Abstract
Quantum theory does not prevent entanglement from being created and observed in macroscopic physical systems, in reality however, the accessible scale of entanglement is still very limited due to decoherence effects. Recently, entanglement has been observed among atoms from thousands to millions levels in extremely lowtemperature and well isolated systems. Here, we create multipartite entanglement of billions of motional atoms in a quantum memory at room temperature and certify the genuine entanglement via Mseparability witness associated with photon statistics. The information contained in a single photon is found strongly correlated with the excitation shared by the motional atoms, which intrinsically address the large system and therefore stimulate the multipartite entanglement. Remarkably, our heralded and quantum memory builtin entanglement generation allows us to directly observe the dynamic evolution of entanglement depth and further to reveal the effects of decoherence. Our results verify the existence of genuine multipartite entanglement among billions of motional atoms at ambient conditions, significantly extending the boundary of the accessible scale of entanglement.
Similar content being viewed by others
Introduction
Quantum technologies, incorporating quantum entanglement^{1} into communication^{2,3}, simulation^{4,5,6,7,8}, computation^{9,10,11}, and metrology^{12}, exert great advantages beyond classical approaches. For a largescale multipartite entangled system, the dimension of Hilbert space will be exponentially expanded as the number of entangled particles increases, which inspires novel approaches of quantum computing or direct simulation for classically intractable problems^{3,11}. The ability to access largescale and more practical multipartite entanglement has been regarded as a benchmark for quantum information processing, like the road map towards quantum supremacy^{13}. However, decoherence resulting from strong internal interactions and coupling with the environment makes entanglement fragile, which limits the expansion of the scale of multipartite entanglement, especially reaching the level of macroscopic physical systems.
So far, significant experimental progresses have been made in realizing different classes of multipartite entanglement in different artificially engineered quantum systems. GreenbergerHorneZeilinger state, a wellknown maximally entanglement state, has already been generated at a scale of up to 12 qubits in superconducting systems^{14}, 18 qubits in photonic systems^{15}, and 20 qubits in ions systems^{16}. The exponentially low efficiency in simultaneously detecting many particles restricts its achievable scale. An interesting way to enhance the collective correlation is to create the twin Fock entanglement state in a BoseEinstein condensate (BEC) system through quantum phase transitions^{17}. W state is another representative multipartite entanglement with only one excitation shared by all particles. The requirement of detecting single excitation for W state is free from the exponential inefficiency of coincidence measurement, which makes it more easily achieved at a largescale, especially in atomic ensembles^{18,19,20,21}.
The achieved largescale entanglement states, however, have to be prepared and detected in the systems that are maintained at extremely low temperatures and well isolated with an environment to eliminate decoherence effects. The decoherence and noiseinduced by the motion and collision of roomtemperature atoms are apparently harmful^{22,23}, and therefore were avoided in purpose in previous endeavors of observing largescale entanglement. Recently, the DuanLukinCiracZoller (DLCZ) protocol operating in a faroffresonance configuration has been found capable of being operated well in roomtemperature conditions. This faroffresonance protocol well eliminates the collisioninduced fluorescence noise and endows a large memory bandwidth, which obviously solves the difficulties that the EIT and near offresonance Raman schemes meet in hot vapor cells.^{24,25,26,27}. Though being challenging, it would be more desirable to explore whether largescale entanglement can exist in ambient conditions and be shared by more motional atoms, not only for the fundamental interest of probing the boundary of quantum to classical transition, but also for future reallife quantum technologies.
Here, we experimentally demonstrate a multipartite entanglement of billions of motional atoms in a roomtemperature quantum memory. The multipartite entanglement W state in a hot atomic vapor cell is heralded by registering a Stokes photon derived from the faroffresonance spontaneous Raman scattering (SRS). In order to certify and quantify the scale of entanglement, we convert the shared excitation of W state into an antiStokes photon by applying another interrogation pulse, and reveal an entanglement depth up to billions of atoms by the witness constructed with the correlated photon statistics. The faroffresonance configuration endorses the broadband feature allowing it to be operated at a high data rate. Furthermore, our heralded and quantum memory builtin fashion of entanglement generation allows us to directly observe the dynamic evolution of entanglement depth in a dissipative environment.
