Study of Filler Flocculation Mechanisms and Floc Properties Induced

The research on filler flocculation and floc properties induced by polyethylenimine was carried out by monitoring the particle chord length distributi...
0 downloads 0 Views 108KB Size
Download PDF
5616

Ind. Eng. Chem. Res. 2005, 44, 5616-5621

Study of Filler Flocculation Mechanisms and Floc Properties Induced by Polyethylenimine Elena Fuente,† Angeles Blanco,*,† Carlos Negro,† Marı´a A. Pelach,‡ Pere Mutje,‡ and Julio Tijero† Chemical Engineering Department, Complutense University of Madrid, Avda Complutense s/n, 28040 Madrid, Spain, and Chemical Engineering Department, University of Gerona, Av. Lluis Santalo´ , s/n, 17071 Girona, Spain

The research on filler flocculation and floc properties induced by polyethylenimine was carried out by monitoring the particle chord length distribution during the flocculation, deflocculation, and reflocculation processes for different polymer doses. Results show that there are two PEI doses that produce a maximum flocculation due to different flocculation mechanisms. Low PEI doses induce a fast flocculation through bridge formation, but high PEI doses induce a slow flocculation by charge neutralization. The proposed flocculation mechanisms, presented in this paper, explain the observed floc properties and their dependence on the polymer dosage. As the flocculation mechanism determines the floc properties and, therefore, the nature of the retention and drainage processes, the findings may allow papermakers to improve the optimization of the wet-end processes by controlling the PEI dosage more accurately. The study of flocculation kinetics and floc stability showed that the optimal dosage point depends strongly on the PEI dosage used. Introduction Polyethylenimines (PEI) are a group of branched polymers with different molecular weights and charge densities whose base structure is shown in Figure 1.1 They are used to accomplish several aims in papermaking, e.g. as retention and drainage aids, as strength additives, as systems to control detrimental substances, and as pitch controllers.2,3 The nonmodified PEI is obtained by the ring-opening cationic polymerization of ethylenimine, according to Dick and Ham.4 However, nonmodified polyethylenimine is not the most appropriate retention system for alkaline papermaking because it loses its charge density at a pH above 5.5.2,5 Therefore, nowadays, modified polyethylenimines are used in the paper industry. These modified PEIs are obtained from other routes of synthesis, differing from the Dick and Ham route. Thus a wide variety of molecular weights, charge densities, and structures can be obtained. They can have a comblike branched chain structure crosslinked with a small quantity of epichlorohydrin or even a linear PEI chain. The charge density depends on the percentage of amine groups that are chemically or sterically blocked.2,6,7 Suty and co-workers measured the stability of a ground calcium carbonate (GCC) suspension with different doses of the unmodified polyethylenimine, and they concluded the dosage that produced the lowest stability corresponded to the isoelectrical point of the suspension. They also concluded that the flocculation mechanism of the suspension under these conditions was charge neutralization, because the stability of the suspension increased significantly when the polymer * To whom correspondence should be addressed. Tel.: +3491-3944247, Fax: +34-91-3944243. E-mail: ablanco@ quim.ucm.es. † Complutense University of Madrid. ‡ University of Gerona.

Figure 1. Nonmodified polyethylenimine structure.

dosage was lower or higher than the one necessary to reach the isoelectrical point.8 Other authors have studied the interactions between papermaking suspensions and unmodified PEI concluding that they are determined by the electrostatic forces,9-12 that PEI acts by charge neutralization,5,10-13 and that the kinetics of PEI adsorption on the fibers is a second order with respect to polymer concentration.14 However, the interaction and flocculation mechanism induced by the high molecular weight modified PEIs is not clear yet, as affirmed by Gess.2 Many authors agree that the high molecular weight PEI acts by patch formation,3,15 but there are several studies about the PEI adsorption on different particles that show that the interaction among PEI and particles depends on many variables such as the pH, the charge density, the ionic strength and the molecular weight of the polymer.16,17 It is well-known that the effect of any polymer on the flocculation kinetics and on floc properties depends on the flocculation mechanism.15,18 Charge neutralization forms small and compact flocs improving the retention without disturbing the sheet formation. The optimal dosage, in this case, would correspond to the isoelectrical point, and the flocs would be reversible if they are

10.1021/ie0503491 CCC: $30.25 © 2005 American Chemical Society Published on Web 06/18/2005

Ind. Eng. Chem. Res., Vol. 44, No. 15, 2005 5617

broken down due to high shear forces or turbulence. Flocculation by patch or bridge formation can take place with lower polymer doses, reducing the negative effect of PEI on the optical properties of the sheet19 and forming larger flocs that favor the retention of fines and fillers. In this case, if hydrodynamic forces break down the flocs, they will not be formed again completely, especially if bridging takes place. Therefore, the aim of this work was to study the flocculation mechanism induced by a high molecular weight modified polyethylenimine at different doses. The study was carried out by monitoring the chord length distribution of the suspended particles during flocculation. Furthermore, deflocculation and reflocculation processes were also studied to determine the process kinetics, the floc properties, and the flocculation mechanisms. This information is crucial to both optimizing the wet-end during papermaking and to controlling it. However, it is necessary to remember the differences with papermaking conditions, where there are many compounds interacting together, different time scales, and higher hydrodynamic forces. Experimental Section Experiments were carried out with a 1% GCC suspension in distilled water, at 25 °C. The mean particle size was around 2 µm, with 99% of the particles being smaller or equal to 7 µm. The specific area was 2.18 m2/ g. The pH of the suspension was around 9.65 and it was constant during the experiments. At this high pH, the solubility of the GCC is very low: only 12 ppm of GCC is dissolved. The flocculation of the suspension was carried out at 250 rpm, because this stirring intensity is sufficient to keep the particles suspended but minimizes floc breakage. The flocculant used was a modified PEI (polymin SK), of high molecular weight (1.3 × 106) and mediumhigh charge density (4.2 meq/g). Its molecular weight is more than double that of the one used by Suty et al., and, in theory, its flocculation mechanism would be patching. A. Polymer Adsorption. To know the maximum adsorption of PEI on GCC particles, the adsorption isotherm at 25 °C was obtained. A 200 mL amount of the suspension of GCC 1% was stirred for 24 h with different PEI doses. After that, the suspensions were centrifuged with a centripetal acceleration of 19600 m/s2 (2000 times gravity) for 5 min, and the dissolved PEI, remaining in the liquid phase, was determined by titration with poly(vinyl sulfate potassium salt) (PVSK) 0.0006 N with a charge detector Mutek PCD03 and an automatic titration. A blank experiment was also carried out, without GCC, to determine the adsorption of the polymer on other surfaces, as for example, the beaker wall, the stirrer, and the sensor probe. The difference between the adsorbed polymer amounts in both cases corresponds to the amount of polymer that is adsorbed on the filler particle surface. To know the amount of adsorbed polymer during flocculation and the effect of deflocculation on it, the amount of adsorbed polymer was measured after the flocculation process and after the deflocculation process. A blank without GCC was also carried out. B. GCC Flocculation. The flocculation process was monitored by a focused beam reflectance measurement probe (FBRM), which measures the particle chord distribution in real time.20-22 The evolution of that

distribution was monitored through statistics, such as the mean chord size and the total number of counts, which are related to size, shape, and concentration of the particles. A full description of the measurement technique, the methodology, and the statistics used has been published in previous works.20,21 The optimization of the PEI dosage to induce filler flocculation was carried out by flocculation trials (at 250 rpm) with different doses of flocculant. It was considered that an optimal dosage corresponds to a maximum in the mean chord. Flocculation-deflocculation-reflocculation trials were carried out with different doses of polymer, including the optimal dosage obtained with the FBRM and the one equal to the cationic demand of the suspension. Flocculation was carried out at 250 rpm, and the evolution of the flocs was monitored at this stirring intensity for 4 min. Then the stirring intensity was increased to 750 rpm for 4 min, to break down the formed flocs. Finally, the reflocculation of the particles was studied by decreasing the impeller speed to 250 rpm again. The cationic demand was obtained by titration of the GCC suspension with the PEI solution, by means of a charge detector MUTEK PCD03. The results were analyzed using a modified model of the flocculation kinetics based on the Smoluchowski theory.18,23 This model describes the kinetics of the evolution of the particle concentration (particles/L) when there are two processes that tend toward the equilibrium: the flocculation process, with a second-order kinetics, and the formed floc breakage, with a first-order kinetics. With the FBRM we measure the number of counts per second, which is the number of particles whose chords the sensor has measured per second. The device generates a laser beam that is focused in a focal point in the probe sapphire window. The focal point describes circles at very high speed (2000 rpm). The detector measures the time duration of the light pulses produced by the light backscattered by the particles when they intercept the focal point path. The movement of the particles does not affect the measurement because of the high focal point speed. Therefore, the number of particles that cross the focal point path increases with the number of particles per volume unit. Thus, the number of counts per second reflects the particle concentration.24,25 Therefore, the flocculation kinetics can be studied by using the model represented by eq 1.26

dnc ) -kc1nc2 + kc2nc dt

(1)

where nc is the number of counts measured per second, t is the time (s), and kc1 and kc2 are kinetic constants. This allows us to compare the flocculation kinetics and the deflocculation kinetics of different flocculation trials. The relationship between both sets of kinetics allows us to obtain the equilibrium situation toward which the system tends. The ratio between kc2 and kc1 is a constant, K, that indicates the number of counts toward which the system tends when t tends to infinity. Therefore, higher kc1 values represent a faster aggregation process and lower kc2 values represent more stable flocs. Results and Discussion Flocculation and Reflocculation. The enhancement of the mean chord size after flocculation and after

5618

Ind. Eng. Chem. Res., Vol. 44, No. 15, 2005

Figure 2. Evolution of the mean chord size during PEI addition. Determination of the optimal PEI dosage.

Figure 4. Adsorption of PEI on GCC at the equilibrium. Table 1. Adsorption of PEI on GCC after Flocculation and after Deflocculation PEI dosage (ppm) (mg/m2) after flocculation after deflocculation

Figure 3. Evolution of the mean chord size after PEI addition.

reflocculation of GCC with different doses of PEI is shown in Figure 2. These are calculated as the difference between the maximum mean chord size of the suspension, reached after adding the flocculant, and the initial value of the mean chord size, and as the difference between the mean chord size value after the reflocculation process and the initial value, respectively. It is observed that the optimal dosage of PEI, from the point of view of the flocculation process, is 5-10 ppm. This corresponds to the maximum mean chord size enhancement. Doses higher than that reduced the mean chord size enhancement after the flocculation and reflocculation processes. However, higher doses, around 50 ppm, produced another maximum in the mean chord enhancement. The theoretical dosage of PEI necessary to neutralize the charge of the particles was 64 ppm. It is observed that, for that dosage, a high flocculation degree was achieved and a remarkable reflocculation was obtained, corresponding to the maximum reversibility of the flocs. These results show that the mechanism of the GCC flocculation induced by PEI depends on the flocculant dosage. On the other hand, Figure 3 shows that when the flocculant dosage was very low, 1.25 ppm, flocculation also took place, producing stable flocs and duplicating the mean chord size of the suspended particles. In this case, flocs were only partially reversible (Figure 2). Adsorption. Figure 4 shows the amount of PEI adsorbed on GCC particles in the equilibrium as a function of the PEI dosage. These results are expressed as mg of PEI per square meter of GCC surface, considering that the specific area of this GCC product is the arithmetic mean value between the two limits given by the manufacturer (2.05 m2/g - 2.3 m2/g) that is 2.18 m2/ g. This figure also shows that the maximum experimental adsorption was around 3.7 mg/m2.

50 100 200 50 100 200

2.27 4.50 8.82 2.27 4.50 8.82

adsorbed adsorbed PEI PEI (% from max. (mg/m2) PEI adsorption) 2.21 2.36 2.57 2.00 2.28 3.17

59 63 73 54 61 86

S. Akari and co-workers27 observed that the average chain diameter for a PEI with a molecular weight of 1.3 × 106 g/mol was about 100 nm. This implies that the mathematical amount of PEI necessary to cover the surface with a monolayer is around 1.5 mg/m2, considering a plane surface of 1 m2 and assuming that PEI chains are spheres with a diameter of 100 nm and with a mass of 2.2 × 10-18 g. It is observed that the maximum adsorption obtained and these results are in the same order of magnitude. Therefore, we can assume that the PEI adsorption of 3.7 mg/m2 corresponds to the necessary amount of polymer to cover the particle surface. The experimental PEI adsorption is higher than the estimated one because of the assumptions made. For example, the PEI chains are not equal spheres, they are branched molecules with different sizes and weights, and the GCC surface is not a plane of 1 m2, but it is distributed in rhombohedrical particles with different sizes. Furthermore, the high pH of the GCC suspension (9.5) reduces the PEI charge density and, therefore the electrostatic repulsion between chains, increasing the adsorption capability, as observed by other authors.1,17 The results from the blank adsorption trials, carried out without GCC, showed that the adsorption of PEI onto surfaces other than the particles was negligible. The amount of adsorbed polymer after the flocculation process and after the deflocculation process is summarized in Table 1. When the PEI dosage was 50 ppm (2.27 mg/m2), almost all the PEI chains were adsorbed on the particles surfaces during the flocculation process. Therefore, we can consider that all the PEI chains were adsorbed during the flocculation stage, when the PEI dosage was lower than 50 ppm. Thus, when the PEI dosage was 1.25 ppm, the covering grade of the particle surface would be around 1.5%, assuming that it was necessary to adsorb 3.7 mg/ m2 (80 ppm) to cover the surface of the particles. This covering grade is very low for a patch model, and consequently the flocculation in such conditions has to be explained by bridge formation. Kinetics and Mechanism. It is necessary to consider that the PEI used in this work has a mean

Ind. Eng. Chem. Res., Vol. 44, No. 15, 2005 5619

Figure 5. Flocculation kinetic constant.

molecular weight of 1.3 × 106 g/mol, being polydisperse. Polymers with this molecular weight can form bridges,15,28 and there are many PEI chains with larger molecular weights than the average that have a higher tendency to form bridges among particles. Furthermore, the high pH of the suspension decreases the cationic charge of the polymer favoring the presence of tails, and therefore its properties move toward the bridging polymers. The optimal dosage obtained in Figure 2 (5 ppm) would correspond to a covering grade of around 6%. This coverage grade is still too low to induce patching flocculation. In this situation, the predominant flocculation mechanism is bridge formation, and it could be possible that the shortest chains would also form some patches among some particles by partially neutralizing their charge. Although the flocculation mechanism is the same, the behavior of the system with 1.25 and 5 ppm is very different as shown in Figure 3, which shows different mean chord size evolutions for these doses. This difference is explained by the competition between several simultaneous processes:15,29 • The aggregation of particles by bridging to form flocs. • The break down of the formed flocs. • The reconformation of the adsorbed polymer toward a flatter configuration (flattening) that reduces the probability of bridging and encourages patching flocculation.27 • The aggregation of particles after the flattening of the adsorbed polymer by the patching mechanism, when the flattening is not enough to avoid the adsorbed polymer being out of the electric double layer. When the PEI dosage was 1.25 ppm, the flocculation rate was low, as shown by the low kc1 value in Figure 5, because the low PEI dosage reduced the collision efficiency. Furthermore, the polymer flattening reduced the overall rate of flocculation. This explains why the mean chord size increases slowly toward a constant value after adding 1.25 ppm of PEI. When the polymer dosage was 4 times higher, the collision efficiency increased and, therefore, flocculation was faster than the flattening. The maximum mean chord size was reached in a few seconds, but some of the formed flocs were destroyed slowly and, due to the reconformation of the polymer, which evolved toward a flat configuration, they were not formed again. Consequently, the flocs were unstable and the mean chord size decreased slowly (Figure 3). Figure 2 shows that a PEI dosage of around 50 ppm also produced a high flocculation. The amount of adsorbed polymer on GCC in this case was 2.2 mg/m2 approximately, with less than the 3% of the polymer remaining in the solution (Table 1). Therefore, assuming

Figure 6. Equilibrium constant toward which the system tends.

Figure 7. Deflocculation kinetic constant. Floc stability.

that the covering grade was total with a PEI adsorption of 3.7 mg/m2, the covering grade when polymer dosage was 50 ppm was around 60%. Figure 2 shows that after the flocculation process, the mean chord size enhancement was very similar to the one obtained with 5 ppm. However, the reflocculation process was more efficient with 50 ppm of PEI. Figures 3 and 5 show that there is a strong difference between the evolution behavior of the suspension with 50 ppm of PEI and with 5 ppm. When the PEI dosage was 5 ppm, the flocculation was faster producing unstable flocs that were broken down slowly. When the polymer dosage was 50 ppm, the flocs were more stable and larger at the end of the experiment. In this case, there was enough polymer to allow flocculation by patch formation and, possibly, also by charge neutralization of some particles. The flocculation by charge neutralization does not require an extended configuration of the PEI chains. Furthermore, the higher covering grade could limit the flattening process. Flocs formed after the flocculation of GCC induced by 64 ppm of PEI (the cationic demand of the suspension) were slightly smaller than the flocs formed by addition of 50 ppm of PEI, but the final number of counts toward which the system tended (K) was lower than the final number of counts with 50 ppm, as shown in Figure 6. Therefore, there are fewer particles at 64 ppm, but the same solid total mass. This means that the addition of 64 ppm of PEI to the suspension produced the aggregation of a larger number of particles, but they form smaller flocs. This could be because with 64 ppm of PEI more particles could aggregate by neutralization and they could form a higher number of smaller, compact flocs, as shown in Figure 6. The flocculation rate was slower than the flocculation rate obtained with 50 ppm, as shown by the kc1 values in Figure 5, but Figure 7 shows that the formed flocs were more stable since kc2 is lower. Figure 8 shows the evolution of the mean chord size during the flocculation-deflocculation-reflocculation trials. Flocs induced by adding 64 ppm of PEI are stronger, as shown by the higher mean chord size value obtained when flocs were

5620

Ind. Eng. Chem. Res., Vol. 44, No. 15, 2005

Figure 8. Mean chord size evolution during flocculation-deflocculation-reflocculation of GCC induced by PEI.

broken at 750 rpm, and they are totally reversible, as shown by the final value of the mean chord size being very similar to the one obtained after the flocculation process. However, the addition of 50 ppm of PEI forms only partially reversible flocs. This could mean that the flocculation mechanism induced by the addition of 50 ppm of PEI was a mixture of patching and neutralization, because of the nonhomogeneous distribution of the polymer on the particle surfaces and the broad molecular weight distribution. Table 1 shows that almost all the polymer was adsorbed after the addition of 50 ppm of PEI. However, after the deflocculation process, 10% of the polymer had been desorbed. Therefore, when the flocs induced by patching flocculation (with 50 ppm of PEI) were broken down, some polymer chains could be desorbed and the conformation of some of the adsorbed polymer chains could became flatter. To interact with other particles, the patches width must be larger than the electrostatic double layer width. Therefore, the polymer flattening could decrease the patching flocculation efficiency. When the polymer dosage was 64 ppm, the covering grade was high enough to limit such flattening and, furthermore, the charge of more particles would be neutralized, decreasing the effect of flattening on the flocculation process. Therefore, particles would be able to interact again, after the deflocculation, forming new flocs. Because of this, flocs were completely reversible when they were formed with 64 ppm of PEI. Figure 2 shows also a minimum flocculation at PEI doses of around 20 ppm. This could indicate that the coverage grade with these doses could be too low to produce an optimal patching flocculation but it could be too high to carry out flocculation by bridging. Polymer doses higher than 64 ppm also produced flocculation but it was slower, and 4 min after the polymer addition, the mean chord size reached was lower. Figure 8 shows that the reflocculation process was too slow to be completed in 4 min, as supported by Figure 9, which was obtained by completing the flocculation and the reflocculation processes. This figure shows that, after reflocculation, the flocs reached the same size as the ones formed after flocculation. Thus, the formed flocs were totally reversible, as they are in a scenario where the charge neutralization mechanism is predominant. The values of kc1 in Figure 5 and the slopes of the mean chord size evolution curves during the flocculation process in Figure 8 show that when the dosage of PEI was high, the flocculation became slower and the final mean chord size enhancement was lower. This could be because the excess of polymer caused electrostatic repulsion among some particles and because the adsorption of polymer chains can be heterogeneous in the

Figure 9. Mean chord size evolution when the flocculation and the reflocculation processes are allowed to reach equilibrium (PEI dosage, 125 ppm).

first moments. Therefore, there could be particles with too much polymer adsorbed, which cannot flocculate unless they collide with a particle with low coverage. The frequency of efficient collisions would decrease because of this, PEI chains would redistribute over time, and the system would evolve toward a situation with a more homogeneous PEI chain distribution, where some particles have been able to aggregate. This redistribution process could limit the flocculation rate. Figures 8 and 9 show that when the stirring speed was increased, after flocculation at PEI doses higher than 50 ppm, the deflocculation process became predominant during the first few seconds but, after a while, the mean chord size increased again. Furthermore, results show that the increase of the stirring intensity produced the desorption of some PEI chains during floc breakage. When the PEI dosage was 100 ppm (Table 1), the available GCC surface increased due to polymer desorption and due to floc breakage but the external diffusion was improved by the high stirring and that encouraged the PEI redistribution. Consequently, when the concentration of flocs was low enough, the flocculation rate became higher than the deflocculation rate and the mean chord size started to increase slowly. Conclusions The flocculation mechanism of the GCC suspension induced by the high molecular weight modified PEI depends on the polymer dosage. The complete study of flocculation kinetics and floc properties allows us to propose a scenario to explain the behavior of the suspension at different PEI doses. When the polymer dosage is lower than 50 ppm, bridging flocculation is the predominant mechanism and formed flocs are not reversible. Patching flocculation participation increases with the polymer dosage, and, in this case, it is the predominant mechanism for polymer doses around 50 ppm. Neutralization flocculation is predominant for polymer doses of 64 ppm or higher. PEI doses of 100 ppm or higher produce a slow flocculation, where the kinetics could be limited by the PEI chains redistribution and by the electrostatic repulsion between the adsorbed polymer chains. The reversibility of the flocs depends strongly on the flocculation mechanism. It also depends on the polymer dosage, as the covering grade affects the flocculation mechanism. The flocs induced by the addition of a PEI dosage between 64 and 125 ppm are reversible, because the predominant flocculation mechanism is charge neutralization. The flocs induced by the addition of doses lower than 64 ppm are partially reversible.

Ind. Eng. Chem. Res., Vol. 44, No. 15, 2005 5621

It is hazardous to extrapolate the results obtained under laboratory conditions, with a model suspension, to full-scale papermaking conditions. In the real pulp suspension there are fibers, fines, fillers, and dissolved and colloidal material interacting with each other and with the polymer. Furthermore, the hydrodynamic forces are more intense in the mill during formation. However, it is possible to conclude that the PEI dosage could be the key to controlling and to modifying the flocculation process, the floc properties, and their effects on retention, drainage, and formation during papermaking. Small changes in polymer dosage could produce very important effects on retention and formation, especially when the dosage corresponds to the optimum to produce bridging flocculation. These low doses should be added after the headbox lips to avoid floc breakage because these flocs are not reversible. Doses similar to the one necessary to neutralize the charge of the suspension induce a slow flocculation. Consequently, they should be added before the headbox, but they could lead to a good retention because these doses induce the formation of totally reversible flocs. Therefore, the optimization of PEI dosage and the addition point, when they are considered together as dependent variables, could be a way to control the retention and the formation processes. Literature Cited (1) Goethals, E. J. Polymeric Amines and Ammonium Salts; Pergamon Press: Oxford, 1980. (2) Gess, J. M. Retention of Fines and Fillers during Papermaking; Tappi Press: Atlanta, 1998; pp 61-176. (3) Stange, A.; Moormann-Schmitz, A.; Blum, R.; Robert, D. New Polymers to Overcome Pitch Problems. Rev. ATIP 2001, 55 (4), 32-41. (4) Dick, C. R.; Ham, G. E. Characterisation of Polyethylenimine. J. Macromol. Sci.-Chem. A. 1970, 4 (6), 1301-1314. (5) Litchfield, E. Dewatering Aids for Paper Applications. Appita J. 1994, 47 (1), 62-65. (6) Zhuk, D. S.; Gembitsky, P. A.; Chmarin, A. I. Method of Producing Linear Polyethylenimine. US 4032480, 1997. (7) Schmeider, J. G.; Dick, C. R.; Ham, G. E. Controlled Molecular Weight Aziridine Polymers. US 3519687, 1970. (8) Suty, S.; Alince, B.; van de Ven, T. G. M. Stability of Ground and Precipitated Calcium Carbonate Suspensions in the Presence of Polyethylenimine and Salt. J. Pulp. Pap. Sci. 1996, 22 (9), J321-J326. (9) Alince, B.; van de Ven, T. G. M. Kinetics of Colloidal Particle Deposition on Pulp Fibers 2. Deposition of Clay on Fibers in the Presence of Poly(ethylenimine). Colloids Surf. A. 1993, 71 (1), 105-114. (10) Porubska, J.; Alince, B.; van de Ven, T. G. M. Homo- and Heteroflocculation of Papermaking Fines and Fillers. Colloids Surf., A. 2002, 210 (2-3), 97-104. (11) Suty, S.; Luzakova, V. Role of Surface Charge in Deposition of Filler Particles onto Pulp Fibres. Colloids Surf. A. 1998, 139, 271-278.

(12) Alince, B. Clay-Fiber Interaction in the Presence of Polyethylenimine and Anionic Contaminants. Colloids Surf. 1988, 39, 279-288. (13) Vanerek, A.; Alince, B.; van de Ven, T. G. M. Colloidal Behaviour of Ground and Precipitated Calcium Carbonate Fillers: Effects of Cationic Polyelectrolytes and Water Quality. J. Pulp. Pap. Sci. 2000, 26 (4), J135-J139. (14) Nedelcheva, M. P.; Stoilkov, G. V. Polyethylenimine Adsorption by Cellulose. J. Appl. Polym. Sci. 1976, 20 (8), 21312141. (15) Blanco, A.; Negro, C.; Tijero, J. Developments in Flocculation; Pira Int: Leatherhead, U.K., 2001. (16) Lindquist, G. M.; Stratton, R. A. The Role of Polyelectrolyte Charge Density and Molecular Weight on the Adsorption and Flocculation of Colloidal Silica with Polyethylenimine. J. Colloid Interface Sci. 1976, 55 (1), 45-59. (17) Pelach, M. A.; Blanco, M. A.; Roux, J. C.; Vilaseca, F.; Mutje, P. Study of the Adsorption of Polyethylenimine on Cellulosic Fibres in an Aqueous Suspension. Appita J. 2001, 54 (5), 460464. (18) Van de Ven, T. G. M. Colloidal Hydrodynamics; Academic Press Inc: London, 1989. (19) Orden, M. U.; Matı´as, M. C.; Urreaga, J. M. Discoloration of Celluloses Treated with Polyethylenimines. Polym. Degrad. Stab. 2004, 85, 697-703. (20) Blanco, A.; Fuente, E.; Negro, C.; Tijero, J. Flocculation Monitoring: Focused Beam Reflectance Measurement as a Measurement Tool. Can. J. Chem. Eng. 2002, 80 (4), 734-740. (21) Blanco, A.; Fuente, E.; Negro, C.; Monte, C.; Tijero, J. Focused Beam Reflectance Measurement as a Tool to Measure Flocculation. Tappi J. 2002, 1 (10), 14-20. (22) Yoon, S.; Deng, Y. Flocculation and Reflocculation of Clay Suspensions by Different Polymer Systems under Turbulent Conditions. J. Coll. Interface Sci. 2004, 278 (1), 139-145. (23) Thomas, D. N.; Judd, J.; Fawcett, N. Flocculation Modelling: A Review. Wat. Res. 1999, 33 (7), 1579-1592. (24) Ruf, A.; Worlitschek, J.; Mazzotti, M. Modeling and Experimental Analysis of PSD Measurements through FBRM. Part. Part. Syst. Charact. 2000, 17, 167-179. (25) Dowding, P. J.; Goodwin, J. W.; Vincent, B. Factors Governing Emulsion Droplet and Solid Particle Size Measurements Performed Using the Focuesd Beam Refectance Technique. Colloids and Surf. A. 2001, 192, 5-13. (26) Negro, C.; Fuente, E.; Blanco, A.; Tijero, J. Flocculation Mechanism Induced by Phenolic Resin/PEO and Floc Properties. AIChE J. 2004, 51 (3), 1022-1031. (27) Akari, S.; Schrepp, W.; Horn, D. Imaging of Single Polyethylenimine Polymers Adsorbed on Negatively Charged Latex Spheres by Chemical Force Microscopy. Langmuir 1996, 12 (4), 857-860. (28) Fleer, G. J.; Lyklema, J. Polymer Adsorption and its Effect on the Stability of Hydrophobic Colloids. II. The Flocculation Process as Studied with the Silver Iodide-Polyvinyl Alcohol System. J. Coll. Interface Sci. 1974, 46 (1), 1-12. (29) Biggs, S.; Habgood, M.; Jameson, G. J.; Yan, Y. Aggregate Structures Formed via a Bridging Flocculation Mechanism. Chem. Eng. J. 2000, 80 (1-3), 13-22.

Received for review March 16, 2005 Revised manuscript received April 25, 2005 Accepted April 26, 2005 IE0503491