pubs.acs.org/Langmuir © 2009 American Chemical Society
Shear-Induced Isotropic to Nematic Transition of Liquid-Crystal Polymers: Identification of Gap Thickness and Slipping Effects Hakima Mendil-Jakani,† Patrick Baroni, and Laurence Noirez* Laboratoire L eon Brillouin (CEA-CNRS), Ce-Saclay, 91191 Gif-sur-Yvette C edex, France. † Present address: CEA/INAC/Structure et Propri et es d’Architectures Mol eculaires, UMR 5819 SPrAM (CEA-CNRS-UJF), Groupe Polym eres Conducteurs Ioniques, CEA-Grenoble, 17, rue des Martyrs, 38054 Grenoble, Cedex 9, France. Received November 20, 2008. Revised Manuscript Received February 21, 2009 The shear-induced isotropic-nematic transition is a generic property of liquid-crystalline polymer melts which is identified by the emergence in the isotropic phase of a strong birefringence above a critical shear rate. Although spectacular, this transition cannot be explained on the basis of a conventional approach (coupling with pretransitional fluctuations or with viscoelastic relaxation times). We investigate the asymptotic rheo-optical behavior of the shearinduced phase. The sample is an unentangled cyanobiphenyl side-chain polyacrylate. We show that the birefringence increases almost linearly and then saturates above a given shear rate depending on the gap thickness. Similarly, the shear stress versus shear rate curve exhibits for the same thickness and using the same substrate (quartz), an overshoot occurring at about the same shear rates followed by the stress plateau (indicating a stress-optical equivalence). In contrast, when the substrate is different, the stress-optical equivalence is no more valid. We interpret the overshoot and the plateau observed in birefringence and in shear stress curves as the entrance in a so far unidentified macroscopic sliding regime. The slipping mechanism is corroborated by the recent identification of macroscopic long-range correlations in unentangled polymer melt and contrasts with a description in terms of a nucleation-growth process as in wormlike micellar solutions.
I. Introduction When referring to the flow properties of liquid crystals, the fundamental issues are usually whether and how the shear flow induces the orientation in the mesomorphic phases.1 At temperatures above the nematic-isotropic transition (TNI), very few experiments concern the study of the isotropic phase since in this phase (except very close to TNI) the molecules are expected to behave as simple Newtonian liquids. However, the study of the flow behavior of the isotropic phase exhibited by both thermotropic and lyotropic liquid crystals has revealed spectacular nonlinear effects assimilated to a nonequilibrium isotropic-nematic (I-N) phase transition.2 In lyotropic systems, the study of the shear-induced transition of solutions of wormlike micellar solutions benefits of an abundant literature3 and is still a subject of active debate. In thermotropic systems, although the shear-induced isotropic-nematic transition is generic and identified since 1995 on main-chain liquid-crystal (LC) polymers4 *Corresponding author: e-mail
[email protected]. (1) de Gennes, P. G.; Prost, J. The Physics of Liquid Crystals; Oxford University Press: New York, 1974; Chapters 3, 5. Leslie, F. M.; Adv. Liq. Cryst. 1979, 4.Chandrasekhar, S. Liquid Crystals; Cambridge University Press: New York, 1992; Chapter 3. (2) Hess, S. Z. Naturforsch. 1976, 31A, 1507. Onuki, A.; Kawasaki, K. Ann. Phys. (N.Y.) 1979, 121, 456. Olmsted, P. D.; Goldbart, P. Phys. Rev. A 1990, 41, 4578. Olmsted, P. D.; Goldbart, P. Phys. Rev. A 1992, 46, 4966. Olmsted, P. D. Rheol. Acta 2008, 47, 283. (3) Schmitt, V.; Lequeux, F.; Pousse, A.; Roux, D. Langmuir 1994, 10, 955. Berret, J. F.; Roux, D. C.; Porte, G.; Linder, P. Europhys. Lett. 1994, 25, 521. Decruppe, J. P.; Cressely, R.; Makhloufi, R.; Cappelaere, E. Colloid Polym. Sci. 1995, 273, 346. Cappelaere, E.; Cressely, R.; Decruppe, J. P. Colloids Surf., A 1995, 104, 353. Cappelaere, E.; Berret, J. F.; Decruppe, J. P.; Cressely, R.; Lindner, P. Phys. Rev. E 1997, 56, 1869. (4) Mather, P. T.; Romo-Uribe, A.; Han, C. D.; Kim, S. S. Macromolecules 1997, 30, 7977. (5) Pujolle-Robic, C.; Noirez, L. Nature (London) 2001, 409, 167. PujolleRobic, C.; Olmsted, P. D.; Noirez, L. Europhys. Lett. 2002, 59, 364. Huskic, M.; Zigon, M. Polymer 2003, 44, 6187. Huski, M.; Zigon, M. Polymer 2003, 44, 6187.
5248
DOI: 10.1021/la803848h
and since 2001 in side-chain LC polymers,5 these spectacular nonequilibrium properties (Figure 1) hidden in the isotropic phase are still poorly known and studied. From a theoretical point of view, the prediction precedes the experimental observation. Hess in 19762 foresees the creation, above a critical shear rate, of a nonequilibrium nematic phase from the isotropic phase and thus coexisting with it. Later, Olmsted and Goldbart in 19922 detailed the mechanism of the transition, originally for thermotropic liquid crystals2 and later for wormlike micellar solutions.6 Finally, recently, Ilg and Hess elaborated a specific theoretical model to take into account the viscoelastic contribution of the polymer component in LC polymers.7 This model is able to predict a low-frequency elastic plateau in the dynamic relaxation spectra. The nonequilibrium transition is generally explained by a flow coupling to the lifetime of the orientational pretransitional fluctuations.2,7 The emergence of the shearinduced nematic phase, in wormlike micelles, is interpreted by a “nucleation-unidirectional growth” process,3 i.e., a coexistence of two defined shear bands related to the isotropic and the shearinduced phases. The proportion of the highly sheared band is supposed to evolve linearly versus applied shear rate in the stress plateau region of the flow curve up to a shear rate where the band invades the whole gap.3,8 The highly sheared band is associated with the shear-induced nucleating band. Similar physics depict the more general case of shear-banding flow of nonmesomorphic systems as entangled polymer solutions or even entangled melts.9 A radically (6) Olmsted, P. D.; David Lu, C. Y. Phys. Rev. E 1999, 60, 4397. Cates, M. E.; Fielding, S. M. Adv. Phys. 2006, 55, 799. (7) Hess, S.; Ilg, P. Rheol. Acta 2005, 44, 465. Ilg, P.; Hess, S. J. Non-Newtonian Fluid Mech. 2006, 134, 2. (8) Salmon, J. B.; Colin, A.; Manneville, S. Phys. Rev. Lett. 2003, 90 228303. (9) Cates, M. E.; Fielding, S. Adv. 2006, 55, 799. Adams, J. M.; Olmsted P. D. Phys. Rev. Lett. 2009, 102, 67801.
Published on Web 3/24/2009
Langmuir 2009, 25(9), 5248–5252
Mendil-Jakani et al.
Figure 1. Scheme of the birefringence setup and photographs of the shear-induced isotropic-nematic phase transition. Right photograph: (velocity, velocity gradient) plane (1 mm gap thickness). Bottom: snapshots taken in the (velocity, neutral axis) plane between crossed polarizers with white light. The first photo corresponds to the isotropic phase at rest. The succession of colors is obtained by increasing shear rate at T > TNI. The thickness of the LC polymer sample is typically 100 μm.
different scenario is however proposed by Callaghan and co-workers,10 performing local velocity measurements using the NMR technique, since they suggest that the highly ordered phase is not the highest sheared one. The flow field suggests a gel-like behavior, fundamentally different from the coexistence of two well-defined shear bands supporting different shear rates. It points out failures on the true nature of the flow behavior of complex systems. In the present article, we examine the shear-induced behavior of the isotropic phase of a side-chain liquid-crystalline polymer, using rheo-optical and shear stress measurements and analyzing, in particular, the asymptotic evolution. The totality of the experiments is carried out by heating the temperature from the nematic-isotropic transition to avoid any misinterpretation in terms of hysteresis effects and is measured significantly above the transition with respect to the temperature accuracy ((0.05 C for the birefringence experiments). Finally, it is important to note that side-chain LC polymers in contrast to main-chain polymers do not exhibit a significant hysteresis. These experiments evidence important new features: the gap thickness dependence of the shear-induced phase transition and the interpretation of the overshoot and the plateau observed in both birefringence and shear stress curves as the occurrence of sliding transitions that confirm the gel-like character of the isotropic phase under flow. II. Experimental Section A. Sample. The polymer studied is a low polydisperse LC polymer of chemical formula
It displays the following mesophase sequence: I (isotropic)119 C-N (nematic)-30 C-glassy state. TNI, the nematicisotropic transition temperature, is determined by optical observation by increasing the temperature (a difference of less than 0.1 C on the transitions is observed by decreasing the temperature). The polymerization degree of the LC polymer (10) Fisher, E.; Callaghan, P. T. Europhys. Lett. 2000, 50, 803; Phys. Rev. E 2001, 64, 11501.
Langmuir 2009, 25(9), 5248–5252
Article (DP = 270) is lower than the polymerization degree between entanglements which is DPe = 440 for a polymer (poly(butyl acrylate)) having the same backbone.11 The hindrance of the liquid crystal moieties of the LC polymer contributes to an additional hardening of the backbone and excludes the possibility to entangle. B. Technique. The rheo-optical experiments were performed in transmission mode under cross-polarized microscopy (Olympus BX60) in the (velocity, vorticity) plane with a magnification 100. The cross-polarizers are oriented at 45 to the flow direction and a wavelength λ = 4700 ( 150 A˚. Steady-state shear flow conditions were ensured using a home-improved CSS450 Linkam shear cell (with a temperature gradient < (0.05 C) equipped with a quartz plate-plate fixture. The transmitted intensity I is measured with a photosensitive diode and normalized to the uncrossed polarizer intensity I0. The averaged birefringence ÆΔnxæ is measured as a function of the shear rate using the formula I/I0 = sin2(ÆΔnxæeπ/λ), where e is the gap thickness. ÆΔnxæ is measured as a function of the shear rate. It is averaged over the whole sample thickness (weighted of the fractions of isotropic and shear induced phases). The rheological measurements have been carried out in steady-state mode with the ARES rheometer (TA Instruments). The thermal environment (air-pulsed oven) ensures a temperature stability of (0.05 C. The accuracy on the temperature with respect to the birefringence data is (1 C. The sample was placed between plate-plate fixtures (10 mm diameter) made of quartz or aluminum.
III. Results A. Low Shear Rates Flow Birefringence Behavior. Figure 2 displays the birefringence evolution of the LC polymer melt at T - TNI = ΔT = +1 C over the nematic-isotropic transition and e = 100 μm gap thickness. First of all, it is important to outline that the birefringence does not originate from the polymer chain. The equivalent chain without liquid crystal moieties is a poly(butyl acrylate) melt of low polymerization degree (270) observed at 90 C over the glass transition. At these temperatures and for the shear rates applied, the poly (butyl acrylate) chain does not exhibit birefringence. The birefringence exhibited in the isotropic phase by the LC polymers under flow originates from the alignment of the liquid crystal moieties. The birefringence curve exhibits two distinct regimes. At low shear rates, the birefringence is close to zero and slightly shear-dependent. This regime is the so-called paranematic : phase2 or flow birefringence.1,2 Above a critical shear rate γ*, a strong increase of the birefringence is observed. This nonlinear behavior is interpreted in agreement with the pretransitional fluctuation coupling model,2,7 as a nonequilibrium first-order phase transition characterized by an alignment of the liquid crystal moieties along the velocity direction. The shear-induced phase is of shear-aligning type. The increase of the birefringence is interpreted by an increase of the proportion of the nematic phase coexisting with the isotropic phase at constant shear stress.2 Because of the invariance of the shear stress during the filling regime, the shear-aligning regime remains stable. The inset of Figure 2 displays the evolution of the : critical shear rate γ* defined by the interception of the basis line to the birefringence curve tangent as a function of the temperature. : The critical shear rate γ* decreases all the more with increasing temperature from TNI, according to the low shear rates : model.1,2 A (nonequilibrium) characteristic time τ* = 1/γ* can (11) Lakrout, H.; Creton, C.; Ahn, D.; Shull, K. R. Macromolecules 2001, 34, 7448.
DOI: 10.1021/la803848h
5249
Article
Mendil-Jakani et al.
Figure 2. Evolution of ÆΔnxæ vs shear rate at 100 μm gap thickness at T = TNI + 1 C (λ = 4700 A˚). The black left bottom inset represents the isotropic phase at rest; the blue right bottom inset represents the shear-induced nematic phase observed with crosspolarized microscopy. The director is aligned along the velocity : direction. The inset represents the critical shear rate γ* as a function of the temperature at 100 μm gap thickness. The accuracy on the temperature determination is smaller than 0.1 C.
be deduced from the critical shear rate. τ* expresses a characteristic time. This critical time τ*, varying from 0.20 ( 0.01 up to 0.040 ( 0.002 s in the temperature range explored, is about 103-105 times larger than the orientational order fluctuations lifetime12 excluding any coupling with the shear flow. B. Thickness Effect on the Flow Birefringence. Figure 3a displays the evolution of the birefringence as a function of the shear rate at ΔT = 1 ( 0.05 C above TNI, for gap thicknesses ranging from 75 up to 125 μm. A first regime is characterized by an apparent superposition of the birefringence curve from the quiescent state up to 20 s-1 : whatever the thickness, up to a second critical shear rate, γc, above which the curves do not longer superimpose. : Above γc, the birefringence curves show an overshoot before leveling into a broad plateau region which is thickness dependent. These crucial observations are unpredicted. The values of : the birefringence plateau and of the critical shear rate γc rise progressively with decreasing gap size. The overshoot is not a transient behavior since measurements are carried out on an integration time range (100-900 s) ensuring a stationary state. Figure 3b displays the birefringence curves measured for increasing (forward curve) and for decreasing (backward curve) shear rates. The “forward” and “backward” curves are superimposed. No hysteresis phenomenon is observed. What is the origin of the birefringence plateau and its thickness dependence? How does one explain the overshoot?
IV. Discussion
: : : Table 1 gathers the critical shear rates γc1, γc2, and γc3 corresponding to the thickness 125, 100, and 75 μm, respectively. By renormalizing the critical shear rate to the gap thickness (v1 = : γce), one obtains a quantity which is constant, independently of the thickness. This quantity has the dimension of a velocity (v1). It indicates that a velocity and not the shear rate, is a critical parameter of the dynamical transition. Since the shear rate is the ratio of the applied velocity (the highest velocity) to the thickness, v1 is the highest velocity in the sample before the dynamical birefringence transition occurs. To demonstrate that this dynamic transition is related to a slip mechanism (slip including apparent, molecular or effective slip, fractures, dynamic heterogeneities), the evolution of the shear stress versus shear rate (12) Reys, V.; Dormoy, Y.; Gallani, J. L.; Martinoty, P.; Le Barny, P.; Dubois, J. C. Phys. Rev. Lett. 1988, 61, 2340.
5250
DOI: 10.1021/la803848h
Figure 3. Evolution of ÆΔnxæ vs shear rate at T = TNI + 1 C.
(a) At different gap thicknesses: (O) 75 μm, (b) 100 μm, and (4) 125 μm. (b) At 100 μm gap thickness, obtained by shearing the isotropic melt from low to high shear rates (b) and from high to low shear rates (O). (c) Shear stress vs shear rate exhibited at T = 122.5 ( 1 C, measured on different substrates: b (100 μm): quartz substrate; 0 (100 μm): aluminum substrate. In the inset, the shear stress vs shear rate is reported at 200 μm (4) and at 100 μm (0) (aluminum substrate). ARESII rheometer data (10 mm diameter plate-plate fixtures).
is measured (Figure 3c). This measurement is in addition carried out on two different substrates: the quartz (which corresponds to conditions identical to the birefringence experiment) and the aluminum (which is the common substrate used in rheology). These data are carried out with the same plate-plate geometry, the same shear rate, and at about the same temperature, 2.5 ( 1 C over the isotropic-nematic transition. The first important information is that a similar transition is exhibited by the shear stress above a critical shear rate. Second, the comparison of the shear stress measurements carried out on quartz substrates shows a remarkable similarity with the birefringence curve (Figure 3a). The birefringence transition observed at 24 s-1 coincides with the overshoot of the shear stress curve. There is a stress-optical equivalence. The shear stress transition is the signature of a heterogeneous flow coherent with a slip regime. Finally, the most important information in favor of a slip mechanism in addition to the gap dependence is the discrepancy at high shear rates between the curves measured on quartz and on aluminum at the same gap thickness (Figure 3c). Since the substrate is the only different parameter between these two curves, the critical shear rate at which the shear stress curve collapses indicates a wall slip transition. The sliding conditions are deterLangmuir 2009, 25(9), 5248–5252
Mendil-Jakani et al.
Article Table 1
e (μm)
: γc (s-1)
vc (mm s-1)
75 100 125
32 24 20
2.4 2.4 2.5
mined by the strength and the dynamic of the interactions between the substrate and the material. Wall slip is favored when the fluid/substrate boundary interactions are weak15 and when the inverse of the characteristic time of adhesion is smaller than the frequency of the solicitation introduced by the flow. In the case of liquid crystals, these moieties are known to easily anchor on many substrates.16 A positional order (smectic-like) of molecules is induced, even in the isotropic phase, at the surface of the substrate.16 It is thus probable that a sliding interface may be created from the first molecular layers close to the substrate. The overshoot (Figure 3a,c) may result from a succession of complex slip states, producing a sharp decrease and then a stabilization to a plateau at high shear rates. This plateau indicates that the additional energy transmitted by increasing the shear flow is integrally dissipated since the birefringence and the shear stress does evolve anymore. v1 is thus the highest velocity that the substrate (quartz) can impose to the material. As illustrated in Figure 3c, this velocity limit differs following the nature of the substrate. The interface created between the material and the substrate plays a crucial role in this dissipation mechanism. This strong dissipative area might involve voids, autophobicity, stickslip,17 etc. It is hardly describable using continuous boundary models that mainly describe “soft” slip effects as measured by methods of local determination of velocity profile and that do not include the wall slip situation.18 Other studies on interfacial mechanisms are needed. The present gap effect is, to our knowledge, so far unknown and unexpected in the frame of the current theoretical considerations.2,7 It also differs from rheological schemes considering that the larger the gap size, the less the wall slip effect is.13 However, the inverse gap effect has been already observed but in the case of microstructured assembly (i.e., a gel-like system).14 The value of the yield stress rises progressively with decreasing gap size. The flow properties change by a progressive microstructural rearrangement. The isotropic phase of the LC polymer behaves as a microstructured assembly. How does one explain the thickness dependence in LC polymer melts? The lifetime of pretransitional orientational fluctuations has been determined just above the nematic-isotropic transition.12 The comparison of these time scales to the flow rate19 shows that the shear-induced birefringence can neither be explained as a direct flow coupling with the lifetime of these pretransitional (13) Henson, D. J.; Mackay, M. E. J. Rheol. 1995, 39, 359. Barnes, H. A. J. NonNewtonian Fluid Mech. 1995, 56, 221. Wang, S.-Q.; Drda, P. Macromol. Chem. Phys. 1997, 198, 673. (14) Clasen, C.; McKinley, G. H. J. Non-Newtonian Fluid Mech. 2004, 124, 1. (15) Zhu, Y.; Granick, S. Phys. Rev. Lett. 2004, 93, 96101. Massey, G.; Hervet, H.; Leger, L. Europhys. Lett. 1998, 43, 83. Pit, R.; Hervet, H.; Leger, L. Tribol. Lett. 1999, 7, 147. Denn, M. M. Annu. Rev. Fluid Mech. 2001, 33, 265.Lauga, E.; Brenner, M. P.; Stone, H. A. The No-Slip boundary Condition. In Handbook of Experimental Fluid Dynamics; Tropea, C., Yarin, A., Foss, J. F., Eds.; Springer: New York, 2007. (16) Jerome, B. Rep. Prog. Phys. 1991, 54, 391. Jer^ome, B.; Commandeur, J.; de Jeu, W. H. Liq. Cryst. 1997, 22, 685. (17) Pujolle-Robic, C.; Noirez, L. Phys. Rev. E 2003, 68, 61706. (18) de Gennes, P.-G. C. R. Acad. Sci. Paris 1979, 288B, 219. Durliat, E.; Hervet, H.; Leger, L. Europhys. Lett. 1997, 38, 383. Joseph, P.; Tabeling, P. Phys. Rev. E 2005, 71, 35303. Reiter, G.; Khanna, R. Phys. Rev. Lett. 2000, 85, 2753. (19) Noirez, L. Phys. Rev. E 2005, 72, 051701. Mendil, H.; Baroni, P.; Noirez, L. Europhys. Lett. 2005, 72, 983.
Langmuir 2009, 25(9), 5248–5252
Figure 4. Dynamic relaxation spectrum displaying the elastic G0 (ω) (b) vs frequency measured at T = TNI + 6 C and at 0.200 mm gap thickness (alumina fixtures, 10 mm diameter, 3% strain amplitude). The inset displays the typical flow behavior expected in the case of a liquid. The complete dynamic analysis is published in ref 19b.
fluctuations nor with the viscoelastic terminal time.20 These conclusions have led us to consider longer relaxation time scales and correlatively a supramolecular cohesion21,22 far above any transitions. Figure 4 illustrates the dynamic relaxation spectrum displayed by a sample of 0.200 mm thickness of this polymer in the isotropic phase. The low-frequency behavior tends to a plateau for both elastic and viscous moduli. The isotropic phase displays neither the dynamic of a simple liquid nor that of a viscoelastic melt that would exhibit a terminal flow behavior (inset of Figure 4),20 but a solidlike behavior. Indeed, the independence of the moduli with respect to the frequency together with an elastic component (shear modulus G0 ) higher than the viscous contribution (loss modulus G00 ) indicates a solidlike response. Long-range elastic correlations are predominant.19,22 The solicitation of these long time scales explains probably the emergence of shear-induced birefringence. But also, the shear flow breaks and fractures the initial “gel” or “solid” into smaller elastic parts. Indeed, the dynamic relaxation has also shown that the increase of the strain produces a weakening of the elastic signal.22,,24,26 A steady-state shear produces a similar effect (the strain is infinite) until reaching a final size of the grains determined by the counterbalance of their elasticity. Such self-assembled elastic unities are prone to produce “lubricity”, slip layers,23 fracture-like planes.
V. Conclusions The spectacular shear-induced birefringence observed in the isotropic phase of LC polymers2 cannot be explained by a coupling to the conventional relaxation times (fluctuation lifetimes, viscoelastic relaxation times). To determine the origin of this nonequilibrium phase transition, we have extended our study to a wide range of shear rates and the possible influence of the gap thickness. A dynamic transition from a linear increase of the shear induced birefringence to a birefringence plateau is identified at (20) Ferry, J. D. Viscoelastic Properties of Polymers, 3rd ed.; John Wiley: New York, 1980.Macosko, C. W. Rheology, Principles, Measurements and Applications; Wiley: New York, 1994. (21) Baroni, P.; Mendil, H.; Noirez, L. Patent 05 10988 (27/11/2005). (22) Mendil, H.; Baroni, P.; Noirez, L. Eur. Phys. J. E 2006, 19, 77. Mendil, H.; Baroni, P.; Grillo, I.; Noirez, L. Phys. Rev. Lett. 2006, 96, 077801. (23) Charitat, T.;Joanny, J.-F. Eur. Phys. J. E 2000, 3, 369.Baumberger, T.; Caroli, C.; Ronsin, O. Phys. Rev. Lett. 2002, 88, 075509-1. (24) Gallani, J. L..; et al. Phys. Rev. Lett. 1994, 72, 2109. Martinoty, P..; et al. Macromolecules 1999, 32, 1746. (25) Decruppe, J. P.; Greffier, O.; Manneville, S.; Lerouge, S. Phys. Rev. E 2006, 73, 61509. (26) Osaki, K.; Inouet, T. Macromolecules 1996, 29, 7622. Pellens, L.; Vermant, J.; Mewis, J. Macromolecules 2005, 38, 1911. Mead, D. W.; Larson, R. G. Macromolecules 1990, 23, 2524.
DOI: 10.1021/la803848h
5251
Article
Mendil-Jakani et al.
high shear rates (Figure 3a). This birefringence plateau looks similar as the saturation plateau described in the shear banding theoretical model4 and can be associated with the shear stress behavior observed by rheological measurements (Figure 3c). However, by varying the sheared thickness, an important and novel property is revealed: the dimensional character of the shear-induced phase. The birefringence curves are characterized by three regimes: a low shear rate apparently independent of the thickness, an overshoot followed by an unpredicted broad thickness-dependent plateau regime. The birefringence plateau is lowered and shifted toward smaller shear rates with increasing the gap size. We interpret the birefringence plateau by the entrance in a strong slippage regime when the shear plate is moving over a velocity limit estimated at 2.4 mm-1 (at 1 C above the nematic-isotropic transition on the quartz substrate). The slip mechanism is confirmed by an identical shear stress transition when the shear stress is measured on the quartz surface. Slippage is unexpected since the LC polymer is a low viscous unentangled fluid sheared at low rates with respect to the conventional relaxation time scales.19 We interpret the slippage ability as the signature of the long-range elastic character of the melt. The isotropic phase of the LC polymer behaves as a soft solid. This elastic cohesion has been already revealed by dynamic relaxation measurements22,24 by applying small strain oscillations that are not damaging the solidlike correlations and the anchoring to the substrate. The thickness effect is interpreted as resulting from the loss of the elasticity with increasing thickness.
5252
DOI: 10.1021/la803848h
The interpretation in terms of elasticity is also coherent with NMR studies of Fisher and Callaghan7 showing that the shear-induced birefringence of micellar solutions results from the solicitation of a gel-like assembly, contrasting with the description of the coexistence at constant stress of shear bands of defined viscosities. Other studies of micellar systems using dynamic light scattering8 and local velocity measurements25 have shown that these latter deviate from a simple Newtonian description and is prone to significant slip in the whole range of shear rate studied, i.e., even in the isotropic branch of the flow curve. The macroscopic shear elasticity identified in the isotropic melt by dynamic relaxation22,24 is very likely at the origin of the shearinduced birefringence and in particular of its thickness dependence. The existence of long-range elastic correlations should shed new light on the shear-induced effects, on the necessity to reconsider the boundary conditions between the material and the substrate, and on the interpretation of some nonlinear effects. We suspect that many antagonistic results in rheophysics, as the nonrelevance of stress-optical rules26 especially when the substrates are different, may originate from miscontrolled boundary conditions. The nonlinear theoretical models reproduce nicely qualitatively the experiments but miss, for the moment, the pronounced dissipative component involved in these phenomena. Concerning the dynamic relaxation (linear rheology), new theoretical models already predict strong anchoring effects on the value of the dynamic moduli in thermotropic nematic polymers.27 (27) Choate, E. P.; Cui, Z.; Forest, M. G. Rheol. Acta 2008, 47, 223.
Langmuir 2009, 25(9), 5248–5252
