1: \documentclass[draft,11pt]{article}

2: %\usepackage{amsmath}

3: %\usepackage{amssymb}

4: \usepackage[dvips]{epsfig}

5: 

6:  \textwidth 8.276in

7:  \advance \textwidth by -2in

8:  \textheight 11.845in

9:  \advance \textheight by -3in

10:  \oddsidemargin 0in

11:  \evensidemargin 0in

12:  \topmargin -.5in

13: 

14: \parindent 10pt

15: \parskip10pt

16: 

17: %\renewcommand{\baselinestretch}{1.6}

18: 

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

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

21: 

22: \def\eps{\epsilon}

23: \def\epso{\epsilon_0}

24: \def\muo{\mu_0}

25: \def\ko{k_0}

26: \def\lambdao{\lambda_0}

27: \def\etao{\eta_0}

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

29: 

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

31: \def\div{\nabla \mbox{ \tiny{$^\bullet$} }}

32: 

33: \def\epsa{\epsilon_a}

34: \def\epsar{\epsilon'_a}

35: \def\epsai{\epsilon''_a}

36: 

37: \def\mua{\mu_a}

38: \def\muar{\mu'_a}

39: \def\muai{\mu''_a}

40: 

41: \def\epsb{\epsilon_b}

42: \def\mub{\mu_b}

43: 

44: 

45: \def\ux{\#{u}_x}

46: \def\uy{\#{u}_y}

47: \def\uz{\#{u}_z}

48: 

49: 

50: \def\le{\left(}

51: \def\ri{\right)}

52: \def\les{\left[}

53: \def\ris{\right]}

54: \def\lec{\left\{}

55: \def\ric{\right\}}

56: 

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

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

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

60: 

61: %for a 4x4 matrix

62: \newcommand{\mat}[1] {\left[\begin{array}{cccc}

63:             #1 \end{array}\right]}

64: 

65: 

66: 

67: \begin{document}

68: 

69: 

70: 

71: \noindent {\Large {\bf On Planewave Remittances and Goos--H\"anchen Shifts

72: of Planar Slabs with Negative Real Permittivity and Permeability}}

73: \vskip 0.2cm

74: 

75: 

76: \noindent  {\bf AKHLESH LAKHTAKIA\/}

77: \vskip 0.2cm

78: \noindent {CATMAS~---~Computational \& Theoretical Materials Sciences Group\\

79: Department of Engineering Science \& Mechanics\\

80: Pennsylvania State University, University Park, PA 16802--6812, USA}

81: 

82: \bigskip

83: 

84: \noindent  {\bf Abstract:} {\em Two results

85: about the planewave response of an isotropic,

86: dielectric--magnetic,

87: homogeneous, planar slab are deduced when the real parts of both the

88: permittivity and the permeability are altered from positive to negative. First,

89:  the reflection and the transmission coefficients suffer phase reversals

90: without change in their respective magnitudes. Second, the Goos--H\"anchen

91: shifts experienced by beams on total reflection reverse their  directions.

92: }

93: 

94: \bigskip

95: 

96: \noindent {\bf Keywords:}  Goos--H\"anchen shifts;

97: Negative permeability,

98: Negative permittivity,  Reflection; Transmission

99: 

100: \bigskip

101: 

102: \section{Introduction}

103: The recent publication of an experimental realization of certain

104: artificial materials~---~that are effectively  isotropic, homogeneous,

105: and possess negative real permittivity and permeability in some

106: frequency range (Shelby, Smith, \& Schultz, 2001)~---~has created

107: a big splash in the electromagnetics research community. Although the

108: first samples of these materials are quite primitive (Lakhtakia, 2001),

109: there is enough experimental as well as theoretical evidence (Dewar, 2001;

110: McCall, Lakhtakia, \& Weiglhofer, 2001)

111: along with the promise of applications (Pendry, 2001), that concerted

112: theoretical research in advance of its experimental counterpart is

113: justified.

114: 

115: This communication focuses on two aspects of the planewave

116: responses of planar slabs whose permittivity and permeability

117: have negative real parts at the frequency of interest. The first aspect comprises the magnitudes and phases

118: of the reflection and the transmission coefficients; and

119: the second encompasses the longitudinal shifts experienced by beams on

120: total reflection. An $\exp(-i\omega t)$ time--dependence is

121: implicit, with $\omega$ as the angular frequency.

