Metric Signature Transitions in Optical Metamaterials Igor Smolyaninov University of Maryland - College Park
Total Page:16
File Type:pdf, Size:1020Kb
Purdue University Purdue e-Pubs Birck and NCN Publications Birck Nanotechnology Center 8-2010 Metric Signature Transitions in Optical Metamaterials Igor Smolyaninov University of Maryland - College Park Evgenii Narimanov Purdue University, [email protected] Follow this and additional works at: http://docs.lib.purdue.edu/nanopub Part of the Nanoscience and Nanotechnology Commons Smolyaninov, Igor and Narimanov, Evgenii, "Metric Signature Transitions in Optical Metamaterials" (2010). Birck and NCN Publications. Paper 703. http://dx.doi.org/10.1103/PhysRevLett.105.067402 This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for additional information. week ending PRL 105, 067402 (2010) PHYSICAL REVIEW LETTERS 6 AUGUST 2010 Metric Signature Transitions in Optical Metamaterials Igor I. Smolyaninov1 and Evgenii E. Narimanov2 1Department of Electrical and Computer Engineering, University of Maryland, College Park, Maryland 20742, USA 2Birck Nanotechnology Center, School of Computer and Electrical Engineering, Purdue University, West Lafayette, Indiana 47907, USA (Received 6 April 2010; revised manuscript received 10 June 2010; published 2 August 2010) We demonstrate that the extraordinary waves in indefinite metamaterials experience an ( þþ) effective metric signature. During a metric signature change transition in such a metamaterial, a Minkowski space-time is created together with a large number of particles populating the space-time. Such metamaterial models provide a tabletop realization of metric signature change events suggested to occur in Bose-Einstein condensates and quantum gravity theories. DOI: 10.1103/PhysRevLett.105.067402 PACS numbers: 78.20.Ci The unprecedented degree of control of the local dielec- change event should be extremely interesting to observe. tric permittivity "ik and magnetic permeability ik tensors Unlike usual phase transitions, in which the physical sys- in electromagnetic metamaterials has fueled the recent tem changes while the background metric is intact, the explosion of novel device ideas, and resulted in the dis- signature change transition affects the underlying back- covery of new physical phenomena. Advances in experi- ground metric experienced by the system. Therefore, sig- mental design and theoretical understanding of nature change events constitute a new kind of phase metamaterials greatly benefited from the field theoretical transition. ideas developed to describe physics in curvilinear space- The equation of motion of a quantum scalar field resid- time. Electromagnetic cloaking [1–3] and electromagnetic ing on some space-time is described by the Klein-Gordon wormholes [4] may be cited as good examples. With the equation [16]: "ik ik ^ i ^ 2 newfound freedom of and manipulation, experi- P Pi’ ¼ m ’; (1) mentalists may create optical models of virtually any met- ^i ric allowed in general relativity [5]. Recent literature [6–9] where P ¼ @=@xi is the momentum operator. For a mass- provides good examples of how it is being achieved in the less field in a ‘‘flat’’ (2 þ 2) four-dimensional 2T space- time, the Klein-Gordon equation can be written as case of optical analogues of black holes. On the other hand, compared to standard general relativity, metamaterial op- @2 @2 @2 @2 þ À À ’ ¼ 0 tics gives more freedom to design an effective space-time 2 2 2 2 (2) @x1 @x2 @x3 @x4 with very unusual properties. Light propagation in all static general relativity situations can be mimicked with positive in the coordinate space, and as 2 2 2 2 "ik ¼ ik [10], while the allowed parameter space of the ðk1 À k2 þ k3 þ k4Þ’k ¼ 0 (3) metamaterial optics is broader. Thus, the flat Minkowski in the k space. To illustrate how dynamics described by space-time with the usual (À; þ; þ; þ) signature does not Eq. (2) can be mimicked in metamaterials, let us start with need to be a starting point. Other effective signatures, such a nondispersive and nonmagnetic uniaxial anisotropic ma- À; À; þ; þ as the ‘‘two times’’ (2T) physics ( ) signature terial with dielectric permittivities, "x ¼ "y ¼ "1 and "z ¼ may be realized [11]. Theoretical investigation of the 2T "2. The wave equation in such a material can be written higher dimensional space-time models had been pioneered [17]as by Dirac [12]. More recent examples can be found in @2E~ [13,14]. Metric signature change events [in which a phase À ¼ $"À1r~ r~ E;~ 2 2 (4) transition occurs between, say, (À; þ; þ; þ) and c @t (À; À; þ; þ) space-time signature] are being studied in where $"À1 is the inverse dielectric permittivity tensor Bose-Einstein condensates and in some modified gravita- calculated at the center frequency of the signal bandwidth. tion theories (see Ref. [15], and the references therein). It is Any electromagnetic field propagating in this material can predicted that a quantum field theory residing on a space- be expressed as a sum of the ‘‘ordinary’’ and ‘‘extraordi- time undergoing a signature change reacts violently to the nary’’ contributions, each of these being a sum of an imposition of the signature change. Both the total number arbitrary number of plane waves polarized in the ordinary and the total energy of the particles generated in a signature (E~ perpendicular to the optical axis) and extraordinary (E~ change event are formally infinite [15]. Therefore, such a parallel to the plane defined by the k vector of the wave and metric signature transition can be called a ‘‘big flash,’’ the optical axis) directions. Let us define our ‘‘scalar’’ which shares some similarities with the cosmological extraordinary wave function as ’ ¼ Ez (so that the ordi- ‘‘big bang.’’ A metamaterial model of a metric signature nary portion of the electromagnetic field does not contrib- 0031-9007=10=105(6)=067402(4) 067402-1 Ó 2010 The American Physical Society week ending PRL 105, 067402 (2010) PHYSICAL REVIEW LETTERS 6 AUGUST 2010 ute to ’). Equation (4) then yields the following form of of dispersion, Eq. (5) in the case of "1 < 0 and "2 > 0 the wave equation: looks like the Klein-Gordon equation (2) for a massless field in a flat (2 þ 2) four-dimensional 2T space-time. @2’ @2’ 1 @2’ @2’ ¼ þ þ : However, metamaterials generally show high dispersion, 2 2 2 2 2 (5) c @t "1@z "2 @x @y which often cannot be neglected. In this case Eq. (4)is replaced by While in ordinary crystalline anisotropic media both "1 " 2 and 2 are positive, this is not necessarily the case in ! ~ ~ ~ ~ ~ $ ~ D! ¼ rrE! and D! ¼ "!E!; (6) metamaterials. In the indefinite metamaterials considered, c2 for example, in [18,19], "1 and "2 have opposite signs. These metamaterials are typically composed of multilayer which results in the following wave equation for !: metal-dielectric or metal wire array structures, as shown in !2 @2’ 1 @2’ @2’ À ’ ¼ ! þ ! þ ! ; Fig. 1(a). Optical properties of such metamaterials are 2 ! 2 2 2 (7) c " @z "2 @x @y quite unusual. For example, there is no usual diffraction 1 limit in an indefinite metamaterial [18–23]. In the absence or equivalently in the k space !2 k2 k2 þ k2 ’ ¼ z ’ þ x y ’ : 2 !;k !;k !;k (8) c "1 "2 Let us consider the case of "1 < 0 and "2 > 0 and assume that this behavior holds in some frequency range. Here we observe that the effective space-time signature as seen by extraordinary light propagating inside the metamaterial has different character depending on the frequency. At high frequencies (above the plasma frequency of the metal), the metamaterial exhibits ‘‘normal’’ Minkowski effective met- ric with a (À; þ; þ; þ) signature, while at low frequencies the metric signature changes to (À; À; þ; þ). After simple coordinate transformation Eq. (8) can be rewritten as 2 2 2 2 ðkt À kz þ kx þ kyÞ’!;k ¼ 0; (9) which coincides with the Klein-Gordon equation for a (2 þ 2) space-time in the k space [see Eq. (3)]. Alternatively, in the case of "1 > 0 and "2 < 0, Eq. (8) can be rewritten as 2 2 2 2 ðkt þ kx þ ky À kzÞ’!;k ¼ 0: (10) As a result, both at the small and the large frequencies the effective metric looks like the Minkowski space-time. However, at small frequencies the z coordinate assumes the role of a timelike variable. Note that causality and the form of Eqs. (9) and (10) place stringent limits on the material losses and dispersion of hyperbolic metamateri- als: a dispersionless and lossless hyperbolic metamaterial would violate causality. On the other hand, such metama- terials indeed enable experimental exploration of the met- ric signature transitions [15] to and from the Minkowski space-time. Unlike the infinite number of particles created in an idealized theoretical signature change event [15], metamaterial losses limit the number of particles created FIG. 1 (color online). (a) Schematic views of the ‘‘wired’’ and during the real signature change transition. Nevertheless, ‘‘layered’’ metamaterials. Panels (b) and (c) show calculated our calculations indicate that the number of created parti- Reð"Þ and Imð"Þ of the wired metamaterial made of silica glass cles is still very large (see Fig. 2). and gallium wires (Im"x;y Im"z, not shown). This medium A composite metamaterial, which exhibits a metric sig- demonstrates metric phase transition in the visible frequency range when gallium wires are converted from the solid to liquid nature phase transition as a function of temperature, can be state. The case of NGa ¼ 0:1 volume fraction of gallium has been designed using materials which exhibit metal-dielectric considered. The range of frequencies for which the effective transitions. A numerical example of such a metamaterial metric signature changes from the Minkowski (À; þ; þ; þ)to is presented in Fig.