Composite flocculants based on magnesium salt–polydiallyldimethylammonium chloride: characterization and flocculation behaviour

Abstract

This study describes preparation, characterization and examines the flocculation characteristics of novel magnesium salt polydiallyldimethylammonium chloride (polyDADMAC) composite flocculants. Five different compositions have been prepared by physical blending between synthesized polyDADMAC and magnesium salt. Fourier transform infrared spectroscopy (FTIR), intrinsic viscosity and zeta potential measurements were conducted to characterize the properties of the composite flocculants produced. Meanwhile jar tests and photometric dispersion analyses (PDA) were carried out to correlate the flocculation behavior and flocs aggregation. The intrinsic viscosity and zeta potential of the composite flocculants increase with increasing polyDADMAC composition. Composite flocculants (sample B and C) show better turbidity removal compared to polyDADMAC alone. In this system, magnesium salt acts as a secondary destabilizer, while polyDADMAC as a primary destabilizer also enhances the bridging mechanism, thus improving the aggregating capacity.

---

1. Introduction

Flocculation plays an important role in wastewater treatment. Increasing demands for greater quantities and high quality from discharged wastewater has encouraged researchers to develop new categories of flocculants such as inorganic–organic polymeric flocculants. These inorganic–organic polymeric flocculants are one of the alternatives to optimize the flocculation process. This category is also known as composite flocculants.

Generally, common inorganic-based flocculants such as aluminum and iron-based substances are widely used in water treatment applications where the source of cationic charge is required. Several studies have examined composite flocculants incorporating aluminum and iron salts.^1–4 This incorporation has been shown to improve flocculation behavior. In this study, magnesium salt is proposed as there are limited literatures available regarding this. According to Semerjian and Ayoub (2003),⁵ the efficacy of water treatment using magnesium compounds dates back to the late 1920s. They claimed that, magnesium salt do achieved significant reduction in the Total Organic Content (TOC) and also reduced the degree of light absorbance in water caused by the presence of suspended solids. Moreover, the price is low which is in the range of RM 0.30 to RM 0.40 per kilogram. This will contribute reduction of price since synthetic organic polymeric flocculants are expensive.

PolyDADMAC is widely use in wastewater treatment, since it is very effective in flocculating, discoloring, and removing organics such as humus and algae.^6,7 PolyDADMAC is a high-charge-density cationic polymer,⁸ which makes it well-suited for the flocculation process. Moreover, it is pH-insensitive as well as chlorine-resistant. Thus, polyDADMAC has been chosen as one of the components for the hybrid composite flocculants used in this study. However the price is quite expensive which it is in the range of RM 3.99 to RM 5.44 per liter.

Composite flocculants have been emerged as it poses better performance compared to conventional flocculants.⁹ It is believed the performance depends on the aggregating power of composite among inorganic- and organic-based flocculants.¹⁰ Since the magnesium salt and polyDADMAC is considered as a new composite flocculants, thus, a particular emphasis was given to the flocculation mechanism and kinetics. The kinetics were examined by online photometric dispersion analyzer since PDA offers a symptom of changes in aggregation (flocculation) or disaggregation (dispersion, de-flocculation).¹¹

2. Experimental

2.1 Materials

DADMAC monomer (65%) was purchased from the Sigma-Aldrich. Tetrasodium ethylenediaminetetraacetate, known as Versene (Na₄EDTA) in solid form, was also purchased from the Sigma-Aldrich. Versene was used as a chelating agent for the polymerization. Ammonium persulfate (APS) in solid form was purchased from the Riedel-de Haen Chemical Company to use as a free radical initiator. Sodium metabisulfite (MBS) was also purchased from the Riedel-de Haen Chemical Company. The solid MBS was used as a free radical scavenger agent. Magnesium chloride from Bendosen was used as an inorganic flocculants to incorporate with polyDADMAC. Distilled water was used to prepare the polymer solutions. All reagents were used without any further purification.

2.2 Preparation of polyDADMAC

First, 1.87 M of 65% DADMAC monomer and 10 mL of deionized water and 0.20 g of Na₄EDTA were charged into a polymerization reactor which was equipped with a condenser, a thermometer, and an overhead agitator. Then, the polymerization mixture was purged with nitrogen gas for 45 min; at the same time, the mixture was stirred at 80 rpm while maintaining a temperature of 60 °C. An aqueous solution of 1.50 g of APS was prepared in 30 mL of deionized water. The aqueous solution was slowly charged into the polymerization mixture drop wise via a syringe pump at 0.07 mL min⁻¹. This feeding process was carried out for 300 min. The reaction temperature was held at 80 to 85 °C, while the reaction mixture was stirred at 110 to 130 rpm for 10 min.

Next, an aqueous solution containing 0.30 g of MBS was prepared in 5 mL of deionized water. The aqueous solution was fed into the reactor dropwise via a syringe pump at 0.1 mL min⁻¹. The reaction mixture was then cooled to room temperature for 15 h and stored in an airtight container. The whole experiment was conducted at pH of 7 to 9. Thus, in order to keep the crosslinking reaction at a desired pH during the course of the additional initiator feed, NaOH solution was added to the polymerization mixture.

2.3 Preparation of magnesium salt–polyDADMAC

Aqueous solutions of composite polymers (MgCl₂–polyDADMAC) were prepared with the desired compositions as described in Table 1. The concentration of MgCl₂ while blending with polyDADMAC was 10 000 ppm. The aqueous solutions of blending were stirred at 250 rpm for about 4 h. It were then allowed to age for 24 h at room temperature prior to characterization.

Table 1 Blending composition of polyDADMAC–MgCl₂ hybrid-based flocculants

Sample	Compositions (MgCl2 : polyDADMAC)
A	10 : 90
B	30 : 70
C	50 : 50
D	70 : 30
E	90 : 10

---

2.4 Characterization of polyDADMAC and composite flocculants

2.4.1 Structure analysis. The structure of monomer, polyDADMAC and different composite flocculants compositions were scanned by Perkin-Elmer system 2000 FT-IR spectroscopy. All samples were in slurry forms.

2.4.2 Molecular weight analysis. Molecular weight of polyDADMAC was analysis by using Agilent gel permeation chromatography (GPC) model 1200 infinity. This machine includes with PL aquagel –OH mixed, 8 μm (particle size) and 7.5 (inner diameter) × 300 mm (length). High performance liquid chromatography (HPLC) grade water was used as the solvent, with a flow rate at 1.0 mL min⁻¹. The values were calibrated with polystyrene standard to establish the molecular weight distribution (MWD) and polydispersity index (PDI).

2.4.3 Intrinsic viscosity determination. Intrinsic viscosity was determined using RPV-1 at 25 °C and 1 M NaCl as a solvent. The ubbelohde tube was equilibrated in the viscometer bath at the vertical position for 10 minutes at 25 °C. All data taken were calculated using Hagenbach correction.

2.4.4 Zeta potential measurements. Zeta potential of polyDADMAC and blending samples were determined by using a Malvern zetasizer nano series machine. The concentration of all samples is 1000 ppm.

2.5 Preparation of kaolin suspension

Ten grams of kaolin were dispersed in 500 mL of distilled water. To obtain full dispersion, it was necessary to raise the pH of the suspension to about 7 by adding 1 mL of 0.1 M NaOH. After blending at 600 rpm for 3 h, the clay suspension was diluted to 1 L with distilled water and allowed to stand overnight. The top 800 mL was decanted, and its solids content was determined gravimetrically as 2 g L⁻¹.

2.6 Flocculation test

For the flocculation tests, 50 mL of stock suspension were diluted with 450 mL of deionized water. The flocculation test of kaolin turbid water was carried out in a standard flocculation jar apparatus (Jar Test FC6S, Velp Scientifica). The beakers were placed on the flocculator and stirred at 200 rpm for 30 min. Immediately after the addition of the flocculants, the suspensions were stirred at a fast mix at 200 rpm for 2 min, slow stirred at 45 rpm for 10 min, and then settled for 30 min. Different pH of kaolin suspension for study of effect on flocculation behavior were prepared by using 0.1 M of HCl and NaOH.

2.7 Analytical test

After the suspension was settled for 30 min, 10 mL of supernatant samples were withdrawn for turbidity measurements. The samples were injected into the glass tube. Then, the liquid tested was measured using a HACH spectrophotometer model DR/2000.

2.8 Kinetic study of flocculation process

The photometric dispersion analyzer (PDA 2000, Rank Bros Ltd., Cambridge), UK monitored the analysis of flowing suspensions. The PDA was connected via flexible tubing to one jar during testing. A Masterflex® peristaltic pump circulated the sample water at 20 mL min⁻¹. The pump was located after the PDA to avoid deterioration of the flocs (Fig. 1). The flocculation index (FI) graphs were constructed based on the ratio results obtained. The FI graphs were structured into three different phases; growth phase, plateau and settling. Different parameters were introduced to extract useful flocculation information from the FI graphs. The gradient of the first minute will be referred to as initial floc aggregation (IFA). Another parameter was variance. It is an indicator of flocs break-up and is related to flocs size distribution. It is based on the fluctuations of the FI data after the floc has reached a steady state plateau. The plateau was determined by the averaged FI values during the last 5 minutes of slow mixing. The variance is calculated according to the following equation

Fig. 1 Schematic diagram of instrument arrangement for PDA assessment.

3. Results and discussion

3.1 Characterization of polyDADMAC

Fig. 2 shows the differences in the infrared spectra of the DADMAC monomer and the radical synthesized polyDADMAC. The synthesized polyDADMAC was characterized by FTIR analysis to ensure that it contained only carbon single bonds instead of double bonds. The band at 959.00–876.59 cm⁻¹ is attributed to the double bond =CH₂ wag vibration in the molecular structure of DADMAC. It is indicates that DADMAC was successfully converted into polyDADMAC since there is no peaks appeared in this region for polymer infrared spectra.

Fig. 2 Infrared spectra for DADMAC monomer and polyDADMAC.

The molecular weight distribution (MWD) of polyDADMAC produced is shown in Fig. 3. Based on Fig. 3, the polyDADMAC produced showed unimodal profiles with the M_w is in the range of 160 000 g mol⁻¹. The PDI (M_w/M_n) for this polymer is 1.06, indicates that the molecular weight distribution is relatively narrow. This polyDADMAC almost similar with polyDADMAC synthesized by Wandrey et al. (1999) with the Mn is 170 000 g mol⁻¹ (PDI is 1.5).⁶

Fig. 3 Molecular weight distribution of synthesis polyDADMAC.

3.2 Characterization of magnesium salt–polyDADMAC

3.2.1 Structure analysis. The compositions of the magnesium salt–polyDADMAC composite polymers were determined by FTIR. Fig. 4 shows the spectra of the magnesium salt–polyDADMAC composite polymers. In the spectra, 3365.57–3330.77 cm⁻¹, 1638.02–1637.60 cm⁻¹ and 1474.48–1474.29 cm⁻¹ were observed for –OH stretch, −CH asymmetrical bending, and –CH in-plane bending, respectively.¹² Clearly, 1474.48–1474.29 cm⁻¹ increases in transmittance percentage as the polyDADMAC concentration increases. This indicates that –CH₂ in-plane bending in the polyDADMAC dominates as the concentration increases. The FTIR spectra indicates that no new chemical bond was created when magnesium salts were blended with polyDADMAC.¹³

Fig. 4 Infrared spectra for magnesium salt–polyDADMAC.

In the magnesium salt–polyDADMAC composite polymers, magnesium chloride was completely miscible with polyDADMAC. Magnesium chloride dissociated into Mg²⁺ and Cl⁻ ions when it blended with polyDADMAC. This suggests that Cl⁻ ions approach the electronegative nitrogen atom of polyDADMAC, while Mg²⁺ appears as a free ion.

3.2.2 Intrinsic viscosity. Fig. 5 shows the effect of composition on the intrinsic viscosity. Note that the intrinsic viscosities of the hybrid increase with the increasing percentage of polyDADMAC. The coulomb forces between cations increase with increasing of the polymer thus enhances chains become more expanded. This expanded subsequently increase the intrinsic viscosity.¹⁴

Fig. 5 Intrinsic viscosity for polyDADMAC and composite flocculants at different compositions.

The intrinsic viscosity results also can be explained by loading of MgCl₂. MgCl₂ is a multivalent salt with a higher concentration of ions (2⁺). Increase loading of MgCl₂ enhanced repulsion energy between salt and polymers and also between polymers. This contributes to reducing of polymer chain expanded thus further reduced the intrinsic viscosity. Salts having high solubility and surface area were preferentially selected to reduce the viscosity.¹⁵

3.2.3 Zeta potential measurement. The zeta potential value for MgCl₂, polyDADMAC and blended samples are shown in Fig. 6. Based on Fig. 6 it shows that incorporating MgCl₂ into polyDADMAC has reduced the values of zeta potential from 42.70 mV into 34.10 mV (sample A). This is contributes from reduction of repulsion magnitude in the sample. The decreasing of repulsion might be due to less cationic charge bring by MgCl₂ as the zeta potential values is only 2.43 mV. The reduction of zeta potential value further extends with increasing of MgCl₂ potion until the composition become 50 : 50 (sample C). The zeta potential value increase from 28.60 mV (sample B) to 41.30 (sample C). The value of zeta potential for sample C is much similar with polyDADMAC as the different is 1.40 mV only. It is believe the repulsion of Mg²⁺ with each other become stronger and it also involve the repulsion between positive charges of polyDADMAC. Increase of repulsion energy in the sample enhances the zeta potential value become higher.

Fig. 6 Zeta potential measurement of polyDADMAC, magnesium salt and composite flocculants at different compositions.

3.3 Flocculation test

3.3.1 Effect of dosage. Generally, the flocculants dosage depends on the content of colloids or suspended solids in the fluid. Treatment efficiency is assumed to increase with the increasing input of flocculants. Increments in the flocculants dosage are expected to encourage charge neutralization. However, over-dosage can lead to reversal of the particle surface charge caused by the excess of inorganic coagulant, resulting in particle re-stabilization which in turn decreases treatment efficiency.

As seen in Fig. 7(a), all samples show an increment of turbidity removal when increasing the dosage to 0.6 ppm and then decrement of turbidity removal when increasing dosage to 1 ppm except for MgCl₂, A and E. This can be further explained by looking at the composition effects in these composite flocculants. For sample A, which consists of 10% magnesium salt and 90% polyDADMAC, the optimum dosage was sustained at 0.4 ppm; for sample E, which consists of 90% magnesium salt and 10% polyDADMAC, optimum dosage was sustained at 1.0 ppm or higher. In other words, higher magnesium salt content (above 70%) in a composite flocculants may require more input of dosages in order to obtain similar to traditional flocculation performance.

Fig. 7 Effect of composite flocculants dosage on turbidity removal.

The incorporation of magnesium salt and polyDADMAC induced better surface charge neutralization in suspension. All blended samples show turbidity removal improvement compared to MgCl₂ which is in the range of 75–90%. In addition, sample B, C and D show similar or even higher of removal compared with polyDADMAC (Fig. 7(b)). The proposed mechanism of MgCl₂, polyDADMAC and composite flocculants samples are shown in Fig. 8.

Based on Fig. 8, polyDADMAC have both mechanism which are charge neutralization and bridging while MgCl₂ only have charge neutralization. This is reason to turbidity removal of kaolin suspension by MgCl₂ was not good as polyDADMAC. Incorporation of MgCl₂ about 30–70% (sample B, C and D) into polyDADMAC system achieved similar and higher performance compared to polyDADMAC. It is indicates that MgCl₂ acts as secondary destabilizer instead of polyDADMAC and the particles destabilized (by MgCl₂) been adsorb by polyDADMAC chains. It also showed that MgCl₂ can reduces dependency on the synthetic polymer, which is highly expensive.¹⁶ It was also observed that sample C shows better flocculation performance in the range of dosage 0.2 ppm to 1.0 ppm compared to others. It showed similar turbidity removal value within the determined range of dosage. Due to the synergetic effect of two distinctive components in one hybrid matrix, the dosage needed to achieve maximum treatment efficiency can be reduced. This is useful in domestic application, where flocculation performance depends on dosage effects. Thus, sample C and dosage at 0.6 ppm are chosen for further investigation on effects of pH.

Fig. 8 Proposed mechanism of MgCl₂, polyDADMAC and composite flocculants.

3.3.2 Effect of pH. Fig. 9 shows optimum flocculation occurred at pH 7. This is supported by Duan and Gregory,¹⁷ who found that the effective charge neutralizing species may be the positively charged colloidal particles at pH of less than 8. When acid is added to the suspension, it dissociates and releases a hydrogen ion which carries out the positive charge. Since kaolin has a negative charge, the hydrogen ion from acid tends to destabilize the charge, which leads to floc formation. Hence, adding acid to the kaolin suspension promotes flocculation. However, as addition of the hybrid material may promote charge correlation in the water system, excessive charge encourages repulsion and thus reduces the flocculation performance in this case.

On the other hand, an alkali dissociates in the suspension to form hydroxide ion, OH⁻. This hydroxide ion has a negative charge, which can assist in maintaining the charge stability by increasing the repulsion force, since the kaolin is already in the negative charge state. To destabilize the charge, a higher dosage of cationic or a high positive concentration is needed to counteract or neutralize the high negative charge due to the addition of alkali. Generally, it also can be concluded that pH does not effect much the flocculation performance where low turbidity values are observed in the range of 99.18% to 99.64%. Obviously, this is useful in real applications since the flocculation performance of hybrid materials is independent of the pH of the kaolin suspension.

3.4 PDA assessment

Flocculation kinetic was assessed at 0.6 ppm and is shown in Fig. 10. All samples showed sharp increment during the first few minutes after the addition of flocculants. The gradient of the slope during the first few minutes (30–40 s) was used to calculate the initial flocculation index. The addition of flocculants induced surface charge neutralization by encouraging surface charge reduction. This promoted agglomeration and induced flocculation in the process. During fast mixing, perikinetic flocculation took place where it shows rapid increment in the flocculation index. This indicates the floc growth in the initial stage. The fast agitation induced Brownian motion and encouraged the particles to progressively coalesce. Then, orthokinetic flocculation arose due to the velocity gradient; thus, floc formation occurred. The last five minutes of slow stirring (420 to 720 s) contributed to the steady plateau region, which shows further variance. Both parameter can be observed and discussed further by assessing Fig. 11.

Fig. 9 Effect of kaolin suspension pH on turbidity removal.

Fig. 10 Flocculation index of polyDADMAC and samples A, B, C, D and E at 0.6 ppm.

From Fig. 11, it can be observed that all samples do have similar initial floc aggregation values; polyDADMAC has 0.149, sample A has 0.160, sample B has 0.147, sample C has 0.155, sample D has 0.150 and sample E has 0.120. Perikinetic flocculation due to Brownian movement encourages colloidal particles to overcome the potential energy barrier between them. This destabilization stage is necessary to induce primary particles to approach and to progressively form larger agglomerates.

Sample A shows the highest value of initial floc aggregation, which indicates the highest growth rate of flocs among all tested samples. The addition of magnesium chloride at 10% weight shows synergetic effects where additional chargeable ions encourage neutralization mechanism in floc growth during fast mixing.

Sample E shows the lowest value, which indicates the lowest growth rate of flocs compared to the other samples. This might be due to the extra chloride ions contributed by the magnesium salt, which can reduce the flocculation performance by having charge repulsion and can cause antagonistic effects.

Fig. 12 shows that the variance value changes for different compositions of flocculants: 0.010 for polyDADMAC, 0.007 for sample A, 0.009 for sample B, 0.008 for sample C, 0.011 for sample D, and 0.006 for sample E. It also can be observed that the turbidity results for different samples at dosage 0.6 ppm correspond to the variance results. According to Staaks et al. (2011), variance can be used to assess floc structural differences.¹⁸ High variance shows larger flocs and better floc distribution, leads to higher turbidity removal values where better flocculation performance occurs, and vice versa. The decrement in variance from polyDADMAC to sample A also contributed to the decrement in turbidity removal. This correlation proves the validity of the variance calculation in the flocculation application.

Fig. 11 Initial floc aggregation of polyDADMAC and samples A, B, C, D and E at 0.6 ppm.

4. Conclusion

Composite flocculants based on magnesium salt–polyDADMAC with five different compositions have been successfully prepared and characterized. Elucidation of the FTIR spectra indicates that no new chemical bond was created when magnesium salts was blended with polyDADMAC. The composite flocculants become viscous with increasing of polyDADMAC composition. The aggregating of flocs occurred more in composite flocculants (sample B and D) compared to single polyDADMAC based on turbidity removal and PDA results. Moreover, pH did not significantly affect the flocculation performance of this composite flocculants. Based on PDA results, it is suggested that additional chargeable ions (MgCl₂) with certain level in composite flocculants encourage more aggregating capacity in floc growth.

Fig. 12 Variance and turbidity removal of polyDADMAC and samples A, B, C, D and E at 0.6 ppm.

Acknowledgements

M. A. A. Razali and M. R. H. Rosdi gratefully acknowledge the personal financial support under the MyBrain15 Scheme provided by the Malaysian Ministry of Education. A. Ariffin would like to thank the MOSTI Science Fund (1001/pbahan/814131) for their support.

References

1. N. D. Tzoupanos and A. I. Zouboulis, Water Res., 2011, 45, 3614–3626.
2. P. Moussas and A. Zouboulis, Sep. Purif. Technol., 2012, 96, 263–273.
3. J. Wei, B. Gao, Q. Yue and Y. Wang, Chem. Eng. J., 2009, 151, 176–182.
4. K. L. Yeap, T. T. Teng, B. T. Poh, N. Morad and K. E. Lee, Chem. Eng. J., 2014, 243, 305–314.
5. L. Semerjian and G. Ayoub, Adv. Environ. Res., 2003, 7, 389–403.
6. C. Wandrey, J. Hernandez-Barajas and D. Hunkeler, in Radical Polymerisation Polyelectrolytes, Springer, 1999, pp. 123–183.
7. Y. Zhang, X. Zhao, X. Li, C. Liu and L. Zhu, Water Sci. Technol., 2010, 62, 330–339.
8. A. Ariffin, M. Razali and Z. Ahmad, Chem. Eng. J., 2012, 179, 107–111.
9. Z. Yang, X. Lu, B. Gao, Y. Wang, Q. Yue and T. Chen, J. Mater. Sci., 2014, 49, 4962–4972.
10. K. E. Lee, T. T. Teng, N. Morad, B. T. Poh and Y. F. Hong, Sep. Purif. Technol., 2010, 75, 346–351.
11. B. Moudgil and P. Somasundaran, Flocculation, sedimentation and consolidation, New York, 1986.
12. F. L. I. Xamena, C. O. Areán, S. Spera, E. Merlo and A. Zecchina, Catal. Lett., 2004, 95, 51–55.
13. B.-Y. Gao, Y. Wang, Q.-Y. Yue, J.-C. Wei and Q. Li, Sep. Purif. Technol., 2007, 54, 157–163.
14. W. M. Kulicke and C. Clasen, Viscosimetry of polymers and polyelectrolytes, Springer Verlag, 2004.
15. A. Ariffin, M. S. Musa, M. B. H. Othman, M. A. A. Razali and F. Yunus, Colloids Surf., A, 2014, 441, 306–311.
16. T. Tripathy and B. R. De, J. Phys. Sci., 2006, 10, 93–127.
17. J. Duan and J. Gregory, Adv. Colloid Interface Sci., 2003, 100, 475–502.
18. C. Staaks, R. Fabris, T. Lowe, C. W. Chow, J. A. van Leeuwen and M. Drikas, Chem. Eng. J., 2011, 168, 629–634.

---