Abstract
Dipole–dipole interactions, which govern phenomena such as cooperative Lamb shifts, superradiant decay rates, Van der Waals forces and resonance energy transfer rates, are conventionally limited to the Coulombic nearfield. Here we reveal a class of realphoton and virtualphoton longrange quantum electrodynamic interactions that have a singularity in media with hyperbolic dispersion. The singularity in the dipole–dipole coupling, referred to as a superCoulombic interaction, is a result of an effective interaction distance that goes to zero in the ideal limit irrespective of the physical distance. We investigate the entire landscape of atom–atom interactions in hyperbolic media confirming the giant longrange enhancement. We also propose multiple experimental platforms to verify our predicted effect with phonon–polaritonic hexagonal boron nitride, plasmonic superlattices and hyperbolic metasurfaces as well. Our work paves the way for the control of cold atoms above hyperbolic metasurfaces and the study of manybody physics with hyperbolic media.
Introduction
Dipole–dipole interactions (DDIs) are instrumental in mediating entanglement and superradiance in cold atoms^{1,2,3}, as well as coherent coupling between single molecules or atoms^{4,5,6,7}. Often described by real and virtual photon exchange, they also cause cooperative frequency shifts between superconducting qubits in circuit quantum electrodynamic (QED) systems^{8,9} and Förster resonance energy transfer (FRET) between dye molecules or quantum dots^{10,11}. There are two fundamental ways of controlling the strength and length scales of DDIs. The first method involves the tuning of intrinsic atomic properties such as transition dipole moments and transition frequencies (cf. highly excited Rydberg atoms and superconducting qubits^{7,12,13}). The second method involves the tuning of the QED vacuum, achieved through cavities, waveguides and photonic bandgaps^{14,15,16,17}. Up to now, these electrodynamic methods have relied on resonant effects that require large quality factors along with extensive nanofabrication steps. It is an open question, however, whether there exists alternative nonresonant techniques for controlling DDIs that would be robust to broad spectral lineshapes of atoms or molecules with possible room temperature applications. Here we present work related to this new avenue of research.
In this study, we reveal a class of divergent excitedstate atom–atom interactions that can occur in natural and artificial media with hyperbolic dispersion. Unlike the above mentioned approaches, which engineer radiative coupling, we show that the homogeneous hyperbolic medium itself fundamentally alters the Coulombic nearfield. The resultant singular longrange interaction, referred to as a superCoulombic interaction, is described by an effective interaction distance that goes to zero (r_{e}→0) along a materialdependent resonance angle. We show that this interaction affects the entire landscape of real photon and virtual photon phenomena such as the cooperative Lamb shift (CLS), the cooperative decay rate (CDR), resonance energy transfer rates and frequency shifts, as well as resonant interatomic forces. Although we find that the singularity is curtailed by material absorption, it still allows for interactions with much larger magnitudes and longer ranges than those found in any conventional media. We also show that atoms in a hyperbolic medium will exhibit a strong orientational dependence that can effectively switch the dipolar interaction off or on, providing an additional degree of freedom to control DDI. Our investigation reveals a marked contrast between groundstate and excitedstate interactions which can be used to distinguish the superCoulombic effect in experiment. Finally, we provide a unified perspective for controlling DDIs on multiple experimental platforms for hyperbolic media including plasmonic superlattices, hyperbolic metasurfaces and natural hyperbolic media such as hexagonal boron nitride (hBN).
We emphasize that the materials platform we introduce in this study, to enhance DDIs, is fundamentally different from the cavity QED^{18,19} or waveguide QED regimes^{5,8,20,21} (see Supplementary Table 1). We do not rely on atom confinement^{2,5,6,7,19}, cavity resonances or modal effects such as the quasi transverse electromagnetic (TEM) mode in circuit QED^{8}, the bandedge slow light as in PhC waveguides^{5,6,22}, the lowmode volume of plasmonic waveguides^{21,23} or the infinite phase velocity at the cutoff frequency of epsilonnearzero (ENZ) waveguides^{14,24}. We also stress that the superCoulombic effect engineers the conventional nonradiative (longitudinal) nearfields as opposed to radiative (transverse) modes and will occur over a broad range of frequencies due to the broadband nature of the hyperbolic dispersion relation^{25,26,27,28}. Figure 1 depicts a schematic of the proposed superCoulombic DDI using hBN^{29,30,31,32} and two dopant atoms. In the infrared spectral range, hBN is a uniaxial material that supports ordinary waves (polarization perpendicular to the optic axis) and extraordinary waves (polarization along the optic axis). Extraordinary waves satisfy the hyperbolic dispersion relation when .
