\documentclass[11pt,letterpaper]{article} \usepackage[T1]{fontenc} \usepackage{lmodern} \usepackage[left=.95in,right=.95in,top=.89in,bottom=.91in]{geometry} \usepackage{amsmath,amssymb,amsthm,mathtools,microtype} \usepackage{xcolor} \usepackage[colorlinks=true,linkcolor=blue!40!black,citecolor=blue!40!black, urlcolor=blue!40!black]{hyperref} \hypersetup{pdftitle={The alternating sum of reciprocal Fibonacci cubes is irrational}, pdfauthor={twinheterodox},pdfsubject={Number theory}} \newtheorem{theorem}{Theorem}[section] \newtheorem{corollary}[theorem]{Corollary} \newtheorem{proposition}[theorem]{Proposition} \newtheorem{lemma}[theorem]{Lemma} \numberwithin{equation}{section} \newcommand{\Q}{\mathbb Q} \newcommand{\Z}{\mathbb Z} \newcommand{\Ord}{\mathcal O} \DeclareMathOperator{\Nm}{N} \title{\fontsize{16}{20}\selectfont\mbox{The alternating sum of reciprocal Fibonacci cubes is irrational}} \author{twinheterodox} \date{18 November 2026} \begin{document} \maketitle \begin{abstract} We prove that the alternating sum of the reciprocal cubes of the Fibonacci numbers is irrational. The result extends to the Lucas sequences of the first kind associated with real quadratic units greater than one, including the Pell numbers. For every such sequence and every fixed positive integer \(k\), the reciprocal cubes with indices \(k,2k,3k,\ldots\), taken with signs \(+,-,+,\ldots\), also have an irrational sum. \end{abstract} \section{The alternating sum} Let \(F_1=F_2=1\) and \(F_{n+2}=F_{n+1}+F_n\). Write \[ A_F^+=\sum_{n\ge1}\frac{1}{F_n^3},\qquad A_F^-=\sum_{n\ge1}\frac{(-1)^{n-1}}{F_n^3}. \] Absolute convergence gives \(A_F^++A_F^-=2\sum_{n\ge1}F_{2n-1}^{-3}\). Duverney, Ke.~Nishioka, Ku.~Nishioka and Shiokawa proved that the odd-index sum is transcendental \cite{DNNS}\cite[Theorem~4, p.~95]{Kyoto}. Thus at least one of \(A_F^+\) and \(A_F^-\) is transcendental. Krisman proved \(A_F^+\) irrational \cite{KFib} and extended that result to ordinary reciprocal cube sums of Lucas sequences with a positive dominant unit \cite[Theorem~1.1]{KLucas}. These results leave open whether \(A_F^-\) is rational \cite[p.~3]{KLucas}. For integers \(P,Q\), let \(U_n=U_n(P,Q)\) be the Lucas sequence of the first kind, \[ U_0=0,\qquad U_1=1,\qquad U_{n+2}=P U_{n+1}-Q U_n. \] \begin{theorem}\label{thm:main} Suppose that \(Q=-1\) and \(P\ge1\), or that \(Q=1\) and \(P\ge3\). Then \[ A^-_{P,Q}=\sum_{n\ge1}\frac{(-1)^{n-1}}{U_n(P,Q)^3} \quad\text{is irrational}. \] \end{theorem} \noindent Changing the sign of the trace gives, directly from the recurrence, \[ U_n(-P,Q)=(-1)^{n-1}U_n(P,Q)\qquad(n\ge1). \] Cubing preserves this sign. Thus \(A^-_{P,Q}\) is the ordinary reciprocal cube sum for trace \(-P\). In the Fibonacci case \((P,Q)=(1,-1)\), the negative-trace sequence begins \(1,-1,2,-3,5,\ldots\). Its dominant root is negative, explaining the signed reciprocal root in Section~\ref{sec:moments}. The pair \((2,-1)\) gives the Pell case. For \(Q=-1\), Elsner, Shimomura and Shiokawa treat alternating reciprocal \emph{even} powers of first- and second-kind Lucas sequences \cite[Theorems~2 and~4]{ESS}. Here the power is three. \begin{corollary}\label{cor:stride} Under the hypotheses of Theorem~\ref{thm:main}, for each integer \(k\ge1\), \[ \sum_{n\ge1}\frac{(-1)^{n-1}}{U_{kn}(P,Q)^3} \quad\text{is irrational}. \] \end{corollary} The determinant method produces polynomials in a single unknown value. Under a rationality assumption, clearing their denominators gives algebraic integers in a real quadratic field. A nonzero algebraic integer has absolute norm at least one, so estimates at \emph{both} real embeddings can yield a contradiction. Zudilin used polynomial Hankel determinants in an irrationality criterion \cite[Proposition~2]{Zudilin}. Uvarov related the orthogonal polynomials before and after a rational change of weight \cite[\S1]{Uvarov}. Krattenthaler gave a determinant identity for the modified moments \cite[Theorem~1]{Kratt}. The particular moving denominator and interpolation argument used here come from the \(q\)-zeta construction \cite[Sections~2 and~5]{QZeta} and Krisman's adaptation of it to reciprocal cubes \cite{KFib,KLucas}. The alternating sign amounts to changing the sign of the small unit in Krisman's construction. His cyclotomic denominator is an identity in a formal variable, so it still clears every coefficient. The main analytic point is at the conjugate embedding. The conjugate pole values turn out to be evaluations of a \emph{single} holomorphic function formed from the original signed moments. A contour integral combines them before their sizes are estimated. For a determinant of size \(K\), measure logarithms in units of \(K^3\log\eta\), where \(\eta>1\) is the dominant unit. The first embedding saves \(2/9\), the conjugate embedding saves \(1\), and clearing the denominators costs at most \(4393/3675\). The margin is \begin{equation}\label{eq:margin} \frac29+1-\frac{4393}{3675}=\frac{296}{11025}. \end{equation} We prove the estimates for the actual signed value and show that infinitely many of the corresponding determinants are nonzero. \section{The signed unit and its moments}\label{sec:moments} Put \(D=P^2-4Q\), \(\eta=(P+\sqrt D)/2\), and \(u=\eta^{-1}\). The two roots of \(x^2-Px+Q\) are \(\eta>1\) and \(Q/\eta\). They are algebraic units. A rational root would be an integer dividing \(Q=\pm1\), so the field \(E=\Q(\sqrt D)\) is quadratic. Conversely, the trace of a real quadratic unit \(\eta>1\) of norm \(Q\) is the integer \(\eta+Q/\eta\), which is positive for \(Q=-1\) and exceeds two for \(Q=1\). This gives exactly the parameter ranges in Theorem~\ref{thm:main}. Let \(\sigma\) denote the nontrivial automorphism of \(E\) and put \[ \tau=-u,\qquad w=Q\tau^2,\qquad \rho=Q\tau. \] Here \(|w|<1\), \(|\rho|<1\), \(w=\rho\tau\), and \(\rho w=\tau^3\). Changing \(\tau\) from \(u\) to \(-u\) fixes the nodes \(w^n\) and multiplies the weights \(\rho^n\) by \((-1)^n\). Binet's formula gives \(U_n=\eta^n(1-w^n)/\sqrt D\). Thus all the \(U_n\) with \(n\ge1\) are positive and \begin{equation}\label{eq:period} \psi_N=\frac{\tau^{3N}}{(1-w^N)^3},\qquad S_a=\sum_{N=1}^{a}\psi_N,\qquad \Xi=\sum_{N\ge1}\psi_N,\qquad A^-_{P,Q}=-D\sqrt D\,\Xi. \end{equation} Here \(\tau^{3N}=(-1)^N\eta^{-3N}\), which accounts for the minus sign in the formula for \(A^-_{P,Q}\). The series is absolutely convergent. The finite quantities satisfy \begin{equation}\label{eq:conj} \sigma\tau=Q/\tau,\qquad \sigma w=w^{-1},\qquad \sigma\rho=\tau^{-1},\qquad \sigma\psi_N=-\psi_N . \end{equation} For the last identity, factor \(-w^{-N}\) out of \(1-w^{-N}\). We shall use \(\sigma\Xi=-\Xi\) only under the hypothesis \(A^-_{P,Q}=v/m\), \(v\in\Z,\ m\ge1\). In that case \begin{equation}\label{eq:rational} mD^2\Xi=-v\sqrt D\in\Ord=\Z[\eta], \qquad \sigma\Xi=-\Xi , \end{equation} since \(\sqrt D=2\eta-P\). This argument uses the value of the convergent series as a whole. We want polynomial moments in \(E\) and simple-pole moments that recover tails of \(\Xi\). The Euler operator \(\theta=z\,d/dz\) supplies the cubed denominator in \eqref{eq:period} through the identity \(\theta(\theta+1)(1-cz)^{-1}=2cz/(1-cz)^3\). At the node \(w^n\), weighting the numerator by \(\rho^n\) supplies \((\rho w)^n=\tau^{3n}\). Thus for \(g\) holomorphic on a neighborhood of \(|z|\le |w|\), set \[ L(g)=\sum_{n\ge1}\rho^n[\theta(\theta+1)g](w^n). \] The derivatives of \(g\) are bounded on the indicated disk, so the sum converges absolutely. With \(\alpha_j=\rho w^j\), summing gives \begin{align} L(z^j)&=\nu_j=\frac{j(j+1)\alpha_j}{1-\alpha_j} &&(j\ge0),\label{eq:monomials}\\ L\big((1-w^az)^{-1}\big)&=2\rho^{-a}(\Xi-S_a) &&(a\ge1).\label{eq:poles} \end{align} Indeed, in the second identity put \(N=n+a\) and use \(w^a\tau^{-3a}=\rho^{-a}\). This is the signed specialization of Krisman's moment identities \cite[Section~2]{KLucas}. For \(K\ge2\) let \(D_K(z)=\prod_{a=1}^K(1-w^a z)\). On rational functions of the form \(p/D_K\), define \(L_X\) by ordinary partial fractions. Use \eqref{eq:monomials} on the polynomial quotient and \eqref{eq:poles} with \(\Xi\) replaced by \(X\) on the simple fractions. Set \begin{equation}\label{eq:det} \Delta_{K,\tau}(X) =\det_{0\le i,j0\). The product rule and \(\sum_n|\rho|^n<\infty\) now give \[ |L(V_iV_j/D_K)| \le C(i+j+1)^2\,2^{i+j}|w|^{b_i+b_j}. \] Extract \(|w|^{b_i}\) and \(|w|^{b_j}\) from the row and column bounds. The remaining factors contribute \(O(K^2)\) to the logarithm by Hadamard's inequality. As \(b_i=i^2/6+O(i)\), we have \(2\sum_{i0\), and \(\Gamma\) is bounded there. Thus \eqref{eq:contour} is bounded independently of \(j,K\). This uses the signed analogue of the common pole function in \cite[Proposition~5.1]{KLucas}. It remains to bound the polynomial quotient when \(j\ge K\). On expanding \(1/F_K(z)\) at infinity its coefficients after the leading \(z^{-K}\) are \(h_\ell=h_\ell(r_1,\ldots,r_K)\), defined by \(\sum_{\ell\ge0}h_\ell t^\ell=\prod_{a=1}^K(1-r_a t)^{-1}\). The polynomial quotient is \(\sum_{\ell=0}^{j-K}h_\ell z^{j-K-\ell}\), and \(\sum_{\ell\ge0}|h_\ell|\le\prod_{a\ge1}(1-|w|^a)^{-1}\). Also \(\sigma\alpha_i=Q/\alpha_i\), so \(\sigma\nu_i=Q i(i+1)/(\alpha_i-Q)\) is \(O((i+1)^2)\), using \(|\alpha_i|\le u<1\). Consequently the entire moment \(Y_j\) is \(O((K+1)^2)\). Returning to \eqref{eq:denconj} multiplies the absolute determinant of the \(Y_{i+j}\) by \(|w|^{K^2(K+1)/2}\). Hadamard's inequality and \(|w|=\eta^{-2}\) give the result. \end{proof} \section{Nonvanishing at arbitrarily large sizes}\label{sec:nonzero} The Cauchy transform of \(L\) is the function from Lemma~\ref{lem:gamma} in a reciprocal variable. For \(|z|>|w|\), \[ \mathcal C(z)=L_s\big((z-s)^{-1}\big) =\frac{1}{z}L_s\big((1-s/z)^{-1}\big) =\frac{\Gamma(1/z)}{z}. \] The points \(r_a=w^a\) of Section~\ref{sec:second} correspond to \(x_a=w^{-a}\). These exterior points are the zeros of \(D_K\) for \(1\le a\le K\), and \(|x_a|\to\infty\). Kronecker proved that the generating function of a fixed moment sequence is rational exactly when its Hankel determinants eventually vanish \cite[Arts.~VII and~IX]{Kronecker}. Zudilin recalls this criterion in the irrationality setting \cite[Section~5]{Zudilin}. Here the moments change with \(K\), but every zero of \(D_K\) is also a zero of \(D_{K+1}\). We use these shared points in a rational interpolation argument, following \cite[Proposition~5.1]{QZeta} and \cite[Proposition~6.1]{KLucas}, for the actual signed functional. \begin{proposition}\label{prop:nonzero} For arbitrarily large \(K\), one has \(\Delta_{K,\tau}(\Xi)\ne0\). \end{proposition} \begin{proof} The series for \(\Gamma\) in Lemma~\ref{lem:gamma} gives the continuation \[ \mathcal C(z)=\sum_{n\ge1}\frac{2\rho^n w^n z}{(z-w^n)^3}. \] The series converges normally on compact subsets away from the nodes and their accumulation point \(0\). Its cubic pole coefficient at \(w^n\) is \(2\rho^n w^{2n}\ne0\). It follows that \(\mathcal C\) has infinitely many poles and is not rational. Suppose now that all determinants from some size onward vanish. At each such size \(k\), take a nonzero polynomial \(R_k\) of degree \(1\) and \(m,D\) are fixed. This proves irrationality. \end{proof} \begin{proof}[Proof of Corollary~\ref{cor:stride}] Write \(\bar\eta=Q/\eta\) and \(P_k=\eta^k+\bar\eta^k\). This is an integer, by the recurrence for the traces of the powers of \(\eta\). Binet's formula gives \[ U_{kn}(P,Q)=U_k(P,Q)\,U_n(P_k,Q^k). \] If \(Q^k=1\), then \(P_k=\eta^k+\eta^{-k}>2\). If \(Q^k=-1\), then \(P_k=\eta^k-\eta^{-k}>0\). Thus the new pair is admissible. Apply Theorem~\ref{thm:main} and divide by the nonzero integer \(U_k^3\). For Fibonacci, \(U_2(1,-1)=1\) and \(P_2=3\), so \(F_{2n}=U_n(3,1)\). This gives the irrational sum \(F_2^{-3}-F_4^{-3}+F_6^{-3}-\cdots\). \end{proof} \section*{Verification} Another AI model independently audited the mathematics and exposition of the 18 November 2026 working proof, including the fixed-stride result. Two independent reviews subsequently examined the exposition of the seven-page manuscript of the same date. \clearpage \begin{thebibliography}{99} \bibitem{DNNS} D.~Duverney, Ke.~Nishioka, Ku.~Nishioka and I.~Shiokawa, Transcendence of Rogers--Ramanujan continued fraction and reciprocal sums of Fibonacci numbers, \emph{Proc.\ Japan Acad.\ Ser.~A Math.\ Sci.} \textbf{73} (1997), 140--142. \href{https://doi.org/10.3792/pjaa.73.140}{doi:10.3792/pjaa.73.140}. \bibitem{Kyoto} D.~Duverney, Ke.~Nishioka, Ku.~Nishioka and I.~Shiokawa, Transcendence of Rogers--Ramanujan continued fraction and reciprocal sums of Fibonacci numbers, \emph{RIMS K\^oky\^uroku} \textbf{1060} (1998), 91--100. \href{https://www.kurims.kyoto-u.ac.jp/~kyodo/kokyuroku/contents/pdf/1060-10.pdf\#page=5}{Theorem~4, p.~95}. \bibitem{ESS} C.~Elsner, S.~Shimomura and I.~Shiokawa, Algebraic relations for reciprocal sums of Fibonacci numbers, \emph{Acta Arith.} \textbf{130} (2007), 37--60. \href{https://doi.org/10.4064/aa130-1-3}{doi:10.4064/aa130-1-3}. \bibitem{KFib} S.~Krisman, The sum of the reciprocal cubes of the Fibonacci numbers is irrational, \href{https://hexagonmath.org/2611.00473v1}{Hexagon 2611.00473v1} (2026). \bibitem{KLucas} S.~Krisman, The reciprocal cube sums of Lucas sequences of quadratic units are irrational, \href{https://hexagonmath.org/2611.00491v1}{Hexagon 2611.00491v1} (2026). \bibitem{Zudilin} W.~Zudilin, A determinantal approach to irrationality, \emph{Constr.\ Approx.} \textbf{45} (2017), 301--310. \href{https://doi.org/10.1007/s00365-016-9333-7}{doi:10.1007/s00365-016-9333-7}. \bibitem{Uvarov} V.~B.~Uvarov, The connection between systems of polynomials orthogonal with respect to different distribution functions, \emph{USSR Comput.\ Math.\ Math.\ Phys.} \textbf{9} (1969), no.~6, 25--36. \href{https://doi.org/10.1016/0041-5553(69)90124-4}{doi:10.1016/0041-5553(69)90124-4}. \bibitem{Kratt} C.~Krattenthaler, A determinant identity for moments of orthogonal polynomials that implies Uvarov's formula for the orthogonal polynomials of rationally related densities, \href{https://arxiv.org/abs/2103.03969}{arXiv:2103.03969} (2021). \bibitem{Kronecker} L.~Kronecker, Zur Theorie der Elimination einer Variabeln aus zwei algebraischen Gleichungen, \emph{Monatsber.\ K\"onigl.\ Preuss.\ Akad.\ Wiss.\ Berlin} (1881), 535--600. \href{https://www.e-rara.ch/download/pdf/5761162.pdf\#page=42}{Art.~IX in the collected works}. \bibitem{QZeta} uplift-universal-inc, Irrationality of the Lambert-series value \(\zeta_{1/p}(3)\), research draft (2026). \href{https://github.com/uplift-universal-inc/qzeta_3-irrationality}{Manuscript and accompanying sources}. \bibitem{checks} twinheterodox, Verification programs accompanying \emph{The alternating sum of reciprocal Fibonacci cubes is irrational} (2026). Included with the source as electronic supplementary material. \href{https://github.com/twinheterodox/alternating-lucas-cubes}{Project repository address}. \end{thebibliography} \end{document}