Mathematics 5335 Problem Set 5 Due October 2

Solutions by Duane Broline

• Problem 1, page 96: Reduce the following congruences to equivalent congruences of degree less
than or equal to 6.

(a) x11 + x8 + 5 ≡ 0 (mod 7).

(b) x20 + x13 + x7 +x≡2 (mod 7).
(c) x15 − x10 + 4x − 3 ≡ 0 (mod 7).

Solution:

(a) By long division, x11 + x8 + 5 = (x4 + x)(x7 − x) + x5 + x2 + 5. Thus the original

congruence is equivalent to x5 + x2 + 5 ≡ 0 (mod 7).
(b) It very easy to make errors in long division. However, remember that x7 ≡ x (mod 7)
Hence

x20 + x13 + x7 + x − 2 ≡ (x7 )2 · x6 + x7 · x6 + x7 + x − 2 ≡ x2 + 3x − 2 (mod 7).

(c) Again, replacing x7 by x in this congruence gives that the original congruence is equiv-
alent −x4 + x3 + 4x − 3 (mod 7)

• Problem 2, page 96: Prove that 2x3 + 5x2 + 6x + 1 ≡ 0 (mod 7) has three solutions by the
use of Theorem 2.29.
Solution: By long division

x7 − x = (x3 + 6x2 + 3x + 4)(x4 − 6x3 + 33x2 + 184x + 1029) + 7(−822x2 − 336x − 588).

This was a very poor choice of a problem on my part. Too computationally messy.

• Problem 4, page 96: Prove that if f (x) ≡ 0 (mod p) has j solutions

x ≡ a1 , x ≡ a2 , · · · , x ≡ aj (mod p),

there is a polynomial q(x) such that

f (x) ≡ (x − a1 )(x − a2 ) · · · (x − aj )q(x) (mod p).

(A hint is provided.)
Solution: The proof is by induction on j. If j = 1, there is a root a1 of f (x). By long
division,
f (x) = (x − a1 )q(x) + r
where r is a constant and q(x) is a polynomial. Setting x = a1 and taking the result modulo
p gives
0 ≡ f (a1 ) ≡ 0 · q(a1 ) + r (mod p).
Problem Set 5 Solutions by Duane Broline 2

Hence r ≡ 0 (mod p) and f (x) ≡ (x − a1 )q(x) (mod p). Hence the result is true for j = 1.
Assume that the result is true for j = n − 1. Suppose that f (x) has n roots,

a1 , a2 , . . . , an .

Since f has n − 1 roots, a1 , a2 , . . . , an−1 , the inductive assumption implies that there is a
polynomial k(x) with

f (x) ≡ (x − a1 )(x − a2 ) · · · (x − an−1 )k(x) (mod p).

But an is also a root of f (x)

0 ≡ f (an ) ≡ (an − a1 )(an − a2 ) · · · (an − an−1 )k(an ) (mod p).

Since p is a prime, either

(an − a1 )(an − a2 ) · · · (an − an−1 ) ≡ 0 (mod p)

or
k(an ) ≡ 0 (mod p).
The first cannot occur since the roots are distinct. Hence the second occurs and by the
argument of the first paragraph,

k(x) ≡ (x − an )q(x) (mod p),

for some polynomial q(x). and thus

f (x) ≡ (x − a1 )(x − a2 ) · · · (x − an−1 )(x − an )q(x) (mod p).

• Problem 6, page 97: Let m be composite. Prove that Theorem 2.26 is false if “mod p” is
replaced by “mod m”.
Solution: As a counterexample, the congruence x2 ≡ 1 (mod 8) has 4 solutions modulo
8. These solutions are 1,3,5,7 mod 8.

• Problem 1, page 106: Find a primitive root for the prime 3; for the prime 5; for the prime 7;
the prime 13.
Solution: Except for 7, 2 is a primitive root. This can either be done by showing there are
p − 1 different powers of 2 modulo p, p = 3, 5, 13 or by showing:
for 3: 21 6≡ 1 (mod 3).
for 5: 22 6≡ 1 (mod 3).
for 13: 26 6≡ 1 (mod 13) and 24 6≡ 1 (mod 13)
For 7, 3 is a primitive root (since 32 6≡ 1 (mod 7) and 33 6≡ 1 (mod 7).
Problem Set 5 Solutions by Duane Broline 3

• Problem 3, page 106: How many primitive roots does the prime 13 have? Find them all.
Solution: Since 2 has order 12 modulo 13, 2h where (h, 12) = 1 will also have order 12.
Thus, there are φ(12) = 4 primitive roots modulo 13. These are

2, 25 mod 13 = 6, 27 mod 13 = 11, 211 mod 13 = 6.

• Problem 6, page 106: If a belongs to the exponent h modulo m, prove that no two of

a, a2 , . . . , ah

are congruent modulo m.

Solution: Since a has an exponent mod m, it must be the case that (a, m) = 1. Suppose,
by way of contradiction that 1 ≤ k1 < k2 ≤ h such that ak1 ≡ ak2 (mod m). Therefore,
ak2 −k1 ≡ 1 (mod m). Since a belongs to the exponent h mod m, h|(k2 − k1 ). But this is
impossible since 0 < k2 − k1 < h.

