- Email: [email protected]
Flocculation by cationic amphiphilic polyelectrolyte: Relating efficiency with the association of polyelectrolyte in the initial solution
Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147
Contents lists available at SciVerse ScienceDirect
Colloids and Surfaces A: Physicochemical and Engineering Aspects journal homepage: www.elsevier.com/locate/colsurfa
Flocculation by cationic amphiphilic polyelectrolyte: Relating efficiency with the association of polyelectrolyte in the initial solution Luminita Ghimici ∗ , Marieta Nichifor “Petru Poni” Institute of Macromolecular Chemistry, Aleea Grigore Ghica Voda, 41A, 700487 Iasi, Romania
h i g h l i g h t s
g r a p h i c a l
Amphiphilic polymer based on nontoxic polysaccharide is studied as a new flocculant. Flocculation efficiency was studied as function of the polymer initial solution concentration, ci . Flocculation window width and floc size significantly increased when ci was above cac. Flocculation mechanisms depended on the electrostatic and hydrophobic forces competition. We show for the first time the importance of the polymer association state in the initial solution for flocculation efficacy.
Particle size distribution.
a r t i c l e
a b s t r a c t
i n f o
Article history: Received 20 July 2012 Received in revised form 14 September 2012 Accepted 17 September 2012 Available online 26 September 2012 Keywords: Amphiphilic polysaccharide Flocculation Polymer concentration Turbidity Zeta potential
a b s t r a c t
Flocculation ability of an amphiphilic cationic polysaccharide with N-octyl-N,N-dimethyl-N-(2hydroxypropyl)ammonium chloride groups attached to a dextran backbone, was evaluated in clay dispersions with respect to polycation dose and initial solution concentration and, consequently, its association state. According to turbidimetric results, the concentration of initial solution of polymer influenced, mainly, the width of the flocculation window (the polycation dose range where the minimum of the residual turbidity was obtained) since this flocculation characteristic increased with increasing this parameter. A significant change in flocculation performances was noticed when polymer concentration was above its critical aggregation concentration. Possible reasons for this dependence are discussed. The negative value of the zeta potential in the whole flocculation window and the floc size distribution point to contributions from both patch/charge neutralization and bridging mechanisms for the flocculation process. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Interactions between small particles and polymers represent a subject of great interest in the industries in which flocculation process plays a key role, for instance in papermaking, mining, sludge
∗ Corresponding author. Tel.: +40 232 260333; fax: +40 232 211299. E-mail address: [email protected] (L. Ghimici). 0927-7757/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.colsurfa.2012.09.016
dewatering, domestic/wastewater treatment. Polymeric flocculants are available in uncharged or charged (cationic, anionic) forms and may be synthetic and natural [1,2]. For effective flocculation, polymers need to be adsorbed on particles by electrostatic and/or hydrophobic attractions, hydrogen bonding and ion binding [1]. These interactions lead to various conformations of the adsorbed chains, such as trains, loops, tails and hence, to different flocculation mechanisms: chain bridging, charge neutralization and charge patch [2]. Quite often, more
L. Ghimici, M. Nichifor / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147
than one mechanism may operate simultaneously, depending on the polymer chain conformation (size and shape), both at the solid/liquid interface and in solution as well as on the particle properties (size and charge) [3]. There is a large number of studies where the flocculation mechanism and hence the efficiency of the process has been related to the polymer characteristics and environment [4–19]. We also dealt with this aspect in our previously published papers, where we reported the results of the flocculation behavior of some hydrophilic [20] and amphiphilic [21] cationic polysaccharides based on dextran with N-alkyl-N,Ndimethyl-2-hydroxypropyl ammonium chloride pendent groups. The investigations were performed with respect to the polycation dose, the charge density, the molar mass, the settling time, the length of the alkyl substituent in the pendant groups. Another parameter with a great influence on the flocculation process is the concentration of polymer in its initial solution, i.e., the concentration of polyelectrolyte solution added to the clay suspension. Despite this fact, the relationship between the polymer concentration and the flocculating performance is less studied. Thus, Zheng et al. [22] investigated the flocculating properties of three kind of polyacrylamide-linear, star and hybrid-with different polymer concentration, called parent solution concentration (cp ), in kaolin suspension and found out the flocculation performance of all polymers enhanced at the beginning and then decreased with increasing cp . For each polymer an optimum parent solution concentration (cop ) was identified. The impact of the polymer concentration in the parent solution, varied between 0.25 and 3 g L−1 , was also examined for the above mentioned hydrophilic cationic dextran derivatives [20] and grafted cationic polysaccharides based on pullulan [23] in a clay suspension; concentrations close to the overlap concentration, c* were found as the optimum parent solution concentrations for the samples investigated. It is well known the amphiphilic polymer chains may exhibit different conformation as a function of concentration due to occurrence, besides electrostatic interactions, of intra/inter-molecular associations between hydrophobic moieties. Hence, one may expect an influence of association state of the polymer in the initial solution on its adsorption on the particle surface and further aggregation of particles (flocculation). There are several studies describing flocculation performances of polymers with a certain hydrophobicity [10], but this aspect has never been covered. Therefore, in this paper we investigate the effect of variation of the initial solution concentration, ci , between 1 and 4 g L−1 , corresponding to polymer non-associated and associated state, respectively, on the flocculation activity of an amphiphilic polyelectrolyte based on dextran carrying pendent quaternary ammonium groups, N-octylN,N-dimethyl-N-(2-hydroxypropyl)ammonium chloride randomly distributed along the polymer backbone in a clay suspension. We consider that the relationship between polymer characteristics, some experimental parameters (polycation dose and initial solution concentration) and flocculation efficiency, in terms of turbidity removal, obtained in the present study on a model clay suspension could be further applied to the ceramic industry wastewaters purification processes. The flocculation process was followed by turbidity and zeta potential measurements. The floc size distributions at polycation doses located in the flocculation window were also evaluated by DLS measurements.
2. Materials and methods 2.1. Materials Cationic polysaccharide D40-Oct30 was synthesized by chemical modification of dextran sample with a molar mass
143
Fig. 1. General chemical structure of polycation D40-Oct30.
Mw = 40 kg/mol (Sicomed S.A., Bucharest), as previously described [24]. Briefly, the polysaccharide was dissolved in deionized water and a mixture of epichlorohydrin and N,N-dimethyl-N-octylamine (both from Aldrich) was added, and the resulting solution was stirred for 6 h at 70 ◦ C. The polymer was recovered by precipitation with acetone, and purified by repeated precipitation and sequential dialysis against 0.1 N HCl and water. Dialysis tubing with a cut-off of 12,000 from Sigma was used for this purpose. The dialyzed solution was freeze dried (24 h at −57 ◦ C and 0.04 mbar) and a white powder was obtained. The modification process proceeded in a homogeneous medium and no micellization of the reagents took place. These reaction conditions allowed an even (statistical) distribution of the charged groups along the polysaccharide chains. Chemical structure, shown in Fig. 1, was proved by 1 H-NMR and elemental analysis. The polymer code is D40-Oct30, where D means dextran, 40 is dextran molar mass in kg mol−1 , Oct is the alkyl substituent at the amino group, and 30 = DS ± 2 mol/% (DS is expressed as moles of amino groups per 100 glucopyranosidic units; DS = 100x/(x + y), where x and y are the molar fraction of substituted and unsubstituted glucosidic units, respectively). The content in amino groups (DS) was determined from the nitrogen content (elemental analysis, by means of a CHNS 2400 II PerkinElmer analyzer) and the chloride ion content was measured potentiometrically by titration with 0.02 N AgNO3 aqueous solution, Cli = 1.27 mequiv g−1 . Other characteristics of the sample investigated were 20 ◦ C ,water = 0.42, the charge density parameter calculated according to Manning [25]; [] = 6.89 dl g−1 , the intrinsic viscosity obtained by fitting the Rao equation [26] to the viscosity data (sp /c versus c) of the cationic polyelectrolyte in salt free aqueous solution [27]; cac = 3 g L−1 , the critical aggregation concentration, determined by fluorescence measurements [28]. Clay powder-SSM Blend 14888 (gift sample from Romanceram Co., Romania) was used to prepare model suspensions composed of 47% kaolin, 22% montmorillonite, 31% quartz with a chemical composition of (wt%), SiO2 , 66–72; Al2 O3 , 23–26; TiO2 , max. 1.8; Fe2 O3 , max. 1.5; Na2 O, 0.1; K2 O, 1.9–2.4 and a particle size distribution of <20 m 90–95%; <10 m 82–89%; <6 m 70–80%; <2 m 55–65; <1 m 45–57%. 2.2. Methods The aqueous polycation solutions, stabilized at room temperature for 1 day before use, and clay dispersions were prepared with distilled water. The hydrodynamic size of the polymer associates, Z-average (d, nm), determined by DLS measurements were 269 and 740 in solutions of 1 g L−1 and 4 g L−1 , respectively. The concentration of the model suspension was 1 g L−1 . Five-hundred milliliter of the clay suspension was placed into 600 ml glass beakers, sonicated for 30 min using an ultrasonicator (Bandelin
144
L. Ghimici, M. Nichifor / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147
Sonorex RK51OH, Berlin, Germany) followed by vigorous stirring for 15 min at 1000 rpm to fully disperse the clay powder. The flocculation experiments were conducted at room temperature. Different volumes of polycation solutions were added at 1000 rpm and stirring was continued with the same speed for about 3 min and then decreased to about 200 rpm for 15 min. After settling times of 15 min, the turbidity of the supernatant was measured with a HACH 2100 AN turbidimeter (HACH Company, Dusseldorf, Germany). A blank experiment was performed in the absence of cationic polysaccharide to evaluate “natural” sedimentation of the suspension under the selected experimental conditions (pH, concentration of suspended mater). The residual turbidity (T%) was expressed as percent of the initial turbidity of the clay suspension, at time zero, in the absence of polymer. All experiments were performed in triplicate and the mean turbidity values were calculated. Standard deviation determined for the experiments was ±4%. The residual turbidity for the blank experiment was high, about 90% and 83% after settling times of 15 min and 120 min, respectively (see Ghimici et al. [20]). The zeta potential of clay suspensions in the presence and absence of polycation was measured after 15 min of settling time with a Zetasizer Nano-ZS, ZEN-3500 model (Malvern Instruments, Malvern, England). The electrophoretic mobility of the particles was converted to the zeta potential using the von Smoluchowski equation [6]. Particle dimensions were measured by laser diffraction with a Mastersizer 2000 system (version 5.31) (Malvern Instruments, Malvern, England) consisting of an optical bank with a He–Ne 632 nm/2 mW laser light, a sample dispersion unit (Hydro 2000A) equipped with stirrer, a recirculation pump, ultrasonics and software to record and process results on the computer. Dispersions obtained with dextran derivatives corresponding to the polycation doses located in the flocculation window were allowed to settle for 15 min, and the flocs were collected and rapidly transferred to the dispersion unit. In order to avoid the agglomeration of the flocs, the sample under investigation is gentle stirred by means of the stirrer integrated in the dispersion unity. The dispersion force in this unit is much milder than during the flocculation experiment, therefore the integrity of the flocs is not affected. In the following, the initial solution concentration of polymer, ci , means the concentration of polyelectrolyte solution added to the clay suspension, and polymer dose refers to the polyelectrolyte concentration in its mixture with clay suspension.
3. Results and discussion 3.1. Turbidity measurements A very useful approach to assess the efficiency of flocculation process is measuring supernatant turbidity; the lower the residual turbidity, the better the flocculating polymer. Fig. 2 displays the residual turbidity dependence on the polycation dose for different initial solution concentrations of the cationic polysaccharide D40Oct30, ci : 1–4 g L−1 . Some common features were observed for all the systems, irrespective of the polymer initial solution concentrations, as presented in the following: (i) there was a range of polymer dose where the supernatant clarity was reasonable (flocculation window); (ii) below and above the flocculation window, the residual turbidity values were more or less higher, depending on the polymer dose, than those recorded in the flocculation window. The presence of cationic charges in the chemical structure of the D40-Oct30 sample suggests that the electrostatic attractions between the positively charged polymer and negative sites of the particle surface played a key role in the reduction of the residual turbidity of the supernatant.
Fig. 2. Residual turbidity (T%) dependence on the polycation dose for different initial solution concentration, (ci ): (circle) 1 g L−1 ; (inverted triangle) 3 g L−1 ; (star) 4 g L−1 ; settling time 15 min.
Once the particle charge was sufficiently reduced, the repulsive electrostatic forces between the partially covered particles with polyelectrolyte were reduced allowing the van der Waals attraction to dominate when the surfaces are in close proximity; the result was the clarification of the supernatant. Above flocculation window the polyelectrolyte adsorption continued and overcharging of the clay particles took place. According to some reports, the additional polycation adsorption might be caused by the electrostatic [29,30] as well as the hydrophobic attractive forces [31]. The hydrophobic group (octyl) and quaternary N atom in the chemical structure of the sample investigated (Fig. 1) make plausible the assumption that both types of attractions contributed to the additional polycation adsorption, followed by the restabilization of the suspension due to electrostatic and/or steric chain repulsions. Besides the common features, there are some differences indicating the impact of the initial polyelectrolyte solution concentration on the width of the flocculation window, as illustrated in Fig. 2. This parameter increased when ci increased. Thus, maximum flocculation occurred at the following dose ranges: 7–11 mg L−1 for ci = 1 g L−1 , 7–16 mg L−1 for ci = 3 g L−1 and 7–18 mg L−1 for ci = 4 g L−1 . The different flocculation window could be explained by various conformations of the macromolecular chains in solution as a function of concentration. Thus, it is well known that in aqueous solution amphiphilic polymers tend to form associates when the polymer concentration increases beyond a critical concentration, called critical aggregation concentration. Previous fluorescence studies performed on this polymer revealed the occurrence of intra- and intermolecular hydrophobic associations at a polymer concentration of about 3 g L−1 [28] (see also Section 2). The ci values under investigation are located below and above cac. At lower ci (ci = 1 g L−1 ), the macromolecular chains are in a rather extended conformation because of the intramolecular electrostatic repulsive forces between the charged groups, which overcomes the hydrophobic association between the alkyl chains; the easier accessibility of cationic groups resulted in more rapid separation (see the rapid decrease of the residual turbidity values for polymer doses < 7 mg L−1 ) and redispersion. At ci ≥ 3 g L−1 , the association of alkyl side chains resulted in hydrophobic microdomains with a hydrophobic core surrounded by positive charges, formed either inside a single chain (unimers), or among different chains (multimers). The previous studies revealed the tendency of this polymer to form multimers’ associates, which is confirmed by the larger hydrodynamic radius of the polymer at ci = 4 g L−1 . Based on Yu and Somasundaran’s findings [32], namely, that at the initial moment
L. Ghimici, M. Nichifor / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147
145
Fig. 3. Zeta potential () dependence on the polycation dose for different initial solution concentration, (ci ): (circle) 1 g L−1 ; (star) 4 g L−1 ; settling time 15 min.
