Abstract

For a class of multiparameter statistical models based on 𝑁2×𝑁2 braid matrices, the eigenvalues of the transfer matrix 𝐓(𝑟) are obtained explicitly for all (𝑟,𝑁). Our formalism yields them as solutions of sets of linear equations with simple constant coefficients. The role of zero-sum multiplets constituted in terms of roots of unity is pointed out, and their origin is traced to circular permutations of the indices in the tensor products of basis states induced by our class of 𝐓(𝑟) matrices. The role of free parameters, increasing as 𝑁2 with N, is emphasized throughout. Spin chain Hamiltonians are constructed and studied for all N. Inverse Cayley transforms of the Yang-Baxter matrices corresponding to our braid matrices are obtained for all N. They provide potentials for factorizable S-matrices. Main results are summarized, and perspectives are indicated in the concluding remarks.

1. Introduction

Statistical models “exact” in the sense of Baxter [1] satisfy “star-triangle” relations leading to transfer matrices commuting for different values of the spectral parameter. Crucial in the study of such models is the spectrum of eigenvalues of these matrices. But even for the extensively studied 6-vertex and 8-vertex models based on 4×4 braid matrices (the braid property guaranteeing star-triangle relations), after the first few simple steps, one has to resort to numerical computations. (We cannot adequately discuss here the vast associated literature but refer to texts citing major sources [2, 3].) But the basic reason for such situation is that, starting with 2×2 blocks 𝑇𝑎𝑏 of the transfer matrix and constructing 2𝑟×2𝑟 blocks via coproduct rules for order 𝑟  (𝑟=1,2,3,), one faces increasingly complicated nonlinear systems of equations to be solved in constructing eigenstates and eigenvalues. Even when a systematic approach is available, such as the Bethe Ansatz for the 6-vertex case, it only means that the relevant nonlinear equations can be written down systematically. The task of solving them remains. For the 8-vertex case one explores analytical properties to extract informations (See [4], e.g.). Also the number of free parameters remains strictly limited for such models, including the multistate generalization of the 6-vertex one [2].

Our class of models exhibits the following properties.(1)A systematic construction for all dimensions. One starts from 𝑁2×𝑁2 braid matrices leading to 𝑁𝑟×𝑁𝑟 blocks of 𝐓(𝑟), the transfer matrix of order 𝑟 for 𝑁=2,3,4,5,;𝑟=1,2,3, with no upper limit.(2)The number of free parameters increases with 𝑁 as 𝑁2. This is a unique feature of our models.(3)For all (𝑁,𝑟) one solves sets of linear equations with systematically obtained simple constant coefficients to construct eigenvalues of 𝐓(𝑟)=𝑁𝑎=1𝑇(𝑟)𝑎𝑎. This is a consequence of our starting point: braid matrices constructed on a basis of “nested sequence” of projectors [5].

The total dimension of the base space on which such equations have to be solved increases as 𝑁𝑟. This will be seen to break up into subspaces closed under the action of 𝐓(𝑟), thus reducing the work considerably. Evidently one cannot continue to display the results explicitly, as 𝑁𝑟 increases. But it is possible to implement a general and particularly efficient approach. This will be illustrated for all 𝑁 and 𝑟=1,2,3,4,5. The generalization to 𝑟>5 will be clearly visible. This is the central purpose of our paper. Certain other features will be explored with comments in conclusion.

2. Braid and Transfer Matrices

The roots of our class of braid matrices and the transfer matrices they generate are to be found in the nested sequence of projectors [5]. In [5] it was noted that the 4×4 projectors providing the basis of the 6-vertex and 8-vertex models (i.e., the braid matrices leading to those models) can be generalized to higher dimensions—all higher ones (𝑁 odd or even). Braid matrices for odd dimensions were exhaustively constructed on such a basis [6], and related statistical models were studied [7]. Then even dimensional ones were presented [8], and the corresponding braid matrices and statistical models were studied [9]. In previous works the eigenvalues and eigenfunctions of the transfer matrices were presented mostly for the lowest values of 𝑁, namely, the 4×4 and 9×9 braid matrices. We present below a general approach for all 𝑁.

Let us just mention that systematic study of the exotic bialgebras that arise from the 9×9 unitary braid matrices was constructed in [10]. The dual bialgebra of one of these exotic bialgebras was also presented.

We start now by recapitulating the construction of our nested sequence of projectors, leading to the remarkable properties of our solutions.

2.1. Even Dimensions

Let 𝑁=2𝑛  (𝑛=1,2,3,). Define ((𝑎𝑏) denoting a matrix with unity on row 𝑎 and column 𝑏) 𝑃(𝜖)𝑖𝑗=12(𝑖𝑖)(𝑗𝑗)+𝑖𝑖𝑗𝑗𝑖+𝜖𝑖𝑗𝑗+𝑖𝑖,𝑗𝑗(2.1) where 𝑖,𝑗{1,,𝑛}, 𝜖=±, 𝑖=2𝑛+1𝑖, 𝑗=2𝑛+1𝑗. Interchanging 𝑗𝑗 on the right one obtains 𝑃𝑖(𝜖)𝑗. One thus obtains a complete basis of projectors satisfying (with 𝑃(𝜖)𝑖𝑗=𝑃(𝜖)𝑖𝑗, 𝑃𝑖(𝜖)𝑗=𝑃(𝜖)𝑖𝑗 by definition) 𝑃(𝜖)𝑎𝑏𝑃(𝜖)𝑐𝑑=𝛿𝑎𝑐𝛿𝑏𝑑𝛿𝜖𝜖𝑃(𝜖)𝑎𝑏,𝑎,𝑏,𝑐,𝑑{1,,𝑁=2𝑛},𝑛𝜖=±𝑖,𝑗=1𝑃(𝜖)𝑖𝑗+𝑃𝑖(𝜖)𝑗=𝐼(2𝑛)2×(2𝑛)2.(2.2) Let 𝑚(𝜖)𝑖𝑗 be an arbitrary set of parameters satisfying the crucial constraint 𝑚(𝜖)𝑖𝑗=𝑚𝑖(𝜖)𝑗.(2.3) Define the 𝑁2×𝑁2 matrix 𝑅(𝜃)=𝜖𝑖,𝑗𝑒𝑚(𝜖)𝑖𝑗𝜃𝑃(𝜖)𝑖𝑗+𝑃𝑖(𝜖)𝑗.(2.4) The 𝜃-dependence entering exclusively through the exponentials as coefficients as above. Such construction guarantees [79] the braid property 𝑅12𝑅(𝜃)23𝑅(𝜃+𝜃)12𝑅(𝜃)=23𝑅(𝜃)12𝑅(𝜃+𝜃)23(𝜃),(2.5) where, in standard notations, 𝑅12=𝑅𝐼 and 𝑅23𝑅=𝐼, 𝐼 denoting the 𝑁×𝑁 identity matrix.

Define the permutation matrix 𝐏=𝑎,𝑏(𝑎𝑏)(𝑏𝑎). Then 𝑅(𝜃)=𝐏𝑅(𝜃)(2.6) satisfies the Yang-Baxter equation. The monodromy matrix of order 1 (𝑟=1) is given by 𝑇(1)(𝜃)=𝑅(𝜃).(2.7) The 𝑁×𝑁 blocks are (with 𝑎,𝑏{1,,2𝑛}) 𝑇(1)𝑎𝑏1(𝜃)=2𝑒𝑚(+)𝑏𝑎𝜃+𝑒𝑚()𝑏𝑎𝜃1(𝑏𝑎)+2𝑒𝑚(+)𝑏𝑎𝜃𝑒𝑚()𝑏𝑎𝜃𝑏𝑎𝑓(+)𝑏𝑎(𝜃)(𝑏𝑎)+𝑓()𝑏𝑎(𝜃)𝑏𝑎,(2.8) with 𝑚(𝜖)𝑏𝑎=𝑚𝑏(𝜖)𝑎=𝑚(𝜖)𝑏𝑎=𝑚(𝜖)𝑏𝑎.(2.9) Higher orders (𝑟>1) are obtained via coproducts 𝑇(𝑟)𝑎𝑏(𝜃)=𝑐1,,𝑐𝑟1𝑇(1)𝑎𝑐1(𝜃)𝑇𝑐(1)1𝑐2(𝜃)𝑇𝑐(1)𝑟2𝑐𝑟1(𝜃)𝑇𝑐(1)𝑟1𝑏(𝜃).(2.10) For some essential purposes, it is worthwhile to express (2.8) as (with 𝑖,𝑗{1,,𝑛}) 𝑇(1)𝑖𝑗1(𝜃)=2𝜖𝑒𝑚(𝜖)𝑗𝑖𝜃(𝑗𝑖)+𝜖𝑗𝑖,𝑇(1)𝑖𝑗1(𝜃)=2𝜖𝑒𝑚(𝜖)𝑗𝑖𝜃𝑗𝑖,𝑇+𝜖(𝑗𝑖)𝑖(1)𝑗1(𝜃)=2𝜖𝑒𝑚(𝜖)𝑗𝑖𝜃𝑗𝑗𝑖+𝜖𝑖,𝑇(1)𝑖𝑗1(𝜃)=2𝜖𝑒𝑚(𝜖)𝑗𝑖𝜃𝑗𝑖+𝜖.𝑗𝑖(2.11) The transfer matrix is the trace on (𝑎,𝑏): 𝐓(𝑟)(𝜃)=𝑎𝑇(𝑟)𝑎𝑎(𝜃).(2.12) The foregoing construction assures the crucial commutativity 𝐓(𝑟)(𝜃),𝐓(𝑟)(𝜃)=0.(2.13) This implies that the eigenstates of 𝐓(𝑟)(𝜃) are 𝜃-independent (sum of basis vectors with 𝜃-independent, constant relative coefficients).

2.2. Odd Dimensions

From the extensive previous studies [6, 7], we select the essential features. For 𝑁=2𝑛1 (𝑛=2,3,), 𝑛=𝑁𝑛+1=𝑛(2.14) with the same definition as in (2.1). This is the crucial new feature. For (𝑖,𝑗)𝑛, one has the same 𝑃(𝜖)𝑖𝑗, 𝑃𝑖(𝜖)𝑗 as before. But now 𝑃(𝜖)𝑖𝑛=12(𝑖𝑖)+𝑖𝑖𝑖+𝜖𝑖+𝑃𝑖𝑖(𝑛𝑛),(𝜖)𝑛𝑖=12(𝑛𝑛)(𝑖𝑖)+𝑖𝑖𝑖+𝜖𝑖+,𝑃𝑖𝑖(𝜖)𝑛𝑛=(𝑛𝑛)(𝑛𝑛).(2.15) Normalizing the coefficient of 𝑃(𝜖)𝑛𝑛 to unity, 𝑅(𝜃) of (2.4) now becomes 𝑅(𝜃)=𝑃(𝜖)𝑛𝑛+𝑖,𝜖𝑒𝑚(𝜖)𝑛𝑖𝜃𝑃(𝜖)𝑛𝑖+𝑒𝑚(𝜖)𝑖𝑛𝜃𝑃(𝜖)𝑖𝑛+𝑖,𝑗𝜖𝑒𝑚(𝜖)𝑖𝑗𝜃𝑃(𝜖)𝑖𝑗+𝑃𝑖(𝜖)𝑗(2.16) (𝑖,𝑗{1,,𝑛1}, 𝑖,𝑗{2𝑛1,,𝑛+1}, 𝜖=±). There are evident parallel modifications concerning 𝑇(𝑟)𝑎𝑏(𝜃). Reference [7] contains detailed discussions and results for 𝑁=3, 𝑟=1,2,3,4. We will not repeat them here but will reconsider them later in the context of the general methods to be implemented below.

2.3. Free Parameters

A unique feature of our constructions is the number of free parameters 𝑚(𝜖)𝑖𝑗, increasing as 𝑁2 with 𝑁. For our choice of normalizations (apart from a possible altered choice of an overall factor, irrelevant for (2.5)), the exact numbers are 12𝑁2=2𝑛2for1𝑁=2𝑛,2𝑛(𝑁+3)(𝑁1)=221for𝑁=2𝑛1.(2.17) This is to be contrasted with the multistate generalization of the 6-vertex model where the parametrization remain fixed at the 6-vertex level ([2] and sources cited here).

One of our principal aims is to display the roles of our free parameters concerning the basic features of our models.

3. Explicit Eigenvalues of Transfer Matrices for All Dimensions 𝑁 and Orders 𝑟

We present below a unified approach for all 𝑁 and illustrate it in some detail for 𝑟=1,2,3,4,5. For 𝑟>5 the extension will be evident.

3.1. Even Dimensions

In Section 5 of [9] it was pointed out that the antidiagonal matrix 𝐾=2𝑛𝑎=1𝑎𝑎=𝑛𝑖=1𝑖𝑖+𝑖𝑖(3.1) relates the blocks of 𝑇(1) as follows: 𝐾𝑇(1)𝑎𝑏=𝑇𝑎(1)𝑏,𝑇(1)𝑎𝑏𝐾=𝑇(1)𝑎𝑏,𝐾𝑇(1)𝑎𝑏𝐾=𝑇(1)𝑎𝑏.(3.2) For 𝑁=2 this led to an iterative construction of eigenstates and eigenvalues of the transfer matrix 𝐓(𝑟) for increasing 𝑟. This works efficiently only for 𝑁=2 (the 4×4  𝑅(𝜃)-matrix).

For our present approach the crucial ingredient shows up immediately in the structure (arising from those of our projectors and the coproducts rules) 𝐓(𝑟)=12𝑟𝑎1,𝑎2,,𝑎𝑟𝜖21,𝜖32,,𝜖1𝑟𝑒(𝑚)𝑎2𝑎1(𝜖21+𝑚)𝑎3𝑎2(𝜖32++𝑚)𝑎1𝑎𝑟(𝜖1𝑟)𝜃×𝑎2𝑎1+𝜖21𝑎2𝑎1𝑎3𝑎2+𝜖32𝑎3𝑎2𝑎1𝑎𝑟+𝜖1𝑟𝑎1𝑎𝑟,(3.3) where each 𝜖 is independently (±). From (3.3), one obtains immediately 𝐓(𝑟)1=12𝑟𝑎1,𝑎2,,𝑎𝑟𝜖21,𝜖32,,𝜖1𝑟𝑒(𝑚)𝑎2𝑎1(𝜖21+𝑚)𝑎3𝑎2(𝜖32++𝑚)𝑎1𝑎𝑟(𝜖1𝑟)𝜃×𝑎1𝑎2+𝜖21𝑎1𝑎2𝑎2𝑎3+𝜖32𝑎2𝑎3𝑎𝑟𝑎1+𝜖1𝑟𝑎𝑟𝑎1.(3.4) Consider the basis state ||𝑏1||𝑏2||𝑏𝑟||𝑏1𝑏2𝑏𝑟.(3.5) (1) If any one (even a single one) of the indices 𝑏𝑖𝑎𝑖+1 or 𝑎𝑖+1 (cyclic), then the coefficient of 𝑒(𝑚)𝑎2𝑎1(𝜖21++𝑚)𝑎1𝑎𝑟(𝜖1𝑟)𝜃, namely, [(𝑎2𝑎1)+𝜖21(𝑎2𝑎1)],[(𝑎1𝑎𝑟)+𝜖1𝑟(𝑎1𝑎𝑟)], will annihilate it. On the other hand (2.9) plays a crucial role. Gathering together all the terms with 𝑒(𝑚)𝑏2𝑏1(𝜖21+𝑚)𝑏3𝑏2(𝜖32++𝑚)𝑏𝑟𝑏(𝜖𝑟,𝑟1𝑟1+𝑚)𝑏1𝑏𝑟(𝜖1𝑟)𝜃 as coefficient, one obtains the action of 𝐓(𝑟) on (3.5). All essential features for the general case can be read off the first few simples examples: 22𝐓(2)||𝑏1𝑏2=𝜖21,𝜖12𝑒(𝑚)𝑏2𝑏1(𝜖21+𝑚)𝑏1𝑏2(𝜖12)𝜃1+𝜖21𝜖12||𝑏2𝑏1+𝜖21|||𝑏2𝑏1,(3.6)23𝐓(3)||𝑏1𝑏2𝑏3=𝜖21,𝜖32,𝜖13𝑒(𝑚)𝑏2𝑏1(𝜖21+𝑚)𝑏3𝑏2(𝜖32+𝑚)𝑏1𝑏3(𝜖13)𝜃1+𝜖21𝜖32𝜖13×||𝑏2𝑏3𝑏1+𝜖21|||𝑏2𝑏3𝑏1+𝜖32|||𝑏2𝑏3𝑏1+𝜖13|||𝑏2𝑏3𝑏1,(3.7)24𝐓(4)||𝑏1𝑏2𝑏3𝑏4=𝜖21,𝜖32,𝜖43,𝜖14𝑒(𝑚)𝑏2𝑏1(𝜖21+𝑚)𝑏3𝑏2(𝜖32+𝑚)𝑏4𝑏3(𝜖43+𝑚)𝑏1𝑏4(𝜖14)𝜃1+𝜖21𝜖32𝜖43𝜖14×||𝑏2𝑏3𝑏4𝑏1+𝜖21|||𝑏2𝑏3𝑏4𝑏1+𝜖32|||𝑏2𝑏3𝑏4𝑏1+𝜖43|||𝑏2𝑏3𝑏4𝑏1+𝜖14|||𝑏2𝑏3𝑏4𝑏1+𝜖21𝜖32|||𝑏2𝑏3𝑏4𝑏1+𝜖21𝜖43|||𝑏2𝑏3𝑏4𝑏1+𝜖21𝜖14|||𝑏2𝑏3𝑏4𝑏1.(3.8) Note that the bars correspond to the indies of the 𝜖’s. One has thus 𝜖𝑖𝑗|𝑏𝑖𝑏𝑗𝑏𝑘𝑏𝑙, 𝜖𝑖𝑗𝜖𝑘𝑖|𝑏𝑖𝑏𝑗𝑏𝑘𝑏𝑙=𝜖𝑖𝑗𝜖𝑘𝑖|𝑏𝑖𝑏𝑗𝑏𝑘𝑏𝑙 since 𝑏𝑖=𝑏𝑖. Consequently only even number of additional bars appear on the right. Starting with even or odd number of bars on the left leads to “even” and “odd” closed subspaces. 25𝐓(5)||𝑏1𝑏2𝑏3𝑏4𝑏5=𝜖21,𝜖32,𝜖43,𝜖14𝑒(𝑚)𝑏2𝑏1(𝜖21+𝑚)𝑏3𝑏2(𝜖32+𝑚)𝑏4𝑏3(𝜖43+𝑚)𝑏54(𝜖54𝑏𝑎+𝑚)𝑏1𝑏5(𝜖15)𝜃1+𝜖21𝜖32𝜖43𝜖54𝜖15×||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖21|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖32|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖43|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖54|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖15|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖21𝜖32|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖21𝜖43|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖21𝜖54|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖21𝜖15|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖32𝜖43|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖32𝜖54|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖32𝜖15|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖43𝜖54|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖43𝜖15|||𝑏2𝑏3𝑏4𝑏5𝑏1+𝜖54𝜖15|||𝑏2𝑏3𝑏4𝑏5𝑏1.(3.9)

(2) It is important to note that the preceding features are independent of 𝑁—for a given 𝑟 they are valid for all 𝑁. For 𝑟>𝑁 all the indices (𝑏1,,𝑏𝑟) cannot be distinct. But even for 𝑟<𝑁, some or all of them can coincide. This general validity permits a unified treatment for all 𝑁.

(3) Note also the crucial cyclic permutation of the indices under the action of 𝐓(𝑟): (𝑏1,𝑏2,𝑏3,,𝑏𝑟1,𝑏𝑟)(𝑏2,𝑏3,𝑏4,,𝑏𝑟,𝑏1). This is a consequence of the trace condition in (2.12) incorporated in (3.3). It will be seen later how this leads to an essential role of the 𝑟th roots of unity 𝑒𝐢((𝑘2𝜋)/𝑟) (𝑘=0,1,,𝑟1) in our construction of eigenstates and also of eigenvalues as factors accompanying exponentials of the type appearing in (3.6)–(3.9).

We continue the study of the particular cases 𝑟=1,2,3,4,5 each one for any 𝑁. General features will be better understood afterwards.

(i)𝑟=1
𝐓(1)=12𝑎,𝜖𝑒𝑚(𝜖)𝑎𝑎𝜃(𝑎𝑎)+𝜖𝑎𝑎=12𝑖,𝜖(1+𝜖)𝑒𝑚(𝜖)𝑖𝑖𝜃(𝑖𝑖)+𝜖𝑖𝑖,(3.10) where 𝑎{1,,2𝑛}, 𝑖{1,,𝑛}. This is already in diagonal form. Eigenstates and values are obtained trivially.

(ii)𝑟=2
We start with (3.6). Separating “even” and “odd” spaces, define (for 𝑖𝑗 when (𝑖,𝑗,𝑖,𝑗) are all distinct) 𝐓(2)𝑐1||𝑖𝑗+𝑐2||𝑖𝑗+𝑐3||𝑗𝑖+𝑐4||𝑗𝑖=𝜐𝑒𝑐1||𝑖𝑗+𝑐2||𝑖𝑗+𝑐3||𝑗𝑖+𝑐4||𝑗𝑖,𝐓(2)𝑑1||𝑖𝑗+𝑑2||𝑖𝑗+𝑑3||𝑗𝑖+𝑑4||𝑗𝑖=𝜐𝑜𝑑1||𝑖𝑗+𝑑2||𝑖𝑗+𝑑3||𝑗𝑖+𝑑4||.𝑗𝑖(3.11) The solutions are, for (𝑐1,𝑐2,𝑐3,𝑐4)=(1,1,1,1),(1,1,1,1),(1,1,1,1),(1,1,1,1) respectively, 𝜐𝑒=𝑒(𝑚(+)𝑖𝑗+𝑚(+)𝑗𝑖)𝜃,𝑒(𝑚(+)𝑖𝑗+𝑚(+)𝑗𝑖)𝜃,𝑒(𝑚()𝑖𝑗+𝑚()𝑗𝑖)𝜃,𝑒(𝑚()𝑖𝑗+𝑚()𝑗𝑖)𝜃,(3.12) and for (𝑑1,𝑑2,𝑑3,𝑑4)=(1,1,1,1),(1,1,1,1),(1,1,1,1),(1,1,1,1), respectively, 𝜐𝑜=𝑒(𝑚(+)𝑖𝑗+𝑚(+)𝑗𝑖)𝜃,𝑒(𝑚(+)𝑖𝑗+𝑚(+)𝑗𝑖)𝜃,𝑒(𝑚()𝑖𝑗+𝑚()𝑗𝑖)𝜃,𝑒(𝑚()𝑖𝑗+𝑚()𝑗𝑖)𝜃.(3.13) For 𝑖=𝑗, with 𝜖=±𝐓(2)|||𝑖𝑖+𝜖𝑖𝑖=𝑒2𝑚(𝜖)𝑖𝑖𝜃|||𝑖𝑖+𝜖𝑖𝑖,𝐓(2)||𝑖𝑖||+𝜖𝑖𝑖=𝜖𝑒2𝑚(𝜖)𝑖𝑖𝜃||𝑖𝑖||+𝜖.𝑖𝑖(3.14) Summing over all 𝑖, consistently with the general rule, one obtains 𝐓Tr(2)=2𝑛𝑖=1𝑒2𝑚(+)𝑖𝑖𝜃(3.15) contributions coming only for 𝑖=𝑗. For 𝑟=2, the 𝑟th roots of unity are ±1 and their role above is not visible very distinctly due to the presence of ±1 also from 𝜖. The role of the roots of unity will be more evident from 𝑟=3 onwards. The simple exercise above gives all eigenvalues and eigenstates for 𝐓(2) for all 𝑁  (=2𝑛). For 𝑁=2, 𝑟=2, writing out the 16×16  𝐓(2) in full and effectively diagonalizing it we have confirmed the results above.

(iii)𝑟=3
Now the starting point is (3.7). Define, for the even subset, 𝑉𝜆=||𝑖1𝑖2𝑖3+𝜖12||𝑖1𝑖2𝑖3+𝜖23||𝑖1𝑖2𝑖3+𝜖31||𝑖1𝑖2𝑖3||𝑖+𝜆3𝑖1𝑖2+𝜖12||𝑖3𝑖1𝑖2+𝜖23||𝑖3𝑖1𝑖2+𝜖31||𝑖3𝑖1𝑖2+𝜆2||𝑖2𝑖3𝑖1+𝜖12||𝑖2𝑖3𝑖1+𝜖23||𝑖2𝑖3𝑖1+𝜖31||𝑖2𝑖3𝑖1,(3.16) where 𝜆=1,𝑒𝐢(2𝜋/3),𝑒𝐢(4𝜋/3),(3.17) and hence 𝜆3=1, and also 1+𝑒𝐢(2𝜋/3)+𝑒𝐢(4𝜋/3)=0. Corresponding to (𝑖1,𝑖2,𝑖3)(𝑖1,𝑖2,𝑖3), define for the odd-subset, 𝑉𝜆=||𝑖1𝑖2𝑖3+𝜖12||𝑖1𝑖2𝑖3+𝜖23||𝑖1𝑖2𝑖3+𝜖31||𝑖1𝑖2𝑖3||+𝜆𝑖3𝑖1𝑖2+𝜖12||𝑖3𝑖1𝑖2+𝜖23||𝑖3𝑖1𝑖2+𝜖31||𝑖3𝑖1𝑖2+𝜆2||𝑖2𝑖3𝑖1+𝜖12||𝑖2𝑖3𝑖1+𝜖23||𝑖2𝑖3𝑖1+𝜖31||𝑖2𝑖3𝑖1.(3.18) The 8 values of the set (𝜖12,𝜖23,𝜖31) are restricted (see (3.7)) by (1+𝜖12𝜖23𝜖31)=2 (0). The remaining 4 possibilities along with 3 for 𝜆 (see (3.17)) yield 4×3=12 solutions for 𝑉𝜆 and 𝑉𝜆 each for distinct indices 𝑖1𝑖2𝑖3𝑖1. The eigenvalues are both 𝑉𝜆 and 𝑉𝜆: 𝑒(𝑚)𝑖1𝑖2(𝜖12+𝑚)𝑖2𝑖3(𝜖23+𝑚)𝑖3𝑖1(𝜖31)𝜃1,𝑒𝐢(2𝜋/3),𝑒𝐢(4𝜋/3).(3.19) For two of the three indices equal (𝑖1=𝑖2𝑖3,) the preceding results can be carried over without crucial change. But for 𝑖1=𝑖2=𝑖3𝑖, say, there is a special situation due to the fact 1+𝜆+𝜆2=(3,0,0),(3.20) for 𝜆=(1,𝑒𝐢(2𝜋/3),𝑒𝐢(4𝜋/3)), respectively. Now 𝑉𝜆=1+𝜆+𝜆2𝜖|𝑖𝑖𝑖+12+𝜖31𝜆+𝜖23𝜆2||𝑖+𝜖𝑖𝑖23+𝜖12𝜆+𝜖31𝜆2||𝑖𝑖𝑖+𝜖31+𝜖23𝜆+𝜖12𝜆2||𝑖𝑖𝑖.(3.21) The eigenfunctions and eigenvalues are (for mutually orthogonal basis states) 𝐓(3)|||𝑖𝑖𝑖+𝑖+||𝑖𝑖𝑖𝑖𝑖+||𝑖𝑖𝑖=𝑒3𝑚(+)𝑖𝑖𝜃|||𝑖𝑖𝑖+𝑖+||𝑖𝑖𝑖𝑖𝑖+||𝑖𝑖𝑖,𝐓(3)||3|𝑖𝑖𝑖𝑖||𝑖𝑖𝑖𝑖𝑖||𝑖𝑖𝑖=𝑒(𝑚(+)𝑖𝑖+2𝑚()𝑖𝑖)𝜃||3|𝑖𝑖𝑖𝑖||𝑖𝑖𝑖𝑖𝑖||𝑖𝑖𝑖,𝐓(3)||𝑖𝑖𝑖+𝑒𝐢(2𝜋/3)||𝑖𝑖𝑖+𝑒𝐢(4𝜋/3)||𝑖𝑖𝑖=𝑒𝐢(2𝜋/3)𝑒(𝑚(+)𝑖𝑖+2𝑚()𝑖𝑖)𝜃||𝑖𝑖𝑖+𝑒𝐢(2𝜋/3)||𝑖𝑖𝑖+𝑒𝐢(4𝜋/3)||𝑖𝑖𝑖,𝐓(3)||𝑖𝑖𝑖+𝑒𝐢(4𝜋/3)||𝑖𝑖𝑖+𝑒𝐢(2𝜋/3)||𝑖𝑖𝑖=𝑒𝐢(4𝜋/3)𝑒(𝑚(+)𝑖𝑖+2𝑚()𝑖𝑖)𝜃||𝑖𝑖𝑖+𝑒𝐢(4𝜋/3)||𝑖𝑖𝑖+𝑒𝐢(2𝜋/3)||𝑖𝑖𝑖.(3.22) For 𝑉𝜆 one follows exactly similar steps and obtains the same set of eigenvalues (with 𝑖𝑖 in the eigenstates). Combining all the solutions above, we obtain complete results for 𝑟=3 and all even 𝑁.

(iv)𝑟=4
The key result (3.8) indicates the following construction (a direct generalization of (3.16)). Define, for the even subset, 𝑉(𝑏1𝑏2𝑏3𝑏4)=||𝑏1𝑏2𝑏3𝑏4+𝜖12|||𝑏1𝑏2𝑏3𝑏4+𝜖23|||𝑏1𝑏2𝑏3𝑏4+𝜖34|||𝑏1𝑏2𝑏3𝑏4+𝜖41|||𝑏1𝑏2𝑏3𝑏4+𝜖12𝜖23|||𝑏1𝑏2𝑏3𝑏4+𝜖12𝜖34|||𝑏1𝑏2𝑏3𝑏4+𝜖23𝜖34|||𝑏1𝑏2𝑏3𝑏4,(3.23) and similarly, implementing circular permutations as in (3.16), 𝑉(𝑏4𝑏1𝑏2𝑏3),𝑉(𝑏3𝑏4𝑏1𝑏2),𝑉(𝑏2𝑏3𝑏4𝑏1), define 𝑉𝜆=𝑉(𝑏1𝑏2𝑏3𝑏4)+𝜆𝑉(𝑏4𝑏1𝑏2𝑏3)+𝜆2𝑉(𝑏3𝑏4𝑏1𝑏2)+𝜆3𝑉(𝑏2𝑏3𝑏4𝑏1),(3.24) where 𝜆=1,𝑒𝐢(2𝜋/4),𝑒𝐢2(2𝜋/4),𝑒𝐢3(2𝜋/4)=(1,𝐢,1,𝐢).(3.25) One obtains for distinct (𝑖1,𝑖2,𝑖3,𝑖4)𝐓(4)𝑉𝜆=𝜆𝑒(𝑚)𝑖1𝑖2(𝜖12+𝑚)𝑖2𝑖3(𝜖23+𝑚)𝑖3𝑖4(𝜖34+𝑚)𝑖4𝑖1(𝜖41)𝜃𝑉𝜆,(3.26) where for nonzero results (1+𝜖12𝜖23𝜖34𝜖41)=2. For the 8 possibilities remaining for the 𝜖’s and the 4 values of 𝜆, one obtains the full set of 32 eigenstates and the corresponding eigenvalues. For the complementary “odd” subspace formed by |||𝑏1𝑏2𝑏3𝑏4,|||𝑏1𝑏2𝑏3𝑏4,|||𝑏1𝑏2𝑏3𝑏4,|||𝑏1𝑏2𝑏3𝑏4,|||𝑏1𝑏2𝑏3𝑏4,|||𝑏1𝑏2𝑏3𝑏4,|||𝑏1𝑏2𝑏3𝑏4,|||𝑏1𝑏2𝑏3𝑏4,(3.27) and the sets related through circular permutations of the indices, one obtains, entirely in evident analogy of (3.24), 𝑉𝜆=𝑉(𝑏1𝑏2𝑏3𝑏4)+𝜆𝑉(𝑏4𝑏1𝑏2𝑏3)+𝜆2𝑉(𝑏3𝑏4𝑏1𝑏2)+𝜆3𝑉(𝑏2𝑏3𝑏4𝑏1).(3.28) One obtains, for distinct indices, the same set of eigenvalues as in (3.26). For some equal indices, but not all, the preceding results can be carried over essentially. For all equal indices one obtains the following spectrum of eigenvalues (grouping together the even and the odd subspaces): 𝑒4𝑚(+)𝑖𝑖𝜃[3(1,1),-times]𝑒2(𝑚(+)𝑖𝑖+𝑚()𝑖𝑖)𝜃1,𝑒𝐢(2𝜋/4),𝑒𝐢2(2𝜋/4),𝑒𝐢3(2𝜋/4),𝑒4𝑚()𝑖𝑖𝜃(1,1),(3.29) (In fact these are the same as in (A.13) of [7], since the distinguishing index for odd dimensions signalled in (2.14) and (2.15) is not involved.) We will not present here the 16 orthogonal eigenstates corresponding to (3.29). But a new feature, as compared to (3.22), should be pointed out. For 𝑟=3, considering together even and odd spaces, one has for identical indices the eigenvalues 𝑒3𝑚(+)𝑖𝑖𝜃[2(1,1),-times]𝑒(𝑚(+)𝑖𝑖+2𝑚()𝑖𝑖)𝜃1,𝑒𝐢(2𝜋/3),𝑒𝐢2(2𝜋/3).(3.30) In fact, for each 𝑟 one obtains, for 𝜖=1, 𝑒𝑟𝑚(+)𝑖𝑖𝜃(1,1) giving 𝐓Tr(𝑟)=2𝑛𝑖=1𝑒𝑟𝑚(𝜖)𝑖𝑖𝜃.(3.31) All other multiplets have each one zero sum. Concerning the zero sum multiplets, comparing (3.30) and (3.29), we note that(1)for 𝑟=3, one has only triplets; (2)for 𝑟=4, one has quadruplets and also a doublet (1,1). This last enters due to factorizability of 𝑟 (4=2×2).Such a feature is worth signalling since it generalizes.
When 𝑟 is a prime number, the zero-sum multiplets appear only as 𝑟-plets (with 𝑟th roots of unity). How they consistently cover the whole base space (along with the doublets giving the trace as in (3.31)) has been amply discussed in our previous papers [7, 9] under the heading “An encounter with a theorem of Fermat.” When 𝑟 has many factors the multiplicity of submultiplets corresponding to each one is difficult to formulate giving, say, an explicit general prescription.

(v)𝑟5
Another problem concerning systematic, complete enumeration one encounters already for 𝑟=5. Starting with (3.9) the generalizations of (3.23)–(3.26) are quite evident. We will not present them here. But as compared to (3.29) one has now for the 32-dimensional subspace (for 𝑖1=𝑖2==𝑖5) 𝑒5𝑚(+)𝑖𝑖𝜃𝑛(1,1),2-times𝑒(3𝑚(+)𝑖𝑖+2𝑚()𝑖𝑖)𝜃1,𝑒𝐢(2𝜋/5),𝑒𝐢2(2𝜋/5),𝑒𝐢3(2𝜋/5),𝑒𝐢4(2𝜋/5),𝑛4-times𝑒(𝑚(+)𝑖𝑖+4𝑚()𝑖𝑖)𝜃1,𝑒𝐢(2𝜋/5),𝑒𝐢2(2𝜋/5),𝑒𝐢3(2𝜋/5),𝑒𝐢4(2𝜋/5),(3.32) where 𝑛2+𝑛4=6. At this stage it is not difficult to present complete construction. But as 𝑟 increases, for prime numbers, one has 𝑒𝑟𝑚(+)𝑖𝑖𝜃𝑛(1,1),2-times𝑒((𝑟2)𝑚(+)𝑖𝑖+2𝑚()𝑖𝑖)𝜃1,𝑒𝐢(2𝜋/𝑟),,𝑒𝐢(𝑟1)(2𝜋/𝑟),𝑛4-times𝑒((𝑟4)𝑚(+)𝑖𝑖+4𝑚()𝑖𝑖)𝜃1,𝑒𝐢(2𝜋/𝑟),,𝑒𝐢(𝑟1)(2𝜋/𝑟),𝑛𝑟1-times𝑒(𝑚(+)𝑖𝑖+(𝑟1)𝑚()𝑖𝑖)𝜃1,𝑒𝐢(2𝜋/𝑟),,𝑒𝐢(𝑟1)(2𝜋/𝑟),(3.33) the sum of the multiplicities of such 𝑟-plets satisfying 𝑛2+𝑛4++𝑛𝑟1=1𝑟(2𝑟2).(3.34) (That the right-hand side is an integer for 𝑟 a prime number is guaranteed by a theorem of Fermat, as has been discussed before.) A general explicit prescription for the sequence (𝑛2,𝑛4,,𝑛𝑟1) is beyond the scope of this paper.

The situation is as above for prime numbers 𝑟. For factorizable 𝑟 the presence, in addition, of submultiplets has already been pointed out. For 𝑟=4, only such submultiplets were (1,1). For 𝑟=6, one can have (1,1), (1,𝑒𝐢(2𝜋/3),𝑒𝐢2(2𝜋/3)) corresponding to the factors 6=2×3, respectively. For 𝑟=30=2×3×5 one can have also submultiplets corresponding to 5. And so on.

But apart from the above-mentioned limitations concerning multiplicities of zero-sum (“roots of unity”) multiplets and submultiplets we can claim to have elucidated the spectrum of eigenvalues for all (𝑁,𝑟). For the generic case with distinct indices for any (𝑁,𝑟), one has eigenvalues 𝜆𝑘𝑒(𝑚)𝑖1𝑖2(𝜖12+𝑚)𝑖2𝑖3(𝜖23++𝑚)𝑖𝑖𝑟(𝜖𝑟1,𝑟𝑟1+𝑚)𝑖1𝑖𝑟(𝜖1𝑟)𝜃,(3.35) where (1+𝜖12𝜖23𝜖𝑟1,𝑟𝜖1,𝑟)=2 and 𝜆𝑘=𝑒𝐢𝑘(2𝜋/𝑟)(𝑘=0,1,2,,𝑟1). Here each 𝑚(𝜖𝑘𝑙)𝑖𝑘𝑖𝑙 is a free parameter. We solve only sets of linear equations with quite simple constant coefficients to obtain the eigenstates and the eigenvalues.

3.2. Odd Dimensions

We refer back to (2.14)–(2.16). When the special index 𝑛 (𝑛=𝑛) for 𝑁=2𝑛1 is not present in the basis state |𝑏1𝑏2𝑏𝑟 and hence in the subspace it generates via cyclic permutations as in the forgoing examples, the foregoing constructions can be taken over wholesale. When 𝑛 is present, the modifications are not difficult to take into account. The trace is now Tr(𝐓(𝑟))=2𝑛1𝑖=1𝑒𝑟𝑚(𝜖)𝑖𝑖𝜃+1, since the central state |𝑛|𝑛|𝑛|𝑛𝑛𝑛 contributes with our normalization 1 to the trace. Various aspects have been studied in considerable detail in our previous papers [6, 7] on odd 𝑁. Here we just refer to them.

4. Spin Chain Hamiltonians

Spin chains corresponding to our braid matrices have already been studied in our previous papers [7, 8]. Here we formulate a unified approach for all dimensions (𝑁=2𝑛1,2𝑛).

The basic formula (see sources cited in [7, 8]) is 𝐻=𝑟𝑘=1̇𝑅𝐼𝐼𝑘,𝑘+1(0)𝐼𝐼,(4.1) where for circular boundary conditions, 𝑘+1=𝑟+11. For even 𝑁 (𝑁=2𝑛), from (2.4),̇𝑑𝑅(0)=|||𝑑𝜃𝑅(𝜃)𝜃=0=𝜖,𝑖,𝑗𝑚(𝜖)𝑖𝑗𝑃(𝜖)𝑖𝑗+𝑃𝑖(𝜖)𝑗(4.2) the projectors being given by (2.1) and the remark below (2.1). For 𝑁 odd (𝑁=2𝑛1; 𝑛=2,3,) we introduce a modified overall normalization factor to start with. Such a factor is trivial concerning the braid equation, but not for the Hamiltonian (since a derivative is involved) if the factor is 𝜃-dependent. Multiply (2.16) by 𝑒𝑚𝜃 and redefine 𝑚+𝑚(𝜖)𝑖𝑗,𝑚+𝑚(𝜖)𝑛𝑖,𝑚+𝑚(𝜖)𝑖𝑛𝑚(𝜖)𝑖𝑗,𝑚(𝜖)𝑛𝑖,𝑚(𝜖)𝑖𝑛(4.3) since (𝑚(𝜖)𝑖𝑗,) are arbitrary to start with. Now for odd 𝑁 (with the ranges of (𝑖,𝑗), (𝑖,𝑗) of (2.16)), ̇𝑅(0)=𝑚𝑃𝑛𝑛+𝜖,𝑖𝑚(𝜖)𝑛𝑖𝑃(𝜖)𝑛𝑖+𝑚(𝜖)𝑖𝑛𝑃(𝜖)𝑖𝑛+𝜖,𝑖,𝑗𝑚(𝜖)𝑖𝑗𝑃(𝜖)𝑖𝑗+𝑃𝑖(𝜖)𝑗.(4.4) This extends the 𝑚=0 case by including the presence of 𝑃𝑛𝑛.

In 𝐻, ̇𝑅(0)𝑘,𝑘+1 acts on the basis |𝑉(𝑘)|𝑉(𝑘+1). One can denote, using standard ordering of spin components for each 𝑘, ||𝑉(𝑘)=|||||||||||||||||||𝑛1/2𝑘||1/2𝑘||1/2𝑘||𝑛+1/2𝑘,||𝑉(𝑘)=||||||||||||||||||||||𝑛1𝑘||1𝑘||0𝑘||1𝑘||𝑛+1𝑘,(4.5) for 𝑁=2𝑛,2𝑛1, respectively. Without being restricted to spin, one can consider more generally any system with 𝑁 orthogonal states. We will continue, however, to use the terminology of spin. At each site one can consider a superposition at each level such as (𝑙𝑐(𝑘)𝑖𝑙|𝑙𝑘). To keep the notation tractable, we will just denote ||𝑉(𝑘)=|||||||||||||||||1𝑘||2𝑘|||2𝑘|||1𝑘,(4.6) and keep possible significances of |𝑖𝑘 in mind. Each index 𝑘 is acted upon twice by 𝐻, namely, by ̇𝑅𝑘1,𝑘(0), ̇𝑅𝑘,𝑘+1(0), and, for closed chains, the index 1 is also thus involved in ̇𝑅12(0) and ̇𝑅𝑟1(0). Some simple examples are

(i)𝑁=2
̇𝑅||||||||||||(0)=̂𝑎+00̂𝑎0̂𝑎+̂𝑎00̂𝑎̂𝑎+0̂𝑎00̂𝑎+||||||||||||,(4.7) where ̂𝑎±=(1/2)(𝑚(+)11±𝑚()11) and hence in evident notations ̇𝑅||(0)𝑉(𝑘)||𝑉(𝑘+1)=||||||||||||̂𝑎+||11+̂𝑎|||11̂𝑎+|||11+̂𝑎|||11̂𝑎|||11+̂𝑎+|||11̂𝑎||11+̂𝑎+|||11(𝑘,𝑘+1),(4.8)

(ii)𝑁=4
In the notation of Section 7 of [9] 𝑅||||||||||||𝐷(𝜃)=11𝐴00110𝐷22𝐴2200𝐴22𝐷220𝐴11𝐷0011||||||||||||,(4.9) where 𝐷11=𝐷11=̂𝑎+0̂𝑏000+̂𝑏0000+0000̂𝑎+,𝐷22=𝐷22=̂𝑐+0𝑑000+𝑑0000+0000̂𝑐+,𝐴11=𝐴11=000̂𝑎̂𝑏0000̂𝑏00̂𝑎,𝐴00022=𝐴22=000̂𝑐𝑑0000𝑑00̂𝑐,000̂𝑎±=12𝑚(+)11±𝑚()11,̂𝑏±=12𝑚(+)12±𝑚()12,̂𝑐±=12𝑚(+)21±𝑚()21,𝑑±=12𝑚(+)22±𝑚()22.(4.10) Hence, ̇𝑅||(0)𝑉(𝑘)||𝑉(𝑘+1)=||||||||||||||𝐷111||+|||𝑉1𝐴11||||𝐷𝑉222||+|||𝑉2𝐴22||||𝐴𝑉222||+|||𝑉2𝐷22||||𝐴𝑉111||+|||𝑉1𝐷11||𝑉(𝑘,𝑘+1),(4.11) where for each 𝑘, ||||||𝑉=||1|2|21

(iii)𝑁=3
Adapting the results of Section 1 of [7] and Section 11 of [8] to notations analogous to the cases above, one can write ||||||||||||||||||,𝑅(𝜃)=𝐷0𝐴0𝐶0𝐴0𝐷(4.12) where 𝐷=̂𝑎+0̂𝑏00+000̂𝑎+,𝐴=00̂𝑎0̂𝑏0̂𝑎00,𝐶=̂𝑐+0̂𝑐0𝑚0̂𝑐0̂𝑐+,(4.13) the central element 𝑚 corresponding to 𝑚𝑃𝑛𝑛 of (4.4) and ̂𝑎±=12𝑚(+)11±𝑚()11,̂𝑏±=12𝑚(+)12±𝑚()12,̂𝑐±=12𝑚(+)21±𝑚()21.(4.14)

Denote the basis factors for each 𝑘 as, |𝑉(𝑘)=||||||1|21(𝑘) and ||𝑉(𝑘)||𝑉(𝑘+1)=||||||||||||||||||||1𝑉2𝑉1||𝑉(𝑘,𝑘+1).(4.15) One obtains ̇||𝑅(0)𝑉(𝑘)||𝑉(𝑘+1)=||||||||||||||||||||||||||||||||||||||𝐷0𝐴0𝐶0𝐴0𝐷1𝑉2𝑉1||𝑉(𝑘,𝑘+1)=|||||||||||𝐷||+|||1𝑉1𝐴||||𝐶||||𝐴||+|||𝑉2𝑉1𝑉1𝐷||𝑉(𝑘,𝑘+1).(4.16) For 𝑁=5,6,, one can generate such results systematically. They, considering the action of all the terms of (4.1), furnish the transition matrix elements and expectation values for possible states of the chain and permit a study of correlations.

One can also consider higher-order conserved quantities (Section 1.5 of [10]) given by 𝐻𝑙=𝑑𝑙𝑑𝜃𝑙log𝐓(𝑟)||||(𝜃)𝜃=0.(4.17) For 𝑙=2, (4.1) is generalized by the appearance of factors of the type (apart from non-overlapping derivatives) ̇𝑅(0)𝑘1,𝑘̇𝑅(0)𝑘,𝑘+1,̈𝑅(0)𝑘,𝑘+1.(4.18) For 𝑙>2 this generalizes in an evident fashion. A study of spin chains for the “exotic” SØ3 can be found in [11]. It is interesting to compare it with the 4×4 (for 𝑁=2) case briefly presented above by exploring the latter case in comparable detail.

One may note that for our class of braid matrices, for even 𝑁 as compared to (4.2), are 𝑑𝑙𝑑𝜃𝑙𝑅||||(𝜃)𝜃=0=𝜖𝑖,𝑗𝑚(𝜖)𝑖𝑗𝑙𝑃(𝜖)𝑖𝑗+𝑃𝑖(𝜖)𝑗,(4.19) and for odd 𝑁, as compared to (4.4), 𝑑𝑙𝑑𝜃𝑙𝑅||||(𝜃)𝜃=0=𝑚𝑙𝑃𝑛𝑛+𝜖,𝑖𝑚(𝜖)𝑛𝑖𝑙𝑃(𝜖)𝑛𝑖+𝑚(𝜖)𝑖𝑛𝑙𝑃(𝜖)𝑖𝑛+𝜖,𝑖,𝑗𝑚(𝜖)𝑖𝑗𝑙𝑃(𝜖)𝑖𝑗+𝑃𝑖(𝜖)𝑗.(4.20)

5. Potentials for Factorizable 𝑆-Matrices

Such potentials can be obtained as inverse Cayley transforms of the Yang-Baxter matrices of appropriate dimensions (see, for example, Section 3 of [12] and Section 1 of [2]). Starting with the Yang-Baxter matrix 𝑅(𝜃)=𝐏𝑅(𝜃), the required potential 𝐕(𝜃) is given by 𝐢𝐕(𝜃)=(𝑅(𝜃)𝜆(𝜃)𝐼)1(𝑅(𝜃)+𝜆(𝜃)𝐼).(5.1) We have emphasized in our previous studies (Section 5 of [7], Section 8 of [8]) that 𝜆(𝜃) cannot be arbitrary, a set of values must be excluded for the inverse (5.1) to be well defined. Here we generalize our previous results to all 𝑁. As compared to other well-known studies [13] of fields corresponding to factorizable scatterings, here our construction starts with the braid matrices (2.4) and (2.16).

Define 𝑋(𝜃)=(𝑅(𝜃)𝜆(𝜃)𝐼)1(5.2) when 𝐢𝐕(𝜃)=𝐼+2𝜆(𝜃)𝑋(𝜃).(5.3) We give below immediately the general solution and then explain the notations more precisely. The solution was obtained via formal series expansions. But the final closed form can be verified directly. The solution is 1𝑋(𝜃)=2𝑁𝜖=±𝑎,𝑏=11𝜆2(𝜃)𝑒(𝑚(𝜖)𝑎𝑏+𝑚(𝜖)𝑏𝑎)𝜃𝑎𝜆(𝜃)(𝑎𝑎)(𝑏𝑏)+𝜖𝑎𝑏𝑏+𝑒𝑚(𝜖)𝑏𝑎𝜃𝑎(𝑎𝑏)(𝑏𝑎)+𝜖𝑏𝑏𝑎,(5.4) where 𝜆(𝜃)±𝑒(1/2)(𝑚(𝜖)𝑎𝑏+𝑚(𝜖)𝑏𝑎)𝜃.(5.5) For 𝑁=2𝑛, (2.9) is implicit in this result. For 𝑁=2𝑛1, when 𝑛=𝑛, the conventions (2.16) (or (4.4)) are to be implemented when 𝑎 or 𝑏 or both (𝑎,𝑏) are 𝑛. From (5.3) and (5.4), 𝐢𝐕(𝜃)=2𝑁𝜖=±𝑎,𝑏=1𝜆2(𝜃)+𝑒(𝑚(𝜖)𝑎𝑏+𝑚(𝜖)𝑏𝑎)𝜃𝜆2(𝜃)𝑒[𝑚(𝜖)𝑎𝑏+𝑚(𝜖)𝑏𝑎]𝜃𝑎(𝑎𝑎)(𝑏𝑏)+𝜖𝑎𝑏𝑏+𝑒𝑚(𝜖)𝑎𝑏𝜃𝜆2(𝜃)𝑒(𝑚(𝜖)𝑎𝑏+𝑚(𝜖)𝑏𝑎)𝜃𝑏(𝑏𝑎)(𝑎𝑏)+𝜖𝑎𝑎𝑏𝑎𝑏,𝑐𝑑𝐕(𝜃)(𝑎𝑏,𝑐𝑑)(𝑎𝑏)(𝑐𝑑).(5.6) Now one can write down the Lagrangians for scalar and spinor fields. For the spinor case, for example, 𝐢=𝑑𝑥𝜓𝑎𝛾𝜈𝜕𝜈𝜓𝑎𝚐𝜓𝑎𝛾𝜈𝜓𝑐𝐕𝑎𝑏,𝑐𝑑𝜓𝑏𝛾𝜈𝜓𝑑.(5.7) The simpler scalar case can be written analogously. The scalar Lagrangian has an interaction term of the form (𝜙𝑎𝜙𝑐)𝐕𝑎𝑏,𝑐𝑑(𝜙𝑏𝜙𝑑). For 𝑁=2𝑛, one obtains the nonvanishing elements of 𝐕 as 𝐕𝑏𝑏,𝑑𝑑𝐢=2𝜖𝜆2(𝜃)+𝑒(𝑚(𝜖)𝑏𝑑+𝑚(𝜖)𝑑𝑏)𝜃𝜆2(𝜃)𝑒(𝑚(𝜖)𝑏𝑑+𝑚(𝜖)𝑑𝑏)𝜃,𝐕𝑏𝑏,𝑑𝑑𝐢=2𝜖𝜖𝜆2(𝜃)+𝑒(𝑚(𝜖)𝑏𝑑+𝑚(𝜖)𝑑𝑏)𝜃𝜆2(𝜃)𝑒(𝑚(𝜖)𝑏𝑑+𝑚(𝜖)𝑑𝑏)𝜃,𝐕𝑑𝑏,𝑏𝑑𝐢=2𝜖𝑒𝑚(𝜖)𝑏𝑑𝜃𝜆2(𝜃)𝑒(𝑚(𝜖)𝑏𝑑+𝑚(𝜖)𝑑𝑏)𝜃,𝐕𝑑𝑏,𝑏𝑑𝐢=2𝜖𝜖𝑒𝑚(𝜖)𝑏𝑑𝜃𝜆2(𝜃)𝑒(𝑚(𝜖)𝑏𝑑+𝑚(𝜖)𝑑𝑏)𝜃,(5.8) where 𝑎,𝑏{1,,𝑁} and (𝑎,𝑏)(𝑛,𝑛) if 𝑁=2𝑛1. When 𝑁 is odd (𝑁=2𝑛1), one has, for the special index 𝑛, 𝐕𝑛𝑛,𝑛𝑛𝜆=𝐢2(𝜃)+2𝜆2,(𝜃)1(5.9) where 𝑚(𝜖)𝑛𝑛 are taken to be zero. For our models, the scattering process can be schematically, presented as (see in Figure 1)

6. Remarks

We have obtained explicit eigenvalues of the transfer matrix 𝐓(𝑟) corresponding to our class of 𝑁2×𝑁2 braid matrices for all (𝑟,𝑁). Starting with our nested sequence of projectors, we obtained a very specific structure of 𝐓(𝑟). Exploiting this structure fully, we obtained the eigenvalues and eigenstates. The zero-sum multiplets parametrized by sets of roots of unity arose from the circular permutations of the indices in the tensor products of the basis states under the action of 𝐓(𝑟). The same structure led to systematic constructions of multiparameter Hamiltonians of spin chains related to our class of braid matrices for all 𝑁. Finally we constructed the inverse Cayley transformation for the general case (any 𝑁) giving potentials compatible with factorizability of 𝑆-matrices. The contents of such matrices (generalizing Section 5 of [7]) will be further studied elsewhere. In the treatment of each aspect we emphasized the role of a remarkable feature of our formalism—the presence of free parameters whose number increase as 𝑁2 with 𝑁. A single class of constraints (𝜃0, 𝑚(+)𝑎𝑏>𝑚()𝑎𝑏, 𝑎,𝑏{1,,𝑁}) assures nonnegative Boltzmann weights in the statistical models. But our constructions also furnish directly some important properties of such models. Thus Tr(𝐓(𝑟))=2𝑛𝑖=1𝑒𝑟𝑚(+)𝑖𝑖𝜃 (𝑁=2𝑛) and Tr(𝐓(𝑟))=2𝑛1𝑖=1𝑒𝑟𝑚(+)𝑖𝑖𝜃+1 (𝑁=2𝑛1) (for the normalization (2.16)). The eigenvalues can be ordered in magnitude by choosing the order of values of the parameters. Thus choosing 𝑚(+)11>𝑚(+)22>, the largest eigenvalue of 𝐓(𝑟) is for 𝜃>0, say 𝑒𝑟𝑚(+)11𝜃, the next largest is 𝑒𝑟𝑚(+)22𝜃 and so on. We intend to study elsewhere the properties of our models more thoroughly.

In previous papers [8, 14] we pointed out that for purely imaginary parameters (𝐢𝑚(±)𝑎𝑏 with 𝑚(±)𝑎𝑏 real) our braid matrices are all unitary, providing an entire class with free parameters for all 𝑁. Here we have not repeated this discussion. But the fact that they generate parametrized entangled states is indeed of interest.

Unitary matrices provide valid transformations of a basis (of corresponding dimension) of quantum states. One may ask the following question. If such a matrix, apart from being unitary, also satisfies the braid equation what consequence might be implied? Link between quantum and topological entanglements has been discussed by several authors [15, 16] cited in our previous papers [8, 14]. For the braid property to be relevant, a triple tensor product (𝑉𝑉𝑉) of basis space is essential. We hope to explore elsewhere our multiparameter unitary matrices in such a context.

In a following paper we will present a quite different aspect of our multiparameter braid matrices. It will be shown how for imaginary parameters (when the matrices are unitary) they can be implemented to generate quantum entanglements. Topological and quantum entanglements will thus be brought together in the two papers.

Note Added
Professor J. H. H. Perk has kindly pointed out that a class of multiparameter generalization of the 6-vertex model is provided by 𝑠𝑙(𝑚𝑛) ones [17, 18]. As the Perk-Schultz models they have been studied by many authors and have led to various important applications. (Relevant references can be easily found via ARXIV.) In this class the sources of parameters are multi-component rapidities. The study of eigenvectors and eigenvalues was pioneered in [18] the most recent followup being [19], where other references can be found. Such studies may be compared to our systematic explicit constructions for all (𝑟,𝑁).