Average Size Of Flocs Research Articles

Two-stage injection of polymer and microsand during ballasted flocculation for treating kaolin waters with or without humic acid: Floc evolutional characteristics, performance and mechanisms

Ballasted flocculation is regarded as a most promising water treatment technology in aspects of retrofit and high-rate applications. To deep understand the incorporation behaviors of ballasting agent into ballasted floc growth, two distinct injection modes (namely a two-stage injection of polyacrylamide (PAM) alone, and a two-stage injection of both PAM and microsand) were developed in this study. Then, ballasted flocculation tests of kaolin and kaolin-HA (humic acid) waters were conducted at varying split ratios for fixed total dosages of both PAM and microsand. The experimental results showed that for either two-stage injection mode, the higher the second percentage of each split ratio, the greater the average size of maturated flocs at the second sub-stage of maturation. Meanwhile, the turbidity and UV254 values of settled water became lower at 30 and 180 s of sedimentation, suggesting that varying split ratios significantly affected the kinetics of ballasted floc growth. Moreover, it was suggested that the selection of either two-stage injection mode or corresponding split ratios played a more pronounced role in the HA removal than the total dosage of PAM. This suggestion was supported by SEM, FTIR and XPS analyses for surface morphological details, functional groups and chemical states of maturated flocs eventually formed in the kaolin-HA water through both two-stage injection modes. Accordingly, newly-established conceptual models of ballasted floc growth were proposed to explore the potential influencing mechanisms of varying split ratios on the ballasted flocculation performance. At each sub-stage of maturation, an appropriate dosage ratio between PAM and microsand was of great importance to effectively incorporate microsand particles into ballasted floc formation, besides the hydrolyzed produces of AS coagulant formed at the coagulation stage of ballasted flocculation. This study is expected to provide valuable insights for making ballasted flocculation more effective, economical and sustainable in water treatment engineering.

Response Surface Optimization and Floc Structure Analysis of Magnetic Flocculation Technology for Anaerobic Digestion Reject Water

In order to improve the removal efficiencies of SS and Fe3+ in anaerobic digestion reject water for the subsequent biological treatment process, on the basis of the single factor test in the early stage, the response surface method was used, and the structure of the formed floc was analyzed by magnetic flocculation. The optimum amounts of magnetic powder, polyaluminum chloride (PAC) and polyacrylamide (PAM) were 40.51 mg/L, 31.31 mg/L and 4.05 mg/L, respectively. At this time, the removal efficiencies of SS and Fe3+ were 97.84% and 98.35%. The effects of floc particle size, scanning electron microscopy, infrared spectroscopy and two-dimensional fractal dimension of flocs on the flocculation ability showed that: compared with conventional coagulation, the average particle size of flocs treated by magnetic flocculation was 76.56 μm, the Fe-O-Al absorption peak appeared at 984 cm−1, the flocculation ability was significantly improved, the surface of the floc was rough and porous, and the structure was dense, and the sedimentation performance was significantly improved also.

Estimating the Shear Resistance of Flocculated Kaolin Aggregates: Effect of Flocculation Time, Flocculant Dose, and Water Quality.

The resistance of kaolin aggregates to shearing in water clarification and recovery operations is a critical input in designing thickener feed wells. A recently formulated but already available criterion is used to determine the shear strength of flocculated kaolin aggregates. The flocculant is a high molecular weight anionic polyelectrolyte. The resistance of the aggregates is evaluated as a function of flocculation time, flocculant dosage, and water quality. The determination is based on a standardized experimental method. First, the time evolution of the average size of kaolin flocs is measured when aggregates are exposed to incremental shear rates from a predetermined base value. Then, the results are fitted to a pseudo-first-order model that allows deriving a characteristic value of the shear rate of rupture associated with the upper limit of the strength of the aggregates. In seawater, at a given dose of flocculant, the strength of the aggregates increases with time up to a maximum; however, at longer times, the resistance decreases until it settles at a stable value corresponding to stable aggregates in size and structure. A higher flocculant dosage leads to stronger aggregates due to more bridges between particles and polymers, leading to a more intricate and resistant particle network. In industrial water with very low salt content, the resistance of the kaolin aggregates is higher than in seawater for the same dose of flocculant. The salt weakens the resistance of the aggregates and works against the efficiency of the flocculant. The study should be of practical interest to concentration plants that use seawater in their operations.

In-situ microscopy investigation of floc development during coagulation-flocculation with chemical and natural coagulants

ABSTRACT This study aimed to include in-situ microscopy in the analysis of floc development during coagulation-flocculation for drinking water treatment. To this end, jartest series were carried out for natural and synthetic waters using aluminum sulfate, sulfate chloride, ferric chloride, and Opuntia sp. as coagulants. Coagulation under optimized conditions was monitored by an in-situ microscope in conjunction with image analysis. Obtained results enabled some insights on the coagulation process. Images captured different stages of initial floc development, including flocs exhibiting heterogeneous, branched, and irregular surface structures. From image analysis, wide distributions of flocculated particle sizes were found for both natural (19–15834 μm) and synthetic water (19–21607 μm), suggesting the occurrence collisions by adhesion and transport between particles, thus influencing floc formation rates depending on the medium. Average size and number of flocs, as determined by the image analysis algorithm as a function of the time, showed inverse correlation of floc growth with water clarification. The microscopic images also illustrated how different coagulants in different water sources undergo breaking through fragmentation or erosion. Our findings also highlight the importance of investigating additional aspects that involve conditions of mixing, development, breaking, regrouping, and resistance of flocs.

Impact of pH on interaction between the polymeric flocculant and ultrafine coal with atomic force microscopy (AFM)

Solution pH plays a vital role in the interactions between polymer flocculants and solid colloid particles which determines the flocculation performance. The influence of solution pH on the flocculation of Magnafloc5250 on ultrafine coal suspension and the interactions between the polymers and solid surfaces were investigated by using atomic force microscopy (AFM), hydrodynamic radius measurements, particle floc size analyses, and flocculation dewatering tests. Our results revealed that the settling velocity and supernatant transmittance of ultrafine coal suspension without flocculant were low in alkaline conditions due to the strong repulsion, which agreed with the extended DLVO theory and direct force measurements by AFM. In the presence of Magnafloc5250, the highest transmittance was observed at pH 4 due to the compressed electric double layer among the coal particles, while the settling velocity reached a maximum value at pH 10, which was consistent with the largest average size of flocs (41.80 µm) and more extended conformation of flocculant. The adhesion force measured by AFM was divided into two parts: flocculation force and bare force. And the flocculation force reached a maximum value with the broadest range of action at pH 10, which supported the settling tests and floc size measurements well. The strengthened bridging interaction (e.g., via hydrogen bonding) with increasing solution pH is attributed to the increase of effective binding sites between more extended polymer chains and coal particles, thus contributing to an enhanced flocculation behavior.

Interactions between magnetic particles and polyaluminum chloride on the coagulation behavior in humic acid-kaolin synthetic water treatment

Understanding the interactions between magnetic particles (MPs) and polyaluminum chloride (PACl) is essential to elucidate the magnetic seeding coagulation (MSC) process. However, little is known about how MPs interact with the different Al species coexisting in the PACl. Here, the relationships among pollutants removal, residual Al distribution, and floc properties were comparatively studied in the MSC and traditional coagulation (TC) processes to address this issue. The response surface analysis indicated that the interaction between PACl and MPs dosages exhibited significant effects on turbidity and DOC removal. Negligible changes of dissolved Al after MPs addition indicated the weak connection between Ala and MPs. The formation of MPs-Alb-HA complexes resulted in the increase of turbidity removal from 90.2% to 96.0% and the reduction of colloidal Al from 0.67 to 0.30 mg L−1. Humic-like components could be adsorbed on MPs forming MPs-HA complexes, which enhanced the DOC removal from 55% to 58.5%. MPs addition produced loose flocs with a small floc fractal dimension value (1.74), so the average size and strength of flocs in the MSC process (425 μm and 49.7%) were lower than that in the TC process (464 μm and 58.3%). The cumulative volume percentage of large flocs (>700 μm) was decreased from 29.7% to 20.7% with MPs addition, indicating the disruption of large flocs and the reproduction of more fragments. The effective separation of these fragments by magnetic attraction maintained the efficient coagulation performance. This study provides new insights into the interaction mechanism of MPs and PACl in the MSC process.

Performance enhancement of a submerged forward osmotic membrane bioreactor by aerobic granules

Forward osmotic membrane bioreactors (FOMBRs) have broad prospects for application in wastewater reuse and resource recovery. Problems such as concentration polarization, salinity accumulation, and membrane fouling have led to a rapid flux decline, which limit their practical application. In this study, we introduced aerobic granules into a FOMBR to clarify the effect of sludge morphology on reactor performance. The results showed that the average removal efficiency of total organic carbon (TOC) and phosphate-phosphorus (PO43−-P) were maintained above 95% during the operation of activated sludge-based FOMBR (R1) and aerobic granule-based FOMBR (R2), respectively. R2 achieved higher removal efficiency of ammonia–nitrogen (NH4+-N) with an average removal efficiency of 77%, which was significantly higher than R1 (57%). Over the entire operation, the content of the soluble microbial products (SMP) in R1 was generally higher than that in R2, which resulted in more polysaccharides and proteins in the fouling layer of the membrane. The membrane flux of R2 was 1.0 L·m−2·h−1 higher than that of R1. The average size of flocs in R1 decreased from 137.24 μm to 76.84 μm, and that of the aerobic granules in R2 decreased from 2.35 mm to 1.20 mm and then remained stable. The results of microbial analysis showed that the membrane surface had lower microbial diversity than that of the mixed liquid, and the microorganisms enriched at the membrane surface were mainly Proteobacteria and Firmicutes with better salt tolerance.

The variation of flocs activity during floc breakage and aging, adsorbing phosphate, humic acid and clay particles

The mechanism of removal of humic acid, phosphate and kaolin particles by coagulation with alum and PACl or adsorption by their pre-formed precipitates was investigated, and it was found that the coagulation mechanisms for monomeric Al at neutral pH and polymeric Al13 at alkaline pH were very similar. The removal of phosphate and humic acid by coagulation with alum or PACl did not change with stirring time (between 1 min and 15 min), independent of the dose and species of coagulants. However, for adsorption of these impurities by pre-formed precipitates, the results were significantly different. Both Al3+ and nano-sized Al13 could precipitate and form aggregates at pH 7 and pH 9, respectively, and their precipitates became less active (fewer binding sites on the surface of precipitate) with the increase of shear time or shear rates before adsorbing pollutants. Thus, although the total surface area increased (the average size of flocs became smaller) at higher applied shear rates or longer shear time, the removal efficiency of humic acid and phosphate decreased. Also, from the MW distributions, it was confirmed that less humic acid was removed by the adsorption on alum precipitate pre-formed with longer shear time. Chemical groups (OH2and OH) on the surface of precipitate determined the removal efficiency of phosphate and humic acid, and the activity of precipitate become lower as a result of higher applied shear and longer shear time. This is confirmed be due to some crystallization of the amorphous precipitate, forming inactivated hydroxyl. When kaolin was added 10 min after the alum or PACl precipitate formed, the precipitates captured kaolin particles only on their surface, whereas when alum was added to kaolin suspensions particles were trapped within the growing flocs. When alum/kaolin flocs were broken at high shear rate re-growth of flocs decreased with increasing shear time, but after a short breakage period, long aging of broken flocs had little effect on floc regrowth.

Hydrolysis of polyaluminum chloride prior to coagulation: Effects on coagulation behavior and implications for improving coagulation performance

The effects of polyaluminum chloride (PACl) hydrolysis prior to coagulation on both the coagulation zone and coagulation performance of a kaolin suspension were investigated by a novel jar test named the “reversed coagulation test”. The tests showed that PACl hydrolysis prior to coagulation decreased the performance of charge neutralization coagulation in the case of short-time slow mixing (10min; G=15sec−1) and increased the optimal dosage for charge neutralization and sweep coagulation. Moreover, the hydrolysis time had insignificant effects on the size and zeta potential of PACl precipitates and the residual turbidity of the raw water. However, PACl hydrolysis prior to coagulation and the size of PACl precipitates had a negligible effect on the performance of sweep coagulation. The results imply that, in practice, preparing a PACl solution with deionized water, rather than tap water or the outlet water from a wastewater treatment unit, can significantly save PACl consumption and improve the performance of charge neutralization coagulation, while preparing the PACl solution with tap or outlet water would not affect the performance of sweep coagulation. In addition, the optimal rapid mixing intensity appears to be determined by a balance between the degree of coagulant hydrolysis before contacting the primary particles and the average size of flocs in the rapid mixing period. These results provide new insights into the role of PACl hydrolysis and will be useful for improving coagulation efficiency.

Effect of aluminum speciation on fouling mechanisms by pre-coagulation/ultrafiltration process with different NOM fractions

Ultrafiltration is an emerging technology for drinking water production, but the membrane fouling is still a challenge. This study was carried out to investigate the effect of aluminum speciation on UF membrane fouling behavior by different NOM fractions-humic substances and proteins, as represented by humic acid (HA) and bovine serum albumin (BSA), respectively. The interesting results showed that the total fouling resistance of the mixture of HA-BSA-kaolinite solution without coagulant demonstrated a slight decrease in comparison with those of the individually filtered substances, indicating a mitigatory fouling effect. The hydrolysis of aluminum products was various as pH and membrane fouling was related to aluminum speciation. The average size of flocs dramatically increased and fractal dimension of flocs decreased with the increasing of pH value independent on water quality, which indicated that aluminum speciation had a significant impact on floc properties. For the mixture of HA-BSA-kaolinte, the slightly larger of flocs average size in comparison with the individual organic fraction after coagulation was probably attributing that BSA was encapsulated by HA to enlarge the molecular length and floc size further increased. The membrane performance also showed that coagulation effluent of HA-BSA-kaolinite mitigated membrane fouling. The strong linear relationship was observed between flocs fractal dimension and final membrane flux in this research. From the results, the control of flocs fractal dimension should be considered as a new technique for traditional hybrid coagulation/ultrafiltration system, which resulted in minimized total and irreversible fouling and has a meaningful engineering application value.

Stochastic Flocculation Model for Cohesive Sediment Suspended in Water

Existing flocculation models for cohesive sediments are classified into two groups: population balance equation models (PBE) and floc growth models. An FGM ensures mass conservation in a closed system. However, an FGM determines only the average size of flocs, whereas a PBE has the capability to calculate a size distribution of flocs. A new stochastic approach to model the flocculation process is theoretically developed and incorporated into a deterministic FGM in this study in order to calculate a size distribution of flocs as well as the average size. A log-normal distribution is used to generate random numbers based on previous laboratory experiments. The new stochastic flocculation model is tested with three laboratory experiment results. It was found and validated with measured data that the new stochastic flocculation model has the capability to replicate a size distribution of flocs reasonably well under different sediment and carrier flow conditions. Three more distributions (normal; Pearson type 3; and generalized extreme value distributions) were also tested. From the comparison with results of different distribution functions, it is shown that a stochastic FGM using a log-normal distribution has a comparative advantage in terms of simplicity and accuracy.

Influence of flocs breakage process on membrane fouling in coagulation/ultrafiltration process—Effect of additional coagulant of poly-aluminum chloride and polyacrylamide

The additional coagulant dosing strategy was applied in a coagulation–ultrafiltration (UF) process for treatment of drinking water. When poly-aluminum chloride (PACl) dose was set as the optimal coagulant dosage (alum 0.15mM), the additional PACl and polyacrylamide (PAM) were added after breakage to further examine their benefits and impact on membrane fouling. The average size of flocs re-formed with additional non-ionic PAM was larger than that of additional PACl. The fractal dimension of flocs (dpf) was proved to play an important role on membrane fouling. Broken flocs re-formed with additional non-ionic PAM has more convoluted boundaries and lower effective density than the one with additional PACl, which is available to form loose and porosity cake layer on surface of membrane. Microscopic observations of membrane surface demonstrated that there was obvious correlation between the structure of cake layer and the fractal dimension of flocs. Besides, reversible fouling (RF) and irreversible fouling (IF) were also investigated in our study. The additional PACl dosing could lower membrane fouling, but irreversible fouling did not reduce because much more positive charged of flocs may adsorb on the surface of membrane under electrostatic attraction. As for non-ionic PAM, both of RF and IF were simultaneously decreased. As a result, the structural characteristics and surface electrical of flocs are two main mechanisms caused irreversible fouling.

Composition of extracellular polymeric substances in a partial nitrification reactor treating high ammonia wastewater and nitrous oxide emission

The objective of this study was to characterize the composition of extracellular polymeric substances (EPS) during the achievement of partial nitrification and subsequent nitrous oxide (N2O) emission treating high ammonia wastewater. After operation of 120days, the reactor achieved high ammonia removal efficiency and stable nitrite accumulation. The average size of sludge flocs in the reactor increased from 102.6 to 258.5μm. The main compositions of EPS, including protein (PN) and polysaccharide (PS), increased to 65.46±3.27 and 21.63±1.08mg/g VSS, respectively. Results of three-dimensional excitation-emission matrix spectroscopy implied that EPS transferred to tryptophan PN-like and humic acid-like substrates. N2O emission accounts for 11.67% of removed nitrogen during the steady state of partial nitrification reactor. The obtained results could contribute a better understanding the achievement of partial nitrification through the composition changes of EPS, and provide more information to determine nitrogen removal by considering N2O emission.

Effect of Enteromorpha polysaccharides on coagulation performance and kinetics for dye removal

The effect of Enteromorpha polysaccharides (Ep) as a new coagulant aid on coagulation performance and kinetics was investigated for synthetic dye wastewater treatment. Color removal efficiency was used to evaluate coagulation effects and meanwhile zeta potential was measured to analyze coagulation mechanism. Floc growth kinetics and re-growth characteristics were investigated by photometric dispersion analyzer (PDA) in terms of floc growth rate, size, strength and reversibility. Results showed that coagulation aid effect of Ep was notable. When Ep was used in combination with alum coagulants, the color removal could be apparently improved, and the optimal solution pH range was 6.0–8.0 for coagulation efficiency. The growth rate and average size of flocs formed by aluminum sulfate-Ep were larger than those by aluminum sulfate in steady-state after floc growth phase, and meanwhile the distribution of floc size had a wider range. In addition, the former was also stronger and had a better recoverability than the latter.

Study of Enteromorpha polysaccharides as a new-style coagulant aid in dye wastewater treatment

Enteromorpha is one of the common fouling green algae, which has brought serious environmental problems in past years. This study was aimed to apply it in water treatment process. Enteromorpha polysaccharides (Ep) were used as a new-style coagulant aid to assess its effect on coagulation behavior and floc characteristics. Color removal was used to evaluate coagulation effects and floc properties were investigated by Photometric Dispersion Analyzer (PDA). Results showed that when Ep was used in combination with aluminum chloride (AC), color removal could be apparently improved, and the optimal solution pH ranged 6.0–8.0. The growth rate and average size of flocs formed by AC–Ep were larger than those by AC in steady-state after floc growth phase, and meanwhile the distribution of floc size had a wider range. Besides, floc recoverability could be significantly improved when Ep was used as coagulant aid.

Investigation of the property of kaolin–alum flocs at acidic pH

The formation, breakage and re-growth of flocs were investigated using aluminium sulfate (alum) at acidic pH to explore the property of flocs and coagulation mechanism by different Al species formed in situ. The broken flocs can fully re-grow to the size before breakage, when monomeric Al or polymeric Al13 forms dominate the coagulant species. However, for relatively unstable small polymeric Al (Al3–Al8) and alum precipitate, the break-up process displayed a distinct irreversibility.There was a distinct increase of fractal dimension (Df) when the pH changed from 5.2 to 6.0, as well as the strength of flocs (the size of broken flocs), which meant that the flocs structure changed significantly. Also the Df of broken flocs was higher than that of flocs before breakage, and the Df values of re-grown flocs were not greatly different from the values before breakage. The number of small flocs after breakage and re-growth was much lower than that before breakage at acidic pH (pH 5.2) when the average size of flocs was chosen at the same value before breakage and after re-growth. An important finding is that the strength of flocs is strongly correlated with Df of flocs.

Effect of Al species in polyaluminum silicate chloride (PASiC) on its coagulation performance in humic acid–kaolin synthetic water

The relative importance of three different Al species (Ala, Alb and Alc) in polyaluminum silicate chloride (PASiC) was investigated in terms of coagulation removal efficiency and floc characteristic during the coagulation of humic acid–kaolin clay (HA–Kaolin) synthetic solution. Coagulation performance of PASiC (Al/Si molar ratio of 10:1) was also compared with that of polyaluminum chloride (PAC). Three PASiC coagulants with Ala, Alb and Alc dominant in one of them were denoted as PASiCa, PASiCb and PASiCc, respectively. The results showed that PASiCb performed the best turbidity and DOC removal, followed by PASiCc and PASiCa, while PASiCb and PASiCc removed UV254 more efficiently than PASiCa. At low dosages (1–4mg/L), flocs formed by PASiCb and PASiCc coagulation had higher zeta potential than those by PASiCa coagulation. However, order of zeta potential was PASiCa>PASiCc>PASiCb at high dosage of 18mg/L. Both coagulation efficiency and zeta potential were observed to be maximal at pH of 6. Zeta potential of formed flocs formed by PASiCc coagulation was the highest followed by PASiCb and PASiCa (in decreasing order) in pH range of 4–8, while PASiCa coagulation showed highest zeta potential followed by PASiCb and PASiCc at pH 9. The average size of flocs formed by three PASiC coagulants varied according to the following order: PASiCa>PASiCb>PASiCc. The floc equivalent diameters obtained by PASiCb coagulation were invariable with pH variation. Strength factors and recovery factors of flocs formed by PASiCc coagulation were larger than those by PASiCa and PASiCb coagulation within all investigated pHs.

Influence of flocs breakage process on submerged ultrafiltration membrane fouling

In order to explore the influence of floc characteristics on membrane fouling, flocs after breakage and re-growth were compared with those formed by conventional coagulation process when alum was used as coagulant. The results showed that, though the average size of flocs after breakage and re-growth was smaller than that without breakage, the fractal dimension of flocs after breakage and re-growth was lower than that without breakage independent of the dosage of alum. The increase of trans-membrane pressure (TMP) was lower when the flocs formed in breakage and re-growth process. When the dosage of alum increased, the fractal dimension ( D 2) of flocs without breakage decreased a little while D 2 of flocs with breakage and re-growth was nearly the same. Moreover, lower dosage of alum (0.1 mM) caused higher increase of TMP with conventional coagulation process, comparing with higher dosage of alum (0.3 mM); while it was nearly the same when the flocs formed with breakage and re-growth process. The fractal dimension is an important factor influencing the increase of TMP compared with the average size of flocs, and a lower frequency of small flocs may cause a smaller increase of TMP. The characteristics of flocs surface changed after breakage and re-growth and this may also affect the increase of TMP.

Effect of two-stage coagulant addition on coagulation-ultrafiltration process for treatment of humic-rich water

A novel two-stage coagulant addition strategy applied in a coagulation-ultrafiltration (UF) process for treatment of humic-rich water at neutral pH was investigated in this study. When aluminum sulfate (alum) doses were set at a ratio of 3:1 added during rapid mix stage and half way through flocculation stage, the integrated process of two-stage alum addition achieved almost the same organic matter removal as that of conventional one-stage alum addition at the same overall dose. Whereas membrane fouling could be effectively mitigated by the two-stage addition exhibited by trans-membrane pressure (TMP) developments. The TMP developments were found to be primarily attributed to external fouling on membrane surface, which was closely associated with floc characteristics. The results of jar tests indicated that the average size of flocs formed in two-stage addition mode roughly reached one half larger than that in one-stage addition mode, which implied a beneficial effect on membrane fouling reduction. Moreover, the flocs with more irregular structure and lower effective density resulted from the two-stage alum addition, which caused higher porosity of cake layer formed by such flocs on membrane surface. Microscopic observations of membrane surface demonstrated that internal fouling in membrane pores could be also remarkably limited by two-stage alum addition. It is likely that the freshly formed hydroxide precipitates were distinct in surface characteristics from the aged precipitates due to formation of more active groups or adsorption of more labile aluminum species. Consequently, the flocs could further connect and aggregate to contribute to preferable properties for filtration performance of the coagulation-UF process. As a simple and efficient approach, two-stage coagulant addition strategy could have great practical significance in coagulation-membrane processes.

Breakage and re-growth of flocs formed by charge neutralization using alum and polyDADMAC

The formation, breakage and re-growth of flocs were investigated using alum and polyDADMAC to explore the reversibility of floc breakage. There is a significant reversibility of the breakage process, i.e. the broken flocs can re-grow to the size before breakage, when charge neutralization dominates the coagulation mechanism. However, for higher alum dosage, the break-up process displayed a distinct irreversibility. When coagulated in charge neutralization, the re-growth process of alum was nearly the same as that of polyDADMAC. The average size, coagulation rate and fractal dimension of flocs before and after breakage were nearly the same, including alum and polyDADMAC. While at higher alum dosage, the average size, coagulation rate and fractal dimension of flocs after breakage were much lower than that before breakage. Most important is that the number of small flocs after breakage and re-growth was much less than before breakage when charge neutralization dominated the coagulation mechanism. On the contrary, at higher alum dosage, the small flocs, after breakage and re-growth, increased. The fractal dimension of flocs with alum increased as coagulation time increased until a limiting floc size was reached, while for higher alum dosage, it decreased, whether before or after breakage. The determining parameter for floc re-growth is probably not the fractal dimension, but rather the chemical characteristics of the flocs surface.