Preprint
Article

This version is not peer-reviewed.

On Series Involving Cubed Catalan Numbers

Submitted:

01 April 2026

Posted:

02 April 2026

You are already at the latest version

Abstract
Using generalized binomial coefficient identities and some results of John Dougall, we derive some families of series involving the cubes of Catalan numbers. We also establish a family of series containing fourth powers of Catalan numbers. Finally, we find a generalization of the Bauer series for \( 1/\pi \) and obtain some Ramanujan-like series for \( 1/\pi^2 \) and~\( 1/\pi^3 \).
Keywords: 
;  ;  ;  ;  

1. Introduction

Catalan numbers are defined for non-negative integers j by
C j = 1 j + 1 2 j j ,
and obey the following recurrence relation:
C j + 1 = 2 ( 2 j + 1 ) j + 2 C j , C 0 = 1 .
Using Dixon’s theorem, Tauraso [5] found the following identity involving cubed Catalan numbers:
k = 0 C k 2 2 k 3 = 8 384 π Γ 4 1 4 ,
where Γ ( z ) is the Gamma function.
In this paper, using identities involving generalized binomial coefficients and results due to Dougall [3] we will derive (3) and the following series having a similar nature:
k = 0 C k 2 2 k k + 2 3 = 152 27 80 81 π 3 Γ 4 1 4 ,
k = 0 ( 1 ) k C k 2 2 k 3 4 k + 3 = 8 16 π ,
k = 0 ( 1 ) k C k 2 2 k k + 2 3 4 k + 5 = 8 9 + 128 27 π ,
k = 0 C k 2 2 k 3 4 k + 3 = 8 + 2 π 3 Γ 4 1 4 ,
and
k = 0 C k 2 2 k k + 2 3 4 k + 5 = 136 9 7168 9 π Γ 4 1 4 .
In fact, each of identities (3)–(8) occurs as a particular member of a family of series stated in Corollaries 1 and 2 and Theorem 3. For instance, (3) is the m = 0 case of the family stated in Corollary 1, namely,
k = 0 C k 2 2 k j = 1 2 m 1 2 k 2 j + 1 3
= 2 6 m + 6 2 m + 2 ! 3 C 2 m + 1 3 1 48 4 m + 1 2 ( 1 ) m π Γ 4 1 4 j = 1 3 m 4 j 1 j = 1 m 4 j 3 3 .
We will also derive families of series involving cubed Catalan numbers and odd harmonic numbers including the following special cases:
k = 0 C k 2 2 k 3 O k = 8 + 64 π π 10 Γ 4 1 4 ,
k = 0 C k 2 2 k k + 2 3 O k = 392 81 8 729 15 π + 32 π 3 Γ 4 1 4 ,
k = 0 ( 1 ) k C k 2 2 k 3 4 k + 3 O k + 1 3 = 16 π 3 3 π ,
and
k = 0 ( 1 ) k C k 2 2 k k + 2 3 4 k + 5 O k + 1 3 = 32 16 5 π 81 π .
As additional results, we will derive a family of series involving fourth powers of Catalan numbers and one involving these numbers and odd harmonic numbers, of which the following are particular cases:
k = 0 C k 2 2 k 4 4 k + 3 = 16 128 π 2 ,
k = 0 C k 2 2 k k + 2 4 4 k + 5 = 176 27 + 16384 243 π 2 ,
k = 0 C k 2 2 k 4 4 k + 3 O k + 1 4 = 12 192 π 2 + 32 2 ln 2 + 1 π 2 ,
and
k = 0 C k 2 2 k k + 2 4 4 k + 5 O k + 1 4 = 220 27 + 75776 729 π 2 8192 243 ln 2 π 2 .
A high point of this paper is the discovery of a family of Ramanujan-like series (Theorem 8):
k = 0 ( 1 ) k 2 2 k 2 k k j = 1 m 1 2 k 2 j + 1 3 4 k 2 m + 1 = m ! 2 m 2 m m 3 2 π , m = 0 , 1 , 2 , ,
which gives the Bauer series [1] at m = 0 :
k = 0 ( 1 ) k 2 2 k 2 k k 3 4 k + 1 = 2 π .
We will establish the following families of Ramanujan-like series for 1 / π 2 and 1 / π 3 :
k = 0 1 2 2 k 2 k k j = 1 m 1 2 k 2 j + 1 4 4 k 2 m + 1 = ( 1 ) m 2 8 m ( m ! ) 4 π 2 m 2 m m 5 , m = 1 , 2 , 3 , ,
and
k = 0 1 2 2 k 2 k k j = 1 2 m 1 2 k 2 j + 1 3
= ( 2 m ) ! 2 2 m 4 m 2 m 3 ( 1 ) m 4 m 1 2 Γ 4 ( 1 / 4 ) 4 π 3 j = 0 3 m 1 4 j 3 j = 0 m 1 4 j 1 3 , m = 0 , 1 , 2 , 3 , ;
with the special cases
k = 0 1 2 2 k 2 k 1 2 k k 4 4 k 1 = 8 π 2
and
k = 0 1 2 2 k 2 k k 3 = Γ 4 1 / 4 4 π 3 .
Binomial coefficients are defined, for non-negative integers i and j, by
i j = i ! j ! ( i j ) ! , i j ; 0 , i < j .
The definition is extended to complex numbers r and s by
r s = Γ ( r + 1 ) Γ ( s + 1 ) Γ ( r s + 1 ) ,
where the Gamma function, Γ ( z ) , is defined for ( z ) > 0 by
Γ ( z ) = 0 e t t z 1 d t = 0 log ( 1 / t ) z 1 d t ,
and is extended to the rest of the complex plane, excluding the non-positive integers, by analytic continuation.
Harmonic numbers, H j , and odd harmonic numbers, O j , are defined for non-negative integers j by
H j = k = 1 j 1 k , O j = k = 1 j 1 2 k 1 , H 0 = 0 = O 0 .
The identity
H n 1 / 2 = 2 O n 2 ln 2 ,
extends harmonic numbers to half-integer arguments and is a consequence of the link between harmonic numbers, odd harmonic numbers and the digamma function.

2. Required Results

Series involving cubed Catalan numbers are given in Theorems 1, 2 and 3 and Corollaries 1 and 2. We require the results stated in Lemmata 1–6.
Lemma 1.
If k and r are non-negative integers, then
r + 1 / 2 k + 1 = ( 1 ) r r + 2 ! C r + 1 2 r + 2 ( 1 ) k C k 2 2 k j = 1 r 1 2 k 2 j + 1 .
Proof. 
This identity and similar ones can be derived using (27) together with the identities
Γ z + 1 2 = π 2 2 z 2 z z Γ z + 1
and
Γ z + 1 2 = ( 1 ) z 2 2 z 2 z z 1 π Γ z + 1 ,
and the Pochhammer relation
Γ z + n Γ z = j = 1 n ( z + j 1 ) = ( z ) n ,
for n a non-negative integer and z a complex number. □
Lemma 2.
If m is an integer, then
Γ 1 4 6 m + 7 Γ 3 1 4 2 m + 5 = 48 2 m + 1 2 Γ 3 4 2 m + 1 Γ 3 1 4 2 m + 1
and
Γ 1 4 6 m + 5 Γ 3 1 4 2 m + 3 = 12 6 m + 1 2 m 1 2 Γ 3 4 2 m 1 Γ 3 1 4 2 m 1 .
Proof. 
Use the recurrence relation Γ ( u ) = u 1 Γ ( u 1 ) . □
Lemma 3.
If m is an integer, then
Γ m + 1 4 = 1 4 m Γ 1 4 j = 1 m 4 j 3
and
Γ m + 3 4 = 1 4 m π 2 Γ 1 4 j = 1 m 4 j 1 .
Proof. 
These follow immediately from (34) and the fact that
Γ 3 4 = π 2 Γ 1 4 ,
which is a consequence of Euler’s reflection formula:
Γ ( z ) Γ ( 1 z ) = π sin ( π z ) .
Lemma 4.
If m is an integer, then
cos π 4 2 m + 1 = ( 1 ) m / 2 2 ,
sin π 4 2 m + 1 = ( 1 ) m + m / 2 2 ,
cos π 4 2 m 1 = ( 1 ) m / 2 2 ,
and
sin π 4 2 m 1 = ( 1 ) m / 2 2 ;
where, as usual, x is the greatest integer less than or equal to x and x is the smallest integer greater than or equal to x.
Proof. 
We prove (41). We have
cos π 4 2 m + 1 = cos π m 2 cos π 4 sin π m 2 sin π 4
= 1 2 cos π m 2 sin π m 2
= 1 2 cos π m 2 , if m is even ; sin π m 2 , if m is odd .
Thus,
cos π 4 2 m + 1 = 1 2 ( 1 ) m / 2 , if m is even ; ( 1 ) ( m + 1 ) / 2 , if m is odd .
= ( 1 ) m / 2 2 .
Lemma 5.
If n is an integer, then O n = O n .
Proof. 
We have
O n = j = 1 n 1 2 j 1 = j = 1 n 0 1 2 j 1 = j = 1 n 1 2 j 2 n 1 .
Thus,
O n = j = 1 n 1 2 n 2 j + 1 = j = 1 n 1 2 n 2 n j + 1 + 1 = j = 1 n 1 2 j 1 = O n .
Each identity in Lemma 6 is a slight re-writing of the original one derived by Dougall [3].
Lemma 6.
If x is a complex number, then
k = 0 ( 1 ) k x k + 1 3 = 1 Γ 1 2 3 x + 2 Γ 3 1 2 x + 2 cos π x 2 , x > 2 / 3 ,
k = 0 x k + 1 3 2 k + 2 x = x sin ( π x ) π , x 1 / 3 ,
and
k = 0 ( 1 ) k x k + 1 3 2 k + 2 x = x + sin π x π Γ 1 2 1 x Γ 1 2 1 + 3 x Γ 2 1 2 1 + x .
Remark 1.
Identities (52)–(54) are also recorded in [4].

3. Series Involving Cubed Catalan Numbers

Theorem 1.
If m is a non-negative integer, then
k = 0 C k 2 2 k j = 1 m 1 2 k 2 j + 1 3
= ( 1 ) m 2 3 m + 6 m + 2 ! 3 C m + 1 3 1 24 2 2 m + 1 2 ( 1 ) m / 2 Γ 3 4 2 m + 1 Γ 3 1 4 2 m + 1 .
Proof. 
Set x = m + 1 / 2 in (52) to obtain
k = 0 ( 1 ) k m + 1 / 2 k + 1 3 = 1 Γ 1 4 6 m + 7 Γ 3 1 4 2 m + 5 cos π 4 2 m + 1 ,
and hence (56) upon using Lemmata 1, 2 and 4. □
Corollary 1.
If m is a non-negative integer, then
k = 0 C k 2 2 k j = 1 2 m 1 2 k 2 j + 1 3
= 2 6 m + 6 2 m + 2 ! 3 C 2 m + 1 3 1 48 4 m + 1 2 ( 1 ) m π Γ 4 1 4 j = 1 3 m 4 j 1 j = 1 m 4 j 3 3
and
k = 0 C k 2 2 k j = 1 2 m 1 1 2 k 2 j + 1 3
= 2 6 m + 3 2 m + 1 ! 3 C 2 m 3 1 ( 1 ) m 4 m 1 2 3 4 π 3 Γ 4 1 4 j = 1 3 m 1 4 j 3 j = 1 m 1 4 j 1 3 .
Proof. 
Write 2 m for m and 2 m 1 for m, in turn, in (56), and apply Lemma 3. □
Remark 2.
Identities (3) and (4) are obtained by evaluating (59) at m = 0 and (61) at m = 1 .
Theorem 2.
If m is a non-negative integer, then
k = 0 ( 1 ) k C k 2 2 k j = 1 m 1 2 k 2 j + 1 3 4 k 2 m + 3
= ( 1 ) m 2 3 m + 6 m + 2 ! 3 C m + 1 3 2 m + 1 ( 1 ) m 2 π .
Proof. 
Use of r = m + 1 / 2 in (53) gives
k = 0 m + 1 / 2 k + 1 3 4 k 2 m + 3 = 2 m + 1 + ( 1 ) m + 1 2 π ,
and hence (63) by Lemma 1. □
Remark 3.
Identities (5) and (6) are obtained at m = 0 and m = 1 in (63).
Theorem 3.
If m is a non-negative integer, then
k = 0 C k 2 2 k j = 1 m 1 2 k 2 j + 1 3 4 k 2 m + 3
= ( 1 ) m 1 2 3 m + 6 m + 2 ! 3 C m + 1 3 2 m + 1 ( 1 ) m + m / 2 24 2 6 m + 1 2 m 1 2 Γ 3 4 2 m 1 Γ 3 1 4 2 m 1 .
Proof. 
Set x = m + 1 / 2 in (54) and apply Lemmata 1, 2 and 4. □
Corollary 2.
If m is a non-negative integer, then
k = 0 C k 2 2 k j = 1 2 m 1 2 k 2 j + 1 3 4 k 4 m + 3
= 2 6 m + 6 2 m + 2 ! 3 C 2 m + 1 3 4 m + 1 ( 1 ) m 12 m + 1 4 m 1 2 3 4 π 3 Γ 4 1 4 j = 1 3 m 1 4 j 3 j = 1 m 1 4 j 1 3
and
k = 0 C k 2 2 k j = 1 2 m 1 1 2 k 2 j + 1 3 4 k 4 m + 5
= 2 6 m + 3 2 m + 1 ! 3 C 2 m 3 4 m 1 + ( 1 ) m 12 m 5 4 m 3 2 48 π Γ 4 1 4 j = 1 3 m 3 4 j 1 j = 1 m 1 4 j 3 3 .
Proof. 
Write 2 m for m and 2 m 1 for m, in turn, in (65), and apply Lemma 3. □
Remark 4.
Identities (7) and (8) correspond to an evaluation of (67) at m = 0 and (69) at m = 1 .

4. Series Involving Cubed Catalan Numbers and Odd Harmonic Numbers

Theorem 4.
If m is a non-negative integer, then
k = m C k 2 2 k j = 1 m 1 2 k 2 j + 1 3 O k m
= k = 0 m C k 2 2 k j = 1 m 1 2 k 2 j + 1 3 O m k + ( 1 ) m 2 3 m + 6 O m + 1 m + 2 ! 3 C m + 1 3
+ ( 1 ) m + m / 2 2 3 m + 7 m + 2 ! 3 C m + 1 3 3 2 2 m + 1 2 Γ 3 4 2 m + 1 Γ 1 4 2 m + 1 3
× ( 1 ) m 1 π 3 + H 3 2 m + 1 4 H 2 m + 1 4 4 O m + 1 .
Proof. 
Write (52) in the equivalent form
k = 0 ( 1 ) k x k + 1 3 = 1 3 x / 2 x / 2 x x / 2 cos π x 2 ,
and differentiate with respect to x to obtain
3 k = 0 ( 1 ) k x k + 1 3 H x H x k 1
= π 2 sin π x 2 3 2 H 3 x / 2 3 2 H x / 2 cos π x 2 Γ 1 2 3 x + 2 Γ 3 1 2 x + 2 .
Set x = m + 1 / 2 and use (30) and Lemma 4 to get
k = 0 ( 1 ) k m + 1 / 2 k + 1 3 O k m
= ( 1 ) m 1 π 3 + H 3 4 2 m + 1 H 1 4 2 m + 1 4 O m + 1 ( 1 ) m / 2 4 2 Γ 1 4 6 m + 7 Γ 3 1 4 2 m + 5 .
Identity (71) now follows upon application of Lemmata 1 and 2. □
Remark 5.
Identities (11) and (12) correspond to evaluating (74) at m = 0 and m = 1 . In deriving (11), note that
H 3 / 4 H 1 / 4 = 1 3 / 4 1 1 / 4 + π cot π 4 = π 8 3 ,
since
H r H 1 r = 1 r 1 1 r + π cot π 1 r ,
when r is not an integer. For (12), note also that
H 9 / 4 = H 5 / 4 + 4 9 = H 1 / 4 + 4 5 + 4 9 = H 1 / 4 + 56 45 ,
since
H r = H r 1 + 1 r ,
when r is not a non-positive integer. Thus
H 9 / 4 H 3 / 4 = H 1 / 4 H 3 / 4 + 56 45 = 8 3 π + 56 45 = 176 45 π .
Theorem 5.
If m is a non-negative integer, then
k = m ( 1 ) k C k 2 2 k j = 1 m 1 2 k 2 j + 1 3 4 k 2 m + 3 O k m + 1 3
= k = 0 m 1 ( 1 ) k C k 2 2 k j = 1 m 1 2 k 2 j + 1 3 4 k 2 m + 3 O m k + 1 3
+ ( 1 ) m 2 3 m + 6 m + 2 ! 3 C m + 1 3 2 m + 1 ( 1 ) m 2 π O m + 1 1 3 .
Proof. 
Differentiate (53) with respect to x to obtain
k = 0 x k + 1 3 2 k + 2 x H x k 1 + 1 3 = H x x sin ( π x ) π 1 3 + cos ( π x ) 3 .
Set x = m + 1 / 2 and use (30) and Lemma 1. □
Remark 6.
Identities (13) and (14) are evaluations of (87) at m = 0 and m = 1 .

5. Series Involving Fourth Powers of Catalan Numbers

In this section, we derive a family of series involving fourth powers of Catalan numbers, based on the identity
k = 0 x k + 1 4 2 k x + 2 = x sin ( π x ) π 2 x x , x > 1 2 ,
which is a variation on [3 Equation (17)] .
Theorem 6.
If m is a non-negative integer, then
k = 0 C k 2 2 k j = 1 m 1 2 k 2 j + 1 4 4 k 2 m + 3
= 2 4 m + 8 m + 2 ! 4 C m + 1 4 2 m + 1 ( 1 ) m 2 4 m + 3 π 2 m + 1 2 m + 1 C m .
Proof. 
At x = m + 1 / 2 , (89) gives
k = 0 m + 1 / 2 k + 1 4 4 k 2 m + 3 = 2 m + 1 ( 1 ) m 4 π 2 m m + 1 / 2 ,
from which (91) follows by (31) and the fact that
2 m m + 1 / 2 = Γ 2 m + 1 Γ m + 3 / 2 Γ m + 1 / 2 = 2 4 m + 1 π m + 1 2 m + 1 C m .
Remark 7.
Identities (15) and (16) are obtained by evaluating (91) at m = 0 and at m = 1 .
Theorem 7.
If m is a non-negative integer, then
k = m C k 2 2 k j = 1 m 1 2 k 2 j + 1 4 4 k 2 m + 3 O k m + 1 4
= k = 0 m 1 C k 2 2 k j = 1 m 1 2 k 2 j + 1 4 4 k 2 m + 3 O m k + 1 4
+ 2 4 m + 8 O m + 1 m + 2 ! 4 C m + 1 4 2 m + 1 + 3 ( 1 ) m + 1 2 4 m + 2 π 2 m + 1 2 m + 1 C m
+ 2 4 m + 8 m + 2 ! 4 C m + 1 4 ( 1 ) m 2 ln 2 + H 2 m + 1 2 4 m + 1 π 2 m + 1 2 m + 1 C m 1 4 .
Proof. 
Differentiating (89) with respect to x gives, after some re-arrangement,
k = 0 x k + 1 4 2 k + 2 x H x k 1 + 1 4
= sin ( π x ) 2 π 2 x x H 2 x 3 H x + x H x 1 4 + 1 4 2 x x cos ( π x ) .
Evaluation at x = m + 1 / 2 , using (30) and some algebra yields
k = 0 m + 1 / 2 k + 1 4 4 k 2 m + 3 O m k + 1 4
= 2 m + 1 + ( 1 ) m + 1 6 π 2 m m + 1 / 2 O m + 1
+ ( 1 ) m π 2 ln 2 + H 2 m + 1 2 m m + 1 / 2 1 4 .
Application of (31) and (93) now produces (97). □
Remark 8.
Identities (17) and (18) correspond to setting m = 0 and m = 1 in (97).

References

  1. Bauer, G. Von den Coefficienten der Reihen von Kugelfunktionen einer Variabeln. Journal für die reine und angewandte Mathematik 1859, 56, 101–121. [Google Scholar]
  2. Chen, H. interesting Ramanujan-like series associated with powers of central binomial coefficients. J. Integer Seq. 2022, 22 22.1.8. [Google Scholar]
  3. Dougall, J. Vandermonde’s theorem and some more general expansions. Proc. Edinb. Math. Soc. 1906, 25, 114–152. [Google Scholar] [CrossRef]
  4. Gould, H.W. Combinatorial Identities; Morgantown, West Virginia, 1972. [Google Scholar]
  5. Tauraso, R. A supercongruence involving cubes of Catalan numbers. Integers 2020, 20, #A44. [Google Scholar]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
Prerpints.org logo

Preprints.org is a free preprint server supported by MDPI in Basel, Switzerland.

Subscribe

Disclaimer

Terms of Use

Privacy Policy

Privacy Settings

© 2026 MDPI (Basel, Switzerland) unless otherwise stated