Thermally sensitive scattering of terahertz waves by coated cylinders for tunable invisibility and masking

Temperature-sensitive scattering of terahertz (THz) waves by infinitely long, cylindrical core-shell structures was theoretically studied. Each structure is a dielectric cylinder coated with an InSb shell illuminated by either a transverse-electric (TE) or a transverse-magnetic (TM) plane wave. InSb is a thermally tunable semiconductor showing a transition from dielectric to plasmonic state at THz frequencies. Accordingly, the total scattering efficiency (TSE) can be thermally tuned for both polarization states of the incident plane wave. The spectral locations of the maxima and minima of the TSE of an InSb-coated cylinder can be exploited for cloaking the core. At least three scenarios lead to the strong suppression of scattering by a single core-shell structure in different spectral regimes when the temperature is fixed. The excitation of localized surface-plasmon resonances is the feature being common for two of them, while the effect of volumetric resonance dominates in the third scenario. Regimes that are either highly or weakly sensitive to the core material were identified. Weak sensitivity enables masking, i.e., the core material cannot be identified by a far-zone observer. The TSE minima are usually significantly sensitive to the polarization state, but ones with weak sensitivity to the polarization state also exist. © 2018 Optical Society of America OCIS codes: (290.5839) Scattering, invisibility; (160.6840) Thermo-optical materials; (240.6680) Surface plasmons. References and links 1. D. Schurig, J. J. Mock, B. J. Justice, S. A. Cummer, J. B. Pendry, A. F. Starr, and D. R. Smith, “Metamaterial electromagnetic cloak at microwave frequencies,” Science 314(5801), 977–980 (2006). 2. N.-A. Nicorovici, G. W. Milton, R. C. McPhedran, and L. C. Botten, “Quasistatic cloaking of two-dimensional polarizable discrete systems by anomalous resonance,” Opt. Express 15(10), 6314–6323 (2007). 3. A. Alù and N. Engheta, “Plasmonic materials in transparency and cloaking problems: mechanism, robustness, and physical insights,” Opt. Express 15(6), 3318–3332 (2007). 4. A. E. Serebryannikov, P. V. Usik, and E. Ozbay, “Non-ideal cloaking based on Fabry-Perot resonances in singlelayer high-index,” Opt. Express 17(19), 16869–16876 (2009). 5. G. Castaldi, I. Gallina, V. Galdi, A. Alù, and N. Engheta, “Power scattering and absorption mediated by cloak/anti-cloak interactions: a transformation-optics route toward invisible sensors,” J. Opt. Soc. Am. B 27(10), 2132–2140 (2010). 6. J. McPhillips, A. Murphy, M. P. Jonsson, W. R. Hendren, R. Atkinson, F. Höök, A. V. Zayats, and R. J. Pollard, “High-performance biosensing using arrays of plasmonic nanotubes,” ACS Nano 4(4), 2210–2216 (2010). 7. Z. Ruan and S. Fan, “Superscattering of light from subwavelength nanostructures,” Phys. Rev. Lett. 105(1), 013901 (2010). 8. G. X. Li, H. L. Tam, F. Y. Wang, and K. W. Cheah, “Half-cylindrical far field superlens with coupled Fabry– Perot cavities,” J. Appl. Phys. 104(9), 096103 (2008). 9. A. Salandrino and N. Engheta, “Far-field subdiffraction optical microscopy using metamaterial crystals: Theory and simulations,” Phys. Rev. B 74(7), 075103 (2006). 10. Q. Cheng, T. J. Cui, W. X. Jiang, and B. G. Cai, “An omnidirectional electromagnetic absorber made of metamaterials,” New J. Phys. 12(6), 063006 (2010). Vol. 26, No. 1 | 8 Jan 2018 | OPTICS EXPRESS 1 #305840 https://doi.org/10.1364/OE.26.000001 Journal © 2018 Received 30 Aug 2017; revised 14 Oct 2017; accepted 1 Dec 2017; published 2 Jan 2018 11. H. Noh, Y. Chong, A. D. Stone, and H. Cao, “Perfect coupling of light to surface plasmons by coherent absorption,” Phys. Rev. Lett. 108(18), 186805 (2012). 12. S. A. Cummer, B.-I. Popa, D. Schurig, D. R. Smith, J. Pendry, M. Rahm, and A. Starr, “Scattering theory derivation of a 3D acoustic cloaking shell,” Phys. Rev. Lett. 100(2), 024301 (2008). 13. D. P. Gaillot, C. Croënne, and D. Lippens, “An all-dielectric route for terahertz cloaking,” Opt. Express 16(6), 3986–3992 (2008). 14. A. Alù and N. Engheta, “Effects of size and frequency dispersion in plasmonic cloaking,” Phys. Rev. E. 78(4), 045602 (2008). 15. P. Alitalo, F. Bongard, J.-F. Zurcher, J. Mosig, and S. Tretyakov, “Expermental verification of broadband cloaking using a volumetric cloak composed of periodically stacked cylindrical transmission-line networks,” Appl. Phys. Lett. 94(1), 014103 (2009). 16. A. E. Serebryannikov and E. Ozbay, “Non-ideal multifrequency cloaking using strongly dispersive materials,” Physica B 405(14), 2959–2963 (2010). 17. R. D. Averitt, S. L. Westcott, and N. J. Halas, “Linear optical properties of gold nanoshells,” J. Opt. Soc. Am. B 16(10), 1814–1823 (1999). 18. T. Christensen, A.-P. Jauho, M. Wubs, and N. A. Mortensen, “Localized plasmons in graphene-coated nanospheres,” Phys. Rev. B 91(12), 125414 (2015). 19. N.-A. P. Nicorovici, R. C. McPhedran, S. Enoch, and G. Tayeb, “Finite wavelength cloaking by plasmonic resonance,” New J. Phys. 10(11), 115020 (2008). 20. A. Mirzaei, I. V. Shadrivov, A. E. Miroshnichenko, and Y. S. Kivshar, “Cloaking and enhanced scattering of core-shell plasmonic nanowires,” Opt. Express 21(9), 10454–10459 (2013). 21. A. E. Miroshnichenko, “Off-resonance field enhancement by spherical nanoshells,” Phys. Rev. A 81(5), 053818 (2010). 22. I. Liberal, I. Ederra, R. Gonzalo, and R. W. Ziolkowski, “Superbackscattering from single dielectric particles,” J. Opt. 17(7), 072001 (2015). 23. Z. Ruan and S. Fan, “Design of subwavelength superscattering nanospheres,” Appl. Phys. Lett. 98(4), 043101 (2011). 24. A. V. Goncharenko, E. F. Venger, Y. C. Chang, and A. O. Pinchuk, “Arrays of core-shell nanospheres as 3D isotropic broadband ENZ and highly absorbing metamaterials,” Opt. Mater. Express 4(11), 2310–2322 (2014). 25. J. Ng, H. Chen, and C. T. Chan, “Metamaterial frequency-selective superabsorber,” Opt. Lett. 34(5), 644–646 (2009). 26. F. Morgan, J. McPhillips, G. Wurtz, S. A. Maier, A. V. Zayats, and R. Pollard, “Fabrication and optical properties of large-scale arrays of gold nanocavities based on rod-in-a-tube coaxials,” Appl. Phys. Lett. 102(10), 103103 (2013). 27. D. de Ceglia, S. Campione, M. A. Vincenti, F. Capolino, and M. Scalora, “Low-damping epsilon-near-zero slabs: Nonlinear and nonlocal optical properties,” Phys. Rev. B 87(15), 155140 (2013). 28. A. Alù and N. Engheta, “Multifrequency optical invisibility cloak with layered plasmonic shells,” Phys. Rev. Lett. 100(11), 113901 (2008). 29. C. A. Valagiannopoulos and P. Alitalo, “Electromagnetic cloaking of cylindrical objects by multilayer or uniform dielectric claddings,” Phys. Rev. B 85(15), 155402 (2012). 30. E. A. Velichko, “Evaluation of a graphene-covered dielectric microtube as a refractive-index sensor in the terahertz range,” J. Opt. 18(3), 035008 (2016). 31. P.-Y. Chen, J. Soric, Y. R. Padooru, H. M. Bernety, A. B. Yakovlev, and A. Alù, “Nanostructured graphene metasurface for tunable terahertz cloaking,” New J. Phys. 15(12), 123029 (2013). 32. M. Riso, M. Cuevas, and R. A. Depine, “Tunable plasmonic enhancement of light scattering and absorption in graphene coated subwavelength wires,” J. Opt. 17(7), 075001 (2015). 33. J. Zhu, J. Han, Z. Tian, J. Gu, Z. Chen, and W. Zhang, “Thermal broadband tunable terahertz metamaterials,” Opt. Commun. 284(12), 3129–3133 (2011). 34. J. Han, A. Lakhtakia, Z. Tian, X. Lu, and W. Zhang, “Magnetic and magnetothermal tunabilities of subwavelength-hole arrays in a semiconductor sheet,” Opt. Lett. 34(9), 1465–1467 (2009). 35. J. Han and A. Lakhtakia, “Semiconductor split-ring resonators for thermally tunable terahertz metamaterials,” J. Mod. Opt. 56(4), 554–557 (2009). 36. S. T. Bui, V. D. Hguyen, X. K. Bui, T. T. Nguyen, P. Lievens, Y. P. Lee, and D. L. Vu, “Thermally tunable magnetic metamaterials at THz frequencies,” J. Opt. 15(7), 075101 (2013). 37. C. Luo, D. Li, J. Yao, and F. Ling, “Direct thermal tuning of the terahertz plasmonic response of semiconductor metasurface,” J. Electromagn. Waves Appl. 29(18), 2512–2522 (2015). 38. M. Kim, J. Jeong, J. K. S. Poon, and G. V. Eleftheriades, “Vanadium-dioxide-assisted digital optical metasurfaces for dynamic wavefront engineering,” J. Opt. Soc. Am. B 33(5), 980–988 (2016). 39. T. G. Mackay and A. Lakhtakia, “Temperature-mediated transition from Dyakonov surface waves to surfaceplasmon-polariton waves,” IEEE Photonics J. 8(5), 4202813 (2016). 40. F. Chiadini, V. Fiumara, T. G. Mackay, A. Scaglione, and A. Lakhtakia, “Temperature-mediated transition from Dyakonov-Tamm surface waves to surface-plasmon-polariton waves,” J. Opt. 19(8), 085002 (2017). 41. L. Kai and A. D’Alessio, “Finely stratified cylinder model for radially inhomogeneous cylinders normally irradiated by electromagnetic plane waves,” Appl. Opt. 34(24), 5520–5530 (1995). 42. I. Gurwich, N. Shiloah, and M. Kleiman, “The recursive algorithm for electromagnetic scattering by tilted infinite circular multilayered cylinder,” J. Quant. Spectrosc. Radiat. Transf. 63(2–6), 217–229 (1999). 43. S. Manickavasagam and M. P. Mengüç, “Scattering-matrix elements of coated infinite-length cylinders,” Appl. Opt. 37(12), 2473–2482 (1998). Vol. 26, No. 1 | 8 Jan 2018 | OPTICS EXPRESS 2

Many materials change their response characteristics due to variation of magnetostatic field, temperature, and intensity of illuminating light.In particular, thermal tunability can be very efficient, if certain constitutive properties are strongly sensitive to temperature variations.Natural materials that fulfil this requirement include InSb, VO 2 and other vanadium oxides.Both InSb [33][34][35][36][37] and VO 2 [38] have been successfully used in tunable metamaterials and metasurfaces.Thermal tunability of electromagnetic surface-wave propagation has also been theoretically demonstrated [39,40].
In this paper, we study scattering of a linearly polarized plane wave by an infinitely long cylinder comprising a solid dielectric core with an InSb coating (shell).The wave vector of the incident plane wave is oriented normal to the axis of the cylinder.Either the plane wave is transverse-magnetically (TM) polarized (i.e., the magnetic field is oriented normally to the cylinder axis) or it is transvserse-electrically (TE) polarized (i.e., the electric field is oriented normally to the cylinder axis).InSb shows a transition from the dielectric state to the plasmonic state in the THz spectral regime as the temperature is increased [33][34][35][36][37]39,40].
Our main goal in this paper is to investigate thermally tunable scattering and, in particular, thermally tunable invisibility and masking.We identify which changes in the total scattering efficiency (TSE) are dynamically achievable by change of the absolute temperature from 295 K to 345 K and the sensitivity of those dynamic changes to the relative permittivity of the core.The changes refer to the spectral locations of the extrema of the TSE.Possible exploitation in novel temperature sensors is beyond the scope of this paper.All results presented in this paper were obtained by using standard analytical treatments [41][42][43], which result from enforcing the continuity of tangential components of the electric and magnetic fields across the core/shell and shell/air interfaces.Despite the studied structures not being scalable [44] due to the specific temperature dependence of the relative permittivity of InSb, we present all results as functions of the normalized frequency ka (k = 2πf/c is the free-space wavenumber, f is frequency, a is the outer radius of the shell, and c is the speed of light in free space) for convenience.

Geometry and material properties
Figure 1 presents the geometry of the boundary-value problem and the relative permittivity of InSb, the coating material.The infinitely long structure comprises (i) the cylindrical core of cross-sectional radius b and relative permittivity ε c and (ii) the InSb shell with outer radius a, as shown in Fig. 1(a).The wave vector of the incident plane wave lies wholly in the xy plane, whereas the cylindrical axis is parallel to the z axis.
The main electromagnetic characteristic used in this paper is the total scattering efficiency where c n stands for the amplitude of the nth-order partial wave in the scattered field.Equation ( 1) is a standard equation [45], in which each term in the sum over n corresponds to the order of cylindrical Bessel and Hankel functions used in the expansions of the field components in the core, shell, and the region beyond the shell [43].A Matlab code was written based on analytical formulas and used to calculate the TSE and field distributions, after rigorous tests of convergence and accuracy.

