1989 AIME Problems

Scroll down and press Start to try the exam! Or, go to the printable PDF, answer key, or professional solutions curated by LIVE by Po-Shen Loh.

All problems are used with official legal permission of the Mathematical Association of America (MAA).

Or jump straight to a single problem with its solution: 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15

Want to learn professionally through interactive video classes?

Learn LIVE

Timed

3:00:00

1.

Compute (31)(30)(29)(28)+1.\sqrt{(31)(30)(29)(28)+1}.

Answer: 869
Concepts:difference of squaresfactoringradical
Difficulty rating: 1650
Small Hint:

Pair the outer factors and the inner factors

Big Hint:

The two paired products are consecutive even integers

Solution:

We have 3128=86831\cdot28=868 and 3029=870.30\cdot29=870. These products lie one below and one above 869,869, so their product is 86921.869^2-1. After the final 11 is added, the radicand is 8692.869^2. Its positive square root is 869.869.

2.

Ten points are marked on a circle. How many distinct convex polygons of three or more sides can be drawn using some (or all) of the ten points as vertices?

Answer: 968
Difficulty rating: 1780
Small Hint:

Each choice of at least three marked points determines one convex polygon

Big Hint:

Count all subsets and remove those of sizes 0,0, 1,1, and 22

Solution:

Every subset of at least three points determines exactly one convex polygon. There are 210=10242^{10}=1024 subsets in all. Of these, (100)=1,\binom{10}{0}=1, (101)=10,\binom{10}{1}=10, and (102)=45\binom{10}{2}=45 have fewer than three points. Therefore the required number is 102411045=968.1024-1-10-45=968.

3.

Suppose nn is a positive integer and dd is a single digit in base 10.10. Find nn if n810=0.d25d25d25.\frac{n}{810}=0.d25d25d25\ldots.

Answer: 750
Difficulty rating: 1970
Small Hint:

Convert the repeating three-digit block into a fraction with denominator 999999

Big Hint:

Use the integrality of nn to determine which digit makes 4d+14d+1 divisible by 3737

Solution:

The repeating decimal is 100d+25999.\frac{100d+25}{999}. Thus n=810(100d+25)999=750(4d+1)37.\begin{aligned}n&=\frac{810(100d+25)}{999}\\&=\frac{750(4d+1)}{37}.\end{aligned} Since 0d9,0\leq d\leq9, the only way 4d+14d+1 can be divisible by 3737 is 4d+1=37,4d+1=37, so d=9d=9 and n=750.n=750.

4.

If a<b<c<d<ea<b<c<d<e are consecutive positive integers such that b+c+db+c+d is a perfect square and a+b+c+d+ea+b+c+d+e is a perfect cube, what is the smallest possible value of c?c?

Answer: 675
Difficulty rating: 2190
Small Hint:

Express both sums in terms of the middle integer cc

Big Hint:

Compare the prime exponents in 3c3c and 5c5c modulo 22 and modulo 33

Solution:

The two sums are 3c3c and 5c.5c. If c=pvp,c=\prod p^{v_p}, then 3c3c being a square and 5c5c being a cube impose conditions on every exponent. For p=3,p=3, the smallest exponent in cc that is odd and divisible by 33 is 3.3. For p=5,p=5, the smallest exponent that is even and congruent to 2(mod3)2\pmod3 is 2.2. Every other prime exponent can be 0.0. Thus the least possible value is c=3352=675.c=3^3\cdot5^2=675.

5.

When a certain biased coin is flipped five times, the probability of getting heads exactly once is not equal to 00 and is the same as that of getting heads exactly twice. Let ij,\frac{i}{j}, in lowest terms, be the probability that the coin comes up heads in exactly 33 out of 55 flips. Find i+j.i+j.