Results
QED theory of hyperbolic media
We begin by formulating the QED theory^{33} of DDIs between two neutral, nonmagnetic atoms in a hyperbolic medium. We focus on dipolar interactions where the electrodynamic field is initially prepared in the vacuum state . Using the multipolar Hamiltonian, the interaction of two neutral atoms [positions r_{j}, transition frequencies ω_{j} and transition electric dipole moments (j=a, b)] is specified by the interaction Hamiltonian
where h.c. stands for the Hermitian conjugate. The matterassisted electric field is given by
and G(r, r′;ω) is the classical dyadic Green function that satisfies the macroscopic Maxwell equations. Here, and are bosonic field operators, which play the role of the creation and annihilation operators of the matterassisted electromagnetic (polaritonic) field. The unique interaction properties are a direct result of the dispersion relation of the hyperbolic polariton, as opposed to the photonic dispersion relation, ω=ck, seen in vacuum. The electric field is defined so that it rigorously satisfies the equaltime commutation relations and fluctuation–dissipation theorem^{33}. We use conventional perturbation theory to calculate the various dipolar interactions in a hyperbolic medium. We emphasize that the QED theory captures both ground state–ground state interactions and excited state–ground state interactions, which a semiclassical approach cannot.
Resonant dipole–dipole interaction
If the initial state of the atomic system is prepared in the symmetric or antisymmetric state, , then one can show that the resonant DDI (RDDI) (see Methods) is given by
where is the transition dipole moment of atom j, assumed to be real. J_{dd} is the CLS (also known as the virtual photon exchange interaction) and γ_{dd} is the CDR commonly associated with superradiant or subradiant effects.
Our result for the RDDI in a hyperbolic medium is
valid when r_{a}≠r_{b}. The first term arises exclusively from extraordinary waves following a hyperbolic dispersion, whereas the second term arises from a combination of ordinary and extraordinary waves. Here we have defined the nearfield and farfield dipole orientation matrix factors and , respectively. Equation (4) reduces to the vacuum RDDI expression when , which is applicable both in the retarded (r≫λ) and nonretarded (r<<λ) regimes. The most unique aspect of DDIs in uniaxial media is the divergence that is predicted from the first term only when the hyperbolic condition is satisfied. In the ideal lossless limit, we find that the effective interaction distance between two atoms, , tends towards the limit
This superCoulombic effect results in the divergence of the DDI strength V_{dd}/ħ along the resonance angle θ_{R}, defined with respect to the optic axis.
Atoms in a hyperbolic medium will then have an associated CLS and CDR
in the limit θ→θ_{R}. Equations (6) and (7) are the dominant factors of the extraordinary wave contribution only.
We now contrast the scaling of CLS with distance when mediated by hyperbolic media as opposed to vacuum modes. In vacuum, for separation distances much larger than the transition wavelength, the CLS scales as and becomes much smaller than the freespace spontaneous emission rate (γ_{o}). On the other hand, for distances much smaller than the wavelength, the CLS scales as , which implies that it can become much larger than the spontaneous emission rate. In contrast, the CLS in a hyperbolic medium is dependent on , , for all interatomic distances. The materialdependent factor 1/ diverges in the lossless case and therefore results in giant CLSs for short and large interatomic distances.
This marked contrast is also revealed in the CDR. At large distances, the CDR in vacuum scales as , therefore becoming weak for distances much larger than the wavelength. For distances much smaller than the wavelength, the CDR becomes independent of position, , and remains on the order of the freespace spontaneous emission rate. In contrast, the CDR in a hyperbolic medium along the resonance angle is not dependent on the effective interaction distance r_{e} and instead it depends crucially on the orientation angle φ of the dipoles, . When both dipoles are oriented perpendicular to the optic axis (φ=π/2), there exists a unique wavelength when the medium can achieve an anisotropic ENZ medium ( and ) resulting in a divergent CDR. Surprisingly, the effect is independent of interatomic distance. When both dipoles are parallel to the optic axis (φ=0), the same anisotropic ENZ condition gives a null CDR between the two atoms, independent of interatomic distance.
We will now consider the role of material absorption ( and ) on atom–atom interactions in a hyperbolic medium. We find that the effective interaction distance is not zero and tends to the finite value . This curtails the singularity of the hyperbolic dipolar interaction but nevertheless allows for very large interaction strengths compared with conventional media whenever r_{e}/r<1 is satisfied. Material absorption will also modify the spatial scaling laws of the RDDI in equation (3) so that both the CDR and Lamb shift will scale as . Another consequence of material absorption on RDDIs is in the transition from nonretarded (r^{−3}) to retarded (r^{−1}) interactions. In vacuum, the transition occurs when the interatomic separation distance is on the order of wavelength,. In an ideal lossless hyperbolic medium, this transition from nearfield to farfield does not occur, as the effective separation distance approaches zero, r_{e}→0 specifically along the resonance angle of a hyperbolic medium. Therefore, we find that RDDI should scale with the characteristic power law of nearfield (longitudinal) nonradiative interactions (r^{−3}) for all interatomic distances. Once material absorption is included, the transition is expected to occur approximately when . The dipolar interactions will transition from the power law to the exponential scaling law , which is valid at large interatomic distances.
Figure 2 shows the result of the CLS and CDR for two zoriented dipoles in a hyperbolic medium that includes material absorption. We compare the RDDIs with the conventional results of a lossy dielectric and vacuum. It is noteworthy that the RDDI peaks near the resonance angle θ_{R} as predicted theoretically. The spatial field plots in the insets clearly demonstrate the distinguishing features of the RDDI in a hyperbolic medium compared with vacuum. Figure 2c,d demonstrate the superCoulombic spatial dependence along the resonance angle. It is noteworthy that the sign of the interaction is dependent on the orientation of the dipoles, as well as the relative position of the dipoles within the hyperbolic medium.
Orientational dependence
We now turn to the unique orientational dependence of the RDDI between two atoms positioned along the resonance angle θ_{R}. In Fig. 3, we plot the normalized CLS of two atoms a full wavelength apart (r=λ) as a function of dipole orientation angle φ. The CLS has a minimum when φ=θ_{R} and a maximum when φ=θ_{R}+π/2. Assuming that , , and , we find that the ratio between the maximum and minimum is
showing that it is proportional to the square of the figure of merit of the hyperbolic medium. In Fig. 3, we use the full Green’s function to calculate the orientational dependence of the dipolar interaction in a hyperbolic medium with material absorption and find excellent agreement with the analytical expression.
Resonance energy transfer
We now consider secondorder superCoulombic QED interactions between nonidentical atoms arising from initial state preparation consisting of atom A in its excited state and atom B in its ground state, . In the weakcoupling regime, an irreversible resonance energy transfer takes place, transferring a photon from atom A to atom B. This process is FRET and the transfer rate given by Fermi’s golden rule is Γ_{ET}=2πħ^{−1}V_{dd}^{2}δ(ħω_{a}−ħω_{b}). Along the resonance angle, FRET is mediated by hyperbolic modes and the rate is given by
which shows a scaling dependence and giant enhancement—the key signature of second order superCoulombic interactions in hyperbolic media.
Casimir–Polder potential
In addition to the FRET rate, there is also a predicted frequency shift that comes from the initial state preparation . This is the excitedstate Casimir–Polder potential, , composed of a resonant and offresonant contribution. The resonant excitedstate Casimir–Polder potential is of the form (ref. 34). We therefore predict that the excitedstate energy potential will also diverge with a scaling dependence similar to the FRET rate.
Figure 4 shows the full numerical results for the secondorder DDIs in a lossy hyperbolic medium, a lossy dielectric and vacuum. In the nonretarded regime , we clearly see the effect of the superCoulombic interaction, which results in a large enhancement of the dipolar interactions U_{eg} and Γ_{ET} (shown in inset). The superCoulombic enhancement occurs only along the asymptotes of the hyperboloid and is unrelated to the suppression of FRET rate of an ensemble of emitters near a conventional metallic surface or hyperbolic medium^{35,36,37}.
It is interesting that the dispersive van der Waals interaction between two groundstate atoms does not diverge in a hyperbolic medium. Using fourthorder perturbation theory^{38}, the interaction energy between two groundstate atoms is given by , where α_{A,B}(ω) is the isotropic electric polarizability of atom A or B. In the nonretarded limit, the dominant contribution is given by
which reduces to the wellknown freespace nonretarded van der Waals interaction energy when . It is important to note that the integral is performed over the entire range of positive imaginary frequencies (η=iω). In general, the hyperbolic condition is only satisfied within a finite bandwidth of the electromagnetic spectrum. We therefore expect that it would not alter the broadband cumulative effect of the entire electromagnetic spectrum and, as a result, we predict that the ground state–ground state interaction energy will not diverge in a hyperbolic medium. From Fig. 4, it is also clear that the ground state–ground state Casimir–Polder potential U_{gg} does not show any type of enhancement for the hyperbolic medium, in agreement with our discussion. It is noteworthy that the distance scaling dependence in the nonretarded regions is in agreement with equations (9) and (10), as expected. In the retarded regime (r≫λ), the excitedstate interactions U_{eg} and Γ_{ET} display an exponential damping behaviour due to material absorption, whereas the groundstate interaction U_{gg} displays the typical Casimir–Polder power law dependence, r^{−7} (Fig. 4).
Discussion
In the following, we discuss multiple experimental platforms for hyperbolic media paving the way for the experimental demonstration of the longrange superCoulombic interactions and unique manybody physics in hyperbolic media.
Figure 5a–b propose a practical plasmonic superlattice system to enhance atomatom interactions taking into account the role of dissipation, dispersion and finite unit cell size. We show the large enhancement of CLS (J_{dd}) for an effective medium model and compare it with a 40layer structure consisting of Ag and TiO_{2} with a total slab thickness of 100 nm. For such a system, effective medium theory predicts a type I response for wavelengths smaller than 492 nm and a type II response for wavelengths larger than 492 nm. Atom A is 4 nm away from the top interface (see Fig. 5 inset), whereas atom B is assumed to be adsorbed to the bottom interface. Atom B has a fixed horizontal displacement of x_{b}=5 nm and therefore there is a fixed separation angle θ_{o} between atom A and atom B with respect to the normal to the interface. The two large peaks seen in Fig. 5 occur when the dispersive resonance angle θ_{R}(λ) is equal to the fixed separation angle, that is, θ_{R}(λ)=θ_{o} in agreement with theory. For the material system shown here, this occurs both in the type I and type II hyperbolic regions. The inset shows the directional sensitivity of the interaction as a function of atom B’s horizontal displacement. It is noteworthy that accurate agreement between the effective medium model and the superlattice structure is achieved when the unitcell size is smaller than the separation distance between atom A and the top interface (Supplementary Note 1).
Figure 5c,d propose a twodimensional van der Waals bonded natural material, hBN, as a candidate material to control optically active vibrational transitions between molecules, or electronic intersubband transitions between quantum wells. hBN is a natural hyperbolic medium in the midinfrared spectral range. We show giant CLSs J_{dd} for the case of two atoms 10 nm away from the top interface of an hBN structure, as well as for two atoms across an hBN film. In the first case, the atom–atom interaction is due to a superCoulombic raylike interaction that reflects from the bottom interface (see insets). In the second case, the interatomic interaction is primarily due to a direct superCoulombic interaction from atom A to atom B. Atom A is 10 nm above the top interface, whereas atom B is assumed to be adsorbed to the bottom interface. It is noteworthy that these longrange DDIs are seen equally in the type I hyperbolic region (λ∼12–13 μm) and in the type II hyperbolic region (λ∼6–7 μm). We have used the experimentally verified permittivities for hBN from Caldwell et al.^{29} for our numerical simulations.
Finally, Fig. 6 proposes a twodimensional material system to enhance RDDIs, using hyperbolic metasurfaces. Our theoretical proposal provides additional future directions for designer metasurfaces based on graphene, black phosphorous, hBN, gold/air or silver/air nanogratings^{39,40,41,42} (see Fig. 6a). We must emphasize that all of the experimental and theoretical studies thus far have focused on Purcell factor enhancements or the photonic spinHall effects. Here we propose hyperbolic metasurfaces to control manybody DDIs. Figure 6a shows the key difference from bulk hyperbolic media where a two dimensional resonance cone mediates giant longrange interactions due to inplane hyperbolic dispersion (x–y plane anisotropy). In Fig. 6b, we show an enhancement of the CLS J_{dd} versus angle θ_{xy} of atom B. The angle θ_{xy} is defined with respect to the optic axis that lies parallel to the interface, such that . A clear enhancement is seen along the resonance angle θ_{R} compared with the vacuum and the dielectric halfspace cases. Furthermore, when the position of atom B lies along the resonance angle we find a clear orderofmagnitude enhancement in the CLS up to distances of 200 nm (Fig. 6b,c). Numerical simulations of the hyperbolic metasurface were done using a dyadic Green function approach (Supplementary Note 2).
To summarize, we have revealed a class of singular excitedstate atom–atom interactions in hyperbolic media that arise from a fundamental modification of the Coulombic nearfield. The experimental observation of such effects will require careful isolation of mediuminduced cooperative interactions between atoms from the effect of independent atoms interacting with the hyperbolic medium. Preliminary results have shown signatures of such interactions between molecules via FRET^{43}. Future work should also focus on understanding the intricate role of nonlocality^{44,45} on DDIs in hyperbolic media. Our work motivates the search for defect centers in natural hyperbolic media such as hBN, where the interaction is mediated by hyperbolic phonon–polaritons. It should also motivate the study of unique manybody physics in atomic lattice quantum metamaterials with hyperbolic response^{46}. Our work also paves the way for studies of longrange entanglement and selforganization^{6}. It is also a first step towards coldatom studies with hyperbolic metasurfaces exhibiting unique effects that are not found in photonic crystals, waveguides or cavities.
Methods
Atomic system
In the following, we only consider the interaction between two identical atoms for the case of RDDIs. We then consider the interaction between two nonidentical atoms for the case of secondorder DDIs such as FRET and the excitedstate Casimir–Polder interaction. For the simulations, we took the transition frequency of atom A to be ω_{a}/2π=500 THz, whereas the transition frequency of atom B was ω_{b}/2π=460 THz.
In the study, we provided equations for the interaction between two twolevel systems for illustrative purposes. The generalized interaction between two Nlevel atoms can be easily extended with the general perturbation results. Furthermore, it is noteworthy that hBN is considered due its highquality factors and its lowloss phonon–polaritonic nature. The superCoulombic effect will occur even in the presence of rapid dispersion in the dielectric constant of hBN as long as the emitter linewidths are not significantly broader than the Reststrahlen bands of hBN where optically active hyperbolic phononpolaritons are found. For our simulations, representative values of loss and dielectric constants have been chosen from recent experiments in the midinfrared spectral range.
Perturbation theory
All DDIs can be calculated from the transition matrix element:
where the summation in the second and higherorder terms runs over all possible intermediate states; the summation can be replaced by integration for the case of continuum states. The energy level shift of the initial state is then given by
where it is understood that the principal value is taken during the integration of continuum intermediate states. The probability transition rate from initial state to final state is given by Fermi’s Golden rule
where the summation runs over all initial and final states.
Resonant dipole–dipole interaction
We now consider the interaction between two identical atoms, labelled atom A and atom B, respectively. The probability transition rate between state and is found through equation (13) to give the CDR
Assuming the dipole moments of both atoms are oriented along the same direction, the total decay rate of two identical atoms will be γ_{tot}=γ_{a}±γ_{dd}, where γ_{a} is the bare spontaneous emission rate of atom A (or atom B). The initial state is and final state . It is noteworthy that represents the singlephoton Fock state with position r and frequency ω.
The firstorder dipole–dipole frequency shift of initial state is then found through evaluation of equation (12), which results in a resonant and offresonant contribution , specified by
and
In the Letter, we only retain the resonant contributions (4) and (5), as they give rise to the superCoulombic DDIs. The results agree with those of ref. 47.
Resonance energy transfer rate
Using equation (3), the resonance energy transfer rate between state and can be calculated to give
where we have taken and final state (ref. 48).
Excited state–ground state interaction
The excitedstate Casimir–Polder potential is given by^{34}
where the resonant component is
and offresonant component is given by
is the isotropic electric polarizability of atom A in the kth energy eigenstate, defined as
Ground state–ground state interaction
The ground state–ground state Casimir–Polder potential is given by^{38}
which is applicable in the retarded and nonretarded regimes. It is noteworthy that we have dropped the spatial coordinate dependence of the Green function in equations (19), (20) and (22).
It is worth noting that for the numerical calculation of the integrals, we considered an anisotropic medium with a Lorentz resonance parallel to the optic axis with parameters ω_{Pz}/2π=550 × 10^{12} Hz, ω_{Tz}/2π=450 × 10^{12} Hz and γ_{z}=0.01ω_{Pz}, and a Lorentz resonance perpendicular to the optic axis with parameters ω_{Px}/2π=770 × 10^{12} Hz, ω_{Tx}/2π=600 × 10^{12} Hz and γ_{z}=0.01ω_{Px}. The isotropic medium had the same relative permittivity as the x axis of the anisotropic medium.
Applicability of perturbation theory
It is noteworthy that the perturbative formalism used in this work is strictly applicable for the case of finite absorption with a sufficiently large interatomic separation distance. This is in agreement with our simulations for practical experimental systems such as plasmonic superlattices and hyperbolic metasurfaces. For the case of low losses and extremely short separation distances, a nonperturbative treatment will be required to treat the dipole–dipole singularity in a selfconsistent manner. It is also noteworthy that the presence of emitters do not alter the hyperbolic polaritonic branches in the weak coupling limit.
Green function in a uniaxial medium
The Green tensor is the unique solution to the homogeneous Helmholtz equation with permittivity tensor ,
and radiation condition G(r, r′; ω)=0 for r−r′→∞. The coordinatefree form of the Green function is given by^{49}
where we have fixed the spatial coordinate of the source at the origin, that is, r′=0. It is noteworthy that this Green function is only applicable when r≠r′, as we have excluded the singularity term that occurs when r=r′.
Data availability
The data that support the findings of this study are available from the corresponding author upon request.
Additional information
How to cite this article: Cortes, C. L. et al. SuperCoulombic atomatom interactions in hyperbolic media. Nat. Commun. 8, 14144 doi: 10.1038/ncomms14144 (2017).
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
 1
Bloch, I. Quantum coherence and entanglement with ultracold atoms in optical lattices. Nature 453, 1016–1022 (2008).
 2
Wang, T. et al. Superradiance in ultracold rydberg gases. Phys. Rev. A 75, 033802 (2007).
 3
de Oliveira, R. A. et al. Singlephoton superradiance in cold atoms. Phys. Rev. A 90, 023848 (2014).
 4
Hettich, C. et al. Nanometer resolution and coherent optical dipole coupling of two individual molecules. Science 298, 385–389 (2002).
 5
Goban, A. et al. Superradiance for atoms trapped along a photonic crystal waveguide. Phys. Rev. Lett. 115, 063601 (2015).
 6
GonzálezTudela, A., Hung, C.L., Chang, D., Cirac, J. & Kimble, H. Subwavelength vacuum lattices and atomatom interactions in twodimensional photonic crystals. Nat. Photonics 9, 320–325 (2015).
 7
Ravets, S. et al. Coherent dipoledipole coupling between two single rydberg atoms at an electricallytuned forster resonance. Nat. Phys. 10, 914–917 (2014).
 8
Van Loo, A. F. et al. Photonmediated interactions between distant artificial atoms. Science 342, 1494–1496 (2013).
 9
Majer, J. et al. Coupling superconducting qubits via a cavity bus. Nature 449, 443–447 (2007).
 10
Andrew, P. & Barnes, W. Energy transfer across a metal film mediated by surface plasmon polaritons. Science 306, 1002–1005 (2004).
 11
Blum, C. et al. Nanophotonic control of the frster resonance energy transfer efficiency. Phys. Rev. Lett. 109, 203601 (2012).
 12
Saffman, M., Walker, T. & MØlmer, K. Quantum information with rydberg atoms. Rev. Modern Phys. 82, 2313 (2010).
 13
Devoret, M. H., Wallraff, A. & Martinis, J. Superconducting qubits: a short review. Preprint at http://arxiv.org/abs/condmat/0411174 (2004).
 14
Fleury, R. & Alù, A. Enhanced superradiance in epsilonnearzero plasmonic channels. Phys. Rev. B 87, 201101 (2013).
 15
Maas, R., Parsons, J., Engheta, N. & Polman, A. Experimental realization of an epsilonnearzero metamaterial at visible wavelengths. Nat. Photonics 7, 907–912 (2013).
 16
MartínCano, D., MartínMoreno, L., GarciaVidal, F. J. & Moreno, E. Resonance energy transfer and superradiance mediated by plasmonic nanowaveguides. Nano Lett. 10, 3129–3134 (2010).
 17
Biehs, S.A. & Agarwal, G. S. Large enhancement of forster resonance energy transfer on graphene platforms. Appl. Phys. Lett. 103, 243112 (2013).
 18
Imamog, A. et al. Quantum information processing using quantum dot spins and cavity qed. Phys. Rev. Lett. 83, 4204 (1999).
 19
Ye, J., Vernooy, D. & Kimble, H. Trapping of single atoms in cavity qed. Phys. Rev. Lett. 83, 4987 (1999).
 20
Zheng, H., Gauthier, D. J. & Baranger, H. U. Waveguide QED: manybody boundstate effects in coherent and fockstate scattering from a twolevel system. Phys. Rev. A 82, 063816 (2010).
 21
Chang, D. E., Sørensen, A. S., Demler, E. A. & Lukin, M. D. A singlephoton transistor using nanoscale surface plasmons. Nat. Phys. 3, 807–812 (2007).
 22
Baba, T. Slow light in photonic crystals. Nat. Photonics 2, 465–473 (2008).
 23
MartínCano, D. et al. Dissipationdriven generation of twoqubit entanglement mediated by plasmonic waveguides. Phys. Rev. B 84, 235306 (2011).
 24
Sokhoyan, R. & Atwater, H. A. Quantum optical properties of a dipole emitter coupled to an nearzero nanoscale waveguide. Opt. Express 21, 32279–32290 (2013).
 25
Milton, G. W., McPhedran, R. C. & Sihvola, A. The searchlight effect in hyperbolic materials. Opt. Express 21, 14926–14942 (2013).
 26
Poddubny, A., Iorsh, I., Belov, P. & Kivshar, Y. Hyperbolic metamaterials. Nat. Photonics 7, 948–957 (2013).
 27
Cortes, C., Newman, W., Molesky, S. & Jacob, Z. Quantum nanophotonics using hyperbolic metamaterials. J. Opt. 14, 063001 (2012).
 28
Krishnamoorthy, H. N., Jacob, Z., Narimanov, E., Kretzschmar, I. & Menon, V. M. Topological transitions in metamaterials. Science 336, 205–209 (2012).
 29
Caldwell, J. D. et al. Subdiffractional volumeconfined polaritons in the natural hyperbolic material hexagonal boron nitride. Nat. Commun. 5, 5221 (2014).
 30
Dai, S. et al. Graphene on hexagonal boron nitride as a tunable hyperbolic metamaterial. Nat. Nanotechnol. 10, 682–686 (2015).
 31
Tran, T. T., Bray, K., Ford, M. J., Toth, M. & Aharonovich, I. Quantum emission from hexagonal boron nitride monolayers. Nat. Nanotechnol. 11, 37–41 (2016).
 32
Wong, D. et al. Characterization and manipulation of individual defects in insulating hexagonal boron nitride using scanning tunnelling microscopy. Nat. Nanotechnol. 10, 949–953 (2015).
 33
Knöll, L., Scheel, S. & Welsch, D.G. Qed in dispersing and absorbing media. Tech. Rep. (2000).