=
is the angular frequency; is the plasma frequency, with e = −1.6 × 10 −19 C, ε 0 = 8.854 × 10 −12 F m -1 , and m * = 1.3665 × 10 −32 kg being the electron charge, free-space permittivity, and effective free-carrier mass, respectively.In the temperature range that extends from 160 K to 350 K, the intrinsic career density N (in cm 3 − ) can be found by using the following formula: where k B = 8.62 × 10 −5 eV K -1 is the Boltzmann constant and T is the absolute temperature.Figure 1 in the vicinity of f = 3 THz.Hence, a significant difference in TSE may be expected at a fixed frequency as the absolute temperature changes from 295 K to 345 K.

Effects of temperature for TE-polarized incidence
Let us begin with the analysis of the effects exerted by variations in T on the scattering of an incident TE-polarized plane wave.Figure 2 presents σ as a function of ka when a = 14 μm, redshifted from ka = 0.88 to ka = 0.58 when T is decreased from 345 K to 295 K.The second minimum located at ka = 1.2 in Fig. 2(a) is redshifted to ka = 0.74 in Fig. 2(d).The third minimum is redshifted from ka = 1.89 to ka = 1.67, which implies lower sensitivity to the temperature, because ε InSb is then closer to ε ∞ and, therefore, varies less quickly with temperature, as can be observed in Fig. 1(b).One can see that all the features remain in the considered range of temperature, but the magnitudes at the maxima and spectral distances (i.e., Δka) between the consecutive minima of σ are changed together with their locations on the ka axis when T is varied.
The minima of σ correspond to three different ranges of ε InSb variation, i.e., , and 1 Re InSb > ε .As shown later, they may arise from different mechanisms of scattering suppression, which all may be possible in one structure.Adjustment of geometric parameters gives one more degree of freedom to control scattering.The main difference between Figs. 2(a)-2(d) and Figs.2(e)-2(h) is that the deep minima with near-zero σ are not observed for the latter at 1.5 ka > . The different spectral distances between the consecutive minima for different values of T are also noticeable.
The spectral location of the first minimum depends on b for fixed a, as seen from the comparison of Figs.2(a)-2(d) and Figs.2(e)-2(f).At the same time, the spectral location of the second minimum is weakly sensitive to b, so that the effect of the interior region on the scattering could be minimal.In certain conditions, this may enable masking of the interior region from a far-zone observer, particularly when the TSE is small.We refer to this phenomenon as masking of the first type.The third minimum of σ is present in the vicinity We next tried to identify those features of the TSE that arise from each of two components (i.e., core and shell) and those which manifest as their combined effects.In order to quantify possible effects of coating, i.e., scattering enhancement and scattering suppression, we calculated the TSE σ c for the coated cylinders and the TSE σ uc for the uncoated cylinders.σ c and σ uc are given by Eq. ( 1) with ε InSb specified by Eq. ( 2) and with ε InSb replaced by 1, respectively.The results of comparison are presented in Fig. 3   When 1 ψ = , masking of the second type can be obtained when the TSE for both coated and uncoated cylinders significantly deviates from zero, as a far-zone observer cannot recognize the difference between the coated and uncoated cylinders, at least if the data for σ are only available.In contrast with masking of the first type, masking of the second type should appear in the vicinity of but not exactly at some minimum of σ c , in order to remain in the scattering-suppression regime.Indeed, the minima of σ c do not coincide with the minima of σ uc , because at least a weak effect of the interior on the depth and location of the minimum of TSE remains.Thus, σ uc may take rather large values at the minima of σ c that enable 1 ψ > .Comparing Figs.2(a) and 3(a), we conclude that (i) 1 ψ < at the first minimum of σ c and (ii) this minimum does not correspond to a peak of ψ .Let us recall that 1 ψ < indicates that scattering by the coated cylinder is stronger than by the uncoated one.In contrast to the first minimum, the second minimum corresponds to a peak of ψ , although 1 ψ > is not achieved for all of T-values.In Fig. 3(a), a temperature threshold T = T th exists for the second minimum of σ c : whereas the InSb shell strengthens scattering for T < T th , it suppresses scattering for T > T th .Strong suppression can be obtained for the third minimum of σ c because 1 ψ > then for every T between 295 K and 345 K, with higher σ c achieved for lower T.Both the suppression and enhancement of scattering are strongly sensitive to the choices of a and b.In Fig. 3(b), an example is presented that shows strong scattering suppression at the first and second minima of σ c when 10 b = m μ , i.e., for the same parameters as for Figs.
2(e)-2(h).At the first minimum of σ c , weakening of scattering of cutoff type is obtained while increasing T; thus, 1 ψ > can be achieved below the temperature At the second minimum of σ c , the situation is opposite, i.e., scattering is suppressed more as T is increased, similarly to the second minimum of σ c in Fig. 3(a), but now 1 ψ > at the peaks for the whole T-range considered.The trend of variation of maxψ with T for the third minimum of σ c in Fig. 3(a) is the same as for the first minimum of σ c in Fig. 3(b).
Whereas the spectral location of the second minimum is insignificantly affected by b for fixed a, the value of ψ can be strongly sensitive to variations in b.In addition, a wideband increase of the contrast is present in Fig. 3  In Fig. 4(a), two LSPRs are asymmetrically excited on the front and the back segments of the core/shell interface.This asymmetry is expected to be connected with rather large values of the TSE.From the results in Fig. 4(b), it follows that the common effect of the LSPRs that are excited symmetrically on the front and the back segments of the core/shell interface is to weaken scattering.In both cases, higher than a six-fold enhancement of the axial magnetic field occurs in the region in which an LSPR is excited, as compared to the incident wave.These results confirm that the first minimum of σ can be connected with the specific field behavior at the core/shell interface, since the spectral location of this minimum is sensitive to variations in b.Indeed, scattering suppression occurs in Fig. 4(b) due to the LSPRs, similarly to the cloaking mechanism of Nicorovici et al. [2].The profile of the electric field in this regime looks like that of a dipole inside an epsilon-negative shell (not shown) [46,47].Figure 4(c) indicates the excitation of an LSPR on the front segment of the shell/air interface; thus, the role of the interior region is weak.This phenomenon occurs at the second maximum of σ in Fig. 2(b).Again, the asymmetry in LSPR excitation can be correlated to strong scattering.Next, Fig. 4(d) corresponds to the second minimum of σ in Fig. 2(b).Now, the resulting magnetic field is jointly due to a weak LSPR on the shell/air interface and a slightly enhanced magnetic field inside the core.These two effects are expected to compensate each other, the compensation being somewhat similar to scattering cancellation observed in plasmonic cloaking [3,14].Clearly, values of the TSE for the core alone and the shell alone are significantly larger than the TSE for the core-shell structure, so that a compensation of some kind should appear.
Finally, Fig. 4(e) presents the spatial profiles of the axial magnetic field at the third minimum of σ in Fig. 2(b).The spatial profile indicates the presence of a volumetric resonance with azimuthally inhomogeneous field distribution in the shell that is affected by the core region.The spatial profile is similar to those associated with Fabry-Perot resonances that enable invisibility of high-permittivity shells [4,16], but it looks here rather like that of an azimuthally propagating wave.However, in both Fig. 4(e) and [4], we have a volumetric resonance in a shell separating air from a dielectric material.As   , because the InSb coating is not thick enough to sustain a volumetric resonance.
We found that all resonances are thermally tunable, so that field maps similar to the ones in Figs. 4 and 5 were obtained for T = 295, 305, and 345 K at the corresponding maxima and minima of the TSE.Hence, there is no need to present field maps at the other three temperatures considered in Fig. 2.

Effects of temperature for TM-polarized incidence
Next, we present our results for TM-polarized incidence.Figure 6 contains plots of σ vs ka for the same geometric parameters and temperatures as in Fig. 2. One can see two minima of σ in all plots in Fig. 6.The spectral locations of the minima and the spectral distances between two consecutive minima vary with T.An increase of b leads to some changes in the spectrum of σ but the basic features remain the same, similarly to TE-polarized incidence discussed in Sec.Regardless of the mentioned differences, in most of these cases, scattering is quite sensitive to temperature.A relatively weak sensitivity to temperature is observed only for the second minima in Figs.6(e)-6(h), for which σ is small but not very close to zero.ψ > is achieved also for T = 295 K and 325 K in Fig. 7(a).Since the spectral locations of these sharp peaks of ψ change with T, thermally tunable invisibility is definitely possible.Neither threshold nor cutoff behavior is observed with changing temperature; rather, one can expect an optimal temperature at which ψ is maximal at the peak.The case of 1 ψ > is obtained also for the second minimum of σ c but with only twofold weakening of the TSE.

Effects of relative permittivity of the core
Let us now consider the effects of ε c on the TSE.Results are presented for T ∈{295,345} K and ε c ∈{−450,4,5,11.4,35.4}.  in Fig. 9(c), the minimum being initially the second one remains insensitive to the core material change, so it is useful for masking when the TSE is small.Indeed, as observed in Figs.9(a)-9(c), the deep minimum of σ in the vicinity of ka = 0.75 is weakly sensitive to ε c in the considered range of that parameter.Although the core material with 450 − = ε c is purely hypothetical since realistic materials with 0 Re < ε c are necessarily dispersive, Fig. 9(c) provides insight into the generality of this regime.
For the TM polarization state, the basic features observed in Figs.9(d)-9(f) are similar to those in Fig. 6.The minima of σ are sensitive to the variations in ε c .The contrast in σ rather than the shift of the minimum could be used to distinguish between two different materials.For example, there is a high contrast between the cases of high ε c [Fig.9(e)] and low ε c [Fig.9(d)] in the vicinities of ka = 0.5 and ka = 1, so that a high-permittivity core material can be distinguished from a low-permittivity one.
Clearly, the weak dependence of σ on ε c at ka = 1 in Fig. 9(d), which can be used for masking at large TSE, is connected with the fact that the scattering is mainly determined by the shell/air interface for ε c ∈{4,5,11.4};but the same is not true for .However, there is no minimum near ka = 0.5 in this case, so that high-permittivity and negative-permittivity core materials can be distinguished from each other.A temperature increase typically results in blueshifts of all minima.This is exemplified by a comparison of Fig. 9 for T = 295 K with Fig. 10 for T = 345 K, all parameters other than the temperature being the same for the two figures.The spectral distances between consecutive minima of σ (near zero) can decrease on temperature increase, as happens with (i)  The TSE is near zero for 450 − = ε c , i.e., when the field is vanishingly small inside the core.Hence, the resulting mechanism to weaken scattering can differ from the aforementioned compensation one, because the weak or vestigial LSPR is enough to obtain a small value of σ , while the field in the core at 1 > ε c can be weakly coupled to the exterior.
Therefore, a more detailed study of the mechanism leading to weak dependence of σ to ε c will soon be undertaken.

Concluding remarks
We theoretically examined the thermal tunability of the total scattering efficiency of an infinitely long dielectric cylinder coated by InSb and illuminated by a linearly polarized plane wave whose electric or magnetic field (but not both) is aligned parallel to the cylinder axis.InSb is a thermally tunable semiconductor showing transition from the dielectric state to the plasmonic state at THz frequencies when the temperature is increased.
For TE-polarized incidence, the first and second minima of the total scattering efficiency can be redshifted by more than 30% in the frequency domain by a 50-K decrease in temperature, when the relative permittivity of core material equals 5. Weakening of scattering at the first minimum for the coated cylinder is typically connected with the excitation of a localized surface-plasmon resonance at the core/shell interface.The electric field appears similar in spatial profile to that of an electric dipole inside a negative-permittivity shell.The second minimum can be associated with compensation of the effects of the different parts of the entire structure, or even with the simultaneous disappearance of these effects.The third minimum of the TSE (that may appear when thickness of the InSb coating is large enough) can be associated with a volumetric resonance in the shell.
For TM-polarized incidence, the first minimum of the TSE is redshifted by more than 35% with a 50-K decrease in temperature for the same core material.This minimum corresponds to the axial electric field that represents a weak zero-order resonance in a circular resonator (core) with a semi-transparent wall (coating).On the other hand, a strong magnetic field localized to the core/shell interface allows to consider this case also in terms of LSPR.
The TSE usually remains rather large at the second minimum, making it difficult to unambiguously identify the underlying mechanism.All of the aforementioned minima are preserved but spectrally shifted when the absolute temperature varies between 295 K and 345 K.The locations of the minima and the spectral distances between consecutive minima appear useful for temperature sensing.
Scattering can be suppressed by the incorporation of the InSb coating.This effect is especially strong for the TM polarization state, leading to more than 80-fold weakening of the scattering compared to the corresponding uncoated core.Hence, thermally tunable invisibility can be realized at THz frequencies.Weakening of scattering may appear either over narrow or broad ranges of temperature.Besides, the coated and the uncoated cylinders can have the same TSE at specific frequencies and temperatures, so that a far-zone observer cannot distinguish between them.In other words, the core can be masked by the InSb shell.Comparison of the results obtained for different core materials indicates the existence of the scattering regimes, which are either weakly or significantly sensitive to the variations in core's relative permittivity.Weak-sensitivity regimes can be exploited to mask the core from a far-zone observer.Strong-sensitivity regimes will allow one to distinguish between materials with significantly different relative permittivity.
The basic scattering features and their sensitivity to temperature endure for a wide variety of core materials, including high-permittivity and negative-permittivity materials.Hence, efficient strategies to achieve tunable invisibility and masking can be developed.Some of the reported effects and regimes can be obtained even for thinner InSb coatings, e.g., for 13 b = m μ .Our conclusions are not restricted to the use of InSb but should hold for coatings of other thermally sensitive materials as well.Furthermore, extensions to metamaterials comprising arrays of thermally tunable cylinders should be possible.Effects of phase as well as those of bright and dark LSPRs are planned for future studies in the context of the discussed scattering scenarios.The richness of the palette of achievable effects is promising for multifunctionality, i.e., when either two functions are combined in one spectral regime, or different functions are realized in different spectral regimes, in one device [48][49][50].

Fig. 1 .
Fig. 1.(a) Geometry of the boundary-value problem.(b) Relative permittivity of InSb at T = 345 K -solid blue lines, (b) 325 K -red dashed lines, 305 K -green dash-dotted lines, and 295 K -dotted black line; Re and Im indicate real and imaginary parts.
in terms of the TSE contrast, σ σ = ψ c uc / .Scattering suppression due to the coating is indicated by 1 ψ > , and scattering enhancement due to the coating is indicated by 1 ψ < .Ideal invisibility is the limiting case of scattering suppression that occurs when ψ → ∞ .Scattering enhancement is the rather usual effect that can arise due to increases in the electrical size and the volume-averaged relative permittivity by the incorporation of the InSb shell.

Fig. 3
Fig. 3. Contrast σ σ = c uc / ψ vs ka for (a) b = 6 μm and (b) b = 10 μm when a = 14 μm, 5 = εc , and the incident plane wave is TE polarized.Solid blue lines are for T = 345 K, red dashed lines for T = 325 K, green dash-dotted lines for T = 305 K, and dotted black lines for T = 295 K.In order to calculate σuc , εInSb was replaced by 1.

Fig. 4 .
Fig. 4. Spatial profiles of the axial magnetic field when 5 = εc , b = 6 μm, a = 14 μm, T = 325 K; (a) ka = 0.74 and 54 0 8 11 InSb .i .+ − = ε , (b) ka = 0.762 and 5 0 23 10 InSb .i .+ − = ε , (c) ka = 0.942 and 26 0 28 1 InSb .i .+ − = ε , (d) ka = 1.0 and 22 0 63 0 InSb .i .+ = ε , and (e) ka = 1.79 and 0385 0 98 10 InSb .i .+ = ε .The dashed white lines indicate the core/shell interface.The structure is illuminated by a TE-polarized plane wave propagating from left to right.To clarify the physics underlying the observed TSE features, Fig. 4 presents the spatial profiles of the axial magnetic field at selected ka-values in Fig. 2(b) for 6 b = m μ and T = 325 K.They correspond to the first maximum [Fig.4(a)], first minimum [Fig.4(b)], second maximum [Fig.4(c)], second minimum [Fig.4(d)], and third minimum [Fig.4(e)] in Fig. 2(b).In Fig.4(a), two LSPRs are asymmetrically excited on the front and the back segments of the core/shell interface.This asymmetry is expected to be connected with rather large values of the TSE.From the results in Fig.4(b), it follows that the common effect of the LSPRs that are excited symmetrically on the front and the back segments of the core/shell interface is to weaken scattering.In both cases, higher than a six-fold enhancement of the axial magnetic field occurs in the region in which an LSPR is excited, as compared to the e), such resonances can appear in the shell; see[4].

Figure 5
Figure 5 presents spatial profiles of the axial magnetic field at selected ka-values in Fig. 2(f) for b = 10 μm and T = 325 K.They correspond to the first maximum [Fig.5(a)], the first minimum [Fig.5(b)], the second maximum [Fig.5(c)], and the second minimum [Fig.5(d)].The basic features of Figs.5(a)-5(c) are the same as of Figs.4(a)-4(c), respectively.Figures 4(d) and 5(d) differ, in particular, in the field-intensity contrast between the LSPR region and the core.There is no analog of Fig. 4(e) for 10 m b μ =, because the InSb coating is not thick enough to sustain a volumetric resonance.We found that all resonances are thermally tunable, so that field maps similar to the ones in Figs.4 and 5were obtained for T = 295, 305, and 345 K at the corresponding maxima and minima of the TSE.Hence, there is no need to present field maps at the other three temperatures considered in Fig.2.

3 .
For instance, the first minimum of σ for 6 b = m μ is redshifted from ka = 1.14 to ka = 0.716 when T is decreased from 345 K to 295 K. Similarly to Figs. 2(a)-2(d), the spectral location of the second deep minimum of σ is less sensitive to the variation of T. The spectral locations of some minima of σ at a fixed temperature are rather weakly sensitive to the polarization state of the incident plane wave, as become clear from comparing Figs.2(a)-2(d) with Figs.6(a)-6(d) for 6 b = m μ .As examples, there are minima at ka = 1.2 and ka = 0.82 for the TE polarization state and at ka = 1.15 and ka = 0.8 for the TM polarization state when T = 345 K and T = 305 K, respectively.Compared to Fig. 2 for the TE polarization state, the spectral locations of the near-zero minima of σ at fixed temperature depend more strongly on the core radius b in Fig. 6 for the TM polarization state.This difference between the two polarization states can be accentuated by a judicious selection of the value of b.Clearly, the locations of the minima in Figs.2(e)-2(h) do not coincide with those in Figs.6(e)-6(h) for 10 b = m μ .Hence, spectral regimes of weak scattering with strong and weak sensitivities to the polarization state are available.

Figure 7
Figure 7 is the analog of Fig. 3 for TM-polarized incidence.In Fig. 7, 1 ψ > at all of the considered minima of σ c ; i.e., the InSb shell weakens scattering.For 6 b = μm, we obtain max 18 ψ ≈ at T = 305 K for the first minimum of σ c , and 15

Fig. 7
Fig. 7. Contrast σ σ = c uc / ψ vs ka for (a) b = 6 μm and (b) b = 10 μm when a = 14 μm, 5 = εc , and the incident plane wave is TM polarized.Solid blue lines are for T = 345 K, red dashed lines for T = 325 K, green dash-dotted lines for T = 305 K, and dotted black lines for T = 295 K.In order to calculate σuc , εInSb was replaced by 1.

Figure 8 μ
Figure 8 presents examples of the spatial profiles of the axial electric field for b ∈{6,10} µm and T ∈{295,325} K.The first minimum of σ is characterized in Figs. 8(a), 8(c), 8(e), and 8(g) by a weak zero-order resonance in a circular resonator (core) with a semi-transparent wall (shell).As shown by a comparison of either Figs.8(a) and 8(c) or Figs.8(e) and 8(g), the spatial profiles are the same for two different values of T. Thus, even though the spectral

Figure 9
presents σ as a function of ka when b = 6 μm, a = 14 μm, T = 295 K, for both TE or TM polarization states.From Fig.9(a) for the TE polarization state, we conclude that the first and the third minima of σ are significantly sensitive to ε c , but the second minimum is almost unaffected by changes in ε c ; moreover, the second maximum is affected weakly.Qualitatively, these inferences are in good agreement with those drawn from Figs.2 and 6; hence, the same mechanisms underlie these results as discussed for Figs.4, 5, and 8.The third minimum of σ is most sensitive to variations in ε c .For instance, changing ε c from 5 to 11.4 leads to a redshift from ka = 1.69 to ka = 1.53.

Fig. 9 .
Fig. 9.Total scattering efficiency σ when b = 6 μm, a = 14 μm, T = 295 K, and the incident plane wave is either (a-c) TE polarized or (d-f) TM polarized; (a,d) 4 = εc (solid lines), 5 = εc (dashed lines), and 4 11.c = ε (dotted lines); (b,e) 4 35.c = ε ; and (c,f) 450 − = εc .For 4 35.c = ε in Fig. 9(b), the first minimum disappears, while the second minimum stays at nearly the same spectral location as in Fig. 9(a).The third minimum is redshifted to ka = 1.13, so the difference in the locations of this minimum for 5 = ε c and 4 35.c = ε is about 0.56 ka Δ = , which corresponds to a redshift of 1.91 THz.For 450 − = ε cin Fig.9(c), the minimum being initially the second one remains insensitive to the core material change, so it is useful for masking when the TSE is small.
weakened in Fig. 9(f) in the vicinity of ka = 1 by the incorporation of the InSb shell for 450 − = ε c
polarization state in Figs.9(b) and 10(b) as also with (ii) for the TM polarization state in Figs.9(d) and 10(d).The minimum of σ , which is weakly sensitive to ε c , is located now in the vicinity of ka = 1.2 (TE polarization state).The difference in the spectral locations of the first deep minimum of σ for high-permittivity and negative-permittivity core materials for the TM polarization state in Figs.10(e) and 10(f) is insignificant.Examples of the spatial profiles of the axial magnetic field for the TE polarization state are presented in Fig. 11.In Figs.11(a)-11(d), the selected ka-values correspond to the third minimum 11(b) and 11(d)] and strong scattering for 5 = ε c [Figs.11(a) and 11(c)] is the field resonance in both the core and the shell regions in the former case, while it occurs only in the shell in the latter case.A comparison of Figs.11(a) and 11(c) and Figs.11(b) and11(d)  shows that a very similar field distribution can be obtained while staying at the corresponding minimum that is redshifted/blueshifted on changing the temperature.

Fig. 11 .
Fig. 11.Spatial profiles of the axial magnetic field when b = 6 μm, a = 14 μm; (a,b) ka = 1.53,T = 295 K and 033 0 2 12 InSb .i .+ = ε ; (c,d) ka = 1.73,T = 345 K and 061 0 48 8 InSb .i .+ = ε ; (eh) ka = 1.2, T = 345 K and 183 0 72 0 InSb .i .+ = ε ; (a,c,e) 5 = εc ; (b,d,f) 4 11.c = ε ; (g) 4 35.c = ε ; and (h) 450 − = εc .The dashed white lines indicate the core/shell interface.The structure is illuminated by a TE-polarized plane wave propagating from left to right.Next, Figs.11(e)-11(h) show the spatial profile of the axial magnetic field at the second minimum of σ in Figs.10(a)-10(c).Its spectral location is weakly sensitive to variations in ε c .The LSPR at the shell/air interface and the field enhancement inside the core are expected to compensate for each other.Although somewhat differently for different values of ε c , a general feature is that the field is rather weak inside the shell but enhanced in the core region, as can be seen in Figs.11(e)-11(g).The TSE is near zero for 450 − = ε c , i.e., when the field is vanishingly small inside the core.Hence, the resulting mechanism to weaken scattering can differ from the aforementioned compensation one, because the weak or vestigial LSPR is enough to obtain a small