Abstract
The efficient transport and engineering of photonic orbital angular momentum (OAM) lie at the heart of various related classical and quantum applications. Here, by leveraging the spatialmodeengineered frequency conversion, we realize the remote transport of highdimensional orbital angular momentum (OAM) states between two distant parties without direct transmission of information carriers. We exploit perfect vortices for preparing highdimensional yet maximal O AM entanglement. Based on nonlinear sumfrequency generation working with a strong coherent wave packet and a single photon, we conduct the Belllike state measurements for highdimensional perfect vortices. We experimentally achieve an average transport fidelity 0.879 ± 0.048 and 0.796 ± 0.066 for a complete set of 3dimensional and 5dimensional OAM mutually unbiased bases, respectively. Furthermore, by exploring the full transverse entanglement, we construct another strategy of quantum imaging with interactionfree light. It is expected that, with the future advances in nonlinear frequency conversion, our scheme will pave the way for realizing truly secure highdimensional quantum teleportation in the upcoming quantum network.
Similar content being viewed by others
Introduction
Since the seminal discovery by Allen et al.^{1}, the highdimensional nature of light’s orbital angular momentum (OAM) has attracted significant attention due to its potential in some classical and quantum information protocols^{2,3,4,5}, e.g., realizing largealphabet optical communication^{6,7} and constructing the highdimensional photonic quantum entanglement^{8}, thus, providing the toolkit for the investigation of quantum fundamental questions^{9}. However, these benefits are only conceivable when we are able to efficiently generate, manipulate, transport, and detect the OAM state. Hitherto, extensive investigations of OAM transport^{6,7,10,11,12,13,14,15,16,17} have been reported, such as OAM transport in freespace, specialized optical fiber, and even the distribution of OAM entanglement. Whereas, we note that most of these previous schemes focused only on the direct information transport.
Recent years have also witnessed a growing interest in the nonlinear frequency conversion of structured light^{5,18}. In spatial domain, it was demonstrated that using OAM light as fundamental frequency light source, one can realize the image processing and mode detection during the process of upconversion imaging^{19,20}, and by shaping the spatial shape of the pump source, one can enhance the performance of highdimensional OAM frequency interface^{21,22,23,24}. In timefrequency domain, by using the dispersionengineered sumfrequency generation, the quantum pulse gate for harnessing and selecting the timefrequency mode has been realized^{25,26}. Also, sumfrequency generation in conjunction with pulseshaping techniques provided a powerful tool for timebin entanglement measurement^{27}. Particularly, we note that there have been several attempts, by means of sumfrequency generation, to perform the coherent transfer of the polarization state between two distant parties without direct transmission of the information^{28}, and to realize the entanglement swapping of timebin entanglement^{29,30}, however, the scenario of highdimensional OAM state has not yet been fully explored. Some interesting questions arise naturally: whether the nonlinear frequency conversion can be used to transport the highdimensional OAM state in an unprecedented way, and what features this might lead to. These also form the major incentive of our present work.
Here, by leveraging the spatialmodeengineered frequency conversion, we realize the remote transport of highdimensional orbital angular momentum (OAM) states at a distance without direct transmission of information carriers. Specifically, by exploiting perfect vortices, we prepare highdimensional yet maximally entangled orbital angular momentum (OAM) states as communication channel. Then, we employ sumfrequency generation (SFG) with a strong coherent state, which is used to bear the highdimensional OAM state, to enhance the frequency conversion, and thus performing the perfectvortexbased highdimensional Belllike state measurement reliably. Furthermore, we report the quantum imaging with truly interactionfree light.
Results
Remote transport of highdimensional OAM states: model and implementation
Twisted photons carrying OAM possesses an inherent capacity for constructing a highdimensional Hilbert space, thus making it a promising candidate for highdimensional quantum information processing. It was Allen and coworkers^{1} that recognized that the LaguerreGaussian (LG) modes of helical phase \(\exp (i\ell \phi )\) carries a welldefined OAM of \(\ell \hslash\). Generally, twophoton OAMentangled state generated via spontaneous parametric downconversion (SPDC) with a fundamental Gaussian pump \(L{G}_{0}^{0}(r,\phi )\) can be written as, \(\varPsi \rangle={\sum }_{\ell }{C}_{\ell,\ell }{\ell \rangle }_{s}{\ell \rangle }_{i}\), where \({C}_{\ell,\ell }\) denotes the probability amplitude of finding the signal photon in the standard LG mode \({{{{{{\rm{LG}}}}}}}_{p=0}^{\ell }\) (often abbreviated to \({\ell \rangle }_{s}\)) and the idler one in \({{{{{{\rm{LG}}}}}}}_{p=0}^{\ell }\) (\({\ell \rangle }_{i}\)). Specifically, we have \({C}_{\ell,\ell }=\int L{G}_{0}^{0}(r,\phi ){[L{G}_{0}^{\ell }(r,\phi )]}^{* }{[L{G}_{0}^{\ell }(r,\phi )]}^{* }\). As for standard LG modes, the radius of the doughnutlike patterns scale with \(\sqrt{\ell }\), a calculation finds that \({C}_{\ell,\ell }\) will decrease as \(\ell \) increases, resulting in a nonmaximally entangled OAM state^{31}, \(\varPsi \rangle\). However, the maximal entanglement is prerequisite for our remote transport of highdimensional OAM states. The Procrustean method of entanglement concertation was proposed to enhance the entanglement. Here, we employ the perfect vortices^{32} for the same purpose. We modify the LG modes as, \({{{{{{\rm{MLG}}}}}}}_{0}^{\ell }(r,\phi )={{{{{{\rm{LG}}}}}}}_{0}^{l}(r,\phi )\exp (i\ell \phi )\) (with \(l\) being a constant, e.g., \(l=5\)), to make different OAM modes yet bear the same radial intensity distribution, i.e., perfect vortices (Methods). Then, according to \({C}_{\ell,\ell }=\int L{G}_{0}^{0}(r,\phi ){[{{{{{{\rm{MLG}}}}}}}_{0}^{\ell }(r,\phi )]}^{* }{[{{{{{{\rm{MLG}}}}}}}_{0}^{\ell }(r,\phi )]}^{* }\), we know that \({C}_{\ell,\ell }\) can naturally be equalized because of the identical overlap probability regardless of different \(\ell\). Besides, it is noted that these modified modes still form an orthogonal and complete basis. In other words, our adoption of perfect vortices enables the additional exploration of radial correlation and ensures the identical twophoton mode amplitudes, thus maximizing the highdimensional OAM entanglement. Without loss of generality, we first consider the transport of 3dimensional OAM states. Assume Alice wishes to transport to Bob the OAM state, \({\varphi \rangle }_{a}={\alpha }_{1}{1\rangle }_{a}+{\alpha }_{2}{0\rangle }_{a}+{\alpha }_{3}{1\rangle }_{a}\), which is carried by a strong coherent light a, and the complex coefficients \({\alpha }_{1}\), \({\alpha }_{2}\), and \({\alpha }_{3}\) fulfill \({{\alpha }_{1}}^{2}+{{\alpha }_{2}}^{2}+{{\alpha }_{3}}^{2}=1\). For transporting such arbitrary 3dimensional states, Alice and Bob need to share the aforementioned highdimensional and maximally entangled state (photons b and c),
Then, Alice mixes the coherent light a and photon b in the nonlinear crystal to conduct the sumfrequency generation (SFG). Owing to the OAM conservation in SFG^{33}, if we detect the SFG photon with a singlemode fiber (SMF), which can only support fundamental mode, and thus, the topological charges of coherent light a and photon b satisfy the relation, \({\ell }_{a}+{\ell }_{b}=0\). Also, owing to the OAM conservation in SPDC, seen in Eq. (1), the coincidence between SFG photon and photon c can only be detected when \({\ell }_{a}={\ell }_{b}={\ell }_{c}\). Accordingly, the OAM state of the coherent light a could be transported to photon c without direct transmission of information carriers. However, due to the nonuniform frequency conversion efficiency for different input OAM eigenstates, e.g., the LG modes, the transport performance will get worse with the increase of OAM value. Here, we adopt the perfect vortices to overcome this obstacle, as they can guarantee the identical conversion efficiency for different pairwise OAM modes \({\ell \rangle }_{a}{\ell \rangle }_{b}\). Thus, the highdimensional OAM states can be transported reliably (Methods).
In other words, after nonlinear projection via SMF, the OAM states of coherent light a and converted photon b satisfy\({\psi \rangle }_{ab}=\frac{1}{\sqrt{3}}({1,1\rangle }_{ab}+{0,0\rangle }_{ab}+{1,1\rangle }_{ab})\). It’s noted that, due to the coherent light a adopted here, \({\psi \rangle }_{ab}\) is just used to characterize the dependence relationship between a and b, and does not represent a regular entanglement state. In other words, it can be seen as a hybrid classical and quantum system. Interestingly, the relationship \({\psi \rangle }_{ab}\) is formally similar with one of the 3dimensional Bell states, \({\psi }_{mn}\rangle=\frac{1}{\sqrt{3}}\mathop{\sum }\nolimits_{\ell=1}^{1}\exp (i2n\ell \pi /3)\ell \rangle \ell \oplus m\rangle\) with \(m,n\in \{0,1,2\}\). Thus here, we termed it as the highdimensional Belllike measurement. It is noted that the nonlinear SFG process has been employed to measure four complete Bell states for teleportation of a polarization state^{28} and for timeenergy entanglement swapping^{29,30}. Inspiringly, if we perform full nine 3dimensional Belllike measurements by selecting the welldesigned OAM modes of SFG photons, we can transport highdimensional OAM state of coherent light a to Bob and implement a corresponding unitary transform simultaneously, that is,
with \(\omega=\exp (i2\pi /3)\).
By a careful analysis on the OAM conversion in SFG process, we find that the desired 9 Belllike states can be equivalently identified. First, Alice can mix coherent light a and photon b in the nonlinear crystal. Owing to OAM conservation in SFG^{33}, we can map the following 7 OAM Belllike states onto 7 singlephoton OAM superposition states, respectively, that is,
Of interest is that, for \({{\psi }_{00}\rangle }_{ab}\), the resultant SFG OAM state is orthogonal to the rest of Belllike states. While for \({{\psi }_{01}\rangle }_{ab}\) and \({{\psi }_{02}\rangle }_{ab}\), the SFG for coherent light a and photon b is forbidden, as a result of the destructive interference. Still, we can identify \({{\psi }_{01}\rangle }_{ab}\) and \({{\psi }_{02}\rangle }_{ab}\) by performing a prior OAMdependent phase shifts on coherent light a. This can be done simply by inserting into coherent light a’s path one Dove prism (DP)^{34}, whose orientation is adjusted at an angle of \(\pi /3\) and \(\pi /3\), respectively, to convert \({{\psi }_{01}\rangle }_{ab}\) and \({{\psi }_{02}\rangle }_{ab}\) to \({{\psi }_{00}\rangle }_{ab}\). By projecting the SFG photons onto \(2\rangle+1\rangle \,{{{{{\rm{and}}}}}}\,2\rangle+1\rangle,\) we can also identify \({{\psi }_{10}\rangle }_{ab}\,{{{{{\rm{ and }}}}}}\,{{\psi }_{20}\rangle }_{ab}\), respectively. Similarly, by performing a prior OAMdependent phase shifts on coherent light a, we can also convert \({{\psi }_{11}\rangle }_{ab}\) and \({{\psi }_{12}\rangle }_{ab}\) to \({{\psi }_{10}\rangle }_{ab}\), \({{\psi }_{21}\rangle }_{ab}\) and \({{\psi }_{22}\rangle }_{ab}\) to \({{\psi }_{20}\rangle }_{ab}\), respectively. Thus all 9 Belllike states can be identified in theory. In our proofofprinciple experimental demonstration, without loss of generality, we have identified three of all the highdimensional OAM Belllike states, respectively. In other words, the transport of highdimensional OAM states could be realized by the perfectvortexencoded and SFGbased highdimensional Belllike state measurement (HDBLSM) in the OAM subspaces. Whereas, limited by the low efficiency of SFG at the singlephoton level^{28,29,30}, if we use the single photon to replace the coherent light a as input, we need to wait a much longer time for information transport. It is expected that, with the future advances in nonlinear frequency conversion, our scheme could be used to perform the truly highdimensional Bell state measurement, and thus paves the way for realizing the secure highdimensional quantum teleportation^{35,36,37,38,39}.
Figure 1 illustrate our experimental setup for remote transport of highdimensional OAM states and the quantum imaging with interactionfree light. The 710 nm laser beam is directed to the Lithium triborate crystal (LBO), via secondharmonic generation, to generate the ultraviolet pulses centered at 355 nm. The residual 710 nm light is guided onto a spatial light modulator (SLM1), that displays suitable holograms, for preparing coherent beam a in the desired highdimensional OAM superpositions to be transported. While the generated 355 nm ultraviolet pulses are directed to pump the βbarium borate crystal (BBO1) to create the nondegenerate OAMentangled photon pairs via SPDC, i.e., photon b and c, centered at 780 nm and 650 nm, respectively. Alice possesses coherent beam a and photon b while Bob holds photon c. In our experiment without any prior concentration, we have achieved the preparation of 3dimensional and 5dimensional entangled states with a high fidelity of 96.7% and 89.5%, respectively (see Supplementary Note 1 for details). Then, Alice performs the HDBLSM on her coherent beam a and photon b, by sending them together to BBO2 to do SFG. The generated 372 nm SFG photon is then coupled to a singlemode fiber (SMF) and detected by a photon multiplier tube (PMT). Thus the singlephoton event from PMT at Alice’s side indicates the successful HDBLSM and heralds a transferred photon c at Bob’s side.
Experimental observations for 3 and 5dimensional OAM states transport
In the first set of experiment, we prepare 12 OAM states to be transported at Alice’s side, \({{\varphi }_{1}\rangle }_{a}={1\rangle }_{a},\) \({{\varphi }_{2}\rangle }_{a}={0\rangle }_{a},\) \({{\varphi }_{3}\rangle }_{a}={1\rangle }_{a},\) \({{\varphi }_{4}\rangle }_{a}=({1\rangle }_{a}+{0\rangle }_{a}+{1\rangle }_{a})/\sqrt{3},\) \({{\varphi }_{5}\rangle }_{a}=({1\rangle }_{a}+\omega {0\rangle }_{a}+{\omega }^{2}{1\rangle }_{a})/\sqrt{3},\) \({{\varphi }_{6}\rangle }_{a}=({1\rangle }_{a}+{\omega }^{2}{0\rangle }_{a}+\omega {1\rangle }_{a})/\sqrt{3},\) \({{\varphi }_{7}\rangle }_{a}=(\omega {1\rangle }_{a}+{0\rangle }_{a}+{1\rangle }_{a})/\sqrt{3},\) \({{\varphi }_{8}\rangle }_{a}=({1\rangle }_{a}+\omega {0\rangle }_{a}+{1\rangle }_{a})/\sqrt{3},\) \({{\varphi }_{9}\rangle }_{a}=({1\rangle }_{a}+{0\rangle }_{a}+\omega {1\rangle }_{a})/\sqrt{3},\) \({{\varphi }_{10}\rangle }_{a}=({\omega }^{2}{1\rangle }_{a}+{0\rangle }_{a}+{1\rangle }_{a})/\sqrt{3},\) \({{\varphi }_{11}\rangle }_{a}=({1\rangle }_{a}+{\omega }^{2}{0\rangle }_{a}+{1\rangle }_{a})/\sqrt{3},\) and \({{\varphi }_{12}\rangle }_{a}=({1\rangle }_{a}+{0\rangle }_{a}+{\omega }^{2}{1\rangle }_{a})/\sqrt{3}\). They constitute the full four sets of 3dimensional mutually unbiased bases (MUB), which are commonly used to testify the universality of transport for all possible superposition states^{40}. The verification of transport results is based on the twofold coincidence detections between the SFG photon (detected by PMT) and photon c (detected by SPCM). After transporting, we employ the generalized GellMann matrix basis^{41} to reconstruct the density matrices of photon c at Bob’s side. Then the performance can be characterized by calculating the transport fidelity, \(F={{{{{\rm{Tr}}}}}}({\rho }_{E}{\rho }_{T})+\sqrt{1{{{{{\rm{Tr}}}}}}({\rho }_{E}^{2})}\sqrt{1{{{{{\rm{Tr}}}}}}({\rho }_{T}^{2})}\), where \({\rho }_{E}\) and \({\rho }_{T}\) represent the normalized density matrices of the transported states and the original ones, respectively. We present in Fig. 2 the transport results for a qutrit OAM state, \({{\varphi }_{4}\rangle }_{a}=({1\rangle }_{a}+{0\rangle }_{a}+{1\rangle }_{a})/\sqrt{3},\) without loss of generality, conditional on three different specific HDBLSM. Specifically, Fig. 2a, b, c show the resultant density matrices conditional on HDBLSM results of \({{\psi }_{00}\rangle }_{ab}\), \({{\psi }_{01}\rangle }_{ab}\), and \({{\psi }_{02}\rangle }_{ab}\), respectively, in each of which the upper rows are the original real and imaginary parts while the bottom are the experimentally reconstructed ones. The fairly good agreement between them can be seen clearly. For quantitative analysis, we also show in Fig. 3 the transport fidelities for all 12 MUB states conditional on these three HDBLSM. The average transport fidelity is \(0.879\pm 0.048\).
Secondly, we conduct the transport of 5dimensional OAM superposition state and show in Fig. 4 the transport fidelities for all 30 MUB states (Methods) conditional on the specific HDBLSM of \({{\psi }_{00}\rangle }_{ab}\). The average transport fidelity is \(0.796\pm 0.066\). Note that all these experimental results are obtained without any background subtraction. The fidelity imperfection mainly arises from two aspects: the prepared states (e.g., entangled states and the tobe transported states) and the imperfect spatiotemporal overlap of photons a and b at BBO2, see Supplementary Note 2 for details.
Quantum imaging with interactionfree light
Our remote highdimensional state transport scheme for the above OAM qutrit and ququint states can be straightforwardly extended to embrace any higherdimensional states. As well known, the vision or image is by far the most universal carrier for transferring information in the real world. According to the digital spiral imaging concept^{42,43}, any 2D optical image can be equivalently represented by a highdimensional state vector encoded in the full transverse spatial degrees of freedom. Actually, the aforementioned three and fivedimensional OAM states can also be seen as the 2D complex images in spatial domain. In this regard, our scheme can also represent another variant of ghost imaging. In the conventional ghost imaging, it necessitates two correlated beams; one beam illuminates an object while the image is recovered from the other beam that has never interacted with the object^{44}. However, these two beams actually share a common past, i.e., created from the same pump via SPDC. In contrast, our scheme uses the coherent state a to bear the image information, and then transport nonlocally to photon c after HDBLSM is done. Obviously, the photons used for object illumination and for image recovery do not share any common past, i.e., truly interactionfree. It’s noted that there have been several attempts, by using entanglementswapped photons, to achieve the ghost imaging^{45}. Here, specifically, for realizing such a quantum imaging with interactionfree light, Alice and Bob share the photon pairs that are entangled in the full transverse spatial modes (rather than OAM entanglement for perfect vortices), which should be written in terms of LG modes as^{31}, \({\varPsi \rangle }_{bc}={\sum }_{\ell,p}{C}_{p,p}^{\ell,\ell }{\ell,p\rangle }_{b}{\ell,p\rangle }_{c}\), where \({C}_{p,p}^{\ell,\ell }\) denotes the probability amplitude of finding photon b in the standard LG mode of \(\ell,p\rangle\) and the photon c in \(\ell,p\rangle\), with \(\ell\) and \(p\) being the azimuthal and radial indices, respectively. Then, at Alice’s side, the realamplitude images, e.g., threeleaf and fourleaf clover, are taken as examples for such a purpose. They are prepared and encoded by coherent beam a with the holograms displayed by SLM1, see the insets of Fig. 1b,c. As mentioned above, the transmission function of any optical image, \(O(r,\phi )\), can be equivalently represented by a highdimensional vector state as, \({\varphi \rangle }_{{{{{{\rm{a}}}}}}}={\sum }_{{\ell }_{a},{p}_{a}}{A}_{{\ell }_{a},{p}_{a}}{\ell,p\rangle }_{a}\), where \({A}_{{\ell }_{a},{p}_{a}}={\int [{{{{{{\rm{LG}}}}}}}_{{p}_{a}}^{{\ell }_{a}}(r,\phi )]}^{* }O(r,\phi )rdrd\phi\). Similarly, Alice mixes coherent beam a (bearing the image information) and photon b in the BBO2 crystal to conduct SFG, and then filters out the fundamental mode of SFG photons using a SMF that is connected to a PMT. Owing to OAM conservation in the process of SFG^{33}, the generated SFG photons can be collected by the SMF only when \({\ell }_{a}={\ell }_{b}\) and \({p}_{a}={p}_{b}\). Consequently, the photon c at Bob’s side is collapsed onto \({\varphi \rangle }_{c}={\sum }_{{\ell }_{c},{p}_{c}}{A}_{{\ell }_{c},{p}_{c}}{\eta }_{{p}_{c},{p}_{c}}^{{\ell }_{c},{\ell }_{c}}{C}_{{p}_{c},{p}_{c}}^{{\ell }_{c},{\ell }_{c}}{\ell,p\rangle }_{c}\), where \({\eta }_{{p}_{c},{p}_{c}}^{{\ell }_{c},{\ell }_{c}}\) denotes the frequency upconversion efficiency for \({{{{{{\rm{LG}}}}}}}_{{p}_{c}}^{{\ell }_{c}}\) and \({{{{{{\rm{LG}}}}}}}_{{p}_{c}}^{{\ell }_{c}}\). For sufficiently large and strong pump beams, both \({\eta }_{{p}_{c},{p}_{c}}^{{\ell }_{c},{\ell }_{c}}\) and \({C}_{{p}_{c},{p}_{c}}^{{\ell }_{c},{\ell }_{c}}\) can be approximately a constant, which means that photon c just takes a perfect replica of the original state of coherent beam a. Then, the singlephoton events from PMT indicates the HDBLSM result \({{\psi }_{00}\rangle }_{ab}\) at Alice’s side, which serves the trigger signal for the ICCD camera at Bob’s side to capture the image that is transported onto photon c.
We present our experimental observations of the interactionfreelightbased ghost imaging of threeleaf and fourleaf Clover images in Fig. 5a,b respectively. As shown by the dotted lines, we define the regions of interest (ROI) for both threeleaf and fourleaf Clover images, which occupy about 432 and 441 pixels, respectively. Here, for a quantitative analysis, we adopt the contrasttonoise ratio^{46}, \({{{{{\rm{CNR}}}}}}=(\langle {G}_{in}\rangle \langle {G}_{out}\rangle )/\sqrt{{\sigma }_{in}^{2}+{\sigma }_{out}^{2}}\) to characterize the image quality of recorded images, where \(\langle {G}_{in}\rangle\) and \(\langle {G}_{out}\rangle\) are the ensemble average of the photon numbers at any pixel inside and outside the ROI, respectively, while \({\sigma }_{in}^{2}\) and \({\sigma }_{out}^{2}\) are the corresponding variances. We have \({{{{{\rm{CNR}}}}}}=\)1.37 and 1.58 for threeleaf and fourleaf Clover images, respectively, which are at the same level with the traditional ghost imaging^{47,48}, therefore confirming the good performance of our system. Note that, the edges of recorded images in Fig. 5a,b are not sharp compared with the original ones (indicated by insets), which mainly caused by both the nonmaximal spatial entanglement for photon pairs generated via SPDC and the nonuniform SFG efficiency for each pixel in the HDBLSM stage. Also, a better time and spatial overlap of all the image planes of the ICCD camera, BBO crystal, and the SLM1 can further improve the CNR.
Discussion
Here, we use a strong coherent light instead of single photon as the input state to enhance the SFG conversion (~4%) to perform the HDBLSM, and thus realizing the remote transport of highdimensional OAM states from a coherent beam to a single photon. Therein the knowledge of the coherent source is not used in our scheme, which is similar with one of the key features of teleportation. However, in the original proposal of quantum teleportation, the input unknown state to be teleported should be encoded with single photon. If we adopt the single photon as input source, limited by the rather low frequency conversion efficiency, we need to wait a much longer time for teleportation to occur. In this regard, it is expected that, with the future advances in structured light nonlinear frequency conversion, our scheme can work in the singlephoton scenario and will pave the way for realizing truly secure highdimensional quantum teleportation in the upcoming quantum network.
In our scheme, Alice only needs to send the HDBLSM result via a classical channel while the photon c received by Bob is always in the singlephoton state. Then Bob is required to conduct a desired unitary operation on photon c, according to the results of HDBLSM (indicated by Eq. (2) to Eq. (10)), to recover the highdimensional OAM states correctly. Thus, under ensuring the eavesdropper has no access to the coherent photons a, such a procedure can offer an additional security for the information transport and ghost imaging protocol. Whereas, in the direct transport of OAM states, the eavesdropper can acquire the information at any position along the communication channel.
In summary, we have demonstrated a remote transport of highdimensional OAM superposition states at a distance, by adopting the perfect vortices as the OAM basis for both a successful preparation of highdimensional maximal entanglement and a better performance of SFGbased HDBLSM. We have experimentally achieved the average transport fidelity of \(0.879\pm 0.048\) and \(0.796\pm 0.066\) for a complete set of 3dimensional and 5dimensional MUB, respectively. Further, by exploring the full transverse spatialmode entanglement, we have succeeded in realizing the ghost imaging of amplitude objects with interactionfree light, which is fundamentally different from previous quantum imaging techniques^{49}, such as the conventional ghost imaging^{44} or quantum imaging with undetected photons^{50}. By calculating the CNR of recorded images, we estimate that the imaging performance of this protocol can reach the same level with the traditional ghost imaging. It can be expected that, with the development of the stateoftheart nonlinear frequency conversion technology, our scheme can be competent to work fully with all single photons in the near future, that is truly highdimensional quantum teleportation.
Methods
Perfect vortices
Photon pairs generated by SPDC have proven to be a reliable entanglement source. However, under the thincrystal approximation and phasematching condition with a Gaussian pump beam, the downconverted twophoton OAM entanglement inevitably suffers from the limited spiral bandwidth^{31}, i.e., lowerorder LG modes appears more frequently than higherorder ones. If the standard LG modes are used to represent the OAM eigenstates \(\ell \rangle\), namely, \(\langle r,\phi \ell \rangle=L{G}_{p=0}^{\ell }(r,\phi )\), the twophoton OAMentangled state can be written as, \({\varPsi \rangle }_{bc}={\sum }_{\ell }{C}_{\ell }{\ell \rangle }_{b}{\ell \rangle }_{c}\), where \({C}_{\ell,\ell }=\int L{G}_{0}^{0}(r,\phi ){[L{G}_{0}^{\ell }(r,\phi )]}^{\ast }{[L{G}_{0}^{\ell }(r,\phi )]}^{\ast }\) represents the probability amplitude of finding photon b in the mode of \(\ell \rangle\) and the photon c in the mode of \(\ell \rangle\). Generally, \({\varPsi \rangle }_{bc}\) is merely a nonmaximally entangled OAM state. However, the maximal entanglement is prerequisite for our present scheme. To overcome this obstacle, we adopt the socalled perfect vortices^{32} to represent the OAM states, instead of the standard LG modes. In our scheme, we prepare the perfect vortices by modifying the LG modes as^{51},
where \(l\) is a constant, e.g., \(l=5\). In such a basis of perfect vortices, we know that \({C}_{\ell,\ell }=\int L{G}_{0}^{0}(r,\phi ){[{{{{{{\rm{MLG}}}}}}}_{0}^{\ell }(r,\phi )]}^{* }{[{{{{{{\rm{MLG}}}}}}}_{0}^{\ell }(r,\phi )]}^{* }\) will become a constant, as they share the same radial intensity distribution such that the overlap probability is identical. Thus, we can obtain the desired maximally entangled OAM state as, \({\varPsi \rangle }_{bc}=\frac{1}{\sqrt{d}}{\sum }_{\ell }{\ell \rangle }_{b}{\ell \rangle }_{c}\), with \(d\) being the dimension of the OAM subspace. Besides, the perfect vortices are also crucial for performing the HDBLSM via SFG. Based on the couplewave equations describing the SFG, we can estimate the frequency conversion efficiency for our perfect vortices as^{24,52,53},
Similarly, we can expect that \({\eta }_{{\ell }_{a},{\ell }_{b}}\) will become a constant even for different OAM modes \({\ell }_{a}\) and \({\ell }_{b}\), and thus realizing a reliant HDBLSM.
Generally, the computergenerated holographic grating displayed by a spatial light modulator (SLM) is used for OAM generation and detection experiments. Here, for making perfect vortices, we need to modulate the standard LG modes to let them carry the same radial intensity profiles regardless of different OAM numbers. We illustrate in Fig. 6 the basic principle for making the desired holograms for the perfect vortices. The resultant holograms addressed by SLM can be mathematically described by^{54},
where \(\varPhi {(r,\phi )}_{{{{{{{\rm{LG}}}}}}}_{0}^{\ell }}\) and \(I{(r,\phi )}_{{{{{{{\rm{LG}}}}}}}_{0}^{l}}\) are desired phase (for different \(\ell\)) and intensity profiles (e.g., \(l=5\)), \(\varPhi {(r,\phi )}_{{{{{{\rm{Linear}}}}}}}\) is the phase of a linear blazed grating, and \({{{{{{\rm{sinc}}}}}}}^{2}(\cdot )\) accounts for the mapping of the phase depth to the diffraction efficiency. From the holographic gratings of Supplementary Fig. 4a–e and the experimental observations of Supplementary Fig. 4f–j, we can see that the different modified LG modes share almost the same intensity profile regardless of carrying different OAM numbers, i.e., the perfect vortices (see Supplementary Note 3 for details).
Experimental setup
Our experimental setup for the remote transport of highdimensional OAM states and the ghost imaging with interactionfree light is sketched in Fig. 1. The 140 fs pulsed laser beam with an average power of 1950 mW, a central wavelength of 710 nm and a repetition rate of 80 MHz is focused by lens L1 \(({f}_{1}=150{{{{{\rm{mm}}}}}})\) and guided to pump a 1.5mmlong LBO crystal to generate the ultraviolet pulse of a central wavelength of 355 nm and an average power of 200 mW via secondharmonic generation (SHG). The residual 710 nm light beam is collimated by L2 \(({f}_{2}=100{{{{{\rm{mm}}}}}})\) and guided onto a spatial light modulator (SLM1), which is used to display suitable holographic gratings for preparing coherent beam a in an arbitrary highdimensional OAM superposition state to be transported. After reflection from SLM1, the light beam has a power of roughly 140 mW. While the generated 355 nm SHG ultraviolet pulses are directed to pump a 3mmlong βbarium borate crystal (BBO1) to create the frequency nondegeneracy OAMentangled photon pairs via SPDC, i.e., photon b and c, whose wavelengths are centered at 780 nm and 650 nm, respectively. A longpass dichroic mirror (DM2) is inserted to separate the downconverted photons from the pump. For performing HDBLSM, the 780 nm photon b reflected from DM3 and coherent beam a are combined through DM4, and then sent to 1.5mmlong BBO2 for implementing SFG. Note that, coherent beam a and photon b are both imaged onto the BBO2 via 4 f imaging systems, L7L9 and L8L9 (\({f}_{7}={f}_{8}=500{{{{{\rm{mm}}}}}}\),\({f}_{9}=100{{{{{\rm{mm}}}}}}\)), to ensure a good spatial overlap of the image planes of BBO2, BBO1 and SLM1. In addition, the temporal overlap between coherent beam a and photon b in BBO2 is realized by accurately adjusting the position of rightangle prism (RAP). Subsequently, we use another 4 f system (\({f}_{10}=500{{{{{\rm{mm}}}}}}\) and \({f}_{11}=300\)) together with a 4 mm collimated lens to couple the generated 372 nm SFG photon into a singlemode fiber (SMF), which is connected to a photon multiplier tube (PMT). The singlephoton event from PMT indicates the successful HDBLSM signal and herald a transported photon c at Bob’s side. For performing the state tomography, photon c is imaged onto SLM2 with a 4 f imaging system (\({f}_{3}=200{{{{{\rm{mm}}}}}}\) and \({f}_{4}=400{{{{{\rm{mm}}}}}}\)) and then reimaged onto the input facet of SMF with another 4 f imaging system (\({f}_{5}=500{{{{{\rm{mm}}}}}}\) and \({f}_{6}=10{{{{{\rm{mm}}}}}}\)). By switching the flipper mirror (FM), we can capture the transported images by using the ICCD camera, which is triggered by the singlephoton events from the PMT. Note that an image preserving optical delay of about 25 m is requested to compensate the electronic delay associated with the PMT and the trigger mechanism in the ICCD camera^{47}.
Generalized protocol for remote transport of quidit
Assume Alice wishes to transport to Bob an arbitrary ddimensional OAM state encoded by coherent beam a, \({\varphi \rangle }_{a}={\sum }_{\ell=\lceil d/2\rceil }^{\lfloor d/2\rfloor }{\alpha }_{\ell }{\ell \rangle }_{a}\), where \(\lfloor \cdot \rfloor\) and \(\lceil \cdot \rceil\) denote the operation of floor and ceil, respectively, and \(\sum {{\alpha }_{\ell }}^{2}=1\). Alice and Bob need to share a ddimensional maximally entangled state between photons b and c, \({\varPsi \rangle }_{bc}=\frac{1}{\sqrt{d}}{\sum }_{\ell=\lceil d/2\rceil }^{\lfloor d/2\rfloor }{\ell \rangle }_{b}{\ell \rangle }_{c}\). Then, Alice performs the HDBLSM, \({{\psi }_{mn}\rangle }_{ab}=\frac{1}{\sqrt{d}}{\sum }_{\ell=\lceil d/2\rceil }^{\lfloor d/2\rfloor }\exp (i2n\ell \pi /d){\ell \rangle }_{a}{\ell \oplus m\rangle }_{b}\), jointly on coherent beam a and photon b, where \(m,n\in \{0,1,2,\cdots d1\}\) and \(\ell \oplus m=(\ell+m){{{{\mathrm{mod}}}}}\,d\). Similarly, by adopting SFG for HDBLSM, we can map all these Belllike states onto the SFG singlephoton states, namely,
Note that for \({{\psi }_{00}\rangle }_{ab}\), we get the OAM state of SFG photon as \(\ell=0\rangle\), which is just orthogonal to those for all the rest Belllike states and can be directly detected with SMF together with PMT. Turning to our specific case of remote transport of OAM ququint states, conditional on \({{\psi }_{00}\rangle }_{ab}\), we have the photon c in the state, \({\varphi \rangle }_{c}={}_{ab}\langle {\psi }_{00}\varPsi \rangle _{abc}={\sum }_{\ell=2}^{2}{\alpha }_{\ell }\ell \rangle\), which is exactly the same with the initial state of coherent beam a.
Mutually unbiased vectors
Generally, the full sets of mutually unbiased bases (MUB) are used to testify the universality of remote transport for all possible superpositions states^{40}. Here, for a 5dimensional system, we need measure 30 states from six mutually unbiased bases (I~VI):
where \(\gamma=\exp (i2\pi /5)\).
Data availability
The datasets generated and analysed during this study are available from the corresponding author upon request. Source data for Figs. 2–4 are provided with this paper. Source data are provided with this paper.
References
Allen, L., Beijersbergen, M. W., Spreeuw, R. J. C. & Woerdman, J. P. Orbital angular momentum of light and the transformation of LaguerreGaussian laser modes. Phys. Rev. A 45, 8185–8189 (1992).
MolinaTerriza, G., Torres, J. P. & Torner, L. Twisted photons. Nat. Phys. 3, 305–310 (2007).
Padgett, M. J. Orbital angular momentum 25 years on [Invited]. Opt. Express 25, 11265–11274 (2017).
Erhard, M., Fickler, R., Krenn, M. & Zeilinger, A. Twisted photons: new quantum perspectives in high dimensions. Light Sci. Appl. 7, 17146 (2018).
Forbes, A., de Oliveira, M. & Dennis, M. R. Structured light. Nat. Photonics 15, 253–262 (2021).
Wang, J. et al. Terabit freespace data transmission employing orbital angular momentum multiplexing. Nat. Photonics 6, 488–496 (2012).
Bozinovic, N. et al. Terabitscale orbital angular momentum mode division multiplexing in fibers. Science 340, 1545–1548 (2013).
Mair, A., Vaziri, A., Weihs, G. & Zeilinger, A. Entanglement of the orbital angular momentum states of photons. Nature 412, 313–316 (2001).
Dada, A. C., Leach, J., Buller, G. S., Padgett, M. J. & Andersson, E. Experimental highdimensional twophoton entanglement and violations of generalized Bell inequalities. Nat. Phys. 7, 677–680 (2011).
Cao, H. et al. Terabit freespace data transmission employing orbital angular momentum multiplexing. Optica 7, 232–237 (2020).
Cozzolino, D. et al. Aircore fiber distribution of hybrid vector vortexpolarization entangled states. Adv. Photon. 1, 046005 (2019).
Zhu, Y. et al. Compensationfree highdimensional freespace optical communication using turbulenceresilient vector beams. Nat. Commun. 12, 1666 (2021).
Zhou, Y. et al. Highfidelity spatial mode transmission through a 1kmlong multimode fiber via vectorial time reversal. Nat. Commun. 12, 1866 (2021).
Nape, I. et al. Revealing the invariance of vectorial structured light in complex media. Nat. Photo 16, 538–546 (2022).
Zhang, R. et al. Turbulenceresilient pilotassisted selfcoherent freespace optical communications using automatic optoelectronic mixing of many modes. Nat. Photonics. 15, 743–750 (2021).
Krenn, M. et al. Twisted light transmission over 143 km. Proc. Natl Acad. Sci. USA 113, 13648–13653 (2016).
Lavery, M. P. J. et al. Freespace propagation of highdimensional structured optical fields in an urban environment. Sci. Adv. 3, e1700552 (2017).
Buono, W. T. & Forbes, A. Nonlinear optics with structured light. OptoElectron. Adv. 5, 210174 (2022).
Qiu, X., Li, F., Zhang, W., Zhu, Z. & Chen, L. Spiral phase contrast imaging in nonlinear optics: seeing phase objects using invisible illumination. Optica 5, 208–212 (2018).
da Silva, B. P. et al. Observation of a triangularlattice pattern in nonlinear wave mixing with optical vortices. Optica 9, 908–912 (2022).
Zhou, Z.Y. et al. Orbital angular momentumentanglement frequency transducer. Phys. Rev. Lett. 117, 103601 (2016).
Liu, S. et al. Highdimensional quantum frequency converter. Phys. Rev. A 101, 012339 (2020).
Wu, H.J. et al. Conformal frequency conversion for arbitrary vectorial structured light. Optica 9, 187–196 (2022).
Qiu, X., Guo, H., Ren, Y. & Chen, L. Highdimensional photonic orbitalangularmomentum frequency interface. Phys. Rev. Appl. 19, 044072 (2023).
