Lec13: Euler’s Theorem

30 Mar 2023, MATH4024-Number Theory, Lecturer: Prof A. Munagi

1 Introduction
We begin with the essential definitions.

Definition 1.1. Let n ∈ Z+ . The Euler phi-function (or totient function) of n, denoted
by φ(n), is the number of positive integers not exceeding n that are relatively prime to
n.

So φ(n) is the cardinality of the set {x ∈ Z+ | x < n and (x, n) = 1}, that is,

φ(n) = |{x ∈ Z+ | x < n and (x, n) = 1}|.

Examples. The values of φ(n) for n = 1, 2, . . . , 18 are shown in the following table.

n 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
φ(n) 1 1 2 2 4 2 6 4 6 4 10 4 12 6 8 8 16 6

Notice, for example, that φ(1) = 1, and φ(p) = p − 1 for any prime p. Thus φ(5) = 4
is the cardinality of {1, 2, 3, 4}, φ(9) = 6 is the cardinality of {1, 2, 4, 5, 7, 8}, etc.

Definition 1.2. A reduced residue system modulo n is any set of φ(n) integers such
that each element of the set is relatively prime to n, and no two different elements of
the set are congruent modulo n.

Example. The set {1, 3, 5, 7} is a reduced residue system mod 8, as well as {−3, −1, 1, 3}.

Proposition 1.3. If {r1 , r2 , . . . , rφ(n) } is a reduced residue system modulo n, and if

a ∈ Z+ with (a, n) = 1, then {ar1 , ar2 , . . . , arφ(n) } is a reduced residue system modulo
n.

1
Prof A. Munagi Three Special Congruences MATH4024-Number Theory

Proof. To show that each arj is relatively prime to n, we assume that (arj , n) > 1.
Then there is a prime divisor p of (arj , n). So p|a or p|rj which implies p|a and p|n, or
p|rj and p|n.
However, we cannot have both p|rj and p|n because rj is a member of a reduced residue
system modulo n, and the condition (a, n) = 1 forbids both p|a or p|n. Hence we
conclude that (arj , n) = 1 for 1 ≤ j ≤ φ(n).
To show that no two arj are congruent modulo n, we assume that arj ≡ ark (mod
n), where j 6= k with 1 ≤ j ≤ φ(n), 1 ≤ k ≤ φ(n).
Since (a, n) = 1, we get rj ≡ rk (mod n). But this is a contradiction since rj and rk
belong to the original set of reduced residues modulo n. Hence rj 6≡ rk (mod n).

Example. R = {1, 2, 4, 5, 7, 8} is a reduced residue system mod 9. By Proposition 1.3

the following sets are also reduced residue systems mod 9.

a = 2 =⇒ 2R = {2, 4, 8, 10, 14, 16}

a = 4 =⇒ 4R = {4, 8, 16, 20, 28, 32}
a = 5 =⇒ 5R = {5, 10, 20, 25, 35, 40}
.. .. .. .. .. .. ..
. . . . . . .

1.1 Euler’s Theorem

We now state our third special congruence (following from Lec12).
The great Swiss mathematician Leonhard Euler published a proof of Fermat’s little
theorem in 1736. In 1760, Euler found a natural generalization of the congruence in
Fermat’s theorem that holds for every integer modulus.

Theorem 1.4 (Euler’s Theorem). If a, m ∈ Z+ with (a, m) = 1, then

aφ(m) ≡ 1 (mod m).

Proof. (similar to the proof of Fermat’s Little Theorem, see Theorem 3.1 in Lec12).

Exercise 1. Prove Theorem 1.4 (Euler’s Theorem).

[Hint: in the proof of Fermat’s Little Theorem replace the sequence 1, 2, . . . , p − 1 with
the sequence r1 , r2 , . . . , rφ(n) .]

1.1.1 Consequences of Euler’s Theorem

These consequences generalize those of Fermat’s Little Theorem.
(1) Finding the least positive residue of integer powers to any modulus. For example,

If 7502 ≡ r (mod 15), where 0 ≤ r < 15, what is r?

2
Prof A. Munagi Three Special Congruences MATH4024-Number Theory

φ(15) = 8; 502 = 8 · 62 + 6. By Euler’s Theorem we know that 78 ≡ 1 (mod 15).

Hence
7502 ≡ (78 )62 · 76 ≡ 76 ≡ 4 (mod 15).

Exercise 2. Let n, k, m ∈ Z+ with m > 1 such that k = φ(m)q + r, 0 ≤ r ≤ φ(m).

Prove that nk ≡ nr (mod m).

(2) If (a, m) = 1, then aφ(m)−1 is an inverse of a modulo m:

a · aφ(m)−1 = aφ(m) ≡ 1 (mod m).

Example, 2φ(9)−1 = 26−1 = 25 = 32 ≡ 5 (mod 9). So 5 is an inverse of 2 modulo 9.

(3) If (a, m) = 1 then the solutions of ax ≡ b (mod m) are given by
x ≡ aφ(m)−1 b (mod m).
Example. To solve 3x ≡ 7 (mod 10), note that φ(10) = 4. The solutions are
x ≡ 3φ(10)−1 · 7 ≡ 33 · 7 ≡ 9 (mod 10).

Exercise 3. What is the remainder when 55142 is divided by 143?

Exercise 4. Show that 55552222 + 22225555 is divisible by 7.

Exercise 5. Show that for any nonnegative integer n, 36n − 26n is divisible by 35.

2 Pseudoprimes
Exercise 6. If b and n are positive integers with (b, n) = 1, then the congruence bn ≡ b
(mod n) is equivalent to the congruence bn−1 ≡ 1 (mod n).

Recall that Fermat’s little theorem says that if p is prime and a is not a multiple of
p, then ap−1 ≡ 1 (mod p), or ap ≡ a (mod p).
The contrapositive of this theorem gives a possible way to detect non-primes: if
we can find an integer a which is coprime to n such that an 6≡ a (mod n), then, by
Fermat’s little theorem n is composite.
Example: Let us check if n = 63 is composite:

263 = (26 )10 · 23 = (64)10 · 23 = 23 ≡ 8 6≡ 2 (mod 63).

Hence 63 is composite.

Exercise 7. Show that 589 is not prime by computing the least nonnegative residue
of 2588 modulo 589.

3
Prof A. Munagi Three Special Congruences MATH4024-Number Theory

A lot of composite numbers can indeed be detected by this test, but there are some
that escape it.
The converse of Fermat’s little theorem would have provided a primality test. But
it turns out to be false.
The ancient Chinese were said to have believed that if 2n ≡ 2 (mod n), then n must be
prime. This statement holds true for 1 ≤ n ≤ 340. However, it fails for n = 341 since
we can show that 2341 ≡ 2 (mod 341) (see Exerc. 8) even though 341 is not prime.
Exercise 8. Let n = 341 = 11 · 31.
(a) Show that 2340 ≡ 1 (mod 11).
(b) Show that 2340 ≡ 1 (mod 31).
(c) Deduce that 2341 ≡ 2 (mod 341).

Pseudoprimes are composite numbers that tend to fulfill the properties of primes.

Definition 2.1. Let b be a positive integer. If n is a composite positive integer and

bn ≡ b (mod n), then n is called a pseudoprime to the base b.

[These are sometimes called Fermat pseudoprimes because they originate from Fermat’s
little theorem.]
For example, we know (from Exerc. 8) that 341 is a pseudoprime to the base 2.

Exercise 9. Prove that 645 = 3 · 5 · 43 is a pseudoprime to the base 2.

There are composite integers that cannot be detected from the contrapositive of
Fermat’s little theorem, because some integers are pseudoprimes to every base, that is,
some integers are Carmichael numbers!

Definition 2.2. A composite number that satisfies bn−1 ≡ 1 (mod n) for all b ∈ Z+
when (b, n) = 1 is called a Carmichael number or an absolute pseudoprime.

[Named after American mathematician Robert D. Carmichael (1879-1967) who studied

them extensively in the early twentieth century..]
Example. We show that 561 = 3 · 11 · 17 is a Carmichael number. If (b, 561) = 1, then
(b, 3) = (b, 11) = 1 = (b, 17) = 1. Hence from Fermat’s little theorem we have
b2 ≡ 1 (mod 3), b10 ≡ 1 (mod 11) and b16 ≡ 1 (mod 17).
Consequently,
b560 ≡ (b2 )280 ≡ 1 (mod 3), b560 ≡ (b10 )56 ≡ 1 (mod 11) and
b560 ≡ (b16 )35 ≡ 1 (mod 17).
Therefore b560 ≡ 1 (mod 561) for all b with (b, n) = 1.

Question 10: Prove that 1105 = 3 · 13 · 17 is a Carmichael number.