, it should be clear that the result is
S S S
F T F T F T
2 1 2 1 2 1
6 6 6
5 3 5 3 5 3
4 4 4
F or T or both.
That is, F ∩ T = F ∪ T . Hence by analogy ¬(F and T ) = ¬F or ¬T .
A practical example may help make it clear. Suppose that a school offers two
languages: French and Japanese. A student who does not study both French and
Japanese, either does not study French, or does not study Japanese, or does not
study any language.
Similar analysis gives ¬(F or T ) = ¬F and ¬T , which is left as an exercise.
AND, OR AND NEGATION: The rules are analogous to the complements of intersections
of sets and unions of sets.
5
• The negation of and is or , so that ¬(F and T ) = ¬F or ¬T .
• The negation of or is and , so that ¬(F or T ) = ¬F and ¬T .
Implication: An implication is the relationship between two statements by which one

is a logical consequence of the other. By way of example, if
n is a multiple of 3
then it logically follows that
2n is also a multiple of 3.
In English, this is often written as a single if ... then ... statement. Thus:
if n is a multiple of 3 then 2n is also a multiple of 3.
Other words used in English for such logical deductions include: hence, thus, so,
so that, consequently, and therefore. These words will have been regularly seen
in the proofs of various theorems given in Year 11 mathematics.
Sometimes the order of the two statements is reversed in English, so care must
be taken to determine which statement is the logical conclusion of the other. For
example, the logical order of the deduction
a2 is odd because a is odd
is made far clearer when it is re-written as:
if a is odd then a2 is odd.
It is good practice when writing mathematics to put the conclusion second in this
way. This order should be followed whenever it is practical to do so.
and this
.
lOMoARcPSD|27395923
Two other words commonly associated with implications are each and all . Here
is a similar example written all three ways.
All odd squares are one more than a multiple of 4.
Each odd square is one more than a multiple of 4.
If a number is odd then its square is one more than a multiple of 4.
The careful reader will have also realised from the examples given so far that
implication is commonly associated with subsets. Thus the odd squares form a
subset of those numbers which are 1 more than a multiple of 4.
This is perhaps more easily seen in a simpler example. S
Consider the prime numbers P and Fibonacci numbers F
F
on a die, as shown in the Venn diagram. Quite clearly: 1 3 4
5
all primes on a die are Fibonacci numbers P 2 6
and it is equally clear that P ⊆ F .
IMPLICATION: An implication is the relationship between two statements by which

6 one is a logical consequence of the other. In English, it is often written as an
if ... then ... statement. Implication is also often associated with subsets.
The mathematical notation used for implication is a double tailed arrow ⇒ which
points towards the conclusion. Thus, returning to the example of odd squares:
a is odd ⇒ a2 is 1 more than a multiple of 4.
Whilst this notation is a useful tool, it should be used sparingly.
Quantifiers: The words all and some have been used in various examples above.
These words, and their synonyms, are called quantifiers because they indicate a
quantity. Thus, from above,
all of the primes on a die are Fibonacci numbers,
whereas, in contrast,
some of the primes on a die are odd. S
Whilst all often indicates subsets, the word some is often D P
associated with intersections of sets. This is made clear 3 2 4
1 5
in the Venn diagram of the second example, where D is 6
the set of odds and P is the set of primes.
QUANTIFIERS: The words all and some, and their synonyms, are called quantifiers.
7 • All is associated with subsets.
• Some is associated with intersections.
An important synonym for some is there exists. Thus, continuing with the odds
and primes on a die, it is also true to state that:
there exists a prime on a die that is odd,
there exists a prime on a die that is not odd.
It was mentioned above that, whilst logic notation is a useful tool, it should be
used sparingly. Indeed it is often clearer to write a statement in words than in
symbols. To make this point, some more useful symbols will now be introduced.
and this
.
lOMoARcPSD|27395923
Recall that:
all odd squares are one more than a multiple of 4.
The symbol ∀ means for all . Thus the above implication might be written as:
∀ a odd, a2 is 1 more than a multiple of 4.
Next observe that the value of a can always be written as a = 2m + 1, where
m ∈ Z. That is, m is an integer. So
∀ m ∈ Z, (2m + 1)2 = (multiple of 4) + 1.
Lastly, the multiple of 4 can always be found. For example, 52 = 6×4+1. To put
it another way, there exists an integer k which is the multiple of 4. The symbol
for there exists is ∃. So finally the statement becomes:
∀ m ∈ Z, ∃ k ∈ Z such that (2m + 1)2 = 4k + 1.
Compare this last jumble of symbols with the clear prose given in the statement
at the top of the page. The symbols have got in the way and hindered an easy
reading of the implication. Here is another pair of statements to show that the
use of symbols can be a distraction.
Not all primes are odd.
∃ x ∈ {primes} such that x 6= 2m + 1, ∀ m ∈ Z.
Of course, these symbols will be needed in certain problems, and students will
be expected to understand them when they are used in questions. Nevertheless,
symbols used in a proof or logical argument in this course should be the exception
rather than the rule.
Sufficient and Necessary: Two other terms are strongly associated with implication
and are often used in mathematical proofs or discussions about proofs. It is easiest
to explain these terms when the statement is written in if ... then ... form. So,
here is the statement about primes on a die written in that way.
If a number on a die is prime then it is a Fibonacci number.
The first part of an if ... then ... statement is called a sufficient condition. That
is, it is sufficient to know that a number on a die is a prime to guarantee that it
is a Fibonacci number. The second part of an if ... then ... statement is called
a necessary condition. That is, it is necessary to know that a number on a die is
a Fibonacci number in order for it to be prime. However, it does not guarantee
that it is a prime. It could be the number 1.
The situation is once again made clear by the fact that on the die, the primes
form a subset of the Fibonacci numbers. The subset corresponds to the sufficient
condition, and the superset corresponds to the necessary condition. A practical
example should help make it clear.
If I travel by bus then I use public transport.
It is sufficient that I travel by bus in order to use public transport, but it is
necessary that I use public transport in order to travel by bus.
SUFFICIENT AND NECESSARY: In the statement if A then B,

8 • A is a sufficient condition for B,
• B is a necessary condition for A.
and this
.
lOMoARcPSD|27395923
Converse Statements: Consider the following two statements.

If x is a multiple of 4 then x is even.
If x is even then x is a multiple of 4.
When the two parts of an if ... then ... statement are swapped in this way,
the result is called the converse. In this case, clearly the first statement is true,
and the converse is false. This is clear because the multiples of 4 form a strict
subset of the evens, so the converse cannot possibly be true. In other words, the
converse of a statement is not the same as the original statement.
In contrast, here are two converse statements from Year 9 geometry.
If a quadrilateral is a rhombus then it has four equal sides.
If a quadrilateral has four equal sides then it is a rhombus.
In this case both statements are true, but it is important to realise that the
converse statement is still not the same as the original. In the first statement, it
is given that the quadrilateral is a rhombus. This is called the premise, and the
conclusion is that it has four equal sides. The second statement differs because
the premise is now that the quadrilateral has four equal sides, and the conclusion
is that it is a rhombus.
CONVERSE STATEMENTS:
9 The converse of the statement if A then B is the statement if B then A.
A statement and its converse may have different logical values.
Equivalent Statements: Two statements are called equivalent if each is a logical

consequence of the other. When solving an equation, the separate steps were
called equivalent equations, which are examples of equivalent statements. Thus
2x + 3 = 11 and 2x = 8
are equivalent. The latter can be obtained by subtracting 3 from both sides of
the first. The former can be obtained by adding 3 to both sides of the second.
So each is a logical consequence of the other. This symmetry means that the two
statements can be written as an if ... then ... statement and its converse, both
of which are true.
If 2x + 3 = 11 then 2x = 8.
If 2x = 8 then 2x + 3 = 11.
EQUIVALENT STATEMENTS: Two statements are equivalent if each is a consequence
10 of the other. This symmetry means that the two statements can be written as
an if ... then ... statement and its converse, both with the same logical value.
Clearly equivalent statements are important and so they have special terminology
and notation. The two implications are often abbreviated into one statement
using the words if and only if . Thus, from the geometry example above:
a quadrilateral is a rhombus if and only if it has four equal sides.
When written, the words if and only if may be abbreviated to iff, or the symbol ⇔
may be used. Here is the same statement written in those two ways.
A quadrilateral is a rhombus iff it has four equal sides.
A quadrilateral is a rhombus ⇔ a quadrilateral has four equal sides.
and this
.
lOMoARcPSD|27395923
Whichever of the new terminology or symbol is used, it is important to remember

that the statement is always an abbreviation of two implications. Thus
n is divisible by 3 iff the sum of its digits is divisible by 3
is a mathematical abbreviation for the two implications
if n is divisible by 3 then the sum of its digits is divisible by 3,
if the sum of the digits of n is divisible by 3 then n is divisible by 3.
Since equivalent statements can be written as a pair of S
implication statements, it follows that each corresponds T E
1 3
to a subset of the other. Consider the the powers of 2 on 2 5
a die, which are 1, 2 and 4, denoted by set T . These are 4 6
also the proper factors of 8, denoted by set E. Thus:
x is a power of 2 on a die iff it is a proper factor of 8,
x ∈ T ⇔ x ∈ E.
Since each set is a subset of the other, it follows that both sets are equal. That
is T = E, as shown in the Venn diagram above.
EQUIVALENT STATEMENTS AND SETS: The sets corresponding to equivalent statements

11
are equal.
Recall that an implication is associated with the words sufficient and necessary.
Writing the last example of equivalence as an implication and its converse:
if x is a power of 2 on a die then it is a proper factor of 8,
if x is a proper factor of 8 on a die then it is a power of 2.
From the first of these, it is sufficient that x is a power of 2 on a die for it to be
a proper factor of 8. From the second, it is necessary that x is a power of 2 on a
die for it to be a proper factor of 8. In other words, x is a power of 2 on a die
is both a necessary and sufficient condition for it to be a proper factor of 8. A
similar situation will hold for any pair of equivalent statements.
NECESSARY AND SUFFICIENT: When two statements are equivalent, each is both a
12
necessary and sufficient condition for the other to be true.
The Contrapositive: There are many examples of pairs of equivalent statements,

but one particular situation plays an important part in many proofs. By way of
example, recall that
if x is a prime on a die then it is a Fibonacci number.
This is logically equivalent to the contrapositive statement
if x is not a Fibonacci number on a die then it is not a prime.
Notice that the two parts of the if ... then ... statement have been negated and
the order swapped. This is always the case with the contrapositive statement. It
is also always the case that an implication and its contrapositive are equivalent.
A practical example may help cement this new concept. Thus
if I own a dog then I have a pet
is logically equivalent to its contrapositive
if I do not have a pet then I do not own a dog.
and this
.
lOMoARcPSD|27395923
This means that an implication and its contrapositive can be written in the form
of an if and only if statement:
A implies B if and only if not B implies not A,
or, writing this entirely in symbols:
(A ⇒ B) ⇔ (∼B ⇒ ∼A).
THE CONTRAPOSITIVE: An implication, if A then B, is logically equivalent to its
13
contrapositive, if not B then not A.
The situation is clear from the Venn diagram of the primes and Fibonacci numbers
on a die. In that case P ⊆ F , and consequently F ⊆ P . That is, if x is not a
Fibonacci number on a die then it is not prime. Here are Venn diagrams with F
and P shaded to demonstrate that F ⊆ P .
S S
F F
1 3 4 1 3 4
5 5
P 2 6 P 2 6
It is further confirmed that F ⊆ P by listing the elements of each set. Thus

F = {4, 6} whilst P = {1, 4, 6}.
Exercise 2A
1. Identify each of these symbols.
(a) = (b) ⇒ (c) ⇔ (d) ∀ (e) ∃
2. Write down the converse of each statement, and state whether the converse is true or false.
(a) If a triangle has two equal sides, then it has two equal angles.
(b) If a number is odd, then its square is odd.
(c) If I am a horse, then I have four legs.
(d) If a number ends with the digit 6, then it is even.
(e) Every square is a rhombus.
√
(f) If n ∈ R, then n ≥ 0.
3. Indicate whether each statement is true or false.
(a) Having four legs is a necessary condition for being a cat.
(b) Having four legs is a sufficient condition for being a cat.
(c) Owning a car is a necessary condition for holding a driver’s licence.
(d) Owning a car is a sufficient condition for holding a driver’s licence.
(e) If two statements are equivalent, then each is a necessary condition for the other to
be true.
(f) If two statements are equivalent, then each is a sufficient condition for the other to be
true.
4. Write down the negation of each statement.
(a) All cars are red. (e) If I live in Tasmania, then I live in Australia.
(b) a>b (f) If Nikhil doesn’t study, then he will fail.
(c) Hillary likes steak and pizza. (g) −3 ≤ x ≤ 8
(d) Bill is correct or Dave is correct. (h) x < −5 or x ≥ 0
and this
.
lOMoARcPSD|27395923
5. Write down the contrapositive of each statement.

(a) If I water my plants, then they will grow.
(b) If you do not live in Australia, then you do not live in Melbourne.
(c) If a triangle has three equal sides, then it has three equal angles.
(d) If I like cycling, then I do not like motorists.
(e) If a number is even, then the next number is odd.
(f) If a and b are both positive and a > b, then a1 < 1b .
6. Write down (in ‘if. . . then’ form) the two converse statements equivalent to each ‘if and
only if’ statement.
(a) A number is divisible by 15 if and only if it is divisible by both 3 and 5.
(b) A triangle has two equal angles if and only if it has two equal sides.
(c) An integer n greater than one is prime if and only if its only divisors are 1 and n.
(d) A quadrilateral is a parallelogram if and only if a pair of opposite sides are equal and
parallel.
DEVELOPMENT
7. State whether each statement is true ∀ x ∈ R. If it is false, provide a counterexample.

(a) x − 3 < x (c) 10x > 0 (e) | − x| = x
√
(b) 3x ≥ x (d) x ≤ x2 (f) |x| = x2
8. State whether each statement is true ∀ a, b ∈ R. If it is false, provide a counterexample.

(a) If a > b, then a2 > b2 (d) If a, b < 0 and a < b, then 1
a > 1b .
(b) If a2 > b2 , then a > b (e) |a + b| ≥ |a| + |b|
(c) If a > b, then a3 > b3 (f) |a − b| ≥ |a| − |b|
9. Assuming that all variables are real variables, insert the correct symbol ⇒ or ⇔ in each
statement below.
(a) it is raining . . . there are clouds in the sky (d) x = 5 . . . x2 = 25
(b) 3a = 6 . . . 5a = 10 (e) x = 5 . . . x3 = 125
(c) a > b . . . −b > −a (f) a is an integer . . . a2 is an integer
10. Answer true or false. Assume that a, b, c 6= 0.
(a) θ = π6 ⇔ sin θ = 21
(b) sin θ = − √12 ⇔ tan θ = ±1
(c) x2 + y 2 < 1 ⇔(x, y) is a point inside the circle x2 + y 2 = 1
(d) α and β are the roots of ax2 + bx + c = 0 ⇔ α1 and β1 are the roots of cx2 + bx + a = 0.
11. Suppose that p is the statement ‘Jack does Extension 2 Mathematics’ and q is the state-
ment ‘Jack is crazy’. Write each of the following as English sentences.
(a) p ⇒ q (c) ∼ p ⇒ q (e) ∼ p ⇒ ∼ q
(b) ∼ (p ⇒ q) (d) ∼ q ⇒ p (f) p 6⇒ q
12. Write each statement as an English sentence, without any use of symbols.
1
(a) ∀ n ∈ Z ∃ m ∈ Z such that m > n (b) a ∈ R and a > 0 ⇒ a + ≥2
a
and this
.
lOMoARcPSD|27395923
66 CHAPTER 2: Proof 2B
13. An implication is false if the premise is true but the conclusion is false. Otherwise it is
true. State whether each statement is true or false.
(a) 1 + 1 = 3 ⇒ 2 + 2 = 4 (c) 1 + 1 = 3 ⇒ 2 + 2 = 3
(b) 1 + 1 = 2 ⇒ 2 + 2 = 3 (d) 1 + 1 = 2 ⇒ 2 + 2 = 4
14. (a) Explain why “1 < 0” and “1 is a negative number” are equivalent statements.
(b) Combine them into an if ... then ... statement. Is the result true?
15. Consider the following statements:
(1) If I do not do my homework then I will fail.
(2) If I study hard then I will pass.
If statements (1) and (2) are both true and I passed, then:
(a) Did I do my homework? (b) Did I study hard?
ENRICHMENT
16. Consider this statement:

‘If either Anna or Bryan passed the exam, then either Anna and Chris both passed or
Bryan and Chris both passed.’
If the statement is false, determine whether Chris passed or failed.
17. On a train, Pender, Sadler and Ward are the the fireman, guard and driver, but NOT
respectively. Also aboard the train are three passengers who have the same names: Dr Pen-
der, Mr Sadler and Mr Ward.
1. Mr Sadler lives in Sydney.
2. The guard lives exactly half way between Melbourne and Sydney.
3. Mr Ward earns exactly $100 000 per year.
4. The guard’s nearest neighbour, one of the passengers, earns exactly three times as much
as the guard.
5. Pender beats the fireman at pool.
6. The passenger whose name is the same as the guard’s lives in Melbourne.
Who is the driver? Clearly explain your reasoning.
2B Number Proofs
Now that the basic concepts, terminology and notation of proof are understood,
they can be put together to write proofs of simple results. The geometry studied
in Years 7 to 10 involved many such proofs, and it would be worth reviewing some
of those in light of this new understanding. In this section and in 2C, however,
attention is focused on some basic results in number theory, that is, the study of
the structures and number patterns in:
Z the integers . . . − 3, −2, −1, 0, 1, 2, 3, . . .
Z+ the positive integers 1, 2, 3, . . .
N the natural numbers 0, 1, 2, 3, . . .
A proper investigation of number theory would occupy an entire book, so only a
few simple results in divisibility are proven here.
and this
.
lOMoARcPSD|27395923
2B Number Proofs 67
Divisibility: It is important to begin with a clear definition of what divisibility means.

Let a, b, m ∈ Z and suppose that b = am. The numbers a and m are called factors
of b. Furthermore b is said to be divisible by a. (Equally, b is divisible by m.)
To put it another way, if b and a are integers and b is divisible by a then there
exists an integer m such that b = am.
14 DIVISIBILITY: If a, b ∈ Z and b is divisible by a then ∃ m ∈ Z such that b = am.
This definition will be essential in the following worked examples, and in the
exercise questions. Although the wording of the definition may seem obscure, in
practice it is quite obvious. For example, 12 is divisible by 4 because 12 = 4 × 3.
WORKED EXAMPLE 1: Let a and b be two integers divisible by 3.

(a) Prove that (a + b) is divisible by 3.
(b) Prove that (ax + by) is divisible by 3, for all x, y ∈ Z.
SOLUTION:
(a) By the definition of divisibility, there exist m, n ∈ Z such that
a = 3m
and b = 3n
Hence (a + b) = (3m + 3n)
= 3(m + n)
That is, (a + b) is divisible by 3.
(b) Likewise, for all x, y ∈ Z,
(ax + by) = (3mx + 3ny)
= 3(mx + ny)
That is, (ax + by) is divisible by 3.
WORKED EXAMPLE 2: Prove that a2 − a is even for all a ∈ Z.

SOLUTION: Factoring, a2 − a = a(a − 1).
Now, if a is odd, then (a − 1) is even,
and if a is even, then (a − 1) is odd.
In either case, the product of an odd and even is even.
Hence a2 − a is even.
Note: this is essentially proof by example, with every case examined.
WORKED EXAMPLE 3: A student claims that if an integer n is divisible by both 4

and 6 then the number is divisible by 4 × 6 = 24.
(a) Disprove this claim by finding a counterexample.
(b) Explain what has gone wrong, and determine the correct conclusion.
SOLUTION:
(a) Clearly 12 = 4 × 3 and 12 = 6 × 2,
hence 12 is divisible by both 4 and 6,
but it is not divisible by 24.
(b) The problem is that 4 and 6 have common factor 2.
Since n is divisible by 4 and 6, there exist integers k and ` such that
and this
.
lOMoARcPSD|27395923
68 CHAPTER 2: Proof 2B
n = 4k
and n = 6`
so 4k = 6`
or 2k = 3` .
Since 2 and 3 are primes, k is divisible by 3 (and ` is even),
so there exists an integer m such that k = 3m.
Hence n = 4 × 3m
= 12m
That is, if n is divisible by both 4 and 6 then it is divisible by 12.
WORKED EXAMPLE 4: Let n = 10x+y, where n, x, y ∈ Z+ , the positive integers.

(a) Prove that if n is divisible by 7 then (x − 2y) is also divisible by 7.
(b) Further, prove that the converse is true.
(c) Write the result as an iff statement.
(d) Hence determine whether or not 3871 is divisible by 7.
SOLUTION:
(a) Since n is divisible by 7, ∃ m ∈ Z+ such that
10x + y = 7m
The key to the proof is that y has been doubled:
so 20x + 2y = 7 × (2m)
thus −x + 2y = 7(2m − 3x) (subtracting 21x from both sides)
or x − 2y = 7(3x − 2m)
Hence (x − 2y) is divisible by 7.
(b) Since the equations in part (a) are equivalent, it follows that
h i h i
10x + y = 7m ⇔ x − 2y = 7(3x − 2m) .
Hence if (x − 2y) is divisible by 7 then n = 10x + y is also divisible by 7.

(c) That is, combining parts (a) and (b),
n = 10x + y is divisible by 7 if and only if (x − 2y) is divisible by 7.
(d) Since the result is an equivalence relation, the divisibility test can be applied
recursively. Thus:
3871 is divisible by 7 iff 387 − 2 × 1 = 385 is divisible by 7.
385 is divisible by 7 iff 38 − 2 × 5 = 28 is divisible by 7.
Now 28 = 7 × 4 is clearly divisible by 7.
Hence 3871 is divisible by 7.
Exercise 2B
Note: In this exercise you may assume that all pronumerals represent integers.
1. (a) If a and b are even, prove that a + b is even.

(b) If a and b are odd, prove that a + b is even.
(c) If a is even and b is odd, prove that a + b is odd.
and this
.
lOMoARcPSD|27395923
2B Number Proofs 69
2. (a) If a and b are even, prove that ab is even.

(b) If a and b are odd, prove that ab is odd.
(c) If a is even and b is odd, prove that ab is even.
3. Prove that:
(a) the square of an even number is even,
(b) the square of an odd number is odd.
4. Prove that:
(a) if b and b + c are both divisible by a, then c is divisible by a,
(b) if b and b − c are both divisible by a, then c is divisible by a.
5. If b and c are both divisible by a, prove that bx + cy is divisible by a for any x, y ∈ Z.
6. If a, b, c and d are consecutive integers, prove that:
(a) a + d = b + c (b) ad = bc − 2 (c) a2 + d2 = b2 + c2 + 4
DEVELOPMENT
7. Prove that if a − b is even, then a − b2 is divisible by 4.

2
8. Suppose that 2a + b and 3a + 2b are both divisible by n. Prove that a and b are both
divisible by n.
9. Suppose that a2 + a and a2 − a are both divisible by 4. Prove that a is even.
10. Prove that a3 − a is divisible by 6 ∀ a ∈ Z.
11. If a is even, prove that a3 + 2a2 is divisible by 8.
12. Prove that a number is divisible by 6 if and only if it is divisible by both 2 and 3.
(Remember that to prove A ⇔ B, you must prove B ⇒ A and A ⇒ B.)
13. Prove that an integer is the sum of 7 consecutive integers if and only if it is divisible by 7.
14. (a) If n is odd, prove that the sum of n consecutive numbers is divisible by n.
(b) If n is even, is the sum of n consecutive numbers divisible by n? Explain your answer.
15. Prove that a 4-digit number is divisible by 3 if and only if the sum of its digits is divisible
by 3.
16. Let n = 10x + y, where n, x, y ∈ Z+ .
(a) Prove that if n is divisible by 13, then x + 4y is also divisible by 13.
(b) Prove that the converse of part (a) is true.
(c) Combine (a) and (b) into an ‘if and only if’ statement using mathematical symbols.
(d) Use part (a) recursively to show that 8112 is divisible by 13.
17. (a) Show that xn − 1 = (x − 1)(xn−1 + xn−2 + xn−3 + · · · + x2 + x + 1).
(b) Hence prove that:
(i) 7n − 1 is divisible by 6 ∀ n ∈ Z+,
(ii) if an − 1 is prime, then a = 2 for a > 0.
18. Suppose that n = pa q b , where p and q are primes and p 6= q.
(a) Use combinatorics (a counting argument) to explain why n has (a + 1)(b + 1) factors.
(b) Hence determine the number of factors of 80 000.
19. Prove the statement: a − c is a divisor of ab + cd ⇒ a − c is a divisor of ad + bc.
and this
.
lOMoARcPSD|27395923
70 CHAPTER 2: Proof 2C
ENRICHMENT
20. (a) Factorise a + 4b by adding and subtracting 4a2 b2.

4 4
(b) Hence prove that 5454 + 4545 is not a prime number.

21. (a) Prove that the square of an even number is divisible by 4.
(b) Prove that the remainder is 1 when the square of an odd number is divided by 8.
(c) Hence prove that if a and b are both odd, then a2 + b2 is not a square.
22. Suppose that p is a prime number greater than 30. Prove that when p is divided by 30,
the remainder is either 1 or prime.
23. Numbers such as 6 and 28 are known as perfect numbers because they are equal to the
sum of their factors, excluding the number itself.
(a) Confirm that 6 and 28 are perfect numbers.
(b) Prove that if 2n − 1 is prime, then 2n−1 (2n − 1) is a perfect number.
[Hint: Separate the factors of 2n−1 (2n − 1) into two groups: those that are strictly
powers of 2 and those that have the factor 2n − 1.]
24. Suppose that I choose six of the first ten positive integers. Prove that I must have chosen
two numbers such that one is a divisor of the other. [Hint: Write each of the 10 numbers
as a power of 2 multiplied by an odd number, then use the pigeonhole principle.]
2C Proof by Contraposition and by Contradiction

Two methods of proof commonly used in mathematics involve negation. These are
called proof by contraposition and proof by contradiction. The latter is also given
the Latin name reductio ad absurdum, which literally means reduce to absurdity .
The reason for this will be explained later.
Proof by Contraposition: This style of proof takes advantage of the fact that an
implication is equivalent to its contrapositive. Thus when an implication is not
easy to prove directly, it may be suitable to use proof by contraposition instead.
It is important to clearly state what is being done at the outset.
WORKED EXAMPLE 5: Prove that if n2 is even then n is even.

SOLUTION: The contrapositive is proven instead. That is:
if n is not even then n2 is not even,
or, more naturally:
if n is odd then n2 is odd.
Let n = 2m + 1, then n is odd and
n2 = (2m + 1)2
= 4m2 + 4m + 1
= 2(2m2 + 2m) + 1
which is one more than a multiple of 2, and is thus odd.
Hence, by the contrapositive, if n2 is even then n is even.
PROOF BY CONTRAPOSITION: An implication and its contrapositive are equivalent.

15
Hence an implication may be proven by proving its contrapositive instead.
and this
.
lOMoARcPSD|27395923
2C Proof by Contraposition and by Contradiction 71
Proof by Contradiction: This style of proof is based on the fact that a statement
can only have one of two values, true or false, and the negated statement must
have the opposite value. That is, if a statement is true then its negation must
be false. Hence proving a statement is equivalent to showing that its negation is
false. Thus when a statement is not easy to prove directly, it may be suitable to
show instead that the negated statement is false.
The first step is to write down the negation as an assumption. It is then shown
that this leads to an absurd statement, like −1 > 2, or it leads to a contradiction
of the initial assumption. This is why the method is also called reductio ad
absurdum . Since the negation is equivalent to a false statement, the negation is
also false. It must therefore hold that the proposition is true.
√
WORKED EXAMPLE 6: Prove that 2 is irrational.
√
SOLUTION: By way of contradiction, assume that 2 is rational.
That is, assume there exist m, n ∈ N (natural numbers) such that
√ m
2=
n
where n ≥ 1 and the HCF (highest common factor) of m and n is 1.
That is, the fraction has been reduced to lowest terms.
Squaring and re-arranging gives
2n2 = m2 .
Thus m2 is divisible by 2.
Now if m were not divisible by 2 then m2 would not be divisible by 2.
Hence m is also divisible by 2. So let m = 2p and write
2n2 = 4p2
or n2 = 2p2 .
Thus n2 is divisible by 2.
Now if n were not divisible by 2 then n2 would not be divisible by 2.
Hence n is also divisible by 2.
That is, 2 is a common factor of m and n.
But the HCF is 1, so there is a contradiction.
√
Hence 2 is irrational.
PROOF BY CONTRADICTION: Proving a statement is equivalent to showing that its

16 negation is false. Begin by writing down the negation and show this leads to
a contradiction.
Notice the careful use of two contrapositive statements in the above proof. Each
√
is a restatement of what was proven in Worked Example 5. The proof that 3
is irrational requires similar steps. That is:
if m were not divisible by 3 then m2 would not be divisible by 3.
A problem arises
√ at this point because that claim has not been proven. Likewise,
the proof that 7 is irrational leads to the claim that
if m were not divisible by 7 then m2 would not be divisible by 7
which also has not been proven. In this course, all such claims will be taken to
be intuitively obvious unless a proof is specifically required by the question.
and this
.
lOMoARcPSD|27395923
72 CHAPTER 2: Proof 2C
The next worked example follows a slightly different argument to show that log2 3
is irrational. This result is extremely important in music. Musicians will know
that when a fifth is played on a piano, say from A to E, it is not a pure fifth but
slightly short. A pure fifth has frequencies in the ratio 2 : 3. Thus using A440
should give E660, but on a piano it is approximately E659. This is done so that
the cycle of fifths works, and it comes about because log2 3 is irrational.
WORKED EXAMPLE 7: Prove that log2 3 is irrational.

SOLUTION: By way of contradiction, assume that log2 3 is rational.
That is, assume there exist m, n ∈ N such that
m
log2 3 =
n
where n ≥ 1 and the HCF of m and n is 1.
That is, the fraction has been reduced to lowest terms.
From the definition of logs, this can be re-written as
m
3 = 2n .
Take the nth power of both sides to get
3n = 2m
Now since n ≥ 1 it follows that 3 is a factor of the LHS.
But clearly 3 is not a factor of the RHS, so there is a contradiction.
Hence log2 3 is irrational.
The Fundamental Theorem of Arithmetic: In the last proof it was assumed that
a positive power of 3 cannot equal a positive power of 2. Intuition and experience
certainly seem to suggest this is true, but it has not been proven. The result is
a specific case of a more general theorem. The theorem effectively states that
there is only one way to write a number in prime factored form. This is such
an important result in number theory that it is given the name the fundamental
theorem of arithmetic . Although any proof of this result is beyond the scope of
this course, it is such an important theorem that a proof has been included in
the appendix to this chapter.
Exercise 2C
1. Prove by contradiction that log7 13 is irrational.
Start by assuming that log7 13 is rational, so log7 13 = mn
, where m, n ∈ Z and m, n have
no common factors other than 1.
√
2. Prove by contradiction that 5 is irrational.
√ √
Start by assuming that 5 is rational, so 5 = m n
, where m, n ∈ Z and m, n have no
common factors other than 1.
3. Consider this statement for a ∈ N: ‘If a2 is odd then a is odd.’
(a) Write down the contrapositive of the statement.
(b) Prove the statement by proving its contrapositive.
4. By proving the contrapositive, prove that if m2 + 4m + 7 is even, then m is odd.
5. Suppose that a, b ∈ Z+ . By proving the contrapositive, prove that if ab is even, then a is
even or b is even.
and this
.
lOMoARcPSD|27395923
2C Proof by Contraposition and by Contradiction 73
DEVELOPMENT
6. (a) Prove that log3 5 is irrational.

(b) Hence prove that log3 15 is irrational.
√
7. (a) Prove that 11 is irrational.
√
(b) Hence prove that 44 is irrational.
8. Suppose that the number n is composite and has two distinct prime√factors p1 and p2 .
Use contradiction to prove that at least one of p1 and p2 is less than n.
9. Suppose that ∃ n ∈ N such that n2 + 2 is divisible by 4.
(a) Deduce that n is even.
(b) Hence prove by contradiction that no such n exists.
10. (a) Explain why every odd number is one more or one less than a multiple of 4.
(b) Prove that the product of any two positive integers of the form 4n + 1, where n is a
positive integer, is also of the form 4n + 1.
(c) Hence prove by contradiction that any composite number of the form 4n − 1 must
have at least one prime factor of the form 4n − 1.
√
11. Prove by contradiction that if n ∈ Z+ , then 4n − 2 is irrational.
√
12. Prove by contradiction that 3 + 1 is irrational.
√
13. (a) Prove that 6 is irrational.
√ √
(b) Hence prove that 3 + 2 is irrational.
14. Prove that if 2n − 1 is prime then n is prime by proving the contrapositive.
15. Prove by contradiction that there are infinitely many prime numbers.
[Hint: Assume that p is the largest prime number and consider the number p! + 1, which
is not divisible by any number from 2 to p.]
ENRICHMENT
16. Consider the following theorem about prime numbers.

If ∀ a, b ∈ Z+ , p|ab ⇒ p|a or p|b then p is prime.

(a) State the contrapositive of the theorem.

(b) Prove the theorem by proving the contrapositive.
p
[Hint: Put a = p1 , where p1 is a prime divisor of p, and b = p1
.]
17. Suppose that a, b, c and d are positive integers and c is not a square.
a d
(a) Prove that if √ + √ is rational, then b2 d = c(a + d).
b+ c c
a d
(b) Hence prove by contradiction that √ + √ is irrational.
1+ c c
18. Suppose that p is a prime number greater than 3 and that for some n ∈ N, pn is a 20-digit
number. Prove that among these 20 digits, there are at least three that are equal.
[Hint: Use proof by contradiction and the test for divisibility by 3.]
19. Prove by contradiction that ∀ a, b ∈ Z+ , (36a + b)(36b + a) is not a power of 2.
[Hint: Assume that 2k is the smallest power of 2 equal to (36a + b)(36b + a).]
and this
.
lOMoARcPSD|27395923
74 CHAPTER 2: Proof 2D
2D Algebraic Inequalities
Two fundamental assumptions about inequalities will be used throughout this
course. They should be intuitively obvious from previous work on equations and
inequations. Those assumptions are, for real numbers a, b and c :
if a > b then a + c > b + c ,
if a > b and c > 0 then ac > bc ,
with similar results for a < b. The first statement means that an inequality is
unchanged when the same amount is added to both sides. In the second case
the inequality is unchanged when both sides are multiplied by the same positive
amount. A particular case of the first statement is when c = −b, which gives
if a > b then a − b > 0 .
That is, if a > b then (a − b) is positive. The converse is also assumed true.
There is an amazing number of types of algebraic inequalities, but most can be
assigned to one of four broad categories for proving. Those are:
• put everything on one side,
• squares of reals cannot be negative,
• combinations of inequalities,
• begin with a known result.
The first category takes advantage of the last statement above.
Put Everything on One Side: In a few instances the inequality can easily be proved
by moving all terms to one side and considering the sign of that expression.
WORKED EXAMPLE 8: Let a < b be real numbers. Prove that the average of the
squares of a and b is greater than the square of the average.
SOLUTION: The corresponding inequality to prove is
2
a2 + b2

a+b
> .
2 2
a2 + b2 a2 + 2ab + b2
Now LHS − RHS = −
2 4
a2 − 2ab + b2
=
4
(a − b)2
=
4
>0 (since squares cannot be negative.)
2 2
2
a +b a+b
Hence − >0
2 2
2
a2 + b2

a+b
and so > .
2 2
Squares Cannot be Negative: The crucial step in the previous worked example was
that the square of a real number cannot be negative. That fact can be deduced
from the assumptions at the start of this section, and the proof is left as an
exercise. This important result can be used to solve numerous other inequalities.
and this
.
lOMoARcPSD|27395923
2D Algebraic Inequalities 75
Arithmetic and Geometric Means: Recall from sequences and series that if a
and b are positive real numbers then the sequence a, x, b will be arithmetic if
a+b
x= .
2
The value of x is called the arithmetic mean of a and b.
Likewise, the sequence a, y, b will be geometric if
√
y = ab .
The value of y is called the geometric mean of a and b.
ARITHMETIC AND GEOMETRIC MEANS: Suppose that a and b are positive.

a+b
17 • The arithmetic mean of a and b is .
√2
• The geometric mean of a and b is ab .
The AM/GM Inequality: One very important result is that the arithmetic mean
is at least as large as the geometric mean. This is sometimes called the AM/GM
inequality. It is proved by noting that squares of real numbers cannot be negative.
WORKED EXAMPLE 9: Prove the AM/GM inequality.

SOLUTION: Let a and b be two positive real numbers.
Since squares cannot be negative,
√ √ 2
a− b ≥ 0.
√
Expanding a − 2 ab + b ≥ 0
√
so a + b ≥ 2 ab
a+b √
or ≥ ab .
2
THE AM/GM INEQUALITY: The arithmetic mean and geometric mean of two positive
numbers a and b are related by the inequality
18
a+b √
≥ ab .
2
The AM/GM inequality can also be proven using circle geometry and that is
done in a worked example in Section 2E.
Combinations of Inequalities: Having established an inequality, it can be restated

using other variables and the results combined to form a new inequality.
WORKED EXAMPLE 10:

(a) Suppose that 1 ≤ k ≤ n.
(i) Show that n ≤ k(n − k + 1).
√ p n+1
(ii) Explain why n ≤ k(n − k + 1) ≤ .
2
(b) Hence prove that, for all positive integers n,
n
√

n+1
nn ≤ n! ≤ .
2
and this
.
lOMoARcPSD|27395923
SOLUTION:
(a) (i) The right hand side is a quadratic in k with negative leading coefficient:
k(n + 1) − k2 .
Hence its graph is concave down, and any minimum will occur at an end
point of the domain. Direct substitution of the two end points gives
1 × n = n × 1 = n.
(ii) The left hand inequality follows directly from part (i).
By the AM/GM inequality,
p k + (n − k + 1)
k(n − k + 1) ≤
2
n+1
≤ .
2
(b) By part (a) it follows that
√ √ n+1
n ≤ 1×n ≤ (k = 1)
2
√ p n+1
n ≤ 2(n − 1) ≤ (k = 2)
2
√ p n+1
n ≤ 3(n − 2) ≤ (k = 3)
2
..
.
√ √ n+1
n ≤ n×1≤ (k = n)
2
Now multiply all these results together to get
√ n
r n + 1 n
n ≤ 1 × 2 × . . . × n × n × (n − 1) × . . . × 1 ≤
2
n
√ √

n + 1
hence nn ≤ n! × n! ≤
2
n
√

n + 1
or nn ≤ n! ≤
2
This example happened to use multiplication of inequalities. Other examples

may require addition, subtraction or division. Multiplication and division should
be avoided unless it is guaranteed that the quantities involved do not change sign.
Otherwise it is not known whether the direction of the inequality is affected.
Begin with a Known Result: Many problems begin with a known result from which
another inequality is to be obtained. An important example of this is the relation
|x| + |y| ≥ |x + y| ≥ |x| − |y|
which is known as the triangle inequality. This was proven in the chapter on
complex numbers by using geometry. In the following Worked Example the left
hand inequality is proven algebraically.
and this
.
lOMoARcPSD|27395923
WORKED EXAMPLE 11: Use the fact that |a| ≥ a to prove that |x| + |y| ≥ |x + y|.
SOLUTION: Square the left hand side to get

(|x| + |y|)2 = x2 + 2|x||y| + y 2
≥ x2 + 2xy + y 2 (since |a| ≥ a)
so (|x| + |y|)2 ≥ (x + y)2 .
Now take the square root of both sides to get
√
|x| + |y| ≥ |x + y| (since a2 = |a| on the RHS.)
The proof of the other part of the triangle inequality is left as an exercise.
In other problems restrictions are placed on the variables, such as a + b + c = 1,

from which it follows that ab + ac + bc < 31 . The usual approach is to begin
manipulating the expression and at crucial steps apply the given restrictions.
In some instances the solution also involves changing the value of a fraction by
altering the numerator or denominator. For example, decreasing the numerator
or increasing the denominator will reduce the value of the fraction. This is clearly
evident in the following numerical example.
3 2 2
> >
5 5 7
The strategy is used twice in the next worked example, along with a restriction.
WORKED EXAMPLE 12: Let a, b and c be three positive real numbers. It is known
that a + b ≥ c.
a+b c
(a) Show that ≥ .
1+a+b 1+c
a b c
(b) Hence show that + − ≥ 0.
1+a 1+b 1+c
SOLUTION:
(a + b)(1 + c) − c(1 + a + b)
(a) LHS − RHS =
(1 + a + b)(1 + c)
(a + b) − c
=
(1 + a + b)(1 + c)
c−c
≥ (since a + b ≥ c)
(1 + a + b)(1 + c)
Thus LHS − RHS ≥ 0 , and hence LHS ≥ RHS .
a b c a b a+b
(b) + − ≥ + − (by part (a))
1+a 1+b 1+c 1+a 1+b 1+a+b
a b a b
= + − −
1+a 1+b 1+a+b 1+a+b
a b a b
≥ + − − ,
1+a 1+b 1+a 1+b
since decreasing the denominators increases the subtracted fractions.
a b c
Hence + − ≥ 0.
1+a 1+b 1+c
and this
.
lOMoARcPSD|27395923
Exercise 2D
1. Given that a > 1, prove that a2 > 1 by proving that LHS − RHS > 0.
2. Prove these inequalities for a, b ∈ R. [Hint: Begin with LHS − RHS.]
2
a2 b2 a2 + b2

2 2 a+b
(a) a + b ≥ 2ab (b) 2 + 2 ≥ 2 (c) ≥
b a 2 2
1
3. Prove that a + ≥ 2 for a > 0.
a
4. Prove, for a, b > 0, that:
√ √
(a) 21 (a + b) ≥ ab (b) 1
3
a + 34 b ≥ ab
5. If a > b > 0, prove that:
(a) a2 − b > b2 − a (b) a3 − b3 > a2 b − ab2
√
6. (a) Given that x and y are non-negative, prove that x + y ≥ 2 xy.
(b) Hence prove that (x + y)(x + z)(y + z) ≥ 8xyz, where z is also non-negative.
DEVELOPMENT
7. Suppose that p, q and r are real and distinct.

(a) Prove that p2 + q 2 > 2pq.
(b) Use part (a) three times to prove that p2 + q 2 + r 2 > pq + qr + rp.
(c) Given that p + q + r = 1, prove that pq + qr + rp < 31 .
[Hint: Begin with (p + q + r)2 = 1 and use part (b).]
8. Suppose that a, b and c are real numbers.
(a) Prove that a4 + b4 + c4 ≥ a2 b2 + a2 c2 + b2 c2 .
(b) Hence show that a2 b2 + a2 c2 + b2 c2 ≥ a2 bc + b2 ac + c2 ab.
(c) Deduce that if a + b + c = d, then a4 + b4 + c4 ≥ abcd.
9. Suppose that a, b and c are positive.
(a) Prove that a2 + b2 ≥ 2ab.
(b) Hence prove that a2 + b2 + c2 ≥ ab + bc + ca.
(c) Given that a3 + b3 + c3 − 3abc = (a + b + c)(a2 + b2 + c2 − ab − bc − ca),
prove that a3 + b3 + c3 ≥ 3abc.
1
(d) If x, y and z are positive, show that x + y + z ≥ 3 (xyz) 3 .
√
10. (a) Prove that 1 + x ≥ 2 x for x > 0.
(b) Suppose that x, y, z > 0 and (1 + x)(1 + y)(1 + z) = 8. Prove that xyz ≤ 1.
4
a b
11. (a) Expand − .
b a
a4 b4
2
b2

a
(b) Hence prove that 4 + 4 + 6 ≥ 4 2 + 2 for a, b ∈ R.
b a b a
2 2

x y x y
(c) Deduce that 2 + 2 + 6 ≥ 4 + for x, y > 0.
y x y x
12. (a) Suppose that a, b, c and d are positive.
a2 + b2 + c2 + d2 √
Use the fact that a2 + b2 ≥ 2ab to show that ≥ abcd.
4
w+x+y+z √
(b) Hence show that ≥ 4 wxyz for w, x, y, z > 0.
4
and this
.
lOMoARcPSD|27395923
13. (a) Show that a2 + b2 ≥ 2ab for all real numbers a and b.
(b) Hence show that a2 + b2 + c2 − ab − bc − ca ≥ 0.
(c) Deduce that for all positive real numbers a, b and c:
(a + b + c)2 ≥ 3(ab + bc + ca)
(d) Suppose that a, b and c are the side lengths of a triangle.
(i) Explain why (b − c)2 ≤ a2 .
(ii) Deduce that (a + b + c)2 ≤ 4(ab + bc + ca).
14. Suppose that a, b and c are positive.
a b
(a) Prove that + ≥ 2.
b a

1 1 1
(b) Hence show that (a + b + c) + + ≥ 9.
a b c

3 3 a b
(c) (i) Prove that a + b ≥ + abc, and write down similar inequalities for b3 + c3
c c
and c3 + a3 .
(ii) Hence prove that a3 + b3 + c3 ≥ 3abc.
a b c
(iii) Deduce that + + ≥ 3.
b c a
a b
15. In the previous question we proved that + ≥ 2 for a, b > 0. Use this result to prove
b a
that ab(a + b) + bc(b + c) + ca(c + a) ≥ 6abc for a, b, c > 0.
16. Suppose that x and y are positive.
1 1 4
(a) Prove that + ≥ .
x y x+y
1 1 2
(b) (i) Prove that 2 + 2 ≥ .
x y xy
1 1 8
(ii) Use part (i) and the square of part (a) to prove that 2
+ 2 ≥ .
x y (x + y)2
17. It is known that |a| ≥ a for any real number a.
Let x and y be any two real numbers. Use the above result to prove the triangle inequality
|x − y| ≥ |x| − |y| .
[Hint: Begin with LHS2 − RHS2 .]
18. [Triangle inequality with complex numbers]
(a) Let z = x + iy be a complex number. Prove algebraically that Re(z) ≤ |z|.
(b) Let z and w be two complex numbers. Prove that |z + w| ≤ |z| + |w|.
Begin by writing |z + w|2 = (z + w)(z + w) .
(c) Under what circumstances is |z + w| = |z| + |w|?
ENRICHMENT
19. Given a, b, c > 0 and abc = 1, use the AM/GM inequality with three terms to prove that
1 + ab 1 + bc 1 + ca
+ + ≥ 3.
1+a 1+b 1+c
and this
.
lOMoARcPSD|27395923
80 CHAPTER 2: Proof 2E
20. (a) Suppose that a, b, c, a + b − c, a + c − b and b + c − a are all positive.

(i) Prove that (a + b − c)(a + c − b) ≤ a2 .
(ii) Hence prove that (a + b − c)(a + c − b)(b + c − a) ≤ abc.
(b) Now consider a triangle ABC with side lengths a, b and c.
(a + b − c)(a + c − b)
(i) Prove that sin2 21 A = .
4bc
1
(ii) Deduce that sin 21 A sin 21 B sin 21 C ≤ .
8
a b c 3
21. If a, b, c > 0, prove that + + ≥ .
b+c a+c a+b 2
[Hint: Begin by adding one to each fraction on the LHS.]
2E Induction
Induction is an important topic in the Mathematics Extension 1 course, and
candidates in Extension 2 are expected to be proficient at the two main styles
already met, sums and divisibility. The Exercise questions include problems
which review this work. Induction will further be applied to inequalities and
verifying formulae for sequences specified recursively.
Review of Induction: Recall that there are three main parts to an induction proof.
First, the statement is verified for any initial terms. The second step proves the
implication that if the statement is true for some integer k then it is true for the
next integer (k + 1). Each proof then concludes with an appeal to the principle
of mathematical induction.
WORKED EXAMPLE 13: Prove by mathematical induction that

n
X
(−1)j j 2 = 12 (−1)n n(n + 1)
j=1
for all positive integers n.

SOLUTION: This is just the nth partial sum of the sequence Tj = (−1)j j 2 ,
Xn
so let Sn = Tj
j=1
= −1 + 22 − 32 + . . . + (−1)n n2 .
2
Also recall that for the partial sums of any sequence

Sn+1 = Sn + Tn+1 .
A. When n = 1, LHS = (−1)1 × 12

= −1
and RHS = 21 (−1)1 × 1 × 2
= −1
= LHS
so the result is true for n = 1.
and this
.
lOMoARcPSD|27395923
2E Induction 81
B. Assume the statement is true for the positive integer n = k.

That is, assume that Sk = 12 (−1)k k(k + 1) (†)
Now prove the statement for n = k + 1. That is, prove that
Sk+1 = 21 (−1)k+1 (k + 1)(k + 2).
But LHS = Sk + Tk+1
so by the induction hypothesis (†):
LHS = 12 (−1)k k(k + 1) + (−1)k+1 (k + 1)2

= 21 (−1)k+1(k + 1) − k + 2(k + 1)
= 21 (−1)k+1 (k + 1)(k + 2)
= RHS.
C. It follows from parts A and B by mathematical induction that the result is

true for all integers n ≥ 1.
Harder questions may involve divisibility problems. In other cases the statement
may only be true for values of n in some specified sequence. The easiest way to
manage this is to replace n with the formula for the mth term of the sequence.
The next Worked Example demonstrates both situations.
WORKED EXAMPLE 14: Prove by mathematical induction that 3n −2n is divisible

by 5 when n is a positive even integer.
SOLUTION: Since n is even, put n = 2m.

Now show that 32m − 22m is divisible by 5 for m ∈ Z+ .
A. When m = 1, 32m − 22m = 9 − 4
= 5×1
so the result is true for m = 1.
B. Assume the statement is true for the positive integer m = k.
That is, assume that 32k − 22k = 5p, for some integer p. (†)
Now prove the statement for m = k + 1.
That is, prove that 32(k+1) − 22(k+1) is divisible by 5.
Now 32(k+1) − 22(k+1) = 32k × 9 − 22k × 4
32(k+1) − 22(k+1) = (5p + 22k ) × 9 − 22k × 4
= 45p + 22k × 5
= 5(9p + 22k )
which is divisible by 5.
true for all integers m ≥ 1, and hence is true for all positive even integers n.
Proving Inequalities: Often it is necessary to prove an inequality by induction, thus

combining the techniques of this section with those presented in Section 2D. In
many cases, the best approach for the inequality is to put everything on one side.
and this
.
lOMoARcPSD|27395923
WORKED EXAMPLE 15: Prove by mathematical induction that 2n > n3 for all
integers n ≥ 10.
SOLUTION:
A. When n = 10, LHS = 210
= 1024
and RHS = 103
= 1000
That is, assume that 2k > k3 (†)
2k+1 > (k + 1)3 .
This will be done by proving LHS − RHS > 0.
LHS − RHS = 2k+1 − (k + 1)3
= 2 × 2k − (k3 + 3k2 + 3k + 1) ,
LHS − RHS > 2k3 − (k3 + 3k2 + 3k + 1)
= k3 − (3k2 + 3k + 1)
> k3 − (3k2 + 3k2 + 3k2 ) (since k > 1)
= k2 (k − 9)
>0 (since k ≥ 10.)
Proving Recursive Formulae: Recall that sequences can be defined using an initial
value and a recursive formula. For example, consider the sequence defined by
Tn = Tn−1 + 2n , for n > 1, where T1 = 2 .
The first few terms of that sequence are 2, 6, 12, 20, 30, 42, . . . It appears that a
simpler formula for this sequence is
Tn = n2 + n
and this can be proved by mathematical induction.
WORKED EXAMPLE 16: For the sequence defined recursively by

Tn = Tn−1 + 2n , for n > 1, where T1 = 2 ,
prove by mathematical induction that
Tn = n2 + n , for all n ≥ 1.
SOLUTION:
A. When n = 1, RHS = 12 + 1
=2
and this
.
lOMoARcPSD|27395923
2E Induction 83

That is, assume that Tk = k2 + k (†)
Tk+1 = (k + 1)2 + (k + 1)
= k2 + 3k + 2 .
From the given recursion formula:
LHS = Tk + 2(k + 1)
LHS = k2 + k + 2(k + 1)
= k2 + 3k + 2
= RHS .
Notice how easy this last induction proof was. It is generally the case that
recursive formulae are easy to prove by induction, but there are exceptions.
Exercise 2E
1. Prove by induction that for all positive integer values of n:
n
X Xn
(a) 1
r = 2 n(n + 1) (d) (2r − 1)2 = 31 n(2n − 1)(2n + 1)
r=1 r=1
n n
X X 1 n
(b) r(r + 1) = 31 n(n + 1)(n + 2) (e) =
r=1 r=1
r(r + 1) n+1
n n
X X 1 n
(c) r 2 = 61 n(n + 1)(2n + 1) (f) =
r=1 r=1
(2r − 1)(2r + 1) 2n + 1

(a) 5n + 3 is divisible by 4 (d) 9n+2 − 4n is divisible by 5
(b) 23n + 6 is divisible by 7 (e) 6n − 5n + 4 is divisible by 5
(c) 5n + 2n+1 is divisible by 3 (f) 4n + 6n − 1 is divisible by 9
DEVELOPMENT

(a) 1 + (1 + 2) + (1 + 2 + 3) + · · · + (1 + 2 + 3 + · · · + n) = 61 n(n + 1)(n + 2)
2
(b) 13 + 23 + 33 + · · · + n3 = (1 + 2 + 3 + · · · + n)
[Hint: Use question 1 part (a).]
4. Prove by mathematical induction that for even positive values of n:
(a) n2 + 2n is a multiple of 8 (b) n3 + 2n is a multiple of 12
5. Prove by mathematical induction that for odd positive values of n:
(a) 7n + 2n is divisible by 9 (b) 7n + 13n + 19n is divisible by 13
6. Prove these inequalities by mathematical induction:
(a) n2 ≥ 3n − 2 for n ≥ 1 (b) 2n ≥ 1 + 3n for n ≥ 4
and this
.
lOMoARcPSD|27395923
7. (a) Prove by induction that (1 + c)n > 1 + cn for all integers n ≥ 2, where c is a non-zero
constant greater than −1.
n
1 1
(b) Hence show that 1 − > for all integers n ≥ 2.
2n 2
8. (a) Solve the inequation x2 > 2x + 1.
(b) Hence prove by induction that 2n > n2 for all integers n ≥ 5.
9. In each case a sequence has been defined recursively and then a formula given for the nth
term. Use mathematical induction to prove each formula.
(a) If T1 = 1 and Tn = Tn−1 + n for n ≥ 2, then Tn = 21 n(n + 1) for n ≥ 1.
(b) If T1 = 1 and Tn = 2Tn−1 + 1 for n ≥ 2, then Tn = 2n − 1 for n ≥ 1.
(c) If T1 = 5 and Tn = 2Tn−1 + 1 for n ≥ 2, then Tn = 6 × 2n−1 − 1 for n ≥ 1.
3Tn−1 − 1 n
(d) If T1 = 1 and Tn = for n ≥ 2, then Tn = for n ≥ 1.
4Tn−1 − 1 2n − 1
d
10. (a) By differentiating from first principles, show that x = 1.
dx
d n
(b) Use mathematical induction and the product rule to show that x = nxn−1 for
dx
all positive integer values of n.
11. Prove by induction that the interior angle sum of a polygon with n sides is (n − 2) × 180◦ .
[Hint: Dissect the (k + 1)-gon into a k-gon and a triangle.]
12. Prove by induction that a polygon with n sides has 21 n(n − 3) diagonals.
13. Prove by induction that n lines in the plane, no two being parallel and no three concurrent,
divide the plane into 21 (n2 + n + 2) regions. [Hint: The (k + 1)th line will cross the other
k lines in k distinct points, and so will add k + 1 regions.]
14. Prove by mathematical induction that every set with n members has 2n subsets. [Hint:
When a new member is added to a k-member set, then every subset of the resulting
(k + 1)-member set either contains or does not contain the new member.]
1 1 1 1 1
(a) 2 + 2 + 2 + · · · + 2 ≤ 2 −
1 2 3 n n
1 × 3 × · · · × (2n − 1) 1
(b) ≥
2 × 4 × · · · × 2n 2n
16. Prove by induction that n3 − n is divisible by 24 for odd positive values of n.
17. [Formulae for APs and GPs] Use mathematical induction to prove each result.
(a) If T1 = a and Tn = Tn−1 + d for n ≥ 2, then Tn = a + (n − 1)d for n ≥ 1.
(b) If T1 = a and Tn = r Tn−1 for n ≥ 2, then Tn = ar n−1 for n ≥ 1.
(c) If S1 = a and Sn − Sn−1 = a + (n − 1)d for n ≥ 2, then
Sn = 21 n 2a + (n − 1)d for n ≥ 1.

(d) If S1 = a and Sn − Sn−1 = ar n−1 for n ≥ 2 and r 6= 1, then

a (r n − 1)
Sn = for n ≥ 1.
r−1
and this
.
lOMoARcPSD|27395923
2E Induction 85
18. Suppose that a, b > 0 and n is a positive integer.

(a) Prove the inequality an+1 + bn+1 ≥ an b + bn a.
n
an + bn

a+b
(b) Hence prove by induction that ≤ for all positive integers n.
2 2
√ √ 1
19. (a) By rationalising the numerator, prove thatn+1− n > √ .
2 n+1
1 1 1 √
(b) Hence prove by induction that 1 + + + · · · + < n, for n ≥ 7.
2 3 n
20. (a) Prove the identity 2 cos A sin B = sin(A + B) − sin(A − B).
(b) Hence prove by induction that for all positive integers n,
1
sin(n − 21 )θ
2 + cos θ + cos 2θ + · · · + cos(n − 1)θ = .
2 sin 21 θ
21. (a) Prove that for positive values of x and y,
x y
+ ≥ 2.
y x
(b) Hence prove by induction that for positive values of a1 , a2 , . . . , an ,

1 1 1
a1 + a2 + · · · + an + +···+ ≥ n2 .
a1 a2 an
(c) Deduce that cosec2 θ + sec2 θ + cot2 θ ≥ 9 cos2 θ.
ENRICHMENT
22. In each case a sequence has been defined recursively and then a formula given for the nth
term. Use a stronger form of induction to prove each formula.
(a) If T1 = 3, T2 = 6, and Tn = 3 Tn−1 − 2 Tn−2 − 1 for n ≥ 3, then Tn = n + 2n for n ≥ 1.
(b) If T1 = 8, T2 = 34, and Tn = 8 Tn−1 − 15 Tn−2 for n ≥ 3, then Tn = 5n + 3n for n ≥ 1.
(c) If T1 = 12, T2 = 30, and Tn = 5 Tn−1 − 6 Tn−2 for n ≥ 3, then
Tn = 3 × 2n + 2 × 3n for n ≥ 1.
(d) If T0 = T1 = 2, and Tn = 2 Tn−1 + Tn−2 for n ≥ 2, then
√ n √ n
Tn = 1 + 2 + 1 − 2 for n ≥ 0.
23. (a) Prove the inequality

1 1 1 1 1 1 1 k
1− + − + − +···+ − > .
2 3 4 5 6 2k − 1 2k (2k + 1)(2k + 2)
(b) Hence prove by induction that for all n ≥ 2,

1 1 1 1 1 1 1
n 1+ + +···+ > n+1 + + +···+ .
3 5 2n − 1 2 4 6 2n
24. Define Sn as the sum of the products of all the pairs of distinct integers that can be formed
from the first n positive integers.
So, for example, S3 = 1 × 2 + 1 × 3 + 2 × 3.
1
Prove by mathematical induction that Sn = 24 (n − 1)(n)(n + 1)(3n + 2) for n ≥ 2.
and this
.
lOMoARcPSD|27395923
86 CHAPTER 2: Proof 2F
25. A squad of n footballers put their training tops out to wash. When the washing has
finished drying, each player takes a training top, but it is found that no-one has their own.
This situation is called a derangement. Let Dn be the number of derangements.
(a) In some derangements, Ben and another player have each other’s top. Explain why
the number of these derangements is (n − 1)Dn−2 , for n > 2.
(b) Find a similar formula for the remaining derangements and hence show that
Dn = (n − 1)Dn−1 + (n − 1)Dn−2 , for n > 2 .
(c) Use the above result to show that

Dn − nDn−1 = (−1) × Dn−1 − (n − 1)Dn−2 , for n > 2 .
(d) Find D1 and D2 , then prove by induction that Dn − nDn−1 = (−1)n , for n > 1.
n
X (−1)r
(e) Hence prove by induction that Dn = n! for all n ∈ Z+ .
r=0
r!
2F Inequalities in Geometry and Calculus

The questions in this section often require the calculus of integration or curve
sketching. Consequently it may be appropriate to delay this section to later in
the course, such as at the end of the Extension 2 Integration chapter.
An Inequality Using Geometry: The AM/GM inequality was proven algebraically

in Section 2D. Here is a geometric proof of this important result. It has the
advantage of allowing readers to see how the inequality works, and identify the
special case of equality.
WORKED EXAMPLE 17: Prove the AM/GM inequality for positive real numbers a
and b using circle geometry. That is, prove that
a+b √
≥ ab
2
and identify when the two sides are equal.
SOLUTION: W
Construct line segment P Q with |P Q| = a + b.
Construct the circle with diameter |P Q| and centre O. x
The radius of this circle is clearly r O V
P Q
a+b a b
r= (the arithmetic mean.)
2
Let V be the point that divides P Q in the ratio a : b.
U
Construct chord U W perpendicular to P Q at V .
Hence V bisects U W , so let U V = V W = x.
Now x2 = ab (product of intercepts of intersecting chords)
√
so x = ab (the geometric mean.)
Finally, the longest chord in a circle is a diameter, hence
r≥x
a+b √
that is ≥ ab
2
with equality when U V is also a diameter. That is, a = b.
and this
.
lOMoARcPSD|27395923
2F Inequalities in Geometry and Calculus 87
An Inequality Using Calculus:

Many inequalities arise through the study of calculus and y
geometry. As a very simple example, using an upper and y = x1
a lower rectangle to approximate the area between the
hyperbola y = x1 and the x-axis for 1 < x < 2 gives
1
1
2
< log 2 < 1 . 1
2
Often knowledge of other topics is also required. The 1 2 x

following worked example makes use of calculus and a
geometric series to find an approximation for log 23 which
is accurate to two decimal places.
WORKED EXAMPLE 18: Consider the geometric series

S2n = 1 − h + h − . . . + h2n ,
2
where 0 < h < 1.

1
(a) Show that S2n−1 < < S2n .
1+h
(b) Integrate the previous result between h = 0 and h = x, where 0 < x < 1,
and hence write down a polynomial inequality for log(1 + x).
(c) Use n = 3 to estimate the value of log 32 .
SOLUTION:
(a) Now S2n−1 = 1 − h + h2 − h3 + . . . − h2n−1
1 − (−h)2n
= (by GP theory)
1 − (−h)
1 − h2n
=
1+h
1
< (increase numerator)
1+h
and S2n = 1 − h + h2 − . . . + h2n
1 − (−h)2n+1
= (by GP theory)
1 − (−h)
1 + h2n+1
=
1+h
1
> (decrease numerator)
1+h
1
hence S2n−1 < < S2n .
1+h
Z x Z x Z x
1
(b) S2n−1 dh < dh < S2n dh
0 0 1+h 0
x x
h2 h2n h2 h2n+1
h ix
or h− +...− < log(1 + h) < h − +...+
2 2n 0 0 2 2n + 1 0
x2 x3 x2n x2 x3 x2n+1
so x − + −...− < log(1 + x) < x − + −...+
2 3 2n 2 3 2n + 1
and this
.
lOMoARcPSD|27395923
1
(c) Put n = 3 and x = 2
to get
1 1 1 1 1 1
2
− 8
+ 24
− 64
+ 160
− 384
< log 23 < 1
2
− 1
8
+ 1
24
− 1
64
+ 1
160
− 1
384
+ 1
896
259
so 640
< log 23 < 909
2240
or 0·4047 < log 23 < 0·4058

.
These calculations suggest that log 23 =. 0·405, correct to three decimal places.
The actual value is 0·40547, correct to five decimal places.
Exercise 2F
1. (a) A regular dodecagon is drawn inside a circle of radius 1 cm
and centre O so that its vertices lie on the circumference, as
shown in the first diagram. Determine the area of 4OAB,
O
and hence find the exact area of the inscribed dodecagon.√
(b) (i) Use the formula for tan 2θ to show that tan 15◦ = 2 − 3.
B
(ii) Another regular dodecagon is drawn with centre O, so
that each side is tangent to the circle, as shown in the A
second diagram. Find the area of 4OGH and hence find
the exact area of the circumscribed dodecagon.
(c) By considering the results in parts (a) and (b), show that
√ . O
3 < π < 12(2 − 3) = . 3·24 .
H
2. (a) Use Simpson’s rule with three function values to approximate
the area under y = sin x between x = 0 and x = π3 . G
. 18
√
(b) Hence show that π = . 13 (4 − 3), which is accurate to two
decimal places.
1 y
3. The points A, P and B on the curve y = have x-coordinates
x
1, 1 21 and 2 respectively. The points C and D are the feet of the 1 A
perpendiculars drawn from A and B to the x-axis. The tangent M
P B
to the curve at P cuts AC and BD at M and N respectively.
N
(a) Find the areas of trapezia ABDC and M N DC. C D
(b) Hence show that 32 < ln 2 < 43 . 1 3
2 2 x
4. The diagram shows the points A(0, 1) and B(−1, e−1 ) on the curve
y = ex , and the point C(−1, 0) on the x-axis. The tangent at y
1
P (− 21 , e− 2 ) intersects OA at M and BC at N .
A
(a) Determine the exact area of the region bounded by the curve, M
BC, CO and OA. P
B
(b) Find the area of trapezium OABC.
N
(c) Show that the area of OM N C is equal to CO×P Q, and hence
C Q O x
find the area of the trapezium.
√
(d) Hence show that 21 (3 + 5) < e < 3.
5. Suppose that m ≤ f (x) ≤ M in the interval a ≤ x ≤ b. Use a diagram to help prove that
Z b
m(b − a) ≤ f (x) dx ≤ M (b − a) .
a
and this
.
lOMoARcPSD|27395923
DEVELOPMENT
1
6. The diagram shows upper rectangles for the graph of y = . y
x
(a) By considering appropriate areas, show that
1 1 1
1+ + + . . . + ≥ log(n + 1) .
2 3 n
1
(b) What do you conclude about the infinite series
1 3 5 x
1 1
1 + + + . . .?
2 3
7. (a) Show, using calculus, that the graph of y = ln x is concave down throughout its
domain.
(b) Sketch the graph of y = ln x, and mark two points A(a, ln a) and B(b, ln b) on the
curve, where 0 < a < b.
(c) Find the coordinates of the point P that divides the interval AB in the ratio 2 : 1.
(d) Using parts (b) and (c), deduce that 31 ln a + 32 ln b < ln( 31 a + 32 b).
8. Let f (x) = xn e−x , where n > 1.
(a) Show that f 0 (x) = xn−1 e−x (n − x).
(b) Show that (n, nn e−n ) is a maximum turning point of the graph of f (x), and hence
sketch the graph for x ≥ 0. (Don’t attempt to find points of inflexion.)
(c) Explain why xn e−x < nn e−n for x > n. Begin by considering the graph of f (x) for
x > n.
(d) Deduce from part (c) that (1 + n1 )n < e.
9. The function f (x) is defined by f (x) = x − loge (1 + x2 ).
(a) Show that f 0 (x) is never negative.
(b) Explain why the graph of y = f (x) lies completely above the x-axis for x > 0.
(c) Hence prove that ex > 1 + x2 , for all positive values of x.
10
x x
10. Consider the function y = e 1 − .
10
(a) Find the two turning points of the graph of the function.
(b) Discuss the behaviour of the function as x → ∞ and as x → −∞.
(c) Sketch the graph of the function.
−10
x
(d) From your graph, deduce that ex ≤ 1 − , for x < 10.
10
10 10
11 10
(e) Hence show that ≤e≤ .
10 9
11. (a) Let a, b and c be the lengths of the sides of a triangle and let 6 A be opposite side a.
Use the cosine rule to help prove that |b − c| ≤ a ≤ b + c.
That is, prove that one side of a triangle is longer than the difference between the
other two sides and shorter than the sum of the other two sides.
(b) Hence prove for any two complex numbers z and w that
|z| − |w| ≤ |z ± w| ≤ |z| + |w| .
(c) Under what circumstances is |z| − |w| = |z + w|?
and this
.
lOMoARcPSD|27395923
12. In this question you may assume that simple exponential curves are concave up.
(a) Show by direct calculation that: (i) 66 < 3 × 56 , (ii) 5 × 66 < 2 × 76 .
1

(b) The points A − 61 , 3− 6 and B(0, 1) lie on the exponential curve y = 3x . The points
1 x
B and C 61 , 52 6 lie on the exponential curve y = 25 .
(i) Use part (a) to show that the gradient of chord AB is greater than 1 and the
gradient of chord BC is less than 1.
(ii) Hence show that 25 < e < 3.
13. Let |t| < 1 and let N be a positive integer.
1
(a) Show that 1 + t2 + t4 + . . . + t2N < .
1 − t2
t2N +2
(b) Show that the difference between the two is .
1 − t2
(c) Integrate the result in part (a) between 0 and x, where 0 < x < 1. Hence show that:
x3 x5 x2N +1

1 1+x
x+ + +...+ < 2 log .
3 5 2N + 1 1−x
Z x 2N +2 Z x 2N +2
t x
(d) Explain why 2
dt ≤ 2
dt .
0 1−t 0 1−t
(e) Use parts (b) to (d) to show that
x3 x5 x2N +1

1 1+x
lim x + + +...+ = 2 log .
N →∞ 3 5 2N + 1 1−x
(f) Hence find log 2 correct to three decimal places.
1
14. The diagram shows the graph of y = , for t > 0. y
t
Let x > 1 and α > 0.
(a) By considering upper and lower rectangles, show that
1
0 < 21 α log x < xα/2 .

log x 1 xa/2 t
(b) Hence show that lim = 0, for all α > 0.
x→∞ xα
15. (a) Let n > 1 and k be positive integers. Use lower rectangles to prove that
Z nk+1
1 1
1− n ≤ dx .
nk x
Z nk
1
(b) Hence prove that dx → ∞ as k → ∞ regardless of the choice of n.
1 x
Z n+x
1
16. Consider the integral dt.
n t
x x

(a) Use upper and lower rectangles to show that x < n log 1 + n < x .
1+ n
x n x

(b) Hence show that lim 1 + n = e for any given value of x.
n→∞
n . x
(c) Use trial and error to determine how big n needs to be so that 1 + nx = . e correct
to three decimal places when x = 0·1.
and this
.
lOMoARcPSD|27395923
ENRICHMENT
17. The diagram shows the curves

y
y = log x and y = log(x − 1) ,
and k−1 rectangles constructed between x = 2 and x = k+1,

where k ≥ 2.
(a) Show that: 1 2 3 4 k k+1 x
Z k+1
(i) log(x − 1) dx = k log k − k + 1
2
Z k+1
(ii) log x dx = (k + 1) log(k + 1) − log 4 − k + 1
2
(b) Deduce that kk < k! ek−1 < 41 (k + 1)k+1 , for all k ≥ 2.
y y = f (x)
18. The diagram on the right shows the curve y = f (x) in the interval
a ≤ x ≤ b where f 00 (x) > 0. The corresponding chord is CD
D
and M N is tangent to y = f (x) at P where x = a+b2
.
N
(a) Use areas to briefly explain why C
P
Z b M
a+b f (a) + f (b) a+b x
(b − a) f < f (x) dx < (b − a) . a 2 b
2 a 2
(b) Hence show that, for n = 2, 3, 4, . . .,

4 1 1 1 1 1
< − < + .
(2n − 1)2 n−1 n 2 (n − 1)2 n2
(c) Deduce that

1 1 1 1 1 1 1
4 + + +··· < 1 < + + + +··· .
32 52 72 2 22 32 42
(d) Show that

1 1 1 1 1 1 1
2
+ 2 + 2 +··· < 2 + 2 + 2 +··· .
2 3 4 5 3 5 7
∞
3 X 1 7
(e) Hence show that < 2
< .
2 n=1 n 4
Z n
19. (a) Show that ln x dx = n ln n − n + 1.
1
(b) Use the trapezoidal rule on the intervals with endpoints 1, 2, 3, . . . , n to show that
Z n
ln x dx > 21 ln n + ln(n − 1)!
1
1
(c) Hence show that n! < nn+√2 e1−n . Note: This is a preparatory lemma in the proof
. 1
of Stirling’s formula n! =
. 2π nn+ 2 e−n , which gives an approximation for n! whose
percentage error converges to 0 for large integers n.
and this
.
lOMoARcPSD|27395923
20. (a) Prove that loge x ≤ x − 1 for x > 0.

(b) Suppose that p1 , p2 , p3 , . . . , pn are positive real numbers whose sum is 1.
Xn
Prove that loge (npr ) ≤ 0.
r=1
(c) Let x1 , x2 , x3 , . . . , xn be positive real numbers.
x1 + x2 + x3 + · · · + xn 1
Prove that ≥ (x1 x2 x3 . . . xn ) n .
n
21. [A proof that e is irrational]
n
X 1
For positive integer values of n, let Sn = 1 + .
r=1
r!
Z 1 n
x −x
(a) Prove by induction that e − Sn = e e dx for all positive integer values of n.
0 n!
3
(b) From (a) deduce that 0 < e − Sn < for all positive integer values of n.
(n + 1)!
(c) Use (b) to deduce that (e − Sn ) n! is NOT an integer for n = 2, 3, 4, . . .
p
(d) Show that there cannot exist positive integers p and q such that e = .
q
and this
.
lOMoARcPSD|27395923
2G Chapter Review Exercise 93
2G Chapter Review Exercise

Exercise 2G
1. Write down the converse of each statement, and state whether the converse is true or false.
(a) If a quadrilateral is cyclic, then its opposite angles are supplementary.
(b) If two numbers are both odd, then their sum is even.
(c) Every rhombus is a parallelogram.
2. Write down the negation of each statement.
(a) All mathematicians are intelligent.
(b) Suzie likes Physics and Chemistry.
(c) If I am on vacation, then I am not working.
3. Write down the contrapositive of each statement.
(a) If I am a bicycle, then I have two wheels.
(b) If a number is odd, then its last digit is not 6.
(c) A square has four equal sides.
4. Write down (in ‘if. . . then’ form) the two converse statements equivalent to each ‘if and
only if’ statement.
(a) A number is even if and only if it is divisible by 2.
(b) A quadrilateral is a parallelogram if and only if its diagonals bisect each other.
(c) a is divisible by b if and only if ∃ c ∈ Z such that a = bc.
5. Prove that:
(a) the sum of three consecutive integers is divisible by 3,
(b) the product of three consecutive even numbers is divisible by 8,
(c) the product of two consecutive even numbers is divisible by 8.
6. Prove that the remainder is 1 when an odd square number is divided by 4.
7. Prove that an integer is divisible by both 3 and 5 if it is divisible by 15.
8. If n is odd, use divisibility arguments to prove that n3 − n is divisible by 24.
9. If the integer n is not divisible by 3, prove that n2 + 2 is divisible by 3.
10. (a) Show by expanding that if n is odd,
xn + 1 = (x + 1)(xn−1 − xn−2 + xn−3 − xn−4 + · · · + x2 − x + 1).
(b) Hence prove that if n is odd:
(i) 2n + 1 is divisible by 3,
(ii) 2mn + 1 is divisible by 2m + 1.
11. Prove by contradiction that:
√
(a) 7 is irrational, (b) log3 7 is irrational.
12. By proving the contrapositive, prove that if a2 is even then a is even.
√
13. (a) Prove that x + y ≥ 2 xy for x, y > 0.
(b) Hence prove that (a + b)(1 + ab) ≥ 4ab for a, b > 0.
and this
.
lOMoARcPSD|27395923
94 CHAPTER 2: Proof 2G

n
X
(a) (r + 1) × 2r = n × 2n+1
r=1
n
X
(b) r 2 (r + 1) = 1
12 n(n + 1)(n + 2)(3n + 1)
r=1

(a) 6n + 4 is divisible by 5,
(b) n3 + 2n is divisible by 3.
16. (a) Prove by induction that for all positive integer values of n:

1 1 1 1 1
1− 2 1− 2 ... 1− = 1+
2 3 (n + 1)2 2 n+1

1 1 1
(b) What is the value of lim 1 − 2 1− 2 ... 1− ?
n→∞ 2 3 (n + 1)2
17. Prove by mathematical induction that:
(a) n(n + 2) is divisible by 4 for even positive values of n,
(b) 3n + 7n is divisible by 10 for odd positive values of n,
(c) 4n + 5n + 6n is divisible by 15 for odd positive values of n.
18. Use mathematical induction to prove that if T1 = 3 and Tn = Tn−1 + 4n for n ≥ 2,
then Tn = 2n2 + 2n − 1 for n ≥ 1.
√
19. For a certain sequence, a1 = 1 and an+1 = 2an + 1 for n ≥ 1.
Prove by induction that an < 3 for n ≥ 1.
20. Prove by induction that n! > 3n for n ≥ 7.
21. Prove by induction that, for n ≥ 1, the nth derivative of xe−x is (−1)n (x − n)e−x .
22. Prove, for n ≥ 3, that the exterior angle sum of a convex n-sided polygon is 360◦ .
23. (a) Prove by induction that 2n > n, for all positive integers n.
√
(b) Hence show that 1 < n n < 2 for n ≥ 2.
√
(c) Suppose that a and n are positive integers. √ It is known that if n a is rational, then
it is an integer. What can we deduce about n n, where n is a positive integer greater
than 1?
x+y √ a+b+c+d √4
24. (a) Use the fact that ≥ xy, for x, y > 0, to prove that ≥ abcd,
2 4
for a, b, c, d > 0.

a+b+c 1 a+b+c
(b) (i) Show that = a+b+c+ .
3 4 3
a+b+c √
3
(ii) Hence prove that ≥ abc.
3
25. Suppose that abc is a 3-digit number, with a − c > 1. If we subtract cba from abc and then
add the result of this subtraction to the number obtained by reversing its digits, prove
that the answer is 1089.
and this
.
lOMoARcPSD|27395923
Appendix: The Fundamental Theorem of Arithmetic 95
Appendix: The Fundamental Theorem of Arithmetic

The fundamental theorem of arithmetic is not part of the Extension 2 course.
Its proof is included here firstly because the theorem is so important in the
study of number theory and secondly to justify its use in several proofs earlier in
this chapter, most notably in proving that log2 3 is irrational. The proof of the
theorem relies on several other results, which also need explanation and proof.
Theorem: Suppose that two integers a and b have HCF = d.
Then (a − bq) is divisible by d.
Proof: Both a and b have factor d, so there exist integers m and n such that
a = md and b = nd .
Thus a − bq = md − ndq
= d(m − nq) .
Hence (a − bq) is divisible by d.
Using the division algorithm, for integers a and b it is always possible to write
a = bq1 + r1 where 0 ≤ r1 < b ,
and where q1 and r1 are integers. It follows from the above proof that both b
and r1 are divisible by d, the HCF of a and b. Now repeat the division algorithm
to get
b = r1q2 + r2 where 0 ≤ r2 < r1 ,
and where q2 and r2 are integers. Once again it follows that both r1 and r2 are
divisible by d. Now continue the process to get a decreasing sequence of positive
remainders r1 , r2, . . . which must therefore terminate. Further, every term in
this sequence is divisible by d. Consequently the last term in this sequence of
remainders is d itself, the HCF of a and b. An example should help convince the
reader of this.
Finding the HCF of 81 and 66:
81 = 66 × 1 + 15 [1]
66 = 15 × 4 + 6 [2]
15 = 6 × 2 + 3 [3]
6 = 3 × 2.
Thus the HCF of 81 and 66 is 3.
More significantly, when the division algorithm is written out like this, the result
is a set of equations that can be used to write the HCF as a sum of multiples of
the original numbers. It is simply a matter of working backwards through the
equations. In this case:
3 = 15 − 6 × 2 (from [3])
= 15 − (66 − 15 × 4) × 2 (from [2])
= 15 × 9 − 66 × 2
= (81 − 66) × 9 − 66 × 2 (from [1])
= 81 × 9 − 66 × 11.
In other words, using these two algorithms, it is always possible to find the values
of x and y such that ax + by = d. In this case, x = 9 and y = −11.
and this
.
lOMoARcPSD|27395923
96 CHAPTER 2: Proof 2H
√
It is appropriate to pause at this point to give a remarkable proof that n is
either an integer or irrational whenever n is a positive integer.
√ √
Theorem: If n ∈ Z then either n ∈ Z or n is irrational.
√
Proof: Suppose that n is rational. Then ∃ a, b ∈ Z such that
√ a
n=
√ b
or b n=a (∗∗)
where b ≥ 1 and the HCF of a and b is 1. Thus, from above, there exist integers x
and y such that
ax + by = 1 .
√
Multiply this equation by n to get
√ √ √
n = (a n)x + (b n)y .
Now use the result of equation (∗∗) to get:
√
n = (bn)x + (a)y .
The RHS of this last equation
√ is a sum of products of integers, and must therefore
√
be an integer. Hence if n is rational then it must be an integer. Otherwise n
is irrational.
Here is another useful theorem that will be needed.
Theorem: If the HCF of prime p and integer a is 1, and if integer ab is divisible
by p, then b is divisible by p.
Proof: Since the HCF of p and a is 1, ∃ x, y ∈ Z such that
ax + py = 1
thus abx + pby = b .
But ab is divisible by p, so put ab = mp to get
mpx + bpy = b
that is p(mx + by) = b ,
by which b is divisible by p.
As a consequence of this proof, if an integer n is divisible by prime p then p
divides at least one factor of n. The proof of this stronger result can be done by
induction, and is left as an exercise. With these tools, it is now possible to prove
the fundamental theorem of arithmetic for positive integers.
Theorem: If a positive integer n can be written as the product of its primes
in two ways, then one is a rearrangement of the other.
Proof: Suppose, by way of contradiction there exists a positive integer n which
can be written as the product of its prime factors in two different ways. Then let
those products be
n = p1 × p2 × . . . × pk = q1 × q2 × . . . × q` ,
where, for the sake of simplicity, it is assumed k < `. Then, by definition,
q1 × q2 × . . . × q` is divisible by p1 .
It follows from above that p1 must divide one of these prime factors. Let this
be qj . But if one prime divides another then they must be equal. That is p1 = qj .
and this
.
lOMoARcPSD|27395923
Appendix: The Fundamental Theorem of Arithmetic 97
Now re-order and re-label the primes so that p1 = q1 . That is:

n = p1 × p2 × . . . × pk = q1 × q2 × . . . × q` with p1 = q1 .
Cancel this prime p1 so that
n
= p2 × p2 × . . . × pk = q2 × q3 × . . . × q` .
p1
Now repeat the above argument with p2 , p3, . . . pk , (or write it out properly using
induction) to get
n
= 1 = qk+1 × qk+2 × . . . × q` .
p1 × p2 × . . . × pk
But a product of primes cannot equal 1. Hence k = ` and each prime pi is equal
to a corresponding prime qi .
THE FUNDAMENTAL THEOREM OF ARITHMETIC: If a positive integer n can be written as

two different products of primes
19 n = p1 × p2 × . . . × pk = q1 × q2 × . . . × q` ,
then k = ` and the primes p1 , p2, . . . , pk are a re-arrangement of the primes
q1 , q2 , . . ., q` .
Note that the above proof can be adapted to include negative integer values of n,
and this is left as an exercise.
and this
.

02-proof

Uploaded by

Copyright:

Available Formats

02-proof

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

02-proof

Uploaded by

Copyright:

Available Formats

lOMoARcPSD|27395923

Scan to open on Studocu

Studocu is not sponsored or endorsed by any college or university

Chapter Overview: There are many opportunities for presenting or reading

2A The Language of Proof

2A The Language of Proof 57

is true if it is known that n = 6, but is clearly false if it is known that n = 7.

1, 1, 2, 3, 5, 8, 13, 21, 34, . . .

A statement which is shown to be true is said to be proven, and the evidence

A PROVEN STATEMENT: A statement which is shown to be true is said to be proven,

Examples and Counterexamples: Whilst it is rarely feasible, a statement can be

In contrast, the statement

A COUNTEREXAMPLE: A statement may be disproved (shown to be false) by a single

NEGATION: The negation of a statement changes its logical value. It is commonly

Notation: It is sometimes convenient to present and manipulate statements in an

And, Or and Negation: In the Advanced and Extension 1 work on probability

2A The Language of Proof 59

Here is the Venn diagram of F ∩ T . And here is the Venn diagram of F ∩ T .

Implication: An implication is the relationship between two statements by which one

IMPLICATION: An implication is the relationship between two statements by which

2A The Language of Proof 61

SUFFICIENT AND NECESSARY: In the statement if A then B,

Converse Statements: Consider the following two statements.

Equivalent Statements: Two statements are called equivalent if each is a logical

2A The Language of Proof 63

Whichever of the new terminology or symbol is used, it is important to remember

EQUIVALENT STATEMENTS AND SETS: The sets corresponding to equivalent statements

The Contrapositive: There are many examples of pairs of equivalent statements,

It is further confirmed that F ⊆ P by listing the elements of each set. Thus

2A The Language of Proof 65

5. Write down the contrapositive of each statement.

7. State whether each statement is true ∀ x ∈ R. If it is false, provide a counterexample.

8. State whether each statement is true ∀ a, b ∈ R. If it is false, provide a counterexample.

16. Consider this statement:

Divisibility: It is important to begin with a clear definition of what divisibility means.

14 DIVISIBILITY: If a, b ∈ Z and b is divisible by a then ∃ m ∈ Z such that b = am.

WORKED EXAMPLE 1: Let a and b be two integers divisible by 3.

WORKED EXAMPLE 2: Prove that a2 − a is even for all a ∈ Z.

WORKED EXAMPLE 3: A student claims that if an integer n is divisible by both 4

WORKED EXAMPLE 4: Let n = 10x+y, where n, x, y ∈ Z+ , the positive integers.

Hence if (x − 2y) is divisible by 7 then n = 10x + y is also divisible by 7.

1. (a) If a and b are even, prove that a + b is even.

2. (a) If a and b are even, prove that ab is even.

7. Prove that if a − b is even, then a − b2 is divisible by 4.

20. (a) Factorise a + 4b by adding and subtracting 4a2 b2.

(b) Hence prove that 5454 + 4545 is not a prime number.

2C Proof by Contraposition and by Contradiction

WORKED EXAMPLE 5: Prove that if n2 is even then n is even.

PROOF BY CONTRAPOSITION: An implication and its contrapositive are equivalent.

2C Proof by Contraposition and by Contradiction 71

PROOF BY CONTRADICTION: Proving a statement is equivalent to showing that its

WORKED EXAMPLE 7: Prove that log2 3 is irrational.

2C Proof by Contraposition and by Contradiction 73

6. (a) Prove that log3 5 is irrational.

16. Consider the following theorem about prime numbers.

(a) State the contrapositive of the theorem.

ARITHMETIC AND GEOMETRIC MEANS: Suppose that a and b are positive.

WORKED EXAMPLE 9: Prove the AM/GM inequality.

Combinations of Inequalities: Having established an inequality, it can be restated

WORKED EXAMPLE 10: