1: % author- ID 51475 ms-ID E08071

2: %%%%%%%%%%%%%%%%%%%%%%% file template.tex %%%%%%%%%%%%%%%%%%%%%%%%%

3: %

4: % This is a template file for The European Physical Journal

5: %

6: % Copy it to a new file with a new name and use it as the basis

7: % for your article

8: %

9: %%%%%%%%%%%%%%%%%%%%%%%% Springer-Verlag %%%%%%%%%%%%%%%%%%%%%%%%%%

10: %

11: \begin{filecontents}{leer.eps}

12: %!PS-Adobe-2.0 EPSF-2.0

13: %%CreationDate: Mon Jul 13 16:51:17 1992

14: %%DocumentFonts: (atend)

15: %%Pages: 0 1

16: %%BoundingBox: 72 31 601 342

17: %%EndComments

18: 

19: gsave

20: 72 31 moveto

21: 72 342 lineto

22: 601 342 lineto

23: 601 31 lineto

24: 72 31 lineto

25: showpage

26: grestore

27: %%Trailer

28: %%DocumentFonts: Helvetica

29: \end{filecontents}

30: %

31: %\documentclass[epj,referee]{svjour}

32: \documentclass[epj]{svjour}

33: 

34: \newcommand{\bea}{\begin{eqnarray}}

35: \newcommand{\eea}{\end{eqnarray}}

36: \newcommand{\spaceint}[2]{\int_{#1} d^3 #2 \;}

37: \newcommand{\vect}[1]{\mathbf{#1}}

38: \newcommand{\vat}{V^{\rm att}}

39: \newcommand{\di}{\displaystyle}

40: 

41: \usepackage{amsmath,amssymb}

42: \usepackage{latexsym}

43: \usepackage{epsfig,color}

44: \usepackage[normalem]{ulem}

45: 

46: %\setlength{\parindent}{0cm}

47: %\setlength{\parskip}{0.5cm}

48: %\setlength{\textwidth}{17cm}

49: %\setlength{\textheight}{24cm}

50: %\addtolength{\oddsidemargin}{-2.5cm}

51: %\addtolength{\evensidemargin}{-2.5cm}

52: %\addtolength{\topmargin}{-2.5cm}

53: 

54: \begin{document}

55: 

56: %\newlength{\mylen}

57: %\setlength{\mylen}{\textwidth}

58: %\addtolength{\mylen}{-1cm}

59: 

60: 

61: %\preprint{{\LARGE \bf DRAFT}}

62: 

63: \title{Effective interactions of colloids on nematic films}

64: 

65: \author{M. Oettel\inst{1}, A. Dom\'\i nguez\inst{2}, M. Tasinkevych\inst{3}

66:  \and S. Dietrich\inst{3}}

67: 

68: \institute{

69:  Johannes--Gutenberg--Universit\"at Mainz, Institut f{\"ur} Physik,

70:   WA 331, D--55099 Mainz, Germany 

71: \and

72:  F\'\i sica Te\'orica, Universidad de Sevilla, Apdo.1065, E-41080

73:   Sevilla, Spain

74: \and 

75:  Max-Planck-Institut f\"ur Metallforschung, Heisenbergstr. 3,

76:   D-70569 Stuttgart and

77:  Institut f\"ur Theoretische und Angewandte Physik, Universit\"at Stuttgart,

78:              Pfaffenwaldring 57, D-70569 Stuttgart, Germany

79: }

80: 

81: \date{Received: / Revised version:}

82: % The correct dates will be entered by Springer

83: %

84: 

85: 

86: \abstract{

87:   The elastic and capillary interactions  between a pair of colloidal particles

88:   trapped on top of a nematic film are studied theoretically 

89:   for large separations $d$. The elastic interaction is repulsive and 

90:   of quadrupolar type, varying as $d^{-5}$.

91:   For macroscopically thick films, the capillary interaction is likewise

92:   repulsive and proportional to $d^{-5}$ as a consequence

93:   of mechanical isolation of the system comprised of the colloids and the interface.

94:   A finite film thickness introduces a nonvanishing force on the system

95:   (exerted by the substrate supporting the film)

96:   leading to logarithmically varying capillary attractions. 

97:   However, their strength turns out to be too small to 

98:   be of importance for the recently observed pattern formation of 

99:   colloidal droplets on

100:   nematic films.

101: }

102: 

103: \PACS{ {82.70.Dd}{Colloids} \and

104:       {68.03.Cd} {Surface tension and related phenomena} \and

105:    {61.30.-v}{Liquid Crystals}

106:      } % end of PACS codes

107: 

108: \maketitle

109: 

110: \section{Introduction}

111: 

112: The interactions of colloidal particles trapped at fluid interfaces have been found to differ

113: significantly from the corresponding interactions in bulk solvents. This has been studied

114: mostly for electrically charged particles trapped at interfaces with water. On 

115: one hand, the presence of the interface gives rise to direct dipolar electrostatic

116: repulsions between the colloids (see Refs.~\cite{Pie80,Ave02,Par07} for some  experimental evidence),

117: on the other hand deformations of the interface may induce longer--ranged

118: capillary attractions (briefly reviewed in Refs.~\cite{Zen06,Bre7r,Oet08}) which 

119: is possibly the source of 

120: pattern formation observed

121: in various experiments \cite{Ghe97,Gar98a,Que01,Gom05,Che06}. 

122: (See, however, Ref.~\cite{Fer04} for an alternative

123: explanation due to interface impurities.)

124: 

125: Recently \cite{Lav04,Lav07}, the experimental observation of ordered structures of glycerol 

126: droplets bound to a 

127: nematic--air interface has been reported and attributed to

128: an effective pair potential between the colloids which contains a repulsive,

129: elastic part due to director deformations in the supporting

130: nematic film and an attractive, capillary part which is long--ranged and mediated 

131: by logarithmically 

132: varying deformations of the nematic--isotropic interface caused by the droplets.   

133: The schematic setup of this experiment is depicted in

134: Fig.~\ref{fig:nem_exp}. According to Ref.~\cite{Lav04}, the colloidal particles 

135: experience an 

136: upward force caused by elastic forces due to director deformations in the supporting

137: nematic film. 

138: %and the resulting surface mediated colloid--colloid interaction can be evaluated 

139: %in the superposition approximation. 

140: This upward force on the colloids is supposed to give rise to

141: the aforementioned logarithmically varying  interface deformation.

142: Applying a superposition approximation for the

143: deformation field, one can show that the ensuing effective capillary interaction

144: potential between two colloids is likewise varying logarithmically. 

145: This is similar to the flotation interaction

146: of mm--sized particles at fluid interfaces for which the force on the colloids is caused

147: by gravity (see, e.g., Ref.~\cite{Kra00}) and also parallels the tentative 

148: explanation given for the experimentally observed attractions

149: between sub-$\mu$m charged colloids at a water-oil interface \cite{Nik02}

150: (for the controversy around this explanation see 

151: Refs.~\cite{Meg03,Nik03,For04,Kra04,Oet05,Oet05a,Kra05,Dan06}). 

152: 

153: \begin{figure}[t]

154:  \begin{center}

155:   \epsfig{file=nem_exp.eps, width=\columnwidth}

156:  \end{center}

157:  \caption{ \label{fig:nem_exp}

158:  Schematic setup of the experiment reported in Ref.~\cite{Lav04}. 

159:  Colloidal glycerol drops ($R = 1 \dots 7$ $\mu$m) are trapped at the surface of a   

160:  thick nematic film ($h \approx 60$ $\mu$m). The director field in the

161:  nematic film is sketched by the black lines and their dotted interpolations. 

162:  Nematic anchoring at the glycerol substrate at the bottom of the film

163:  as well as on the surface of the glycerol drops 

164:  is parallel, whereas it is perpendicular (homeotropic) 

165:  at the deformed nematic--air interface. Each

166:  drop is necessarily accompanied by a topological defect ($\otimes$). 

167:  }

168: \end{figure}

169: 

170: However, it is now well established \cite{Meg03,Nik03,For04,Oet05,Oet05a,Kra05,Dom06}

171: %some of us have shown recently \cite{Oet05} 

172: that interface deformations and effective colloidal interactions varying logarithmically only arise in experimental systems which are not isolated

173: mechanically. For mechanically isolated systems it can be shown \cite{Oet05b,Wue05,Dom05,Dom07}

174: that both the interface deformation around  a single  colloid and the effective

175: interaction between two of them  are shorter--ranged and the latter cannot be 

176: calculated reliably within the

177: superposition approximation. 

178: 

179: In the following we will extend the arguments presented in 

180: Refs.~\cite{Oet05,Oet05b,Wue05,Dom05,Dom07}

181: to systems with colloids at nematic interfaces. We will show that mechanical isolation

182: of the system ``nematic film -- colloid -- air" can be violated through a subtle interplay

183: between the finite thickness of the film and the anchoring conditions at the colloids

184: and at the nematic interfaces with the substrate and with the  air, respectively. 

185: However, for 

186: experimental conditions as the ones described in 

187: Ref.~\cite{Lav04} a quantitative estimate of the strength of the ensuing 

188: logarithmic attraction between the colloids yields that these attractions

189: are unobservably small. Therefore it seems likely that this kind of asymptotic capillary forces

190: cannot be invoked as a relevant mechanism to account for the observations 

191: reported in Ref.~\cite{Lav04}.  

192: 

193: The manuscript is organized as follows:

194: In Sec.~\ref{sec:coarse} the coarse--grained model for the nematic

195: phase is introduced which will serve as the basis for all subsequent

196: calculations. In Subsec.~\ref{sec:thick} we study the case of an

197: infinitely thick nematic film. First, we compute the asymptotic form of

198: the director field and the ensuing elastic force between two

199: particles. Then we calculate the effective force arising from the

200: deformation of the fluid--nematic interface caused by the elastic

201: stresses. In Subsec.~\ref{sec:thin} we consider a nematic film of finite

202: thickness, which models more closely the experimental setup described

203: in Ref.~\cite{Lav04}; for such a system we extend the above calculations 

204: to the two opposite cases

205: of perpendicular and parallel anchoring of the director field at the

206: substrate surface.  In Sec.~\ref{sec:conclusion} we

207: discuss our results.

208: 

209: %that for a  

210: %mechanically isolated experimental system (which is reasonable to assume since

211: %gravity is unimportant for $\mu$m--sized particles) an acting force on the colloid

212: %is necessarily accompanied by a stress on the interface in such a way that 

213: %for small forces the

214: %resulting interface deformation is {\em not} logarithmic but short--ranged:

215: %if the stress field on the interface around one colloid $\Pi(r)$ decays like

216: %$r^{-n}$ the interface deformation $u(r)$ decays like $r^{-n+2}$.   

217: %Consequently, the superposition approximation predicts a short--ranged, 

218: %{\em repulsive} surface mediated potential $V_{\rm men}(d)$ for colloids at distance

219: %$d$ which varies as $d^{-n}$. Although we have investigated 

220: %only charged colloids in more detail, the arguments are quite general and can be applied

221: %to a nematic interface as well which we will demonstrate below. 

222: 

223: 

224: \section{Coarse--grained model}

225: \label{sec:coarse}

226: 

227: In view of the mesoscopic length scales involved we describe the bulk part of the 

228: nematic free energy associated with the director deformations

229: in terms of the Frank free energy expression within the one--coupling approximation

230: \cite{deG74}

231: \bea

232:   {\cal F}^{\rm b}_{\rm ne} &=&  \spaceint{V_{\rm ne}}{r} f^{\rm b} (\vect r) \; \nonumber \\

233:  \label{eq:f_lc}

234:            &=& \frac{K}{2}\spaceint{V_{\rm ne}}{r} \left[ (\nabla \cdot \vect n)^2

235:    + (\nabla \times \vect n)^2 \right]   \\

236:     &=& \frac{K}{2}\spaceint{V_{\rm ne}}{r} \nabla n_{i}\cdot \nabla n_{i} + \nonumber \\

237:  & &   

238:   \frac{K}{2}\spaceint{V_{\rm ne}}{r} \nabla \cdot[ \vect n (\nabla \cdot \vect n) -(\vect n \cdot \nabla) \vect n] \;. \nonumber

239: \eea

240: $V_{\rm ne}$ denotes the volume occupied by the nematic film, $\vect n$ is the

241: director field ($\vect n^2 =1$), and the constant $K$ is of the order of $10^{-11}$ N

242: \cite{deG74}. 

243: The  total divergence term in the last line of Eq.~(\ref{eq:f_lc}) exhibits 

244: the so--called ``$K_{24}$--structure" and is unimportant for the bulk equations describing

245: the equilibrium configuration.

246: The surface free energies associated with the interfaces with air and substrate, 

247: respectively, (see Fig.~\ref{fig:nem_exp}) 

248: are described in terms of the Poulini expression \cite{Lav03}: 

249: \bea

250:  \label{eq:fs_lc}   

251:    {\cal F}^{\rm s}_{\rm ne} &=& \frac{W_1}{2} \int_{A_{\rm air-ne}}

252:   \!\!\!\!\!\!\!\!\!\!\!\! dA \;

253:    (\vect n \cdot \vect e_A)^2 + 

254: %  \nonumber \\ & &

255:    \frac{W_2}{2} 

256:    \int_{A_{\rm sub-ne} \bigcup A_{\rm coll-ne}}

257:   \!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\! dA \;

258:    (\vect n \cdot \vect e_A)^2  \;.

259: \eea 

260: %{\tt New things included from here till Eq.(3)}

261: Here, $\vect e_A$ denotes the local surface normal unit vector pointing outwards from the

262: film or the colloid. Normal alignment is favored

263: for $W_i<0$ and parallel alignment for $W_i>0$. 

264: %

265: %and the director adopts the favored alignment over a length of the

266: %order of $K/|W_i|$.

267: %

268: Typically one has

269: $|W_i| \sim 10^{-5}$ N/m \cite{Lav03} so that the length scale $K/|W_i| \sim 1$ $\mu$m is smaller

270: than the range of droplet radii investigated in Ref.~\cite{Lav04}.

271: Thus, in the ``strong anchoring'' limit which we shall consider, %this case, 

272: the effect of the boundary terms is so strong that

273: as a first approximation it amounts to  

274: fixing the angle between the director and the surface normal. 

275: We shall adopt $W_1<0$ (normal alignment at the nematic--air interface) and $W_2>0$ 

276: (parallel alignment at the nematic--glycerol interfaces). Some consequences of 

277: deviations from the strong anchoring limit will be discussed in App.~\ref{sec:app1}.

278: 

279: These surface contributions to the free energy (``wetting energies'') 

280: are small corrections to the 

281: surface tensions which are mainly due to dispersion interactions.

282: We denote these non--nematic contributions to the surface tensions as  

283: $\gamma_1$ (colloid--air surface

284: tension), $\gamma_2$ (substrate--nematic surface tension) and $\gamma'$ (nematic--air

285: surface tension). Typically, these surface tensions are of the order of  $10^{-2}$ N/m.

286: Therefore, they are much larger than the constants $W_i$ which determine the nematic

287: contributions to the surface tension. We see that due to 

288: the above anchoring conditions the surface tension of the substrate--nematic interface

289: carries no nematic contributions due to $\vect n \cdot \vect e_A=0$ (Eq.~(\ref{eq:fs_lc})) 

290: whereas in the strong anchoring limit the full nematic--air surface tension 

291: is $\gamma= \gamma' + W_1/2 \approx \gamma'$ due to $\vect n \cdot \vect e_A=1$ 

292: (Eq.~(\ref{eq:fs_lc})).\footnote{

293: A genuine contribution to the surface tension of a nematic interface

294: arises if one takes into account the variation of the nematic tensorial order

295: parameter through the interface, described by, e.g., the Landau--de Gennes

296: free energy functional (generalizing Eq.~(\ref{eq:f_lc})). 

297: The magnitude of these contributions

298: can be estimated by the surface tension of the interface between the nematic and

299: the isotropic phase of a liquid crystal. Typically, such a surface tension

300: is also $O(10^{-5}$ N/m) $\sim |W_i|$ and therefore small compared to the dispersion

301: force contribution. Furthermore, a distorted director structure in the {\em bulk}

302: may also give rise to surface energy contributions on the boundaries. For an 

303: example see Ref.~\cite{deG74}, p. 131 and p. 174. Also in this case it can be

304: argued that the corresponding contributions to the surface tension do not exceed

305: $|W_i|$.  } 

306: %

307: %Ref.~\cite{Lav04} identifies the force associated to the anchoring

308: %``wetting energy'' $\propto W_2$ with the ultimate origin of the

309: %capillary attraction. We notice, however, that this argument is

310: %incomplete: if the particle is only partially wetted by the nematic

311: %phase (contact angle $\theta \in (0,\pi)$, as found experimentally),

312: %this force will be counteracted by the force associated to the

313: %``wetting energy'' of the particle--air interface. Therefore, the

314: %particle can be trapped at an undeformed fluid interface

315: %%without the need of a deformation of the meniscus 

316: %and remain in mechanical equilibrium \cite{Pie80} (this effect is

317: %described by the second term in the free energy functional in

318: %Eq.~(\ref{eq:F}) below). In the present case, however, there are also elastic

319: %forces pulling at the interface. Our goal is the computation of the

320: %interfacial deformation and the ensuing effective capillary

321: %interaction due to these elastic forces.

322: 

323:   

324: The canonical stress tensor $\pi_{ij}$ associated with the free energy 

325: expression in Eq.~(\ref{eq:f_lc})

326: is given by 

327: \bea

328:  \label{eq:stresstensor}

329:  \pi_{ij} = \frac{\partial{ f^{\rm b}}}{\partial n_{k,j}}\;n_{k,i} - \delta_{ij}\;f^{\rm b}

330: \eea

331: where $n_{k,i} =\partial n_k/\partial x_i$ (summation over $k$). 

332: The total stress tensor $\Pi_{ij}$ is obtained by

333: adding the contribution of the isotropic pressure $p$:

334: \bea

335:  \Pi_{ij} = \pi_{ij} - \delta_{ij}\;p\;.

336: \eea

337: 

338: \subsection{Macroscopically thick nematic film}

339: \label{sec:thick}

340: 

341: First we consider the limiting case $h \to \infty$ (i.e., very thick nematic films (see 

342: Fig.~\ref{fig:nem_exp})).\footnote{This was 

343: implicitly assumed also  by the authors of  Ref.~\cite{Lav04} in discussing Fig.~2 

344: therein.} Due to 

345: the small values of the elastic coupling constant, $K \ll \gamma\, R$, 

346: %(see Fig.~\ref{fig:nem_exp})

347: and

348: of the anchoring energy, $|W_i| \ll \gamma$, the equilibrium 

349: configuration of  a single colloid at the nematic--air interface deviates 

350: only slightly from

351: the reference configuration depicted in Fig.~\ref{fig:ref}. In this latter configuration, 

352: the interface

353: is flat and the colloid is positioned such that the contact angle fulfills Young's law

354: $\cos \theta = (\gamma_1-\gamma_2)/\gamma$.

355: 

356: \begin{figure}[t]

357:  \begin{center}

358:   \epsfig{file=ref.eps, width=\columnwidth}

359:  \end{center}

360:   \caption{\label{fig:ref}

361:      In the reference configuration the whole system is divided into

362:       volumes $V_1$ and $V_2$. Volume $V_2$ (enclosed by the

363:       upper dashed curve) includes the air and the glycerol drop and volume $V_1$ (enclosed by

364:       the lower dashed curve) includes the nematic.

365:       The arrows indicate the

366:       direction in which the surfaces (including the infinitesimally

367:       displaced ones) are oriented: $S$

368:       encloses the whole system,  $S_{\rm men}$ is the interface

369:       between the nematic phase and air (acting as a meniscus), and $S_1$ is the 

370:       interface between the colloidal

371:       drop and the nematic phase. The director field and the topological 

372:       defect ($\otimes$) are

373:       indicated as in Fig.~\ref{fig:nem_exp}. The radius of the three--phase

374:       contact line is denoted by $\rho_0=R\sin\theta$ where $\theta$ is the contact angle

375:       of the air--nematic interface with the colloid of radius $R$. $\rho$ denotes

376:       the lateral distance from the vertical symmetry axis of the colloidal drop.}

377: \end{figure}

378: 

379: The total force on the whole system reads (the

380: superscript $^{+(-)}$ denotes evaluation on the positive (negative)

381: side of the oriented surface, i.e., on the side the arrows in Fig.~\ref{fig:ref}

382: [which indicate the surface normals] point to (do not point to)):

383: \bea

384:  \oint_S  d{\bf A} \cdot {\mathbf \Pi} &=& \int_{V_1\bigcup V_2} 

385:   \!\!\!\!\!\! dV \; 

386:    (\nabla \cdot {\mathbf \Pi}) + \int_{S_{\rm men}\bigcup S_1} 

387:   \!\!\!\!\!\! d{\bf A} \cdot 

388:   ({\mathbf \Pi}^+ - {\mathbf \Pi}^-) \nonumber \\

389:   &=& - \int_{S_{\rm men}}  dA \; (\pi_{zz}+p_{\rm air}-p)  \, {\bf e}_z  + 

390:   \nonumber \\ &&

391:   \int_{S_1} d{\bf A}\cdot \left[{\boldsymbol \pi}+(p_{\rm air}-p){\bf 1}\right]

392:    \nonumber \\

393:   &=& - \int_{S_{\rm men}}  dA \; \pi_{zz}  \, {\bf e}_z  + 

394:   \int_{S_1} d{\bf A}\cdot {\boldsymbol \pi}  \; .

395:  \label{eq:Ftot}

396: \eea

397: In obtaining this equation we have applied Gauss' theorem. Furthermore

398: we have used the relation

399: $\nabla \cdot {\mathbf \Pi}=0$ in volumina $V_1$ and $V_2$ which is valid because 

400: the reference configuration is taken 

401: to be in force equilibrium. This also implies that the  isotropic pressures above the

402: interface ($p_{\rm air}$) and below it ($p$) are equal 

403: and that the director configuration is given by the 

404: corresponding Euler--Lagrange equilibrium equations following from the functional

405: in Eq.~(\ref{eq:f_lc}).

406: Since at the interface $S_1$ the colloidal drop is rigidly attached to the liquid crystal,

407: we can identify the vertical force $F$ on the colloid and the total force $F_\pi$

408: on the air--nematic interface by

409: \bea

410:  \label{eq:Fdef}

411:   F\;{\bf e}_z &=& \int_{S_1} d{\bf A}\cdot {\boldsymbol \pi}

412: \eea

413: and 

414: \bea 

415:  \label{eq:Fpi}

416:   F_\pi &=& -\int_{S_{\rm men}}  dA \; \pi_{zz} \; ,

417: \eea

418: respectively.

419: Mechanical isolation of the system means that the total force  

420: $\oint_S  d{\bf A} \cdot {\mathbf \Pi}$ acting  on it 

421: is zero which leads to

422: \bea

423:  \label{eq:f_eq_pi}

424:   F&=& F_\pi\;.

425: \eea 

426: 

427: For a given force on the colloid and a given stress on the interface, the

428: interface deformation relative to the reference configuration can be determined perturbatively.

429: To that end, we summarize briefly those results of Ref.~\cite{Oet05} which are pertinent

430: also for the present system. 

431: With the introduction of the  two small, dimensionless parameters 

432: \bea

433:   \label{eq:epsdef}

434:   \varepsilon_F = - F/(2\pi\,\rho_0\,\gamma) \quad \mbox{and} \quad

435:   \varepsilon_\pi = - F_\pi/(2\pi\,\rho_0\,\gamma)

436: \eea

437: one can  expand (up to second order in

438: $\varepsilon_F$ and $\varepsilon_\pi$) the free energy difference ${\cal F}$  

439: associated with the interface deformation $u(\rho\ge \rho_0)$ 

440: around a single colloid (see Fig.~\ref{fig:ref}) 

441: and with a vertical shift $\Delta h$ which is the difference of the  colloid center position

442: relative to that in  the reference configuration: 

443: \bea

444:   {\cal F} & \simeq & 

445:   2 \pi \gamma \int_{\rho_0}^{\infty} d\rho \; \rho \left[ \frac{1}{2}\left(\frac{d u}{d \rho} \right)^2 +

446:     \frac{u^2}{2 \lambda^2} + \frac{1}{\gamma} \pi_{zz} \, u \right] +

447:  \nonumber \\ &&

448:   \label{eq:F}

449:   \pi \gamma [ u(\rho_0) - \Delta h ]^2 -

450:   F \Delta h \;.

451: \eea

452: Here, $\lambda=(\gamma/(g\bar\rho_{\rm m}))^{1/2}$ is the capillary length associated 

453: with the interface where $g$ is the gravitational constant and $\bar\rho_{\rm m}$ is the

454: mass density of the nematic phase. This expression

455: for the free energy contains all surface free energy changes relative to the reference

456: configuration involving the interfaces between air, nematic,

457: or colloid. It also  contains the contributions due to  volume forces acting on the nematic 

458: (associated with $\lambda$)

459: and the energy change of the colloid upon vertical shifts (for further details see

460: Ref.~\cite{Oet05}).

461: Note that 

462: to leading (quadratic) order in $\varepsilon_\pi,\varepsilon_F$

463: the free energy change of the nematic due to the shifted interface

464: and due to a change in the director configuration with respect

465: to the reference configuration is captured 

466: by the term $\propto \int \pi_{zz}\,u$.

467: (The analogous textbook argument for electrostatics \cite{Sch98} can be 

468: easily generalized

469: to the nematic case described by the free energy expression in Eq.~(\ref{eq:f_lc}).) 

470: 

471: Minimizing ${\cal F}$ with respect to $u(\rho)$ and $\Delta h$ and 

472: focussing on  the regime $\rho \ll \lambda$ yields 

473: \bea

474:   \label{eq:sol1}

475:   u (\rho) \simeq \rho_0 (\varepsilon_\pi - \varepsilon_F) \ln\frac{C \lambda}{\rho}

476:   + \frac{1}{\gamma} \int_{\rho}^{+\infty} \!\!\!\! d\sigma \; \sigma \, \pi_{zz}(\sigma) \ln \frac{\sigma}{\rho} \;, \quad

477: \eea

478: with $C \simeq 1.12$. We see that in the case of an isolated system

479: $(\varepsilon_\pi= \varepsilon_F)$ the logarithmic part of $u(\rho)$ vanishes. The 

480: second term on the rhs of Eq.~(\ref{eq:sol1}) leads to 

481: $u(\rho \to \infty) \propto \rho^{-n+2}$

482: if $\pi_{zz} \propto \rho^{-n}$ and thus describes a shorter--ranged power--law

483: decay of the interface deformation. 

484: 

485: The absence of logarithmic deformations for an isolated system has been derived here

486: under certain simplifying conditions (small interfacial deformation everywhere, rotational symmetry).

487: %which allows using the reference configuration. 

488: In App.~\ref{sec:app2} we demonstrate that this conclusion

489: holds in general.

490: % less restrictive case that the interface deformation is small

491: % only far away from the particles.

492:  

493: 

494: \subsubsection{Asymptotic director configuration and elastic force between colloids}

495: 

496: 

497: %Next we derive

498: %the exponent $n$ of the asymptotic stress field $\pi_{zz}$ for our simple 

499: %director model, defined by Eq.~(\ref{eq:f_lc}) and 

500: %perpendicular anchoring at the interface as realized experimentally in 

501: %Ref.~\cite{Lav04}. 

502: In Refs.~\cite{Pou97,Lub98} it has been shown that a colloidal drop 

503: immersed in the bulk of a liquid crystal

504: is accompanied by a single counterdefect such that the total topological charge is zero

505: (here, the volume occupied by the colloid contains a topological charge which

506: may be represented by a virtual defect inside the colloid) and

507: the asymptotic behavior of the director field is of dipolar character.

508: Based on similar considerations we shall show that for a colloidal drop located at the

509: air--nematic interface the boundary conditions for that interface impose a 

510: quadrupole--like  asymptotic 

511: behavior of the director field. Macroscopically far from the colloid the director

512: is oriented parallel to the $z$ axis.

513: Accordingly, at large but finite distances $r$ the director is

514: given by $\vect n(\vect r) \simeq (n_1,n_2,1-O(n_1^2,n_2^2))$ and the bulk

515: free energy corresponding to Eq.~(\ref{eq:f_lc}) is  given by

516: \bea

517:  \label{eq:fasy}

518:   {\cal F}^{\rm b}_{\rm ne} \simeq \frac{K}{2} \spaceint{V_{\rm ne}}{r} 

519:     \left( \sum_{i=1,2} \left( \nabla n_i 

520:     \right)^2 + O (n_i^4) \right)\;.

521: \eea

522: Here we have discarded the total divergence term in the free energy expression

523: (\ref{eq:f_lc}). It is unimportant for the bulk equations and it adds a mere

524: constant to the free energy because the director is anchored normally at 

525: the boundary (the nematic--air interface).

526: Thus for each component

527: $i=1,2$ the equilibrium director field fulfills the Laplace equation

528: \bea

529:  \label{eq:lapl}

530:    \Delta n_i = 0\;.

531: \eea

532: Analogously to electrostatics, the asymptotic solution for $n_i$ can be expanded

533: in terms of multipoles. To this end we  consider the reference configuration in

534:  Fig.~\ref{fig:ref}. 

535: We choose as the origin of the coordinate system the center 

536: of the circle formed by the planar three--phase contact line. 

537: The solution for the director field has to fulfill the following requirements:

538: ($i$) rotational covariance around the $z$-axis\footnote{If $\bf D$ specifies 

539: the transformation matrix for such a rotation then this

540: requirement is given by $\vect n (\bf D\cdot \vect r) = \bf D \cdot \vect n

541: (\vect r)$.} and ($ii$) $n_i(x,y,z=0)=0$.

542: Analyzing the multipole {\em ansatz} (with $\vect r=(r_1, r_2, r_3)$)

543: %\alpha,\;\alpha=1,2,3)$)

544: \bea

545:  \label{eq:es_mul}

546:   n_i &=& q_i\,\frac{1}{r} + \sum_{\alpha=1}^3 P_{i\alpha}\,\frac{r_\alpha}{r^3} +  

547:       \sum_{\alpha,\beta=1}^3 Q_{i\alpha\beta}\,\frac{r_\alpha\,r_\beta}{r^5}

548:  + \dots

549: \eea

550: it follows that rotational covariance requires $q_i=0$, 

551: $ P_{i\alpha} = P\,\delta_{i\alpha} + P_{\rm mag}\,\epsilon_{i\alpha 3}$ 

552: and $Q_{i\alpha\beta}=Q'_\beta\,\delta_{i\alpha} + 

553: Q'_{{\rm mag},\beta}\,\epsilon_{i\alpha 3}$ ($\epsilon_{ijk}$ is the Levi-Civit\`a tensor).

554: $P_{\rm mag}$ and $Q'_{{\rm mag},\beta}$ are dipole and quadrupole moments,

555: respectively, for a director field of ``magnetic" type, i.e., for

556: which ${\rm div}\,n_i=0$ holds. 

557: The boundary condition ($ii$) at the interface with the air further imposes 

558: $P=P_{\rm mag}=0$ and

559: $Q'_\beta= Q\, \delta_{\beta3}$, $Q'_{{\rm mag},\beta}= Q_{\rm mag}\, 

560: \delta_{\beta3}$. It appears to be difficult geometrically to match

561: the asymptotic solution of ``magnetic" type  with a solution near the

562: colloid which obeys parallel anchoring at the colloid surface. 

563: %\textcolor{blue}{\sout{Furthermore,

564: %if we assumed weak anchoring ($|W|R/K \ll 1$), the magnetic

565: %quadrupole would vanish identically as can be shown in a perturbative

566: %calculation along the lines of Ref.~\cite{Ruh97}. }}

567: Therefore we 

568: discard the magnetic quadrupole,

569:  i.e., the leading asymptotic term is given by

570: the remaining, ``electric" quadrupole term:

571: \bea

572:    \label{eq:quadrupole}

573:   n_i  &=& Q\; \frac{z\,r_i}{r^5} +\dots \qquad (z \equiv r_3) \;. 

574: \eea 

575: This is at variance with Ref.~\cite{Lav04} 

576: (see Eq.~(3) therein and the considerations in the paragraph above that equation 

577: which assume

578: a dipole field) but it is consistent with the analysis in Ref.~\cite{Lub98}. 

579: Dimensional analysis yields $Q=O(R^3)$ \cite{Lub98}. 

580: Note that we have derived the asymptotic behavior of the director field at the

581: interface using strong anchoring at the interface ($n_1=n_2=0$). 

582: In App.~\ref{sec:app1}

583: we discuss corrections to strong anchoring which are, however, of subleading

584: character and leave the leading behavior (Eq.~(\ref{eq:quadrupole})) unchanged.

585: 

586: The asymptotic elastic interaction between two colloids in the bulk at distance $d$ accompanied

587: by a quadrupolar director deformation has been analyzed in Ref.~\cite{Ruh97}

588: (for weak anchoring) and in Refs.~\cite{Ram96,Lub98} (using a coarse--graining

589: method, applicable also for strong anchoring) yielding identical results. For the present configuration (distance

590: vector perpendicular to the asymptotic director) the elastic

591: potential is repulsive  and varies as

592: \begin{equation}

593:   V_{\rm el} \propto \frac{K\,Q^2}{d^5} \propto \gamma \rho_0^2\,

594:    \varepsilon_F\left(\frac{\rho_0}{d}\right)^5\;. 

595: \end{equation}

596: We have used that the dimensionless force parameter $\varepsilon_F$ is proportional to

597: $Q^2$

598: which actually follows from Eqs.~(\ref{eq:Fdef}), (\ref{eq:f_eq_pi}), (\ref{eq:epsdef}), and  

599: (\ref{eq:pizz_single}) below.

600: Note that we have simply extrapolated the {\em bulk} results for two interacting

601: colloids which cause asymptotically quadrupolar deformations of the director.

602:  This appears

603: to be reasonable because the asymptotic director field in the nematic phase for 

604: the interface

605: problem is  precisely that of the bulk solution in the lower half plane

606: and because the bulk solution is antisymmetric with respect to $z \to -z$, thus

607: respecting the boundary condition $n_i(x,y,z=0)=0$. However, the precise

608: numerical value of the quadrupole moment $Q$ might be rather different for the case

609: of the colloid trapped at the interface as compared to the bulk case.

610: 

611: \subsubsection{Asymptotic behavior of the stress on the interface and meniscus--induced 

612: effective potential between colloids}

613: 

614: The asymptotic behavior of the stress tensor component $\pi_{zz}$ at the interface

615: follows from inserting 

616: Eq.~(\ref{eq:quadrupole}) into Eq.~(\ref{eq:stresstensor}):

617: \bea

618:  \left. \pi_{zz}\right|_{\rm interface} &=&

619: %  \left.\frac{K}{2} \sum_{i=1}^2 \left(n_{i,z}n_{i,z}-n_{i,r_1}n_{i,r_1}-n_{i,r_2}n_{i,r_2}

620:   \left.\frac{K}{2} \sum_{i=1}^2 \left(n_{i,z}^2-n_{i,r_1}^2-n_{i,r_2}^2

621:    \right)\right|_{z=0} 

622:  \nonumber \\  

623:  \label{eq:pizz_single}

624:   &\stackrel{r\to\infty}{\longrightarrow} &  \frac{K}{2}\,Q^2 \,\frac{1}{\rho^8} \;, 

625: \eea

626: ($\rho^2=r_1^2+r_2^2$).

627: Consequently the interface

628: deformation around a single colloid for a mechanically isolated system obeys

629: $u(\rho\to\infty)\propto \rho^{-6}$ (see Eq.~(\ref{eq:sol1})). 

630: 

631: For the problem of two identical colloids  located at ${\boldsymbol \rho}_1$ and 

632: ${\boldsymbol \rho}_2$ (vectors are defined in the interface plane $z=0$) a distance

633: $d=|{\boldsymbol \rho}_1-{\boldsymbol \rho}_2| \gg R$ apart the expression for the

634: free energy is a straightforward generalization of the one for the single--colloid

635: free energy given in Eq.~(\ref{eq:F}):

636: \bea

637:   \label{eq:F2}

638:   {\hat {\cal F}} &= &

639:   \gamma \int_{S_{\rm men}} \!\!\!\!\!\! d^2 \rho \;

640:   \left[  \frac{|\nabla \hat{u}|^2}{2} + \frac{\hat{u}^2}{2 \lambda^2} -

641:     \frac{\hat{\pi}_{zz}}{\gamma} \, \hat{u} \right] +

642:  \\ \nonumber & &

643:    \sum_{i=1,2} \left\{

644:     \frac{\gamma}{2 \rho_{0}} \oint_{\partial S_i} \!\!\! d\ell \; [\Delta \hat{h}_i - \hat{u}]^2

645:     - \hat{F}_i \Delta \hat{h}_i \right\} .

646: % \nonumber

647: \eea

648: Here, $\hat{F}_i$ denotes the force on colloid $i$ and  $\Delta \hat{h}_i$ is the relative

649: position of its center. The integration domain $S_{\rm men}$ is the whole interface plane

650: except for the two circular disks bordered by the (reference configuration) contact lines  

651: $\partial S_i$. The meniscus--induced effective potential 

652: is the difference between the equilibrium free energy of the two colloids at distance $d$

653: and their free energy at macroscopic distance:

654: \bea

655:   V_{\rm men}(d) = {\hat {\cal F}}_{\rm eq} (d) - {\hat {\cal F}}_{\rm eq} (d\to\infty) \;.

656: \eea

657: As before, minimization with respect to

658: $\hat u(\boldsymbol \rho)$ and $\Delta \hat h_i$ renders the equilibrium free energy.

659: 

660: The behavior of $V_{\rm men}(d)$ has been analyzed  in detail in 

661: Refs.~\cite{Oet05,Dom07}. Here we summarize these results as far as they are relevant

662: for the present problem.

663: The interfacial stress $\hat \pi_{zz}$ % and the interface deformation field $\hat u$ 

664: may be decomposed generally as

665: \bea

666:   \hat \pi_{zz}(\boldsymbol \rho)& = &\pi_{zz}(|\boldsymbol \rho-\boldsymbol \rho_1|) + \pi_{zz}(|\boldsymbol \rho-\boldsymbol \rho_2|) 

667:   + 2\,\pi_{zz, {\rm m}}(\boldsymbol \rho)

668:  \nonumber \\

669:   \label{eq:pihat}

670:  & \equiv & \pi_{zz,1}+\pi_{zz,2}+2\,\pi_{zz, {\rm m}} \;.% \\

671: %  \label{eq:uhat}

672: %  \hat u(\boldsymbol \rho)& = &u(|\boldsymbol \rho-\boldsymbol \rho_1|) + u(|\boldsymbol \rho-\boldsymbol \rho_2|) + u_{\rm m}(\boldsymbol \rho)

673: %  \equiv u_1+u_2+u_{\rm m} \;. 

674: \eea

675: Here, $\pi_{zz, i}$ denotes the stress %and  interface deformation field

676: around  colloid $i$ which pertains to the problem of a single colloid. 

677: To quadratic order the asymptotic director field around 

678: two colloids is given by the superposition of the components $n_i$ of the

679: single--colloid solutions and thus to this order we recover the decomposition

680: in Eq.~(\ref{eq:pihat}) with the mixed

681: component of stress field $\pi_{zz, {\rm m}}$ given by

682: \bea

683:   \pi_{zz, {\rm m}} =

684:  \frac{K}{2}\;Q^2\; \frac{(\boldsymbol \rho-\boldsymbol \rho_1) \cdot (\boldsymbol \rho-\boldsymbol \rho_2)}

685: {|\boldsymbol \rho-\boldsymbol \rho_1|^5\; |\boldsymbol \rho-\boldsymbol \rho_2|^5} \;.

686: \eea

687: 

688: It turns out that  

689: for a system under an external force $(\varepsilon_\pi \not = \varepsilon_F)$

690: the mixed term $\pi_{zz, {\rm m}}$ does not contribute to the leading

691: term in $V_{\rm men}$. %and $u_{\rm m}$. 

692: Thus in the case of a non--vanishing external force this leading

693: contribution to $V_{\rm men}$ is obtained by a superposition 

694: {\em ansatz} which  consists in 

695: approximating the interfacial deformation and the total %deformation and 

696: stress field by the sum of the respective single--colloid quantities only

697: ($\hat \pi_{zz}\approx \pi_{zz,1}+\pi_{zz,2}$,

698:  $\hat u \approx u_1+u_2$) \cite{Oet05,Dom07}:

699: \bea

700:  \label{eq:vmen_noniso}

701:  V_{\rm men}(\rho_0 \ll d \ll \lambda) &\simeq&   - 2\pi\;\gamma\; \rho_0^2\,

702:    (\varepsilon_\pi - \varepsilon_F)^2 \, \ln \frac{C\lambda}{d}  \qquad

703:   \\ && \nonumber

704:      (\varepsilon_\pi \not = \varepsilon_F)\;.

705: \eea

706: For an isolated system  $(\varepsilon_\pi= \varepsilon_F)$ and for the stress

707: given in Eq.~(\ref{eq:pizz_single}) the superposition approximation 

708: $\hat \pi_{zz}\approx \pi_{zz,1}+\pi_{zz,2}$

709: yields $V_{\rm men} \propto \varepsilon_F^2/d^8$ as the dominant term. 

710: However, this is not the leading term, which rather stems from 

711: $\pi_{zz,{\rm m}}$. This term has two peaks around the colloid centers and 

712: therefore close to the colloids it can be approximated by

713: \bea

714:   \label{eq:pimmpeak}

715:   \pi_{zz,{\rm m}} \approx \frac{K \,Q^2}{2\,d^4} \sum_{i=1}^2 (-1)^i

716:   \frac{ \vect e_d \cdot(\boldsymbol \rho-\boldsymbol \rho_i)}

717:   {|\boldsymbol \rho-\boldsymbol \rho_i|^5} 

718: \eea

719: where $\vect e_d = (\boldsymbol \rho_2-\boldsymbol \rho_1)/d$.

720: %This mixed stress gives rise to mixed contributions $u_{\rm m}$ which are also

721: %centered around the colloids and whose overall amplitude is determined by the 

722: %prefactor of $\pi_{zz,{\rm m}}$ and thus we have $u_{\rm m} \propto d^{-4}$.

723: As discussed in Ref.~\cite{Dom07}, the qualitative behavior of  $V_{\rm men}(d)$  

724: is captured by the integral over $\pi_{zz,{\rm m}}$:

725: \bea

726:  \label{eq:vmen_iso}

727:   V_{\rm men}(d) &\propto & \gamma\rho_0^3\,\varepsilon_F \int_{S_{\rm men}} \!\!\! d^2\rho \, 

728:    \pi_{zz,{\rm m}}(\boldsymbol \rho) 

729:   \propto  \gamma\rho_0^2\,\varepsilon_F^2\,\left(\frac{\rho_0}{d}\right)^5 \qquad

730:  \\ && \nonumber

731:   (\varepsilon_\pi = \varepsilon_F)\;.

732: \eea

733: We note that this form of the mixed stress and thus the leading behavior of 

734: $V_{\rm men}(d)$

735: is formally analogous to the contributions of the

736: electric field components parallel to the interface in the case of charged colloids

737: \cite{Dom07}.  

738: The meniscus--induced potential $V_{\rm men}$ is repulsive and 

739: falls off asymptotically\footnote{Superficially one would expect a 

740: leading decay 

741: $V_{\rm men}\propto d^{-4}$ as displayed in Eq.~(\ref{eq:pimmpeak}). However, due to

742: the geometric factor in the numerator of Eq.~(\ref{eq:pimmpeak}), this apparent

743: leading order vanishes upon integration.} $\propto

744: d^{-5}$, as does the likewise repulsive elastic 

745: potential $V_{\rm el}$. We note that 

746: $V_{\rm el} \propto |\varepsilon_F|$ and $V_{\rm men} \propto \varepsilon_F^2$, so that

747: the meniscus--induced potential is small compared to the elastic one for

748: the parameters of the experiment reported in Ref.~\cite{Lav04}.

749: 

750: 

751: 

752: In order to explain the experimentally observed attractions, in Ref.~\cite{Lav04} 

753: a perturbative 

754: picture similar to the one presented

755: above was suggested, but the effect of the interface stress $\pi_{zz}$ was

756: neglected completely. (A similar error has been made in Ref.~\cite{Nik02} which

757: the authors of Ref.~\cite{Lav04} refer to.) 

758: %{\tt The following sentence has been rewritten:} 

759: In a heuristic way, only an ``upward" force on the colloid (perpendicular to the interface)

760:  associated with

761: the anchoring ``wetting" energy at the nematic--particle interface

762: has been invoked in Ref.~\cite{Lav04}, neglecting

763: the force on the interface described by $\pi_{zz}$.

764: % which, as we have seen, is incomplete.

765: % a change of the director

766: % field between the reference and the deformed configuration which can be seen

767: % to be of higher order

768: % in the parameters $\varepsilon_\pi, \varepsilon_F$. 

769: In this way, the unbalance of the force

770: gives rise to a logarithmic term in the meniscus

771: deformation and in the meniscus--induced potential. However, mechanical isolation 

772: (i.e., force balance) renders

773: the meniscus--induced potential actually repulsive and shorter--ranged 

774: (see Eq.~(\ref{eq:vmen_iso})), apparently in contrast to the experimental results.

775: A logarithmically varying potential can only arise if mechanical isolation is violated (see App.~\ref{sec:app2}). 

776: Below we shall investigate whether a net force on the system ``colloid and interface"

777: may appear if the

778: thickness of the nematic phase is finite, as it is the case in the actual experiment. 

779: 

780:  

781: 

782: 

783: 

784: 

785: \subsection{Finite thickness of the nematic film}

786: \label{sec:thin}

787: 

788: In our discussion of a finite film thickness of the nematic phase we shall consider two cases:

789: \begin{enumerate}

790: \item  The anchoring of the nematic director at the surface of the bottom substrate is 

791: perpendicular as it is the case at the upper interface with the air. 

792: This case bears a strong formal resemblance to  charged

793: colloids on water surfaces which have been  discussed in Refs.~\cite{Oet05,Dom07}. 

794: \item The anchoring at the bottom substrate surface is parallel (as in the experiment reported in

795: Ref.~\cite{Lav04}). At large lateral distances from the colloids, this leads to a director

796: field which gradually rotates from parallel orientation at the bottom substrate to the

797: perpendicular orientation at the upper interface. 

798: \end{enumerate}

799: %{\tt new intermediate summary}

800: %In combination with the normalization

801: %condition ${\bf n}^2=1$ this leads to a strong net force

802: %and consequently to a rather strong interface deformation. This effect is peculiar 

803: %to this experimental system.  

804: 

805: For both cases the total force on the system -- comprising air, nematic film, 

806: colloid, and the substrate -- must be zero. This leads to

807: (see Fig.~\ref{fig:ref1}(a) and compare with Eq.~(\ref{eq:Ftot}))

808: \bea

809:  0 &=& \oint_S  d{\bf A} \cdot {\mathbf \Pi} \nonumber \\

810:    &=& \int_{V_1\bigcup V_2\bigcup V_2'} \!\!\!\!\!\!\!\!\! dV \; 

811:    (\nabla \cdot {\mathbf \Pi}) + \int_{S_{\rm men}\bigcup S_1\bigcup S_{\rm sub}} \!\!\!\!\!\!\!\!\! d{\bf A} \cdot 

812:   ({\mathbf \Pi}^+ - {\mathbf \Pi}^-) \nonumber \\

813:   &=& - \int_{S_{\rm men}}  dA \; (\pi_{zz}+p_{\rm air}-p)  \, {\bf e}_z  +

814:  \label{eq:Ftot1} \\

815:  && \int_{S_1} \!\!\!d{\bf A}\cdot \left[{\boldsymbol \pi}+(p_{\rm air}-p){\bf 1}\right]

816:   + \int_{S_{\rm sub}} \!\!\!\!\!\!dA \; (\pi_{zz}+p_{\rm sub}-p)  \, {\bf e}_z  \; .

817:  \nonumber

818: \eea

819: For reasons which will become clear in the  discussion of  the second anchoring case, 

820: we consider the isotropic pressures in the substrate and air, $p_{\rm sub}$ and

821: $p_{\rm air}$, respectively, not necessarily to be equal to the pressure $p$ in the 

822: nematic film. The second equation in Eq.~(\ref{eq:Ftot1}) follows from the equilibrium

823: condition $\nabla \cdot {\mathbf \Pi}=0$ which holds  in all volumina.

824: 

825: \subsubsection{Perpendicular anchoring at both interfaces}

826: 

827: \label{sec:aligned}

828: 

829: \begin{figure}[t]

830:  \begin{center}

831:   \epsfig{file=ref1.eps, width=\columnwidth}

832:  \end{center}

833:   \caption{\label{fig:ref1}

834:     The reference configuration for a nematic film (a) differs from the reference 

835:     configuration

836:     shown in Fig.~\ref{fig:ref} by the addition of the substrate volume $V_2'$.

837:     The interface between the  substrate and the nematic film is denoted by 

838:     $S_{\rm sub}$.

839:     Panel (b) shows the string of image quadrupoles $ Q_b=Q$ which are needed 

840:     to fulfill the boundary conditions for perpendicular anchoring of the director 

841:     field (originating

842:     from the colloid quadrupole $Q$) at both interfaces confining the nematic film. 

843:     The distance between any two nearest neighbor quadrupoles is $2h$.

844:    }

845: \end{figure}

846: 

847: As discussed above, the presence of the  colloid asymptotically generates a 

848: quadrupolar director field  

849: which fulfills the boundary condition at the nematic--air interface.

850: In order to fulfill the

851: boundary condition at the substrate--nematic interface, an image quadrupole

852: of the same strength $Q$ is needed which, however, leads to a violation of the

853: nematic--air interface boundary condition and requires a second image quadrupole etc.

854: Continuation of this process leads to a string of image quadrupoles as 

855: depicted in Fig.~\ref{fig:ref1}(b). This string of quadrupoles

856: generates a stress field ${\boldsymbol \pi}$ which vanishes for large lateral distances.

857: Therefore the isotropic pressures must be equal in all volumina:

858: $p_{\rm sub}=p_{\rm air}=p$. From Eq.~(\ref{eq:Ftot1}) one finds that the difference

859: between the force $F$ on the colloid and the integrated stress 

860: $F_\pi$ at the nematic--air interface,

861: \bea

862:  \label{eq:df1}

863:   F- F_\pi &=& \Delta F = - \int_{S_{\rm sub}}  dA \; \pi_{zz}\;,

864: \eea

865: is given by the integrated stress over the substrate surface, i.e., the total force

866: on all quadrupoles above the substrate surface exerted by the image quadrupoles

867: in the substrate. Expressed in terms of the force $F_{Q-Q}$ between two quadrupoles

868: at distance $2h$, we find

869: \bea

870:  -\Delta F / F_{Q-Q} & =&   \sum_{n=1}^\infty \frac{1}{n^6} + \sum_{n=2}^\infty 

871:  \frac{1}{n^6} + \dots \nonumber \\ 

872:     &  = &  \sum_{n=1}^\infty \frac{1}{n^5} = \zeta(5) \approx 1.04  \;.

873: \eea

874: In this equation, the first sum is the total force (divided by $F_{Q-Q}$) on the first

875: quadrupole above the substrate exerted by all quadrupoles in the substrate,

876: the second sum is the total force (divided by $F_{Q-Q}$) on the second quadrupole

877: above the substrate etc. 

878: For the force between two quadrupoles at distance $2h$ we find, 

879: using Eqs.~(\ref{eq:quadrupole}),

880: (\ref{eq:stresstensor}), and (\ref{eq:df1}):

881: \bea

882:  \label{eq:fqq}

883:    F_{Q-Q} & =& \frac{5}{6}\pi\,K\,\frac{Q^2}{h^6}\;.

884: \eea

885: Thus mechanical isolation for the system ``colloid and \linebreak nematic--air interface" 

886: is violated and the total force $\Delta F$ on this system 

887: is (up to the factor 1.04) given by the quadrupole force $F_{Q-Q}$.

888: Nevertheless the magnitude of the corresponding induced logarithmic capillary potential

889: (see Eq.~(\ref{eq:vmen_noniso}) with $\varepsilon_\pi - \varepsilon_F = \Delta F / (2\pi\gamma\rho_0)$) is small, because $Q \sim R^3$ due to dimensional arguments

890: \cite{Lub98}.

891: For parameters similar

892: to the ones appearing in Ref.~\cite{Lav04} 

893: ($R/h\simeq 10^{-1}$, $K/(\gamma R)\simeq 10^{-4}$)

894: we find $V_{\rm men} \simeq 10^{-14}\, k_B T\;\ln(R/d)$, which is 

895: unimportant for the actual intercolloidal interaction. 

896: %

897: %{\tt I have corrected the numerical estimate of the amplitude, which

898: %  is now much smaller than quoted before. I have done the same in the

899: %  next subsection, and in the Conclusions with the power-law

900: %  dependence on $R/h$. It seems that you inadvertendly quoted the

901: %  amplitude of the \mbox{{\em deformation} ($\propto \Delta F$)}

902: %  whenever you really meant the amplitude of the \mbox{{\em

903: %      interaction energy} ($\propto (\Delta F)^2$)}. Please check.}

904:  

905: The elastic potential $V_{\rm el}(d)$ between two colloids 

906: is the interaction between the second quadrupole 

907: and the first quadrupole together with its string of image quadrupoles.

908: Using the solution given in Refs.~\cite{Ram96,Ruh97} we have checked

909: that $V_{\rm el}(d)$ remains repulsive.  For $d < h$ the overall magnitude of $V_{\rm el}(d)$

910: is somewhat weakened,

911: whereas for $d \gg h$ a crossover to $V_{\rm el}(d) \propto \exp(-d/h)$ is observed\footnote{This result can be obtained more easily by solving the field equations 

912: $\Delta n_i=0$ in cylindrical coordinates rather than by using the image quadrupoles.}.   

913: 

914: 

915: \subsubsection{Parallel anchoring at the bottom substrate }

916: 

917: We assume that the substrate induces a preferred in--plane axis for the director 

918: orientation which we take to be the  $x$--axis. With no colloid present at the nematic--air 

919: interface, the equilibrium director field is given by

920: \bea

921:  \label{eq:efield}

922:   \vect n_0=  \begin{pmatrix}  \sin(-q_0 z) \\ 0 \\ \cos(-q_0 z) \end{pmatrix}

923:   \;, \qquad q_0 = \pi /(2 h) \;,

924: \eea 

925: with the consequence that both at the nematic--air and at the nematic--substrate 

926: interface a constant stress is acting:

927: \bea

928:   \pi_{0,zz} = \frac{K}{2}\,q_0^2\;.

929: \eea

930: For the unperturbed interface to be in equilibrium, the air and substrate pressures

931: differ from the pressure in the liquid crystal: $p-p_{\rm sub\,[air]}=

932: \pi_{0,zz}$. 

933: 

934: We now introduce a single colloid at the nematic--air interface in the reference 

935: configuration. 

936: Due to this pressure difference, the force on the colloid

937: and the integrated stress over the nematic--air interface are

938: given  by (see Eq.~(\ref{eq:Ftot1}))

939: \bea

940:  \label{eq:f1}

941:   F\;{\bf e}_z &=& \int_{S_1} d{\bf A}\cdot ({\boldsymbol \pi}-\pi_{0,zz}\,{\bf 1})\;, \\

942:  \label{eq:fpi1}

943:   F_\pi &=& \int_{S_{\rm men}}  dA \; (\pi_{zz}- \pi_{0,zz}) \; ,

944: \eea

945: and the total excess force on the system ``colloid  and nematic--air interface"

946: is determined by

947: \bea

948:  \label{eq:df2}

949:   F- F_\pi &=& \Delta F = - \int_{S_{\rm sub}}  dA \; (\pi_{zz}-\pi_{0,zz})\;.

950: \eea

951: In order to calculate the director field $\vect n$ and the stress tensor ${\boldsymbol\pi}$

952: in the presence of the colloid, we introduce the auxiliary director deformation 

953: fields $v(x,y,z)$ and $w(x,y,z)$ which parametrize the deviations from the unperturbed 

954: director field $\vect n_0$ and which are small at large distances from the colloid:

955: \bea

956:  \label{eq:nlin}

957:   \vect n &=& \begin{pmatrix} \sin(-q_0 z + v)\cos w \\ \sin w \\ \cos(-q_0 z + v)\cos w

958:             \end{pmatrix} \approx \vect n_0 + \\

959:  \nonumber  & &

960:    \begin{pmatrix}  \cos(q_0 z)\;v + \frac{1}{2}\sin(q_0 z)(v^2+w^2) \\ w \\ \sin(q_0 z) \;v -\frac{1}{2}\cos(q_0 z)(v^2+w^2)  \end{pmatrix}

961:   + O((v,w)^3) \;.\quad

962: \eea

963: The first equality in Eq.~(\ref{eq:nlin}) is a general parametrization of

964: the director field $\vect n$ in terms of the auxiliary fields $v,w$ 

965: which fulfills $\vect n^2=1$.

966: These auxiliary fields are taken to be zero at the substrate and the nematic--air interface,

967: i.e., the boundary conditions are $v(x,y,0)=v(x,y,-h)=w(x,y,0)=w(x,y,-h)=0$.

968: The nematic free energy of the film up to order $O(v^2,w^2)$ is obtained

969: by inserting Eq.~(\ref{eq:nlin}) into Eq.~(\ref{eq:f_lc}) for the Frank  free energy

970: after dropping the total divergence of the $K_{24}$--type.

971: Using the boundary conditions for $v$ and $w$ we obtain \cite{Fuk02}

972: (with the notation introduced in Eq.~(\ref{eq:stresstensor})):

973: \bea

974:   {\cal F}^{\rm film}_{\rm ne} = {\cal F}_0 + \frac{K}{2}\int_{V_{\rm film}} d^3r 

975:     \left( v_{,i}^2 + w_{,i}^2 - q_0^2\;w^2 \right)\;.

976: \eea

977: Here, ${\cal F}_0$ is the free energy of the film without colloid. At first

978: glance it is not evident  that

979: this free energy is positive definite. However, the boundary conditions

980: on $v$ and $w$ ensure positivity \cite{Fuk02}. Upon minimization we find

981: $\Delta v=0$, i.e., the deformation field  $v$ of the director 

982: can again be expanded in terms of electrostatic multipoles (Eq.~(\ref{eq:es_mul})).

983: On the other hand, the solution for $w$ must fulfill the Helmholtz equation

984: $(\Delta + q_0^2)w=0$ and can be expanded in terms of multipoles as follows:

985: \bea

986:  \label{eq:hel_mul}

987:   w(r,\theta,\phi)& =& \frac{1}{\sqrt{q_0 r}}\; \sum_{j=0}^\infty\sum_{m=-j}^{j} 

988:   Y_{jm}(\theta,\phi)\; \\

989:   \nonumber  & & \left( w^J_{jm}\; J_{j+1/2}(q_0 r) + w^Y_{jm}\; Y_{j+1/2}(q_0 r) \right)\;

990:     ,

991: \eea

992: where $J[Y]_{j+1/2}(r)$ are the spherical Bessel functions of the first

993: [second] kind and $Y_{jm}(\theta,\phi)$ 

994: are the usual spherical harmonics for the standard set of spherical coordinates

995: $r,\theta,\phi$. 

996: (The origin is again taken as the center of the circular three--phase contact line.)

997: The coefficients $w^{J[Y]}_{jm}$ are dimensionless multipole moments.

998: 

999: The Dirichlet boundary conditions for $v$ and $w$ at the two interfaces can be fulfilled

1000: as before by constructing the full solution in terms of multipoles around the colloid

1001: and corresponding image multipoles as shown in Fig.~\ref{fig:ref1}. 

1002: Since rotational covariance is broken by the parallel substrate anchoring,

1003: the solution for $v$ contains a nonzero dipole contribution.

1004: Nevertheless, the director field

1005: still obeys a reflection symmetry

1006: with respect to the $xz$--plane: $v(x,y,z)=v(x,-y,z)$ and

1007: $w(x,y,z)=-w(x,-y,z)$. Therefore, the

1008: leading asymptotic behavior for $v$  is given by

1009: \bea

1010:  \label{eq:uasy}

1011:   v = P_v \frac{z}{r^3} + Q_v \frac{zx}{r^5} +\dots + v_{\rm image} \;.

1012: \eea

1013: The dipole contribution should vanish for $h\to \infty$. If one

1014: assumes a power--law dependence on $h$, dimensional analysis for 

1015: $P_v$ leads to 

1016: \bea

1017:  \label{eq:kappadef}

1018:  P_v = O(R^2\;(R/h)^\kappa) \qquad (\mbox{with}\; \kappa>0)\;.

1019: \eea

1020: The precise functional form of $P_v$  turns out to be unimportant for 

1021: the subsequent calculations.

1022: We note that an asymptotic solution with a nonvanishing

1023: $x$--component of the dipole moment cannot appear because it would not fulfill

1024: the boundary conditions and the reflection symmetry $w(x,y,z)=-w(x,-y,z)$ excludes

1025: any dipolar contribution in the solution for $w$.\footnote{

1026: In this respect the pictorial argument given in Ref.~\cite{Lav04} (see Fig.~3(b) therein)

1027: is slightly misleading (at least in an asymptotic sense): there the

1028:  assumed director configuration around the colloids for the

1029: case of a finite film thickness is a tilted dipole with a nonvanishing component

1030: $v \propto P_{v,x} x/z^3$. 

1031: Actually, the broken symmetry in $x$--direction rather

1032: enters through a tilted quadrupole.}  Therefore the leading asymptotic

1033: behavior of $w$ takes the form

1034: \bea

1035:   w &=&  \frac{zy}{r^2}\;\frac{1}{\sqrt{q_0 r}} \left( Q^J_w\;J_{5/2}(q_0 r) +

1036:    Q^Y_w \;Y_{5/2}(q_0 r) \right) + \dots 

1037:   \nonumber \\&& + w_{\rm image} \;.

1038:  \label{eq:vasy1}

1039: \eea

1040: We note that there are two quadrupolar contributions (with dimensionless moments

1041: $Q^J_w$ and $Q^Y_w$)

1042:  which show an oscillatory behavior 

1043: for radial distances $r \gg h$ (i.e., $q_0 r \gg 1$). On the other hand, for 

1044: large film thicknesses

1045: there is an intermediate asymptotic regime $R \ll r \ll h$:

1046: \bea

1047:  \label{eq:vasy2}

1048:   w (q_0 R \ll q_0 r \ll 1) &=& \frac{zy}{r^2}\;\sqrt{\frac{2}{\pi}} \left(

1049:     Q^J_w\;\frac{(q_0 r)^2}{15} -   Q^Y_w \; \frac{3}{(q_0 r)^3} \right)

1050:  \nonumber \\

1051:   && + \dots + 

1052:    w_{\rm image} \;.

1053: \eea

1054: For large film thicknesses the solution for the director field

1055: near the nematic--air interface in the regime $R \ll r \ll h$ should coincide

1056: with the solution for macroscopically thick films (Eq.~(\ref{eq:quadrupole})).

1057: Near the nematic--air interface, one has $v\approx n_x$, $w\approx n_y$. Therefore

1058: one recovers the rotationally covariant quadrupole solution of Eq.~(\ref{eq:quadrupole})

1059: for  $Q_v=Q$ (obtained by equating Eq.~(\ref{eq:uasy}) with Eq.~(\ref{eq:quadrupole})

1060: for $i=1$, i.e., $r_1=x$)  and $-3\sqrt{\di 2/\pi}\,Q^Y_w/q_0^3=Q$ (obtained by equating 

1061: Eq.~(\ref{eq:uasy}) with Eq.~(\ref{eq:quadrupole}) for $i=2$, i.e., $r_2=y$).

1062: Since $Q^Y_w \sim q_0^3 Q$ and $Q = O(R^3)$, $Q^Y_w=O([q_0 R]^3)$ is

1063: a very small number.

1064: 

1065: The Dirichlet boundary conditions for $w$ at the substrate and at the 

1066: nematic--air interface

1067: enforce that the contribution to $w$ due to the quadrupole $Q^J_w$ and all corresponding image

1068: quadrupoles is zero. This holds also for the contribution of  all higher multipoles of 

1069: degree $j$ 

1070: for which the  radial dependence is given by $J_{j+1/2}(r)$. 

1071: The only solution of the Helmholtz equation

1072: which fulfills the boundary conditions and which, as an additional requirement, 

1073: is smooth everywhere,

1074: is $w\equiv 0$. Since the Bessel functions of the first kind are smooth

1075: everywhere, all respective multipole moments must be zero.

1076: (This does not hold for the multipole moments pertaining to the Bessel functions

1077: of the second kind since $Y_{j+1/2}(r)$ is singular at $r=0$.) 

1078: % Note that this is valid

1079: %even for more general boundary conditions at the substrate derived from

1080: %Eq.~(\ref{eq:fs_lc}).

1081: 

1082: The excess force on the system ``colloid and nematic--air interface" follows from

1083: Eq.~(\ref{eq:df2}) and can be expressed as

1084: \bea

1085:  \label{eq:stress_twisted}

1086:  \Delta F = - \frac{K}{2}\int_{S_{\rm sub}} dA\; (v_{,z}^2-2q_0\,v_{,z}+w_{,z}^2) \;,

1087: \eea

1088: utilizing the boundary conditions %and symmetries of 

1089: for the solutions $v$ and $w$.

1090: For the multipoles appearing in the expansion of $v$ the method described in 

1091: Subsec.~\ref{sec:aligned}

1092: may be used. In order to obtain  the quadrupolar contributions to $w$ we perform the summation over the

1093: image multipoles and the integration over the substrate surface numerically. Due to the

1094: slow decay of $Y_{5/2}(r) \propto (\cos r)/r^{1/2}$ the stress integral is superficially

1095: divergent. However, a detailed asymptotic analysis yields convergence with the result  

1096: \bea

1097:  -\Delta F = \pi \,K \left( \frac{3\zeta(3)}{2}\,\frac{P_v^2}{h^4} + \frac{5\zeta(5)}{12}\,

1098:   \frac{Q_v^2}{h^6} + c_w\,(Q^Y_w)^2 \right) \;,\quad

1099: \eea

1100: where $c_w\approx 0.30$ has been determined numerically. 

1101: %{\tt I have rewritten the following sentence:} 

1102: Note that the linear contribution

1103: %s in $P_v$ and $Q_v$ 

1104: due to the term $\sim q_0\,v_{,z}$

1105: in the stress tensor (Eq.~(\ref{eq:stress_twisted})) turns out to be zero, as can be 

1106: easily checked by applying Gauss' theorem and the field equation $\Delta v = 0$.

1107: Using the above considerations

1108: concerning the magnitude of the multipole moments $P_v$, $Q_v$ and $Q^Y_w$ we obtain

1109: \bea

1110:  \label{eq:ftwist}

1111:  -\frac{\Delta F}{K} \sim a_P \,\left(\frac{R}{h}\right)^{4+2\kappa} + 

1112:   a_Q\, \left(\frac{R}{h}\right)^6\;,

1113: \eea

1114: where $a_P$ and $a_Q$ are numerical coefficients of order 1 and $\kappa$ has been

1115: defined in Eq.~(\ref{eq:kappadef}). Again the excess force

1116: on the colloid falls off rapidly with increasing film thickness such that the induced 

1117: logarithmic capillary interaction (see Eq.~(\ref{eq:vmen_noniso})) remains very weak.

1118: For the parameters characterizing the experimental system studied in Ref.~\cite{Lav04} 

1119: ($R/h\sim 10^{-1}$, $K/(\gamma R)\sim 10^{-4}$)

1120: we find $V_{\rm men}(d) \sim$ \linebreak $10^{-11-4\kappa}\, k_B T\;\ln(R/d)$ 

1121: which appears to be undetectably small.

1122: Note that for $d<h$ the direct elastic repulsion remains essentially unchanged

1123: because in this regime the leading term of the elastic interaction is given by the

1124: repulsion between the two quadrupoles located at the colloid sites. In this case the image

1125: quadrupoles can be neglected. 

1126: 

1127: 

1128: 

1129: 

1130: \section{Discussion and conclusion}

1131: \label{sec:conclusion}

1132: 

1133: We have investigated the effective potential between two colloidal microspheres

1134: of radius $R$

1135: floating at asymptotically large distances $d$ on an interface between a nematic film 

1136: of thickness $h$ and air (Fig.~\ref{fig:nem_exp}).

1137: This effective potential is the sum of an elastic interaction caused by

1138: the director distortions around the colloids and of a capillary interaction mediated

1139: by surface deformations. We have analyzed the effective potential for large $d$ employing the

1140: coarse--grained Frank free energy (within the one constant approximation)

1141: for the director distortions and the

1142: linearized Young--Laplace equation for the interface distortions.   

1143: 

1144: In the case of a macroscopically thick nematic film, the director deformation

1145: around a single colloid is of quadrupolar type. Thus the induced elastic interaction

1146: between two colloids at distance $d$ is repulsive and of quadrupolar type  $\propto d^{-5}$.

1147: The capillary interaction is also repulsive and decays $\propto d^{-5}$ but it is much

1148: weaker than the elastic interaction. This rapid decay is a consequence of the mechanical 

1149: isolation of the system,

1150: %

1151: i.e., of the fact that the net force on the colloidal particles and

1152: the surrounding interface vanishes.

1153: 

1154: A finite film thickness $h$ leads to a violation of the mechanical isolation of the

1155: system ``colloid and interface". (The excess force is counteracted by the film substrate

1156: such that the whole experimental system is of course in mechanical equilibrium.)

1157: However, on the thermal energy scale the strength of the ensuing logarithmic 

1158: capillary potential turns out to be very small. 

1159: In the case of homeotropic boundary conditions on both sides of the nematic film

1160: the strength is proportional to $(R/h)^{12}$ (see Eq.~(\ref{eq:fqq}) and the subsequent discussion).

1161:  For  twisted boundary conditions

1162: (parallel at the bottom substrate, perpendicular at the upper nematic--air interface)

1163: there is a qualitatively different asymptotic behavior of the director field

1164: due to a dipole contribution, which is induced by the broken rotational 

1165: symmetry and vanishes for $h\to \infty$.

1166: However, even for these boundary conditions 

1167: %{\tt Here I have dropped the following sentence: ``since this dipole contribution induced by the finite film thickness

1168: %must vanish for $h\to \infty$'', because the amplitude of the log would also vanish even if the dipole would not (Eq.~(43) with $\kappa=0$).}

1169: the strength of the logarithmically varying 

1170: capillary potential remains 

1171: extremely

1172: small, vanishing at least $ \propto (R/h)^8$ (see Eq.~(\ref{eq:ftwist}) and the subsequent discussion).

1173: % and thus seemingly does not offer an explanation of 

1174: %the recent experimental observations made in Ref.~\cite{Lav04}.

1175: 

1176: Thus our analysis based on the mechanical isolation of the

1177: experimental system under consideration rules

1178: out a significant attractive contribution $\sim \ln d$ of capillary type 

1179: to the  effective potential

1180: between two colloids at a nematic interface.

1181: The amplitude of such a logarithmic contribution vanishes rapidly 

1182: for large  film thickness $h$. Therefore, this effective pair potential 

1183: does not provide a mechanism for the stability of isolated colloid clusters

1184: at a nematic interface as reported in Ref.~\cite{Lav04}.

1185: %We have demonstrated that the explanation offered in Ref.~\cite{Lav04}

1186: %for the , based on an effective 

1187: %pair potential between

1188: %the colloids, is not valid.   Also the elastic contribution of dipolar type

1189: %($\propto d^{-3}$) to the effective pair potential is not backed by our asymptotic 

1190: %analysis.   

1191: %Ref.~\cite{Lav04}

1192: %derives an $h$--independent amplitude because only part of the forces

1193: %contributing to the net mechanical balance are identified. 

1194: 

1195: 

1196: Naturally one strives for other explanations for the observation of stable clusters

1197: reported in Ref.~\cite{Lav04}. 

1198: There the mutual center--to--center distance between 

1199: neighboring colloids in the cluster has been found to be between $3R$ and $5R$, 

1200: depending on the

1201: radius $R$ of the colloids. If the cluster stability is attributed to a minimum

1202: in the effective pair potential at this range $3R \dots 5R$ of distances, 

1203: the applicability of the above asymptotic considerations at such distances is doubtful 

1204: for both the elastic and capillary

1205: contributions to the effective pair potential: 

1206: \begin{itemize}

1207:  \item {\em Elastic part:} For the single--colloid problem, the asymptotic, quadrupolar

1208:   form of the director field (Eq.~(\ref{eq:quadrupole})) is based on the assumption

1209:   that the deviations of the director from the preferred direction $\vect e_z$ are small: \linebreak

1210:   $|n_{x[y]}| \ll 1$. At a radial distance $\rho$ from the center

1211:   of the colloid this implies $Q/\rho^3 \ll 1$, which seems to be fulfilled for

1212:   the dimensional estimate for the quadrupole moment $Q = O(R^3)$ and for the

1213:   distances $\rho = 3R \dots 5R$ under discussion. However, the absolute magnitude of the

1214:   director deformations is not fixed by these arguments, so that it might be

1215:   worthwhile to determine the minimum of the Frank free energy 

1216:   by a full numerical calculation in the presence of one or two colloids at

1217:   the nematic--air interface, similar to Ref.~\cite{Sta99} where  the explicit 

1218:   director solution around a single colloid in the bulk has been determined.   

1219:   Our analysis indicates that 

1220:   only if the multipolar expansion of the director field around a single trapped colloid

1221:   fails for distances $\rho = 3R \dots 5R$, there is a chance for attractive elastic

1222:   interactions between two colloids at these distances. 

1223:   Failure of the multipole expansion at these intermediate 

1224:   distances would point to a strong deviation of the magnitude of the 

1225:   multipole moments $M_n$ of order $n$ from the dimensional estimate

1226:   $M_n \sim R^{n+1}$. Thus it is certainly worthwhile to determine

1227:   multipole renormalizations for bulk and interface configurations, taking into

1228:   account the full nonlinearity of the director field equations.

1229: 

1230:  \item {\em Capillary part:} Whenever the colloid--induced nematic stress deviates from the

1231:   asymptotic, quadrupolar form (Eq.~(\ref{eq:pizz_single})), corrections

1232:   to the asymptotic capillary potential will

1233:   arise. However, due to the smallness of the dimensionless

1234:   force $\varepsilon_F$ (see Eq.~(\ref{eq:epsdef}), $\varepsilon_F \propto K/(\gamma R) \approx 10^{-3}$) 

1235:   the perturbative treatment of the ensuing capillary deformations

1236:   is justified. Therefore, one of our main findings, i.e., that the magnitude

1237:   of the capillary potential ($\propto \varepsilon_F^2$) is always smaller

1238:   than the magnitude of the elastic interaction ($\propto |\varepsilon_F|$), 

1239:   is likely to hold also for elastic stresses deviating from the

1240:   asymptotic limit.

1241: %{\tt In case we decide to answer also the claim with the torques by

1242: %  Lavrentovich et al.}: 

1243: 

1244: Another possible, subtle effect related to the 

1245: anisotropy inherent to nematic phases

1246: allows the substrate to exert also a torque on the colloidal particles

1247: and the surrounding interface. This case can be studied similarly

1248: to what we have done here. Actually, the result can be obtained 

1249: straightforwardly if the analogy of capillary deformation with 2D

1250: electrostatics is employed (see, e.g., Ref.~\cite{Dom06}): The net

1251: force creates ``capillary monopoles'' $Q_{\rm cap}$ with an interaction

1252: energy $V_{\rm men}(d) = Q_{\rm cap}^2/(2\pi\gamma)\, \ln( C\lambda/d)$ 

1253: (see Eq.~(\ref{eq:vmen_noniso})). A net torque creates

1254: ``capillary dipoles'' ${\bf P}_{\rm cap}$ and the interaction energy

1255: between them is $V_{\rm men}(d) \sim (|{\bf P}_{\rm cap}|/d)^2$. This decay is

1256: sufficiently

1257: slow to dominate asymptotically the

1258: nematic--mediated repulsion. However, dimensional analysis yields $P_{\rm cap}

1259: \sim R\, Q_{\rm cap}$ so that the amplitude of

1260: this capillary energy, which must vanish for large film thicknesses $h\to\infty$, 

1261: has also a

1262: too small numerical value to explain the experimental results reported in Ref.~\cite{Lav04}.

1263: \end{itemize}

1264: %

1265: %First, we want to emphasize that our strictly asymptotic

1266: %considerations cannot exclude the possibility of stronger,

1267: %non--logarithmical capillary attractions when the colloids are close

1268: %to each other.

1269: %

1270: %Actually, we obtain indications that beyond the small--deformation

1271: %regime of validity of our calculations, a capillary attraction might

1272: %arise which would dominate asymptotically over the nematic--mediated

1273: %repulsion. There is, however, no evidence that the experiments

1274: %conducted in Ref.~\cite{Lav04} involve large interfacial deformations.

1275: 

1276: We recall that for simplicity our calculations have been carried out in the ``strong

1277: anchoring'' limit. ``Weak anchoring'' would lead to a

1278: different director field, and thus to different values of the

1279: dimensionless forces $\varepsilon_F$ and $\varepsilon_\pi$ (see

1280: Eq.~(\ref{eq:epsdef})). It would also lead to an explicit contribution to the

1281: expression~(\ref{eq:F}) for the

1282: free energy accounting for the anchoring ``wetting''

1283: energies~(Eq.~(\ref{eq:fs_lc})), which in the ``strong

1284: anchoring'' limit considered so far were subsumed in a (quantitatively

1285: negligible) additive renormalization of the surface tensions. The consequences of 

1286: this more general case of no strong anchoring have to

1287: be explored yet, but we do not expect that this alters our

1288: conclusion that a logarithmically varying capillary attraction is

1289: ruled out by mechanical isolation ($\varepsilon_F = \varepsilon_\pi$),

1290: because this latter result is based on very general principles.

1291: 

1292: 

1293: Finally we note that for a two--dimensional system, the interactions between 

1294: two upright circular colloids trapped at the nematic--isotropic interface have been

1295: studied numerically \cite{Tas05} using a Landau--de Gennes free energy. 

1296: In three dimensions, this corresponds to

1297: long, cylindrical colloids which are aligned side--to--side at the interface.

1298: For boundary conditions that yield a similar, quadrupolar behavior of the director 

1299: field around a single colloid \cite{Pet98}, 

1300: the effective interaction is found to be repulsive

1301: and consistent with a power--law decay (for intermediate distance $d/R$ up to 7). 

1302: In this particular two--dimensional situation, the repulsive interactions appear

1303: to be longer--ranged $\propto d^{-1}$ and in the numerical results 

1304: for the director field there is no trace of a sizeable interface

1305: deformation which would lead to capillary attractions.

1306: 

1307: In summary, we have presented several arguments 

1308: % that point to the

1309: % absence of an effective potential between two colloids at a nematic interface

1310: % which consists of a repulsive elastic and an attractive capillary part. 

1311: which rule out an asymptotic attraction of capillary origin in the

1312: effective interaction potential between two colloids at a nematic interface.

1313: If one discards effective pair potentials as the source of the

1314: stability of colloidal mesostructures as the ones found in Ref.~\cite{Lav04}, 

1315: the question arises whether this stability is a consequence

1316: of genuine many--body effects \cite{Lav07a}. Nematic surfaces (without colloids)

1317: can stabilize regular patterns of surface defects \cite{deG74}. If these  defects

1318: persist in the presence of colloids, the experimentally observed regular order of 

1319: the colloids might

1320: be attributed to them. This issue calls for further experimental and theoretical

1321: investigations.   

1322: 

1323: 

1324: 

1325: {\bf Acknowledgment:}

1326: We acknowledge useful discussions with S.~Chernyshuk, O.~Lavrentovich,

1327: B.~Lev, P.~Patr\'icio, and J.~M.~Romero--Enrique. M.~O. thanks the German 

1328: Science Foundation (DFG) for financial

1329: support through the Collaborative Research Centre (SFB-TR6) 

1330: ``Colloids in External Fields", project section D6-NWG.

1331: %

1332: A.~D. acknowledges financial support from the Junta de Andaluc{\'\i}a

1333: (Spain) through the program ``Retorno de Investigadores''.

1334: 

1335: 

1336: 

1337: 

1338: 

1339: 

1340: \begin{appendix}

1341: 

1342: \section{Corrections to the strong anchoring limit}

1343: 

1344: \label{sec:app1}

1345: 

1346: Here we consider the asymptotic director field around a single colloid at the nematic--air

1347: interface in the case of finite anchoring strength. For large distances from the colloid,

1348: the overall deviations from the preferred director $\vect n =(n_1,n_2,n_3) 

1349: = (0,0,1)$ are small

1350: and the total free energy of the nematic phase is given by the harmonic approximation

1351: for the bulk part (Eq.~(\ref{eq:fasy})) and the Poulini expression for the surface part:

1352: \bea

1353:   {\cal F}_{\rm ne} & = & \frac{K}{2} \spaceint{V_{\rm ne}}{r} 

1354:      \sum_{i=1,2} \left( \nabla n_i \right)^2 +

1355:      \frac{W_1}{2} \int_{A_{\rm air-ne}}\!\!\! dA \;

1356:    (\vect n \cdot \vect e_A)^2 \nonumber  \\

1357:    & =& \frac{K}{2} \spaceint{V_{\rm ne}}{r} \left( \left( \nabla n_1 \right)^2 +

1358:   \left( \nabla n_2 \right)^2 \right) + 

1359:  \nonumber \\ &&  

1360:   \frac{W_1}{2} \int_{A_{\rm air-ne}}\!\!\! dA \;

1361:   (1-n_1^2-n_2^2)  \;.

1362: \eea   

1363: The second equation holds because the normal $\vect e_A$ of the nematic--air interface

1364: $A_{\rm air-ne}$ is parallel to the $z$--direction and $\vect n^2=1$. 

1365: For particles with radii $R$ in the $\mu$m range, such as those investigated 

1366: in Ref.~\cite{Lav04}, 

1367: the dimensionless parameter $\alpha = K/(|W_1| R)$ is smaller than 1. Minimizing

1368: the free energy renders the Laplace equation in the bulk,

1369: \bea

1370:    \Delta n_i = 0 \qquad (i=1,2)\,,

1371: \eea 

1372: supplemented with the Robin boundary condition:

1373: \bea

1374: %     K \frac{\partial}{\partial z} n_i (x,y,z=0) - W_1 n_i (x,y,z=0) & = & 0 

1375: %    \quad \to \nonumber \\

1376:    \alpha R\, \frac{\partial}{\partial z} n_i (x,y,z=0) - n_i (x,y,z=0) & = & 0

1377:   \quad (i=1,2) \;. \nonumber \\

1378:  \label{eq:boun}

1379: \eea

1380: The general, asymptotic solution  is given by the multipole {\em ansatz} 

1381: (Eq.~(\ref{eq:es_mul}))

1382: \bea

1383:   n_i &=& q_i\,\frac{1}{r} + \sum_{\beta=1}^3 P_{i\beta}\,\frac{r_\beta}{r^3} +  

1384:       \sum_{\beta,\gamma=1}^3 Q_{i\beta\gamma}\,\frac{r_\beta\,r_\gamma}{r^5}

1385:  + \dots

1386: \eea

1387: with the requirement of rotational covariance around the $z$--axis. In view of

1388: the boundary condition (Eq.~(\ref{eq:boun})) we note that the derivative

1389: in Eq.~(\ref{eq:boun}) applied to a multipole of order $m$ generates 

1390: a multipole term of order

1391: $m+1$. Thus, to leading order in $1/r$, the boundary condition can only be met if

1392: the leading multipole ($n_i^{\ell}$) fulfills the strong anchoring condition

1393: $n_i^{\ell}(z=0)=0$ and is accompanied by a subleading multipole ($n_i^{s\ell}$)

1394: which is connected to the leading multipole through the condition

1395: $\alpha R\, \partial_z n_i^{\ell}(z=0) = n_i^{s\ell} (z=0)$.

1396: Both the monopole and the dipole do not fulfill the strong anchoring condition

1397: $n_i^{\ell}(z=0)=0$.

1398: Therefore, as before, we find that the multipole expansion starts with the quadrupole solution

1399: and that it is necessarily accompanied by a hexapole:

1400: \bea

1401:  \label{eq:app_sol1}

1402:   n_i & = & n_i^{\ell} + n_i^{s\ell} \;, \\

1403:   n_i^{\ell} &=& Q\,\frac{z r_i}{r^5} + Q^{\rm mag} \epsilon_{i\alpha 3}\, \frac{z r_\alpha}{r^5}\;, \\

1404: %  \mbox{and} \quad 

1405:   n_i^{s\ell} &=& \alpha QR\, \frac{r_i(r^2-5z^2)}{r^7} +

1406:                \alpha Q^{\rm mag}R \epsilon_{i\alpha 3}\, \frac{r_\alpha(r^2-5z^2)}{r^7}

1407:   \;. \nonumber \\

1408:  \label{eq:app_sol2}

1409: \eea

1410: Since $\partial_z n_i^{s\ell}(z=0)=0$, this solution 

1411: (Eqs.~(\ref{eq:app_sol1})--(\ref{eq:app_sol2})) exactly satisfies the boundary condition in 

1412: Eq.~(\ref{eq:boun}). Thus

1413: the asymptotically dominant director field consists of a quadrupole--hexapole

1414: superposition. 

1415: Furthermore, the magnitude of the accompanying hexapole moment is small due to

1416: the factor $\alpha$, therefore the results reported before for strong anchoring 

1417: (based on the

1418: leading quadrupole only) are unaffected by a finite anchoring strength at the nematic--air

1419: interface.   

1420: 

1421: 

1422: 

1423: \section{Force balance in a general configuration} 

1424: 

1425: \label{sec:app2}

1426: 

1427: \begin{figure}[t]

1428:  \begin{center}

1429:   \epsfig{file=forcebalance.eps, width=\columnwidth}

1430:  \end{center}

1431:  \caption{ 

1432:    \label{fig:forcebalance}

1433:    A general configuration of the particle and the fluid interface.

1434:    The arrows indicate the orientation of the corresponding surfaces:

1435:    $S_{\rm men}$ is the interface, $S^{1(2)}_{\rm part}$ is the

1436:    surface of the particle in contact with the lower (upper) fluid

1437:    phase, $S_{\rm wall}$ is the surface of the container of the system

1438:    (sketched here as a quadrangular box for simplicity). The

1439:    three--phase contact line between the particle and the interface is

1440:    denoted as $C_0$.}

1441: \end{figure}

1442: 

1443: For the benefit of the reader, we discuss in this Appendix some

1444: previous results \cite{Dom05,Dom06} concerning the  force balance of a general

1445: equilibrium configuration and demonstrate how the amplitude of an 

1446: asymptotic, logarithmically varying interfacial deformation is

1447: determined solely by this mechanical condition of force balance.

1448: 

1449: Figure~\ref{fig:forcebalance} represents a colloidal particle in equilibrium at

1450: the deformed interface; in general the deformation is not small. 

1451: The condition of mechanical equilibrium implies that locally

1452: %not only that the net force on the whole system , but also 

1453: the net force on any part of the system must vanish. Thus one

1454: has:

1455: \begin{enumerate}

1456: \item Each of the fluid phases is in equilibrium. 

1457: %The lower (nematic) phase 

1458: %   , so that the integral of the

1459: %   stress ${\mathbf \Pi}$ over the surface enclosing the phase must

1460: %   vanish:

1461: %   \begin{equation}

1462: %     \int_{S_{\rm men} \cup S^{\rm wall}_{1} \cup (\mbox{}-S_1)}

1463: %     d{\bf A} \cdot {\mathbf \Pi} = {\bf 0} .

1464: %   \end{equation}

1465:   We introduce the %(elastic and hydrostatic)

1466:   forces exerted {\em by} each fluid phase  on the particle, on the whole

1467:   fluid meniscus, and on the wall of the container as

1468:   %

1469:   \bea

1470:   \label{eq:fpart}

1471:   {\mathbf F}_{\rm part}^{1(2)} & := & \int_{S_{\rm part}^{1(2)}} d{\bf A} \cdot {\mathbf \Pi}^{1(2)} , \\

1472:   \label{eq:fint}

1473:   {\mathbf F}_{\rm men}^{1(2)} & := & \mbox{} -(+) \int_{S_{\rm men}} d{\bf A} \cdot {\mathbf \Pi}^{1(2)} , \\

1474:   {\mathbf F}_{\rm wall}^{1(2)} & := & \mbox{} - \int_{S_{\rm wall}^{1(2)}} d{\bf A} \cdot {\mathbf \Pi}^{1(2)} ,  

1475:   \eea

1476:   % 

1477:   respectively, in terms of the stress tensor ${\bf \Pi}^{1(2)}({\bf r})$ in each

1478:   fluid phase with due account for the orientation of the surfaces

1479:   (see Fig.~\ref{fig:forcebalance}). The superscript $^{1(2)}$

1480:   indicates the lower (upper) phase in Fig.~\ref{fig:forcebalance}. 

1481:   The condition of mechanical equilibrium

1482:   of each phase under the influence of these three forces reads

1483:   \begin{equation}

1484:     \label{eq:netFfluid}

1485:     {\mathbf F}_{\rm part}^{1(2)} + {\mathbf F}_{\rm men}^{1(2)} +

1486:     {\mathbf F}_{\rm wall}^{1(2)} = {\bf 0} .

1487:   \end{equation}

1488: 

1489:   The total force exerted {\em by} the fluids on the particle is

1490:   ${\mathbf F}_{\rm part} := {\mathbf F}_{\rm part}^{1} + {\mathbf

1491:     F}_{\rm part}^{2}$, and on the meniscus it is ${\mathbf F}_{\rm

1492:     men} := {\mathbf F}_{\rm men}^{1} + {\mathbf F}_{\rm men}^{2}$.

1493:   The expressions for these forces reduce to those given in Eqs.~(\ref{eq:Fdef}) and

1494:   (\ref{eq:Fpi}), respectively, upon evaluation in the reference 

1495:    configuration depicted

1496:   in Fig.~\ref{fig:ref}.

1497:   %

1498: %   {\tt Note: strictly speaking, Eq.(\ref{eq:Fpi}) should be defined

1499: %     with the opposite sign}

1500:   

1501:   If the condition of mechanical equilibrium is applied 

1502:   locally to an infinitesimal volume in

1503:   the bulk of each of the fluid phases, it turns into $\nabla\cdot{\bf

1504:     \Pi}^{1(2)} = {\bf 0}$. In the nematic phase this condition yields

1505:   the field equations determining the director field.

1506: 

1507: \item The particle is in mechanical

1508:   equilibrium under the combined action of ${\mathbf F}_{\rm part}$

1509:   and the tension exerted {\em by} the interface on the particle at the

1510:   three--phase contact line $C_{0}$. This tension can be expressed in terms of a line

1511:   integral involving the surface tension $\gamma$:

1512:   \begin{equation}

1513:     {\mathbf F}_{\rm contact} := \mbox{} - \gamma \oint_{C_0} d\ell 

1514:     \,{\mathbf e}_c , %\times {\mathbf e}_n ,

1515:   \end{equation}

1516:   where ${\mathbf e}_c$ is the unit vector tangent to the interface,

1517:   normal to the contact line, and oriented towards the particle side. 

1518:   Therefore, the condition of mechanical equilibrium reads

1519:   \begin{equation}

1520:     \label{eq:netFpart}

1521:     {\mathbf F}_{\rm part} + {\mathbf F}_{\rm contact} = {\bf 0} .

1522:   \end{equation}    

1523: 

1524: \item Any piece $S_{\rm int} \subset S_{\rm men}$ of the fluid

1525:   interface is in mechanical equilibrium. The force on $S_{\rm int}$

1526:   exerted by the fluid phases and the tension exerted on this piece at

1527:   its boundary, $\partial S_{\rm int}$, are balanced:

1528:   \begin{equation}

1529:     \label{eq:netFint}

1530:     \int_{S_{\rm int}} d{\bf A} \cdot 

1531:     [{\mathbf \Pi}^2 - {\mathbf \Pi}^1]

1532:     + \gamma \oint_{\partial S_{\rm int}} d\ell \,{\mathbf e}_c 

1533:     = {\bf 0} ,

1534:   \end{equation}

1535:   with ${\mathbf e}_c$ oriented towards the exterior of $S_{\rm int}$.

1536:   If this expression is applied locally to an infinitesimal piece of

1537:   interface, it turns into an equation for the interfacial deformation

1538:   relating the mean curvature of the interface to the pressure jump

1539:   accross it. If, in addition, the interface deviates only slightly

1540:   from a flat interface (identified with the plane $z=0$), this equation reduces

1541:   in turn to the well--known equation~(\ref{eq:linearYLeq}) below

1542:   %$\gamma \nabla^2 u = \Pi_{zz}^1 - \Pi_{zz}^2$ 

1543:   for the local height $u(x,y)$ over the plane $z=0$.

1544: 

1545: \end{enumerate}

1546: 

1547: The net force balance of the whole system follows %of course 

1548: from the three separate balance conditions~(\ref{eq:netFfluid}),

1549: (\ref{eq:netFpart}), and (\ref{eq:netFint}): with $S_{\rm int} =

1550: S_{\rm men}$ (so that $\partial S_{\rm int} = C_0 \cup C_{\rm wall}$;

1551: $C_{\rm wall}$ is the three--phase contact line between phase 1, phase 2, and the

1552: container enclosing the system),

1553: one finds that, as expected, in equilibrium the net force of the outer environment on

1554: the system must vanish\footnote{The reasoning can be

1555:   easily generalized to the case that in addition to the

1556:   short--ranged influence of the wall there are also external fields

1557:   (gravity, electric force) acting on any part of the system.}:

1558: % on the fluid phases by the wall of the container

1559: % must be balanced by a tension at the contact line of the fluid

1560: % interface with the wall

1561: \begin{equation}

1562:   \label{eq:netFtotal}

1563:   {\mathbf F}_{\rm wall}^{1} + {\mathbf F}_{\rm wall}^{2} -

1564:   \gamma \oint_{C_{\rm wall}} d\ell \,{\mathbf e}_c = {\bf 0} ,

1565: \end{equation}

1566: where ${\bf e}_c$ points to the exterior of the system.

1567: 

1568: The condition of mechanical equilibrium can be applied advantageously

1569: to obtain a precise statement about the amplitude of an

1570: interfacial deformation $u(\rho,\phi)$ varying logarithmically with

1571: the lateral distance $\rho \gg R$ from the particle  with radius $R$. 

1572: Far away from the particle, interface deformations and their gradients are

1573: small, so that the linearized equation holds.

1574: %Since a logarithmic dependence

1575: %is a solution of the 2D Laplace equation, it can only arise provided

1576: %there is a region where the deformation from a flat interface is small

1577: %and accordingly the linearized equation holds.

1578: % and there

1579: % is no pressure force deforming the interface, so that $\gamma \nabla^2

1580: % u =0$. Physically, this requires a clear spatial separation between

1581: % the regions where the forces deforming the interface are applied, in

1582: % our case the neighbourhood of the particle and that one of the wall.

1583: Thus there is a distance $\xi$ beyond which the linear theory is applicable:

1584: \begin{equation}

1585:   \label{eq:linearYLeq}

1586:   \gamma \nabla^2 u (\vect r) = \Pi_{zz}^1 (\vect r) - \Pi_{zz}^2(\vect r)  \qquad

1587:   ({\bf r} \in S_{\rm ext}) \;,  %\xi < \rho ,

1588: \end{equation}

1589: where the piece of interface $S_{\rm ext}$ is enclosed by the

1590: circle $\rho=\xi$ and the line $C_{\rm wall}$.  The general solution 

1591: to this inhomogeneous Laplace equation can be written as

1592: \begin{eqnarray}

1593:   \label{eq:generalu}

1594:   u(\rho,\phi) & = & A_0 + B_0 \ln\frac{\xi}{\rho} +  \\ && 

1595:   \sum_{m=1}^{+\infty} \left[ A_m \left(\frac{\rho}{\xi}\right)^m 

1596:     + B_m \left(\frac{\xi}{\rho}\right)^m \right] \cos m(\phi-\phi_m) +

1597:    \nonumber\\

1598:   & &  \frac{1}{2\pi\gamma} \int_{S_{\rm ext}} \!\!\! d\phi' d\rho' \rho'\;

1599:   [\Pi_{zz}^1({\bf r}') - \Pi_{zz}^2({\bf r}')] \, 

1600:   \ln\frac{|{\bf r}-{\bf r}'|}{\rho}\; . \nonumber

1601: \end{eqnarray}

1602: The fixed values of the constants $A_m$, $B_m$, and $\phi_m$ are determined by

1603: the boundary conditions. This expression reduces to

1604: Eq.~(\ref{eq:sol1}) in the particular case of rotational symmetry and

1605: a wall located at infinity.

1606: 

1607: We can apply Eq.~(\ref{eq:netFint}) to the piece $S_{\rm int}=S_{\rm

1608:   men}\backslash S_{\rm ext}$ enclosed by the contact line $C_0$

1609: and the circle $\rho=\xi$. Invoking Eq.~(\ref{eq:netFpart}), one has

1610: \bea

1611:   {\bf F}_{\rm part} + \int_{S_{\rm int}} d{\bf A} \cdot [{\mathbf \Pi}^2 - {\mathbf \Pi}^1]

1612:   &=& \mbox{} - \gamma \oint_{\rho = \xi} d\ell \,{\mathbf e}_c

1613:   \\ \nonumber

1614:   &\approx& \mbox{} - \gamma \, {\bf e}_z \int_0^{2\pi} d\phi \; \rho 

1615:   \left.\frac{\partial u}{\partial \rho} \right|_{\rho=\xi} ,

1616: \eea

1617: where the last, approximate equality involves the leading order term in

1618: $\nabla u$ of the line integral. (This approximation is justified because the interface 

1619: deviates only slightly from a flat interface at the circle $\rho=\xi$.) 

1620: Evaluating this latter term for the general

1621: solution~(\ref{eq:generalu}) one finally finds that the amplitude of

1622: the logarithmic term in Eq.~(\ref{eq:generalu}) is proportional to the

1623: force exerted by the upper and the lower fluid on the particle and on the meniscus:

1624: \begin{equation}

1625:   \label{eq:amplitudelog}

1626:   B_0 = \frac{1}{2\pi\gamma} {\bf e}_z \cdot \left[ 

1627:     {\bf F}_{\rm part} + {\bf F}_{\rm men} \right] .

1628: \end{equation}

1629: In obtaining Eq.~(\ref{eq:amplitudelog}) we have used that

1630: in the region $S_{\rm ext}$ (where the interface is almost flat)

1631: one has to leading order

1632: \begin{equation}

1633:   \int_{S_{\rm ext}} d{\bf A} \cdot [{\mathbf \Pi}^2 - {\mathbf \Pi}^1]

1634:   \cdot {\bf e}_z

1635:   \approx 

1636:   \int_{S_{\rm ext}} d\phi\, d\rho\, \rho \;

1637:   [\Pi_{zz}^2({\bf r}) - \Pi_{zz}^1({\bf r})] .

1638: \end{equation}

1639: In conclusion, if the stress $\Pi_{zz}^1({\bf r}) - \Pi_{zz}^2({\bf

1640:   r})$ decays sufficiently fast with the distance $\rho$ from the

1641: particle, the solution in Eq.~(\ref{eq:generalu}) will be dominated

1642: asymptotically by the logarithm with an amplitude given by

1643: Eq.~(\ref{eq:amplitudelog}), provided this amplitude does not

1644: vanish \footnote{To which extent this asymptotic regime is actually observable in

1645:   a particular experimental realization depends on the precise

1646:   functional form of ${\bf \Pi}^{1(2)}({\bf r})$ and the values of the

1647:   constant parameters $A_m$, $B_m$, and $\phi_m$ entering the solution given by

1648:   Eq.~(\ref{eq:generalu}).}. As one can infer from

1649: Eq.~(\ref{eq:netFfluid}), this latter condition means physically that

1650: the walls exert a non-vanishing force: a

1651: logarithmic term can only arise if mechanical isolation of the system ``colloid + interface'' is violated.

1652: (Note that Eq.~(\ref{eq:amplitudelog}) follows actually from

1653: Eq.~(\ref{eq:netFtotal}) if the interfacial deformations are small.)

1654: 

1655: We emphasize the generality of this result: it only requires that the

1656: interface departs slightly from a flat one for distances sufficiently far

1657: from the particle. (Otherwise it is actually not useful to

1658: speak of a logarithmically varying deformation to begin with.) In

1659: particular, the interfacial deformation close to the particle may be

1660: arbitrarily large, the particle itself need not be perfectly spherical,

1661: or it may even consist of a many--body configuration lacking any kind

1662: of symmetry.

1663: 

1664: 

1665: 

1666: \end{appendix}

1667: 

1668:  

1669: 

1670: 

1671: \begin{thebibliography}{99}

1672: 

1673: \bibitem{Pie80} P.~Pieranski,

1674: % Two--dimensional interfacial colloidal crystals,

1675:          Phys.~Rev.~Lett. {\bf 45}, 569 (1980).

1676: 

1677: \bibitem{Ave02} %R.~Aveyard {\em et al.},

1678:  R.~Aveyard, B.~P.~Binks, J.~H.~Clint, P.~D.~I.~Fletcher, T.~S.~Horozov, B.~Neumann, V.~N.~Paunov, J.~Annesley, S.~W.~Botchway, D.~Nees, A.~W.~Parker, A.~D.~Ward, and A.~N.~Burgess,

1679: % Measurement of long-range repulsive forces between charged particles at an oil-water interface,

1680:           Phys.~Rev.~Lett. {\bf 88}, 246102 (2002).

1681: 

1682: \bibitem{Par07} B.~J.~Park, J.~P.~Pantina, E.~Furst, M.~Oettel, S.~Reynaert, and J.~Vermant,

1683: % Direct measurements of the effects of salt and surfactant on interaction forces between colloidal particles at water-oil interfaces,

1684:          Langmuir {\bf 24}, 1686 (2008).

1685: 

1686: \bibitem{Zen06} C.~Zeng, H.~Bissig, and A.~D.~Dinsmore,

1687: % Particles on droplets: From fundamental physics to novel materials,

1688:          Solid State Comm. {\bf 139}, 547 (2006).

1689: 

1690: \bibitem{Bre7r} F.~Bresme and M.~Oettel,

1691: % Nanoparticles at fluid interfaces,

1692:           J.~Phys.: Condens.~Matter {\bf 19}, 413101 (2007).

1693: 

1694: \bibitem{Oet08} M.~Oettel and S.~Dietrich,

1695: Langmuir {\bf 24}, 1425 (2008).

1696: 

1697: 

1698: \bibitem{Ghe97} F.~Ghezzi and J.~C.~Earnshaw,

1699: % Formation of meso--structures in colloidal monolayers,

1700:          J.~Phys.: Condens.~Matt. {\bf 9}, L517 (1997).

1701: 

1702: \bibitem{Gar98a} J.~Ruiz-Garc\'\i a and B.~I.~Ivlev,

1703: % Formation of colloidal clusters and chains at the air/water interface,

1704:          Mol.~Phys. {\bf 95}, 371 (1998).

1705: 

1706: \bibitem{Que01} %M.~Quesada--P\'erez {\em et al.},

1707:  M.~Quesada--P\'erez, A.~Moncho--Jord\'a, F.~Mart\'\i nez--L\'opez, and R.~Hidalgo--Alvarez,

1708: % Probing interaction forces in colloid monolayers: Inversion of structural data ,

1709:          J.~Chem.~Phys. {\bf 115}, 10897 (2001).

1710: 

1711: \bibitem{Gom05} O.~G\'omez--Guzm\'an and J.~Ruiz--Garc\'ia,

1712: % Attractive interactions between like-charged colloidal particles at the air/water interface,

1713:          J.~Colloid~Interface Sci. {\bf 291}, 1 (2005).

1714: 

1715: \bibitem{Che06} W.~Chen, S.~S.~Tan, Z.~S.~Huang, T.~K.~Ng, W.~T.~Ford, and P.~Tong,

1716: % Measured long--ranged attractive interaction between charged polystyrene latex spheres at a water-air interface,

1717:          Phys.~Rev.~E {\bf 74}, 021406 (2006).

1718: 

1719: \bibitem{Fer04} J.~C.~Fern\'andez--Toledano, A.~Moncho--Jord\'a, F.~Mart\'inez--L\'opez, and R.~Hidalgo--Alvarez,

1720: % Spontaneous formation of of mesostructures in colloidal monolayers trapped at the air-water interface: A simple explanation

1721:          Langmuir {\bf 20}, 6977 (2004).

1722: 

1723: \bibitem{Lav04}

1724:  I.~I.~Smalyukh, S.~Chernyshuk, B.~I.~Lev, A.~B.~Nych, U.~Ognysta, V.~G.~Nazarenko, and

1725:  O.~D.~Lavrentovich,

1726: % Ordered droplet structures at the liquid crystal surface and elastic-capillary colloidal interactions

1727:             Phys.~Rev.~Lett. {\bf 93}, 117801 (2004).

1728: 

1729: \bibitem{Lav07}

1730:  A.~B.~Nych, U.~M.~Ognysta, V.~M.~Pergamenshchik, B.~I.~Lev, V.~G.~Nazarenko, I.~Musevic, M.~Skarabot, and O.~D.~Lavrentovich,

1731: %  Coexistence of Two Colloidal Crystals at the Nematic-Liquid-Crystal–Air Interface,

1732:             Phys.~Rev.~Lett. {\bf 98}, 057801 (2007).

1733: 

1734: \bibitem{Kra00} P.~A.~Kralchevsky, K.~Nagayama,

1735: % Capillary interactions between particles bound to interfaces, liquid films and biomembranes,

1736:          Adv. Colloid Interface Sci. {\bf 85}, 145 (2000).

1737: 

1738: \bibitem{Nik02} %M.~G.~Nikolaides {\em et al.},

1739:  M.~G.~Nikolaides, A.~R.~Bausch, M.~F.~Hsu, A.~D.~Dinsmore, M.~P.~Brenner, C.~Gay, and D.~A.~Weitz,

1740: % Electric--field induced capillarity attraction between like--charged particles at liquid interfaces,

1741:          Nature {\bf 420}, 299 (2002).

1742: 

1743: \bibitem{Meg03} M.~Megens and J.~Aizenberg,

1744: % Like--charged particles at liquid interfaces: comment on Nikolaides et al., Nature {\bf 420} (2002) 299

1745:           Nature {\bf 424}, 1014 (2003).

1746: 

1747: \bibitem{Nik03} %M.~G.~Nikolaides {\em et al.},

1748:  M.~G.~Nikolaides, A.~R.~Bausch, M.~F.~Hsu, A.~D.~Dinsmore, M.~P.~Brenner, C.~Gay, and

1749: D.~A.~Weitz,

1750: % Electric--field induced capillarity attraction between like--charged particles at liquid interfaces: Reply on Comment,

1751:     Nature  {\bf 424}, 1014 (2003).

1752: 

1753: \bibitem{For04} L.~Foret and A.~W{\"u}rger,

1754: % Electric--field induced capillary interaction of charged particles at a polar interface,

1755:           Phys.~Rev.~Lett.~{\bf 92}, 058302 (2004). %cond-mat/0310657.

1756: 

1757: \bibitem{Kra04} K.~D.~Danov, P.~A.~Kralchevsky, and M.~P.~Boneva,

1758: % Electrodipping force acting on solid particles at a fluid interface,

1759:           Langmuir {\bf 20}, 6139 (2004).

1760: 

1761: \bibitem{Oet05} M.~Oettel, A.~Dom\'inguez, and S.~Dietrich,

1762: % Effective capillary interaction of spherical particles at  fluid interfaces,

1763:          Phys.~Rev.~E {\bf 71}, 051401 (2005). %cond-mat/0411329.

1764: 

1765: \bibitem{Oet05a} M.~Oettel, A.~Dom\'inguez, and S.~Dietrich,

1766: % Comment on Electrodipping force acting on solid particles at a fluid interface (Langmuir {\bf 20}, 6139 (2004).)

1767:          Langmuir {\bf 22}, 846 (2006).

1768: 

1769: \bibitem{Kra05}  K.~D.~Danov and P.~A.~Kralchevsky,

1770: % Reply to Comment

1771:          Langmuir {\bf 22}, 848 (2006).

1772: 

1773: \bibitem{Dan06} K.~D.~Danov, P.~A.~Kralchevsky, and M.~P.~Boneva,

1774: % Shape of the capillary meniscus around an electrically charged particle at a fluid interface: Comparison of theory and experiment,

1775:           Langmuir {\bf 22}, 2653 (2006).

1776: 

1777: \bibitem{Dom06} A.~Dom\'inguez,  M.~Oettel, and S.~Dietrich,

1778:          J.~Chem.~Phys. {\bf 128}, 114904 (2008).

1779: 

1780: \bibitem{Oet05b} M.~Oettel, A.~Dom\'inguez, and S.~Dietrich,

1781: % Attractions between charged colloids at water interfaces,

1782:          J.~Phys.: Condens.~Matter {\bf 17}, L337 (2005).

1783: 

1784: \bibitem{Wue05} A.~W{\"u}rger and L.~Foret,

1785: % Capillary atraction of colloidal particles at an aqueous interface,

1786:          J.~Phys.~Chem.~B {\bf 109}, 16435 (2005).

1787: 

1788: \bibitem{Dom05} A.~Dom\'inguez,  M.~Oettel, and S.~Dietrich,

1789: % Capillary induced interactions between colloids at an interface,

1790:          J.~Phys.: Condens.~Matter {\bf 17}, S3387 (2005).

1791: 

1792: \bibitem{Dom07} A.~Dom\'inguez,  M.~Oettel, and S.~Dietrich,

1793: % Theory of capillary-induced interactions beyond the superposition approximation

1794:          J.~Chem.~Phys. {\bf 127}, 204706 (2007).

1795: 

1796: \bibitem{deG74} P.~G.~de Gennes,

1797: {\em The Physics of Liquid Crystals} (Clarendon, Oxford 1993).

1798: 

1799: \bibitem{Lav03} M.~Kleman and O.~D.~Lavrentovich,

1800: {\em Soft Matter Physics: An Introduction} (Springer, New York 2003).

1801: 

1802: \bibitem{Sch98} J.~Schwinger, L.~L.~DeRaad Jr., K.~A.~Milton, and W.~Tsai,

1803: {\em Classical Electrodynamics} (Perseus, Reading 1998), Chap. 11.2.

1804: 

1805: \bibitem{Pou97} P.~Poulin, H.~Stark, T.~C.~Lubensky, and D.~A.~Weitz,

1806: % Novel colloidal interactions in anisotropic fluids,

1807:          Science {\bf 275}, 1770 (1997).

1808: 

1809: \bibitem{Lub98} T.~C.~Lubensky, D.~Pettey, N.~Currier, and H.~Stark,

1810: % Topological defects and interactions in nematic emulsions,

1811:          Phys.~Rev.~E {\bf 57}, 610 (1998).

1812: 

1813: \bibitem{Ruh97} R.~W.~Ruhwandl and E.~M.~Terentjev,

1814: % Long-range forces and aggregation of colloid particles in a nematic liquid crystal,

1815:          Phys.~Rev.~E {\bf 55}, 2958 (1997).

1816: 

1817: \bibitem{Ram96} S.~Ramaswamy, R.~Nityananda, V.~A.~Raghunathan, and J.~Prost,

1818: % Power-law forces between particles in a nematic,

1819:          Mol.~Cryst.~Liq.~Cryst. {\bf 288}, 175 (1996).

1820: 

1821: \bibitem{Fuk02} J.~Fukuda, B.~I.~Lev, K.~M.~Aoki, and H.~Yokoyama,

1822: % Interaction of particles in a deformed nematic liquid crystal,

1823:          Phys.~Rev.~E {\bf 66}, 051711 (2002).

1824: 

1825: \bibitem{Sta99} H.~Stark,

1826: % Director field configurations around a spherical particle in a nematic liquid crystal,

1827:          Eur.~Phys.~J.~B {\bf 10}, 311 (1999).

1828: 

1829: \bibitem{Tas05} D.~Andrienko, M.~Tasinkevych, and S.~Dietrich,

1830: % Effective pair interaction between colloidal particles at a nematic-isotropic interface, 

1831:          Europhys.~Lett. {\bf 70}, 95 (2005).

1832: 

1833: \bibitem{Pet98} D.~Pettey, T.~C.~Lubensky, and D.~R.~Link,

1834: % Topological inclusions in 2D semctic C films,

1835:          Liquid Crystals {\bf 25}, 579 (1998).

1836: 

1837: \bibitem{Lav07a} O.~Lavrentovich, private communication. 

1838: 

1839: \end{thebibliography}

1840: 

1841: 

1842: \end{document}

1843: