Fiyana Kusuma Dewi Alih Jalur Ganjil

16250525

Strategy For Minimization Of Excess Sludge Production From

The Activated Sludge Process
Yu Liu*, Joo-Hwa Tay
Environmental Engineering Research Center, School of Civil and Structural Engineering, Nanyang
Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore

1. Introduction
Removal of organic materials by biological oxidation is a core technology in
wastewater treatment process. New cells (sludge), carbon dioxide, soluble microbial products
and water are the end products for this process. Activated sludge process has been applied
worldwide in municipal and industrial wastewater treatment practice. Daily production of
excess sludge from conventional activated sludge process is around 15±100 L kg ÿ 1 BOD5
removed, in which over 95% is water (Metcalf and Eddy, 1991; Bitton, 1994). It is evident that
the general purpose of activated sludge process is in the removal of organic pollutants rather in
cultivation of excess sludge. McCarty (1966) anticipated a quasi-exponential growth of excess
sludge production in USA. In 1984, the excessive sludge to be treated in the European Union
countries reached 5.56 millions dry materials (Chudoba, 1991). With the expansion of
population and industry, the increased excess sludge production is generating a real challenge
in the field of environmental engineering technology. So far the regulations of food safety,
agriculture and sludge disposal in most countries are being more and more stringent in relation
to application of biosolids in agriculture and dumping into the sea. It should be realized that
biomass production is an important economic factor because the sludge generated is a
secondary waste that must be disposed in an environmentally sound and cost-effective manner.
Currently, production of excess sludge from activated sludge process is one of the most serious
problems encountered in wastewater aerobic treatment. The treatment of the excess sludge may
account for 25±65% of the total plant operation cost (Horan, 1990; Zhao and Kugel, 1997).
One has been looking for appropriate ways to reuse the excess sludge produced from the
activated sludge process (Tay and Show, 1997; Tay et al., 2000). An ideal way to solve sludge-
associated problems is to reduce sludge production in the wastewater purification process
rather than the post-treatment of the sludge generated. Strategies for minimization of excess
sludge production from activated sludge process are becoming a very practical and urgent isse.
Therefore, it appears to be very necessary to review techniques that can be applied for reducing
sludge production from industrial scale activated sludge process.
2. Theoretical consideration
2.1 Maintenance and endogenous metabolism

According to Pirt (1965), part of energy source would be used for maintaining living
functions of microorganisms, which is so-called maintenance metabolism. It is generally
considered that the maintenance energy includes energy for turnover of cell materials, active
transport, motility and so on. The importance of maintenance metabolism is that the
maintenance-associated substrate consumption is not synthesized to new cellular mass. Thus,
the sludge production should be inversely related to the activity of maintenance metabolism
(Chang et al., 1993; Liu and Tay, 2000b). On the other hand, to account for the decrease in
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

biomass production that is usually observed when the specific growth rate decreases, Herbert
et al. (1956) postulated that the maintenance energy requirement could be satisfied through
endogenous metabolism. In this case, part of cellular components is oxidized to produce the
energy for maintenance functions. In the environmental engineering literature, endogenous
respiration is usually the so-called autodigestion of biomass. In a continuous culture growth,
endogenous metabolisms could occur concurrently. The major advantage of the endogenous
metabolism is that the incoming substrate could be finally respired to carbon dioxide and water,
while results in a lower biomass production (Gaudy and Gaudy, 1980; Martinage and Paul,
2000). As Gaudy and Gaudy (1980) noted, ``the aim of both design and operation is to foster
as much of this biological incineration as possible, i.e. to enhance the aerobic decay leg of the
carbon-oxygen cycle''. It should be realized that the control of endogenous respiration would
have much practical significance as does the control of microbial growth and substrate removal
in wastewater treatment processes. In the following section, it can be seen that the theory of
endogenous metabolism provides a useful method for minimization of sludge production by
controlling sludge age.
2.2 Energy Spilling
Metabolism is the sum of biochemical transformations that includes interrelated
catabolic and anabolic reactions, and the behaviors of a microbial culture are determined by
the catabolism and anabolism. Microbiologists have assumed that the yield of cells is directly
proportional to the amount of energy produced via catabolism (Lehninger, 1975; Bitton, 1994).
However, there is evidence that under some conditions, such as existence of organic
protonophores, heavy metals, abnormal temperature and alternative aerobic±anaerobic cycle,
the variation in respiration would be far greater than the amount that could be ascribed to ATP
production, i.e. catabolism no longer couples to anabolism (Tsai and Lee, 1990; Cabrero et al.,
1998; Mayhew and Stephenson, 1998; Liu, 2000). There is evidence to show that under energy
uncoupling conditions, the microorganisms are able to overconsume substrate and a higher
substrate consumption rate is observed (Cook and Russell, 1994; Liu, 1999; Low et al., 2000).
Cook and Russell (1994) reported that bacteria utilized an energy source one-third as fast as
exponentially growing cells in the complete absence of growth (Cook and Russell, 1994). This
indicates that cells could dispose of their intracellular energy by dissipation of membrane
potential, ATP hydrolysis and futile cycles, and a major part of the substrate would be oxidized
to carbon dioxide under energy uncoupling conditions. In this case, the energy generated was
used to drive futile cycles of the energy, but no significant effect was imposed on the substrate
removal rate (Mayhew and Stephenson, 1998; Liu and Tay, 2000a; Low et al., 2000). It must
be realized that the specific feature of energy uncoupling is the breakdown and reformation of
substrate by microorganisms, but without a corresponding change in cell mass. In an
environmental engineering sense, the concept of energy uncoupling can be extended to the
phenomenon in which the rate of substrate consumption is higher than that required for growth
and maintenance. As a result, under energy uncoupling conditions the observed growth yield
of activated sludge would be reduced markedly. In theory, reduction in the growth yield means
that sludge production can be cut down by an equivalent percentage. This exhibits a promising
way to reduce excessive sludge production by controlling metabolic state of microorganisms
in order to maximize dissociation of catabolism from anabolism.
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

3. Practical Strategies
Intensive attention has been given to develop practical strategy for minimizing sludge
production from aerobic process of wastewater treatment through maximization of energy
spilling, endogenous and physicochemical lysis of cells. In the following, we are going to Y.
Liu, J.-H. Tay / Biotechnology Advances 19 (2001) 97±107 99 discuss the most promising
process techniques that can be applied to industrial scale operation for control of excess sludge
production in the activated sludge process.

3.1 Oxic-Settling-Anaeerobic Process

Oxic-settling-anaerobic process is a modification of conventional activated sludge
technology by inserting an anaerobic reactor in the recycling bypass of sludge. This process
has been successfully employed to repress the growth of filamentous organisms. Westgarth et
al. (1964) possibly for the first time reported that the insertion of a period of anaerobiosis in
the high-rate activated sludge process could reduce the rate of excess sludge production by half
as compared with that conventional process without anaerobic reactor. Since then, an
alternative fasting/feasting approach based on oxic and anaerobic cycling has been seriously
considered for minimizing excessive sludge production in the activated sludge process
(Chudoba, 1991; Chudoba et al., 1992; Ghiglizza et al., 1996; Copp and Dold, 1998; Chen et
al., 2000). For aerobic microorganisms, ATP is generated from the oxidation of exogenous
organic substrate. When the microorganisms are subject to anaerobic condition without food
supply, they are no longer able to produce the energy and have to use their ATP reserves as
energy source. During the anaerobic starvation period, the ATP would be exhausted. After
microorganisms return to food-enriched aerobic reactor, they have to rebuild necessary energy
reserves prior to biosynthesis because cellular synthesis could not proceed without a certain
intracellular stock of ATP. In this case, the substrate consumption should thus go to catabolic
metabolism to satisfy the energy requirement of microorganisms (Chudoba et al., 1991).
Therefore, it appears that alternative aerobic± anaerobic cycling of activated sludge would
stimulate catabolic activity, and make catabolism dissociate from anabolism. As pointed out
earlier, maximized energy uncoupling would result in a minimized sludge yield. The energy
uncoupling induced by alternative aerobic and anaerobic treatment constructs theoretical basis
of oxic-settlinganaerobic technique designed for excess sludge minimization. Chudoba and
Capdeville (1991), Chudoba et al. (1991) compared sludge production in oxic-settling-
anaerobic and conventional activated sludge processes, and found that in the oxic-settling-
anaerobic process, the specific sludge production was reduced by 20±65% as compared with
conventional process. On the other hand, the SVI value in the oxic-settlinganaerobic process
was much lower than that observed in conventional process, i.e. the oxicsettling-anaerobic
process could also improve the settleability of activated sludge. Another successful application
of oxic-settling-anaerobic strategy is in sequencing batch reactor (SBR) system, in which lower
sludge production and excellent settleability were obtained (Ghiglizza et al., 1996). With
increasing sludge loading rates, the sludge production in conventional activated sludge process
increases, however, the sludge production in the oxic-settling-anaerobic process shows a
reducing trend (Chudoba and Capdeville, 1991). This in turn implies that the oxic-settling-
anaerobic process is capable of handling high-strength organic pollutants without serious
sludge-associated problems. In view of industrial scale application, the oxicsettling-anaerobic
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

process provides a promising technique for efficiently reducing sludge production, while
improving the stability of process operation.

3.2 High Dissolved Oxygen Process

Different opinions can be found in the literature with regard to the effect of dissolved
oxygen concentration on sludge production (McWhirter, 1978; Roques et al., 1984; Metcalf
and Eddy, 1991; Bitton, 1994; Abbassi et al., 2000). It is generally recognized that in full scale
activated sludge process, supply of dissolved oxygen plays a limiting role for further increasing
loading rates of treatment facility. Results from purified oxygenation activated sludge process
show that the growth yield can be lowered by up to 54% as compared with conventional air-
activated sludge system even at high sludge loading rates (McWhirter, 1978). Boon and
Burgess (1974) compares the sludge production in oxygen and air-activated sludge systems,
and found that for the similar sludge retention time, the observed sludge yield in the pure
oxygen system was only 60% of that in the air system. Abbassi et al. (2000) recently reported
that the excess sludge production decreased from 0.28 to 0.2 mg MLSS mg ÿ 1 BOD5 as the
reactor DO was increased from 1.8 to 6.0 mg L ÿ 1 in a laboratory-scale conventional activated
sludge reactor. So far, the mechanism of reduced sludge production with high DO operation is
not yet thoroughly clear. In order to explain the observed DO-associated yield reduction,
McWhirter (1978) postulated that the high DO concentration would produce a higher level of
active biomass, and hence, a lower true sludge loading rate that would be measured relative to
a low DO system. The lower true sludge loading rate in turn results in a relatively lower sludge
production rate at the same apparent measured value of the sludge loading rate. It is most likely
that such explanation does not fully reveal the cause because lower sludge production can be
obtained even while operating at high sludge loading rate (Abbassi et al., 2000). In this case,
Abbassi et al. (2000) further proposed that the increase of the oxygen concentration in the bulk
liquid resulted in a deep diffusion of oxygen that subsequently led to an enlargement of the
aerobic volume inside the flocs. As a result, the hydrolyzed biomass in the floc matrix can then
be aerobically degraded and sludge quantity is thus reduced. As shown above, the yield values
obtained by Abbassi et al. (2000) seem to be 2-fold lower than those, typically 0.6 mg MLSS
mg ÿ 1 BOD5, reported by Boon and Burgess (1974), McWhirter (1978) in pure oxygen
processes. It is hard to imagine that over half of the sludge produced is decomposed through
autohydrolysis of biomass at high DO concentrations, thus the mechanism proposed by Abbassi
et al. (2000) needs further investigation. In fact, there is some evidence to show that the cell
surface hydrophobicity, microbial activity and exopolymer production are linked to dissolved
oxygen level in reactor (Roques et al., 1984; Mishima and Nakamura, 1991; Palmegren et al.,
1998; Pena et al., 2000). These in turn imply that dissolved oxygen could have some impacts
on energy metabolism of activated sludge, which further determines carbon flux flowing
between anabolism and catabolism. It is most likely that the reduced sludge production at high
DO concentration somehow would be a consequence of DO-induced metabolic changes of
activated sludge, however, a detailed study is required. Sludge bulking is one of the most
serious operation problems encountered in activated sludge processes. It has been demonstrated
sufficiently that low levels of dissolved oxygen in aeration basins favor the growth of
filamentous organisms that cause sludge bulking (Richard et al., 1985; Nowak et al., 1986). On
the contrary, high DO-activated sludge process can Y. Liu, J.-H. Tay / Biotechnology
Advances 19 (2001) 97±107 101 efficiently repress development of filamentous organisms in
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

