1: %% submitted to PRL on 21st Jan 2005

2: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

3: \documentclass[bibnotes,twocolumn,showpacs,amsmath,amssymb,nobalancelastpage,dvips]{revtex4}

4: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

5: \usepackage{bm}

6: \usepackage{graphics,psfig,epsfig,subfigure,afterpage}

7: \usepackage{amsfonts}

8: \usepackage{psfrag}

9: 

10: \newcommand{\be}{\begin{equation}}

11: \newcommand{\ee}{\end{equation}}

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

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

14: \newcommand{\gdot}{\dot{\gamma}}

15: \newcommand{\gdotbar}{\bar{\dot{\gamma}}}

16: \newcommand{\ie}{{\it i.e.\/}}

17: \newcommand{\eg}{{\it e.g.\/}}

18: \newcommand{\etc}{{\it etc.\/}}

19: \newcommand{\versus}{{\it vs.\/}}

20: \newcommand{\etal}{{\it et al.\/}}

21: \newcommand{\bw}{\begin{widetext}}

22: \newcommand{\ew}{\end{widetext}}

23: \newcommand{\lae}{\stackrel{<}{\scriptstyle\sim}}

24: 

25: \newcommand{\tmax}{t_{\rm max}}

26: \newcommand{\Tband}{T_{\rm b}}

27: \newcommand{\Nbase}{N_{\rm base}}

28: 

29: \newcommand{\ommax}{\omega_{\rm max}}

30: 

31: %\newcommand{\vecv}[1]{\underline{#1}}

32: %\newcommand{\tens}[1]{\underline{\underline{#1}}}

33: %\newcommand{\nablu}{\vecv{\nabla}}

34: 

35: \newcommand{\vecv}[1]{\mathbf{{#1}}}

36: \newcommand{\tens}[1]{\mathbf{{#1}}}

37: \newcommand{\nablu}{{\bf \nabla}}

38: 

39: 

40: \begin{document}

41: 

42: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

43: \title{Linear instability of planar shear banded flow}

44: \author{S. M. Fielding}

45: \email{s.m.fielding@leeds.ac.uk}

46: \affiliation{Polymer IRC and

47:   School of Physics \& Astronomy, University of Leeds, Leeds LS2

48:   9JT, United Kingdom} 

49: 

50: \date{\today} 

51: \begin{abstract}

52:   We study the linear stability of planar shear banded flow with

53:   respect to perturbations with wavevector in the plane of the banding

54:   interface, within the non local Johnson Segalman model.  We find

55:   that perturbations grow in time, over a range of wavevectors,

56:   rendering the interface linearly unstable. Results for the unstable

57:   eigenfunction are used to discuss the nature of the instability. We

58:   also comment on the stability of phase separated domains to shear

59:   flow in model H.

60: 

61: \end{abstract}

62: \pacs{{47.50.+d} {Non-Newtonian fluid flows}--

63:      {83.60.Wc} {Flow instabilities} --

64:      {47.55.Kf} {Multi-phase and particle laden flows} --

65:      {61.25.Hq} {Macromolecular and polymer solutions; polymer melts; swelling}

66: }

67: \maketitle

68: 

69: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

70: 

71: 

72: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

73: 

74: Complex fluids such as wormlike micellar

75: surfactants~\cite{BritCall97}, lamellar onion phases~\cite{diat93},

76: polymer solutions~\cite{HilVla02} and soft

77: glasses~\cite{RayMouBauBerGuiCou02} commonly undergo flow

78: instabilities and flow-induced transitions that result in spatially

79: heterogeneous ``shear banded'' states. This effect is captured by

80: several notable rheological

81: models~\cite{doiedwards}

82: %~\cite{doiedwards,cates90,doi81,olmsted99c} 

83: in which the underlying constitutive curve of shear stress \versus\ 

84: shear rate, $T_{xy}(\gdot)$, is non-monotonic

85: (Fig.~\ref{fig:flowCurve}), allowing the coexistence of bands of

86: differing shear rate at common shear stress, Fig.~\ref{fig:profile}.

87: However, most theoretical studies have considered only one spatial

88: dimension (1D)~\cite{spenley93,lu99},

89: %~\cite{spenley93,olmsted90,lu99}

90: normal to the interface between the bands (the flow gradient

91: direction, $y$).  The stability of 1D banded profiles in higher

92: dimensions has been implicitly assumed, but is in fact an open

93: question.  In this Letter, therefore, we study numerically the linear

94: stability of 1D planar shear banded profiles with respect to

95: perturbations with wavevectors in the  interfacial

96: plane $(x,z) = ({\rm flow},{\rm vorticity})$.

97: %in the plane of the interface, $x,z$

98: %(the flow, vorticity plane of the unperturbed profile).

99: 

100: We work within the Johnson Segalman (JS) model~\cite{johnson77},

101: modified to include non local diffusive terms~\cite{olmsted99a}. These

102: account for gradients in the order parameters across the banding

103: interface, conferring a surface tension. This ``dJS'' model is often

104: taken as a paradigm of shear banding systems. Our main result will be

105: that interfacial fluctuations typically grow in time, rendering the 1D

106: banded profile linearly unstable.  This potentially opens the way to

107: non trivial interfacial dynamics and could form a starting point for

108: understanding an emerging body of data revealing erratic fluctuations

109: of shear banded flows~\cite{WunColLenArnRou01}.

110: %~\cite{WunColLenArnRou01,BanBasSoo00,BritCall99,HolLopCal03,HBP98,FisWheFul02}

111: This work is a timely counterpart to new techniques for measuring

112: interfacial dynamics~\cite{ManSalCol04}.

113: %~\cite{ManSalCol04,SalBecManCol03,callnote}

114: It is also relevant industrially, to processing instability and oil

115: extraction.

116: 

117: The model is defined as follows. The generalised Navier Stokes

118: equation for a viscoelastic material in a Newtonian solvent of

119: viscosity $\eta$ and density $\rho$ is:

120: %

121: \be

122: \label{eqn:NS}

123: \rho(\partial_t + \vecv{V}.\nablu)\vecv{V} = \nablu .(\tens{\Sigma} +

124: \eta\nablu \vecv{V} -P\tens{I}), \ee

125: %

126: where $\vecv{V}(\vecv{R})$ is the velocity field and

127: $\vecv{\Sigma}(\vecv{R})$ the viscoelastic part of the stress. For

128: homogeneous planar shear, $\vecv{V}=y\gdot\vecv{\hat{x}}$, the total

129: shear stress $T_{xy}=\Sigma_{xy}(\gdot)+\eta\gdot$. The pressure $P$

130: is determined by incompressibility,

131: %

132: \be

133: \nablu.\vecv{V}=0.

134: \label{eqn:incomp}

135: \ee

136: %

137: The viscoelastic stress evolves with dJS

138: dynamics~\cite{johnson77,olmsted99a}

139: %

140: \be

141: \label{eqn:dJS}

142:  \stackrel{\diamondsuit}{\tens{\Sigma}} = 2 G\tens{D}-\frac{\tens{\Sigma}}{\tau}+ \frac{l^2}{\tau }\nablu^2 \tens{\Sigma},

143: \ee

144: %

145: with plateau modulus $G$ and relaxation time $\tau$.  The non local

146: diffusive term accounts for spatial gradients across the interface

147: between the bands. It arises naturally in models of liquid crystals,

148: and diffusion of strained polymer molecules~\cite{elkareh89}.

149: %~\cite{elkareh89,goveas98}

150: The time derivative

151: %

152: \be 

153: \label{eqn:tderiv}

154: \stackrel{\diamondsuit}{\tens{\Sigma}} = (\partial_t +\vecv{V}\cdot\nablu )\,\tens{\Sigma} -

155: a(\tens{D}\cdot\tens{\Sigma}+\tens{\Sigma}\cdot\tens{D}) - (\tens{\Sigma}\cdot\tens{\Omega} - \tens{\Omega}\cdot\tens{\Sigma}), \nonumber

156: \ee

157: %

158: in which $\tens{D}$ and $\tens{\Omega}$ are the symmetric and

159: antisymmetric parts of the velocity gradient tensor, $(\nablu

160: \vecv{V})_{\alpha\beta}\equiv \partial_\alpha v_\beta$. The ``slip

161: parameter'' $a$ measures the non-affinity of deformation of the

162: viscoelastic component~\cite{johnson77}. Slip occurs for $|a|<1$.  The

163: underlying constitutive curve $T_{xy}(\gdot)$ is then capable of the

164: non-monotonic behaviour of Fig.~\ref{fig:flowCurve}.

165: 

166: Within this model we consider planar shear between infinite, flat

167: parallel plates at $y=0,L$. We use units in which $G=1$, $\tau=1$ and

168: $L=1$; and boundary conditions at $y=0,1$ of

169: $\partial_y\Sigma_{\alpha\beta}=0\;\forall\;\alpha,\beta$ for the

170: viscoelastic stress, with no slip and no penetration for the velocity.

171: 

172: For an imposed shear rate $\gdotbar$ in the region of decreasing

173: stress, $d T_{xy}/d\gdot<0$, homogeneous flow is

174: unstable~\cite{Yerushalmi70}.  A 1D analysis in the flow gradient

175: dimension then predicts a separation into two bands of differing shear

176: rates $\gdot_1, \gdot_2$ at common shear stress, $\Tband$, separated

177: by an interface of width $O(l)$.  As the applied shear rate $\gdotbar$

178: is tracked across the banding regime, the relative width-fraction of

179: the bands adjusts to maintain the constraint $\int dy \gdot(y) =

180: \gdotbar$, while $\gdot_1, \gdot_2$ and $\Tband$ stay constant,

181: leading to a plateau in the steady state flow curve

182: (Fig.~\ref{fig:flowCurve}).

183: 

184: %

185: \begin{figure}[t]

186: %\includegraphics[scale=0.3]{figures/flowCurvea0.3.eps}

187: \includegraphics[scale=0.3]{flowCurvea0.3.eps}

188: \caption{Underlying constitutive curve; steady state flow curve. $a=0.3$, $\eta=0.05$. Banding occurs on the plateau. 

189: \label{fig:flowCurve} } 

190: \end{figure}

191: %

192: 

193: %

194: \begin{figure}[t]

195:   \vspace{-0.5cm} 

196: %\includegraphics[scale=0.3]{figures/profile.eps}

197: \includegraphics[scale=0.3]{profile.eps}

198: \caption{1D banded profile, with spatial gradients restricted  to the flow gradient direction, $y$. $\gdotbar=2.0$, towards the left of the plateau in Fig.~\ref{fig:flowCurve}. $l=0.01$, $\Nbase=800$.

199: \label{fig:profile} } 

200: \end{figure}

201: %

202: 

203: We verified this 1D scenario by numerically evolving

204: Eqns.~\ref{eqn:NS} to~\ref{eqn:dJS}, allowing spatial variations only

205: in the flow-gradient direction $y$.  We used a Crank Nicholson

206: algorithm~\cite{numericalrecipes} within a finite difference scheme on

207: a uniform mesh of ``full'' points $y_0, y_1 ...y_ {\Nbase}$ for

208: $\tens{\Sigma}$ and staggered ``half'' points

209: $y_{\frac{1}{2}},y_{\frac{3}{2}} ...  y_{\Nbase-\frac{1}{2}}$ for

210: $\vecv{V}$. We evolved with time-step $Dt$ for a time $t_{\rm max}$ to

211: steady state, checking for convergence to the limit $\Nbase\to\infty,

212: Dt\to 0, \tmax\to\infty$.

213: 

214: The resulting flow curve is shown in Fig.~\ref{fig:flowCurve}.  A

215: typical steady state shear banded profile $\vecv{V}(y)$,

216: $\tens{\Sigma}(y)$ is given in Fig.~\ref{fig:profile}. The velocity

217: normal to the interface $V_y=0$ in this 1D profile. The smooth

218: variation of the order parameters across the interface results from

219: the diffusive term in Eqn.~\ref{eqn:dJS}, which confers an interface

220: width $O(l)$.  This is in contrast to local models ($l=0$) in which

221: the interface is a sharp discontinuity. In fact, local models are

222: pathological in the sense that the banded state is not uniquely

223: selected, but depends on flow history~\cite{lu99,olmsted99a}.

