Abstract
We formulate a density matrix to fully describe twophoton state within a thermal light source in the photon orbital angular momentum (OAM) Hilbert space. We prove the separability, i.e., zero entanglement of the thermal twophoton state. Still, we reveal the hidden quantum correlations in terms of geometric measures of discord. By mimicking the original protocol of quantum teleportation, we demonstrate that the nonzero quantum discord can be utilized to transmit a highdimensional OAM state at the singlephoton level. It is found that albeit the low fidelity of teleportation due to the inherent component of maximally mixed state, the information of all parameters that characterize the original state can still be extracted from the teleported one. Besides, we demonstrate that the multiple repetitions of the protocol, enable the transmission of a complexamplitude light field, e.g., an optical image, regardless of being accompanied with a featureless background. We also distinguish our scheme of optical image transmission from that of ghost imaging.
Introduction
In a seminal paper in 1993^{1}, Bennett and coworkers described the quantum protocol of teleportation to transfer any unknown quantum state from one location to another distant one, without physical traveling of the object itself. Its first experimental realization was achieved by Bouwmeester and his colleagues in 1997 using photonic polarization entanglement^{2}. Since then, quantum teleportation has been playing an active role in the progress of both quantum information science and technology^{3,4}. A lot of research efforts have been triggered to demonstrate numerous variants of teleportation schemes, such as entanglement swapping and quantum repeaters^{5}, opendestination teleportation^{6}, quantum gate teleportation and quantum computing^{7,8}, portbased teleportation^{9}, and teleportation of a single photon’s multiple degrees of freedom^{10}. The photons are an optimal choice for carrying information in the form of flying qubits and for showcasing nonlocality at a distance such that longhaul experimental teleportation was generally demonstrated with photons^{11,12}. Recent progress has been made to realize the teleportation from a transmitter on Earth to a receiver on a satellite, towards a global scale^{13}.
In the original scheme of quantum teleportation^{1}, two communicating parties, Alice and Bob, must initially share an entangled pair of quantum particles that serves as the quantum channel. For twolevel quantum systems, e.g., polarization qubits as in the 1997 experiment^{2}, the entangled state generated by parametric downconversion can be chosen as, \(\left {\Psi} \right\rangle _{A,B}^{(  )} = \frac{1}{{\sqrt {\mathrm{2}} }}\left( {\left H \right\rangle _A\left V \right\rangle _B  \left V \right\rangle _A\left H \right\rangle _B} \right)\), where H and V represent the horizontal and vertical polarizations, respectively. In the meantime, Alice is given another third photon C whose polarization state \(\left \psi \right\rangle _C\) will be teleported from Alice to Bob. Alice performs a joint Bell measurement on her photons A and C and obtains one possible outcome of four Bell states, namely, \(\left {\Psi} \right\rangle _{A,C}^{( \pm )} = \frac{1}{{\sqrt {\mathrm{2}} }}\left( {\left H \right\rangle _A\left V \right\rangle _C \pm \left V \right\rangle _A\left H \right\rangle _C} \right)\) and \(\left {\Phi} \right\rangle _{A,C}^{( \pm )} = \frac{1}{{\sqrt {\mathrm{2}} }}\left( {\left H \right\rangle _A\left H \right\rangle _C \pm \left V \right\rangle _A\left V \right\rangle _C} \right)\). Then Alice communicates her measurement outcome to Bob via a classical channel. Accordingly, Bob applies a unitary operation, namely the identity operator I or Pauli operators σ_{x}, σ_{y}, σ_{z} on his photon B and recovers the original state. Quantum teleportation has been already generalized to highdimensional discretevariable^{14,15} or infinitedimensional continuousvariable systems^{16,17}. Besides, it was demonstrated that some mixed states that contain noisy entanglement, rather than a pure entanglement, can also be exploited as quantum channels for teleportation^{18,19}. Moreover, teleportation using only classical channels has been explored and quantified in terms of fidelity bounds^{20}. Here, by mimicking the standard protocol of quantum teleportation, we propose a new scheme of using a pseudothermal light source to realize the transmission of a highdimensional orbital angular momentum (OAM) state and even an optical image with quantum discord. Our scheme is based on the formulation of a density matrix on the basis of OAM eigenstates to fully describe twophoton state, which reveals the inherently highdimensional spatial correlation of nonzero quantum discord. At the singlephoton level, the fidelity of the transferred sate is quantified, and for multiple repetitions of our protocol, the distinction of our scheme for optical image transmission is also discussed from ghost imaging.
Results
Thermal twophoton state in the OAM space
Our theory is based on the adoption of OAM eigenstates^{21,22}, in terms of Laguerre−Gaussian (LG) modes, to formulate the density matrix describing twophoton states within a thermal light source. The quantum theory of the second and higherorder coherence of light was introduced in 1963 by Glauber^{23,24}. Here, we adopt ref. ^{25} to model the thermal radiation as a collection of independent atoms emitting radiation randomly, and its density operator is written as, \(\rho _0 = \mathop {\sum}\nolimits_{\{ n_{\mathbf{k}}\} } {P_{\{ n_{\mathbf{k}}\} }\left {\{ n_{\mathbf{k}}\} } \right\rangle \left\langle {\{ n_{\mathbf{k}}\} } \right}\), where \(P_{\{ n_{\mathbf{k}}\} } = \mathop {\prod}\nolimits_{\mathbf{k}} {P(n_{\mathbf{k}})}\) with P(n_{k}) being the probability for n_{k} photons in the mode k, and the symbol {n_{k}} denotes a set of numbers n_{k1}, n_{k2}, n_{k3},…, etc, of photons excited in very mode. For twophoton case, \(n = \mathop {\sum}\nolimits_{\mathbf{k}} {n_{\mathbf{k}}} = 2\), then we can specify \(\rho _0 = \mathop {\sum}\nolimits_{{\mathbf{k}},{\mathbf{k}}^\prime } {P\left( {\mathbf{k}} \right)P\left( {{\mathbf{k}}^\prime } \right)\left {\mathbf{k}} \right\rangle \left {{\mathbf{k}}^\prime } \right\rangle \left\langle {{\mathbf{k}}^\prime } \right\left\langle {\mathbf{k}} \right}\), where \(\left {\mathbf{k}} \right\rangle = \left {1_{\mathbf{k}}} \right\rangle\) and \(\left {{\mathbf{k}}^\prime } \right\rangle = \left {1_{{\mathbf{k}}^\prime }} \right\rangle\), denoting only one excited photon in each mode \(\left {\mathbf{k}} \right\rangle\) and \(\left {{\mathbf{k}}^\prime } \right\rangle\), respectively. As the LG modes form an orthogonal and complete computation basis^{26}, the light field of any mode \(\left {\mathbf{k}} \right\rangle\) can be expressed in terms of LG modes \(\left {\ell ,p} \right\rangle\), where \(\ell\) and p are the azimuthal and radial mode indices, respectively. Accordingly, we have
which indicates that the chaotic source, e.g., thermal light, can be thought of as incoherent statistical mixtures of photon pairs, \(\left {\ell ,p} \right\rangle \left {\ell ^\prime ,p^\prime } \right\rangle\).
In our proposed scheme sketched in Fig. 1, the light derived from a thermal source is directed to a nonpolarizing beam splitter (BS) and then is divided into two paths. The interaction with BS changes the twophoton state to,
where
Subsequently, we use coincident measurement between photons in two paths, which means that it is just this joint detection that post selects the following twophoton state,
where \(\left {{\Psi}(\ell ,p;\ell ^\prime ,p^\prime )} \right\rangle _{ab} = \left( {\left {\ell ,p} \right\rangle _a\left {\ell ^\prime ,p^\prime } \right\rangle _b + \left {\ell ^\prime ,p^\prime } \right\rangle _a\left {\ell ,p} \right\rangle _b} \right)/\sqrt 2\). It is found that Eq. (4) can be equivalently written as the following density matrix (see Supplementary Information for details),
where
with subscript, a and b denote the photon a and b in two paths, respectively. We show the physically intuitive quantum−mechanical structure of Eq. (5), which describes the twophoton state within a thermal light source: \(\rho ^0_{ab}\) of Eq. (6) is merely a diagonal fully separable state, corresponding to the maximally mixed state. In contrast, \(\left {\Psi} \right\rangle _{ab}\) of Eq. (7) is inherently a pure entangled state, which behaves in a very analogous way to the OAM entanglement^{27}.
To further derive the specific expression of \(P_{\ell ,p}\), we consider the chaotic thermal light source’s crossspectrum density function. Because of its generality and validity, the Gaussian−Schell model was extensively employed, which can be written as^{28},
where G_{0} is a constant, σ_{S} is the transverse size, and σ_{g} is the transverse coherence width of the source. In biphoton case, twophoton amplitude, \({\Phi}({\mathbf{r}}_1{\mathbf{,r}}_2)\), can be expressed as the singlesum expansion based on the Schmidt decomposition, \({\Phi}({\mathbf{r}}_1{\mathbf{,r}}_2) = \mathop {\sum}\nolimits_n {\sqrt {\lambda _n} } u_n({\mathbf{r}}_1)v_n({\mathbf{r}}_2)\), where u_{n}(r_{1}) and v_{n}(r_{2}) are Schmidt modes for signal and idler photons, respectively, and λ_{n} is the eigenvalue^{29}. We perform a similar mode decomposition on W(r_{1}, r_{2}) in the full set of normalized LG modes, i.e., W(r_{1}, r_{2}) = ∑_{ℓ,p,ℓ′,p′}ƒ_{ℓ},_{ℓ′,p,p′} LG_{ℓ,}_{p} (r_{1})LG_{ℓ′,p′}(r_{2}). After some algebra (see Supplementary Information for details), we can find that,
where tan β = 2σ_{S}/σ_{g}. The main incentive of our present work is to explore the thermal twophoton OAM state of Eq. (5) for the possibility of mimicking teleportation to realize the transmission of a highdimensional OAM state and even a twodimensional complexamplitude optical image, regardless of no prior entanglement.
Proof of the separability of thermal twophoton OAM state
Based on the concept of robustness of entanglement, we first prove the separability of twophoton state of Eq. (5). Following the procedure presented in ref. ^{30}, we know the robustness of entanglement of Eq. (7) \(\left( {\rho _Q = \left {\Psi} \right\rangle _{ab}\left\langle {\Psi} \right} \right)\) is \(R = \left( {\mathop {\sum}\nolimits_{\ell ,p} {P_{\ell ,P}} } \right)^2  1\), and the local noise, \(\rho_S^  = \frac{1}{R}\mathop {\sum}\nolimits_{\ell \ne \ell ^\prime {\mathrm{,}}\;p \ne p^\prime } {P_{\ell ,p}P_{\ell ^\prime ,p^\prime }\left {\ell ,p} \right\rangle _A\left\langle {\ell ,p} \right \otimes \left {\ell ^\prime ,p^\prime } \right\rangle _B\left\langle {\ell ^\prime ,p^\prime } \right}\), which is a separable state by construction. In other words, all entanglement in ρ_{Q} can be washed out after being mixed with \(\rho _S^ \), and the required minimal amount of \(\rho _S^ \) is R. Thus the mixture, \(\rho _S^ + = \frac{1}{{1 + R}}\left( {\rho _Q + R\rho _S^  } \right)\), becomes separable. Besides, by considering the fact that \(\rho _S^  = \frac{1}{R}\left( {\hat \rho _{ab}^0  \mathop {\sum}\nolimits_{\ell ,p} {P_{\ell ,p}^2} \left {\ell ,p} \right\rangle _A\left\langle {\ell ,p} \right \otimes \left {\ell ,p} \right\rangle _B\left\langle {\ell ,p} \right} \right)\), we can rewrite Eq. (5) as,
which is obviously separable. Thus we have proved that the thermal twophoton source is not entangled per se. However, our theoretical formulation also provides an intuitive picture to reveal the quantum correlations hidden in thermal twophoton OAM state.
The revelation of the quantumness of thermal twophoton OAM state
When quantum correlation is mentioned, we might be immediately tempted to think of entanglement. But it deserves our attention that entanglement is not the only aspect of quantum correlations, namely, some quantum systems even without entanglement are capable of possessing correlations that cannot be simulated by classical physics^{31}. In the OAM Hilbert space, we have formulated the density operator of Eq. (5) to fully describe the thermal twophoton source. It is crucial to distinguish quantum correlations beyond entanglement. In parallel with the entanglement measure used for entanglement vs separability criteria, discord was proposed as a figure of merit for characterizing the nonclassical resource in the quantum vs classical scenario^{32,33} It is defined as the discrepancy between quantum mutual information (total correlation) and classical correlation in a bipartite system. However, the evaluation of quantum discord requires considerable numerical minimization. From a geometric perspective, a distancebased notion of discord^{34}, was introduced instead. For an arbitrary state ρ, this geometric discord is defined as \(D\left( \rho \right) = \mathop {{\min }}\limits_{\chi \in {\Omega}_0} \left\ {\rho  \chi } \right\^2\), where Ω_{0} denotes the set of zerodiscord states, and \(\left\ {\rho  \chi } \right\^2 = {\mathrm{tr}}\left( {\rho  \chi } \right)^2\) is the square norm in the HilbertSchmidt space. Progress in computing D(ρ) has been made by showing that^{35},
where the minimization is taken over all local bases \(\left {\eta _k^B} \right\rangle\) on photons in path B.
After a lengthy algebra based on Eq. (11), we can obtain the analytic geometric discord for Eq. (5) as (see Supplementary Information),
We have plotted D(ρ) in Fig. 2 in different OAM subspaces, whose dimension is specified by \(d = \left( {2L + 1} \right)\left( {P + 1} \right)\), with \(\ell\) ranging from −L to L and p from 0 to P. One can see that, regardless of zero entanglement, the quantum discord is surprisingly nonzero, and therefore revealing the quantumness of correlations hidden in the thermal twophoton state. We also plot the geometric discord D(ρ_{Q}) for the pure entangled state of Eq. (7) within the same subspaces. Two extreme cases, σ_{g} = ∞ and σ_{g} = 0, deserve our special attention. For σ_{g} = ∞, the source is completely coherent, and none of the quantum correlations can be extracted from either the thermal state or the entangled state, as \(D( \rho ) = D({\rho}_Q) = 0\). For L = P = ∞, we can also obtain, \(D_\infty \left( \rho \right) = 1/\left( {\sigma _{\mathrm{g}}/\sigma _{\mathrm{s}} + 2\sigma _{\mathrm{s}}/\sigma _{\mathrm{g}}} \right)^4\), which is shown by the purple curves in Fig. 2. The other extreme is the case of the completely incoherent light source, where ρ_{Q} is maximally entangled and \(P_{\ell ,p}\) becomes a constant. For comparison, if we also apply the geometric discord to the initial state of Eq. (1), a straightforward analysis finds D(ρ_{0}) = 0. Thus we are allowed to conclude that it is just the function of BS and the postselection of joint detection that extract the nonclassical correlations. As we will demonstrate below, we can still utilize such a nonentangled yet nonclassical thermal twophoton state to realize the transmission of a highdimensional OAM state at the singlephoton level and that of an optical image with multiple repetitions of our protocol.
Mimicking teleportation to transmit the highdimensional OAM state of a single photon
As shown in Fig. 1, we use an additional mirror (M) in the path of photon b to flip the OAM value. For simplicity and without loss of generality, we consider a completely incoherent source with σ_{g} = 0, which leads to \(\tan \frac{\beta }{2} = 1\) and indicates that \(P_{\ell ,p}\) becomes a constant, as it is independent of \(\ell\) or p. If we further consider a subspace with the radial index p ranging from 0 to P and the azimuthal index \(\ell\) from −L to +L, we have \(d = \left( {P + 1} \right)\left( {2L + 1} \right)\) denotes the dimension. Besides, for convenience, we reindex all the eigenstates of LG modes by assigning \(\left j \right\rangle = \left {\ell ,p} \right\rangle\), with \(j = \ell + \left( {2L + 1} \right)p + L\) ranging from j = 0 to d − 1. As a result, we can rewrite Eqs. (6) and (7) as,
Here we explore the above quantum correlations hidden in the thermal twophoton state to serve as a quantum bridge to accomplish the transmission of a highdimensional OAM state for a single photon, akin to the standard teleportation protocol. In Fig. 1, the photons a and b in two paths are spatially separated. Assume Alice holds photon a while Bob possesses photon b. To mimic the teleportation, Alice has a third photon c at hand, which is encoded with the highdimensional OAM state, mathematically corresponding to the complex light field of an optical image, ψ(r). This is realized by passing the photon c in an initial fundamental Gaussian mode through an encoded holographic mask. According to the concept of digital spiral imaging^{26,36}, we can rewrite the complex light field in terms of LG modes as, \(\left \psi \right\rangle _c = \mathop {\sum}\nolimits_j {A_j} \left j \right\rangle _c\), where \(A_j = A_{\ell ,p} = {\int} {\left[ {{\mathrm{LG}}_{\ell, p} ({\mathbf{r}})} \right]^ \ast \psi ({\mathbf{r}}){\mathrm{d}}{\mathbf{r}}}\). In our proposed experiment, we use a series of imaging lenses to image the plane of a chaotic thermal light source to the plane of an ICCD camera (via 4f_{1} for photon b) and to the plane of a highdimensional Bell state measurement device (dBSM) (via 4f_{2} for photon a), respectively. Meanwhile, another single imaging lens (4f_{3}) is used to image the mask onto the dBSM as well in the path of photon c. Then the wavefunction of the whole system can be written as, \(\rho _{abc} = \left( { \rho _{ab}^0 + \left {\Psi} \right\rangle _{ab}\left\langle {\Psi} \right} \right)\left( {\left \psi \right\rangle _c\left\langle \psi \right} \right)\). Here the dBSM realizes the function of projecting photons a and c onto the generalized highdimensional Bell states in the Hilbert space spanned by the LG modes. Then, we are allowed to rewrite the whole state \(\rho_{abc}\), in terms of the generalized Bell states of photons a and c, as
where \(\left {\psi ^{m,n}} \right\rangle _{ac} = \frac{1}{{\sqrt d }}\mathop {\sum}\nolimits_j {{\mathrm{e}}^{{\mathrm{i}}2\pi jm/d}} \left j \right\rangle _a\left {\left( {j + n} \right){\mathrm{mode}} \ {d}} \right\rangle _c\) are twophoton generalized Bell states, and \(\left \varphi \right\rangle _b = U_{mn}^ + \left \psi \right\rangle _b\) with \(U_{mn} = \mathop {\sum}\nolimits_k {{\mathrm{e}}^{{\mathrm{i}}2\pi km/d}} \left k \right\rangle \left\langle {\left( {k + n} \right){\mathrm{mode}} \ {d}} \right\) being a unitary matrix. Equation (15) forms the key result of the present work, which suggests the possibility of using a thermal light source to mimic the teleportation protocol for realizing the transmission of a highdimensional OAM state of a single photon and even a complexamplitude optical image. After performing the generalized Bell state measurement on photons a and c, Alice gets one possible result out of d^{2} outcomes, k = (m,n), where \(m,n = 0,1,2,...,d  1\). Then Alice communicates classically her measurement result k = (mn) with Bob. Accordingly, Bob applies the conditional unitary matrix U_{mn} to the photon b, namely
This suggests that Bob can retrieve an output state of photon b, which is just a mixture of an exact replica of the original state of a photon a, \(\left \psi \right\rangle _a\), and a maximally mixed state, \(\frac{1}{d}\mathop {\sum}\nolimits_j {\left j \right\rangle } \left\langle j \right\).
Discussion
Our protocol at the singlephoton level
We assume that the state encoded by photon c is represented by a highdimensional vector, \(\psi \rangle_c = \mathop {\sum}\nolimits_j A_j \left j \right\rangle_c = \mathop {\sum}\nolimits_{\ell,p} A_{\ell,p} \left\ell,p\right\rangle_c\), which can be equivalently characterized by two spectra, \(a_{\ell ,p} = \left {A_{\ell ,p}} \right\) and \(\phi _{\ell ,p} = \arg (A_{\ell ,p})\), respectively. It is noted that in classical optics, such a highdimensional state just corresponds to a complexamplitude optical image, namely, \(\psi \left( {\mathbf{r}} \right) = \mathop {\sum}\nolimits_{\ell ,p} {A_{\ell ,p}} {\mathrm{LG}}_{\ell ,p}\left( {\mathbf{r}} \right)\). The calculation of \(a_{\ell ,p}\) and \(\phi _{\ell ,p}\) for a Clover image is shown in the section of “Materials and methods”. From Eq. (15), without loss of generality, assume the dBSM results at Alice’s side are \(k = (m,n) = (0,0)\), (2,0) and (10,10), namely, \(\left {\psi ^{0,0}} \right\rangle _{ac}\), \(\left {\psi ^{2,0}} \right\rangle _{ac}\), and \(\left {\psi ^{10,10}} \right\rangle _{ac}\), respectively. Accordingly, photon b will be projected onto the state of \(U_{0,0}^ + \left \psi \right\rangle _b\), \(U_{2,0}^ + \left \psi \right\rangle _b\), and \(U_{10,10}^ + \left \psi \right\rangle _b\). For an easy visualization, we illustrate in Fig. 3 the spectra \(a_{\ell ,p}\) and \(\phi _{\ell ,p}\) of these states, in which the trivial background of a maximally mixed state \(\frac{1}{d}\mathop {\sum}\nolimits_j {\left j \right\rangle _b\left\langle j \right}\) is omitted. In Fig. 3a, b, as \(U_{00}^ + = I\), photon b just acquires the original state of photon c, accompanying with the trivial background. In contrast, conditional on \(\left {\psi ^{2,0}} \right\rangle _{ac}\) and \(\left {\psi ^{10,10}} \right\rangle _{ac}\) and without performing the desired unitary operations, we can only obtain the incorrect highdimensional state, \(U_{2,0}^ + \left \psi \right\rangle _b\), and \(U_{10,10}^ + \left \psi \right\rangle _b\), respectively. According to \(U_{mn} = \mathop {\sum}\nolimits_k {{\mathrm{e}}^{{\mathrm{i}}2\pi km/d}} \left k \right\rangle \left\langle {(k + n) \ {\mathrm{mode}} \ {d}} \right\), we have \(U_{mn}^ + = \mathop {\sum}\nolimits_k {{\mathrm{e}}^{  {\mathrm{i}}2\pi km/d}} \left {(k + n) \ {\mathrm{mode}} \ {d}} \right\rangle \left\langle k \right\). Therefore, we have \(U_{2,0}^ + = \mathop {\sum}\nolimits_k {{\mathrm{e}}^{  {\mathrm{i}}4\pi k/d}} \left k \right\rangle \left\langle k \right\) such that the spectrum of \(a_{\ell ,p}\) remains unchanged while each phase \(\phi _{\ell ,p}\) is individually shifted by an amount of \({\Delta}\phi =  4\pi k/d\), see Fig. 3c, d. While for \(U_{10,10}^ + = \mathop {\sum}\nolimits_k {{\mathrm{e}}^{  {\mathrm{i}}20\pi k/d}} \left {k + 10} \right\rangle \left\langle k \right\), as shown by Fig. 3e, f, we can see that both spectra of \(a_{\ell ,p}\) and \(\phi _{\ell ,p}\) are changed. In other words, the two latter cases without desired unitary operations can only lead to the incorrect transferred states.
As well known, quantum teleportation is distinct in that it allows a single copy of a state to be teleported with unity fidelity. Therefore, it is crucial to investigate the fidelity of the transferred state in our protocol, which can be written as,
The particular case of d = 2 leads to F = 3/4, which agrees to the optimal fidelity achievable with a two spin1/2 Werner stat^{37}. However, high dimensions are essential for encoding an optical image. From Eq. (17), we find that the fidelity decreases inversely as the dimensionality increases, approaching to the limit of F = 1/2 for \(d \to \infty\), due to the inherent existence of the background maximally mixed state of \(\frac{1}{d}\mathop {\sum}\nolimits_j {\left j \right\rangle } \left\langle j \right\) in Eq. (16). In an actual experiment, the background can only be subtracted by sending multiple copies of the same state, which changes the resources used in the scheme since Alice performing the Bellstate measurement needs to communicate d bits to Bob for each repetition of the protocol, corresponding to one of the d possible outcomes of the Bellstate measurement. This classical communication itself may exceed the number of bits used to describe the image. However, if the density matrix of Eq. (16) could be reconstructed, e.g., via quantum state tomography, one can see that the information about all the parameters of the original state, \(a_{\ell ,p}\) and \(\phi _{\ell ,p}\), could still be fully known, Albeit a low fidelity of teleportation.
Our protocol with multiple repetitions
Reconstruction of an optical image requires multiple singlephoton events. Thus it is also interesting to consider the multiple repetitions of our protocol. As mentioned above, the highdimensional state just corresponds to a complexamplitude optical image of a Clover, see the inset of Fig. 1. If we postselect the generalized dBSM result of \(\left {\psi ^{0,0}} \right\rangle _{ac}\), we can record the single photons to reconstruct the Clover image with the ICCD camera. Then we can plot the intensity and phase profiles of the Clover image in Fig. 4a, b, respectively, which corresponds to the highdimensional state characterized by Fig. 3a, b. Note that the background is trivially omitted. It can be seen that the reconstruction of the pure Clover image is ensured by the second term of Eqs. (15) or (16), which when viewed separately is due to the spatial correlations established by the entangled state of Eq. (7) or Eq. (14). This is just physically equivalent to using a quantum source of biphoton pure entangled state. In contrast, when the ICCD camera is triggered by another dBSM results, e.g, \(\left {\psi ^{2,0}} \right\rangle _{ac}\) and \(\left {\psi ^{10,10}} \right\rangle _{ac}\), we could only record a fuzzy image, see Fig. 4c–f, respectively, as a result of the incorrect transmissions of the highdimensional states, see Fig. 3c–f, respectively. In principle, one should perform a desired unitary operations for each repetition in order to recover the correct image.
Comparison with ghost imaging
As it is known, another wellestablished application of thermal light source, is the technique of ghost imaging. The ghost imaging could be traced back to the experiment conducted in 1995 that exploited the nonlocal correlations of entangled photon pairs created by parametric downconversion^{38}. However, it was demonstrated that a coincidence image can also be obtained with thermal light source^{39,40,41,42}, thus raising a question whether entanglement was truly necessary for ghost imaging^{43,44,45}. Several efforts have contributed to providing some insightful understanding on this historic argument^{46,47,48}. Here our scheme uses thermal light, and the recorded images in both two cases are embedded in a featureless background. However, we distinguish our scheme of image teleportation from that of ghost imaging. In ghost imaging, only a pair of photons are employed; one photon in the pair illuminates the object and is collected by a bucket detector while the other photon is utilized to record the images with a spatially resolved detector, e.g., an intensive CCD camera. In contrast, here we employ three photons; two photons (a and b) in a pair derived from the thermal light source serve as the channel for teleportation, while the third photon (c) is directed to illuminate the object and acquires the image information to be teleported. We then follow the standard procedure of teleportation to perform the generalized Bell state measurements (dBSM) on photons a and c, and use photon b to record the teleported images with the ICCD camera nonlocally. Besides, it is shown that the correct image could be obtained only after the desired unitary transformation, conditional on the dBSM results of photons a and c.
Besides, the formulation of thermal twophoton OAM state, \(\rho=\rho_Q+\rho ^0_{ab}\), offers an intuitive picture of twophoton quantum interference. If we look ρ_{Q} separately, it just accounts for the existence twophoton OAM interference. In contrast, if we look ρ wholly, the interference of ρ_{Q} has been already canceled out by \(\rho^0_{ab}\), thus leading to the inexistence of twophoton interference. In this regard, our work can also shed new light on the historical debate as to whether the origin of ghost imaging is quantummechanical or classical.
In summary, we have formulated a density matrix on the basis of OAM eigenmodes to fully describe the thermal twophoton state and reveal the quantumness with nonzero discord. We also demonstrate that albeit a low teleportation fidelity, such a thermal twophoton state can be explored for realizing the transmission of a highdimensional OAM state at the singlephoton level and that of an optical image with multiple repetitions of our protocol. This is obviously distinguished from the traditional method of using a prior entangled source. However, as in the case of using entanglement, the practical implementation of our scheme is hindered by the highdimensional Bell state measurement (dBSM). For our demonstration, it requires the manipulation of twophoton dlevel states encoded by LG modes with both azimuthal and radial indices, which remains more challenging with the current experimental technique. In the future, our work can also call for further studies of using thermal multiphoton state to demonstrate some new quantum information tasks, such as remote state preparation^{49} and novel imaging with undetected photons^{50}.
Materials and methods
The calculation of both intensity and phase spiral spectra for the Clover image
In Fig. 1, the Clover object together with the singlemode fiber and the singlephoton detector (not shown) inversely defines the state of photon c, \(\left \psi \right\rangle _c = \mathop {\sum}\nolimits_{\ell ,p} {A_{\ell ,p}\left {\ell ,p} \right\rangle _c}\), where \(A_{\ell ,p} = {\int} {\left( {{\mathrm{LG}}_{\ell, p} ({\mathbf{r}})} \right)^ \ast \psi ({\mathbf{r}}){\mathrm{d}}{\mathbf{r}}}\) denotes the overlap probability amplitude and is related to digital spiral imaging of the Clover object. Accordingly, we present in Fig. 3a, b the intensity spectrum, \(a_{\ell ,p}\), and the phase spectrum, \(\phi _{\ell ,p}\). Due to the threefold rotational symmetry of the Clover considered here, one can see that only OAM components of \(\ell = 0,\; \pm {\mathrm{3,}}\; \pm {\mathrm{6}}\), dominate in the spiral spectrum. Besides, as the adopted beam waist of the LG modes w = 0.4469 mm is comparable to the radius of the Clover object R = 1 mm, only those lowerorder LG modes with a small p index contribute together to form the Clover object.
References
 1.
Bennett, C. H. et al. Teleporting an unknown quantum state via dual classical and Einstein−Podolsky−Rosen channels. Phys. Rev. Lett. 70, 1895–1899 (1993).
 2.
Bouwmeester, D. et al. Experimental quantum teleportation. Nature 390, 575–579 (1997).
 3.
Pan, J. W. et al. Multiphoton entanglement and interferometry. Rev. Mod. Phys. 84, 777–838 (2012).
 4.
Pirandola, S. et al. Advances in quantum teleportation. Nat. Photonics 9, 641–652 (2015).
 5.
Żukowski, M. et al. “Eventreadydetectors” Bell experiment via entanglement swapping. Phys. Rev. Lett. 71, 4287–4290 (1993).
 6.
Zhao, Z. et al. Experimental demonstration of fivephoton entanglement and opendestination teleportation. Nature 430, 54–58 (2004).
 7.
Gottesman, D. & Chuang, I. L. Demonstrating the viability of universal quantum computation using teleportation and singlequbit operations. Nature 402, 390–393 (1999).
 8.
Huang, Y. F. et al. Experimental teleportation of a quantum controlledNOT gate. Phys. Rev. Lett. 93, 240501 (2004).
 9.
Ishizaka, S. & Hiroshima, T. Asymptotic teleportation scheme as a universal programmable quantum processor. Phys. Rev. Lett. 101, 240501 (2008).
 10.
Wang, X. L. et al. Quantum teleportation of multiple degrees of freedom of a single photon. Nature 518, 516–519 (2015).
 11.
Gisin, N. & Thew, R. Quantum communication. Nat. Photonics 1, 165–171 (2007).
 12.
Northup, T. E. & Blatt, R. Quantum information transfer using photons. Nat. Photonics 8, 356–363 (2014).
 13.
Ren, J. G. et al. Groundtosatellite quantum teleportation. Nature 549, 70–73 (2017).
 14.
Werner, R. F. All teleportation and dense coding schemes. J. Phys. A: Math. Gen. 34, 7081–7094 (2001).
 15.
Goyal, S. K. et al. Quditteleportation for photons with linear optics. Sci. Rep. 4, 4543 (2014).
 16.
Braunstein, S. L. & Kimble, H. J. Teleportation of continuous quantum variables. Phys. Rev. Lett. 80, 869–872 (1998).
 17.
Furusawa, A. et al. Unconditional quantum teleportation. Science 282, 706–709 (1998).
 18.
Horodecki, M., Horodecki, P. & Horodecki, R. General teleportation channel, singlet fraction, and quasidistillation. Phys. Rev. A 60, 1888–1898 (1999).
 19.
Verstraete, F. & Verschelde, H. Optimal teleportation with a mixed state of two qubits. Phys. Rev. Lett. 90, 097901 (2003).
 20.
Henderson, L., Hardy, L. & Vedral, V. Twostate teleportation. Phys. Rev. A 61, 062306 (2000).
 21.
Padgett, M., Courtial, J. & Allen, L. Light’s orbital angular momentum. Phys. Today 57, 35–40 (2004).
 22.
MolinaTerriza, G., Torres, J. P. & Torner, L. Twisted photons. Nat. Phys. 3, 305–310 (2007).
 23.
Glauber, R. J. Photon correlations. Phys. Rev. Lett. 10, 84–86 (1963).
 24.
Glauber, R. J. Coherent and incoherent states of the radiation field. Phys. Rev. 131, 2766–2788 (1963).
 25.
Loudon, R. The Quantum Theory of Light 3rd edn, (Oxford University Press, 2000).
