arXiv subjects

A. J. Macfarlane

Publications and source records attributed to A. J. Macfarlane.

At least 19 recordsLinked to original sources

arXiv2024

On the structure of the complement $\overline{\Mfib}$ of the set $\Mfib$ of fibbinary numbers in the set of positive natural numbers

The set $\Mfib$ of fibbinary numbers is defined via a bijection between the set $\BB{N}$ of natural numbers and $\Mfib$. Since the elements of $\Mfib$ do not exhaust $\BB{N}$, the structure of the complement $\overline{\Mfib}$ of $\Mfib$ in $\BB{N}$ is of interest. An explicit expression $\overline{\Mfib}=\bigcup_{k \geq 1}^\infty Φ_k$ is obtained in terms of certain well-defined sets $Φ_k, \; k \geq 1$. The key to its proof lies in first considering the odd numbers involved in this statement: a general treatment, with full justification, of the binary representations of the odd numbers is developed, and exploited in showing the expression quoted for $\overline{\Mfib}$ to be correct. The main results of the article can also be viewed as providing partitions of the set $\BB{N}$ of natural numbers, and also of its subset of odd numbers, that follow from the introduction of the set $\Mfib$, and of its subset of odd integers.

math.NT↗

arXiv2024

On the fibbinary numbers and the Wythoffarray

This paper defines the set fib of fibbinary numbers and displays its structure in the form of a table of a specialised type, and in array form. It uses the Zeckendorf representation $n \in \mathbf{N}$ to define a bijection $\mathcal{Z}$ between $\mathbf{N}$ and fib. It is proved that the fibbinary array is the image under $\mathcal{Z}$ of the famous Wythoff array. The fibbinary table proves useful pictorial insight into the fractal defined by the Wythoff array. The Wythoff table, obtained as the image under the inverse of $\mathcal{Z}$ of the fibbinary table, leads to a simpler view of the fractal, and may be compared with the (1938) Steinhaus tree.

math.CO↗

arXiv2024

Pi Formulas: some smooth stones on the beach of rough numbers

This article is about Pi Formulas, infinite series of fractions which sum to multiples of Pi. Each such one can be associated with a unique set $S_k$ of rough numbers, where $k$ is a prime number. Given $S_k$ for any prime $k$, the set $S_{k^{\prime}}$, where $k^{\prime}$ is the smallest prime greater than $k$, can be constructed easily. From this it follows that Pi Formulas occur in disjoint families. In any family, there is a first member, a series of least prime number $k_{min}$, which must be summed from first principles. Then from the series of some $k > k_{min}$ already summed, the series for $k^{\prime}$ can be summed by a simple algebraic procedure. A good number of Pi Formulas, belonging to a variety of families and giving results believed to be new, are presented here.

math.NT↗

arXiv2003

Representations of the exceptional and other Lie algebras with integral eigenvalues of the Casimir operator

The uniformity, for the family of exceptional Lie algebras g, of the decompositions of the powers of their adjoint representations is well-known now for powers up to the fourth. The paper describes an extension of this uniformity for the totally antisymmetrised n-th powers up to n=9, identifying (see Tables 3 and 6) families of representations with integer eigenvalues 5,...,9 for the quadratic Casimir operator, in each case providing a formula (see eq. (11) to (15)) for the dimensions of the representations in the family as a function of D=dim g. This generalises previous results for powers j and Casimir eigenvalues j, j<=4. Many intriguing, perhaps puzzling, features of the dimension formulas are discussed and the possibility that they may be valid for a wider class of not necessarily simple Lie algebras is considered.

math-ph↗

arXiv2002

Development of a unified tensor calculus for the exceptional Lie algebras

The uniformity of the decomposition law, for a family F of Lie algebras which includes the exceptional Lie algebras, of the tensor powers ad^n of their adjoint representations ad is now well-known. This paper uses it to embark on the development of a unified tensor calculus for the exceptional Lie algebras. It deals explicitly with all the tensors that arise at the n=2 stage, obtaining a large body of systematic information about their properties and identities satisfied by them. Some results at the n=3 level are obtained, including a simple derivation of the the dimension and Casimir eigenvalue data for all the constituents of ad^3. This is vital input data for treating the set of all tensors that enter the picture at the n=3 level, following a path already known to be viable for a_1. The special way in which the Lie algebra d_4 conforms to its place in the family F alongside the exceptional Lie algebras is described.

math-ph↗

arXiv2001

Compilation of relations for the antisymmetric tensors defined by the Lie algebra cocycles of $su(n)$

This paper attempts to provide a comprehensive compilation of results, many new here, involving the invariant totally antisymmetric tensors (Omega tensors) which define the Lie algebra cohomology cocycles of $su(n)$, and that play an essential role in the optimal definition of Racah-Casimir operators of $su(n)$. Since the Omega tensors occur naturally within the algebra of totally antisymmetrised products of $λ$-matrices of $su(n)$, relations within this algebra are studied in detail, and then employed to provide a powerful means of deriving important Omega tensor/cocycle identities. The results include formulas for the squares of all the Omega tensors of $su(n)$. Various key derivations are given to illustrate the methods employed.

math-ph↗

arXiv2001

Hidden supersymmetries in supersymmetric quantum mechanics

We discuss the appearance of additional, hidden supersymmetries for simple 0+1 $Ad(G)$-invariant supersymmetric models and analyse some geometrical mechanisms that lead to them. It is shown that their existence depends crucially on the availability of odd order invariant skewsymmetric tensors on the (generic) compact Lie algebra $\cal G$, and hence on the cohomology properties of the Lie algebra considered.

hep-th↗

arXiv2001

Lie algebra and invariant tensor technology for g2

Proceeding in analogy with su(n) work on lambda matrices and f- and d-tensors, this paper develops the technology of the Lie algebra g2, its seven dimensional defining representation gamma and the full set of invariant tensors that arise in relation thereto. A comprehensive listing of identities involving these tensors is given. This includes identities that depend on use of characteristic equations, especially for gamma, and a good body of results involving the quadratic, sextic and (the non-primitivity of) other Casimir operators of g2.

math-ph↗

arXiv2000

Symplectic and orthogonal Lie algebra technology for bosonic and fermionic oscillator models of integrable systems

To provide tools, especially L-operators, for use in studies of rational Yang-Baxter algebras and quantum integrable models when the Lie algebras so(N) (b_n, d_n) or sp(2n) (c_n) are the invariance algebras of their R matrices, this paper develops a presentation of these Lie algebras convenient for the context, and derives many properties of the matrices of their defining representations and of the ad-invariant tensors that enter their multiplication laws. Metaplectic-type representations of sp(2n) and so(N) on bosonic and on fermionic Fock spaces respectively are constructed. Concise general expressions (see (5.2) and (5.5) below) for their L-operators are obtained, and used to derive simple formulas for the T operators of the rational RTT algebra of the associated integral systems, thereby enabling their efficient treatment by means of the algebraic Bethe ansatz.

math-ph↗

arXiv2000

Monopole supersymmetries and the Biedenharn operator

The hidden supersymmetry of the monopole found by De Jonghe et al. is generalized to a spin $\2$ particle in the combined field of a Dirac monopole plus a $λ^2/r^2$ potential [considered before by D'Hoker and Vinet], and related to the operator introduced by Biedenharn a long time ago in solving the Dirac-Coulomb problem. Explicit solutions are obtained by diagonalizing the Biedenharn operator

hep-th↗

arXiv2000

Optimally defined Racah-Casimir operators for su(n) and their eigenvalues for various classes of representations

This paper deals with the striking fact that there is an essentially canonical path from the $i$-th Lie algebra cohomology cocycle, $i=1,2,... l$, of a simple compact Lie algebra $\g$ of rank $l$ to the definition of its primitive Casimir operators $C^{(i)}$ of order $m_i$. Thus one obtains a complete set of Racah-Casimir operators $C^{(i)}$ for each $\g$ and nothing else. The paper then goes on to develop a general formula for the eigenvalue $c^{(i)}$ of each $C^{(i)}$ valid for any representation of $\g$, and thereby to relate $c^{(i)}$ to a suitably defined generalised Dynkin index. The form of the formula for $c^{(i)}$ for $su(n)$ is known sufficiently explicitly to make clear some interesting and important features. For the purposes of illustration, detailed results are displayed for some classes of representation of $su(n)$, including all the fundamental ones and the adjoint representation.

math-ph↗

arXiv2000

Fermionic realisations of simple Lie algebras

We study the representation ${\cal D}$ of a simple compact Lie algebra $\g$ of rank l constructed with the aid of the hermitian Dirac matrices of a (${\rm dim} \g$)-dimensional euclidean space. The irreducible representations of $\g$ contained in ${\cal D}$ are found by providing a general construction on suitable fermionic Fock spaces. We give full details not only for the simplest odd and even cases, namely su(2) and su(3), but also for the next (${dim} \g$)-even case of su(5). Our results are far reaching: they apply to any $\g$-invariant quantum mechanical system containing ${\rm dim} \g$ fermions. Another reason for undertaking this study is to examine the role of the $\g$-invariant fermionic operators that naturally arise. These are given in terms of products of an odd number of gamma matrices, and include, besides a cubic operator, (l-1) fermionic scalars of higher order. The latter are constructed from the Lie algebra cohomology cocycles, and must be considered to be of theoretical significance similar to the cubic operator. In the (${\rm dim} \g$)-even case, the product of all l operators turns out to be the chirality operator $γ_q, q=({{\rm dim} \g+1})$.

hep-th↗

arXiv1999

Yang-Baxter Algebra for the n-Harmonic Oscillator Realisation of sp(2n,R)

Using a rational R-matrix associated with the 4 x 4 defining matrix representation of c_2=sp(4), the Lie algebra of Sp(4), a one-site operator solution of the associated Yang-Baxter algebra acting in the Fock space of two harmonic oscillators is derived. This is used to define N-site integrable systems, which are soluble by a version of the algebraic Bethe ansatz method without nesting. All essential aspects of the work generalise directly from c_2 to c_n.

solv-int↗

arXiv1999

On characteristic equations, trace identities and Casimir operators of simple Lie algebras

Two approaches are developed to exploit, for simple complex or compact real Lie algebras g, the information that stems from the characteristic equations of representation matrices and Casimir operators. These approaches are selected so as to be viable not only for `small' Lie algebras and suitable for treatment by computer algebra. A very large body of new results emerges in the forms, a) of identities of a tensorial nature, involving structure constants etc. of g, b) of trace identities for powers of matrices of the adjoint and defining representations of g, c) of expressions of non-primitive Casimir operators of g in terms of primitive ones. The methods are sufficiently tractable to allow not only explicit proof by hand of the non-primitive nature of the quartic Casimir of g2, f4, e6, but also e.g. of that of the tenth order Casimir of f4.

math-ph↗

arXiv1999

Construction of supercharges for the one-dimensional supersymmetric nonlinear sigma model

This paper addresses an issue essential to the study of hidden supersymmetries (meaning here ones that do not close on the Hamiltonian) for one-dimensional non-linear supersymmetric sigma models. The issue relates to ambiguities, due to partial integrations in superspace, both in the actual definition of these supersymmetries and in the Noether definition of the associated supercharges. The unique consistent forms of both these definitions have to be determined simultaneously by a process that adjusts the former definitions so that the associated supercharges do indeed correctly generate them with the aid of the canonical formalism. The paper explains and illustrates these matters and gives some new results.

hep-th↗

arXiv1999

Higher order BRST and anti-BRST operators and cohomology for compact Lie algebras

After defining cohomologically higher order BRST and anti-BRST operators for a compact simple algebra {\cal G}, the associated higher order Laplacians are introduced and the corresponding supersymmetry algebra $Σ$ is analysed. These operators act on the states generated by a set of fermionic ghost fields transforming under the adjoint representation. In contrast with the standard case, for which the Laplacian is given by the quadratic Casimir, the higher order Laplacians $W$ are not in general given completely in terms of the Casimir-Racah operators, and may involve the ghost number operator. The higher order version of the Hodge decomposition is exhibited. The example of su(3) is worked out in detail, including the expression of its higher order Laplacian W.

hep-th↗

arXiv1998

Effective actions, relative cohomology and Chern Simons forms

The explicit expression of all the WZW effective actions for a simple group G broken down to a subgroup H is established in a simple and direct way, and the formal similarity of these actions to the Chern-Simons forms is explained. Applications are also discussed.

hep-th↗

arXiv1997

Invariant tensors for simple groups

The forms of the invariant primitive tensors for the simple Lie algebras A_l, B_l, C_l and D_l are investigated. A new family of symmetric invariant tensors is introduced using the non-trivial cocycles for the Lie algebra cohomology. For the A_l algebra it is explicitly shown that the generic forms of these tensors become zero except for the l primitive ones and that they give rise to the l primitive Casimir operators. Some recurrence and duality relations are given for the Lie algebra cocycles. Tables for the 3- and 5-cocycles for su(3) and su(4) are also provided. Finally, new relations involving the d and f su(n) tensors are given.

math-ph↗