the aeration tank (McWhirter, 1978; Bitton, 1994; Abbassi et al., 2000). As compared with
conventional air aeration process, the pure oxygen process has the advantage such that ability
to maintain a higher MLVSS concentration in the aeration tank; better sludge settling and
thickening; lower net sludge production; higher oxygen transfer efficiency per horsepower and
more stable operation (Benefield and Randall, 1980). Consequently, high oxygen process
shows great industrial potential for minimization of excess sludge production as well as in
improvement of system operation.

3.3 Uncoupler-containing activated sludge process

For most aerobic bacteria, ATP is generated by oxidative phosphorylation, in which
process electrons are transported through the electron transport system (ETS) from a source of
electrons at elevated energy levels (substrate) to a final electron acceptor (oxygen). According
to Mitchell (1961), the molecules directly using the proton gradient built up by electron
transport is considered proton-ATPase pumps that can be forced to reversibly operate. If
operating in forward direction, the pumps would use energy released by ATP hydrolysis to
drive proton across a membrane, but in cellular systems producing ATP, the pumps are driven
in reverse by the magnitude of the proton gradient produced by electron transport. It has been
known that such chemiosmotic mechanisms of oxidative phosphorylation can be effectively
uncoupled by organic protonophores, such as 2,4-dinitrophenol (dNP), para-nitrophenol (
pNP), pentachlorophenol and 3,30 ,40 ,5-tetrachlorosalicylanilide (TCS). In the presence of
organic protonophores, the majority of organic substrate would be oxidized to carbon dioxide
rather than used for biosynthesis. As a result, the growth efficiency is much lowered in
uncoupler-containing activated sludge process. Based on the above principle, much research
has been focused on development of uncoupler-induced energy spilling process for
minimization of excess sludge production (Okey and Stensel, 1993; Low and Chase, 1998;
Mayhew and Stephenson, 1998; Chen et al., 1999; Liu, 2000; Low et al., 2000). In a laboratory
scale activated sludge system, Low et al. (2000) reported that biomass production was reduced
by 49% when pNP was added to the culture. At a pNP concentration of 120 mg L ÿ 1 , no
sludge production was observed (Low and Chase, 1998). However, it should be pointed out
that most of the organic protonophores are xenobiotic and potentially harmful to the
environment, thus their application in wastewater treatment practice would not be very prudent.
Among those uncouplers, TCS seems to be more environmentally sound. In fact, TCS has been
often used as a component in formulation of soaps, rinses, shampoos and so on (Budavari et
al., 1989). Cook and Russell (1994) found that the majority of organic substrate was oxidized
via energy spilling and did not go to biosynthesis functions in TCS-containing microbial
culture. In a batch culture of activated sludge, Chen et al. (1999) also reported that the observed
growth yield was reduced by 78% at a TCS concentration of 0.8 mg L ÿ 1 as compared with
control culture without addition of TCS, but no significant effect on substrate removal kinetics.
It seems certain that the dissipation of energy via uncoupling biochemical processes can be
regarded as a promising technique for reduction of excessive production of activated sludge.
Industrial application of energy uncoupling induced by organic protonophores in two full102
Y. Liu, J.-H. Tay / Biotechnology Advances 19 (2001) 97±107 scale activated sludge plants in
Phoenix and Arizona had been reported by Okey and Stensel (1993). The negative effects of
protonophore-induced energy uncoupling at large scale would be an unexpected increase in the
oxygen requirement (Mayhew and Stephenson, 1998). Thus, it is expected that the combination
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

of pure oxygen aeration process with the metabolic uncoupling technique would generate a
novel and efficient biotechnology for the minimization of excess sludge production.

3.4 Ozonation-combined activated sludge process

To date, the disposal of excess sludge from biological wastewater treatment processes
causes a lot of economic and environmental problems. As pointed out earlier, the best way to
solve the sludge-associated problems is to minimize excess sludge production from processes
rather than the post-treatment of sludge after generation. For achieving such a purpose, a
combined system of activated sludge process and intermittent ozonation has been successfully
developed (Yasui and Shibata, 1994; Sakai et al., 1997; Kamiya and Hirotsuji, 1998). Similar
to oxic-settlinganaerobic process, recycling sludge is ozonated in the sludge circulation line.
Research by Kamiya and Hirotsuji (1998) showed that the excess sludge production was
reduced by 50% at an ozone dose of 10 mg g ÿ 1 MLSS in aeration tank per day. While the
ozone dose was kept as high as 20 mg g ÿ 1 MLSS in aeration tank per day, no excess sludge
was produced. A working principle for the ozonation-combined activated sludge process has
been proposed by Kamiya and Hirotsuji (1998) as follows, (i) a part of activated sludge in the
aeration tank is ozonated in the ozone reactor. Most activated sludge microorganisms in the
ozonation reactor would be killed and oxidized to organic substances; (ii) those organic
substances produced from the sludge ozonation can then be degraded in the subsequent
biological treatment. It should be stressed that after introduction of ozonation into activated
sludge process, the effluent quality in terms of dissolved organic carbon concentration is not
influenced significantly, however, the sludge settleability in terms of SVI was highly improved
as compared with control test without ozonation (Kamiya and Hirotsuji, 1998). Consequently,
ozonation-combined activated sludge process would be a useful technology for reducing excess
sludge production and further improving sludge settleability. Future research should be focused
on optimization of ozone dosage and dosing mode.
3.5 Control of sludge retention time and biodegradation of sludge
In the current activated sludge theory, sludge retention time (qc) is defined as the
average time a unit of biomass remains in the treatment system. Much research has shown that
qc is the most operational parameter in activated sludge process. For steady state system, the
qc is inversely related to the specific growth rate. It has been demonstrated that the relationship
between sludge yield ( Yobs) and sludge retention time can be described by the following
expression (Lawrence and McCarty, 1970). where Ymax is the true growth yield; and Kd is
specific endogenous rate. Eq. (1) shows that the observed growth yield is inversely dependent
on the sludge retention time and endogenous rate in steady state activated sludge process. This
equation also provides a theoretical basis for in-plant engineers to control the total sludge
production by adjusting the qc during the wastewater biological treatment. Much research has
been done to verify the relationship between the qc and Yobs as given by Eq. (1) (Sherrard and
Schroeder, 1973; Stall and Sherrard, 1976). Stall and Sherrard (1976) reported that excess
sludge production was reduced by 60% when the qc was increased from 2 to 18 days, but no
effect on COD removal efficiency was observed. On the other hand, results reported by
Wunderlich et al. (1985) clearly showed that in high-purity oxygen activated sludge system,
the sludge production was reduced from 0.38 to 0.28 mg VSS mg ÿ 1 COD removed as the qc
increased from 3.7 to 8.7 days. It appears from these results that the pure oxygen aeration
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

process operated at a relatively long qc would be much more beneficial for reduction of
excessive sludge production. Another way to achieve a lower sludge production is to enhance
cryptic growth, i.e. microbial growth on lysates. Research shows that microbial cell lysis can
be amplified by prolongation of sludge retention time or through physicochemical treatments
of sludge, such as thermal, alkaline and acid (Carrio et al., 1985; Yamamoto et al., 1989;
Canales et al., 1994; Rocher et al., 1999). In a membrane separation-combined activated sludge
reactor, 100% of the sludge can be kept in the reactor and the sludge retention time should be
long enough. Material balance on COD in this type of reactor shows that around 90% of the
influent COD is oxidized to carbon dioxide and suspended solid concentration in the reactor is
almost constant, without sludge wastage (Yamamoto et al., 1994). On the other hand, Rocher
et al. (1999) developed physicochemical techniques (thermal, alkaline and acid) to enhance
sludge lysis, and found that in thermal±chemical hydrolysis, sodium hydroxide was the most
efficient for inducing cell lysis. More recently, Ghyooy and Verstraete (2000) proposed a two-
stage membrane-assisted bioreactor to reduce sludge production. The first stage was a
completely mixed reactor without sludge retention for the stimulation of dispersed bacterial
growth, and the second stage was an activated sludge system in which growth of protozoa and
metazoa was stimulated. Solid±liquid separation was achieved by submerged membrane
filtration in the second stage. Results showed that such a system yielded a 20±30% lower sludge
production than the conventional activated sludge system, and this may be due to higher
amount of predators in the second stage of membrane-assisted reactor configuration.
Consequently, it is a reasonable consideration that activated sludge process combining
membrane separation and physicochemical lysis has great potential in reduction of sludge
production.
4 Prospects
The most promising process techniques for minimization of excess sludge production
were discussed. This review shows that the combination of the different processes would
generate novel and more efficient techniques for control of excess sludge production, such as
uncoupler-containing oxic-settling-anaerobic process, combined pure oxygen aeration and 104
Y. Liu, J.-H. Tay / Biotechnology Advances 19 (2001) 97±107 metabolic uncoupling process
and membrane-assisted physicochemical lysis process. These approaches would be interesting
and attractive for future research and process development.
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

Effect Of Ultrasonic, Thermal And Ozone Pre-Treatments On Waste

Activated Sludge Solubilisation And Anaerobic Biodegradability
C. Bougrier a, C. Albasi b, J.P. Delgenes` a, H. Carrere ` a,∗
a Institut National de la Recherche Agronomique (INRA), Laboratoire de Biotechnologie de
l’Environnement, Avenue des Etangs, 11100 Narbonne, France b Laboratoire de G´enie
Chimique (LGC), BP 1301, 5 rue Paulin Talabot, 31106 Toulouse Cedex 1, France Received
4 November 2005; received in revised form 14 February 2006; accepted 15 February 2006
Available online 6 March 2006

Abstract
In order to enhance the efficiency of anaerobic digestion, the effects of ultrasounds,
ozonation and thermal pre-treatment have been studied on waste activated sludge. The feature
of this study was to carry out the comparison of the three pre-treatments in the same conditions
and on the same sludge sample. Each treatment was tested in two conditions close to optimum
conditions to maximise batch anaerobic sludge biodegradability. All treatments led to chemical
oxygen demand and matter solubilisation and had little influence on mineral matter. In terms
of solubilisation thermal pre-treatment was better than sonication or ozonation. But, in terms
of batch anaerobic biodegradability, best results were obtained with ultrasounds with an energy
of 6250 or 9350 kJ/kg TS and a thermal treatment at 170 or 190 ◦C. Moreover, treatments had
effects on physicochemical characteristics of sludge samples: apparent viscosity decreased
after all treatments but the reduction was more important with thermal treatment. Median
diameter of sludge flocs were reduced after sonication, increased after thermal treatment and
did not change after ozonation. Finally, capillary suction time (CST) increased after ozonation,
increased highly after sonication and was reduced after thermal treatment. © 2006 Elsevier
B.V. All rights reserved. Keywords: Ultrasounds; Ozone; Thermal treatment; Solubilisation;
Methanisation; Particle size; Viscosity; Capillary suction time; Sewage sludge

1. Introduction

At present in France, sludge production is increasing, due to the enforcement of the

European legislation regarding the Urban Wastewater Treatments Directive (91/271/EEC). The
research of more efficient treatment is therefore necessary. Indeed, the Directive sets limits for
some of the established sanitary determinants (e.g. biological oxygen demand, suspended
solids and nutrients) and imposes small town to build plant in order to treat their wastewater.
This leads to an important increase in sewage sludge production. In the same time, disposal
routes are subject to more legal and social constraints: land disposal is now restricted in France,
incineration is quite expensive and land application (or agricultural use) is highly debated. This
causes a large problem to communities and wastewater treatment plant operators.
In order to solve this problem, it is necessary to reduce sludge production to the source
that is to say in the wastewater treatment plant. This is possible with anaerobic digestion. This
treatment, which allows a reduction of sludge quantity of about 40–50%, has become one
common method of sludge stabilisation, due to the production of biogas that makes the process
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

profitable. In wastewater treatment plants, anaerobic digestion is generally applied to mixture

of primary and secondary (waste activated) sludge. But waste activated sludge (WAS) are
known to be more difficult to digest than primary sludge [1]. Anaerobic digestion process is
achieved through several stages: hydrolysis, acidogenesis, methanogenesis. For WAS
degradation, the rate-limiting step is the hydrolysis [2]. In order to improve hydrolysis and
anaerobic digestion performance, one possibility is to use cell lyse pre-treatments. Several pre-
treatments can be considered: mechanical, thermal, chemical or biological treatments [3–5].
The aim of these treatments is to solubilise and/or to reduce the size of organic compounds,
and specially refractory compounds, in order to make them more easily biodegradable [6,7].
Final quantity of residual sludge and time of digestion can thus be reduced and biogas
production can be increased [8–10].
This paper deals with the comparison of three pre-treatments: ultrasounds, ozonation
and thermal pre-treatment. Ultrasonic pre-treatment leads to cavitation bubbles formation in
the liquid phase [11]. These bubbles grow and then violently collapse when they reach a critical
size. Cavitational collapse produces intense local heating and high pressure at the liquid–gas
interface, turbulence and high shearing phenomena in the liquid phase, but also formation of
radicals [12,13]. Moreover, it has been proved that the degradation of excess sludge is more
efficient using low frequencies: mechanical effects facilitate particles solubilisation [11]. Due
to its strong oxidative properties, ozone has been used for water and wastewater treatment.
During sludge ozonation, because of the complex composition of sludge, ozone decomposes
itself into radicals and reacts with the whole matter: soluble and particular fractions, organic or
mineral fractions[14,15]. Using ozone for sludge reduction has been widely studied. Optimal
consumed ozone dose ranges from 0.05 and 0.5 g O3/g of total solid: there is a phenomenon of
mineralisation for higher ozone doses [10,16]. Moreover, ozonation modifies viscosity and
settlement of sludge [17]. With thermal pre-treatment, cells are broken due to pressure
differences. These treatments also have the advantage of sanitising the sludge and enhancing
its dewatering properties [18]. According to several authors, optimal temperature is around
170–200 ◦C and treatment time has little effect [2,9,19]. Moreover, when combined with
anaerobic digestion, energy required to perform thermal treatment can be positively balanced
by biogas production [20]. Beside, thermal treatment has effects on the viscosity and the
filterability of sludge [21]. Previous works in the laboratory [22–24] were devoted to determine
optimum conditions of ultrasounds, ozonation and thermal treatment. For all treatments sludge
solubilisation increased with the treatment (ultrasonic specific applied energy, transferred
ozone dose and temperature). In all cases, solubilisation of matter was focused on organic
solids: mineral solids solubilisation was lower than organic solids solubilisation. During
ozonation, total COD and total solids remained constant for an ozone dose lower than 0.15 g
O3/g TS. For higher ozone dose, total concentrations seemed to decrease. This could suggest a
mineralisation phenomenon for ozone doses higher than 0.18 g O3/g TS. Sludge
biodegradability was assessed by batch anaerobic digestion tests. In all cases, biogas
production for treated samples was higher than for untreated sludge, but treated sludge was not
completely biodegradable in 16–20 days. For ultrasonic specific energy between 0 and 7000
kJ/kg TS, biogas production increased with energy supplied. But for energy supplied of 7000
and 15,000 kJ/kg TS, biogas production was almost the same. The optimum ultrasonic energy
was thus about 7000 kJ/kg TS [22]. Biogas production increased with ozone dose until 0.15 g
O3/g TS and then decreased. This was probably due to inappropriate acidic condition for micro-
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

organisms or due to the formation of recalcitrant ozonation by-products. Therefore, an ozone

dose of 0.15 g O3/g TS seemed to be the most interesting [23].
In the range of temperature tested (90–210 ◦C), optimal temperature seemed to be 170
or 190 ◦C: sludge biodegradability of treated sludge increased with temperature up to 170 ◦C.
The increase in biogas production was almost the same for 170 and 190 ◦C and was lower for
210 ◦C [24]. The objective of this study was to compare the effects of these three pre-treatments
(ultrasounds, ozonation and thermal pre-treatment) on waste activated sludge, in order to
improve anaerobic digestion. The feature of this work was to carry out the comparison of the
three pre-treatments in the same conditions and on the same sludge sample. Solubilisation of
chemical oxygen demand and matter were measured, as well as physicochemical
characteristics of sludge (pH, particles size, viscosity and filterability) and improvement of
biogas production during batch anaerobic digestion. For each treatment, we applied two
conditions close to the optimal ones [22–24]: • a specific energy of 6250 and 9350 kJ/kg TS,
for sonication, • an ozone dose of 0.1 and 0.16 g O3/g TS, for ozonation, • a temperature of
170 and 190 ◦C, for thermal treatment.

2. Experimental
2.1 Waste Activated Sludge Characteristics
Experiments were carried out using flottation-thickened WAS collected from the
municipal WWTP of Carcassonne (South of France). This plant had a capacity of 90,000
people equivalent, that is to say that the pollution in entrance was equivalent to 4.5 tonnes of
suspended solids per day. This plant treated domestic and industrial wastewater and was
operated with a high loaded aeration tank. For the experiments, sludge was diluted in order to
obtain a total solid concentration (TS) of 20 g/l. The organic solids (or total volatile solids VS)
content was 76% of TS.

2.2 Pre-Treatments Conditions

The ultrasonic apparatus used was an ultrasonic homogenizer Autotune 750 W
(Bioblock Scientific), working with a standard probe, an operating frequency of 20 kHz and a
supplied power of about 225 W. Batch experiments were carried out in beakers without
temperature regulation (no cooling). Treated samples had a volume of 0.5 l. Specific supplied
energy (Es) was 6250 and 9350 kJ/kg TS. Specific energy (Es) is defined using ultrasonic
power (P), sonication time (t), sample volume (v) and initial total solid concentration (TS0):
Es = (Pt)/(vTS0) (1) Sludge was ozonized in batches in a bubble column. Ozone was generated
from pure oxygen using an OZAT® type CFS-1 generator (Ozonia). Ozone concentrations in
gas phase, before and after reaction with sludge, were measured, during the oxidation every 30
s, with UV analysers BMT 96 3 in order to calculate the amount of O3 that was transferred.
Gas flow rate was 1 l/min, ozone inlet concentration was about 30 mg/l. Reactor volume was
1 l, sample volume was 0.3 l. The transferred ozone dose was 0.10 and 0.16 g O3/g TS. The
reactor used for thermal treatment was a Zipperclave (Autoclave France) which temperature
was controlled by PID (proportional integral derivative). The autoclave was equipped with an
Hastelloy C tank. Sample volume was 0.5 l. Tested temperatures were 170 and 190 ◦C. For 190
◦C, treatment duration corresponded to the rise in temperature (about 60 min) and for 170 ◦C
experiments lasted 30 min more.
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

2.3 Samples Analyses

2.3.1 Solubilisation evaluation
In order to determine sludge solubilisation, several measurements were carried
out on samples, according to Standard Methods [25]. In fact, the term “solubilisation”
represents the transfer (of COD or solids) from the particulate fraction of the sludge (solids
after centrifugation) to the soluble fraction of the sludge (supernatant after centrifugation).
First, soluble and particulate fractions were obtained after centrifugation (Beckman J2 MC
25,000 × g, 15 min, 5 ◦C). Then, solubilisation was characterised by chemical oxygen
demand (COD), solids repartition (soluble/particulate or organic/mineral). COD was
measured on the total sludge and on the supernatant, using the normalised method [25].
For this paper, COD measured on supernatant will be called “soluble COD” (CODs) and
COD measured on the “solids of centrifuge” will be called “particulate COD” (CODp).
COD solubilisation (SCOD) was calculated using the difference between soluble COD
(CODs) and initial soluble COD (CODs0), compared to the initial particulate COD
(CODp0). SCOD = (CODs − CODs0)/CODp0 × 100% (2) Solids concentrations were
estimated by heating (105 ◦C during 24 h for total solids and 550 ◦C during 2 h for mineral
solids concentrations). Volatile solids concentrations were deduced. Measures of total and
organic solids (TS and VS) were realised on sludge and on solids after centrifugation (total
and volatile suspended solids: TSS and VSS). Solids concentration of the supernatant, that
is to say the soluble phase, was then deduced. That led to the composition in the different
parts in the sludge. Matter solubilisation (STS) was calculated.

STS = (TSS0 − TSS)/TSS0 × 100%

2.3.2 Sludge Characteristics

Particle size measurements were realised using a laser diffraction sensor
(Mastersizer 2000, from the Malvern firm). Particle size was determined using a sphere of
same volume. Results were expressed in median diameter (d). Viscosity measurements
were carried out using a High Resolution C-VOR viscosimeter (Bohlin Instrument)
connected to a computer. The system was a “plane-cone” with a diameter of 60 mm. The
angle of the cone was 2◦ and the gap measured 70m. Measures were realised by increasing
the shear stress from 0.01 to 8 Pa and were duplicated. Their reproductibility was good.
Capillary suction time measurements (CST) were realised using a Triton type 319 Multi-
CST (Triton Electronics Ltd.). The CST permits to estimate the sludge ability to
dewater:water is absorbed by CST paper by capillarity. The CST measure corresponds to
the time needed for water to cross a fixed distance in the filter paper. It is often used to
estimate sludge filterability.

2.3.3 Characterisation of Anaerobic Biodegradability

Batch anaerobic digestion tests were carried out to assess sludge biodegradability.
For these experiments the inoculum used was a sludge treating a mixture of wine
effluents (80%) and sludge (20%). The inoculum was diluted to 4 g/L of volatile
suspended solids (VSS-equivalent to organic suspended solids). For each pre-treatment,
samples of treated or untreated sludge were added to 400 ml of inoculum. The pollution
to degrade was equivalent to 0.5 g COD/g VSS of inoculum. Plasma bottles were
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

agitated (200 rpm) at 35◦C. Several control samples were realised: a blank (water), an
ethanol sample (completely biodegradable compound) and a untreated sample (raw
sludge added). Biogas volume produced was measured by movement of liquid (water,
pH 2, NaCl 10%). Enhancement of biodegradability was evaluated by comparison of
biogas volumes produced by treated and untreated samples. Biogas ratio is biogas
volume produced with treated sludge divided by biogas volume produced with
untreated sludge. Moreover, the biodegradability percentage was estimated by
comparing the biogas volume produced with sludge (treated or not) to the biogas
volume produced with ethanol. For all substrates, volumes of biogas were reported to
the introduced COD (quantity of pollution to degrade).

3. Results and Discussion

3.1 Effects of Pre-Treatments on Matter and COD Solubilisation
During these comparative tests, total COD and TS concentrations remained almost
constant for all treatments, except for thermal treatment at 170 ◦C (sludge slightly stuck on the
lyse reactor walls). COD concentration was 15.0 g O2/l (standard deviation: 1.0 g O2/l); TS
one 18.9 g/l (S.D.: 1.1 g O2/l). Treatments led to a transfer from particles to supernatant. Fig.
1 presents results obtained in terms of COD and TS solubilisation. For each technique, COD
solubilisation and TS solubilisation were quite similar: almost 15% for sonication, 20–25% for
ozonation and 40–45% for thermal treatment. Thus solubilisation results were very different
according to used techniques. Solubilisation is much more higher with thermal treatment than
with sonication or ozonation. Moreover, for a given technique, there were few differences
between the two chosen operating conditions, which were relatively close. Pre-treatments led
to a modification of the repartition of the solids (Fig. 2). For all treatments, total mineral solids
concentration was almost constant: sonication, ozonation or thermal treatment did not lead to
a mineralisation phenomenon in those conditions. Moreover, mineral concentration in particles
seemed to be almost constant to: 3.8 g/l (S.D.: 0.3 g/l). Therefore, it means that mineral solids
were only slightly affected by pretreatments: mineral solids solubilisation was very low (less
than 10%). At the opposite, organic solids were highly affected by treatment. Organic solids
concentration in particles decreased strongly from 14.4 g/l in raw sludge to 11–12 g/l for sludge
treated with ultrasounds, 9–10 g/l for sludge treated with ozone and around 7 g/l for thermally
treated sludge. Thus, specially for thermal treatment, particular fraction of sludge became more
mineral. The ratio VSS/TSS decreased from 78% for raw sludge to 73% for sludge treated with
ultrasounds or ozone and 66% for thermally treated sludge. In conclusion, thermal treatment
was the most efficient treatment in terms of matter solubilisation and sonication and ozonation
led to almost the same results (lower than for thermal treatment).

3.2 Effects of Pre-Treatments on Physico-chemical Characteristics of Waste Activated

Sludge
Lyse pre-treatments led to modification of the physicochemical characteristics of
sludge. For instance, pH decreased with ozonation or thermal treatment, in order to reach a
value of 5.8. This can be explained by the formation of acidic compounds. On the other hand,
for sonication, pH was not modified. Thus, it seems that sonication, ozonation and thermal
treatment do not act in the same way. Ozonation and thermal treatment led to the modification
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

of sludge composition: organic compounds were directly affected by treatment. In fact, it seems
that lipids were degraded in order to form volatile fatty acids, which decreased the pH [24].
Treatments also had effects on rheology of sludge. WAS are generally non-Newtonian fluids:
the shear stress (τ) is not linearly related to the shear rate (γ˙) as shown in Fig. 3. The treated
sludge curves were lower than the curve for untreated sludge. That means that apparent
viscosity (τ/γ˙), for a given shear rate, decreased with treatments. For instance, for a shear rate
of 100 s−1, the apparent viscosity of raw sludge was 0.034 Pa s, and it decreased to 0.009–
0.014 Pa s for sonicated or ozonized sludge and to 0.003 Pa s for thermally treated sludge. In
the same time, it seemed that rheology curve for thermally treated sludge became more linear.
This could mean that sludge tended to become more Newtonian after thermal treatment. In
order to verify this hypothesis, we used a rheology model. According to literature, the Ostwald
de Waele model is the most used model in the case of sludge [26,27]. τ = kγ˙ n (4) In this
model, k is a constant, linked to the structure: the higher the value is, the higher the apparent
viscosity. The structure index n is linked to the apparent viscosity dependence to shear rate: n
equals to 1 for Newtonian fluids. Table 1 presents model parameters, which were determined
by fitting experimental data. The table also presents the apparent viscocity measured for two
different shear rates. So, during experiment, k constant strongly decreased with treatments: k
was divided by 10 for sonicated and ozonized sludge and by 100 for thermally treated sludge.
In the same time, the structure index n increased from 0.42 to 0.89 for thermally treated sludge.
Sludge became therefore more Newtonian and apparent viscosity decreased strongly: viscosity
was divided by 16, for a shear rate of 56 s−1, for sludge thermally treated at 170 ◦C. Thus
treated sludge should be easier to pump. An other modification due to pre-treatment was
particles size.
For the applied energies, sonication decreased median diameter from 36m for raw
sludge to 10m: that is to say a decrease of 70%. On the contrary, thermal treatment led to an
increase in the diameter. Ozonation did not seem to affect particles size. These modification
meant that sonication broke aggregates, flocs and maybe cells [29]. On the other hand, thermal
treatment led to particles agglomeration. This could suggest that the rise in temperature led to
the creation of chemical bonds. Treatments also led to a modification of the sludge filterability.
Table 3 shows CST measures for the different treatments. The three treatments did not have
the same effect on filterability. Sonication and ozonation increased strongly the CST value,
whereas, thermal treatment decreased it. This can maybe be linked to the particles size. But, in
priori, particles size is not the only parameter having an impact on CST. Indeed, particles size
remained constant with ozonation. Sonication, by decreasing particles size, led to the damage
of filterability. This confirms results obtained by Chu et al. [28].
They observed an increase of the CST, which they explain by an increase of the bound
water linked to the particles surface: as the particles diameters were lower, the surface of
contact was higher and the quantity of bound water too. On the other hand, thermal treatment
led to the release of more water, by breaking the sludge structure. The reached temperature
could have effect on hydrogen bonds, which gave structure to sludge. By modifying this
structure, it was possible to release a part of the initial bound water. Moreover, thermal
treatment was initially used as a dewatering pre-treatment [18].
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

3.3 Effects of Pre-Treatments on Batch Anaerobic Digestion

The aim of pre-treatments is to improve sludge anaerobic digestion. Therefore, batch
anaerobic digestion tests were realised in order to choose the best treatment. Fig. 4 presents
results of batch anaerobic digestion tests. All pre-treatments allowed a biogas production
equal or higher than for untreated sludge, but total produced biogas volume remained
lower than for ethanol (totally biodegradable substrate). All sludge samples were not
completely biodegradable in 24 days. However, for thermal treatment and sonication, final
biogas volumes produced with treated sludge were only slightly lower than the biogas
volume produced with ethanol. For sonication, results obtained for the two conditions of
treatment were identical during the whole test. It was the same for thermal treatment but
both samples of ozonized sludge showed little difference. Table 4 presents the biogas
production enhancement, the methane content and the biodegradability percentage, at the
end of the test (day 24). For sonication and thermal treatment, results obtained in terms of
biogas production enhancement were similar to those obtained in previous work with the
same sludge and in almost the same conditions [22,24]. On the contrary, results obtained
with ozone treatment were surprising. At the end of the experiments (day 24), biogas
volume produced with sludge treated with an ozone dose of 0.16 g O3/g TS was only 1.25
times higher than volume produced with untreated sludge. Previously, this enhancement
was equal to 2.58 [23]. This could be due to inhibitory conditions (to much ozone remained
in the soluble phase), to the formation of refractory compounds, to a not well-adapted
inoculum or to ozone consumption by reduced compounds of the sludge. An other
explanation could also be the initial biodegradability percentage of raw sludge. For the
previous studies with ozone [23], initial biodegradability was around 35%. For this
comparative series of tests, initial biodegradability was 63%. This variation of anaerobic
biodegradability could be explained by variations in sludge composition and/or in used
inoculum. Moreover, it seems that thermal treatment slightly decreased methane content:
it passed from 75% for raw sludge to 70% for thermal treatment. On the other hand,
sonication and ozonation did not affect the methane content. After 24 days of tests, sludge
was almost totally biodegradable with sonication and thermal treatment: biogas volume
produced with treated sludge was equal to almost 95% of the volume produced with
ethanol, but it has to be reminded that initial sludge biodegradability was high. Treatments
also permitted to accelerate sludge degradation. For untreated sludge, it took 24 days to
produced 300 ml of biogas per gram of COD added, and only 3 days for thermally treated
sludge, 6 days for sonicated sludge and around 15–18 days for ozonized sludge (depending
transferred ozone dose). These results can have important consequences in continuous
anaerobic digestor operation. Indeed, sludge lysis pre-treatments can be used either to
maximise biogas production (and to minimise residual sludge amount) or to accelerate
sludge anaerobic digestion and to treat more sludge in a given digestor by reducing sludge
retention time (SRT). The 24 days needed to produce 300 ml of biogas per gram of COD
added in batch test are of the same order of magnitude as classical SRT of raw sludge
(generally 20 days). Fig. 4 shows that the same volume of biogas could be produced with
a SRT of 3 and 5 days with a thermal or ultrasonic pre-treatment. However, such low SRT
could lead to methanogenic micro-organisms washing out. Finally, Fig. 4 shows that a
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

SRT of 15 days could be a good choice if thermal or ultrasound is applied. It would allow
to reduce classical SRT of 25% and to increase biogas production.
Fig. 5 presents methane production versus COD solubilisation, for the three treatments
(day 24). A COD solubilisation of 20% led to a methane production increase of 147 ml
CH4/ g CODadded for sonication and around 40–45 ml CH4/g CODadded for ozonation
or thermal treatment. Slopes of curves have been calculated, despite of the little number of
experimental points. Indeed, previous separate studies have shown that, for all the
treatments used, methane or biogas production was linearly correlated with COD
solubilisation [24]. Therefore, ultrasounds allowed a weak solubilisation of COD and a
high biodegradability, whereas, ozonation allowed a weak solubilisation and a weak
biodegradability and thermal treatment a strong solubilisation and a strong
biodegradability. The values determined for slopes allowed us to make suppositions about
biodegradability of treated sludge. So, it seemed that treatments were controlled by
different mechanisms.
In the case of sonication, the COD solubilised (approximately 20%) was almost totally
and rather quickly biodegradable. At the same time, ultrasounds acted on the particular
material. Particles size decreased and exchange area between particles and liquid phase
increased. Particular COD thus became more easily available to micro-organisms and the
biodegradability of particles improved but remained rather slow. For thermal treatment,
solubilised COD (approximately 50%) was also almost totally and rather quickly
biodegradable, but a small fraction remained slowly biodegradable. Moreover, considering
the particular part of sludge, thermal treatment had less impact than the sonication: the
accessibility of CODp seemed to be less modified than for the sonication. In the case of
ozonation, COD was weakly solubilised (approximately 20%) and CODp biodegradability
was little modified (same level as for thermal treatment). So, the sonication main action
was focused on particles accessibility, whereas, thermal treatment was mainly centred on
compounds solubilisation.

4. Conclusion
All three techniques led to solids solubilisation and to anaerobic biodegradability
enhancement. In terms of solubilisation, thermal treatment was the most efficient. In the same
time, thermal treatment led to a strong decrease of apparent viscosity, a strong increase in
filterability and an increase in particles diameter. Sonication led to a decrease in particles size,
in apparent viscosity and in filterability. Ozonation also led to a decrease in apparent viscosity
and filterability, but had no effect on particles size. In terms of anaerobic biodegradability, pre-
treatments led to an enhancement of biogas production. Nevertheless, for ozonation, this
enhancement was low, but raw sludge biodegradability was very high (60%). This was
probably due to inhibitory conditions or refractory compounds. On the contrary, sludge became
almost completely biodegradable within a 24 days batch with sonication (6250 or 9350 kJ/kg
TS) and thermal treatment (170 or 190 ◦C). This was the result of two different mechanisms.
Ultrasounds led to little solubilisation of sludge and to particle size reduction, which allowed
to enhance the particulate fraction biodegradability. Thermal treatment main effect was solid
solubilisation and did not affect remaining particulate biodegradability. It was therefore
impossible to find out the best treatment among sonication and thermal treatment: there was
not enough difference between sludge biodegradability results. So, in order to chose one
Fiyana Kusuma Dewi Alih Jalur Ganjil
16250525

treatment rather than the other, it is necessary to consider other parameters like costs, sanitation,
simplicity of installations.