Answer: 283
Difficulty rating: 2110
Small Hint:

Let pp be the probability of heads and equate the two binomial probabilities

Big Hint:

Cancel the nonzero common factors before solving for pp

Solution:

Let pp be the probability of heads. The condition gives 5p(1p)4=10p2(1p)3.5p(1-p)^4=10p^2(1-p)^3. The stated nonzero condition permits cancellation, yielding 1p=2p,1-p=2p, so p=13.p=\frac{1}{3}. The probability of exactly three heads is (53)(13)3(23)2=40243.\binom53\left(\frac13\right)^3\left(\frac23\right)^2=\frac{40}{243}. Therefore i+j=40+243=283.i+j=40+243=283.

6.

Two skaters, Allie and Billie, are at points AA and B,B, respectively, on a flat, frozen lake. The distance between AA and BB is 100100 meters. Allie leaves AA and skates at a speed of 88 meters per second on a straight line that makes a 6060^\circ angle with AB.AB. At the same time Allie leaves A,A, Billie leaves BB at a speed of 77 meters per second and follows the straight path that produces the earliest possible meeting of the two skaters, given their speeds. How many meters does Allie skate before meeting Billie?

Answer: 160
Difficulty rating: 2270
Small Hint:

After tt seconds, Allie is 8t8t meters from AA and Billie can be 7t7t meters from BB

Big Hint:

Apply the Law of Cosines to the triangle and select the smaller positive time

Solution:

Suppose the skaters meet after tt seconds. Their distances from AA and BB are 8t8t and 7t,7t, and the included angle at AA is 60.60^\circ. The Law of Cosines gives (7t)2=(8t)2+10022(8t)(100)cos60.\begin{aligned}(7t)^2&=(8t)^2+100^2\\&\quad-2(8t)(100)\cos60^\circ.\end{aligned} Hence 3t2160t+2000=0,3t^2-160t+2000=0, whose roots are 2020 and 1003.\frac{100}{3}. The earliest meeting occurs at t=20,t=20, so Allie skates 8(20)=1608(20)=160 meters.

7.

If the integer kk is added to each of the numbers 36,36, 300,300, and 596,596, one obtains the squares of three consecutive terms of an arithmetic series. Find k.k.

Answer: 925
Difficulty rating: 2250
Small Hint:

Write the three arithmetic-sequence terms as x,x, x+r,x+r, and x+2rx+2r

Big Hint:

Subtract adjacent square equations, then subtract those two resulting equations

Solution:

Let the three terms be x,x, x+r,x+r, and x+2r.x+2r. Subtracting the square equations gives r(2x+r)=30036=264r(2x+r)=300-36=264 and r(2x+3r)=596300=296.r(2x+3r)=596-300=296. Their difference is 2r2=32,2r^2=32, so r=±4.r=\pm4. Negating all three terms does not change their squares, so take r=4.r=4. Then 4(2x+4)=264,4(2x+4)=264, giving x=31.x=31. Therefore k=x236=96136=925.k=x^2-36=961-36=925.

8.

Assume that x1,x_1, x2,x_2, ,\ldots, x7x_7 are real numbers such that

x1+4x2+9x3+16x4+25x5+36x6+49x7=1,4x1+9x2+16x3+25x4+36x5+49x6+64x7=12,9x1+16x2+25x3+36x4+49x5+64x6+81x7=123.\begin{aligned}x_1+4x_2+9x_3&\\+16x_4+25x_5&\\+36x_6+49x_7&=1,\\4x_1+9x_2+16x_3&\\+25x_4+36x_5&\\+49x_6+64x_7&=12,\\9x_1+16x_2+25x_3&\\+36x_4+49x_5&\\+64x_6+81x_7&=123.\end{aligned}

Find the value of

16x1+25x2+36x3+49x4+64x5+81x6+100x7.\begin{aligned}16x_1+25x_2+36x_3&\\+49x_4+64x_5&\\+81x_6+100x_7.&\end{aligned}

Answer: 334
Difficulty rating: 2230
Small Hint:

Let S(t)=i=17(i+t)2xiS(t)=\sum_{i=1}^7(i+t)^2x_i

Big Hint:

Because S(t)S(t) is quadratic in t,t, its second finite differences are constant

Solution:

Define S(t)=i=17(i+t)2xi.S(t)=\sum_{i=1}^7(i+t)^2x_i. This is a quadratic polynomial in t,t, and the equations say S(0)=1,S(0)=1, S(1)=12,S(1)=12, and S(2)=123.S(2)=123. Its first two differences are 1111 and 111,111, so the constant second difference is 100.100. The next first difference is therefore 211,211, giving S(3)=123+211=334.S(3)=123+211=334.

9.

One of Euler’s conjectures was disproved in the 19601960s by three American mathematicians when they showed there was a positive integer nn such that 1335+1105+845+275=n5.133^5+110^5+84^5+27^5=n^5. Find the value of n.n.

Answer: 144
Difficulty rating: 2090
Small Hint:

Estimate the fifth root to narrow the possible integer values of nn

Big Hint:

Evaluate the fifth powers by repeated squaring and multiplication, then compare their sum with the nearby candidate

Solution:

Direct integer arithmetic gives 1335=41,615,795,893,1105=16,105,100,000,845=4,182,119,424,275=14,348,907.\begin{aligned}133^5&=41{,}615{,}795{,}893,\\110^5&=16{,}105{,}100{,}000,\\84^5&=4{,}182{,}119{,}424,\\27^5&=14{,}348{,}907.\end{aligned} Their sum is 61,917,364,224.61{,}917{,}364{,}224. Repeated multiplication also gives 1445=61,917,364,224,144^5=61{,}917{,}364{,}224, so the positive integer nn is 144.144.

10.

Let a,a, b,b, cc be the three sides of a triangle, and let α,\alpha, β,\beta, γ\gamma be the angles opposite them. If a2+b2=1989c2,a^2+b^2=1989c^2, find cotγcotα+cotβ.\frac{\cot\gamma}{\cot\alpha+\cot\beta}.

Answer: 994
Difficulty rating: 2530
Small Hint:

Simplify cotα+cotβ\cot\alpha+\cot\beta using α+β=πγ\alpha+\beta=\pi-\gamma

Big Hint:

Use the Law of Sines for the resulting sine factors and the Law of Cosines for cosγ\cos\gamma

Solution:

First, cotα+cotβ=sin(α+β)sinαsinβ=sinγsinαsinβ.\begin{aligned}\cot\alpha+\cot\beta&=\frac{\sin(\alpha+\beta)}{\sin\alpha\sin\beta}\\&=\frac{\sin\gamma}{\sin\alpha\sin\beta}.\end{aligned} Hence the desired ratio is cosγsinαsinβsin2γ=abcosγc2,\frac{\cos\gamma\sin\alpha\sin\beta}{\sin^2\gamma}=\frac{ab\cos\gamma}{c^2}, where the Law of Sines was used in the final equality. By the Law of Cosines, 2abcosγ=a2+b2c2=1988c2.\begin{aligned}2ab\cos\gamma&=a^2+b^2-c^2\\&=1988c^2.\end{aligned} Therefore the ratio is 19882=994.\frac{1988}{2}=994.

11.

A sample of 121121 integers is given, each between 11 and 10001000 inclusive, with repetitions allowed. The sample has a unique mode (most frequent value). Let DD be the difference between the mode and the arithmetic mean of the sample. What is the largest possible value of D?\lfloor D\rfloor? (For real x,x, x\lfloor x\rfloor is the greatest integer less than or equal to x.x.)

Answer: 947
Difficulty rating: 3270
Small Hint:

By symmetry, place the mode at the low endpoint and push every other entry as high as the frequency restriction allows

Big Hint:

If the mode occurs ff times, every other value may occur at most f1f-1 times; optimize separately over ff

Solution:

By reflecting every value xx to 1001x,1001-x, it suffices to maximize the mean minus the mode. For a fixed modal frequency f,f, the extremal sample has ff copies of 1,1, then fills the largest available integers with at most f1f-1 copies each.

Put 121f=q(f1)+r,121-f=q(f-1)+r, where 0r<f1.0\leq r<f-1. The nonmodal entries are f1f-1 copies of each of 1000,1000, 999,999, ,\ldots, 1001q,1001-q, followed by rr copies of 1000q.1000-q. For f=2,f=2, f=3,f=3, f=4,f=4, f=5,f=5, and f=6,f=6, this formula gives floors 924,924, 945,945, 947,947, 944,944, and 939,939, respectively. If f7,f\geq7, there are at most 114114 nonmodal terms, so even the weaker bound D114(999)121<942D\leq\frac{114(999)}{121}<942 suffices. Thus the maximum occurs at f=4.f=4.

The extremal sample contains four 11’s and three copies of every integer from 962962 through 1000.1000. Put T=962+963++1000.T=962+963+\cdots+1000. Since T=38259,T=38259, D=3T+41211=114660121=947+73121.\begin{aligned}D&=\frac{3T+4}{121}-1\\&=\frac{114660}{121}\\&=947+\frac{73}{121}.\end{aligned} Therefore the largest possible floor is 947.947.

12.

Let ABCDABCD be a tetrahedron with AB=41,AB=41, AC=7,AC=7, AD=18,AD=18, BC=36,BC=36, BD=27,BD=27, and CD=13,CD=13, as shown in the figure. Let dd be the distance between the midpoints of edges ABAB and CD.CD. Find d2.d^2.

Answer: 137
Difficulty rating: 2560
Small Hint:

Represent the vertices by vectors and write the vector between the two midpoints

Big Hint:

Expand A+BCD2\lVert A+B-C-D\rVert^2 in terms of the six edge lengths

Solution:

Let the vertex names also denote their position vectors. The vector between the midpoints is A+BCD2.\frac{A+B-C-D}{2}. Expanding squared lengths gives 4d2=AC2+AD2+BC2+BD2AB2CD2.\begin{aligned}4d^2={}&AC^2+AD^2\\&+BC^2+BD^2\\&-AB^2-CD^2.\end{aligned} Therefore 4d2=72+182+362+272412132=548,\begin{aligned}4d^2={}&7^2+18^2+36^2+27^2\\&-41^2-13^2=548,\end{aligned} so d2=137.d^2=137.

13.

Let SS be a subset of {1,2,3,,1989}\{1,2,3,\ldots,1989\} such that no two members of SS differ by 44 or 7.7. What is the largest number of elements SS can have?

Answer: 905
Difficulty rating: 2930
Small Hint:

Within any eleven consecutive integers, the forbidden-difference graph is an 1111-cycle

Big Hint:

For a matching construction, look for five allowable residue classes modulo 1111

Solution:

On any eleven consecutive integers, join two numbers when their difference is 44 or 7.7. Because 7=114,7=11-4, this graph is an 1111-cycle, whose largest independent set has size 5.5. Similarly, any ten consecutive integers induce a path on ten vertices and contribute at most 5.5. Since 1989=17911+210,1989=179\cdot11+2\cdot10, this gives S179(5)+2(5)=905.|S|\leq179(5)+2(5)=905.

This bound is attained by taking every integer whose residue modulo 1111 is 1,1, 3,3, 4,4, 6,6, or 9.9. No two selected residues differ by 44 or 77 modulo 11,11, and each of these five residues occurs 181181 times from 11 through 1989.1989. Thus the maximum is 5(181)=905.5(181)=905.

14.

Given a positive integer n,n, it can be shown that every complex number of the form r+si,r+si, where rr and ss are integers, can be uniquely expressed in the base n+i-n+i using the integers 0,0, 1,1, ,\ldots, n2n^2 as digits. That is, the equation

r+si=am(n+i)m+am1(n+i)m1++a1(n+i)+a0\begin{aligned}r+si={}&a_m(-n+i)^m\\&+a_{m-1}(-n+i)^{m-1}\\&+\cdots+a_1(-n+i)\\&+a_0\end{aligned}

is true for a unique choice of nonnegative integer mm and digits a0,a_0, a1,a_1, ,\ldots, ama_m chosen from the set {0,1,2,,n2},\{0,1,2,\ldots,n^2\}, with am0.a_m\ne0. We write

r+si=(amam1a1a0)n+ir+si=(a_ma_{m-1}\ldots a_1a_0)_{-n+i}

to denote the base n+i-n+i expansion of r+si.r+si. There are only finitely many integers k+0ik+0i that have four-digit expansions

k=(a3a2a1a0)3+ia30.k=(a_3a_2a_1a_0)_{-3+i}\qquad a_3\ne0.

Find the sum of all such k.k.

Answer: 490
Difficulty rating: 2840
Small Hint:

Compute the second and third powers of 3+i-3+i and set the imaginary part of the expansion equal to zero

Big Hint:

The digit bounds leave only two possible triples (a3,a2,a1)(a_3,a_2,a_1); then let a0a_0 range over all digits

Solution:

Let b=3+i.b=-3+i. Then b2=86ib^2=8-6i and b3=18+26i.b^3=-18+26i.

The imaginary part of a3b3+a2b2+a1b+a0a_3b^3+a_2b^2+a_1b+a_0 is 26a36a2+a1.26a_3-6a_2+a_1. Thus a1=6a226a3.a_1=6a_2-26a_3. With 1a391\leq a_3\leq9 and 0a1,a29,0\leq a_1,a_2\leq9, the only possibilities are (a3,a2,a1)=(1,5,4),(a3,a2,a1)=(2,9,2).\begin{gathered}(a_3,a_2,a_1)=(1,5,4),\\(a_3,a_2,a_1)=(2,9,2).\end{gathered}

The corresponding real parts are 10+a010+a_0 and 30+a0,30+a_0, respectively. As a0a_0 ranges from 00 through 9,9, the required sum is (10+11++19)+(30+31++39)=145+345=490.\begin{aligned}&(10+11+\cdots+19)\\&\quad+(30+31+\cdots+39)\\&=145+345=490.\end{aligned}

15.

Point PP is inside triangle ABC.ABC. Line segments APD,APD, BPE,BPE, and CPFCPF are drawn with DD on BC,BC, EE on AC,AC, and FF on ABAB (see the figure). Given that AP=6,AP=6, BP=9,BP=9, PD=6,PD=6, PE=3,PE=3, and CF=20,CF=20, find the area of triangle ABC.ABC.

Answer: 108
Difficulty rating: 3060
Small Hint:

Use the two known cevian ratios to find the barycentric weights of AA and BB at PP

Big Hint:

Place PP at the origin; the resulting vector relation determines the angle between PAPA and PBPB

Solution:

Since AP=PD,AP=PD, the barycentric weight of AA at PP is 12.\frac{1}{2}. Since BP:PE=3:1,BP:PE=3:1, the weight of BB is 14,\frac{1}{4}, so the weight of CC is also 14.\frac{1}{4}. Along CF,CF, this means PFCF=14,\frac{PF}{CF}=\frac{1}{4}, hence PF=5PF=5 and CP=15.CP=15.

Place PP at the origin and denote the position vectors A,A, B,B, and CC by the same letters. The barycentric relation is 2A+B+C=0.2A+B+C=0. Thus C=2AB.C=-2A-B. Using A=6,|A|=6, B=9,|B|=9, and C=15,|C|=15, 225=2A+B2=4(36)+81+4AB,\begin{aligned}225&=|2A+B|^2\\&=4(36)+81+4A\mathbin{\cdot}B,\end{aligned} so AB=0.A\mathbin{\cdot}B=0. Therefore PAPB.PA\perp PB. Expanding the cross product gives (BA)×(CA)=4(A×B).(B-A)\mathbin{\times}(C-A)=4(A\mathbin{\times}B). Consequently, [ABC]=2A×B=2(6)(9)=108.\begin{aligned}{}[ABC]&=2|A\mathbin{\times}B|\\&=2(6)(9)=108.\end{aligned}