122: 

123: \section{Theoretical Preliminaries}

124: Suppose the region $0 \leq z \leq L$ is occupied by an

125: isotropic, homogeneous, dielectric--magnetic medium

126: with relative permittivity $\epsa=\epsar+i\epsai$ and

127: relative permeability $\mua=\muar+i\muai$. The medium

128: being passive, the principle of energy conservation requires that

129: $\epsai \geq 0$ and $\muai\geq 0$. The half--spaces

130: $z < 0$ and $z>L$ are vacuous. A plane wave is incident

131: on the face $z=0$ of the slab; accordingly, a plane wave

132: is reflected into the half--space $z<0$ and another plane

133: wave is transmitted into the half--space $z>L$.

134: 

135: \subsection{s--polarization}

136: Let the incident plane wave be $s$--polarized. Then, the

137: electric field phasor in the two half--spaces can be

138: expressed as follows:

139: \begin{equation}

140: \label{eq1}

141: {\bf E}(z) =

142: \uy \exp(i\kappa x)\,\lec

143: \begin{array}{ll}

144:  \exp(i\ko\cos\theta_{inc}z) + r_s\,\exp(-i\ko\cos\theta_{inc}z) 

145: \,, & \;\; z\leq 0\\

146: t_s \exp\les i\ko\cos\theta_{inc}(z-L)\ris \,, & \;\; z\geq L

147: \end{array}

148: \right.\,.

149: \end{equation}

150: Here and hereafter, $\ko=\omega\sqrt{\epso\muo}$ is the wavenumber in vacuum;

151: $\epso$ and $\muo$ are the permittivity and the

152: permeability of vacuum, respectively; $\etao=\sqrt{\muo/\epso}$

153: is the intrinsic impedance

154: of vacuum; the

155: quantity

156: \begin{equation}

157: \kappa =\ko\,\sin\theta_{inc}\,

158: \end{equation}

159: involves $\theta_{inc}$ as the angle of incidence measured from the $+z$ axis;

160: $\ux$, $\uy$ and $\uz$ are cartesian unit vectors;

161: while $r_s$ and $t_s$ are the reflection and the transmission

162: coefficients, respectively.

163: 

164: Following standard practice, and after defining the 2$\times$2 matrixes

165: \begin{equation}

166: \label{s-defmp}

167: [P_s] =\les\begin{array}{cc}

168: 0 & -\omega\muo\mua \\-\omega\epso\epsa +\frac{\kappa^2}{\omega\muo\mua} & 0

169: \end{array}\ris\,,\quad

170: [M_s] = \exp\lec i [P_s]\,L\ric\,,

171: \end{equation}

172: we obtain the following matrix equation for the boundary

173: value problem described in the previous two paragraphs:

174: \begin{equation}

175: \label{s-sys}

176: t_s\,

177: \les\begin{array}{c}

178:  1\\ -\,\etao^{-1}\, \cos\theta_{inc} \end{array}\ris

179: =[M_s]\,

180: \les\begin{array}{c}

181: r_s +1 \\ (r_s-1)\,\etao^{-1}\, \cos\theta_{inc}\end{array}\ris\,.

182: \end{equation}

183: This equation's solution yields $r_s$ and $t_s$, from which

184: the remittances $R_s=\vert r_s\vert^2$ and $T_s=\vert t_s\vert^2$ 

185: can be calculated.

186: 

187: \subsection{p--polarization}

188: If the incident plane wave is $p$--polarized, the magnetic

189: field phasor outside the slab $0\leq z\leq L$ can be written

190: as

191: \begin{equation}

192: {\bf H}(z) =

193: -\,\uy\,\etao^{-1}\,\exp(i\kappa x)\,\lec

194: \begin{array}{ll}

195:  \exp(i\ko\cos\theta_{inc}z) + r_p\,\exp(-i\ko\cos\theta_{inc}z) 

196: \,, & \;\; z\leq 0\\

197: t_p \exp\les i\ko\cos\theta_{inc}(z-L)\ris \,, & \;\; z\geq L

198: \end{array}\right.

199: \,.

200: \end{equation}

201: The solution of the boundary value problem involves the calculation

202: of the coefficients $r_p$ and $t_p$ from the matrix

203: equation

204: \begin{equation}

205: \label{p-sys}

206: -\,t_p\,

207: \les\begin{array}{c}

208:  \cos\theta_{inc}\\ \etao^{-1}\end{array}\ris

209: =[M_p]\,

210: \les\begin{array}{c}

211: (r_p -1)\,\cos\theta_{inc}\\ -\,(r_p+1)\,\etao^{-1}\end{array}\ris\,,

212: \end{equation}

213: where

214: \begin{equation}

215: \label{p-defmp}

216: [M_p]= \exp\lec i [P_p]\,L\ric\,,\quad

217: [P_p]= \les\begin{array}{cc}

218: 0 &\omega\muo\mua-\frac{\kappa^2}{\omega\epso\epsa}\\

219: \omega\epso\epsa & 0\end{array}\ris\,.

