Abstract
Ultrafast optical excitation of photocarriers has the potential to transform undoped semiconductor superlattices into semiconductor hyperbolic metamaterials (SHMs). In this paper, we investigate the optical properties associated with such ultrafast topological transitions. We first show reflectance, transmittance, and absorption under TE and TM plane wave incidence. In the unpumped state, the superlattice exhibits a frequency region with high reflectance (>80%) and a region with low reflectance (<1%) for both TE and TM polarizations over a wide range of incidence angles. In contrast, in the photopumped state, the reflectance for both frequencies and polarizations is very low (<1%) for a similar range of angles. Interestingly, this system can function as an all-optical reflection switch on ultrafast timescales. Furthermore, for TM incidence and close to the epsilon-near-zero point of the longitudinal permittivity, directional perfect absorption on ultrafast timescales may also be achieved. Finally, we discuss the onset of negative refraction in the photopumped state.
© 2015 Optical Society of America
1. Introduction
Hyperbolic metamaterials (HMs) [1, 2] provide great flexibility for tailoring the optical properties of surfaces and are utilized in a wide variety of optical devices [3–6]. Furthermore, the high density of states associated with the hyperbolic isofrequency wavevector surface can greatly enhance spontaneous emission [7, 8], enhance near-field thermal energy transfer [9, 10], lead to enhanced absorption processes [11, 12], and generate negative refraction [13–16].
At mid-infrared frequencies, highly doped semiconductors behave like metals. Thus, a highly doped layer can be used as a “metal” layer, and, along with an undoped layer, can form the unit cell of a semiconductor hyperbolic metamaterial (SHM) [13, 17–19]. We have recently shown in Ref [19]. that it might be possible to use ultrafast optical excitation to generate high densities of photocarriers in nominally undoped quantum wells [20, 21] of semiconductor superlattices. With sufficient pumping strength, the wells will become metallic in the frequency range of interest, and an ultrafast topological transition to the SHM state will occur. This might in turn give rise to ultrafast appearance of the very large density-of-states signatures and enable optical gating and modulation for many applications, such as imaging, lensing, near-field energy transfer, negative refraction, and all-optical switching.
In this paper, we investigate the optical properties of finite-thickness semiconductor superlattices under transient excitation. As a first step, we consider uniform distribution of pump light within the semiconductor superlattice to understand the attainable optical properties; more refined models will be analyzed in the future. However, to a first approximation, a uniform excitation can be obtained by choosing the pump wavelength such that the absorption per quantum well is small. We observe that the combination of filling factors, superlattice thickness, and materials allows for the presence of impedance matching conditions, Fabry-Perot resonances, and epsilon-near-zero conditions that give rise to reflectionless features under plane wave illumination. Remarkably, these conditions can be exploited to obtain ultrafast, wide-angle, all-optical switches and directional perfect absorbers. We also show that we can obtain negative refraction on ultrafast timescales.
2. Modeling of finite-thickness superlattices under plane wave incidence
Consider the experimentally achievable [22], finite-thickness semiconductor superlattice made of alternating InAs and undoped GaSb layers (Fig. 1). For sufficiently thin layers, compared to the operating wavelength, such a structure can be homogenized as a uniaxial material with a uniaxial permittivity tensor of the kind , where is the transverse dielectric constant along the transverse direction (parallel to the layers) and is the longitudinal dielectric constant along the longitudinal direction (perpendicular to the layers). In these expressions, and are the permittivity and thickness of the InAs layers, and and are the permittivity and thickness of the GaSb layers. Although this slab could be simply modeled considering the superlattice implementation, the homogeneous interpretation provides a better insight on the physical origin of the observed optical properties and may thus lead to device design guidelines.
For the case of lightly doped InAs wells (Fig. 1(a)), is positive in the frequency range of interest and the homogenized superlattice acts as an anisotropic dielectric (Fig. 1(b)). If intense optical excitation with above bandgap photons creates a sufficiently large population of carriers (~2x1019 cm−3) in the InAs quantum wells (Fig. 1(c)), becomes negative and the homogenized superlattice acts as a hyperbolic material (Fig. 1(d)). After initial pumping, the superlattice will remain in the hyperbolic state for a brief period (~1 ns) before relaxation processes deplete the photocarriers and the system returns to the anisotropic dielectric state. This represents the basic working principle that might enable interesting optical properties on ultrafast timescales.
Let us now explicitly analyze the structures shown in Fig. 1(a, c) via a transmission line formalism [23, 24] in which each layer can be modeled using the ABCD matrix that relates the input electric and magnetic fields to the output electric and magnetic fields as
where d is the layer thickness, and is the longitudinal wavenumber in the layer with the transverse wavenumber, the free space wavenumber, the angular frequency and the speed of light. Note that assumes different values in the different layers (InAs, Gasb, and homogenized) as their relative permittivities are different. The layer transmission line impedance is given by for TM polarization and for TE polarization, where is the relative permittivity of the layer, and and are the absolute permittivity and permeability of free space. The monochromatic time harmonic convention, , is implicitly assumed.The total ABCD parameters , , , and for the finite-thickness superlattice are given by the matrix multiplication
where N indicates the number of layers composing the semiconductor superlattice as shown in Fig. 1(a) and 1(c).The reflection coefficient is evaluated as
with for TM polarization and for TE polarization. Also, , where and for TM and TE polarizations, respectively.We note that the reflectance of the homogenized structures can also be obtained using the transmission line formalism, however in this case , , , with and is the impedance of the homogenized superlattice, being for TM polarization, and and for TE polarization.
3. Reflectionless features on ultrafast timescales
We show in this section that the reflection and transmission properties obtained using the explicit superlattice and the homogenized slab are in excellent agreement. We analyze both the TE and TM properties of the superlattice. Although in the literature the TE polarization is seldom considered for SHMs as it does not couple to the hyperbolic branch, we find it is important for the purpose of gathering a complete understanding of the optical properties of transiently excited superlattices. We consider in this section a superlattice with total thickness t = 600 nm (10 pairs). The unbounded media are considered to have a relative permittivity equal to unity (i.e. ). [Note that results similar to those shown in this paper can be achieved by using a bottom space with , e.g. a barium fluorite substrate, making the results here discussed experimentally achievable.]
Before presenting the transmission and reflection properties, we describe two important impedances that dictate the reflection behavior of the SHM stack. The first is the “material impedance”, defined by the properties of the homogenized superlattice as for TM polarization and for TE polarization. Note that is a bulk material property of the effective medium and does not depend on the overall thickness. The slab will appear reflectionless when is matched to both the upper and bottom impedances. The second impedance is the “slab impedance”, which depends on both superlattice and bottom space impedance as . This impedance depends on all of the parameters of the superlattice, including the layer thickness and number of layers. All the parameters were defined in Sec. 2 for both TE and TM polarizations.
The reflectance and transmittance under TE plane wave incidence for the structures of Fig. 1(a) and Fig. 1(c), evaluated using the effective medium approximation, are shown in Fig. 2. For the unpumped case, one can notice that around 40 THz, and for wide range of incidence angles, the reflectance (transmittance) is fairly high (low), meaning that most of the incoming signal will be reflected. In contrast, for the photopumped case, a reflectionless, mainly transmissive feature appears on ultrafast timescales around 40 THz. The origin of this latter feature is due to a material impedance match with the effective transverse permittivity [25], i.e. , and can be tuned by changing the materials and filling fractions of the superlattice. Interestingly, the feature at 40 THz also corresponds to the m = 0 Fabry-Perot mode of the SHM stack. This correspondence arises only because the slab is very thin.
We also notice for both unpumped and photopumped cases an angle-independent reflectionless feature around 70 and 80 THz, respectively. This is due to an m = 1 Fabry-Perot resonance of the SHM stack that can be tuned with thickness. To show the correspondences with Fabry-Perot resonances, we compute the angular dependence of the m = 0, 1 resonances using . The results are shown as dashed lines in Fig. 2. We observe that the dispersions of these modes are rather independent of the incidence angle. While for the feature at 40 THz this is because , for the m = 1 resonance this can be justified by rewriting the dispersion relation for TE waves below Eq. (3) as
Thus we see that if the effective transverse permittivity is large in comparison to the term , the term dependence in Eq. (4) can be neglected, leading to the angle-independent properties of the features shown in Fig. 2. This is indeed the case, as and at ~70 THz for the unpumped case and at ~80 THz for the photopumped case.
The reflectance and transmittance under TM plane wave incidence for the structures of Fig. 1(a) and 1(c), evaluated using the effective medium approximation, are shown in Fig. 3. Features similar to those observed in Fig. 2 are observed for both unpumped and photopumped cases. The reflectionless feature at 40 THz for the photopumped case is due to a slab impedance match, i.e. [25], and can be tuned by changing the materials and filling fractions of the superlattice. Interestingly, the feature at 40 THz also corresponds to the m = 0 Fabry-Perot mode of the SHM stack, which ends at the material impedance match with the effective transverse permittivity for , i.e. . The features around 70 and 80 THz for unpumped and photopumped cases are due to the m = 1 Fabry-Perot resonance of the SHM stack. We thus overlap on the reflectance maps of this figure the dispersion of the Fabry-Perot resonances retrieved with m = 0, 1. We observe that the dispersion of the m = 1 mode is rather independent of angle of incidence. This can be justified in a manner similar to the TE case.
To demonstrate the excellent agreement between the superlattice and effective medium models, we show in Fig. 4 the reflectance of the photopumped state for both TE and TM plane wave incidence calculated using both models. To avoid unnecessary duplication, we will utilize the effective medium model for the remainder of this paper.
It is interesting to note that the transient optical properties presented above can be used to demonstrate an all-optical switch. We show in Fig. 5 the TE and TM reflectances for the “OFF” (photopumped) and “ON” (unpumped) states of the reflection mode switch for an incidence angle of 20 degrees. These curves correspond to horizontal cuts of the maps in Fig. 2 and Fig. 3. It is remarkable that near 40 THz, and for both polarizations, the ON state exhibits a reflectance of about 80% while the OFF state is almost 0%. We can define a modulation parameter given by , , indicated by an arrow in Fig. 5 around 40 THz, to describe the quality of the switching device. For TE polarization, ; for TM polarization, , indicating a good switching performance for these semiconductor superlattices. However, we note that such performance could also be obtained through appropriate photopumping of a single semiconductor layer as has been demonstrated for bulk semiconductor switches [26–29].
4. Directional perfect absorption on ultrafast timescales
In Sec. 3, we have observed the presence of a zero transmittance feature for TM incidence in the photopumped state. This feature corresponds to an epsilon-near-zero condition and may lead to directional perfect absorption [30, 31]. We thus show in Fig. 6 the absorption under TM plane wave incidence for the structures of Fig. 1(a) and Fig. 1(c), evaluated using the effective medium approximation. It is peculiar that near 70 THz the nearly zero absorption of the unpumped state becomes almost unit absorption for the photopumped state. This ultrafast directional perfect absorption is due to the occurrence of an epsilon-near-zero transition of in the photopumped state.
5. Negative refraction on ultrafast timescales
We analyze in this section the possibility of obtaining negative refraction on ultrafast timescales. We consider two cases: (i) the interface between a medium with unit permittivity and an infinitely extended superlattice; and (ii) a superlattice with total thickness t = 6 µm (100 pairs). The unbounded media are considered to have a relative permittivity equal to unity in case (ii), i.e. . The dispersion diagrams of the superlattice for unpumped and photopumped conditions at 68 THz are reported in Fig. 7(a) and Fig. 7(b), respectively. One can notice a transition from elliptic to hyperbolic dispersion upon carrier excitation. In the hyperbolic state, the group velocity will now point inward, allowing for negative refraction, as explicitly indicated in Fig. 7(b).
To show such a feature, we use full-wave simulation [32] to analyze the refraction of a Gaussian beam in case (i). The unpumped condition is shown in Fig. 7(c), where it is apparent that phase and group velocities are along the same direction, a signature of a standard positive refraction. On the contrary, the photopumped condition of Fig. 7(d) nicely shows that phase and group velocities are in opposite directions, a signature of a negative refraction. Thus, using this approach, negative refraction can now be accessed on ultrafast timescales. Similar results can be achieved also for the situation of case (ii) as shown in Fig. 7(e) and Fig. 7(f) for unpumped and photopumped conditions. Notice the transition between positive and negative refraction within the slab.
6. Conclusion
In conclusion, we have analyzed the optical properties of transiently-excited semiconductor superlattices. Under ultrafast optical excitation of photocarriers, we can optically induce electrons to populate the quantum wells [20, 21] of the superlattice, thereby transforming undoped semiconductor superlattices into SHMs and allowing the rise of interesting properties on ultrafast timescales. These intriguing results suggest that transient excitation of SHMs might enable the ultrafast appearance of the very large density-of-states that are characteristic of hyperbolic metamaterials. This will provide a route to transiently enable phenomena relying on the hyperbolic isofrequency surfaces such as imaging, lensing, near-field energy transfer, and negative refraction. Finally, we stress that experimental verification of such a photopumping scheme will be extremely challenging, but should be achievable.
Acknowledgments
This work was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering and performed, in part, at the Center for Integrated Nanotechnologies, an Office of Science User Facility operated for the U.S. Department of Energy (DOE) Office of Science. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.
References and links
1. H. N. S. Krishnamoorthy, Z. Jacob, E. Narimanov, I. Kretzschmar, and V. M. Menon, “Topological transitions in metamaterials,” Science 336(6078), 205–209 (2012). [CrossRef] [PubMed]
2. A. Poddubny, I. Iorsh, P. Belov, and Y. Kivshar, “Hyperbolic metamaterials,” Nat. Photonics 7(12), 948–957 (2013). [CrossRef]
3. Y. Guo, W. Newman, C. L. Cortes, and Z. Jacob, “Applications of hyperbolic metamaterial substrates,” Adv. OptoElectronics 2012, 9 (2012).
4. L. Cheng, L. Wei, J. Xunya, and C. Juncheng, “Far-field super-resolution imaging with a planar hyperbolic metamaterial lens,” Europhys. Lett. 105(2), 28003 (2014). [CrossRef]
5. J. Bénédicto, E. Centeno, and A. Moreau, “Lens equation for flat lenses made with hyperbolic metamaterials,” Opt. Lett. 37(22), 4786–4788 (2012). [CrossRef] [PubMed]
6. D. Lu and Z. Liu, “Hyperlenses and metalenses for far-field super-resolution imaging,” Nat. Commun. 3, 1205 (2012). [CrossRef] [PubMed]
7. X. Ni, G. V. Naik, A. V. Kildishev, Y. Barnakov, A. Boltasseva, and V. M. Shalaev, “Effect of metallic and hyperbolic metamaterial surfaces on electric and magnetic dipole emission transitions,” Appl. Phys. B 103(3), 553–558 (2011). [CrossRef]
8. L. Gu, T. U. Tumkur, G. Zhu, and M. A. Noginov, “Blue shift of spontaneous emission in hyperbolic metamaterial,” Sci. Rep. 4, 4969 (2014). [CrossRef] [PubMed]
9. Y. Guo and Z. Jacob, “Thermal hyperbolic metamaterials,” Opt. Express 21(12), 15014–15019 (2013). [CrossRef] [PubMed]
10. S. A. Biehs, M. Tschikin, and P. Ben-Abdallah, “Hyperbolic Metamaterials as an analog of a blackbody in the near field,” Phys. Rev. Lett. 109(10), 104301 (2012). [CrossRef] [PubMed]
11. C. Guclu, S. Campione, and F. Capolino, “Hyperbolic metamaterial as super absorber for scattered fields generated at its surface,” Phys. Rev. B 86(20), 205130 (2012). [CrossRef]
12. D. Ji, H. Song, X. Zeng, H. Hu, K. Liu, N. Zhang, and Q. Gan, “Broadband absorption engineering of hyperbolic metafilm patterns,” Sci. Rep. 4, 4498 (2014). [CrossRef] [PubMed]
13. A. J. Hoffman, L. Alekseyev, S. S. Howard, K. J. Franz, D. Wasserman, V. A. Podolskiy, E. E. Narimanov, D. L. Sivco, and C. Gmachl, “Negative refraction in semiconductor metamaterials,” Nat. Mater. 6(12), 946–950 (2007). [CrossRef] [PubMed]
