On a BBP-type formula for π 2 \pi^{2} in the golden ratio base

B. Cloitre · 2025

Read On a BBP-type formula for π 2 \pi^{2} in the golden ratio base by B. Cloitre (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-03743.txt.

Introduction

Benoit Cloitre proved a digit-extraction formula for pi squared in two thousand twenty-five. BBP-type formulas pluck digits of a constant without computing earlier ones. This one works in the golden ratio base. Cloitre had found it by experiment two decades earlier. The proof rests on a geometric identity. That identity ties the golden ratio to fifth roots of unity.

The paper presents the proof in self-contained form. It then extends the same method further. A Machin-like formula for another constant emerges. Its series converge fast and nest in layers. Both formulas also use the golden ratio. Such formulas serve high-precision computation checks. They also reveal hidden links between classic constants.

Catalog record

CREATORB. Cloitre
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE22 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-03743

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On a BBP-type formula for π 2 \pi^{2} in the golden ratio base B. Cloitre Abstract This paper presents a detailed, self-contained proof of a BBP-type formula for π 2 \pi^{2} expressed in the golden ratio base, ϕ \phi . The formula was discovered empirically by the author in 2004. The proof presented herein is built upon a fundamental geometric identity connecting ϕ \phi to the fifth roots of unity, offering an intuitive and direct path to the result. The power of the underlying methodology is then demonstrated by extending it to establish a new, computationally efficient Machin-like formula for ζ ⁡ ( 3 ) \zeta(3) , expressed through rapidly converging, hierarchical series involving the golden ratio. 1 Introduction and Context The search for explicit formulas for mathematical constants has long been a source of fascination. A significant breakthrough in this area came with the discovery of ”BBP-type” formulas, named after Bailey, Borwein, and Plouffe [ 1 ] . A BBP formula for a constant C is a series representation of the form: C=\sum_{k=0}^{\infty}\frac{1}{b^{k}}\sum_{j=1}^{n}\frac{a_{j}}{(nk+j)^{s}} where b is an integer base and a_{j} are integers. Their primary importance lies in their ability to allow the computation of the d -th digit of C in base b without computing the preceding digits. The archetypal example is the formula for \pi : \pi=\sum_{k=0}^{\infty}\frac{1}{16^{k}}\left(\frac{4}{8k+1}-\frac{2}{8k+4}-\frac{1}{8k+5}-\frac{1}{8k+6}\right) The original discovery for \pi sparked a flurry of research. This led to many new BBP-type formulas, first for other weight-1 constants, and later for constants of higher weight, such as \zeta(3) or Catalan’s constant, as extensively documented in Bailey’s compendium [ 3 ] . A parallel line of inquiry explored the existence of such formulas in non-integer algebraic bases. The golden ratio base, \phi=(1+\sqrt{5})/2 , proved to be a particularly fertile ground, yielding formulas for \pi and other weight-1 constants [ 4 ] . This paper addresses a question at the confluence of these two research streams: do BBP-type formulas exist for constants of higher weight in the golden ratio base? It is in this context that in December 2004, while experimenting with the lindep algorithm in PARI/GP, I discovered the novel BBP-type formula for \pi^{2} —a constant of weight 2—that is the subject of this paper. While, as noted, numerous formulas for weight-1 constants have been found in the golden ratio base, those for constants of higher weight remain exceptionally rare. The formula for \pi^{2} was subsequently featured as an exercise in Experimental Mathematics in Action [ 2 ] , where the hints provided are sufficient to construct a proof by specializing a general polylogarithmic identity. In contrast, the demonstration provided herein offers a more fundamental and intuitive perspective. It originates from a single geometric identity (Lemma 2 ), which, through the properties of the dilogarithm, establishes a direct connection between a remarkable feature of the regular pentagon and the formula for \pi^{2} . The entire proof is thus constructed from first principles, highlighting the underlying algebraic structure in a self-contained way. Finally, the investigation was extended to \zeta(3) , a constant of weight 3. While the experimental search for a pure BBP-type formula for \zeta(3) in the golden ratio base did not succeed, it led to the discovery of a different, noteworthy result: a non-linear Machin-like formula for \zeta(3) . This identity, which expresses \zeta(3) as a polynomial of rapidly converging series in terms of \phi , is established here as a valuable and computationally efficient result in its own right. 2 Mathematical Preliminaries To make this paper self-contained, this section gathers the key definitions and identities used. 2.1 The Polylogarithm Function The polylogarithm function, whose study was initiated by Euler and later developed by Landen and Lewin, provides a bridge between series, integrals, and special values of functions like the Riemann zeta function. For instance, \mathrm{Li}_{2}(1)=\sum_{k=1}^{\infty}1/k^{2}=\zeta(2)=\pi^{2}/6 . These functions also appear in various contexts in theoretical physics, such as in the calculation of Feynman diagrams. The polylogarithm of order s , denoted \mathrm{Li}_{s}(z) , is defined for any complex number z with |z|<1 by the power series: \mathrm{Li}_{s}(z)=\sum_{k=1}^{\infty}\frac{z^{k}}{k^{s}} For s=2 , this is the dilogarithm , and for s=3 , the trilogarithm . Lemma 1 (Functional Equations for the Dilogarithm) . For |z|\leq 1 and z not 0 or 1 , the dilogarithm satisfies the following functional equations, which are standard results found in [ 5 ] : 1. Landen’s Identity (Reflection Formula): \mathrm{Li}_{2}(z)+\mathrm{Li}_{2}(1-z)=\frac{\pi^{2}}{6}-\ln(z)\ln(1-z) (1) 2. Inversion Formula: \mathrm{Li}_{2}(z)+\mathrm{Li}_{2}(1/z)=-\frac{\pi^{2}}{6}-\frac{1}{2}\ln^{2}(-z) (2) Here, \ln(z) denotes the principal branch of the complex logarithm. 2.2 The Golden Ratio and the Regular Pentagon The golden ratio, \phi=(1+\sqrt{5})/2 , is deeply connected to the geometry of the regular pentagon. The vertices of a regular pentagon centered at the origin can be represented by the fifth roots of unity, \omega^{k}=e^{2\pi ik/5} for k=0,1,2,3,4 . The geometry of this figure dictates the exact trigonometric values needed for our proof. For instance, the ratio of a diagonal to a side in a regular pentagon is \phi . This property can be used to derive the following standard results [ 6 ] : \cos(2\pi/5)=\frac{\phi-1}{2}=\frac{1}{2\phi}\quad\text{and}\quad\cos(\pi/5)=\frac{\phi}{2} The point z=\phi^{-1}e^{2\pi i/5} , which is central to our proof, lies within this geometric context, as shown in Figure 1 . \Re \Im \omega^{0} \omega^{1} \omega^{2} \omega^{3} \omega^{4} z=\phi^{-1}\omega^{1} Figure 1: The fifth roots of unity \omega^{k}=e^{2\pi ik/5} on the unit circle and the key point z in the complex plane. 3 The main results This paper establishes two main theorems. Theorem 1 (BBP formula for \pi^{2} ) . Let \phi=\frac{1+\sqrt{5}}{2} be the golden ratio. The following identity holds: \frac{\pi^{2}}{50}=\sum_{k=0}^{\infty}\frac{1}{\phi^{5k}}\left(\frac{\phi^{-2}}{(5k+1)^{2}}-\frac{\phi^{-1}}{(5k+2)^{2}}-\frac{\phi^{-2}}{(5k+3)^{2}}+\frac{\phi^{-5}}{(5k+4)^{2}}+\frac{2\phi^{-5}}{(5k+5)^{2}}\right) (3) Theorem 2 (Machin-like formula for \zeta(3) ) . Let \phi=\frac{1+\sqrt{5}}{2} . The constant \zeta(3) can be expressed as the following polynomial of rapidly converging series: \zeta(3)=\frac{5}{4}M_{3}(\phi^{-2})+\Big(M_{2}(\phi^{-1})+M_{2}(\phi^{-2})\Big)M_{1}(\phi^{-2})+\frac{7}{6}\Big(M_{1}(\phi^{-2})\Big)^{3} where the constants M_{s}(w) are defined by hierarchical sums designed for rapid convergence: M_{s}(w)=\sum_{j=0}^{\infty}5^{-sj}\left(\sum_{r=1}^{4}\sum_{k=0}^{\infty}\frac{(w^{5^{j}})^{5k+r}}{(5k+r)^{s}}\right) This series converges for |w|<1 , which is the case for the arguments \phi^{-1} and \phi^{-2} used here. 4 Proof of the \pi^{2} formula (theorem 1) The proof is presented in two stages: we first prove a core geometric identity and its consequence for a polylogarithm series, and then we demonstrate that the BBP formula is an algebraic rearrangement of this result. 4.1 The Geometric and Polylogarithmic Foundation Lemma 2 (A key geometric identity) . Let \phi=(1+\sqrt{5})/2 and z=\phi^{-1}e^{2\pi i/5} . Then, the following identity holds: 1-z=e^{-i\pi/5} . Proof. We prove the identity by showing that the real and imaginary parts of both sides are equal. First, the real part of 1-z : \displaystyle\Re(1-z) \displaystyle=1-\phi^{-1}\cos(2\pi/5)=1-\phi^{-1}\left(\frac{1}{2\phi}\right)=1-\frac{\phi^{-2}}{2} \displaystyle=1-\frac{1-\phi^{-1}}{2}=\frac{2-1+\phi^{-1}}{2}=\frac{1+\phi^{-1}}{2}=\frac{\phi}{2} This is precisely \cos(\pi/5)=\Re(e^{-i\pi/5}) . Next, the imaginary part of 1-z : \Im(1-z)=-\phi^{-1}\sin(2\pi/5) Using the double-angle formula \sin(2\theta)=2\sin\theta\cos\theta : \displaystyle\phi^{-1}\sin(2\pi/5) \displaystyle=\phi^{-1}\left(2\sin(\pi/5)\cos(\pi/5)\right) \displaystyle=\phi^{-1}\left(2\sin(\pi/5)\frac{\phi}{2}\right)=(\phi^{-1}\phi)\sin(\pi/5)=\sin(\pi/5) Thus, \Im(1-z)=-\sin(\pi/5) . Since the real and imaginary parts match, the identity is proven. ∎ Geometric Interpretation . This identity expresses a remarkable metric relationship within the geometry of the pentagon. In the complex plane, the numbers 0,1,z form a triangle. The identity 1-z=e^{-i\pi/5} states that the vector from z to 1 has length 1 and makes an angle of -\pi/5 with the real axis. This specific configuration is a direct consequence of the golden ratio’s role in the pentagon’s construction. \Re \Im 0 1 z 1-z e^{-i\pi/5} Figure 2: Geometric visualization of the identity 1-z=e^{-i\pi/5} . Lemma 3 . Let \phi=\frac{1+\sqrt{5}}{2}

Read the complete text