Multiregion Shear Thinning for Subsequent Static Self-Thickening in

Article pubs.acs.org/JPCB

Multiregion Shear Thinning for Subsequent Static Self-Thickening in Chitosan-graf t-polyacrylamide Aqueous Solution Lei Jin,† Yeqiang Tan,† Yonggang Shangguan,*,† Yu Lin,† Bo Xu,‡ Qiang Wu,§ and Qiang Zheng*,† †

MOE Key Laboratory of Macromolecular Synthesis and Functionalization, Department of Polymer Science and Engineering, Zhejiang University, Hangzhou 310027, China ‡ School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332, United States § College of Engineering, Zhejiang Agricultural and Forestry University, Hangzhou 311300, China S Supporting Information *

ABSTRACT: A special shear thinning phenomenon followed by static self-thickening in chitosan-graf t-polyacrylamide (GPAM) aqueous solutions was investigated. This multiregion shear thinning can be defined as the first stage of the recently reported shear induced self-thickening (SIT) in our previous work. The three thinning regions (labeled as N1, N2, and N3) are considered very important, and they can reflex the complex variations of intermolecular interactions among and inside the aggregates in solution with increasing shear rate. To verify this multiregion shear thinning, a critical concentration of GPAM for this three-region shear thinning was first investigated. Shear recovery tests with the maximal shear rates located in the N1−N3 were carried out to ascertain the crucial role of shear thinning in SIT. The mechanisms of these three shear thinning regions were proposed based on the dependence of shear rheological behavior on various conditions in each region, including GPAM concentration, grafting ratio, temperature, added hydrogen bonding breaker, and salt. The above results confirm that N1 is due to the breakage of the interactions among hydrogen bonding aggregates, while N2 and N3 are attributed to the progressive destruction of the aggregates. As the first stage of SIT, shear thinning can markedly break the original aggregate and expose additional hydrogen bonding stickers to reform more aggregates with bigger size, resulting in the final higher viscosity.

■

INTRODUCTION Rheopexy and shear-thickening behavior have been found in various associative polymer solutions possessing hydrophobic association,1−4 hydrogen bond,5,6 host−guest complexation,7 and metal−ligand bond8,9 since Eliassaf et al. reported the shear-thickening phenomenon of poly(methacry1ic acid) solution due to intermolecular association.10 Usually, these shear-thickening behaviors are transient and reversible and could be ascribed to two possible mechanisms including shearinduced increase in the number of elastically active chains and shear-induced nonlinear high tension along stretched polymer chains beyond the Gaussian range.9 Recently, a special shear induced self-thickening (SIT) behavior was observed in chitosan-graf t-polyacrylamide (GPAM) aqueous solution.11 When this GPAM solution is subjected to a high-rate shear for several minutes, it presents shear thinning. Once the shear ceases, its viscosity can recover rapidly and surpass the original viscosity finally. Differing from rheopexy and shear-thickening behavior reported previously, SIT presents three distinct features. (i) The first is different thickening process. This thickening behavior consists of an initial shear-thinning and a subsequent thickening region after removing strong shear, while the thickening of conventional shear-thickening systems © 2013 American Chemical Society

only happens under shear. (ii) The second is maintainable the thickening effect. The resulting thickening could be held until undergoing strong shear once again, while for the conventional shear-thickening most of the increased viscosity may spontaneously recover. (iii) Controllable thickening extent. The viscosity increment of SIT is related to the shear conditions and could repeat well. Thus, this SIT would be very effective to be used as rheology modifiers,12,13 flocculants for wastewater treatment,14,15 and oil-displacing agents for enhanced oil recovery (EOR),16 etc. Furthermore, GPAM can overcome the two main drawbacks of polyacrylamide (PAM) in applications, i.e. nondegradability and low viscosity under strong shear.17 The semiflexible chemical structure of GPAM in AcOH aqueous solution is depicted in Scheme 1. In acidic medium, primary amine groups in CS could be easily fully protonated18 and the groups in PAM could be partially protonated19,20 (as shown in eqs 1 and 2), making GPAM a special associative polymer with cationic charges. In our work, each CS chain (Mw Received: September 2, 2013 Revised: October 31, 2013 Published: November 15, 2013 15111

dx.doi.org/10.1021/jp408782e | J. Phys. Chem. B 2013, 117, 15111−15121

The Journal of Physical Chemistry B

Article

Scheme 1. Illustration of Semiflexible Chemical Structure of GPAMa

0.615 is the deacetylation degree of CS, and G is the ratio of grafted units of the polysaccharide. −NH3⊕ denotes protonated −NH2 in acidic medium. a

= 1 × 106 Da; mean Mw of repeating unit is about 205 according to deacetylation degree) possesses about 3000 active amine groups that can be grafted by PAM side chains and the mean length of PAM side chains only ranging from 15 to 40 depends on grafting ratio shown in Table 1. Furthermore, many

original solution and it should be responsible for SIT. Figure 1C presents a comparison of steady flow curves of original and thickened GPAM, and an obvious increase in ηa at low shear rates can be observed after SIT. In Figure 1D, SIT is depicted as a rapid response process to shear and the viscosity increment seems to approach saturation after several shear treatments, and then a good repeatability appears, showing a rapid switchable thickening. In conclusion, SIT can be considered as a hidden self-thickening because GPAM only shows shear thinning with increasing shear rate during the steady shear process, without any shear thickening phenomenon. SIT cannot be found easily if we did not use the proper shear cycle test. In fact, a beforehand shear thinning process is very important and necessary for SIT. Only following a strong shear thinning process which indicates the severe destruction of the original aggregates, the re-formation of the aggregative structure could occur after ceasing shear. Furthermore, the shear thinning of GPAM shows a special multiregion process that may play a key role in SIT. For traditional polymer solution, the mechanism of shear thinning has been widely studied, and it is usually attributed to the faster breaking rate of intermolecular structure than the rebuilding rate. Shear thinning is a useful method to investigate the variation of interactions in associative polymer system, like the interfacial interactions among chitosan nanoparticles.21 However, few studies are focused on the special associative polymer systems exhibiting unusual multiple and distinct shear-thinning except the polymer liquid crystal (PLC) system.22 In previous studies of associative polymers, Jenkins23 observed shear thinning with multiple regions at high shear stress. Tam et.al.24 found that HEUR solution shows special three shear thinning regions after a mild shear thickening, and they successfully examined the relaxations in each shear thinning regions of associative polymer using superposition-of-oscillation experiments on steady shear flows.25 Shear thinning with multiple regions were also found in other associative polymer systems like hydrophobic modified alginate with added cyclodextrin compounds.26 In our work, GPAM solution shows special multiregion shear thinning and all regions are more distinct than these in the above studies.

