Bioresource Technology 98 (2007) 1199–1207

Experiences with the dual digestion of municipal sewage sludge

Sebastian Borowski *, Jozef Stanisław Szopa

Technical University of Lodz, Institute of Fermentation Technology and Microbiology, ul. Wolczanska 171/173, 90-924 Łodz, Poland

Received 16 February 2005; received in revised form 27 February 2006; accepted 9 May 2006Available online 10 July 2006

Abstract

The dual digestion process was investigated using sludge samples collected at the WWTP of Tomaszow Mazowiecki (Poland). Mixedsludge was treated in a laboratory setup under batch and semi-continuous conditions. Dual digestion with a 1 d SRT aerobic thermo-philic pretreatment followed by an anaerobic step with 20 d of SRT turned out to be optimal, since a 44–46% VS reduction and a biogasyield of 480 dm3/kg VS fed were achieved. In the course of the process, the concentration of nitrogen in supernatant increased over 5times and its major portion was converted into ammonia. Phosphorus also entered the supernatant, reaching over 200 g/m3. The dualdigestion noticeably deteriorated the sludge dewaterability. Following completion of the process, capillary suction time measurementsaveraged 64 s for the raw sludge, 400 s for aerobically pretreated sludge and 310–360 for the anaerobically digested sludge. Aerobic pre-treatment consistently reduced Enterobacteriaceae content to below detectable limits.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Sludge digestion; Sludge stabilization; Dual digestion; Aerobic thermophilic pretreatment; Pathogen inactivation; Nutrient control; Sludgedewatering

1. Introduction

The principal objective of sewage sludge treatment is itsstabilization, that is a controlled decomposition of easilydegradable organic matter resulting in a significant reduc-tion of volatile solids content, a change of an unpleasantsmell into an earthy one, and an elimination of sludgeputrescibility. The most common methods of sludge stabil-ization are biological processes of anaerobic mesophilicdigestion and aerobic digestion at ambient conditions, bothof which are not without disadvantages. Anaerobic meso-philic digestion, due to relatively long SRT, high sensitivityand biogas production, needs large reaction volumes,gas collecting tanks and complex instrumentation (Kelly,1989). Aerobic biological stabilization at ambient condi-tions also needs large operation volumes as well as energyconsuming aeration devices and has traditionally beenundertaken for small communities (Kelly, 1989).

0960-8524/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.biortech.2006.05.017

* Corresponding author. Tel.: +48 42 631 34 84; fax: +48 42 636 59 76.E-mail address: sebasbor@poczta.onet.pl (S. Borowski).

An alternative system that overcomes the above men-tioned problems is autothermal thermophilic aerobic diges-tion (ATAD). This unique and relatively new process wasdeveloped by Hubert Fuchs, and has been operated success-fully in several countries including Germany, USA andCanada (Fuchs and Schwinning, 1997; Kelly, 1989; Kellyet al., 1993; Schwinning et al., 1997). ATAD represents abiological system that converts biodegradable organic mat-ter to lower-energy forms (mainly CO2 and H2O) throughaerobic and fermentative processes. Part of the energyobtained during the oxidation of organic matter is dissipatedinto sludge as heat, which, if conserved, will yield operationat thermophilic temperatures (Kelly and Warren, 1995).

A modification of the ATAD system, known as the dualdigestion process, utilizes an aerobic thermophilic pretreat-ment (ATP) prior to anaerobic digestion. In the first stageof this technology, sludge is ‘‘pretreated’’ through efficientsolubilization and partial digestion (acidification) of partic-ulate organic matter. Due to a very short retention time,usually 12–24 h, very little sludge stabilization takes placein the aerobic reactor – only to the degree that the heat gen-erated biologically maintains thermophilic temperatures.

Nomenclature

AD anaerobic digestionATAD autothermal thermophilic aerobic digestionATP aerobic thermophilic pretreatmentCOD chemical oxygen demand (gO2/kg TS)CST capillary suction time (s)HRT hydraulic retention time (d) or (h)MPN most probable numberN–NH4 ammonium nitrogen (gN/m3)PE population equivalentSRT solids retention time (d)

TAL total alkalinity (gCaCO3/m3)TC total carbon (gC/kg TS)TKN total Kjeldahl nitrogen (gN/kg TS) or (gN/m3)TOC total organic carbon (gC/kg TS)TP total phosphorus (gP/kg TS) or (gP/m3)TS total solids (g/kg)VFA volatile fatty acids (gCH3COOH/m3)VS volatile solids (g/kg)WAS waste activated sludge (excessive sludge)WWTP wastewater treatment plant