• Problem 8, page 106: Use Theorem 2.37 to determine how many solutions each of the following
congruences has:

(a) x12 ≡ 16 (mod 17)

(b) x48 ≡ 9 (mod 17)
(c) x20 ≡ 13 (mod 17)
(d) x11 ≡ 9 (mod 17)

Solution:

(a) Note that (12, 16) = 4 and 1616/4 ≡ 1 (mod 17). Hence, there are 4 solutions modulo
17.
(b) Note that (48, 16) = 16 and 916/16 6≡ 1 (mod 17). Hence, there are no solutions modulo
17.
(c) Note that (20, 16) = 4 and 1316/4 ≡ 1 (mod 17). Hence, there are 4 solutions modulo
17.
(d) Note that (11, 16) = 1 and 916 ≡ 1 (mod 17). Hence, there is a unique solution modulo
17.

• Problem 10, page 106: Find the sequence of powers of 3 (mod 17). Use this information to
find the solutions of the congruences in Problem 8.
Solution: The powers of 3 are

n 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
3n 31 32 33 34 35 36 37 38 39 310 311 312 313 314 315 316
3n mod 17 3 9 10 13 5 15 11 16 14 8 7 4 12 2 6 1
Problem Set 5 Solutions by Duane Broline 4

(a) x12 ≡ 16 (mod 17) is equivalent to 312i ≡ 38 (mod 17), by setting x = 3i . Since
3 is a primitive root modulo 17, this congruence is equivalent to 12i ≡ 8 (mod 16).
This last congruence is equivalent to 3i ≡ 2 (mod 4). Hence i = 2, 6, 8, 10 mod 16 and
x = 9, 15, 16, 4 mod 17.
(b) x48 ≡ 9 (mod 17) is equivalent to 348i ≡ 32 (mod 17), by setting x = 3i . Since 3 is a
primitive root modulo 17, this congruence is equivalent to 48i ≡ 2 (mod 16). This last
congruence is equivalent to 0 ≡ 2 (mod 16) which has no solution.
(c) x20 ≡ 13 (mod 17) is equivalent to 320i ≡ 34 (mod 17), by setting x = 3i . Since 3 is
a primitive root modulo 17, this congruence is equivalent to 20i ≡ 4 (mod 16). This
last congruence is equivalent to 5i ≡ 1 (mod 4). Hence, i = 1, 5, 9, 13 mod 16. and
x = 3, 5, 14, 4 mod 17.
(d) x11 ≡ 9 (mod 17) is equivalent to 311i ≡ 32 (mod 17), by setting x = 3i . Since 3 is a
primitive root modulo 17, this congruence is equivalent to 11i ≡ 2 (mod 16). This last
congruence has a unique solution modulo 16, namely i = 6 mod 16. Hence x = 15 mod
17.

• Problem 11, page 106: Using the data in the preceding problem, decide which of the congru-
ences
x2 ≡ 1, x2 ≡ 2, . . . , x2 ≡ 16 (mod 17),
have solutions.
Solution: x2 ≡ a has as solution if and only if

a ≡ 9, 13, 15, 16, 8, 4, 2, or , 1 (mod 17).

These are the even powers of the primitive root.

• Problem 15, page 106: Prove that if a belongs to the exponent h modulo a prime p, and if h
is even, then ah/2 ≡ −1 (mod p).
Solution: If h is even, then h/2 is an integer. Let b = ah/2 . Since b2 ≡ ah ≡ 1 (mod p),
Lemma 2.10 shows that either b ≡ 1 (mod p) or b ≡ −1 (mod p). Since the first would
imply that ah/2 ≡ 1 (mod p), contrary to the assumption that a has belongs to the exponent
h modulo p, it must be that b ≡ −1 (mod p), as desired.

• Problem 18, page 107: Show that if g and g 0 are primitive roots modulo an odd prime p, then
gg 0 is not a primitive root of p.
Solution: Now g belongs to the exponent p − 1 modulo p. Since p − 1 is even, the previous
problem implies
g (p−1)/2 ≡ −1 (mod p).
Similarly,
(g 0 )(p−1)/2 ≡ −1 (mod p).
Hence,
(gg 0 )(p − 1)/2 ≡ 1 (mod p).
Problem Set 5 Solutions by Duane Broline 5

Thus, the order of gg 0 modulo p divides (p − 1)/2. In particular, gg 0 cannot be a primitive

root modulo p.

• Use the facts that 12 is a primitive root modulo 103 and that 1214 ≡ 2 (mod 103) to find all
solutions to the congruence
x10 ≡ 2 (mod 103).

Solution: Letting x = 12i the given congruence is equivalent to

1210i ≡ 1214 (mod 103).

Since φ(103) = 102, this congruence is equivalent to

10i ≡ 14 (mod 102)

which, in turn, is equivalent to

5i ≡ 7 (mod 51).
It is a simple matter to find the unique solution modulo 51 to this congruence, namely 32
mod 51. Hence the solutions to

10i ≡ 14 (mod 102)

are 32 and 83 mod 102. The solutions to the original congruence are

1232 and 1283 mod 103.

There are 29 and 74 mod 103.