---
title: Hyperbolic Calogero-Sutherland System
url: https://www.emergentmind.com/topics/hyperbolic-calogero-sutherland-system
type: topic
---

# Hyperbolic Calogero-Sutherland System

Searching arXiv for recent and foundational papers on the hyperbolic Calogero–Sutherland system and close variants.
The hyperbolic Calogero–Sutherland system is a class of classical and quantum integrable many-body systems whose defining interactions are inverse hyperbolic-square kernels, typically of \(1/\sinh^2\) type, and whose realizations are organized by root systems such as \(A_{n-1}\), \(C_n\), and \(BC_n\). In current work the family encompasses spinless and spin models, supersymmetric extensions, boundary-coupled \(BC_n\) systems, Hamiltonian and Hitchin reductions, Lax and \(r\)-matrix formulations, and exact quantum wave functions described by Baxter operators and Heckman–Opdam hypergeometric functions [2508.18864] [1205.1029] [2305.00791].

## 1. Standard hyperbolic Hamiltonians

For the \(A_{n-1}\) quantum model, a standard nonrelativistic Hamiltonian is
\[
\mathcal H = -\sum_{j=1}^n \partial_{x_j}^2 + \sum_{1\le j\ne k\le n}\frac{\pi^2 g(g-1)}{\sinh^2\!\pi(x_j-x_k)},
\]
together with a commuting family obtained after gauge transformation by the weight \(w(x)=|2\sinh \pi x|^g\) [2508.18864]. In the same setting, the basic kernel \(\mathcal K(x)=(2\cosh \pi x)^{-g}\) appears as the hyperbolic limit of the Ruijsenaars double-sine kernels, so the hyperbolic interaction is tied from the outset to a bispectral integral-operator structure rather than only to a Schrödinger operator [2508.18864].

In classical gauge-theoretic realizations, the same hyperbolic interaction emerges after diagonalizing a positive Hermitian matrix variable \(X\), eliminating off-diagonal gauge fields, and passing to logarithmic coordinates \(x_a=e^{2q_a}\). The resulting bosonic action is
\[
S=\frac12\int dt\left[ \sum_a \dot q_a^2 -\sum_{a\neq b}\frac{c^2}{4\sinh^2(q_a-q_b)} \right],
\]
which is the standard hyperbolic Calogero–Sutherland form of \(A_{n-1}\) type [1902.08023]. From the nonrelativistic limit of the trigonometric Ruijsenaars–Schneider model one also recovers the classical Lax matrix
\[
L^{CS}_{ij}=\delta_{ij}p_j+(1-\delta_{ij})\frac{v}{\sinh(q_i-q_j)},
\]
so the hyperbolic model sits naturally as a limit of a relativistic system with the same spectral data [1510.07509].

## 2. Root-system realizations and boundary terms

Beyond the \(A_{n-1}\) case, the hyperbolic Calogero–Sutherland family is organized by non-\(A\) root systems through additional sum-type and boundary interactions. A standard \(BC_n\) Hamiltonian is
\[
H(q,p)=\frac12\sum_{c=1}^n p_c^2 +\sum_{1\le a<b\le n}\left( \frac{g^2}{\sinh^2(q_a-q_b)}+\frac{g^2}{\sinh^2(q_a+q_b)} \right) +\sum_{c=1}^n\left( \frac{g_1^2}{\sinh^2(q_c)}+\frac{g_2^2}{\sinh^2(2q_c)} \right),
\]
on the Weyl chamber
\[
C_n=\{q=(q_1,\dots,q_n)\in \mathbb{R}^n \mid q_1>\cdots>q_n>0\},
\]
with three independent couplings \((g,g_1,g_2)\) in the repulsive regime [1205.1029]. In this formulation the pair terms \(q_a\pm q_b\) encode the roots \(\pm e_a\pm e_b\), while the one-body terms encode \(e_c\) and \(2e_c\).

The \(BC_n\) framework also supports more structured particle content. One Hamiltonian-reduction model interprets the system as two species of particles on the half-line, with same-species interactions of \(\sinh^{-2}(q_j\pm q_k)\) type, cross-species interactions of \(\cosh^{-2}(q_j\pm q_k)\) type, and species-dependent boundary terms involving \(\sinh^{-2}(2q_j)\), \(\sinh^{-2}(q_j)\), and \(\cosh^{-2}(q_j)\) [1105.4552]. In the three-coupling \(BC_n\) model, setting \(g_2=0\) gives the hyperbolic \(B_n\) Sutherland model, while \(g_1=0\) gives the \(C_n\) model [1109.0446]. This root-system description is the mechanism by which hyperbolic Calogero–Sutherland dynamics acquires walls, mirror interactions, and boundary fields.

## 3. Reduction, Lax representations, and \(r\)-matrices

A central structural feature of the hyperbolic Calogero–Sutherland system is its derivation by reduction from higher-dimensional free dynamics. For the \(BC_n\) case, symplectic reduction of free geodesic motion on \(U(n,n)\) with a shifted momentum map produces the physical phase space, the hyperbolic Hamiltonian, a Lax matrix \(L(q,p)\), and a dynamical \(r\)-matrix. The reduced bracket takes the form
\[
\{L_1,L_2\} = [r_{12}(q),L_1]-[r_{21}(q),L_2],
\]
with \(r_{12}(q)\) depending only on the configuration variables \(q\), not on the momenta \(p\), and not explicitly on the coupling parameters \((\mu,\nu,\kappa)\) used in the reduction [1205.1029]. The corresponding Lax equation is equivalent to Hamilton’s equations, so integrability is encoded directly at the matrix level [1205.1029].

A different but equally explicit construction appears in the generalized two-spin hyperbolic model. There the Hamiltonian is
\[
H = \frac12\sum_{j=1}^N v_j^2 + \sum_{j<k} \frac{ S_{jk}^2 + T_{jk}^2 - 2S_{jk}T_{jk}\cosh(u_j-u_k) }{ \sinh^2(u_j-u_k) },
\]
where \(S\) and \(T\) are two skew-symmetric spin matrices on commuting copies of \(\mathfrak{so}^*(N)\) [1706.08793]. The model is realized as a Hitchin system on a singular curve formed by gluing two copies of \(\mathbb{CP}^1\); the reduced phase space is
\[
\mathcal R^{\mathrm{red}}=\mathbb R^{2N-2}\times \mathcal O_{\mathfrak{so}(N)}\times \mathcal O_{\mathfrak{so}(N)},
\]
and complete integrability is established by an explicit Lax pair and a spectral-parameter-dependent classical \(r\)-matrix on that singular curve [1706.08793]. When one spin variable vanishes, the construction reduces to the standard hyperbolic spin Calogero–Sutherland model; for \(N=2\), it reproduces the \(BC_1\) potential [1706.08793].

## 4. Duality, Ruijsenaars relatives, and action-angle structure

Hyperbolic Calogero–Sutherland systems are part of a larger reduction-theoretic hierarchy in which Sutherland and Ruijsenaars-type models appear as dual coordinate charts on the same reduced space. For the \(C_n\) root system, one common symplectic reduction yields both the hyperbolic \(C_n\) Sutherland model
\[
H_S=\frac12\sum_{c=1}^n p_c^2 +\sum_{1\le a<b\le n}\left(\frac{g^2}{\sinh^2(q_a-q_b)}+\frac{g^2}{\sinh^2(q_a+q_b)}\right) +\sum_{c=1}^n\frac{g_2^2}{\sinh^2(2q_c)}
\]
and a rational \(C_n\) Ruijsenaars–Schneider–van Diejen model, with global Darboux coordinates on both sides and a symplectomorphism under which the action variables of one system become the position variables of the other [1106.2943]. The same pattern holds for the three-coupling \(BC_n\) case: the hyperbolic \(BC_n\) Sutherland model and the rational \(BC_n\) Ruijsenaars–Schneider–van Diejen model arise from one reduction picture, and the positive eigenvalues of the Sutherland Lax matrix provide the action variables on the Sutherland side [1109.0446].

A relativistic companion to this picture is obtained by Poisson–Lie reduction on the Heisenberg double of \(SU(n,n)\). The reduced Hamiltonian is a \(BC_n\)-symmetric Ruijsenaars-type model with multiplicative hyperbolic interactions and one-body boundary terms, and in the cotangent bundle or nonrelativistic limit it reduces to the standard three-parameter hyperbolic \(BC_n\) Sutherland Hamiltonian [1311.4641]. This places the hyperbolic Calogero–Sutherland system inside a hierarchy
\[
\text{Poisson--Lie reduction} \;\Longrightarrow\; \text{Ruijsenaars-type } BC_n \text{ model} \;\Longrightarrow\; \text{hyperbolic } BC_n \text{ Sutherland model},
\]
so the hyperbolic system functions both as an autonomous integrable model and as a nonrelativistic shadow of a relativistic one [1311.4641].

## 5. Quantum wave functions and special-function theory

At the quantum level, one major development is the proof that two distinct integral representations of the \(GL_n\) hyperbolic Calogero–Sutherland wave function are actually equivalent. One representation is Euler-type, built recursively from raising operators \(\Lambda_n\); the other is Mellin–Barnes type. The equality
\[
\Psi_n(x_n)=\hat\Psi_n(x_n)
\]
is established through the analysis of two families of Baxter operators, one acting in spatial variables and one in spectral variables, together with dominated-convergence arguments for the nonrelativistic limit from the hyperbolic Ruijsenaars system [2508.18864]. The same work identifies the resulting wave function with the renormalized Heckman–Opdam \(\mathfrak{gl}_n\) hypergeometric function, fixes its normalization at the origin, and derives asymptotic formulas in the Weyl chamber [2508.18864]. This closes the gap between Baxter-operator constructions, bispectrality, and the standard Heckman–Opdam framework.

A separate line of work treats a Morse-deformed hyperbolic Calogero–Sutherland operator,
\[
L_x^{cs} = \sum_{j=1}^n \frac{\partial^2}{\partial x_j^2} - \sum_{j=1}^n \Bigl(g_S e^{-x_j}-a e^{-2x_j}\Bigr) - \sum_{1\le j<k\le n}\frac{2g_M(g_M-1)}{\sinh^2\!\bigl(\tfrac12(x_j-x_k)\bigr)},
\]
as a partial confluent limit of the \(BC_n\) Heckman–Opdam hypergeometric function [2305.00791]. In that construction, the \(BC_n\) Harish–Chandra series is translated and its couplings rescaled so that the \(BC_n\) recurrence degenerates to the Morse-deformed hyperbolic Calogero–Sutherland recurrence [2305.00791]. The same paper shows that the resulting wave functions satisfy dual difference equations in the spectral variable and extend analytically in that variable, so their dependence on \(\xi\) is holomorphic rather than merely meromorphic [2305.00791].

## 6. Spin and supersymmetric extensions

Supersymmetric hyperbolic Calogero–Sutherland systems are obtained by gauging matrix superfield models. For \(\mathcal N=2\), gauging a \(U(n)\) matrix system yields a model whose bosonic core is exactly the standard \(A_{n-1}\) hyperbolic Calogero–Sutherland system, while the fermionic sector consists of matrix fermions rather than a minimal particle-by-particle superpartner set [1902.08023]. The subsequent Hamiltonian analysis produces classical and quantum supercharges and an explicit Lax pair. A notable structural difference then appears: classically, the interaction terms in the supercharges are proportional to off-diagonal fermions, whereas quantum mechanically the supersymmetry generators admit a consistent invariant subsector without off-diagonal fermion operators [1910.07348].

For \(\mathcal N=4\), the bosonic sector becomes spinful. One gauged matrix construction uses \(SU(2)\)-spinor semi-dynamical variables and odd matrix fields so that the bosonic core is the \(U(2)\)-spin hyperbolic Calogero–Sutherland system, with spin matrices
\[
(S_a)_{ik}=\bar Z_{a i} Z_{a k}
\]
obeying the \(U(2)\) algebra [2007.11424]. The model admits explicit \(\mathcal N=4\) supercharges, a Lax representation, and an invariant reduction to the spinless hyperbolic Calogero–Sutherland system by setting one \(SU(2)\) component of the spinor variables to zero [2007.11424]. More generally, Hamiltonian constructions starting from the classical \(A_n\), \(B_n\), \(C_n\), and \(D_n\) series produce \(\mathcal N=2\) and \(\mathcal N=4\) hyperbolic or trigonometric Calogero–Sutherland cousins whose bosonic cores are the standard systems; within that framework, the hyperbolic and trigonometric models appear to saturate at \(\mathcal N=4\) [2002.03929].

## 7. Related constructions and terminological boundaries

The family also appears under closely related nomenclature. In one formulation the same hyperbolic many-body system is treated as the hyperbolic Calogero–Moser system; the dynamics are derived from pole reduction of the semi-discrete KP equation, and the main result is a Lagrangian 1-form structure in both discrete and continuous time [1601.04799]. The discrete Lagrangians involve \(\ln|\sinh(x_i-\tilde x_j)|\), the temporal Lax matrices satisfy compatibility relations yielding a discrete closure relation, and two successive continuum limits produce a continuous hierarchy together with the continuous closure relation \(\partial L^{(2)}/\partial t_3=\partial L^{(3)}/\partial t_2\) [1601.04799]. In this usage, the distinction between “Calogero–Moser” and “Calogero–Sutherland” is terminological rather than structural [1601.04799].

The adjective “hyperbolic” is also used in distinct senses across adjacent literature. One direction replaces line dynamics by motion on the hyperboloid \(H^N\), with Hamiltonian
\[
H_{H^N} = \frac{p_\chi^2}{2r_0^2} + \frac{I}{r_0^2\sinh^2\chi} + V(\tanh\chi),
\]
where \(I\) is the angular Hamiltonian of a generalized rational Calogero model; the resulting hyperbolic Calogero-oscillator and Calogero-Coulomb systems on \(H^N\) are maximally superintegrable [1409.8288]. A different construction, the “hyperbolic Kac–Moody Calogero model,” is based not on \(\sinh^{-2}\) pair potentials on a line but on the real roots of the hyperbolic Lie algebra \(AE_3\), with a modular potential on the upper half-plane and Dunkl operators whose commutativity is obstructed by rank-2 hyperbolic subsystems [2203.06519]. That paper is careful not to claim nonintegrability outright; it states only that the usual Dunkl-based sufficient criterion fails [2203.06519]. The standard hyperbolic Calogero–Sutherland system is therefore best understood as the integrable \(\sinh^{-2}\)-interaction family, while neighboring “hyperbolic” constructions may refer either to ambient hyperbolic geometry or to hyperbolic Weyl groups.

Source: https://www.emergentmind.com/topics/hyperbolic-calogero-sutherland-system