Results
Creating W states among atoms
To herald multipartite entanglement W state, we adopt the SRS regime as proposed in the DuanLukinCiracZoller (DLCZ) protocol originally aiming at realizing applicable quantum repeaters^{24}. However, we have to conceive a faroffresonance scheme to avoid the huge fluorescence noise in roomtemperature atomic ensemble^{25,26,27}, which does not exist in cold ensembles and diamonds^{28,29}. The energy levels of the Λtype configuration are shown in Fig. 1a. This process will generate the correlated photonatoms pair, which can be expressed as^{29}
where ∣ε_{s}∣^{2} is the excitation probability of Stokes photon, \(\leftvac\right\rangle =\leftva{c}_{opt}\right\rangle \bigotimes \leftva{c}_{ato}\right\rangle\) is the initial product state of a photonatoms system, S and a are the annihilation operators of spinwave and Stokes photon, respectively. Here, we set the intensity of the control light pulse so weak that the excitation probability ∣ε_{s}∣^{2} is much smaller than unity. Therefore, we can ignore the higherorder terms in creating of \(\left{\psi }_{s}\right\rangle\) with extremely small probability^{24,30}. With the creation operators acting on the initial state of an atomic ensemble, the W state is written as^{22,26}\(\left{W}_{1}\right\rangle =\frac{1}{\sqrt{N}}\mathop{\sum }\limits_{j=1}^{N}{e}^{i{{\Delta }}\overrightarrow{k}\cdot {\overrightarrow{r}}_{j}}\left{g}_{1}{g}_{2}...{s}_{j}...{g}_{N}\right\rangle\), where N is the number of involving atoms, \({{\Delta }}\overrightarrow{k}\) is the wave vector of spinwave, \({\overrightarrow{r}}_{j}\) is the position vector of j_{th} excited atom. This generating process of entanglement is shown in Fig. 1b, and the W state can be heralded through the detection of one scattering Stokes photon. \(\left{W}_{1}\right\rangle\) contains only one excitation shared by all motional atoms illustrated in Fig. 1c, where every atom posesses the equal probability of being excited with spin up.
It is inevitable that the SRS in the generating process may produce highorder excitations with a comparably low probability, and such terms would change the structure of our desired multipartite entanglement. The entangled ensemble with the twoexcitation events can be generally expressed as \(\left{W}_{2}\right\rangle =\sqrt{\frac{2}{N(N1)}}\mathop{\sum }\limits_{i < j}^{N}{e}^{i{{\Delta }}\overrightarrow{k}\cdot \left({\overrightarrow{r}}_{j}+{\overrightarrow{r}}_{i}\right)}\left{g}_{1}{g}_{2}...{s}_{i}...{s}_{j}...{g}_{N}\right\rangle\), and higherorder events have negligible contributions. To certify and qualify the W state, we need to apply another optical probe pulse to convert the shared single excitation in the atomic ensemble into an antiStokes photon, as shown in Fig. 1a, b. In order to obtain the information of entanglement depth, we analyze the photon number statistics of the correlated Stokes and antiStokes photons via a Hanbury Brown–Twiss interferometer (as Fig. 1b shows and more experimental details see Methods). Due to the decoherence effects, as Fig. 1d shows, the atomic ensemble with highorder excitations will evolve to several subgroups, where each part shares single excitation^{21}. Suppose that the whole atomic ensemble is in a pure state, containing M separable parts, which can be expressed in a product form \(\left\psi \right\rangle =\left{\psi }_{1}\right\rangle \otimes \left{\psi }_{2}\right\rangle ...\otimes \left{\psi }_{M}\right\rangle\), where M is the number of separable subgroups, \(\left{\psi }_{i}\right\rangle (i=1,...,M)\) represents each separable group that may contain individual multipartite entanglement, while different subgroups are independent of the others. For an extremely largescale multipartite entanglement, it is convenient to adopt the concept of entanglement depth to characterize its scale information. Here, we can define entanglement depth as D = N/M^{21}, which represents the smallest number of genuinely entangled particles among the subgroups, and N is the number of total atoms participating in the interaction.
Verifying the W states by witness
In order to quantify the multipartite entanglement, we adopt the entanglement witness compatible with photon number statistics of the correlated photons pair. Such witness is efficient especially when the vacuum component of the state is dominant^{21}. For a given M value, we ought to determine the lower bound of the entanglement state with D particles entangled (see Methods). The witness operator can be expressed as
where the two key parameters p_{1} and p_{2} are projective probabilities in the forms of \({\left\left\langle {W}_{1} \psi \right\rangle \right}^{2}\), \({\left\left\langle {W}_{2} \psi \right\rangle \right}^{2}\), and \({p}_{2}^{bound}({p}_{1},M)\) stands for the theoretical minimal value of p_{2} under the condition of fixed p_{1} and M value. For the density matrix ρ of experimental state, tr(ρω^{M}) < 0 means that the entanglement depth is at least D = N/M. In actual experiments, p_{1} should be defined as the conditional probability of detecting a correlated antiStokes photon with a heralding forward Stokes photon, and the probability p_{2} stands for the probability of two excitation events, which is deduced by the autocorrelation function \({g}_{A{S}_{1}\cdot A{S}_{2} S}^{(2)}=2{p}_{2}/{p}_{1}^{2}\) measured by a Hanbury Brown–Twiss interferometer.
