MATH55 HOMEWORK #5

CHANG-YEON CHO

I can be reached at [email protected]

Notation and Remark In what follows, Ill use a m b rather than a b (mod m) to denote the mod m
equivalence between a and b. And I recommend everyone reads the solution for 4.1.44. I put a little bit of
comments on modulo arithmetic.
1. Divisibility and Modular Arithmetic
4.1.16 Let m be a positive integer. Show that a mod m = b mod m if a m b
Solution: Say a b = mc for some integer c. Apply division algorithm to b:
b = b0 m + r
where 0 r < m. Then
a = b + mc = b0 m + r + mc = (b0 + c)m + r
By the uniqueness of the quotient and remainder, r must be the remainder when dividing a by m. So a
mod m = b mod m = r

4.1.34 Show that if a m b and c m d, where a, b, c, d, and m are integers with m 2, then a c m b d.
Solution: Say a b = md1 and c d = md2 for some integers d1 and d2 . Then
(a c) (b d) = (a b) (c d) = md1 md2 = m(d1 d2 )
Then the result follows because d1 d2 is an integer.

4.1.36 Show that if a, b, c, and m are integers such that m 2, c > 0, and a m b, then ac mc bc.
Solution: Put a b = md for some integer d. Then
ac bc = c(a b) = mcd
Since d is an integer, this implies the result.

4.1.40 Prove that if n is an odd positive integer, then n2 8 1.

Solution: Since n is odd, n 4 1 or n 4 3. If n = 4m + 1 for some integer m, then
n2 = (4m + 1)2 = 16m2 + 8m + 1
Since m and m2 are integers, n2 8 1. The case for m 4 3 is similar.

4.1.44 Show that the distributive property of multiplication over addition holds for Zm , where m 2 is an
integer.
Solution: This is to prove that a(b + c) = ab + ac in Zm . Here the sum and multiplication are those of
Zm . First of all, I object to the notation Zm . It is somewhat widely used but its inappropriate and there
is no reason for using it. It should be written as Z/nZ or Z/n for short. Zp will be used later in number
theory.1 The meaning of / would be clear once you learn a basic algebra, especially quotient groups, rings,
1I also object to use A to denote the complement of a set A. The bar notation is for a closure in topology.
1

or modules. Secondly, an element in Z/nZ should be denoted by a

. Here a
denotes an equivalence class with
the choice of representative a. You start with the set of integers and make a partition out of it. Namely, you
cut the set of integers into n number of subsets A0 , , An1 where Ai is the set of integers whose remainder
is i when divided by n. Another way is to define a relation on Z. We define
a b if a n b
2

This gives you an equivalence relation. Recall that what you need to check is that
(1) a a for every a Z
(2) If a b, then b a for all a, b Z
(3) If a b and b c, then a c for all a, b, c Z
Read them in terms of modulo n and prove it. In this way, you somehow defined your own equality. Now
Id like to have a short-handed or better notion for Ai s. For any integer i, consider
i = {m Z : m n i}
That is, you pick i and collect all integers that is equal to i in view of your newly defined equality.
(Warning: The author of our textbook uses the bar notation to denote a multiplicative inverse
a
of an element a Zn . This is completely nonstandard and is not a good notation.3)
It follows then that
i = j if and only if i n j
(Check this!). What does that mean? You have a single set Ai . However, there are many different choices
for j that Ai = j. In fact, any j with j n i works.
What am I doing, huh? Im going to define a sort of arithmetic on the set
Z/nZ := {A0 , A1 , . . . , An1 }
Namely, Id like to add Ai with Aj and so on. This is pretty meaningful and useful. This is called a quotient
object and is a fundamental object when studying any sort of mathematical object. So lets define Ai plus
Aj . Remember 0 i, j n. You may feel like Ai +Aj should be Ai+j . But the index should be in between
0 and n. So there is an issue like this. Lets use bar notations which give you a better understanding for
this issue. What is the sum of i and j? Your innate mathematical genes tell you i + j. Aha! This might not
be well-defined! That is, what if I denote i + n for i? They are the same set! This arithmetic depends on
the choice of representatives for each i. So whenever you define some operation, you should be very careful.
Of course, math is beautiful and these kinds of arithmetic indeed define an arithmetic. This is so called a
quotient ring structure on Z/nZ.
Lets dive into the problem. Well prove that
a
(b + c) = a
b + a
c
for any integers a, b, and c. The key is that the addition and multiplication are well-defined as you expect.
Namely, define
a
+ b := a + b
a
b := ab
You do need to check that these are well-defined in a sense that these they are independent of choices of
representatives. Once you do this, all the remaining parts are pretty straightforward. For example,
a
(b + c) = a
b + c = a(b + c) = ab + ac = ab + ac = a
b + a
c
2An equivalence relation is one of the key concepts in this course. Well discuss it later in section 9.5
3Why do I prefer Z/nZ and a
? First of all, Z/nZ is best understood as a quotient object of Z. They contain the idea of

defining a arithmetic on cosets. And the slash and the bar are standard notations in this case. (People sometimes use [a] rather
than a
, too.) In particular, it suggest how we should define arithmetic on Z/nZ. Namely, it should be inherited arithmetic from
Z. From this point of view, the bar notation is fantastic. Wanna add elements? Just connect bars. Wanna multiply elements?
Connect bars! Whatever you wanna do, just connect bars. Secondly, its pretty annoying to work with {0, 1, , n 1} when
doing arithmetic. You need to worry about the size of numbers every time. For example, 2+3 should be 5 in Z5 but 5 is not an
element of Z5 . Theres no such worries in Z/nZ because we play with equivalence classes. Theres no way to avoid considering
modulo computation and going over equivalence classes. Finally, a
sheds some light on equivalence relation which is one of the
fundamental idea in math. This is the moment you think about equivalence relation.
2

Or if you want to stick with the textbook, here your solution. We need to prove that
a n (b +n c) = a n b +n a n c
Recall also that
a +n b = (a + b) mod n
And similarly for multiplication. But then
a n (b +n c) = (a ((b + c) mod n)) mod n
= a(b + c) mod n
= (ab + ac) mod n
= ((ab mod n) + (ac mod n)) mod n
= (ab mod n) +n (ac mod n)
= a n b +n a n c
You now know why I dont like the authors notation.
2. Integer Representations and Algorithms
4.2.32 Show that a positive integer is divisible by 11 if and only if the difference of the sum of its decimal
digits in even-numbered positions and the sum of its decimal digits in odd-numbered positions is divisible
by 11.
Solution: In the language of modulo arithmetic, an integer a is divisible by 11 if and only if a 11 0. So
let a be an integer. We can express a by
an an1 a1 a0
Pn
i
i
That is, a = i=0 ai 10 . Lets analyze 10 modulo 11. Since 10 11 1, 10i 11 (1)i . In particular,
n
X

ai 10i 11

i=0

Therefore, a is divisible by 11 if and only if

n
X

(1)i ai

i=0

Pn

i=0 (1)

ai 11 0 as desired.

3. Primes and Greatest Common Divisors

m

4.3.10 Show that if 2 + 1 is an odd prime, then m = 2n for some nonnegative integer n.
Solution: For every positive odd4 integer t, we have the equality of polynomials
xt + 1 = (x + 1)(xt1 xt2 + x + 1)
Why? Just expand it. Or better, observe that 1 is a zero of the polynomial xt + 1 and then apply division
algorithm for polynomials. Now use the prime factorization to express m = kt where k is a power of 2, say
2n for some nonnegative integer n, and t is odd allowing 1. Remark that the product of odd numbers is still
odd. Then the above identity says
xm + 1 = (xk + 1)(xk(t1) + xk(t2) + xk + 1)
simply by replacing x with xk . Now assume 2m is an odd prime. Plug-in 2 to the above identity to get
2m + 1 = (2k + 1)(2k(t1) + 2k(t2) + 2k + 1)
If k > 1, 2k + 1, which is strictly greater than 1, divides 2m + 1. This contradicts to the assumption. So
k = 1 and were done.
4.3.12 Prove that for every positive integer n, there are n consecutive composite integers.
Solution: Following the hint, for any positive integer n, consider the n consecutive integers starting with
(n + 1)! + 2. That is,
(n + 1)! + 2, (n + 1)! + 3, , (n + 1)! + (n 1)
4Why not even numbers? Do you see why?
3

We claim that none of them are prime numbers. This is because (n + 1)! + i is divisible by i but strictly
greater than i for each 2 i n 1.

4.3.16 Determine whether the integers in each of theses sets are pairwise relatively prime.
a) 21, 34, 55
b) 14, 17, 85
c) 25, 41, 49, 64
d) 17, 18, 19, 23
Solution: Only b) is not the case since 17 divides 85. For the others, think about their prime factorizations
and compare powers of each prime.

4.3.24 What are the greatest common divisors of these pairs of integers?
a) 22 33 55 , 25 33 52
b) 2 3 5 7 11 13, 211 39 11 1714
c) 17, 1717
d) 22 7, 53 13
e) 0, 5
f) 2 3 5 7, 2 3 5 7
Solution: Take the smallest powers for each common prime. a)22 33 52 , b)2 3 11, c)17, d)1, e)5, f )2 3 5 7.

4.2.32 Use the Euclidean algorithm to find

a) gcd(1,5)
b) gcd(100,101)
c) gcd(123,277)
d) gcd(1529, 14309)
e) gcd(1529, 14038)
f) gcd(11111,111111)
Solution: Ill only explain c). Apply the division algorithm to 123 and 277 to get
277 = 123 2 + 31
We then know that gcd(277, 123) = gcd(123, 31). At this time, apply the division algorithm to 123 and 31:
123 = 31 3 30
This tells us gcd(123, 31) = gcd(31, 30). To sum up,
gcd(277, 123) = gcd(123, 31) = gcd(31, 30) = 1
In other words, 123 and 277 are relatively prime. For the other parts,
gcd(1, 5) = 5,

gcd(1529, 14309) = 139

gcd(100, 101) = gcd(1529, 14038) = gcd(11111, 111111) = 1

4.3.40 Express the greatest common divisor of each of these pairs of integers as a linear combination of
these integers.
a) 9, 11
b) 33, 44
c) 35, 78
d) 21, 55
e) 101, 203
f) 124, 323
4

g) 2002, 2339
h) 3457, 4669
i) 10001, 13422
Solution: Ill only explain part c). Apply division algorithm to 78 by dividing by 35.
78 = 35 2 + 8

(1)
Then iterate with 35 and 8.
(2)

35 = 8 4 + 3

And then with 8 and 3.

(3)

8=32+2

We now know that gcd(78, 35) = gcd(3, 2) = 2. Lets go backward like salmon swimming upstream. From
the last equation (3), we have
2=832
Replace the 3 by 35 8 4 using the equation (2). So
2 = 8 3 2 = 8 (35 8 4) 2 = 9 8 2 35
Replace the 8 by 78 35 2 using the equation (1). Then you have
2 = 9 8 2 35 = 9 (78 35 2) 2 35 = 9 78 20 35
For other parts, do the same calculation. Remember that when you express gcd(a, b) as a linear combinations
of a and b, the choices of coefficients are not unique. For example, gcd(2, 4) = 2 and we have two different
expressions:
2=3214=7234

4.3.52 Prove or disprove that p1 p2 pn + 1 is prime for every positive integer n, where p1 , p2 , , pn are
the n smallest prime numbers.
Solution: This is not true. For n = 4, 2 3 5 7 + 1 = 201 is not prime because 201 = 3 67.

4.3.54 Prove that there are infinitely many primes of the form 3k + 2, where k is a nonnegative integer.
Solution: Assume that there are only finitely many primes of the form 3k + 2. Say
q1 , q2 , . . . , qn
Consider the number N := 3q1 q2 qn 1 and its prime factorization
N = p1 p2 pn
where primes pi s are not necessarily distinct. I claim that there exists at least one prime factor of the form
3k + 2. First of all, there are no i such that pi 3 0. Indeed, if pi 3 0 for some i, then N 3 0 as well. But
then we also have N 3 = 1 from the definition of N . This leads to a contradiction that 0 3 1. So far,
we know that pi is 1 ore 2 modulo 3. Now assume pi 3 1 for all 1 i . Then we get N 3 1 from the
definition and N 3 1 from the prime factorization. Again, this is a contradiction. Therefore, there exists
some i such that pi 3 2. With that pi , it is prime of the form 3k + 2 but is different from any of qj s. This
is because pi divides N but none of qj does. So were done.