Gravitationally Induced Entanglement Between Particles in Harmonic Traps: Limits for Gaussian States
Julia Tokarska, Andrzej Dragan · 2025
Read Gravitationally Induced Entanglement Between Particles in Harmonic Traps: Limits for Gaussian States by Julia Tokarska, Andrzej Dragan (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-2512-24312.txt.
Introduction
Julia Tokarska and Andrzej Dragan test a proposed probe of quantum gravity. If gravity is quantum, it should entangle two masses that interact only through gravitation. Their system holds two particles in harmonic traps. The particles start in thermal states or in two-mode squeezed states. The authors compute how much entanglement gravity can build.
For thermal states, a ceiling appears. Above a maximum temperature, no entanglement forms at all. For fixed settings, one trap frequency works best. It maximizes the logarithmic negativity, their entanglement measure. Squeezing the start state does not add entanglement. It does widen the temperature range where entanglement survives. A general bound for all Gaussian states is also derived. Ground and squeezed states saturate that bound. The sober message is scale: the entanglement is extremely small. Observing it will challenge experimenters severely.
Catalog record
| CREATOR | Julia Tokarska, Andrzej Dragan |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-24312
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Gravitationally Induced Entanglement Between Particles in Harmonic Traps: Limits for Gaussian States Julia Tokarska, Andrzej Dragan Abstract Gravitationally induced entanglement has been proposed as a probe of the quantum nature of gravity. This work analyzes a system of two particles in harmonic traps interacting only through gravity, considering thermal and two-mode squeezed initial states. For thermal states, a maximum temperature is identified above which entanglement cannot be generated, and for fixed system parameters an optimal trap frequency that maximizes the logarithmic negativity is found. Squeezing the initial state does not further enhance the entanglement generation, but increases the temperature range over which it can be observed. Extending the analysis to general Gaussian states, an upper bound on the achievable entanglement is derived and shown to be saturated, for example, by ground and squeezed states. The results show that the amount of entanglement generated in this setup is extremely small, highlighting the experimental challenges of observing gravitationally induced quantum effects. I Introduction Although much effort has gone into formulating a quantum theory of gravity, it remains unclear whether gravity is fundamentally quantum, making experimental verification of its nature essential. A natural approach is to study gravitons, which has long been considered beyond reach 17 . Recent work, however, has proposed ways to detect individual quanta of gravitational radiation 21 , though the results could still be explained by a classical description of gravity 7 . An alternative approach is to study systems in which gravity mediates interactions between quantum objects 4 ; 15 . In such setups, the appearance of entanglement between two systems interacting only via gravity would indicate that it acts as a quantum mediator, since such correlations cannot arise classically. While the proposals in 4 ; 15 focused on massive objects placed in superpositions with large spatial separations, there is growing interest in continuous-variable systems, which are simpler to implement experimentally. Recent studies have investigated gravitational entanglement between particles in harmonic traps prepared in the ground state 5 , coherent states 23 , squeezed states 16 ; 14 , or thermal states 14 ; 9 . Other work has explored the possibility of detecting gravity-induced squeezing 9 . This work analyzes a system of two particles of mass m , each trapped in a separate harmonic potential with frequency \omega and separated by a distance d , interacting only via gravity (Fig. 1 ). The Hamiltonian is \begin{split}\hat{H}=\frac{\hat{p}_{1}^{\;2}+\hat{p}_{2}^{\;2}}{2m}+\frac{1}{2}m\omega^{2}(\hat{x}_{1}^{\;2}+\hat{x}_{2}^{\;2})\\ -\frac{Gm^{2}}{|d+\hat{x}_{1}-\hat{x}_{2}|},\end{split} (1) where \hat{p}_{i} and \hat{x}_{i} are the momentum and position operators of the i th particle. The gravitational interaction, following 15 , is approximated as Newtonian. m , \omega m , \omega -\frac{Gm^{2}}{|d+\hat{x}_{1}-\hat{x}_{2}|} Figure 1: Experimental setup: two particles of mass m trapped in harmonic potentials with frequency \omega , interacting only via gravity. This work analyzes entanglement generation from thermal and two-mode squeezed states, quantifying the entanglement produced during their evolution. For given values of d , \omega and m , a temperature is identified above which thermal states remain separable. The analysis further reveals that, for fixed d , m and temperature, an optimal trapping frequency exists that maximizes the logarithmic negativity. Extending the study to general Gaussian initial states, an upper bound on entanglement enhancement is established, along with examples of states that saturate this bound. The paper is organized as follows. Section II presents the time evolution of the system. Section III discusses entanglement generated from thermal initial states, while Section IV focuses on two-mode squeezed states. Section V derives an upper bound on entanglement enhancement. Finally, Section VI summarizes the main results. II Time evolution The analysis focuses on masses prepared near their ground state in the harmonic traps, where relative position fluctuations are small compared to the separation d . In this regime, the Hamiltonian ( 1 ) can be approximated as \begin{split}\hat{H}\approx&\;\frac{\hat{p}_{1}^{2}+\hat{p}_{2}^{2}}{2m}+\frac{1}{2}m\omega^{\prime 2}\left[\left(\hat{x}_{1}+\delta\right)^{2}+\left(\hat{x}_{2}-\delta\right)^{2}\right]\\ &+\frac{2Gm^{2}}{d^{3}}\left(\hat{x}_{1}+\delta\right)\left(\hat{x}_{2}-\delta\right)+\text{const},\end{split} (2) with the parameters \omega^{\prime} and \delta defined as \omega^{\prime 2}=\omega^{2}-\frac{2Gm}{d^{3}},\quad\delta=\frac{Gmd}{\omega^{2}d^{3}-4Gm}. (3) and a constant term that can be ignored. The Hamiltonian is conveniently expressed in terms of the dimensionless operators \hat{q}_{i} and \hat{k}_{i} , defined as \begin{split}\hat{q}_{1,2}&=\sqrt{\frac{m\omega^{\prime}}{\hbar}}\;\left(\hat{x}_{1,2}\pm\delta\right),\\ \hat{k}_{1,2}&=\frac{1}{\sqrt{\hbar m\omega^{\prime}}}\;\hat{p}_{1,2},\end{split} (4) which satisfy the commutation relations: [\hat{q}_{i},\hat{k}_{j}]=i\delta_{ij}. (5) In this basis, the Hamiltonian takes the form: \begin{split}\hat{H}=\frac{1}{2}\hbar\omega^{\prime}(\hat{q}_{1}^{2}+\hat{q}_{2}^{2}+\hat{k}_{1}^{2}+\hat{k}_{2}^{2})\\ +\frac{Gm\hbar}{d^{3}\omega^{\prime}}(\hat{q}_{1}\hat{q}_{2}+\hat{q}_{2}\hat{q}_{1}).\end{split} (6) This expression can be written more compactly by introducing the vector \hat{\boldsymbol{X}} and the matrix \boldsymbol{H} : \hat{\boldsymbol{X}}=\begin{bmatrix}\hat{q}_{1}\\ \hat{k}_{1}\\ \hat{q}_{2}\\ \hat{k}_{2}\end{bmatrix},\quad\boldsymbol{H}=\frac{1}{2}\hbar\omega^{\prime}\begin{bmatrix}1&0&\varepsilon&0\\ 0&1&0&0\\ \varepsilon&0&1&0\\ 0&0&0&1\end{bmatrix}, (7) where \varepsilon=\frac{2Gm}{d^{3}\omega^{\prime 2}} . In this notation, the commutation relations are given by: [\hat{X}_{j},\hat{X}_{k}]=i\Omega_{jk}, (8) where \boldsymbol{\Omega}=\bigoplus_{i=1}^{N}\boldsymbol{\omega},\quad\boldsymbol{\omega}=\begin{bmatrix}0&1\\ -1&0\end{bmatrix}, (9) and the Hamiltonian then takes the form: \hat{H}=\hat{\boldsymbol{X}}^{T}\boldsymbol{H}\hat{\boldsymbol{X}}. (10) The initial states considered here are Gaussian, and since the Hamiltonian is quadratic in the operators \hat{q}_{i} and \hat{k}_{i} , they remain Gaussian under time evolution. Such states are fully described by their vector of first moments and the covariance matrix of the \hat{q}_{i} and \hat{k}_{i} operators. The time evolution of the covariance matrix is given by 3 : \boldsymbol{\sigma}(t)=\boldsymbol{S}\,\boldsymbol{\sigma}(0)\,\boldsymbol{S}^{T}, (11) where 1 1 1 In 3 the symplectic matrix is obtained as \boldsymbol{S}=e^{\frac{t}{\hbar}\boldsymbol{\Omega H}} (Eq. (76), p. 22); however, the calculation yields an additional factor of 2 in the exponent. \boldsymbol{S}=e^{\frac{2t}{\hbar}\boldsymbol{\Omega H}} . For the Hamiltonian ( 10 ), the corresponding symplectic matrix takes the form \boldsymbol{S}_{G}=\exp\left(\omega^{\prime}t\begin{bmatrix}0&1&0&0\\ -1&0&-\varepsilon&0\\ \ 0&0&0&1\\ -\varepsilon&0&-1&0\\ \end{bmatrix}\right)\\ . (12) Its elements can be expressed explicitly as combinations of trigonometric functions; however, due to their complexity, the full expressions are omitted here. Entanglement is quantified using the negativity and the logarithmic negativity, which are well suited for bipartite Gaussian states 12 : \mathcal{N}(\rho)=\frac{\|\tilde{\rho}\|-1}{2},\qquad E_{\mathcal{N}}=\log_{2}\|\tilde{\rho}\|. (13) Here \rho denotes the density matrix of the bipartite system, \tilde{\rho} its partial transpose, and \|A\|=\mathrm{Tr}\sqrt{A^{\dagger}A} the trace norm. For bipartite Gaussian states, positivity of the partially transposed density matrix is equivalent to separability, ensuring that these measures detect all entangled states 20 . For a bipartite Gaussian state with covariance matrix \boldsymbol{\sigma}=\begin{bmatrix}\boldsymbol{\alpha}&\boldsymbol{\gamma}\\ \boldsymbol{\gamma}^{T}&\boldsymbol{\beta}\end{bmatrix}, (14) where \boldsymbol{\alpha} , \boldsymbol{\beta} , and \boldsymbol{\gamma} are 2\times 2 matrices, the entanglement is determined by the symplectic eigenvalue 2 \nu=\sqrt{2\left(\Delta-\sqrt{\Delta^{2}-4\det\boldsymbol{\sigma}}\right)}, (15) with \Delta=\det\boldsymbol{\alpha}+\det\boldsymbol{\beta}-2\det\boldsymbol{\gamma} . The negativity and logarithmic negativity are then given by 2 : \begin{split}\mathcal{N}&=\max\left(0,\frac{1-\nu}{2\nu}\right),\\ E_{\mathcal{N}}&=\max\left(0,-\log_{2}\nu\right).\end{split} (16) The state is entangled if \nu<1 . III Thermal states We start by examining thermal states, which naturally arise in realistic experimental settings where perfect ground-state preparation cannot be achieved. A thermal state at temperature T is described by the density matrix \begin{split}\rho(T)&=N\sum_{n,m=0}^{\infty}e^{-\frac{\hbar\omega}{k_{B}T}(n+m)}\lvert n,m\rangle\langle n,m\rvert,\\ N&=\left(1-e^{-\frac{\hbar\omega}{k_{B}T}}\right)^{2},\end{split} (17) where \lvert n,m\rangle denote the eigenstates of the uncoupled Hamiltonian \hat{H}_{\omega}=\hat{H}_{1}+\hat{H}_{2},\qquad\hat{H}_{i}=\frac{\hat{p}_{i}^{2}}{2m}+\frac{1}{2}m\omega^{2}\hat{x}_{i}^{2}. (18) These eigenstates satisfy \begin{split}\hat{H}_{1}\lvert n,m\rangle&=\hbar\omega\left(n+\tfrac{1}{2}\right)\lvert n,m\rangle,\\ \hat{H}_{2}\lvert n,m\rangle&=\hbar\omega\left(m+\tfrac{1}{2}\right)\lvert n,m\rangle.\end{split} (19) The initial covariance matrix is given by \boldsymbol{\sigma}_{T}(0)=\tfrac{1}{2}\,\theta\,\mathds{1}, (20) where \theta=\coth\!\left(\frac{\hbar\omega}{2k_{B}T}\right) . After time evolution, the covariance matrix becomes \boldsymbol{\sigma}_{T}(t)=\boldsymbol{S}_{G}\,\boldsymbol{\sigma}_{T}(0)\,\boldsymbol{S}_{G}^{T}=\tfrac{1}{2}\,\theta\,\boldsymbol{S}_{G}\boldsymbol{S}_{G}^{T}. (21) Evaluating Eq. ( 15 ) with this covariance matrix provides an expression for the generated entanglement. To extract the dominant physical contribution, we expand the result to first order in \varepsilon . Under the assumption \omega^{\prime 2}\simeq\omega^{2} , this leads to 2 2 2 The treatment of thermal states in 14 produces results compatible with Eq. ( 22 ). The maximum temperature allowing for entanglement was previously determined in 9 , which examined both gravitationally induced entanglement and squeezing. In contrast, the present analysis further reveals the presence of an optimal trap frequency for fixed mass and temperature, a feature that has not been explicitly discussed in the prior literature. : \nu_{T}(t)\approx{\theta}\left(1-\frac{2Gm}{d^{3}\omega^{2}}|\sin(\omega