1200 S. Borowski, J.S. Szopa / Bioresource Technology 98 (2007) 1199–1207

However, the raw sludge should be externally preheated tothe thermophilic temperatures. The aerobic reactor is oper-ated under oxygen limiting conditions, which, in conjunc-tion with the short HRT, results in the formation ofsignificant concentrations of soluble products includingvolatile fatty acids, through the fermentative metabolismof thermophilic bacteria (Messenger et al., 1993; Masonet al., 1987; Haner et al., 1994; McIntosh and Oleszkiewicz,1997; Pitt and Ekama, 1996). In the second anaerobic meso-philic digestion step final and full stabilization takes place.By aerobic thermophilic pretreatment, the anaerobic dig-estion efficiency is enhanced when compared to a con-ventional single stage process, resulting in shorter SRT(usually 8–15 d), higher VS destruction, greater pH stabilitythrough the production of alkalinity, and complete patho-gen inactivation (production of Class A biosolids accordingto the EPA regulations) (McIntosh and Oleszkiewicz, 1997;Messenger et al., 1993; Pagilla et al., 1996, 2000; Ward et al.,1998).

This study was initiated to evaluate the dual digestion ofmunicipal mixed sewage sludge by the determination of theimpact of aerobic thermophilic pretreatment on the anaer-obic digestion performance in the second step (VS reduc-tion, SRT and biogas production). Of particular interestwas the determination of changes in nutrient concentra-tions in the sludge supernatant as well as changes in sludgedewaterability during digestion. Pathogen inactivation effi-ciency of the stabilization process was also evaluated.

2. Methods

2.1. Materials

Laboratory scale experiments were conducted usingmixed sludge, primary and waste activated in the averageproportion of 1:2 respectively. The sludge was taken fromthe Municipal Wastewater Treatment Plant at TomaszowMazowiecki, Poland, serving a population equivalent ofabout 70,000 inhabitants. This plant was operated with aconventional activated sludge process for BOD removalonly (without enhanced biological nutrient removal).Before the experiments the sludge was pre-thickened to

approximately 5% TS. The elemental composition of rawsludge is shown in Table 1.

2.2. Experiments

The laboratory scale experiments were conducted usinga 12 dm3 aerobic thermophilic reactor with a working vol-ume of 7 dm3, followed by a 10 dm3 anaerobic mesophilicdigester (working volume of 7 dm3). The control systemconsisted of a 10 dm3 anaerobic mesophilic digester (work-ing volume of 7 dm3). The aerobic thermophilic reactorwas operated at 55 ± 2 �C whereas the anaerobic digesterswere operated at 35 ± 1 �C. The units were placed insideconstant temperature chambers to maintain consistent tem-peratures. The aerobic reactor was supplied with air at aflow rate of 40 dm3/dm3 h by mean of a Secoh Air Pumplaboratory blower. The reactor and digesters were operatedwith stirring and with no recycle, so that the system SRTand HRT were equal.

In batch experiments the ATP reactor was filled with theheated raw sludge and aerated under thermophilic condi-tions for 12, 24, 36 and 48 h depending on the process ser-ies. The aerobically pretreated sludge was then fed to theanaerobic digester where the mesophilic digestion tookplace, without addition of seed material. The anaerobicdigestion was continued to the point when only a residualbiogas production was found (no more than 20 cm3/dm3

digester active volume per day). The control anaerobic pro-cess was operated in the same manner with the exceptionthat sludge was not aerobically pretreated (HRT of aerobicstep was 0 h).

In semi-batch experiments the ATP reactor was fed anequal volume of heated raw sludge every 12 h, with effluentsludge removal before feeding. The operating SRT of theaerobic reactor was determined in batch experiments.Two anaerobic digesters were batch fed constant portionsof aerobically pretreated sludge (depending on a selectedSRT for each digester) every 24 h, also with effluent sludgeremoval before feeding. Digesters were operated at SRTsof 20, 30 and 40 d. To ensure steady state operation digest-ers were operated for a minimum of three SRTs at eachretention time.

Table 1Characterization of raw sludges used for the investigations

Parameter Unit Range Average Standard deviationa

pH – 6.10–6.61 6.34 0.20TS % 4.67–6.06 5.42 1.04VS % 3.84–4.85 4.41 0.84

% of TS 80.07–83.77 81.41 0.92COD g/kg TS 896–1184 1018 130TC gC/kg TS 449.4–488.3 464.1 16.6TOC gC/kg TS 444.1–468.7 445.8 14.5TKN gN/kg TS 70.27–88.68 81.66 6.53TP gP/kg TS 10.62–15.84 14.03 1.95Potassium gK/kg TS 2.36–3.03 2.72 0.26Sodium gNa/kg TS 2.27–2.49 2.35 0.11Calcium gCa/kg TS 4.48–19.49 11.80 4.25Magnesium gMg/kg TS 1.87–3.16 2.67 0.46Iron gFe/kg TS 6.67–8.26 7.36 1.40Lead mgPb/kg TS 139.17–160.46 152.82 38.11Zinc mgZn/kg TS 1385–1558 1482 208Cadmium mgCd/kg TS 10.42–13.86 12.04 1.57Copper mgCu/kg TS 111.64–165.47 142.55 35.50CST s 55–70 64 4Salmonella sp. In 25 g Present – –Enterobacteriaceae MPN/g (8–22) · 105 17 · 105 7Coliforms MPN/g (4–9) · 105 7 · 105 2Escherichia coli MPN/g (2–3) · 105 2.4 · 105 0.6

Sludge Supernatant

TAL gCaCO3/m3 120–1500 1060 200VFA gCH3COOH/m3 550–1200 810 253N–NH4 gN/m3 290–313 299 9

% TKN 71.5–80.6 72.6TKN gN/m3 360–426 411 29TP gP/m3 98–104 102 3

a No. of replicates = 10.