220: \end{equation}

221: The remittances $R_p=\vert r_p\vert^2$ and $T_p=\vert t_p\vert^2$

222: can be obtained thereafter.

223: 

224: \section{Reflection and Transmission  Analysis}

225: Both matrix equations \r{s-sys} and \r{p-sys}

226: can be stated together compactly as

227: \begin{equation}

228: \label{sys}

229: t\, [f] + r\, [M]\,[g] = [M]\,[h]\,, \quad [M] = \exp\lec i[P]\,L\ric\,,

230: \end{equation}

231: where $[f]$, $[g]$ and $[h]$ are 2$\times$1

232: matrixes.

233: 

234: 

235: Now, $\epsar > 0$ and $\muar>0$ if the region $0

236: \leq z\leq L$ is filled with a conventional medium.

237: Suppose the transformation 

238: \begin{equation}

239: \label{tfm1}

240: \lec\epsar\,\rightarrow\,-\,\epsar\,,\,\muar\,\rightarrow\,-\,\muar\ric\,

241: \end{equation}

242: occurs; i.e., the conventional medium is replaced by a counterpart

243: the real parts of whose permittivity and permeability are exactly

244: negative of those of the conventional medium, but both $\epsai$

245: and $\muai$ remain unchanged. 

246: 

247: 

248: 

249: \subsection{Magnitude Invariance and Phase Reversal}

250: According to \r{s-defmp} and

251: \r{p-defmp}, the transformation

252: \begin{equation}

253: \label{tfm2}

254: \lec [P]\,\rightarrow\,-\,[P]^\ast\ric\,

255: \end{equation}

256: must follow, with the asterisk denoting the complex conjugate. As

257: shown in the Appendix, we get

258: \begin{equation}

259: \label{tfm3}

260: \lec [M]\,\rightarrow\,\,[M]^\ast\ric\,

261: \end{equation}

262: in consequence. The matrixes $[f]$, $[g]$ and $[h]$ in

263: \r{sys} being real--valued,  \r{tfm3} in turn mandates

264: the transformation

265: \begin{equation}

266: \label{tfm4}

267: \lec r\,\rightarrow\,r^\ast,\, t\,\rightarrow\,t^\ast\ric \,

268: \end{equation}

269: for the reflection and the transmission coefficients,

270: and

271: \begin{equation}

272: \label{tfm5}

273: \lec R\,\rightarrow\,R\, , T\,\rightarrow\,T\ric \,

274: \end{equation}

275: for the remittances.

276: 

277: The symmetry inherent in the

278: relationship \r{tfm1}$\Rightarrow$\r{tfm4} is of

279: significance for experiments on planar, isotropic dielectric--magnetic

280: slabs.

281: When the signs of the real parts of both the

282: permittivity and the permeability of a medium are altered,

283: the planewave reflection and transmission coefficients remain

284: unchanged in magnitude but suffer a phase reversal.

285: The alteration cannot be detected if only the planewave remittances

286: are measured.

287: 

288: \subsection{Goos--H\"anchen Shifts}

289: A major consequence of the relationship

290: \r{tfm1}$\Rightarrow$\r{tfm4} can be appreciated when the incident

291: plane wave is replaced by a beam. By definition,

292: a plane wave has infinite transverse extent; but a beam

293: is of finite transverse extent.  Newton

294: had conjectured that if a beam were to impinge on a

295: planar interface with an electromagnetically rarer medium and total reflection were

296: to occur, the reflected beam would emerge with a longitudinal displacement

297: (along the $+x$ axis). Such

298: displacements are often called the Goos--H\"anchen shifts after the two

299: scientists who reported the first measurements in 1947

300: (Goos \& H\"anchen, 1947). 

301: Subsequently, these shifts have been estimated as well as measured for planar

302: interfaces between many different pairs

303: of materials (Lakhtakia, 2002; Lotsch, 1970a,b, 1971a,b),

304:  and

305: several applications have emerged

306: as well (Jackson, 1999; Chauvat et al., 2000; de Fornel, 2000).

307: 

308: Analytical treatment of the Goos--H\"anchen shifts {\em essentially}

309: requires the solution of the same problem as dealt with

310: in Section 2 (Haibel, Nimtz,  \&   Stahlhofen, 2001). 

311: The vacuous half--spaces are, however,

312: filled with a medium possessing $\epsb$ and $\mub$

313: as its relative permittivity and relative permeability,

314: both positive and  real--valued. Equations \r{eq1}--\r{p-defmp}

315: still apply, but with (i) $\ko$ replaced by $\ko\sqrt{\epsb\mub}$,

316: and (ii) $\etao$ by $\etao\sqrt{\mub/\epsb}$. The relationship

317: \r{tfm1}$\Rightarrow$\r{tfm4} still holds.

318: 

319: \bigskip

320: \begin{center}

321: --------------------------------------------------------------------------------------------------------

322: \bigskip

323: \begin{figure}[!ht]

324: \centerline{ \psfull

325: \epsfig{file=GH1.eps,width=2in}} 

326: 

327: \caption{Schematics for Goos--H\"anchen shifts.}

328: \end{figure}

329: \bigskip

330: --------------------------------------------------------------------------------------------------------\end{center}

331: \bigskip

332: 

333: With the assumption of a Gaussian profile for the incident beam,

334: and after denoting the reflection coefficient $r = \vert r\vert\, \exp(i\varphi)$,

335: the longitudinal shift is estimated

336: as

337: \begin{equation}

338: \label{def-d}

339: d=-\,\frac{\partial\varphi}{\partial\kappa}\,.

340: \end{equation}

341: This quantity is meaningful, in the present

342: context, only when $\theta_{inc}$ exceeds

343: the critical angle; i.e., when total

344: reflection occurs. A depth of penetration equal to $(d/2)\cot\theta_{inc}$

345: is also sometimes calculated.

346: 

347: Two Goos--H\"anchen shifts are possible in the present scenario:

348: $d_s$ and $d_p$ for $s$-- and $p$--polarized beams,

349: respectively. Their

350: characteristics are well--known (Jackson, 1999; de Fornel, 2000)

351: and do not bear repetition here. It suffices to note

352: here that 

353: \begin{equation}

354: \label{GH}

355: \lec\epsar\,\rightarrow\,-\,\epsar\,,\,\muar\,\rightarrow\,-\,\muar\ric\,\Rightarrow

356: \lec d_s\,\rightarrow\,-\, d_s,\,d_p \,\rightarrow\,-\, d_p\ric\,,

357: \end{equation}

358: this relationship being an immediate consequence 

359: of

360: the relationship

361: \r{tfm1}$\Rightarrow$\r{tfm4} and the definition \r{def-d}.

362: 

363: 

364: The result \r{GH} can be understood 

365: with the help of Figure 1 as follows: When total reflection

366: occurs and the electromagnetically rarer medium is of the conventional

367: type (i.e., $\epsar>0$ and $\muar>0$), the reflected beam appears

368: as if the incident beam has skipped forward. On the contrary,

369: according to \r{GH}, the reflected beam must appear to skip

370: backward, if the electromagnetically rarer medium is of the kind

371: reported by Shelby, Smith, \& Schultz (2001) (i.e., 

372: $\epsar<0$ and $\muar<0$). No violation of causality

373: would occur in the latter instance, because some time

374: must elapse during the traversal of the distance 

375: that is apparently skipped. This result is also

376: consistent with the propagation velocity being oppositely

377: directed to the energy velocity when $\epsar<0$ and $\muar<0$

378: (McCall, Lakhtakia, \& Weiglhofer, 2001).

379: 

380: 

381: 

382: \section{Conclusions}

383: 

384: Examining the alteration in the planewave response of an isotropic,

385: dielectric--magnetic,

386: homogeneous, planar slab when the real parts of both the

387: permittivity and the permeability are changed from positive to negative,

388: we arrived at the following deductions:

389: \begin{itemize}

390: \item[(i)]

391:  The planewave reflection and transmission coefficients suffer phase reversals.

392: \item[(ii)] 

393: The planewave remittances remain unaffected.

394: \item[(iii)] When conditions for total reflection prevail, the longitudinal

395: shift experienced by a beams reverses its direction.

396: \end{itemize}

397: \bigskip

398: 

399: \section*{Appendix}

400: Let a matrix $[P]$ of size $m\times m$

401: be diagonalizable. Then it can be decomposed

402: as 

403: \begin{equation}

404: [P] = [U]\,[G]\,[U]^{-1}\,,

405: \end{equation}

406: where the matrix $[G]={\rm diag}(g_1,\,g_2,\cdots \,g_m)$ contains the 

407: consecutive eigenvalues

408: of $[P]$, while the successive columns of $[U]$ are the corresponding

409: eigenvectors of $[P]$. It follows that (Hochstadt, 1975)

410: \begin{equation}

411: \exp\lec i[P]z\ric = [U]\,\exp\lec i[G]z\ric\,[U]^{-1}\,.

412: \end{equation}

413: 

414: Let the matrix $[Q]=-[P]^\ast$ so that

415: \begin{equation}

416: [Q] = [U]^\ast\,\Big(-[G]^\ast\Big)\,\Big([U]^{-1}\Big)^\ast\,.

417: \end{equation}

418: Consequently,

419: \begin{eqnarray}

420: \nonumber

421: \exp\lec i[Q]z\ric &=& [U]^\ast\,\exp\Big(-i[G]^\ast z\Big)\,\Big([U]^{-1}\Big)^\ast

422: \\

423: \nonumber

424: &=& \Big([U]\,\exp\lec i[G]z\ric\,[U]^{-1}\Big)^\ast

425: \\

426: &=& \Big( \exp\lec i[P]z\ric\Big)^\ast\,.

427: \end{eqnarray}

428: 

429: 

430: 

431: \bigskip\bigskip\bigskip

432: 

433: \newpage

434: 

435: Chauvat, D., O. Emile, F. Bretenaker, \& A. Le Floch. 2000.

436: Direct measurement of the Wigner delay associated

437: with the Goos--H\"anchen effect.

438: {\em Phys. Rev. Lett.\/}  84:71--74.

439: 

440: de Fornel, F. 2000.

441: {\em Evanescent waves~---~From Newtonian optics to

442: atomic optics.\/} 

443: Berlin: Springer; pp. 12--18.

444: 

445: Dewar, G.A. 2001.

446: Candidates for $\mu < 0$, $\epsilon < 0$ nanostructures.

447: {\em Int. J. Mod. Phys. B\/} 15:3258--3265.

448: 

449: Goos, F. \& H. H\"anchen. 1947.

450: Ein neuer und fundamentaler Versuch zur Total\-reflexion.

451: {\em Ann. Phys. Lpz.\/} 1:333--346.

452: 

453: Haibel, A., G. Nimtz, \& A.A. Stahlhofen. 2001.

454: Frustrated total reflection: the double--prism revisited.

455: {\em Phys. Rev. E\/} 63:047601.

456: 

457: Hochstadt, H. 1975

458: {\em Differential equations~---~A modern approach.\/}

459: New York, NY: Dover Press; p. 57.

460: 

461: Jackson, J.D. 1999.

462: {\em Classical electrodynamics, 3rd ed.\/}

463: New York: Wiley; pp. 306--309.

464: 

465: 

466: Lakhtakia, A. 2001.

467: An electromagnetic trinity from ``negative permittivity"

468: and ``negative permeability".

469: {\em Int. J. Infrared Millim. Waves\/} 22:1731--1734.

470: 

471: Lakhtakia, A. 2002.

472: Truncation of angular spread of Bragg zones by total reflection,

473: and Goos--H\"anchen shifts

474: exhibited by chiral sculptured thin films.

475: {\em Arch. Elektr. \"Uber.\/} 56:000--000 (accepted

476: for publication).

477: 

478: Lotsch, H.K.V. 1970a.

479: Beam displacement at total reflection: The Goos--H\"anchen

480: effect, I.

481: {\em Optik\/} 32: 116--137.

482: 

483: Lotsch, H.K.V. 1970b.

484: Beam displacement at total reflection: The Goos--H\"anchen

485: effect, II.

486: {\em Optik\/} 32: 189--204.

487: 

488: Lotsch, H.K.V. 1971a.

489: Beam displacement at total reflection: The Goos--H\"anchen

490: effect, III.

491: {\em Optik\/} 32: 299--319.

492: 

493: Lotsch, H.K.V. 1971b.

494: Beam displacement at total reflection: The Goos--H\"anchen

495: effect, IV.

496: {\em Optik\/} 32: 553--569.

497: 

498: 

499: McCall, M.W., A. Lakhtakia, \& W.S. Weiglhofer. 2001.

500: The negative index of refraction demystified. Report No. 2001/30.

501: Department of Mathematics, University of Glasgow, UK.

502: 

503: Pendry, J. 2001.

504: Electromagnetic materials enter the negative age.

505: {\em Physics World\/} 14(9):47--51.

506: 

507: Shelby, R.A., D.R. Smith, \& S. Schultz. 2001.

508: Experimental verification of a negative index of refraction.

509: {\em Science\/} 292:77--79.

510: 

511: \end{document}

512: 

513: 

514: