1: \documentclass[prl,twocolumn,showpacs]{revtex4}

2: %\documentclass[aps,showpacs]{revtex4}

3: %\documentclass[prb]{revtex4}% Physical Review B

4: %\documentclass[preprint,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}

5: 

6: \usepackage{graphicx}

7: \usepackage{amsmath}

8: \usepackage{amssymb}

9: 

10: 

11: \begin{document}

12: \def\a{\alpha}

13: \def\b{\beta}

14: \def\e{\varepsilon}

15: \def\d{\delta}

16: \def\l{\lambda}

17: \def\m{\mu}

18: \def\t{\tau}

19: \def\n{\nu}

20: \def\o{\omega}

21: \def\r{\rho}

22: \def\S{\Sigma}

23: \def\G{\Gamma}

24: \def\D{\Delta}

25: \def\O{\Omega}

26: % Macro varie

27: 

28: \def\ra{\rightarrow}

29: \def\ua{\uparrow}

30: \def\da{\downarrow}

31: \def\pd{\partial}

32: \def\bg{{\bf g}}

33: \def\br{{\bf r}}

34: \def\bm{{\bf m}}

35: \def\bz{{\bf z}}

36: 

37: \def\be{\begin{equation}}

38: \def\ee{\end{equation}}

39: \def\bea{\begin{eqnarray}}

40: \def\eea{\end{eqnarray}}

41: \def\nn{\nonumber}

42: \def\lb{\label}

43: \def\pref#1{(\ref{#1})}

44: 

45: \title{Origin of four-fold anisotropy in square lattices of circular

46: ferromagnetic dots}

47: 

48: 

49: \author{G.N.~Kakazei,$^{1,2}$ Yu.G.~Pogorelov,$^3$ M.D.~Costa,$^4$ T.~Mewes,$^{5}$

50: P.E.~Wigen,$^2$ P.C.~Hammel,$^2$  V.O.~Golub,$^{1}$  T.~Okuno,$^6$

51: V.~Novosad$^7$}

52: 

53: \affiliation{$^1$Institute of Magnetism National Academy of Sciences

54: of Ukraine, 36b Vernadskogo Blvd., 03142 Kiev, Ukraine}

55: 

56: \affiliation{$^2$Department of Physics, Ohio State University, 191

57: West Woodruff Avenue, Columbus, OH 43210}

58: 

59: \affiliation{$^3$IFIMUP/Departamento de F\'{\i}sica, Universidade do

60: Porto, R. Campo Alegre, 687, Porto, 4169, Portugal}

61: 

62: \affiliation{$^4$CFP/Departmento de F\'{\i}sica, Universidade do

63: Porto, R. Campo Alegre, 687, Porto, 4169, Portugal}

64: 

65: \affiliation{$^5$MINT/Department of Physics and Astronomy,

66: University of Alabama, Box 870209, Tuscaloosa, AL 35487}

67: 

68: \affiliation{$^6$Institute for Chemical Research, Kyoto University,

69: Kyoto 611-0011, Japan}

70: 

71: \affiliation{$^7$Materials Science Division, Argonne National

72: Laboratory, Argonne, IL 60439}

73: 

74: \begin{abstract}

75: We discuss the four-fold anisotropy of in-plane ferromagnetic

76: resonance (FMR) field $H_r$, found in a square lattice of circular

77: Permalloy dots when the interdot distance $a$ gets comparable to the

78: dot diameter $d$. The minimum $H_r$, along the lattice

79: $\langle11\rangle$ axes, and the maximum, along the

80: $\langle10\rangle$ axes, differ by $\sim$ 50 Oe at $a/d$ = 1.1. This

81: anisotropy, not expected in uniformly magnetized dots, is explained

82: by a non-uniform magnetization $\bm(\br)$ in a dot in response to

83: dipolar forces in the patterned magnetic structure. It is well

84: described by an iterative solution of a continuous variational

85: procedure.

86: \end{abstract}

87: \pacs{75.10.Hk; 75.30.Gw; 75.70.Cn; 76.50.+g}

88: \maketitle

89: 

90: Magnetic nanostructures are of increasing interest for technological

91: applications, such as patterned recording media \cite{moser}, or

92: magnetic random access memories \cite{allwood}. One of the most

93: important issues for understanding their collective behavior is the

94: effect of long-range dipolar interactions between the dots

95: \cite{demokritov}. For the single-domain magnetic state of a dot,

96: the simplest approximation is that dots are uniformly magnetized and

97: interactions only define relative orientation of their magnetic

98: moments \cite{guslienko}. If so, the system of dipolar coupled dots

99: in a square lattice should be magnetically isotropic.

