1: \documentclass[prl,twocolumn,floatfix,superscriptaddress,byrevtex]{revtex4}
2: \usepackage[english]{babel}
3: \usepackage{bm,amsmath,amssymb}
4: \usepackage{dcolumn}
5: \usepackage{graphicx}
6: \begin{document}
7: \title{Dynamical slowdown of polymers in laminar and random flows}
8: \author{A. Celani}
9: \author{A. Puliafito}
10: \affiliation{CNRS--INLN, 1361 Route des Lucioles, 06560 Valbonne, France}
11: \author{D. Vincenzi}
12: %\email{dv47@cornell.edu}
13: \affiliation{Max-Planck-Institut f\"ur Dynamik und Selbstorganisation,
14: Bunsenstra{\ss}e 10, 37073 G\"ottingen, Germany}
15: \affiliation{Sibley School of Mechanical \& Aerospace Engineering and
16: LASSP,
17: Cornell University, Ithaca, NY 14853}
18: \begin{abstract}
19: The influence of an external flow on the relaxation dynamics of a
20: single polymer is investigated theoretically and numerically.
21: We show that a pronounced dynamical slowdown occurs in the vicinity of
22: the coil--stretch transition, especially when the dependence on
23: polymer conformation of the drag is accounted for.
24: For the elongational flow, relaxation times are
25: exceedingly larger than the Zimm relaxation time,
26: resulting in the observation of conformation hysteresis.
27: For random smooth flows hysteresis is not present. Yet,
28: relaxation dynamics is significantly
29: slowed down because of the large variety of accessible polymer
30: configurations. The implications of these results for the modeling
31: of dilute polymer solutions in turbulent flows are addressed.
32: \begin{center}
33: {\fontfamily{ppl}\selectfont{\itshape Phys. Rev. Lett.}
34: \textbf{97}, 118301 (2006)
35: \\ \texttt{http://link.aps.org/abstract/PRL/v97/e118301}
36: }
37: \end{center}
38: \end{abstract}
39: %\pacs{47.57.Ng,47.27.-i}
40: \pacs{}
41: \maketitle
42: The dynamics of an isolated polymer in a flow field
43: forms the basis of
44: constitutive models for dilute polymer solutions~\cite{chu,larson,shaqfeh}.
45: The modeling of drag-reducing flows, for instance,
46: requires an appropriate description
47: of single polymer deformation in turbulent velocity fields~\cite{95GB}.
48: In the last decade, major advances in fluorescence
49: microscopy offered the possibility
50: of tracking isolated polymers both in laminar~\cite{shaqfeh}
51: and random flows~\cite{05GCS}.
52: The dynamics of a polymer in thermal equilibrium with the
53: surrounding solvent is commonly described in terms of
54: normal modes and relaxation times associated with them.
55: The analytical form of the relaxation spectrum was first obtained
56: by Rouse under the assumption that the polymer could be described
57: as a series of beads connected by Hookean springs~\cite{53R}.
58: The Rouse model was subsequently improved by Zimm to include hydrodynamical
59: interactions between segments of the polymer~\cite{zimm}. In Zimm's
60: formulation,
61: the equations of motion are decoupled into a normal mode structure by
62: preaveraging the distances between the beads over the distribution of polymer
63: configurations.
64: The relaxation time associated with the fundamental mode, $\tau$, determines
65: the typical time that it takes for a deformed polymer to recover the
66: equilibrium configuration in a solvent.
67: The normal mode theory was confirmed
68: by the analysis of the oscillatory motion of a DNA molecule
69: immersed in a solvent and held in a partially extended state
70: by means of optical tweezers~\cite{97QBC}. An alternative approach
71: to examine polymer relaxation consists in stretching a
72: tethered DNA molecule in a flow and measuring
73: its relaxation after cessation of the flow~\cite{94PQSC,96MCVCC}.
74: The theoretical predictions for these experimental conditions
75: are provided by the scaling theory~\cite{95MB}
76: and the static dynamics formalism~\cite{02RZ}.
77:
78: The aforementioned studies all consider
79: the internal dynamics of a polymer floating in a solvent under the influence
80: of thermal noise only ---
81: the interaction of the polymer with an external flow is not taken into account.
82: One of the aspects highlighted by experiments is that polymer dynamics
83: in a moving fluid is strongly influenced by the carrier velocity field.
84: The coil--stretch transition is the most noticeable example:
85: as the strain rate
86: exceeds a threshold value, the polymer undergoes a transition
87: from the coiled, equilibrium configuration to an almost fully extended
88: one~\cite{74DG}.
89: Therefore, when a polymer is freely transported by a non-uniform flow,
90: we expect that the time scales describing its dynamics
91: may be significantly different from the Zimm time~$\tau$.
92: Simple models for polymer stretching indeed
93: suggest deviations from Zimm's theory near the
94: coil--stretch transition~\cite{05CMV,05MV,05VB}.
95: Discrepancies in the definition of the
96: correct relaxation time are also encountered in drop formation
97: experiments~\cite{TKCW}.
98:
99: In this Letter we investigate polymer relaxation dynamics
100: both in elongational and random smooth flows. Our analysis
101: brings to evidence an important slowdown
102: of dynamics with respect to
103: the Zimm timescales, in the vicinity of the coil--stretch transition.
104: For the elongational flow, this is related
105: to conformation hysteresis~\cite{03Schetal,04SSC,05HL}.
106: For random flows, we show that
107: hysteresis is not present. Nonetheless, the amplification
108: of the relaxation time persists, albeit to a lesser extent,
109: due to the large heterogeneity of polymer
110: configurations. In both cases, the dependence of the drag force on
111: the polymer configuration plays a prominent role. This suggests
112: the necessity of improving current models
113: of polymer solutions in turbulent flows to account for such effect.
114:
115: The dumbbell approximation is the basis of the
116: most common models of single polymer dynamics
117: and viscoelastic models of dilute polymer solutions \cite{Bird}.
118: Its validity relies on the fact that the slowest deformational mode
119: of the polymer is the most influential in producing
120: viscoelasticity~\cite{95GB}.
121: However, when attention is directed to non-equilibrium dynamics it
122: is often too crude to assume
123: that~$\tau$ is independent of the conformation of the molecule~\cite{99HQ}.
124: Therefore, following de Gennes~\cite{74DG},
125: Hinch~\cite{74H}, and Tanner's~\cite{75T} approach, we consider a model where
126: the polymer is described as two beads connected by an elastic spring
127: and the probability density function of the end-to-end vector, $\Psi(\bm{R},t)$,
128: satisfies the diffusion equation:
129: \begin{equation} \label{eq:dumbbell}
130: \frac{\partial\Psi}{\partial t}=
131: -\frac{\partial}{\partial \bm R}\cdot\bigg[\bm\kappa(t)\cdot\bm R\,\Psi-\frac{f(R)\bm R}{2\tau\nu(R)}\,\Psi
132: -\frac{R_0^2}{2\tau \nu(R)}\,\frac{\partial\Psi}{\partial\bm R}
133: \bigg]
134: %\dot{\bm{R}}=\bm{R}\cdot\nabla\bm{v}(t)-\frac{f(R)}{2\tau\nu(R)}{\bm{R}}+\sqrt{\frac{R_0^2}{\tau \nu(R)}}
135: %\,\bm{\xi}(t),
136: \end{equation}
137: where $\kappa_{ij}(t)=\partial_jv_i(t)$~is the velocity gradient, $R_0$ is the mean extension at
138: thermal equilibrium and~$R=|\bm{R}|$.
139: The function~$f(R)$ defines the entropic force restoring stretched molecules
140: into the coiled configuration. Synthetic polymers
141: are properly described
142: by the Warner law, $f(R)=1/(1-R^2/L^2)$, where~$L$ is the contour length of
143: the polymer; biological macromolecules
144: are better characterized by the Marko--Siggia law,
145: $f(R)=2/3-L/(6R)+L/[6R(1-R/L)^2]$~\cite{larson}.
146: The flow strength relative to the polymer tendency to recoil is
147: expressed by the Weissenberg number~\textit{Wi},
148: defined as the product of the Zimm time~$\tau$ by a characteristic
149: extension rate of the flow.
150: The function~$\nu(R)$ encodes for the dependence on the polymer
151: conformation of the drag exerted by the fluid:
152: a spherical coil offers a smaller resistance with respect to a
153: long rod-like configuration.
154: We utilize the expression~$\nu(R)=1+(\zeta_{s}/\zeta_{c}-1)R/L$
155: that interpolates linearly between these extremes (see Refs.~\cite{97L,Doyle}). Here, $\zeta_{c}= 3\sqrt{6\pi^3}R_0\eta/8$ and
156: $\zeta_{s}= 2\pi L \eta/\ln(L/\ell)$ are the friction coefficients
157: for the coiled and the stretched configuration, respectively,
158: $\eta$ is the solvent viscosity and $\ell$ is the diameter of the molecule.
159: Strictly speaking, the above form of~$\nu(R)$ was deduced from
160: experiments and microscopic simulations of laminar flows.
161: To our knowledge, measures of~$\nu(R)$ in random flows
162: are not available yet, and the study of its functional dependence lies beyond
163: the scope of the present work.
164: Assuming that the polymer is aligned with the local
165: stretching direction of the flow for the most part of its evolution,
166: we shall use a linear~$\nu(R)$ for a random flow as well.
167:
168: To define the relaxation time in the presence of an arbitrary
169: external flow, we consider the probability density function of the
170: rescaled extension, $P(r,t)$ with $r=R/L$,
171: and identify the dynamical relaxation time, $t_{\mathrm{rel}}$,
172: as the characteristic time needed for~$P(r,t)$ to attain its stationary
173: form~$P_{\mathrm{st}}(r)$. In the cases examined in this Letter,
174: as we shall see,
175: the probability density function of~$r$ satisfies the
176: Fokker--Planck equation
177: \begin{equation}\label{eq:fp}
178: \partial_{t'} P=-\partial_r(D_1(r)P)+
179: \partial_rD_2(r)\partial_r P,
180: \end{equation}
181: where the form
182: of~$D_1(r)$ and~$D_2(r)$ depends on the flow and $t'=t/\tau$.
183: The stationary solution to Eq.~\eqref{eq:fp}
184: takes the potential form
185: $P_{\mathrm{st}}(r)=N\exp{[-E(r)/K_BT]}$, where
186: $N$ is the normalization constant and
187: $E(r)=-K_BT\int^r D_1(\rho)/D_2(\rho)\,d\rho$.
188: The finite-time solution admits the expansion
189: \begin{equation}\label{eq:tdfp}
190: P(r,t')=P_{\mathrm{st}}(r)
191: +\sum_{n=1}^{\infty}a_n \mathit{p}_n(r)\, e^{-t'/\sigma_n},
192: \end{equation}
193: where the coefficients~$a_n$ are fixed by~$P(r,0)$,
194: $p_n(r)$ are the eigenfunctions of the Fokker--Planck operator,
195: and~$\sigma_n$ are the reciprocals of its (strictly positive) eigenvalues,
196: arranged in descending order ($\sigma_n > \sigma_{n+1}$).
197: The relaxation time is thus defined as $t_{\mathrm{rel}}\equiv\sigma_1 \tau$.
198: \begin{figure}
199: \includegraphics[width=.45\columnwidth]{t_wi.eps}\hfill
200: \includegraphics[width=.45\columnwidth]{t_max.eps}
201: \caption{Elongational flow: (a)
202: rescaled relaxation time vs.~$\textit{Wi}$ for $b=400$.
203: The entropic force is given by the Warner law.
204: For small~$\textit{Wi}$, $t_{\mathrm{rel}}/\tau$
205: grows as $1/(1-2\textit{Wi})$, as can be seen by replacing~$\hat{f}(r)$
206: with~1 in Eq.~\eqref{eq:fp}.
207: For~$\textit{Wi}\gg\textit{Wi}_{\text{crit}}$,
208: the time needed to reach the asymptotic regime is set by the time
209: scale of the flow, $\gamma^{-1}$, and~$t_{\mathrm{rel}}/\tau$ decreases
210: as~$\textit{Wi}^{-1}$.
211: (b)~Rescaled maximum relaxation time~$t_{\text{max}}/\tau$
212: vs.~$\zeta_{s}/\zeta_{c}$ ($b=400$).}
213: \label{fig:relax}
214: \end{figure}
215:
216: As a first example, we examine the steady planar
217: elongational flow~$\bm v=\gamma(x,-y)$.
218: By assuming that the polymer extension in the $x$-direction is much greater
219: than in the $y$-direction, it is easy to derive from Eq.~\eqref{eq:dumbbell}
220: an equation of the form~\eqref{eq:fp}
221: with $\displaystyle D_1(r)=\mathit{Wi}\,r-\hat{f}(r)r/[2
222: \hat{\nu}(r)]$, $\displaystyle D_2(r)=[2b\hat{\nu}(r)]^{-1}$,
223: $\hat{f}(r)=f(rL)$, $\hat{\nu}(r)=\nu(rL)$, $b=L^2/R_0^2$,
224: and~$\textit{Wi}=\gamma\tau$~\cite{note}.
225: For~$\zeta_{s}=\zeta_{c}$, $t_{\mathrm{rel}}$ can be computed
226: from~\eqref{eq:tdfp} by solving a central two-point
227: connection problem for a generalized spheroidal wave
228: equation~\cite{05VB}. In the general case, $\zeta_{s}>\zeta_{c}$,
229: we resorted to a numerical
230: computation based on the variation--iteration method of quantum
231: mechanics~\cite{MF}.
232: In the vicinity of the coil--stretch transition ($\mathit{Wi}=1/2$)
233: $t_{\mathrm{rel}}$ shows a sharp peak (Fig.~\ref{fig:relax}).
234: In this range of~\textit{Wi}
235: there is a critical competition between the entropic force and the
236: velocity gradient
237: that makes the convergence time to the steady state extremely
238: long. This effect is strongly enhanced by the conformation-dependent drag;
239: the peak~$t_{\text{max}}$ indeed increases
240: with~$\zeta_{s}/\zeta_{c}$~(Fig.~\ref{fig:relax}).
241: Those extremely long relaxation times are intimately connected with
242: the observation of finite-time conformation
243: hysteresis~\cite{03Schetal,04SSC,05HL}.
244: For large enough $\zeta_{s}/\zeta_{c}$, there is a
245: narrow range of~\textit{Wi} around the critical value for the
246: coil--stretch transition where~$E(r)$ has a double well
247: structure~\cite{03Schetal,04SSC,05HL}.
248: The barrier height separating the coiled and the stretched state
249: is much greater than the thermal energy~$K_BT$,
250: and therefore the polymer remains trapped in its initial
251: configuration for an exceptionally long time (Fig.~\ref{fig:potential}).
252:
253: The discovery of elastic turbulence has recently
254: allowed the examination of single
255: polymer dynamics in a random smooth flow generated by viscoelastic
256: instabilities~\cite{00GS,05GCS}.
257: The velocity gradient fluctuates along
258: fluid trajectories; the average local deformation rates define the three
259: Lyapunov exponents of the flow.
260: Experimental observations have been accompanied by theoretical and
261: numerical studies based on the dumbbell model~
262: (see, e.g., Refs.~\cite{Chertkov00,05CMV,05MV}).
263: To analytically investigate polymer relaxation dynamics in
264: random flows, we initially
265: assume that the velocity field obeys the Batchelor--Kraichnan
266: statistics~\cite{68K}.
267: The velocity gradient is then a statistically isotropic and
268: parity invariant Gaussian process with zero mean and
269: correlation: $\langle \kappa_{ij}(t)\kappa_{kl}(s)\rangle=
270: 2\lambda\delta(t-s)[(d+1)\delta_{ik}\delta_{jl}-\delta_{ij}\delta_{kl}
271: -\delta_{il}\delta_{jk}]/[d(d-1)]$, where~$d$ is the dimension of the
272: flow and~$\lambda$ denotes the largest Lyapunov exponent.
273: In this context, we indicate by~$P(r,t')$
274: the probability density function of the extension
275: both with respect to thermal noise and the realizations
276: of the velocity field.
277: For the elongational flow Eq.~\eqref{eq:fp} was obtained under the
278: uniaxial approximation. On the contrary,
279: for the isotropic random flow Eq.~\eqref{eq:fp}
280: holds exactly and can be obtained by a Gaussian integration by parts followed
281: by integration over angular variables.
282: The drift and diffusion coefficients take the form
283: $\displaystyle D_1(r)=(d-1)/d\,\textit{Wi}
284: \;r-\hat{f}(r)r/
285: [2\hat{\nu}(r)]+(d-1)/[2b\hat{\nu}(r)r]$ and
286: $\displaystyle D_2(r)=\textit{Wi}\;
287: r^2/d+[2b\hat{\nu}(r)]^{-1}$ with~$\mathit{Wi}=\lambda\tau$.
288: The stationary distribution admits once more a potential form.
289: \begin{figure}[t]
290: \includegraphics[width=0.49\columnwidth]{pot_ext_random.eps}\hfill
291: \includegraphics[width=0.465\columnwidth]{pot_random_wi.eps}
292: \caption{(a) Effective energy at the coil--stretch
293: transition for a polyacrylamide (PAM) molecule
294: ($b=3953$, $\zeta_s/\zeta_c=6.87$) in the elongational flow (dashed line)
295: and the 3D Batchelor--Kraichnan flow (solid line).
296: The left vertical axis refers to the dashed line; the right vertical axis
297: refers to the solid line; (b) effective energy
298: in the random flow
299: for a PAM molecule with constant drag ($\zeta_s$=$\zeta_c$)
300: (from top to bottom~$\textit{Wi}=0.4,0.7,1.0,1.3$); (c) the same as (b),
301: but with~$\zeta_s=6.87\zeta_c$
302: (from top to bottom~$\textit{Wi}=0.28,0.33,0.38,0.43$).
303: When~$\zeta_s>\zeta_c$
304: the transition occurs in a much narrower range of~\textit{Wi}.
305: }
306: \label{fig:potential}
307: \end{figure}
308: For~$d=3$ and~$\zeta_{s}>\zeta_{c}$, the potential
309: displays a very wide well,
310: the effect of the
311: conformation-dependent drag being to increase the probability of large
312: extensions. % and hence make the potential flatter.
313: There is no evidence of pronounced double wells (Fig.~\ref{fig:potential}).
314: Near the coil--stretch transition, the effective
315: barrier heights separating the coiled and
316: stretched states
317: are indeed at most comparable to thermal energy.
318: For realistic~$\zeta_{s}/\zeta_{c}$
319: no conformation hysteresis is therefore expected to be observed in
320: random flows. The behavior of~$t_{\mathrm{rel}}$
321: vs.~$\mathit{Wi}$ is however analogous to the one encountered
322: in the elongational
323: flow: $t_{\mathrm{rel}}/\tau$
324: increases as $[1-\mathit{Wi}(d+2)/d]^{-1}$ at small~\textit{Wi} and
325: decreases as~$\mathit{Wi}^{-1}$ at large~\textit{Wi}.
326: A peak near the transition is present,
327: that becomes more and more pronounced
328: with increasing~$\zeta_{s}/\zeta_{c}$,
329: attaining values as large as about thirty times~$\tau$
330: (Fig.~\ref{fig:tdiz}).
331: The reason for this
332: behavior is the breadth of~$P_{\mathrm{st}}(r)$ and
333: the consequent large heterogeneity of accessible polymer configurations.
334: \begin{figure}[b]
335: \includegraphics[width=0.45\columnwidth]{t_wi_random.eps}\hfill
336: \includegraphics[width=0.435\columnwidth]{t_max_random.eps}
337: \caption{3D Batchelor--Kraichnan flow:
338: (a) $t_{\mathrm{rel}}/\tau$ vs.~$\mathit{Wi}$
339: for a PAM molecule ($b=3953$); (b)~$t_{\text{max}}/\tau$ for the following polymers:
340: DNA ($\bullet$, $b=191.5$; $\circ$, $b=260$; $\square$,
341: $b=565$; $+$, $b=2250$),
342: polystyrene ($\times$, $b=673$),
343: polyethyleneoxide (PEO) ($\blacktriangle$, $b=1666$),
344: Escherichia Coli DNA ($\vartriangle$, $b=9250$),
345: PAM ($\blacksquare$). Measures of~$b$ and~$\zeta_{s}/\zeta_{c}$
346: can be found in~\cite{97L,03Schetal,04SSC,05HL}.
347: Synthetic polymers are modeled
348: by the Warner law, whereas biological molecules are described by the
349: Marko--Siggia law. Relaxation times were computed by means of
350: the variation--iteration method \cite{MF}.
351: For~$\zeta_{s}=\zeta_{c}$ they
352: can be obtained by solving
353: an eigenvalue problem for a Heun equation~\cite{05MV}.}
354: \label{fig:tdiz}
355: \end{figure}
356:
357: To corroborate the results obtained in the context of
358: the short-correlated flow, we performed Brownian Dynamics
359: simulations of dumbbell molecules~\cite{Ito} in the random flow
360: introduced by Brunk et al.~\cite{brunk}. This model reproduces
361: the small-scale structure of a turbulent flow by means of a
362: statistically isotropic Gaussian velocity gradient.
363: The autocorrelation times of components of the strain and rotation tensors
364: are set to be multiple of the Kolmogorov time~$\tau_\eta$
365: by comparison with direct numerical simulations of 3D isotropic turbulence
366: ($\tau_S=2.3\tau_\eta$, $\tau_R=7.2\tau_\eta$).
367: The Lyapunov exponent of this flow is
368: $\lambda\simeq 10\tau_{\eta}^{-1}$. We computed~$t_{\mathrm{rel}}$ as
369: the time of convergence of the moments $\langle r^n(t)\rangle$ to
370: their stationary value~$\langle r^n\rangle_\mathrm{st}$:
371: $t_{\mathrm{rel}}^{-1}=-\lim_{t\to\infty}\ln{[\langle r^n(t)\rangle
372: -\langle r^n\rangle_{\mathrm{st}}]}/t$, where the averages were taken
373: over an ensemble of realizations of the flow,
374: all with the same initial extension~$r(0)$.
375: The numerical difficulty arising from the singularity of the entropic
376: force at~$R=L$ has been overcome by exploiting the algorithm introduced
377: in~\cite{05CPT}.
378: The results shown in Fig.~\ref{fig:numerics} confirm
379: the scenario depicted in the context of the
380: short-correlated flow.
381: It is worth noting that the above definition
382: provides an operational method to measure~$t_{\mathrm{rel}}$ that can be
383: implemented in experiments.
384: \begin{figure}[t]
385: \includegraphics[width=0.47\columnwidth]{t_wi_BD_peo.eps}\hfill
386: \includegraphics[width=0.47\columnwidth]{t_wi_BD_pam.eps}
387: \caption{Brunk--Koch--Lion flow:
388: $t_{\mathrm{rel}}/\tau$ vs.~$\textit{Wi}=\lambda\tau$
389: from Brownian Dynamics simulations; \label{fig:numerics} (a)~PEO;
390: (b)~PAM.}
391: \end{figure}
392:
393: In summary, we have shown that the
394: equilibrium configuration of a polymer in a flow,
395: as well as the time a deformed molecule
396: takes on average to recover that configuration, depend sensitively
397: on the properties of the flow.
398: In the vicinity of the coil--stretch transition
399: the characteristic relaxation time
400: is much longer than the Zimm time,
401: both in elongational and random flows.
402: In other words, the effective~\textit{Wi} differs considerably
403: from the ``bare'' one~\cite{note_2}. This effect is strongly amplified when the
404: drag coefficient depends on the conformation of the polymer,
405: and may play an important role in
406: drag-reducing turbulent flows,
407: where the strain rate often fluctuates around values typical of the
408: coil--stretch transition~\cite{00SW}.
409: Our conclusions thus suggest
410: that the conformation-dependent drag should be included as
411: a basic ingredient of continuum models of polymer solutions,
412: calling for further theoretical, experimental and numerical
413: study.
414:
415: \acknowledgments
416: The authors gratefully acknowledge inspiring discussions with M. Chertkov,
417: S.~Gerashchenko, M.~Martins Afonso and V.~Steinberg.
418: This work has been partially supported by the EU (contract
419: HPRN-CT-2002-00300).
420:
421: \begin{thebibliography}{29}
422:
423: \bibitem{chu}
424: S.~Chu, Phil. Trans. R. Soc. Lond. A \textbf{361}, 689 (2003).
425:
426: \bibitem{larson}
427: R.~G.~Larson, J. Rheol. \textbf{49}, 1 (2005).
428:
429: \bibitem{shaqfeh}
430: E.~S.~G.~Shaqfeh, J. Non-Newton. Fluid Mech. \textbf{130}, 1 (2005).
431:
432: \bibitem{95GB}
433: A.~Gyr and H.~W.~Bewersdorff, {\itshape Drag reduction of Turbulent Flows
434: by Additives} (Kluwer Academic, Dordrecht, Boston, 1995)
435:
436: \bibitem{05GCS}
437: S.~Gerashchenko, C.~Chevallard, and V.~Steinberg, Europhys. Lett. \textbf{71}, 221 (2005).
438:
439: \bibitem{53R}
440: P.~E.~Rouse, Jr., J.~Chem.~Phys. \textbf{21}, 1272 (1953).
441:
442: \bibitem{zimm}
443: B.~H.~Zimm, J. Chem. Phys. \textbf{24}, 269 (1956).
444:
445: \bibitem{97QBC}
446: S.~R.~Quake, H.~Babcock, and S.~Chu, Nature \textbf{388}, 151 (1997).
447:
448: \bibitem{94PQSC}
449: T.~T.~Perkins {\itshape et al.}, Science \textbf{264}, 822 (1994).
450:
451: \bibitem{96MCVCC}
452: S.~Manneville {\itshape et al.}, Europhys. Lett. \textbf{36}, 413 (1996).
453:
454: \bibitem{95MB}
455: Y.~Marciano and F.~Brochard-Wyart, Macromolecules \textbf{28}, 985 (1995).
456:
457: \bibitem{02RZ}
458: R.~Rzehak and W.~Zimmermann, Europhys. Lett. \textbf{59}, 779 (2002).
459:
460: \bibitem{74DG}
461: P.~G.~de~Gennes, J. Chem. Phys. \textbf{60}, 5030 (1974).
462:
463: \bibitem{05CMV}
464: A.~Celani, S.~Musacchio, and D.~Vincenzi, J. Stat. Phys. \textbf{118}, 531 (2005).
465:
466: \bibitem{05MV}
467: M.~Martins~Afonso and D.~Vincenzi, J. Fluid Mech. \textbf{540}, 99 (2005).
468:
469: \bibitem{05VB}
470: D.~Vincenzi and E. Bodenschatz, J. Phys. A: Math. Gen. \textbf{39}, 10691
471: (2006)
472:
473: \bibitem{TKCW}
474: V.~Tirtaatmadja, G.~H.~McKinley, and J.~J.~Cooper-White, Phys. Fluids
475: \textbf{18}, 043101 (2006).
476:
477: \bibitem{03Schetal}
478: C.~M.~Schroeder {\it et al.}, Science \textbf{301}, 1515 (2003).
479:
480: \bibitem{04SSC}
481: C.~M.~Schroeder, E.~S.~G.~Shaqfeh, and S.~Chu, Macromolecules \textbf{37},
482: 9242 (2004).
483:
484: \bibitem{05HL}
485: C.-C.~Hsieh and R.~G.~Larson, J. Rheol. \textbf{49}, 1081
486: (2005).
487:
488: \bibitem{Bird}
489: R. B. Bird \textit{et al.},
490: {\itshape Dynamics of Polymeric Liquids}, vol. 2, 2nd edition
491: (Wiley, New York, 1987).
492:
493:
494: \bibitem{99HQ}
495: J.~W.~Hatfield and S.~R.~Quake, Phys. Rev. Lett. \textbf{82}, 3548 (1999).
496:
497: \bibitem{74H}
498: E.~J.~Hinch, in {\itshape Proceedings of
499: Colloques Internationaux du CNRS No. 233}
500: (CNRS Editions, Paris, 1974), p.~241.
501:
502: \bibitem{75T}
503: R. I. Tanner, Trans. Soc. Rheol. \textbf{19}, 557 (1975)
504:
505: \bibitem{97L}
506: R.~G.~Larson {\it et al.}, Phys. Rev. E \textbf{55}, 1794 (1997).
507:
508: \bibitem{Doyle}
509: P. S. Doyle {\itshape et al.},
510: J. Non-Newton. Fluid Mech. \textbf{76}, 79 (1998).
511:
512: \bibitem{note}
513: Our definition of~\textit{Wi} follows the literature on elongational
514: flows and therefore is half the one commonly adopted in the context of random flows.
515:
516: \bibitem{MF}
517: P.~M. Morse and H.~Feshbach, {\it Methods of Theoretical Physics}
518: (McGraw--Hill, New York, 1953).
519:
520: \bibitem{00GS}
521: A.~Groisman and V.~Steinberg, Nature \textbf{405}, 53 (2000);
522: New J. Phys. \textbf{6}, 29 (2004);
523: T.~Burghelea, E.~Segre and V.~Steinberg,
524: Phys. Fluids \textbf{17}, 103101 (2005).
525:
526: \bibitem{Chertkov00}
527: M.~Chertkov, Phys. Rev. Lett. \textbf{84}, 4761 (2000);
528: E.~Bal\-kovsky, A.~Fouxon, and V.~Lebedev, Phys. Rev. Lett. \textbf{84}, 4765
529: (2000);
530: J.-L. Thiffeault, Phys. Lett. A \textbf{308}, 445 (2003);
531: M.~Chertkov {\it et al.}, J. Fluid Mech. \textbf{531}, 251 (2005);
532: A.~Puliafito and K.~Turitsyn, Physica D \textbf{211}, 9 (2005);
533: J.~Davoudi and J. Schumacher, Phys. Fluids \textbf{18}, 025103 (2006).
534:
535: \bibitem{68K}
536: R.~H.~Kraichnan, Phys. Fluids \textbf{11}, 945 (1968).
537:
538: \bibitem{Ito}
539: The integration algorithm is based on the It\^o stochastic differential equation
540: equivalent to Eq.~\eqref{eq:dumbbell}, $d\bm R=
541: \{\bm\kappa(t)\cdot\bm{R}-f(R){\bm R}/[2\tau\nu(R)]-R_0^2\nu'(R)\bm R/[2\tau\nu^2(R)R]\}dt+
542: \sqrt{R_0^2/[\tau \nu(R)]}\,d\bm W(t)$~\cite{Oettinger}; $\bm W(t)$ is the Brownian motion.
543:
544: \bibitem{Oettinger}
545: H. C. \"Ottinger, \textit{Stochastic Processes in Polymeric Fluids}
546: (Springer--Verlag, Berlin, Heidelberg, 1996).
547:
548: \bibitem{brunk}
549: B.~K.~Brunk, D.~L.~Koch, and L.~W.~Lion, Phys. Fluids \textbf{9}, 2670 (1997);
550: J. Fluid Mech. \textbf{364}, 81 (1998).
551:
552: \bibitem{05CPT}
553: A.~Celani, A.~Puliafito, and K.~Turitsyn, Europhys. Lett. \textbf{70},
554: 464 (2005).
555:
556: \bibitem{note_2}
557: This fact is related to the overestimation of the drag reducing~\textit{Wi}
558: encountered in numerical simulations~\cite{04D}.
559:
560: \bibitem{04D}
561: Y.~Dubief et al., J. Fluid Mech. \textbf{514}, 271 (2004).
562:
563: \bibitem{00SW}
564: K.~R.~Sreenivasan and C.~M.~White, J. Fluid Mech. \textbf{409}, 149 (2000);
565: E.~Balkovsky, A.~Fouxon, and V.~Lebedev, Phys. Rev. E \textbf{64}, 056301
566: (2001);
567: G. Boffetta, A.~Celani, and S. Musacchio, Phys. Rev. Lett. \textbf{91},
568: 034501 (2003); V.~S.~L'vov
569: {\it et al.}, {\it ibid.} \textbf{92}, 244503 (2004).
570:
571:
572: \end{thebibliography}
573: \end{document}
574:
575:
576:
577:
578:
579:
580:
581:
582:
583:
584:
585:
586:
587: \end{thebibliography}
588:
589:
590: \end{document}
591:
592: