Maths Olympiad Prep

Library / USA IMO /10 of 29

USA IMO

Number theory Difficulty 8.2 Shortlist Prove it United States

Let a>b>c>da > b > c > d be positive integers and suppose
ac+bd=(b+d+a−c)(b+d−a+c). ac + bd = (b + d + a - c)(b + d - a + c).
Prove that ab+cdab + cd is not prime.

Solution

First Solution. For the sake of contradiction, assume that ab+cdab + cd is prime. Note that
ab+cd=(a+d)c+(b−c)a=m⋅gcd⁡(a+d,b−c) ab + cd = (a + d)c + (b - c)a = m \cdot \gcd(a + d, b - c)
for some positive integer mm. Writing g=gcd⁡(a+d,b−c)g = \gcd(a+d, b-c), we have
m=a+dg⋅c+b−cg⋅a≥c+a>1. m = \frac{a+d}{g} \cdot c + \frac{b-c}{g} \cdot a \geq c+a > 1.
Therefore, because ab+cdab+cd is prime, g=1g=1.
Substituting ac+bd=(a+d)b−(b−c)aac+bd = (a+d)b - (b-c)a for the left-hand side of the given condition, we obtain
(a+d)b−(b−c)a=(a+d)(b+d−a+c)+(b−c)(b+d−a+c), (a+d)b - (b-c)a = (a+d)(b+d-a+c) + (b-c)(b+d-a+c),
or
(a+d)(a−c−d)=(b−c)(b+c+d). (a+d)(a-c-d) = (b-c)(b+c+d).
Hence, there exists a positive integer kk such that
a−c−d=k(b−c), a-c-d = k(b-c),
b+c+d=k(a+d). b+c+d = k(a+d).
Adding these equations, we obtain a+b=k(a+b−c+d)a+b = k(a+b-c+d) and thus k(c−d)=(k−1)(a+b)k(c-d) = (k-1)(a+b). Recall that a>b>c>d>0a > b > c > d > 0. If k=1k=1, then c=dc=d, a contradiction. If k≥2k \ge 2, then
2≥kk−1=a+bc−d>2bc>2, 2 \ge \frac{k}{k-1} = \frac{a+b}{c-d} > \frac{2b}{c} > 2,
a contradiction.
Therefore, our original assumption was wrong, and ab+cdab+cd is not prime.

Second Solution. (By Yonggao Chen, China) We give a proof by contradiction. Assume that p=ab+cdp = ab+cd is prime. Then ab≡−cd(modp)ab \equiv -cd \pmod p. By (1),
b2(b2+bd+d2)=b2(a2−ac+c2)=(ab)2−ab(bc)+b2c2. b^2(b^2 + bd + d^2) = b^2(a^2 - ac + c^2) = (ab)^2 - ab(bc) + b^2c^2.
It follows that
b2(b2+bd+d2)≡(ab)2−ab(bc)+b2c2≡(cd)2+cd(bc)+b2c2≡c2(b2+bd+d2)(modp), \begin{aligned} b^2(b^2 + bd + d^2) &\equiv (ab)^2 - ab(bc) + b^2c^2 \\ &\equiv (cd)^2 + cd(bc) + b^2c^2 \\ &\equiv c^2(b^2 + bd + d^2) \pmod p, \end{aligned}
implying that p∣(b2−c2)(b2+bd+d2)p \mid (b^2 - c^2)(b^2 + bd + d^2). Observe that 0<b2−c2<b2<ab<p0 < b^2 - c^2 < b^2 < ab < p. Thus, pp and b2−c2b^2 - c^2 must be relatively prime, so
p∣(b2+bd+d2).(2) p \mid (b^2 + bd + d^2). \qquad (2)
Because
0<b2+bd+d2<ab+ab+cd=2ab+cd<2p, 0 < b^2 + bd + d^2 < ab + ab + cd = 2ab + cd < 2p,
we must have b2+bd+d2=p=ab+cdb^2 + bd + d^2 = p = ab + cd or, equivalently,
b(b+d−a)=d(c−d). b(b+d-a) = d(c-d).
Because ab+cdab + cd is prime, bb must be relatively prime to dd, so b∣c−db \mid c - d. This is impossible, because 0<c−d<b0 < c - d < b.

Third Solution. (By Zhiqiang Zhang, China) Let x=a−cx = a - c, y=a+cy = a + c, u=b−du = b - d, and v=b+dv = b + d. By the given condition, we have
y2−x2+v2−u2=4(ac+bd)=4[(b+d)+(a−c)][(b+d)−(a−c)]=4(v+x)(v−x)=4(v2−x2), \begin{aligned} y^2 - x^2 + v^2 - u^2 &= 4(ac + bd) \\ &= 4[(b+d) + (a-c)][(b+d) - (a-c)] \\ &= 4(v+x)(v-x) = 4(v^2 - x^2), \end{aligned}
or
y2−u2=3(v2−x2).(3) y^2 - u^2 = 3(v^2 - x^2). \qquad (3)
Let s=a+b+c+ds = a + b + c + d, x1=s−2dx_1 = s - 2d, x2=s−2cx_2 = s - 2c, x3=s−2bx_3 = s - 2b, and x4=s−2ax_4 = s - 2a. Then x1=y+ux_1 = y + u, x2=v+xx_2 = v + x, x3=y−ux_3 = y - u, x4=v−xx_4 = v - x. Because a>b>c>da > b > c > d, x1>x2>x3>x4x_1 > x_2 > x_3 > x_4. Now (3) reads
x1x3=3x2x4.(4) x_1x_3 = 3x_2x_4. \qquad (4)
Because
xu+vy=(a−c)(b−d)+(a+c)(b+d)=2(ab+cd) xu + vy = (a - c)(b - d) + (a + c)(b + d) = 2(ab + cd)
and
x1x2+x3x4=(y+u)(v+x)+(y−u)(v−x)=2(xu+yv), x_1x_2 + x_3x_4 = (y + u)(v + x) + (y - u)(v - x) = 2(xu + yv),
we have
ab+cd=14(x1x2+x3x4).(5) ab + cd = \frac{1}{4}(x_1x_2 + x_3x_4). \qquad (5)
Let g=gcd⁡(x1,x4)g = \gcd(x_1, x_4). It is clear that s≡xi(mod2)s \equiv x_i \pmod 2 for i=1,2,3,4i = 1, 2, 3, 4. We consider the following cases.
(i) s≡1(mod2)s \equiv 1 \pmod 2. First suppose that g=1g = 1. Then by (4), there exists some positive integer kk such that x3=kx4x_3 = kx_4 and, consequently, kx1=3x2kx_1 = 3x_2. Because x3>x4x_3 > x_4, k>1k > 1; because x1>x2x_1 > x_2, k<3k < 3. Therefore, k=2k = 2. But then x3x_3 is even, contradicting the assumption that each xix_i is odd.
It follows that g>1g > 1, and that gg divides x1x2+x3x4=4(ab+cd)x_1x_2+x_3x_4 = 4(ab+cd).
Because x1x_1 and x4x_4 are odd, gg is odd as well, implying that g∣(ab+cd)g \mid (ab+cd). Also observe that x1x2+x3x4≥3x1+2x4≥3g+2g=5gx_1x_2+x_3x_4 \ge 3x_1+2x_4 \ge 3g+2g = 5g, so that g<ab+cdg < ab+cd. Therefore, ab+cdab+cd is divisible by a number strictly between 1 and ab+cdab+cd, implying that it is composite.
(ii) s≡xi≡0(mod2)s \equiv x_i \equiv 0 \pmod{2}. Let xi′=xi/2x_i' = x_i/2 for i=1,2,3,4i = 1, 2, 3, 4, and let g′=gcd⁡(x1′,x2′)g' = \gcd(x_1', x_2'). Then g=2g′≥2g = 2g' \ge 2 and x1′>x2′>x3′>x4′x_1' > x_2' > x_3' > x_4'. Note also that (4) and (5) become
x1′x3′=3x2′x4′andab+cd=x1′x2′+x3x4′,(4′) x_1'x_3' = 3x_2'x_4' \quad \text{and} \quad ab + cd = x_1'x_2' + x_3x_4', \quad (4')
respectively.
If g′>1g' > 1, then g′∣(ab+cd)g' \mid (ab + cd). Since ab+cd>x2′≥g′ab + cd > x_2' \ge g', ab+cdab + cd must be composite.
If g′=1g' = 1, then by (4'), x3′=kx4′x_3' = kx_4' and kx1′=3x2′kx_1' = 3x_2' for some positive integer kk. Then again by x3′>x4′x_3' > x_4' and x1′>x2′x_1' > x_2', k=2k = 2. Hence 2 divides both x2′x_2' and x3′x_3' implying that ab+cdab+cd is even (by (4')). Since ab+cd>a>2ab+cd > a > 2, ab+cdab+cd is composite.
From the above arguments, we conclude that ab+cdab + cd is not prime.

Fourth Solution. (By Andrei Vorobiev, Russia) Let x=b+d+a−cx = b+d+a-c. It is clear that x>1x > 1. We have c≡a+b+d(modx)c \equiv a+b+d \pmod x and d≡c−a−b(modx)d \equiv c-a-b \pmod x. These congruences, combined with the given condition, yield
0≡ac+bd≡a(a+b+d)+bd≡(a+b)(a+d)(modx) \begin{aligned} 0 &\equiv ac + bd \equiv a(a + b + d) + bd \\ &\equiv (a + b)(a + d) \pmod x \end{aligned}
and
0≡ac+bd≡ac+b(c−a−b)≡(a+b)(c−b)(modx). \begin{aligned} 0 &\equiv ac + bd \equiv ac + b(c - a - b) \\ &\equiv (a + b)(c - b) \pmod x. \end{aligned}
Hence, x∣(a+b)(a+d)x \mid (a+b)(a+d) and x∣(a+b)(c−b)x \mid (a+b)(c-b).
Because a+b>(a+b)−(c−d)=xa+b > (a+b) - (c-d) = x and 2x=2[a+(b−c)+d]>2a>a+b2x = 2[a + (b-c) + d] > 2a > a+b, a+ba+b is not divisible by xx. Thus, there is a prime pp that divides each of xx, (a+d)(a+d), and (c−b)(c-b). To finish, we only need to prove that pp is a proper divisor of ab+cdab+cd. In fact, ab+cd>a+d≥pab+cd > a+d \ge p and
p∣(a+d)b+(c−b)d=ab+cd, p \mid (a+d)b + (c-b)d = ab + cd,
as desired.

Fifth Solution. Let ABCDABCD be the quadrilateral with AB=a,BC=d,CD=b,AD=c,∠BAD=60∘AB = a, BC = d, CD = b, AD = c, \angle BAD = 60^\circ, and ∠BCD=120∘\angle BCD = 120^\circ. Such a quadrilateral exists in view of (1) and the Law of Cosines; the common value in (1) is BD2BD^2. Let ∠ABC=α\angle ABC = \alpha, so that ∠CDA=180∘−α\angle CDA = 180^\circ - \alpha. Applying the Law of Cosines to triangles ABCABC and ACDACD gives
a2+d2−2adcos⁡α=AC2=b2+c2+2bccos⁡α. a^2 + d^2 - 2ad \cos \alpha = AC^2 = b^2 + c^2 + 2bc \cos \alpha.
Hence, 2cos⁡α=(a2+d2−b2−c2)/(ad+bc)2 \cos \alpha = (a^2 + d^2 - b^2 - c^2)/(ad + bc), and
AC2=a2+d2−ada2+d2−b2−c2ad+bc=(ab+cd)(ac+bd)ad+bc. AC^2 = a^2 + d^2 - ad \frac{a^2 + d^2 - b^2 - c^2}{ad + bc} = \frac{(ab + cd)(ac + bd)}{ad + bc}.
Because ABCDABCD is cyclic, the Ptolemy's Theorem yields
(AC⋅BD)2=(AB⋅CD+AD⋅BD)2=(ab+cd)2 (AC \cdot BD)^2 = (AB \cdot CD + AD \cdot BD)^2 = (ab + cd)^2
It follows that
(ac+bd)(a2−ac+c2)=(ab+cd)(ad+bc).(6) (ac + bd)(a^2 - ac + c^2) = (ab + cd)(ad + bc). \quad (6)
(Note that straightforward algebra can also be used to obtain (6) from (1).) Observe that
ab+cd>ac+bd>ad+bc.(7) ab + cd > ac + bd > ad + bc. \quad (7)
The first inequality follows from (a−d)(b−c)>0(a-d)(b-c) > 0, and the second from (a−b)(c−d)>0(a-b)(c-d) > 0.
Now assume that ab+cdab+cd is prime. It then follows from (7) that ab+cdab+cd and ac+bdac+bd are relatively prime. Hence, from (6), it must be true that ac+bdac+bd divides ad+bcad+bc. However, this is impossible by (7). Thus, ab+cdab+cd must not be prime.

Sixth Solution. (By Reid Barton and Gabriel Carroll) Let ω=e2πi3\omega = e^{\frac{2\pi i}{3}}. Then
ω3=1and1+ω+ω2=0.(8) \omega^3 = 1 \quad \text{and} \quad 1 + \omega + \omega^2 = 0. \qquad (8)
We are going to use two fundamental facts about the ring Z[ω]\mathbb{Z}[\omega]:
* Fact 1. Z[ω]\mathbb{Z}[\omega] is a unique factorization domain (UFD);
* Fact 2. the units in Z[ω]\mathbb{Z}[\omega] are ±1,±ω,±ω2\pm 1, \pm\omega, \pm\omega^2.
Factoring (1) in Z[ω]\mathbb{Z}[\omega] gives
(c+ωa)(c+ω2a)=(b−ωd)(b−ω2d).(9) (c + \omega a)(c + \omega^2 a) = (b - \omega d)(b - \omega^2 d). \qquad (9)
Lemma 1. If a>b>c>da > b > c > d are positive integers satisfying (9), and ab+cdab + cd is prime, then c+ωac + \omega a and b−ωdb - \omega d are not relatively prime.
Proof. Assume for the sake of contradiction that c+ωac + \omega a and b−ωdb - \omega d are relatively prime. Since complex conjugation is an automorphism of Z[ω]\mathbb{Z}[\omega] sending ω\omega to ω2\omega^2, c+ω2ac + \omega^2 a and b−ω2db - \omega^2 d must also be relatively prime. From the two facts, we conclude that c+ωa=u(b−ω2d)c + \omega a = u(b - \omega^2 d) for some unit u∈{±1,±ω,±ω2}u \in \{\pm 1, \pm \omega, \pm \omega^2\}.
If u=±1u = \pm 1, then c+ωa=±(b−ω2d)=±(b+d)±ωdc + \omega a = \pm (b - \omega^2 d) = \pm (b + d) \pm \omega d (by the second part of (8)), contradicting a≠±da \neq \pm d.
If u=±ωu = \pm \omega, then c+ωa=±ω(b−ω2d)=∓d±ωbc + \omega a = \pm \omega(b - \omega^2 d) = \mp d \pm \omega b (by the first part of (8), contradicting both a≠±ba \neq \pm b and c≠∓dc \neq \mp d).
If u=±ω2u = \pm \omega^2, then c+ωa=±ω2(b−ω2d)=±(ω2b−ωd)=∓b∓(b+d)ωc + \omega a = \pm \omega^2(b - \omega^2 d) = \pm(\omega^2 b - \omega d) = \mp b \mp (b+d)\omega (by (8)), contradicting c≠∓bc \neq \mp b.
In all cases, we reach a contradiction, so c+ωac + \omega a and b−ωdb - \omega d are not relatively prime. ■
Lemma 2. If a>b>c>da > b > c > d are positive integers satisfying (1) and ab+cdab + cd is prime, then ad=cb+cdad = cb + cd.
Proof. Since a,b,c,da, b, c, d satisfy (1), they also satisfy (9), so by Lemma 1, there exists some prime p=q+rω∈Z[ω]p = q + r\omega \in \mathbb{Z}[\omega] such that p∣c+ωap \mid c + \omega a and p∣b−ωdp \mid b - \omega d. Then p‾=q+rω2∣b−ω2d\overline{p} = q + r\omega^2 \mid b - \omega^2 d, so pp‾∣(c+ωa)(b−ω2d)p\overline{p} \mid (c + \omega a)(b - \omega^2 d). Note that
N(p)=pp‾=q2−qr+r2 N(p) = p\overline{p} = q^2 - qr + r^2
and that
(c+ωa)(b−ω2d)=bc+ωab−ω2dc−ω3ad=(−ad+bc+dc)+(ab+cd)ω. (c + \omega a)(b - \omega^2 d) = bc + \omega ab - \omega^2 dc - \omega^3 ad \\ = (-ad + bc + dc) + (ab + cd)\omega.
Therefore, N(p)∣[(−ad+bc+dc)+(ab+cd)ω]N(p) \mid [(-ad + bc + dc) + (ab + cd)\omega]. Since N(p)∈ZN(p) \in \mathbb{Z}, we must have N(p)∣(−ad+bc+dc)N(p) \mid (-ad + bc + dc) and N(p)∣ab+cdN(p) \mid ab + cd. Since ab+cdab + cd is prime, N(p)=ab+cdN(p) = ab + cd, and so ab+cd∣(−ad+bc+cd)ab + cd \mid (-ad + bc + cd). But
ab+cd−(−ad+bc+dc)=(ab−bc)+ad>0 ab + cd - (-ad + bc + dc) = (ab - bc) + ad > 0
and
ab+cd+(−ad+bc+dc)>ab−ad>0, ab + cd + (-ad + bc + dc) > ab - ad > 0,
so ∣−ad+bc+cd∣<ab+cd|-ad+bc+cd| < ab+cd. Hence, we must have −ad+bc+cd=0-ad+bc+cd = 0, that is, ad=cb+cdad = cb + cd.
Now suppose that a>b>c>d>0a > b > c > d > 0 are integers satisfying (1). Then
(a−c)2+(a−c)c+c2=a2−2ac+c2+ac−c2+c2=a2−ac+c2=b2+bd+d2. \begin{aligned} (a-c)^2 + (a-c)c + c^2 &= a^2 - 2ac + c^2 + ac - c^2 + c^2 \\ &= a^2 - ac + c^2 \\ &= b^2 + bd + d^2. \end{aligned}
Since c>d>0c > d > 0, we must have a−c<ba - c < b implying (a−c)d<bd<bc(a - c)d < bd < bc, or ad<cb+cdad < cb + cd. By Lemma 2, ab+cdab + cd cannot be prime, and we are done.

Want a route through all this instead of an archive? The track puts 2,604 problems in a working order, from Junior Challenge level to the IMO shortlist.

Source: MathNet, licensed CC-BY-4.0. Statement reproduced verbatim; metadata (topic, difficulty) added by this project.