S. Borowski, J.S. Szopa / Bioresource Technology 98 (2007) 1199–1207 1201

2.3. Analysis

The product sludge samples collected from both aerobicreactor and anaerobic digester were analyzed for total sol-ids (TS), volatile solids (VS), COD, capillary suction time(CST), supernatant ammonia nitrogen (N–NH4), volatilefatty acids (VFA), total alkalinity concentration (TAL)and pH level. The raw and anaerobically digested sludgeswere additionally analyzed for total and organic carbon(TC and TOC), total Kjeldahl nitrogen (TKN), total phos-phorus (TP) and supernatant TKN and TP. In raw andanaerobically digested sludge samples (in semi-batch exper-iments) the concentrations of metals, Enterobacteriaceaebacteria density and the presence of Salmonella sp. werealso determined.

The concentrations of metals (K, Na, Ca, Mg, Fe, Cu,Zn, Pb and Cd) were determined using an atomic absorp-tion spectrophotometer (SOLAAR 969 Unicam). A COU-LOMAT 702, Strohlein Ltd. detector was used to measureTC and TOC content. Biogas production from the anaero-bic mesophilic digester was measured by the water displace-ment method. Methane content of the sludge biogas wasanalyzed with a Shimadzu GC-14B gas chromatograph fit-ted with a chromosorb 102 packed column (2 m · 4 mmand mesh size 60–80) and a thermal conductivity detector

(TCD). Column temperature was kept at 50 �C while thosefor injector and detector were set at 100 �C. Helium wasused as a carrier gas, at a flow rate of 40 cm3/min.

Capillary suction time was measured on a capillary suc-tion apparatus as described by Chen et al. (1996). Viablecell counts of Enterobacteriaceae, coliforms and Esche-

richia coli were conducted using PETRIFILM plates(Trafalgar Scientific Ltd., UK): MB 03-type for the enu-meration of E. coli and coliforms, and MB 05-type forthe enumeration of Enterobacteriaceae. The plates wereincubated at a temperature of 37 �C in the traditional man-ner. The presence of Salmonella sp. was detected usingTecra Unique Salmonella immune enzymatic test afterpre-enrichment of 25 g sludge samples. The other para-meters (pH, TS, VS, COD, TAL, VFA, N–NH4, TKNand TP) were measured according to Standard Methods(APHA, 1995).

3. Results and discussion

3.1. Digestion performance

To evaluate how an aerobic thermophilic pretreatmentprocess affects anaerobic mesophilic digestion, batch exper-iments were undertaken. Size and operating parameters

1202 S. Borowski, J.S. Szopa / Bioresource Technology 98 (2007) 1199–1207

and results of these experiments are summarized in Tables2 and 3 and Fig. 1. The dual digestion batch process withSRT = 24 h of ATP was found to be the most effective interms of VS reduction, gas production and time of anaero-bic digestion. In such conditions the SRT of anaerobicdigestion was 42 h, almost 30% shorter than that requiredin control anaerobic digestion. In spite of shorter SRT,the VS reduction (of about 46%) was slightly greater by2% points for dual digestion with 24 h of aerobic step,while the biogas production of 414 dm3/kg VS fed and1202 dm3/kg VS removed turned out to be the highest(6% greater than in the control process). High biogas yieldin the dual process correlated with the maximum produc-tion of volatile fatty acids in the aerobic thermophilic reac-tor (net VFA increase 3818 g/m3). Simultaneously, only apartial destruction of organic matter took place in the aer-obic step (15% VS reduction) and the principal degradationof organic matter proceeded in the anaerobic digester. Ashorter SRT of 12 h in the aerobic reactor also showedthe limitation of the time of anaerobic digestion. However,the biogas production and VS reduction were lower than

Table 2Size and operating parameters of laboratory scale treatment systems – batch e

Parameter Unit Control AD Dual digwith 12

ATP

Duration time d 60 0.5Feed TS content % 4.86 5.60Feed VS content % of TS 83.77 81.44net VFA increase gCH3COOH/m3 160 3016net N–NH4 production gN/m3 1595 1103Biogas yield dm3/kg VSfed 416 –

dm3/kg VSreduced 940 –VS reduction % 44.22 7.34VS reduction in two steps % – 43.

Table 3Characteristics of sludges after the dual digestion batch experiments

Parameter Unit Control AD Dua

12 h

pH – 8.10 8.40TS % 3.02 3.57VS % of TS 75.2 71.6COD g/kg TS 1056 1028TC gC/kg TS 459.7 468.TOC gC/kg TS 403.8 381.TKN gN/kg TS 137.7 134.TP gP/kg TS 24.17 23.7CST s 400 350

Sludge Supe

TAL gCaCO3/m3 7200 8600VFA g CH3COOH/m3 1160 960N–NH4 gN/m3 1885 2109

% TKN 87.3 71.9TKN gN/m3 2160 2935TP gP/m3 194 210

those achieved in the dual digestion with a 24 h aerobicSRT, as well as in the control anaerobic process. Aerobicpretreatment SRT greater than 24 h resulted in a shorten-ing of the time of anaerobic digestion and an increasein VS degradation, but lower biogas yields were alsoachieved. At the 48 h aerobic SRT, the overall reductionof VS was greater than 50% but the biogas production of320 dm3/kg VS fed was considerably lower than in theother batch experiments.

As mentioned above, the highest biogas yield wasaccompanied by the maximum VFA production in the aer-obic step (around 4000 g/m3). The aerobic thermophilicreactor also contributed very little to sludge stability byway of VS removal. At thermophilic temperatures greaterthan 55 �C, oxygen solubility is so low that its concentra-tion in an aerobic tank is close to zero, and both oxidativeand fermentative processes take place there. In the aerobicthermophilic reactor, the main reactions are solubilizationof organic polymers and formation of fermentation prod-ucts such as volatile fatty acids. However, these productsare not oxidized further due to a relatively short SRT

xperiments

estionh ATP

Dual digestionwith 24 h ATP

Dual digestionwith 36 h ATP

Dual digestion with48 h ATP

AD ATP AD ATP AD ATP AD

39 1 42 1.5 40 2 355.02 5.29 4.66 5.85 4.76 5.78 4.3980.26 80.11 77.66 81.43 80.13 82.84 81.29410 3818 122 3362 650 2775 2391812 1430 1847 1383 1887 1107 1847369 – 441 – 378 – 3201010 – 1202 – 1134 – 94737.87 14.67 36.7 19.93 33.31 25.48 33.83

96 45.98 46.60 50.69

l digestion

ATP 24 h ATP 36 h ATP 48 h ATP

8.05 7.98 8.063.34 3.54 3.21

4 68.55 71.87 73.571166 963 961

1 436.8 455.8 476.41 378.2 386.3 411.35 125.4 128.2 140.55 23.94 21.18 21.9

380 400 360

rnatant

6400 7920 8000672 1200 8642144 2184 214074.1 76.1 74.22895 2870 2885215 200 185

0

200

400

600

800

1000

1200

0

duration time [d]

Bio

gas

prod

uctio

n [d

m3 /

m3 ·

d]

control anaerobic digestiondual digestion with 12 h ATPdual digestion with 24 h ATPdual digestion with 36 h ATPdual digestion with 48 h ATP

20 30 40 50 6010

Fig. 1. Daily biogas yield from anaerobic digesters – batch experiments.

S. Borowski, J.S. Szopa / Bioresource Technology 98 (2007) 1199–1207 1203

(Mason et al., 1987; Haner et al., 1994; McIntosh and Oles-zkiewicz, 1997). Experience on a full scale ATAD processin Salmon Arm (Kelly, 1989) showed that in the first stagedigester the production of volatile fatty acids can be as highas 10,000 g/m3. It was also reported that acetate consti-tuted 70–90% of the total concentration of VFA in aerobicthermophilic pretreatment plants (Chu et al., 1994; Masonet al., 1987; Mavinic et al., 2001; Fothergill and Mavinic,2000). Since acetate serves as preferred substrate for meth-anogenic bacteria, its high concentration in the aerobicallypretreated sludge is one of the keys for enhanced degrada-tion efficiency of the anaerobic mesophilic step.

To establish the optimal conditions for anaerobic diges-tion in the dual process, semi-batch experiments were per-formed. Size and operating parameters and results of theseexperiments are shown in Tables 4–6 and Fig. 2. The firsttwo runs (at retention times of 41 and 31 d) started withfilling the digesters with the aerobically treated sludge.During the first month of these runs a growth and acclima-tion of biocenosis (responsible for anaerobic decomposi-tion) proceeded in the digesters. Considerable changes in

Table 4Size and operating parameters of laboratory scale treatment systems – semi-b

Parameter Unit Total SRT =

ATP

Temperature �C 55 ± 2SRT d 1Organic sludge loading kg VS/m3 d 38.4–48.5Feed TS content % 4.67–6.06Feed VS content % 3.84–4.85Average biogas yield dm3/kg VS fed –

dm3/kg VS reduced –Average VS reduction % 13.31

Average VS reduction in two steps % 47.4

the biogas productions and VFA concentrations were evi-dence of this. After this period, the concentrations of vola-tile fatty acids and total alkalinity were relatively constantshowing the VFA/TAL indicator not exceeding 0.16 (thiswas much lower than the acceptable value of 0.25). Thethird run with the SRT = 21 d was performed using theanaerobic digester from the first run (after its completion)so no larger changes of daily biogas yield, VFA and TALconcentrations were observed.

