Lecture 5 2 2.3 The Greatest Common Divisor The Divisibility Relation. The Division Algorithm states: If a and b are integers with b > 0, then there exist unique integers q and r such that a = bq + r where 0 ≤ r < b. If r=0, then a = b∙q+0 = bq. We then say that “b divides a” or “a is divisible by b”. 3 2.3 The Greatest Common Divisor The Divisibility Relation. Let a and b be two integers with b≠0. Then a is said to be divisible by b if there exists an integer c such that a = b∙c and we write b|a (b divides a). If b|a, we also say b is a factor of a, or a is a multiple of b. If b does not divide a, then we write 𝑏 ∤ 𝑎. For example, 3|12 since 12 = 3∙4. Also 5|(-30) since -30 = 5∙(- 6) But 6 ∤ 15, since there is no integer q which makes the equation 15 = 6q true. Remark: Since the divisor must always be non-zero, so whenever the notation b|a is employed, it is understood that b is non-zero. 4 Theorem (Divisibility Properties) Let a, b, c, d be integers. Then 1. a|0, 1|a, -1|a, a|a [Reflexive property] Proof: The equation 0 = a∙0 is true for all integers a, which implies a|0. Similarly, the equations a = 1∙a implies 1|a and and a = (-1)∙(-a) implies -1|a. Finally, a = a∙1 implies a|a. 5 2. a|1 if and only if a = ±1. Proof: Let a|1 then 1= a∙q for some q ∈ ℤ. If a>0, then q>0 so the equation is true only when a=q=1. If a<0, then q<0 so the equation is true only when a=q=-1. Thus, either a=1 or a= -1.
Conversely, if a = ±1, then the equation
1 = (±1)∙(±1) implies ±1|1 and so a|1. 6 3. If a|b and c|d then ac|bd Proof: Suppose a|b and c|d. The there exist integers q1, q2 such that b = a∙q1 and d = c∙q2 Now bd = (a∙q1)(c∙q2) = (a∙c)(q1∙q2) where q1∙q2 ∈ ℤ which implies ac|bd. 7 4. If a|b and b|c then a|c [Transitive property] Proof: Suppose a|b and b|c. Then there exist integers t1, t2 such that b = a∙t1 and c = b∙t2. Now c = b∙t2 = (a∙t1)∙t2 = a(t1∙t2) where t1∙t2 ∈ ℤ This implies a|c. 8 5. a|b and b|a if and only if a = ±b. Proof: Suppose a|b and b|a. Then by convention both a and b are non-zero integers. Hence, there exist non-zero integers s1, s2 such that b=a∙s1 and a=b∙s2. Now a = b∙s2 = (a∙s1)∙s2 = a(s1∙s2) Cancelling a ≠0 on both sides, we have 1 = s1∙s2 This is possible if and only if s1 = s2 = 1 or s1 = s2 = -1. This in turn implies a = b or a = -b. That is a = ±b. 9 6. If a|b and b≠0, then |a| ≤ |b|. Proof: If a|b, then there exists an integer q such that b = a∙q. By convention a≠0. Also given that b≠0. Hence the above equation implies that q≠0, whence |q| ≥ 1. Taking absolute value of the equation b = a∙q we have |b| = |a∙q| = |a|∙|q| ≥ |a|∙1 = |a| That is |a| ≤ |b|. 10 7. If a|b and a|c then a|(bx+cy) for arbitrary integers x and y. In general, if a|bk for k=1, 2, …, n then a|(b1x1+ b2x2+…+ bnxn) for all integers x1, x2,…,xn. Proof: Suppose a|b and a|c. The there exist integers q1, q2 such that b = a∙q1 and c = a∙q. Now bx+cy = (a∙q1)x + (a∙q2)y = a(q1x+q2y) for all x, y ∈ ℤ Thus a|(bx+cy), as desired. 11 The general result can be proved using induction. Basis Step: For n=2, the statement is: If a|bk for k=1, 2, then a|(b1x1+ b2x2) for all integers x1, x2. This is true as proved. Inductive Step: Suppose the result is true for n=m. That is If a|bk for k=1, 2, …, m, then a|(b1x1+ b2x2+…+ bmxm) for all xi ∈ ℤ. Suppose further a|bm+1. Then a|[(b1x1+ b2x2+…+ bmxm)∙1 + bm+1 xm+1] for any xm+1∈ ℤ. That is a|(b1x1+ b2x2+…+ bmxm + bm+1xm+1) which verifies the inductive step and completes the proof. 12 Problems 2.3 Q2. Given integers a, b, c, verify that i. if a|b, then a|bc; Solution: Suppose a|b. Then there exists an integer q such that b = a∙q. Multiplying both sides by c we get bc = a∙(qc) Since c, q ∈ ℤ so qc ∈ ℤ. Hence above equation implies a|bc. 13 ii. if a|b and a|c, then a2|bc; Solution: Suppose a|b and a|c. Then there exist integers q1, q2 such that b = a∙q1 and c = a∙q2 Thus bc = (a∙q1)∙(a∙q2) = a2(q1q2) where q1q2 ∈ ℤ. This shows that a2|bc. 14 iii. a|b if and only if ac|bc where c≠0. Solution: Suppose a|b. Then b = a∙q for some q ∈ ℤ. Multiplying both sides by c≠0 we get bc = (ac)∙q This implies ac|bc. Conversely, assume ac|bc. Then there exists an integer q such that bc = (ac)∙q Since c≠0, we cancel c on both sides to get b = a∙q which implies a|b. 15 Q3. Prove or disprove: if a|(b+c), then either a|b or a|c. Solution: The statement is not true, since 3|(2+4) but neither 3|2 nor 3|4. Or 2|(3+5) but neither 2|3 nor 2|5. Create your own counter examples. 16 Q4. For n ≥ 1, use mathematical induction to establish each of the following divisibility statements: i. 8|52n + 7 Solution: By definition of divisibility, 8|52n + 7 is equivalent to 52n + 7 = 8∙(an int); Let P(n): 52n + 7 = 8∙(an integer); n ≥ 1 Basis Step LHS of P(1) = 52(1) + 7 = 25 + 7 = 32 = 8∙4 which confirms RHS form. Hence P(1) is true. 17 Inductive Step Suppose P(k): 52k + 7 = 8∙(an integer) is true for some integer k ≥ 1. We show that P(k+1): 52(k+1) + 7 = 8∙(an integer) is also true. Now LHS of P(k+1) = 52(k+1) + 7 = 52k ∙52 + 7 = 52k ∙(1+24) + 7 = 52k + 7 + 24∙52k = 8∙(an integer) + 24∙52k = 8∙(an integer + 3∙52k) = 8∙(an integer) = RHS Hence induction is complete and so the statement is true for all integers n ≥1. 18 Q4 v. 24|2∙7n + 3∙5n – 5 Solution: By definition of divisibility, 24|2∙7n + 3∙5n – 5 is equivalent to 2∙7n + 3∙5n – 5 = 24∙(an integer); Let P(n): 2∙7n + 3∙5n – 5 = 24∙(an integer); n ≥ 1 Basis Step LHS of P(1) = 2∙71 + 3∙51 – 5 = 14 + 15 – 5 = 24 = 24∙1 which is in confirmation with the RHS form. Hence P(1) is true. 19 Inductive Step Suppose P(k): 2∙7k + 3∙5k – 5 = 24∙(an integer) is true for k ≥1. We show that P(k+1): 2∙7k+1 + 3∙5k+1 – 5 = 24∙(an integer) is also true. Now LHS of P(k+1) = 2∙7k+1 + 3∙5k+1 – 5 = (2∙7k)∙7 + (3∙5k)∙5 – 5 = (2∙7k)∙(1+6) + (3∙5k)∙(1+4) – 5 = (2∙7k + 3∙5k – 5) + 12∙7k + 12∙5k = (24∙an integer) + 12∙(7k + 5k) Since 5 and 7 are odd, so any positive integer power of these i.e., 5k and 7k will also be odd. Consequently, 7k + 5k is even say 2m, m ∈ ℤ. 20 ∴ LHS of P(k+1) = (24∙an integer) + 12∙(7k + 5k) = (24∙an integer) + 12∙2m = 24∙(an integer + m) = 24∙(an integer) = RHS of P(k+1) Hence induction is complete and the statement is true for all integers n ≥1. 21 Q8. Prove the following: i. The sum of the squares of two odd integers cannot be a perfect square. Solution: We know that the square of an integer is of the form 4k or 4k+1. Let a = 2m+1 and b = 2n+1, where m, n ∈ ℤ, be odd integers. Now a2 + b2 = (2m+1)2 + (2n+1)2 = (4m2+4m+1) + (4n2+4n+1) = 4(m2+n2+m+n) + 2 = 4k+2, which cannot be a perfect square. 22 Q8. ii The product of four consecutive integers is 1 less than a perfect square. Solution: Let the four consecutive integers be n, n+1, n+2, n+3. We consider n(n+1)(n+2)(n+3) = n(n+3)(n+1)(n+2) = (n2+3n)(n2+3n+2) = (m)(m+2) where n2+3n = m = m2+2m +1 – 1 = (m+1)2 – 1 = (n2+3n+1)2 – 1 = A perfect square minus 1. Practice Questions Problems 2.3 Q1-9