The American Mathematical Monthly
ISSN: 0002-9890 (Print) 1930-0972 (Online) Journal homepage: https://www.tandfonline.com/loi/uamm20
A Fresh Look at Highly Composite Numbers
Harold W. Ellingsen Jr.
To cite this article: Harold W. Ellingsen Jr. (2019) A Fresh Look at Highly Composite Numbers,
The American Mathematical Monthly, 126:8, 740-741, DOI: 10.1080/00029890.2019.1632631
To link to this article: https://doi.org/10.1080/00029890.2019.1632631
Published online: 19 Sep 2019.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=uamm20
A Fresh Look at Highly Composite Numbers
Harold W. Ellingsen, Jr.
Abstract. In this note we provide fresh ways to prove some of the already known properties
of highly composite numbers.
Define τ : Z+ → Z+ by τ (n) is the number of divisors of n. For any n ∈ Z+ , if
b b b
q1 1 q2 2 · · · ql l is its prime-power factorization, then τ (n) = li=1 (bi + 1) [2, p. 252].
A number n is called highly composite provided that it has more divisors than any
number less than it, that is, for all m ∈ Z+ if m < n, then τ (m) < τ (n). The following
table shows that 1, 2, 4, 6, and 12 are highly composite while 3, 5, 7, 8, 9, 10, and 11
are not.
n 1 2 3 4 5 6 7 8 9 10 11 12
τ (n) 1 2 2 3 2 4 2 4 3 4 2 6
For each i ∈ Z+ we will denote the ith prime by pi . One of the first properties
proved about highly composite numbers is their form: If n is a highly composite
number greater than 1, then there is k ∈ Z+ such that its prime-power factorization
a a a
is p11 p22 · · · pk k where a1 ≥ a2 ≥ · · · ≥ ak ≥ 1 [1, p. 113]. Notice that the prime divi-
sors of a highly composite number are the first k primes, so that p1 = 2, p2 = 3,
p3 = 5, etc. We use this property along with Bertrand’s postulate (namely, that for all
n > 1, there is a prime between n and 2n [2, p. 83]) in the following proofs.
Theorem 1 ([1, Inequality (54), p. 117]). If n is a highly composite number, then
a1 ≤ 2(c − 1) where c = log2 pk+1 .
Proof. We prove this theorem by contrapositive. Suppose that n is a number whose
a a a
prime-power factorization is p11 p22 · · · pk k where a1 ≥ 2c − 1 with c = log2 pk+1 .
pk+1
Let m = 2c n. Since p1 = 2, we have that the prime-power factorization of m
a −c a a a
is p11 p22 p33 · · · pk k pk+1 . As c > log2 pk+1 , we have 2c > pk+1 , and so m < n.
Since a1 ≥ 2c − 1, we get a1 + 1 ≤ 2(a1 − c + 1). Then τ (n) = ki=1 (ai + 1) ≤
2(a1 − c + 1)ki=2 (ai + 1) = τ (m). Thus n is not a highly composite number. There-
fore, if n is a highly composite number, then a1 ≤ 2(c − 1) where c = log2 pk+1 .
Theorem 2 ([1, Inequality (34), p. 114]). If n is a highly composite number, then
ak ≤ 2.
Proof. We prove this theorem by contrapositive. Suppose n is a number whose prime-
a a a p
power factorization is p11 p22 · · · pk k where ak ≥ 3. Let m = pk+1
2 n. By Bertrand’s pos-
k
tulate, pk+1 is between pk and pk2 , and so m < n. Since ak ≥ 3, we get ak + 1 ≤
2(ak − 1). Then τ (n) = ki=1 (ai + 1) ≤ 2(ak − 1)k−1
i=1 (ai + 1) = τ (m). Thus n is
not a highly composite number. Therefore, if n is a highly composite number, then
ak ≤ 2.
doi.org/10.1080/00029890.2019.1632631
MSC: Primary 11A25
740
c THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 126
Lemma 1. Let 3 ≤ p < q < r be consecutive primes. Then r < p 2 .
Proof. Suppose that 3 ≤ p < q < r are consecutive primes. If p = 3, then q = 5 and
r = 7, and so r = 7 < 9 = p 2 . Suppose p ≥ 5. Then by Bertrand’s postulate, there is
a prime between p and 2p and another one between 2p and 4p. As the given primes
are consecutive, r < 4p. Since p ≥ 5, r < 4p < p 2 .
Theorem 3 ([1, Inequality (54), p. 117]). If n is a highly composite number with
k ≥ 3, then ak−1 ≤ 2.
Proof. We prove this theorem by contrapositive. Suppose that n is a number whose
a a a p
prime-power factorization is p11 p22 · · · pk k where k ≥ 3 and ak−1 ≥ 3. Let m = pk+1
2 n.
k−1
Since k ≥ 3, we have that pk−1 ≥ 3, and so by Lemma 1, pk+1 < pk−1 2
. Thus m < n.
Since ak−1 ≥ 3, we have that ak−1 + 1 ≤ 2(ak−1 − 1). Then τ (n) = ki=1 (ai + 1) ≤
2(ak−1 − 1)(ak + 1)k−2 i=1 (ai + 1) = τ (m). Thus n is not a highly composite number.
Therefore, if n is a highly composite number with k ≥ 3, then ak−1 ≤ 2.
Theorem 4 ([1, Inequality (41), p. 115]). If n is a highly composite number with
k ≥ 3, then ak = 1.
Proof. We prove this theorem by contrapositive. Suppose that n is a number whose
a a a
prime-power factorization is p11 p22 · · · pk k where k ≥ 3 and ak = 2. If ai−1 < ai for
some 2 ≤ i ≤ k, then we know that n is not highly composite. Assume that a1 ≥
a2 ≥ · · · ≥ ak−1 ≥ ak . Since Theorem 3 tells us that ak−1 ≤ 2, we must have that
pk+1
ak−1 = 2. Let m = 2p k
n. By Bertrand’s postulate, pk+1 < 2pk , and so m < n. Then
τ (n) = 9i=1 (ai + 1) and τ (m) = 12a1 k−2
k−2
i=2 (ai + 1). In order that τ (n) ≤ τ (m), we
must have that 9(a1 + 1) ≤ 12a1 . This occurs when a1 ≥ 3. Thus if a1 ≥ 3, then n is
not a highly composite number.
Since ak = 2, we now only have to consider the case when a1 = 2, which then
d
gives us that each of the exponents is 2. Now we let m = p2 k n where d = log2 pk .
As d < log2 pk , we have that 2d < pk , and so m < n. Then τ (n) = 3k and τ (m) =
2(d + 3) · 3k−2 . Since k ≥ 3, pk ≥ 5. Thus d ≥ 2, and so we have 2(d + 3) ≥ 10.
Hence τ (m) > 9 · 3k−2 = 3k = τ (n). Hence if a1 = 2, then n is not a highly composite
number.
Therefore, if n is a highly composite number with k ≥ 3, then ak = 1.
ACKNOWLEDGMENTS. I would like to thank my colleague Dr. Blair Madore for suggesting that I submit
this note for publication. I would like to thank the reviewers and editors for their valuable comments and
suggestions.
REFERENCES
[1] Ramanujan, S. (1915). Highly composite numbers. Proc. Lond. Math. Soc. (2). 14: 347–409.
[2] Rosen, K. H. (2005). Elementary Number Theory and Its Applications. Boston, MA: Addison-Wesley.
SUNY Potsdam, 44 Pierrepont Ave., Potsdam, NY 13676
[email protected]October 2019] NOTES 741