224: 

225: The linear stability of the sharply banded profiles of local models

226: was studied by previous authors. Renardy~\cite{renardy} found

227: instability with respect to interfacial fluctuations of high

228: wavevector, $q_x\to\infty$, in the local JS model restricted to the

229: case of a thin high shear band. McLeish~\cite{Mcleish87} studied

230: capillary flow, for general band thickness.  He demonstrated a long

231: wavelength ($q_x\to 0$) instability due to the jump in normal stresses

232: across the interface.  This mechanism was also discussed in

233: Ref.~\cite{hinch}.

234: 

235: 

236: %

237: Here we study numerically the {\em non local} case, in which the 1D

238: banded profile is uniquely selected~\cite{lu99,olmsted99a}. The non

239: zero interfacial width, $l\ll L$, now confers a surface tension, which

240: was absent from the local case.  We study general band thicknesses and

241: the full $({\rm velocity}, {\rm vorticity})$ plane of perturbation

242: wavevectors $(q_x, q_z)$.

243: 

244: We linearised the model equations~\ref{eqn:NS} to~\ref{eqn:dJS} for

245: small perturbations (lower case) about the (upper case) base profile,

246: $\vecv{\tilde{\Phi}}(x, y, z, t)=\vecv{\Phi} (y)+

247: \vecv{\phi}_{\vecv{q}}(y)\exp(\omega_{\vecv{q}}t+ iq_x x+iq_z z)$.

248: The vector $\vecv{\Phi}$ comprises all components

249: $\vecv{\Phi}=(\Sigma_{\alpha\beta}, V_\alpha)$, the pressure being

250: eliminated by incompressibility. This linearisation results in an

251: eigenvalue equation with an operator $\mathcal{L}$, which acts linearly

252: on the perturbation $\vecv{\phi}_{\vecv{q}}(y)$:

253: %

254: \be

255: \label{eqn:eigen}

256: \omega_{\vecv{q}}\vecv{\phi}_{\vecv{q}}(y)=\mathcal{L}({\vecv{\Phi}(y), {\vecv{q}}}, \partial_y, \partial_y^2...)\vecv{\phi}_{\vecv{q}}(y).

257: \ee

258: %

259: %

260: 

261: 

262: 

263: %

264: For numerical study, we discretized this equation on a staggered mesh.

265: The 1D base profile $\vecv{\Phi}(y)$ was read in from the calculation

266: already described.  For narrow interfaces, its uniform mesh had too

267: many nodes for use in the eigenvalue problem,

268: %($\Nbase=3200$ for $l=0.0025$)

269: so we adapted it to put most attention near the interface. We then

270: used a NAG routine~\cite{NAG} to find the eigenmodes of this

271: discretized problem.

272: 

273: The results, discussed below, were checked as follows: (i) for

274: convergence with respect to mesh structure; (ii) that for a

275: homogeneous base state on the underlying constitutive curve our

276: results match those of Ref.~\cite{FieOlm03}; (iii) that for $a=0$,

277: $l=0$ (the local Oldroyd B model), our method gives results consistent

278: with Fig. 3 of Ref.~\cite{WilRenRen99}; (iv) that linearisation about

279: a semi-evolved (non-steady) banded state using the {\em analytically}

280: derived Eqn.~\ref{eqn:eigen} gives the same results in the limit

281: $q_x=0, q_z\to 0$ as a particular direct {\em numerical} linearisation

282: performed about the same profile in the code that evolves the 1D base

283: state; (v) for robustness with respect to first evolving the base

284: state on either a uniform or adapted grid, using either a

285: semi-implicit or explicit algorithm; (vi) that two different methods

286: of eliminating the pressure (using the Oseen tensor, and the curl

287: operator) agree.

288: 

289: 

290: %

291: \begin{figure}[t]

292: %\includegraphics[scale=0.3]{figures/maxEigen_gdot2.0.eps}

293: \includegraphics[scale=0.3]{maxEigen_gdot2.0.eps}

294: \caption{Real part of the eigenvalue of the most unstable mode. $a=0.3$,  $\eta=0.05$,  $\gdotbar=2.0$, Reynolds number $\rho/\eta=0$. The data for $l=0.01$ correspond to the base profile in Fig.~\ref{fig:profile}. Symbols: data. Solid lines: cubic splines.

295: \label{fig:maxEigen} } 

296: \end{figure}

297: %

298: \begin{figure}[t]

299: %\includegraphics[scale=0.3]{figures/maxima.eps}

300: \includegraphics[scale=0.3]{maxima.eps}

301: \caption{Peak of the dispersion relation, {\it i.e.} $\Re \ommax$ at $d\Re\ommax/dq_x=0$.  Parameters as for Fig.~\ref{fig:maxEigen}.  Limits of the banding regime shown by vertical lines. Symbols: data. Dotted lines: cubic splines, as a guide to the eye.

302: \label{fig:maxima} } 

303: \end{figure}

304: %

305: 

306: 

307: \iffalse

308: \begin{figure}[t]

309:   \includegraphics[scale=0.3]{figures/maxEigen_many_gdot.eps}

310: \caption{Real part of the eigenvalue of the most unstable mode. The slip parameter $a=0.3$, solvent viscosity $\eta=0.05$, interface width parameter $l=0.005$, Reynolds number $\rho/\eta=0$.

311: \label{fig:maxEigen_many_gdot} } 

312: \end{figure}

313: \fi

314: 

315: 

316: %

317: %

318: %\begin{widetext}

319: \begin{figure}[ht]

320: %\includegraphics[scale=0.7]{figures/flow.eps}

321: %\includegraphics[scale=0.5]{figures/flow_q2.0_l0.01_sigxx.eps}

322: \includegraphics[scale=0.5]{flow_q2.0_l0.01_sigxx.eps}

323: \caption{ Perturbation to  flow field $s_1\Re\vecv{v}(y)e^{iq_xx}$

324:   (arrows), and contour lines of perturbed normal stress

325:   $\tilde{\Sigma}_{xx}(x,y)=\Sigma_{xx}(y)+s_2\Re

326:   \sigma_{xx}(y)e^{iq_xx}$ (dotted lines), corresponding to the

327:   eigenvalue of Fig.~\ref{fig:maxEigen} with $l=0.01$, $q_x=2.0$.

328:   Contours downwards: 0.45, 0.60, 0.75, 0.90, 1.05, 1.20, 1.35 (middle

329:   value shown thicker). Arbitrary scale factors $s_1=1.5$ and

330:   $s_2=0.3$.

331: \label{fig:eigenVector1} } 

332: \end{figure}

333: %\end{widetext}

334: %

335: 

336: For any base profile $\vecv{\Phi}(y)$ and wavevector $\vecv{q}$, the

337: number of eigenmodes is equal to the number of order parameters summed

338: over all mesh points. In this Letter, we only consider the eigenvalue

339: $\ommax(\vecv{q})$ with the largest real part,

340: $\Re\,\ommax(\vecv{q})$. In particular, we ask if this mode is stable,

341: $\Re\,\ommax<0$, or unstable, $\Re\,\ommax>0$.  All results given are

342: for a low solvent viscosity $\eta=0.05\ll G\tau\equiv 1$, consistent

343: with experiment. We set $a=0.3$, although our findings are

344: qualitatively robust to variations in $a$. This leaves the applied

345: shear rate $\gdotbar$ as the tunable parameter.

346: 

347: The dispersion relation $\Re\,\ommax(q_x, q_z=0)$ for fluctuations

348: with wavevector confined to the direction of the unperturbed flow is

349: shown in Fig.~\ref{fig:maxEigen} for $\gdotbar=2.0$.  At any $q_x$,

350: $\Re\ommax$ increases with decreasing $l$, and for small enough $l$

351: the dispersion relation is positive over a range of wavevectors,

352: rendering the 1D profile unstable. For small $l$ this applies to shear

353: rates right across the stress plateau of Fig.~\ref{fig:flowCurve}, as

354: shown in Fig.~\ref{fig:maxima}.  Because the $l$ values accessed here

355: -- $l=O(1-10\mu {\rm m})$ for a 1mm rheometer gap -- are even larger than

356: those expected physically, $l=O(100{\rm nm})$, our results suggest

357: that, experimentally, the entire stress plateau will be unstable.

358: 

359: In the limit $l\to 0$, $q_x\to 0$, the corresponding eigenfunction

360: $\{\partial_y v_x, v_y=0, \sigma_{\alpha\beta}(y)\}$ tends to the

361: spatial derivative of the base state, $\partial_y\{\partial_y V_x,

362: V_y=0, \Sigma_{\alpha\beta}\}$, representing a simple displacement of

363: the interface in the flow-gradient direction, with small corrections

364: in the bulk phases to maintain $\gdotbar={\rm constant}$. As $q_x$

365: increases from zero, this displacement is modulated by a wave of

366: wavevector $q_x\hat{\vecv{x}}$ with an eigenvalue $\ommax(q_x) =

367: \omega_0 + iq_x \omega_1 + q_x^2 \omega_2$ with $\omega_2>0$,

368: signifying instability. A natural question is whether this instability

369: has the same origin as that described by McLeish for the local

370: model~\cite{Mcleish87}. It is not obvious, a priori, that this should

371: be true because, for the base state at least, the limit $l\to 0$ is

372: singular~\cite{lu99}. Indeed, a detailed analysis (work in progress)

373: is more complicated in this case, and deferred to a longer

374: publication. However, the numerical results of

375: Fig.~\ref{fig:eigenVector1} are qualitatively consistent with the

376: mechanism of McLeish, as follows. A wavelike interfacial displacement

377: with extrema at $q_x x/2\pi = 0.0, 0.5, 1.0$ causes an interfacial

378: tilt near $q_xx/2\pi=0.25, 0.75$, exposing the normal stress jump

379: $\Delta\Sigma_{xx}$ across the interface (recall

380: Fig.\ref{fig:profile}).  This triggers a horizontal perturbation to

381: the flow field $\Im v_x$ in these regions, which recirculates, giving

382: an $O(q_x^2)$ vertical velocity $\Re v_y$ at $q_x x/2\pi = 0.0, 0.5,

383: 1.0$. This enhances the original displacement and so causes

384: instability.  Stability is restored for higher $q_x$

385: (Fig.~\ref{fig:maxEigen}), a feature that is absent in the local case.

386: 

387: The eigenvalue $\Re\,\ommax(\vecv{q})$ over the $(q_x,q_z)$ plane is

388: shown in Fig.~\ref{fig:xzPlane}. 

389: %(We give results only for the upper

390: %right quadrant because the plot is symmetric with respect to

391: %reflections in the $q_x$, $q_z$ axes.)  

392: Modes with wavevector along the $q_x$ axis are much more prone to

393: instability than those along the $q_z$ axis.  Nonetheless, for smaller

394: values of $l$ (not shown), modes along the line $q_x=0$ can go

395: unstable as well.

396: %

397: 

398: We note finally an important bound on the validity of our calculation.

399: The expansion used to obtain Eqn.~\ref{eqn:eigen} is valid for

400: perturbations that are small at any point in space. For example, for

401: the stress components we require $\sigma_{\alpha\beta} \ll 1$.

402: Displacement of the interface by a distance $\Delta$ gives

403: $\sigma_{\alpha\beta}=\Delta\, d\Sigma_{\alpha\beta}/dy$, which is

404: $O(\Delta/l)$, because the base profile $\Sigma_{\alpha\beta}$ changes

405: by $O(1)$ over the interfacial width $O(l)$. We are thus restricted to

406: small displacements, $\Delta\ll l$.  In future work, we will consider

407: $\Delta\gg l$.

408: 

409: \begin{figure}[t]

410: \vspace{0.5cm}

411: %\includegraphics[scale=0.30]{figures/xzPlaneContours.eps}

412: \includegraphics[scale=0.30]{xzPlaneContours.eps}

413: \caption{Real part of the most unstable eigenvalue.  $a=0.3$,  $\eta=0.05$,  $\gdotbar=2.0$, Reynolds number $\rho/\eta=0.01$ (negligible),  $l=0.01$. Contours are $-0.45, -0.40 ...$ (dotted line), $0.00$ (dashed) and $... 0.25, 0.30$  (solid).

414: \label{fig:xzPlane} } 

415: \end{figure}

416: 

417: We comment briefly on the stability of a sheared interface between two

418: phases of a binary fluid in ``model H''~\cite{hohenberg77}. Although

419: this was studied in Ref.~\cite{BraCavTra02}, that work integrated over

420: space to get a simple equation for the position of the interface. Such

421: an approach neglects changes in the interface's profile, and any fluid

422: flow normal to the interface, so is not guaranteed to agree with ours.

423: Nonetheless, we found the interface to be stable, as in

424: Ref.~\cite{BraCavTra02}.  This supports the idea that normal stresses

425: (absent in model H) cause the instability described above.

426: 

427: To conclude, we have found 1D planar shear banded flow to be linearly

428: unstable to fluctuations with wavevector in the plane of the banding

429: interface, within the non local Johnson Segalman model. This applies

430: to shear rates right across the stress plateau, suggesting that the

431: instability is ubiquitous and therefore that the existing theoretical

432: picture of two stable shear bands separated by a steady interface

433: needs further thought.  Indeed, our finding is consistent with

434: accumulating evidence for erratic

435: fluctuations~\cite{WunColLenArnRou01}

436: %~\cite{BanBasSoo00} 

437: and band breakup~\cite{lerouge_thesis} in several systems.  Future

438: work will study the fate of the interface in the non linear regime,

439: beyond the validity of this linear study. One possibility is that the

440: instability is self limiting beyond a critical amplitude set by $l$

441: ({\it e.g.}, $l^{1/2}$). This would be consistent with a narrowly

442: localized but still unsteady interface, which might be interpreted as

443: steady in experiments that did not have high spatial resolution. This

444: might even reconcile early data showing apparently steady interfaces

445: with recent work revealing fluctuations. 

446: 

447: By contrast, if the instability were found not to be self limiting,

448: and yet ubiquitous in existing banding models (work in progress), one

449: would then need a new theoretical picture of (reasonably) steady shear

450: bands that could still accommodate the required normal stress jump

451: across the interface. Other open questions include the status of the

452: instability in curved Couette geometry; and the relative importance of

453: instabilities at non-zero $\vecv{q}$ (as studied here) to those found

454: at zero $\vecv{q}$ in recent models of spatio-temporal

455: rheochaos~\cite{FieOlm04}.

456: %~\cite{FieOlm04,AraCat04}

457: %and the wider significance of the instability found here to other

458: %shear banding models~\cite{doiedwards,spenley93,lu99},

459: %~\cite{olmsted99c,spenley93}

460: %including possible instability of ``vorticity'' shear bands that occur

461: %at common shear stress.

462: 

463: %

464: 

465: The author thanks Paul Callaghan, Mike Cates, Tanniemola Liverpool,

466: Tom McLeish, Peter Olmsted and Helen Wilson for useful discussions and

467: feedback; and the EPSRC GR/S29560/01 for funding.

468: 

469: \vspace{0.4cm}

470: 

471: 

472: \begin{thebibliography}{36}

473: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi

474: \expandafter\ifx\csname bibnamefont\endcsname\relax

475:   \def\bibnamefont#1{#1}\fi

476: \expandafter\ifx\csname bibfnamefont\endcsname\relax

477:   \def\bibfnamefont#1{#1}\fi

478: \expandafter\ifx\csname citenamefont\endcsname\relax

479:   \def\citenamefont#1{#1}\fi

480: \expandafter\ifx\csname url\endcsname\relax

481:   \def\url#1{\texttt{#1}}\fi

482: \expandafter\ifx\csname urlprefix\endcsname\relax\def\urlprefix{URL }\fi

483: \providecommand{\bibinfo}[2]{#2}

484: \providecommand{\eprint}[2][]{\url{#2}}

485: 

486: \bibitem[{\citenamefont{Britton and Callaghan}(1997)}]{BritCall97}

487: \bibinfo{author}{\bibfnamefont{M.~M.} \bibnamefont{Britton}} \bibnamefont{and}

488:   \bibinfo{author}{\bibfnamefont{P.~T.} \bibnamefont{Callaghan}},

489:   \bibinfo{journal}{Phys. Rev. Lett.} \textbf{\bibinfo{volume}{78}},

490:   \bibinfo{pages}{4930} (\bibinfo{year}{1997}).

491: 

492: \bibitem[{\citenamefont{Diat et~al.}(1993)\citenamefont{Diat, Roux, and

493:   Nallet}}]{diat93}

494: \bibinfo{author}{\bibfnamefont{O.}~\bibnamefont{Diat}},

495:   \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Roux}}, \bibnamefont{and}

496:   \bibinfo{author}{\bibfnamefont{F.}~\bibnamefont{Nallet}},

497:   \bibinfo{journal}{J.~Phys.~II (France)} \textbf{\bibinfo{volume}{3}},

498:   \bibinfo{pages}{1427} (\bibinfo{year}{1993}).

499: 

500: \bibitem[{\citenamefont{Hilliou and Vlassopoulos}(2002)}]{HilVla02}

501: \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Hilliou}} \bibnamefont{and}

502:   \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Vlassopoulos}},

503:   \bibinfo{journal}{Ind.\ Eng.\ Chem.\ Res.} \textbf{\bibinfo{volume}{41}},

504:   \bibinfo{pages}{6246} (\bibinfo{year}{2002}).

505: 

506: \bibitem[{\citenamefont{Raynaud et~al.}(2002)\citenamefont{Raynaud, Moucheront,

507:   Baudez, Bertrand, Guilbaud, and Coussot}}]{RayMouBauBerGuiCou02}

508: \bibinfo{author}{\bibfnamefont{J.~S.} \bibnamefont{Raynaud}},

509:   \bibinfo{author}{\bibfnamefont{P.}~\bibnamefont{Moucheront}},

510:   \bibinfo{author}{\bibfnamefont{J.~C.} \bibnamefont{Baudez}},

511:   \bibinfo{author}{\bibfnamefont{F.}~\bibnamefont{Bertrand}},

512:   \bibinfo{author}{\bibfnamefont{J.~P.} \bibnamefont{Guilbaud}},

513:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{P.}~\bibnamefont{Coussot}},

514:   \bibinfo{journal}{J.\ Rheol.} \textbf{\bibinfo{volume}{46}},

515:   \bibinfo{pages}{709} (\bibinfo{year}{2002}).

516: 

517: \bibitem[{\citenamefont{Doi and Edwards}(1989)}]{doiedwards}

518: \bibinfo{author}{\bibfnamefont{M.}~\bibnamefont{Doi}} \bibnamefont{and}

519:   \bibinfo{author}{\bibfnamefont{S.~F.} \bibnamefont{Edwards}},

520:   \emph{\bibinfo{title}{The Theory of Polymer Dynamics}}

521:   (\bibinfo{publisher}{Clarendon}, \bibinfo{address}{Oxford},

522:   \bibinfo{year}{1989}).

523: %\bibitem[{\citenamefont{Cates}(1990)}]{cates90}

524: \bibinfo{author}{\bibfnamefont{M.~E.} \bibnamefont{Cates}},

525:   \bibinfo{journal}{J.~Phys. Chem.} \textbf{\bibinfo{volume}{94}},

526:   \bibinfo{pages}{371} (\bibinfo{year}{1990}).

527: %\bibitem[{\citenamefont{Doi}(1981)}]{doi81}

528: \bibinfo{author}{\bibfnamefont{M.}~\bibnamefont{Doi}},

529:   \bibinfo{journal}{J.~Poly. Sci: Poly. Phys.} \textbf{\bibinfo{volume}{19}},

530:   \bibinfo{pages}{229} (\bibinfo{year}{1981}).

531: %\bibitem[{\citenamefont{Olmsted and Lu}(1999)}]{olmsted99c}

532: \bibinfo{author}{\bibfnamefont{P.~D.} \bibnamefont{Olmsted}} \bibnamefont{and}

533:   \bibinfo{author}{\bibfnamefont{C.-Y.~D.} \bibnamefont{Lu}},

534:   \bibinfo{journal}{Phys. Rev.} \textbf{\bibinfo{volume}{E60}},

535:   \bibinfo{pages}{4397} (\bibinfo{year}{1999}).

536: 

537: \bibitem[{\citenamefont{Spenley et~al.}(1993)\citenamefont{Spenley, Cates, and

538:   McLeish}}]{spenley93}

539: \bibinfo{author}{\bibfnamefont{N.~A.} \bibnamefont{Spenley}},

540:   \bibinfo{author}{\bibfnamefont{M.~E.} \bibnamefont{Cates}}, \bibnamefont{and}

541:   \bibinfo{author}{\bibfnamefont{T.~C.~B.} \bibnamefont{McLeish}},

542:   \bibinfo{journal}{Phys. Rev. Lett.} \textbf{\bibinfo{volume}{71}},

543:   \bibinfo{pages}{939} (\bibinfo{year}{1993}).

544: %\bibitem[{\citenamefont{Olmsted and Goldbart}(1990)}]{olmsted90}

545: \bibinfo{author}{\bibfnamefont{P.~D.} \bibnamefont{Olmsted}} \bibnamefont{and}

546:   \bibinfo{author}{\bibfnamefont{P.~M.} \bibnamefont{Goldbart}},

547:   \bibinfo{journal}{Phys.~Rev.} \textbf{\bibinfo{volume}{A41}},

548:   \bibinfo{pages}{4578} (\bibinfo{year}{1990}).

549: 

550: \bibitem[{\citenamefont{Lu et~al.}(2000)\citenamefont{Lu, Olmsted, and

551:   Ball}}]{lu99}

552: \bibinfo{author}{\bibfnamefont{C.-Y.~D.} \bibnamefont{Lu}},

553:   \bibinfo{author}{\bibfnamefont{P.~D.} \bibnamefont{Olmsted}},

554:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{R.~C.} \bibnamefont{Ball}},

555:   \bibinfo{journal}{Phys. Rev. Lett.} \textbf{\bibinfo{volume}{84}},

556:   \bibinfo{pages}{642} (\bibinfo{year}{2000}).

557: 

558: \bibitem[{\citenamefont{Johnson and Segalman}(1977)}]{johnson77}

559: \bibinfo{author}{\bibfnamefont{M.}~\bibnamefont{Johnson}} \bibnamefont{and}

560:   \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Segalman}},

561:   \bibinfo{journal}{J. Non-Newt. Fl. Mech} \textbf{\bibinfo{volume}{2}},

562:   \bibinfo{pages}{255} (\bibinfo{year}{1977}).

563: 

564: \bibitem[{\citenamefont{Olmsted et~al.}(2000)\citenamefont{Olmsted, Radulescu,

565:   and Lu}}]{olmsted99a}

566: \bibinfo{author}{\bibfnamefont{P.~D.} \bibnamefont{Olmsted}},

567:   \bibinfo{author}{\bibfnamefont{O.}~\bibnamefont{Radulescu}},

568:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{C.-Y.~D.} \bibnamefont{Lu}},

569:   \bibinfo{journal}{J. Rheology} \textbf{\bibinfo{volume}{44}},

570:   \bibinfo{pages}{257} (\bibinfo{year}{2000}).

571: 

572: \bibitem[{\citenamefont{Wunenburger et~al.}(2001)\citenamefont{Wunenburger,

573:   Colin, Leng, Arneodo, and Roux}}]{WunColLenArnRou01}

574: \bibinfo{author}{\bibfnamefont{A.~S.} \bibnamefont{Wunenburger}},

575:   \bibinfo{author}{\bibfnamefont{A.}~\bibnamefont{Colin}},

576:   \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Leng}},

577:   \bibinfo{author}{\bibfnamefont{A.}~\bibnamefont{Arneodo}}, \bibnamefont{and}

578:   \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Roux}},

579:   \bibinfo{journal}{Phys.\ Rev.\ Lett.} \textbf{\bibinfo{volume}{86}},

580:   \bibinfo{pages}{1374} (\bibinfo{year}{2001}).

581: %\bibitem[{\citenamefont{Bandyopadhyay et~al.}(2000)\citenamefont{Bandyopadhyay,

582: %  Basappa, and Sood}}]{BanBasSoo00}

583: \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Bandyopadhyay}},

584:   \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Basappa}}, \bibnamefont{and}

585:   \bibinfo{author}{\bibfnamefont{A.~K.} \bibnamefont{Sood}},

586:   \bibinfo{journal}{Phys.\ Rev.\ Lett.} \textbf{\bibinfo{volume}{84}},

587:   \bibinfo{pages}{2022} (\bibinfo{year}{2000}).

588: %\bibitem[{\citenamefont{Britton and Callaghan}(1999)}]{BritCall99}

589: %\bibinfo{author}{\bibfnamefont{M.~M.} \bibnamefont{Britton}} \bibnamefont{and}

590: %  \bibinfo{author}{\bibfnamefont{P.~T.} \bibnamefont{Callaghan}},

591: %  \bibinfo{journal}{European Physical Journal B} \textbf{\bibinfo{volume}{7}},

592: %  \bibinfo{pages}{237} (\bibinfo{year}{1999}).

593: %\bibitem[{\citenamefont{Holmes et~al.}(2003)\citenamefont{Holmes,

594: %  Lopez-Gonzalez, and Callaghan}}]{HolLopCal03}

595: \bibinfo{author}{\bibfnamefont{W.~M.} \bibnamefont{Holmes}},

596:   \bibinfo{author}{\bibfnamefont{M.~R.} \bibnamefont{Lopez-Gonzalez}},

597:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{P.~T.}

598:   \bibnamefont{Callaghan}}, \bibinfo{journal}{Europhys.\ Lett.}

599:   \textbf{\bibinfo{volume}{64}}, \bibinfo{pages}{274} (\bibinfo{year}{2003}).

600: %\bibitem[{\citenamefont{Hu et~al.}(1998)\citenamefont{Hu, Boltenhagen, and

601: %  Pine}}]{HBP98}

602: \bibinfo{author}{\bibfnamefont{Y.~T.} \bibnamefont{Hu}},

603:   \bibinfo{author}{\bibfnamefont{P.}~\bibnamefont{Boltenhagen}},

604:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{D.~J.} \bibnamefont{Pine}},

605:   \bibinfo{journal}{J. Rheol.} \textbf{\bibinfo{volume}{42}},

606:   \bibinfo{pages}{1185} (\bibinfo{year}{1998}).

607: %\bibitem[{\citenamefont{Fischer et~al.}(2002)\citenamefont{Fischer, Wheeler,

608: %  and Fuller}}]{FisWheFul02}

609: \bibinfo{author}{\bibfnamefont{P.}~\bibnamefont{Fischer}},

610:   \bibinfo{author}{\bibfnamefont{E.~K.} \bibnamefont{Wheeler}},

611:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{G.~G.}

612:   \bibnamefont{Fuller}}, \bibinfo{journal}{Rheol.\ Acta}

613:   \textbf{\bibinfo{volume}{41}}, \bibinfo{pages}{35} (\bibinfo{year}{2002}).

614: 

615: \bibitem[{\citenamefont{Manneville et~al.}(2004)\citenamefont{Manneville,

616:   Salmon, and Colin}}]{ManSalCol04}

617: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Manneville}},

618:   \bibinfo{author}{\bibfnamefont{J.~B.} \bibnamefont{Salmon}},

619:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{A.}~\bibnamefont{Colin}},

620:   \bibinfo{journal}{Eur.\ Phys.\ J.\ E} \textbf{\bibinfo{volume}{13}},

621:   \bibinfo{pages}{197} (\bibinfo{year}{2004}).

622: %\bibitem[{\citenamefont{Salmon et~al.}(2003)\citenamefont{Salmon, Becu,

623: %  Manneville, and Colin}}]{SalBecManCol03}

624: \bibinfo{author}{\bibfnamefont{J.~B.} \bibnamefont{Salmon}},

625:   \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Becu}},

626:   \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Manneville}},

627:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{A.}~\bibnamefont{Colin}},

628:   \bibinfo{journal}{Eur.\ Phys.\ J.\ E} \textbf{\bibinfo{volume}{10}},

629:   \bibinfo{pages}{209} (\bibinfo{year}{2003}).

630: %\bibitem[{cal()}]{callnote}

631:   \bibinfo{author}{\bibfnamefont{M.~R.}~\bibnamefont{L{\'{o}}pez-Gonz{\'{a}}lez}},

632: \bibinfo{author}{\bibfnamefont{W.~M.} \bibnamefont{Holmes}},

633:   \bibinfo{author}{\bibfnamefont{P.~T.}~\bibnamefont{Callaghan}},

634:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{P.~J.}~\bibnamefont{Photinos}},

635:   \bibinfo{journal}{Phys.\ Rev.\ Lett.} \textbf{\bibinfo{volume}{93}},

636:   \bibinfo{art. no.}{268302} (\bibinfo{year}{2004}).

637: 

638: \bibitem[{\citenamefont{El-Kareh and Leal}(1989)}]{elkareh89}

639: \bibinfo{author}{\bibfnamefont{A.~W.} \bibnamefont{El-Kareh}} \bibnamefont{and}

640:   \bibinfo{author}{\bibfnamefont{L.~G.} \bibnamefont{Leal}},

641:   \bibinfo{journal}{J. Non-Newt. Fl.~Mech.} \textbf{\bibinfo{volume}{33}},

642:   \bibinfo{pages}{257} (\bibinfo{year}{1989}).

643: %\bibitem[{\citenamefont{Goveas and Fredrickson}(1998)}]{goveas98}

644: \bibinfo{author}{\bibfnamefont{J.~L.} \bibnamefont{Goveas}} \bibnamefont{and}

645:   \bibinfo{author}{\bibfnamefont{G.~H.} \bibnamefont{Fredrickson}},

646:   \bibinfo{journal}{Europhys. J. B} \textbf{\bibinfo{volume}{2}},

647:   \bibinfo{pages}{79} (\bibinfo{year}{1998}).

648: 

649: \bibitem[{\citenamefont{Yerushalmi et~al.}(1970)\citenamefont{Yerushalmi, Katz,

650:   and Shinnar}}]{Yerushalmi70}

651: \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Yerushalmi}},

652:   \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Katz}}, \bibnamefont{and}

653:   \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Shinnar}},

654:   \bibinfo{journal}{Chemical Engineering Science}

655:   \textbf{\bibinfo{volume}{25}}, \bibinfo{pages}{1891} (\bibinfo{year}{1970}).

656: 

657: \bibitem[{\citenamefont{Press et~al.}(1992)\citenamefont{Press, Teukolsky,

658:   Vetterling, and Flannery}}]{numericalrecipes}

659: \bibinfo{author}{\bibfnamefont{W.~H.} \bibnamefont{Press}},

660:   \bibinfo{author}{\bibfnamefont{S.~A.} \bibnamefont{Teukolsky}},

661:   \bibinfo{author}{\bibfnamefont{W.~T.} \bibnamefont{Vetterling}},

662:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{B.~P.}

663:   \bibnamefont{Flannery}}, \emph{\bibinfo{title}{Numerical recipes in C}}

664:   (\bibinfo{publisher}{Cambridge University Press},

665:   \bibinfo{address}{Cambridge}, \bibinfo{year}{1992}).

666: 

667: \bibitem[{\citenamefont{Renardy}(1995)}]{renardy}

668: \bibinfo{author}{\bibfnamefont{Y.~Y.} \bibnamefont{Renardy}},

669:   \bibinfo{journal}{The. Comp. Fl.~Dyn.} \textbf{\bibinfo{volume}{7}},

670:   \bibinfo{pages}{463} (\bibinfo{year}{1995}).

671: 

672: \bibitem[{\citenamefont{McLeish}(1987)}]{Mcleish87}

673: \bibinfo{author}{\bibfnamefont{T.~C.~B.} \bibnamefont{McLeish}},

674:   \bibinfo{journal}{J.~Poly. Sci. B-Poly. Phys.} \textbf{\bibinfo{volume}{25}},

675:   \bibinfo{pages}{2253} (\bibinfo{year}{1987}).

676: 

677: \bibitem[{\citenamefont{Hinch}(1992)}]{hinch}

678: \bibinfo{author}{\bibfnamefont{E.~J.} \bibnamefont{Hinch}},

679: \bibinfo{author}{\bibfnamefont{O.~J.} \bibnamefont{Harris}},

680: \bibinfo{author}{\bibfnamefont{J.~M.} \bibnamefont{Rallison}},

681:   \bibinfo{journal}{J.\ Non-Newton.\ Fluid Mech.} \textbf{\bibinfo{volume}{43}},

682:   \bibinfo{pages}{311} (\bibinfo{year}{1992}).

683: 

684: 

685: \bibitem[{NAG()}]{NAG}

686: \bibinfo{note}{Numerical Algorithms Group Ltd., Wilkinson House, Jordan Hill

687:   Road, Oxford, OX2 8DR, UK.}

688: 

689: \bibitem[{lon()}]{FieOlm03}

690: \bibinfo{author}{\bibfnamefont{S.~M.} \bibnamefont{Fielding}},

691: \bibinfo{author}{\bibfnamefont{P.~D.} \bibnamefont{Olmsted}},

692:   \bibinfo{journal}{Phys.\ Rev.\ E} \textbf{\bibinfo{volume}{68}},

693:   \bibinfo{art. no.-}{036313} (\bibinfo{year}{2003}).

694: 

695: \bibitem[{\citenamefont{Wilson et~al.}(1999)\citenamefont{Wilson, Renardy, and

696:   Renardy}}]{WilRenRen99}

697: \bibinfo{author}{\bibfnamefont{H.~J.} \bibnamefont{Wilson}},

698:   \bibinfo{author}{\bibfnamefont{M.}~\bibnamefont{Renardy}}, \bibnamefont{and}

699:   \bibinfo{author}{\bibfnamefont{Y.}~\bibnamefont{Renardy}},

700:   \bibinfo{journal}{J.\ Non-Newton.\ Fluid Mech.}

701:   \textbf{\bibinfo{volume}{80}}, \bibinfo{pages}{251} (\bibinfo{year}{1999}).

702: 

703: \bibitem[{\citenamefont{Hohenberg and Halperin}(1977)}]{hohenberg77}

704: \bibinfo{author}{\bibfnamefont{P.~C.} \bibnamefont{Hohenberg}}

705:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{B.~I.}

706:   \bibnamefont{Halperin}}, \bibinfo{journal}{Rev. Mod. Phys.}

707:   \textbf{\bibinfo{volume}{49}}, \bibinfo{pages}{435} (\bibinfo{year}{1977}).

708: 

709: \bibitem[{\citenamefont{Bray et~al.}(2002)\citenamefont{Bray, Cavagna, and

710:   Travasso}}]{BraCavTra02}

711: \bibinfo{author}{\bibfnamefont{A.~J.} \bibnamefont{Bray}},

712:   \bibinfo{author}{\bibfnamefont{A.}~\bibnamefont{Cavagna}}, \bibnamefont{and}

713:   \bibinfo{author}{\bibfnamefont{R.~D.~M.} \bibnamefont{Travasso}},

714:   \bibinfo{journal}{Phys.\ Rev.\ E} \textbf{\bibinfo{volume}{6501}},

715:   \bibinfo{pages}{art. no.} (\bibinfo{year}{2002}).

716: 

717: \bibitem[{ler()}]{lerouge_thesis}

718: \bibinfo{note}{S Lerouge, PhD Thesis, University of Metz, 2000.}

719: 

720: \bibitem[{\citenamefont{Fielding and Olmsted}(2004)}]{FieOlm04}

721: \bibinfo{author}{\bibfnamefont{S.~M.} \bibnamefont{Fielding}} \bibnamefont{and}

722:   \bibinfo{author}{\bibfnamefont{P.~D.} \bibnamefont{Olmsted}},

723:   \bibinfo{journal}{Phys.\ Rev.\ Lett.} \textbf{\bibinfo{volume}{92}},

724:   \bibinfo{pages}{art. no.} (\bibinfo{year}{2004}).

725: %\bibitem[{Ara()}]{AraCat04}

726: \bibinfo{note}{Aradian, A and Cates, M E, Pre-print cond-mat/0410509.}

727: 

728: \end{thebibliography}

729: 

730: 

731: 

732: %\bibliography{ackerson,actin,articles,banding,barham,berrportdecruppe,books,bray,callaghan,chandcits,crystal,fredrickson,LCtheory,lifshitz,line_tension,malkus,master,mine,noirez,notes,onions,psolutions,rheochaos,rheofolks,ryan,salmon,schoot,semenov,sood,sriram,stein,wilson,worms2,worms3,worms}

733: 

734: \vspace{-0.3cm}

735: 

736: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

737: 

738: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

739: \end{document}

740: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

741: 

742: 

743: