Effect of Shear on Intramolecular and Intermolecular Association

Jun 3, 2005 - The growth of clusters has been investigated at various stages in the course of the cross-linking process by quenching the reaction mixt...
0 downloads 10 Views 258KB Size
Download PDF
J. Phys. Chem. B 2005, 109, 12329-12336

12329

Effect of Shear on Intramolecular and Intermolecular Association during Cross-Linking of Hydroxyethylcellulose in Dilute Aqueous Solutions Atoosa Maleki, Anna-Lena Kjøniksen,* and Bo Nystro1 m Department of Chemistry, UniVersity of Oslo, P.O. Box 1033, Blindern, N-0315, Oslo, Norway ReceiVed: March 18, 2005; In Final Form: April 22, 2005

Intramolecular and intermolecular associations of dilute aqueous alkali solutions of hydroxyethylcellulose (HEC) in the presence of a chemical cross-linker agent (divinyl sulfone, DVS) are studied with the aid of dynamic light scattering (DLS) and rheological methods. At quiescent state, DLS detected only interchain aggregation of HEC during the cross-linker reaction, and the magnitude and start of this effect depend on the cross-linker concentration. The growth of clusters has been investigated at various stages in the course of the cross-linking process by quenching the reaction mixture to a lower pH. After quenching, no further association of the species occurred. When the dilute reaction mixtures are subjected to shear, intrapolymer cross-linking with contraction of the molecules is observed, and at moderate shear rates this effect is followed by interpolymer cross-linking and the formation of aggregates at longer times. The rate of the growth of the multichain aggregates decreases with increasing shear rate, and at sufficiently high shear rates no cross-linking effect is observed. Depending on the shear rate, the aggregates continue to grow until they reach a certain size where an incipient breakup of interaggregate chains can be observed. The delicate interplay between intramolecular and intermolecular association effects is governed by factors such as the magnitude of the shear rate, polymer concentration, and cross-linker density.

Introduction The gelation of a semidilute polymer solution results from the formation of a connected network via intermolecular crosslinks that can be established by various mechanisms, such as attractive interactions between hydrophobic groups,1-2 temperature-induced attraction between polymer chains,3-5 and crosslinks mediated by a chemical cross-linker agent.6-8 However, if a dilute polymer solution is, for example, chemically crosslinked, the connectivity is lost and nonlinked aggregates of finite size are formed; these may be referred to as microgels.9 In colloidal suspensions, aggregation processes inevitably generate clusters or flocs10 containing a large number of individual particles connected into complex structures, which usually affect the physical properties of the system in which they have been formed. For example, aggregation rates as well as the sizes and structures of the aggregates will influence the dynamical and rheological features of the system. The tendency to form colloidal aggregates will depend on factors such as the stickiness of the associating species and shear flow.11,12 It is frequently found13-16 that shear flows tend to bring particles more rapidly in contact with each other than Brownian motion does, leading to a change in the aggregation dynamics by speeding up the process. The formation of flocs in the presence of a shear flow is known as “orthokinetic” aggregation, and this phenomenon has mostly been investigated in dilute suspensions of colloidal particles. In the same way as colloid particles with a low surface charge can form strong, rigid bonds and aggregate into large, tenuous flocs, it is an interesting analogy to follow the association phenomena in dilute polymer solutions in the presence of a chemical cross-linker agent. There is a lack of shear flow studies where association complexes are created through chemical cross-linking of polymer chains. * Corresponding author. E-mail: [email protected].

In the present work, the effect of shear rate on the crosslinking in dilute alkaline (pH ≈ 11.8) water solutions of hydroxyethylcellulose (HEC) in the presence of the difunctional cross-linker divinyl sulfone (DVS) will be examined. HEC is a hydrophilic polymer with a typical polysaccharide structure (see Figure 1a) that is not inclined to self-associate in dilute aqueous solutions, but intramolecular and/or intermolecular associations can be generated by the DVS cross-linker through reaction with the hydroxyl groups on the HEC backbone (see Figure 1b) at alkali conditions. The conjecture is that the hydroxyl groups at the C6 position are more reactive because of the reduced steric hindrance. The chemical structures of HEC and DVS, as well as the proposed scheme for intramolecular and intermolecular cross-linking of HEC, are depicted in Figure 1. Steady shear experiments at various pre-set shear rates on dilute HEC solutions with different amounts of DVS are conducted in this study. To gain insight into the dynamics of the cross-linking process of dilute HEC solutions at the quiescent state, dynamic light scattering measurements are carried out to monitor the growth of the aggregates during the chemical reaction, and the behavior of quenched solutions is also probed. Under quiescent conditions, Brownian motion dominates the aggregation kinetics. The aim of this work is to learn more about the influence of shear rate on the cross-linking process in dilute polymer solutions and to shed some light on the intricate interplay between intra- and interpolymer cross-linking. To the best of our knowledge, this is the first experimental rheological study that demonstrates both the intra- and intermolecular associations. Background In the past, several studies have been reported on the aggregation of colloidal particles in suspensions, mainly under the influence of Brownian dynamics; papers have also been published11-16 on the effect of shear flow. The degree of

10.1021/jp0514271 CCC: $30.25 © 2005 American Chemical Society Published on Web 06/03/2005

12330 J. Phys. Chem. B, Vol. 109, No. 25, 2005

Maleki et al.

Figure 1. (a) Chemical structure of HEC and the cross-linker DVS. (b) Mechanism of the cross-linker reaction and schematic representation of intra- and interpolymer cross-linking.

aggregation of colloidal particles exposed to a shear flow depends on factors such as the stickiness of the particles, collision frequency of the particles, and the magnitude of the shear stress. In the present investigation, the scenario is different with polymer chains with hydroxyl functions that can react with each other via the cross-linker agent. There are two possible types of cross-linking processes,17-19 that is, inter- and intrapolymer cross-linking (see Figure 1b). The former involves a cross-linked polymer via the coupling of two or more polymer chains, and this allows for the formation of interchain complexes; however, the latter does not alter the molecular weight of the cross-linked polymer but affects quantities related to the polymer chain dimension, because the cross-linking takes place

within the same polymer chain. In this case, the sizes of the individual polymer coils are expected to shrink.17,18 At quiescent conditions, or low shear rates, the aggregation dynamics will be governed by diffusion of chains; thereby, contacts between chains will be established and cross-linking reactions can proceed with the formation of multichain aggregates. When a sufficiently high shear rate is imposed on the system, a different scenario emerges in which shear viscosity stresses and hydrodynamic forces become the dominant transport mechanisms and the number of collisions may increase; however, at the same time the mechanical perturbations from the shear flow may reduce the number of cross-links between the chains. Initially, during the shear rate process it may be

Shear and Association during Cross-Linking of HEC more favorable for the chains to make intramolecular crosslinks rather than connections between polymer chains exposed to shear forces. The reason may be that rotations of chains can promote contacts between hydroxy groups on the same chain. However, after some time the many collisions between compressed entities through mechanical convections produce interpolymer cross-links and favor growth of aggregates. When the clusters are sufficiently large, they break up under the action of shear stresses, and as a result the aggregation rate levels off or decreases. The breakup process is usually considered to consist of two steps: one is an instantaneous breakup due solely to fluid stresses, and the other is a kinetic breakup generated by shear-induced collisions between interaggregate chains. When a high shear rate is applied to a dilute solution with an added cross-linker agent, the strong shear forces will rotate and stretch the chains and thereby suppress both the intra- and interpolymer cross-linking reactions. Experimental Section Materials and Solution Preparation. A sample of hydroxyethylcellulose with the commercial name Natrosol 250 GR (Lot no. V-0403) was obtained from Hercules, Aqualon Division. The degree of substitution of hydroxyethyl groups per repeating anhydroglucose unit of the polymer is 2.5 (given by the manufacturer). The weight average molecular weight of this sample was determined to be approximately 400 000 by means of intensity light scattering.20 From shear viscosity measurements (conducted on the rheometer described below) on dilute aqueous solutions of HEC, the intrinsic viscosity ([η]) is found to be 4.02 wt %-1, and the overlap concentration c* (estimated from c* ) 1/[η]) is 0.25 wt %, which is much higher than the concentrations considered in this investigation. All the measurements have been carried out in the dilute concentration regime, where no interconnected network can be formed. The crosslinking agent DVS was purchased from Merck and utilized without further purification. Millipore water was used for the preparation of all solutions. Dilute HEC solutions were dialyzed against Millipore water for at least 1 week to remove low-molecular-weight impurities and were thereafter freeze-dried. As a dialyzing membrane, regenerated cellulose with a molecular weight cutoff of 8000 (Spectrum Medical Industries) was used. After freeze-drying, the polymer was redissolved in aqueous alkaline (NaOH) media with pH of 11.8 (at this pH the cross-linker reaction proceeds); solutions with the desired polymer concentrations were prepared by weighing the components, and the samples were homogenized by stirring at room temperature for 1 day. At these conditions, the polymer and the cross-linker concentrations had no effect on the pH of the solution. Shortly before the commencement of experiment, the prescribed amount of DVS (in the range 10-60 µL/g, or 0.1-0.6 molal) was added to the sample, and a fast homogenization of the solution was performed. The same procedure to obtain homogeneous solutions was repeated for all samples to ensure good reproducibility of the measurements. All experiments were carried out at 25 °C. In the reported quenching experiments, the solutions were quenched rapidly by adding a few drops of concentrated HCl to the samples to lower the pH to acid conditions (pH ≈ 1.5) and thereby stop the cross-linker reaction. Turbidity Measurements. Time evolution of the transmittance of 0.1 wt % alkali solutions of HEC in the presence of various amounts of the cross-linker DVS was measured with a temperature-controlled Helios Gamma (Thermo Spectronic, Cambridge, UK) spectrophotometer at a wavelength of 500 nm.

J. Phys. Chem. B, Vol. 109, No. 25, 2005 12331 The apparatus is equipped with a temperature unit (Peltier plate) that gives a good temperature control over an extended time. The turbidity τ of the samples can be determined from the following relationship: τ ) (-1/L)ln(It/I0) where L is the light path length in the cell (1 cm), It is the transmitted light intensity, and I0 is the incident light intensity. The results from the spectrophotometer will be presented in terms of turbidity. The cloud point (CP) of the considered sample was determined as the temperature at which the first deviation from the baseline occurred. Dynamic Light Scattering. Dynamic light scattering measurements were conducted with the aid of a standard laboratorybuilt light-scattering spectrometer, with vertically polarized incident light of wavelength λ ) 514.5 nm supplied by an argonion laser (Lexel laser, model 95). The beam was focused onto the sample cell through a temperature-controlled chamber (temperature-controlled to within (0.05 °C) filled with refractive-index-matching silicone oil. The sample solutions were filtered through 0.22-µm filters (Millipore) directly into precleaned 10-mm NMR tubes (Wilmad Glass Company) of highest quality. The light-scattering process defines a wave vector q ) (4πn/ λ)sin(θ/2), where θ is the scattering angle and n is the refractive index of the medium. The value of n was determined for all samples at λ ) 514.5 nm by employing an Abbe´ refractometer. In the present study the full homodyne intensity autocorrelation function g2(t) was mostly measured at a scattering angle of 70° with an ALV-5000 multiple-τ digital correlator. If the scattered field obeys Gaussian statistics (as for all the solutions considered in this study), the measured correlation function g2(t) can be related to the theoretically amenable first-order electric field correlation function g1(t) by the Siegert relationship g2(t) ) 1 + B|g1(t)|2, where B is an instrumental parameter. The correlation functions were recorded in the real time “multipleτ” mode of the correlator, in which 256 time channels are logarithmically spaced over an interval ranging from 0.2 µs to almost 1 h. Experiment duration was in the range 5-10 min. Despite the low polymer concentrations employed in this work, the decays of the correlation functions were always found to be bimodal:21,22 initially a single exponential, followed at longer times by a stretched exponential.