The decoherence effects can be revealed by the observation of entanglement depth’s evolution via adjusting the delay of optical probe pulse in our quantum memory builtin configuration. The p_{1} can be influenced by the retrieval efficiency of quantum memory, the photon loss of the channels and detectors. Therefore, the final experimental data should be handled as the following two stages: the raw measured data and the processed data after subtracting the loss of the channels and detectors. The latter reflects the genuine entanglement state at the moment just after applying the probe pulse. The M values of entanglement state evolving with storage time are shown in Fig. 2, where the energy of the light pulses is 225 pJ. The data point below the boundary curve of fixed M value indicates that there is an entanglement depth at least N/M. Our experimental data show that the number of entanglement subgroups increases as the memory time elapsing, i.e. the decoherence effects caused by the thermal motions of atoms^{22,26} will tremendously influence the structure of entanglement, which is consistent with the physical picture depicted in Fig. 1d.
During the process of verifying the existence of entanglement, the collective enhancement effect contributes to the transducing of Stokes photon due to the phase coherence of W state^{24}. The variation of p_{1} with the delay of the probe pulse is shown in Fig. 3a. Due to the decoherence of phase mainly resulting from the motions of warm atoms, the effect of collective enhancement becomes deteriorative, which results in the exponential decay of p_{1}. What is more, we measure the crosscorrelation between the correlated photons and the autocorrelation of the retrieved antiStokes photon as shown in Fig. 3b. The degrade of quantum correlation and single photon characteristic implies the variation of the structure of multipartite entanglement according to the relation of p_{2} and \({g}_{A{S}_{1}\cdot A{S}_{2} S}^{(2)}\) (see Methods), which are consistent with the deduced M values in Fig. 2. Our results well exhibit the quantumtoclassical transition in multipartite entanglement of billions of motional atoms heralded by a single photon.
From another perspective, we can demonstrate how the entanglement depth varies with the delay time. In order to determine the information of entanglement depth, the number of total cesium atoms involved in the interaction is the key parameter that should be measured precisely. The atomic density can be obtained by fitting the measured transmission rate of light with different frequencies according to the absorption model, and the total atomic number is evaluated by the coherence analysis of the collective enhancement effect according to the collection optics (see Methods). The results show that there are nearly at least 1.77 billion motional atoms sharing one excitation constituting the W state. As is shown in Fig. 3c, despite the fast decrease of entanglement depth resulting from the increased noise and the destructive effects of decoherence, there is still considerable entanglement depth in the warm atomic ensemble after storing for microseconds level.
It is also accessible to manipulate the size of the largescale entanglement state in the macroscopic ensemble by changing our experimental parameters. What dominantly influences the p_{1} probability in our experiment is the energy of the addressing light pulse, which determines the excitation probability of the Stokes photon during the SRS process. The M values of multipartite entanglement created by different pulse energies, 115.5 pJ, 225 pJ, 330 pJ respectively, are shown in Fig. 4a. The dependence of entanglement depth on excitation probability, the light pulse energy is also analyzed and shown in Fig. 4b. The results show that the stronger light pulse energy has a higher p_{1} probability because of the higher converting efficiency of the W state, but has a smaller M value, which indicates that the structure of multipartite entanglement has not been deteriorated by noise. Limited by the power of our laser setup, we can’t further explore the qualitative relationship between the entanglement depth and the pulse energy. Based on the fact that the crosscorrelation between Stokes and antiStokes photons would decrease when the pulse energy becomes stronger^{26}, we may deduce that the entanglement depth would also meet the turning point because of the increasing noise level.
Discussion
In summary, we have experimentally demonstrated a multipartite entanglement of billions of motional atoms heralded by a single photon. With the quantum memory builtin and broadband capacities, we have efficiently displayed the dynamical evolution of entanglement depth and decoherence effects. Remarkably, this certifies the feasibility for testing quantumtoclassical transition with multipartite entanglement at the ambient environment, which is prominently different from the other platforms with rigorous conditions. Our work has certified that quantum entanglement can be observed in a macroscopic roomtemperature atomic ensemble with motional atoms and demonstrated the accessibility of feasible manipulations of entanglement depth, which expands the bound of operating largescale multipartite entanglement and may stimulate a wide spectrum of applications for the future quantum information science and technologies.
Creating a largescale multipartite entanglement of more atoms is possible, a larger beam waist and stronger energy of light pulse will be helpful with the prerequisite of wellcontrolled levels of noise. What’s more, the larger beam waist can mitigate the detrimental effect of decoherence brought by the thermal motions of atoms, since there is a broader space to prevent warm atoms from escaping from the interaction region, which leads to a longer lifetime of multipartite entanglement. Recent works also show that the antirelaxation coating of vapor cells will preserve the coherence for longer lifetime^{31,32}, which may be beneficial for improving the maintenance of the heralded multipartite entanglement. Remarkably, recent proposals and experimental developments about transferring the single collective excitation of electronic spins to noblegas nuclear spins by spinexchange regime may exceedingly prolong the lifetime of the W state even up to several hours^{33,34}.
