Ultrasonics Sonochemistry 24 (2015) 19–26

Contents lists available at ScienceDirect

Ultrasonics Sonochemistry
journal homepage: www.elsevier.com/locate/ultson

Characterization of drinking water treatment sludge after ultrasound

treatment
Zhiwei Zhou, Yanling Yang ⇑, Xing Li, Yang Zhang, Xuan Guo
Key Laboratory of Beijing for Water Quality Science and Water Environment Recovery Engineering, Beijing University of Technology, Beijing 100124, PR China

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

Article history: Ultrasonic technology alone or the combination of ultrasound with alkaline or thermal hydrolysis as pre-
Received 7 June 2014 treatment for anaerobic digestion of activated sludge has been extensively documented. However, there
Received in revised form 17 October 2014 are few reports on ultrasound as pretreatment of drinking water treatment sludge (DWTS), and thereby
Accepted 5 November 2014
the characteristic variability of sonicated DWTS has not been fully examined. This research presents a
Available online 13 November 2014
lab-scale study on physical, chemical and biological characteristics of a DWTS sample collected from a
water plant after ultrasonic treatment via a bath/probe sonoreactor. By doing this work, we provide
Keywords:
implications for using ultrasound as pretreatment of enhanced coagulation of recycling sludge, and for
Ultrasound
Drinking water treatment sludge
the conditioning of water and wastewater mixed sludge by ultrasound combined with polymers. Our
Sludge disintegration results indicate that the most vigorous DWTS disintegration quantified by particles’ size reduction and
Organic solubilization organic solubilization is achieved with 5 W/ml for 30 min ultra-sonication (specific energy of
Microorganisms inactivation 1590 kWh/kg TS). The Brunauer, Emmett and Teller (BET) specific surface area of sonicated DWTS flocs
Recycling pretreatment increase as ultra-sonication prolongs at lower energy densities (0.03 and 1 W/ml), while decrease as
ultra-sonication prolongs at higher energy densities (3 and 5 W/ml). Additionally, the pH and zeta poten-
tial of sonicated DWTS slightly varies under all conditions observed. A shorter sonication with higher
energy density plays a more effective role in restraining microbial activity than longer sonication with
lower energy density.
Ó 2015 Published by Elsevier B.V.

1. Introduction
Drinking water treatment sludge (DWTS), characterized with
elevated concentrations of inorganic or organic substance, as well
as precipitation hydroxide that is derived from inorganic alum-
based or ferric-based coagulants, is largely produced during floccu-
lation–sedimentation or flotation process. DWTS and solids within
filter backwash water are known as the main components of water
treatment residual. Okuda et al. [1] report that global production of
solid residuals ‘might be totally 10,000 tons per day’, of which the
European countries including Ireland, Germany, Netherland, Uni-
ted Kingdom and Portugal take up 10.38%, and the United States
and Chinese Taiwan accounts for 72.6% and 0.003%, respectively
[2]. The increasing generation of water treatment residues, coupled
with environmental restraints regarding their current disposal out-
lets, has prompted increased research toward their reuse.
The reuse of DWTS is an important approach to realize the
reduction and reclamation of the total waste residues. Chemical
safety and bio-stability are the two main concerns of reuse process
⇑ Corresponding author. Tel./fax: +86 10 67391726.
E-mail address: yangyanling@bjut.edu.cn (Y. Yang).
[3], thus limiting the concentration of organic matter and metal
ions, and controlling microbial activity to a low level within sludge
before recycling can help ensure the safety of the recycle outputs.
The reuse of DWTS can contribute toward an improvement of
treatment performance regarding the particulates and phosphorus
removal [3,4], and an improvement in settling and dewatering
characteristics of the re-coagulation flocs as well. However, the
enhanced removal of pollutions, i.e., organic matter of recycling
processes sometimes is not desirable due to the release of pollu-
tions (humic-like and protein-like substances) that are derived
from sludge [2,5,6]. We recently recommend one idea that to dis-
integrate DWTS, and simultaneously to maximum inactivate the
pathogenic microorganism by ultrasound to enhance organic mat-
ter removal in sludge recycling process [7]. A higher disintegration
degree of DWTS is expected prior to recycling, since intracellular or
extracellular polymers, i.e., proteins and polysaccharide among the
solubilized organics that are difficult to be removed by coagulation
are discharged in the pre-sonicated recycling DWTS, thereby low-
ering the burden of recycling process.
The ultrasonic technique is well known for disrupting sludge
flocs and lysing biological cells, which can lead to the solubilization
of organic matter, reduction of particle size and inactivation of
sludge microorganisms [8–11]. The effect of sonication parameters,

http://dx.doi.org/10.1016/j.ultsonch.2014.11.007
1350-4177/Ó 2015 Published by Elsevier B.V.
20 Z. Zhou et al. / Ultrasonics Sonochemistry 24 (2015) 19–26
including ultra-sonication time, frequency and energy input, and
sludge characteristics (solids content, origin of sludge) on the level
of sludge disintegration, as well as the increase in biogas produc-
tion in anaerobic digestion have been well investigated [12,13].
Low-frequency and high-intensity ultrasound facilitates activated
sludge disintegration, due to the elevated mechanical shear forces
produced by the implosion of the cavitation bubbles [14,15]. Low
energy density ultrasound causes the release of substances capable
of solubilizing, such as nucleic acids, lipids, humic acids, and pro-
teins [16]. Furthermore, increasing ultrasound intensity leads to
the lyses of cells of the present microorganisms, with the release
of cellular liquid and components of cytoplasm [17].
Previous results [8–13] are obtained on the basis of using ultra-
sound or the combination of ultrasound with alkaline or thermal
hydrolysis treatment as pretreatment for anaerobic digestion of
activated sludge. However, there are few reports about the use of
ultrasound as pretreatment of DWTS, and thereby the characteris-
tic variability of sonicated DWTS has not been investigated. In pre-
vious research, the soluble organics that were derived from
activated sludge mainly consisted of intracellular or extracellular
polymers, i.e., proteins and polysaccharide [18,19], which were dif-
ferent from DWTS that fulvic acid was the main constitution that
also contained humic and hymathomelanic acids [2]. Furthermore,
the organic concentration quantified by soluble chemical oxygen
demand (sCOD) in the supernatant of activated sludge was about
10-folds or 50-folds higher [20,21] than that in DWTS of our previ-
ous study [6]. Since the organic constitution and characteristics of
DWTS are different from those of activated sludge, the insights into
characterization of ultrasonically DWTS are needed.
In this study, the sonication parameters, including ultrasound
frequency and energy density, as well as type of sonoreactor on
physical, chemical and biological characterization of a DWTS after
ultrasonic treatment are systematically investigated. The main
issues focused upon are: (1) to evaluate the effect of ultrasound
frequency at 25 and 40 kHz with low energy density of 0.03 W/
ml on organic solubilization and the flocs characteristics, as well
as the microbial activity via a bath (C)/probe (T) sonoreactor; (2)
to elucidate the DWTS disintegration degree and microbial activity
after the DWTS subjected to higher energy densities (1, 3 and 5 W/
ml) with frequency of 25 kHz using the probe sonoreactor.
2. Materials and methods
2.1. The raw DWTS used in the experiments
The raw DWTS sample used in the experiments is collected
from a water treatment plant (Beijing, China), which handles
1,500,000 m3/day using a coagulation–flocculation–sedimenta-
tion–sand filtration process. Poly-aluminum chloride and ferric
chloride is employed as coagulant. The alum and ferric sludge set-
tled in the sedimentation basin and is discharged to the thickener.
The obtained DWTS samples are transferred immediately to the
laboratory and stored at 4 °C. Before each experiment, DWTS sam-
ples are warmed to ambient temperature. All of the experiments
are performed within a week from the date of sampling. The micro-
bial parameters are firstly investigated for each sets of experiment.
The characteristics of raw DWTS used in the experiments are dis-
played in Table 1.
2.2. Sonification experiments
We conduct two sets of sonification experiments (seen in
Fig. 1), one of which is to investigate the impact of low energy den-
sity of 0.03 W/ml using both bath and probe-type sonoreactor. The
second experiment is to examine the impact of high energy densi-
ties (1, 3, and 5 W/ml) using the probe sonoreactor. Each sample is
sonicated for 5, 10, 15, 20 and 30 min in sequence. For each exper-
iment, the chemical parameters (total chemical oxygen demand
(TCOD), sCOD, proteins and polysaccharide in the supernatant of
sonicated DWTS), microbial parameters (counts of total bacteria
and total coliforms), as well as physicochemical parameters of
DWTS flocs (pH, zeta potential, average size and BET specific sur-
face area) are systematically assessed. All experiments are run
twice and the average values of organic, microbial and physico-
chemical parameters are recorded.
15 L of raw DWTS sample is sonicated in a rectangular bath
(250 mm (L)  250 mm (B)  300 mm (H)) reactor where the
transducers are located at each side of the vessel. Ultrasound fre-
quencies of 25 and 40 kHz, corresponding energy density as low
as 0.03 W/ml are evaluated. The ultrasonic power dissipated into
the suspension is measured using calorimetry [22]. Such low
energy density is used, due to the limitation of the dimensional
configuration and the maximum power input of 450 W for batch
sonoreactor. At each sampling point, firstly the sonicated DWTS
is homogeneously mixed, and then about 250 mL sonicated DWTS
is withdrawn from 150 mm below the upper surface of the vessel.
During the experiments, we find that the temperature of the DWTS
is slightly improved from 20.4 ± 0.8 °C to 22.5 ± 0.5 °C at the end of
the sonication of 30 min. So, the impact of the raise in temperature
of DWTS matrix on DWTS characterization is not investigated.
As to the probe-type sonoreactor, under low energy density of
0.03 W/ml, 150 mL of raw DWTS sample is placed in a double-wall,
jacketed glass container and subject to continuous ultrasound irra-
diation emitted through a 18 mm in diameter tip at power input of
4.5 W. At each sampling point, about 25 mL sonicated DWTS is
withdrawn from the center of the container. The temperature of
DWTS medium is controlled with a water bath coupled to a circu-
lator, and is kept constant at 20 ± 1 °C. In order to minimize con-
tamination of the sample, the probe is immersed in the DWTS
through a silicon rubber plug. The DWTS level inside the container
is 4 cm and the probe (total length of 10 cm) is positioned in the
middle of the container, with its tip 2 cm from the bottom. When
the energy density is 1, 3 and 5 W/ml, 50 mL of raw DWTS sample
is put into the container and the responding power input is 50, 150
and 250 W accordingly.
2.3. Analytical methods
All analyses are made using chemicals of analytical grade. pH is
determined by Thermo (Shanghai, China) pH meter, which is cali-
brated daily using pH buffer solution. The TS, SS, VSS, TCOD and
sCOD are determined by the Standard Methods [23]. The proteins
and polysaccharide in the supernatant of DWTS are determined
spectrophotometrically with a UV/Visible spectrophotometer
(UV2600, China) after centrifuging at 4000 rpm for10 min. Fluores-
cence measurements of the solubilized organics in the supernatant
are performed using a spectrofluorometer (F-4500, Hitachi, Japan)
equipped with a 150 W xenon lamp at ambient temperature of
24 °C. Zeta potential is analyzed with a Zetasizer Nano2000 (Mal-
vern Instruments, UK). BET specific surface area of samples is
determined by the BET/N2-adsorption method, before measure-
ment the samples are prepared as described by Wang et al. [24].
The average flocs size of the DWTS is calculated using image
analysis, as describe in the reference [7]. Briefly, after transferring
the sonicated DWTS sample to a flat microscope slide, three images
of the flocs are captured by an optical microscope (Olympus,
BX51TF, Japan) equipped with a video camera. The floc dimensions
are obtained by processing the flocs images using Image J software
to obtain values for the average size. The total number of the
aggregations larger than 14 pixels (10 lm) for a given flocs image
is about 30–50.
 Z. Zhou et al. / Ultrasonics Sonochemistry 24 (2015) 19–26 21

Table 1
Main physicochemical characteristics of raw DWTS.

Analytes (Units) Range Mean ± SD Analytes (Units) Range Mean ± SD

Temperature (°C) 19.7–20.4 20.4 ± 0.8 VSS/TS (%) 26.7–38.5 31.2
pH 6.99–8.37 7.60 ± 0.54 sCOD/TCOD (%) 2.87–4.24 3.72
TS (g L1) 1.33–2.13 1.57 ± 0.22 sCOD in supernatant (mg L1) 10.53–12.04 11.43 ± 0.56
VSS (g L1) 0.39–0.60 0.49 ± 0.07 Proteins in supernatant (mg L1) 0.23–0.40 0.37 ± 0.05
TCOD (mg L1) 298–331 307 ± 10 Polysaccharide in supernatant (mg L1) 10.10–14.46 11.82 ± 1.30

Note: SD means standard deviation. Number of measurements (n): for temperature and pH, n = 5; for total solids (TS), suspended solids (SS), volatile suspended solids (VSS),
total chemical oxygen demand (TCOD), sCOD, proteins and polysaccharide, n = 10.

Fig. 1. The arrangement of ultra-sonication experiments.

The reduction rate of the flocs’ size is calculated according to
the Eq. (1). Where R0, Ri is the average size of the flocs before
and after ultrasonic treatment at time t (min).
 
Rt
Reduction rate of flocs’ size ¼ 1   100%
R0
A dilution series (103–105) of each DWTS sample is prepared
by serially transferring a 1 mL portion of raw or sonicated DWTS.
Specifically, 1 mL raw or sonicated DWTS is first added into 9 mL
sterile water, and then 1 mL of the 1:10 diluted DWTS sample is
added into another 9 mL sterile water to prepare the 1:100 diluted
DWTS sample. Such procedure repeats once or thrice again to make
a dilution of 103 or 105. 1 mL of serially diluted sample is trans-
ferred to a plate containing Nutrient Agar medium. Incubation time
is 24 h at 35 °C and then counted. The concentration of total coli-
forms is determined by membrane filter technique. Total coliforms
are enumerated after the plates with agar medium are incubated at
37 °C for 48 h. The inactivation efficiency (IE) of total bacteria or
total coliforms is calculated according to the following Eq. (2).
Where N0, Nt is the number of total bacteria or total coliforms
before and after ultrasonic treatment at time t (min).
 
Nt
IE ¼ 1   100%
N0
3. Results and discussion
3.1. Effect of low energy density of two sonoreactors at frequency of 25
and 40 kHz
3.1.1. Disintegration and solubilization effect
As indicated in Table 1, the average TS, VSS and TCOD concentra-
tion of raw DWTS is 1.57 g L1, 0.49 g L1, 307 mg L1 respectively,
and the sCOD, proteins and polysaccharide in supernatant of raw
DWST is 11.43 mg L1, 0.37 mg L1, 11.82 mg L1 respectively.
The low VSS to TS ratio in the raw DWTS, 31.2%, which indicates
raw DWTS mainly consists of inorganic substances. The sCOD to
ð1Þ TCOD ratio is 3.72%, indicating that a large part of COD in raw DWTS
is associated with the insoluble phase.
The extent of DWTS disintegration quantified by the release of
sCOD is investigated, as shown in Fig. 2. The sCOD increase repre-
sents sludge disintegration that releases organic matters from
sludge flocs to water. For the bath sonoreactor, the release of sCOD
continuously increases with ultrasonic time, and it exhibits simi-
larity at the initial 15 min for the two frequencies, afterwards it
boosts and more apparently at 40 kHz than that at 25 kHz. At the
end of the sonication of 30 min, the release of sCOD is 70.73 and
59.67 mg L1 for 40 and 25 kHz, respectively. Comparatively, as
to the probe sonoreactor, the release of sCOD slightly differs within
20 min ultra-sonication for both frequencies, but it is much higher
at 25 kHz than that at 40 kHz at 30 min. As a whole, the extent of
DWTS disruption using bath sonoreactor is greater than the probe
one, and 40 kHz with 15 min or more has greater potential than
25 kHz to solubilize the carbon organics from the DWTS into the
bulk liquid. Several studies [25–27] have demonstrated that the
ð2Þ release of the organics from activated sludge using ultrasound is
mainly attributed to intensive cavitation which results in vigorous
and effective sludge disruption leading to release of intra-cellular
materials, and transformation of insoluble particulate organics into
soluble ones. Here these two aspects also play significant role in
the release of carbon organics from DWTS.
Generally, the reduction in flocs size of sludge is accompanied
with the release of sCOD [5,13,18]. The relationship of average size
of DWTS flocs and sCOD in the supernatant after ultrasound treat-
ment for two sonoreactors is plotted in Fig. 3. It is found that the
flocs’ average size of sonicated DWTS is significantly negatively
22 Z. Zhou et al. / Ultrasonics Sonochemistry 24 (2015) 19–26

80
0.8
(a) Probe sonoreactor
Energy density:0.03W/mL
16
70 25C
40C energydensity: 0.03W/mL
25T

Polysaccharide (mg/L)
60 0.7 14
40T

Proteins(mg/L)
sCOD(mg/L)

50
0.6
40 12

30 0.5

20 25 kHz-poteins 10
0.4 40 kHz-poteins
10 25 kHz-polysacchdide
40 kHz-polysacchdide
0 5 10 15 20 25 30 0.3 8
ultrasonication time(min) 0 5 10 15 20 25 30
Ultrasonication time(min)
Fig. 2. Effect of frequencies of 25 and 40 kHz of two sonoreactors on the release of
sCOD (energy density: 0.03 W/ml). Note: 25C–25 kHz, bath sonoreactor; 40C–
40 kHz, bath sonoreactor; 25T–25 kHz, probe-type sonoreactor; 40T–40 kHz, 2.0
probe-type sonoreactor. (b)
Bath sonoreactor 20
correlated with corresponding sCOD in the supernatant, suggesting 1.6 Energy density:0.03W/mL
that a greater average size of the flocs varies, a higher released

Polysaccharide (mg/L)
sCOD could be. Moreover, from the slope of the fitting line (3.21
Proteins(mg/L) 18
against 0.89), it can be deduced that the disintegration degree for 1.2
bath sonoreactor is greater than probe one, which is consistent
with the results from Fig. 2. Here under the low energy density
of 0.03 W/ml, the highest reduction rate of average size of soni- 16
0.8
cated flocs reaches up 23.32%, 37.69% respectively for bath and 25 kHz-poteins
40 kHz-poteins
probe sonoreactor at 25 kHz with 30 min ultra-sonication.
25 kHz-polysacchdide
Fig. 4 shows the concentration of soluble proteins and polysac- 0.4 40 kHz-polysacchdide 14
charide in the supernatant of DWTS after ultrasound treatment.
The increasing of soluble proteins and polysaccharide represent
0 5 10 15 20 25 30
cell lysis that transfers them into the supernatant. In present study,
the relatively higher polysaccharide concentration indicates that Ultrasonication time(min)
polysaccharide takes up a larger portion of the DWTS organic mat- Fig. 4. Concentration of released proteins and polysaccharide in the supernatant;
ters as compared to proteins. Meanwhile, the very low proteins for probe sonoreactor (a) and bath sonoreactor (b); energy density: 0.03 W/ml;
and polysaccharide content detected in the raw DWTS reflects frequencies: 25 and 40 kHz.
low concentration of extracellular polymeric substances.
extent of polysaccharide at 25 kHz is lower than that at 40 kHz.
As indicated in Fig. 4(a), the release of proteins into supernatant
The opposite trend of proteins and polysaccharide under different
basically continuously increases as the ultra-sonication time is pro-
frequencies is not consistent with the finding of Gallipoli and
longed, which might be due to the release of cell components dur-
Braguglia [28] that the increase of soluble proteins in solution
ing cells lysis. Meanwhile, the release extent at 25 kHz is higher
was significantly higher than that of carbohydrates. The concentra-
than that at 40 kHz. Additionally, the trend of polysaccharide with
tion of released proteins and polysaccharide in bath sonicator is
ultra-sonication time is similar as the proteins, and the release
comparatively displayed in Fig. 4(b). It is found that the concentra-
30 31 32 33 34 35 36 37 38 39 40 41 tion of released proteins and polysaccharide increases as the pro-
26 80 longing of the ultra-sonication time. What’s more, the release
extent of both proteins and polysaccharide at 40 kHz is greater than
that at 25 kHz. As compared to the probe sonoreactor, the highest
70 released proteins and polysaccharide can reach up to 1.85 and
24
20.23 mg L1 from 0.4 and 14.46 mg L1, respectively, indicating
bath sonoreactor has greater capacity to disrupt the raw DWTS
sCOD(mg/L)
sCOD(mg/L)

for probe reactor 60

which is well consistent with the findings of Figs. 2 and 3.
22 Y=-0.89X+49.55, Adj. R2=0.91 Additionally, the released proteins and polysaccharide for both
for bath reactor
sonoreactors in this study are very low, which can be explained by
Y=-3.21X+165.46, Adj. R2=1 50
two reasons. Firstly, the low energy density is applied here. Sec-
ondly, DWTS contains mainly of non-degradable substances, i.e.,
20 sand and metal hydroxides that are not easily disrupted, while
40
the activated sludge is composed mainly of biological solids that
are readily disrupted.
18 30
28 29 30 31 32 33 34
3.1.2. Fluorescence characteristics of solubilized organics
Average size of the flocs(um)
In order to clarify the composition and distribution of released
Fig. 3. Relationship of average size of DWTS flocs and released sCOD in the organics in the supernatant, the fluorescence characteristics are
supernatant for two sonoreactors. analyzed using the bath sonoreactor at 25 and 40 kHz with
 Z. Zhou et al. / Ultrasonics Sonochemistry 24 (2015) 19–26
sonication time of 5, 15 and 30 min as an illustration. The detail
data can be seen in our former study [6], which also can be found
in the supplementary file (Seen in Fig. 1S and Table 1S). There are
commonly five key fluorescence peaks referred to as fluorophores
A, C, T1, T2 and B in water samples [29–31]. Peak A related to
humic-like substance derived from the breakdown of plant mate-
rial exhibit maximums at emission (Em) wavelength of 420–
450 nm from excitation (Ex) at 230–260 nm. Protein-like fluoro-
phores including tryptophan-like (Peak T) and tyrosine-like (Peak
B) materials are usually detected at enhanced levels in water
impacted by domestic sewage. Peak T1 which is tryptophan-like
(protein-like) exhibits a maximum at an Em wavelength of
340 nm from Ex at the 275 nm, while Peak T2-tryptophan-like
(kex/em = 225–237/340–381 nm) is detected [29].
In this study, the peaks (A, T1 and T2) distinctly identified are
selected because fluorescence intensity of peaks B and C is rela-
tively lower. Briefly, as to the fluorescence spectra of organic mat-
ter in supernatant of raw DWTS, Peak T1, T2 and A is located at Ex/
Em wavelengths of 280.0/335.0 nm, 235.0/345.0 nm and 255.0/
425.0 nm, respectively. Fluorescence EEM spectra of organic mat-
ter in the supernatant after sonication at 25 or 40 kHz with sonica-
tion time of 5, 15 and 30 min are similar to that of raw DWTS
accordingly. The maxima of peak A, T1 and T2 is at the Ex/Em
250.0–255.0/420.0–445.0 nm, 280.0/325.0–335.0 nm and 230.0/
335.0–340.0 nm, respectively. The fluorescence intensities (Seen
in Table 1S) of the three Peaks of A, T1 and T2 firstly increase
and then decrease with the increase of sonication time at 25 kHz,
while significantly increase as sonication time increases at
40 kHz, indicating that both humic-like and protein-like sub-
stances are largely released into the supernatant during sonication.
3.1.3. pH, zeta potential and BET specific surface area variation of
DWTS flocs
The effect of sonication conditions applied here on changes in
physicochemical characteristics of DWTS flocs, i.e., pH, zeta poten-
tial and BET specific surface area is studied, as summarized in
Table 2. Here the change in pH and zeta potential is very small
under the investigated conditions (seen in Table 2). The 4pH is
in the range of 0.19–0.30. The 4zeta potential is in the range of
1.4 to +2.7 mV. For a given DWTS matrix, after sonication the
flocs are broken, leading to the interior encapsulated hydrolyzates
of poly-aluminum chloride or ferric chloride shift to the surface of
the flocs. As a result, the zeta potential of flocs is changed. But
there is no additional coagulant, so the zeta potential cannot signif-
icantly vary. The slight variance of pH demonstrates that the DWST
has a strong acid and alkalinity buffering capacity. The zeta poten-
tial variance indicate that US cannot effectively change the surface
charge of sludge particles, which is well consistent with Chu et al.
[7] who pointed out that US had no effects on the surface charges
of the suspended particles of activated sludge regardless of sonica-
tion time under ultrasonic energy of 0.11 and 0.33 W/ml.
According to our recent findings [32], the BET specific surface
area of flocs can be calculated based on the following Eq. (3).
Where S is the specific surface area of flocs, j is surface roughness
of flocs, Rpart is the radius of a single floc, q is the density of the
flocs. It can be found that specific surface area of flocs is propor-
tional to the surface roughness and is inversely proportional to
the particle size. Thus, when j is a constant, a larger average size
of flocs responds to a smaller specific surface area.
3j
S¼
Rpart q
As indicated in Table 2, the BET specific surface area of soni-
cated DWTS flocs for the two sonoreactors gradually increased as
the increasing of ultra-sonication time, and the improvement at
23
40 kHz is higher than at 25 kHz. Additionally, with comparison of
probe sonoreactor, the BET specific surface area of sonicated DWTS
flocs using bath sonoreactor is lower, responding to the smaller
average size of flocs and higher DWTS disintegration. The maxi-
mum BET specific surface area of 113.04 m2/g can be achieved
from 80.28 m2/g of raw DWTS under the frequency of 40 kHz with
30 min ultra-sonication using probe sonoreactor.
3.1.4. Microbial inactivation
Fig. 5 presents the inactivation efficiency of total bacteria and
total coliforms of two sonoreactors. A higher inactivation efficiency
of total bacteria or total coliforms reflects a lower microbial activ-
ity in DWTS. Schläfer et al. [33] assumed that under low energy
density, the microbial activity might increase since the declumping
effect was predominant instead of inactivating. As seen in Fig. 5(a),
for the probe sonoreactor, the inactivation efficiency of total bacte-
ria initially increases, and then reaches a plateau as the increasing
of ultra-sonication time. While for bath sonoreactor, the inactiva-
tion efficiency of total bacteria continuously increases. The highest
inactivation efficiency of total bacteria reaches up to 50% with
25 kHz and 30 min ultra-sonication of bath sonoreactor. The inac-
tivation efficiency of total coliforms shows the similarity with total
bacteria as the sonication prolongs. The highest inactivation effi-
ciency reaches up to 29.4% with 40 kHz and 30 min ultra-sonica-
tion of bath sonoreactor, which is about 9.4% higher than the
findings of Chu et al. [10] who observed that at frequency of
20 kHz under ultrasound density of 0.11 W/ml with sonication
time within 30 min, the inactivation rate of total coliforms was
about 20% in activated sludge. Additionally, it is comparatively
found that 25 kHz is better than 40 kHz for inactivating of total
bacteria, which is opposite for total coliforms’ inactivation, and
the bath sonoreactor is uniformly more beneficial to disinfect
microorganisms. Moreover, the inactivation efficiency of total col-
iforms is much lower than that of total bacteria, suggesting that
inactivation capacity of co-existence of gram-positive and gram-
negative bacteria is greater, though total coliforms is relatively less
resistant to sonication (as displayed in Table 3).
3.2. Effect of high energy density using probe sonoreactor at frequency
of 25kHz
3.2.1. Reduction in average size of sonicated DWTS flocs and release of
polysaccharide
It is stated that sonication density plays a critical role in the dis-
ruption for the formation and behavior of cavitation bubbles, and
the transient bubbles is regarded as the major contributor [13].
As mentioned in the Section 3.1.1, it is feasible to use the reduction
in flocs’ size to assess the disruption of DWTS. The effect of energy
density with frequency of 25 kHz using probe sonoreactor on mean
flocs’ size and concentration of released polysaccharide is investi-
gated. The result is shown in Fig. 5.
It can be seen in Fig. 6 that the average size of the sonicated
DWTS flocs decreases as the ultra-sonication prolongs. The higher
energy density attributes to a greater reduction in average size. The
highest reduction rate of flocs of 42.67% is achieved under the
energy density of 5 W/ml with 30 min ultra-sonication. Mean-
while, it is also indicated in Fig. 6 that a critical density level exist
beyond which the flocs can be sufficiently disintegrated. The result
is not well consistent with the findings of Show et al. [13] that at
the same specific energy of 10 kWh/kg DS (dried solid), the mean
particle size of activated sludge could reduce to 9, 13, and 19 lm
ð3Þ from 49 lm with the density of 0.52, 0.33 and 0.18 W/ml,
respectively. Additionally, the polysaccharide in the supernatant
of sonicated DWTS continuously increases as the increasing of
ultra-sonication time. At the highest energy density investi-
gated (5000 kW/m3), the highest concentration of released
24 Z. Zhou et al. / Ultrasonics Sonochemistry 24 (2015) 19–26

Table 2 60
pH, zeta potential and BET specific surface area of DWTS flocs before and after ultra- (a) energydensity: 0.03W/mL
sonication under low energy density of 0.03 W/ml.

Inactivation efficiency (%)

total bacteria
25T
US conditions Characteristics of sonicated DWTS flocs
25C
Frequencies Duration DpHa Dzeta BET specific surface 40 40T
(kHz) (min) potential area (m2/g) 40C
(mV)b
25C 5 0.22 0.5 87.34
10 0.07 0.7 91.53 20
15 0.01 0.3 95.63
20 0.02 0.7 101.34
30 0.02 0.2 100.44
40C 5 0.04 0.5 85.79 0
10 0.15 0.6 91.97
15 0.03 0.3 92.57
20 0.04 1.1 98.33 0 5 10 15 20 25 30
30 0.19 1.1 103.34 Ultrasonication time (min)
25T 5 0.01 0.6 85.76
10 0.03 0.7 91.09
15 0.03 0.6 99.68 30 (b) energydensity: 0.03W/mL
20 0.02 0.8 98.36 total coliforms

Inactivation efficiency (%)

30 0.05 1.3 107.44 25T
40T 5 0.22 1.4 87.36 25C
10 0.28 0.3 96.25 20 40T
15 0.30 2.7 102.57 40C
20 0.22 0.3 108.33
30 0.29 0.9 113.04

The pH, zeta potential and BET specific surface area of raw DWTS flocs is 7.60, 10
15.7 mV and 80.28 m2/g, respectively.
a
DpH = pHt  pH0;
b
Dzeta potential = zeta potentialt  zeta potential0.
0

polysaccharide reaches up to 23.61 mg/L, as compare to13.42 mg/L 0 5 10 15 20 25 30

with 30 kW/m3. As reported by Lehne et al. [34], 3000 kJ kg1 TS
were required for floc size reduction and decreased the median
size of the particles down to 10–15 lm, and with a higher amount
of specific energy (from 3000 to 100000 kJkg1 TS) a further sludge
disintegration could be achieved with the particle size as small as
5 lm. In his study, the maximum specific energy investigated for 1,
3 and 5 W/ml is 4000, 12,000 and 20,000 kJ kg1 TS, respectively.
Thus, the DWTS flocs can be effectively disintegrated with the
highest energy density of 5 W/ml, accompanying the most releas-
ing of polysaccharide.
3.2.2. pH, zeta potential and BET specific surface area variations of
DWTS flocs
The effect of energy density using probe sonoreactor with 5, 15
and 30 min ultra-sonication on the pH, zeta potential and BET spe-
cific surface area of sonicated DWTS is also investigated. Here the
change in pH and zeta potential is very small. The 4pH and 4zeta
potential is in the range of 0.13 to 0.38, 2.7 to 1.6 mV, respec-
tively, which indicates that the high energy densities also play no
significant effect on pH and surface charge of sonicated DWTS.
Interestingly, the BET specific surface area of sonicated DWTS flocs
increases as the duration prolongs at relatively lower energy den-
sity (1 W/ml), while it decreases as the increasing of ultra-sonica-
tion time at higher energy densities of 3 and 5 W/ml. From the Eq.
(3), it is known that the BET specific surface area is proportional to
the surface roughness of the flocs, and is inversely proportional to
the particle size. At relatively higher energy density (3 and 5 W/
ml), the surface roughness of the flocs firstly increases and then
decreases after a long exposure to sonication, which is confirmed
by the scanning electron microscope of flocs in our former study
[32]. At the same time, the flocs size continuously decreases and
the reduction rate becomes slower at later stage of the sonication.
Consequently, the BET specific surface area of sonicated DWTS
flocs reaches a peak value, and then decreases. Additionally, a
Ultrasonication time (min)
Fig. 5. Inactivation efficiency of total bacteria (a); and total coliforms (b) (Concen-
trations of total bacteria and total coliforms in raw DWTS is 6.5  104 CFU/mL, and
800,000 total coliforms/100 mL, respectively). Note: 25C–25 kHz, bath sonoreactor;
40C–40 kHz, bath sonoreactor; 25T–25 kHz, probe-type sonoreactor; 40T–40 kHz,
probe-type sonoreactor.
shorter ultra-sonication time can achieve a larger BET specific sur-
face area at higher energy densities. The results can be explained
that at higher energy density, i.e., 5 W/ml, the largely released
polysaccharide (in Fig. 6) has greater stickiness that bonds the dis-
rupted flocs together. The faster increasing of aggregations mass
surpassed the increasing of aggregations’ area, leading to a decline
of BET specific surface area.
3.2.3. Microbial inactivation
As indicated in Fig. 7 that the inactivation efficiency of the total
bacteria initially increases, and then reaches a plateau as the
increasing of ultra-sonication time for all cases of energy density
ranging from 0.03 to 5 W/ml. Additionally, the ultra-sonication
time that the inactivation efficiency reaches the plateau under
the highest energy density (5 W/ml) is much less than the lower
ones (0.03, 1 and 3 W/ml). The inactivation efficiency under rela-
tively higher energy density is higher than that of the lower ones.
The results indicate that shorter sonication time (5 min) with
higher (3 and 5 W/ml) energy density plays a more effective role
in controlling the microbial activity than longer sonication time
(20 min or more) with lower energy density(0.03 and 1 W/ml).
3.3. Specific energy analysis
The specific energy is widely used to reflect the energy con-
sumption and required electricity cost during ultra-sonication.
 Z. Zhou et al. / Ultrasonics Sonochemistry 24 (2015) 19–26 25

Table 3
pH, zeta potential and BET specific surface area of DWTS flocs with 5, 15 and 30 min ultra-sonication.

Duration (min) Energy density (W/ml) DpH 4zeta potential (mV) BET specific surface area (m2/g)
5 1 0.19 1.6 93.16
15 0.23 0.6 96.55
30 0.38 1.6 105.77
5 3 0.23 0.4 96.76
15 0.14 2.7 90.42
30 0.18 0.1 85.24
5 5 0.24 1.0 100.31
15 0.14 0.7 86.39
30 0.13 0.4 87.63

Note: The pH, zeta potential and BET specific surface area of raw DWTS flocs was the same as in Table 2.

250 W, as well as the initial raw DWTS volume of 50 mL. Based

26 1 24
on the experimental results above, as it is expected. The most vig-
3 W/ml 25kHZ
5 orous DWTS disintegration quantified by particles’ size reduction
24 22
and organic solubilization is achieved with 5 W/ml for 30 min

Polysacchride(mg/L)
ultra-sonication, responding to the specific energy of 1590 kWh/
Average size(um)

20
22 kg TS. Considering the DWTS disintegration and bacteria inactiva-
18 tion together, the low specific energy input using bath sonoreactor
20 is recommended in the future work to clarify the feasibility of
16 using ultrasound as pretreatment of sludge recycling process for
18 ballast flocculation of low-turbidity water.
14
16
12 4. Conclusions
14
10
0 5 10 15 20 25 30 In this paper, the frequency at 25 and 40 kHz of a bath and a
probe sonoreactor under low energy density of 0.03 W/ml on the
Ultrasonication time (min)
physical, chemical and biological characterization of a DWTS is
Fig. 6. Effect of energy density on flocs’ average size and concentration of examined. The energy densities (1, 3 and 5 W/ml) of the probe
polysaccharide in the supernatant of sonicated DWTS. sonoreactor with frequency of 25 kHz on the characteristics vari-
ance of DWTS are also investigated. The main conclusions can be
drawn as follows:

0.03 (1) With energy density as low as 0.03 W/ml, DWTS is limitedly
100 25 kHz
1 disrupted quantified by the release of sCOD, proteins and
3 W/mL
polysaccharide, as well as reduction rate of sonicated DWTS
Inactivation efficiency (%)

5
80
60
40
20
0 sonication, responding to the maximum released polysac-
0 5
Ultrasonication time (min)
Fig. 7. Effect of energy density on total bacteria inactivation efficiency (Mean
concentrations of total bacteria in raw DWTS is 5.0  105 CFU/mL).
For the bath sonoreactor, the specific energy applied is approxi-
mately in the range of 1.61–10.06 kWh/kg TS, with the conditions
of 450 W of energy input, 5–30 min of ultra-sonication time, 15 L
of initial volume of DWTS sample and 250 mL of successional sam-
pling volume respectively, as well as 1.57 g/L of mean TS concen-
tration of raw DWTS. As to the probe sonoreactor, the highest
specific energy is approximately 1590 kWh/kg TS, with the longest
ultra-sonication time of 30 min and the highest energy input of
flocs. From the perspective of disintegration degree and
microbial activity, the bath sonoreactor shows a little supe-
riority than the probe one, and 25 kHz displays nearly equal
behavior as 40 kHz does.
(2) The solubilized organics are mainly humic-like and protein-
like substances. The flocs’ average size of sonicated DWTS is
significantly negatively correlated with released sCOD.
(3) The highest reduction rate (42.67%) of sonicated DWTS flocs
is achieved with energy density of 5 W/ml for 30 min ultra-
charide of 23.61 mg/L compared to13.42 mg/L of 0.03 W/ml.
10 15 20 25 30
(4) The pH and zeta potential of sonicated DWTS slightly varies,
even the energy density is as high as 5 W/ml. The BET spe-
cific surface area moderately increases as the ultra-sonica-
tion prolongs at relatively lower energy density (0.03 and
1 W/ml), while it decreases with the increasing of ultra-son-
ication time at higher energy densities of 3 and 5 W/ml.
(5) The inactivation efficiency of total coliforms is much lower
than that of total bacteria with 0.03 W/ml. The highest inac-
tivation efficiency of total bacteria of 81.25% can be obtained
with 5 W/ml for 5 min ultra-sonication using probe sonore-
actor, suggesting a shorter sonication time with higher
energy density plays a more effective role in controlling
the microbial activity than longer sonication time with
lower energy density.
26 Z. Zhou et al. / Ultrasonics Sonochemistry 24 (2015) 19–26
Acknowledgements
This research is jointly funded by the National Natural Science
Foundation (51278005), Beijing Natural Science Foundation
(8132007) of China and Doctoral Fund of Innovation of Beijing Uni-
versity of Technology. We would like to give our sincere thanks to
the peer-reviews for their suggestions and discussions.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ultsonch.2014.
11.007.
References
[1] T. Okuda, W. Nishijima, M. Sugimoto, N. Saka, S. Nakai, K. Tanabe, J. Ito, K.
Takenaka, M. Okada, Removal of coagulant aluminum from water treatment
residuals by acid, Water Res. 60 (2014) 75–81.
[2] A.O. Babatunde, Y.Q. Zhao, Constructive approaches towards water treatment
works sludge management: an international review of beneficial re-uses, Crit.
Rev. Environ. Sci. Technol. 37 (2007) 129–164.
[3] Z.W. Zhou, Y.L. Yang, X. Li, W. Gao, H. Liang, G.B. Li, Coagulation efficiency and
flocs characteristics of recycling sludge during the treatment of low
temperature and micro-polluted water, J. Environ. Sci. China 24 (2012)
1014–1020.
[4] C.C. Wu, C.P. Huang, Effects of recycling-sludge operation on the structure and
moisture content of floc in water treatment plant, Sep. Sci. Technol. 32 (1997)
2873–2882.
[5] X.H. Guan, G.H. Chen, C. Shang, Re-use of water treatment works sludge to
enhance particulate pollutant removal from sewage, Water Res. 39 (2005)
3433–3440.
[6] Z.W. Zhou, Y.L. Yang, X. Li, Y. W. Liu, X. Guo, Characteristics and recycling effect
of solubilized organics derived from drinking water treatment sludge using
ultrasound, Desalin. Water Treat., doi: <http://dx.doi.org/10.1080/
19443994.2014.935804>.
[7] Z.W. Zhou, Y.L. Yang, X. Li, W.Q. Wang, Y. Wu, C.Y. Wang, J.L. Luo, Coagulation
performance and flocs characteristics of recycling pre-sonicated condensate
sludge for low-turbidity surface water treatment, Sep. Purif. Technol. 123
(2014) 1–8.
[8] H. Li, Y.Y. Jin, B.M. Rasool, Z.Y. Wang, Y.F. Nie, Effects of ultrasonic
disintegration on sludge microbial activity and dewaterability, J. Hazard.
Mater. 161 (2009) 1421–1426.
[9] P.Y. Zhang, G.M. Zhang, W. Wang, Ultrasonic treatment of biological sludge:
floc disintegration, cell lysis and inactivation, Bioresour. Technol. 98 (2007)
207–210.
[10] E. Joyce, S.S. Phull, J.P. Lorimer, T.J. Mason, The development and evaluation of
US for the treatment of bacterial suspensions: a study of frequency, power and
sonication time on cultured Bacillus species, Ultrason. Sonochem. 10 (2003)
315–318.
[11] C.P. Chu, Observations on changes in ultrasonically treated waste activated
sludge, Water Res. 35 (2001) 1038–1046.
[12] Sridhar Pilli, Puspendu Bhunia, Song Yan, R.J. LeBlanc, R.D. Tyagi, R.Y.
Surampalli, Ultrasonic pretreatment of sludge: a review, Ultrason.
Sonochem. 18 (2011) 1–18.
[13] K.Y. Show, T.H. Mao, D.J. Lee, Optimisation of sludge disruption by sonication,
Water Res. 41 (2007) 4741–4747.
[14] T.J. Mason, D. Peters, Practical Sonochemistry: Power Ultrasound and
Applications, second ed., Horwood Publishing Ltd., Chichester, U.K., 2002.
[15] T.J. Mason, A. Tiehm, Advanced in sonochemistry, Ultrasound in
Environmental Protection, vol. 6, JAI, Lightning Source UK Ltd., United
Kingdom, 2001.
[16] E. Gonze, S. Pillot, E. Valette, S. Gonthier, A. Bernis, Ultrasonic treatment of
aerobic activated sludge in batch reactor, Chem. Eng. Process. 42 (2003) 965–
975.
[17] B. Eder, F.W. Gunthert, Practical experience of sewage sludge disintegration by
ultrasound, Ultrasound in Environmental Engineering II, in: U. Neis (Ed.),
Reports on Sanitary Engineering, vol. 35, Technical University of Hamburg,
2002, pp. 173-188.
[18] D.H. Kim, E. Jeong, S.E. Oh, H.S. Shin, Combined (alkaline + ultrasonic)
pretreatment effect on sewage sludge disintegration, Water Res. 44 (2010)
3093–3100.
[19] G.H. Yu, P.J. He, L.M. Shao, Y.S. Zhu, Extracellular proteins, polysaccharides and
enzymes impact on sludge aerobic digestion after ultrasonic pretreatment,
Water Res. 42 (2008) 1925–1935.
[20] X. Feng, H.Y. Lei, J.C. Deng, Q. Yu, Hualiang Li, Physical and chemical
characteristics of waste activated sludge treated ultrasonically, Chem. Eng.
Process. 48 (2009) 187–194.
[21] S. Na, Y.U. Kim, J. Khim, Physiochemical properties of digested sewage sludge
with ultrasonic treatment, Ultrason. Sonochem. 14 (2007) 281–285.
[22] T. Kimura, T. Sakamoto, J.M. Leveque, H. Sohmiya, M. Fujita, S. Ikeda, T. Ando,
Standardization of ultrasonic power for sonochemical reaction, Ultrason.
Sonochem. 3 (1996) 157–161.
[23] APHA, AWWA, WEF, Standard Methods for the Examination of Water and
Wastewater, 19th ed., American Public Health Association, Washington, DC,
1995 (pp. 253–258).
[24] Y.L. Wang, B.Y. Du, J. Liu, J. Lu, B.Y. Shi, H.X. Tang, Surface analysis of
cryofixation–vacuum–freeze-dried polyaluminum chloride–humic acid (PACl–
HA) flocs, J. Colloid Interface Sci. 316 (2007) 457–466.
[25] K.K. Jyoti, A.B. Pandit, Effect of cavitation on chemical disinfection efficiency,
Water Res. 38 (2004) 2248–2257.
[26] U. Neis, K. Nickel, A. Tiehm, Enhancement of anaerobic sludge digestion by
ultrasonic disintegration, Water Sci. Technol. 42 (2000) 73–80.
[27] T. Mao, S.Y. Hong, K.Y. Show, J.H. Tay, D.J. Lee, A comparison of ultrasound
treatment on primary and secondary sludges, Water Sci. Technol. 50 (2004)
91–97.
[28] A. Gallipoli, C.M. Braguglia, High-frequency ultrasound treatment of sludge:
combined effect of surfactants removal and floc disintegration, Ultrason.
Sonochem. 19 (2012) 864–871.
[29] P.G. Coble, Characterization of marine and terrestrial DOM in seawater using
excitation-emission matrix spectroscopy, Mar. Chem. 51 (1996) 325–346.
[30] F.L. Rosario-Ortiz, S.A. Snyder, I.H. Suffet, Characterization of dissolved organic
matter in drinking water sources impacted by multiple tributaries, Water Res.
41 (2007) 4115–4128.
[31] T. Liu, Z.L. Chen, W.Z. Yu, S.Y. You, Characterization of organic membrane
foulants in a submerged membrane bioreactor with pre-ozonation using
three-dimensional excitation–emission matrix fluorescence spectroscopy,
Water Res. 45 (2011) 2111–2121.
[32] Y. Wu, Y.L. Yang, X. Li, Z.W. Zhou, Z.Y. Su, X. Guo, Effects of ultrasound on flocs
characteristics of drinking water precipitated sludge and its impact on
coagulation efficiency of sludge recycling process, China Environ. Sci. 34
(2014) 1166–1172 (In Chinese).
[33] O. Schläfer, M. Sievers, H. Klotzbuecher, T.I. Onyeche, Improvement of
biological activity by low energy ultrasound assisted bioreactor, Ultrasonics
38 (2000) 711–716.
[34] G. Lehne, A. Muller, J. Schwedes, Mechanical disintegration of sewage sludge,
Water Sci. Technol. 43 (2001) 19–26.