Controlling the entanglement of mechanical oscillators in composite optomechanical system
Zhang Jun1, Mu Qing-Xia2, Zhang Wen-Zhao3, †
School of Mathematics and Statistics, Guizhou University of Financeand Economics, Guiyang 550025, China
Mathematics and Physics Department, North China Electric Power University, Beijing 102206, China
Beijing Computational Science Research Center (CSRC), Beijing 100193, China

 

† Corresponding author. E-mail: zhangwz@csrc.ac.cn

Abstract

A controllable entanglement scheme of two mechanical oscillators is proposed in a composite optomechanical system. In the case of strong driving and high dissipation, the dynamics of the movable mirror of the optomechanical cavity is characterized by an effective frequency in the long-time evolution of the system. Considering the classical nonlinear effects in an optomechanical system, we investigate the relationship between the effective frequency of the movable mirror and the adjustable parameters of the cavity. It shows that the effective frequency of the movable mirror can be adjusted ranging from ωm (the resonance frequency of the coupling oscillator) to −ωm. Under the condition of experimental realization, we can generate and control steady-state entanglement between two oscillators by adjusting the effective frequency of the movable mirror and reducing the effective dissipation by selecting the parameter of the cavity driving laser appropriately. Our scheme provides a promising platform to control the steady-state behavior of solid-state qubits using classical manipulation, which is significant for quantum information processing and fundamental research.

1. Introduction

Quantum entanglement generation and control has been studied in various physical systems[1] such as electronic,[2,3] photonic,[4,5] and atomic and molecular systems,[6,7] ranging from microscopic systems to mesoscopic devices.[8] It can help us understand the fundamentals of quantum phenomena, such as quantum decoherence[9,10] and quantum-classical boundary.[11] In addition, it has wide applications in quantum information and quantum computation.[12,13]

Recently, there has been a great deal of interest in optomechanical systems which provides us with a convenient platform to realize macroscopic quantum effects.[14] Optomechanics explores the coupling between photons and phonons via radiation pressure. Because of the special coupling structure, optomechanics act as a coupling medium for diverse systems.[1518] The interaction between photons and phonons can be used in mass sensors,[19] force and displacement detectors,[20] and the hardware for realizing quantum information processing.[12] Long distance photon entanglement can be used to construct quantum communication channels for secure information transmission.[21] The entanglement between the photons and solid bits,[22] such as mechanical oscillators and intracavitary atoms can realize the information interaction between flying bits and local bits,[23,24] as well as quantum information storage.[13]

Generating and controlling the entanglement of mechanical oscillators are also important subjects in the study of macroscopic quantum effects.[25,26] It is difficult for us to directly realize the parametric down-conversion (PDC) interaction between two mechanical oscillators in a system containing the interaction Hamiltonian of . To achieve the effective mechanical interaction, the frequency difference between the mechanical oscillators should not be too large. Under this condition, the PDC interaction is very weak in comparison to the beam-splitter (BS) interaction in the dynamic system. In this case, it is hard to generate steady-state entanglement between oscillators. More recently, in order to prepare strongly entangled mechanical modes in the steady state, one can engineer the dissipation of the mechanical modes such that its steady state is the desired target state,[2729] and we can also construct bilinear coupling in modulated optomechanics to create steady-state entanglement.[30] Usually, these schemes need to introduce extra quantum control, which increases the difficulty of the practical implementation of the entanglement proposals.

In this paper, we propose a convenient steady-state entanglement generating and controlling scheme for mechanical oscillators in a composite optomechanical system, in which the movable mirror of the optoemchanical cavity is coupled to a mechanical oscillator. By controlling the adjustable parameters of the classical driving laser, we are able to control the steady-state entanglement between two oscillators. In our analyses, the nonlinear optomechanical interaction, caused by radiation pressure, results in an effective frequency of the movable mirror in the optomechanical system. By choosing appropriate parameters, the effective frequency of the movable mirror can be reversed to −ωm. In this case, even though the frequency of two oscillators are the same, we can still enhance the PDC interaction and suppress the BS interaction, which generates steady-state entanglement.

2. Model and Hamiltonian

Considering an optomechanical system in which the movable mirror is coupled with a mechanical oscillator, as shown in Fig. 1, the Hamiltonian reads ( ) where a, b1, and b2 are the bosonic operators for the optical cavity, movable mirror, and coupled mechanical oscillator with frequencies ωc, ωm1, and ωm2 respectively. The single-photon coupling coefficient of the optomechanical interaction is . Hd describes the standard continuous-wave driving, ωd is the angular frequency of the laser and ϵ is the cavity driving strength, given by , with P being the input power of the laser and being the input rate of the cavity. In the rotating frame under input laser frequency ωd, we obtain the quantum Langevin equations, For the purposes of discussion, we split the optical and mechanical field operators into the classical and quantum components: and , where , . The time evolution of the annihilation operators of the system in the Heisenberg picture is then governed by where is the detuning between the driving field and the cavity frequency, is the effective detuning of the system. The nonlinear equations (5) dominate the coherent mean-field amplitude evolution of the system which depends on adjustable parameters Δ, κ, and ϵ. These parameters can be controlled by the input driving laser. Therefore, it is possible to control the quantum behavior of the system by utilizing the classical laser field. In blue sideband detuning , the optomechanical system can be used to cool the mechanical oscillator.[31] In red sideband detuning , the optomechanical system can be used to generate entanglement between photons and phonons.[1] In strong driving condition , the optical mode of the system can exhibit bistable and chaos behavior.[32] The nonlinear equations (6) dominate the quantum evolution of the system. The terms and denote the field-enhanced linear interaction between optical mode and mechanical mode b1. The linearized coupling rate depends on the dynamics of the classical part. The terms and denote the nonlinear interaction between the optical mode and the mechanical mode b1. In addition, the average displacement of the mechanical oscillator, caused by the radiation pressure, , will lead to a small shift of the detuning frequency . It also adds a term of as an equivalent driving on the coupled mechanical oscillator.

Fig. 1. (color online) Schematic diagram for the composite optomechanical system, where the optomenchnical oscillator is coupled to an additional mechanical oscillator. This three-mode bosonic system is used to prepare steady-state entanglement between modes b1 and b2. The cavity mode a is driven by a laser with frequency ωd and amplitude ϵ.

Currently, most experimental realizations of cavity optomechanics are still limited in the single-photon weak coupling with a strong driving condition.[3335] Under this condition, for the quantum part, we can introduce the so-called “linearized” approximate description[14] of optomechanics when the nonlinear terms and can be omitted since they are relatively small with respect to the factor α. Thus, equation (6) can be formally integrated as

We consider the case of , i.e., the mode is rapidly oscillating in evolution. Then the time evolution of can be approximately described as , where . Substituting it into the equation of , we have where denotes the effective dissipation. Under the condition of , the term containing is a fast decaying term, thus it can be neglected. Then we have We consider the case of , where the second term can be ignored as a higher order small quantity. Substituting into Eqs. (6), we find that the dynamics of modes b1 and b2 are independent of mode a. From Eqs. (6) we obtain the approximated expressions, where denotes the effective frequency of the movable mirror, denotes the effective dissipation of the movable mirror, is the effective amplification factor caused by the optomechanical interaction, is the effective damping of the movable mirror. Thus, the effective Hamiltonian is In addition, the mean value β causes an effective driving on the coupled oscillator, i.e. . This equivalent driving is a coherent driving which can be used to enhance the mean number of the coupled oscillator. The effective frequency ωeff is no longer a fixed parameter. Instead, it can be controlled by the laser driving strength ϵ and detuning Δ. Thus, it is possible to control the quantum behavior between the two oscillators.

3. Controlling the entanglement of mechanical oscillators

As one can tell, the effective frequency ωeff is a function of G and frequency detuning Δ. Thus, the optical driving field can indirectly modulate the frequency of the movable mirror through the interaction from radiation pressure. When , analogous to the anti-Stokes regime, BS interaction is on resonance in the system, i.e., . This interaction will cause the state transfer between the oscillators, sideband cooling, etc.[35,36] When , analogous to the Stokes regime, PDC interaction is on resonance in the system, i.e. . This interaction will cause the entanglement of mechanical oscillators, squeeze of mechanical oscillators, etc.[1,37] Noting that since we select approximately equal to ωm1, ξ should be large enough to make , so that the effective dissipation will also be magnified. In order to get a large value of ξ and reduce the effective dissipation, a larger G is needed, which means, a strong driving strength ϵ and a small dissipation rate κ are needed.

The interaction between light and mechanical motion due to radiation pressure is intrinsically nonlinear. Several theoretical studies of the nonlinear effect in optomechanical systems have been reported recently.[13,32,38] Although we have used the linearized approximation in the quantum part, for rigorous consideration, especially for strong driving condition, we should also examine the nonlinear behavior of the optomechanical system when discussing the quantum effect between the two oscillators. Without losing generality, we set in the following discussion. To study the nonlinear behavior of the system, we first consider the classical part of the dynamics, i.e., Eqs. (5). In the steady state they read, Combining the above equations we obtain a third-order polynomial root equation for the mean-field cavity occupation, where is the mean photon number in the cavity, and is the nonlinear parameter. This equation is commonly used to discuss the bistability of the optomechanical system.[13,32,38] Equation (14) indicates that the mean-field equations have either one or three solutions depending on the number of real roots of the polynomial. As shown in Fig. 2, we exhibit the bistability behavior of the linearized coupling rate G as a function of driving strength ϵ, single-photon coupling rate g, detuning Δ, and cavity dissipation rate κ (the black dashed line denotes the metastable state solution,[32] the same as in Fig. 3). The same behavior has also been discussed in Ref. [32]. We can obtain a strong linearized coupling rate , in weak single-photon coupling rate and large dissipation rate regime. When using ϵ and G to adjust the value of ωeff, we need to ensure that the condition is satisfied to ensure the rationality of the approximation. The solution corresponding to the blue-solid line in Figs. 2(a) and 2(b) meets our needs. Therefore, in the following discussion, we will only deal with the steady state solution corresponding to the blue-solid line and ignore others.

Fig. 2. (color online) (a) The coupling rate G as a function of driving strength ϵ, where , , , and . (b) G as a function of single-photon coupling rate g, where , , , and . (c) G as a function of detuning frequency Δ, where , , , and . (d) G as a function of dissipation rate κ, where , , , and .
Fig. 3. (color online) (a) and (c) Effective frequency ωeff and effective dissipation rate γeff as a function of driving strength ϵ. The detuning frequency . (b) and (d) ωeff and γeff as a function of detuning frequency Δ. The driving strength . Other parameters are , , and .

The corresponding bistable behavior of the effective frequency ωeff and the effective dissipation rate are shown in Fig. 3. In Figs. 3(a) and 3(b), we plot the effective frequency ωeff as a function of ϵ and Δ, showing an obvious bistable behavior of ωeff. We can adjust the effective frequency from ωm2 to by increasing driving strength and frequency detuning. In Figs. 3(c) and 3(d), we plot the effective frequency γeff as a function of ϵ and Δ, showing an obvious bistable behavior of γeff. The effective damping rate increases with the increase of the driving strength, and decreases with the increase of the detuning frequency.

Considering the effect of cavity dissipation rate κ, we plot the density graph of ωeff as a function of ϵ, κ, and Δ, κ in Figs. 4(a) and 4(b), respectively. The blue-dashed line indicates , and the red-dashed line indicates . As shown in Fig. 4(a), with the increase of the driving strength, the effective frequency will shift from the intrinsic frequency of the oscillator to the negative frequency. It is worth noting that the value of driving strength ϵ = 0 cannot be achieved in our approximation. We draw from ϵ = 0 only for the completeness of the figure. As shown in the blue-dashed line, when the driving strength is large enough, the effective frequency becomes , and the original BS interaction turns into PDC interaction. In addition, to achieve , we need to increase the driving strength with the increase of dissipation rate. Shown as the red-dashed line in Fig. 4(b), when the detuning frequency of the cavity is equal to the oscillator frequency ωm2, the resonance amplification occurs. Under this condition, the amplification rate ϵ reaches maximum, but the effective frequency is not shifted, which can also be told from the expression of ωeff. Shown as the blue-dashed line, the effective frequency can reach the value −ωm under the proper detuning frequency Δ. However, when the dissipation is too large, the optomechanical interaction is not sufficient to make the effective frequency equal to −ωm. Thus, under the given driving strength, the effective frequency can reach −ωm with the appropriate detuning frequency. Considering the results in 3 comprehensively, by properly increasing the detuning frequency, the effective frequency ωeff can achieve −ωm and the effective dissipation γeff can be reduced under the given driving strength. The above analysis shows that, it is possible to adjust the effective frequency from ωm2 to by controlling the driving strength ϵ and detuning Δ. Especially for the case , the system can be used to generate entanglement between two oscillators.

Fig. 4. (color online) Effective frequency ωeff as a function of ϵ, Δ, and κ. (a) ωeff as a function of ϵ and κ, the red-dashed line denotes and the blue-dashed line denotes . The detuning frequency . (b) ωeff as a function of Δ and κ, the red-dashed line denotes and the blue-dashed line denotes . The driving strength . Other parameters are the same as those in Fig. 3.

With the help of the effective Hamiltonian, we investigate the entanglement between the two oscillators. For a nonchaotic dynamical process, the properties of the quantum fluctuations can be determined from a 4 × 4 covariance matrix, , where the elements of the covariance matrix are defined as , , and

Thus, the degree of quantum entanglement between the mechanical and optical modes can be assessed by calculating the so-called logarithmic negativity,[39] defined as , where and . The matrix VA, VB, and VC are 2 × 2 matrices related to the covariance matrix V as

We plot the logarithmic negativity En as a function of optical driving strength ϵ and frequency detuning Δ with different oscillator coupling strength V in Fig. 5. Note that the linearized optomechanical interaction can lead to a cooling process.[31] Combining with the phonon–phonon parametric coupling, the system can cool the mechanical modes b1 and b2 into an entangled state in the steady state. Therefore, we have the reason to choose the thermal phonon number of the movable mirror to be equal to zero. As shown in Fig. 5(a), for a particular coupling strength V, the peak of entanglement appears when the driving strength is around . Under this driving strength, the effective frequency (this result can be obtained from Fig. 4(a)) which is consistent with the conclusions we have discussed above. It means that we can achieve the purpose of controlling the entanglement of the oscillators by adjusting the driving strength of the cavity. In addition, the degree of the entanglement increases with the oscillator coupling strength V. As shown in Fig. 5(b), for a particular coupling strength V, the peak of entanglement appears at the frequency detuning around . Under this frequency detuning, the effective frequency (this result can be obtained from Fig. 4(b)) which is consistent with the conclusions we have discussed above. This means that we can achieve the purpose of controlling the entanglement of the oscillators by adjusting the frequency detuning of the optical cavity. Again the entanglement increases with the oscillator coupling strength V.

Fig. 5. (color online) (a) Entanglement En as a function of ϵ with different coupling V. The detuning frequency . (b) En as a function of ϵ with different coupling V. The driving strength . Other parameters are , , , .
4. Discussion and conclusion

In the present experimental conditions, the mass and scale of the mechanical oscillator can achieve macroscopic level in suspended mirror optomechanical system.[8,14] The coupling between mechanical resonators can be realized in geometrically interconnected beams[40] and a charged mechanical resonator via Coulomb interaction.[41] Therefore, our system has potential applications for the generation and controling of macroscopic entanglement. Under the current experimental conditions, the coherent phonon manipulation in coupled mechanical resonators is difficult to realize. One can utilize two GaAs-based mechanical resonators by using a piezoelectric transducer[40] or two superconducting lumped-element electrical resonators using an RF SQUID-mediated tunable coupler.[42] In our scheme, a more convenient way to manipulate the mechanical resonators has been proposed, with which one only needs to control the adjustable parameters of the classical driving laser.

In our result, we proposed a scheme to generate and control the entanglement between two oscillators in a composite optomechanical system. It shows that, under the condition , the radiation pressure from the optical mode can cause an effective frequency ωeff of the mechanical mirror. The mean value of the movable mirror acts as an effective driving on the coupled oscillator. In addition, this effective frequency can be controlled by the classical optical driving, i.e., driving strength ϵ and frequency detuning Δ. In the case of anti-Stokes regime , the effective frequency of the movable mirror can be adjusted from to −ωm and the phonon can be cooled to the ground state. Then, we can generate the steady state entanglement between two oscillators by setting the effective frequency , which corresponds to the parametric down-conversion process. Manipulating the effective frequency ωeff and coupling strength V we can control the strength of the entanglement. Our scheme provides a more convenient way to generate and control the steady state entanglement of mechanical oscillators, which has potential applications in quantum control and fundamental research.

Reference
[1] Cheng J Zhang W Z Zhou L Zhang W 2016 Sci. Rep. 6 23678
[2] Cerletti V Gywat O Loss D 2005 Phys. Rev. 72 115316
[3] Wang C Zhang Y Jin G S 2011 Phys. Rev. 84 032307
[4] Sangouard N Simon C Coudreau T Gisin N 2008 Phys. Rev. 78 050301
[5] Wang G Huang L Lai Y C Grebogi C 2014 Phys. Rev. Lett. 112 110406
[6] Slodička L Hétet G Röck N Schindler P Hennrich M Blatt R 2013 Phys. Rev. Lett. 110 083603
[7] Nicacio F Furuya K Semião F L 2013 Phys. Rev. 88 022330
[8] Liao J Q Tian L 2016 Phys. Rev. Lett. 116 163602
[9] Beau M Kiukas J Egusquiza I L del Campo A 2017 Phys. Rev. Lett. 119 130401
[10] Pikovski I Zych M Costa F Brukner Č 2015 Nat. Phys. 11 668
[11] Zurek W H 2003 Rev. Mod. Phys. 75 715
[12] Chen T Y Zhang W Z Fang R Z Hang C Z Zhou L 2017 Opt. Express 25 10779
[13] Zhang W Z Cheng J Liu J Y Zhou L 2015 Phys. Rev. 91 063836
[14] Aspelmeyer M Kippenberg T J Marquardt F 2014 Rev. Mod. Phys. 86 1391
[15] Chen X Liu Y C Peng P Zhi Y Xiao Y F 2015 Phys. Rev. 92 033841
[16] Fong K Y Fan L Jiang L Han X Tang H X 2014 Phys. Rev. 90 051801
[17] Dong Y Ye J Pu H 2011 Phys. Rev. 83 031608
[18] Heikkilä T T Massel F Tuorila J Khan R Sillanpää M A 2014 Phys. Rev. Lett. 112 203603
[19] He Y 2015 Appl. Phys. Lett. 106 121905
[20] Zhang W Z Han Y Xiong B Zhou L 2017 New J. Phys. 19 083022
[21] Černotík O C V Hammerer K 2016 Phys. Rev. 94 012340
[22] Liao J Q Wu Q Q Nori F 2014 Phys. Rev. 89 014302
[23] Lian J Liu N Liang J Q Chen G Jia S 2013 Phys. Rev. 88 043820
[24] Stannigel K Rabl P Sørensen A S Lukin M D Zoller P 2011 Phys. Rev. 84 042341
[25] Thompson J D Zwickl B M Jayich A M Marquardt F Girvin S M Harris J G E 2008 Nature 452 72
[26] Mu Q Zhao X Yu T 2016 Phys. Rev. 94 012334
[27] Wang Y D Clerk A A 2013 Phys. Rev. Lett. 110 253601
[28] Tan H Li G Meystre P 2013 Phys. Rev. 87 033829
[29] Woolley M J Clerk A A 2014 Phys. Rev. 89 063805
[30] Wang M X Y Wang Y D You J Q Wu Y 2016 Phys. Rev. 94 053807
[31] Liu Y C Liu R S Dong C H Li Y Gong Q Xiao Y F 2015 Phys. Rev. 91 013824
[32] Ghobadi R Bahrampour A R Simon C 2011 Phys. Rev. 84 033846
[33] Gröblacher S Hammerer K Vanner M R Aspelmeyer M 2009 Nature 460 724
[34] Chan J Alegre T P M Safavi-Naeini A H Hill J T Krause A Gröblacher S Aspelmeyer M Painter O 2011 Nature 478 89
[35] Teufel J D Donner T Li D Harlow J W Allman M S Cicak K Sirois a J Whittaker J D Lehnert K W Simmonds R W 2011 Nature 475 359
[36] U S S Narayanan A 2013 Phys. Rev. 88 033802
[37] Qu K Agarwal G S 2015 Phys. Rev. 91 063815
[38] Doolin C Hauer B D Kim P H MacDonald A J R Ramp H Davis J P 2014 Phys. Rev. 89 053838
[39] Plenio M B 2005 Phys. Rev. Lett. 95 090503
[40] Okamoto H Gourgout A Chang C Y Onomitsu K Mahboob I Chang E Y Yamaguchi H 2013 Nat. Phys. 9 598
[41] Ma P C Zhang J Q Xiao Y Feng M Zhang Z M 2014 Phys. Rev. 90 043825
[42] Tian L Allman M S Simmonds R W 2008 New J. Phys. 10 115001