Table 1. Structural Parameters Calculated by the Results of Elemental Analysis (EA) and Gel Permeation Chromatography (GPC) for GPAM Samples sample GPAM1 GPAM2 grafting ratio G% a (%) CS component (%) Mw b (kDa) mean number of AM units in side chains

316 24.04 4160 15

410 19.61 5100 19

GPAM3 GPAM4 563 15.08 6630 26

633 13.64 7330 30

GPAM5 856 10.46 9560 40

a

Calculated by EA measurement (in Supporting Information). Determined on the basis of the Mw of CS (∼1000 kDa) calculated by grafting ratio. b

hydrogen bonding or hydrophobic stickers exist along the long chains and there may be enough less-charged segments so that some visible aggregates occur. The dominating drive forces of these aggregates in GPAM solutions were proven as hydrogen bonding effects, which have been discussed in detail in the previous work.11 chitosan−NH 2 + H3O+ ↔ chitosan−NH3+ + H 2O

(1)

PAM−NH 2 + H3O+ ↔ PAM − NH3+ + H 2O

(2)

Figure 1 comprehensively presents the SIT phenomenon in GPAM solution. Figure 1A gives the variations of shear rate with time during a whole SIT process. Once strong shear ceases, the apparent viscosity (ηa) of GPAM would recover soon, then it increases sequentially, and the final ηa is much higher than the original one, as shown in Figure 1B. In addition, as indicated by TEM graphs in Figure 1B, the scale of GPAM aggregate in sheared solution is obviously larger than that in 15112



Article

Figure 1. (A) Design of the shear rate in SIT test. (B) SIT of 1 wt % GPAM1 solution monitored at 0.01 s−1 after being sheared at 1000 s−1 for 2 min and TEM observations of 0.4 wt % GPAM1 in 0.2% (v/v) AcOH solution before and after SIT for 30 min. For clear image, 1 wt % solution was diluted to 0.4 wt % before observation (0.4 wt % solution shows similar SIT as 1 wt % solution). (C) Steady flow curves of 1 wt % GPAM1 solution before and after SIT for 30 min. (D) Controllable thickening for apparent viscosity at 0.01 s−1 with tuned shear rate for 1 wt % GPAM1. ηa′/ηa means the thickening ratio, and ηa′/ηa > 1 is a sign that the observed viscosity surpasses the origin.

Sample Preparation. Graft polymerization of PAM onto CS was carried out using CAN as the initiator under nitrogen (N2) atmosphere at 45 °C following the method published17 except for some small modifications. CS was first dissolved in 1 wt % aqueous AcOH solution. After magnetic stirring and bubbling N2 for 30 min, CAN was added into CS solution. The solution was then kept for 5 min as pretreatment for the CS by the initiator, followed by the dropwise addition of acrylamide monomer aqueous solution into the reactive solution for 10 min. The pretreatment of CS by the initiator could efficiently suppress the formation of PAM homopolymer. After 4 h of reaction under N2 atmosphere at 45 °C, copolymerization was stopped, and the samples were precipitated in massive acetone. The solid product was filtered, washed, and extracted using acetone as solvent in Soxhlet extraction for 48 h to remove impurities in the sample. The resulting product, GPAM, was dried at room temperature until constant weight was obtained. Measurements. GPC measurements were carried out on Waters 2690 (Waters, USA) with acetic acid aqueous solution as eluent. CS standard samples were used for calibration. Elemental analysis (EA) tests were done using Flash EA 1112 (Thermo Finnigan, USA), and each sample was measured at least three times to ensure accuracy. Morphology observations were done using a JEM-2100EX (JEOL, JPN) transmission electron microscope (TEM). The GPAM aqueous solutions were cast onto copper grids coated with carbon films followed by evaporation. Rheological experiments were performed on a stresscontrolled rotational rheometer AR-G2 (TA Instruments, USA). A 40 mm cone−plate geometry with 2° cone angle

Undoubtedly, each shear thinning region should be an intrinsic reflection to the transformation of intermolecular structures. Therefore, it is necessary to investigate the shear thinning behavior of GPAM aqueous solution in order to illuminate the mechanism of whole SIT and clarify the relationships between the rheological properties and aggregates. In this article, we focused on the multiregion shear thinning and try to put forward the proper mechanisms for each region through a systemic rheological investigation, which may help us to comprehend the whole SIT better. On the basis of these, the mechanisms of each shear thinning region are proposed as an approach to catch the physical images of SIT for GPAM solution systems.

■

EXPERIMENTAL SECTION

Materials. Chitosan (CS) with weight-average molecular weight (Mw) of 1 × 106 Da estimated by gel permeation chromatography (GPC) in 0.2 wt % acetic acid and acrylamide were obtained from Aladdin Co. Ltd., China. The deacetylation degree of CS is 0.615, determined by titration and conductimetry. Ceric ammonium nitrate (CAN, Sinopharm Co. Ltd., China) was used as initiator, and acetone (Sinopharm) was used as precipitator and extractant. Acetic acid (AcOH, Sinopharm), sodium chloride (NaCl, Sinopharm), and ammonium acetate (AcONH4, Sinopharm) were used to adjust the conditions of solution samples. All materials above were used without further purification. All experiments were carried out with distilled water (conductivity is less than 5 μS/ cm). 15113



Article

and 50 μm gap size were chosen for steady shear tests. Defined amount of sample solution was directly poured very slowly onto Peltier region in order to avoid shear thinning effect and small air bubbles caused by using pipettes. A simple solvent trap near the margin of Peltier region was used to prevent the evaporation of solvent. All tests except the special marked ones were taken at 20 °C, and all solution samples were equilibrated for at least 5 min before testing. In all rheological tests, original samples were dissolved by 0.2% (v/v) AcOH solution. Steady flow tests were performed in the shear rate range from 10−2 to 103 or 2.5 × 103 s−1 with 5 min as the maximum point time to attain equilibrium at each shear rate point.

hydrogen bonding and hydrophobic interactions.32 The aggregates have been proven to be nanogels with a denser core and highly swollen shell.33,34 In dilute solution, such core− shell aggregates could be observed in TEM with a treatment by negative stain with uranyl acetate.35 Besides, the result of bimodal distribution of Rh in DLS reveals that single macromolecule (unimer) and aggregates can coexist in dilute solution.35,36 We can conjecture that only unimers exist in the extreme dilute solution in which aggregates cannot form, while the single aggregates can adjoin and then make contact together with increasing polymer concentration. Therefore, additional interactions among the neighboring aggregates could occur. According to previous results, aggregates exist in GPAM solutions even at concentrations as low as 0.005 wt %.11 To clarify the particulars of aggregates and how the aggregates affect viscosities of GPAM, the influence of concentration is necessary to be investigated. Figure 3 shows the dependence of specific viscosity (ηsp, ηsp = η0/ηsolvent − 1) on concentration of GPAM in 0.2% (v/v) AcOH aqueous solutions. As shown in Figure 3A), there are two critical concentrations (C01, C02) that divide the plots into three regions. In the lowest concentration region (C < C01), the solutions almost are Newtonian fluid and ηsp is less than 1, suggesting that it is dilute solution.12 In the second region (C01 < C < C02), ηsp obviously increases with the concentration and the curves show little shear thinning, which is due to the formation of aggregates. Direct evidence can be found from bimodal Rh distribution in 0.005 wt % GPAM solution by DLS measurements.11 In the third region (C > C02), ηsp increases more significantly with concentration. Besides, the flow curves start to present N1 when C > C02 as shown in Figure 3B, which is similar to the results that the first shear thinning region (region I) of PLC system appears suddenly as polymer concentration reaches one critical value.21,22 The fast increased ηsp may be attributed to the additional interfacial interactions among the aggregates with the increasing amount of aggregates. It should be noted that the η0 of GPAM sample above C02 is difficult to measure correctly because of the fact that remarkable shear thinning starts since low shear rate, which is beyond the measurable range. Therefore, we use mean ηa in N2 plateau (much lower than η0) instead of η0 in Figure 3A. In fact, the real slope of plots above C02 should be much higher. However, it should not affect the existence and exact critical value of C02. Besides, Figure 3C−F give the comparisons of flow curves before and after a shear treatment shown in Figure 1A. As presupposed, ηa of sheared samples is larger than that of origin for the GPAM solutions with concentrations above C02 (0.3 and 0.75 wt %), i.e., it shows SIT, while there is no SIT for the diluter samples (0.001 and 0.1 wt %). Therefore, only the samples with amounts of aggregates can show N1 and SIT. In summary, we first give a conjecture that the interfacial interactions among aggregates will lead to the occurrence of N1, which should be crucial for SIT. According to the structure of GPAM, the possible interactions may include positive charge repulsion, weak entanglements of surficial chains, and bridges along long chains involved in more than one aggregate. However, the detailed information on the interactions among the aggregates is still not fully understood. Shear Thinning as a Key Role for SIT. To obtain an indepth understanding and explore the role of multiregion shear thinning on SIT, we would like to obtain more information about the shear thinning stage. Therefore, a series of shearrecovery experiments were carried out by setting various

■

RESULTS AND DISCUSSIONS Three-Region Shear Thinning. Figure 2 gives the steady state flow curves for GPAM, CS, and PAM aqueous solutions.

Figure 2. Comparison of steady state flow curves for GPAM, CS, and PAM 0.2% (v/v) AcOH aqueous solutions (CGPAM = CCS = CPAM1 = 1 wt %, CPAM2 = 2 wt %). The inset indicates three regions marked as N1, N2, and N3.

Compared with CS, PAM, and other linear polyelectrolyte solutions, all GPAMs display obvious three-region shear thinning in the flow curves. For clarity, the three regions are labeled as N1, N2, and N3, respectively, as shown in the inset in Figure 2. In accordance with relative reported results,27−31 both CS and PAM display common shear thinning. CS presents a long Newtonian plateau, and no obvious shear thinning occurs when the shear rate is below about 10 s−1 due to the rapid relaxation of semiflexible chains. Relatively, PAM displays severe shear thinning since about 0.1 s−1 due to the disentanglement of linear flexible chains. As shown in Figure 2, GPAMs display relative flexural curves with much larger extent of shear thinning than CS in the whole shear rate range, indicating a more complex structure in GPAM. On the other hand, ηa of GPAM at high shear rates is obviously higher than that of PAM. That is, GPAMs exhibits the higher shear stability at high shear rate which is favorable to water treatment or enhanced oil recovery (EOR) fields. To demonstrate the higher ηa of GPAM under strong shear, the flow curves of an industrial PAM product with ultrahigh molecular weight (∼20 000 kDa, PAM1) and a PAM sample with comparative Mw but higher concentration (2 wt %, ∼8000 kDa, PAM2) are also given in Figure 2. Both of the two PAM samples display preconceived remarkable shear thinning, and their ηa values are lower than that of GPAM with increasing shear rate since about 50 s−1. Critical Concentration for Aggregates, Three-Region Shear Thinning, and SIT. In general, CS and graft-modified CS tend to form aggregates in aqueous solutions because of 15114



Article

Figure 3. Influence of concentration on viscosities of GPAM1 samples in 0.2% (v/v) AcOH solutions. (A) Dependence of GPAM concentrations on ηsp. (B) Steady state flow curves of GPAM1 with various concentrations. Flow curves of GPAM1 with various concentrations before and after SIT: (C) 0.001 wt %; (D) 0.1 wt %; (E) 0.3 wt %; (F) 0.75 wt %.

aggregate and expose additional stickers to rebuild more new aggregates. In brief, a relative strong shear with the shear rate above N2 is required to achieve SIT and the shear thinning process could be established as the first stage of SIT. Furthermore, the obvious transition from N1 to N2 in the curves may suggest a transform of breakage from the interactions among the aggregates to them inside the aggregates. Influence Factors of Three-Region Shear Thinning. Since three-region shear thinning should be related to the GPAM aggregates, some factors that influence the states of aggregate would impact the shear thinning. Relative factors are listed and discussed as follows. 1. Concentrations and Grafting Ratios. To clarify the mechanisms of each shear thinning region, the influences of concentrations and grafting ratios on steady flow curves were first investigated by the power law function.

maximal shear rates in these three regions to reveal the shear rate requirement in the first step of SIT. As shown in Figure 4A, the shear rate first increases from 0.01 to 0.1 s−1 (in N1) and then decreases to the starting point in the same way. This suggests that the variations of the structure are almost completely reversible upon the mild shear deformation. In Figure 4B, the maximal shear rate was set to 1 s−1 and it comes to the lower critical shear rate of N2, indicating that almost all interactions among aggregates are destructed and would not recover in a short time. Therefore, a remarked hysteresis is observed in the down-ramp curve. In Figure 4C, the maximal shear rate was set to 50 s−1 and it comes to the middle of N2. Surprisingly, the ηa measured in the down-ramp curve is slightly higher than that in the up-ramp curve, which suggests that there are more aggregates with larger sizes formed after the damaging of original structure by strong shear, as in the results of TEM in our last work.11 In Figure 4D, the maximal shear rate is up to 2500 s−1 in order to destroy the aggregates more dramatically. As a result, a remarkable ηa increase in the down-ramp curve is observed because sufficiently strong shear force can break the

η = Kγ ṅ − 1 15115

(3) dx.doi.org/10.1021/jp408782e | J. Phys. Chem. B 2013, 117, 15111−15121


Article

Figure 4. Shear and recovery of 1 wt % GPAM1 in 0.2% (v/v) AcOH solution by different upper limit shear rate: (A) 0.1 s−1; (B) 1 s−1; (C) 50 s−1; (D) 2500 s−1.

Here K is a constant and n is rate index. In the shear thinning region, a larger n means a milder shear thinning. For some polymeric solution system, the concentration is considered as the major factor affecting n, i.e., the higher the concentration is, the smaller n tends to be. In the concentrated system, there are many interaction sites like physical entanglements that are easy to break, resulting in a remarkable decrease in ηa with increasing shear rate.37 On the other hand, dilute system would perform as a Newtonian fluid when undergoing shear because no intermolecular contact exists. Concentrations above C02 and grafting ratios (G%) of GPAM were chosen as the variations to investigate the rate index labeled as n1−n3 for three shear thinning regions. As indicated by the results in Figure 5, n1 and n3 show the same trends while n2 displays an opposite one. For concentrations, n1

in Figure 5A and n3 in Figure 5C decrease with increasing concentration. As in the above conjecture, we could consider the dominative cause in N1 as the interfacial interactions among these aggregates. Therefore, with the increase of concentrations, the sites of interactive aggregates increase, resulting in more obvious shear thinning (smaller n1). No matter what types of interactions dominate the linkages among the aggregates, they may make a structure as many micelle spheres which are absorbed together, like the reported flocculated gel particles.38 In flocculated gel particles, enough amounts of particular micelles exist and they could make contact, forming flocculated gel through interfacial interaction among particles. In flocculated gel systems, η0 increases exponentially with the depth of interfacial interactions and leads to an increase in the relaxation time required for particles to rearrange their positions.39 Differing from n1 and n3, Figure 5B shows independence of n2 on concentration, indicating that the main contribution to ηa in N2 is uncommon. For shear thinning system with different concentrations, only in the Newtonian platform does the rate index keep constant against concentrations, i.e., n = 1. In Newtonian platform of polymeric system, there is a dynamic equilibrium of destruction and re-formation under shear. Similar to Newtonian platform, N2 has almost constant n2 which is closer to 1 than n2 or n3, suggesting that there may be a similar dynamic equilibrium in N2. On the basis of the relationship of viscosity and concentration in Figure 3, the dynamic equilibrium in N2 may refer to the breakage inside an individual aggregate (following destruction of interfacial interactions among aggregates) and recovery of a broken aggregate. As for N3, it is attributed to the faster breakage rate of aggregates, surpassing the re-forming rate with increasing shear rate. Similar to n1, n3 would be smaller with increasing concentration in which more interactive sites exist to be broken by shear force.

Figure 5. (A−C) Dependence of rate indexes n1 (A), n2 (B), and n3(C) on concentration for GPAM1 in 0.2% (v/v) AcOH solutions. (D−F) Dependence of n1 (D), n2 (E), and n3 (F) on grafting ratio for 1 wt % GPAM in 0.2% (v/v) AcOH solutions. 15116



Article

Part D, E, and F of Figure 5 display the dependenc of n1, n2, and n3 on grafting ratio, respectively. Contrary to concentration, n1 and n3 almost keep constant with G%, while n2 decreases with increasing G%. According to the structure features of GPAM, higher G% means lower content of CS and longer PAM side chains in solution with fixed GPAM mass concentration. In general, we may deduce that higher G% will lead to weaker interactions inside the aggregates because of less stickers located in CS and more space between two GPAM backbones due to longer side chains. Therefore, n2 may be related to the strength inside aggregates and it supports the hypothesis of dynamic equilibrium between aggregates breakage and reconstruction of aggregates in N2, which will be further discussed as follows. 2. Added H-Bond Breaker. To explore possible relationship between hydrogen bonding effect and N2, Figure 6A presents a

increases) as shown in Figure 6A. That is, shear thinning becomes not obvious when adding AcONH4 because of the decrease of intermolecular interactions amounts. Compared with the obvious decrease of ηa in N2 and N3, ηa in N1 almost keeps constant after adding AcONH4. There may be two reasons: (1) the breaker can only break the aggregates and lessen the size of them, but the amount of aggregates may increase to ηa in N1; (2) the addition of NH4+ can boost additional hydrophobic effects. Besides, as a result of the weakening of intermolecular interactions, the effect of SIT obviously decreases for the sample added AcONH4 as shown in Figure 6B. In summary, the hydrogen bonding effect dominates the intermolecular interactions inside the aggregates, i.e., the ηa in N2 and N3. 3. Ultrasonic Treatment. Ultrasonic treatment is one of the most widely applied methods for dispersion and chain scission,42 and it is expected to break GPAM aggregates here. A mild ultrasonic treatment for 10 min was carried out to determine the role of aggregates. The flow curve of ultrasonic treated sample is given in Figure 7. Different from the original

Figure 7. Comparison of steady state flow curves for GPAM1 before and after ultrasonic treatment (300 kHz, 10 min).

sample, ultrasonic treated sample shows a remarkable ηa decrease in the whole shear rate range, suggesting a serious breakage of aggregate or chain scission. It should be pointed out that the decrease of ηa under high shear rate is not obvious compared with that of the samples with added AcONH4 as shown in Figure 6A. As we know, the strength of chemical bonds is remarkably stronger than physical intermolecular interactions. Therefore, ultrasonic treatment should break aggregates completely before it is able to break the chemical bonds. The ηa difference under high shear rate in Figure 6A and Figure 7 suggests that obvious chain scission may not occur and the breakage of aggregates is the main reason for ηa decrease in Figure 7. Therefore, Figure 7 gives the flow curves after breaking hydrogen bonding aggregate without the additional hydrophobic effect compared with the results in Figure 6A. In other words, the decrease of ηa in N1 reflects the decrease in either aggregate size or aggregates amount after ultrasonic treatment. Furthermore, the breakage of aggregates induced by ultrasonic treatment should be more significant than strong shear so that it needs much longer time to recover. 4. Added Salts and Heating. Because of the molecular structure, GPAM in solution could form two types of aggregates associated with hydrogen bonding and hydrophobic effects. In general, added salt could destroy the water molecular protections and improve the polarity of solvent,43 while heating could enhance the hydrophobic effect while weakening the

Figure 6. Influence of added hydrogen bond breaker AcONH4 on steady state shear (A) and SIT (B) for 1 wt % GPAM1 solution.

comparison of flow curves of the original sample and the samples containing added hydrogen bonding breaker, ammonium acetate (AcONH4).40,41 In Figure 6A, the most significant change induced by the addition of AcONH4 is that ηa in N2 decreases obviously and the rate index n2 also decreases. The remarkable decrease in ηa indicates that considerable hydrogen bonding aggregates still exist in N2 for the original solution. It is reasonable evidence for the above conjecture of dynamic equilibrium between hydrogen bonding aggregate destruction and reformation. Furthermore, as indicated by the DLS results in previous work,11 Rh of the aggregative unit decreases with the increasing concentration of AcONH4 while Rh of unimers keeps constant, indicating the breakage of hydrogen bonds inside GPAM aggregates. Therefore, the decreasing ηa in N2 and N3 may be linked to the decreasing intermolecular interactions inside the aggregates and among the GPAM chains after aggregates are broken, respectively. In addition, the boundaries between N2 and N3 almost diminish (n2 decreases while n3 15117



Article

hydrogen bonding effect.44 Figure 8 gives the flow curves of the samples after added salt and being heated. In Figure 8A, NaCl

N3 in the measurement range. In fact, the stretched N1 indicates some stronger interactions among aggregates whose breakage needs stronger shear force. Taking hydrophobic effects into consideration, increasing ηa in N1 may be the result of increasing amounts of aggregates dominated by hydrophobic force. On the other hand, the samples with added salts or heating appear to lower ηa under high-rate shear than the original ones. The reason may be that the strength inside hydrophobic aggregates is weaker than the strength inside hydrogen bonding aggregates, in according with the above conclusion that hydrogen bonding interaction dominates ηa in N2 and N3. Mechanisms of Three-Region Shear Thinning. According to above rheological behavior of GPAM solution, the mechanisms of three-region shear thinning should be related to the aggregates. Similar to CS and its derivatives, GPAM should form core−shell like aggregates based on the various hydrophilicities of different parts of macromolecules in solution when reaching a critical concentration (i.e., C01 in Figure 3). The hydrophobic part will huddle together forming the core while the hydrophilic part will be pulled around the core and form a shell between water and the core. Finally, the core−shell like aggregates will be micelle-like in water. According to the structure feature of GPAM, two types of driving forces may dominate the micelles. As shown in Scheme 1, there are many kinds of groups related to forming hydrogen bonds, such as −OH, −NH2, CO, etc. In fact, the hydrogen bond groups distribute along both CS and PAM chains. Therefore, these associative GPAM chains are inclined to aggregate to an anomalous spherical clew through abundant intermolecular hydrogen bonds, which may be described by an entangled sticky chains model as shown in Scheme 2.45 Besides the spherical clew hydrogen bonding core, some flexible chains with more positive charge and less hydrogen bond groups may form a shell distributed outside the core. On the other hand, the possible hydrophobic groups only exist in the CS chain, i.e., acetyl in the undeacetylated CS units.46 Therefore, hydrophobic aggregate occurs among these CS segments, which

Figure 8. Influence of added salt (A) and temperature (B) on steady state shear for 1 wt % GPAM1 in 0.2% (v/v) AcOH solutions.

was added to GPAM solution and ηa increases obviously in low shear rate region but decreases a lot under high shear rate. The similar change is found in the curve of the heated sample in Figure 8B. The only reason for the thickening in low shear rate region should be the enhancement of the hydrophobic effect among GPAM chains. Besides increasing viscosity under low shear rate, the change of curve shapes is also observed, i.e., the extended N1 results in a shorter N2 and near disappearance of

Scheme 2. Mechanism of Shear Thinning with Three Regions and the Transformations from Hydrogen Bonding Aggregates to Hydrophobic Aggregatesa

a

Green coils mean GPAM molecules. Red short lines mean PAM side chains. Green dash circles mean cores of hydrogen bonding aggregates forming by an anomalous spherical clew through abundant intermolecular hydrogen bonds. Blue coils mean tight cores of hydrophobic aggregates forming by hydrophobic moiety in CS. Blue circles mean shell border of the aggregates. The crossover of blue circles (shells) means the interfacial interactions among the aggregates. 15118



Article

analysis, we consider that there is a dynamic equilibrium between hydrogen bonding destruction and re-formation in N2. Some evidence would as follows: (i) Shear thinning rate in N2 approaches 0, indicating an equilibrium or semiequilibrium state involving certain dynamic breakage and re-form process in GPAM solution. (ii) In Figure 5, shear thinning in N2 turns more obvious (n2 decreases) with increasing grafting ratio, which should be related to the decreasing amount of hydrogen bonding groups. (iii) ηa in N2 decreases remarkably with added hydrogen bond breaker, as shown in Figure 6, suggesting that the dynamic interaction is related to the hydrogen bonding effect. (iv) N2 platform would be less obvious when heating or adding salt into GPAM solutions (as shown in Figure 8) in which hydrophobic effect would surpass the hydrogen bonding effect, supporting that the hydrogen bonding effect dominates the dynamic process. As for N3, it may be clearer that the remaining interactions of chains following the broken aggregates dominate the ηa. The ηa under high shear rate should be related to the hydrogen bonding aggregates, which has been proven as the main contribution for ηa under high shear rate in Figure 6 and Figure 7. Finally, the schematic diagrams are given in Scheme 2 to conclude the mechanisms discussed above. Besides, the other factors like methods of preparing solution samples, pH, grafting ratio, etc. can also affect the observations of SIT significantly, which will be studied and discussed in the future.

could form some tight core rather than anomalous spherical clews. The chains without hydrophobic moiety may be dragged by hydrophobic aggregate and adjoin around the tight aggregates, forming the similar core−shell structure with a thick shell consisting of the chains except the hydrophobic segments, as reported hydrophobic modified CS.35 At room temperature, hydrogen bonding effects are the dominating driving force, as proposed in our previous work.11 The discussions about the mechanisms of three-region shear thinning will be given based on the aggregates in GPAM solution. 1. N1. The three concentration regions shown in Figure 3 can be correlated to the GPAM aggregates. In GPAM solution with C > C02, based on the existence of aggregate particles, the dramatic increase of ηa in N1 should be attributed to the interface interactions among of aggregates. This interface interaction may be similar to reported flocculated gel system.38 There are many properties in flocculated gel system, including the following. (i) The plateau viscosity is only accessed at very low shear rate, as low as 10−5−10−3 s−1.47 (ii) There is a sudden decrease in viscosity under a higher shear stress compared to critical shear stress which is often referred to as yield.48 (iii) The dependence of the viscoelasticities on particle size, particle concentration, and particle−particle interactions is of obvious importance for the processing of colloidal gels,49 etc. All of the above properties could also be found in the GPAM system,11 especially the yield stress (σy) existing before and after SIT as shown in Figure 9, indicating that N1 of GPAM is a

■

CONCLUSIONS A multiregion shear thinning is observed in GPAM solution, and it can be defined as the first stage of SIT. This flow curve displays more different shapes than CS and PAM, indicating a more complex shear thinning mechanism. By detection of the relationship between ηsp and concentration, two critical concentrations (C01, C02) were observed, and they should refer to the onset of forming aggregate and the start of forming interactions among aggregates, respectively. It needs to emphasize that three-region shear thinning and SIT can only be observed in the samples above the second limited concentrations (C02). Shear recovery tests confirm that SIT of GPAM can only be observed after strong shear by which aggregates can be damaged. That also confirms that the shear thinning process plays a crucial role on SIT, i.e., the first stage of SIT. Mechanisms of three-region shear thinning for GPAM solutions are proposed according to a series of rheological experiments. Mechanism of N1 is determined as the breakage of interfacial interactions among aggregates. Mechanism of N2 is ascribed to the damage of hydrogen bonding interaction inside individual aggregate, and that of N3 is linked to the remanent hydrogen bonding interactions among GPAM chains after the aggregates are broken. It also should be noted that the aggregates of GPAM dominated by hydrogen bonding and hydrophobic effects present two different types of shear thinning curves in the measurement range.

Figure 9. Existence of yield stress for 1 wt % GPAM1 in 0.2%(v/v) AcOH solutions before and after SIT.

consequence of the flocculated gel particles structures. Furthermore, the enhanced σy after SIT indicates that there are stronger interfacial interactions among the aggregates with larger amount or size, which is in accordance with the TEM results shown in Figure 1B. Besides, on the basis of these flocculated gel particle structures, GPAM shows the concentration dependence on n1 in Figure 5A. 2. N2 and N3. When C > C02, two kinds of intermolecular interactions may coexist including the interfacial interactions among aggregates and the interactions inside aggregate. In the steady shear process (shear rate sweep), the shear rate generally increases so that the shear preferentially destroys some weaker interactions. Therefore, N1 could be recognized as the breaking process of weak interactions, while N2 and N3 are reasonably thought to be related to the breakage inside the aggregates. Because of the existence of additional shear thinning in N3, we can infer that the aggregates could be further broken or deformed under high shear rate. In the previous conjecture and

■

ASSOCIATED CONTENT

S Supporting Information *

Results of elemental analysis. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*Y.S.: e-mail, [email protected]. *Q.Z.: e-mail, [email protected]. 15119



Article

Notes

(18) Rinaudo, M.; Pavlov, G.; Desbrieres, J. Influence of Acetic Acid Concentration on the Solubilization of Chitosan. Polymer 1999, 40, 7029−7032. (19) Chen, W.; Tomalia, D. A.; Thomas, J. L. Unusual pHDependent Polarity Changes in PAMAM Dendrimers: Evidence for pH-Responsive Conformational Changes. Macromolecules 2000, 33, 9169−9172. (20) Jin, G.-w.; Koo, H.; Nam, K.; Kim, H.; Lee, S.; Park, J.-S.; Lee, Y. PAMAM Dendrimer with a 1,2-Diaminoethane Surface Facilitates Endosomal Escape for Enhanced pDNA Delivery. Polymer 2011, 52, 339−346. (21) Roux, R.; Ladavière, C.; Montembault, A.; David, L.; Delair, T. Shear Thinning 3D-Colloidal Assemblies of Chitosan and Poly(lactic acid) Nanoparticles. J. Phys. Chem. B 2013, 117, 7455−7464. (22) Walker, L. M.; Wagner, N. J.; Larson, R. G.; Mirau, P. A.; Moldenaers, P. The Rheology of Highly Concentrated PBLG Solutions. J. Rheol. 1995, 39, 925−952. (23) Jenkins, R. D. The Fundamental Thickening Mechanism of Associative Polymers in Latex Systems: A Rheological Study. Ph.D. Thesis, Lehigh University, 1990. (24) Tam, K.; Jenkins, R.; Winnik, M.; Bassett, D. A Structural Model of Hydrophobically Modified Urethane-Ethoxylate (HEUR) Associative Polymers in Shear Flows. Macromolecules 1998, 31, 4149−4159. (25) Tirtaatmadja, V.; Tam, K.; Jenkins, R. Superposition of Oscillations on Steady Shear Flow as a Technique for Investigating the Structure of Associative Polymers. Macromolecules 1997, 30, 1426−1433. (26) Burckbuchler, V.; Kjøniksen, A.-L.; Galant, C.; Lund, R.; Amiel, C.; Knudsen, K. D.; Nyströ m, B. Rheological and Structural Characterization of the Interactions between Cyclodextrin Compounds and Hydrophobically Modified Alginate. Biomacromolecules 2006, 7, 1871−1878. (27) Kienzle-Sterzer, C.; Rodriguez-Sanchez, D.; Rha, C. Flow Behavior of a Cationic Biopolymer: Chitosan. Polym. Bull. 1985, 13, 1−6. (28) Nyström, B.; Kjøniksen, A.-L.; Iversen, C. Characterization of Association Phenomena in Aqueous Systems of Chitosan of Different Hydrophobicity. Adv. Colloid Interfac. 1999, 79, 81−103. (29) Desbrieres, J. Viscosity of Semiflexible Chitosan Solutions: Influence of Concentration, Temperature, and Role of Intermolecular Interactions. Biomacromolecules 2002, 3, 342−349. (30) Li, Y.; Kwak, J. C. Rheology of Hydrophobically Modified Polyacrylamide-co-Poly(acrylic acid) on Addition of Surfactant and Variation of Solution pH. Langmuir 2004, 20, 4859−4866. (31) Martinez, A.; Chornet, E.; Rodrigue, D. Steady-Shear Rheology of Concentrationed Chitosan Solutions. J. Texture Stud. 2004, 35, 53− 74. (32) Philippova, O. E.; Korchagina, E. V.; Volkov, E. V.; Smirnov, V. A.; Khokhlov, A. R.; Rinaudo, M. Aggregation of Some Water-Soluble Derivatives of Chitin in Aqueous Solutions: Role of the Degree of Acetylation and Effect of Hydrogen Bond Breaker. Carbohyd. Polym. 2012, 87, 687−694. (33) Bao, H.; Li, L.; Gan, L. H.; Ping, Y.; Li, J.; Ravi, P. Thermo- and pH-Responsive Association Behavior of Dual Hydrophilic Graft Chitosan Terpolymer Synthesized via ATRP and Click Chemistry. Macromolecules 2010, 43, 5679−5687. (34) Korchagina, E. V.; Philippova, O. E. Effects of Hydrophobic Substituents and Salt on Core−Shell Aggregates of Hydrophobically Modified Chitosan: Light Scattering Study. Langmuir 2012, 28, 7880− 7888. (35) Korchagina, E. V.; Philippova, O. E. Multichain Aggregates in Dilute Solutions of Associating Polyelectrolyte Keeping a Constant Size at the Increase in the Chain Length of Individual Macromolecules. Biomacromolecules 2010, 11, 3457−3466. (36) Hao, J.; Yuan, G.; He, W.; Cheng, H.; Han, C. C.; Wu, C. Interchain Hydrogen-Bonding-Induced Association of Poly(acrylic acid)-graft-poly(ethylene oxide) in Water. Macromolecules 2010, 43, 2002−2008.

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Nature Science Foundation of China (Grant No. 51173165), the Key Program of the National Natural Science Foundation of China (Grant No. 50633030), the Fundamental Research Funds for the Central Universities (Grant No. 2013QNA4048), and Zhejiang Provincial Innovative Research Team (Grant No. 2009R50004).

■

REFERENCES

(1) Cadix, A.; Chassenieux, C.; Lafuma, F.; Lequeux, F. Control of the Reversible Shear-Induced Gelation of Amphiphilic Polymers through Their Chemical Structure. Macromolecules 2005, 38, 527−536. (2) Hu, Y.; Wang, S.; Jamieson, A. Rheological and Rheooptical Studies of Shear-Thickening Polyacrylamide Solutions. Macromolecules 1995, 28, 1847−1853. (3) Lele, A.; Shedge, A.; Badiger, M.; Wadgaonkar, P.; Chassenieux, C. Abrupt Shear Thickening of Aqueous Solutions of Hydrophobically Modified Poly(N,N′-dimethylacrylamide-co-acrylic acid). Macromolecules 2010, 43, 10055−10063. (4) Takeda, M.; Kusano, T.; Matsunaga, T.; Endo, H.; Shibayama, M.; Shikata, T. Rheo-SANS Studies on Shear-Thickening/Thinning in Aqueous Rodlike Micellar Solutions. Langmuir 2011, 27, 1731−1738. (5) Yang, Y.; Xie, X.; Yang, Z.; Wang, X.; Cui, W.; Yang, J.; Mai, Y.W. Controlled Synthesis and Novel Solution Rheology of Hyperbranched Poly(urea-urethane)-Functionalized Multiwalled Carbon Nanotubes. Macromolecules 2007, 40, 5858−5867. (6) Tho, I.; Kjøniksen, A.-L.; Nyström, B.; Roots, J. Characterization of Association and Gelation of Pectin in Methanol−Water Mixtures. Biomacromolecules 2003, 4, 1623−1629. (7) Zhu, L.; Shangguan, Y.; Sun, Y.; Ji, J.; Zheng, Q. Rheological Properties of Redox-Responsive, Associative Ferrocene-Modified Branched Poly(ethylene imine) and Its Modulation by B-Cyclodextrin and Hydrogen Peroxide. Soft Matter 2010, 6, 5541−5546. (8) Yount, W. C.; Loveless, D. M.; Craig, S. L. Small-Molecule Dynamics and Mechanisms Underlying the Macroscopic Mechanical Properties of Coordinatively Cross-Linked Polymer Networks. J. Am. Chem. Soc. 2005, 127, 14488−14496. (9) Xu, D.; Hawk, J. L.; Loveless, D. M.; Jeon, S. L.; Craig, S. L. Mechanism of Shear Thickening in Reversibly Cross-Linked Supramolecular Polymer Networks. Macromolecules 2010, 43, 3556. (10) Eliassaf, J.; Silberberg, A.; Katchalsky, A. Negative Thixotropy of Aqueous Solutions of Polymethacrylic Acid. Nature 1955, 176, 1119. (11) Jin, L.; Shangguan, Y.; Ye, T.; Yang, H.; An, Q.; Zheng, Q. Shear Induced Self-Thickening in Chitosan-Grafted Polyacrylamide Aqueous Solution. Soft Matter 2013, 9, 1835−1843. (12) Boris, D. C.; Colby, R. H. Rheology of Sulfonated Polystyrene Solutions. Macromolecules 1998, 31, 5746−5755. (13) Jiang, H.; Taranekar, P.; Reynolds, J. R.; Schanze, K. S. Conjugated Polyelectrolytes: Synthesis, Photophysics, and Applications. Angew.Chem., Int.Ed. 2009, 48, 4300−4316. (14) Petzold, G.; Buchhammer, H.-M.; Lunkwitz, K. The Use of Oppositely Charged Polyelectrolytes as Flocculants and Retention Aids. Colloids Surf. A 1996, 119, 87−92. (15) Tilton, R.; Murphy, J.; Dixon, J. The Flocculation of Algae with Synthetic Polymeric Flocculants. Water Res. 1972, 6, 155−164. (16) Wever, D.; Picchioni, F.; Broekhuis, A. Polymers for Enhanced Oil Recovery: A Paradigm for Structure−Property Relationship in Aqueous Solution. Prog. Polym. Sci. 2011, 36, 1558−1628. (17) Lu, Y.; Shang, Y.; Huang, X.; Chen, A.; Yang, Z.; Jiang, Y.; Cai, J.; Gu, W.; Qian, X.; Yang, H. Preparation of Strong Cationic Chitosan-Graft-Polyacrylamide Flocculants and Their Flocculating Properties. Ind. Eng. Chem. Res. 2011, 50, 7141−7149. 15120



Article

(37) Kjøniksen, A.-L.; Beheshti, N.; Kotlar, H. K.; Zhu, K.; Nyström, B. Modified Polysaccharides for Use in Enhanced Oil Recovery Applications. Eur. Polym. J. 2008, 44, 959−967. (38) Larson, R. G. The Structure and Rheology of Complex Fluids; Oxford University Press: Oxford, U.K., 1999; Vol. 1. (39) Buscall, R.; McGowan, J. I.; Morton-Jones, A. J. The Rheology of Concentrated Dispersions of Weakly Attracting Colloidal Particles with and without Wall Slip. J. Rheol. 1993, 37, 621. (40) Lamarque, G.; Lucas, J.-M.; Viton, C.; Domard, A. Physicochemical Behavior of Homogeneous Series of Acetylated Chitosans in Aqueous Solution: Role of Various Structural Parameters. Biomacromolecules 2005, 6, 131−142. (41) Schatz, C.; Viton, C.; Delair, T.; Pichot, C.; Domard, A. Typical Physicochemical Behaviors of Chitosan in Aqueous Solution. Biomacromolecules 2003, 4, 641−648. (42) Tayal, A.; Khan, S. A. Degradation of a Water-Soluble Polymer: Molecular Weight Changes and Chain Scission Characteristics. Macromolecules 2000, 33, 9488−9493. (43) Recillas, M.; Silva, L. L.; Peniche, C.; Goycoolea, F. M.; Rinaudo, M.; Argüelles-Monal, W. M. Thermoresponsive Behavior of Chitosan-g-N-isopropylacrylamide Copolymer Solutions. Biomacromolecules 2009, 10, 1633−1641. (44) Woodward, P. J.; Hermida Merino, D.; Greenland, B. W.; Hamley, I. W.; Light, Z.; Slark, A. T.; Hayes, W. Hydrogen Bonded Supramolecular Elastomers: Correlating Hydrogen Bonding Strength with Morphology and Rheology. Macromolecules 2010, 43, 2512− 2517. (45) Leibler, L.; Rubinstein, M.; Colby, R. H. Dynamics of Reversible Networks. Macromolecules 1991, 24, 4701−4707. (46) Philippova, O. E.; Volkov, E. V.; Sitnikova, N. L.; Khokhlov, A. R.; Desbrieres, J.; Rinaudo, M. Two Types of Hydrophobic Aggregates in Aqueous Solutions of Chitosan and Its Hydrophobic Derivative. Biomacromolecules 2001, 2, 483−490. (47) Patel, P.; Russel, W. The Rheology of Polystyrene Latices Phase Separated by Dextran. J. Rheol. 1987, 31, 599. (48) Barnes, H.; Walters, K. The Yield Stress Myth? Rheol. Acta 1985, 24, 323−326. (49) Buscall, R.; McGowan, I. J.; Mumme-Young, C. A. Rheology of Weakly Interacting Colloidal Particles at High Concentration. Faraday Discuss. 1990, 90, 115−127.

15121


Multiregion Shear Thinning for Subsequent Static Self-Thickening in

Recommend Documents