login
A174973
Numbers whose divisors increase by a factor of at most 2.
55
1, 2, 4, 6, 8, 12, 16, 18, 20, 24, 28, 30, 32, 36, 40, 42, 48, 54, 56, 60, 64, 66, 72, 80, 84, 88, 90, 96, 100, 104, 108, 112, 120, 126, 128, 132, 140, 144, 150, 156, 160, 162, 168, 176, 180, 192, 196, 198, 200, 204, 208, 210, 216, 220, 224, 228, 234, 240, 252, 256
OFFSET
1,2
COMMENTS
That is, if the divisors of a number are listed in increasing order, the ratio of adjacent divisors is at most 2. The only odd number in this sequence is 1. Every term appears to be a practical number (A005153). The first practical number not here is 78.
Let p1^e1 * p2^e2 * ... * pr^er be the prime factorization of a number, with primes p1 < p2 < ... < pr and ek > 0. Then the number is in this sequence if and only if pk <= 2*Product_{j < k} p_j^e_j. This condition is similar to, but more restrictive than, the condition for practical numbers.
The polymath8 project led by Terry Tao refers to these numbers as "2-densely divisible". In general they say that n is y-densely divisible if its divisors increase by a factor of y or less, or equivalently, if for every R with 1 <= R <= n, there is a divisor in the interval [R/y,R]. They use this as a weakening of the condition that n be y-smooth. - David S. Metzler, Jul 02 2013
Is this the same as numbers k with the property that the symmetric representation of sigma(k) has only one part? If not, where is the first place these sequences differ? (cf. A237593). - Omar E. Pol, Mar 06 2014
Yes, the sequence so defined is the same as this sequence; see proof in the links. - Hartmut F. W. Hoft, Nov 26 2014
Saias (1997) called these terms "2-dense numbers" and proved that if N(x) is the number of terms not exceeding x, then there are two positive constants c_1 and c_2 such that c_1 * x/log_2(x) <= N(x) <= c_2 * x/log_2(x) for all x >= 2. - Amiram Eldar, Jul 23 2020
Weingartner (2015, 2019) showed that N(x) = c*x/log(x) + O(x/(log(x))^2), where c = 1.224830... As a result, a(n) = C*n*log(n*log(n)) + O(n), where C = 1/c = 0.816439... - Andreas Weingartner, Jun 22 2021
LINKS
Alois P. Heinz, Table of n, a(n) for n = 1..20000 (first 1000 terms from T. D. Noe)
José Manuel Rodríguez Caballero, Jordan's Expansion of the Reciprocal of Theta Functions and 2-densely Divisible Numbers, Integers, Vol. 20 (2020), Article A2.
Hartmut F. W. Hoft, Proof of a conjecture.
Hartmut F. W. Hoft, Proof of a second conjecture.
Eric Saias, Entiers à diviseurs denses 1, Journal of Number Theory, Vol. 62, No. 1 (1997), pp. 163-191.
Terence Tao, A Truncated Elementary Selberg Sieve of Pintz. (blog entry defining y-densely divisible)
Terence Tao et al., Polymath8 home page.
Andreas Weingartner, Practical numbers and the distribution of divisors, The Quarterly Journal of Mathematics, Vol. 66, No. 2 (2015), pp. 743-758; arXiv preprint, arXiv:1405.2585 [math.NT], 2014-2015.
Andreas Weingartner, On the constant factor in several related asymptotic estimates, Math. Comp., Vol. 88, No. 318 (2019), pp. 1883-1902; arXiv preprint, arXiv:1705.06349 [math.NT], 2017-2018.
Andreas Weingartner, The number of prime factors of integers with dense divisors, arXiv:2101.11585 [math.NT], 2021.
Andreas Weingartner, Uniform distribution of alpha*n modulo one for a family of integer sequences, arXiv:2303.16819 [math.NT], 2023.
FORMULA
a(n) = A047836(n) / 2. - Reinhard Zumkeller, Sep 28 2011
a(n) = C*n*log(n*log(n)) + O(n), where C = 0.816439... (see comments). - Andreas Weingartner, Jun 23 2021
EXAMPLE
The divisors of 12 are 1, 2, 3, 4, 6, 12. The ratios of adjacent divisors is 2, 3/2, 4/3, 3/2, and 2, which are all <= 2. Hence 12 is in this sequence.
Example from Omar E. Pol, Mar 06 2014: (Start)
The symmetric representation of sigma(6) = 12 in the first quadrant looks like this:
y
.
._ _ _ _
|_ _ _ |_
. | |_
. |_ _ |
. | |
. | |
. . . . . |_| . . x
.
6 is in the sequence because the symmetric representation of sigma(6) = 12 has only one part. The 6th row of A237593 is [4, 1, 1, 1, 1, 4] and the 5th row of A237593 is [3, 2, 2, 3] therefore between both symmetric Dyck paths there is only one region (or part) of size 12.
70 is not in the sequence because the symmetric representation of sigma(70) = 144 has three parts. The 70th row of A237593 is [36, 12, 6, 4, 3, 2, 2, 2, 1, 1, 1, 1, 1, 1, 2, 2, 2, 3, 4, 6, 12, 36] and the 69th row of A237593 is [35, 12, 7, 4, 2, 3, 1, 2, 1, 1, 1, 1, 1, 1, 2, 1, 3, 2, 4, 7, 12, 35] therefore between both symmetric Dyck paths there are three regions (or parts) of size [54, 36, 54]. (End)
MAPLE
a:= proc() option remember; local k; for k from 1+`if`(n=1, 0,
a(n-1)) while (l-> ormap(x-> x, [seq(l[i]>l[i-1]*2, i=2..
nops(l))]))(sort([(numtheory[divisors](k))[]])) do od; k
end:
seq(a(n), n=1..100); # Alois P. Heinz, Jul 27 2018
MATHEMATICA
OK[n_] := Module[{d=Divisors[n]}, And@@(#<=2& /@ (Rest[d]/Most[d]))]; Select[Range[1000], OK]
dif2Q[n_]:=AllTrue[#[[2]]/#[[1]]&/@Partition[Divisors[n], 2, 1], #<=2&]; Select[Range[300], dif2Q] (* Requires Mathematica version 10 or later *) (* Harvey P. Dale, Nov 29 2020 *)
PROG
(Haskell)
a174973 n = a174973_list !! (n-1)
a174973_list = filter f [1..] where
f n = all (<= 0) $ zipWith (-) (tail divs) (map (* 2) divs)
where divs = a027750_row' n
-- Reinhard Zumkeller, Jun 25 2015, Sep 28 2011
(PARI) is(n)=my(d=divisors(n)); for(i=2, #d, if(d[i]>2*d[i-1], return(0))); 1 \\ Charles R Greathouse IV, Jul 06 2013
(Magma) [k:k in [1..260]|forall{i:i in [1..#Divisors(k)-1]|d[i+1]/d[i] le 2 where d is Divisors(k)}]; // Marius A. Burtea, Jan 09 2020
(Python)
from sympy import divisors
def ok(n):
d = divisors(n)
return all(d[i]/d[i-1] <= 2 for i in range(1, len(d)))
print(list(filter(ok, range(1, 257)))) # Michael S. Branicky, Jun 22 2021
CROSSREFS
Subsequence of A196149 and of A071562. A000396 and A000079 are subsequences.
Cf. A027750, A047836, A237593, A365429 (characteristic function).
Column 1 of A240062.
First differs from A103288 and A125225 at a(23). First differs from A005153 at a(24).
Sequence in context: A125225 A092903 A005153 * A365270 A238443 A325795
KEYWORD
nonn
AUTHOR
T. D. Noe, Apr 02 2010
EXTENSIONS
Edited by N. J. A. Sloane, Sep 09 2023
Edited by Peter Munn, Oct 17 2023
STATUS
approved