1: \documentclass[12pt]{article}

2: \usepackage{amsmath, amssymb}

3: \usepackage[dvips]{epsfig}

4: 

5: 

6: 

7: \setlength{\textwidth}{145mm}

8: \setlength{\textheight}{200mm}

9: \setlength{\topmargin}{0mm}

10: 

11: \def\refname{\normalsize \centerline{\bf References}}

12: \def\abstractname{\ }

13: 

14: \def\##1{\underline #1}

15: \def\=#1{\underline{\underline #1}}

16: \def\4#1{\underline{\underline{\underline{\underline #1}}}}

17: 

18: \def\.{\mbox{ \tiny{$^\bullet$} }}

19: 

20: \def\le{\left(}

21: \def\ri{\right)}

22: \def\les{\left[}

23: \def\ris{\right]}

24: \def\lec{\left\{}

25: \def\ric{\right\}}

26: 

27: \def\c#1{\cite{#1}}

28: \def\l#1{\label{#1}}

29: \def\r#1{(\ref{#1})}

30: 

31: 

32: \def\del{\partial     }

33: \def\curl{\nabla\times}

34: \def\Re{{\rm Re}}

35: 

36: \def\eps{\epsilon}

37: \def\epso{\eps_0}

38: \def\muo{\mu_0}

39: \def\ko{k_0}

40: \def\etao{\eta_0}

41: \def\epsr{\eps_r}

42: \def\mur{\mu_r}

43: \def\epsrr{\eps_r'}

44: \def\epsri{\eps_r''}

45: \def\murr{\mu_r'}

46: \def\muri{\mu_r''}

47: \def\tde{\tan\delta_\eps}

48: \def\tdm{\tan\delta_\mu}

49: \def\nr{n'}

50: \def\ni{n''}

51: 

52: \renewcommand{\thefootnote}{\fnsymbol{footnote}}

53: 

54: \pagestyle{empty}

55: 

56: \date{}

57: 

58: \begin{document}

59: 

60: \baselineskip .582cm   %.82

61: 

62: \begin{center}

63: \vskip 0.2cm

64: {\large\bf ON PERFECT LENSES AND NIHILITY}\\[20pt]

65: {\bf Akhlesh Lakhtakia}

66: \\[20pt]

67: {\em CATMAS -- Computational \& Theoretical Materials Sciences Group\\

68: Department of Engineering Science and Mechanics\\

69: Pennsylvania State University, University Park, PA 16802--6812, USA}\\[25pt]

70: \end{center}

71: 

72: \begin{abstract}

73: The canonical problem of a perfect lens with linear bianisotropic

74: mater\-ials is formulated. Its solution is shown

75: to be directly connected with

76: the concept of nihility, the electromagnetic nilpotent. Perfect lenses as well as

77: nihility remain unrealizable.

78: \\

79: 

80: KEYWORDS: Anti--vacuum, Lens, Negative permeability, 

81: Negative permittivity,   Nihility, Perfect lens

82: \end{abstract}

83: 

84: 

85: \section{Introduction}

86: This communication has been inspired by a report on the

87: theory of a perfect lens by Pendry \c{P}. This lens

88: is supposedly constructed of a material whose permittivity

89: and permeability, respectively, are exactly negative of the

90: permittivity and the permeability of free space. In  

91:  a predecessor paper \c{Lakh1},

92: that material was postulated as the {\em anti--vacuum\/}.

93: 

94: \section{Canonical Formulation for a Lens}

95: As geometrical optics is applicable for lenses, the canonical

96: formulation  for a lens merely involves a linear homogeneous material

97: confined to the region between two parallel planes.

98: Let us, however, generalize the situation in order to understand

99: the issue at greater depth by involving two linear homogeneous

100: materials and four interfaces, as shown in Figure 1. The regions

101: $0 \leq z \leq d_1$ and $d_1+d_2 \leq z \leq d_1+d_2+d_3$

102: are occupied by a material labelled $a$, and the region

103: $d_1 \leq z \leq d_1+d_2$ by a material labelled $b$.

