Bioresource Technology 100 (2009) 3799–3807

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /b ior tech

Comparison of static, in-vessel composting of MSW with thermophilicanaerobic digestion and combinations of the two processes

Lee Walker *, Wipa Charles, Ralf Cord-RuwischFaculty of Sustainability, Environmental and Life Sciences, School of Biological Sciences and Biotechnology, Murdoch University, South Street, Murdoch, Western Australia 6150,Australia

a r t i c l e i n f o

Article history:Received 4 August 2008Received in revised form 5 February 2009Accepted 9 February 2009Available online 3 April 2009

Keywords:Thermophilic anaerobic digestionIn-vessel compostingHydrolysis limitationMSW stabilisationDiCOM�

0960-8524/$ - see front matter � 2009 Elsevier Ltd. Adoi:10.1016/j.biortech.2009.02.015

* Corresponding author. Tel.: +61 8 9360 2815; faxE-mail address: L.Walker@murdoch.edu.au (L. Wa

a b s t r a c t

The biological stabilisation of the organic fraction of municipal solid waste (OFMSW) into a form stableenough for land application can be achieved via aerobic or anaerobic treatments. To investigate the ratesof degradation (e.g. via electron equivalents removed, or via carbon emitted) of aerobic and anaerobictreatment, OFMSW samples were exposed to computer controlled laboratory-scale aerobic (static in-ves-sel composting), and anaerobic (thermophilic anaerobic digestion with liquor recycle) treatment individ-ually and in combination. A comparison of the degradation rates, based on electron flow revealed thatprovided a suitable inoculum was used, anaerobic digestion was the faster of the two waste conversionprocess. In addition to faster maximum substrate oxidation rates, anaerobic digestion (followed by post-treatment aerobic maturation), when compared to static composting alone, converted a larger fraction ofthe organics to gaseous end-products (CO2 and CH4), leading to improved end-product stability andmaturity, as measured by compost self-heating and root elongation tests, respectively. While not compa-rable to windrow and other mixed, highly aerated compost systems, our results show that in the thermo-philic, in-vessel treatment investigated here, the inclusion of a anaerobic phase, rather than usingcomposting alone, improved hydrolysis rates as well as oxidation rates and product stability. The combi-nation of the two methods, as used in the DiCOM� process, was also tested allowing heat generation tothermophilic operating temperature, biogas recovery and a low odour stable end-product within 19 daysof operation.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Historically, strategies for the management of municipal solidwaste (MSW) included burial, burning and ocean dumping; prac-tices which are now known to lead to contamination of land, airand sea (Earle et al., 1995). Until recently landfill was the mainwaste treatment method utilised. Waste management has becomeone of the largest environmental concerns in recent decades, withthe problems in disposal compounded by the ever-increasingquantity of refuse to be managed. The scarcity of land and theuncontrolled contamination with gas and leachate emissions havemade landfill, particularly of organics, no longer a sustainable op-tion (Hartmann and Ahring, 2006). It is now accepted that no singlesolution exists for the management of MSW, with an integrated ap-proach most likely to succeed (Earle et al., 1995).

Approximately 50% of MSW consists of organic matter, with thecomposition of the organic fraction of MSW (OFMSW) being animportant parameter in determining the most appropriate methodfor its treatment. Typically, food waste, which is too wet and lacks

ll rights reserved.

: +61 8 9360 6303.lker).

the structure for composting, is treated via anaerobic digestionwhereas green waste (plant material) is composted (Edelmannand Engeli, 1993; Braber, 1995).

Both composting and anaerobic digestion have their own spe-cific advantages and disadvantages (Table 1), with compostinggenerally accepted as being a more rapid process than anaerobictreatment (Lopes et al., 2004; Mohaibes and Heinonen-Tanski,2004). However, based on an energy balance, anaerobic digestionhas an advantage over composting, incineration, a combinationof composting and digestion (Edelmann et al., 2005) or land-filling(Haight, 2005), with anaerobic digestion capable of being energysufficient if only one quarter of the biogenic waste is digested tobiogas (Edelmann et al., 2000).

One recognised disadvantage of anaerobic digestion is the factthat the solids produced are not typically suitable for direct landapplication as they tend to be odorous, too wet and too high in vol-atile fatty acid (VFA) concentration, which are phytotoxic. In addi-tion, if the digestion is not performed under thermophilicconditions, the solids are not sanitised. Consequently, a post treat-ment of these solids is required (Poggi-Varaldo et al., 1999) withcomposting providing an appropriate management solution (Frickeet al., 2005; Meissl and Smidt, 2007).

Table 1Advantages and disadvantages of composting and anaerobic digestion. References: (a)Edelmann and Engeli (1993); (b) Braber (1995); (c) De Baere (1999); (d) Edelmannet al. (1999); (e) Smet et al. (1999); (f) De Baere (2000); (g) Edelmann et al. (2000); (h)Mata-Alvarez et al. (2000); (i) Edelmann et al. (2005) and (j) Hartmann and Ahring(2006).

Composting Anaerobic Digestion References

Advantage Disadvantage Advantage Disadvantage

Simple More complex a, b, f, hInexpensive More expensive

Larger area Smaller area a, bOdourpollution

Reduced odourvia biogascombustion

a, b, c, e

Uncontrolledleachateemission

High strengthwastewaterformed

j

UncontrolledCH4

production

b, c, g, i

Net energyconsumer

Net energyproducer

d

Table 2Phase lengths of trials.

Trial description Length ofinitial aeration(days)

Length ofanaerobictreatment (days)

Length of finalaeration(days)

Thermophilic anaerobicdigestion (B)

0 12 7

Combined aerobic andthermophilic anaerobicdigestion (C)

1 10 7

DiCOM� (A) 5 7 7Static in-vessel composting

(D)12 0 7

Thermophilic staticcomposting (E)

12 0 7

3800 L. Walker et al. / Bioresource Technology 100 (2009) 3799–3807

Providing aerobic maturation for the solid digestate obtainedfrom anaerobic treatment of solid waste has been found to im-prove the quality of the end-product (Edelmann and Engeli,1993) and reduce odour emission (De Baere, 1999) by reducingthe emission of volatile compounds (Smet et al., 1999). Recentyears have seen the development of numerous integrated wastetreatment systems (Six and De Baere, 1992; Kayhanian andTchobanoglous, 1993; Kübler and Schertler, 1994; ten Brummeler,2000; Müller et al., 2003; Teixeira et al., 2004; Walker et al., 2006a)utilising post-anaerobic digestion composting to mature anaerobicdigestate.

During composting, readily degradable substrates (which in-clude residual VFA from anaerobic solids) are rapidly consumedwith the significant energy released heating up the material.Depending on the degradability of the organic substrate, the oxy-gen (O2) supply and heat loss, the temperature of the materialcan rise to 70 �C or more which contributes to eliminating patho-gens from the material (Neklyudov et al., 2006). It is therefore log-ical and advantageous to develop processes which combineanaerobic digestion and composting to provide improved wasteprocessing. The DiCOM� process (Walker et al., 2006a,b; DiCOM,2009), developed and patented by AnaeCo Ltd. (Perth, WesternAustralia), is one such process. It exposes OFMSW to 5 days of pres-surised aeration, followed by 7 days of thermophilic anaerobicdigestion (with liquor transfer and recirculation) and 7 days of aer-obic maturation within a single completely sealed reactor. The fi-nal products of this process are a composted end-product andrenewable energy in the form of biogas.

The aim of this paper is to compare rates of degradation ofOFMSW under static, in-vessel aerobic composting and thermo-philic anaerobic digestion conditions and a combination of thetwo processes. As an example of a full-scale process that combinesaerobic composting and thermophilic anaerobic digestion, thetreatment regime of the DiCOM� process was also investigated.Data produced are expected to be helpful in designing the mostefficient combination of the two methods for the degradation ofOFMSW.

2. Methods

2.1. Experimental design

Mixed MSW collected in the metropolitan area of Perth, Wes-tern Australia in March, 2005 was mechanically sorted using a

screen aperture of 50 mm. Larger inert objects (plastic, metal andglass) in the sorted OFMSW were removed by hand before theorganics were further shredded (<25 mm). Portions of the wellmixed batch of OFMSW were combined with shredded paper(Hygenex� 2187951) and Jarrah wood chips (trapped between 1and 5 mm screens) in the ratio of 1000:17:67 (w/w) (to replacethat removed during the mechanical sorting process and providea solid matrix) and frozen (�20 �C) to provide an identical startingmaterial for all trials and reproducible outcomes. Prior to use, sam-ples (2.4 kg; wet bulk density 578 kg/m3; free air space 55%; C:Nwas 18:1; 55% moisture content; 56% total volatile solids content(VS); 8.3% protein; 4.3% fat and 45% carbohydrate) were thawedat room temperature and deionised water (�400 mL) added to pro-vide a positive ‘fist test’, as described in Australian Standard 4454(2003).

The OFMSW was treated in an insulated cylindrical, 7 L hightemperature PVC computer controlled laboratory-scale reactor asdescribed previously by Walker et al. (2006a). The reactor wasoperated as a sequencing batch reactor capable of providing in-vessel composting, anaerobic digestion or combinations of both.

Trials consisted of at most 12 days of treatment (aerobic, anaer-obic or a combination of both) followed by 7 days of aerobic mat-uration (Table 2). During aerobic operation, pressurised air wasintroduced into the reactor until the internal pressure was raisedto a predetermined level. The internal pressure was maintained(10 min) before being released and the aeration regime repeated.This aeration regime was used to prevent channelling of airthrough the essentially unmixed material. Small scale compostingtrials typically underperform due to limited heat build-up (highsurface to volume ratio causing increased heat loss). To preventthe heat loss typical for small scale composting experiments, ahighly insulated vessel (heat loss coefficient 0.0912 h�1) was usedand the external reactor temperature controlled, by means of aheat tape, to that of the reactor core, but not beyond 60 �C.

For those trials where a thermophilic anaerobic phase was pres-ent (Trials A, B and C; Table 2), digestion was initiated by sealingthe reactor and allowing aerobic microbial metabolism to consumeresidual oxygen and establish an anaerobic environment. Onceanaerobic conditions had been established, the reactor was floodedwith 4.1 L liquid (anaerobic inoculum obtained from a laboratory-scale DiCOM� reactor) (NHþ4 –N = 1400 mg/L). Then the liquor wasre-circulated (70 mL/min (max)) from the reactor top to its baseand maintained at 55 ± 2 �C. At the conclusion of digestion, theanaerobic liquid was drained and the solids mechanically squeezed(to remove excess moisture and provide a positive ‘‘fist test” (AS4454, 2003)).

Aerobic post-digestion maturation consisted of 7 days of aera-tion as described above. Again, to limit heat loss from this smallscale reactor, external heating was used to ensure the core reactortemperature did not fall below 35 ± 2 �C.

L. Walker et al. / Bioresource Technology 100 (2009) 3799–3807 3801

2.2. Specific trial information

� Trial A: The DiCOM� trial was conducted as previously describedby Walker et al. (2006a).

� Trial B: For the anaerobic trial, the OFMSW was heated to 55 �Cin an incubator (3 h) prior to being introduced to the reactor.

� Trial C: The combined aerobic/anaerobic trial was conducted asper Trial A except pre-aeration was reduced to 1 day whileanaerobic digestion was operated for 10 days.

� Trial D: For the static in-vessel composting trial the introducedair was heated and moistened by sparging through water (Van-derGheynst et al., 1997) at 55 ± 2 �C to avoid evaporative cool-ing and drying of the OFMSW (Mata-Alvarez et al., 1993;Mason and Milke, 2005). During days 5–12, the oxygen wasadded at a frequency (pressurising and releasing) such thatthe O2 concentration in the exit gas was between approxi-mately 1% and 3%.

� Trial E: Thermophilic composting was conducted as per Trial Dexcept during days 5–12 where the core temperature was forcedto remain above 55 ± 2 �C. This was to allow a fair comparison ofcomposting at the same temperature as the thermophilic anaer-obic digestion.

2.3. Data collection and chemical analysis

O2 concentration, airflow rate, internal pressure, core and out-side reactor temperature (digital thermocouples (one placed inthe core and the other on the reactor circumference)), pH andbiogas generation rate were logged by computer using a NationalInstruments data acquisition card (NI PCI 6224 M Series DAQ)and National Instruments LabView 7 control software. O2 con-centration in the exit gas during the aeration phases was mea-sured using a polarographic (Mettler Toledo InPro6100) O2

sensor. Biogas production was determined by the downward dis-placement of oil (Dow Corning 200 Fluid 50CS) with real-timecarbon dioxide (CO2) and methane (CH4) concentrations in theexit gas logged by a gas analyzer (Geotechnical Instruments GA2000).

A Varian Star 3400 gas chromatograph (GC) fitted with a Varian8100 auto-sampler was used to analyze the VFA concentration ofliquid samples. Samples were acidified with formic acid (to 1%(v/v)) before 1 lL samples were injected onto an Alltech ECONO-CAPTM ECTM 1000 (15 m � 0.53 mm (i.d.) 1.2 lm) column. The carriergas (N2) was set at a flow rate of 5 mL/min. The oven temperaturewas programmed as follows: initial temperature 80 �C; tempera-ture ramp 40 �C/min to 140 �C, hold for 1 min; temperature ramp50 �C/min to 230 �C hold for 2 min. Injector and detector tempera-tures were set at 200 and 250 �C, respectively. The peak area of theFlame Ionisation Detector (FID) output signal was computed viaintegration using STAR Chromatography Software (� 1987–1995).

The moisture content of OFMSW was obtained by heating over-night at 105 �C. Liquid extracts from compost samples were ob-tained and used to determine NHþ4 content, pH and conductivityas per Australian Standard (AS 4454 – 2003). NHþ4 concentrationin extracts was determined as described by American Public HealthAssociation (1992). Total organic and inorganic carbon and Kjel-dahl nitrogen were analyzed by Marine and Freshwater ResearchLaboratory, Murdoch, Western Australia. Protein, Lipid and Carbo-hydrate analysis was performed by Chemistry Centre (W.A.), Perth,Western Australia.

Compost stability was assessed using the self-heat test. Thetests were carried out according to the Australian Standard (AS4454 – 2003), except 1.9 L stainless steel thermos flasks (AladdinAustralia) were substituted for Dewar flasks. Temperature datawas computer logged using National Instruments LabView 7 con-

trol software and a thermocouple connected via a National Instru-ments Data Acquisition Card.

Compost maturity was determined using a phyto-toxicity assay(root elongation) adapted from Tiquia et al. (1996). Aqueous com-post extracts were prepared by mechanically shaking a fresh com-post sample with deionised water at 1:10 w/v for 10 min. Themixture was centrifuged at 1000 rpm for 10 min before the super-natant was filtered under vacuum through a Whatman #41 ashlessfilter paper. Ten seeds of Long Scarlet Radish (Fothergills, SouthWindsor, N.S.W., Australia) (AS 4454 – 2003) were incubated in a90 mm petri dish containing a Whatman Number #1 filter paperand 8 mL of extract. After 5 days incubation at 22 �C, in the dark,root length was determined and compared to a control grown indeionised water. Tests and controls were performed in triplicate.In preparation, seeds were soaked overnight in deionised waterat ambient conditions.

3. Results and discussion

3.1. Operation as a sequential aerobic–anaerobic–aerobic system

As an example of a full-scale hybrid aerobic/anaerobic OFMSWtreatment process, the aerobic–anaerobic–aerobic regime, as usedin the DiCOM� process, was carried out and monitored (Trial A).Significant biological oxidation commenced during the initial5 days of aerobic treatment where oxygen was consumed and eas-ily degradable organics, present within the waste, were minera-lised (Fig. 1). The heat generated brought the reactor tothermophilic conditions for the ensuing thermophilic anaerobicdigestion.

Anaerobic conditions initiated CH4 production with the onset ofmethanogenesis facilitated by the transfer of anaerobic inoculumin the liquor obtained from a laboratory-scale DiCOM� reactor.CH4 generation was noted within 4 h of liquor addition, with peakgeneration (32 L/day/kg VS) attained within 36 h and a total meth-ane production of 103 L/kg VS (270 Lbiogas/kg VS). These values arelower than many semi/continuous (450 Lbiogas/kg VS (DRANCO):Six and De Baere, 1992; 280 Lbiogas/kg VS (BTA): Kübler and Scher-tler, 1994; 630–710 Lbiogas/kg VS: Hartmann and Ahring, 2005a;640–790 Lbiogas/kg VS: Hartmann and Ahring, 2005b) and batch(550 Lbiogas/kg VS: Hartmann and Ahring, 2005b) anaerobic diges-tion systems reported in the literature but higher than other con-tinuous (230 Lbiogas/kg VS: Bolzonella et al., 2003) and batchsystems (50 L/kg VS: Forster-Carneiro et al., 2008). These datahighlight that the nature of the organic substrate and the method-ology applied have an important influence on the biodegradationprocess and the CH4 yield (Hartmann and Ahring, 2006; Forster-Carneiro et al., 2008).

Volatile fatty acid (VFA) accumulated (total VFA = 75 mM) dur-ing the initial 2 days of anaerobic treatment however minimal pHvariation occurred during this time (7.3–6.6). Process acidificationdid not occur, presumably due to the high buffer capacity, and therelease of ammonia. The strong correlation between the reductionin CH4 generation rate (4.5 L/day/kg VS) and residual VFA (Fig. 2A),during the final 24 h of digestion, indicated the exhaustion of easilydegradable organics within the OFMSW. After draining of the li-quor, the subsequent aerobic decomposition did not cause a signif-icant increase in temperature also indicating exhaustion of easilydegradable organics (Fig. 1).

3.2. Operation as static, in-vessel composting of OFMSW

In general, composting is considered to be a faster but less en-ergy efficient degradation system. To verify whether a purely aer-obic treatment provides benefits in terms of degradation rate or

3802 L. Walker et al. / Bioresource Technology 100 (2009) 3799–3807

product stability, static in-vessel composting was also carried out.Fig. 3 displays temperature and oxygen uptake rate (OUR) profilesfor a completely aerobic in-vessel composting trial (Trial D). A sec-ond composting trial was carried out under the temperature con-straints of the DiCOM� process (Trial E) to investigate whether

0

10

20

30

40

50

60

70

80

0 2 4 6 8 10Reactor Run T

Tem

pera

ture

(°C

)

Temperature

Fig. 1. Oxygen uptake rate (OUR) during static, in-vessel aeration phases and temperatuReactor temperature was maintained through aerobic microbial activity during aeration

0

20

40

60

80

0 2 4 6Reactor Run T

Con

cent

ratio

n (m

M)

Acetate Propionat

0

20

40

60

80

0 2 4 6Reactor Run Tim

Con

cent

ratio

n (m

M)

Acetate Propionate

0

20

40

60

80

0 2 4 6Reactor Run T

Con

cent

ratio

n (m

M)

Acetate Propionate

Trial A: DiCOM® 5 days Aerobic Pre

Trial B: 0 days Aerobic Pre-Treatme

Trial C: 1 day Aerobic Pre-Treatmen

Fig. 2. Effect of (A) 5, (B) 0 and (C) 1 days of static in-vessel aerobic pre-treatment on vOFMWS treatment.

the efficiencies of microbial degradation are significantly influ-enced by the different temperature regimes.

In order to directly compare the degradation rate under aerobicand anaerobic conditions the aerobic OUR and the anaerobic CH4

formation rate were converted to a molar electron flux defined as

12 14 16 18ime (days)

0

10

20

30

40

50

60

70

80

OU

R (m

mol

/h/k

g VS

)

OUR

re profile of a laboratory-scale DiCOM� reactor (Trial A) during OFMSW processing.and via an external heat exchanger during anaerobic processing.

8 10 12ime (days)

5

6

7

8

9

pH

e Butyrate pH

8 10 12e (days)

5

6

7

8

9

pH

Butyrate

8 10 12ime (days)

5

6

7

8

9

pH

Butyrate pH

-Treatment

nt

t

olatile fatty acid build-up during the subsequent anaerobic phase of thermophilic

L. Walker et al. / Bioresource Technology 100 (2009) 3799–3807 3803

the rate at which electrons are removed from the OFMSW. Consid-ering that an O2 molecule, can accept four electrons and that a CH4

molecule represents eight electron equivalents, the oxidation of or-ganic molecules into the final gaseous end-products, CO2 and CH4,can be directly compared via the flow of electrons from the solid(Fig. 4).

In terms of overall mineralisation rates the hybrid system (Di-COM�) showed a greater overall degradation than static aerobiccomposting, even though the switch to anaerobic conditions dur-ing the DiCOM� trial temporarily (for 1 day) slowed the degrada-tion rate as measured by electron flux. Interestingly, despite thereduced electron flow during days 6–12 of the fully compostedtrial, the product (based on electron flow) appeared to be as stableas the product from the hybrid process.

3.3. Effect of duration of the anaerobic treatment

Considering the above result, that during anaerobic treatmentthe rate of mineralisation of OFMSW did not slow down but wasenhanced, the effect of longer durations of anaerobic treatmentwas investigated by extending the anaerobic phase from 7 to 10(Trial C) and 12 days (Trial B) at the expense of the initial aerobicphase (Table 2).

0

10

20

30

40

50

60

70

80

0 2 4 6 8 10Reactor Run

Tem

pera

ture

(°C

)

Temperatur

Fig. 3. Temperature and oxygen uptake rate (OUR) profile

0

100

200

300

400

500

0 2 4 6 8

Reactor

Mol

ar E

lect

ron

Flo

w(m

mol

/h/k

g V

S)

Trial A

Fig. 4. Molar electron flow of OFMSW stabilised using the DiCOM� proces

During the anaerobic phases of all trials (A, B and C) acetateaccumulated initially, accompanied by H2 production (data notshown) and followed by propionate accumulation, which degradedby day 11 (Fig. 2). Butyrate accumulation, only observed in the trialwhere prior aerobic treatment was absent (Trial B), indicated avail-ability of easily fermentable material. Without adequate buffercapacity, and VFA degrading microbial communities, such abuild-up of butyrate is known to cause acidification and digesterfailure.

When comparing the overall microbial conversion as electronflow to either O2 or CH4, it can be shown that by reducing the initialaerobic phase a greater amount of electrons flow towards CH4,both in terms of total amount and maximum rates produced(Fig. 5). The low electron flow rate in the final aerobic phase ofall trials is in-line with completed anaerobic digestion of VFA(Fig. 2). The low level of VFA and electron flow towards the endof the purely anaerobic trial (Trial B) indicated methanogenic sub-strates were being depleted and that the treatment may be furthershortened.

The duration of the initial aerobic phase had a clear effect onoverall CH4 recovery, showing that the shorter it was the moreCH4 gas could be recovered (Table 3) which is in-line with the lit-erature (Krzystek et al., 2001).

12 14 16 18Time (days)

0

10

20

30

40

50

60

70

80

OU

R (m

mol

/h/k

g VS

)

e OUR

s of static in-vessel composting of OFMSW (Trial D).

10 12 14 16 18

Run Time (days)

Trial D Trial E

s (Trial A) and two static in-vessel composting trials (Trials D and E).

Table 3The effect of the length of aeration on the proportion of electron flow conserved asCH4 during anaerobic digestion of OFMSW. Electron flows represent treatment duringthe initial 12 days only.

Trial Aeration(days)

Total electronflow(mmol/kg VS)

Electron flowto CH4

(mmol/kg VS)

Electronflow toCH4 (%)

Thermophilic anaerobicdigestion (B)

0 56,400 56,400 100

Combined aerobic andthermophilicanaerobic digestion (C)

1 49,000 42,700 87

DiCOM� (A) 5 58,700 33,500 57Static in-vessel

composting (D)12 42,800 0 0

Thermophilic staticcomposting (E)

12 42,000 0 0

3804 L. Walker et al. / Bioresource Technology 100 (2009) 3799–3807

3.4. Compared to composting, anaerobic treatment enhancedhydrolysis rates

Hydrolysis has been reported in the literature as being the lim-iting factor in some anaerobic digestion processes (Pavlostathisand Giraldo-Gomez, 1991). In an effort to determine whetherhydrolysis is also a limiting factor encountered during static in-vessel composting, the likely carbon hydrolysis rates of both aero-bic and anaerobic processes were estimated. As no significant sol-uble COD had accumulated in the solid after aeration (data notshown), it can be assumed that, during microbial heat generation,hydrolysed substrates were completely oxidised to CO2 and re-leased from the reactor. Therefore, under the assumption thatany CO2 produced from non-soluble material originated from theoxidation of monomers that were hydrolysed from non-solublemacromolecules, during aeration the minimum hydrolysis rate isdirectly indicated by the CO2 production rate:

C� hydrolysis rate ¼ CO2 release rate

Under anaerobic conditions the hydrolysis rate can be recon-structed considering the CH4 generation and VFA accumulationrates and estimated to be:

C-hydrolysis rate ¼ VFA C-accumulation rate

þ CH4 production rateþ CO2 release rate

where VFA C-accumulation rate = (2 � acetate) + (3 � propionate) +(4 � butyrate) accumulation rates. (note: a negative VFA accumula-tion rate results when VFA consumption exceeds production).

The dropping respiration activity after 5 days of aerobic treat-ment indicates that easily degradable substrates were exhaustedand aerobic composting became hydrolysis limited (Figs. 5 and 6Trial D). The reduction in hydrolytic rate was unlikely to be dueto the available moisture as the moisture content of the OFMSW(at the conclusion of the fully aerobic trials) was found to be 60%and 62% (w/w) which is in optimum range for composting (50–70%: Richard et al., 2002). The continued decrease in C-hydrolysisrate over the following days of treatment indicated that the solidwas approaching stabilisation.

The described, early drop in carbon hydrolysis rate under aero-bic conditions could be overcome by switching the conditions toanaerobic and flooding the reactor with anaerobic liquor contain-ing an active inoculum (Fig. 6, Trials A and D). The improved hydro-lysis observed may be due to one or more of the following:

0

100

200

300

400

500

600

0 2 4 6 8 1Reactor Ru

Mol

ar E

lect

ron

Flow

(mm

ol/h

/kg

VS)

0 days Aerobic 1 Day Aerobic

Fig. 5. Aerobic and anaerobic molar electron flows, as measured by oxygen uptake rate (fof CH4 formed), for Trials B, C, A and D having 0, 1, 5 and 12 days of static in-vessel init

� The anaerobic liquid contained residual exo-enzymes fromprevious uses and increased the concentrations of activeenzymes.

� The enzymes generated during the composting phase werewater limited. In an aerobic composting system, the water avail-able for enzyme activity is found as a film on the surface of thesolid. It is also within this liquid film that water-soluble metab-olites accumulate, reducing the free-water available for enzy-matic activity. Flooding the reactor may provide the free-waternecessary for optimal activity.

� The enzymes had completely degraded the substrate in theirimmediate vicinity and the addition of liquid enabled contactwith fresh substrate (mass transfer improvement).

� Bacteria producing hydrolytic enzymes may have been intro-duced with the anaerobic inoculum.

A comparison of total electron flow (62,000; 51,300; 64,400 and46,400 mmol/kg VS for Trials B, C, A and D, respectively; Fig. 5) andC-hydrolysis (11,300; 10,000; 10,500 and 8000 mmol C/kg VS,respectively; Fig. 6) suggests that treatment containing an anaero-bic period, during which the solid waste was completely sub-merged, increased the degree of solid degradation in the waste.The waste in Trials A, B and C appeared to reach stability, indicatedby low molar electron flows (Fig. 5) and C-hydrolysis rates (Fig. 6),between days 10 and 12; yet each of these trials was exposed to adifferent length of aerobic treatment. It can therefore be deducedthat it is not the duration of the aerobic treatment that is critical

0 12 14 16 18 20n Time (Days)

5 Days Aerobic 12 Days Aerobic

our electrons per mol of O2 used) and CH4 production rate (eight electrons per moleial aerobic treatment.

L. Walker et al. / Bioresource Technology 100 (2009) 3799–3807 3805

in rapid waste stabilisation in hybrid systems but the presence ofan effective anaerobic component.

While both processes could potentially be further improved bycontrolling environmental conditions, there is an additional limita-tion on aerobic static in-vessel degradation when up-scaling theprocess. This limitation can be derived theoretically and has beentested practically (data not shown). Assuming that the heat gener-ated by respiration is 18 MJ/m3 O2 (Kaiser, 1996); negligible heatloss from a large, full-scale reactor; a cooling effect of dry air at20 �C of 36 MJ/m3 of air (leaving the reactor fully saturated withmoisture and at 60 �C); it can be estimated that, to avoid overheat-ing (>60 �C) of the reactor, a maximum of 28 L O2/m3 reactor/h canbe used for degradation. This would result in a maximum hydroly-sis rate of 19 mmol C/h/kg VS which is 3 and 2 times less than theaverage anaerobic and aerobic degradation rates observed in ourlaboratory trials (Fig. 6, Trials B and D: average hydrolysis ratesduring the initial 5 days for anaerobic Trial B: 62 mmol C/h/kg VSand aerobic Trial D: 40 mmol C/h/kg VS). While in industrial appli-cations the use of cooling systems can overcome this problem, it

0

20

40

60

80

100

120

0 2 4 6 8

Reactor Run

Hyd

roly

sis

Rat

e (m

mol

C/h

/kg

VS

)

0 Days Aerobic 1 Day Aerobic

Fig. 6. Aerobic and anaerobic hydrolysis rates, estimated from CO2 release, CH4 productiovessel initial aerobic treatment corresponding to Trials B, C, A and D.

0

50

100

150

200

250

300

350

0 1Length of Aeration

Ger

min

atio

n In

dex

Radish Cabb

Minimum Required forMaturity

Maximum Required for Stability

CB

Fig. 7. Effect of initial aeration phase duration on compost maturity indicators, self-heatCabbage for Trials B, C, A and D. *A GI > 85 indicates the disappearance of phyto-toxicitythe test when compared to the control (deionised water). Mathematical parameters (mea(I) Control: �x, r; Test: �x, r. (II) Control: �x, r: Test: �x, r where: (I) and (II) refer to CabbageTest: 123, 38. Trial B: (I) Control: 27, 10; Test: 25, 8. (II) Control: 76, 28: Test: 128, 52. TControl: 32, 11; Test: 30, 9. (II) Control: 47, 28: Test: 91, 49.

would add additional expense to the process which is avoided byswitching to anaerobic conditions.

3.5. Effect of treatment regimes on compost stability and maturity

Apparent product stability can be inferred from the final aerobicphase showing little electron flow for all treatments indicating thatadequate product stability had been obtained (Fig. 5). However,this assumption is not supported by data obtained from self-heat-ing tests (Fig. 7) which showed that static in-vessel composting didnot adequately stabilise the product, with the trial failing the Aus-tralian Standard (AS 4454 – 2003) self-heat test (Fig. 7). To meetthe requirements of this Standard, the compost must not self-heatto temperatures higher than 40 �C. The failure to meet this require-ment indicated that readily available substrate was still presentwithin the solid and the end-product was not yet stable, which isin-line with static aerobic composting providing the lowest totalamount of degradation as measured by cumulative electron flowand total C-hydrolysis. This lack of stability is in-line with the lit-

10 12 14 16 18

Time (Days)

5 Days Aerobic 12 Days Aerobic

n and intermittent VFA accumulation, for trials with 0, 1, 5 and 12 days of static in-

5 12 Phase (days)

0

10

20

30

40

50

60

Max

imum

Tem

pera

ture

Dur

ing

Hea

t Tes

t (ºC

)

age Heat Test

A D

ing test and germination index (GI) with Radish Long Scarlet and Chinese Flowering(Tiquia et al., 1996). In the above, a GI > 100 resulted from enhanced root growth inn (�x) and standard deviation (r)) for each trial are given in the form: Trial Number:and Radish, respectively. Trial A: (I) Control: 24, 7; Test: 28, 9. (II) Control: 41, 26:

rial C: (I) Control: 25, 9; Test: 29, 8. (II) Control: 135, 54: Test: 128, 48. Trial D: (I)

3806 L. Walker et al. / Bioresource Technology 100 (2009) 3799–3807

erature where OFMSW treated (3 months) in a static pile withforced aeration, when compared to a turned pile, and turned pilewith forced aeration, showed the lowest level of product stability(Ruggieri et al., 2008). Szanto et al. (2007) found a reduction(17%) in organic matter degradation when straw-rich pig manurewas composted (4 months) in a static pile as compared to piles thatwhere turned monthly. The reduction in product stability, and or-ganic matter degradation, in both of these studies was attributed tocompaction of the material resulting in a reduction in air perme-ability within the aggregates formed; a deduction that may alsobe applicable in this study.

It could be suggested that the lack of heat generation duringself-heating tests, for material treated anaerobically, is the resultof aerobic micro-organisms having been destroyed during anaero-bic treatment periods. However, an increase in electron flow (Fig. 5,Trials A, B and C)) and hydrolysis rate (Fig. 6, Trials A, B and C) atthe onset of post-digestion aeration indicated that aerobic micro-organisms had survived anaerobic treatment and were active whenexposed to oxygen.

Root elongation and germination index (Tiquia et al., 1996)have been found to be sensitive indicators of the phyto-toxicityand maturity of compost (Tiquia et al., 1996). Shortening the initialaerobic component of the first 12 days of processing had no appar-ent adverse effects on the germination index of the seed species se-lected (Fig. 7), suggesting that a longer anaerobic phase couldprovide a greater biogas yield without compromising the maturityof the final end-product.

4. Conclusions

Results from laboratory experiments indicated that:

� In the case of static in-vessel treatment of OFMSW, the inclusionof anaerobic steps may enhance the rate of degradation, in addi-tion to the generation of useful energy as CH4 gas. Possible rea-sons that may lead to the limitation of the aerobic processinclude:

o A decrease in porosity of the solid, leading to a limitationin oxygen availability.

o Limitation of water for hydrolysis processes.o Accumulation of soluble metabolites (e.g. ammonia and

VFA) under oxygen limiting conditions, in the smallamounts of water present.

� Static in-vessel aerobic composting may undergo hydrolysislimitation that can be overcome by flooding the reactor andshifting to anaerobic conditions.

� Fully anaerobic treatment, followed by a period of aerobic mat-uration, provided the most rapid processing.

� The DiCOM� process can be optimised by allowing a shorter aer-obic and a longer anaerobic phase thus speeding up the processand providing a greater CH4 yield.

Acknowledgements

The authors would like to acknowledge the Environmental Bio-technology Cooperative Research Centre, Australia (EBCRC) andAnaeCo Ltd. for the support of this project.

References

Bolzonella, D., Innocenti, L., Pavan, P., Traverso, P., Cecchi, F., 2003. Semi-drythermophilic anaerobic digestion of the organic fraction of municipal solidwaste: focusing on the start-up phase. Bioresour. Technol. 86, 123–129.

Braber, K., 1995. Anaerobic digestion of municipal solid waste: a modern wastedisposal option on the verge of breakthrough. Biomass Bioenerg. 9 (1–5), 365–376.

De Baere, L., 1999. Anaerobic digestion of solid waste: state-of-the-art. In: Mata-Alvarez, J., Tilche, A., Cecchi, F. (Eds.), Proceedings of the Second InternationalSymposium on Anaerobic Digestion of Solid Wastes, 15th–18th June, vol. 1.Barcelona, Spain, pp. 290–299.

De Baere, L., 2000. Anaerobic digestion of solid waste: state-of-the-art. Water Sci.Technol. 41 (3), 283–290.

DiCOM. <http://www.anaeco.com/index.php?option=com_content&task=view&id=15&Itemid=37> (accessed 10.01.2009).

Earle, J.F.K., Chynoweth, D.P., Nordstedt, R.A., 1995. Anaerobic bioconversion:biogasification of municipal solid waste. Florida Cooperative Extension Service,Institute of Food and Agricultural Sciences, University of Florida. Bulletin 267.

Edelmann, W., Engeli, H., 1993. Combined digestion and composting of organicindustrial and municipal wastes in Switzerland. Water Sci. Technol. 27 (2), 169–182.

Edelmann, W., Joss, A., Engeli, H., 1999. Two step anaerobic digestion of organicsolid wastes. In: Mata-Alvarez, J., Tilche, A., Cecchi, F. (Eds.), Proceedings of theSecond International Symposium on Anaerobic Digestion of Solid Wastes, 15th–18th June, vol. 1. Barcelona, Spain, pp. 153–160.

Edelmann, W., Schleiss, K., Joss, A., 2000. Ecological, energetic and economiccomparison of anaerobic digestion with different competing technologies totreat biogenic wastes. Water Sci. Technol. 41 (3), 263–273.

Edelmann, W., Baier, U., Engeli, H., 2005. Environmental aspects of the anaerobicdigestion of the organic fraction of municipal solid wastes. Water Sci. Technol.52 (1–2), 203–208.

Forster-Carneiro, T., Pérez, M., Romero, L.I., 2008. Thermophilic anaerobic digestionof source-sorted organic fraction of municipal solid waste. Bioresour. Technol.99, 6763–6770.

Fricke, K., Santen, H., Wallmann, R., 2005. Comparison of selected aerobic andanaerobic procedures for MSW treatment. Waste Manage. 25, 799–810.

Haight, M., 2005. Assessing the environmental burdens of anaerobic digestion incomparison to alternative options for managing the biodegradable fraction ofmunicipal solid wastes. Water Sci. Technol. 52 (1–2), 553–559.

Hartmann, H., Ahring, B.K., 2005a. Anaerobic digestion of the organic fraction ofmunicipal solid waste: influence of co-digestion with manure. Water Res. 39,1543–1552.

Hartmann, H., Ahring, B.K., 2005b. A novel process configuration for anaerobicdigestion of source-sorted household waste using hyper-thermophilic post-treatment. Biotechnol. Bioeng. 90 (7), 830–837.

Hartmann, H., Ahring, B.K., 2006. Strategies for the anaerobic digestion of theorganic fraction of municipal solid waste: an overview. Water Sci. Technol. 53(8), 7–22.

Kaiser, J., 1996. Modelling composting as a microbial ecosystem: a simulationapproach. Ecol. Model. 91, 25–37.

Kayhanian, M., Tchobanoglous, G., 1993. Innovative two-stage process for therecovery and compost from the organic fraction of municipal solid waste. WaterSci. Technol. 27 (2), 133–143.

Krzystek, L., Ledakowicz, S., Kahle, H., Kaczorek, K., 2001. Degradation of householdbiowaste in reactors. J. Biotechnol. 92 (2), 103–112.

Kübler, H., Schertler, C., 1994. Three-phase anaerobic digestion of organic wastes.Water Sci. Technol. 30 (12), 367–374.

Lopes, W.S., Leite, V.D., Prasad, S., 2004. Influence of inoculum on performance ofanaerobic reactors for treating municipal solid waste. Bioresour. Technol. 94,261–266.

Mason, I.G., Milke, M.W., 2005. Physical modelling of the composting environment:a review. Part 1: reactor systems. Waste Manage. 25, 481–500.

Mata-Alvarez, J., Cecchi, F., Pavan, P., Bassetti, A., 1993. Semi-dry thermophilicanaerobic digestion of fresh and pre-composted organic fraction ofmunicipal solid waste (MSW): digester performance. Water Sci. Technol.27 (2), 87–96.

Mata-Alvarez, J., Mace, S., Llabres, P., 2000. Anaerobic digestion of organic solidwastes. An overview of research achievements and perspectives. Bioresour.Technol. 74, 3–16.

Meissl, K., Smidt, E., 2007. High quality composts from anaerobic digestion residues.Biocycle 48 (3), 55–58.

Mohaibes, M., Heinonen-Tanski, H., 2004. Aerobic thermophilic treatment of farmslurry and food wastes. Bioresour. Technol. 95, 245–254.

Müller, H., Schmidt, O., Daxl, A., Bergmair, J., 2003. 3A-biogas; three stepfermentation of solid state biowaste for biogas production and sanitation. In:Pullammanappallil, P., McComb, A., Diaz, L.F., Bidlingmaier, W. (Eds.),Proceedings of the Fourth International Conference of ORBIT Association, 30thApril–2nd May, Perth, Australia. Part 1, pp. 327–332.

Neklyudov, A.D., Fedotov, G.N., Ivankin, A.N., 2006. Aerobic processing of organicwaste into composts. Appl. Biochem. Microbiol. 42 (4), 341–353.

Pavlostathis, S.G., Giraldo-Gomez, E., 1991. Kinetics of anaerobic treatment. WaterSci. Technol. 24 (8), 35–59.

Poggi-Varaldo, H.M., Gómez-Cisneros, E., Fernández-Villagómez, G., Esparza-García,F., Rinderknecht-Seijas, N., 1999. Aerobic post-composting of digestates fromanaerobic digestion of paper mill sludge and the organic fraction of municipalwastes. In: Mata-Alvarez, J., Tilche, A., Cecchi, F. (Eds.), Proceedings of theSecond International Symposium on Anaerobic Digestion of Solid Wastes, 15th–18th June, vol. 1. Barcelona, Spain, pp. 258–265.

Richard, T.L., Hamelers, H.V.M., Veeken, A., Silva, T., 2002. Moisture relationships incomposting processes. Compost Sci. Util. 10 (4), 286–302.

Ruggieri, L., Gea, T., Mompeó, M., Sayara, T., Sánchez, A., 2008. Performance ofdifferent systems for the composting of the source-selected organic fraction ofsolid waste. Biosyst. Eng. 101, 78–86.

L. Walker et al. / Bioresource Technology 100 (2009) 3799–3807 3807

Six, W., De Baere, L., 1992. Dry anaerobic conversion of municipal solidwaste by means of the DRANCO process. Water Sci. Technol. 25 (7),295–300.

Smet, E., Van Langenhove, H., De Bo, I., 1999. The emission of volatile compoundsduring the aerobic and the combined anaerobic/aerobic composting ofbiowaste. Atmos. Environ. 33, 1295–1303.

Szanto, G.L., Hamelers, H.V.M., Rulkens, W.H., Veeken, A.H.M., 2007. NH3, N2O andCH4 emissions during passively aerated composting of straw-rich pig manure.Bioresour. Technol. 98 (14), 2659–2670.

Teixeira, A.A., Chynoweth, D.P., Owens, J.M., Rich, E.R., Dedrick, A.L., Haley, P.J., 2004.Prototype space mission SEBAC biological solid waste management system. In:34th International Conference on Environmental Systems (ICES). Paper 2004-ICES-098. <http://www.test.abe.ufl.edu/chyn/download/Publications_DC/Refereed/2004-ICES_SEBAC2004.pdf> (24.10.07).

ten Brummeler, E., 2000. Full scale experience with the BIOCEL process. Water Sci.Technol. 41 (3), 299–304.

Tiquia, S.M., Tam, N.F.Y., Hodgkiss, I.J., 1996. Effects of composting on phytotoxicityof spent pig-manure sawdust litter. Environ. Pollut. 93 (3), 249–256.

VanderGheynst, J.S., Gossett, J.M., Walker, L.P., 1997. High-solids aerobicdecomposition: pilot-scale reactor development and experimentation. ProcessBiochem. 32 (5), 361–375.

Walker, L., Charles, W., Cord-Ruwisch, R., 2006a. Performance of a laboratory-scaleDiCOM� reactor – a novel hybrid aerobic/anaerobic municipal solid wastetreatment process. In: Kraft, E., Bidlingmaier, W., de Bertoldi, M., Diaz, L.F.,Barth, J. (Eds.), Proceedings of the International Conference ORBIT 2006Biological Waste Management: From Local to Global, 13th–15th September,Weimar, Germany. Part 3, pp. 317–326.

Walker, L., Charles, W., Cord-Ruwisch, R., 2006b. The effect of direct transfer ofanaerobic Inoculum on the performance of a laboratory-scale DiCOM� reactor.In: Biomass and Waste to Energy Symposium, 29th November–1st December,Venice, Italy.