The W state with phase information encoded in billions of atoms exists in the form of a spinwave, which resembles tremendous networked quantum sensors with an entanglement between each element. The phase information of spinwave is not only related to the position information of motional atoms, but also sensitive to some other physical parameters related to atomic internal states, like magnetic field^{35}, which makes the W state become a promising candidate for quantum sensing. Interestingly, the W state is also robust for the purpose of metrology, because the remaining particles are still entangled while one particle is traced out. Furthermore, the multipartite entanglement constructed between these quantum sensors may significantly enhance the precision of multiparameter estimation^{36}. Due to the collective enhancement effect in the readout of spinwave, the huge scale may become an advantage in metrology. The nonclassical correlations contained in the W state among huge entangled particles may endow quantum advantages over classical states, though it is still an open question how to sufficiently exploit the distinct features of such a largescale W state.
Methods
Experimental details
Figure 1a shows the energy levels of faroffresonance DLCZ protocol, where the faroffresonance regime means that the detuning is much larger than the Doppler broadening in the warm vapor cell. The large detuning in the generation and verification process can prevent the scattering signal photons from being immersed in the fluorescence noise, which endows a low unconditional noise level and a high crosscorrelation^{26}. The pump and probe pulses are generated by the same distributed Bragg reflector laser(SLAVE laser), whose frequency is locked to an external cavity diode laser(MASTER laser). The MASTER laser provides a fixed frequency reference in the transition of 6S_{1/2}, F = 4 → 6S_{3/2}, F = 4 co 5 lines of cesium atoms. This frequency stabilizing system is not only important for W state’s creating and verifying, but also related to the performance of subsequent filtering systems. In this scheme, the pump and probe light are collinear with horizontal polarization, but the scattering correlated Stokes and antiStokes photons are in the vertical polarization because of the selection rules of the SRS process. To split the collinear Stokes and antiStokes photons, we have built two sets of highperformance FabryPérot cavities to direct different photons into different counting modules, as Fig. 1b shows.
There are several elements that will influence the entanglement depth in the actual experiment, such as the beam waist, the detuning, and the energy of addressing light. For the faroffresonance DLCZ protocol, we have chosen a “sweet point” (Δ = 4 GHz) for the detuning, which has been experimentally demonstrated to have the lowest unconditional noise^{25,26}. As for the beam waist, it is not appropriate to utilize a too large beam waist to provide enough addressing energy density. Note that there is no problem with using a larger beam waist while stronger addressing light is equipped, whose advantage is that there will be more atoms involved in the creation of an entanglement state. In our experiment, we choose a beam waist of 100 μm for providing a sufficient excitation rate. Our generating scheme of the W state is operated in a broadband feature benefiting from the condition of far detuning, so we can use 2 ns pulses to generate and verify the W state. To generate this highspeed light pulse with enough intensity in Fig. 1b, we have developed a system to satisfy the needs of tunable central frequency, broad bandwidth, and more importantly, generation time in a programable fashion. For observing the decoherence effects of the multipartite entanglement, we have chosen the appropriate pulse energy of 225 pJ. The whole time sequence for initializing, generating, and the verifying process can be depicted in Fig. 5. More information about the experimental parameters can be found in Supplementary Note 1.
We can define the heralded probability of the retrieving antiStokes photon as p_{1} in the verifying process, which can be expressed as \({p}_{1}=\frac{{N}_{S\cdot AS1}+{N}_{S\cdot AS2}}{{N}_{S}}\) with N_{S⋅AS1(2)} represents the coincidence of the Stokes photons and the antiStokes photons of two paths. To analyze the higherorder excitations, we using the standard HBT configuration to measure the threefold coincidence, so the \({p}_{2}=\frac{2{N}_{S\cdot AS1\cdot AS2}}{{N}_{S}}\), where factor 2 is derived from the fact that there is a 50% chance for the two photons to go into different paths of the beam splitter. Based on the above definitions, we can construct a simple relationship between p_{1} and p_{2}, i.e. \({p}_{2}=\frac{1}{2}{g}_{A{S}_{1}\cdot A{S}_{2} S}^{(2)}{p}_{1}^{2}\), where the autocorrelation function \({g}_{A{S}_{1}\cdot A{S}_{2} S}^{(2)}=({N}_{s}{N}_{S\cdot AS1\cdot AS2})/({N}_{S\cdot AS1}{N}_{S\cdot AS2})\) with the approximation N_{S⋅AS1} ≈ N_{S⋅AS2}. Actually, it is reasonable for us to remove the loss of detection when we deals with the raw experimental data, because the genuinely verified entanglement depth can’t be varied with the detection loss and the loss can be accurately measured in our experiment. Therefore, the p_{1}, p_{2} can be modified as \({p}_{1}^{\prime}=\frac{{N}_{S\cdot AS1}/{\eta }_{AS1}+{N}_{S\cdot AS2}/{\eta }_{AS2}}{2{N}_{S}}\),\({p}_{2}^{\prime}=\frac{2{N}_{S\cdot AS1\cdot AS2}/{\eta }_{AS1}{\eta }_{AS2}}{{N}_{S}}\) (where the factor 2 in the denominator of \({p}_{1}^{\prime}\) comes from that \({p}_{1}^{\prime}\) is the sum of two antiStokes paths, and the detection losses η_{AS1} = 1.95%, η_{AS2} = 1.85% based on the experimental measurement), which indicates the photons statistics at the moment of having converted the W state into a mappingout photon. The raw experimental data about p_{1} = 0.0123 and p_{2} = 2.49 × 10^{−5} with a memory time of 30 ns are shown in Fig. 2. After subtracting the detection inefficiencies, the genuine p_{1} = 0.32 and p_{2} = 0.017 with a memory time of 30 ns.
Witness for Mseparability
In the certifying process, we apply an optical pulse to interrogate the state of the atomic ensemble, and analyze the correlated photons statistics. We assume that the atomic state can be described in a pure state, which can be decomposed into an Mseparable form like equation \(\left\psi \right\rangle\) in the main text. Due to the imperfect experimental conditions and decoherence, the representation of the state in every independent subgroup may be a superposition of many possible states^{37}. Since we consider at most two excitations in the spontaneous Raman scattering process, the specific form of each state in the subgroup can be expressed in the following state
where \(\left{W}_{1}\right\rangle\), \(\left{W}_{2}\right\rangle\) are the Dicke states in each subgroup; \(\left{W}_{0}\right\rangle\) is a vacuum state. Thus, the whole state of the ensemble is the product of all subgroups:
The two probability p_{1}, p_{2} can be calculated specifically,
where \(D=\frac{N}{M}\) is the entanglement depth. Actually, the entanglement depth should be defined as the largest scale of all the entangled subgroups, but we can take N/M as the representation of entanglement depth to avoid getting a very large subgroup size^{21}.
In order to determine the entanglement depth in the experiment, we need to determine the lower bound for p_{2} with a fixed M value. Obviously, we need the probability p_{2} as low as possible in the actual experiment, which means that the fidelity of the target W_{1} state is high. The bound can be calculated by
Note that the constraint between the coefficients of superposition in eq. (3) is, \({\left{a}_{i}\right}^{2}+{\left{b}_{i}\right}^{2}+{\left{c}_{i}\right}^{2}\le 1\), and we can take the approximation ∣a_{i}∣^{2} + ∣b_{i}∣^{2} + ∣c_{i}∣^{2} = 1 owing to the neglectable higherorder excitations. Utilizing the Lagrange multiplier method deduced in the supplementary notes of^{21}, the conclusion is that the symmetric solution gives the global minimal value of p_{2} for M≥5. This symmetric solution requires that \({a}_{i}=a,{b}_{i}=b,c=\sqrt{1{a}^{2}{b}^{2}}\). In this condition, the optimal values for p_{1}, p_{2} are
The final form of function \({p}_{2}^{bound}({p}_{1},M)\) is,
For fixed p_{1} and M, we need to calculate minimal value of p_{2} with a ∈ (0, 1). The theoretical bound of eq. (10) can be obtained by taking all values of p_{1} ∈ (0, 1), which is shown in Fig. 2a and Fig. 3b with different M values.
Number of atoms involved in the creation of multipartite entanglement
Theoretically, the transmission rate of probe light passing through the atomic ensemble has the following form^{38,39}
where n is the density of atoms, k is the wave vector of the probe light, L is the length of our vapor cell, d is the reduced dipole matrix element, S_{i} is the strength of relative coupling from the hyperfine level F = 3 of the ground state to \(F^{\prime} =2,3,4\) in the excited state^{40}, l_{i}(ω) is the normalization lineshape. For more precisely fitting, the normalization lineshape l_{i}(ω) should be considered as Vigot lineshape^{39}. More details about the theoretical absorption model and experimental fitting are in Supplementary Note 2. According to the fitting coefficients, the number density of atoms in the ^{133}Cs cell is nearly 1.21 × 10^{18} m^{−3}.
To determine how many atoms participating in the interaction, we need to know the volume of the interaction area illuminated by light in the creating and certifying process. Considering the coherence length of the atomic ensemble, the actual volume of the atoms involved in the W state may be delocalized in a subensemble. There is a simple model to analyze the coherence volume of the ensemble according to the collective enhancement effect in the superposition of the transformed antiStokes photon. While the probe light is applied for transforming the W state into an antiStokes photon, many atoms will collectively participate in this process. This collective enhancement of the wave function of the antiStokes photon is derived from the coherent superposition of the whole ensemble, which can be expressed as
where k_{w} is the wave vector of the W state correlated to the pumping process with the form of k_{w} = k_{p} − k_{s}, k_{p} is the wave vector of the pump or probe light, k_{s} and k_{as} are the wave vectors of the Stokes and antiStokes photons respectively, N is the number of involving atoms, \(\left{\phi }_{0}\right\rangle\) is the spherical wave function of the single photon. Therefore, the coherence of the antiStokes photon is related to the mismatch wave vector Δk = 2k_{p} − k_{s} − k_{as}. In the ideal case, the excitation process satisfies the matching condition of wave vectors, i.e. Δk = 0, then the \(\left{{{\Psi }}}_{as}\right\rangle\) is in a state of coherent superposition, which will amplify the excitation probability of the antiStokes photon being proportional to the atomic number N. However, the collections of the Stokes and antiStokes emission are defined with a range of kvectors at slightly different angles, so there may be small mismatching of the wave vectors in the actual experiment, which means that the uncertainty of the wave vectors implies a limit on the scale of the atomic ensemble that can be considered in the coherent superposition. The maximum mismatching angle is determined by the maximum collection angle of the singlemode fiber. Even though the singlemode fiber can define a well spatial mode with a very small divergence angle, the actual mismatching situation still sets the limit of the spatial scale of the ensemble to compose the W state.
Supposing the maximum collection angle defined by the singlemode fiber is θ, we can build a simple analyzing model to determine the spatial bound of the subensemble which contributes to the coherent superposition. Figure 6 shows the maximum mismatching of Δk_{max}, decomposed into two components, i.e. the longitudinal component \({{\Delta }}{k}_{max}^{\parallel }\) and the transverse component \({{\Delta }}{k}_{max}^{\perp }\), which can determine the length l and radius r of the coherent subensemble. For satisfying the coherent superposition of eq. (12), the phase accumulation by the mismatching of wave vectors cannot be much larger than 2π, so the components \({{\Delta }}{k}_{max}^{\parallel }\) and \({{\Delta }}{k}_{max}^{\perp }\) should meet the constraints: \({{\Delta }}{k}_{max}^{\parallel }\cdot l\le 2\pi\), \({{\Delta }}{k}_{max}^{\perp }\cdot 2r\le 2\pi\). Due to that the collection angle θ is extremely small, we can approximately take \({{\Delta }}{k}_{max}^{\parallel }={k}_{p}\cdot {\theta }^{2}\) and \({{\Delta }}{k}_{max}^{\perp }=2{k}_{p}\cdot \theta\), then we can evaluate the coherent bound of the subensemble l ≈ λ/θ^{2}, r ≈ λ/4θ, and the coherent volume can be calculated as V_{coh} = πr^{2} ⋅ l = λ^{3}/(16θ^{4}). The maximum mismatching angle is determined by the maximum collection angle of the singlemode fiber. For our experimental parameters, the maximum collection angle is θ_{SM} = 2.7 × 10^{−3} rad. However, considering the spontaneous Raman scattering process in the generation process, the forward signal mode, correlated with the W state, is distributed inside a small cone, whose spatial structure can be read as^{41}
where \({\eta }_{0}={r}_{0}^{2}/({r}_{0}^{2}+{R}_{0}^{2})\) is the structure factor of the actual atomic ensemble, r_{0}, and R_{0} characterize the radius of the pump beam and the radius of the atomic ensemble, L is the length of the ensemble, k_{0} is the wave vector of the signal photon. Therefore, the forward signal photons characterized by the experimental parameters will be distributed below the angle of \({\theta }_{FS}=\frac{1}{2}min[1/({k}_{0}{R}_{0}\sqrt{{\eta }_{0}}),1/\sqrt{{k}_{0}L}]\) and the actual bound of the angle of phase mismatching should be determined by θ_{bound} = min[θ_{FS}, θ_{SM}] = 6.7 × 10^{−4} rad. According to the maximum angle θ_{bound} of phase mismatching, the maximum size of a coherent atomic ensemble can be calculated as l_{bound} = 3.5 m, r_{bound} = 429.2 µm, while our vapor cell used in the actual experiment has the length of 7.5 cm and the radius of 2.5 cm. So, the real atoms involved in the W state may be delocalized in a subensemble with the bound of transverse radius being 429.2 µm.
In order to make the evaluation of atomic number more reasonable, we have also taken the variations of the Gaussian beam radius into consideration. Along the propagating direction of light, the different beam radiuses of the pumping Gaussian beam at different positions and the transverse radius determined by the uncertainty of wave vector need to be considered while calculating the volume of the coherent subensemble. The condition of maximum phase mismatching delivers the transverse bound of a coherent ensemble, which can not excess r_{bound} = 429.2 µm. In other words, the transverse radius of the interaction area should be limited by r = λ/4θ when the radius of the Gaussian beam excesses r_{bound}. Therefore, we can define the effective coherent volume as
where the beam radius \(W(z)={W}_{w}{(1+{z}^{2}/{z}_{w}^{2})}^{1/2}\) (W_{w} is the beam waist and z_{w} is Rayleigh length of a laser beam), \({{{\Gamma }}}_{0}=\int\nolimits_{0}^{{\theta }_{bound}}{f}_{{{\Omega }}}(\theta )d\theta\) is the normalization factor of the spatial weight function f_{Ω}(θ). Here, we define the effective area of the Gaussian beam by the amplitude decreasing to 1/10 of maximum magnitude. Actually, the witness applied to certify the W state doesn’t require equal amplitude for the superposition of \(\left{W}_{1}\right\rangle\)^{21}, therefore more atoms can be taken into calculations owing to the extension of the Gaussian beam’s intensity. If the Gaussian beam radius is much smaller than the coherent bound r_{bound} in the whole atomic ensemble, the effective coherent volume defined as eq. (14) can be reduced into a simple form \({V}_{eff}=\pi ln10\int\nolimits_{\frac{L}{2}}^{\frac{L}{2}}{W}_{z}^{2}dz\).
In our experiment, the Gaussian beam has the beam waist of W_{w} = 100 × 10^{−6} m, the Rayleigh length of z_{w} = 3.69 × 10^{−2} m, and the length of the cesium cell is L = 75.3 × 10^{−3} m, so the maximum beam radium located in the end of the cesium cell is W(L/2) = 142.8 µm, which is much smaller than the limited transverse radium by the coherence. Therefore, the volume determined by the pumping Gaussian beam can be regarded as a welldefined coherent ensemble inside the whole atomic ensemble. As the main text said, the volume defined by Gaussian beam reads as \({V}_{Gaus}=\pi ln10\int\nolimits_{\frac{L}{2}}^{\frac{L}{2}}{W}_{w}^{2}(1+{z}^{2}/{z}_{w}^{2})dz\), and we can define a localized factor ζ to characterize the locality of the entangled volume, which can be expressed as ζ = V_{Gaus}/V_{coh} = 1.5 × 10^{−15} < <1. Finally, the total number of atoms involved in the creation of entanglement is N = nV = 8.85 × 10^{9}.
Data availability
The data that support the findings of this study are available from the corresponding author on request.
References
Einstein, A., Podolsky, B. & Rosen, N. Can quantummechanical description of physical reality be considered complete?. Phys. Rev. 47, 777 (1935).
Gisin, N. & Thew, R. Quantum communication. Nat. Photonics 1, 165–171 (2007).
Zoller, P. et al. Quantum information processing and communication. Eur. Phys. J. D. 36, 203 (2005).
Lloyd, S. Universal quantum simulators. Science 273, 1073 (1996).
Zhang, J. et al. Observation of a manybody dynamical phase transition with a 53qubit quantum simulator. Nature 551, 601–604 (2017).
Eckert, K. et al. Quantum non demolition detection of strongly correlated systems. Nat. Phys. 4, 50–54 (2008).
Hauke, P. et al. Measuring multipartite entanglement through dynamic susceptibilities. Nat. Phys. 12, 778–782 (2016).
Qi, X.L. Does gravity come from quantum information? Nat. Phys. 14, 984–987 (2018).
Raussendorf, R., Briegel, HansJ. & OneWay, A. Quantum computer. Phys. Rev. Lett. 86, 5188 (2001).
Jeremy, L. OaBrien, optical quantum computing. Science 318, 1567 (2007).
Ladd, T. D. et al. Quantum computers. Nature 464, 45–53 (2010).
Pezzè, L., Smerzi, A., Oberthaler, M. K., Schmied, R. & Treutlein, P. Quantum metrology with nonclassical states of atomic ensembles. Rev. Mod. Phys. 90, 035005 (2018).
Preskill. Quantum computing and the entanglement frontier. Preprint at http://arXiv.org//abs/1203.5813 (2012).
Ming, G. et al. Genuine 12qubit entanglement on a superconducting quantum processor. Phys. Rev. Lett. 122, 110501 (2019).
XiLin, W. et al. 18qubit entanglement with six photons? Three degrees of freedom. Phys. Rev. Lett. 120, 260502 (2018).
Nicolai, F. et al. Observation of entangled states of a fully controlled 20qubit system. Phys. Rev. X 8, 021012 (2018).
XinYu, L. et al. Deterministic entanglement generation from driving through quantum phase transitions. Science 355, 620 (2017).
Haas, F., Volz, J., Gehr, R., Reichel, J. & Estève, J. Entangled states of more than 40 atoms in an optical fiber cavity. Science 344, 180 (2014).
Hosten, O., Engelsen, N. J., Krishnakumar, R. & Kasevich, M. A. Measurement noise 100 times lower than the quantumprojection limit using entangled atoms. Nature 529, 505 (2016).
McConnell, R., Zhang, H., Hu, J., Cuk, S. & Vuletic, V. Entanglement with negative Wigner function of almost 3,000 atoms heralded by one photon. Nature 519, 439 (2015).
Florian, F. et al. Experimental certification of millions of genuinely entangled atoms in a solid. Nat. Commun. 8, 907 (2017).
Zhao, B. et al. A millisecond quantum memory for scalable quantum networks. Nat. Phys. 5, 95–99 (2009).
Manz, S., Fernholz, T., Schmiedmayer, J. & Pan, J.W. Collisional decoherence during writing and reading quantum states. Phys. Rev. A. 75, 040101(R) (2007).
Duan, L.M., Lukin, M. D., Cirac, J. I. & Zoller, P. Long distance quantum communication with atomic ensembles and linear optics. Nature 414, 413–418 (2001).
JianPeng, D. et al. Direct observation of broadband nonclassical states in a roomtemperature lightmatter interface. npj Quantum Inf. 31, 55 (2018).
JianPeng, D. et al. A broadband DLCZ quantum memory in roomtemperature atoms. Comms. Phys. 1, 55 (2018).
Pang, X.L. et al. A hybrid quantum memory enabled network at room temperature. Sci. Adv. 6, eaax1425 (2020).
Chou, C. W. et al. Measurementinduced entanglement for excitation stored in remote atomic ensembles. Nature 438, 828–832 (2005).
Lee, K. C. et al. Entangling macroscopic diamonds at room temperature. Science 334, 1253 (2011).
Chou, C. W., Polyakov, S. V., Kuzmich, A. & Kimble, H. J. Singlephoton generation from stored excitation in an atomic ensemble. Phys. Rev. Lett. 92, 213061 (2004).
Balabas, M. V. et al. High quality antirelaxation coating material for alkali atom vapor cells. Opt. Express 18, 5825 (2010).
Michael, Z. et al. Longlived nonclassical correlations towards quantum communication at room temperature. Comms. Phys. 1, 76 (2018).
Dantan, A. et al. Longlived quantum memory with nuclear atomic spins. Phys. Rev. Lett. 95, 123002 (2005).
Katz, O. et al. Quantum interface for noblegas spins based on spinexchange collisions. Preprint at http://arXiv.org//abs/1905.12532 (2019).
Bing, C. et al. Atomlight hybrid interferometer. Phys. Rev. Lett. 115, 043602 (2015).
Proctor, T. J., Knott, P. A. & Dunningham, J. A. Multiparameter estimation in networked quantum sensors. Phys. Rev. Lett. 120, 080501 (2018).
Yunfei, P. et al. Experimental entanglement of 25 individually accessible atomic quantum interfaces. Sci. Adv. 4, 3931 (2018).
Michael Roger Sprague. Quantum memory in atomic ensembles. Ph.D. thesis, University of Oxford, (2014).
Siddons, P., Adams, C. S., Ge, C. & Hughes, I. G. Absolute absorption on the rubidium D lines: comparison between theory and experiment. J. Phys. B 41, 155004 (2008).
D. A. Steck, Cesium D Line Data (2003), http://steck.us/alkalidata.
Duan, L.M., Cirac, J. I. & Zoller, P. Threedimensional theory for interaction between atomic ensembles and freespace light. Phys. Rev. A. 66, 023818 (2002).
Acknowledgements
The authors thank JianWei Pan for helpful discussions. This research was supported by the National Key R&D Program of China (2019YFA0308700, 2019YFA0706302, 2017YFA0303700), the National Natural Science Foundation of China (61734005, 11761141014, 11690033), the Science and Technology Commission of Shanghai Municipality (STCSM) (20JC1416300, 2019SHZDZX01, 17JC1400403), the Shanghai Municipal Education Commission (SMEC) (201701070002E00049), and China Postdoctoral Science Foundation (2020M671091). X.M.J. acknowledges additional support from a Shanghai talent program and support from Zhiyuan Innovative Research Center of Shanghai Jiao Tong University.
Author information
Authors and Affiliations
Contributions
X.M.J. and J.M.L. conceived the project. H.L., J.P.D. and X.M.J. design of the experiment. H.L., J.P.D., X.L.P., C.N.Z., Z.Q.Y., T.H.Y., J.G., performed the experiment. H.L. and J.P.D. developed the theory. X.M.J. and H.L. analysed the data and wrote the paper.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no Competing Financial or NonFinancial Interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Li, H., Dou, JP., Pang, XL. et al. Multipartite entanglement of billions of motional atoms heralded by single photon. npj Quantum Inf 7, 146 (2021). https://doi.org/10.1038/s41534021004761
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41534021004761
This article is cited by

Entangling motional atoms and an optical loop at ambient condition
npj Quantum Information (2023)