sequential anaerobic/aerobic digestion of waste activated sludge: analysis of the process...

ResearchPap

er

New Biotechnology �Volume 29, Number 1 �December 2011 RESEARCH PAPER

Sequential anaerobic/aerobic digestion ofwaste activated sludge: analysis of theprocess performance and kinetic study

M. Concetta Tomei, Sara Rita and Giuseppe Mininni

Water Research Institute, C.N.R., Via Salaria km 29,300, C.P. 10 - 00015 Monterotondo Stazione (RM), Italy

Sequential anaerobic–aerobic digestion was applied to waste activated sludge (WAS) of a full scale

wastewater treatment plant. The study was performed with the objective of testing the sequential

digestion process on WAS, which is characterized by worse digestibility in comparison with the mixed

sludge. Process performance was evaluated in terms of biogas production, volatile solids (VS) and COD

reduction, and patterns of biopolymers (proteins and polysaccharides) in the subsequent digestion

stages. VS removal efficiency of 40%, in the anaerobic phase, and an additional removal of 26%, in the

aerobic one, were observed. For total COD removal efficiencies of 35% and 25% for anaerobic and

aerobic stage respectively, were obtained. Kinetics of VS degradation process was analyzed by assuming

a first order equation with respect to VS concentration. Evaluated kinetic parameters were

0.44 � 0.20 d�1 and 0.25 � 0.15 d�1 for the anaerobic stage and aerobic stage, respectively. With regard

to biopolymers, in the anaerobic phase the content of proteins and polysaccharides increased to 50%

and 69%, respectively, whereas in the subsequent aerobic phase, a decrease of 71% for proteins and 67%

for polysaccharides was observed. The average specific biogas production 0.74 m3/(kgVS destroyed),

was in the range of values reported in the specialized literature for conventional anaerobic mesophilic

WAS digestion.

IntroductionAlthough biological processes are an effective way of treating

wastewater and ensuring minimum residual impact on the aquatic

environment, they have the serious drawback of producing high

amounts of excess sludge. Historically, it was common to see plant

layouts that showed the water treatment scheme in detail with all

of the process units and an arrow at the end that simply said

‘sludge to disposal’. This approach does not represent anymore the

reality and today it is recognized that without a reliable disposal

method for the produced sludge, the actual concept of water

protection would fail. Currently, production of excess sludge is

one of the most serious challenges in biological wastewater treat-

ment. Treatment and disposal of sewage sludge from wastewater

treatment plants (WWTPs) account for about half, even up to 60%,

of the total cost of wastewater treatment [1]. In addition to the

Corresponding author: Tomei, M.C. ([email protected])

1871-6784/$ - see front matter � 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.nbt.2011.03.006

high costs, the current legal constraints and public sensitivity for

sewage sludge disposal have provided considerable interest to

explore and develop strategies and technologies for minimization

of sludge production. To this aim one of the considered strategies

is the optimization of the digestion stage. Anaerobic mesophilic

digestion is extensively applied for sludge stabilization and was

demonstrated to be effective in reducing pathogens and destroy-

ing organic matter, but solid reduction above 50% is often difficult

to achieve [2]. A promising solution to improve the sludge diges-

tion performance, in terms of solid abatement, is the combined

anaerobic–aerobic process that has been investigated in the recent

years by several researchers in different configurations [2–4]. The

principle of operation, in the combined digestion, is the potential

improvement of biosolid degradation, due to the specificity

of metabolic pathways, aerobic and anaerobic, required by the

different sludge fractions. Moreover, combining the two digestion

methods can also be advantageous to reduce the drawbacks

www.elsevier.com/locate/nbt 17

mailto:[email protected]

http://dx.doi.org/10.1016/j.nbt.2011.03.006

RESEARCH PAPER New Biotechnology �Volume 29, Number 1 �December 2011

Research

Pap

er

attributed to the two single processes. In fact, aerobic digestion is

characterized by high energy demand but is simple to manage.

Instead anaerobic digestion is more complex to manage and less

stable, but allows the energy recovery with the produced methane.

Concerning the sequence of the process phases, both anaerobic–

aerobic and aerobic–anaerobic schemes have been proposed.

Pagilla et al. [5] observed a positive effect in terms of solid reduc-

tion and coliform abatement by adding a thermophilic aerobic

pre-treatment stage before the anaerobic digestion. Subramanian

et al. [6] focused their investigations on the dewatering properties

of the digested sludge and found a significant improvement when

the anaerobic digested sludge was post-treated in a subsequent

aerobic phase.

Advantages of the anaerobic–aerobic sequential digestion were

also highlighted in previous studies [4,7]. In comparison to the

conventional anaerobic digestion, the combined process con-

firmed the better solid removal efficiency, the improvement in

dewatering characteristics and a marked removal of the ammonia

nitrogen in the aerobic phase. Nitrogen removal in combined

digestion was also investigated by Zupancic and Ros [8] at different

intervals of temperatures and utilizing pure oxygen and air. They

found that the addition of an aerobic stage with pure oxygen

aeration to the conventional anaerobic digestion enhances ammo-

nium nitrogen removal that reached 85% efficiency at 8 days of

hydraulic retention time for the aerobic stage.

The objective of this study was to verify the performance of the

sequential anaerobic–aerobic digestion on real waste activated

sludge (WAS) by evaluating:� the removal efficiencies of volatile solids (VS) and COD in the

anaerobic and aerobic stages,� the removal efficiencies of colloidal proteins and polysacchar-

ides as specific sludge components determining the demand for

polymer conditioning agents,� the biogas production,� the VS degradation kinetics in both anaerobic and aerobic

phases.

Elements of novelty in the present paper are the testing of the

sequential digestion process on WAS, which is characterized by

worse digestibility in comparison with the mixed (primary and

WAS) sludge [9], and the evaluation of the kinetic parameters for

VS degradation in the two digestion phases.

Materials and methodsSludgeThe WAS utilized in this work was provided by the Rome North

wastewater treatment plant. The plant is a conventional activated

sludge system including screening, primary clarification and sec-

ondary treatment, and serves about 700,000 P.E. It is operated with

a relatively high sludge age (20 d) and it is characterized by a

diluted influent sewage. The influent COD average value is

200 mg/L that is in the very low end of the range of values,

210–740 mg/L COD, reported in Henze et al. [10] for settled waste-

water in EU countries.

Secondary sludge was obtained for each feed step from

the recycle stream, and then thickened for 18 h before utiliza-

tion, whereas the anaerobic inoculum was taken from the full

scale digester of the plant fed with primary and secondary

sludges.

18 www.elsevier.com/locate/nbt

ReactorsThe reactors utilized in this study are cylindrical glass vessels of

7.4 L volume.

The reactors were operated in series, the first reactor was oper-

ated under anaerobic conditions and was maintained at

37 � 0.58C by a thermostated jacket. The work volume was 7 L

and the sludge retention time (SRT) was controlled at 15 days. The

second reactor was operated under aerobic conditions with a work

volume of 4.5 L. Air was supplied by a compressor able to maintain

the concentration of dissolved oxygen at levels �3 mg/L. The

aerobic reactor worked at room temperature and the SRT was

controlled at 12 days.

In both reactors mixing was ensured by mechanical stirrers

equipped with helicoidal blades.

WAS was fed to the anaerobic reactor once per day, whereas an

equivalent volume of digested sludge was extracted from the

reactor and fed to the following aerobic reactor.

AnalysisRegular sample collection and analysis were started after one week

from the start up. Feed, anaerobic digested sludge and aerobic

digested sludge were analyzed for total solids (TS), VS, COD,

proteins and polysaccharides. Analytical methods and devices

are reported in the following.

Total and volatile solids – TS and VS were measured according to

Standard Methods APHA 2540B and APHA 2540E, respectively

[11].

Chemical oxygen demand (COD) – COD Cell Tests (MERCK-refer-

ring to EPA 410.4 method), based on potassium dichromate oxida-

tion and spectrophotometric determination (Spectroquant

Nova30), have been employed.

Proteins and polysaccharides – Samples were centrifuged for

10 min at 4000 rpm then the supernatant was filtered at 0.2 mm

and analyzed for the biopolymer concentration. Protein concen-

tration was determined according to the spectroscopic Bradford

method [12], based on the addition and reaction with the Comas-

sie Brilliant Blue and absorbance detection at 595 nm wavelength.

Polysaccharide concentration was evaluated by the Dubois

method [13] based on the reaction of the sample with phenol

and sulphuric acid. Absorbance of the treated sample was mea-

sured at 490 nm wavelength.

Biogas detection – Flow rate of biogas produced by the anaerobic

reactor was measured by a volumetric counter using a closed water

displacement system with electrical contacts and with an electro-

magnetic valve to discharge the produced biogas to the atmo-

sphere [14]. The measurement device is controlled by a

Programmable Logic Controller that also provides the recording

of signals.

Methane – Methane in the biogas was determined by a gas-

chromatograph PerkinElmer AutoSystem equipped with a Car-

boxen 1000 (Supelco) column and a TCD detector. Sample volume

was 50 mL, transport gas was helium (3 bar) and operating tem-

peratures were 1808C for the oven, 1508C for the injector and

2508C for the detector.

Results and discussionThe two reactors were operated in series in semi-continuous mode

for �2.5 months to have a work period long enough to be repre-


0

0.5

1

1.5

2

2.5

50454035302520time (d)

Vola

tile

Solid

s(%

)

feed anaerobic effluent aerobic effluent

FIGURE 2

Volatile solids concentration profiles in the fed WAS and in effluent from the

anaerobic and aerobic digesters. ResearchPap

er

sentative of the achievable performance in a real system. The start

up phase, that is the time required to have stable performance in

the two reactors, was quite short (�15 d), and this could be

reasonably explained with the use in both anaerobic and aerobic

reactors of biomass inocula already acclimatized. After the start-

up, even if the removal efficiencies are affected by the variability of

the fed sludge, the performance of the system confirmed the

beneficial effect of the additional aerobic stage on the VS and

COD removal for the entire operation period.

One of the key parameters affecting the performance of the

anaerobic digestion is the SRT, and different criteria can be con-

sidered in its choice, depending on the goals to be achieved. In

previous investigations on sequential digestion [2,4] the SRT of the

anaerobic phase was in the range of 10–15 d for mesophilic and 15–

20 d for thermophilic digestion. Optimal SRT values reported in

[15] for conventional mesophilic anaerobic digestion are in the

range of 15–18 days. The 15 d SRT value utilized in this study was

chosen taking into account these data with the objectives of

ensuring a good performance of the anaerobic digestion and, at

the same time, having a reduced reactor volume.

Concerning the aerobic phase, Kumar et al. [4], in their experi-

ments on secondary aerobic digestion of anaerobically digested

mixed sludge, worked with aerobic SRT values in the range of 3–9

d, so in this study, considering the poorer digestibility of WAS, a

more cautionary SRT value of 12 d was utilized.

Moreover, being the anaerobic SRT utilized in this study in the

range of optimal values for the conventional anaerobic digestion,

the characteristics of the anaerobically digested sludge in our

experiments are comparable to the ones obtainable with the single

anaerobic digestion. As a consequence, all the performance

improvements achieved with the additional aerobic phase can

be considered as advantages of the sequential digestion process.

VS and TS removalTS and VS concentration profiles, detected in the fed and digested

sludge, from anaerobic and aerobic reactors, are shown in Figs 1

and 2. Reported data are referring to one-month period (from the

20th to 50th day) whereas the removal efficiencies, as discussed

below, refer to the entire operation period.

The results must be read considering the low VS/TS ratio (0.4–

0.5) in the influent WAS sludge. Quite stable performance was

achieved in the anaerobic phase and the observed variation of VS

0

1

2

3

4

5

6

50454035302520time (d)

Tota

l Sol

ids

(%)


FIGURE 1

Total solids concentration profiles in the fed WAS and in effluent from theanaerobic and aerobic digesters.

removal efficiency can be reasonably attributed to the marked

variability of the VS concentration in the feed that, in the period of

reference, was 1.54 � 0.17 (expressed as %). Average anaerobic VS

removal efficiency, evaluated after the first start up period, was

40 � 10% within the range of values, 30–45%, reported for the

digestion of waste activated sludge [16]. In the subsequent aerobic

stage an additional VS removal of 26 � 9% was achieved, so the

global VS removal efficiency was significantly improved reaching

levels (56% average value) comparable with the values (60–65%),

reported in [2,4], for mixed sludge. The slightly lower efficiency, in

our case, is justified by the poorer WAS digestibility with respect to

the mixed sludge and by the operating conditions of the WWTP

that, because of the high SRT, produces a partially digested sec-

ondary sludge.

A similar trend in terms of removal efficiency was observed for

TS with a percentage of 31 � 8% for the anaerobic phase and

21 � 9% for the aerobic one.

Kinetic analysisIt is worth nothing that, in spite of the large diffusion of anaerobic

digestion as preferential treatment of sludge stabilization for med-

ium–high size plants, there is a lack of kinetic parameter data for

this substrate and this makes difficult the setting up of reliable

process models. To give a contribution to this aspect, the kinetic

analysis of VS removal in the anaerobic phase was performed by

correlating the experimental data with the following equation:

dXs

dt¼ �k � Xs

that is a first order kinetics with respect to the VS concentration,

where XS is the particulate substrate concentration (VS) and k the

kinetic constant. In spite of its simplicity, the first order equation is

the most utilized for modeling anaerobic sewage sludge degrada-

tion. According to [17] this empirical equation, if adequately

supported by experimental data, is potentially able to consider

the cumulative effects of many processes playing a role in a so

complex system where is really difficult to distinguish the active

biomass from the sludge volatile solids representing the substrate.

The kinetic coefficient k was evaluated by integrating the equation

over the subsequent time intervals. From VS data correlation

resulted an average k value of 0.44 � 0.20 d�1 comprised within



0

5000

10000

15000

20000

25000

30000

35000

50454035302520

time (d)

Tota

l CO

D (m

g/L)


FIGURE 3

Total COD concentration profiles in the fed WAS and in effluent from the

anaerobic and aerobic digesters.

0

10000

20000

30000

40000

50000

COD totTSVS

VS, T

S, T

otal

CO

D (m

g/L)


FIGURE 4

Overview of VS, TS and COD concentration patterns in the fed WAS and in the

effluent from the anaerobic and aerobic digesters.

Research

Pap

er

the range of values (0.17–0.60 d�1) reported in [18] for anaerobic

digestion of secondary sludge.

For the aerobic phase the VS degradation was modeled as a

biomass decay process with the same first order equation. The

evaluated average k value was 0.25 � 0.15 d�1 that is in the range

of values of the endogenous decay rate reported for activated

sludge in [10].

Both kinetic coefficients are characterized by a high standard

deviation value because of the marked variability of the VS con-

centration in the feed, but, in any case, they can be considered

acceptable for a biological process and constitute the first step to

model the sequential digestion process.

COD removalIn Fig. 3 the total COD profile, referring to the same period, is

shown. In the first anaerobic phase the removal efficiency was

35 � 8% whereas in the subsequent aerobic phase, an additional

removal of 25 � 14% was achieved.

If we consider the soluble COD fraction, a different pattern was

observed. WAS fed to the anaerobic reactor was characterized by a

low soluble COD content (about 30 mg/L), which increased after

anaerobic treatment up to 300 mg/L, then, in the aerobic phase,

the soluble COD is partially removed (43 � 11%).

On the basis of these results, we can conclude that the total COD

removal observed in the anaerobic phase (Fig. 3) was due to the

hydrolysis of the particulate matter (VS) that generated soluble

products not completely degraded in anaerobic conditions, there-

fore we found a fraction of them as soluble COD in the anaerobic

digester effluent.

To have a complete picture of the sequential digestion perfor-

mance, with reference to the classical evaluation parameters,

TABLE 1

Concentration of biopolymers in WAS and digested sludge samples

Day Proteins (mg/L)

WAS Anaerobically digested Aerobically digested

38 14.70 52.80 16.10

42 24.08 45.49 4.01

44 56.10 73.60 48.20

49 29.40 68.20 8.02


patterns of VS, TS and total COD concentration in the feed,

anaerobic effluent and aerobic effluent are summarized in Fig. 4.

Biopolymers (proteins and polysaccharides)It has been shown [19] that both in anaerobic and in aerobic

digestion of sewage sludge the biofloc destruction results in the

release of proteins and polysaccharides in colloidal form into

solution. These biopolymers cause the deterioration of the dewa-

terability properties of the sludge thus determining an increase of

the polymer conditioning demand. Novak and Park [20] found a

direct correlation between the concentration of biopolymers in

solution and the polymer conditioning dose both for anaerobic

and for aerobic digestion, so the reduction of the biopolymer

content in the digested sludge is certainly an advantage in terms

of cost reduction for sludge conditioning. The amount of biopo-

lymers released and the ratio of proteins to polysaccharides in

solution depend on the digester reaction environment (anaerobic

or aerobic) and on the operating conditions (i.e. SRT of the

digestion).

In this study, the concentrations of proteins and polysacchar-

ides were measured in the fed WAS and in the anaerobically and

aerobically digested sludge over a period of 11 days. The results are

shown in Table 1. An average release of 50 � 20% for proteins and

69 � 1% for polysaccharides was observed in anaerobic phase,

followed by an average removal of 71 � 26% for proteins and

67 � 15% for polysaccharides, in the subsequent aerobic phase

(Fig. 5). This is in agreement with the soluble COD pattern during

the subsequent digestion phases. In a former study [4] on sequen-

tial anaerobic aerobic digestion, with anaerobic SRT of 15 d and

aerobic SRTs in the range of 3–9 days, a similar protein pattern was

observed, while for polysaccharides appreciable removal was

Polysaccharides (mg/L)

WAS Anaerobically digested Aerobically digested

51.67 175.00 37.50

40.23 124.17 62.50

38.33 123.33 24.17

48.33 157.50 63.33


-100-80

-60-40-20

0

204060

80100

aerobic phaseanaerobic phase

% in

crem

ent o

r red

uctio

n

proteins polysaccharides

FIGURE 5

Percent release and removal of biopolymers in the anaerobic phase and

aerobic phase, respectively.

00.010.020.030.040.050.060.070.080.09

50403020100

time(d)

Bio

gas

(m3 )

0

0.4

0.8

1.2

1.6

2

50403020100

time (d)

Bio

gas

(m3 /k

gVS

dest

roye

d)

(a)

(b)

FIGURE 6

Cumulative biogas production (a) and specific daily biogas production

referred to the destroyed VS unit (b).

ResearchPap

er

observed only at aerobic SRT of 9 days. So, the marked removal of

proteins and polysaccharides observed in this study is consistent

with the employed aerobic SRT of 12 days.

BiogasBiogas production was continuously monitored during the experi-

ments as the relevant parameter to evaluate the efficiency of the

anaerobic digestion process. In Fig. 6a is reported the cumulative

biogas production and in Fig. 6b the specific production referred to

the destroyed VS unit. The cumulative biogas production shows a

regular increase thus confirming the stable performance of the

anaerobic stage during the experimental period. The average

specific biogas production is 0.74 � 0.15 m3/(kgVS destroyed),

and is comprised in the range of values reported in the specialized

literature, 0.6–1 m3/(kgVS destroyed), for mesophilic digestion of

activated sludge at SRT �20 days [16,21]. The biogas produced can

also be correlated with the added VS. In our work, this correlation

gives a specific production of 0.23 � 0.045 m3/(kgVS added), that

is also in the range of values for WAS reported in [21].

The methane content of the produced biogas was measured over

a one week period and was found to be equal to 65 � 4%, which is

consistent with literature data [15,16].

ConclusionsSequential anaerobic–aerobic digestion of WAS was extensively

investigated to verify the potential achievable advantages with

respect to the single conventional anaerobic or aerobic stabiliza-

tion processes.

Results of the study can be summarized as follows:� sequential anaerobic–aerobic digestion provided efficient VS

degradation: 40% efficiency in the anaerobic phase plus an

additional percent removal of 26% the aerobic one; VS

degradation process was also kinetically characterized by

evaluating the kinetic parameters of a first order equation

(with respect to the VS concentration) for both the anaerobic

and aerobic phases of the sequential digestion;� total COD showed removal efficiencies of 35% and 25% in the

anaerobic phase and aerobic phase respectively; the soluble

COD fraction increased in anaerobic phase (due to the

hydrolytic processes), and was removed with an efficiency of

43% in the subsequent aerobic one;� biopolymers (proteins and polysaccharides) released in the

anaerobic stage were effectively removed in the following

aerobic stage (71% and 67% efficiency for proteins and

polysaccharides respectively). The reduced biopolymer content

in the digested sludge is advantageous in that results in a

markedly reduced polymer demand for sludge conditioning;� biogas production was 0.74 �0.15 m3/(kgVS destroyed), which

is consistent with the values reported in the specialized

literature for conventional anaerobic digestion of WAS. The

value is indicative of a satisfactory performance (and energy

recovery) of the anaerobic digestion stage.

References

1 Wei, Y. et al. (2003) Minimization of excess sludge production for biological

wastewater treatment. Water Res. 37, 4453–4467

2 Novak, J.T. et al. (2011) Combined anaerobic–aerobic digestion for increased solids

reduction and nitrogen removal. Water Res. 45, 618–624

3 Hasegawa, S. et al. (2000) Solubilisation of organic sludge by thermophilic

aerobic bacteria as a pretreatment for anaerobic digestion. Water Sci. Technol. 41,

163–169

4 Kumar, N. et al. (2006) Effect of secondary aerobic digestion on properties of

anaerobic digestion biosolids. Water Environmental Federation 79th Annual Technical

Exibition and Conference, Dallas. pp. 6806–6829

5 Pagilla, K.R. et al. (2000) Aerobic thermophilic and anaerobic mesophilic treatment

of swine waste. Water Res. 34, 2747–2753

6 Subramanian, S. et al. (2007) Effect of anaerobic digestion and anaerobic/aerobic

digestion processes on sludge dewatering. J. Residuals Sci. Tech. 4, 17–23

7 Park, C. et al. (2006) The digestibility of waste activated sludges. Water Environ. Res.

78, 59–68

8 Zupancic, G.D. and Ros, M. (2008) Aerobic and two-stage anaerobic–aerobic sludge

digestion with pure oxygen and air aeration. Bioresour. Technol. 99, 100–109

9 Kopp, J. and Dichtl, N. (2001) Influence of the free water content on the

dewaterability of sewage sludges. Water Sci. Technol. 44, 177–183



Research

Pap

er

10 Henze, M. et al. (1987) IAWPRC task group on mathematical modelling for design

and operation of biological wastewater treatment. Activated sludge Model N. 1.

IAWPRC Scientific and Technical Reports n. 1 pp 33.

11 APHA, (1998) Standard Methods for the Examination of Water and Wastewater (20th

edn.),

12 Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of

microgram quantities of protein utilizing the principle of protein-dye binding.

Anal. Biochem. 72, 248–254

13 Dubois, M. et al. (1956) Colorimetric methods for determination of sugars and

related substances. Anal. Chem. 28, 350–356

14 Mata-Alvarez, J. et al. (1986) A simple device to measure biogas production in

laboratory scale digesters. Biotechnol. Lett. 8, 719–720

15 Dohanyos, M. and Zabranska, J. (2001) Anaerobic digestion. In Sludge into

Biosolids–Processing, Disposal, Utilization (Spinosa, L. and Vesilind, P.A., eds), pp.

223–241, IWA Publishing


16 Toreci, I. et al. (2009) Evaluation of continuous mesophilic anaerobic sludge

digestion after high temperature microwave pretreatment. Water Res. 43, 1273–

1284

17 Eastman, J.A. and Ferguson, J.F. (1981) Solubilization of particulate organic carbon

during the acid phase of anaerobic digestion. JWPCF 53, 352–366

18 Vavilin, V.A. et al. (2008) Hydrolysis kinetics in anaerobic degradation of

particulate organic material: an overview. Waste Manage. 28, 939–951

19 Novak, J.T. et al. (2003) Mechanism of floc destruction during anaerobic and

aerobic digestion and the effect on conditioning and dewatering of biosolids.

Water Res. 37, 3136–3144

20 Novak, J.T. and Park, C. (2004) Chemical conditioning of sludge. Water Sci.

Technol. 49, 73–80

21 Bolzonella, D. et al. (2005) Mesophilic anaerobic digestion of waste activated

sludge: influence of the solid retention time in the wastewater treatment process.

Process Biochem. 40, 1453–1460

sequential anaerobic/aerobic digestion of waste activated sludge: analysis of the process...

Documents