Bioresource Technology 142 (2013) 69–76

Contents lists available at SciVerse ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Anaerobic digestion of activated sludge after pressure-assistedozonation

0960-8524/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.biortech.2013.04.058

⇑ Corresponding author. Tel.: +1 801 581 7232; fax: +1 801 585 5477.E-mail address: hong@civil.utah.edu (P.K.A. Hong).

Chia-Jung Cheng, P.K. Andy Hong ⇑Department of Civil and Environmental Engineering, University of Utah, Salt Lake City, UT 84112, USA

h i g h l i g h t s

� Ozonation of activated sludge (AS) improved solids reduction and biogas yield during digestion.� Ozonation with 10 mg O3 g�1 TSS via pressure cycles improved COD solubilization of AS to 18%.� Ozonation via pressure cycles improved VSS reduction by 1.6 folds and biogas yield by 8 folds.

a r t i c l e i n f o

Article history:Received 14 February 2013Received in revised form 15 April 2013Accepted 16 April 2013Available online 2 May 2013

Keywords:Activated sludgeOzonationAnaerobic digestionBiogasSolids

a b s t r a c t

This study was undertaken to examine the benefits of pressure-assisted ozonation (PAO) in enhancingsolids reduction and biogas production in anaerobic digestion. The results showed significant improve-ments in both. With unacclimated inoculum at varied food-to-inoculum (F/I) ratios of 0.5–2, solids andCOD reductions were improved by PAO, as well as by increased F/I ratios even without PAO, which wouldwarrant further optimization of the F/I ratio for an unacclimated inoculum. With acclimated inocula at F/Iratio of 0.8, volatile suspended solids reduction and biogas production were improved by up to 60% and800%, respectively, when the AS had been subjected to 20 cycles of PAO. In extended operation in plantswhere acclimated anaerobes are encountered, PAO pretreatment offers improved digestion of AS in termsof solids and COD removals and biogas production.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Excess wasted activated sludge (AS) is a significant issue world-wide prompting numerous recent studies aiming to alleviate theconsequential environmental burden. Ozonation has shown prom-ising results for improving anaerobic digestibility of AS from benchto full-scale wastewater treatment plants (Bougrier et al., 2007;Carrere et al., 2010; Elliott and Mahmood, 2007; Weemaes et al.,2000b; Yasui et al., 2005; Yeom et al., 2002). Cell solubilizationas manifested in flocs disintegration, solids reduction, and solubleorganics increases is generally achieved when 50 or more mg O3 -g�1 dry solids is applied to AS (Park et al., 2003; Zhang et al., 2009).Studies have shown increased anaerobic biogas production withincreased sludge solubilization by increased ozone dosage (Bougri-er et al., 2006; Carrere et al., 2010; Chu et al., 2009b; Weemaeset al., 2000a). For example, a 155% increase in biogas productionwas observed with a dose of 150 mg O3 g�1 solids, and only 31%increase with 15 mg O3 g�1 solids (Bougrier et al., 2007). However,life cycle assessment of anaerobic digestion has shown that

ozonation may not be beneficial for solubilization of biologicalwastes (e.g., kitchen wastes and sewage sludge), meaning thatthe expended energy may result in a greater environmental burdenthan benefit (Carballa et al., 2011). However, few studies examinedlow dosage for sludge ozonation or demonstrated good solubiliza-tion efficiencies at low dosage (Chu et al., 2009a).

Pressure gradient (DP > 30 bar) was also tried for sludge treat-ment to enhance flocs disintegration, cell rupture, leading to in-creased volatile solids (VS) removal and biogas production(Carrere et al., 2010; Elliott and Mahmood, 2007; Park and Clark,2002; Rai and Rao, 2009). VS removal was increased from 35% to50% when the AS was treated by the plate collision method(DP = 30 bar); biogas production was increased by 18% when theAS was treated by a homogenizer at 600 bars of pressure gradient(Barjenbruch and Kopplow, 2003; Carrere et al., 2010; Choi et al.,1997). Solubilization efficiency was more effective with high pres-sure gradient than with low ones. With pressure gradient of 10 barwith CO2, COD solubilization was merely 5% with little improve-ment in VS removal (Ma et al., 2011). Pressure-assisted ozonation(PAO) was used for disrupting aggregated materials, including AS,sediment, and soil (Cheng et al., 2012; Hong and Nakra, 2009;Hong et al., 2008). A previous study showed when filamentous

70 C.-J. Cheng, P.K.A. Hong / Bioresource Technology 142 (2013) 69–76

AS flocs were subjected to PAO, they became disrupted showingviscous patches with little discernible AS remaining (Cheng et al.,2012). PAO was effective that it delivered dissolved air and ozoneinto cells under hyperbaric pressure which upon pressure releaseproduced expanding gas bubbles from within the cells leading tothe cells’ rupture and release of intracellular materials (Chenget al., 2012). Relative to ozonation with and without the use ofmicrobubble, PAO demonstrated advantages in COD and solids sol-ubilization with low ozone dose and short contact time (Chenget al., 2012; Chu et al., 2008, 2009a; Dogruel et al., 2007; Yeomet al., 2002). The advantages were attributed to microbubblesbeing created ubiquitously throughout the reactor volume andnot dependent on the location of the gas injector (Chu et al.,2008, 2009a; Cha et al., 2010).

In assessing biogas production and biodegradability, feed-to-inoculum ratio (F/I) is an important factor (Braguglia et al., 2006;Jensen et al., 2011; Tomei et al., 2008). In some cases, the F/I ratiobecame a critical factor (Neves et al., 2004). For example, solidsreduction and sludge hydrolysis were increased in sonicated ASas well as in untreated AS when the F/I ratio was increased from0.1 to 4 (Tomei et al., 2008). Although ozonation pretreatment ofAS had been extensively studied, results are not available on batchanaerobic digestion or on its effectiveness as influenced by F/I ofthe fed AS having been subjected to low ozone dose via PAO(<50 mg O3 g�1 dry solids). While the previous study showedadvantages of PAO in AS solubilization, the present study hasinvestigated the biodegradability of PAO-treated sludge by batchanaerobic digestion and the kinetics of solid reduction and biogasproduction of the PAO-treated sludge at different F/I ratios. An ulti-mate goal is to reduce excess sludge volume and enhance biogasproduction during conventional anaerobic digestion.

2. Methods

2.1. Sludge and inoculum sources

Weekly samples of returned AS were taken from Central ValleyWater Reclamation Facility (CVWRF), Salt Lake City, Utah. The ASsamples were kept in a 4-L bottle in the refrigerator (4 �C) untiluse in 48 h. Two inocula were used for the study, the fresh inocu-lum from CVWRF (FI) and the laboratory incubated inoculum (LI).The original inoculum was obtained at the anaerobic digester ofthe CVWRF and used on the same day for the digestion tests asFI, whereas LI from the same source had been incubated in two2-L jacketed beakers at 35 �C for more than 150 d before digestiontests. A digestion period of 21 d was used in order to compare withrelated studies (Bougrier et al., 2007; Carrere et al., 2010; Ma et al.,2011; Rani et al., 2012; Weemaes et al., 2000b). During incubation,LI was fed with AS from CVWRF with both solids retention time(SRT) and hydraulic retention time (HRT) of 21 d. The COD andthe volatile solids to total solids ratio (VS/TS) of AS was10 ± 1 g L�1 and 0.8, respectively. The volatile suspended solids(VSS) concentration of FI and LI was 12–14 and 3–5 g VS L�1,respectively, and VSS to total suspended solids ratio (VSS/TSS) ofFI and LI was 0.7 and 0.8, respectively.

2.2. Sludge pretreatment

In PAO treatment, 1.2 L of AS was placed in a 1.5-L closedreactor and subjected to 10 or 20 cycles of compression (withan O3/air stream) and decompression. PAO treatment and sup-plied ozone dose determination were previously detailed (Chenget al., 2012). The ozone generator (Model T-816, Polymetrics) wasfed with dry, filtered oxygen or compressed air at 105 V at the

flow rate of 2 L min�1. Each pressure cycle began by compressingan ozone-air mixture (0.06% O3 v/v) into the reactor to reach1040 kPa by an air compressor (RIDGID model 45150); oncereaching the target pressure, it was held there for 30 s forequilibration and then quickly vented to the ambient pressurein 3–5 s. For comparison, the same volume of AS was placed inthe same reactor and contacted with an ozone-air mixture(0.9% O3 v/v) bubbling through the suspension at 2 L min�1 for15 min under ambient pressure. In all ozonation treatments,any ejected foam and light particulate matters were collectedfor analysis and mass balance calculation. Solids concentration(TS, VS, TSS and VSS), pH, and COD of sludge were determinedbefore and after treatment per Standard Methods (APHA et al.,2005); soluble COD (sCOD) of sludge was determined for the fil-trate having passed through a 1.5-lm glass filter (Whatman 934-AH) and closed reflux colorimetric method per Standard Methods(APHA et al., 2005).

2.3. Batch anaerobic digestion with LI

Batch anaerobic digestion test with LI was carried out for 20 d in125-mL Erlenmeyer flasks. Prior to placement of the treated (20PAO and conventional ozonation) and untreated AS samples, theflasks were purged with nitrogen and each flask contained 30 mLof LI. To maintain a constant F/I ratio (g VS of AS/g VS of inoculum)of 0.8 in experiments with 30 mL of LI, the volume of AS (withvarying VS concentrations) to be incubated was varied. Forexample, 20 mL of untreated AS inoculated with 30 mL of LI willhave the same F/I ratio as 22 mL of PAO-treated AS (of lower VSconcentration) inoculated with 30 mL of LI. All flasks were sealedby rubber-sleeved stoppers and were incubated in a water-bathshaker agitated at 100 rpm at 35 �C (New Brunswick G76). Eachflask was connected to a DI water-filled gas collection tube, whichmeasured biogas production via liquid displacement. The COD, sol-ids, and pH of the digested AS/inoculum mixtures were measuredin triplicate or more at day 0 and 20 per Standard Methods (APHAet al., 2005). Biogas production was measured at regular timeintervals within the 20-d digestion period.

Separate digestion test was performed at 35 �C for nearly 150 din two jacketed beakers each of 1.5-L working volume approximat-ing complete-mix anaerobic reactors. The reactor was mixedcontinuously by a magnetic stirrer at 200 rpm. The original inocu-lum in both reactors was FI, fed with AS for a SRT of 21 d over10 months prior to experimentation. At day 0, one digester wasfed with untreated AS and the other with treated AS by 20 cyclesof PAO. The F/I ratio was 0.8 ± 0.1 and the SRT was 21 d for bothdigesters. TS, TSS, VS, VSS, COD, sCOD, and pH were measuredweekly per standard methods (APHA et al., 2005). Methane contentin the biogas and the biogas production rate were determined fromday 63 onward. Biogas was collected and measured weekly bywater displacement and its methane content measured periodi-cally by GC-TCD (HP 7890) equipped with a CARBONPLOT column.

2.4. Batch anaerobic digestion with FI

Anaerobic digestion was carried out over 14 d in batch using FI.The 125-mL Erlenmeyer flasks were likewise purged by nitrogenand mixed the AS with inoculum in different amounts to arriveat 49–68 mL of working volume, resulting in different F/I ratiosat 0.5, 1, and 2 g VS/g VS. The COD, sCOD, solids, and pH of the di-gested sludge along with the biogas production rate were mea-sured in duplicate at intervals within the 14-d digestion periodper Standard Methods (APHA et al., 2005).

C.-J. Cheng, P.K.A. Hong / Bioresource Technology 142 (2013) 69–76 71

3. Results and discussion

3.1. AS characteristics after ozonation pretreatments

As in other studies (Bougrier et al., 2006; Braguglia et al., 2006;Cheng et al., 2012; Zhang et al., 2009), the COD solubilization effi-ciency was calculated by: COD solubilization efficiency ¼sCODtreated�sCODinitial

CODinitial� 100%

The average COD solibilization and VSS reduction efficienciesfor AS using 20 pressure cycles of PAO as a pretreatment werefound to be 18% and 18%, respectively, significantly higher thanthose of 8% and 10%, respectively, using 15 min of conventionalbubbling ozonation, albeit the ozone dose expended in the formerpretreatment was five folds less than in the latter. These results aresimilar to the previous study (Cheng et al., 2012).

Ozonation of sludge has long shown benefits in improvedsludge treatability, available sCOD, and biogas production (Bougri-er et al., 2007; Carrere et al., 2010; Yeom et al., 2002; Zhang et al.,2009). The benefits of employing microbubble ozonation sludgepretreatment are greater than conventional bubbling ozonationdue to increased mass transfer from the microbubble gas to the li-quid phase leading to enhanced ozone utilization. Chu et al. (2008)found that when the sludge was treated by ozone supplied viamicrobubbles, more sCOD and cellular contents (source of totalnitrogen and phosphorous) were released from the sludge intothe supernatant than those from samples treated via conventionalbubbling ozonation. Cheng et al. (2012) found 37 folds of increasein sCOD concurrent with 25% reduction of TSS when the sludge wastreated by 10 mg O3 g�1 TSS via 20 PAO cycles; contrarily, lesssCOD and TSS reduction were found with a higher dose of80 mg O3 g�1 TSS delivered via conventional bubbling ozonation.They proposed a mechanism that explained the accelerated disin-tegration of AS beyond increased mass transfer due to the de-creased bubble size, and credited the expanding gas volume thatphysically disrupted the cell membrane for release of cell contents(Cheng et al., 2012). Results of Table 1 are consistent with theirprevious results of enhanced solubilization via PAO. Rani et al.(2012) also suggested that TSS reduction and soluble carbohydrateand protein increases were due to cell rupture of AS and release ofintracellular matter after subjecting AS to thermo-chemical treat-ment (60 �C at pH 12).

3.2. Solids reduction and biogas production during anaerobic digestionwith LI

Following ozonation pretreatment, AS was subjected to anaero-bic digestion with unacclimated FI at 0.8 for 20 d. Table 1 shows AScharacteristics afterward. Throughout the digestion period, the pHwas 7.2–7.4 and the VSS/TSS ratio was 0.7–0.8. At day 0, the sCODconcentrations prior to inoculation with LI were 85 ± 4, 720 ± 33,

Table 1TSS, VSS, sCOD, and sCOD/COD and changes of untreated and ozonated AS after 20 d of an

No treatment

Ozone dose (mg O3 g�1 TSS) –ReductionTSS (%) 17VSS (%) 22sCODa (%) �14

Day 0 Day 20pH 7.2 7.2sCOD (mg L�1) 85 ± 4 97 ± 10COD (mg L�1) 8347 ± 600 7260 ± 764sCOD/COD (#) 0.010 0.013

a The reduction of sCOD was calculated by: sCOD reduction ð%Þ ¼ sCOD0�sCOD20 dsCOD0

� 10

and 470 ± 39 mg L�1 for untreated, PAO-treated, and ozone-bub-bled sludges, respectively, which showed significantly higher sCODin ozonated AS (highest after PAO) than the untreated AS. After20 d of digestion with acclimated LI, the sCOD/COD ratios of allsamples were below 0.02 indicating extensive depletion of sCOD.The extents of TSS, VSS, and sCOD reduction followed the samedecreasing order of PAO-treated AS > ozone-bubbled AS > un-treated AS. Similar specific COD reduction efficiencies, approxi-mately 0.4 g COD g�1 VSinoculum, were found for all three AS;however, much more sCOD disappeared from ozonated AS thanfrom the untreated AS. A higher sCOD removal in ozonated ASbut similar specific COD removals in all three AS were possible be-cause the sCOD accounted for only a small fraction of the totalCOD. Fig. 1 shows specific biogas production rates (mL gas g�1

CODsupplied) of the three AS during 20 d of anaerobic digestionusing F/I of 0.8. Specific biogas production rates from AS pretreatedwith PAO and conventional ozonation were 800% and 240%,respectively, higher than from AS without pretreatment.

In contrast to ozonated AS, sCOD in the untreated AS increasedafter 20 d of digestion, which indicated hydrolysis playing a majorrole in the untreated AS during this period. Furthermore, the in-creased solids and sCOD reductions accompanied by increased bio-gas production with the ozonated AS indicated an enhanceddigestion process via ozonation pretreatment. Higher TSS andVSS reductions in ozonated AS lent support to the thesis that phys-ical disintegration of the flocs resulted in more ruptured biomassthat was more amenable to digestion than untreated ‘‘whole’’ bio-mass. However, when conventional ozonation dose was increasedto 50 mg O3 g�1, AS digestion efficiencies (i.e. solids reduction andbiogas production) were less than those treated by PAO. Refractorymaterials of AS were found to influence not only solids and CODremovals but also anaerobic microbial activities (Bougrier et al.,2006). Specifically, refractory materials were attributed to negligi-ble VS removal after 10 d of anaerobic digestion (Braguglia et al.,2012). Enhanced digestion of PAO-treated AS possibly resultedfrom less or none of the refractory compounds being formed be-cause of the smaller ozone dose and shorter contact time used.Cheng et al. (2012) showed higher ozone utilization rate withPAO treatment than with conventional bubbling ozonation, whichare consistent with higher solids reduction and specific biogas pro-duction in PAO-treated AS.

3.3. Complete-mix digesters with LI

Solids reduction and biogas production efficiencies of PAO-trea-ted AS were further examined in two complete-mix reactors foranaerobic digestion with periodic feeds over 147 d simulating aSRT of 21 d. VS reduction of PAO-treated AS was slightly moreextensive than of AS without treatment, as shown by results ofFig. 2a. VS reduction efficiencies over time of the two reactors were

aerobic digestion inoculated with LI at F/I of 0.8.

PAO of 20 cycles Bubbling O3

10 52

25 2030 2680 71

Day 0 Day 20 Day 0 Day 207.3 7.2 7.4 7.2

719 ± 33 144 ± 20 470 ± 39 134 ± 177920 ± 57 6790 ± 511 7600 ± 481 6480 ± 721

0.091 0.021 0.062 0.021

0%.

0

50

100

150

200

250

0 5 10 15 20

Spec

ific

bio

gas

prod

ucti

on(m

L g

as g

-1 C

OD

supp

lied)

Time (d)

No treatment 20 cycles PAO 15 min Conventional ozonation

Fig. 1. Specific biogas production of batch test with acclimated inocula at 0.8 F/I ratio.

0

5

10

15

20

25

30

35

40

45

50

0 50 100 150

VS

redu

ctio

n (%

)

Time (d)

20 cycles PAONo treatment20 cycles PAO fittingNo treatment fitting

150

200

250

300

105 115 125 135 145

Spec

ific

bio

gas

prod

ucti

on

(mL

bio

gas

g-1 C

OD

supp

lied)

Time (d)

20 cycles PAO

No treatment

(a)

(b)

Fig. 2. Results of 150 d digestion test with LI in terms of (a) VS reduction efficiency (%) calculated by: (VSinf – VSeff)/VSinf and (b) specific biogas production between day 105 today 147 of the test.

72 C.-J. Cheng, P.K.A. Hong / Bioresource Technology 142 (2013) 69–76

fitted by a saturation-type, non-linear equation asY ¼ Yeð1� expð�ktÞÞ; where Y denoted VS the reduction efficiency(%), Ye the VS reduction efficiency at equilibrium (%), k a first-orderrate constant (d�1), and t the digestion time (d). Ye values for

reactors fed with untreated and PAO-treated AS were 35% and40%, respectively. The VS content from the wastewater treatmentplant changed significantly, accounting for the variations in thefigure. Specifically, at days 112 and 116, samples indicated VS of

0

10

20

30

40

50

60

0 5 10 15

Spec

ific

VS

redu

ctio

n (%

)

Time (d)

No treatment, F/I = 0.5

10 cycles PAO, F/I = 0.5

20 cycles PAO, F/I = 0.5

No treatment, F/I = 1

10 cycles PAO, F/I = 1

20 cycles PAO, F/I = 1

No treatment, F/I = 2

10 cycles PAO, F/I = 2

20 cycles PAO, F/I = 2

0

10

20

30

40

50

60

0 5 10 15

Spec

ific

VSS

red

ucti

on (

%)

Time (d)

No treatment, F/I = 0.5

10 cycles PAO, F/I = 0.5

20 cycles PAO, F/I = 0.5

No treatment, F/I = 1

10 cycles PAO, F/I = 1

20 cycles PAO, F/I = 1

No treatment, F/I = 2

10 cycles PAO, F/I = 2

20 cycles PAO, F/I = 2

0

10

20

30

40

50

60

0 5 10 15

Spec

ific

CO

D r

educ

tion

(%

)

Time (d)

No treatment, F/I = 0.5

10 cycles PAO, F/I = 0.5

20 cycles PAO, F/I = 0.5

No treatment, F/I = 1

10 cycles PAO, F/I = 1

20 cycles PAO, F/I = 1

No treatment, F/I = 2

10 cycles PAO, F/I = 2

20 cycles PAO, F/I = 2

Fig. 3. The specific VS, VSS, and COD reduction efficiencies during 14-d digestion of untreated and PAO treated AS at F/I ratio of 0.5, 1, and 2.

C.-J. Cheng, P.K.A. Hong / Bioresource Technology 142 (2013) 69–76 73

13.6 g L�1 and 10.8 g L�1, respectively, which were higher than theaverage of 8.6 g L�1 throughout the entire period. This could affectthe calculated removal of VS, which was calculated by(VSinf – VSeff)/VSinf. A higher influent VS on these particular dayscould result in a higher than usual values of removal. Thus, thespikes in removal were attributed to spikes of VS content in thesamples. The methane content of biogas increased from 51% to

56% when one reactor was switched from untreated AS to PAO-treated feed, while the methane content of the other reactor with-out the feed switch remained unchanged throughout. Duringdigestion at steady state, e.g. day 105 to 147, the average specificbiogas production rates were 202 and 215 mL g�1 CODsupplied foruntreated and PAO-treated AS, respectively (Fig. 2b). It should benoted that LI had been fed and thus acclimated with untreated

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0 5 10 15

Spec

ific

sub

stra

te u

tiliz

atio

n ra

te

(gA

S g-1

inoc

ulum

d-1

)

Initial COD (g L-1)

No treatment

10 cycles PAO

20 cycles PAO

Linear (No treatment)

Linear (10 cycles PAO)

Linear (20 cycles PAO)

Fig. 4. Linear plot of specific substrate utilization rate vs. available substrate.

Table 2Second-order rate constant (k0; L g�1 d�1) and correlation coefficient (R2) during 14-ddigestion of AS inoculated with FI at different F/I ratios.

F/I = 0.5 F/I = 1 F/I = 2

k0 R2 k0 R2 k0 R2

No treatment 0.10 0.90 0.21 0.87 0.23 0.8710 cycles of PAO 0.10 0.73 0.21 0.92 0.35 0.9620 cycles of PAO 0.09 0.90 0.16 0.93 0.24 0.99

74 C.-J. Cheng, P.K.A. Hong / Bioresource Technology 142 (2013) 69–76

AS for 10 months, and was not necessarily adapted to ozonated AS.However, higher digestibility of PAO-treated AS in both batch andcontinuously fed reactor suggested improved anaerobic digestionvia PAO.

3.4. Anaerobic digestion in batch with FI

3.4.1. Degradation efficiencies and kinetics analysisFig. 3 shows the specific reduction efficiencies of solids and COD

(%; 100 g AS g�1 VSinoculum) after 14 d of anaerobic digestion inbatch with FI according to different treatments and F/I ratios. Gen-erally, solids and COD reductions during digestion were higher forAS with PAO treatment, showing positive impact on AS digestionby PAO at doses of 5 and 10 mg O3 g�1 TSS supplied by 10 and20 cycles, respectively. However, the tested PAO conditions werelimited and identification of optimal conditions was not possiblewithout further investigation. In separate experiments, the diges-tion of AS treated by bubbling ozonation at 52 mg O3 g�1 TSS wastested and solids reduction and biogas production were found sim-ilar to those of untreated AS (data is not shown).

COD removals were investigated with varied F/I ratios and AStreated differently; the results were analyzed as specific COD utili-zation rate vs. available COD. COD removal rates were obtained bytaking the instantaneous slope (tangential line) of the monitoredCOD vs. time profile. The linearity of specific utilization rate vs.available substrate in Fig. 4 confirms for COD removal via anaero-bic digestion as first-order with respect to the available substrate,i.e. AS, and first-order with respect to the inoculum amount. Theslopes are determined by linear regression analysis, yieldingfirst-order rate constants of 0.019, 0.022, and 0.026 d�1 foruntreated AS, AS treated by 10 cycles of PAO, and AS treated by

20 cycles of PAO, respectively. Moreover, the second-order rateconstant (k0) for AS digestion by inoculum can be calculated aslisted in Table 2. At the same F/I ratio, the rate constants of diges-tion were similar for variously treated AS indicating little depen-dence on treatment; however, the obtained rate constants variedfor different F/I ratios. Tomei et al. (2008) found the effect of son-ication on anaerobic digestion to be as significant as F/I ratio atthe range of 0.1–4.0. Increasing F/I resulted in an increase of pro-cess kinetics up to 10 folds, while the effect of sonication wasnot appreciable at low F/I of 0.1 and 0.5, which were attributedto low substrate availability. Fig. 3 further shows increasing CODreduction with increasing F/I ratio even without PAO treatment.Given the same PAO treatment, solids and COD reductions also in-creased with increasing F/I ratio from 0.5 to 1; however, the effectof PAO pretreatments on COD or solids reduction at F/I = 2 was lessthan or similar to those at F/I of 1. The relationship between CODremoval and F/I ratio was more complex and might be due to sev-eral interacting factors at work. For example, the degradable CODfractions in the mixture of inoculum and AS were changed withchanges in F/I ratios (Braguglia et al., 2006; Tiehm et al., 2001; To-mei et al., 2008). Besides differences in available substrates to thebacterial consumers, differences in species working on differentsubstrates and differences in the ranges of available substratesand consumers as differentiated by experimental designs (i.e.,varying F/I ratios in batches while keeping the incubation volumeconstant) were considered by various researchers (Bragugliaet al., 2006; Gungor-Demirci and Demirer, 2004; Ma et al., 2011;Patil et al., 2012; Neves et al., 2004).

3.4.2. sCOD changes during digestionFig. 5 shows sCOD (mg L�1) vs. time profiles for untreated and

PAO-treated AS at F/I of 0.5, 1, and 2. Initial sCOD in the treatedsludge just prior to anaerobic digestion was high (1000–2500 mg L�1), which confirmed effective solids and COD solubiliza-tion by 10 and 20 cycles of PAO treatments. In many cases, sCODcontinued to increase at the beginning of digestion. These increasesoccurred in the first 5 d, to be followed by continual decreases to<550 mg L�1 during the rest of the 14-d digestion period. In caseswhen the initial sCOD were very high (e.g. in cases of 10 and 20 -cycles of PAO at F/I of 2), sCOD decreased immediately from thebeginning of digestion without any initial sCOD increases in thefirst 5 days. Braguglia et al. (2006) reported similar results and

0

500

1000

1500

2000

2500

3000

0 3 6 9 12 15

sCO

D (m

g L

-1)

Time (d)

No treatment, F/I = 0.5

10 cycles PAO, F/I = 0.5

20 cycles PAO, F/I = 0.5

0

500

1000

1500

2000

2500

3000

0 3 6 9 12 15

sCO

D (m

g L

-1)

Time (d)

No treatment, F/I = 1No treatment, F/I = 210 cycles PAO, F/I = 110 cycles PAO, F/I = 220 cycles PAO, F/I = 120 cycles PAO, F/I = 2

Fig. 5. sCOD vs. time profiles at F/I = 0.5, 1, and 2 for 14-d anaerobic digestion test.

C.-J. Cheng, P.K.A. Hong / Bioresource Technology 142 (2013) 69–76 75

suggested that the inoculum in the batch reactors was able to startutilizing sCOD in the fermentation phase skipping or greatly short-ening the hydrolysis phase when initial sCOD was abundant. A sig-nificant decrease in soluble carbohydrate concentration wasobserved during 12–24 h of test of hydrogen production when ASwas pretreated by 0.158 g O3 g�1 dry solids (Yang et al., 2012).The changes in soluble carbohydrate were positively related tohydrogen production. It was suggested that the released solubleintracellular matter after ozonation and/or ultrasound pretreat-ments was readily digested by anaerobic bacteria.

3.4.3. Biogas productionAfter 14 d of batch digestion, the specific biogas produced from

untreated AS was 112, 297, and 520 mL g�1 VSinoculum at F/I of 0.5,1, and 2, respectively. For 20 cycles of PAO-treated AS, accumu-lated specific biogas production increased from 171 to 628 mL g�1

VSinoculum when the F/I was increased from 0.5 to 2 after 14 d.Although the accumulated specific biogas production of all sam-ples increased significantly with increasing F/I ratios, the influenceof PAO treatment on enhancing biogas production at a specific F/Iratio was much less than that of F/I ratios. At lower F/I ratio of0.5, PAO treatment of AS exerted a positive effect on gas produc-tion, while no obvious effects were observed at the high F/I ratioof 2. When F/I was 0.5, the specific gas productions for 10 and20 cycles of PAO-treated AS were 190% and 150% more than thatfor untreated AS, respectively.

4. Conclusions

Ozonation and PAO treatments solubilized AS, providing solidsreduction and available sCOD. Both VSS reduction and COD solubi-lization of AS after 20 cycles of PAO (10 mg O3 g�1 TSS) were 18%,higher than those of 10% and 8%, respectively, after conventionalozonation (50 mg O3 g�1 TSS). Digestion of pretreated AS withacclimated inoculum showed significant improvement in solidsreduction and biogas production. At F/I of 0.8, VSS reduction wasimproved by 1.2 and 1.6 folds by ozonation and PAO pretreat-ments, respectively. Specific biogas production with LI was im-proved by 240% with ozonation and by 800% with PAO.

Acknowledgements

We thank the Central Valley Water Reclamation Facility of SaltLake City, Utah for AS and digester samples. We appreciate the par-tial financial support of The University of Utah ResearchFoundation.

References

APHA, AWWA, WEF, 2005. Standard methods for the examination of water andwastewater, 21 ed. APHA, AWWA, and WEF, Maryland.

Barjenbruch, M., Kopplow, O., 2003. Enzymatic, mechanical and thermal pre-treatment of surplus sludge. Adv. Environ. Res. 7 (3), 715–720.

76 C.-J. Cheng, P.K.A. Hong / Bioresource Technology 142 (2013) 69–76

Bougrier, C., Albasi, C., Delgenes, J.P., Carrere, H., 2006. Effect of ultrasonic, thermaland ozone pre-treatment on waste activated sludge solubilization andanaerobic biodegradability. Chem. Eng. Process. 45, 711–718.

Bougrier, C., Battimelli, A., Delgenes, J.-P., Carrere, H., 2007. Combined ozonepretreatment and anaerobic digestion for the reduction of biological sludgeproduction in wastewater treatment. Ozone: Sci. Eng. 29 (3), 201–206.

Braguglia, C.M., Gianico, A., Mininni, G., 2012. Comparison between ozone andultrasound disintegration on sludge anaerobic digestion. J. Environ. Manage. 95(Suppl.), S139–S143.

Braguglia, C.M., Mininni, G., Tomei, M.C., Rolle, E., 2006. Effect of feed/inoculumratio on anaerobic digestion of sonicated sludge. Water Sci. Technol. 54 (5), 77–84.

Carballa, M., Duran, C., Hospido, A., 2011. Should we pretreat solid waste prior toanaerobic digestion? An assessment of its environmental cost. Environ. Sci.Technol. 45 (24), 10306–10314.

Carrere, H., Dumas, C., Battimelli, A., Batstone, D.J., Delgenes, J.P., Steyer, J.P., Ferrer,I., 2010. Pretreatment methods to improve sludge anaerobic degradability: areview. J. Hazard. Mater. 183 (1–3), 1–15.

Cha, Z., Lin, C.-F., Cheng, C.-J., Hong, P.K.A., 2010. Removal of oil and oil sheen fromproduced water by pressure-assisted ozonation and sand filtration.Chemosphere 78 (5), 583–590.

Cheng, C.-J., Andy Hong, P.K., Lin, C.-F., 2012. Improved solubilization of activatedsludge by ozonation in pressure cycles. Chemosphere 87 (6), 637–643.

Choi, H.B., Kwang, K.Y., Shin, E.B., 1997. Effects of anaerobic digestion of sewagesludge pretreatment. Water Sci. Technol. 35 (10), 207–211.

Chu, L., Wang, J., Wang, B., Xing, X.-H., Yan, S., Sun, X., Jurcik, B., 2009a. Changes inbiomass activity and characteristics of activated sludge exposed to low ozonedose. Chemosphere 77 (2), 269–272.

Chu, L., Yan, S., Xing, X.-H., Sun, X., Jurcik, B., 2009b. Progress and perspectives ofsludge ozonation as a powerful pretreatment method for minimization ofexcess sludge production. Water Res. 43 (7), 1811–1822.

Chu, L.-B., Yan, S.-T., Xing, X.-H., Yu, A.-F., Sun, X.-L., Jurcik, B., 2008. Enhancedsludge solubilization by microbubble ozonation. Chemosphere 72 (2), 205–212.

Dogruel, S., Sievers, M., Germirli-Babuna, F., 2007. Effect of ozonation onbiodegradability characteristics of surplus activated sludge. Ozone: Sci. Eng.29 (3), 191–199.

Elliott, A., Mahmood, T., 2007. Pretreatment technologies for advancing anaerobicdigestion of pulp and paper biotreatment residues. Water Res. 41 (19), 4273–4286.

Gungor-Demirci, G., Demirer, G.N., 2004. Effect of initial COD concentration,nutrient addition, temperature and microbial acclimation on anaerobictreatability of broiler and cattle manure. Bioresour. Technol. 93 (2), 109–117.

Hong, P.K.A., Cai, X., Cha, Z., 2008. Pressure-assisted chelation extraction of leadfrom contaminated soil. Environ. Pollut. 153 (1), 14–21.

Hong, P.K.A., Nakra, S., 2009. Rapid extraction of sediment contaminants bypressure cycles. Chemosphere 74 (10), 1360–1366.

Jensen, P.D., Ge, H., Batstone, D.J., 2011. Assessing the role of biochemical methanepotential tests in determining anaerobic degradability rate and extent. WaterSci. Technol. 64 (4), 880–886.

Ma, J., Duong, T.H., Smits, M., Verstraete, W., Carballa, M., 2011. Enhancedbiomethanation of kitchen waste by different pre-treatments. Bioresour.Technol. 102 (2), 592–599.

Neves, L., Oliveira, R., Alves, M.M., 2004. Influence of inoculum activity on the bio-methanization of a kitchen waste under different waste/inoculum ratios.Process Biochem. 39 (12), 2019–2024.

Park, C.B., Clark, D.S., 2002. Rupture of the cell envelope by decompression of thedeep-sea methanogen Methanococcus jannaschii. Appl. Environ. Microbiol. 68(3), 1458–1463.

Park, K.Y., Ahn, K.-H., Maeng, S.K., Hwang, J.H., Kwon, J.H., 2003. Feasibility of sludgeozonation for stabilization and conditioning. Ozone: Sci. Eng. 25 (1), 73–80.

Patil, J.H., Shetty, V., Hosur, M., Muralidhara, P.L., 2012. Impact of incubation periodof primary sludge inoculum on biomethanation of water hyacinth. J. Chem.Pharm. Res. 4 (3), 1719–1724.

Rai, C.L., Rao, P.G., 2009. Influence of sludge disintegration by high pressurehomogenizer on microbial growth in sewage sludge: an approach for excesssludge reduction. Clean Technol. Environ. Policy 11 (4), 437–446.

Rani, R.U., Jumar, S.A., Kaliappan, S., Yeom, I.-T., Banu, J.R., 2012. Low temperaturethermo-chemical pretreatment of dairy waste activated sludge for anaerobicdigestion process. Bioresour. Technol. 103 (1), 415–424.

Tiehm, A., Nickel, K., Zellhorn, M., Neis, U., 2001. Ultrasonic waste activated sludgedisintegration for improving anaerobic stabilization. Water Res. 35 (8), 2003–2009.

Tomei, M.C., Braguglia, C.M., Mininni, G., 2008. Anaerobic degradation kinetics ofparticulate organic matter in untreated and sonicated sewage sludge: role ofthe inoculum. Bioresour. Technol. 99 (14), 6119–6126.

Weemaes, M., Grootaerd, H., Simoens, F., Huysmans, A., Verstraete, W., 2000a.Ozonation of sewage sludge prior to anaerobic digestion. Water Sci. Technol. 42(9), 175–178.

Weemaes, M., Grootaerd, H., Simoens, F., Verstraete, W., 2000b. Anaerobic digestionof ozonized biosolids. Water Res. 34 (8), 2330–2336.

Yang, S.-S., Guo, W.-Q., Cao, G.-L., Zheng, H.S., Ren, N.Q., 2012. Simultaneous wasteactivated sludge disintegration and biological hydrogen production using anozone/ultrasound pretreatment. Bioresour. Technol. 124, 347–354.

Yasui, H., Komatsu, K., Goel, R., Li, Y.Y., Noike, T., 2005. Full-scale application ofanaerobic digestion process with partial ozonation of digested sludge. WaterSci. Technol. 52 (1–2), 245–252.

Yeom, I.T., Lee, K.R., Lee, Y.H., Ahn, K.H., Lee, S.H., 2002. Effects of ozone treatmenton the biodegradability of sludge from municipal wastewater treatment plants.Water Sci. Technol. 46 (4–5), 421–425.

Zhang, G., Yang, J., Liu, H., Zhang, J., 2009. Sludge ozonation: disintegration,supernatant changes and mechanisms. Bioresour. Technol. 100 (3), 1505–1509.