See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/280910343

Dewaterability of five sewage sludges in Guangzhou conditioned with

Fenton's reagent/lime and pilot-scale experiments using ultrahigh pressure
filtration system

Article  in  Water Research · July 2015

DOI: 10.1016/j.watres.2015.07.041 · Source: PubMed

CITATIONS READS

31 272

5 authors, including:

Jialin Liang
GuangDong University of Technology
14 PUBLICATIONS   124 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

sludge dewatering View project

All content following this page was uploaded by Jialin Liang on 04 November 2018.

The user has requested enhancement of the downloaded file.

 Water Research 84 (2015) 243e254

Contents lists available at ScienceDirect

Water Research
journal homepage: www.elsevier.com/locate/watres

Dewaterability of five sewage sludges in Guangzhou conditioned with

Fenton's reagent/lime and pilot-scale experiments using ultrahigh
pressure filtration system
Jialin Liang a, b, 1, Shaosong Huang a, b, 1, Yongkang Dai a, b, Lei Li a, b, Shuiyu Sun a, b, *
a
School of Environmental Science and Engineering, Guangdong University of Technology, Guangzhou 510006, China
b
Key Laboratory of Resources Comprehensive Utilization and Cleaner Production, Guangdong University of Technology, Guangzhou 510006, China

a r t i c l e i n f o a b s t r a c t

Article history: Sludge conditioning with Fenton's reagent and lime is a valid method for sludge dewatering. This study
Received 22 May 2015 investigated the influence of different organic matter content sludge on sludge dewatering and discussed
Accepted 23 July 2015 the main mechanism of sludge conditioning by combined Fenton's reagent and lime. The results indi-
Available online 29 July 2015
cated that the specific resistance to filterability (SRF) of sludge was reduced efficiently by approximately
90%, when conditioned with Fenton's reagent and lime. Through single factor experiments, the optimal
Keywords:
conditioning combinations were found. In addition, the relationship between VSS% and consumption of
Sewage sludges
the reagents was detected. Furthermore, it was also demonstrated that the SRF and filtrate TOC values
Dewaterability
Fenton's reagent
had a significant correlation with VSS% of sludge (including raw and conditioned). The main mechanism
Lime of sludge dewatering was also investigated. Firstly, it revealed that the dewaterability of sludge was
Ultrahigh pressure filtration closely correlated to extracellular polymeric substances (EPS) and bound water contents. Secondly, the
results of scanning electron microscopy (SEM) stated that sludge particles were to be smaller and thinner
after conditioning. And this structure could easily form outflow channels for releasing free water.
Additionally, with the ultrahigh pressure filtration system, the water content of sludge cake conditioned
with Fenton's reagent and lime could be reduced to below 50%. Moreover, the economic assessment
shows that Fenton's reagent and lime combined with ultrahigh pressure filtration system can be an
economical and viable technology for sewage sludge dewatering. Finally, three types of sludge were
classified: (1) Fast to dewater; (2) Moderately fast to dewater; (3) Slow to dewater sludge.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction
Municipal wastewater treatment plants (WWTPs) produce
large amounts of sludge commonly containing over 95% water.
Large amount of sludge has been a great trouble and raised
Abbreviations: DS, dry solids content; PSs, polysaccharide; PNs, protein; RS, raw
sludge; CST, capillary suction time; DSC, differential scanning calorimetry; EPS,
extracellular polymeric substances; SEM, scanning electron microscope; SRF, spe-
cific resistance to filterability; TOC, total organic carbon; TSS, total suspended
solids; VSS, volatile suspended solids; SCOD, soluble chemical oxygen demand;
TCOD, total chemical oxygen demand; WWTPs, wastewater treatment plants; VSS/
TSS%, organic matter contents.
* Corresponding author. School of Environmental Science and Engineering,
Guangdong University of Technology, Guangzhou 510006, China.
E-mail address: sysun@gdut.edu.cn (S. Sun).
1 tions (Qi et al., 2011).
Co-first authors contributed equally to this work.
significant concerns in China (Yang et al., 2015). To reduce the cost
of treatment and disposal, it is necessary to reduce the sludge
volume, making reduction of sludge produced and improving the
dewaterability of paramount importance (Colin and Gazbart,
1995; Neyens and Baeyens et al., 2003a, 2003b; Zhang and
Yang, 2014; Ning and Chen, 2014). Sludge dewatering is gener-
ally achieved through pretreatment with an inorganic coagulant
or a synthetic polyelectrolytes, such as polyacrylamide (PAM),
followed by centrifugation or filter pressing. However, these
methods can only achieve 75e80% water removal. Furthermore,
the PAM is expensive and can cause secondary pollution (Chang
and Abu-Orf, 2005; Ho and Norli, 2010; Citeau and Larue, 2011).
Physical conditioners including coal fine, fly ash and gypsum are
commonly used to reduce the compressibility of sludge. But more
often the addition of a physical conditioner follows coagulation or
flocculation with a chemical conditioner. So it has some limita-

http://dx.doi.org/10.1016/j.watres.2015.07.041
0043-1354/© 2015 Elsevier Ltd. All rights reserved.
244 J. Liang et al. / Water Research 84 (2015) 243e254
However, Neyens and Baeyens et al. (2003a, 2003b) advocated
over a decade ago that Fenton oxidation was effective for dew-
atering. Fenton's reagent is a mixture of ferrous iron, and H2O2 with
the former catalyzing the decomposition of the latter, to yield
highly reactive hydroxyl radicals. The generated radicals react on
the surface of the sludge, subsequently reducing the volume of the
sludge. Since then, numerous researchers have studied sludge
detwatering by Fenton oxidation. For instance, Buyukkamaci
(2004) reported that sludge dewatering was optimal at dosages
of 5000 mg/L Fe2þ and 6000 mg/L H2O2. Dewil and Baeyens (2005)
stated that Fenton peroxidation improved the dewaterability and
drying of waste activated sludge. Debowski and Zielinski (2008)
showed that the application of the Fenton method to the condi-
tioning process enhanced the sludge. Recently, Liu and Yang (2012)
found that sludge conditioning with a combination of Fenton's
reagent and skeleton builders was an efficient means to achieve
deep dewatering. Fenton treatment resulted in partial destruction
of extracellular polymeric substances (EPS) and decreases in sludge
floc size. Subsequently, lime and Portland cement were added to
serve as skeleton builders and transmit the stress to the internal
parts of floc and provided channels for water release under high
pressures (Liu and Yang, 2013). Therefore, sludge conditioning with
Fenton's reagent and skeleton builders has been shown to be a valid
method for sludge dewatering.
Recently, the mechanism of sludge dewatering has become the
focus of the researches in this area. Many investigations have been
conducted on the role of EPS in bioflocculation and dewaterability.
Li and Yang (2007) demonstrated that EPS was essential to sludge-
floc formation. However, excessive EPS could weaken cell attach-
ment and the floc structure, resulting in limited dewaterability. EPS
produced by sludge microorganisms played a definite role in sludge
flocculation (Bala Subramanian and Yan, 2010). From these, sludge
dewaterability is related to EPS content. Alternatively, some re-
searchers focus on the bound water. Vaxelaire and Ce zac. (2004)
reported that the behavior of a molecule of water during the
dewatering process was widely dependent on its proximity to the
solid. Bound water was generally considered a gross estimate of
several states of water including vicinal (or surface) water, water of
hydration and interstitial water. Zhang and Yang (2014) clearly
showed that bound water was released and converted into free
water due to the degradation of EPS. Thus, mechanistic study of the
sludge dewatering would be beneficial to sludge dewatering tech-
nological advances.
Pilot-scale experiments are used to verify the laboratory results.
Zhang and Yang (2015) and Liu and Yang (2013) operated the pilot-
scale filter press dewatering experiment for Fenton treatment.
However, these experiments might lack of representativeness due to
the limitations of their dewatering equipment. In previous studies,
the pressure of the common pressure filtration was lower than 2Mpa
and it did not reach the high pressure requirements, which will cause
dewatering incompletely. In order to simulate the real industrial
condition and obtain better results, a novel spring pressure filtration
was utilized in the pilot-scale experiments. The ultrahigh pressure
filtration could achieve pressure to 40Mpa, which further improved
the sludge dewatering rate and dewatering efficiency.
Generally, the organic content of sludge differs between
WWTPs. Although many experiments have proved that Fenton's
reagent and skeleton builders are effective for sludge dewatering,
but few researchers studied the consistency of this dewatering ef-
fect across sludges of various composition. Furthermore, many
previous studies indicated that sludge dewaterability could be
greatly improved via peroxidation, but the responsible mechanism
was not fully reported. Hence, this study systematically researched
dewaterability of variously composed sewage sludge via condi-
tioning with a combination of Fenton's reagent and lime. Specific
objectives of this study were to: (1) conduct single factor tests in
sludges containing different levels of organic matter conditioned
with Fenton's reagent and lime to find out the optimal conditioning
combination; (2) get more comprehensive insights into the
respective roles of reagents in sludge containing different amounts
of organic matter and research the inner connection between VSS%
and SRF or TOC; (3) investigate the dewatering mechanism of the
EPS, bound water content, wastage rate (iron, calcium) and the
morphological structure of the sludge; (4) conduct pilot-scale ex-
periments with ultrahigh pressure filtration system and verify the
feasibility of these sludge treatment technology; (5) show a sche-
matic changes in sludge flocs treated with Fenton's reagent/lime
and then construct different types of sludge models.
2. Materials and methods
2.1. Materials
2.1.1. Raw sludge
Raw sludge (RS) samples were collected from the secondary
sedimentation tanks of five different municipal WWTPs (GZ-1, GZ-
2, GZ-3, GZ-4 and GZ-5) across Guangzhou, China from August to
October, 2014 (Fig. S1). The various WWTPs are summarized in
Table 1. Before being used in the tests, the sludge sample was
settled by gravity for about 12 h to increase the concentration
(water content about 95%). Since the characteristics of sludge are
volatile, two batches of fresh sludge were used from each plant. All
the experiments were completed within 3 days and the sludge was
placed in a freezer at 4  C. The main characteristics of the RS are
listed in Table 2.
In the WWTPs of China, A/A/O process is a widely used tech-
nology and the ratio of WWTPs that choose A/A/O was 25.45% (Jin
and Zhang, 2014). Therefore, we researched sludge samples of
municipal WWTPs, which chose A/A/O process. And the combined
total effluent volume in WWTPs of GZ-1, GZ-2 and GZ-3 accounts
for more than 50% of the total wastewater volume from Guangzhou.
In order to obtain more information about other types of sludge in
Guangdong that makes the research comprehensively, the other
two sludge types (GZ-1 and GZ-5) were also collected separately.
As shown in Table 2, the sludge samples (GZ-1, GZ-2, GZ-3, GZ-4
and GZ-5) represent different organic matter content sludge. The
reason of the difference might be associated with the influent
CODCr concentration (Table 1). Municipal WWTPs (GZ-1) was
located in suburbs of Guangzhou and the different influent CODCr
was relatively low. Municipal WWTPs (GZ-5) was located in Huadu,
which was the automobile manufacturing center, so the influent
CODCr was high because of 10%e15% industrial wastewater. And
municipal WWTPs (GZ-2, GZ-3, GZ-4) were located in urban center,
which almost 98% of the wastewater was domestic, so the influent
CODCr concentration was relatively high.
2.1.2. Conditioner
In Fenton experiments, H2SO4 (analytical grade, Guangzhou
Chemical Reagent Factory, China) was used to adjust the initial pH
of sludge. Fenton's reagent was prepared by mixing a solution of
FeSO4$7H2O (50 wt.%) (analytical grade, Damao Chemical Reagent
Factory, China.) with H2O2 (30 wt.%) (analytical grade, Guangzhou
Chemical Reagent Factory, China). Lime (analytical grade, Tianjin
Fuchen Chemical Reagents Factory, China) was used as a neutralizer
and also a skeleton builder.
2.2. Sludge conditioning and dewatering
2.2.1. Laboratory-scale investigations
Initially, 300 mL of sludge samples were carefully transferred to
 J. Liang et al. / Water Research 84 (2015) 243e254 245

Table 1
Description of the WWTPs.

WWTPs name Influent CODCr Source of wastewater Description of wastewater Wastewater treatment Processesa Daily sewage sludge
concentration (mg/L) capacity (m3/d) production (t/d)

GZ-1 100e160 Domestic > 98% Rural wastewater 8.00  104 A/A/O 80
GZ-2 180e240 Domestic > 98% City center wastewater 40.00  104 A/A/O 300
GZ-3 200e270 Domestic > 98% City center wastewater 120.00  104 A/A/O 480
GZ-4 240e300 Domestic > 98% City center wastewater 80.00  104 A/A/O 400
GZ-5 280e360 Domestic > 85% 10%e15% automobile 10.00  104 A/A/O 110
manufacturing wastewater
a
A: Anaerobic; O: Aerobic; A/A/O: Anaerobic/Anoxic/Aerobic.

Table 2
Main characteristics of the RS.

WWTPs name Batch pH Water content (%) TSS (g/L) VSS/TSS (%) SCOD (mg/L) TCOD (mg/L) SRF(1012 m/kg)

GZ-1 1 6.82 95.50 ± 0.29 49.48 ± 0.35 23.90 ± 0.22 234.52 ± 40.86 11,600 ± 1000 1.58 ± 0.05
2 6.72 95.81 ± 0.16 48.75 ± 0.15 23.70 ± 0.16 218.67 ± ± 35.02 11,542 ± 1242 1.54 ± 0.03
GZ-2 1 6.53 94.97 ± 0.26 45.59 ± 0.68 30.97 ± 0.26 194.11 ± 21.85 12,904 ± 400 3.57 ± 0.07
2 6.51 94.72 ± 0.19 46.00 ± 0.85 30.99 ± 0.57 208.54 ± 33.93 12,054 ± 589 3.73 ± 0.09
GZ-3 1 6.53 95.47 ± 0.28 50.94 ± 0.73 38.60 ± 0.46 258.93 ± 50.68 18,453 ± 995 3.98 ± 0.10
2 6.56 95.27 ± 0.37 51.49 ± 0.35 38.46 ± 0.48 289.47 ± 15.72 19,388 ± 1000 4.18 ± 0.02
GZ-4 1 6.67 95.24 ± 0.15 45.12 ± 0.28 49.82 ± 0.82 301.80 ± 40.63 23,547 ± 2366 7.06 ± 0.15
2 6.81 95.58 ± 0.10 46.63 ± 0.82 52.22 ± 0.50 295.35 ± 22.91 22,895 ± 1420 6.82 ± 0.12
GZ-5 1 6.67 94.90 ± 0.30 47.18 ± 0.58 61.74 ± 0.63 285.65 ± 34.80 20,175 ± 1240 8.04 ± 0.12
2 6.94 95.35 ± 0.24 46.69 ± 0.44 65.26 ± 0.19 257.99 ± 30.44 23,553 ± 1841 7.80 ± 0.08

500 mL beakers and then a coagulation agitator was used to mix the
sludge. The conditioning process was as follows: (1) Sludge pH was
adjusted by adding 50% H2SO4 with stirring (30 s, 300 rpm). (2) 50%
Fe2þ solution was dosed into the reactor (180 s, 300 rpm). (3) H2O2
was added to the sludge to initiate the Fenton reaction (0e120 min,
300 rpm). (4) Following the Fenton reaction, lime was added to the
sludge (120 s, 200 rpm). Furthermore, temperature control in the
process of the experiment at 25e30  C.
In order to evaluate the effect of each component of Fenton's
reagent combined with lime on sludge dewatering performance, a
series of single-factor experiments were conducted. Single factors
included Fenton reaction time, lime dosages, initial pH, Fe2þ and
H2O2 dosages. The optimal conditions for each single factor
experiment were set as the next test conditions. The final SRF and
filtrate TOC values were the average of the three measured values.
Single-factor experiments with the five different sludges were
carried out respectively.
Sludge dewaterability was determined by SRF, filtrate TOC and
water content of sludge cakes. TOC was measured by a TOC analyzer
(TOC-VCPH, Shimadzu, Japan). The SRF was measured in accor-
dance with standard methods (Qi et al., 2011).
2PbA2
SRF ¼
mu
where SRF is the specific resistance to filtration (m/kg), P is the
filtration pressure (0.05 MPa), b is the slope of the filtrate discharge
curve (t/V versus V) (s/m6), A is the filtration area (m2), m is the
viscosity of the filtrate (Ns/m2) and u is the weight of cake solids
per unit volume of filtrate (kg/m3).
In addition, water content and pH value of sludge were
measured according to standard methods (APHA, 1998).
2.2.2. Pilot-scale experiments of ultrahigh pressure filtration system
The main characteristic of this system (Foshan JKD Filtration
Equipment Company) is the spring filter press, which is mainly
steel and can achieve the squeezing pressure up to 40 MPa. As
shown in Fig. 1b, the spring filter plate decreased from 80 mm to
65 mm by secondary compression. It could further improve sludge
dewatering efficiency and depth. The diagram of the sludge dew-
atering process in pilot scale was shown in Fig. 1a. First, raw sludge
was transferred to a 100 L conditioning tank and then mix it with
Fenton's reagent and lime. Second, conditioned sludge was pumped
into the spring filter press by a piston pump. The dewatering pro-
cess had taken place since the pressure gradually increased to
1.6 MPa. Last, after all the sludge putting into the spring filter plates,
compression process began with the hydraulic system and the
pressure increased gradually from 1.6Mpa to 40Mpa. The filtrate
was collected and weighed regularly in the process. The sludge
cakes were the final products as the pressure declined. The water
content of sludge cakes will be to assess dewatering performance.
The compression time and compression pressure was to assess
compressibility. And the filtrate rate was measured to assess
permeability.
2.3. Mechanism of sludge dewatering
2.3.1. EPS extraction and analysis
A heat-extraction method was modified to extract the LB-EPS
and TB-EPS from the sludge (Li and Yang, 2007). Firstly, the
(1) sludge samples were centrifuged in a 50 mL tube at 4000 g for
5 min. The centrate liquor was removed and the precipitant in the
tube was used for EPS extraction. EPS were analyzed for protein
(PN) and polysaccharide (PS). The PN was determined by the
Coomassie Brilliant Blue G-250 method (Bradford, 1976) using
bovine serum albumin as the standard, while the PS was deter-
mined by the Anthrone method (Gaudy, 1962) using glucose as the
standard.
2.3.2. The wastage rate of iron and calcium
In this experiment, reagents played an important role in the
sludge dewatering, so it was necessary to understand how Fenton's
reagent and lime affected the sludge dewatering and how many
reagents were lost. In order to obtain the liquid after dewatering,
the sludge was filtered using a device for measuring sludge SRF.
Diluted filtrate was measured for iron and calcium content by
246 J. Liang et al. / Water Research 84 (2015) 243e254
Fig. 1. Schematic diagram of the ultrahigh pressure filtration system: (a) Diagram of the experimental setup for sludge conditioning and dewatering process; (b) Diagram of the
spring filter plate.
Atomic Absorption Spectroscopy (Z  2000, Hitachi). In addition,
the iron and calcium content of the sludge cakes were determined
using standard analytical methods (EPA, 2005). The wastage rate of
iron and calcium ions was calculated by Eq. (2).
 
Filtrate½Fe or Ca2þ ðmg=g DSÞ
The wastage rateð%Þ ¼    100%
Total ½Feor Ca2þ ðmg=g DSÞ
where Total [Fe] or [Ca2þ] (mg/g DS) is dosage content.
2.3.3. Bound water content
The bound water content was determined by the differential
scanning calorimetry (DSC) method (Lee and Lee, 1995) permitting
direct thermal analysis for phase changes of free water. Sample
weight for the DSC test was kept at approximately 8 mg. To achieve
uniform sample quality, the sludge was first filtered to remove most
of the free water and the wet filtration cake blended thoroughly for
15 min in a small two-arm kneader. A DSC analyzer (DSC Q2000,
USA) was used to record the thermogram data. The sample tem-
perature was initially decreased at a rate of 5  C/min to 50  C
and then raised back to 25  C at the same rate. Heat absorption was
determined by integrating the peak area under the endothermic
curve. Then, the bound water content was calculated in accordance
with Eq. (3):
WB ¼ WT  DH=DH0
where WB is the bound water content in the sludge samples (g/
g),WT is the total water content in the sludge samples, D H is the
DSC enthalpy of the sludge samples, and D H0 is the standard
melting heat of ice, which was determined by obtaining the ther-
mogram of pure water of known mass and measuring the curve
area.
2.3.4. Floc morphology and morphological structure
To analyze the microstructural characteristic of raw sludge and
dewatered sludge, floc morphology was investigated through laser
diffraction, zeta instruments and scanning electron microscopy
(SEM).
A laser diffraction instrument (Malvern Mastersizer 2000,
Malvern, UK) was used to measure dynamic floc size and structure,
which enabled measurements in the range 0.02e2000 mm. Zeta-
potential was obtained from a Zeta Master, DelsaNano C, Beck-
man Coulter Instruments. To observe the microscopic structure of
(2) the sludge, the samples were immersed in liquid nitrogen for
30 min, then freeze-dried and coated with gold before analysis.
Then the sludge samples were observed by a SEM (Se3400 N,
Hitachi, Japan).
3. Results and discussion
3.1. Effect of conditioning on dewatering in various VSS% sludge
Based on our previous study (Mo and Huang, 2015), the
optimization of initial pH was adjusted to around 3. The results
were consistent with other researches (Neyens and Baeyens
et al., 2003a, 2003b; Buyukkamaci, 2004; and Ning and Chen,
2014). The effects of sludge dewaterability were investigated at
a Fe2þ/H2O2 dosage ratio of 50/30 mg/g DS, a lime dosage of
50 mg/g DS and a Fenton reaction time of 10 min. The SRF of
sludge (GZ-2) was from 3.80  1012 to 1.85  1011 m/kg and SRF
reduction efficiency reached 95.13%. To evaluate the effect of each
component of the Fenton's reagent and lime on sludge dew-
(3) atering performance, a series of single-factor experiments were
conducted.
3.1.1. Optimization of Fenton reaction time
Fenton reaction time on sludge conditioning were examined at a
Fe2þ/H2O2 dosage ratio of 50/30 mg/g DS, a lime dosage of 50 mg/g
DS with an initial pH of 3. This experiment used different VSS%
sludge from batch 1. The SRF and filtrate TOC values of the samples,
taken every 10 min were shown in Fig. 2a and b. The SRF and filtrate
TOC values increased, when the VSS% of sludge increased from
23.80% to 63.50%. In addition, the SRF values were all decreasing
 J. Liang et al. / Water Research 84 (2015) 243e254
Fig. 2. Effects of reaction time and lime dosages of each component of the composite conditioner on sludge conditioning: (a) Effect of Fenton reaction time on sludge SRF in
different VSS% sludge; (b) Effect of Fenton reaction time on sludge filtrate TOC in different VSS% sludge; (c) Effect of lime dosage on sludge SRF in different VSS% sludge; (d) Effect of
lime dosage on sludge filtrate TOC in different VSS% sludge.
initially, before slowly rising with time (Fig. 2a). A similar result was
observed by Guo and Feng (2014). The results showed that over the
optimal reaction time, sludge dewaterability did not be improved
remarkably. This optimal reaction time might be dependent on VSS
% of the sludge. The optimal reaction times and SRFs were GZ-
1.10 min (reaction time), 1.56  1011 m/kg (SRF); GZ-2, 30 min,
1.85  1011 m/kg; GZ-3, 50 min, 2.48  1011 m/kg; GZ-4, 70 min,
3.66  1011 m/kg; GZ-5, 110 min, 5.05  1011 m/kg. This showed that
the optimal Fenton reaction time and the SRF increased with the
rise in organic matter content. This could attribute to the length of
time period required to more reactive hydroxyl radicals to the level
required for high VSS% sludge. Similarly, at longer reaction times,
the filtrate TOC values were slightly positively affected. A possible
explanation for this is that the reactive hydroxyl radicals were
released gradually and caused the disruption of EPS (Niu and
Zhang, 2013).
3.1.2. Optimization of lime dosage
Lime as a skeleton builder can reduce the compressibility of
sludge and improve the mechanical strength and permeability of
solids existing in sludge during compression (Mowla and Tran,
2013). In accordance with the optimization above, the reaction
times were set at 10 min (GZ-1), 30 min (GZ-2), 50 min (GZ-3),
70 min (GZ-4) and 110 min (GZ-5). Meanwhile, lime was added at
dosages from 10 to 100 mg/g DS (pH ¼ 3, Fe2þ ¼ 50 mg/g DS;
H2O2 ¼ 30 mg/g DS; sludge samples from batch 1). The results
indicated that in high VSS% sludge samples, SRF and filtrate TOC
values were also high, but in low VSS% sludge samples the opposite
was turn. It can be seen in Fig. 2c and d, with the addition of the
lime, both the SRF and filtrate TOC values decreased initially and
then increased gradually. At higher dosages of lime, sludge dew-
atering was slightly negatively affected. This could be due to the
247
strong alkaline disturbing the coagulation of sludge fragments and
inhibiting further dewatering. Furthermore, the conditioned sludge
with higher final pHs tended to have higher filtrate TOC concen-
trations and worse dewatering. Therefore, it was found that 70 mg/
g DS to be the optimal dosage of lime, for GZ-1 sludge dewater-
ability (the ultimate pH ¼ 5.20). And under these conditions, the
SRF and filtrate TOC values were 1.65  1011 m/kg and 4.21 mg/g DS,
respectively. The SRF reduction efficiency was not high when using
only Fenton's treatment and combined Fenton's reagent and lime
were 85.86 and 90.92%, respectively. In other sludge, similar results
were observed. The optimal lime dosages for GZ-2, GZ-3, GZ-4, GZ-
5 were 70, 70, 50 and 60 mg/g DS, respectively and the SRF value
was reduced from 3.57  1012, 3.98  1012, 7.06  1012,
8.04  1012 m/kg to 1.91  1011, 2.49  1011, 2.99  1011,
5.57  1011 m/kg, respectively. Although the optimal lime dosages
in different VSS% sludge were different, the ultimate pH remained
about 5e6. The addition of lime could promote the chemical
coagulation of sludge by raising the final pH of Fenton treatment
and thus lead to termination of the oxidation reaction. In our study,
the lime dosage was about 50e70 mg/g DS. This dosage of lime is
quite small when compared with Liu and Yang (2012) (lime
dosage ¼ 300e400 mg/g DS).
3.1.3. Optimization of initial pH
It was found that Fenton process was most efficient when the
reaction pH was around 3 (Fig. S2). Thus, the results showed that an
initial pH of 3 achieved the best effect, the SRF values were
1.70  1011 m/kg (GZ-1), 1.88  1011 m/kg (GZ-2), 2.64  1011 m/kg
(GZ-3), 3.28  1011 m/kg (GZ-4), 5.32  1011 m/kg (GZ-5) and
filtrate TOC values were 4.03 mg/g DS, 8.58 mg/g DS, 10.31 mg/g DS,
12.31 mg/g DS, 14.76 mg/g DS, respectively.
248 J. Liang et al. / Water Research 84 (2015) 243e254

3.1.4. Effect of Fe2þ/H2O2 dosage on the dewaterability of sludge
The dosages of Fe2þ and H2O2 have the most direct impact on
Fenton oxidation factors. To investigate the effect of varying Fe2þ/
H2O2 dosage on the dewaterability of sludge, a series of experi-
ments were conducted (Fe2þ from 10 to 90 mg/g DS, H2O2 from 15
to 120 mg/g DS). The experiments were based on the optimal re-
sults shown above (sludge samples from batch 2). Fig. 3 shows the
relationship between the dosages of Fe2þ/H2O2 and the SRF values.
It was revealed that a molar Fe2þ/H2O2 ratio of more than 1 was
enough to achieve a good dewatering effect. This showed that the
SRF value decreases with increased Fe2þ/H2O2 dosage until a
certain ratio and above this the reverse results were obtained.
Moreover, the SRF values increased with increasing H2O2 dosage
within a certain range. This might be attributed to the amount of
hydroxyl radicals. It had some negative effects when H2O2 con-
centration exceeded a critical concentration. Similar studies re-
ported by Tony and Zhao (2008) and Liu and Yang (2012). As shown
in Fig. 3, sludge dewaterability was stably improved when Fe2þ
addition varied between 10 and 70 mg/g DS. But, if ferrous ions
were excessive, the additional ferrous ions could change to ferric
ions easily, and the ferric ions could form ferric hydroxide colloid
which would negatively affect the dewaterability of the sludge
(Guo and Feng, 2014). In Fig. 3a, under 30 mg/g DS Fe2þ and 15 mg/g
DS H2O2, the GZ-1 sludge dewaterability was optimal resulting in
an SRF value 1.54  1011 m/kg. Over the optimal addition of Fe2þ
and H2O2, the sludge dewaterability would get worse. In Fig. 3b,
sludge dewaterability was stably improved when Fe2þ addition
varied from 10 to 50 mg/g DS and H2O2 concentration from 15 to
60 mg/g DS. At dosages of 50 mg/g DS Fe2þ and 30 mg/g DS H2O2,
the lowest SRF value was achieved in GZ-2 sludge (1.92  1011 m/
kg). As illustrated in Fig. 3cee, the SRF values were reduced from 10
to 70 mg/g DS Fe2þ and from 15 to 60 mg/g DS H2O2. The optimal
dosages were found at 70 mg/g DS Fe2þ and 30 mg/g DS H2O2 in
which the SRF values were 2.04  1011 m/kg (GZ-3), 2.44  1011 m/
kg (GZ-4) and 4.48  1011 m/kg (GZ-5). In the results, the SRF values
of the best conditioning increased as the VSS% increased. Similarly,

Fig. 3. Different sludge dewatering capacity at various Fenton's reagent dosages: (a) GZ-1 sludge sample (Operating parameters: reaction time/lime/pH ¼ 10 min, 70 mg/g DS, 3.0);
(b) GZ-2 sludge sample (30 min, 70 mg/g DS, 3.0); (c) GZ-3 sludge sample (50 min, 70 mg/g DS, 3.0); (d) GZ-4 sludge sample (70 min, 70 mg/g DS, 3.0); (e) GZ-5 sludge sample
(110 min, 70 mg/g DS, 3.0); (f) Optimal conditioning of sludge filtrate TOC (Operating parameters: GZ-1, Fe2þ/H2O2 ¼ 30, 15 mg/g DS; GZ-2, Fe2þ/H2O2 ¼ 50, 30 mg/g DS; GZ-3, Fe2þ/
H2O2 ¼ 70, 30 mg/g DS; GZ-4, Fe2þ/H2O2 ¼ 70, 30 mg/g DS; GZ-5, Fe2þ/H2O2 ¼ 70, 30 mg/g DS).
 J. Liang et al. / Water Research 84 (2015) 243e254 249

as shown in Fig. 3f, the VSS% and the filtrate TOC values showed
positive correlation relationship in the best conditioning.
3.2. Initial exploration of the dewatering mechanism
3.2.1. Relationship between VSS% and sludge dewatering
Based on the above results, the optimal dewatering conditioning
for sludges with various VSS% was concluded. Meanwhile, experi-
ments were conducted in the optimal conditioning to investigate
the relationship between sludge VSS% and sludge dewatering. The
sludge samples were from batch 2 and each sample was tested
three times. The results are listed in Table 3.
As shown in Fig. S3a, strong correlations were found between
VSS% and optimal Fenton reaction time. Besides, the effect of VSS%
on sulfuric acid consumption is plotted in Fig. S3b. Strong correla-
tions were observed between VSS% and sulfuric acid consumption.
The results of the wastage rates of iron and calcium are shown in
Table S1. The wastage rates of iron and calcium were about 1.30%
and 40.14%, respectively. The results demonstrated that most of the
iron stayed in the filter cake and about 40% calcium was lost to the
filtrate.
The SRF values showed significant correlations (Raw sludge:
R21 ¼ 0.9789, y1 ¼ 2.08  109x21 þ 3.33  1011x1  5.06  1012;
Conditioned sludge: R22 ¼ 0.9722, y2 ¼ 2.16  108x22
 1.33  1010x2 þ 3.78  1011) with the VSS% of sludge as exhibited
in Fig. 4a. Results implied that both the SRF of raw sludge and
conditioned sludge increased rapidly when VSS% increased. As
shown in Table 3, the SRF decreased markedly after conditioning,
while the filtrate TOC increased. In addition, VSS% of conditioned
sludge decreased slightly. This might be due to Fenton reaction.
Fenton oxidation would relatively damage cells. Thus, the EPS could
easily remove and dissolve. Dissolve EPS would release to the
filtrate. Besides, residual reagents might affect the filtrate TOC.
As shown in Fig. 4b, the VSS% was significantly correlated to
filtrate TOC values (R21 ¼ 0.9893, y1 ¼ 0.0001x21
þ 0.0383x1  0.4562; Conditioned sludge: R22 ¼ 0.9967,
y2 ¼ 0.007x22 þ 1.0429x2  17.535). These results indicated that
dewatering in high VSS% sludge is slow and hard. Inversely, dew-
atering is relatively fast and easy in low VSS% sludge. This might be
due to high VSS% sludge containing higher concentrations of EPS.
Some research has indicated that high EPS content has a negative
effect on sludge dewaterability, due to a large number of hydro-
philic sludge floc structures (Bala Subramanian and Yan, 2010).
3.2.2. Effect of conditioning on EPS content within sludge of
different VSS%
There is a clear release of EPS from different VSS% sludge
when conditioned by Fenton's reagent and lime (Fig. 4c). PNs and
PSs are believed to be the major constituents of EPS (Li and Yang,
2007). In our study, the impact of PNs and PSs on sludge det-
watering was researched. The PS content increased from 1.15 to
2.46 mg/g DS and the PNs contents increased from 4.03 to
11.55 mg/g DS as the VSS% of sludge increased from 23.70% to
65.26%. Meanwhile, it was shown that both PN and PS content
decreased after conditioning. The results revealed that the
reduction efficiency of PS reached over 80% and the PN reduced
over 28%. Sludge particles always carry negative charge due to
the ionization of functional groups. Electrostatic repulsion, pre-
venting destabilization and flocculation of sludge particles might
be aroused by high negative surface charge density (Zhang and
Yang, 2015). In addition, the reduction efficiency of PN was not
high because of the strong attached water capacity. PNs in the
EPS might present a large quantity of exoenzymes, which were
easy to get rid of.
The relationship between VSS% and EPS is presented in Fig. 4d
(Raw sludge: R21 ¼ 0.8744, y1 ¼ 0.1973x1 þ 1.6782; Conditioned
sludge: R22 ¼ 0.9528, y2 ¼ 0.12825x2 þ 0.23579). Previous studies
indicated that improving sludge dewatering would cause a reduc-
tion in EPS and make sludge more compact (Chen and Yang, 2001).
Our results demonstrated that EPS had the closer relationship with
the VSS%. Thus, removing the EPS to a certain degree is the key to
sludge dewatering.
It can be seen in Fig. 4c and e that after sludge conditioning with
Fenton's reagent and lime, the EPS and bound water contents
decreased regularly. The results are closely related to the organic
content of sludge. Firstly, sludge samples contain biomass, which
wrap in EPS and bound water. It means high VSS% sludge contain
many bound water and EPS. Secondly, excessive EPS were hard to
mineralize completely by Fenton treatment. It could only remove a
certain amount of EPS and bound water.
3.2.3. Effect of conditioning on bound water content for different
VSS% sludge
The standard melting heat of ice is 334.0 ± 2.5 kJ/kg, which was
determined by obtaining the thermograms of pure water (9.2 mg)
and measuring the curve area. The bound water content of sludge
samples were calculated using Eq. (3). The bound water content of
raw sludge and conditioned sludge samples was determined and
was shown in Fig. 4e. The bound water in the raw sludge were
1.38 ± 0.02 kg/kg DS (GZ-1), 1.54 ± 0.02 kg/kg DS (GZ-2),
1.89 ± 0.13 kg/kg DS (GZ-3), 2.05 ± 0.01 kg/kg DS (GZ-4) and
3.16 ± 0.03 kg/kg DS (GZ-5). Similar values for bound water of the
sewage sludge were reported by Heukelekian and Weisberg (1956);
Colin and Gazbart (1995); and Chu and Chang (2001). From our
results, it was found that there were high content of bound water in
high VSS% sludge. Thus, excessive bound water might limit
dewaterability.
After conditioning, the bound water content of sludge decreased
sharply and fell by over 70%. Again, this replicated previous
research (Vaxelaire and Ce zac, 2004; Zhang and Yang, 2014). In
these previous studies, they indicated that chemical conditioning
could reduce the bound water in sludge either by replacing water
molecules adsorbed on the particle surface or by affecting the water
binding capacity of the particles. Therefore, sludge conditioning

Table 3
The sludge dewaterability of different VSS% sludge.

Sludge VSS/TSS (%) Optimal conditioning combination SRF (1011 m/kg) TOC (mg/g DS)
source
Raw sludge Conditioned Fenton time Lime pH Fenton's reagent (mg/g DS) Raw Conditioned Raw Conditioned
sludge (min) (mg/g DS) sludge sludge sludge sludge
Fe2þ H2O2

GZ-1 23.70 ± 0.16 18.72 ± 0.09 10 70 3 30 15 15.47 ± 0.31 1.71 ± 0.13 0.35 ± 0.01 3.54 ± 0.24
GZ-2 30.99 ± 0.57 26.38 ± 0.24 30 70 3 50 30 37.34 ± 0.90 1.89 ± 0.19 0.65 ± 0.03 7.36 ± 0.12
GZ-3 38.46 ± 0.48 30.46 ± 0.11 50 70 3 70 30 41.85 ± 0.20 2.02 ± 0.09 0.88 ± 0.09 12.51 ± 0.09
GZ-4 51.49 ± 0.50 43.30 ± 0.15 70 50 3 70 30 68.21 ± 1.27 2.42 ± 0.21 1.13 ± 0.11 17.64 ± 0.44
GZ-5 65.26 ± 0.19 55.24 ± 0.20 110 60 3 70 30 78.01 ± 0.83 4.40 ± 0.17 1.55 ± 0.05 20.54 ± 0.59
250 J. Liang et al. / Water Research 84 (2015) 243e254

Fig. 4. Correlation between sludge VSS% and dewaterability: (a) Relationship between VSS% and SRF (both raw and conditioned sludge); (b) Relationship between VSS% and TOC
(both raw and conditioned sludge); (c) PN and PS content in different VSS% sludge; (d) Relationship between the VSS% and the EPS (both raw and conditioned sludge); (e) Bound
water content in different VSS% sludge; (f) Relationship between the VSS% and the bound water content.

with Fenton's reagent and lime could convert at least a portion of
the bound water to free water.
As shown in Fig. 4f, strong correlations were found between the
VSS% of sludge and the concentration of bound water (Raw sludge:
R21 ¼ 0.9701, y1 ¼ 0.0013x21  0.0719x1 þ 2.4183; Conditioned
sludge: R22 ¼ 0.9882, y2 ¼ 0.0117x2 þ 0.0379). The above results
show that the SRF values had a significant correlation with the VSS%
of sludge. Thus, removal of the bound water might be one of the key
factors to sludge dewatering.
3.3. Floc morphology and morphological structure
To fully comprehend the possible dewatering mechanisms, the
morphology of sewage sludge flocs was characterized through SEM.
As shown in Fig. 5a, the raw sludge contained mainly large rough
floc blocks surrounded by lots of particles. However, after condi-
tioning with Fenton's regent and lime, large flocs broke into small
pieces and became thinner. And the ferric hydroxide complexes
mixed with calcium might rebuild a more stable structure, which

Fig. 5. SEM images of dewatered sludge cakes: (a) Raw sludge (GZ-2); (b) Conditioned sludge (GZ-2).
 J. Liang et al. / Water Research 84 (2015) 243e254
Table 4
Effects of conditioning on particle size and zeta-potential.
Sludge samples Median particle size (mm)
Raw sludge
GZ-1 45.6
GZ-2 54.7
GZ-3 51.8
GZ-4 64.4
GZ-5 57.2
was benefit to removing the EPS and releasing the bound water (as
shown in Fig. 5b). Moreover, the microstructural results were
supported by the analyses of particle size and zeta potential (results
listed in Table 4). The particle size of conditioned sludge became
smaller and the particle sizes were between 40 and 60 m m. The zeta
potentials of the sludge were negative. After conditioning, the zeta
potential had a tendency to increase. The results illustrated that
Fenton's reagent and lime could be regarded as an effective inor-
ganic coagulant, was able to regroup the disrupted EPS and sludge
fragments, thus improving sludge dewatering performance. Similar
conclusions were drawn by Liu and Yang (2013); Ning and Chen
(2014) and Zhang and Yang (2014).
3.4. Pilot-scale experiments
To obtain more reliable results for real industrial condition, a
novel filter press which could achieve ultrahigh pressure was
applied to improve dewaterability of sludge. In these tests, all re-
agents can meet the industrial standard. The diagram of the sludge
dewatering process in pilot scale is shown in Fig. 1a. And 70 L of raw
sludge samples (water content about 95%) were transferred to a
100 L conditioning tank and then treat it with the following pro-
cedure: (1) adding H2SO4 (300 rpm, 5 min); (2) adding FeSO4$7H2O
(300 rpm, 5 min); (3) adding H2O2 (300 rpm, optimal reaction
time); (4) adding lime (200 rpm, 5 min). Sludge cakes (Fig. 6) were
obtained after feed and compressed.
Fig. 6. The photo of sludge cake.
251
Zeta potential (mV)
Conditioned sludge Raw sludge Conditioned sludge
40.0 10.5 5.1
50.5 7.3 2
42.2 19.1 11.5
58.9 15.8 12.6
46.5 20.5 10.8
3.4.1. Effect of feed time and total filtrate time
It can be seen in Fig. 7, the sludge dewatering process of ultra-
high pressure filtration system is divided into three stages: (1) Feed
stage; (2) Compression stage; (3) Spring Compression stage. In
stage 1, pressure was gradually increased to 1.6 MPa. Meanwhile,
conditioned sludge was feed into the spring filter plates and
removed most of the free water. In stage 2, pressure was increased
rapidly and sludge cakes were compressed with a high pressure. In
stage 3, pressure was increased slowly and sludge cakes were
compress again with a higher pressure.
Fig. 7f and Table 5 revealed that with SRF decreased, both total
filtrate mass and the filtration rate are gradually increased. The
results indicated that total filtrate mass and the filtration rate were
related to SRF. High VSS% sludge samples needed longer time to
feed and compress because of the high SRF value. As shown in
Fig. 7aee, in GZ-1 sludge, the dewatering efficiency was highest
and the compressibility was best. But dewaterability and
compressibility became worse when the VSS% of sludge increased.
For instance, in GZ-5 sludge samples, it took 50 min to compress
(High pressure: 28 Mpa) and 60 min to feed. Thus, high VSS% sludge
might be contain large flocs (mainly EPS and bound water), are
suitable for high pressure dewatering.
The results of pilot-scale experiments were given in Table 5. On
the dewatered cake, the percentage of the water contents that
conditioned with Fenton's reagent and lime are low, at
36.85 ± 0.21% (GZ-1), 38.71 ± 0.18% (GZ-2), 41.88 ± 0.30% (GZ-3),
44.56 ± 0.11% (GZ-4), 49.90 ± 0.25% (GZ-5), respectively. All the
water contents of the dewatered sludge cakes were lower than 50%,
reaching the standard of the sludge disposal. This clearly demon-
strated that the optimal conditions could shape a porous and stable
dewatered sludge cake, which was able to maintain high perme-
ability during the dewatering process. The results showed that the
water contents of the dewatered sludge cakes (GZ-1) were lowest.
It might be related to the EPS and the bound water contents. Low
VSS% sludge (less EPS and bound water) that conditioning could be
relatively easy to remove the water and to rebuild the skeleton.
These results confirmed that Fenton's reagent and lime performed
much better in dewatering than traditional conditioning methods.
Thus, the optimal Fenton's reagent and lime combined with ultra-
high pressure filtration system could become a rapid and practical
sludge treatment technique. And different VSS% of sludges might
utilize their specific optimal conditioning combinations.
3.4.2. Cost estimation of Fenton's reagent and lime
It was illustrated the cost estimation of Fenton's reagent and
lime in Table 6. The estimated prices of the reagents using the
optimal Fenton's reagent and lime are about 27.39 (GZ-1), 42.36
(GZ-2), 49.82 (GZ-3), 49.06 (GZ-4), 51.39 USD/t DS (GZ-5). It indi-
cated that even in GZ-5 sludge (VSS% ¼ 65.26), the water content of
conditioned sludge cake can also be reduced to below 50% and the
cost of reagents is 51.39 USD/t DS. Compared with the previous
studies (Liu and Yang, 2012, 2013), the cost of reagents is relatively
low in our study. Besides, the cost of traditional polyacrylamide
252 J. Liang et al. / Water Research 84 (2015) 243e254

Fig. 7. The results of the feed time and the filtrate time: (a) GZ-1 sample; (b) GZ-2 sample; (c) GZ-3 sample; (d) GZ-4 sample; (e) GZ-5 sample; (f) the total filtrate time of different
VSS% sludge.

Table 5
Results of pilot-scale experiments.

Sludge Water H2SO4a dosage FeSO4,7H2Ob H2O2c dosage Limed dosage SRF (1011 m/kg) Sludge cake water Feed time Compression Pressure
samples content (%) (kg/t DS) dosage (kg/t DS) (kg/t DS) (kg/t DS) content (%) (min) time (min) (MPa)

GZ-1 95.67 ± 0.12 25.8 155.3 50 90 1.79 ± 0.08 36.85 ± 0.21 25 20 15

GZ-2 95.19 ± 0.10 28.3 258.9 100 90 1.96 ± 0.03 38.71 ± 0.18 30 30 20
GZ-3 95.31 ± 0.22 38.1 362.5 100 90 2.34 ± 0.07 41.88 ± 0.30 40 35 28
GZ-4 94.98 ± 0.15 49.0 362.5 100 70 2.61 ± 0.02 44.56 ± 0.11 45 40 28
GZ-5 94.60 ± 0.20 68.5 362.5 100 80 4.45 ± 0.09 49.90 ± 0.25 60 50 28
a
Sulfuric acid (industrial-grade, 90%).
b
FeSO4$7H2O (industrial-grade, 95%).
c
H2O2 (industrial-grade, 30%).
d
Lime (industrial-grade, 80%).

(PAM) can be as high as 65 USD/t DS in the WTTPs of China (Zhang
and Yang, 2014). The water content of dewatered sludge condi-
tioned with PAM is about 80%. Compare with our results, the water
contents are less than 50% and the volume reduced by about 70%,
which is convenient for sludge transport and disposal. Meanwhile,
it can save much money on transportation expenses.
Therefore, it is an economical and viable way to apply the
technique (Fenton's reagent and lime combined with ultrahigh
pressure filtration system) to sludge dewatering in the full-scale
WWTPs of Guangzhou.
3.5. Model of the conditioning mechanism
From the above results and analysis, a preliminary understand
of the mechanism of sludge dewatering was obtained, which
combined EPS content, bound water content, the wastage rates of
 J. Liang et al. / Water Research 84 (2015) 243e254 253

Table 6
Cost estimation of the Fenton and lime treatment.

Sludge samples H2SO4a cost (USD/t DS) FeSO4$7H2Ob cost (USD/t DS) H2O2 costc (USD/t DS) Lime costd (USD/t DS) Total cost (USD/t DS)

GZ-1 2.01 10.05 8.07 7.26 27.39

GZ-2 2.21 16.75 16.14 7.26 42.36
GZ-3 2.97 23.45 16.14 7.26 49.82
GZ-4 3.82 23.45 16.14 5.65 49.06
GZ-5 5.34 23.45 16.14 6.46 51.39
a
Unit price of sulfuric acid is 78.0 dollar per ton.
b
Unit price of FeSO4$7H2O is 64.7 dollar per ton.
c
Unit price of H2O2 is 161.4 dollar per ton.
d
Unit price of lime is 80.7 dollar per ton.

Fig. 8. The possible dewatering mechanisms of sewage sludge flocs: (a) Model of fast to dewate sludge; (b) Model of moderately fast to dewater sludge; (c) Model of slow to dewater
sludge; (d) Schematic of changes in sludge flocs under Fenton's reagent and lime.

iron and calcium, SEM, particle sizes and zeta potential. The EPS, bound water and cellular contents; moderate feed and
possible dewatering mechanisms are shown in Fig. 8d. It is divided compression time (feed time: 25e45 min, compression time:
into two major processes: (1) Fenton oxidation; (2) Skeleton 20e40 min); pressure (15e28 Mpa).
structure construction. Through Fenton treatment, the flocs were (3) Slow to dewater sludge: high VSS% (more than 50%); high
oxidized and flexible EPS was broken down, which led to release of SRF (more than 6.9  1012 m/kg); more EPS, bound water and
the cellular contents and bound water. That made iron combined cellular contents; slow feed and compression time (feed time
with the flocs easily. Then, the addition of lime neutralized the pH more than 45 min, compression time more than 40 min);
and acted as a skeleton builder. Furthermore, the number of ferric pressure (more than 28 Mpa).
hydroxo complexes increased, resulting in fracturing large sludge
and making sludge particles become smaller. As a result, the sludge
4. Conclusions
dewaterability improved significantly after conditioning with
Fenton's reagent and lime.
In this study, it was demonstrated that the dewaterability of
Eventually, according to the laboratory-scale and pilot-scale
different VSS% sludge (including raw and conditioned) were
experiments, the sludges were classified as three types: (1) Fast
inconsistent. Through a series of single-factor experiments, the
to dewater; (2) Moderately fast to dewater; (3) Slow to dewater
optimal conditioning combinations of different VSS% sludges were
sludge. And the diagram of these models is shown in Fig. 8aec.
found. The results indicated that sludge dewatering performance
was effectively improved by conditioning with Fenton's reagent
(1) Fast to dewater sludge: low VSS% (less than 25%); low SRF
and lime. As the relationship between VSS% of sludge and sludge
(less than 1.5  1012 m/kg); less EPS, bound water and
dewatering shown, it was seen that high VSS% sludge samples
cellular contents; less than 25 min to feed and 20 min to
required longer reaction times and consumed more conditioner
compress; pressure less than 15 Mpa.
including sulfuric acid, Fe2þ and H2O2. Meanwhile, it was demon-
(2) Moderately fast to dewater sludge: moderate VSS% (25%e
strated that most of the iron stayed in the filter cake and about 40%
50%); moderate SRF (1.5  1012 e 6.9  1012 m/kg); moderate
calcium was lost to the filtrate. Besides, it was found that the SRF
254 J. Liang et al. / Water Research 84 (2015) 243e254
and filtrate TOC values had a significant correlation with the VSS%
of sludge (including raw and conditioned).
The dewaterability mechanism was also investigated. The re-
sults exhibited that EPS contents and the bound water decreased
after dewatering. In addition, it was also detected that EPS contents
and bound water were associated with the VSS% (including raw and
conditioned). And it was demonstrated that reducing EPS content
suitably and releasing the bound water was the key to sludge
dewatering. Moreover, the microstructural (SEM) results implied
sludge particles became smaller and built the skeleton. The pilot-
scale experiments revealed that Fenton's reagent and lime com-
bined with ultrahigh pressure filtration system is feasible and
economical to enhance the dewatering of the sludge. Consequently,
novel mechanistic models were put forward.
Acknowledgments
The authors gratefully thank the financial support of this study
by Education Special Funds of University Discipline Construction of
Guangdong Province (Grant 2014KTSP022), the Special Foundation
for Circular Economy Development of Guangdong Province
(12ZK0058) and Fundamental Research Funds for the Guangzhou
(2013J4100010). The authors would also like to acknowledge the
Analytical and Testing Center of Guangdong University of Tech-
nology for providing the experimental facilities and five different
municipal WWTPs (GZ-1, GZ-2, GZ-3, GZ-4 and GZ-5) for supplying
the raw sludge samples. And the authors would like to thank the
Foshan JKD Filtration Equipment Company for providing the ul-
trahigh pressure filtration system.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.watres.2015.07.041.
References
APHA Standard Methods, 1998. In: Clesceri, L.S., Greenberg, A.E., Eaton, A.D. (Eds.),
Standard Methods for the Examination of Water and Wastewater. American
Public Health Association, Washington.
Bala Subramanian, S., Yan, S., 2010. Extracellular polymeric substances (EPS) pro-
ducing bacterial strains of municipal wastewater sludge: Isolation, molecular
identification, EPS characterization and performance for sludge settling and
dewatering. Water Res. 44 (7), 2253e2266.
Bradford, M., 1976. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal. Bio-
chem. 72 (1e2), 248e254.
Buyukkamaci, Nurdan, 2004. Biological sludge conditioning by Fenton's reagent.
Process Biochem. 39 (11), 1503e1506.
Chang, J., Abu-Orf, M., 2005. Dentel alkylamine odors from degradation of flocculant
polymers in sludges. Water Res. 39 (14), 3369e3375.
Chen, Y., Yang, H., 2001. Effect of acid and surfactant treatment on activated sludge
View publication stats
dewatering and settling. Water Res. 35 (11), 2615e2620.
Chu, C., Chang, B., 2001. Observations on changes in ultrasonically treated waste-
activated sludge. Water Res. 35 (4), 1038e1046.
Citeau, M., Larue, O., 2011. Influence of salt, pH and polyelectrolyte on the pressure
electro-dewatering of sewage sludge. Water Res. 45 (6), 2167e2180.
Colin, F., Gazbart, S., 1995. Distribution of water in sludges in relation to their
mechanical dewatering. Water Res. 29 (8), 2000e2005.
Debowski, M., Zielinski, M., 2008. Efficiency of sewage sludge conditioning with the
Fenton's method. Ochr. Srodowiska 30 (2), 43e47.
Dewil, R., Baeyens, J., 2005. Fenton peroxidation improves the drying performance
of waste active sludge. J. Hazard. Mater. 117 (2e3), 161e170.
EPA Standard Methods, 2005. CJ/T221-2005 Determination Method for Municipal
Sludge in Wastewater Treatment Plant. Environmental Protection Administra-
tion, China.
Gaudy, A., 1962. Colorimetric determination of protein and carbohydrate. Ind. Water
Wastes 7 (1), 17e22.
Guo, H., Feng, S., 2014. Application of Fenton's reagent combined with sawdust on
the dewaterability of oily sludge. Environ. Sci. Pollut. Res. 21 (18), 10706e10712.
Heukelekian, H., Weisberg, E., 1956. Bound water and activated sludge bulking. Sew.
Ind. Wastes 28 (4), 558e574.
Ho, Y.C., Norli, I., 2010. Morad characterization of biopolymeric flocculant (pectin)
and organic synthetic flocculant (PAM): a comparative study on treatment and
optimization in kaolin suspension. Bioresour. Technol. 101 (4), 1166e1174.
Jin, L., Zhang, G., 2014. Current state of sewage treatment in China. Water Res. 66,
85e98.
Lee, D., Lee, S., 1995. Measurement of bound water content in sludge: the use of
differential scanning calorimetry (DSC). J. Chem. Technol. Biotechnol. 62,
359e365.
Li, X., Yang, S., 2007. Influence of loosely bound extracellular polymeric substances
(EPS) on the flocculation, sedimentation and dewaterability of activated sludge.
Water Res. 41 (5), 1022e1030.
Liu, H., Yang, J., 2012. Conditioning of sewage sludge by Fenton's reagent combined
with skeleton builders. Chemosphere 88 (2), 235e239.
Liu, H., Yang, J., 2013. A comprehensive insight into the combined effects of Fenton's
reagent and skeleton builders on sludge deep dewatering performance.
J. Hazard. Mater. 258, 144e150.
Mo, R., Huang, S., 2015. A rapid Fenton treatment technique for sewage sludge
dewatering. Chem. Eng. J. 269, 391e398.
Mowla, D., Tran, H., 2013. A review of the properties of biosludge and its relevance
to enhanced dewatering processes. Biomass Bioenergy 58, 365e378.
Neyens, E., Baeyens, J., 2003a. A review of classic Fenton's peroxidation as an
advanced oxidation technique. J. Hazard. Mater. 98 (1e3), 33e50.
Neyens, E., Baeyens, J., 2003b. Pilot-scale peroxidation(H2O2) of sewage sludge.
J. Hazard. Mater. 98 (1e3), 91e106.
Ning, X., Chen, H., 2014. Effects of ultrasound assisted Fenton treatment on textile
dyeing sludge structure and dewaterability. Chem. Eng. J. 242, 102e108.
Niu, M., Zhang, W., 2013. Correlation of physicochemical properties and sludge
dewaterability under chemical conditioning using inorganic coagulants. Bio-
resour. Technol. 144, 337e343.
Qi, Y., Thapa, K.B., Hoadley, A.F.A., 2011. Application of filtration aids for improving
sludge dewatering properties e a review. Chem. Eng. J. 171, 373e384.
Tony, M., Zhao, Y., 2008. Conditioning of aluminium-based water treatment sludge
with Fenton's reagent: effectiveness and optimising study to improve dew-
aterability. Chemosphere 72 (4), 673e677.
Vaxelaire, J., Cezac, P., 2004. Moisture distribution in activated sludges: a review.
Water Res. 38 (9), 2215e2230.
Yang, G., Zhang, G., Wang, H., 2015. Current state of sludge production, manage-
ment, treatment and disposal in China. Water Res. 78, 60e73.
Zhang, H., Yang, J., 2014. Mechanism of red mud combined with Fenton's reagent in
sewage sludge conditioning. Water Res. 59, 239e247.
Zhang, W., Yang, P., 2015. Insights into the respective role of acidification and
oxidation for enhancing anaerobic digested sludge dewatering performance
with Fenton process. Bioresour. Technol. 181, 247e253.