g1(t) ) Af exp(-t/τf) + As exp[-(t/τse)β]

(1)

with Af + As ) 1. The parameters Af and As are the amplitudes for the fast and the slow relaxation mode, respectively. Analyses of the time correlation functions of the concentration fluctuations in the domain qRh < 1 (Rh is the hydrodynamic radius) have shown21-23 that the first term (short-time behavior) on the righthand side of eq 1 is related to the mutual diffusion coefficient Dm (τf-1 ) Dmq2). The second term (long-time feature) is expected to be associated with disengagement relaxation of individual chains24,25 or cluster relaxation.26 In this study, the fast mode is diffusive, whereas the inverse slow relaxation time exhibits a somewhat stronger q dependence, probably due to the contribution from internal modes. This behavior is illustrated in Figure 2, where the q dependence of the correlation functions, at conditions indicated, is depicted in the form of a reduced plot. We can see that the correlation function data at different scattering angles collapse onto a single curve at short times, while a close scrutiny of the curves reveal a small deviation at long times. This suggests that also the slow relaxation mode is almost diffusive. Since the principal aim of this study is to follow the time evolution of the correlation function under various cross-linking conditions, we have chosen to perform

12332 J. Phys. Chem. B, Vol. 109, No. 25, 2005

Maleki et al.

Figure 2. Plot of the first-order electric field correlation function as a function of q2t (every third data point is shown) at the scattering angles indicated for a 0.1 wt % HEC solution with 30 µL/g DVS that has been quenched at different times in the course of the cross-linker reaction. The quenching is carried out by adding a few drops of HCl to the solution to lower the pH to 1.5.

Figure 3. Time evolution of the turbidity for 0.1 wt % HEC solutions with different cross-linker concentrations. The inset plot shows the values of the cloud point determined from the incipient rise of the turbidity curve.

most of the measurements at a fixed scattering angle (70°) and to analyze the relaxation results in terms of relaxation times (τf and τs). The variable τse is some effective relaxation time, and β (0 < β e 1) is a measure of the width of the distribution of relaxation times. The distribution of relaxation times for the present systems displays values of β in the range β ) 0.5-0.9. The mean relaxation time is given by

τs )

τse 1 Γ β β

()

(2)

where Γ(β-1) is the gamma function of β-1. In the analysis of the correlation function data, a nonlinear fitting algorithm (a modified Levenberg-Marquardt method) was used to obtain best-fit values of the parameters Af, τf, τse, and β appearing on the right-hand side of eq 1. Rheological Experiments. Steady shear measurements were performed in a Paar-Physica MCR 300 rheometer using a coneand-plate geometry, with a cone angle of 1° and a diameter of 75 mm. This geometry was employed in all rheological experiments. To prevent evaporation of the solvent, the free surface of the sample was always covered with a thin layer of low-viscosity silicone oil (the viscoelastic response of the sample is virtually not affected by this layer). In the steady shear measurements, a fixed shear rate was set and the time evolution of the viscosity was monitored during the cross-linker reaction. The sample was homogenized for a short time and was then rapidly transferred to the plate; the measurements were started immediately. The measuring unit is equipped with a temperature unit (Peltier plate) that gives a very good temperature control over an extended time. Results and Discussion Turbidity and Dynamic Light Scattering. It is evident from Figure 3 that the solutions become turbid in the course of the cross-linking reaction at all cross-linking densities, except at the lowest concentration. The results indicate that the formation of large aggregates starts at earlier times when the cross-linker concentration increases and the value of the cloud point time decreases as the level of cross-linker addition increases (see inset). These results demonstrate that, at quiescent conditions, the interpolymer cross-linking behavior constitutes a dominant

Figure 4. Plot of the first-order electric field correlation function versus time (every third data point is shown) at a scattering angle of 70° for 0.1 wt % HEC solutions with various cross-linker concentrations at different stages during the cross-linking reaction.

feature in the aggregation process. In view of these findings, the growth of the aggregates has not been monitored over very long times in the light scattering experiments, because we try to avoid floc sizes for which multiple scattering effects13 may be significant. This effect is usually accompanied with a decrease of the value of B in the Siegert relation. Figure 4 shows the time evolution of the correlation function for alkali HEC solutions (pH ) 11.8) of a polymer concentration of 0.1 wt % in the presence of different amounts of the crosslinker. The general trend is that the relaxation function is shifted toward longer times as the cross-linking process proceeds, and this effect is more pronounced with increasing cross-linker concentration. The slower decay of the relaxation function at longer reaction times announces the growth of large clusters. The time dependences of the reduced intensity (the scattered intensity divided by the intensity of the incoming light) and the fast (τf) and the slow (τs) relaxation times are displayed in Figure 5 for an alkali 0.1 wt % HEC concentration in the presence of various amounts of cross-linker. At the lowest DVS concentration, only a slight increase of the reduced intensity at very long times can be traced (Figure 5a), whereas for the other cross-linking densities abrupt upturns can be observed, suggesting that large association complexes are formed. The rise of the reduced intensity occurs at earlier times when the cross-

Shear and Association during Cross-Linking of HEC

Figure 5. Time evolution of the reduced intensity, fast relaxation time, and slow relaxation time during the cross-linker process for a 0.1 wt % HEC solution in the presence of different amounts of cross-linker.

linker concentration increases. Similar trends are displayed for the relaxation times. The fast relaxation time probably reflects the motion of individual molecules and small associations of chains, whereas τs monitors the growth of large clusters. The growth of the aggregates starts at earlier times for high crosslinker densities, and the rise of the relaxation times is stronger as the DVS concentration increases. These findings clearly demonstrate that at sufficiently high cross-linker concentrations, large intermolecular complexes are evolved during the reaction in dilute alkali HEC solutions at quiescent conditions, and the kinetic features depend on the cross-linker concentration. To study the growth of the clusters under stationary conditions, 0.1 wt % solutions of HEC in the presence of a fixed amount of cross-linker (30 µL/g DVS) were quenched by a rapid acidification of the reaction mixture at different stages during the association process, and the correlation functions were recorded (see Figure 6a). To scrutinize the fitting procedure and to endorse the functional form of eq 1 that is used to portray the correlation functions, residual plots at two different conditions for 0.1 wt % solutions of HEC are displayed in Figure 6b. The random distribution and small values of the residuals indicate good agreement between the fitting expression and the correlation function data. The results in Figure 6c show that the fast and the slow relaxation times increase with increasing quenching time, as well as the hydrodynamic radius (see the inset plot), calculated from the diffusion coefficient of the fast mode via the Stokes-Einstein relationship. These results demonstrate that the clusters grow as long as the cross-linker reaction continues. In Figure 7, the q dependence of the reduced intensity for 0.1 wt % solutions of HEC is illustrated at different quenching times during the cross-linker reaction. In the limited q range covered in these measurements, the wave vector dependence of the reduced intensity Ired can be described by a power law Ired ∼ q-1.0, where the value of the scaling exponent may reflect the rodlike behavior of the semirigid polysaccharide chains.27 It is interesting to note that the value of the exponent does not vary as the association complex grows, which suggests that on this dimensional scale the structure is not affected by the

J. Phys. Chem. B, Vol. 109, No. 25, 2005 12333

Figure 6. (a) First-order electric field correlation function versus time for 0.1 wt % HEC solutions without cross-linker and with 30 µL/g DVS at different times of quenching (every third data point is shown). (b) Residuals obtained by fitting the correlation function data with the aid of eq 1 at different stages of quenching. (c) Evolution of the fast and slow relaxation time at different times of quenching. The inset plot shows the effect of quenching time on the hydrodynamic radius, calculated from the fast relaxation time via the Stokes-Einstein relationship.

Figure 7. Wave vector dependence of the reduced intensity for 0.1 wt % HEC solutions without cross-linker and with 30 µL/g DVS at different times of quenching. The inset shows the reduced intensity at a scattering angle of 30° at different times of quenching.

interpolymer cross-linking process. The inset of Figure 7 shows that the reduced intensity rises strongly with the quenching time, announcing significant interchain aggregation and increased molecular weight of the clusters because for a given polymer concentration the reduced intensity is proportional to Mw. To examine whether the aggregation process continues after the cross-linking reaction was arrested by quenching, correlation functions were recorded over a long time (Figure 8). It was suspected that the quenched species might possess an enhanced stickiness that could lead to further aggregation. However, the collapse of the correlation function data onto a single curve suggests that the quenched clusters exhibit no tendency to further associate into larger complexes. This means that decrease of

12334 J. Phys. Chem. B, Vol. 109, No. 25, 2005

Figure 8. First-order electric field correlation function versus time for a 0.1 wt % HEC solution with 30 µL/g DVS (every second data point is shown). The correlation functions have been recorded over a long time after the sample was quenched. This demonstrates that no further aggregation takes place after quenching.

Figure 9. Time dependence of the shear viscosity for a 0.1 wt % solution of HEC in the presence of 30 µL/g DVS that is exposed to the shear rates indicated. The inset shows a magnification of the data around the transition zone.

pH is an efficient method to cut off the interchain aggregation process. Shear Viscosity Measurements. Figure 9 shows the time evolution of the viscosity for a reaction mixture of 0.1 wt % HEC and DVS of a concentration of 30 µL/g under the influence of different shear rates. At the highest shear rate, the rotation and stretching of the chains induced by strong mechanical shear stresses prevent the intra- and interpolymer cross-linking reactions from occurring, but the slight increase of the viscosity at very long times may be a harbinger of incipient interchain aggregation. At lower shear rates, the general behavior is characterized by a minimum followed by a rise of the viscosity (a magnification of the transition area is depicted in the inset) at longer times. Since the viscosity is constant for more than an hour before the onset of the minimum, it is unlikely that it is caused by shear thinning. This therefore suggests contraction of the species, and this behavior is a signature of intramolecular cross-linking, whereas the following rise in the viscosity announces interchain aggregation. Intrachain compaction of the molecules was never detected from the DLS measurements above, probably because of the dominance of interpolymer cross-linking at quiescent conditions. At moderate shear rates, the rotation of the chains may initially be more favorable to intrachain contacts than interchain reactions because the large

Maleki et al.

Figure 10. Time evolution of the shear viscosity in the regime where multichain aggregates are formed, at the shear rates indicated for a 0.1 wt % HEC solution with 30 µL/g DVS. The inset shows the shear rate dependence of the power law exponent R, describing the time dependence of the viscosity (η ∼ tR) in the region of formation of intermolecular aggregates.

number of hydroxyl groups on a single chain and its rotational motion may facilitate intramolecular contacts and cross-linking. At longer times, the repeated collisions between the contracted species may lead to the formation of multichain aggregates. A scrutiny of the magnitude of the viscosity increase for the lower two shear rates shows that the breakup of the interconnected aggregates depends on the shear rate. It is obvious from Figure 9 that a lower shear rate promotes the buildup of larger clusters before rupture sets in. The scatter of the data points at the peaks of the viscosity curves indicates the competition between the buildup and breakup of interaggregate chains when the complexes are sufficiently large. This size-limitation step begins as soon as aggregates reach a size large enough to be broken by the viscous forces exerted by the surrounding medium. The enhancement of the viscosity takes place at earlier times at low shear rate, and the steepness of the rise becomes weaker as the shear rate increases. This result is expected because a lower shear rate is advantageous for interpolymer association. A close inspection of the viscosity (η) enhancement at various shear rates discloses that the time dependence of the viscosity increase can be portrayed by a power law η ∼ tR (see Figure 10). The inset plot shows that the increase of the viscosity becomes weaker as the shear rate increases. It can be argued that augmented shear rates should inhibit the growth of large association complexes since mechanical disturbances will obstruct the cross-linker reaction between functional groups on the polymer chains. In Figure 11, the time dependence of the viscosity at a constant shear rate (20 s-1) for a 0.1 wt % HEC solution in the presence of different cross-linker concentrations is depicted. The profile of the viscosity curves at the higher two DVS concentrations is similar to that observed above, that is, a minimum followed by an enhancement of the viscosity. The upturn of the viscosity takes place at an earlier time for the sample with the highest level of DVS addition because the probability of forming intermolecular cross-links is higher. A close inspection of the viscosity curves representing the two lower cross-linker densities reveals a decreasing tendency at long times, which may reflect an incipient intramolecular crosslinking effect. In this case, the cross-linker concentrations are too low to provoke interchain aggregation over the time interval considered.Figure 12 shows the time evolution of the viscosity at a fixed shear rate (20 s-1) and a constant amount of DVS

Shear and Association during Cross-Linking of HEC

Figure 11. Time dependence of the shear viscosity for 0.1 wt % solutions of HEC in the presence of the cross-linker densities indicated and at a shear rate of 20 s-1. The inset shows a magnification of the data around the transition zone.

J. Phys. Chem. B, Vol. 109, No. 25, 2005 12335

Figure 13. Time evolution of the shear viscosity during cross-linker reaction for mixtures of HEC and DVS at the cross-linker and polymer concentrations indicated and at a fixed shear rate of 20 s-1. These compositions give a constant value of the HEC/DVS ratio. The inset shows a magnification of the data around the transition zone.

concentrations, only a slight decrease of the viscosity can be traced over time but no intermolecular association is visible. The reason is probably that the polymer concentration is too low to build up interconnected aggregates. At the higher polymer concentrations, compacted intramolecular entities are formed and we observe the subsequent creation and breakup of interaggregate chains. Again, the magnitude of the viscosity increases, and a short induction time before the growth of the aggregates is promoted by a high polymer concentration. Conclusions

Figure 12. Time evolution of the shear viscosity during cross-linker reaction for dilute solutions of HEC of the concentrations indicated in the presence of a cross-linker concentration of 30 µL/g and at a shear rate of 20 s-1. The inset shows a magnification of the data around the transition zone.

(30 µL/g) in solutions of different HEC concentration. The general picture that emerges is that the polymer concentration plays an important role in the shear-induced development of intramolecular and intermolecular structures. At the lowest HEC concentration, only a slight increase of the viscosity is visible at long times. This trend may be a message of incipient aggregation after several collisions of the entities. At the two higher polymer concentrations, the general appearances of the viscosity curves are similar as the features discussed above, with minimum and viscosity enhancement. The increase of the viscosity occurs at an earlier time for the higher polymer concentration because the probability that interchain cross-links are developed is increased. We note from the results in Figure 12 that the rheometer is capable of detecting viscosity differences between these low polymer concentrations. The average size of the interpolymer complex before it breaks up is larger for the highest polymer concentration. The conjecture is that a high polymer concentration promotes the creation of strong multichain aggregates. Finally, in Figure 13, the time dependence of the viscosity at a constant shear rate (20 s-1) and at the same HEC/DVS composition is illustrated. By keeping the value of the HEC/ DVS ratio fixed, the state of the cross-linker reaction should be the same for all samples. At the lower two polymer

In this work, we have provided some novel information about the influence of steady shear flows on intramolecular and intermolecular associations in dilute aqueous solutions of hydroxyethylcellulose in the presence of a cross-linker agent. Dynamic light scattering results at quiescent conditions reveal no intrachain contraction but only multichain aggregation. The weak perturbation caused by the Brownian dynamics will favor intermolecular association, and possible intramolecular association is overshadowed. The growth of the aggregates starts at earlier times when the cross-linker concentration increases. After quenching to a lower pH, the growth of the species is arrested. We have demonstrated that shear flow can give rise to contracted intrapolymer cross-linked structures and the subsequent buildup of multichain aggregates. The conjecture is that, at moderate shear rates, the large number of hydroxide groups on the individual chain will come close to each other through rotation of the chains and form intramolecular cross-links. As time goes by, the moieties will collide with each other many times, and gradually large aggregates will be built up via interpolymer cross-linking. The commencement and magnitude of these features depend on factors such as shear rate, polymer concentration, and cross-linker density. The rate of the viscosity increase in the process of forming interaggregate chains decreases when the shear rate increases because of shear stresses. The capacity to build up large aggregates is reduced when strong shear forces operate in the solution. These findings have shown that by tuning shear rate, cross-linker density, and polymer concentration, it is possible to observe effects of both intrapolymer- and interpolymer cross-linking. Acknowledgment. B.N. gratefully acknowledges support from the Norwegian Research Council through a NANOMAT

12336 J. Phys. Chem. B, Vol. 109, No. 25, 2005 project (158550/431). We thank Prof. K. D. Knudsen for valuable comments. References and Notes (1) Nystro¨m, B.; Kjøniksen, A.-L.; Iversen, C. AdV. Colloid Interface Sci. 1999, 79, 81. (2) Aamer, K. A.; Sardinha, H.; Bhatia, S. R.; Tew, G. N. Biomaterials 2004, 25, 1087. (3) Kjøniksen, A.-L.; Nystro¨m, B.; Lindman, B. Macromolecules 1998, 31, 1852. (4) Gilsenan, P. M.; Richardson, R. K.; Morris, E. R. Carbohydr. Polym. 2000, 41, 339. (5) Barbier, V.; Herve´, M.; Sudor, J.; Brulet, A. Hourdet, D.; Viovy, J.-L. Macromolecules 2004, 37, 5682. (6) Kjøniksen, A.-L; Nystro¨m, B. Macromolecules 1996, 29, 5215. (7) Sannino, A.; Madaghiele, M.; Conversano, F.; Mele, G.; Maffezzoli, A.; Netti, P. A.; Ambrosio, L.; Nicolais, L. Biomacromolecules 2004, 5, 92. (8) Boyko, V.; Richter, S. Macromol. Chem. Phys. 2004, 205, 724. (9) Omari, A.; Chauveteau, G.; Tabary, R. Colloids Surf., A: Physicochem. Eng. Aspects 2003, 225, 37. (10) Colloid-Polymer Interactions: From Fundamentals to Practice; Farinato, R. S., Dubin, P. L., Eds.; John Wiley & Sons: New York, 1999. (11) Sonntag, R. C.; Russel, W. B. J. Colloid Interface Sci. 1986, 113, 399.

Maleki et al. (12) Wessel, R.; Ball, R. C. Phys. ReV. A 1992, 46, R 3008. (13) Torres, F. E.; Russel, W. B.; Schowalter, W. R. J. Colloid Interface Sci. 1991, 142, 554. (14) Potanin, A. A. J. Colloid Interface Sci. 1991, 145, 140. (15) Le Berre, F.; Chauveteau, G.; Pefferkorn, E. J. Colloid Interface Sci. 1998, 199, 13. (16) Serra, T.; Casamitjana, X. J. Colloid Interface Sci. 1998, 206, 505. (17) Ulanski, P.; Bothe, E.; Rosiak, J. M.; Sonntag, C. Macromol. Chem. Phys. 1994, 195, 1443. (18) Wang, B.; Mukataka, S.; Kodama, M.; Kokufuta, E. Langmuir 1997, 13, 6108. (19) Schmitz, K. S.; Wang, B.; Kokufuta, E. Macromolecules 2001 34, 8370. (20) Maleki, A. Intensity light scattering measurements on dilute HEC solutions. Unpublished data. (21) Kjøniksen, A.-L.; Nystro¨m, B.; Lindman, B. Langmuir 1998, 14, 5039. (22) Lauten, R. A.; Nystro¨m, B. Colloids Surf., A: Physicochem. Eng. Aspects 2003, 219, 45. (23) Nystro¨m, B.; Thuresson, K.; Lindman, B. Langmuir 1995, 11, 1994. (24) Wang, C. H.; Zhang, X. Q. Macromolecules 1993, 26, 707. (25) Douglas, J. F.; Hubbard, J. B. Macromolecules 1991, 24, 3163. (26) Ngai, K. L. AdV. Colloid Interface Sci. 1996, 64, 1. (27) Buhler, E.; Boue´, F. Macromolecules 2004, 37, 1600.