Bi-Hamiltonian Structures and Equivalent Representations of the Pais-Uhlenbeck Model.
Bethan Turner 1 · 2025
Read Bi-Hamiltonian Structures and Equivalent Representations of the Pais-Uhlenbeck Model. by Bethan Turner 1 (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-2509-26477.txt.
Introduction
A classic model in physics pairs two oscillators in a way that can leave its energy math unbounded below. Bethan Turner, writing alone in 2025, sorts every way that model, the Pais-Uhlenbeck oscillator, can sit inside a two-dimensional space. The study also maps the model's paired energy frames, called bi-Hamiltonian structures.
New frames can be grown from the model's own symmetry moves, Turner shows. Those extra frames offer a route around the unbounded energy snag that dogs higher-derivative theories. The study checks what happens when forces between parts are added. The map may help theorists tame a model that has long looked both rich and risky.
Catalog record
| CREATOR | Bethan Turner 1 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-26477
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Bi-Hamiltonian Structures and Equivalent Representations of the Pais-Uhlenbeck Model. Bethan Turner 1 Abstract We provide a complete classification of all the ways the Pais-Uhlenbeck osicllator might be embedded in two dimensional space. We discuss the Bi-Hamiltonian structures of this model, and examine how alternative Hamiltonian structures might be generated from the dynamical Lie symmetries of the theory. We then examine how the Bi-Hamiltonian strucutre may be exploited to evade the problem of unbounded Hamiltonians that is usually associated with Higher Time Derivative Theories. The effect of interactions on this Bi-Hamiltonian structure is also considered. 1 Introduction The question of whether or not a viable physical theory can be a Higher Time Derivative Theory (HTDT), meaning a theory where the Lagrangian is at least second order in time derivatives, continues to be of some interest. While such theories are generally discounted as unphysical on the grounds that they contain fundamental instabilities associated with the absence of a lower bound on the Hamiltonian, they continue to be of interest due to the fact that they tend to be renormalisable, most notably in the case of higher derivative quantum gravity [ 1 ] . Other potential benefits are the avoidance of the self interaction problem in classical electromagnetism [ 2 , 3 ] , or the avoidance of the cosmological singularity [ 4 ] . There is a price to pay for these appealing properties however. In [ 1 ] , it is shown that the renormalisable gravity theory comes at the cost of the introduction of either negative norm or negative energy states, a charecteristic feature of quantum HTDTs. The model proposed in [ 4 ] is known to lead to the unstable perturbing of flat space [ 5 ] . The model proposed in [ 2 ] leads to an energy momentum tensor with a negative part [ 3 ] . The fundamental reason for these problems is that, even at the classical level, the energy does not have a global minimum [ 6 ] . This in turn leads to a quantum picture where an unbounded energy eigenvalue spectrum can only be avoided at the price of the introduction of negative norm states, indeed this is what happens in [ 1 ] . Given the appealing properties of HTDTs, a large amount of work has been done on solving the problems discussed above. Ideas include imposing constraints on the system [ 7 ] , restricting our interest to low energy effective theories [ 8 ] and distinguishing between benign and malicous theories, where we search for systems with stable dynamics despite the presence of higher time derivatives [ 9 , 10 ] . One of the most prototypical HTDTs is a system known as the Pais-Uhlenbeck oscillator, first studied in [ 11 ] . There is already a large amount written about this model, including many works on how the ”ghost problem” (the name typically given to the problems associated with HTDTs) might be resolved and/or circumvented. For instance it is sometimes suggested that it is possible to embed the model in a two dimensional space in such a way as to preserve the dynamics but give a positive definite Hamiltonian [ 12 , 13 , 14 ] . Other suggestions include generating a positive definite Hamiltonian from the conserved quantities of the model [ 15 ] , or that the existence of a positive definite conserved quantity stabilises the model [ 16 ] . Many other works have of course been written, but it is not possible to cite them all here. In this article we will review and classify all the different ways this higher derivative model may be equivalently represented as a first order two dimensional model. We will also show how the Bi-Hamiltonian structure of this model, which as far as we can tell was first derived in [ 17 ] , can be found via consideration of the Lie symmetries of this model. I will then discuss how attempts to write down a positive definite two dimensional Hamiltonian of this system only make sense in relation to the Bi-Hamiltonian structure. In the last section I will review how this is changed by the inclusion of interactions. A more detailed discussion of the work in this paper can be found in [ 18 ] . 2 The Pais-Uhlenbeck Oscillator, Classical Dynamics and Quantisation The Pais-Uhlenbeck oscillator is probably one of the most extensively studied HTDTs, beginning with its first consideration in [ 11 ] . In its simplest version it is a linear theory that is quadratic in derivatives such that the lagrangian and equation of motion are given by where we define the parameters \alpha:=\omega_{1}^{2}+\omega_{2}^{2} , \beta:=\omega_{1}^{2}\omega_{2}^{2} , \omega_{1},\omega_{2}\epsilon\mathbb{R} . The phase space of this system can be constructed using an old method devised by Ostorgradsky in the 19^{th} century, which gives a higher derivative generalisation of the Legendre transform, see [ 19 , 6 ] . From the four dimensional configuration space described in ( 1 ), we get a four dimensional phase space with coordinates From the coordinates in ( 2 ) we can construct the Hamiltonian and Poisson tensor \displaystyle J_{1}=\frac{\partial}{\partial X_{1}}\wedge\frac{\partial}{\partial P_{1}}+\frac{\partial}{\partial X_{2}}\wedge\frac{\partial}{\partial P_{2}},=-\frac{\partial}{\partial q}\wedge\frac{\partial}{\partial q^{(3)}}+\frac{\partial}{\partial\dot{q}}\wedge\frac{\partial}{\partial\ddot{q}}+\alpha\frac{\partial}{\partial\ddot{q}}\wedge\frac{\partial}{\partial q^{(3)}} (3) \displaystyle H_{1}=P_{2}\ddot{q}\left(P_{1},P_{2},X_{1},X_{2}\right)+P_{1}X_{2}-L=-\dot{q}q^{(3)}+\frac{1}{2}\ddot{q}^{2}-\frac{1}{2}\beta q^{2}-\frac{1}{2}\alpha\dot{q}^{2}. (4) From the Hamiltonian we can see the origin of the problem with HTDTs. It is linear in the coordinate P_{1} , meaning that, as mentioned above, it does not posses a global minimum. This is what is thought to be the origin of the instability, allowing for runaway solutions even at the classical level. What is noteworthy about the Pais-Uhlenbeck oscillator is that in its free form it does not exhibit runaway behaviour. In fact, as we will now see, if we solveequation ( 1b ) we find oscillatory solutions that essentially matches the sum of two harmonic oscillators, \displaystyle q=a_{1}^{*}e^{i\omega_{1}t}+a_{1}e^{-i\omega_{1}t}+a_{2}e^{-i\omega_{2}t}+a_{2}^{*}e^{i\omega_{2}t} , \displaystyle a_{i}=a_{i}\left(q,\dot{q},\ddot{q},q^{(3t)}\right)\big\rvert_{t=0}. (5) Therefore in its free form this theory is perfectly well behaved. This can change in the presence of interactions, but this issue will be dealt in section 5 . Before proceeding a brief comment on the quantum theory is warranted. We canonically quantise in the usual way, promoting the coordinates to operators q^{i}\lvert_{t=0}\to\hat{q}^{(i)} , which in turn means the coefficients in ( 5 ) are also promoted to operators, a_{i}\to\hat{a}_{i} , a^{*}_{i}\to\hat{a}_{i}^{\dagger} . The classical Poisson brackets are also promoted to commutators, allowing us to derive the relations, where R_{i}:=\omega_{i}^{2}\left(\omega_{1}^{2}-\omega_{2}^{2}\right) . The quantum Hamiltonian can then be expressed as \displaystyle\hat{H}=2\left(R_{1}\hat{a}_{1}^{\dagger}\hat{a}_{1}-R_{2}\hat{a}_{2}^{\dagger}\hat{a}_{2}\right)+\frac{1}{2}\left(\omega_{1}+\omega_{2}\right). (7) To see the quantum analogue of the classical problem discussed above, our two options for the choice of ground state. In the first option, we treat both modes \omega_{1} and \omega_{2} as positive energy modes, such that the ground state vanishes under the action of both annihalation operators \displaystyle\hat{a}_{i}\ket{0} \displaystyle=0, \displaystyle\bra{0}\hat{a}_{i}\hat{a}_{i}^{\dagger}\ket{0} \displaystyle=\frac{(-1)^{i-1}}{2R_{i}}, \displaystyle\hat{H}\hat{a}_{i}^{\dagger}\ket{0} \displaystyle=\frac{\left(\omega_{1}+\omega_{2}\right)}{2}+\omega_{i}. (8) In this choice, the Hamiltonian eigenvalue spectrum is bounded, the ground state energy is the lowest energy. However, it also means we have negative norm states, such as the one shown in for ( 9b ). Alternatively, we can choose \omega_{1} and \omega_{2} to be positive and negative energy modes respectively, in which case we have, \displaystyle\hat{a}_{1}\ket{0} \displaystyle=0, \displaystyle\bra{0}\hat{a}_{1}\hat{a}_{1}^{\dagger}\ket{0} \displaystyle=\frac{1}{2R_{1}}, \displaystyle\hat{H}\hat{a}_{2}\ket{0} \displaystyle=\frac{\left(3\omega_{1}-\omega_{2}\right)}{2} (9a) \displaystyle\hat{a}_{2}^{\dagger}\ket{0} \displaystyle=0, \displaystyle\bra{0}\hat{a}_{2}^{\dagger}\hat{a}_{2}\ket{0} \displaystyle=\frac{1}{2R_{2}}, \displaystyle\hat{H}\hat{a}_{2}\ket{0}