of adsorption to the surface, the macromolecular coil will retain its conformation in solution and then attempts to relax toward the surface, we assume that the polymer associated chains are probably adsorbed as such on the particle surface, without significant changes in size and conformation, that means no chain extension by dilution and spreading over the particle surface. This leads to the formation of larger polymer patches, which increase the flocculation efficiency at partial surface coverage and consequently, enlarge the flocculation window. The effect of larger hydrodynamic radius in increasing flocculation window was previously demonstrated by Schwarz et al. [10] for polyelectrolytes with different molar mass. In the present case, the higher Rh is the result of a higher apparent molar mass due to formation of multi-chain associates. The above findings allow us to conclude that, in this particular case, the changing of ci and, consequently, of association state of the polymer, led to changing of the flocculation window width. This result is for the first time observed, as in the case of other systems where hydrophilic polymers were used, only the residual turbidity value at optimum polymer dose was modified, while the optimum polymer dose range was almost the same irrespective of ci [20,22,23]. 3.2. Assessment of the flocculation mechanism It is generally believed that charge neutralization and/or patch mechanisms occur when polymers with medium and high charge content and low molar mass adsorb onto oppositely charged surface while the bridging one operates for nonionic polymers and linear polyelectrolytes with low charge density and high molar masses [2]. The chain characteristics of the sample investigated, namely, the charge content (30%) and the molar mass (40 kg mol−1 ) plead for the former mechanisms. But, in order to elucidate the flocculation mechanisms proposed above, zeta potential as well as particle size measurements have been performed and shown in the following. 3.2.1. Zeta potential measurements Plotted in Fig. 3 is the evolution of the zeta potential as a function of the polymer dose for ci of 1 and 4 g L−1 . Independent of the initial solution concentration, increasing polycation dose led to increase from the initially negative value of the suspension (−25.6 mV) until positive values for higher polymer doses. At lower polymer doses, increases monotonously, but ← just at the end of flocculation window for each experiment (at polymer dose > 11 mg L−1 and 18 mg L−1 for ci = 1 and 4 g L−1 , respectively), a sharp increase in was observed, leading to a rapid charge reversal. In both cases reached the charge neutralization point for a polycation dose which was higher (about 13 mg L−1 for
Fig. 4. Floc size distribution for different initial solution concentration, (ci ) and polycation doses located in the flocculation window: (inverted triangle) 1 g L−1 , polycation dose = 7 mg L−1 ; (star) 4 g L−1 , polycation dose = 11 mg L−1 ; (circle) no polymer. The particle size distribution is presented as volume fraction versus particle diameter.
ci = 1 g L−1 and 19 mg L−1 for ci = 4 g L−1 ) than those corresponding to the end of the flocculation window. The variation of with polymers doses, for the system with different ci , confirmed the turbidimetric results about flocculation window width. The earlier sharp increase in for the lower ci can be explained by the difference in the activity of the supplementary polymer chains adsorbed on the particle surface above 7 mg L−1 . The more extended single chains determine a rapid charge reversal, but large associates are still able to interact with each other by inter-particle hydrophobic aggregation. The rapid decrease of the residual turbidity over the range of polymer doses from 1 to 7 mg L−1 , accompanied by a monotonous change in for both polymer concentrations, until the end of the flocculation window, together with the negative values of in the whole flocculation window can be correlated with the patch mechanism for clay removal, irrespective of the initial solution concentration. An optimum dose around that needed to give a zeta potential close to zero is usually attributed to simple charge neutralization [33]. Nevertheless, the bridging mechanism can also be taken under consideration for this system. 3.2.2. Particle size measurements The particle size distribution, presented as volume fraction versus particle diameter, for D40-Oct30 at polycation doses located in the flocculation window, 7 mg L−1 for ci = 1 g L−1 and 11 mg L−1 for ci = 4 g L−1 as well as of the initial clay suspension is shown in Fig. 4. For each of the initial solution concentration a multimodal distribution was observed: (i) one peak corresponds to particles of small size, which points to the neutralization/patch mechanism for the particle clay flocculation; (ii) one or two peaks for particles of large size, which indicates that a certain portion of the particles was flocculated via bridging mechanism, because the hydrophobic chains could participate to inter-particle associations, which resulted in bigger flocs. The high volume percentage of peak (i) in the case of ci = 1 g L−1 suggests that the mechanisms based on the electrostatic attractions are the dominant ones. Peaks (ii) occurs for both ci = 1 g L−1 and ci = 4 g L−1 , but the much higher volume percentage (about 60%) of these peaks observed when ci = 4 g L−1 supports a preponderance of the bridging mechanism in this case. Perhaps, the polycation associated chains existing in solution with ci = 4 g L−1 and adsorbed at the particle surface, have also some free hydrophobic units (not included in the hydrophobic
146
L. Ghimici, M. Nichifor / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147
Fig. 5. Floc size distribution for D40-Oct30 with ci = 4 g L−1 at different polymer dose: (star) 11 mg L−1 ; (square) 18 mg L−1 ; (inverted triangle) 20 mg L−1 ; (circle) no polymer. The particle size distribution is presented as volume fraction versus particle diameter.
intra- and intermolecular associations), which could participate to inter-particle hydrophobic associations, resulting in bigger flocs. In order to obtain more evidence about the flocculation mechanisms, the flocculated suspensions with polymer doses of 7 mg L−1 , residual turbidity of 2.47% (ci = 1 g L−1 ) and 11 mg L−1 , residual turbidity 1.53% (ci = 4 g L−1 ) were stirred for 60 min at 1000 rpm. After a settling time of 15 min, the supernatant residual turbidity values were 25% and 55% for ci = 1 g L−1 and 4 g L−1 , respectively. According to Yan et al. [34] flocs which are initially formed via charge neutralization and then broken up by shearing can easily re-form upon removing the shearing forces, while polymer-bridged particles stay apart once broken up, since polymer tails and loops bridging across two or more particles are physically disrupted by the shearing forces. This chain breakage leaves residual polymer on both surfaces. Surface rearrangements of these polymeric components can lead to particle restabilization against further aggregation by the presence of an electro-steric polymer layer on the surface. Based on these findings, the more (for ci = 1 g L−1 ) and less (for ci = 4 g L−1 ) stable flocs resulted after the prolonged shearing sustain our assumption that the mechanism based on the electrostatic attractive forces prevails in the former case while bridging flocculation mechanism predominates in the latter case. The particle size distribution for D40-Oct30, ci = 4 g L−1 at the polycation doses located in the flocculation window (11 mg L−1 ) and higher (polymer dose: 18 and 20 mg L−1 ) (Fig. 5) supports both turbidimetric and zeta potential results. The smaller floc sizes recorded for the higher polymer doses than 11 mg L−1 might be explained by the adsorption of an excess positive polymer chains onto the clay particles leading to an increase in positive charges on the floc. These charges, in turn, increase the repulsion among particles, thereby hindering microflocs from growing into larger ones. 4. Conclusions For the first time we demonstrated that the association state of an amphiphilic polymer in its initial solution can significantly influence the flocculation of a clay suspension. The results suggest that polymer associated chains are adsorbed as such on the particle surface, without significant changes in size and conformation. The examination of the clay particle flocculation features when added polymer is in associated form (ci = 4 g L−1 ) led us to several conclusions:
䊉 The occurrence of a wider flocculation window, revealed by turbidimetric measurements, can be attributed to the larger hydrodynamic radius of the polymer associates. This effect is similar to that observed when polymer with higher molecular weight were used [10], but in our case the increase of apparent Mw (and, consequently, of Rh ) is due to intermolecular hydrophobic associations (multimers’ associations). Moreover, the changing of the flocculation window width with changing of ci was for the first time observed, as in the case of other systems where hydrophilic polymers were used, only the residual turbidity value at optimum polymer dose was modified (the optimum polymer dose range was almost the same irrespective of ci [20,22,23]). 䊉 The zeta potential variation with polymers doses, for the systems with different ci , confirmed the turbidimetric results about flocculation window width. The monotonous increase of until the end of the flocculation window, together with its negative values in the whole flocculation window can be correlated with the patch mechanism for clay removal, irrespective of the initial solution concentration. An optimum dose around that needed to give a zeta potential close to zero is usually attributed to simple charge neutralization [33]. 䊉 The significant increase in the volume fraction of flocs with bigger size supports the involvement of the bridging mechanism for particle aggregation and flocculation. The big flocs are disrupted under shear, and this is considered as a supplementary proof for the bridging mechanism [34]. 䊉 The above findings have implications for separation processes where amphiphilic polymers are used; in this particular case, the presence of large associates in the initial polymer solution determines a significant improvement of the flocculation process by enlargement of the flocculation window and floc size. Moreover, a large flocculation window is preferred in the real flocculation processes where significant fluctuations of the medium parameters (pH, temperature, ionic strength) take place. Acknowledgements The authors thank the group of Micro-and nanostructures characterization Laboratory (“P. Poni” Institute of Macromolecular Chemistry, Iasi, Romania) for performing zeta potential and particle dimension measurements. This work was supported by a grant of the Romanian National Authority for Scientific Research, CNCSUEFISCDI, project number PN-II-ID-PCE-2011-3-0622. References [1] J. Gregory, Particle in Water: Properties and Processes, CRC, Press, Boca Raton, Fl, USA, 2006. [2] B. Bolto, J. Gregory, Organic polyelectrolytes in water treatment, Water Res. 41 (2007) 2301–2324. [3] V. Runkana, P. Somansundaran, P.C. Kapur, Mathematical modeling of polymerinduced flocculation by charge neutralization, J. Colloid Interface Sci. 270 (2004) 347–358. [4] J.A. Caskey, R.J. Primus, The effect of anionic polyacrylamide molecular conformation and configuration on flocculation effectiveness, Environ. Prog. 5 (1986) 98–103. [5] S. Sakohara, T. Kimura, K. Nishikawa, Flocculation mechanism of suspended particles using the hydrophlic/hydrophobic transition of a thermosensitive polymer, KONA 20 (2002) 246–250. [6] S.P. Strand, K.M. Vårum, K. Østgaard, Interactions between chitosans and bacterial suspensions: adsorption and flocculation, Colloids Surf. B: Biointerfaces 27 (2003) 71–81. [7] J. Roussy, M. Van Vooren, E. Guibal, Influence of chitosan characteristics on coagulation and flocculation of organic suspension, J. Appl. Polym. Sci. 98 (2005) 2070–2079. [8] J. Labille, F. Thomas, M. Milas, C. Vanhaverbeke, Flocculation of colloidal clay by bacterial polysaccharides: effect of macromolecule charge and structure, J. Colloid Interface Sci. 284 (2005) 149–156. [9] G. Petzold, M. Mende, N. Kochurova, Polymer–surfactant complexes as flocculants, Colloids Surf. A: Physicochem. Eng. Aspects 298 (2007) 139–144.
L. Ghimici, M. Nichifor / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147 [10] S. Schwarz, W. Jaeger, B.-R. Paulke, S. Bratskaya, N. Smolka, J. Bohrisch, Cationic flocculants carrying hydrophobic functionalities: applications for solid/liquid separation, J. Phys. Chem. B 111 (2007) 8649–8654. [11] Y. Wei, F. Cheng, H. Zheng, Synthesis and flocculating properties of cationic starch derivatives, Carbohydr. Polym. 74 (2008) 673–679. [12] L.-J. Wang, J.-P. Wang, S.-J. Zhang, Y.-Z. Chen, S.-J. Yuan, G.-P. Sheng, H.-Q. Yu, A water-soluble cationic flocculant synthesized by dispersion polymerization in aqueous salts solution, Sep. Purif. Technol. 67 (2009) 331–335. [13] R. Gaudreault, N. Di Cesare, D. Weitz, T.G.M. van de Ven, Flocculation kinetics of precipitated calcium carbonate, Colloids Surf. A: Physicochem. Eng. Aspects 340 (2009) 56–65. [14] A. Hebeish, A. Higazi, S. El-Shafei, Synthesis of carboxymethyl cellulose (CMC) and starch-based hybrids and their applications in flocculation and sizing, Carbohydr. Polym. 29 (2010) 60–69. [15] I.C. Ho, I. Norli, A.F.M. Alkarkhi, N. Morad, Characterization of biopolymeric flocculant (pectin) and organic synthetic flocculant (PAM): a comparative study on treatment and optimization in kaolin suspension, Bioresour. Technol. 101 (2010) 1166–1174. [16] M.G. Rasteiro, F.A.P. Garcia, P.J. Ferreira, E. Antunes, D. Hunkeler, C. Wandrey, Flocculation by cationic polyelectrolytes: relating efficiency with polyelectrolyte characteristics, J. Appl. Polym. Sci. 116 (2010) 3603–3612. [17] R. Rojaj-Reyna, S. Schwarz, G. Heinrich, G. Petzold, S. Schutze, J. Bohrisch, Flocculation efficiency of modified water soluble chitosan versus commonly used commercial polyelectrolytes, Carbohydr. Polym. 81 (2010) 317–322. [18] D. Tian, X. Wu, C. Liu, H.-Q. Xie, Synthesis and flocculation behavior of cationic konjac glucomannan containing quaternary ammonium substituents, J. Appl. Polym. Sci. 115 (2010) 2368–2374. [19] S. Piriyaprasarth, P. Sriamornsak, Flocculating and suspending properties of commercial citrus pectin and pectin extracted from pomelo (Citrus maxima) peel, Carbohydr. Polym. 83 (2011) 561–568. [20] L. Ghimici, S. Morariu, M. Nichifor, Separation of clay suspension by ionic dextran derivatives, Sep. Purif. Technol. 68 (2009) 165–171. [21] L. Ghimici, M. Nichifor, Novel biodegradable flocculanting agents based on cationic amphiphilic polysaccharides, Bioresour. Technol. 101 (2010) 8549–8554.
147
[22] B.Q. Zheng, J.Q. Qian, X.K. Li, Z.H. Zhu, Effect of the parent solution concentration on the flocculation performance of PAAm flocculants and the relation between the optimal parent solution concentration and critical concentrations, J. Appl. Polym. Sci. 103 (2007) 1585–1592. [23] L. Ghimici, M. Constantin, G. Fundueanu, Novel biodegradable flocculanting agents based on Pullulan, J. Hazard. Mater. 181 (2010) 351–358. [24] M. Nichifor, M.C. Stanciu, B.C. Simionescu, New cationic hydrophilic and amphiphilic polysaccharides synthesized by one pot procedure, Carbohydr. Polym. 82 (2010) 965–975. [25] G.S. Manning, Limiting laws and counterion condensation in polyelectrolyte solutions. I. Colligative properties, J. Chem. Phys. 51 (1969) 924–933. [26] M.V.S. Rao, Viscosity of dilute to moderately concentrated polymer solutions, Polymer 34 (1993) 592–596. [27] M. Nichifor, M.C. Stanciu, L. Ghimici, B.C. Simionescu, Hydrodynamic properties of some cationic amphiphilic polysaccharides in dilute and semi-dilute aqueous solutions, Carbohydr. Polym. 83 (2011) 1887–1894. [28] M. Nichifor, S. Lopes, M. Bastos, A. Lopes, Self-aggregation of amphiphilic cationic polyelectrolytes based on polysaccharides, J. Phys. Chem. B 108 (2004) 16463–16472. [29] V. Lobaskin, K. Qamhich, Effective macroion charge and stability of highly asymmetric electrolytes at various salt conditions, J. Phys. Chem. B 107 (2003) 8022–8029. [30] G. Gillies, W. Lin, M. Borkovec, Charging and aggregation of positively charged latex particles in the presence of anionic polyelectrolytes, J. Phys. Chem. B 111 (2007) 8626–8633. [31] P.M. Claesson, M.A.G. Dahlgren, L. Eriksson, Forces between polyelectrolytecoated surfaces: relations between surface interaction and floc properties, Colloids Surf. A 93 (1994) 293–303. [32] X. Yu, P. Somasundaran, Role of polymer conformation in interparticle bridging dominated flocculation, J. Colloid Interface Sci. 177 (1996) 283–287. [33] J. Kleimann, C. Gehin-Delval, H. Auweter, M. Borkovec, Super-stoichiometric charge-neutralization in particle– polyelectrolyte systems, Langmuir 21 (2005) 3688–3698. [34] Y.D. Yan, S.M. Glover, G.J. Jameson, S. Biggs, The flocculation efficiency of polydisperse polymer flocculants, Int. J. Miner. Process. 73 (2004) 161–175.
Contents lists available at SciVerse ScienceDirect
Colloids and Surfaces A: Physicochemical and Engineering Aspects journal homepage: www.elsevier.com/locate/colsurfa
Flocculation by cationic amphiphilic polyelectrolyte: Relating efficiency with the association of polyelectrolyte in the initial solution Luminita Ghimici ∗ , Marieta Nichifor “Petru Poni” Institute of Macromolecular Chemistry, Aleea Grigore Ghica Voda, 41A, 700487 Iasi, Romania
h i g h l i g h t s
g r a p h i c a l
Amphiphilic polymer based on nontoxic polysaccharide is studied as a new flocculant. Flocculation efficiency was studied as function of the polymer initial solution concentration, ci . Flocculation window width and floc size significantly increased when ci was above cac. Flocculation mechanisms depended on the electrostatic and hydrophobic forces competition. We show for the first time the importance of the polymer association state in the initial solution for flocculation efficacy.
Particle size distribution.
a r t i c l e
a b s t r a c t
i n f o
Article history: Received 20 July 2012 Received in revised form 14 September 2012 Accepted 17 September 2012 Available online 26 September 2012 Keywords: Amphiphilic polysaccharide Flocculation Polymer concentration Turbidity Zeta potential
a b s t r a c t
Flocculation ability of an amphiphilic cationic polysaccharide with N-octyl-N,N-dimethyl-N-(2hydroxypropyl)ammonium chloride groups attached to a dextran backbone, was evaluated in clay dispersions with respect to polycation dose and initial solution concentration and, consequently, its association state. According to turbidimetric results, the concentration of initial solution of polymer influenced, mainly, the width of the flocculation window (the polycation dose range where the minimum of the residual turbidity was obtained) since this flocculation characteristic increased with increasing this parameter. A significant change in flocculation performances was noticed when polymer concentration was above its critical aggregation concentration. Possible reasons for this dependence are discussed. The negative value of the zeta potential in the whole flocculation window and the floc size distribution point to contributions from both patch/charge neutralization and bridging mechanisms for the flocculation process. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Interactions between small particles and polymers represent a subject of great interest in the industries in which flocculation process plays a key role, for instance in papermaking, mining, sludge
∗ Corresponding author. Tel.: +40 232 260333; fax: +40 232 211299. E-mail address: [email protected] (L. Ghimici). 0927-7757/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.colsurfa.2012.09.016
dewatering, domestic/wastewater treatment. Polymeric flocculants are available in uncharged or charged (cationic, anionic) forms and may be synthetic and natural [1,2]. For effective flocculation, polymers need to be adsorbed on particles by electrostatic and/or hydrophobic attractions, hydrogen bonding and ion binding [1]. These interactions lead to various conformations of the adsorbed chains, such as trains, loops, tails and hence, to different flocculation mechanisms: chain bridging, charge neutralization and charge patch [2]. Quite often, more
L. Ghimici, M. Nichifor / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147
than one mechanism may operate simultaneously, depending on the polymer chain conformation (size and shape), both at the solid/liquid interface and in solution as well as on the particle properties (size and charge) [3]. There is a large number of studies where the flocculation mechanism and hence the efficiency of the process has been related to the polymer characteristics and environment [4–19]. We also dealt with this aspect in our previously published papers, where we reported the results of the flocculation behavior of some hydrophilic [20] and amphiphilic [21] cationic polysaccharides based on dextran with N-alkyl-N,Ndimethyl-2-hydroxypropyl ammonium chloride pendent groups. The investigations were performed with respect to the polycation dose, the charge density, the molar mass, the settling time, the length of the alkyl substituent in the pendant groups. Another parameter with a great influence on the flocculation process is the concentration of polymer in its initial solution, i.e., the concentration of polyelectrolyte solution added to the clay suspension. Despite this fact, the relationship between the polymer concentration and the flocculating performance is less studied. Thus, Zheng et al. [22] investigated the flocculating properties of three kind of polyacrylamide-linear, star and hybrid-with different polymer concentration, called parent solution concentration (cp ), in kaolin suspension and found out the flocculation performance of all polymers enhanced at the beginning and then decreased with increasing cp . For each polymer an optimum parent solution concentration (cop ) was identified. The impact of the polymer concentration in the parent solution, varied between 0.25 and 3 g L−1 , was also examined for the above mentioned hydrophilic cationic dextran derivatives [20] and grafted cationic polysaccharides based on pullulan [23] in a clay suspension; concentrations close to the overlap concentration, c* were found as the optimum parent solution concentrations for the samples investigated. It is well known the amphiphilic polymer chains may exhibit different conformation as a function of concentration due to occurrence, besides electrostatic interactions, of intra/inter-molecular associations between hydrophobic moieties. Hence, one may expect an influence of association state of the polymer in the initial solution on its adsorption on the particle surface and further aggregation of particles (flocculation). There are several studies describing flocculation performances of polymers with a certain hydrophobicity [10], but this aspect has never been covered. Therefore, in this paper we investigate the effect of variation of the initial solution concentration, ci , between 1 and 4 g L−1 , corresponding to polymer non-associated and associated state, respectively, on the flocculation activity of an amphiphilic polyelectrolyte based on dextran carrying pendent quaternary ammonium groups, N-octylN,N-dimethyl-N-(2-hydroxypropyl)ammonium chloride randomly distributed along the polymer backbone in a clay suspension. We consider that the relationship between polymer characteristics, some experimental parameters (polycation dose and initial solution concentration) and flocculation efficiency, in terms of turbidity removal, obtained in the present study on a model clay suspension could be further applied to the ceramic industry wastewaters purification processes. The flocculation process was followed by turbidity and zeta potential measurements. The floc size distributions at polycation doses located in the flocculation window were also evaluated by DLS measurements.
2. Materials and methods 2.1. Materials Cationic polysaccharide D40-Oct30 was synthesized by chemical modification of dextran sample with a molar mass
143
Fig. 1. General chemical structure of polycation D40-Oct30.
Mw = 40 kg/mol (Sicomed S.A., Bucharest), as previously described [24]. Briefly, the polysaccharide was dissolved in deionized water and a mixture of epichlorohydrin and N,N-dimethyl-N-octylamine (both from Aldrich) was added, and the resulting solution was stirred for 6 h at 70 ◦ C. The polymer was recovered by precipitation with acetone, and purified by repeated precipitation and sequential dialysis against 0.1 N HCl and water. Dialysis tubing with a cut-off of 12,000 from Sigma was used for this purpose. The dialyzed solution was freeze dried (24 h at −57 ◦ C and 0.04 mbar) and a white powder was obtained. The modification process proceeded in a homogeneous medium and no micellization of the reagents took place. These reaction conditions allowed an even (statistical) distribution of the charged groups along the polysaccharide chains. Chemical structure, shown in Fig. 1, was proved by 1 H-NMR and elemental analysis. The polymer code is D40-Oct30, where D means dextran, 40 is dextran molar mass in kg mol−1 , Oct is the alkyl substituent at the amino group, and 30 = DS ± 2 mol/% (DS is expressed as moles of amino groups per 100 glucopyranosidic units; DS = 100x/(x + y), where x and y are the molar fraction of substituted and unsubstituted glucosidic units, respectively). The content in amino groups (DS) was determined from the nitrogen content (elemental analysis, by means of a CHNS 2400 II PerkinElmer analyzer) and the chloride ion content was measured potentiometrically by titration with 0.02 N AgNO3 aqueous solution, Cli = 1.27 mequiv g−1 . Other characteristics of the sample investigated were 20 ◦ C ,water = 0.42, the charge density parameter calculated according to Manning [25]; [] = 6.89 dl g−1 , the intrinsic viscosity obtained by fitting the Rao equation [26] to the viscosity data (sp /c versus c) of the cationic polyelectrolyte in salt free aqueous solution [27]; cac = 3 g L−1 , the critical aggregation concentration, determined by fluorescence measurements [28]. Clay powder-SSM Blend 14888 (gift sample from Romanceram Co., Romania) was used to prepare model suspensions composed of 47% kaolin, 22% montmorillonite, 31% quartz with a chemical composition of (wt%), SiO2 , 66–72; Al2 O3 , 23–26; TiO2 , max. 1.8; Fe2 O3 , max. 1.5; Na2 O, 0.1; K2 O, 1.9–2.4 and a particle size distribution of <20 m 90–95%; <10 m 82–89%; <6 m 70–80%; <2 m 55–65; <1 m 45–57%. 2.2. Methods The aqueous polycation solutions, stabilized at room temperature for 1 day before use, and clay dispersions were prepared with distilled water. The hydrodynamic size of the polymer associates, Z-average (d, nm), determined by DLS measurements were 269 and 740 in solutions of 1 g L−1 and 4 g L−1 , respectively. The concentration of the model suspension was 1 g L−1 . Five-hundred milliliter of the clay suspension was placed into 600 ml glass beakers, sonicated for 30 min using an ultrasonicator (Bandelin
144
L. Ghimici, M. Nichifor / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147
Sonorex RK51OH, Berlin, Germany) followed by vigorous stirring for 15 min at 1000 rpm to fully disperse the clay powder. The flocculation experiments were conducted at room temperature. Different volumes of polycation solutions were added at 1000 rpm and stirring was continued with the same speed for about 3 min and then decreased to about 200 rpm for 15 min. After settling times of 15 min, the turbidity of the supernatant was measured with a HACH 2100 AN turbidimeter (HACH Company, Dusseldorf, Germany). A blank experiment was performed in the absence of cationic polysaccharide to evaluate “natural” sedimentation of the suspension under the selected experimental conditions (pH, concentration of suspended mater). The residual turbidity (T%) was expressed as percent of the initial turbidity of the clay suspension, at time zero, in the absence of polymer. All experiments were performed in triplicate and the mean turbidity values were calculated. Standard deviation determined for the experiments was ±4%. The residual turbidity for the blank experiment was high, about 90% and 83% after settling times of 15 min and 120 min, respectively (see Ghimici et al. [20]). The zeta potential of clay suspensions in the presence and absence of polycation was measured after 15 min of settling time with a Zetasizer Nano-ZS, ZEN-3500 model (Malvern Instruments, Malvern, England). The electrophoretic mobility of the particles was converted to the zeta potential using the von Smoluchowski equation [6]. Particle dimensions were measured by laser diffraction with a Mastersizer 2000 system (version 5.31) (Malvern Instruments, Malvern, England) consisting of an optical bank with a He–Ne 632 nm/2 mW laser light, a sample dispersion unit (Hydro 2000A) equipped with stirrer, a recirculation pump, ultrasonics and software to record and process results on the computer. Dispersions obtained with dextran derivatives corresponding to the polycation doses located in the flocculation window were allowed to settle for 15 min, and the flocs were collected and rapidly transferred to the dispersion unit. In order to avoid the agglomeration of the flocs, the sample under investigation is gentle stirred by means of the stirrer integrated in the dispersion unity. The dispersion force in this unit is much milder than during the flocculation experiment, therefore the integrity of the flocs is not affected. In the following, the initial solution concentration of polymer, ci , means the concentration of polyelectrolyte solution added to the clay suspension, and polymer dose refers to the polyelectrolyte concentration in its mixture with clay suspension.
3. Results and discussion 3.1. Turbidity measurements A very useful approach to assess the efficiency of flocculation process is measuring supernatant turbidity; the lower the residual turbidity, the better the flocculating polymer. Fig. 2 displays the residual turbidity dependence on the polycation dose for different initial solution concentrations of the cationic polysaccharide D40Oct30, ci : 1–4 g L−1 . Some common features were observed for all the systems, irrespective of the polymer initial solution concentrations, as presented in the following: (i) there was a range of polymer dose where the supernatant clarity was reasonable (flocculation window); (ii) below and above the flocculation window, the residual turbidity values were more or less higher, depending on the polymer dose, than those recorded in the flocculation window. The presence of cationic charges in the chemical structure of the D40-Oct30 sample suggests that the electrostatic attractions between the positively charged polymer and negative sites of the particle surface played a key role in the reduction of the residual turbidity of the supernatant.
Fig. 2. Residual turbidity (T%) dependence on the polycation dose for different initial solution concentration, (ci ): (circle) 1 g L−1 ; (inverted triangle) 3 g L−1 ; (star) 4 g L−1 ; settling time 15 min.
Once the particle charge was sufficiently reduced, the repulsive electrostatic forces between the partially covered particles with polyelectrolyte were reduced allowing the van der Waals attraction to dominate when the surfaces are in close proximity; the result was the clarification of the supernatant. Above flocculation window the polyelectrolyte adsorption continued and overcharging of the clay particles took place. According to some reports, the additional polycation adsorption might be caused by the electrostatic [29,30] as well as the hydrophobic attractive forces [31]. The hydrophobic group (octyl) and quaternary N atom in the chemical structure of the sample investigated (Fig. 1) make plausible the assumption that both types of attractions contributed to the additional polycation adsorption, followed by the restabilization of the suspension due to electrostatic and/or steric chain repulsions. Besides the common features, there are some differences indicating the impact of the initial polyelectrolyte solution concentration on the width of the flocculation window, as illustrated in Fig. 2. This parameter increased when ci increased. Thus, maximum flocculation occurred at the following dose ranges: 7–11 mg L−1 for ci = 1 g L−1 , 7–16 mg L−1 for ci = 3 g L−1 and 7–18 mg L−1 for ci = 4 g L−1 . The different flocculation window could be explained by various conformations of the macromolecular chains in solution as a function of concentration. Thus, it is well known that in aqueous solution amphiphilic polymers tend to form associates when the polymer concentration increases beyond a critical concentration, called critical aggregation concentration. Previous fluorescence studies performed on this polymer revealed the occurrence of intra- and intermolecular hydrophobic associations at a polymer concentration of about 3 g L−1 [28] (see also Section 2). The ci values under investigation are located below and above cac. At lower ci (ci = 1 g L−1 ), the macromolecular chains are in a rather extended conformation because of the intramolecular electrostatic repulsive forces between the charged groups, which overcomes the hydrophobic association between the alkyl chains; the easier accessibility of cationic groups resulted in more rapid separation (see the rapid decrease of the residual turbidity values for polymer doses < 7 mg L−1 ) and redispersion. At ci ≥ 3 g L−1 , the association of alkyl side chains resulted in hydrophobic microdomains with a hydrophobic core surrounded by positive charges, formed either inside a single chain (unimers), or among different chains (multimers). The previous studies revealed the tendency of this polymer to form multimers’ associates, which is confirmed by the larger hydrodynamic radius of the polymer at ci = 4 g L−1 . Based on Yu and Somasundaran’s findings [32], namely, that at the initial moment
L. Ghimici, M. Nichifor / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147
145
Fig. 3. Zeta potential () dependence on the polycation dose for different initial solution concentration, (ci ): (circle) 1 g L−1 ; (star) 4 g L−1 ; settling time 15 min.
of adsorption to the surface, the macromolecular coil will retain its conformation in solution and then attempts to relax toward the surface, we assume that the polymer associated chains are probably adsorbed as such on the particle surface, without significant changes in size and conformation, that means no chain extension by dilution and spreading over the particle surface. This leads to the formation of larger polymer patches, which increase the flocculation efficiency at partial surface coverage and consequently, enlarge the flocculation window. The effect of larger hydrodynamic radius in increasing flocculation window was previously demonstrated by Schwarz et al. [10] for polyelectrolytes with different molar mass. In the present case, the higher Rh is the result of a higher apparent molar mass due to formation of multi-chain associates. The above findings allow us to conclude that, in this particular case, the changing of ci and, consequently, of association state of the polymer, led to changing of the flocculation window width. This result is for the first time observed, as in the case of other systems where hydrophilic polymers were used, only the residual turbidity value at optimum polymer dose was modified, while the optimum polymer dose range was almost the same irrespective of ci [20,22,23]. 3.2. Assessment of the flocculation mechanism It is generally believed that charge neutralization and/or patch mechanisms occur when polymers with medium and high charge content and low molar mass adsorb onto oppositely charged surface while the bridging one operates for nonionic polymers and linear polyelectrolytes with low charge density and high molar masses [2]. The chain characteristics of the sample investigated, namely, the charge content (30%) and the molar mass (40 kg mol−1 ) plead for the former mechanisms. But, in order to elucidate the flocculation mechanisms proposed above, zeta potential as well as particle size measurements have been performed and shown in the following. 3.2.1. Zeta potential measurements Plotted in Fig. 3 is the evolution of the zeta potential as a function of the polymer dose for ci of 1 and 4 g L−1 . Independent of the initial solution concentration, increasing polycation dose led to increase from the initially negative value of the suspension (−25.6 mV) until positive values for higher polymer doses. At lower polymer doses, increases monotonously, but ← just at the end of flocculation window for each experiment (at polymer dose > 11 mg L−1 and 18 mg L−1 for ci = 1 and 4 g L−1 , respectively), a sharp increase in was observed, leading to a rapid charge reversal. In both cases reached the charge neutralization point for a polycation dose which was higher (about 13 mg L−1 for
Fig. 4. Floc size distribution for different initial solution concentration, (ci ) and polycation doses located in the flocculation window: (inverted triangle) 1 g L−1 , polycation dose = 7 mg L−1 ; (star) 4 g L−1 , polycation dose = 11 mg L−1 ; (circle) no polymer. The particle size distribution is presented as volume fraction versus particle diameter.
ci = 1 g L−1 and 19 mg L−1 for ci = 4 g L−1 ) than those corresponding to the end of the flocculation window. The variation of with polymers doses, for the system with different ci , confirmed the turbidimetric results about flocculation window width. The earlier sharp increase in for the lower ci can be explained by the difference in the activity of the supplementary polymer chains adsorbed on the particle surface above 7 mg L−1 . The more extended single chains determine a rapid charge reversal, but large associates are still able to interact with each other by inter-particle hydrophobic aggregation. The rapid decrease of the residual turbidity over the range of polymer doses from 1 to 7 mg L−1 , accompanied by a monotonous change in for both polymer concentrations, until the end of the flocculation window, together with the negative values of in the whole flocculation window can be correlated with the patch mechanism for clay removal, irrespective of the initial solution concentration. An optimum dose around that needed to give a zeta potential close to zero is usually attributed to simple charge neutralization [33]. Nevertheless, the bridging mechanism can also be taken under consideration for this system. 3.2.2. Particle size measurements The particle size distribution, presented as volume fraction versus particle diameter, for D40-Oct30 at polycation doses located in the flocculation window, 7 mg L−1 for ci = 1 g L−1 and 11 mg L−1 for ci = 4 g L−1 as well as of the initial clay suspension is shown in Fig. 4. For each of the initial solution concentration a multimodal distribution was observed: (i) one peak corresponds to particles of small size, which points to the neutralization/patch mechanism for the particle clay flocculation; (ii) one or two peaks for particles of large size, which indicates that a certain portion of the particles was flocculated via bridging mechanism, because the hydrophobic chains could participate to inter-particle associations, which resulted in bigger flocs. The high volume percentage of peak (i) in the case of ci = 1 g L−1 suggests that the mechanisms based on the electrostatic attractions are the dominant ones. Peaks (ii) occurs for both ci = 1 g L−1 and ci = 4 g L−1 , but the much higher volume percentage (about 60%) of these peaks observed when ci = 4 g L−1 supports a preponderance of the bridging mechanism in this case. Perhaps, the polycation associated chains existing in solution with ci = 4 g L−1 and adsorbed at the particle surface, have also some free hydrophobic units (not included in the hydrophobic
146
L. Ghimici, M. Nichifor / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147
Fig. 5. Floc size distribution for D40-Oct30 with ci = 4 g L−1 at different polymer dose: (star) 11 mg L−1 ; (square) 18 mg L−1 ; (inverted triangle) 20 mg L−1 ; (circle) no polymer. The particle size distribution is presented as volume fraction versus particle diameter.
intra- and intermolecular associations), which could participate to inter-particle hydrophobic associations, resulting in bigger flocs. In order to obtain more evidence about the flocculation mechanisms, the flocculated suspensions with polymer doses of 7 mg L−1 , residual turbidity of 2.47% (ci = 1 g L−1 ) and 11 mg L−1 , residual turbidity 1.53% (ci = 4 g L−1 ) were stirred for 60 min at 1000 rpm. After a settling time of 15 min, the supernatant residual turbidity values were 25% and 55% for ci = 1 g L−1 and 4 g L−1 , respectively. According to Yan et al. [34] flocs which are initially formed via charge neutralization and then broken up by shearing can easily re-form upon removing the shearing forces, while polymer-bridged particles stay apart once broken up, since polymer tails and loops bridging across two or more particles are physically disrupted by the shearing forces. This chain breakage leaves residual polymer on both surfaces. Surface rearrangements of these polymeric components can lead to particle restabilization against further aggregation by the presence of an electro-steric polymer layer on the surface. Based on these findings, the more (for ci = 1 g L−1 ) and less (for ci = 4 g L−1 ) stable flocs resulted after the prolonged shearing sustain our assumption that the mechanism based on the electrostatic attractive forces prevails in the former case while bridging flocculation mechanism predominates in the latter case. The particle size distribution for D40-Oct30, ci = 4 g L−1 at the polycation doses located in the flocculation window (11 mg L−1 ) and higher (polymer dose: 18 and 20 mg L−1 ) (Fig. 5) supports both turbidimetric and zeta potential results. The smaller floc sizes recorded for the higher polymer doses than 11 mg L−1 might be explained by the adsorption of an excess positive polymer chains onto the clay particles leading to an increase in positive charges on the floc. These charges, in turn, increase the repulsion among particles, thereby hindering microflocs from growing into larger ones. 4. Conclusions For the first time we demonstrated that the association state of an amphiphilic polymer in its initial solution can significantly influence the flocculation of a clay suspension. The results suggest that polymer associated chains are adsorbed as such on the particle surface, without significant changes in size and conformation. The examination of the clay particle flocculation features when added polymer is in associated form (ci = 4 g L−1 ) led us to several conclusions:
䊉 The occurrence of a wider flocculation window, revealed by turbidimetric measurements, can be attributed to the larger hydrodynamic radius of the polymer associates. This effect is similar to that observed when polymer with higher molecular weight were used [10], but in our case the increase of apparent Mw (and, consequently, of Rh ) is due to intermolecular hydrophobic associations (multimers’ associations). Moreover, the changing of the flocculation window width with changing of ci was for the first time observed, as in the case of other systems where hydrophilic polymers were used, only the residual turbidity value at optimum polymer dose was modified (the optimum polymer dose range was almost the same irrespective of ci [20,22,23]). 䊉 The zeta potential variation with polymers doses, for the systems with different ci , confirmed the turbidimetric results about flocculation window width. The monotonous increase of until the end of the flocculation window, together with its negative values in the whole flocculation window can be correlated with the patch mechanism for clay removal, irrespective of the initial solution concentration. An optimum dose around that needed to give a zeta potential close to zero is usually attributed to simple charge neutralization [33]. 䊉 The significant increase in the volume fraction of flocs with bigger size supports the involvement of the bridging mechanism for particle aggregation and flocculation. The big flocs are disrupted under shear, and this is considered as a supplementary proof for the bridging mechanism [34]. 䊉 The above findings have implications for separation processes where amphiphilic polymers are used; in this particular case, the presence of large associates in the initial polymer solution determines a significant improvement of the flocculation process by enlargement of the flocculation window and floc size. Moreover, a large flocculation window is preferred in the real flocculation processes where significant fluctuations of the medium parameters (pH, temperature, ionic strength) take place. Acknowledgements The authors thank the group of Micro-and nanostructures characterization Laboratory (“P. Poni” Institute of Macromolecular Chemistry, Iasi, Romania) for performing zeta potential and particle dimension measurements. This work was supported by a grant of the Romanian National Authority for Scientific Research, CNCSUEFISCDI, project number PN-II-ID-PCE-2011-3-0622. References [1] J. Gregory, Particle in Water: Properties and Processes, CRC, Press, Boca Raton, Fl, USA, 2006. [2] B. Bolto, J. Gregory, Organic polyelectrolytes in water treatment, Water Res. 41 (2007) 2301–2324. [3] V. Runkana, P. Somansundaran, P.C. Kapur, Mathematical modeling of polymerinduced flocculation by charge neutralization, J. Colloid Interface Sci. 270 (2004) 347–358. [4] J.A. Caskey, R.J. Primus, The effect of anionic polyacrylamide molecular conformation and configuration on flocculation effectiveness, Environ. Prog. 5 (1986) 98–103. [5] S. Sakohara, T. Kimura, K. Nishikawa, Flocculation mechanism of suspended particles using the hydrophlic/hydrophobic transition of a thermosensitive polymer, KONA 20 (2002) 246–250. [6] S.P. Strand, K.M. Vårum, K. Østgaard, Interactions between chitosans and bacterial suspensions: adsorption and flocculation, Colloids Surf. B: Biointerfaces 27 (2003) 71–81. [7] J. Roussy, M. Van Vooren, E. Guibal, Influence of chitosan characteristics on coagulation and flocculation of organic suspension, J. Appl. Polym. Sci. 98 (2005) 2070–2079. [8] J. Labille, F. Thomas, M. Milas, C. Vanhaverbeke, Flocculation of colloidal clay by bacterial polysaccharides: effect of macromolecule charge and structure, J. Colloid Interface Sci. 284 (2005) 149–156. [9] G. Petzold, M. Mende, N. Kochurova, Polymer–surfactant complexes as flocculants, Colloids Surf. A: Physicochem. Eng. Aspects 298 (2007) 139–144.
L. Ghimici, M. Nichifor / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 142–147 [10] S. Schwarz, W. Jaeger, B.-R. Paulke, S. Bratskaya, N. Smolka, J. Bohrisch, Cationic flocculants carrying hydrophobic functionalities: applications for solid/liquid separation, J. Phys. Chem. B 111 (2007) 8649–8654. [11] Y. Wei, F. Cheng, H. Zheng, Synthesis and flocculating properties of cationic starch derivatives, Carbohydr. Polym. 74 (2008) 673–679. [12] L.-J. Wang, J.-P. Wang, S.-J. Zhang, Y.-Z. Chen, S.-J. Yuan, G.-P. Sheng, H.-Q. Yu, A water-soluble cationic flocculant synthesized by dispersion polymerization in aqueous salts solution, Sep. Purif. Technol. 67 (2009) 331–335. [13] R. Gaudreault, N. Di Cesare, D. Weitz, T.G.M. van de Ven, Flocculation kinetics of precipitated calcium carbonate, Colloids Surf. A: Physicochem. Eng. Aspects 340 (2009) 56–65. [14] A. Hebeish, A. Higazi, S. El-Shafei, Synthesis of carboxymethyl cellulose (CMC) and starch-based hybrids and their applications in flocculation and sizing, Carbohydr. Polym. 29 (2010) 60–69. [15] I.C. Ho, I. Norli, A.F.M. Alkarkhi, N. Morad, Characterization of biopolymeric flocculant (pectin) and organic synthetic flocculant (PAM): a comparative study on treatment and optimization in kaolin suspension, Bioresour. Technol. 101 (2010) 1166–1174. [16] M.G. Rasteiro, F.A.P. Garcia, P.J. Ferreira, E. Antunes, D. Hunkeler, C. Wandrey, Flocculation by cationic polyelectrolytes: relating efficiency with polyelectrolyte characteristics, J. Appl. Polym. Sci. 116 (2010) 3603–3612. [17] R. Rojaj-Reyna, S. Schwarz, G. Heinrich, G. Petzold, S. Schutze, J. Bohrisch, Flocculation efficiency of modified water soluble chitosan versus commonly used commercial polyelectrolytes, Carbohydr. Polym. 81 (2010) 317–322. [18] D. Tian, X. Wu, C. Liu, H.-Q. Xie, Synthesis and flocculation behavior of cationic konjac glucomannan containing quaternary ammonium substituents, J. Appl. Polym. Sci. 115 (2010) 2368–2374. [19] S. Piriyaprasarth, P. Sriamornsak, Flocculating and suspending properties of commercial citrus pectin and pectin extracted from pomelo (Citrus maxima) peel, Carbohydr. Polym. 83 (2011) 561–568. [20] L. Ghimici, S. Morariu, M. Nichifor, Separation of clay suspension by ionic dextran derivatives, Sep. Purif. Technol. 68 (2009) 165–171. [21] L. Ghimici, M. Nichifor, Novel biodegradable flocculanting agents based on cationic amphiphilic polysaccharides, Bioresour. Technol. 101 (2010) 8549–8554.
147
[22] B.Q. Zheng, J.Q. Qian, X.K. Li, Z.H. Zhu, Effect of the parent solution concentration on the flocculation performance of PAAm flocculants and the relation between the optimal parent solution concentration and critical concentrations, J. Appl. Polym. Sci. 103 (2007) 1585–1592. [23] L. Ghimici, M. Constantin, G. Fundueanu, Novel biodegradable flocculanting agents based on Pullulan, J. Hazard. Mater. 181 (2010) 351–358. [24] M. Nichifor, M.C. Stanciu, B.C. Simionescu, New cationic hydrophilic and amphiphilic polysaccharides synthesized by one pot procedure, Carbohydr. Polym. 82 (2010) 965–975. [25] G.S. Manning, Limiting laws and counterion condensation in polyelectrolyte solutions. I. Colligative properties, J. Chem. Phys. 51 (1969) 924–933. [26] M.V.S. Rao, Viscosity of dilute to moderately concentrated polymer solutions, Polymer 34 (1993) 592–596. [27] M. Nichifor, M.C. Stanciu, L. Ghimici, B.C. Simionescu, Hydrodynamic properties of some cationic amphiphilic polysaccharides in dilute and semi-dilute aqueous solutions, Carbohydr. Polym. 83 (2011) 1887–1894. [28] M. Nichifor, S. Lopes, M. Bastos, A. Lopes, Self-aggregation of amphiphilic cationic polyelectrolytes based on polysaccharides, J. Phys. Chem. B 108 (2004) 16463–16472. [29] V. Lobaskin, K. Qamhich, Effective macroion charge and stability of highly asymmetric electrolytes at various salt conditions, J. Phys. Chem. B 107 (2003) 8022–8029. [30] G. Gillies, W. Lin, M. Borkovec, Charging and aggregation of positively charged latex particles in the presence of anionic polyelectrolytes, J. Phys. Chem. B 111 (2007) 8626–8633. [31] P.M. Claesson, M.A.G. Dahlgren, L. Eriksson, Forces between polyelectrolytecoated surfaces: relations between surface interaction and floc properties, Colloids Surf. A 93 (1994) 293–303. [32] X. Yu, P. Somasundaran, Role of polymer conformation in interparticle bridging dominated flocculation, J. Colloid Interface Sci. 177 (1996) 283–287. [33] J. Kleimann, C. Gehin-Delval, H. Auweter, M. Borkovec, Super-stoichiometric charge-neutralization in particle– polyelectrolyte systems, Langmuir 21 (2005) 3688–3698. [34] Y.D. Yan, S.M. Glover, G.J. Jameson, S. Biggs, The flocculation efficiency of polydisperse polymer flocculants, Int. J. Miner. Process. 73 (2004) 161–175.