100: 

101: However, in all known experimental studies of closely packed arrays

102: of circular dots, a four-fold anisotropy (FFA) was found, either by

103: Brillouin light scattering \cite{mathieu}, ferromagnetic resonance

104: (FMR) \cite{jung} or magnetization measurements (from hysteresis

105: loops) \cite{natali,zhu}. It is important to note that FFA exists in

106: both unsaturated samples and saturated ones (i.e. above vortex

107: annihilation point on the hysteresis loop). Hence it cannot be only

108: associated with vortex formation suggested in Ref. \cite{natali}. It

109: was instead qualitatively related to stray fields from unsaturated

110: parts of magnetization inside the dots \cite{mathieu}. However no

111: quantitative description of FFA in such systems was given up to now.

112: So the aim of this study is to explain quantitatively the deviations

113: from isotropy in terms of modified demagnetizing effect in a

114: patterned planar system at decreasing inter-dot distance, from the

115: limit of isolated dot to that of continuous film. The choice of

116: \emph{X}-band FMR techniques for this study has an advantage in

117: eliminating possible interference from domain (vortex) structure

118: \cite{kakazei03}. The variational theoretical analysis is followed

119: by micromagnetic simulations.

120: 

121: Permalloy (Py) dots were fabricated with electron beam lithography

122: and lift-off techniques, as explained elsewhere \cite{novosad}. The

123: dots of thickness $t$ = 50 nm and diameter $d$ = 1 $\mu$m were

124: arranged into square arrays with the lattice parameter $a$ (center

125: to center distance) varying from 1.1 $\mu$m to 2.5 $\mu$m. The

126: dimensions were confirmed by atomic force microscopy and scanning

127: electron microscopy. Room temperature FMR studies were performed at

128: 9.8 GHz using a standard \emph{X}-band spectrometer. The dependence

129: of the FMR field $H_r$ on the azimuthal angle $\varphi_H$ of applied

130: field $\mathbf{H}$ with respect to the lattice [10] axis for almost

131: uncoupled dots ($a$ = 2.5 $\mu$m) is shown in Fig. \ref{fig1}a. Only

132: a weak uniaxial anisotropy of $H_r(\varphi_H)$ is present here,

133: which can be fitted by the simple formula $H_{r}(\varphi_H) =

134: H_{r,av} + H_{2}\cos2\varphi_H$. For the $a$ = 2.5 $\mu$m sample, we

135: found the average peak position $H_{r,av} \approx 1.13$ kOe and the

136: uniaxial anisotropy field $H_{2} \approx 5$ Oe. The latter value

137: remains the same for the rest of our samples, so this uniaxial

138: anisotropy is most probably caused by some technological factors.

139: 

140: 

141: \begin{figure}

142: \includegraphics[bb=100bp 20bp 220bp 235bp, scale=.7]{Fig1.eps}

143: \caption{In-plane FMR field in square lattices of 1 $\mu$m circular

144: Py dots as a function of field angle $\varphi_H$. a) The data for

145: lattice parameter $a = 2.5 \mu$m  are well fitted by uniaxial

146: anisotropy (solid line).  b) At $a = 1.1 \mu$m, the best fit (solid

147: line) is a superposition of FFA and uniaxial anisotropy (separately

148: shown by dashed line).} \lb{fig1}

149: \end{figure}

150: 

151: With decreasing distance $a$ between dots, two changes are observed

152: in the $H_r(\varphi_H)$ dependence. First, $H_{r,av}$ decreases to

153: $\approx 1.09$ kOe at $a$ = 1.1 $\mu$m (Fig. \ref{fig1}b). Second, a

154: four-fold anisotropy (FFA) is detected in the samples with

155: $a\leq$1.5 $\mu$m by pronounced minima of $H_{r}(\varphi_H)$ at

156: $\varphi_H$ close to the lattice $\langle11\rangle$ axes. This

157: behavior is fitted by $H_r(a,\varphi_H) = H_{r,av}(a) + H_4(a)

158: \cos4\varphi_H + H_2 \cos2\varphi_H$, as shown in Fig. \ref{fig1}b.

159: The interdot distance dependence of $H_{r,av}$ and FFA field $H_{4}$

160: is shown in Fig. \ref{fig2}. Also such anisotropy is detected in the

161: FMR linewidth, smaller for $\langle11\rangle$ than for

162: $\langle10\rangle$ case (reaching $\sim$30\% at $a/d$ = 1.1).

163: 

164: \begin{figure}

165: \includegraphics[bb=20bp 10bp 180bp 120bp, scale=1.3]{Fig2.eps}

166: \caption{ a) Average FMR field $H_{r,av}$ and FFA field $H_{4}$ as

167: functions of interdot spacing $a$. The points are the experimental

168: data and the solid lines present the 1st iteration theory (the

169: limits mark $H_r$ of isolated dot and continuous film). b) The same

170: data plotted against $(d/a)^2$ for $H_{r,av}$ and $(d/a)^4$ for

171: $H_{4}$ give excellent linear fits (dashed lines).} \lb{fig2}

172: \end{figure}

173: 

174: The FFA effect, which could not arise in uniformly in-plane

175: magnetized cylindrical dots, is evidently related to a non-uniform

176: distribution of the magnetization $\bm(r,\varphi,z)$ (in cylindric

177: coordinates $0 \leq r \leq R = d/2, 0 \leq \varphi < 2\pi, 0\leq z

178: \leq t$). A similar effect was discovered using Brillouin light

179: scattering \cite{mathieu} and magnetization reversal

180: \cite{natali,zhu} in such systems under weak enough external fields,

181: which displace vortices in each dot. This can be modeled by

182: displacements of two oppositely in-plane magnetized uniform domains

183: \cite{guslienko01}. But in the presence of external fields strong

184: enough to observe FMR, one has to assume a continuous (and mostly

185: slight) deformation of $\mathbf m(r,\varphi)$. The simplest model

186: for such deformation uses a variational procedure with respect to a

187: single parameter \cite{metlov}. However, as will be shown below, the

188: non-uniform magnetic ground state of this coupled periodic system

189: results from a rather complicate interplay between intra-dot and

190: inter-dot dipolar forces, which requires a more general variational

191: procedure.

192: 

193: Assuming fully planar and $z$-independent dot magnetization with the

194: 2D Fourier amplitudes $\bm_\bg = \int \textrm{e}^{i\bg \cdot \br}

195: \bm(\br)d\br$, the total (Zeeman plus dipolar) magnetic energy (per

196: unit thickness of a dot) can be written as (see Appendix)

197: 

198: \be

199: \lb{eq1}

200:     E = -\mathbf{H} \cdot \bm_0 +

201: \frac{2 \pi} {a^2} \sum_{\bg\neq 0} \frac{f(g t)}{g^2} |\bm_\bg

202: \cdot \bg|^2, \ee

203: 

204: \noindent where $f(u)=1-\left(1-\textrm{e}^{-u}\right)/u$

205: \cite{guslienko} and the vectors of the 2D reciprocal lattice are

206: $\varphi_H$-dependent: $\bg = (2\pi/a) (n_1 \cos\varphi_H - n_2 \sin

207: \varphi_H, n_1 \sin \varphi_H + n_2 \cos \varphi_H)$ (for

208: $\mathbf{H}\parallel x$ and integer $n_{1,2}$). The variation of

209: exchange energy at deformations on the scale of whole sample is of

210: the order of stiffness constant ($\sim 10^{-6} \textrm{ erg/cm}$ for

211: Py) and it can be neglected beside the variation $\sim H M_s d^2

212: \sim 10^{-2} \textrm{ erg/cm}$ of terms included in Eq. \ref{eq1}.

213: If the dot magnetization has constant absolute value: $\bm (\br) =

214: M_s \left (\cos \varphi(\br), \sin \varphi(\br) \right)$, its

215: variation: $\d \bm (\br) = \hat{\bz} \times \bm (\br)

216: \d\varphi(\br)$ (where $\hat{\bz}$ is unit vector normal to plane),

217: is only due to the angle variation $\d \varphi (\br)$. Using the

218: Fourier transform $\d\bm_\bg = \hat{\mathbf{z}} \times

219: \sum_{\bg^\prime} \bm_{\bg - \bg^\prime}\d\varphi_{\bg^\prime}$ in

220: the condition $\d E = 0$ leads to the equilibrium equation for the

221: Fourier amplitudes:

222: 

223: \bea

224:  m_{\bg,y} &  = & \frac{4\pi}{H a^2} \sum_{\bg^\prime \neq 0}

225: \frac{f(g^\prime t)} {{g^\prime}^2}

226: (\bm_{\bg^\prime} \cdot \bg^\prime)   \notag  \\

227: & & \qquad\qquad\qquad (\bm_{\bg - \bg^\prime} \times \bg^\prime)

228: \cdot \hat{\bz}. \lb{eq2} \eea

229: 

230: \noindent It can be suitably solved by iterations:

231: 

232: \bea

233:  m^{(n)}_{\bg,y} &=& \frac{4\pi}{H

234: a^2}\sum_{\mathbf{g}^\prime\neq 0} \frac{f(g^\prime

235: t)}{{g^\prime}^2} (\bm^{(n-1)}_{\bg^\prime} \cdot

236: \bg^\prime)\nn \\

237: & & \,\qquad\qquad (\bm^{(n-1)}_{\bg - \bg^\prime}

238: \times\bg^\prime)\cdot\hat{\bz}, \lb{eq3} \eea

239: 

240: \noindent starting from uniformly magnetized dots as zeroth

241: iteration: $m^{(0)}_{\bg,y} = 0$, $m^{(0)}_{\bg,x} = 2\pi R M_s

242: J_1(g R)/g$  (with the Bessel function $J_1$). Already the 1st

243: iteration (including the inverse Fourier transform):

244: 

245: \bea

246: {m}^{(1)}_y(\br)& =& -\theta(d-2r) \frac{8\pi^2 R^2 M^2_s}{H

247: a^2} \sum_{\bg \neq 0}  \frac{f(g t) g_x g_y}{g^3}\,\nn \\

248: & &\qquad\qquad\qquad\qquad J_1(g R) \cos(\bg\cdot\br), \lb{eq5}

249: \eea

250: 

251: \noindent (with the Heavyside $\theta$ function) reveals the FFA

252: behavior, due to the rotationally non-invariant product $g_x g_y $.

253: The calculated maximum variation of $\varphi(\br) = \arcsin [m_y

254: (\br)/M_s]$ in $\langle10\rangle$  field geometry is $\sim60\%$

255: bigger than in the $\langle11\rangle$  geometry (Fig. \ref{fig3},

256: upper row). This expected behavior persists upon further iterations.

257: Our analytic approach was checked, using the micromagnetic OOMMF

258: code \cite{donahue} on a 9$\times$9 array of considered disks (Fig.

259: \ref{fig3}, lower row) at standard values of $M_{s} = 0.83$ kOe and

260: exchange stiffness $1.3 \cdot 10^{-6}$ erg/cm \cite{kakazei04} for

261: Py. The distributions obtained in this way for the central disk in

262: the array are within $\sim10\%$ to the analytic results of the 1st

263: iteration.

264: 

265: \begin{figure}

266: \includegraphics[width=8cm]{Fig3.eps}

267: \end{figure}

268: \begin{figure}

269: \includegraphics[width=8cm]{Fig4.eps}

270: \caption{ Density plots of the equilibrium magnetization angle

271: $\varphi(\mathbf r)$ for two field geometries, $\varphi_H = 0$

272: ($\langle10\rangle$) and $\varphi_H = \pi/4$ ($\langle11\rangle$).

273: Upper row: calculated from the sum, Eq. \ref{eq5}, over

274: 100$\times$100 sites of reciprocal lattice at parameter values $a =

275: 1.1\, \mu$m, $H=$ 1.1 kOe, $M_s=$ 0.83 kOe. Lower row: micromagnetic

276: calculation by OOMMF code for the central disc in the 9$\times$9

277: array.} \lb{fig3}

278: \end{figure}

279: 

280: The FMR precession of $\bm(\br)$ is defined by the internal field

281: $\mathbf{H}_i(\br) = \mathbf{H} + \mathbf{h}(\br)$ through the local

282: dipolar field

283: 

284: \bea h_z(\br)&=&-\frac{4\pi}{a^2}\sum_{\bg}\left[1-f(g

285: t)\right]m_{\bg,z}\cos\left(\bg\cdot\br\right),\nn\\

286: h_\a(\br)&=&-\frac{4\pi}{a^2}\sum_{\b,\bg\neq 0} \frac{f(g t)g_\a

287: g_\b}{g^2}\tilde{m}_{\bg,\b}\cos\left(\bg\cdot\br\right), \lb{eq6}

288: \eea

289: 

290: \noindent ($\a,\b=x,y$). The 1st iteration for $\mathbf{h}(\br)$

291: corresponds to the zeroth iteration for $\bm(\br) =

292: (M_s,\m_y,\m_z)$, which now includes the uniform FMR amplitudes

293: $\m_y,\,\m_z$. Then the local demagnetizing factors $N_x(\br) = -

294: h_x(\br) / M_s,\,N_{y,z}(\br) = -h_{y,z}(\br)/\m_{y,z}$ define the

295: local FMR field $H_r(\br)$:

296: 

297: \bea

298: H_r(\br) &=& \sqrt{H_0^2+M_s^2[N_z(\br)-N_y(\br)]^2/4}\nn \\

299:    &-& M_s[N_z(\br)+N_y(\br)-2N_x(\br)]/2\lb{eq7}

300: \eea

301: 

302: \noindent (here $H_{0}\approx 3.3$ kOe). The average FMR field is

303: defined by the isotropic averaged demagnetizing factors

304: 

305: \begin{eqnarray}

306: \overline{N}_{x,y}& = &\left(\pi R^2 \right)^{-1} \int_{r < R}

307: N_{x,y}

308: \left(\br\right)d\br \nonumber\\

309: &=& (8\pi^2/a^2)\sum_{\bg\neq 0}f(g t)J_1^2(g R)/g^2, \label{eq8}

310: \end{eqnarray}

311: 

312: \noindent and $\overline{N}_z = 4\pi-2\overline{N}_x$. At

313: $a\to\infty$, they tend to the single dot values \cite{joseph} which

314: are for $t/R=0.1$: $N_{x,y}^{(d)}\approx0.776$ and $N_z^{(d)}

315: \approx 11.01$. Using $N_i^{(d)}$ instead of $N_i(\br)$ in Eq.

316: \ref{eq7} accurately reproduces the single dot FMR limit

317: $H_{r}^{(d)} \approx 1.15$ kOe (estimated from Fig. \ref{fig2}b).

318: Otherwise, for decreasing interdot distance, $a\to d$, the 1st

319: iteration values, Eq. \ref{eq8}, used in Eq. \ref{eq7} well describe

320: the tendency of $H_{r,av}(a)$ towards the continuous film limit

321: $H_r^{(f)}=\sqrt{H_0^2+4\pi^2M_s^2}-2\pi M_s\approx0.96$ kOe (Fig.

322: \ref{fig2}a).

323: 

324: Finally, by calculating the true local FMR fields $H_{r}(\br)$ from

325: Eqs. \ref{eq6} and \ref{eq7}, the field dependent absorption is

326: obtained as $I(H)\propto\int_{r<R} \d(H-H_{r}(\br))d\br$. Then the

327: FMR fields, $H_r$ defined from maximum of $I(H)$ in two geometries,

328: display FFA in a good agreement with the experimental data (Fig.

329: \ref{fig2}). This effect is due to the fact that stronger

330: deformation of magnetization stronger suppresses the demagnetizing

331: effect (the differences $N_z-N_{x,y}$) and thus enhances $H_r$. Also

332: it produces a bigger spread of local resonance fields $H_r(\br)$ and

333: thus broadens the FMR line, again in agreement with our

334: observations.

335: 

336: In conclusion, it is shown that under in-plane magnetic fields,

337: $\mathbf{H}$, even strong enough for FMR, the dipolar coupling in a

338: dense lattice of circular magnetic dots is able to produce a

339: continuous deformation of the dot magnetization, strongest for the

340: field orientation along lattice axes.

341: 

342: Work at ANL was supported by the U.S. Department of Energy, BES

343: Materials Sciences under Contract No. W-31-109-ENG-38; MDC was

344: supported by FCT (Portugal) and the European Union, through POCTI

345: (QCA III) grant No. SFRH/BD/7003/2001.

346: 

347: 

348: \begin{thebibliography}{10}

349: \bibitem{moser}A. Moser, K. Takano, D.T. Margulies, M. Albrecht, Y. Sonobe,

350: Y. Ikeda, S. Sun, and E.E Fullerton, J. Phys. D: Applied Physics

351: \textbf{35}, R157 (2002); S. Sun, D. Weller, J. Magn. Soc. Jpn. 25,

352: 1434 (2001).

353: \bibitem{allwood}D.A. Allwood, Gang Xiong, M.D. Cooke, C.C. Faulkner,

354: D. Atkinson, N. Vernier, and R. P. Cowburn, Science \textbf{296},

355: 2003 (2002).

356: \bibitem{demokritov}S.O. Demokritov, B. Hillebrands, and A.N. Slavin,

357: Phys. Rep. \textbf{348}, 441 (2001).

358: \bibitem{guslienko}K.Yu. Guslienko and A.N. Slavin, J. Appl. Phys. \textbf{87}, 6337 (2000);

359: J. Magn. Magn. Mat. \textbf{215}, 576 (2000).

360: \bibitem{mathieu}C. Mathieu, C. Hartmann, M. Bauer, O. B\"{u}ttner, S. Riedling,

361: B. Roos, S.O. Demokritov, and B. Hillebrands, Appl. Phys. Lett.

362: \textbf{70}, 2912 (1997).

363: \bibitem{jung} S. Jung, B. Watkins, L. DeLong, J.B. Ketterson, and

364: V.  Chandrasekhar, Phys. Rev. B \textbf{66}, 132401 (2002).

365: \bibitem{natali} M. Natali, A. Lebib, Y. Chen, I.L. Prejbeanu, and K. Ounadjela,

366: J. Appl. Phys. \textbf{91}, 7041 (2002).

367: \bibitem{zhu}X. Zhu, P. Grutter, V. Metlushko, and B. Ilic, Appl. Phys. Lett.

368: \textbf{80}, 4789 (2002).

369: \bibitem{kakazei03}G.N. Kakazei, P.E. Wigen, K.Y. Guslienko, R.W. Chantrell,

370: N.A. Lesnik, V. Metlushko, H. Shima, K. Fukamichi, Y. Otani, and V.

371: Novosad, J. Appl. Phys. \textbf{93}, 8418 (2003).

372: \bibitem{novosad} V. Novosad, K. Yu. Guslienko, H. Shima, Y. Otani, S. G. Kim,

373: K. Fukamichi, N. Kikuchi, O. Kitakami, and Y. Shimada, Phys. Rev. B

374: 65, 060402 (2002).

375: \bibitem{guslienko01}K.Yu. Guslienko, Phys. Lett. \textbf{278}, 293 (2001).

376: \bibitem{metlov} K.L. Metlov, Phys. Stat. Sol. (a) \textbf{189}, 1015 (2002);

377: K.L. Metlov and K.Yu. Guslienko, Phys. Rev. B \textbf{70}, 052406

378: (2004).

379: \bibitem{donahue}M.J. Donahue and D.G. Porter, URL: http://math.nist.gov/oommf

380: \bibitem{kakazei04}G.N. Kakazei, P.E. Wigen, K.Y. Guslienko, V. Novosad,

381: A.N. Slavin, V.O. Golub, N.A. Lesnik, and Y. Otani, Appl. Phys.

382: Lett. \textbf{85}, 443 (2004).

383: \bibitem{joseph}R.I. Joseph and E. Schl\"{o}mann, J. Appl. Phys.  \textbf{36}, 1579 (1965).

384: 

385: \end{thebibliography}

386: 

387: \section{Appendix}

388: For fully planar and \emph{z}-independent dot magnetization, the

389: dipolar energy per unit thickness of a dot in the lattice is:

390: 

391:  \bea

392:  E_d & = & \frac 1 {2t} \int_{-t/2}^{t/2} dz \int_{-t/2}^{t/2}

393: dz^\prime \int_c d\br \int d\br^\prime \sum_{\a,\b} m_\a(\br)\nn\\

394: & \times& \frac{\pd^2}{\pd r_\a \pd r_\b}\frac {m_\b(\br^\prime)}

395: {\sqrt{\left|\br - \br^\prime\right|^2 + \left(z -

396: z^\prime\right)^2}},

397:  \nn

398:  \eea

399: 

400: \noindent where the 2D integrations $\int_c d\br$ and $\int d\br$

401: are respectively over the unit cell and over the entire plane. It

402: can be also presented as

403: 

404: \bea E_d & = & \frac 1 {2t} \int_{-t/2}^{t/2} dz \int_c d\br

405: \sum_{\a} m_\a(\br) h_\a(\br,z) \nn\\

406: & =& \frac 1 {4\pi t a^2} \int_{-t/2}^{t/2} dz

407: \int_{-\infty}^{\infty} dq {\rm e}^{-i q z} \sum_{\a,\bg} \tilde

408: m_{\a,\bg} \tilde h_{\a,\bg,q},

409:  \nn

410:  \eea

411: 

412: \noindent where the Fourier amplitudes of the dipolar field are:

413: 

414:  \bea

415:   h_{\a,\bg,q} & = & \int_c d\br \int_{-\infty}^{\infty} dz^\prime

416:  {\rm e}^{i (\bg \cdot \br + qz^\prime)} h_\a(\br,z^\prime) \nn\\

417:  & = & \int_c d\br \int_{-\infty}^{\infty} dz^\prime {\rm e}^{i (\bg \cdot

418:  \br + q z^\prime)} \int d\br^\prime \int_{-t/2}^{t/2} dz^{\prime\prime} \nn\\

419:  & \times &

420:  \sum_\b \frac{\pd^2}{\pd r_\a  \pd r_\b} \frac {m_\b(\br^\prime)}

421:  {\sqrt{\left|\br - \br^\prime\right|^2 + \left(z^\prime -

422: z^{\prime\prime}\right)^2}}.

423:  \nn

424:  \eea

425: 

426: \noindent To calculate them, we express the lattice magnetization

427: $m_\b(\br^\prime)$ through its Fourier amplitudes:

428: 

429:  \[m_\b(\br^\prime) = \frac 1 {a^2} \sum_{\bg^\prime} {\rm e}^{-i \bg^\prime

430:  \cdot \br^\prime}  m_{\b,\bg^\prime},\]

431: 

432:  \noindent and then introduce the factor ${\rm e}^{i(\bg^\prime \cdot \br -

433:  qz^{\prime\prime})}$ into the integral in $d\br^\prime dz^{\prime}$,

434:  and the compensating factor ${\rm e}^{-i(\bg^\prime \cdot \br -

435:  qz^{\prime\prime})}$ into the integral in $d\br dz^{\prime\prime}$.

436:  Then the spatial integrations in $E_d$ are done accordingly to the formulas:

437: 

438:  \bea

439: &&\int_c d\br {\rm e}^{i (\bg - \bg^\prime) \cdot \br} = a^2

440: \d_{\bg,\bg^\prime},\nn\\

441: &&\,\quad\int_{-t/2}^{t/2} dz \int_{-t/2}^{t/2}

442: dz^{\prime\prime}{\rm e}^{i q (z^{ \prime

443: \prime} - z)} = \frac 4 {q^2} \sin^2 \frac {qt} 2,\nn\\

444:  && \qquad\quad \int_{-\infty}^{\infty} dz^\prime \int d\br^\prime {\rm

445: e}^{i\bg \cdot (\br - \br^\prime) + i q(z^\prime -

446: z^{\prime\prime})}\nn\\

447: && \qquad\qquad \times\frac{\pd^2}{\pd r_\a \pd r_\b} \frac 1

448: {{\sqrt{\left| \br

449: - \br^\prime\right|^2 + \left(z^\prime - z^{\prime\prime}\right)^2}}}\nn\\

450: & & \qquad\qquad\qquad\qquad\qquad = \frac {4 \pi g_\a g_\b}{g^2 +

451: q^2}. \nn \eea

452: 

453: \noindent Finally, the momentum integration

454: 

455:  \[\int_{-\infty}^{\infty} \frac{\sin^2 (qt/2)} {q^2({g}^2 +

456:  q^2)} dq = \frac{\pi t}{2 {g}^2} f(g t)\]

457: 

458: \noindent leads to the result included in Eq. \ref{eq1}.

459: 

460: 

461: \end{document}

462: