Third-order Jacobsthal 3 3 -Parameter Generalized Quaternions

GAMALIEL MORALES · 2025

Read Third-order Jacobsthal 3 3 -Parameter Generalized Quaternions by GAMALIEL MORALES (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-2504-03646.txt.

Introduction

Gamaliel Morales, writing alone in two thousand twenty-five, joins two known math objects into a new, large number system. One is a set of third-order counts in the Jacobsthal line. The other is a broad form of quaternions set by three terms.

Morales sets out the new system and works through its core rules. He finds its step-by-step rule for the next term, and closed forms that give each term at once. He also gives a way to make all its terms from a seed rule, sums of its terms, and two classic links that tie near terms to each other. The work grows both lines of study at once.

Catalog record

CREATORGAMALIEL MORALES
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE25 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-03646

Browse the full library · About the Tower of Babel

Full text

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

Third-order Jacobsthal 3 3 -Parameter Generalized Quaternions GAMALIEL MORALES Abstract . The purpose of this article is to bring together the third-order Jacobsthal numbers and 3-parameter generalized quaternions, which are a general form of the quaternion algebra according to 3-parameters. With this purpose, we introduce and examine a new type of quite big special numbers system, which is called third-order Jacobsthal 3-parameter generalized quaternions (shortly, third-order Jacobsthal 3PGQs). Further, we compute both some new equations and classical well-known equations such as: linear recurrence, Binet formulas, generating function, sum formulas, Cassini identity and d’Ocagne identity. 1991 Mathematics Subject Classification Mathematical subject classification: 11B37, 11K31, 11R52, 11Y55. Instituto de Matemáticas, Pontificia Universidad Católica de Valparaíso, Blanco Viel 596, Valparaíso, Chile. E-mail: [email protected] 1. Introduction Third-order Jacobsthal numbers and modified third-order Jacobsthal numbers are well-known numbers among integer sequences. Third-order Jacobsthal numbers satisfy the third order recurrence relation J_{0}^{(3)}=0,\ J_{1}^{(3)}=J_{2}^{(3)}=1,\ J_{n}^{(3)}=J_{n-1}^{(3)}+J_{n-2}^{(3)}+2J_{n-3}^{(3)},\ n\geq 3. Similarly, modified third-order Jacobsthal numbers are defined the following recurrence relation K_{0}^{(3)}=0,\ K_{1}^{(3)}=1,\ K_{2}^{(3)}=3,\ K_{n}^{(3)}=K_{n-1}^{(3)}+K_{n-2}^{(3)}+2K_{n-3}^{(3)},\ n\geq 3. Generating functions for the sequences \{J_{n}^{(3)}\}_{n\geq 0} and \{K_{n}^{(3)}\}_{n\geq 0} are J(x)=\sum_{n=0}^{\infty}J_{n}^{(3)}x^{n}=\frac{x}{1-x-x^{2}-2x^{3}} and K(x)=\sum_{n=0}^{\infty}K_{n}^{(3)}x^{n}=\frac{3-2x-x^{2}}{1-x-x^{2}-2x^{3}}, respectively. Binet formulae for third-order Jacobsthal numbers and modified third-order Jacobsthal numbers are J_{n}^{(3)}=\frac{1}{7}\left[2^{n+1}+X_{n}-2X_{n+1}\right] and K_{n}^{(3)}=2^{n}+X_{n}+2X_{n+1}, respectively, where \{X_{n}\}_{n\geq 0} is the cyclic sequence defined by X_{n}=\frac{\omega_{1}^{n}-\omega_{2}^{n}}{\omega_{1}-\omega_{2}}=\left\{\begin{array}[]{ccc}0&\textrm{if}&n\equiv 0\kern-5.0pt\pmod{3}\\ 1&\textrm{if}&n\equiv 1\kern-5.0pt\pmod{3}\\ -1&\textrm{if}&n\equiv 2\kern-5.0pt\pmod{3}\end{array}\right. and \omega_{1}=\omega_{2}^{2}=\frac{1}{2}(-1+i\sqrt{3})\in\mathbb{C} are the roots of the equation x^{2}+x+1=0 . For more details, the reader may review [ 4 , 8 ] . On the other hand, Şentürk and Ünal [ 11 ] introduced a generalized quaternions algebra called 3-parameter generalized quaternions (shortly, 3PGQs). The set of 3PGQs is denoted by \mathbb{H}_{\lambda_{1},\lambda_{2},\lambda_{3}} and defined as \{\psi_{0}+\sum_{l=1}^{3}\psi_{l}\textbf{e}_{l}:\ \textbf{e}_{1}^{2}=-\lambda_{1}\lambda_{2},\ \textbf{e}_{2}^{2}=-\lambda_{1}\lambda_{3},\ \textbf{e}_{3}^{2}=-\lambda_{2}\lambda_{3},\ \textbf{e}_{1}\textbf{e}_{2}\textbf{e}_{3}=-\lambda_{1}\lambda_{2}\lambda_{3}\}, where \lambda_{l\in\{1,2,3\}}\in\mathbb{R} . The vectors \{\textbf{e}_{1},\textbf{e}_{2},\textbf{e}_{3}\} satisfy the multiplication rules (1) \textbf{e}_{1}\textbf{e}_{2}=-\textbf{e}_{2}\textbf{e}_{1}=\lambda_{1}\textbf{e}_{3},\ \ \textbf{e}_{1}\textbf{e}_{3}=-\textbf{e}_{3}\textbf{e}_{1}=-\lambda_{2}\textbf{e}_{2},\ \ \textbf{e}_{2}\textbf{e}_{3}=-\textbf{e}_{3}\textbf{e}_{2}=\lambda_{3}\textbf{e}_{1}. Table 1 includes some special cases of 3PGQs. One can observe that bringing together the various types of quaternions and special recurrence sequence components is quite an attractive concept for several researchers in the literature. In [ 5 ] , Halıcı and Karataş studied the generalized Fibonacci quaternions, Morales [ 9 ] introduced generalized Tribonacci quaternions and the third-order Jacobsthal generalized quaternions are presented in [ 10 ] , Akyiğit et al. [ 1 , 2 ] examined the split Fibonacci quaternions and generalized Fibonacci quaternions, and Bród introduced the split Horadam quaternions. Recently, Kızılateş and Kibar [ 7 ] determined the 3PGQs with higher order generalized Fibonacci numbers components, and İşbilir and Gürses [ 6 ] introduced the 3PGQs with Horadam numbers components. \lambda_{1} \lambda_{2} \lambda_{3} Type of 3PGQs 1 \lambda_{2} \lambda_{3} Generalized quaternions 1 1 -1 Split quaternions 1 1 1 Hamilton quaternions 1 1 0 Semi-quaternions 1 -1 0 Split semi-quaternions Table 1. Some types of 3PGQs. In this study, we intend to combine the third-order Jacobsthal numbers and 3PGQ; namely, we investigate 3PGQ with third-order Jacobsthal numbers components. Also, we examine the recurrence relation, Binet formula, generating function, summing formulas, d’Ocagne identity, Cassini identity, and some special equations as well. 2. Third-order Jacobsthal 3-parameter generalized quaternions We define third-order Jacobsthal 3-parameter (TJ3p) generalized quaternions and modified third-order Jacobsthal 3-parameter (MTJ3p) generalized quaternions in following. Definition 2.1 . For n\geq 0 , the n -th TJ3p and MTJ3p generalized quaternions are J\mathcal{G}_{n}^{(3)}=J_{n}^{(3)}+J_{n+1}^{(3)}\textbf{e}_{1}+J_{n+2}^{(3)}\textbf{e}_{2}+J_{n+3}^{(3)}\textbf{e}_{3} and K\mathcal{G}_{n}^{(3)}=K_{n}^{(3)}+K_{n+1}^{(3)}\textbf{e}_{1}+K_{n+2}^{(3)}\textbf{e}_{2}+K_{n+3}^{(3)}\textbf{e}_{3}, where J_{n}^{(3)} and K_{n}^{(3)} are the classical third-order Jacobsthal and modified third-order Jacobsthal numbers, \{\textbf{e}_{1},\textbf{e}_{2},\textbf{e}_{3}\} satisfy the multiplication rules in Eq. ( 1 ). For n\geq 3 , TJ3p and MTJ3p generalized quaternions satisfy the following recurrence relations (2) J\mathcal{G}_{n}^{(3)}=J\mathcal{G}_{n-1}^{(3)}+J\mathcal{G}_{n-2}^{(3)}+2J\mathcal{G}_{n-3}^{(3)} and (3) K\mathcal{G}_{n}^{(3)}=K\mathcal{G}_{n-1}^{(3)}+K\mathcal{G}_{n-2}^{(3)}+2K\mathcal{G}_{n-3}^{(3)}, respectively. The next theorem gives generating functions for the sequences \{J\mathcal{G}_{n}^{(3)}\}_{n\geq 0} and \{K\mathcal{G}_{n}^{(3)}\}_{n\geq 0} . Since the proofs are very straightforward and easy, we do not prefer to give them. Theorem 2.2 . Generating functions for the sequences \{J\mathcal{G}_{n}^{(3)}\}_{n\geq 0} and \{K\mathcal{G}_{n}^{(3)}\}_{n\geq 0} are \sum_{n=0}^{\infty}J\mathcal{G}_{n}^{(3)}x^{n}=\frac{1}{\sigma(x)}\left[\textnormal{{e}}_{1}+\textnormal{{e}}_{2}+2\textnormal{{e}}_{3}+(1+\textnormal{{e}}_{2}+3\textnormal{{e}}_{3})x+2(\textnormal{{e}}_{2}+\textnormal{{e}}_{3})x^{2}\right] and \sum_{n=0}^{\infty}K\mathcal{G}_{n}^{(3)}x^{n}=\frac{1}{\sigma(x)}\left\{\begin{array}[]{c}3+\textnormal{{e}}_{1}+3\textnormal{{e}}_{2}+10\textnormal{{e}}_{3}-(2-2\textnormal{{e}}_{1}-7\textnormal{{e}}_{2}-5\textnormal{{e}}_{3})x\\ -(1-6\textnormal{{e}}_{1}-2\textnormal{{e}}_{2}-6\textnormal{{e}}_{3})x^{2}\end{array}\right\}, where \sigma(x)=1-x-x^{2}-2x^{3} . An interesting property of this type of quaternions is presented in the following theorem. To prove this result, we will use the relations that satisfy the third-order Jacobsthal and modified third-order Jacobsthal numbers. Theorem 2.3 . For any non-negative integer n , we have J\mathcal{G}_{n+3}^{(3)}=J\mathcal{G}_{n}^{(3)}+2^{n+1}\Theta and K\mathcal{G}_{n+3}^{(3)}=K\mathcal{G}_{n}^{(3)}+7\cdot 2^{n}\Theta where \Theta=1+2\textnormal{{e}}_{1}+4\textnormal{{e}}_{2}+8\textnormal{{e}}_{3} . Proof. Considering the Definition 2.1 and the relation J_{n+3}^{(3)}-2^{n+1}=J_{n}^{(3)} for n\geq 0 , we get \displaystyle J\mathcal{G}_{n}^{(3)} \displaystyle=J_{n}^{(3)}+J_{n+1}^{(3)}\textbf{e}_{1}+J_{n+2}^{(3)}\textbf{e}_{2}+J_{n+3}^{(3)}\textbf{e}_{3} \displaystyle=J_{n+3}^{(3)}-2^{n+1}+\left[J_{n+4}^{(3)}-2^{n+2}\right]\textbf{e}_{1}+\left[J_{n+5}^{(3)}-2^{n+3}\right]\textbf{e}_{2}+\left[J_{n+6}^{(3)}-2^{n+4}\right]\textbf{e}_{3} \displaystyle=J_{n+3}^{(3)}+J_{n+4}^{(3)}\textbf{e}_{1}+J_{n+5}^{(3)}\textbf{e}_{2}+J_{n+6}^{(3)}\textbf{e}_{3}-2^{n+1}(1+2\textnormal{{e}}_{1}+4\textnormal{{e}}_{2}+8\textnormal{{e}}_{3}) \displaystyle=J\mathcal{G}_{n+3}^{(3)}-2^{n+1}\Theta. The last equation gives the first equation in theorem. The proof of the second equation is very similar using the identity K_{n+3}^{(3)}-7\cdot 2^{n}=K_{n}^{(3)} for modified third-order Jacobsthal numbers. ∎ Binet formulas for the TJ3p and MTJ3p generalized quaternions can be found in the following theorem. Theorem 2.4 . For n\geq 0 , the n -th TJ3p and MTJ3p generalized quaternions are, respectively \displaystyle J\mathcal{G}_{n}^{(3)} \displaystyle=\frac{1}{7}\left[2^{n+1}\Theta+X_{n}\textnormal{{A}}-X_{n+1}\textnormal{{B}}\right] \displaystyle=\frac{1}{7}\left\{\begin{array}[]{ccc}2^{n+1}\Theta-\textnormal{{B}}&\textrm{if}&n\equiv 0\kern-5.0pt\pmod{3}\\ 2^{n+1}\Theta+\textnormal{{A}}+\textnormal{{B}}&\textrm{if}&n\equiv 1\kern-5.0pt\pmod{3}\\ 2^{n+1}\Theta-\textnormal{{A}}&\textrm{if}&n\equiv 2\kern-5.0pt\pmod{3}\end{array}\right. and \displaystyle K\mathcal{G}_{n}^{(3)} \displaystyle=2^{n}\Theta+X_{n}\textnormal{{C}}-X_{n+1}\textnormal{{D}}=\left\{\begin{array}[]{ccc}2^{n}\Theta-\textnormal{{D}}&\textrm{if}&n\equiv 0\kern-5.0pt\pmod{3}\\ 2^{n}\Theta+\textnormal{{C}}+\textnormal{{D}}&\textrm{if}&n\equiv 1\kern-5.0pt\pmod{3}\\ 2^{n}\Theta-\textnormal{{C}}&\textrm{if}&n\equiv 2\kern-5.0pt\pmod{3}\end{array}\right., where \Theta=1+2\textnormal{{e}}_{1}+4\textnormal{{e}}_{2}+8\textnormal{{e}}_{3} , \textnormal{{A}}=1+2\textnormal{{e}}_{1}-3\textnormal{{e}}_{2}+\textnormal{{e}}_{3} , \textnormal{{B}}=2-3\textnormal{{e}}_{1}+\textnormal{{e}}_{2}+2\textnormal{{e}}_{3} , \textnormal{{C}}=1-2\textnormal{{e}}_{1}+\textnormal{{e}}_{2}+\textnormal{{e}}_{3} and \textnormal{{D}}=-2+\textnormal{{e}}_{1}+\textnormal{{e}}_{2}-2\textnormal{{e}}_{3} . Proof. From the Definition 2.1 for TJ3p generalized quaternions, the Binet formula for the third-order Jacobsthal numbers and recurrence relation X_{n+2}=-X_{n+1}-X_{n} for n\geq 0 , we have \displaystyle 7J\mathcal{G}_{n}^{(3)} \displaystyle=7J_{n}^{(3)}+7J_{n+1}^{(3)}\textbf{e}_{1}+7J_{n+2}^{(3)}\textbf{e}_{2}+7J_{n+3}^{(3)}\textbf{e}_{3} \displaystyle=2^{n+1}+X_{n}-2X_{n+1} \displaystyle\ \ +\left[2^{n+2}+X_{n+1}-2X_{n+2}\right]\textbf{e}_{1} \displaystyle\ \ +\left[2^{n+3}+X_{n+2}-2X_{n+3}\right]\textbf{e}_{2} \displaystyle\ \ +\left[2^{n+4}+X_{n+3}-2X_{n+4}\right]\textbf{e}_{3} \displaystyle=2^{n+1}(1+2\textbf{e}_{1}+4\textbf{e}_{2}+8\textbf{e}_{3})+X_{n}(1+2\textbf{e}_{1}-3\textbf{e}_{2}+\textbf{e}_{3}) \displaystyle\ \ -X_{n+1}(2-3\textbf{e}_{1}+\textbf{e}_{2}+2\textbf{e}_{3}) \displaystyle=2^{n+1}\Theta+X_{n}\textnormal{{A}}-X_{n+1}\textnormal{{B}}, where \Theta=1+2\textbf{e}_{1}+4\textbf{e}_{2}+8\textbf{e}_{3} , \textnormal{{A}}=1+2\textbf{e}_{1}-3\textbf{e}_{2}+\textbf{e}_{3} and \textnormal{{B}}=2-3\textbf{e}_{1}+\textbf{e}_{2}+2\textbf{e}_{3} . The last equation gives the first equation in theorem. The proof of the second equation is very similar. ∎ The notation \Theta=1+2\textnormal{{e}}_{1}+4\textnormal{{e}}_{2}+8\textnormal{{e}}_{3} , \textnormal{{A}}=1+2\textnormal{{e}}_{1}-3\textnormal{{e}}_{2}+\textnormal{{e}}_{3} , \textnormal{{B}}=2-3\textnormal{{e}}_{1}+\textnormal{{e}}_{2}+2\textnormal{{e}}_{3} , \textnormal{{C}}=1-2\textnormal{{e}}_{1}+\textnormal{{e}}_{2}+\textnormal{{e}}_{3} and \textnormal{{D}}=-2+\textnormal{{e}}_{1}+\textnormal{{e}}_{2}-2\textnormal{{e}}_{3} of the previous theorem will be used in the rest of the paper. It should be noted that B A \displaystyle=(2-3\textbf{e}_{1}+\textbf{e}_{2}+2\textbf{e}_{3})(1+2\textbf{e}_{1}-3\textbf{e}_{2}+\textbf{e}_{3}) \displaystyle=2+4\textbf{e}_{1}-6\textbf{e}_{2}+2\textbf{e}_{3} \displaystyle\ \ -3\textbf{e}_{1}+6\lambda_{1}\lambda_{2}+9\textbf{e}_{3}+3\lambda_{2}\textbf{e}_{2} \displaystyle\ \ +\textbf{e}_{2}-2\lambda_{1}\textbf{e}_{3}+3\lambda_{1}\lambda_{3}+\lambda_{3}\textbf{e}_{1} \displaystyle\ \ +2\textbf{e}_{3}+4\lambda_{2}\textbf{e}_{2}+6\lambda_{3}\textbf{e}_{1}-2\lambda_{2}\lambda_{3} \displaystyle=2+6\lambda_{1}\lambda_{2}+3\lambda_{1}\lambda_{3}-2\lambda_{2}\lambda_{3}+\textbf{e}_{1}-5\textbf{e}_{2}+4\textbf{e}_{3}+7(\lambda_{3}\textbf{e}_{1}+\lambda_{2}\textbf{e}_{2}+\lambda_{1}\textbf{e}_{3}) \displaystyle=\Psi+7\Omega, where \Psi=2+6\lambda_{1}\lambda_{2}+3\lambda_{1}\lambda_{3}-2\lambda_{2}\lambda_{3}+\textbf{e}_{1}-5\textbf{e}_{2}+4\textbf{e}_{3} and \Omega=\lambda_{3}\textbf{e}_{1}+\lambda_{2}\textbf{e}_{2}+\lambda_{1}\textbf{e}_{3} . Then, using the multiplicative rule in Eq. ( 1 ), we obtain the next identities (4) \textnormal{{B}}\textnormal{{A}}=\Psi+7\Omega,\ \ \textnormal{{A}}\textnormal{{B}}=\Psi-7\Omega and (5) \textnormal{{D}}\textnormal{{C}}=\Phi+3\Omega,\ \ \textnormal{{C}}\textnormal{{D}}=\Phi-3\Omega, where \displaystyle\Psi \displaystyle=2+6\lambda_{1}\lambda_{2}+3\lambda_{1}\lambda_{3}-2\lambda_{2}\lambda_{3}+\textbf{e}_{1}-5\textbf{e}_{2}+4\textbf{e}_{3}, \displaystyle\Phi \displaystyle=-2+2\lambda_{1}\lambda_{2}-\lambda_{1}\lambda_{3}+2\lambda_{2}\lambda_{3}+5\textbf{e}_{1}-\textbf{e}_{2}-4\textbf{e}_{3}, \displaystyle\Omega \displaystyle=\lambda_{3}\textbf{e}_{1}+\lambda_{2}\textbf{e}_{2}+\lambda_{1}\textbf{e}_{3}. Eqs. ( 4 ) and ( 5 ) will be used for simplification in the next section. It should be noted that (6) \textnormal{{B}}\textnormal{{A}}=\textnormal{{A}}\textnormal{{B}}+14\Omega,\ \ \textnormal{{D}}\textnormal{{C}}=\textnormal{{C}}\textnormal{{D}}+6\Omega. 3. Main Results In this section, we present generalizations for the some well-known identities of third-order Jacobsthal and modified third-order Jacobsthal numbers. We begin with the Vajda’s identities for the TJ3p and MTJ3p generalized quaternions given in the following theorem. Theorem 3.1 . For integers n , a and b , we have J\mathcal{G}_{n+a}^{(3)}J\mathcal{G}_{n+b}^{(3)}-J\mathcal{G}_{n}^{(3)}J\mathcal{G}_{n+a+b}^{(3)}=\frac{1}{49}\left\{\begin{array}[]{c}2^{n+1}\left[\Theta Y_{n+b}(a)-2^{b}Y_{n}(a)\Theta\right]\\ +X_{a}\left[X_{b}\varXi-14X_{b+2}\Omega\right]\end{array}\right\} and K\mathcal{G}_{n+a}^{(3)}K\mathcal{G}_{n+b}^{(3)}-K\mathcal{G}_{n}^{(3)}K\mathcal{G}_{n+a+b}^{(3)}=\left\{\begin{array}[]{c}2^{n}\left[\Theta Y_{n+b}^{*}(a)-2^{b}Y_{n}^{*}(a)\Theta\right]\\ +X_{a}\left[X_{b}\varXi^{*}-6X_{b+2}\Omega\right]\end{array}\right\}, where Y_{n}(a)=X_{n}(2^{a}\textnormal{{A}}+X_{a+2}\textnormal{{A}}-X_{a}\textnormal{{B}})-X_{n+1}(2^{a}\textnormal{{B}}+X_{a}\textnormal{{A}}-X_{a+1}\textnormal{{B}}) , Y_{n}^{*}(a)=X_{n}(2^{a}\textnormal{{C}}+X_{a+2}\textnormal{{C}}-X_{a}\textnormal{{D}})-X_{n+1}(2^{a}\textnormal{{D}}+X_{a}\textnormal{{C}}-X_{a+1}\textnormal{{D}}) , \varXi=\textnormal{{A}}^{2}+\textnormal{{A}}\textnormal{{B}}+\textnormal{{B}}^{2} and \varXi^{*}=\textnormal{{C}}^{2}+\textnormal{{C}}\textnormal{{D}}+\textnormal{{D}}^{2} . Proof. Let us consider Z_{n}=X_{n}\textnormal{{A}}-X_{n+1}\textnormal{{B}} . Using the Vajda’s identity for sequence \{X_{n}\} , X_{n+a}X_{n+b}-X_{n}X_{n+a+b}=X_{a}X_{b} , we have \displaystyle Z_{n+a}Z_{n+b} \displaystyle-Z_{n}Z_{n+a+b} \displaystyle=\left[X_{n+a}\textnormal{{A}}-X_{n+a+1}\textnormal{{B}}\right]\left[X_{n+b}\textnormal{{A}}-X_{n+b+1}\textnormal{{B}}\right] \displaystyle\ \ -\left[X_{n}\textnormal{{A}}-X_{n+1}\textnormal{{B}}\right]\left[X_{n+a+b}\textnormal{{A}}-X_{n+a+b+1}\textnormal{{B}}\right] \displaystyle=\left[X_{n+a}X_{n+b}-X_{n}X_{n+a+b}\right]\textnormal{{A}}^{2} \displaystyle\ \ -\left[X_{n+a}X_{n+b+1}-X_{n}X_{n+a+b+1}\right]\textnormal{{A}}\textnormal{{B}} \displaystyle\ \ -\left[X_{n+a+1}X_{n+b}-X_{n+1}X_{n+a+b}\right]\textnormal{{B}}\textnormal{{A}} \displaystyle\ \ +\left[X_{n+a+1}X_{n+b+1}-X_{n+1}X_{n+a+b+1}\right]\textnormal{{B}}^{2} \displaystyle=X_{a}X_{b}\textnormal{{A}}^{2}-X_{a}X_{b+1}\textnormal{{A}}\textnormal{{B}}-X_{a}X_{b-1}\textnormal{{B}}\textnormal{{A}}+X_{a}X_{b}\textnormal{{B}}^{2} \displaystyle=X_{a}X_{b}\textnormal{{A}}^{2}-X_{a}X_{b+1}\textnormal{{A}}\textnormal{{B}}-X_{a}X_{b-1}(\textnormal{{A}}\textnormal{{B}}+14\Omega)+X_{a}X_{b}\textnormal{{B}}^{2} \displaystyle=X_{a}X_{b}(\textnormal{{A}}^{2}+\textnormal{{A}}\textnormal{{B}}+\textnormal{{B}}^{2})-14X_{a}X_{b+2}\Omega \displaystyle=X_{a}\left[X_{b}\varXi-14X_{b+2}\Omega\right], where \varXi=\textnormal{{A}}^{2}+\textnormal{{A}}\textnormal{{B}}+\textnormal{{B}}^{2} . From the Binet formula for the TJ3p generalized quaternions in Theorem 2.4 , we obtain \displaystyle 49\left[J\mathcal{G}_{n+a}^{(3)}J\mathcal{G}_{n+b}^{(3)}-J\mathcal{G}_{n}^{(3)}J\mathcal{G}_{n+a+b}^{(3)}\right] \displaystyle=\left[2^{n+a+1}\Theta+Z_{n+a}\right]\left[2^{n+b+1}\Theta+Z_{n+b}\right] \displaystyle\ \ -\left[2^{n+1}\Theta+Z_{n}\right]\left[2^{n+a+b+1}\Theta+Z_{n+a+b}\right] \displaystyle=2^{n+a+1}\Theta Z_{n+b}+2^{n+b+1}Z_{n+a}\Theta \displaystyle\ \ -2^{n+1}\Theta Z_{n+a+b}+2^{n+a+b+1}Z_{n}\Theta \displaystyle\ \ +Z_{n+a}Z_{n+b}-Z_{n}Z_{n+a+b} \displaystyle=2^{n+1}\left[\Theta Y_{n+b}(a)-2^{b}Y_{n}(a)\Theta\right] \displaystyle\ \ +X_{a}\left[X_{b}\varXi-14X_{b+2}\Omega\right], where Y_{n}(a)=X_{n}(2^{a}\textnormal{{A}}+X_{a+2}\textnormal{{A}}-X_{a}\textnormal{{B}})-X_{n+1}(2^{a}\textnormal{{B}}+X_{a}\textnormal{{A}}-X_{a+1}\textnormal{{B}}) . The last equation gives the first equation in theorem. The proof of the second equation is very similar. ∎ Taking b=-a in the Theorem 3.1 with the identities X_{-n}=-X_{n} and X_{n}=X_{n+3} gives Catalan’s identities for the TJ3p and MTJ3p generalized quaternions. Corollary 3.2 . Let n and a be two integers such that n\geq a , we have J\mathcal{G}_{n+a}^{(3)}J\mathcal{G}_{n-a}^{(3)}-\left[J\mathcal{G}_{n}^{(3)}\right]^{2}=\frac{1}{49}\left\{\begin{array}[]{c}2^{n+1}\left[\Theta Y_{n-a}(a)-2^{-a}Y_{n}(a)\Theta\right]\\ -X_{a}\left[X_{a}\varXi-14X_{a+1}\Omega\right]\end{array}\right\} and K\mathcal{G}_{n+a}^{(3)}K\mathcal{G}_{n-a}^{(3)}-\left[K\mathcal{G}_{n}^{(3)}\right]^{2}=\left\{\begin{array}[]{c}2^{n}\left[\Theta Y_{n-a}^{*}(a)-2^{-a}Y_{n}^{*}(a)\Theta\right]\\ -X_{a}\left[X_{a}\varXi^{*}-6X_{a+1}\Omega\right]\end{array}\right\}, with Y_{n}(a) , Y_{n}^{*}(a) , \varXi and \varXi^{*} as in Theorem 3.1 . Similarly, taking

Read the complete text