IOQM (2024) Practice Sheet-3

Corporate Office : Aakash Tower, 8, Pusa Road, New Delhi-110005

Ph.: 011-47623456

Indian Olympiad Qualifier in Mathematics

Practice Sheet-3
Answers
1. (07)
3. (80)
5. (07)
6. (12)
7. (42)
8. (02)
9. (36)
10. (52)
11. (26)
12. (87)
13. (03)
14. (17)
15. (03)
16. (56)
17. (90)
18. (18)
19. (08)

Corporate Office : Aakash Tower, 8, Pusa Road, New Delhi-110005 Page [3]
IOQM (2024) Practice Sheet-3

Solution

1. Answer (07) Now since 19 is a prime, 18!  –1 (mod 19)

By Fermat’s little theorem 558  1 (mod 59), so immediately follows by Wilson’s theorem. Also,
5116  1 (mod 59), so 5119  53  7 (mod 59). noting that 23 is a prime, we have
2. Answer () 24 × 18! = 18! × (–1) × (–2) × (–3) × (–4)
By Fermat’s little theorem, 10p – 1 – 1 is divisible  18! × 19 × 20 × 21 × 22
by p, say pa = 10p – 1 – 1. So Wilson's
= 22!  –1  – 24 (mod 23)
k ka  1 1 1 
 p 1  ka  p 1    ...  And thus 18! º –1 (mod 23) as well.
p 10  1 2  p 1 3  p 1
 10 10 10 
Therefore 437 | (18! + 1).
Since ka < pa < 10p – 1, write ka as a(p – 1)-digit 5. Answer (07)
number (with Os in the front if necessary). The Rewrite the equation as
above expansion shows that the (p – 1)-digit
23! 23! 23!
number will repeat in the decimal expansion. a  23!   ... 
2 3 23
For example, let k = 2 and p = 7. Then Clearly all the terms in the right side are integer.
6
10  1 23!
a  142857 , so ka = 285714. So Also except all the quotients contain the
7 13
2 285714  1 1 1  factor 13 and thus are divisible by 13. Therefore,
  285714  6  12  18  ... 
7 999999  10 10 10  we get
= 0.285714 + 0.000000285714 + 23! 3! 23!
a  23!   ... 
0.0000000000000285714 + … 2 3 23
= 0.285714 23!
  1 2  ...  12  14  ...  23
3. Answer (80) 13
There are f(100) = (4 – 2)(25 – 5) = 40 coprime  (1 × 2 × … × 12) × (1 × 2 × … × 10)
numbers to 100 in {1, 2, …, 100} Wilson's
= 12! × 10!  12 × 10! (mod 13)
Since gcd(a, b) = gcd(a – b, b)
So, we know that a  12 × 10! (mod 13). Now, we
gcd(101, 100) = gcd(1, 100)
use 11!  1  66 (mod 13), which follows by
gcd(102, 100) = gcd (2, 100)
Wilson’s. Hence 10!  6 (mod 13). Finally, we

gcd(200, 100) = gcd (100, 100) have

So, there are f(100) = 40 coprime numbers to 100 a  12 × 10!  12 × 6  7 (mod 13)
in {101, 102, …, 200}. So the answer is 40 + 40 = 6. Answer (12)
80. The solutions are (0, 1, 83) and (1, 2, 3) up to
4. Answer () permutation. First, we do the case where at least
First notice that 437 = 19 × 23 and so it suffices one of a, b, c is 0. WLOG, say a = 0. Then we
to prove that have 1 + bc = 84 ⇒ bc = 83. As 83 is prime, the
18! –1 (mod 19), (mod 23) only solution is (0, 1, 83) up to permutation.
Otherwise, we claim that at least one of a, b, c is

Corporate Office : Aakash Tower, 8, Pusa Road, New Delhi-110005 Page [4]
IOQM (2024) Practice Sheet-3

