Number Theory B Solutions

Number Theory B Solutions
1. Find the number of pairs of integers x and y such that x2 + xy + y 2 = 28.

Solution. 12. We multiply both sides by 4, and rewrite the equation as (2x + y)2 + 3y 2 = 112.
From there, we can simply take possible values for y sequentially to get equations that work,
i.e. if we are considering some y, we check that 112 − 3y 2 is a perfect square, and if it is, it is
immediately an admissible answer. We end up with three possible equations: 102 +3·22 = 112,
82 + 3 · 42 = 112 and 22 + 3 · 62 = 112. For each of these, we can find a possible x and y,
all nonzero, satisfying the equation. But note that, given that x and y, ±x and ±y serve the
purpose just as well. So it follows that for each of those three equations, there are four possible
values of (x, y). Hence the total number of pairs of integers is 12.
2. Suppose you are given that for some n ∈ N, the expression 1! + 2! + ... + n! is a perfect square.
Find the sum of all possible values of n.
Solution. 4. Note that it is easy to check the result for n = 1, 2, 3, 4, which yield the results
1, 3, 9 and 33 respectively. Out of these, obviously, two are squares, and so 1 and 3 do satisfy
the condition. Now, note that 5! has a factor of 10, coming from one 2 and one 5. So, the last
digit of 5! is 0. In fact, the last digit of any factorial after 4 is 0 for precisely the same reason.
Hence, the sum of some consecutive factorials after 4 is going to give us a number with last
digit 0. Adding that to 1! + 2! + 3! + 4!, we get a number whose last digit is a 3, but that
cannot be a perfect square. Hence, the given expression is not a perfect square for n ≥ 5, and
in fact we have checked the rest of the values ourselves. It follows that 1 and 3 are the only
solutions, and so our answer is 1 + 3 = 4.
3. You are given that
17! = 355687ab8096000
for some digits a and b. Find the two-digit number ab that is missing above.
Solution. 42. First note that the number is divisible by 11 as well as 9. We simply apply
the divisibility criteria for these two numbers, and immediately obtain two simultaneous linear
equations:
9|34 + a + b + 23
and
11|(16 + a + 17) − (18 + b + 6)
which give the following possibilities: (a + b) ∈ {6, 15} and (a − b) ∈ {−9, 2, 13}, where a and
b are digits. Now, a − b = −9 iff b = 9, a = 0, which does not satisfy any of the relations on
a + b. So, that possibility is eliminated. Furthermore, note that a + b and a − b added together
gives an even number 2a, so the parities of a + b and a − b must be equal. It follows that
either we have a + b = 6, a − b = 2, or a + b = 15, a − b = 13. Solving the first equation gives
1
(a, b) = (4, 2) and the second gives (a, b) = (14, 1), but since a and b are digits, it follows that
our required solution is 42.
4. Find the number of ordered pairs (a, b) of positive integers that are solutions of the following
equation:
a2 + b2 = ab(a + b)
.
Solution. 1. Suppose at first that a = b. Then we get 2a2 = 2a3 , which yields a = 0 or
a = 1. Since we must have positive integers, this yields a = b = 1 as a possible solution. Now
if a 6= b, then suppose WLOG that a > b. Then, the equation reduces to ab + ab = a + b. But,
1 > ab , and a + b ≥ a + 1, because we have positive integers only. Then, we get
a a b
+1> + =a+b≥a+1
b b a
i.e. ab > a, which means 1 > b, which is a contradiction. Note that we can divide throughout
freely by any of the two variables because they are positive integers.
5. Find the sum of all prime numbers p which satisfy
p = a4 + b4 + c4 − 3
for some primes (not necessarily distinct) a, b and c.