104: Both materials are linear, homogeneous, bianisotropic

105: and necessarily dispersive, their frequency--domain constitutive relations being as follows:

106: \begin{equation}

107: \left. \begin{array}{l}

108: {\bf D}(x,y,z,\omega) = \epso\, \les \=\eps^{a,b}(\omega)\.{\bf E}\,(x,y,z,\omega) +

109: \=\alpha^{a,b}(\omega)\.{\bf H}\,(x,y,z,\omega)\ris\\[10pt]

110: {\bf B}(x,y,z,\omega) = \muo\, \les \=\beta^{a,b}(\omega)\.{\bf E}\,(x,y,z,\omega) +

111: \=\mu^{a,b}(\omega)\.{\bf H}\,(x,y,z,\omega)\ris

112: \end{array}\ric\,.

113: \end{equation}

114: In these relations, the dielectric properties are delineated by 

115: $\=\eps^{a,b}(\omega)$,

116: the magnetic properties by $\=\mu^{a,b}(\omega)$, and the magnetoelectric properties

117: by $\=\alpha^{a,b}(\omega)$ as well as $\=\beta^{a,b}(\omega)$, all of these

118: dyadics being functions of the angular frequency $\omega$.

119: The permittivity and permeability of free space (i.e., vacuum) are denoted

120: by $\epso$ and $\muo$, respectively.

121: 

122: %%%%%%%%% Figure 1 begins %%%%%%%%%%%

123: \begin{center}

124: \begin{figure}[!ht]

125: \centering \psfull \epsfig{file=PLNFig1.eps,width=3.1in}

126: \end{figure}

127: \bigskip 

128: {\bf Figure 1:}

129: Schematic for the canonical lens formulation.

130: \end{center}

131: \bigskip

132: %%%%%%%%% Figure 1 ends %%%%%%%%%%%

133: 

134: 

135: Without loss of essential generality, we can take the spatial Fourier transformations

136: of all electromagnetic phasors with respect to $x$ and $y$; thus,

137: \begin{equation}

138: {\bf E}(x,y,z,\omega) = {\bf e}(z,\kappa,\psi,\omega) \exp\les i\kappa

139: (x\cos\psi + y\sin\psi)\ris\,,

140: \end{equation}

141: etc., suffice for the present purposes. Wave propagation in the two materials

142: can then be cast in terms of 4$\times$4 matrix ordinary differential equations

143: as follows \c{Lakh2}:

144: \begin{equation}

145: \frac{d}{dz}\,[\#f(z,\kappa,\psi,\omega)]= i [\=P_{\,a,b}(\kappa,\psi,\omega)]\.

146: [\#f(z,\kappa,\psi,\omega)]\,.

147: \end{equation}

148: Here, $[\#f] \equiv [e_x;\,e_y;\,h_x;\,h_y]^T$ is a column 4--vector

149: with the superscript $^T$ denoting the transpose, while

150: $[\=P_{\,a,b}]$ are 4$\times$4 matrixes. Accordingly,

151: we obtain the basic relation

152: \begin{equation}

153: [\#f(d_1+d_2+d_3,\kappa,\psi,\omega)]= [\=M(d_1+d_2+d_3,\kappa,\psi,\omega)]\.

154: [\#f(0,\kappa,\psi,\omega)]\,,

155: \end{equation}

156: where

157: \begin{eqnarray}

158: \nonumber

159: [\=M(d_1+d_2+d_3,\kappa,\psi,\omega)]&=& \exp\lec id_3[\=P_{\,a}(\kappa,\psi\,\omega)]\ric\.

160: \exp\lec id_2[\=P_{\,b}(\kappa,\psi\,\omega)]\ric\\

161: &&\quad \.

162: \exp\lec id_1[\=P_{\,a}(\kappa,\psi\,\omega)]\ric\,.

163: \end{eqnarray}

164: 

165: Within the confines of continuum electromagnetics,

166: the canonical lens problem  involves finding the material $b$ and the

167: thickness $d_2$ for a specified material $a$ and thicknesses $d_1$ and $d_3$,

168: such that

169: \begin{equation}

170: \label{lens}

171: [\=M(d_1+d_2+d_3,\kappa,\psi,\omega)]= 

172: \les\begin{array}{cccc}

173: 1 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 1

174: \end{array}\ris

175: \end{equation}

176: for {\em all\/} $\kappa$, $\psi$ and $\omega$.

177: 

178: \section{Analysis and Discussion}

179: Obviously, that would be a fruitless endeavor in practice. Hence, lens designers 

180: {\em effectively\/} settle

181: for some acceptable ranges of $\kappa$, $\psi$ and $\omega$

182: in which \r{lens} holds, when material $a$ is air. Deviations from

183: an ideal match introduce aberrations \c{Sin}.

184: 

185: Mathematically, and at first glance, an excellent

186: candidate for ideal

187: match is the following:

188: \begin{equation}

189: \label{ideal}

190: \left.\begin{array}{ll}

191: \=\eps^{b}(\omega) = -\=\eps^{a}(\omega)\,, &\quad

192: \=\alpha^{b}(\omega) = -\=\alpha^{a}(\omega)\, \\

193: \=\beta^{b}(\omega) = -\=\beta^{a}(\omega)\,, &\quad

194: \=\mu^{b}(\omega) = -\=\mu^{a}(\omega)\, \\

195: d_2=d_1+d_3

196: \end{array}\ric\,.

197: \end{equation}

198: Close inspection of \r{lens}, however, shows that \r{ideal}

199: is suitable for all $\kappa$,

200: only if material $a$ has orthorhombic symmetry, i.e., 

201: \begin{equation}

202: \label{ideal1}

203: \=\xi^{a} = \les \begin{array}{ccc}

204: \xi_1^a & 0 & 0 \\ 0 & \xi_2^a & 0 \\ 0 & 0 & \xi_3^a

205: \end{array}\ris\,,\qquad \xi = \eps\,,\,\alpha\,,\beta\,,\mu\,.

206: \end{equation}

207: 

208: A sandwich of equal thicknesses of materials $a$ and $b$

209: then constitutes a planar realization of a medium named {\em nihility}

210: earlier in this journal \c{Lakh1}. Nihility is the postulated electromagnetic nilpotent, with the following

211: constitutive relations:

212: \begin{equation}

213: \left. \begin{array}{l}

214: {\bf D}(x,y,z,\omega) = {\bf 0} \\[10pt]

215: {\bf B}(x,y,z,\omega) = {\bf 0}

216: \end{array}\ric\,.

217: \end{equation}

218: Wave propagation cannot occur in nihility, because $\curl {\bf E}(x,y,z,\omega) ={\bf 0}$

219: and $\curl {\bf H}(x,y,z,\omega)  = {\bf 0}$ in the absence of sources therein.

220: Whereas the phase velocity and the

221: wavevector

222: of a plane wave in vacuum/anti--vacuum are co--parallel/anti--parallel,

223: the directionality of the phase velocity relative to the

224: wave\-vector in nihility is a non--issue.

225: 

226: 

227: Physically, \r{ideal} and \r{ideal1} are still deficient because

228: the principle of energy conservation has

229: not been considered. With the normal constraint that material $a$ be

230: passive, it follows from \r{ideal} that material $b$ has to be active.

231: Although their electromagnetic response  can be simulated {\em via\/} composites

232: containing

233: active circuit elements \c{AZ},

234: I do not think that active materials will provide a realistic option 

235: in the near future. Therefore, the passivity constraint on material

236: $b$ would lead to aberrations due to absorption. To those must be

237: added chromatic aberrations, which would arise from the non--fulfilment

238: of \r{ideal} outside some limited range of $\omega$.

239: 

240: Suppose next that material $a$ is isotropic and non--magnetoelectric, i.e.,

241: $\=\eps^a=\eps^a\,\=I$, $\=\mu^a=\mu^a\,\=I$, and $\=\alpha^a=\=\beta^a = \=0$,

242: where $\=I$ is the identity dyadic and $\=0$ is the null dyadic. Then, a reasonable match

243: in some frequency range could conceivably be

244: provided by materials that supposedly possess a negative index of refraction \c{SSS},

245: so long as absorption is acceptably low.

246: 

247: Pendry \c{P} actually took material $a$ to be air, which

248: is optically indistinguishable from vacuum ($\=\eps^a= \=I$, $\=\mu^a= \=I$, and $\=\alpha^a=\=\beta^a = \=0$)

249: for most purposes. Thus, the requirements on material

250: $b$ became very simple, {\em viz.\/}, $\=\eps^b=- \=I$, $\=\mu^b= -\=I$, and $\=\alpha^b=\=\beta^b = \=0$. (In other words,

251: material $b$ has to be anti--vacuum.) But

252: Ziolkowski \c{Z} has recently concluded from

253: two--dimensional computer simulations

254: that even these simple

255: requirements (in some narrow frequency range) cannot be met by realistic meta--materials. 

256: Recent correspondence between Pendry and others adds to the debate \c{P1}--\c{P4}.

257: 

258: A perfect lens

259: remains unrealizable in my opinion, as does nihility. 

260: 

261: 

262: \small{

263: 

264: \begin{thebibliography}{99}

265: 

266: 

267: \bibitem{P}

268: J.B. Pendry,

269: Negative refraction makes a perfect lens.

270: {\em Phys. Rev. Lett.\/} {\bf 85}, 3966--3969 (2001).

271: 

272: \bibitem{Lakh1}

273: A. Lakhtakia,

274: An electromagnetic trinity from ``negative permittivity" and ``negative permeability".

275: {\em Int. J. Infrared Millim. Waves\/} (at press); physics/0112003 (2001).

276: 

277: \bibitem{Lakh2}

278: A. Lakhtakia,

279: General schema for the Brewster conditions.

280: {\em Optik\/} {\bf 90}, 184--186 (1992).

281: 

282: \bibitem{Sin}

283: D.C. Sinclair, Optical design software.

284: In: M. Bass, E.W. Van Stryland, D.R. Williams,

285: \& W.L. Wolfe (eds),

286: {\em Handbook of Optics. Vol. I\/}. New York: McGraw--Hill (1995);

287: Chap. 34.

288: 

289: \bibitem{AZ}

290: F. Auzanneau \& R.W. Ziolkowski,

291: Theoretical study of synthetic bianisotropic materials.

292: J. Electromagn. Waves Appl. {\bf 12}, 353--370 (1998).

293: 

294: \bibitem{SSS}

295: R.A. Shelby, D.R. Smith \& S. Schultz,

296: Experimental verification of a negative index of refraction.

297: {\em Science\/} {\bf 292}, 77--79 (2001).

298: 

299: \bibitem{Z}

300: R.W. Ziolkowski,

301: Wave propagation in media having negative permittivity

302: and permeability.

303: {\em Phys. Rev. E\/} {\bf 64}, 056625 (2001).

304: 

305: \bibitem{P1}

306: G.W. 't Hooft,

307: Comment on ``Negative refraction makes a perfect lens".

308: {\em Phys. Rev. Lett.\/} {\bf 87}, 249701 (2001).

309: 

310: \bibitem{P2}

311: J. Pendry,

312: Reply.

313: {\em Phys. Rev. Lett.\/} {\bf 87}, 249702 (2001).

314: 

315: \bibitem{P3}

316: J.M. Williams,

317: Some problems with negative refraction.

318: {\em Phys. Rev. Lett.\/} {\bf 87}, 249703 (2001).

319: 

320: \bibitem{P4}

321: J. Pendry,

322: Reply.

323: {\em Phys. Rev. Lett.\/} {\bf 87}, 249704 (2001).

324: 

325: \end{thebibliography}

326: }

327: 

328: 

329: 

330: 

331: \end{document}

332: 

333: 

334: 

335: