institute of STEM education

POTW Handout 20 answers

§1 problems
Problem 1.1. Find all real numbers x, y, z so that
x2 y + y 2 z + z 2 = 0
1
z 3 + z 2 y + zy 3 + x2 y = (x4 + y 4 ).
4
answer. Notice we can derive two relations from the first equation, which are:
2 4
(I) : (z + y2 )2 = y4 − x2 y
2 4
(II) : ( x2 + yz)2 = x4 − z 3
2 2 4 4
Summing these two, we get: (z+ y2 )2 +( x2 +yz)2 = x +y 3 2 2 3
4 −z −x y = z y+zy = −(xy)
2
2 2
Hence, (z + y2 )2 + ( x2 + yz)2 + (xy)2 = 0 and now is just conclude that x = y = z = 0

Problem 1.2. If x is a real number satisfying the equation

9 log3 x − 10 log9 x = 18 log27 45,
√
then the value of x is equal to m n, where m and n are positive integers, and n is not
divisible by the square of any prime. Find m + n.
9 log x 5 log x 6 log 45
answer. We have log 3 −
log 3 = log 3 , so 4√
log x = 6 log 45.
√
Removing the logs we have x4 = 6
45 , or x = 45 45 =⇒ 135 5 =⇒ 135 + 5 = 140

Problem 1.3. Find all primes p, such that there exist positive integers x, y which satisfy
(
p + 49 = 2x2
p2 + 49 = 2y 2
answer. Even removing the requirement that p is prime, this is an elliptic curve and it is
possible to determine all integral points on it.
More precisely, we can rewrite it as (2x2 − 49)2 = 2y 2 − 49. Multiplying by 2x2 gives us
Y 2 = X 3 − 98X 2 + 2450X
where X = 2x2 and Y = 2xy.
A computer algebra system can be used to compute the Mordell-Weil rank, which is
equal to 2, and to determine all the integral points (up to sign and ignoring the infinity
point):
(X, Y ) = (0, 0), (25, 125), (32, 104), (49, 49), (50, 50), (72, 204), (98, 490), (9800, 965300).
Translating back into (x, y, p), we find the following solutions:
(x, y, p) = (0, 35, −49), (4, 13, −17), (5, 5, 1), (6, 17, 23), (7, 35, 49), (70, 6895, 9751).
Among these, only (6, 17, 23) gives prime value of p.

1
POTW Handout 20 answers () institute of STEM education

Problem 1.4. Suppose x and y are real numbers satisfying

(
x3 − y 3 = 493.
x2 y − y 2 x = 50.

What is the positive difference between x and y?

answer.
(x − y)3 = x3 − 3x2 y + 3y 2 x − y 3
(x − y)3 = (x3 − y 3 ) − 3(x2 y − y 2 x)
(x − y)3 = 493 − 150 = 343
x−y =7

(
x+y =2
Problem 1.5. Find all pairs (x, y) of real numbers satisfying the system :
x4 − y 4 = 5x − 3y

answer. From the first equation, x + y = 2 =⇒ y = 2 − x. Substituting this to the

second equation gives
x4 − (2 − x)4 = 5x − 3(2 − x)
8x3 − 24x2 + 24x − 10 = 0
8(x − 3)3 − 2 = 0
1
This shows us that x = 3 + √ 3
4
 
1 1
The only pair is 3 + √ 3
, −1 − √
3
.
4 4


x + y + xy = 4

Problem 1.6. Solve the equation in R, the system y + z + yz = 7

x + z + xz = 9


answer. (x + 1)(y + 1) = 5 (y + 1)(z + 1) = 8 (x + 1)(z + 1) = 10

Multiplying this all together, (x+1)2 (y+1)2 (z+1)2 = 400 =⇒ (x+1)(y+1)(z+1) = ±20.
Dividing by the original equations, z + 1 = ±8, x + 1 = ± 52 , y + 1 = ±2.
This leads to the solutions ( 23 , 1, 7) and ( −7
2 , −3, −9).

Problem 1.7. If a dan b are positive numbers and satisfy, a log4 =b log10 =a−b log25
What are the value of a and b?

answer. Given loga 4 = logb 10 = loga−b 25. Then log4 a = log10√b = log25 (a − b) = t ⇒
√ √ √
4t = a and 10t = b and 25t√= a−b, 2t = a and 10t = b and 5t = a − b. a· a − b = b.
a2 − ab − b2 = 0. a = b · 5+1
2 (1). From log4 a = log10 b, we have ln a ln b
ln 4 = ln 10 (2).

Problem 1.8. Solve

x4x−6
 
logx = 2x − 3.
2

2
POTW Handout 20 answers () institute of STEM education

1
answer. 4x − 6 − logx 2 = 2x − 3 2x − 3 = log2 x (2x − 3) log2 x = 1 x = 2

√ √ √
Problem 1.9. Let a, b, and c be distinct positive integers such that a + b = c and
c is not a perfect square. What is the least possible value of a + b + c?
√
answer. Squaring gives a + b + 2 ab = c, so ab must be a perfect square.
Obviously ab = 1 doesn’t work, and neither does ab = 4, 9.
If ab = 16, then a = 2, b = 8 gives c = 18 and an answer of 28 .

Problem 1.10. Solve this system of equations

(
x2 = y 3 + 1
y 2 = x3 − 23
answer. p
x2 = y 3 + 1 =⇒ x = ± y 3 + 1
p
Now, plugging x = y 3 + 1 in y 2 = x3 − 23
p
y 2 = ( y 3 + 1)3 − 23 =⇒ y = 2
3
y 2 = (y 3 + 1) 2 − 23 =⇒ (y 2 + 23)2 = (y 3 + 1)3
y 4 + 46y 2 + 529 = (y 3 + 1)3
(y − 2)(y 8 + 2y 7 + 4y 6 + 11y 5 + 22y 4 + 43y 3 + 89y 2 + 132y + 264) = 0
(
y − 2 = 0 =⇒ y = 2
(y 8 + 2y 7 + 4y 6 + 11y 5 + 22y 4 + 43y 3 + 89y 2 + 132y + 264) = 0 has no solution ∈ R
p
For x = − y 3 + 1 in y 2 = x3 − 23, there is no solution ∈ R.
Now, for getting x substitute with y in one of these equations:

x2 = 8 + 1 =⇒ x = 3

x=3 y=2

Problem 1.11. Starting with a 5 × 5 grid, choose a 4 × 4 square in it. Then, choose
a 3 × 3 square in the 4 × 4 square, and a 2 × 2 square in the 3 × 3 square, and a 1 × 1
square in the 2 × 2 square. Assuming all squares chosen are made of unit squares inside
the grid. In how many ways can the squares be chosen so that the final 1 × 1 square is
the center of the original 5 × 5 grid?

answer. It doesn’t matter which 4 × 4 square you choose. WLOG let’s assume you chose
the bottom left 4 × 4 square. Then you have 4 cases. Top right 3 × 3 square Top left 3 × 3
square Bottom right 3 × 3 square Bottom left 3 × 3 square Thus, there are 4 + 2 + 2 + 1 = 9
ways. Since there are 4 possible 4 × 4 squares our answer is 36

Problem 1.12. Let ABCD be a rectangle with AB = 10 and AD = 5. Suppose points

P and Q are on segments CD and BC, respectively, such that the following conditions
hold: BD ∥ P Q ∠AP Q = 90◦ . What is the area of △CP Q?

3
POTW Handout 20 answers () institute of STEM education

1 15 15
answer. Slope of P Q is 2 thus, the slope of AP is −2. Therefore P C = 2 and CQ = 4
225
so our answer is
16

Problem 1.13. How many real roots does this log equation have?
2
log(x2 −3x)3 4 =
3
Should I use the fundamental theorem of algebra for this problem?
2 2 2
answer. log(x2 −3x)3 4 = 3 ⇔ 3 log(x2 −3x) 2 = 3 ⇔ log(x2 −3x) 2 = 1 ⇔ x2 − 3x = 2 ⇔
x2 − 3x − 2 = 0

Problem 1.14. In trapezoid ABCD, leg BC is perpendicular √ to bases AB and

√ CD,
and diagonals AC and BD are perpendicular. Given that AB = 11 and AD = 1001,
find BC 2 .

answer. Label the intersection of the diagonals E. AE 2 + BE 2 = 11, BE 2 + CE 2 = BC 2 ,

AE 2 + DE 2 = 1001, so we have 11 − BE 2 = 1001 − DE 2 ⇒ BE 2 = DE 2 − 990.
Substituting, CE 2 + √
DE 2 − 990 = BC 2 = CD2 − 990. Drawing
√ 2 the perpendicular √AF to
2 2 2
DC, we have CF = 11. AF = BC √ = 1001 − (CD − 11) √= 990 − CD + 2 11 =
CD2 − 990, so we have CD2 − CD 11 − 990 = 0 ⇒ CD = 10 11. Therefore, we have
BC 2 = CD2 − 990 = 110 .

Problem 1.15. Find

2π 4π 2010π 2012π
cos + cos + · · · + cos + cos
2013 2013 2013 2013
nπ z n +z 2013−n
answer. Let z = eiπ/2013 be a 2013th root of unity. We have cos( 2013 ) = 2 .
Thus,
1 2 2013−2
2π
cos( 2013 4π
)+cos( 2013 )+...+cos( 2012π
2013 ) = 2 [z +z +z 4 +z 2013−4 +...z 2012 +z 2013−2012 ]

= 12 [z + z 2 + z 3 + ... + z 2012 ]
= 12 [1 + z + z 2 + z 3 + ... + z 2012 ] − 1
2

z 2013 − 1 1
= −
2(z − 1) 2
−1
=
2

Problem 1.16. Find all triples (a, b, c) of real numbers such that ab + bc + ca = 1 and

a2 b + c = b2 c + a = c2 a + b.

4
POTW Handout 20 answers () institute of STEM education

answer. First assume that none of the numbers is equal to zero. Then considering the
first equation we have

a(ab − 1) = c(b2 − 1)
By the condition we have ab − 1 = −c(a + b) so dividing by c we get

a(a + b) = 1 − b2 =⇒ a2 + ab = 1 − b2
Now add the other two similar relations to find 2(a2 + b2 + c2 ) = 3 − ab − bc − ca = 2, so
a2 + b2 + c2 = 1. Then (a − b)2 + (b − c)2 + (c − a)2 = 0 implying a = b = c = √13 , which
works.
Now wlog assume a = 0, then we get c = b2 c = b, so in particular b = c and
ab + bc + ca = b2 = 1, giving b = c = 1 or b =
 c = −1, and both work.
1 1 1
In summary the solutions are √3 , √3 , √3 , (0, 1, 1) and (0, −1, −1), along with cyclic
permutations.
2 2 2 2
Problem 1.17. Solve over R the equation 4(sinx) + 3(tanx) = 4(cosx) + 3(cotanx) .
1−t t
answer. Setting t = cos2 x ∈ (0, 1), this is 41−t + 3 t = 4t + 3 1−t LHS is decreasing
while RHS is increasing and so at most one real root over (0, 1) t = 12 is a trivial root
π π
and so is the only one. And so cos2 x = 12 which is x = + k
4 2

Problem 1.18. Two distinct, real, infinite geometric series each have a sum of 1 and
have the same second term. The third term of one √
of the series is 1/8, and the second
m−n
term of both series can be written in the form p , where m, n, and p are positive
integers and m is not divisible by the square of any prime. Find 100m + 10n + p.
1
answer. Let the second term of each series be x. Then, the common ratio is 8x , and the
first term is 8x2 .
So, the sum is
8x2
1 =1
1 − 8x
. Thus,
√
3 2 1 −1 ± 5
64x − 8x + 1 = (4x − 1)(16x + 4x − 1) = 0 ⇒ x = ,
4 8
. √
5−1
The only solution in the appropriate form is x = 8 . Therefore, 100m + 10n + p =
518 .

Problem 1.19. Let a, b, c, and d be real numbers that satisfy the system of equations
a + b = −3
ab + bc + ca = −4
abc + bcd + cda + dab = 14
abcd = 30.
There exist relatively prime positive integers m and n such that
m
a2 + b2 + c2 + d2 = .
n
Find m + n.

5
POTW Handout 20 answers () institute of STEM education

30
answer. Let d = abc . Equation (3) becomes :

30(ab + bc + ca) 120

abc + = abc − = 14
abc abc
Hence (abc)2 − 14(abc) − 120 = 0. Solving the above we get abc = −6 or abc = 20.
ab + bc + ca = ab − 3c = −4 =⇒ ab = 3c − 4. If abc = −6, then c(3c − 4) = −6. This
gives does not give real solutions for c. Hence abc = 20. c(3c − 4) = 20 =⇒ c = −2 or
c = 10
3 . If c = 10/3, then the system a + b = −3 and ab = 6. Does not have real solutions
for both a and b. Hence c = −2.
9 9 141
a2 + b2 + c2 + d2 = 9 + − 2ab + 4 = 9 + 20 + 4 + =
4 4 4

The answer is 145 .

Problem 1.20. An infinite geometric series has sum 2005. A new series, obtained by
squaring each term of the original series, has 10 times the sum of the original series. The
common ratio of the original series is m n where m and n are relatively prime integers.
Find m + n.

answer. Let’s call the first term of the original geometric series a and the common ratio
r, so 2005 = a + ar + ar2 + . . ..
a
Using the sum formula for infinite geometric series, we have 1−r = 2005. Then we form
a new series,
a2 + a2 r 2 + a2 r 4 + . . .
a 2 a
We know this series has sum 20050 = 1−r 2 . Dividing this equation by 1−r , we get
a
10 = 1+r . Then a = 2005 − 2005r and a = 10 + 10r so 2005 − 2005r = 10 + 10r,
1995
1995 = 2015r and finally r = 2015 = 399
403 , so the answer is 399 + 403 = 802 .