equal to 1. Otherwise, all are at least 2, so (1 + last two digits when squared: 0, 1, 25, 76. We can
ab)(1 + bc)(1 + ac) ≥ 53 > 84. So WLOG, set now go backwards, repeatedly solving equations
a = 1. We now need (b + 1)(c + 1)(bc + 1) = 84. of the form x2  n (mod 100) where n is a number
Now, WLOG, say b ≤ c. If b = 1, then (c + 1)2 = that already satisfies the condition.
42, which has no solution. If b ≥ 3, then (b + 1)(c 0 and 25 together gives all multiples of 5, for 20
+ 1)(bc + 1) ≥ 42  10 = 160 > 84. So we need b = numbers in total.
2. Then we need (c + 1)(2c + 1) = 21. Solving this 1 gives 1, 49, 51, 99, and 49 then gives 7, 43, 57,
gives c = 3, for the solution (1, 2, 3). 93. Similarly 76 gives 24, 26, 74, 76 and 24 then
Therefore, the answer is 6 + 6 = 12. gives 18, 32, 68, 82 for 16 numbers in total.
7. Answer (42)
Hence there are 20 + 16 = 36 such numbers in
The sum ab + bc + cd + da = (a + c)(b + d) = total.
437 = 19  23, so {a + c, b + d} = {19, 23} as 10. Answer (52)
having either pair sum to 1 is impossible. Then
Note that the remainder when n2 is divided by
the sum of all 4 is 19 + 23 = 42. (In fact, it is not
5 must be 0, 1 or 4. Then we have 25p = n2 or
difficult to see that the only possible solutions
25p = n2 − 1 or 25p = n2 − 4. In the first case
are (a, b, c, d) = (8, 8, 11, 15) or its cyclic
there are no solutions. In the second case, if 25p
permutations and reflections.)
= (n − 1)(n + 1), then we must have n − 1 = 25 or
8. Answer (02)
n + 1 = 25 as n − 1 and n + 1 cannot both be
We want ab|a2017 + b. This gives that a|b. divisible by 5, and also cannot both have a
Therefore, we can set b = b2017a. Substituting factor besides 25. Similarly, in the third case,
this gives b2017a2|a2017 + b2017a, so b2017a|a2016 + 25p = (n − 2)(n + 2), so we must have n − 2 = 25
b2017. Once again, we get a|b2017, so we can set or n + 2 = 25. Therefore the n we have to check
b2017 = b2016a. Continuing this way, if we have are 23, 24, 26, 27. These give values of p = 21,
bi+1a|ai + bi+1, then a|bi+1, so we can set bi+1 = bia p = 23,
and derive bia|ai−1 + bi. Continuing down to i = 1, p = 27, and p = 29, of which only 23 and 29 are
we would have b = b1 a2017 so ab1|1 + b1. If a ≥ 3, prime, so the answer is 23 + 29 = 52.
then ab1 > 1 + b1 for all b1 ≥ 1, so we need either 11. Answer (26)
a = 1 or We claim that all the weird numbers are all the
a = 2. If a = 1, then b|b + 1, so b = 1. This gives prime numbers and 4. Since no numbers
the pair (1, 1). If a = 2, we need 2b|b + 22017. between 1 and p –2 divide prime p, (p –2)! will not

Therefore, we get b|22017, so we can write b = 2k be divisible by p. We also have 2! = 2 not being a
multiple of 4.
for 0 ≤ k ≤ 2017. Then we need 2k+1|2k + 22017.
Now we show that all other numbers are not
As k ≤ 2017, we need 2|1 + 22017−k. This can only
weird. If n = pq where p  q and p, q  2, then
happen is k = 2017. This gives the pair (2, 22017)
since p and q both appear in 1, 2, …, n – 2 and
9. Answer (36)
are distinct, we have pq | (n – 2)!. This leaves the
It suffices to consider the remainder mod 100. We
only case of n = p2 for prime p  3. In this case,
start with the four numbers that have the same

Corporate Office : Aakash Tower, 8, Pusa Road, New Delhi-110005 Page [5]
IOQM (2024) Practice Sheet-3

we can note that p and 2p are both less than Because no number from 2007, 2008,….4012 is
p2 – 2, so 2p2| (n – 2)! and we are similarly done. twice another such number, p(2007),
p(2008,…,p(4012) are 2006 distinct odd positive
Since there are 25 prime numbers not
integers. Also note that these odd numbers
exceeding 100, there are 25 + 1 = 26 weird
belong to the set {1, 3, 5,…4011}, which also
numbers.
consists of exactly 2006 elements. It follows that
12. Answer (87) {p(2007), p(2008),….,p(4012)} = {1, 3, …., 4011}.
The factorization of 1547 is 7 · 13 · 17, so we wish Hence the desired sum is equal to
to find the number of positive integers less than p(2006) + 1 + 3 + …+4011 = 1003 + 20062
1998 that are not divisible by 7, 13, or 17. By the
Principle of Inclusion-Exclusion, we first subtract = 1003  4013 = 4025039.
the numbers that are divisible by one of 7, 13, and 15. Answer (03)
17, add back those that are divisible by two of 7, By Fermat’s little theorem we have a5  a(mod 5)
13, and 17, then subtract those divisible by three for any integer a. Modulo 5, we have
of them. That is, 15 + 25 + 35 + … + 20075  1 + 2 + 3 + … + 2007
 1997   1997  1997  1997  = 2007.1004  2.4  3.
1997      
 7   13   17   7  13  Thus, the desired remainder is 3.
 1997   1997   1997  16. Answer (56)
   
 7  17  13  17   7  13  17  Given 49a + 7b + c = 28
= 1997 – 285 – 153 – 117 + 21 + 16 + 9 – 1  7(7a + b) + c = 286 … (i)

= 1487. A number ‘7’ divides 7(7a + b)

13. Answer (03)  Remainder on L.H.S = c

For p  3, p2  1 (mod 3) and so 3 | (p2 + 11). Remainder obtained by dividing 286 by ‘7’ gives
Similarly, for p  2, p2  1 (mod 4) and so 4 | (p2 + the remainder = 6
11). Except in these two cases, then, 12 | (p2 +  c=6
11); since 12 itself has 6 divisors (1, 2, 3, 4, 6, 12) 49a + 7b + 6 = 286
and p2 + 11 > 12 for p > 1. p2 + 11 must have
 49a + 7b + 6 = 286
more than 6 divisors. The only cases to check are
 49a + 7b = 280
p = 2 and p = 3. If p = 2, then p2 + 11 = 15, which
has only 4 divisors (1, 3, 5, 15), while if p = 3, then  7a + b = 40 … (ii)
p2 + 11 = 20, which indeed has 6 divisors {1, 2, 4, ‘7a’ is divisible by ‘7’
5, 10, 20}. Hence, p = 3 is the only solution.  Remainder on L.H.S = b
14. Answer (17) Remainder obtained by dividing ‘40’ by 7 gives
For a positive integer n, let p(n) denote its the remainder = 5
greatest odd divisor. We can write n = 2k  p(n) for
 b=5
some nonnegative integer k. If two positive
 7a = 35 from (ii)
integers n1 and n2 are such that p(n1) = p(n2), then
one is at least twice the other.  a=5
 Value of 10b + c = 10 × 5 + 6

Corporate Office : Aakash Tower, 8, Pusa Road, New Delhi-110005 Page [6]
IOQM (2024) Practice Sheet-3

= 50 + 6 = (n + 16)! – (n + 15)! + (n + 15)! – (n + 14)! + …

= 56 + (n + 1)! – n!
17. Answer (90) = (n + 15)!(n + 15) + (n + 14)!(n + 14) + … + (n +
It is clear that n = 1 and n = 2 work so assume 1)!(n + 1) + n!n.
that n > 2. If 2n – 3 is composite then its smallest This shows that the factorial base expansion of
2n  3 (n + 6)! – n! is
prime factor is at most  n so will be
2 (0, 0, …, 0, n, n + 1, …, n + 14, n + 15).
coprime to 15.n!2 + 1. Therefore assume that Which begins with a block of n – 1 zeros. The
2n – 3 = p is prime. We can rewrite the numerator factorial base expansion of 16! Is (0, 0, …, 0, 1),
as so the requested expansion is
p  3   p  3 p 1
 1n .15.  1.2... . .

...  p  1   1mod p  (0, 0, …, 0; 1; 0, …, 0; 32, 33, …, 47; 0, …, 0; 64,
 2  2 2 
…, 79; ….; 1984, …, 1999).
By Wilson’s theorem, (p – 1)!  –1 (mod p), so the
Notice that starting in position thirty-two, the
expression simplifies to expansion contains groups of sixteen nonzero
n 1 p  3 p 1 p 1 p  3
 1 .15. . . . 1 numbers alternating with groups of sixteen zeros.
2 2 2 2
With the exception of f16 = 1, each nonzero fi is i.
n 1 135
  1 .  1  mod p  Each of the 62 groups of sixteen nonzero
16
numbers contributes 8 to the alternating sum and
If p  3 (mod 4), then we have
f16 contributes –1, so the requested value is 8.62
135  16 151
  0  mod p  – 1 = 495.
16 16
19. Answer (08)
If p  1 (mod 4), then we have
Notice first that the given hypothesis implies that
135  16 119
  0  mod p  {a–1} = a–1 (since 1 < a and 0 < a–1 < 1) and {a2} =
16 16
a2 – 2. Hence a must satisfy the equation a–1 = a2
So p must be a prime divisor of 151 or 119,
– 2, or a3 – 2a – 1 = 0. This factors as
which means that p  {7, 17, 151}. All of these
(a + 1)(a2 – a – 1) = 0
numbers work aside from 7 (because 7  (mod
1 5
4)) and the corresponding values of n are 10 Whose only positive root is a  . Now use
2
and 77. The sum of the solutions is then 1 + 2 +
the relations a2 = a + 1 and a3 = 2a + 1 to calculate
10 + 77 = 90.
a6 = 8a + 5, a12 = 144a + 89 and a13 = 233a + 144
18. Answer (18)
From which it follows that
Because (n + 1)! – n! = n!(n + 1) – n! = n!n, it
a13  144
follows that (n + 16)! – n! a12  144a 1   233 .
a



Corporate Office : Aakash Tower, 8, Pusa Road, New Delhi-110005 Page [7]