X-ray Photoelectron Spectroscopy

The X-ray photoelectron spectroscopy (XPS) was applied to determine surface composition of Al-flocs to a depth of less than 10 nm. XPS was performed using an X-ray photoelectron spectrometer, with a monochromatized K X-ray beam at 3.8 kW was generated from an Al rotating anode. All samples were stored in a nitrogen atmosphere to prevent atmospheric contamination and oxidation. Each analysis started with a survey scan in the binding energy rang from 0 to 1000 eV at a step of 1 eV. The binding energies of the photoelectrons were calibrated by the aliphatic adventitious hydrocarbon C (1s) peak at 284.6 eV.

CHAPTER IV

EFFECT OF Al SPECIES TRANSFORMATION ON COLLOID DESTABILIZATION MECHANISMS

In this section, standard jar tests were carried out to evaluate the efficiency of turbidity removal. First, three types of PACl coagulants, namely PACl-C, PACl-E and PACl-Al13, were used to coagulate the suspended kaolin particles. The effects of pH on coagulation were also evaluated. Zeta potential and residual turbidity for each coagulant at 1 mg/L were determined for this evaluation. In addition, three PACl coagulants were characterized by Ferron as well as ²⁷Al-NMR methods, and the effects of pH on Al species transformation was examined to evaluate the destabilization mechanisms of various PACl coagulants. On the other hand, the effects of dosage on particle destabilization mechanisms were further investigated through the analysis of the reactive Al species within flocs by solid-state ²⁷Al-NMR.

4.1 Effects of pH on Coagulation

The pH of solution principally influences the surface charge of particles such as zeta potential that rules the stability of particles in suspension system. Likewise, the Al species and their surface charge are significantly affected by pH values of solution, which leads to the variation in destabilization mechanism of particles. Literatures have indicated that pH change may have a profound impact on PACl coagulation as a result of motley hydrolyzed Al species (Wang et al., 2002; Hu et al., 2006).

Therefore, the Al species distributions of various PACl coagulants applied in this study were determined. Several coagulation experiments were conducted to study the effects of pH on the Al species transformation as well as coagulation efficiency.

4.1.1 Characterization of Coagulants

The results of Al speciation for three coagulants used in this study are summarized in Table 4.1. By Ferron method, the major Al species of PACl-C are the Ala

(monomeric Al) and Alc (colloidal Al(OH)3), while the majorities of Al species in PACl-E and PACl-Al13 are found to be Alb. The distributions of Alm and Al13

measured by ²⁷Al-NMR correspond to the distributions of Ala and Alb obtained by Ferron method, respectively. Although there is a little discrepancy between the composition of Alb and Al13, both methods has confirmed that PACl-E contains high contents of Al13 exceeding 60% of total Al concentration, and the PACl-Al13

possesses about 96 % Al13 with a little amount of Alu. Other studies also have indicated that Al13 can be roughly represented by Alb (Parker and Bertsch, 1992; Liu et al., 1999; Shi et al., 2007). With increasing the contents of polymeric Al, the ratio of Al13/Alb closes to 1, as indicated in Table 4.1, and there are more Al13 amounts in PACl prepared at higher γ, as evidenced in Fig. 4.1. Therefore, the Ferron method can be used to estimate the Al speciation of various PACl coagulants in further studies.

Table 4.1 Characteristics of coagulants by Ferron assay and ²⁷Al-NMR methods Ferron assay (Al %) ²⁷Al-NMR (Al %) ratio Coagulant AlT (M)
γ pH

Ala Alb Alc Alm Al13 Alu Al13/Alb

PACl-C 0.04 1.4 3.8 42.3 8 49.7 38.4 6.8 54.8 0.85

PACl-E 0.04 2.1 4.06 19 65.8 15.2 16.6 60 23.4 0.91

PACl-Al13 0.05  4.7 3.7 96.3 0 0 95.8 4.2 0.99

γ^{: [OH-/Al] ratio}

0 -20 40 20

80 60 100

(ppm)

Fig. 4.1 ²⁷Al-NMR spectra of (a) PACl-C (b) PACl-E (c) PACl-Al13. (a)

(b) (c) Al13

Alm

4.1.2 Effects of pH on Turbidity Removal

The coagulant dosages for 50 NTU turbidity water coagulated by PACl-C, PACl-E and PACl-Al13 are held at 1 mg/L, which are applied subsequently in the jar tests to evaluate the optimum pH of coagulation for turbidity removal. The changes of residual turbidity along with zeta potentials during coagulation at various pH values are shown in Fig. 4.2. A marked difference between the coagulation by three coagulants is observed. For PACl-C coagulation, the efficient pH region ranges from 7 to 10. The pH for the optimum turbidity removal coincides with the pH at which the zeta potential closes to zero. The efficient turbidity removal also occurs at pH 7.5 and pH 9, where the zeta potential of particles is around 10 mV and -20 mV, respectively.

For PACl-E coagulation, the optimum turbidity removal has been obtained when the pH is at around 9 where the zeta potential closes to zero, which is due to charge neutralization. However, the turbidity removal is also obviously effective when pH is higher than 10 where the relatively negative zeta potential is found, which implies the particles could be destabilized by electrostatic patch. Similarly, the efficient turbidity removal for PACl-Al13 coagulation also occurs when pH is higher than 9.

Since the coagulation performance is directly affected by the effective Al species, the effects of pH on coagulation can be explained by its effects on the transformation of Al species.

(a) PACl-C

4.1.3 Effects of pH on Al Speciation in Coagulation

Al speciation of PACl-C, PACl-E and PACl-Al13 at various pH values, as measured by Ferron assay, is shown in Fig. 4.3, respectively. For PACl-C, the Ala

drops sharply in the range of pH 4 to 5. Very little monomeric Al is detected at neutral pH, but concentrations then rise rapidly at higher pH. The disappearance of Ala at neutral pH is compensated by the rise in Alb and Alc, in which the Alc is substantially more than Alb. By contrast, pH has little effect on any of the three species of PACl-E. The most dramatic observation is that the Ala of PACl-E stays very low, especially at neutral pH, while the Alb content is quite high, ranging from 60 to 70%. Only a little Ala of PACl-E transforms into Alb at neutral pH. In addition, Alc content of PACl-E decreases slightly in the basic pH region. Likewise, PACl-Al13 intrinsically remains the extremely high Alb content at acidic pH except that very little Alb transforms into Ala at alkaline pH as a result of dissolution of Al13.

Study has indicated that a discrepancy between the stability of the Alb species pre-formed before the dosing and those formed in-situ after dosing, suggesting different Alb structures (Wang et al., 2004). Although the maximum Alb is observed at neutral pH for PACl-C, the in-situ formed Alb species is metastable, and thus probably transforms into Alc or another Al species immediately after dosing (Hu et al., 2006). By contrast, the pre-hydrolyzed Alb in PACl-E as well as PACl-Al13 remains more stable with varying pH except the content of Alb slightly drops down when pH is higher than 8. With the increase in pH, the Al13 becomes unstable, and thus the Al13 transforms into other larger Al species such as Al13 aggregates or precipitates at alkaline pH (Van Benschoten and Edzwald, 1990). However, the Al13 aggregates could not be identified clearly by Ferron method. Thus, the amount of Alb could not reflect the real stability of Al13 at alkaline pH.

(b) PACl-E pH values under the concentration of 2×10^-4 mol Al/L.

4.1.4 Effects of Al Speciation on Particle Destabilization Mechanisms

Previous studies have indicated that Al(III) species can transform into voluminous amorphous aluminum hydroxide (Al(OH)3(am)) at neutral pH (Duan and Gregory, 2003;

Van Benschoten and Edzwald, 1990). Large amounts of Alc formed at neutral pH for PACl-C, indicating the formation of significant amount of aluminum hydroxide (Al(OH)3(am) and colloidal Al(OH)3(s)), and consequently the occurrence of precipitation. In addition, because the in-situ formed Alb that occurs at neutral pH can quickly transform into Alc, Alc dominate over PACl-C coagulation at neutral pH.

As a result, enmeshment or sweep flocculation by aluminum hydroxide is responsible for the turbidity removal by PACl-C at neutral pH, even though the charge reversal of particles occurs at such condition, as seen in Fig. 4.2 (a). At alkaline pH, the particles could be destabilized by electrostatic patch of Al(OH)3 precipitates due to adsorption of precipitates, and then the growth of flocs is formed through precipitation charge neutralization (Dentel, 1988; Ye et al., 2007). Therefore, the particles are destabilized effectively at alkaline pH even though the zeta potential of particles is negative. However, the positive charge of aluminum hydroxide precipitates becomes gradually neutralized with increasing pH, and then the influence of electrostatic patch lessens, which results in the increase of residual turbidity with pH.

Because the Alb content remains relatively stable throughout the entire pH range studied, as shown in Fig. 4.3 (b), the PACl-E containing more than 60% Al13 of total Al concentration, indicating the PACl-E has strong charge neutralization ability, most likely destabilize particles predominately by charge neutralization. Thus, the particles can be rapidly restabilized at low pH, while still keep coagulation efficient at high pH. Meanwhile, the similar results are found for PACl-Al13 coagulation at pH

10. On the other hand, the coagulation efficiency of PACl-E is superior to that of PACl-C when pH is higher than 10 where the zeta potential is rather negative, which implies the more particles are favorably destabilized by electrostatic patch. Since Al13 favors aggregation when pH is over 6 (Furrer et al., 1992), Al13 aggregates are thus easily formed at high alkaline pH at which the Al13 molecules are mainly absorbed onto the isolated regions of particles with negative surface charge and then could further restructure and aggregate to form Al13 aggregates (Ye et al., 2007). At such condition, some regional charge heterogeneity can occur, which results in electrostatic patch as polymer flocculation (Gregory, 1973). As a result, PACl-E coagulation could be favorable for particle destabilization at alkaline pH to form larger flocs by electrostatic patch with larger Al13 aggregates, resulting in more efficient turbidity removal. In the case of PACl-Al13 coagulation, the tendency of changes in the zeta potentials and residual turbidity is similar to PACl-E coagulation;

however, the coagulation efficiency of PACl-Al13 is worse, which is ascribed to higher negative zeta potential of particles that results in the stronger repulsion among particles coagulated by PACl-Al13, as indicated in Fig. 4.2 (c). This impedes the occurrence of sufficient particles destabilization and aggregation.

4.2 Effects of Dosage on Coagulation Efficiency

The dosage of coagulant can influence the amounts of hydrolyzed Al species adsorbed onto the surface of particles and the conformation of hydrolyzed polymeric Al species adhered to the surface of particles; as a consequence, the destabilization mechanism induced by polymeric species are affected (Zhou and Franks, 2006).

PACl-C favors enmeshment or sweep flocculation at neutral pH, and in order to testify whether Al13 aggregates formed at alkaline pH are able to cause other mechanisms such as interparticle bridging in addition to electrostatic patch as high dosage of PACl-E and PACl-Al13 is applied. The effects of dosage on the efficiency of particle destabilization were evaluated at pH 7.5 for PACl-C coagulation and pH 10 for PACl-E and PACl-Al13 coagulation, respectively.

4.2.1 Effects of Dosage on Particle Destabilization

The changes in the residual turbidity and zeta potential with dosage during coagulation are illustrated in Fig. 4.4. For PACl-C coagulation, as shown in Fig. 4.4 (a), the removal of turbidity increases with dosage, which accompanies the increasing zeta potentials. The efficient turbidity removal is still found at high dosage, even though the charge reversal of particles takes place, which is attributed to the occurrence of enmeshment or sweep flocculation that is considered to be not affected by repulsion between particles.

By contrast, the turbidity decreases rapidly after the addition of PACl-E as well as PACl-Al13 at pH 10, and then reaches a minimum. The magnitude of the negative zeta potential sharply decreases with dosage at low dosage (< 2 mg/L), while the negative zeta potential nearly levels off, and approaches to zero when the dosage is higher than 8 mg/L where the coagulant dosage reaches the plateau of adsorption, as

shown in Fig. 4.4 (b) and Fig. 4.4 (c). The plateau of almost zero zeta potential implies that the charge neutralization ability of Al13 is reduced. Wu et al. (2007) have indicated that highly positively charged Al13 adsorbs onto the surface of particles by monolayer adsorption. Then, the saturation of surface coverage occurs rapidly and the retabilization of particles is due to the strong repulsion between polycations.

However, the destabilized particles with zero zeta potential at high dosage are still found in this study, which could be ascribed to the low charge neutralization ability of Al13 aggregates (Bottero et al., 1987; Bertsch, 1987).

Al13, which has a +7 valence, could be modified by deprotonation that release the protons in pairs and yield the weakly positively charged Al13 such as Al135+ Al133+ and Al13+ while pH is raised, and the aggregation and precipitation of Al13 occurs above pH 6, at which the charge of aggregated Al13 became weak (Furrer et al., 1992). In addition, the size and the electrophoretic mobility of Al13 aggregates principally vary with pH of solution and the concentration of Al13 (Furrer et al., 1999; Dubbin and Sposito, 2005). Although Rakotonarivo et al. (1988) have addressed that Al13

aggregates formed at various Al concentrations almost have zero surface charge at pH 8, Al13 aggregates may form at alkaline pH and still have certain charge neutralization ability to cause efficient coagulation (Chen et al., 2006). This is due to the fact that Al13 aggregates formed at different concentration of Al13 and pH values responds to their different neutralization charge ability.

As shown in Fig. 4.5, the zeta potential of Al species for PACl-Al13 coagulation without kaolin particles at pH 10 is around zero and almost insensitive to dosage, which implies that the Al13 aggregates with nearly zero surface charge formed at alkaline pH and high dosage for PACl-E and PACl-Al13 coagulation could not cause the strong charge neutralization. However, different concentrations of Al13 applied

can influence the formation rate of Al13 aggregates in kaolin suspension. Because Al13 require time to form Al13 aggregates, the weakly positively charged Al13 may form at pH 10 and low dosage before the formation of Al13 aggregates and then be adsorbed rapidly onto particles to neutralize their partial negative charge. At high dosage, because the formation of Al13 aggregates is faster, almost neutrally charged Al13 aggregates can quickly cloud the particles by adsorption, which leads to nearly zero zeta potential of the particles, even though the dosage is continuously increased.

This demonstrates that at alkaline pH the different coagulation mechanisms could be induced by PACl-E and PACl-Al13 at low and high dosage, respectively.

Zeta Potential (mV)

Fig. 4.4 Dosage effects on the residual turbidity and the zeta potential for (a) PACl-C (7% Al13) coagulation at pH 7.5 (b) PACl-E (60% Al13)

Dosage (mg/L as Al)

0 1 2 3 4 5 6 7 8 9 10 11

Zeta Potential (mV)

-1.5 -1.0 -0.5 0.0 0.5 1.0

Fig. 4.5 The zeta potential of Al species for PACl-Al13 coagulation at pH 10 and various dosages without kaolin particles (rapid mixing: 200 rpm; reaction time: 1 min).

4.2.2 Reactive Al Species of Flocs

In order to further verify that the predominant coagulation mechanism of three coagulants, the solid-state ²⁷Al-NMR was used to identify the Al speciation within the flocs. Fig. 4.6 shows the ²⁷Al-NMR spectra of flocs coagulated by PACl-C at neutral pH and by PACl-E as well as PACl-Al13 at pH 10 at various dosages. An intensive peak at 0 ppm is found within PACl-C flocs, while a weak peak at 33 ppm is only observed when the dosage is increased. On the other hand, the spectra for PACl-E flocs formed at 8 mg/L exhibits two distinct resonances at about 0 ppm and 63 ppm, whereas the spectra for PACl-E and PACl-Al13 flocs formed at 1 mg/L displays one prominent resonance at 0 ppm and very weak resonance at 63 ppm.

The symmetric peak appears at 0 ppm corresponds to octahedrally coordinated aluminum and the peak occurs at 63 ppm corresponds to tetrahedral aluminum (Bottero et al., 1980). The 33 ppm signal of PACl-C flocs, indicating a five-coordinate Al, is attributed to the process of freeze-drying of sediments (Wood et al., 1990). Only the octahedral aluminum is found in PACl-C flocs even at high dosage, which suggests enmeshment or sweep flocculation plays an important role for PACl-C coagulation at neutral pH. For PACl-E and PACl-Al13 coagulation, because of the weaker surface charge of Al13 at alkaline pH, less Al13 could react with the negatively charged kaolin particles, which leads to the unobvious 63 ppm signal.

However, the unambiguous resonance at 63 ppm is detected when the dosage of coagulant increases, which could be attributed to the aggregation of the Al13. At high dosage, more and larger Al13 aggregates form and adsorb onto the particles, resulting in PACl-E and PACl-Al13 flocs with extensive characteristic resonance of Al13. Since the central AlO4 is almost unreactive when the aggregation of Al13

occurs (Phillips et al., 2000), the 63 ppm signal remains intensive even though the

aggregated Al13 is formed. Although previous studies have proved that the signal at around 62 ppm is still obtained from solid-state ²⁷Al-NMR experiments of Al precipitates formed at high [OH^-]/[Al] ratio (Bottero et al., 1980; Bottero et al., 1987;

Furrer et al., 2002), the conditions of preparation such as concentration of total Al and [OH^-]/[Al] ratio could significantly affect the characteristics of Al species, including noncrystalline and crystalline Al species. At different conditions, the noncrystalline and crystalline Al species may show the same chemical properties, but differ in particle size and reactive surface. The results of Al species distribution in coagulation suggests that the flocs formation by PACl-C coagulation at neutral pH relies on Al(OH)3 precipitates regardless of dosage applied, while Al13 and Al13

aggregates are the predominant species to interact with kaolin particles for PACl-E and PACl-Al13 coagulation at alkaline pH.

-200 -100

0 100

200
(ppm)

Al_m

Al₁₃

(a) (b) (c) (d) (e) (f)

five-coordinate

Fig. 4.6 Solid-state ²⁷Al MAS-NMR spectra of freeze-dried kaolin flocs coagulated by three coagulants at various dosages. PACl-C:(a) 1 mg/L (b) 8 mg/L;

PACl-E:(c) 1 mg/L (d) 8 mg/L; PACl-Al13:(e) 1 mg/L (f) 8 mg/L

4.3 Summary

The major Al species of PACl-C are monomeric and colloidal Al, while the majority of Al species in PACl-Eand PACl-Al13 is Alb (i.e., Al13), especially for PACl-Al13 that contains more than 95% Al13 of total Al concentration. It is found that Al13 can be roughly represented by Alb. PACl-C responds to pH differs from PACl-E as well as PACl-Al13. Large amounts of Alc form at neutral pH for PACl-C, while the pre-hydrolyzed Alb in PACl-E as well as PACl-Al13 remains more stable with varying pH except alkaline pH. The flocs formation by PACl-C coagulation at neutral pH relies on Al(OH)3 precipitates regardless of the dosage applied. In addition, either electrostatic patch or interparticle bridging induced by Al13 aggregates may be responsible for particle destabilization and aggregation for PACl-E and PACl-Al13 coagulation at alkaline pH, depending on the dosage. At low dosage, the partially neutralized particles by electrostatic patch are easier to occur for PACl-E and PACl-Al13 coagulation at alkaline pH. Contrary to low dosage, the plateau of most zero zeta potential occurs even though high dosage is applied. At such condition, more and larger Al13 aggregates could form at such condition and adsorb onto the surface of kaolin particles, which could increase the opportunities of interparticle

The major Al species of PACl-C are monomeric and colloidal Al, while the majority of Al species in PACl-Eand PACl-Al13 is Alb (i.e., Al13), especially for PACl-Al13 that contains more than 95% Al13 of total Al concentration. It is found that Al13 can be roughly represented by Alb. PACl-C responds to pH differs from PACl-E as well as PACl-Al13. Large amounts of Alc form at neutral pH for PACl-C, while the pre-hydrolyzed Alb in PACl-E as well as PACl-Al13 remains more stable with varying pH except alkaline pH. The flocs formation by PACl-C coagulation at neutral pH relies on Al(OH)3 precipitates regardless of the dosage applied. In addition, either electrostatic patch or interparticle bridging induced by Al13 aggregates may be responsible for particle destabilization and aggregation for PACl-E and PACl-Al13 coagulation at alkaline pH, depending on the dosage. At low dosage, the partially neutralized particles by electrostatic patch are easier to occur for PACl-E and PACl-Al13 coagulation at alkaline pH. Contrary to low dosage, the plateau of most zero zeta potential occurs even though high dosage is applied. At such condition, more and larger Al13 aggregates could form at such condition and adsorb onto the surface of kaolin particles, which could increase the opportunities of interparticle