 26.
Torner, L., Torres, J. P. & Carrasco, S. Digital spiral imaging. Opt. Express 13, 873–881 (2005).
 27.
Mair, A. et al. Entanglement of the orbital angular momentum states of photons. Nature 412, 313–316 (2001).
 28.
Foley, J. T. & Zubairy, M. S. The directionality of Gaussian Schellmodel beams. Opt. Commun. 26, 297–300 (1978).
 29.
Law, C. K. & Eberly, J. H. Analysis and interpretation of high transverse entanglement in optical parametric down conversion. Phys. Rev. Lett. 92, 127903 (2004).
 30.
Vidal, G. & Tarrach, R. Robustness of entanglement. Phys. Rev. A 59, 141–155 (1999).
 31.
Modi, K. et al. The classicalquantum boundary for correlations: discord and related measures. Rev. Mod. Phys. 84, 1655–1707 (2012).
 32.
Ollivier, H. & Zurek, W. H. Quantum discord: a measure of the quantumness of correlations. Phys. Rev. Lett. 88, 017901 (2001).
 33.
Henderson, L. & Vedral, V. Classical, quantum and total correlations. J. Phys. A: Math. Gen. 34, 6899–6905 (2001).
 34.
Dakić, B., Vedral, V. & Brukner, Č. Necessary and sufficient condition for nonzero quantum discord. Phys. Rev. Lett. 105, 190502 (2010).
 35.
Luo, S. L. & Fu, S. S. Geometric measure of quantum discord. Phys. Rev. A 82, 034302 (2010).
 36.
Chen, L. X., Lei, J. J. & Romero, J. Quantum digital spiral imaging. Light.: Sci. Appl. 3, e153 (2014).
 37.
Popescu, S. Bell’s inequalities versus teleportation: what is nonlocality? Phys. Rev. Lett. 72, 797–799 (1994).
 38.
Pittman, T. B. et al. Optical imaging by means of twophoton quantum entanglement. Phys. Rev. A 52, R3429–R3432 (1995).
 39.
Cheng, J. & Han, S. S. Incoherent coincidence imaging and its applicability in Xray diffraction. Phys. Rev. Lett. 92, 093903 (2004).
 40.
Gatti, A. et al. Ghost imaging with thermal light: comparing entanglement and classical correlation. Phys. Rev. Lett. 93, 093602 (2004).
 41.
Valencia, A. et al. Twophoton imaging with thermal light. Phys. Rev. Lett. 94, 063601 (2005).
 42.
Ferri, F. et al. Highresolution ghost image and ghost diffraction experiments with thermal light. Phys. Rev. Lett. 94, 183602 (2005).
 43.
Scarcelli, G., Berardi, V. & Shih, Y. Can twophoton correlation of chaotic light be considered as correlation of intensity fluctuations? Phys. Rev. Lett. 96, 063602 (2006).
 44.
Gatti, A. et al. Comment on “can twophoton correlation of chaotic light be considered as correlation of intensity fluctuations?”. Phys. Rev. Lett. 98, 039301 (2007).
 45.
Scarcelli, G., Berardi, V. & Shih, Y. H. Scarcelli, Berardi, and Shih reply. Phys. Rev. Lett. 98, 039302 (2007).
 46.
Zhang, M. H. et al. Subwavelength Fouriertransform imaging of a purephase object with thermal light. Phys. Lett. A 366, 569–574 (2007).
 47.
Meyers, R., Deacon, K. S. & Shih, Y. Ghostimaging experiment by measuring reflected photons. Phys. Rev. A 77, 041801 (2008).
 48.
Erkmen, B. I. & Shapiro, J. H. Unified theory of ghost imaging with Gaussianstate light. Phys. Rev. A 77, 043809 (2008).
 49.
Dakić, B. et al. Quantum discord as resource for remote state preparation. Nat. Phys. 8, 666–670 (2012).
 50.
Lemos, G. B. et al. Quantum imaging with undetected photons. Nature 512, 409–412 (2014).
Acknowledgements
We thank Professor Jeffrey Shapiro at MIT and Robert Boyd at the University of Ottawa for some useful discussions. This work is supported by the National Natural Science Foundation of China (NSFC) (12034016, 61975169), the Fundamental Research Funds for the Central Universities at XMU (20720190057, 20720200074), and the program for New Century Excellent Talents in University of China (NCET130495).
Author information
Affiliations
Contributions
L. C. conceived the idea, performed the simulation, and wrote the paper.
Corresponding author
Ethics declarations
Competing interests
The author declares no competing interests.
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
Chen, L. Quantum discord of thermal twophoton orbital angular momentum state: mimicking teleportation to transmit an image. Light Sci Appl 10, 148 (2021). https://doi.org/10.1038/s41377021005858
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41377021005858
Further reading

Special issue on the 100th anniversary of Xiamen University
Light: Science & Applications (2021)