The influence of multivalent cations on the flocculation of activated sludge with different sludge retention times

w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 2 2 5 e2 3 2

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/watres

The influence of multivalent cations on the flocculation of activated sludge with different sludge retention times Haisong Li a,b, Yue Wen a,*, Asheng Cao a, Jingshui Huang a, Qi Zhou a a

State Key Laboratory of Pollution Control and Resource Reuse, College of Environmental Science and Engineering, Tongji University, Shanghai 200092, PR China b School of Chemical Engineering and Energy, Zhengzhou University, Zhengzhou 450001, PR China

article info

abstract

Article history:

The mechanism governing the flocculation of activated sludge (AS) with different sludge

Received 25 September 2013

retention times (SRTs) was studied in this paper. AS samples were cultivated in 8 lab-scale

Received in revised form

reactors with SRTs of 5 d, 7.5 d, 10 d, 12.5 d, 15 d, 20 d, 30 d, and 40 d. The bulk solution,

4 December 2013

loosely bound extracellular polymeric substances (LB-EPS), tightly bound EPS (TB-EPS), and

Accepted 3 February 2014

pellet were extracted for all 8 AS samples. There was a clear trend that the effluent

Available online 15 February 2014

turbidity decreased as the SRT increased, and we deduced that this is because AS samples with longer SRTs have lower interaction energy barriers and lower LB-EPS content.

Keywords:

Furthermore, the concentrations of multivalent cations (especially trivalent cations) in the

Activated sludge

pellets were found to be closely correlated to the AS flocculability, total interaction energy

Flocculation

(Wtot), and LB-EPS content. The multivalent (especially trivalent) cations possess greater

Sludge retention time (SRT)

binding ability, and this ability to bind tightly to AS in large quantities is responsible for the

Cations

superior flocculability of AS samples with longer SRTs. Hence, the concentrations of

Extracellular polymeric substances

multivalent cations in the pellets are an important indicator of AS flocculability. We

(EPS)

deduced that variations in the quantities of multivalent cations that tightly bind with the AS rather than remaining in the influent are the core reason behind observed fluctuations in the AS flocculability with different SRTs. ª 2014 Elsevier Ltd. All rights reserved.

1.

Introduction

Activated sludge (AS) systems are the most widely used biological wastewater treatment process (Sanin et al., 2006). One of the key elements to the successful operation of an AS system is the efficient solideliquid separation achieved by bioflocculation.

The sludge retention time (SRT) is not only considered one of the most important operating parameters but also one of the most complicated elements to manipulate due to its widespread effects on the AS system (Maharajh, 2010). The SRT affects the performance, including the removal of slowly biodegradable organics, nitrogen, and phosphorus. In addition, AS flocculation is also considerably influenced by the

* Corresponding author. Room 301, Mingjing Building, College of Environmental Science and Engineering, Tongji University, Shanghai 200092, PR China. Tel./fax: þ86 21 65982697. E-mail addresses: [email protected], [email protected] (Y. Wen). 0043-1354/$ e see front matter ª 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.watres.2014.02.014

226

w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 2 2 5 e2 3 2

SRT. A higher content of pin-point flocs is often found to be related to a shorter SRT, and a minimum SRT is required for acceptable bioflocculation (Maharajh, 2010). It is a commonly accepted idea that AS flocculation, sedimentation, and dewaterability improve considerably with a longer SRT (Chao and Keinath, 1979; Liao et al., 2006; Sanin et al., 2006; Li and Yang, 2007; Xie and Yang, 2009). The extracellular polymeric substances (EPS) content also has a significant influence on the bacterial surface characteristics and flocculability (Liu et al., 2007). However, the exact role the EPS content plays in AS flocculation is currently still under debate (Masse et al., 2006). For example, Wile´n et al. (2003) suggested that increased amounts of total extracted EPS are negatively correlated to AS flocculability, whereas another study showed exactly the reverse trend (Ng and Hermanowicz, 2005); meanwhile, Liao et al. (2001) found that the surface properties, hydrophobicity, surface charge, and EPS composition, rather than EPS quantity, govern bioflocculation. Recently, researchers separated EPS into two parts, loosely bound EPS (LB-EPS) and tightly bound EPS (TB-EPS), and found that LB-EPS had a negative effect on bioflocculation and sludge-water separation and also that both the flocculability and LB-EPS content of AS decreased as the SRT increased (Li and Yang, 2007; Yang and Li, 2009; Wu et al., 2011). Due to the high negative charge density of EPS, cations also play an important role in AS flocculation (Wile´n et al., 2008). The theories pertaining to the role of cations in AS flocculation are divalent cation bridging (DCB) theory (include the alginate theory) and Derjaguin-Landau-Verwey-Overbeek (DLVO) theory (Sobeck and Higgins, 2002). According to DCB theory, divalent cations serve as bridges that connect negatively charged functional groups within the EPS and this helps to promote AS flocculation. Therefore, it is a well accepted idea that divalent cations are much more beneficial to AS flocculation than monovalent cations, and a ratio of divalent:monovalent cations exceeding 1:2 (expressed on an equivalent basis) is necessary for good bioflocculation (Higgins and Novak, 1997). The classical DLVO theory was first proposed to describe the stability of colloidal suspensions in colloid chemistry, and the theory takes into account attractive energy (the van der Waals force (WA)) and repulsive energy (the electric double layer (WR)) (Yen, 1999). Later, Lewis acidebase interactions (WAB) were also added to the extended DLVO theory; thus, the total interaction energy (Wtot) in the extended DLVO theory is the sum: Wtot ¼ WR þ WA þ WAB (Wu et al., 1999). Both DLVO theory and extended DLVO theory have been widely used in different fields including aggregation of sludge flocs (Zita and Hermansson, 1994; Liu et al., 2007, 2010; Chia et al., 2011). Al3þ and Fe3þ are widely adopted in wastewater treatment plants (WWTPs) as coagulating agents because of their remarkable effects in promoting AS flocculation (Subramanian et al., 2010). Limited information indicates that there is a connection between the flocculation and the content of trivalent cations in the AS, and that trivalent cations contribute to a better effluent quality (Murthy and Novak, 2001; Park et al., 2006). Despite the importance of trivalent cations to bioflocculation, the role that trivalent ions such as Al3þ and Fe3þ play in AS flocculation has been investigated far

less than the roles of monovalent and divalent cations, yet researchers have suggested they may contribute greater floc stability due to their higher charge valence (Park et al., 2006; Maharajh, 2010). However, too high concentration of aluminum and iron is detrimental for the activated sludge flocculation, because their precipitate could block the transfer between microorganisms and their surroundings, and the impact is more sever for iron, due to its ability to form oxygen reactive species which could attacking the molecules of microorganisms (De Freitas and Meneghini, 2001; Philips et al., 2003; Agridiotis et al., 2007). Recent studies have shown that both longer SRTs and lower contents of LB-EPS are beneficial to AS flocculation; furthermore, they indicated that higher contents of multivalent cations also have some relationship with the bioflocculation. Thus, it is reasonable to speculate that the relationships between the abovementioned phenomena will definitely be helpful in understanding AS flocculation. In this study, we have found a relationship between the SRT, the LBEPS content, the concentrations of multivalent cations, and the AS flocculation, and this has provided a new perspective on the mechanisms governing AS flocculation with different SRTs.

2.

Materials and methods

2.1.

AS cultivation

The AS used in this study was cultivated in 8 parallel sequencing batch reactors (SBRs), and each of them had a volume of 4 L. The reactors were seeded with AS from Qu Yang WWTP with a nitrogen and phosphorus removal process in Shanghai, China. Each reactor was equipped with a paddle mixer operating at 100 rpm to prevent the AS from settling; air was introduced through stone air diffusers in order to maintain the dissolved oxygen (DO) in the range of 2e3 mg/L; and the temperatures of the reactors were maintained at 201  C. SRTs of 5 d, 7.5 d, 10 d, 12.5 d, 15 d, 20 d, 30 d, and 40 d were achieved by wasting of the AS at the end of the last phase of aeration. The SBRs were operated at a cycle time of 4 h; thus, 6 cycles were performed each day. Each cycle was divided into six phases: 1) mixing þ filling (anaerobic) for 20 min; 2) mixing þ aerating (aerobic) for 120 min; 3) mixing (anoxic) for 30 min; 4) mixing þ aerating (aerobic) for 30 min; 5) settling for 30 min; and 6) decanting for 10 min. The chemical oxygen demand (COD) values of the influent and effluent for the 8 reactors are shown in Table S1, and the COD:N:P ratio of the influent was maintained at 100:7.5:1.5. Glucose, NH4Cl, and KH2PO4 were the carbon, nitrogen, and phosphorus sources, respectively. The types and concentrations of feeding micronutrients (including iron) for the reactor with the SRT of 15 d were in accordance with the protocol that was given by Liao et al. (2001); and the ratios of micronutrient concentrations between all 8 reactors were the same as the ratios of their influent CODs; thus, the micronutrient concentrations were higher in the AS with shorter SRTs. The pH of all reactors was controlled in the range of 6.8e7.2 by adding NaHCO3 to the influent. After 80 d of operation under good conditions,

w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 2 2 5 e2 3 2

parameters such as the COD, mixed liquor suspended solids (MLSS), sludge volume index (SVI), and SS in the effluent were monitored on a daily basis for 10 d. The AS was collected for the experiments on the condition that the fluctuation of these parameters was less than 20%. The MLSS were kept at a level of 2300 mg/L by varying the influent COD. All of the AS samples, whether for the purpose of experimental research or for controlling the SRTs, were wasted during the last 10 min of the second aeration phase. The major cations in the influent of the 8 reactors are listed in Table S2.

2.2.

The flocculation and fractionation protocol of AS

Eight AS samples, each of which had a volume of 100 mL, were collected separately from 8 SBRs, and each of the AS samples was placed in a 250-mL beaker. Then, the 8 AS samples underwent a two-step flocculation process: rapid mixing with a stirrer at a speed of 117 rpm for 5 min, followed by slow mixing at 50 rpm for 5 min; the operational conditions were identical for the 8 beakers. After flocculation, the bulk solution, LB-EPS, TB-EPS, and the pellet were extracted using the following methods. The stratification structure and extraction protocol for the AS in this study were modified based on previous research (Morgan et al., 1990; Li and Yang, 2007; Yu et al., 2008). A 25-mL sample of sludge suspension was centrifuged at 4000 g for 5 min at 4  C, and the supernatant that was carefully collected was the bulk solution. A NaCl solution with the same conductivity as the AS sample was prepared and preheated to 70  C; then, it was used to re-suspend the AS material in the tube at its original volume before the bulk solution was removed. With no delay, the AS suspension was sheared by a vortex mixer (S25, IKA, Germany) for 1 min; then, it was centrifuged at 4000 g for 10 min at 4  C, and the supernatant was collected as LB-EPS. The AS sample left in the tube was resuspended again to its original volume of 25 mL with the NaCl solution and then put in a water bath at 60  C for 30 min. It was centrifuged again at 4000 g for 15 min at 4  C, and the supernatant collected was TB-EPS, and the AS sample left was the pellet.

2.3.

Analytical techniques

The total organic carbon (TOC) contents of the bulk solution, LB-EPS, TB-EPS, and pellet were determined by a TOC analyzer (TMM-1, SHIMADZU, Japan). The pellet was lyophilized (LGJ10, Xinzhi Co., China) for 48 h and carefully weighed before TOC analysis. The proteins (PNs) and humic-like substances (HSs) were analyzed following the modified Lowry method (FrØlund et al., 1995) using bovine serum albumin (Sigma) and humic acid (Fluka) as the standards, respectively. The polysaccharides (PSs) content was measured by the anthrone method (Gerhardt et al., 1994) using glucose as the standard. The DNA contents in the EPS were determined by the diphenylamine colorimetric method (Sun et al., 1999) using deoxyribonucleic acid sodium salt from calf thymus (Sigma, 50 mg, D-1501) as the standard. The cation concentrations in the bulk solution, LB-EPS, TB-EPS, and pellet were analyzed by inductively coupled plasma optical emission spectrometry (ICPAES, Optima 2100 DV, Perkin Elmer, USA). Aqua regia digestion

227

was employed to digest the pellet following the standard method (USEPA, 1998) before measuring its cation concentrations. The turbidities of effluents from the 8 AS reactors were measured by a turbidity meter (2100P, HACH, USA). The sizes of the AS flocs were determined by the Eyetech particle size and shape analyzer (KS-100, ANKERSMID, Holland). 100 mL of sludge suspension was taken, and the SVI was measured in a graduated cylinder. The dewaterabilities of the AS flocs were determined using a CST instrument (model 319, Triton, UK). Zeta potential was quantified by a Zetasizer (Nano Z, Malvern, UK). The AS interaction energy curves were determined by a contact angle approach; the contact angle was determined by a contact angle analyzer (JC2000 D, Powereach Co., Shanghai, China) using the sessile drop technique, which is detailed elsewhere (Liu et al., 2010). The MLSS in the reactors and the effluent SS were analyzed by standard laboratory methods (APHA et al., 1998). All measurements were carried out five times (except for the ones indicated).

3.

Results and discussion

3.1. The influence of SRT on the flocculation-related characteristics of AS The effluent turbidity clearly declined as the SRT increased (Table S1); this trend is in agreement with previous findings (Li and Yang, 2007; Liao et al., 2001). Moreover, the average value of the effluent turbidity for long SRTs (20 d, 30 d, and 40 d) only accounted for 45% of that for short SRTs (5 d, 7.5 d, and 10 d), which supported a study by Liao et al. (2001) showing that the effluent suspended solids (ESS) for long SRTs (16 d and 20 d) only accounted for 47% of that for short SRTs (4 d, 9 d, and 12 d). Furthermore, a sharp reduction in effluent turbidity occurred when the SRT was 20 d, whereas a similar sharp reduction in ESS occurred at an SRT of 16 d in the study by Liao et al. (2001). Thus, the effluent turbidities of reactors with different SRTs seem to fall into two types: for shorter SRTs (5 de15 d), the effluent turbidity decreases steadily as the SRT increases; as the SRT further increases, the effluent turbidity remains approximately constant. There was a similar general trend in that the floc size, SVI, zeta potential, and capillary suction time (CST) decreased as the SRT increased, although this trend was not as clear as the trend for effluent turbidity (Table S1). It should also be pointed out that all four of the above parameters were much larger at lower SRTs (5 d, 7.5 d, and 10 d) in comparison to higher SRTs.

3.2. Variations in the interaction energy of AS at different SRTs The interaction energy curves point to a general trend that AS with longer SRTs possess lower energy barriers (Fig. S1). The energy barriers ranged from 254 KT (SRT ¼ 30 d) to 1932 KT (SRT ¼ 5 d). However, changes in the energy barriers were not uniform: some changed little, such as AS with an SRT of 5 d (1932 KT) and 7.5 d (1837 KT); some changed significantly, such as AS with an SRT of 12.5 d (1339 KT) and 15 d (711 KT); in some

228

w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 2 2 5 e2 3 2

cases, AS with a longer SRT exhibited a higher energy barrier, such as AS with an SRT of 30 d (254 KT) and 40 d (281 KT). Liu et al. (2007) have studied the flocculation of a type of bacterial suspension, and have found that the interaction energy barrier of the suspensions decreased from 1981 KT to 533 KT with an increase in electrolyte concentration; therefore, the flocculability of that bacterial suspension improved with an increase in ionic strength. In our study, the concentrations of the major ions added to the influent of the 8 reactors are listed in Table S2. It is clear that the influent ionic strength decreased with increasing SRT, and this is further proved by the conductivities of the AS in the 8 reactors, which decreased as the SRT increased (Table S1). However, the energy barriers decreased as the electrolyte concentrations decreased when the SRT increased, and this contradicts the aforementioned finding of Liu et al. (2007). It was suspected that there might be mechanisms governing the interaction energy of AS other than ionic strength, and these will be explored in this study.

3.3. The quantities and components of LB-EPS and TBEPS of AS with different SRTs

3.4.1. of AS

The distribution of different cations in different parts

AS samples with SRTs of 5 d, 10 d, 15 d, and 30 d were used to examine the cation distributions in them (Fig. S3). It is clear that the distributions of the different cations were significantly different, whereas the differences of given type cation distributions in these four AS samples were limited; thus, the average values for AS samples with different SRTs were used in the comparison. The trivalent cations (aluminum and iron) exhibited a considerable preference for the AS, as nearly 100% were concentrated in the pellet. In comparison, fewer bivalent cations (magnesium (34% on average) and calcium (16%) were found in the pellet. Compared to trivalent and bivalent cations, the lowest percentage (10%) of monovalent cations (potassium) was found in the pellet, and the distribution percentages of cations in the pellet followed the order: trivalent cations > bivalent cations > monovalent cations. Since the pellet is the core of the AS, a greater percentage of cations concentrated in the pellet indicates higher binding ability of that type of cation to the AS.

The quantities of LB-EPS and TB-EPS

The compositions of LB-EPS and TB-EPS

For both LB-EPS and TB-EPS, the PNs and HSs accounted for the major part of the EPS, whereas PSs and DNA only accounted for a small portion of the total organic substances (Fig. S2), and this is consistent with a previous study (Liao et al., 2001; Li and Yang, 2007; Wile´n et al., 2008). There was a tendency for the contents of all the components both in LBEPS and TB-EPS (except PSs in LB-EPS) to decrease as the SRT increased; thus, the sum of the four components decreased as the SRT lengthened. This is again consistent with the reduction in EPS content for AS of longer SRTs.

20 18

Effluent turbidity (NTU)

The content of the organic substances (in terms of TOC) in TBEPS was much higher than that in LB-EPS; this is consistent with other studies (Li and Yang, 2007; Yang and Li, 2009; Liu et al., 2010). In general, The TOC content in both LB-EPS and TB-EPS of AS decreased as the SRT increased (Fig. 1). In addition, the LB-EPS contents were reduced to a greater extent than the TB-EPS contents: the ratio of the least to the most LBEPS content was 42%, whereas it was 65% for TB-EPS. This indicates that SRT had a more distinct effect on LB-EPS than TB-EPS. Both LB-EPS and TB-EPS contents positively correlated to the effluent turbidity, but the correlation between LB-EPS and the effluent turbidity was considerably higher (Fig. 1). This supports the notion that higher LB-EPS content has a negative effect on AS flocculation (Li and Yang, 2007). It is also interesting to note that the effluent turbidities from different AS reactors with longer SRTs (20 d, 30 d, 40 d) did not vary as much as those with shorter SRTs (5 de15 d), but the LB-EPS of both groups of AS decreased with no significant differences between those with longer and shorter SRTs. Therefore, LB-EPS seems to have a different influence on AS flocculation depending on the SRT; for AS with longer SRTs, the effects of LB-EPS on the AS flocculation were smaller.

3.3.2.

The role of multivalent cations in AS flocculation

(a)

16

2

R =0.88

14 12 10 8 6 4 2 0

6

8

10

12

14

16

18

20

22

24

LB-EPS (mg TOC/g SS) 20 18 16 Effluent turbidity (NTU)

3.3.1.

3.4.

14

(b) 2

R = 0.50

12 10 8 6 4 2 0 20

25

30

35

40

45

50

55

60

TB-EPS (mg TOC/g SS) Fig. 1 e The influence of (a) LB-EPS and (b) TB-EPS on AS flocculability indicated by effluent turbidity. Error bars represent mean values ± one standard deviation.

229

w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 2 2 5 e2 3 2

3.4.2. The influence of the concentrations of multivalent cations in the pellet on the AS flocculability

20

(a)

18 Effluent turbidity (NTU)

The variations in concentrations of the different cations in the pellet of AS as the SRT increased were different (Fig. S4). There was not any clear trend for potassium as the SRT lengthened whereas, for the bivalent cations (calcium and magnesium), there was a general trend that cation concentration increased with increasing SRT, and, for the trivalent cations (especially aluminum), it was clear that concentrations in the pellet increased more regularly than those of bivalent cations as the SRT increased. The concentrations of trivalent cations in the pellet increased proportionately as the SRT lengthened although the total trivalent cation concentrations in the influent decreased. This was because AS with a longer SRT remained in the reactor for a longer period, and all of the reactors experienced the same volume of influent per day, so that AS in reactors with longer SRTs interacted overall with a larger volume of influent water, which provided a greater source of trivalent cations. It is interesting to note that the concentrations of the different types of multivalent cations in the pellet showed different relationships to the AS flocculability (Fig. 2). The concentrations of trivalent cations in the pellet were highly correlated to the AS flocculability (R2 ¼ 0.90); in comparison, the concentrations of bivalent cations (calcium and magnesium) in the pellet had less effect on AS flocculability (R2 ¼ 0.49). Therefore, it is clear that trivalent cations play a

2

16

R = 0.49

14 12 10 8 6 4 2 0.30

0.35

0.40

0.45

0.50

0.55

0.60

0.65

Bivalent cations' concentration (meq/g SS)

20

(b)

18

2

Effluent turbidity (NTU)

R =0.90 16 14 12 10 8 6 4

0.15

0.20

0.25

0.30

0.35

0.40

2500

(a)

Trivalent cations' concentration (meq/g SS)

The considerably greater abilities of the trivalent cations to bind to the AS should be closely related to their characteristics. The trivalent cations (aluminum and iron) can exist as several species, including colloidal trivalent cation hydroxides, poly-trivalent cations, and trivalent cations (Nielsen and Keiding, 1998). Since trivalent cations possess higher valence states, they have a higher polarizability and relatively smaller degree of hydration compared to bivalent or monovalent cations. Therefore, if the trivalent cations are interacting with the AS, they will indeed form stronger bonds and thus be more difficult to separate from the AS. As a result, most trivalent cations will be present in the pellet of AS. If the cations exist in the form of colloidal trivalent cation hydroxides, then precipitation will occur (Nielsen and Keiding, 1998). As these colloids are positively charged and possess a very reactive surface, they are better flocculants than bivalent or monovalent cations. Hence, they will still precipitate and be tightly bound to the AS.

Energy barrier (KT)

2000

2

R = 0.82

1500 1000 500 0 24 22 LB-EPS (mg TOC/ g SS)

Fig. 2 e The effects of different cations concentration in the pellet of AS with different SRTs on their flocculability indicated by effluent turbidity; the cations are: (a) bivalent cations (magnesium and calcium) and (b) trivalent cations (aluminum and iron). Error bars represent mean values ± one standard deviation.

(b)

20

2

R = 0.88

18 16 14 12 10 8 6

0.15

0.20

0.25

0.30

0.35

0.40

Trivalent cations' concentration (meq/g SS) Fig. 3 e The influence of trivalent cations in the pellet on interaction energy barrier (a) and LB-EPS content (b).

w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 2 2 5 e2 3 2

much more important role in improving AS flocculation than bivalent cations. It has already been mentioned in Section 3.1 that the effluent turbidity decreased by 33% when the SRT increased from 15 d to 20 d, which was the most significant change between two reactors with sequential SRTs; at the same time, the concentrations of trivalent cations in the pellet increased by 30%, which was also the most considerable increase between two reactors with sequential SRTs. This further proves that the concentrations of trivalent cations in the AS pellet play a vital role in AS flocculation. Liao et al. (2002) found that sludge flocs at higher SRTs (16 and 20 d) were physically more stable than those at lower SRTs (4 and 9 d), and the similar phenomenon in Section 3.1 also showed that sludge flocculationrelated parameters exhibit two different types of characteristics due to their different SRTs. This indicates that higher concentrations of multivalent cations (especially trivalent cations) in AS pellets with longer SRTs possibly account for their better flocculability.

3.4.3. The influence of the concentrations of multivalent cations on the total interaction energy and LB-EPS content As mentioned in Sections 3.2 and 3.3, both the interaction energy and LB-EPS content play important roles in AS flocculation, and it is most interesting to find that the concentrations of multivalent cations (particularly trivalent cations) in the AS pellets were closely correlated to both of these parameters (Fig. 3), as will be discussed in the following.

3.4.3.1. The influence of multivalent cations on the total interaction energy. A question about the interaction energy barrier was posed in Section 3.2 regarding why the AS energy barriers were reduced when the ionic strengths were decreased. The answer can be found in Fig. 3a: the increasing concentrations of trivalent cations in the AS pellet contributed significantly to the reduction in Wtot of AS (R2 ¼ 0.82); the bivalent cations in the pellet were also negatively correlated to Wtot (Fig. S5), but this correlation was much weaker than that of the trivalent cations. Although the ionic strengths and concentrations of multivalent cations in the pellet can have opposite influences on the Wtot, obviously the concentrations of multivalent cations in the pellet were the predominant factor under our experimental conditions. Multivalent cations (especially trivalent cations) located in the pellet can effectively reduce the zeta potential of AS, which leads to a reduction in Wtot. As mentioned earlier in Section 3.4.1, no matter what form of trivalent cations (trivalent cations or their hydroxides) exist in the AS system, they can bind tightly to the AS, thus effectively neutralizing the negative charges in the EPS and, therefore, reducing the zeta potential so that, as a result, the Wtot is reduced. It has been widely accepted that bivalent cations (mainly Ca2þ and Mg2þ) play an important role in ion bridging between AS flocs (Sobeck and Higgins, 2002; Higgins and Novak, 1997; Sheng et al., 2006; Kara et al., 2008); the bridging process can also lead to close contact between bivalent cations and the AS, which is helpful in reducing the Wtot. The more multivalent cations bind tightly with the AS, the more Wtot is reduced, thus the concentrations of multivalent cations in the pellet are an important parameters indicating the ability to reduce the Wtot.

Researchers have found that AS surfaces are more negatively charged at lower SRTs than at higher SRTs (Liao et al., 2001, 2002; Cao et al., 2010). It is suggested according to this study that the greater binding abilities of multivalent (especially trivalent) cations to AS and their higher concentrations in AS pellets with longer SRTs could account for the less charged AS surfaces. Besides, the bivalent:monovalent cation ratio in the influent also probably affects the Wtot. Although the ionic strength of influent decreased as the SRT lengthened, this was mainly due to the decrease in monovalent cations (especially Naþ), whereas the concentrations of bivalent cations in the influent did not change substantially as the SRT increased (Table S2). Thus, the ratio of bivalent:monovalent cations increased significantly as the SRT increased. It is interesting to find that the ratio of bivalent:monovalent cations in the influent positively correlated with the AS energy barriers (R2 ¼ 0.89) (Fig. 4), which means that the higher ratio of bivalent : monovalent cations is beneficial to the reduction of Wtot. It has been reported that, since the monovalent cations can replace the bivalent cations in AS flocs, this will be deleterious to the structure of the AS flocs and will lead to deterioration in AS flocculation. Therefore, a higher bivalent : monovalent cation ratio was considered to be crucial for good flocculation of the AS (Higgins and Novak, 1997; Sobeck and Higgins, 2002). However, we propose that “the higher bivalent:monovalent cation ratio being favorable to better AS flocculation” can also be explained by a reduction in Wtot rather than a bridging effect of the bivalent cations (Fig. 4). Liu et al. (2007) reported that, at the same level of ionic strength, the zeta potential of a bacterial suspension in a CaCl2 solution was much higher than that in a NaCl solution. This result, combined with the discussion in this section, indicates that, although ionic strength plays an important role in the variations of Wtot, the types, distribution, and ratios of the cations can also have a crucial influence on Wtot.

2500 2

R =0.89 2000 Energy barrier (KT)

230

1500 1000 500 0 0.4

0.5

0.6

0.7

0.8

0.9

1.0

Ratio of bivalent/monovalent cations Fig. 4 e The influence of bivalent/monovalent cations ratio in the influent on interaction energy barrier of AS. The data were collected from the influents and AS of all the 8 reactors with different SRT, and the cations were expressed on an equivalent basis. Error bars represent mean values ± one standard deviation (n [ 5).

w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 2 2 5 e2 3 2

3.4.3.2. The influence of multivalent cations on the LB-EPS content. The LB-EPS content showed a nearly linear decrease with the concentration of trivalent cations in the pellet (R2 ¼ 0.88) (Fig. 3b). However, there was a much poorer correlation between the concentration of bivalent cations in the pellet and the LB-EPS content (R2 ¼ 0.58) (Fig. S6). Moreover, the LB-EPS content exhibited a sudden decrease from 15.43 mg TOC/g SS (SRT ¼ 15 d) to 12.82 mg TOC/g SS (SRT ¼ 20 d), whereas the trivalent cation concentration increased from 0.245 meq/g SS to 0.317 meq/g SS, and this also was the most significant increase for any two AS samples with sequential SRTs. This further proved the connection between LB-EPS content and the concentration of trivalent cations in the pellet. There are several factors that might account for the influence of the concentration of trivalent cations in the pellet on the LB-EPS content. Firstly, as mentioned earlier in Section 3.4.1, because of the higher valence states, higher polarizability, and relatively smaller degree of hydration of trivalent cations compared to bivalent cations, the trivalent cations will definitely bind tightly with the EPS, so that the well bridged EPS will be harder to extract and, as a result, the LB-EPS will be reduced. Secondly, the colloidal trivalent cation hydroxides are better flocculants than the bivalent cations. This has been shown for many macromolecules, including polysaccharides and humic substances (Gecsey and Jang, 1989; Nielsen and Keiding, 1998). Thus, the trivalent cations can flocculate the EPS, which is made up of organic macromolecules, and tightly bind them to the AS pellet, and this could also contribute to the LB-EPS reduction. Furthermore, a study conducted by Murthy and Novak (2001) showed that ferric hydroxide can retain protein within the AS through possible adsorptive interactions and may be responsible for some biopolymer retention in the AS flocs. Since protein and iron are major components of LB-EPS and trivalent cations, respectively, ferric hydroxide could account for the reduction in LB-EPS content. This was supported by the negative correlation (R2 ¼ 0.76) between the PNs in LB-EPS and iron in the pellet of AS (Fig. S7). Considering the close relationship between the concentrations of multivalent cations in the pellet, Wtot, and LB-EPS content, the greater binding abilities of the multivalent (especially trivalent) cations and the higher quantities that bind tightly to the AS are the real factors governing AS flocculation by affecting the interaction energy and LB-EPS. This is consistent with our former study about the variation AS flocculability with different multivalent cations as additives (Li et al., 2012). As the quantities of multivalent cations in the pellet represent the quantities of cations tightly bound to the AS, the concentration of multivalent cations in the pellet is an important parameter indicating the AS flocculability. The concentration of multivalent cations was higher in AS pellets with longer SRTs as discussed earlier in Section 3.4; therefore, AS possesses better flocculability when the SRT is longer.

4.

Conclusions

The AS flocculability improved as SRT increased, simultaneously, the AS with longer SRT possess lower energy

231

barriers and lower level of LB-EPS content. Both of the Wtot and LB-EPS contributed a lot to the differences in flocculability of AS with different SRT. Furthermore, the Wtot, LBEPS, and the AS flocculability showed much closer correlation to the trivalent cations concentration in the pellet of AS than other bivalent or monovalent cations. Since AS with longer SRT can bind more trivalent cations to the pellet, thus the AS flocculability becoming better with longer SRT. Hence, it is proposed that the trivalent cations concentration in pellet rather than in the influent is the fundamental reason governing the differences in AS flocculability with different SRT.

Acknowledgments This work was supported by the Major Science and Technology Program for Water Pollution Control and Treatment of China (2011ZX07318-001) and the National Science Foundation of China (51078284).

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.watres.2014.02.014.

references

Agridiotis, V., Forster, C.F., Carliell-Marquet, C., 2007. Addition of Al and Fe salts during treatment of paper mill effluents to improve activated sludge settlement characteristics. Bioresour. Technol. 98, 2926e2934. APHA, AWWA, WEF, 1998. Standard Methods for the Examination of Water and Wastewater, twentieth ed. APHA, Washington, DC, USA. Cao, Z., Zhang, J., Zhang, H., 2010. Effects of sludge retention time on the characteristics of mixed liquor and membrane fouling in membrane bioreactor. Adv. Mater. Res. 113e114, 2057e2061. Chao, A.C., Keinath, T.M., 1979. Influence of process loading intensity on sludge clarification and thickening characteristics. Water Res. 13, 1213e1223. Chia, T.W.R., Nguyen, V.T., McMeekin, T., Fegan, N., Dykes, G.A., 2011. Stochasticity of bacterial attachment and its predictability by the extended Derjaguin-LandauVerwey-Overbeek theory. Appl. Environ. Microbiol. 77 (11), 3757e3764. De Freitas, J., Meneghini, R., 2001. Iron and its sensitive balance in the cell. Mutat. Res. 475, 153e159. FrØlund, B., Griebe, T., Nielsen, P.H., 1995. Enzymatic activity in the activated-sludge floc matrix. Appl. Microbiol. Biotechnol. 43, 755e761. Gerhardt, P., Murray, R.G.E., Wood, W.A., Krieg, N.R., 1994. Methods for General and Molecular Bacteriology. American Society for Microbiology, Washington, DC, USA. Gecsey, G.G., Jang, L., 1989. Interactions between metal ions and capsular polymers. In: Bcrveridge, T.J., Doyle, R.J. (Eds.), Metal Ions and Bacteria. Wiley, New York, pp. 325e357. Higgins, M.J., Novak, J.T., 1997. The effect of cations on the settling and dewatering of activated sludges: laboratory results. Water Environ. Res. 69 (2), 215e224.

232

w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 2 2 5 e2 3 2

Kara, F., Gurakan, G.C., Sanin, F.D., 2008. Monovalent cations and their influence on activated sludge floc chemistry, structure, and physical characteristics. Biotechnol. Bioeng. 100, 231e239. Li, H.S., Wen, Y., Cao, A.S., Huang, J.S., Zhou, Q., Somasundaran, P., 2012. The influence of additives (Ca2þ, Al3þ, and Fe3þ) on the interaction energy and loosely bound extracellular polymeric substances (EPS) of activated sludge and their flocculation mechanisms. Bioresour. Technol. 114, 188e194. Li, X.Y., Yang, S.F., 2007. Influence of loosely bound extracellular polymeric substances (EPS) on the flocculation, sedimentation and dewaterability of activated sludge. Water Res. 41 (5), 1022e1030. Liao, B.Q., Allen, D.G., Droppo, I.G., Leppard, G.G., Liss, S.N., 2001. Surface properties of sludge and their role in bioflocculation and settleability. Water Res. 35 (2), 339e350. Liao, B.Q., Allen, D.G., Leppard, G.G., Droppo, I.G., Liss, S.N., 2002. Interparticle interactions affecting the stability of sludge flocs. J. Colloid Interface Sci. 249, 372e380. Liao, B.Q., Droppo, I.G., Leppard, G.G., Liss, S.N., 2006. Effect of solids retention time on structure and characteristics of sludge flocs in sequencing batch reactors. Water Res. 40, 2583e2591. Liu, X.M., Sheng, G.P., Luo, H.W., Zhang, F., Yuan, S.J., Xu, J., Zeng, R.J., Wu, J.G., Yu, H.Q., 2010. Contribution of extracellular polymeric substances (EPS) to the sludge aggregation. Environ. Sci. Technol. 44 (11), 4355e4360. Liu, X.M., Sheng, G.P., Yu, H.Q., 2007. DLVO approach to the flocculability of a photosynthetic H-2-producing bacterium, Rhodopseudomonas acidophila. Environ. Sci. Technol. 41 (13), 4620e4625. Maharajh, N., 2010. Effect of Feed Rate and Solid Retention Time (SRT) on Effluent Quality and Sludge Characteristics in Activated Sludge Systems Using Sequencing Batch Reactors Quality and Sludge Characteristics in Activated Sludge Systems Using Sequencing Batch Reactors. M. S. Thesis. Virginia Polytechnic Institute and State University, Blacksburg, Virginia. Masse, A., Sperandio, M., Cabassud, C., 2006. Comparison of sludge characteristics and performance of a submerged membrane bioreactor and an activated sludge process at high solids retention time. Water Res. 40, 2405e2415. Morgan, J.W., Forster, C.F., Evison, L., 1990. A comparative-study of biopolymers extracted form anaerobic and activated sludges. Water Res. 24 (6), 743e750. Murthy, S.N., Novak, J.T., 2001. Influence of cations on activatedsludge effluent quality. Water Environ. Res. 73 (1), 30e36. Ng, H.Y., Hermanowicz, S.W., 2005. Specific resistance to filtration of biomass from membrane bioreactor reactor and activated sludge: effects of exocellular polymeric substances and dispersed microorganisms. Water Environ. Res. 77, 187e192. Nielsen, P.H., Keiding, K., 1998. Disintegration of activated sludge flocs in presence of sulfide. Water Res. 32, 313e320. Park, C., Muller, C.D., Abu-Orf, M.M., Novak, J.T., 2006. The effect of wastewater cations on activated sludge characteristics: effects of aluminum and iron in floc. Water Environ. Res. 78 (1), 31e40.

Philips, S., Rabaey, K., Verstraete, W., 2003. Impact of iron salts on activated sludge and interaction with nitrite or nitrate. Bioresour. Technol. 88, 229e239. Sanin, F.D., Vatansever, A., Turtin, I., Kara, F., Durmaz, B., Sesay, M.L., 2006. Operational conditions of activated sludge: influence on flocculation and dewaterability. Dry. Technol. 24, 1297e1306. Sheng, G.P., Yu, H.Q., Li, X.Y., 2006. Stability of sludge flocs under shear conditions: roles of extracellular polymeric substances (EPS). Biotechnol. Bioeng. 93, 1095e1102. Sobeck, D.C., Higgins, M.J., 2002. Examination of three theories for mechanisms of cation-induced bioflocculation. Water Res. 36 (3), 527e538. Subramanian, S.B., Yan, S., Tyagi, R.D., Surampalli, R.Y., 2010. Extracellular polymeric substances (EPS) producing bacterial strains of municipal wastewater sludge: isolation, molecular identification, EPS characterization and performance for sludge settling and dewatering. Water Res. 44 (7), 2253e2266. Sun, Y., Clinkenbeard, K.D., Clarke, C., Cudd, L., Highlander, S.K., Dabo, S.M., 1999. Pasteurella haemolytica leukotoxin induced apoptosis of bovine lymphocytes involves DNA fragmentation. Veterinary Microbiol. 65, 153e166. USEPA, 1998. Method 3051A. Microwave Assisted Acid Digestion of Sediments, Sludges, Soils and Oils. Revision 1. US Environmental Protection Agency (USEPA), Washington, DC. Wile´n, B.M., Jin, B., Lant, P., 2003. The influence of key chemical constituents in activated sludge on surface and flocculating properties. Water Res. 37 (9), 2127e2139. Wile´n, B.M., Lumley, D., Mattsson, A., Mino, T., 2008. Relationship between floc composition and flocculation and settling properties studied at a full scale activated sludge plant. Water Res. 42 (16), 4404e4418. Wu, B., Yi, S., Fane, A.G., 2011. Microbial behaviors involved in cake fouling in membrane bioreactors under different solids retention times. Bioresour. Technol. 102, 2511e2516. Wu, W., Giese, R.F., van Oss, C.J., 1999. Stability versus flocculation of particle suspensions in water e correlation with the extended DLVO approach for aqueous systems, compared with classical DLVO theory. Coll. Surf. B-Biointerf. 14 (1e4), 47e55. Xie, B., Yang, S.F., 2009. Analyses of bioflocculation and bacterial communities in sequencing batch reactors. Environ. Eng. Sci. 26, 481e487. Yang, S.F., Li, X.Y., 2009. Influences of extracellular polymeric substances (EPS) on the characteristics of activated sludge under non-steady-state conditions. Process Biochem. 44 (1), 91e96. Yen, T.F., 1999. Environmental Chemistry: Essentials of Chemistry for Engineering Practice. Indianapolis, IN, Prentice Hall PTR, pp. 126e127. Yu, G.H., He, P.J., Shao, L.M., He, P.P., 2008. Stratification structure of sludge flocs with implications to dewaterability. Environ. Sci. Technol. 42 (21), 7944e7949. Zita, A., Hermansson, M., 1994. Effects of ionic-strength on bacterial adhesion and stability of flocs in a waste-water activated-sludge system. Appl. Environ. Microbiol. 60 (9), 3041e3048.