Solution. 719. If a, b and c are all odd, then the right hand side is even (and it’s greater than
2, which can be easily checked), and so this forces p to be an even number greater than 2, a
contradiction. So exactly one or three of a, b and c is 2. Again, if all three are 2, then p = 45,
which is not a prime, hence inadmissible. So exactly one of a, b and c is 2, say a. Then we
have p = b4 + c4 + 15. Now, if none of b and c is 3, then they are each of the form 6k ± 1 for
some integer k. Then, their fourth powers are of the form 6k 0 + 1, and hence, adding them
together, the right hand side becomes divisible by 3, which is inadmissible. So, one of b and c
must be 3 (they cannot both be 3, because then p = 175, not a prime. So suppose b is 3. Then
we get p = c4 + 94. The last deduction is as follows: if c 6= 5, then c must end in 1, 3, 7 or 9.
The fourth power of each number of this form ends in the digit 1. Then, adding that to 94,
we will get a number divisible by 5, a contradiction. So c must be 5. We have to finally check
that p = 719 is indeed a prime. This is checked easily, and hence we get our unique solution.
6. Find the sum of all integers x for which there is an integer y, such that x3 − y 3 = xy + 61.
Solution. 6. It is easy to see that one or more of x and y cannot be zero, because then the
equation cannot hold true because of the constant term. So assume that x and y are positive
integers. Also, x 6= y, because then we would have x2 + 61 = 0, which is inadmissible. If
x < y, then the left hand side becomes negative while the right remains positive. So clearly,
x > y ≥ 1. We will use the identity x3 − y 3 = (x − y)(x2 + xy + y 2 ) = xy + 61, from which we
have
61 = (x − y)(x2 + y 2 ) + (x − y − 1)xy
2
.
Then, if x − y ≥ 3, then we must have x ≥ 3 + y = 4, and 61 ≥ 3(x2 + y 2 ) + (x − y −
1)xy ≥ 3(x2 + y 2 ), so that x2 + y 2 ≤ 20, and then x = 4. Then, y = 1 or y = 2, but
x − y ≥ 3 =⇒ (x, y) = (4, 1), which does not satisfy the original equation.
If x − y = 2, then we get
61 = 2(x2 + y 2 ) + xy = 2((y + 2)2 + y 2 ) + (y + 2)y = 5y 2 + 10y + 8

which has no solution for y that is positive and integer. So it follows that we must have
x − y = 1, and so
61 = (x2 + y 2 ) = (y + 1)2 + y 2
solving which we easily get (x, y) = (6, 5), which is the unique solution.
7. Suppose that for some positive integer n, the first two digits of 5n and 2n are identical. Find
the number formed by these two digits.
Solution. 31. Suppose a is the number formed by the two digits. From the condition, we
must have 10k a < 2n < 10k (a + 1) and 10l a < 5n < 10l (a + 1) for some positive integers k and
l. Then, we can multiply these two equations together to obtain
10k+l a2 < 10n < 10k+l (a + 1)2

.
Now, note that a is a two-digit number, and so 10 ≤ a ≤ 99. so, 102 ≤ a2 , and in fact,
(a + 1)2 ≤ 1002 = 104 .
So, our equation now has two additional bounds, and becomes
10k+l+2 ≤ 10k+l a2 < 10n < 10k+l (a + 1)2 ≤ 10k+l+4

2 3
and hence it follows
√ that n = k + l + 3. Once we have this, it becomes easy to see a < 10 <
2
(a + 1) , i.e. a < 1000 < a + 1, whence a = 31.
8. Let s(m) denote the sum of the digits of the positive integer m. Find the largest positive
integer that has no digits equal to zero and satisfies the equation
2s(n) = s(n2 )
Solution. 1111. Suppose n has k digits, that is 10k−1 ≤ n < 10k , then k ≤ s(n) by the
condition. Also, n2 < 102k , hence n2 has at most 2k digits, and so s(n2 ) ≤ 18k. Thus
2k ≤ 2s(n) = s(n2 ) ≤ 18k
which implies k ≤ 6, and so 2s(n) ≤ 18 × 6 or also s(n) ≤ 6. If s(n) = 6 then n is divisible by
3 so n2 and s(n2 ) = 2s(n) is divisible by 3, which is impossible. Hence 1 ≤ s(n) ≤ 5 and so
from the equation, the possible values of s(n2 ) are 2,4,8,16, or 32. But the remainder of s(n2 )
modulo 9 is the same as the remainder of n2 modulo 9, which can be only 0,1,4 or 7. Hence
s(n) is either 2 or 4, and the greatest number satisfying the conditions of the problem is 1111.

Number Theory B Solutions

Uploaded by

Copyright:

Available Formats

Number Theory B Solutions

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Number Theory B Solutions

Uploaded by

Copyright:

Available Formats

Number Theory B Solutions

1. Find the number of pairs of integers x and y such that x2 + xy + y 2 = 28.

3. You are given that

11|(16 + a + 17) − (18 + b + 6)

for some primes (not necessarily distinct) a, b and c.

61 = 2(x2 + y 2 ) + xy = 2((y + 2)2 + y 2 ) + (y + 2)y = 5y 2 + 10y + 8

10k+l a2 < 10n < 10k+l (a + 1)2

10k+l+2 ≤ 10k+l a2 < 10n < 10k+l (a + 1)2 ≤ 10k+l+4

You might also like