Eckstein, A., Brecht, B. & Silberhorn, C. A quantum pulse gate based on spectrally engineered sum frequency generation. Opt. Express 19, 13770–13778 (2011).
Brecht, B. et al. Demonstration of coherent timefrequency Schmidt mode selection using dispersionengineered frequency conversion. Phys. Rev. A 90, 030302(R) (2014).
Donohue, J. M., Agnew, M., Lavoie, J. & Resch, K. J. Coherent ultrafast measurement of timebin encoded photons. Phys. Rev. Lett. 111, 153602 (2013).
Kim, Y., Kulik, S. & Shih, Y. Quantum teleportation of a polarization state with a complete bell state measurement. Phys. Rev. Lett. 86, 1370–1373 (2001).
Sangouard, N. et al. Faithful entanglement swapping based on sumfrequency generation. Phys. Rev. Lett. 106, 120403 (2011).
Li, Y. et al. Multiuser timeenergy entanglement swapping based on dense wavelength division multiplexed and sumfrequency generation. Phys. Rev. Lett. 123, 250505 (2019).
Torres, J. P., Alexandrescu, A. & Torner, L. Quantum spiral bandwidth of entangled twophoton states. Phys. Rev. A 68, 050301(R) (2003).
Ostrovsky, A. S., RickenstorffParrao, C. & Arrizón, V. Generation of the “perfect” optical vortex using a liquidcrystal spatial light modulator. Opt. Lett. 38, 534–536 (2013).
Qiu, X.D. et al. Optical vortex copier and regenerator in the Fourier domain. Photonics Res. 6, 641–646 (2018).
Wang, F. et al. Generation of the complete fourdimensional Bell basis. Optica 4, 1462–1467 (2017).
Bennett, C. H. et al. Teleporting an unknown quantum state via dual classical and EinsteinPodolskyRosen Channels. Phys. Rev. Lett. 70, 1895–1899 (1993).
Bouwmeester, D. et al. Experimental quantum teleportation. Nature 390, 575–579 (1997).
Luo, Y. et al. Quantum teleportation in high dimensions. Phys. Rev. Lett. 123, 070505 (2019).
Hu, X.M. et al. Experimental highdimensional quantum teleportation. Phys. Rev. Lett. 125, 230501 (2020).
Walborn, S. P., Ether, D. S., de Matos Filho, R. L. & Zagury, N. Quantum teleportation of the angular spectrum of a singlephoton field. Phys. Rev. A 76, 033801 (2007).
Ivanovic, I. D. Geometrical description of quantal state determination. J. Phys. A Math. Gen. 14, 3241–3245 (1981).
Bertlmann, R. A. & Krammer, P. Bloch vectors for qudits. J. Phys. A Math. Theor. 41, 235303 (2008).
Torner, L., Torres, J. P. & Carrasco, S. Digital spiral imaging. Opt. Exp. 13, 873–881 (2005).
Chen, L., Lei, J. & Romero, J. Quantum digital spiral imaging. Light Sci. Appl. 3, e153 (2018).
Pittman, T. B., Shih, Y. H., Strekalov, D. V. & Sergienko, A. V. Optical imaging by means of twophoton quantum entanglement. Phys. Rev. A 52, R3429–R3432 (1995).
Bomman, N. et al. Ghost imaging using entanglementswapped photons. NPJ Quantum Inf. 5, 63 (2019).
Zerom, P. et al. Thermal ghost imaging with averaged speckle patterns. Phys. Rev. A 86, 063817 (2012).
Qiu, X., Zhang, D., Zhang, W. & Chen, L. StructuredPumpEnabled quantum pattern recognition. Phys. Rev. Lett. 122, 123901 (2019).
Qiu, X. et al. Parallel ghost imaging. Adv. Quantum Technol. 3, 2000073 (2020).
Moreau, P. A., Toninelli, E., Gregory, T. & Padgett, M. J. Imaging with quantum states of light. Nat. Rev. Phys. 1, 367–380 (2019).
Lemos, G. B. et al. Quantum imaging with undetected photons. Nature 512, 409–412 (2014).
Chen, L., Zhang, W., Lu, Q. & Lin, X. Making and identifying optical superpositions of high orbital angular momenta. Phys. Rev. A 88, 053831 (2013).
Zhou, Z. Y. et al. Orbital angular momentum photonic quantum interface. Light Sci. Appl. 5, e16019–e16019 (2016).
Ni, R. et al. Topological charge transfer in frequency doubling of fractional orbital angular momentum state. Appl. Phys. Lett. 109, 15103 (2016).
Leach, J., Dennis, M. R., Courtial, J. & Padgett, M. J. Vortex knots in light. N. J. Phys. 7, 55 (2005).
Acknowledgements
This work was supported by the National Natural Science Foundation of China (12034016, 61975169), the National Key R&D Program of China (2023YFA1407200), China Postdoctoral Science Foundation (2021M691891), the Natural Science Foundation of Fujian Province of China (2021J02002), and the program for New Century Excellent Talents in University of China (NCET130495).
Author information
Authors and Affiliations
Contributions
L.C. conceived the idea. L.C. and X.Q. designed the experiment. X.Q. conducted the experiment with the help from H.G and L.C. X.Q. and L.C. analyzed the data and cowrote the manuscript. L.C. supervised the project.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Jonathan Leach and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Source data
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
Qiu, X., Guo, H. & Chen, L. Remote transport of highdimensional orbital angular momentum states and ghost images via spatialmodeengineered frequency conversion. Nat Commun 14, 8244 (2023). https://doi.org/10.1038/s41467023439504
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467023439504
This article is cited by

Orbital angular momentum lasers
Nature Reviews Physics (2024)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.