Photonic doping beyond two dimensions: exact additive dopant rules, loss and dispersion limits, and dense-doping regimes in epsilon-near-zero media

Photonic doping establishes that, in a two-dimensional transverse-magnetic scattering configuration, arbitrarily placed nonmagnetic dielectric inclusions (dopants) inside an epsilon-near-zero (ENZ) host endow it with a position-independent, additive effective permeability $\mu_{\mathrm{eff}} = 1 + \sum_d \Delta\mu_d$, exactly reproducing all external fields while the effective permittivity stays near zero. The open question is whether the defining law of photonic doping — the exact, geometry- and position-independent additive dopant rule — extends to generic fully three-dimensional ENZ media, and how it is quantitatively modified by realistic material losses, strong dispersion, and high dopant filling fractions. No established three-dimensional additive rule is known in the literature; three-dimensional zero-index systems instead exhibit impurity-immunity or connectivity-governed rather than additive behavior, and the three-dimensional rigorous result available (spectral perturbation theory for a high-contrast Maxwell core–shell resonator, 2026) both identifies geometry-invariant regimes and constructs geometry-sensitive counterexamples. The three-dimensional dopant law, its loss/dispersion corrections, and the dense-filling collective limit remain open, while the optical-frequency practical realization has recently been achieved in an all-dielectric ENZ platform.

Background

An epsilon-near-zero (ENZ) medium is a material whose relative permittivity $\varepsilon$ approaches zero at a design frequency, so that the effective wavelength inside it is stretched and the fields become spatially quasi-static. In 2017, Liberal, Mahmoud, Li, Edwards and Engheta showed that such media admit a doping mechanism directly analogous to semiconductor doping: a host ENZ body containing a small number of nonmagnetic inclusions ("photonic dopants") behaves, to any external observer, exactly like a homogeneous body whose effective permeability is shifted by an additive, dopant-specific amount $\Delta\mu_d$, while its effective permittivity remains near zero (Liberal et al., Science 2017). For a two-dimensional scattering configuration with transverse-magnetic (TM) polarization — meaning the magnetic field is perpendicular to the plane of the cross-section, so that time-harmonic Maxwell equations reduce to a single scalar 2D Helmholtz equation for the out-of-plane magnetic field — the rule is exact: the external near and far fields, including those of nearby sources, are reproduced by $\mu_{\mathrm{eff}} = 1 + \sum_d \Delta\mu_d$ independent of the dopants' positions and mutual interactions. For a circular dielectric rod of radius $r_d$ and permittivity $\varepsilon_d$ in a host of area $A$, the contribution is $\Delta\mu_d = \frac{A_d}{A}\left\{\frac{2}{k_d r_d}\frac{J_1(k_d r_d)}{J_0(k_d r_d)} - 1\right\}$, with $A_d = \pi r_d^2$, $k_d = \omega\sqrt{\varepsilon_d}/c$, and $J_n$ the cylindrical Bessel functions. Tuning $\Delta\mu_d$ to make $\mu_{\mathrm{eff}}\to 0$ yields epsilon-and-mu-near-zero (EMNZ) behavior, and $\mu_{\mathrm{eff}}\to\infty$ realizes a perfect magnetic conductor (PMC), both without any intrinsic magnetic material; the same mechanism underlies geometry-invariant "supercoupling", the tunneling of waves through arbitrarily narrow and sharply bent ENZ channels with near-zero phase delay, demonstrated experimentally at microwave frequencies in the 2017 work. The concept seeded a large applications literature: reconfigurable ENZ metasurfaces (Liberal, Li and Engheta, Nanophotonics 2018), substrate-integrated printed-circuit implementations (Zhou et al., Nature Communications 2019), non-Hermitian doping with gain and loss (Coppolaro et al., PNAS 2020), and multi-dopant "dispersion coding" in which sparse dopants act as independent spectral bits (Zhou et al., Light: Science & Applications 2022). A parallel rigorous-mathematics line put the 2017 heuristic on firm ground: Kohn and Venkatraman proved that, for the 2D TM configuration, the scattered field depends complex-analytically on the ENZ parameter $\delta$ (the small, generally complex-valued host permittivity), which explains why photonic doping is only mildly affected by material loss and persists when $\varepsilon$ is merely small rather than zero, and they characterized the leading-order electric field that the physics literature had left undetermined (Communications on Pure and Applied Mathematics, 2023; see also their analysis of 2D ENZ resonators, Archive for Rational Mechanics and Analysis, 2024). Two facts delimit what is known. First, all exact treatments — heuristic and rigorous — are two-dimensional (TM, cylindrical, or quasi-2D waveguide settings with field uniformity along one axis); the effective-permeability dopant law has never been established for generic fully three-dimensional hosts and dopants. Second, three-dimensional zero-index systems behave differently: a 3D photonic-crystal "void space" with simultaneous zero permittivity and permeability exhibits impurity-immunity rather than the doping effect of its 2D counterpart (Xu et al., Physical Review Letters 2021), and coherent perfect absorption in 3D zero-index media is governed by the long-range connectivity of defects rather than by an additive dopant rule (Dongyang Yan et al., Nanophotonics 2023). The most recent rigorous three-dimensional result — analytic spectral perturbation theory for a high-contrast dielectric core–shell structure in a perfectly conducting cavity (Kohn and Venkatraman, 2026) — proves that leading-order resonance asymptotics are independent of the ENZ shell's geometry under identifiable conditions but also constructs shell-geometry-sensitive counterexamples, and it addresses resonances of core–shell cavities rather than the position-independent additive external-field equivalence that defines photonic doping. On the practical side, photonic doping remained confined to microwave and far-infrared platforms for a decade because optical ENZ materials are intrinsically lossy; only in 2026 was optical-frequency photonic doping demonstrated, using dielectric Mie-resonator nanoparticles embedded in an ultra-low-loss all-dielectric ENZ Bragg-microcavity platform rather than a lossy plasmonic ENZ host (Panahpour, Kelavuori and Huttunen, Nanophotonics 2026).

Problem Statement

The problem is to determine the law of photonic doping outside the two-dimensional, effectively lossless setting in which it was established, with one overarching objective: an exact effective-medium description of a doped epsilon-near-zero (ENZ) body that is valid for generic three-dimensional geometries, realistic materials, and dense dopant configurations. Concretely, the originating work (Liberal et al., Science 2017) and its rigorous confirmation (Kohn and Venkatraman, Comm. Pure Appl. Math. 2023) establish, for 2D TM scattering with out-of-plane magnetic polarization, that arbitrary nonmagnetic dopants imprint a position-independent, mutually independent, additive effective permeability $\mu_{\mathrm{eff}} = 1 + \sum_d \Delta\mu_d$ on an ENZ host of area $A$, exactly reproducing all external fields with $\varepsilon_{\mathrm{eff}} \approx 0$. The open questions are: (1) Dimensional generalization — does an analogous exact additive dopant rule hold in generic fully three-dimensional ENZ media (beyond quasi-2D cases with field uniformity along one axis), where the vector Maxwell system admits no scalar uniform-field limit; if additivity or position-independence fails, what is the correct replacement law, and under what precise hypotheses on host, dopant, and frequency does the external observer remain unable to infer the dopants' positions? Refuting additive or position-independent dopant behavior and determining the correct replacement law are distinct sub-questions: the replacement law and its precise hypotheses do not follow from a refutation of additivity. (2) Loss and dispersion limits — a quantitative characterization, valid beyond the moderate-loss regime numerically sampled in the original work, of how absorption, strong dispersion, and realistic frequency-dependent material responses modify $\Delta\mu_d$, the dopant resonance positions, and the robustness of geometry-invariant EMNZ tunneling in three-dimensional hosts. (3) Dense-doping limit — the collective effective response and its emergent phenomena (including the asymptotic approach to EMNZ transparency at high dopant filling fraction) when dopant filling factors become large, where the sparse-regime independence demonstrated for multiple dopants no longer applies. These three components are jointly supported by the single objective of completing the effective-medium law of doped ENZ media: component (1) supplies the geometrical scope of the law, component (2) its material validity range, and component (3) its configurational range, and together they determine the regime in which a designer could predict, exactly and without geometry-specific simulation, the external response of a 3D, lossy, arbitrarily doped ENZ body.

References

1. Iñigo Liberal, Ahmed M. Mahmoud, Yue Li, Brian Edwards, Nader Engheta (2017). Photonic doping of epsilon-near-zero media. Science 355(6329):1058-1062. DOI 10.1126/science.aal2672. https://doi.org/10.1126/science.aal2672
2. Iñigo Liberal, Yue Li, Nader Engheta (2018). Reconfigurable epsilon-near-zero metasurfaces via photonic doping. Nanophotonics 7(6):1117-1127. DOI 10.1515/nanoph-2018-0012. https://doi.org/10.1515/nanoph-2018-0012
3. Ziheng Zhou, Yue Li, Hao Li, Wangyu Sun, Iñigo Liberal, Nader Engheta (2019). Substrate-integrated photonic doping for near-zero-index devices. Nature Communications 10:4132. DOI 10.1038/s41467-019-12083-y. https://doi.org/10.1038/s41467-019-12083-y
4. Marino Coppolaro, Massimo Moccia, Giuseppe Castaldi, Nader Engheta, Vincenzo Galdi (2020). Non-Hermitian doping of epsilon-near-zero media. Proceedings of the National Academy of Sciences 117(25):13921-13928. DOI 10.1073/pnas.2001125117. https://doi.org/10.1073/pnas.2001125117
5. Michaël Lobet, Iñigo Liberal, Erik N. Knall, M. Zahirul Alam, Orad Reshef, Robert W. Boyd, Nader Engheta, Eric Mazur (2020). Fundamental Radiative Processes in Near-Zero-Index Media of Various Dimensionalities. ACS Photonics 7(8):1965-1970. DOI 10.1021/acsphotonics.0c00782. https://doi.org/10.1021/acsphotonics.0c00782
6. Changqing Xu, Hongchen Chu, Jie Luo, Zhi Hong Hang, Ying Wu, Yun Lai (2021). Three-Dimensional Electromagnetic Void Space. Physical Review Letters 127(12):123902. DOI 10.1103/PhysRevLett.127.123902. https://doi.org/10.1103/PhysRevLett.127.123902 (arXiv:2012.13547, https://arxiv.org/abs/2012.13547)
7. Robert V. Kohn, Raghavendra Venkatraman (2023). Complex analytic dependence on the dielectric permittivity in ENZ materials: The photonic doping example. Communications on Pure and Applied Mathematics 77(2):1278-1301. DOI 10.1002/cpa.22138. https://doi.org/10.1002/cpa.22138 (arXiv:2203.08911, https://arxiv.org/abs/2203.08911)
8. Ziheng Zhou, Hao Li, Wangyu Sun, Yijing He, Iñigo Liberal, Nader Engheta, Zhenghe Feng, Yue Li (2022). Dispersion coding of ENZ media via multiple photonic dopants. Light: Science & Applications 11:207. DOI 10.1038/s41377-022-00892-8. https://doi.org/10.1038/s41377-022-00892-8
9. Robert V. Kohn, Raghavendra Venkatraman (2024). Transverse Magnetic ENZ Resonators: Robustness and Optimal Shape Design. Archive for Rational Mechanics and Analysis 248(6):110. DOI 10.1007/s00205-024-02023-6. https://doi.org/10.1007/s00205-024-02023-6 (arXiv:2403.11242, https://arxiv.org/abs/2403.11242)
10. Wendi Yan, Ziheng Zhou, Hao Li, Yue Li (2023). Transmission-type photonic doping for high-efficiency epsilon-near-zero supercoupling. Nature Communications 14:6154. DOI 10.1038/s41467-023-41965-5. https://doi.org/10.1038/s41467-023-41965-5
11. Wendi Yan, Ziheng Zhou, Hao Li, Wangyu Sun, Qihao Lv, Yue Li (2023). Low-Loss Epsilon-Near-Zero Metamaterials. Laser & Photonics Reviews 17(8):2201000. DOI 10.1002/lpor.202201000. https://doi.org/10.1002/lpor.202201000
12. Wendi Yan, Hao Li, Xu Qin, Peihang Li, Pengyu Fu, Kaifeng Li, Yue Li (2024). Fano Resonance in Epsilon-Near-Zero Media. Physical Review Letters 133(25):256402. DOI 10.1103/PhysRevLett.133.256402. https://doi.org/10.1103/PhysRevLett.133.256402
13. Robert V. Kohn, Raghavendra Venkatraman (2026). Analytic spectral perturbation theory for a high-contrast Maxwell operator. arXiv:2601.13408. https://arxiv.org/abs/2601.13408
14. Ali Panahpour, Jussi Kelavuori, Mikko Huttunen (2026). Photonic Doping of Epsilon-Near-Zero Bragg Microcavities. Nanophotonics 15(13):e70186. DOI 10.1002/nap2.70186. https://doi.org/10.1002/nap2.70186 (arXiv:2511.01658, https://arxiv.org/abs/2511.01658)
15. Lin Jiang, Shufang Dong, Kui Tang, Shaojie Wang, Ke Chen, Junming Zhao, Yijun Feng (2026). Realization of significant magnetic-field enhancement in a photonic-doped epsilon-near-zero medium. Optics Express 34(20):38042. DOI 10.1364/oe.612828. https://doi.org/10.1364/oe.612828
16. Zhicheng Xiong, Fei Sun, Weiqi Yuan, Yichao Liu, Shuai Zhang, Zhihui Chen, Mingda Zhang (2026). Topological-Charge-Enabled Photonic Doping in ENZ Media. arXiv:2607.16751. https://arxiv.org/abs/2607.16751
17. Dongyang Yan, Ran Mei, Mingyan Li, Zhikai Ma, Zhi Hong Hang, Jie Luo (2023). Controlling coherent perfect absorption via long-range connectivity of defects in three-dimensional zero-index media. Nanophotonics 12(22):4195-4204. DOI 10.1515/nanoph-2023-0485. https://doi.org/10.1515/nanoph-2023-0485
18. Alessandra Contestabile, Angelo Galante, Giuseppe Castaldi, Angela Capocefalo, Marcello Alecci, Vincenzo Galdi, Carlo Rizza (2026). Mietronic magnetism for deep-subwavelength wave manipulation. Newton, article 100619. DOI 10.1016/j.newton.2026.100619. https://doi.org/10.1016/j.newton.2026.100619