 34
Safari, H. & Karimpour, M. R. Bodyassisted Van der Waals interaction between excited atoms. Phys. Rev. Lett. 114, 013201 (2015).
 35
Tumkur, T. U. et al. Control of forster energy transfer in the vicinity of metallic surfaces and hyperbolic metamaterials. Farad. Discuss. 178, 395–412 (2015).
 36
Poddubny, A. N. Collective forster energy transfer modified by a planar metallic mirror. Phys. Rev. B 92, 155418 (2015).
 37
Potemkin, A. S., Poddubny, A. N., Belov, P. A. & Kivshar, Y. S. Green function for hyperbolic media. Phys. Rev. A 86, 023848 (2012).
 38
Safari, H., Buhmann, S. Y., Welsch, D.G. & Dung, H. T. Bodyassisted van der waals interaction between two atoms. Phys. Rev. A 74, 042101 (2006).
 39
Jacob, Z. & Narimanov, E. E. Optical hyperspace for plasmons: Dyakonov states in metamaterials. Appl. Phys. Lett. 93, 221109 (2008).
 40
High, A. A. et al. Visiblefrequency hyperbolic metasurface. Nature 522, 192–196 (2015).
 41
GomezDiaz, J. S., Tymchenko, M. & Alù, A. Hyperbolic plasmons and topological transitions over uniaxial metasurfaces. Phys. Rev. Lett. 114, 233901 (2015).
 42
Gangaraj, S., Low, T., Nemilentsau, A. & Hanson, G. W. Black phosphorus and twodimensional hyperbolic materials: tunable surface plasmons, green function and complexplane analysis. Preprint at http://arxiv.org/abs/1509.01544 (2015).
 43
Newman, W. D. et al. in Conference on Lasers and ElectroOptics FTh4D.1 (Optical Society of America (2016).
 44
Yan, W., Wubs, M. & Mortensen, N. A. Hyperbolic metamaterials: nonlocal response regularizes broadband supersingularity. Phys. Rev. B 86, 205429 (2012).
 45
CorreasSerrano, D., GomezDiaz, J., Tymchenko, M. & Alù, A. Nonlocal response of hyperbolic metasurfaces. Opt. Express 23, 29434–29448 (2015).
 46
Jha, P. K. et al. Coherencedriven topological transition in quantum metamaterials. Phys. Rev. Lett. 116, 165502 (2016).
 47
Dung, H. T., Knöll, L. & Welsch, D.G. Resonant dipoledipole interaction in the presence of dispersing and absorbing surroundings. Phys. Rev. A 66, 063810 (2002).
 48
Dung, H. T., Knoll, L. & Welsch, D.G. Intermolecular energy transfer in the presence of dispersing and absorbing media. Phys. Rev. A 65, 043813 (2002).
 49
Chen, H. C. Theory of electromagnetic waves a coordinatefree approach McGrawHill Book Company (1983).
Acknowledgements
Research was funded by the National Sciences and Engineering Research Council of Canada (NSERC) as well as the National Science Foundation (NSF).
Author information
Affiliations
Contributions
C.L.C developed the theoretical formalism, performed the analytic calculations and performed the numerical simulations. Both authors contributed to the final version of the manuscript. Z.J supervised the project.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Supplementary information
Supplementary Information
Supplementary Figure 1, Supplementary Table 1, Supplementary Notes 12 and Supplementary References. (PDF 594 kb)
Rights and permissions
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
About this article
Cite this article
Cortes, C., Jacob, Z. SuperCoulombic atom–atom interactions in hyperbolic media. Nat Commun 8, 14144 (2017). https://doi.org/10.1038/ncomms14144
Received:
Accepted:
Published:
Further reading

Abnormal Wave Propagation in Tilted Linear‐Crossing Metamaterials
Advanced Photonics Research (2021)

Collective nearfield coupling and nonlocal phenomena in infraredphononic metasurfaces for nanolight canalization
Nature Communications (2020)

Hyperbolic metamaterials: From dispersion manipulation to applications
Journal of Applied Physics (2020)

Single photon pulse induced transient entanglement force
New Journal of Physics (2020)

CircuitBased Magnetic Hyperbolic Cavities
Physical Review Applied (2020)
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.