14. A. Fang, T. Koschny, and C. M. Soukoulis, “Optical anisotropic metamaterials: Negative refraction and focusing,” Phys. Rev. B 79(24), 245127 (2009). [CrossRef]
15. C. Argyropoulos, N. M. Estakhri, F. Monticone, and A. Alù, “Negative refraction, gain and nonlinear effects in hyperbolic metamaterials,” Opt. Express 21(12), 15037–15047 (2013). [CrossRef] [PubMed]
16. K. V. Sreekanth, A. De Luca, and G. Strangi, “Negative refraction in graphene-based hyperbolic metamaterials,” Appl. Phys. Lett. 103(2), 023107 (2013). [CrossRef]
17. A. J. Hoffman, A. Sridhar, P. X. Braun, L. Alekseyev, S. S. Howard, K. J. Franz, L. Cheng, F.-S. Choa, D. L. Sivco, V. A. Podolskiy, E. E. Narimanov, and C. Gmachl, “Midinfrared semiconductor optical metamaterials,” J. Appl. Phys. 105(12), 122411 (2009). [CrossRef]
18. P. Shekhar and Z. Jacob, “Strong coupling in hyperbolic metamaterials,” Phys. Rev. B 90(4), 045313 (2014). [CrossRef]
19. S. Campione, S. Liu, T. S. Luk, and M. B. Sinclair, “Realizing high-quality, ultra-large momentum states and ultrafast topological transitions using semiconductor hyperbolic metamaterials,” J. Opt. Soc. Am. B 32(9), 1809–1815 (2015). [CrossRef]
20. W. S. Pelouch, R. J. Ellingson, P. E. Powers, C. L. Tang, D. M. Szmyd, and A. J. Nozik, “Comparison of hot-carrier relaxation in quantum wells and bulk GaAs at high carrier densities,” Phys. Rev. B Condens. Matter 45(3), 1450–1453 (1992). [CrossRef] [PubMed]
21. S. Lutgen, R. A. Kaindl, M. Woerner, T. Elsaesser, A. Hase, H. Künzel, M. Gulia, D. Meglio, and P. Lugli, “Nonequilibrium dynamics in a quasi-two-dimensional electron plasma after ultrafast intersubband excitation,” Phys. Rev. Lett. 77(17), 3657–3660 (1996). [CrossRef] [PubMed]
22. M. Razeghi, Y. Wei, A. Hood, D. Hoffman, B. M. Nguyen, P. Y. Delaunay, E. Michel, and R. McClintock, “Type-II superlattice photodetectors for MWIR to VLWIR focal plane arrays,” Proc. SPIE 6206, 62060N (2006). [CrossRef]
23. D. M. Pozar, Microwave Engineering (John Wiley and Sons, 2011).
24. P. Yeh, Optical Waves in Layered Media (Wiley, 2005).
25. N. Mattiucci, M. J. Bloemer, N. Aközbek, and G. D’Aguanno, “Impedance matched thin metamaterials make metals absorbing,” Sci. Rep. 3, 3203 (2013). [CrossRef] [PubMed]
26. H. S. Loka and P. W. E. Smith, “Ultrafast all-optical switching in an asymmetric Fabry-Perot device using low-temperature-grown GaAs,” IEEE Photonics Technol. Lett. 10(12), 1733–1735 (1998). [CrossRef]
27. L. Qian, P. W. E. Smith, M. A. Matin, and B. J. Robinson, “Ultrafast all-optical asymmetric Fabry-Perot switch based on bulk beryllium-doped InGaAsP grown by He-plasma-assisted epitaxy,” in Conference on Lasers and Electro-Optics,380–381 (2000). [CrossRef]
28. A. J. Alcock, P. B. Corkum, and D. J. James, “A fast scalable switching technique for high‐power CO2 laser radiation,” Appl. Phys. Lett. 27(12), 680–682 (1975). [CrossRef]
29. P. B. Corkum and D. Keith, “Controlled switching of 10-micrometer radiation using semiconductor étalons,” J. Opt. Soc. Am. B 2(12), 1873–1879 (1985). [CrossRef]
30. T. S. Luk, S. Campione, I. Kim, S. Feng, Y. C. Jun, S. Liu, J. B. Wright, I. Brener, P. B. Catrysse, S. Fan, and M. B. Sinclair, “Directional perfect absorption using deep subwavelength low permittivity films,” Phys. Rev. B 90(8), 085411 (2014). [CrossRef]
31. K. Halterman and J. M. Elson, “Near-perfect absorption in epsilon-near-zero structures with hyperbolic dispersion,” Opt. Express 22(6), 7337–7348 (2014). [CrossRef] [PubMed]
32. FDTD Solutions by FDTD Lumerical Inc, https://www.lumerical.com/.