A Divisor Diversion

Find the sum of all positive integers n n , each of which has precisely 16 positive divisors and a successor n + 1 n + 1 that has precisely 3 divisors.

If you think that there are an infinite number of such integers, then enter 99999 as your answer.


Inspiration .


The answer is 288.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

3 solutions

The only positive integers that have precisely 3 3 positive divisors are squares of primes p p , (the 3 3 divisors being 1 , p 1, p and p 2 p^{2} ). So we are looking for positive integers n n that equal p 2 1 p^{2} - 1 for some prime p p and have a total of 16 16 positive divisors.

For p = 2 p = 2 we have p 2 1 = 3 p^{2} - 1 = 3 which has only 2 2 divisors, and for p = 3 p = 3 we have p 2 1 = 8 = 2 3 p^{2} - 1 = 8 = 2^{3} , which has only 4 4 divisors. Now all primes p > 3 p \gt 3 are of the form 6 k ± 1 6k \pm 1 for some positive integer k k . Then p 2 1 = ( 6 k ± 1 ) 2 1 = 36 k 2 ± 12 k = 12 k ( 3 k ± 1 ) p^{2} - 1 = (6k \pm 1)^{2} - 1 = 36k^{2} \pm 12k = 12k(3k \pm 1) . Now if k k is even then 24 12 k 24 | 12k and if k k is odd then 3 k ± 1 3k \pm 1 is even and so 24 12 ( 3 k ± 1 ) 24 | 12(3k \pm 1) . Either way we have that 24 p 2 1 24 | p^{2} - 1 . Now 24 = 2 3 × 3 24 = 2^{3} \times 3 has ( 3 + 1 ) ( 1 + 1 ) = 8 (3 + 1)(1 + 1) = 8 divisors, so for p 2 1 p^{2} - 1 to end up with a total of 16 16 divisors we can introduce precisely one of the factors 2 4 , 3 2 2^{4}, 3^{2} or a prime (to the power 1 1 ) other than 2 2 or 3 3 .

Now since p 2 = ( p 1 ) ( p + 1 ) p^{2} = (p - 1)(p + 1) , and p 1 p - 1 and p + 1 p + 1 differ by 2 2 , if the added factor were:

  • (i) 2 4 2^{4} then we would need to be able to write 2 7 × 3 2^{7} \times 3 as a product of two integers that differ by 2 2 . The closest we can get in this case is 16 × 24 16 \times 24 ;

  • (ii) 3 2 3^{2} then we would need to be able to write 2 3 × 3 3 2^{3} \times 3^{3} as a product of two integers that differ by 2 2 . The closest we can get in this case is 12 × 18 12 \times 18 .

So we are left with the added prime option to look at. Now one of p 1 p - 1 or p + 1 p + 1 will be divisible by at least 6 6 and the other by at least 2 2 , and since they are successive even numbers one will also be divisible by 4 4 . So say one is divisible by 12 12 and the other by 2 2 . Then for p 1 p - 1 and p + 1 p + 1 to differ by 2 2 the introduced prime must be either 5 5 or 7 7 , yielding 2 × 5 = 10 2 \times 5 = 10 along with 12 12 as one pair, and 2 × 7 = 14 2 \times 7 = 14 along with 12 12 as another pair. If one was divisible by 6 6 and the other by 4 4 then only by introducing the factor 2 2 could we get successive even numbers 6 6 and 2 × 4 = 8 2 \times 4 = 8 , but the prime we need to introduce must be other than 2 2 or 3 3 .

So after looking at all the cases, the only positive integers n n that satisfy the conditions are

10 × 12 = 2 3 × 3 × 5 = 120 = 1 1 2 1 10 \times 12 = 2^{3} \times 3 \times 5 = 120 = 11^{2} - 1 and 12 × 14 = 2 3 × 3 × 7 = 168 = 1 3 2 1 12 \times 14 = 2^{3} \times 3 \times 7 = 168 = 13^{2} - 1 ,

the sum of which is 120 + 168 = 288 120 + 168 = \boxed{288} .

Ashish Gupta
Feb 8, 2017

Mark Hennings
Jan 29, 2017

A slightly different presentation...

We want σ 0 ( n ) = 16 \sigma_0(n) = 16 and σ 0 ( n + 1 ) = 3 \sigma_0(n+1) = 3 . The second of these tells us that n = p 2 1 n = p^2 - 1 where p p is prime. Since n = 3 n=3 is not a solution to the problem, p p must be odd. Thus both p 1 , p + 1 p-1,p+1 are even, and exactly one of them is a multiple of 4 4 . Thus we can write p ± 1 = 2 a q p 1 = 2 r p\pm1 \; = \; 2^{a}q \hspace{2cm} p \mp1 \; = \; 2r where a 2 a \ge 2 and p , q p,q are coprime odd integers. Thus n = p 2 1 = 2 a + 1 q r n = p^2-1 = 2^{a+1}qr and so 16 = σ 0 ( n ) = ( a + 2 ) σ 0 ( q ) σ 0 ( r ) 16 = \sigma_0(n) = (a+2)\sigma_0(q)\sigma_0(r) . Since a + 2 4 a+2 \ge 4 , there are three options:

  1. If a + 2 = 16 a+2 = 16 , then a = 14 a=14 and q = r = 1 q=r=1 . But then p ± 1 = 2 14 p \pm1 = 2^{14} , p 1 = 2 p \mp1 = 2 , which is impossible.

  2. If a + 2 = 8 a+2 = 8 , then a = 6 a=6 and σ 0 ( q ) σ 0 ( r ) = 2 \sigma_0(q)\sigma_0(r) = 2 . Now n = 8 n=8 is not a solution, so p = 3 p=3 is not allowed, and hence r r cannot be equal to 1 1 . Thus r r must be prime, and q = 1 q=1 . But then p = 2 6 ± 1 p = 2^6 \pm 1 is either 63 63 or 65 65 , which is impossible.

  3. if a + 2 = 4 a+2 = 4 , then a = 2 a=2 and σ 0 ( q ) σ 0 ( r ) = 4 \sigma_0(q)\sigma_0(r) = 4 . If q = 1 q=1 then p = 3 , 5 p = 3,5 , and so n = 8 , 24 n = 8,24 , neither of which are possible. If r = 1 r=1 then p = 3 p = 3 , which is impossible. Thus we deduce that both q q and r r are prime.

This last case now breaks down into two final subcases:

  • If p = 4 q + 1 = 2 r 1 p = 4q + 1 = 2r - 1 , then r = 2 q + 1 r = 2q+1 . If q 1 ( m o d 3 ) q \equiv 1 \pmod{3} , then r 0 ( m o d 3 ) r \equiv 0 \pmod{3} , so that r = 3 r=3 and q = 1 q=1 , which is impossible. If q 2 ( m o d 3 ) q \equiv 2 \pmod{3} then p 0 ( m o d 3 ) p \equiv 0 \pmod{3} , and hence p = 3 p=3 , which is impossible. Thus q 0 ( m o d 3 ) q \equiv 0 \pmod{3} , and hence q = 3 q=3 , so that p = 13 p=13 , r = 7 r=7 , and hence n = 168 n=168 is one solution.

  • If p = 4 q 1 = 2 r + 1 p = 4q-1 = 2r+1 , then r = 2 q 1 r = 2q-1 . If q 1 ( m o d 3 ) q \equiv 1 \pmod{3} , then p 0 ( m o d 3 ) p \equiv 0 \pmod{3} , so that p = 3 p=3 , which is impossible. If q 2 ( m o d 3 ) q \equiv 2 \pmod{3} then r 0 ( m o d 3 ) r \equiv 0 \pmod{3} , and hence r = 3 r=3 , so that q = 2 q=2 , which is impossible. Thus q 0 ( m o d 3 ) q \equiv 0 \pmod{3} , and hence q = 3 q=3 , so that p = 11 p=11 , r = 7 r=7 , and hence n = 120 n=120 is the other solution.

Thus the answer is 168 + 120 = 288 168 + 120 = \boxed{288} .

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...