The dual process with total SRT = 21 d was found to bethe most effective one. The average gas production fromthe digester was 480 dm3/kg VS fed and 1136 dm3/kg VSremoved, with an average total VS reduction (in two steps)of more than 44%. This is greater than the minimum 38%VS reduction required by the 40 CFR 503 (EPA, 1999).The VS reduction at 41 d SRT was greater by around 5%points; however, the biogas yield from the digester showedonly 373 dm3/kg VS fed and 1038 dm3/kg VS reduced. Ata total SRT of 31 d the VS degradation rate was the lowestof all the runs, but the biogas production from 1 kg VSremoved turned out to be the highest one (average

atch experiments

41 d Total SRT = 31 d Total SRT = 21 d

AD ATP AD ATP AD

35 ± 1 55 ± 2 35 ± 1 55 ± 2 35 ± 140 1 30 1 200.76–1.12 38.4–48.5 1.02–1.50 38.4–48.5 1.53–2.253.81–5.75 4.67–6.06 3.81–5.75 4.67–6.06 3.81–5.753.06–4.49 3.84–4.85 3.06–4.49 3.84–4.85 3.06–4.49373 – 419 – 4801038 – 1307 – 113639.34 13.31 33.85 13.31 35.61

1 42.65 44.18

Table 6Sludge characteristics after the dual digestion–semi-batch experiments

Parameter Unit Total SRT = 41 d Total SRT = 31 d Total SRT = 21 d

Range Average Range Average Range Average

pH – 7.62–8.18 7.88 7.48–8.01 7.78 7.63–7.88 7.75TS % 2.86–3.32 3.12 3.19–3.58 3.41 3.27–3.41 3.37VS % TS 72.04–74.11 73.0 71.84–73.54 72.79 70.99–72.55 71.69COD g/kgTS 1081–1307 1132 999–1140 1105 1081–1211 1169TC gC/kg TS 432.4–498.8 461.7 399.6–457.4 441.6 410.8–484.4 456.9TOC gC/kg TS 380.6–453.9 411.7 359.6–420.8 402.4 369.7–448.1 417.2TKN gN/kg TS 128.5–135.2 132.9 122.2–136.0 128.8 131.7–140.9 135.0TP gP/kg TS 19.93–28.64 22.95 19.11–26.62 21.84 20.31–21.06 20.54CST s 300–350 310 290–480 360 310–360 335

Sludge Supernatant

TAL g/m3 6800–9200 7800 6600–9000 7680 7400–8400 8175VFA g/m3 528–1300 903 720–1200 883 624–1104 807VFA/TAL – 0.06 – 0.16 0.11 0.09–0.16 0.11 0.08–0.13 0.10N–NH4 gN/m3 1840–2427 2050 1736–2342 2060 1913–2380 2198

% TKN 73.2–76.3 74.5 64.4–80.2 73.6 69.6–85.0 78.8TKN gN/m3 2453–3181 2750 2697–2923 2798 2750–2800 2788TP gP/m3 188–226 207 189–260 212 173–192 181

Table 5Sludge characteristics after the aerobic thermophilic pretreatment – semi-batch experiments

Parameter Unit Range Average

pH – 6.05–7.04 6.52TS % 3.81–5.75 4.72VS % 3.06–4.49 3.75

% of TS 78.67–80.76 79.55COD g/kg TS 1180–1444 1300CST s 390–405 400

Sludge Supernatant

TAL g/m3 3800–5500 4400VFA g/m3 2880–4224 3632N–NH4 gN/m3 1098–1885 1343

1204 S. Borowski, J.S. Szopa / Bioresource Technology 98 (2007) 1199–1207

1307 dm3). The biogas produced in anaerobic digesters con-tained relatively large amounts of methane, from 70% to77%, irrespective of the SRT. It should also be noted thatfurther shortening of anaerobic digestion SRT (under

0

100

200

300

400

500

600

700

0 20 40 60 80 100 120 140 160

duration time [d]

Bio

gas

yiel

d[d

m3 /k

g V

S fe

d]

SRT = 41 d

0 20

Fig. 2. Daily biogas yield from anaerobi

20 d) resulted in process breakdown as the gas productiondeclined and VFA increased sharply (data not shown).

The minimal SRT of the anaerobic step seems thereforeto have been relatively long compared to other research.

40 60 80 100 120

duration time [d]

SRT = 31 d

0 20 40 60 80

duration time [d]

SRT = 21d

c digesters – semi-batch experiments.

S. Borowski, J.S. Szopa / Bioresource Technology 98 (2007) 1199–1207 1205

This was probably due to the specific nature of the sludge,which contained around 70% waste activated sludge rela-tively resistant to biological decomposition. Andersonet al. (1995) and Nellenschulte and Kayser (1997) maintainthat the degradation rate of organic solids mass is higher inprimary sludges than in waste activated sludges (WAS). Insimplified terms, the microorganisms of WAS consist ofcytoplasm, which is surrounded by a cell membrane resis-tant to biodegradation. Despite a rather long SRT of theanaerobic digestion step, the biogas yield and VS reductionseemed to be relatively high. However, when compared tothe control anaerobic digestion the biogas production wasonly 6% greater and VS reduction slightly higher by 2%points for dual digestion with the optimal total SRT of21 d. This was consistent with the observations of Wardet al. (1998). They reported a slightly higher productionof gas by 1–2% points in dual digestion systems than in acontrol digester, which corresponded to a slightly higheroverall VS removal efficiency (60% and 57% respectively).An increase in biogas production and VS destruction afterATP application has also been reported by Pagilla et al.(1996, 2000).

3.2. Changes of nutrient content

In all experiments nitrogen and phosphorus werereleased to the supernatant as well as being accumulatedin sludge dry matter (Tables 2, 3, and 6). The concentrationof nitrogen in sludge supernatant increased by over 5 timesduring digestion in both aerobic and anaerobic steps. Themaximum nitrogen content of 2870–2935 gN/m3 was foundin digested sludge supernatant after batch experiments. Amajor portion of nitrogen was in the ammonium form,and its concentration ranged from 2109–2184 gN/m3 inbatch experiments, and from an average 2050–2198 gN/m3 in semi-batch runs. This corresponds to 72–76% TKNand 74–79% TKN respectively. It is interesting that forcontrol anaerobic digestion the percentage of ammoniumnitrogen (87% TKN) was greater than for the two-step pro-cesses. This was probably attributed to loss of nitrogenthrough its stripping in the aerobic step (Mason et al.,1987). A substantial ammonium release was observed dur-ing aerobic thermophilic pretreatment. In all experimentsthe net ammonium production ranged from 1103 to1430 gN/m3 during the aerobic step to give ammoniumconcentrations of 1098–1885 gN/m3 in aerobically pre-treated sludge supernatant. Phosphorus (like nitrogen)entered the sludge supernatant, consistently reaching aconcentration over 200 gP/m3. However no correlationbetween the phosphorus release level and applied SRTwas shown.

Both nutrients were also accumulated in the sludge drymatter during the experimental processes, consistentlyincreasing the sludge manurial value. The concentrationof nitrogen in the sludge mass increased by around 50–70% to give average concentrations after the treatmentsof 125.4–140.5 gN/kg TS in batch and 128.8–135.0 gN/

kg TS in semi-batch experiments. No correlation betweennitrogen accumulation rate and SRT of the process wasfound. A similar accumulation rate of phosphorus was alsoobserved. The content of this nutrient in the sludge massincreased during digestion to 21.18–23.94 gP/kg TS inbatch and 20.54–22.95 gP/kg TS in semi-batch experi-ments. In contrast to nitrogen, longer SRT favored phos-phorus accumulation in the sludge dry matter. No impactof aerobic thermophilic digestion on nutrient accumulationwas found.

The concentrations of nitrogen and phosphorus in sludgesupernatants measured in this study were relatively highwhen compared to other investigations. In South Africantwo-step digestion installations, concentrations of ammo-nium nitrogen in anaerobic sludge supernatant averaged650–990 g/m3 (Messenger et al., 1993; Pitt and Ekama,1996). Wedi and Konig (1993) examined anaerobic sludgesupernatant, which had ammonia concentrations of 900–1000 gNH4/m3, whereas Sung and Santha (2003) reportedeven greater values of TKN and ammonium nitrogen,which averaged 3840 gN/m3 and 2330 gN/m3 respectively.The concentrations of phosphorus in supernatants of anaer-obic digesters depend on the sludge origin, and can evenexceed 300 g/m3 in EBPR plants (Doyle and Parson,2002; Popel and Jardin, 1993). Large amounts of nutrientsin sludge supernatant after digestion can be explained byintensive ammonification processes and almost completesolubilization of stored phosphorus because of the elevatedtemperatures of the stabilization processes (mesophilic orthermophilic) (Jardin and Popel, 1996; Mavinic et al.,2001; Popel and Jardin, 1993; Fothergill and Mavinic,2000).

3.3. Changes of sludge dewaterability

Dewaterability of sludge, as measured by CST, appearedto change considerably through the process, and showed ageneral deterioration by a factor of 6 times following diges-tion, irrespective of the SRT. The most visible change insludge dewaterability was observed in the first step of theprocess, when the CST indicator exceeded 400 s (Table5). The anaerobic digestion resulted in some improvementin dewaterability, as the CST indicator decreased to anaverage 335 s after the digestion, which therefore indicateda poorly dewaterable sludge. However, the CST valueswere no worse than that for the control anaerobic digestion(400 s). Furthermore, no correlation was found betweenCST values and SRTs of either aerobic or anaerobic steps.

To explain a deterioration of the sludge during diges-tion, Nellenschulte and Kayser (1997) introduced particlesize as the new parameter describing the dewatering behav-iour of sludge. During digestion, both aerobic and anaero-bic, the content of fine particles increases, which results inan increase of CST value. Deterioration of sludge dewater-ability is greater in the case of aerobic processes because ofmuch higher bacterial growth following decay as well asmechanical stress, which leads to greater disintegration

1206 S. Borowski, J.S. Szopa / Bioresource Technology 98 (2007) 1199–1207

and formation of fine particles. On the contrary, the dis-integration of sludge during anaerobic digestion can beexplained mostly by the degradation of exopolymersresponsible for floc formation.

3.4. Pathogen inactivation

According to the Polish regulations Salmonella and eggsof helminthes are indicator pathogens in municipal sludge,and their absence in the sludge is considered to be sufficientfor its agricultural reuse. In the following experiments,apart from the Salmonella, total viable counts of coliforms,E. coli and Enterobacteriaceae-family were also investi-gated (Table 1). No pathogen bacteria were detected insamples of digested sludge. This however does not indicateunambiguously that treated sludge can be reused asfertilizer, because eggs of helminthes are generally moreresistant to environmental conditions including the temper-ature and the time of its exposure. The effect of aerobicthermophilic pretreatment on pathogen density has beeninvestigated by many scientists (Baier and Zwiefelhofer,1991; Cheunbarn and Pagilla, 1999; Jepsen et al., 1997;Pagilla et al., 1996, 2000; Ward et al., 1998). They all con-firmed that aerobic–anaerobic systems produced Class Abiosolids that could be reused without any restrictions.

4. Conclusions

Based on the results obtained from the laboratory scaleexperiments to investigate aerobic thermophilic-anaerobicmesophilic dual digestion treatment of municipal sludge,the following conclusions can be made:

1. The aerobic thermophilic pretreatment with optimalSRT of 24 h resulted in a shortening of the anaerobicdigestion retention time by around 30% compared tocontrol anaerobic digestion.

2. The optimal SRTs for aerobic and anaerobic stepsturned out to be 24 h and 20 d respectively. Under suchconditions, an overall VS reduction of 44.18% and a bio-gas yield of 480 dm3/kg VS fed were obtained. However,the gas production was only 6% greater and VS reduc-tion slightly higher by 2% points for dual digestion whencompared to the control anaerobic process.

3. The dual digestion process consistently produced nutri-ent-rich supernatant and biosolids. The concentration ofnitrogen in sludge supernatant increased by over 5 timesduring the process to give almost 3000 gN/m3 afterdigestion, and a major portion of nitrogen was con-verted into ammonia. Phosphorus also entered thesupernatant, reaching a concentration in excess of200 g/m3. Both nutrients were also accumulated in thesludge dry matter to increase its manurial value.

4. In all experiments, the sludge dewaterability deterio-rated by a factor of 6 times following digestion, irrespec-tive of the SRT. The CST value exceeded around 400 sin the ATP step and only decreased slightly to average

335 s after the digestion. No impact of aerobic thermo-philic pretreatment on sludge dewaterability wasobserved.

5. The Enterobacteriaceae density was reduced to belowdetectable limits in all experiments.

Acknowledgements

The authors would like to thank the staff of the Munici-pal Wastewater Treatment Plant at Tomaszow Mazowieckifor preparation and supplying sewage sludge.

References

Anderson, B.C., Mavinic, D.S., Oleszkiewicz, J.A., 1995. Stabilization ofcombined wastewater sludge: anaerobic processes. Canadian Journalof Civil Engineering 22 (2), 223–234.

APHA, 1995. Standard Methods for the Examination of Water andWastewater, 19th ed. American Public Health Association, Washing-ton, DC, USA.

Baier, U., Zwiefelhofer, H.P., 1991. Effects of aerobic thermophilicpretreatment. Water Science and Technology 23, 56.

Chen, G.W., Lin, W.W., Lee, D.J., 1996. Capillary suction time as ameasure of sludge dewaterability. Water Science and Technology 34(3–4), 443–448.

Cheunbarn, T., Pagilla, K.R., 1999. Temperature and SRT effects onaerobic thermophilic sludge treatment. Journal of EnvironmentalEngineering 125 (7), 626–629.

Chu, A., Mavinic, D.S., Kelly, H.G., Ramey, W.D., 1994. Volatile fattyacids production in thermophilic aerobic digestion of sludge. WaterResearch 28 (7), 1513–1522.

Doyle, J.D., Parson, S.A., 2002. Struvite formation control and recovery.Water Research 36, 3925–3940.

EPA, 1999. Control of pathogens and vector attraction in sewage sludge.EPA/625/R-92/013, US EPA, Office of Research and Development,Washington, DC 20460, USA. Available from: <http://www.epa.gov>.

Fothergill, S., Mavinic, PE., 2000. VFA production in thermophilicaerobic digestion of municipal sludges. Journal of EnvironmentalEngineering 126 (5), 389–396.

Fuchs, L., Schwinning, H.G., 1997. Zum stand der aerob-thermophilenstabilisierung und entseuchung von Klarschlamm. KorrespondenzAbwasser 44 (10), 1834–1842.

Haner, A., Mason, A.C., Hamer, G., 1994. Aerobic thermophilic wastesludge biotreatment: carboxylic acid production and utilization duringbiodegradation of bacterial cells under oxygen limitation. AppliedMicrobiology and Biotechnology 40 (6), 904–909.

Jardin, N., Popel, H.J., 1996. Behavior of waste activated sludge fromenhanced biological phosphorus removal during sludge treatment.Water Environmental Research 68 (6), 965–973.

Jepsen, S.E., Krause, M., Gruttner, H., 1997. Reduction of fecalStreptococcus and Salmonella by selected treatment methods forsludge and organic waste. Water Science and Technology 36 (11),203–210.

Kelly, H., 1989. Aerobic thermophilic digestion of liquid composting ofmunicipal sludges. In: Environmental Engineering Proceedings ofSpecial Conference/EE Div./ASCE, Austin, TX, USA, pp. 650–662.

Kelly, H., Warren, R., 1995. What’s in a name? – Flexibility. WaterEnvironmental Technology 7 (7), 46–50.

Kelly, H.G., Melcer, H., Mavinic, D.S., 1993. Autothermal thermophilicaerobic digestion of municipal sludges: a one-year, full-scale demon-stration project. Water Environmental Research 65 (4), 849–861.

Mason, A.C., Hamer, G., Fleischmann, T., Lang, C., 1987. Aerobicthermophilic biodegradation of microbial cells – some effects of

S. Borowski, J.S. Szopa / Bioresource Technology 98 (2007) 1199–1207 1207

dissolved oxygen and temperature. Applied Microbiology and Biotech-nology 25 (6), 568–576.

Mavinic, D.S., Mahendraker, V., Sharma, A., Kelly, H.G., 2001. Effect ofmicroaerophilic conditions on autothermal thermophilic aerobic diges-tion process. Journal of Environmental Engineering 127 (4), 311–316.

McIntosh, K.B., Oleszkiewicz, J.A., 1997. Volatile fatty acid production inaerobic thermophilic pre-treatment of primary sludge. Water Scienceand Technology 36 (11), 189–196.

Messenger, J.R., de Villiers, H.A., Laubscher, S.J.A., Kenmuir, K.,Ekama, G.A., 1993. Evaluation of the dual digestion system: Part 1:Overview of the Milnerton experience. Water SA 19 (3), 185–191.

Nellenschulte, T., Kayser, R., 1997. Change of particle structure of sewagesludges during mechanical and biological processes with regard to thedewatering result. Water Science and Technology 36 (4), 293–306.

Pagilla, K.R., Craney, K.C., Kido, W.H., 1996. Aerobic thermophilicpretreatment of mixed sludge for pathogen reduction and Nocardia

control. Water Environmental Research 68 (7), 1093–1098.Pagilla, K., Kim, H., Cheunbarn, T., 2000. Aerobic thermophilic and

anaerobic mesophilic treatment of swine waste. Water Research 34(10), 2747–2753.

Pitt, A.J., Ekama, G.A., 1996. Dual digestion of sewage sludge using airand pure oxygen. In: Proceedings of the 69th Water EnvironmentFederation Annual Conference and Exposition, Dallas, TX, USA, vol.2, pp. 69–82.

Popel, H.J., Jardin, N., 1993. Influence of enhanced biological phosphorusremoval on sludge treatment. Water Science Technology 28 (1), 263–271.

Schwinning, H.G., Deeny, K.J., Hong, Sun-Nan, 1997. Experiencewith autothermal thermophilic aerobic digestion (ATAD) in theUnited States. In: Proceedings of the 70th Water EnvironmentFederation Annual Conference and Exposition, Chicago, IL, USA,pp. 275–285.

Sung, S., Santha, H., 2003. Performance of temperature-phased anaerobicdigestion (TPAD) system treating dairy cattle wastes. Water Research37, 1628–1636.

Ward, A., Stensel, D.H., Ferguson, J.F., 1998. Effect of autothermaltreatment on anaerobic digestion in the dual digestion process. WaterScience Technology 38 (8–9), 435–442.

Wedi, D., Konig, E., 1993. Elimination of nitrogen and phosphorus fromsludge liquor. Water Science Technology 28 (1), 283–287.