w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 5 6 e6 5

Available online at www.sciencedirect.com

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

Contribution of stratified extracellular polymeric substances to the gel-like and fractal structures of activated sludge D.Q. Yuan, Y.L. Wang*, J. Feng College of Environmental Science and Engineering, Beijing Key Lab for Source Control Technology of Water Pollution, Beijing Forestry University, Beijing 100083, China

article info

abstract

Article history:

The gel-like and fractal structures of activated sludge (AS) before and after extracellular

Received 1 September 2013

polymeric substances (EPS) extraction as well as different EPS fractions were investigated.

Received in revised form

The contributions of individual components in different EPS fractions to the gel-like

9 February 2014

behavior of sludge samples by enzyme treatment were examined as well. The centrifu-

Accepted 12 February 2014

gation and ultrasound method was employed to stratify the EPS into slime, loosely and

Available online 24 February 2014

tightly bound EPS (LB- and TB-EPS). It was observed that all samples behaved as weak gels with weak-link. TB-EPS and AS after LB-EPS extraction showed the strongest elasticity in

Keywords:

higher concentrations and highest mass fractal dimension, which may indicate the key

Activated sludge

role of TB-EPS in the gel-like and fractal structures of the sludge. Effects of protease or

Stratified extracellular polymeric

amylase on the gel-like property of sludge samples differed in the presence of different EPS

substances

fractions.

Gel-like structure

ª 2014 Elsevier Ltd. All rights reserved.

Fractal structure Conceptual sludge model

1.

Introduction

Three-dimensional, gel-like, highly hydrated extracellular polymeric substances (EPS) is considered to act as a biological glue, which is of significant importance in binding floc components together, containing polysaccharides (PS) and proteins (PN) as dominant components together with some lipid, nucleic acids, and humic-like substances (HS) (Nielsen et al., 1996). Some researchers stated that a significant proportion of the EPS was hydrophobic, such as PN, amino acids and lipids, while PS was hydrophilic (Dignac et al., 1998; Jorand et al., 1998). PS are composed of long chains of

* Corresponding author. Tel.: þ86 10 62336528; fax: þ86 10 62336596. E-mail address: [email protected] (Y.L. Wang). http://dx.doi.org/10.1016/j.watres.2014.02.028 0043-1354/ª 2014 Elsevier Ltd. All rights reserved.

monosaccharide units bound together by glycosidic bonds with linear to highly branched structure. They can readily mix in water because of the presence of the strong electrostatic interactions and hydrogen bond forces. While PN consisting of one or more chains of amino acid residues and lipids, a group of molecules, including phospholipids, monoglycerides and others, are non-polar and less readily bound with water (Raszka et al., 2006). On the other hand, the hydrophobic EPS was involved in the formation and organization of microbial aggregates (Jin et al., 2003; Liu et al., 2004), which maintains the structural and functional integrity of flocs/aggregates (Mu and Yu, 2006). Both the hydrophobic and hydrophilic components of EPS need to be accounted for the dewatering process

57

w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 5 6 e6 5

(Raszka et al., 2006; Mikkelsen and Keiding, 2002). Recently, EPS stratification has generated great research interest. Yu and co-workers (Yu et al., 2008, 2009) employed a novel EPS fractionation approach, the centrifugation and ultrasound approach, to stratify the EPS into different fractions including slime, loosely and tightly bound EPS (LB- and TB-EPS), and found that PN and PN/PS in the slime layers markedly influenced the sludge dewaterability as well as the TB-EPS fraction possessed higher flocculating rate to kaolin suspension in comparison with the other EPS fractions. In addition, Li and his co-workers (Li and Yang, 2007, 2009) employed a heat extraction method to fractionate EPS into LB- and TB-EPS and suggested that LB-EPS had a negative effect on bioflocculation and sludge dewatering. Wastewater sludge is a non-Newtonian fluid, which possesses both viscous and elastic properties, namely, also called viscoelastic (VE) properties (gel-like structure) originating from EPS (Legrand et al., 1998), which can be measured by dynamic testing where samples are subjected to oscillatory motion (Steffe, 1996; Ayol et al., 2006). Due to the existence of negatively charged macromolecules, EPS have been previously reported to provide cohesion and gel-like behavior to the floc structure (Keiding et al., 2001; Wile´n et al., 2003). Seviour et al. (2009a) also revealed that EPS are responsible for the gel-like properties of the granules, which can be referred to as hydrogels. EPS components, like exopolysaccharides or glycosides were the gelling agent in aerobic granules, confirmed to be important in determining the gel-like properties of sludge as well (Seviour et al., 2009a). In addition, Yuan and Wang (2013) found that at a TSS content of 54 g/L, AS after LB-EPS extraction exhibited the strongest gel-like structure. The result that TB-EPS may be the key fraction to the gel-like properties of AS was also obtained when an assumption was proposed to the formula derivation of rheological parameters for identifying the contribution of stratified EPS to the gel-like properties of AS. However, the direct experimental evidences for the contribution of stratified EPS to the gel-like behavior of AS at different TSS contents are still absent. On the other hand, EPS can be formed from simply aggregated polymeric chains, or their structure can be reinforced by cross-linking (Dursun, 2007). Several studies on the gel-like properties of EPS have also been reported (Seviour et al., 2009b; Lewin, 1956; Moreno et al., 2000). Seviour et al. (2009b) observed that the sol-gel transition of the granule EPS occurred at pH 9.0 to 12.0. At pH < 9 the granule EPS exhibited as a strong gel. The transition to a weak gel for the EPS from sludge flocs was found at pH 4.0 to 5.0. Seviour et al. (2010) also described the full structure of the exopolysaccharide, indicating that the gel-forming properties of granule EPS was attributed to a key exopolysaccharide. In addition, PN and PS are both able to form gels under very specific structural, stereochemical and environmental conditions (Seviour et al., 2009b). The result that the exopolysaccharides produced by

green algae (Lewin, 1956) or from cyanobacterium Anabaena sp. (Moreno et al., 2000) behaved as a weak gel was obtained by rheological measurements. It has ever been reported that gellike behavior is significant in restricting water mobility (Poxon and Darby, 1997). The gel structure of EPS results in fairly tenacious retention of water within the sludge (Steffe, 1996). Hence, it’s may be relatively easier to dewater after weakening or destroying the gel-like structure of the EPS or the AS. However, as the aforementioned research was performed based on EPS as a whole, the gel-like behavior of different EPS fractions is rather limited. Aside from rheological characterization of the structure of wastewater sludge, fractal structure characterization is another method commonly used to describe sludge properties (Li and Ganczarczyk, 1990; Wu et al., 2002; Wang et al., 2011). An important parameter used to describe the fractal structure of the sludge is the mass fractal dimension, Df. Generally, Df could indicate the floc density, floc strength, and the compactness/looseness of flocs/aggregate structures (Wang et al., 2009). While fractal analysis is widely used in studies on sludge flocs/aggregate structures (Wu et al., 2002), an apparent lack of knowledge about the influence of different EPS fractions on the AS and about the EPS themselves with respect to the fractal structure remains. The main objective of the present study is to explore the gel-like behavior at different TSS contents and fractal structure of the AS before and after EPS extraction as well as different EPS fractions. The contributions of individual components in different EPS fractions to the gel-like behavior of sludge samples by enzyme treatment were examined as well. The present study also proposes a conceptual sludge model to elucidate the significance of EPS on the AS structure. In this case, the EPS layer which mainly contributes to the gel-like structure of AS would be indentified, and would become a key fraction to be weakened during the dewatering and drying process. Therefore, these results obtained will provide further information on the influence of EPS on the dewatering properties of the AS.

2.

Materials and methods

2.1.

Characteristics of the AS

The AS was sampled from a municipal wastewater treatment plant in Beijing, China. Sludge samples kept on crushed ice were transferred to the laboratory within 2 h and immediately passed through a 1.2 mm sieve. Filtered samples were subsequently stored at 4  C. Table 1 lists the main characteristics of the AS. All of the tests were conducted within one week. The pH was measured using a pH meter (PB-10, Sartorius Stedim Biotech Co., Ltd., Beijing, China). Conductivity was measured using a conductivity meter (EC215, Beijing

Table 1 e Characteristics of the AS. pH 6.73  0.02

Conductivity (mS/cm)

TSS (g/L)

VSS (g/L)

VSS/TSS (%)

COD (mg/L)

SCOD (mg/L)

Zeta potential (mV)

1.46  0.01

8.86  0.30

6.13  0.21

69.15  0.03

6020.0  32.4

78  5

18.9  1.1

58

w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 5 6 e6 5

Kanggaote Science and Technology Co., Ltd., China). The total suspended solids (TSS g/L) and volatile suspended solids (VSS g/L) were assessed according to the standard method (APHA et al., 2005). The VSS/TSS ratios of about 75% were observed for all the sludge samples. The chemical oxygen demand (COD) of the filtrate was referred as soluble COD (SCOD). COD and SCOD analyses were conducted using a COD expedited testing apparatus (HATO CTL-12, Huatong Environmental Protecting Instruments Co., Ltd., Chengde, China). The Zeta potentials for the dilute suspension of the raw AS and different EPS solutions were recorded by a Malvern Zetasizer instrument (Nano Z, Malvern Co., UK). All chemical analyses were carried out in triplicate using chemicals of analytical grade.

2.2.

Stratification and analytical protocols for EPS

The different EPS layers: the slime, LB- and TB-EPS of the AS was extracted by centrifugation and ultrasound method (Yu et al., 2008). The slime contained organic materials obtained from the supernatant after low-speed centrifugation. The LBEPS and TB-EPS were dissolved in a buffer solution (pH 7) containing Na3PO4, NaH2PO4, NaCl, and KCl at a molar ratio of 2:4:9:1. The conductivities of the buffers were then adjusted with distilled water to match those of the filtrated sludge samples listed in Table 1. Briefly, the screened sludge samples were initially centrifuged at 2000g for 15 min at 4  C. The organic matter in the supernatant was considered to be slime. The settled sludge samples were subsequently resuspended to their original volume using the aforementioned buffer solution. Next, the suspensions (AS obtained after slime extraction) were subjected to centrifugation at 5000g for another 15 min at 4  C, after which the bulk solution and solid phase obtained separately. The organic matter in the supernatant comprised the LB-EPS. The collected sediments were again resuspended in the buffer solution (AS obtained after LB-EPS extraction) to predetermined volumes and then exposed to ultrasound at 20 kHz and 480 W for 10 min in an ice bath. The extracted solutions were centrifuged at 20,000g for 20 min at 4  C. The organic matter in the bulk solution was designated as the TB-EPS and residues resuspended in the buffer solution to their original volumes were designated as the pellet (AS obtained after TB-EPS extraction). Membranes (0.45 mm) were applied to filter out the particulates present in the slime, LBand TB-EPS solutions after all EPS fractions had been extracted. Two folds volumes of ethanol was added to the EPS fractions and maintained at 4  C for 24 h to precipitate soluble substances. The collected precipitate was referred to as the slime, LB- and TB-EPS, accordingly. The TSS of EPS fractions was measured according to the standard method (APHA et al., 2005). The PN, PS, HS and DNA were measured according the method descried in Yuan and Wang. (2013). In brief, the PN and HS of different EPS fractions were measured by the modified Lowry method using bovine serum albumin (Beijing Aoboxing Biotechnology Co., Ltd., China) and humic acid (Sigma, America) as standards, respectively. Polysaccharides (PS) and DNA were determined by the anthrone method, using glucose as the standard, and the diphenylamine colorimetric method, using 2-deoxy-D-ribose (Beijing Ruibo Biotechnology

Co., Ltd., China) as the standard, respectively. All of the chemicals used were of analytical grade, and all of the tests were performed in triplicate.

2.3.

Dynamic rheological measurements

Rheological tests were conducted by a rheometer (Physica MCR 300, Anton Paar, Austria) at constant T, 25  C. A PP 50 plate and plate sensor 49.94 mm of diameter and with a 2.0 mm gap was used. The strain amplitude sweep (SAS) test indicates variations in strain amplitude over time while the frequency (f) is maintained at 1 Hz. The moduli (storage modulus G0 , loss modulus G00 , complex modulus G*) on the logarithmic are usually plotted as a function of strain (g) on the logarithmic (Mezger, 2006). These moduli are independent of g until a critical strain level (gcD), above which the linear domain is approached. Then the linear viscoelastic (LVE) range could be obtained. Beyond this point, its nonlinear behavior appears. The critical storage modulus ðG00 Þ and gcD could be determined according to the transition. G00 represents the critical elasticity of the sludge network and can be adopted to indicate the gel-like structure of materials. After determining the G00 by SAS test, frequency sweep (FS) test was conducted at 0.1% of the strain value. The corresponding G0 and G00 values were measured during the test. The rheological tests were performed in duplicate.

2.4.

Enzymatic degradation of EPS

Stock solutions of six enzymes were prepared by dissolving the crude powder or fluid in the phosphate buffer. Protease (15e27 kDa, 250 mg/l) from Streptomyces griseus was used to degrade PN by hydrolyzing peptide bonds. The a- (51 kDa, 200 mg/l) b- (300 mg/l) amylases from Bacilus subtilis were used to cleave (1 / 4) -a- and b-glycosidic linkages, respectively, reducing the molecular size of glucans. Lipase (67 kDa, 200 mg/l) from Candida was applied to breakdown lipids, while RNase (- 13 kDa, 50 mg/l) and DNase (38 kDa, 10 mg/l) from Bovine pancreas were used to degrade any nucleic acids. Prior to the enzyme treatment, the stock solutions of enzymes were diluted into 20 ml with supernatant of various sludge samples, respectively. Afterwards, these enzymes solutions were added and mixed with the corresponding sludge samples (45 g/L, 5 ml). All samples were incubated at 37  C for up to 72 h without stirring or other agitation to mimic the physiological conditions. Following the incubation period, the samples were harvested regularly to measure their gel-like behaviors by SAS test. Control samples were also prepared in the corresponding supernatant alone, in comparisons with the enzyme-treated samples. Each experiment was conducted in duplicate to determine consistency in the results obtained.

2.5.

Determination of mass fractal dimension, Df

Sanin (2002) indicated that non-Newtonian behavior exists because of the colloidal properties of solids rather than the molecular properties of the suspension. According to the model proposed by Shih et al. (1990), the corresponding relationship between the gel-like properties for colloid gels and the particle concentration (F) can be described as K w F(3þx)/

w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 5 6 e6 5

(3D) 1/(3D) f in the strong-link regime and as K w F f in the weaklink regime.The K is the elastic constant such as: storage modulus G0 (Shih et al., 1990). Df is the fractal dimension of the flocs/aggregates, and x is the backbone fractal dimension of the flocs/aggregates, which varies in the range of 1.0e1.3 when particle concentration is lowered (Shih et al., 1990). Then the Df could be assessed by regression analysis of the lg (K) versus lg (TSS) plot.

3.

Results

3.1. Contents and characteristics of different EPS fractions Table 2 presents the contents and characteristics of different EPS fractions of the AS. PN and PS were mainly dispersed in the TB-EPS (48.6%, 43.2%) and slime (39.2%, 36.0%) fractions, with the lowest percentage (12.3%, 20.8%) detected in the LBEPS fraction. HS was mainly found in the slime fraction, though some was also observed in the TB-EPS fraction. The corresponding mass concentration ratios of HS in slime and TB-EPS to the sum of slime, LB- and TB-EPS in terms of HS were 51.1% and 36.4%, respectively. Compared with other fractions, the LB-EPS fraction showed the lowest concentrations of PN, PS, HS, and DNA. The order of all of the chemical component contents in terms of amounts was PS > PN > HS > DNA. Table 2 lists PN and PS as the major constituents of EPS. In addition, the three EPS fractions displayed different but negative surface charges, as measured by the Zeta potential.

3.2.

Gel-like structure

3.2.1.

SAS and FS tests

The difference among dilute solutions, entanglement network systems, weak gels and strong gels can be assessed by means of oscillatory dynamic experiments using parallel or cone and plate geometries (Clark and Ross-Murphy, 1987; Shoemaker et al., 1992). In dynamic rheological tests, G0 is a measure of the deformation energy stored in the sample and represents its elastic behavior during the shear process. Contrarily, G00 represents the viscous dissipation of energy used to deform the material (Mezger, 2006). Generally, if G0 > G00 , the material behaves a gel-like structure, that is, material deformations is essentially elastic or recoverable. If G00 > G0 , viscous behavior dominates over elastic behavior and the material behavior is liquid-like. The critical strain or frequency (gc or fc) of G0 ¼ G00 can be useful to study the structure of different samples. As

Table 2 e Contents and characteristics of the different EPS fractions of the AS. Slime PN (mg/g-VSS) PS (mg/g-VSS) HS (mg/g-VSS) DNA (mg/g-VSS) Zeta potential (mV)

16.34  36.533  4.59  0.13  15.1 

0.34 0.004 0.02 0.03 1.12

LB-EPS 5.12  21.11  1.12  0.036  11.6 

1.03 0.18 0.12 0.001 1.03

TB-EPS 20.25 43.89 3.27 0.28 17.9

 0.34  3.08  0.02  0.03  1.16

59

shown in Fig. 1a, in the LVE range during the SAS test, G0 was much greater than G00 , implying a gel-like behavior of sludge at low strains, while an opposite trend could be observed beyond a gc. G0 and G00 did not show any strain dependence at low strains as well (for brevity, raw AS is presented as an example). The rest of sludge samples in the whole range of investigated TSS concentration showed very similar trends. The gc values of G0 equaling to G00 for the AS before and after EPS extraction showed a decrease trend with an increase of TSS concentration (Fig. 1b). In the FS test, two moduli remained almost constant and G0 was slightly higher than G00 (Fig. 1c), indicating that the material exhibited a solid-like behavior before the fc reached at which G0 equals to G00 . The result was as similar as that of other samples. The fc values were observed and fluctuated with TSS concentration (Fig. 1d). In both dynamic rheological measurements, the loss factor, tan d ¼ (G00 /G0 ) was greater than 0.1 and smaller than 1.0 for all measured samples before which gel-sol transition reached, thus these samples can be classified into weak gels (Farahnaky et al., 2010). Weak gels show G0 higher than G00 with two modui almost parallel to each other. This type of gel exhibits a less difference between two modui and tends to flow under the high enough shear stress condition. Strong gels may remain in the LVE range over greater strain than weak gels. Entanglement network systems behave more solid behavior at higher frequencies (Clark and Ross-Murphy, 1987).

3.2.2.

Effect of TSS concentration on the gel-like structure

In the SAS test mode, G00 values were obtained at the end of the plateau (LVE range) of the complex modulus, corresponding to a state of suspension right before flow occurs (Mori et al., 2006) without damage to the floc structures of the sludge (Comte et al., 2006). Fig. 2a presents logarithmic plots of G00 as function of TSS content for the AS before and after EPS extraction. It can be seen that the G00 values increased with an increase of TSS content for all the sludge samples. AS after LB-EPS extraction exerted much higher and the three other sludge samples showed minor difference among each other in terms of the G00 values, which is consistent with the result obtained from our previous study (Yuan and Wang, 2013). Following this order, the enhancement in the elasticity of sludge may be obtained after LB-EPS extraction. Additionally, good linear relationships were found for lg G00 -lg TSS, the determination coefficients of the linear regression were greater than 0.80 (R2 > 0.80). The effect of TSS content on the gel-like structure of different EPS fractions was investigated. The result of the existence of the LVE range, 0.1 < tan d < 1.0 before the gel-sol transition reached, was also observed at higher TSS content in the SAS testing for different EPS fractions. Thus, the EPS behaved as weak gels as well (Farahnaky et al., 2010). The G00 values of different EPS fractions are depicted in Fig. 2b. As can be observed, G00 showed a logarithmic increase as lg TSS increased, and the parameter increased gradually for LB-EPS, and increased dramatically for two other samples, especially for TB-EPS. At relatively small TSS concentrations, the ranking order of G00 was LB-EPS > slime > TB-EPS. However, the order was opposite above which the critical TSS content appeared. In addition, the G00 values correlated well with the TSS content (R2 > 0.81). In the process of sludge treatment, the sludge

60

w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 5 6 e6 5

3.0

Raw sludge Sludge after slime extraction Sludge after LB-EPS extraction Sludge after TB-EPS extraction

(b)

60 55 50

2.5

45 γc(%)

lgG* or lgG' or lgG'' (Pa)

65

G* G' G''

(a)

Raw sludge

2.0

40 35 30 25 20

1.5 -2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

15

2.0

20

30

40

50

lgγ 6

Raw sludge

90

100

(d) 50

0

40

-2

30

-4

20

-6

10

Raw sludge Sludge after slime extraction Sludge after LB-EPS extraction Sludge after TB-EPS extraction

0

-0.5

80

60

2

-8 -1.0

70

70

(c)

fc(Hz)

lgG' or lgG'' (Pa)

4

G' G''

60

TSS (g/L)

0.0

0.5

1.0

1.5

2.0

20

2.5

30

40

50

lgf

60

70

80

90

100

TSS (g/L)

Fig. 1 e FS and SAS tests results. (a) and (c) typical rheograms TSS [ (46.82 ± 0.08) g/L; (b) and (d) the gc and fc for AS before and after EPS extraction as a function of TSS (g is the relative shear strain, and its value is the percentage of the shear strain amplitude).

would be condensed at higher concentrations. In that case, TB-EPS and sludge with TB-EPS showed the strongest elasticity, which may reveal the crucial role of TB-EPS in the gellike property of the sludge. The subsequent research should focus on the effect of individual component of TB-EPS on the gel-like property to find out the key component. It has been reported that PS (exopolysaccharides or glycosides) are the major gel-forming components of granlular EPS (Seviour et al., 2009a,b). Seviour et al. (2010) described the gel-forming exopolysaccharide as a complex heteropolysaccharide consisting of eight sugar residues, which was called Granulan (Seviour

4.5

lgG0'(Pa)

4.0 3.5

Raw sludge Sludge after slime extraction Sludge after LB-EPS extraction Sludge after TB-EPS extraction

y=7.0851x-8.4131 2 R =0.9916

3.0

4.0 3.5

y=3.5072x-3.3025 2 R =0.8005 y=3.9107x-4.1501 2 R =0.9989

2.5 2.0 1.5 1.4

4.5

(a)

lgG0'(Pa)

5.0

1.6

1.7

lgTSS (g/L)

1.8

slime LB-EPS TB-EPS

3.0 2.5 y=3.8063x-3.6767 2 R =0.9768

1.5

1.9

2.0

(b)

y=1.1140x+1.2532 2 R =0.8117

2.0

y=4.1912x-4.8146 2 R =0.9524 1.5

et al., 2011). While Lin et al. (2010) stated that it is alginatelike, which is consistent with the finding of Dursun (2007) that the synthetic slurry consisting of alginate showed gellike property. In addition, Seviour et al. (2012) found that the interactions between branches, carboxyl group of the terminal hexuronic acid of the sugar branches, were mediated by Ca2þ and could strengthen the gel-like structure. The gel structure of EPS could result in fairly tenacious retention of water within the sludge (Dursun, 2007). Hence, the TB-EPS fraction would probably highly influence sludge dewatering and drying. However, in a published study, Li and

1.0 1.2

1.4

1.6

1.8

y=6.9348x-11.5155 2 R =0.9376

2.0

2.2

2.4

lgTSS (g/L)

Fig. 2 e Evolution of G00 with TSS content for AS before/after EPS extraction and EPS fractions.

w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 5 6 e6 5

Yang (2007) demonstrated that the performance of sludgewater separation was much more closely correlated with the amount of LB-EPS than TB-EPS. The discrepancy might be attributed in part to the differences in EPS extraction and analytical methods. Specifically, in the previous study, the EPS was extracted by the heat method and the decanted part by 4000g for 5 min corresponded to summation of the slime and parts of LB-EPS in this study, while the centrifugation and ultrasound method was used in this study. In addition, the influence of the EPS on the sludge dewatering was investigated by the statistical analysis on the correlations between the sludge-water separation and the LB- and TB-EPS contents in the published study, but investigated by the research on the gel-like structure of sludge and EPS in this study. According to the established dewatering technique and the water distribution in sludge from a report by Zhou et al. (2001), we can know that the conventional dewatering technique was only able to remove some water in slime and LB-EPS fractions. Additionally, the rate of water content in the thickened sludge is around 95%. After mechanical dewatering, the rate of water content will decrease to 70e80%. The rate of water content would be 10e20% after sludge drying (Ma, 2008). According to our further research, the rate of water content in different EPS fractions was obtained (unpublished data). The sum of the water content rate (g/g-wet sludge) from slime and LB-EPS was approximately 74%, among the range of 70e80%. This implies that the mechanical dewatering process may not completely remove water from the LB-EPS fraction, and it seems that the mechanical dewatering process did not involve the water

61

contained in the TB-EPS fraction. So to be exact, the TB-EPS layer may be a key fraction that influences the drying process.

3.2.3.

Enzyme degradation

Fig. 3 shows the G00 variation in response of incubation time for AS before and after EPS extraction. At incubation time 0, four sludge samples exhibited initial G00 of 936.5  67.2 Pa for raw AS, 716.5  48.8 Pa for AS after slime extraction, 1115.0  49.5 Pa for AS after LB-EPS extraction, and 728.5  50.2 Pa for AS after TB-EPS extraction, respectively. It’s also possible to observe that the degradation profiles were generally typically biphasic. The first part of the profiles showed a slight decrease of G00 (decrease down to minimal value) from the incubation time of 0e48 h, which may be due to the fact that the dissolution of fractions created by enzymatic hydrolysis (Dalev et al., 1998; Dursun, 2007) could reduce the gel-like behavior of the EPS or sludge. Then slight increase in G00 were observed with increasing incubation time, which may be due to the fact that the enhancement in the gellike property resulting from mineral phase (Ca2þ, Mg2þ) association with degradation products from EPS (Dursun, 2007), is advantageous over the depression by enzyme degradation. Regarding these phenomena, it was very likely that no collapse of gel structure occurred during the incubation period. A sudden rupture of the sludge into a discrete number of fragments should result in a dramatic reduction in G00 (gellike behavior). Apparently, this was not the case, which confirms indirectly that degradation of sludge samples is a surface-controlled mechanism. Additionally, the control

Fig. 3 e Time response of G00 for AS before/after EPS extraction with different enzymes.

62

w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 5 6 e6 5

samples (enzyme-untreated) showed a lower G00 loss for raw AS (around 25.7%) in comparison with AS samples after EPS extraction (around 46.6%, 56.1%, and 40.5%, respectively) at the incubation time of 48 h. For AS after TB-EPS extraction, the small particles out of the microorganisms may occur although the integrity of the floc is maintained (Hamdi et al., 1998). Then the higher G00 loss for AS after EPS extraction may indicate that it’s more prone to release the soluble low-molecularmass degradation products under the influence of the built-in enzymes for AS after EPS extraction than for raw AS. For raw AS and the AS after LB-EPS extraction, the drop in G00 for most of the enzyme-treated samples was greater than that in the control, while opposite behaviors occurred in the two other sludge samples during the incubation time of 48 h. This may be due to the fact that diffusion of enzymes into the sludge samples was strongly decreased due to steric hindrance for AS after slime and TB-EPS extraction. The phenomenon may indicate that the enzymes as the polymer may have a positive influence on the gel-like behavior of the AS after slime and TB-EPS extraction. For AS after LB-EPS extraction (enzyme e treated and e untreated), a dramatic reduction in G00 was observed in the first 24 h, after which reduction continued at a slower rate at the incubation time 48 h. The most dramatic reduction in G00 was observed in the presence of DNase up to the incubation time 48 h, compared to other enzyme treatment, for raw AS and AS after TB-EPS extraction, amounting to 63% and 40.4%, respectively. However, as can been seen from Table 2, the DNA contents were majorly presented in TB-EPS, followed by slime. This could indicate that the DNA had the greatest impact in the gel-like behavior in the raw AS, but had no much effect for the AS after LB-EPS extraction. Generally, PN and PS have been proved to play an important role in the network of sludge (Adav et al., 2008; Seviour et al., 2009a; Forster, 1983). Adav et al. (2008) found that the b-polysaccharides formed the backbone of a network-like outer layer with embedded proteins, lipids, a-polysaccharides, and cells to support the mechanical stability of granules. Proteins and a-polysaccharides were confirmed to be the major granule structural materials in the study of Seviour et al. (2009a). Forster (1983) summarized that PS, and PN were important in determining rheological characteristics of AS. In the present study, aside from the effect of the DNase on some sludge samples, the hydrolysis of PN and PS also led to the loss of G00 to some extent. After 48 h hydrolysis, the G00 loss reached minimal, approximately 66.4% (in the presence of a-amylase) for raw AS, 79.2% (in the presence of b-amylase) for AS after slime extraction, 32.5% and 35.8% (in the presence of protease and b-amylase, respectively) for AS after TB-EPS extraction. While for AS after LB-EPS extraction, the minimal value was 74.8% and 74.3% (in the presence of a- and bamylase, respectively) at the incubation of 24 h. This meant

that the effective enzymes such as protease or amylase differed in the AS samples with different EPS fractions and chemical structures within different EPS fractions showed somewhat differences.

3.3.

Fractal structure

The gel networks of non-Newtonian fluids are considered to be a collection of closely packed fractal flocs/aggregates (Vreeker et al., 1992). The Df values of the AS before/after EPS extraction and different EPS fractions deduced from SAS test are presented in Table 3. In the SAS test, provided that samples exhibited strong linking behaviors, the calculated Df values were less than 2.0 or higher than 3.0 (data not shown) though, except for TB-EPS (strong-link: Df ¼ 2.37). This result does not conform to the 3D structure of the aggregates and implies that the AS obtained before/after EPS extraction, slime and LB-EPS are dominated by the weak - links region, in which the strength of the link between flocs or polymer is weaker than that within flocs or polymer. Both G00 and the gcD in the LVE range increased with increasing TSS content at most TSS content for all samples. Thus, the AS obtained before/after EPS extraction, slime and LB-EPS, even TB-EPS can be considered as weak-link gels (Shih et al., 1990). The ranking order of the Df derived from Fig. 2 is AS obtained after LB-EPS extraction (2.86) > AS obtained after slime extraction (2.76) zAS obtained after TB-EPS extraction (2.74) z raw AS (2.71). The result indicates that the AS obtained after LB-EPS extraction had a relatively compact and dense gel-like network structure in which the porous configuration was more compressed and collapsed. On the other hand, the Df values of EPS fractions were 2.74 for slime, 2.10 for LB-EPS, and 2.86 for TB-EPS, which implies that TB-EPS presents the most compact and denser structure, followed by slime. The same degree of fractal structure between the AS after LB-EPS extraction and TB-EPS may indicate the key role of TB-EPS structure to the fractal structure of the sludge. This result may be attributed to the retention of the TB-EPS structure of a certain shape and its tight and stable binding to the cell surface (Wingender et al., 1999).

4.

Discussion

Yu et al. (2009) believed that EPS in sludge flocs could be subdivided into soluble EPS (slime) and bound EPS. The slime fraction is weakly bound to cells and dispersed, moving freely among the flocs, and readily removed by washing. In contrast, bound EPS represents sludge fractions strongly bound to cells with a distinct margin outside the cell wall. These two layers are highly similar to the double layer of the ionic: the stern layer and the diffusion layer. Thus, the slime and bound EPS layers are assumed to be the dispersed layer and stable layer,

Table 3 e Mass fractal dimensions of AS before/after EPS extraction and EPS fractions. Mass fractal dimension lgG00 elg TSS

Raw sludge

Sludge after slime extraction

Sludge after LB-EPS extraction

Sludge after TB-EPS extraction

2.71

2.76

2.86

2.74

Slime LBEPS 2.74

2.10

TBEPS 2.86

w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 5 6 e6 5

respectively. The latter exhibits a double-layered EPS structure, composed of LB- and TB-EPS (Wingender et al., 1999; Li and Yang, 2007). The LB-EPS may function as the primary surface for cell attachment and flocculation (Li and Yang, 2007), while the TB-EPS layer has a specific shape and is bound tightly and stably to the cell surface (Wingender et al., 1999). Li and Ganczarczyk (1990) confirmed the highly porous fractal-like interior structure of the AS flocs. Snidaro et al. (1997) reported that the size of the bacterium was around 2.5 mm. Bacteria were enmeshed in the gel-like matrix of EPS, forming a microcolony of around 13 mm. These microcolonies were then linked together by EPS to form the sludge floc of about 125 mm. Snidaro et al. (1997) also found that the microcolony was confirmed to be a fractal object with a mass fractal dimension of around 3. Therefore, a microcolony can be adopted as the fundamental unit of a fractal floc. The single bacterium level, microcolony level and the sludge floc level are illustrated in Fig. 4. At the sludge floc level, when sludge flocs approach one another, the two layers of the EPS from sludge flocs compress because of the high sludge floc concentration, according to the well-known DLVO theory. Sludge flocs then form into the aggregate. As the separation distance x reaches a certain value, Van der Waals interactions exceed electrostatic interactions, and the sum of the Van der Waals interaction energy (VA) and the electrostatic interaction energy (VR), the total interaction energy (VT), approaches the second minimum energy (the second Vmin). These flocs are bound to each other reversibly, as predicted by the second Vmin flocculation theory. The model in the present study is different from those reported by Li and Logan (1997) and Gregory (1997) whose models were ideal and had spheres in orderly rows; the model in the present work represents actual scenarios. According to results obtained from the SAS test mode, critical strains of the G00 exceeding the G0 were obtained

63

beyond the LVE range (data not displayed). That G0 > G00 was shown in the LVE range, indicating that the sludge samples had a gel structure and exhibited the behavior of a VE solid (Mezger, 2006). Mezger (2006) reported that the behavior of VE solids can be illustrated using the Kelvin/Voigt model with a spring and dashpot in parallel connection. The spring expresses G0 and the dashpot expresses the viscous behavior. The AS before and after EPS extraction have been demonstrated to be in the weak-link regime, in which the links between flocs (interfloc links) have a lower elastic constant than those of flocs (Shih et al., 1990). The red spring represents the link within the flocs and is longer than the blue spring, representing the links between the flocs. The longer the spring, the stronger the link. A weak-link structure model of the aggregate is displayed in Fig. 4. According to the Df obtained from the rheological measurement, the geometric structure of AS after LB-EPS extraction was relatively more compact and dense while the three other sludge samples showed almost the same degrees of compactness and density. Here, the two-plate model is proposed to explain the SAS test results (Mezger, 2006). The upper plate is deflected by the shear force while the lower plate is immovable. The sample is sheared in the gap between the plates, and the angle of rotation is 90 (g ¼ gA). The elastic deformation and viscous behavior are described as G00 and h, respectively, the order of which is AS after LB-EPS extraction > raw AS > AS after TB-EPS extraction > sludge AS after slime extraction. This order indicates that AS obtained after LB-EPS extraction exhibited stronger gel-like behavior.

5.

Conclusions

AS after LB-EPS extraction and the TB-EPS fraction showed higher G00 and Df, suggesting that sludge obtained after LB-EPS

Fig. 4 e Structual models of AS before and after EPS extraction.

64

w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 5 6 e6 5

extraction and TB-EPS exhibited a stronger gel-like and more compact structure. TB-EPS may be the key fraction to the gellike and dyring process of AS. AS samples before and after EPS extraction behaved as weak-link gels. The minimal G00 loss occurred in the presence of a-amylase for raw AS, of bamylase for AS after slime extraction, of protease and bamylase for AS after TB-EPS extraction at the incubation of 48 h, while in the presence of a- and b-amylase for AS after LBEPS extraction at the incubation of 24 h. Protease or amylase showed different activity in the sludge samples with different EPS fractions.

Acknowledgments This work was supported by the National Natural Science Foundation of China (Nos. 51078035 and 21177010), the Fundamental Research Funds for the Central Universities (Nos. JC2011-1 and TD2010-5), and the Ph.D. Programs Foundation of the Ministry of Education of China (No. 20100014110004).

references

Adav, S.S., Lee, D.J., Tay, J.H., 2008. Extracellular polymeric substances and structural stability of aerobic granule. Water Res. 42 (6e7), 1644e1650. APHA, AWWA, WEF, 2005. Standard Methods for the Examination of Water and Wastewater, twenty-first ed. (Washington DC, USA). Ayol, A., Dentel, S.K., Filibeli, A., 2006. Toward efficient sludge processing using novel rheological parameters: dynamic rheological testing. Water Sci. Technol. 54 (5), 17e22. Clark, A., Ross-Murphy, S., 1987. Biopolymers. Springer, Berlin Heidelberg, pp. 57e192. Comte, S., Guibaud, G., Baudu, M., 2006. Relations between extraction protocols for activated sludge extracellular polymeric substances (EPS) and EPS complexation properties. Enzyme Microb. Technol. 38 (1e2), 237e245. Dalev, P., Vassileva, E., Mark, J.E., Fakirov, S., 1998. Enzymatic degradation of formaldehyde-crosslinked gelatin. Biotechnol. Tech. 12 (12), 889e892. Dignac, M.F., Urbain, V., Rybacki, D., Bruchet, A., Snidaro, D., Scribe, P., 1998. Chemical description of extracellular polymers: implication on activated sludge floc structure. Water Sci. Technol. 38 (8), 45e53. Dursun, D., 2007. Gel-like Behavior of Biosolids in Conditioning and Dewatering Processes. University of Delaware. Farahnaky, A., Askari, H., Majzoobi, M., Mesbahi, G., 2010. The impact of concentration, temperature and pH on dynamic rheology of psyllium gels. J. Food Eng. 100 (2), 294e301. Forster, C.F., 1983. Bound water in sewage sludges and its relationship to sludge surfaces and sludge viscosities. J. Chem. Technol. Biotechnol. 33 (1), 76e84. Gregory, J., 1997. The density of particle aggregates. Water Sci. Technol. 36 (4), 1e13. Hamdi, G., Ponchel, G., Duchene, D., 1998. An original method for studying in vitro the enzymatic degradation of cross-linked starch microspheres. J. Control. Release 55 (2e3), 193e201. Jin, B., Wile´n, B.W., Lant, P., 2003. A comprehensive insight into floc characteristics and their impact on compressibility and settleability of activated sludge. Chem. Eng. J. 95 (1), 221e234.

Jorand, F., Boue-Bigne, F., Block, J.C., Urbain, V., 1998. Hydrophobic/hydrophilic properties of activated sludge exopolymeric substances. Water Sci. Technol. 37 (4), 307e315. Keiding, K., Wybrandt, L., Nielsen, P.H., 2001. Remember the water e a comment on EPS colligative properties. Water Sci. Technol. 43 (6), 17e23. Legrand, V., Hourdet, D., Audebert, R., Snidaro, D., 1998. Deswelling and flocculation of gel networks: application to sludge dewatering. Water Res. 32 (12), 3662e3672. Lewin, R.A., 1956. Extracellular polysaccharides of green algae. Can. J. Microbiol. 2 (7), 665e672. Li, D.H., Ganczarczyk, J.J., 1990. Structure of activated sludge floes. Biotechnol. Bioeng. 35 (1), 57e65. Li, X.Y., Logan, B.E., 1997. Collision frequencies of fractal aggregates with small particles by differential sedimentation. Environ. Sci. Technol. 31 (4), 1229e1236. 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. Lin, Y., de Kreuk, M., van Loosdrecht, M.C.M., Adin, A., 2010. Characterization of alginate-like exopolysaccharides isolated from aerobic granular sludge in pilot-plant. Water Res. 44 (11), 3355e3364. Liu, Y.Q., Liu, Y., Hay, J.H., 2004. The effects of extracellular polymeric substances on the formation and stability of biogranules. Appl. Microbiol. Biotechnol. 65 (2), 143e148. Ma, X.W., 2008. Research on Drying Characteristics and Technology of Sewage Sludge. Zhejiang Univerisy (In Chinese). Mezger, T.G., 2006. The Rheology Handbook: for Users of Rotational and Oscillatory Rheometers. Vincentz Network. Mikkelsen, L.H., Keiding, K., 2002. Physico-chemical characteristics of full scale sewage sludges with implications to dewatering. Water Res. 36 (10), 2451e2462. Moreno, J., Vargas, M.A., Madiedo, J.M., Munoz, J., Rivas, J., Guerrero, M.G., 2000. Chemical and rheological properties of an extracellular polysaccharide produced by the cyanobacterium Anabaena sp. ATCC 33047. Biotechnol. Bioeng. 67 (3), 283e290. Mori, M., Seyssiecq, I., Roche, N., 2006. Rheological measurements of sewage sludge for various solids concentrations and geometry. Process Biochem. 41 (7), 1656e1662. Mu, Y., Yu, H.Q., 2006. Rheological and fractal characteristics of granular sludge in an upflow anaerobic reactor. Water Res. 40 (19), 3596e3602. Nielsen, P.H., Frolund, B., Keiding, K., 1996. Changes in the composition of extracellular polymeric substances in activated sludge during anaerobic storage. Appl. Microbiol. Biotechnol. 44 (6), 823e830. Poxon, T.L., Darby, J.L., 1997. Extracellular polyanions in digested sludge: measurement and relationship to sludge dewaterability. Water Res. 31 (4), 749e758. Raszka, A., Chorvatova, M., Wanner, J., 2006. The role and significance of extracellular polymers in activated sludge. Part I: literature review. Acta hydrochim. Hydrobiol. 34 (5), 411e424. Sanin, F.D., 2002. Effect of solution physical chemistry on the rheological properties of activated sludge. Water SA 28 (2), 207e212. Seviour, T., Pijuan, M., Nicholson, T., Keller, J., Yuan, Z.G., 2009a. Understanding the properties of aerobic sludge granules as hydrogels. Biotechnol. Bioeng. 102 (5), 1483e1493. Seviour, T., Pijuan, M., Nicholson, T., Keller, J., Yuan, Z.G., 2009b. Gel-forming exopolysaccharides explain basic differences between structures of aerobic sludge granules and floccular sludges. Water Res. 43 (18), 4469e4478. Seviour, T., Lambert, L.K., Pijuan, M., Yuan, Z.G., 2010. Structural determination of a key exopolysaccharide in mixed culture

w a t e r r e s e a r c h 5 6 ( 2 0 1 4 ) 5 6 e6 5

aerobic sludge granules using NMR Spectroscopy. Environ. Sci. Technol. 44 (23), 8964e8970. Seviour, T., Lambert, L., Pijuan, M., Yuan, Z., 2011. Selectively inducing the synthesis of a key structural exopolysaccharide in aerobic granules by enriching for Candidatus“Competibacter phosphatis”. Appl. Microbiol. Biotechnol. 92 (6), 1297e1305. Seviour, T., Malde, A.K., Kjelleberg, S., Yuan, Z., Mark, A.E., 2012. Molecular dynamics unlocks atomic level self-assembly of the exopolysaccharide matrix of water-treatment granular. Biomacromolecules 13 (6), 1965e1972. Shih, W.H., Shih, W.Y., Kim, S.I., Liu, J., Aksay, I.A., 1990. Scaling behavior of the elastic properties of colloidal gels. Phys. Rev. A 42 (8), 4772e4779. Shoemaker, C.F., Nantz, J., Bonnans, S., Noble, A.C., 1992. Rheological characterization of dairy-products. Food Technol. 46 (1), 98e104. Snidaro, D., Zartarian, F., Jorand, F., Bottero, J.Y., Block, J.C., Manem, J., 1997. Characterization of activated sludge flocs structure. Water Sci. Technol. 36 (4), 313e320. Steffe, J.F., 1996. Rheological Methods in Food Process Engineering. Freeman Press. Vreeker, R., Hoekstra, L.L., den Boer, D.C., Agterof, W.G.M., 1992. The fractal nature of fat crystal networks. Colloids Surf. 65 (2e3), 185e189. Wang, Y.L., Lu, J., Du, B.Y., Shi, B.Y., Wang, D.S., 2009. Fractal analysis of polyferric chloride-humic acid (PFC-HA) flocs in

65

different topological spaces. J. Environ. Sci. (China) 21 (1), 41e48. Wang, Y.L., Dieude-Fauvel, E., Dentel, S.K., 2011. Physical characteristics of conditioned anaerobic digested sludge e a fractal, transient and dynamic rheological viewpoint. J. Environ. Sci. (China) 23 (8), 1266e1273. 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. Wingender, J., Neu, T.R., Flemming, H.C., 1999. Microbial Extracellular Polymeric Substances: Characterization, Structures and Function. Springer-Verlag, Berlin Heidelberg. Wu, R.M., Lee, D.J., Waite, T.D., Guan, J., 2002. Multilevel structure of sludge flocs. J. Colloid Interface Sci. 252 (2), 383e392. 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. Yu, G.H., He, P.J., Shao, L.M., 2009. Characteristics of extracellular polymeric substances (EPS) fractions from excess sludges and their effects on bioflocculability. Bioresour. Technol. 100 (13), 3193e3198. Yuan, D.Q., Wang, Y.L., 2013. Influence of extracellular polymeric substances on rheological properties of activated sludge. Biochem. Eng. J. 77 (15), 208e213. Zhou, J.X., Liu, Z., She, P., Ding, F.X., 2001. Water removal from sludge in a horizontal electric field. Drying Technol. 19 (3e4), 627e638.