Bioresource Technology 142 (2013) 655–662

Contents lists available at SciVerse ScienceDirect

Bioresource Technology

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

Mesophilic anaerobic co-digestion of sewage sludge and a lixiviationof sugar beet pulp: Optimisation of the semi-continuous process

0960-8524/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.biortech.2013.05.017

⇑ Corresponding author. Tel.: +34 956016158; fax: +34 956016411.E-mail addresses: rocio.montanes@uca.es (R. Montañés), montserrat.perez

@uca.es (M. Pérez), rosario.solera@uca.es (R. Solera).

Rocio Montañés, Montserrat Pérez ⇑, Rosario SoleraDepartment of Environmental Technologies, University of Cádiz, Campus Puerto Real, 11510 Puerto Real, Cádiz, Spain

h i g h l i g h t s

� Lixiviation of sugar beet pulp used as co-substrate.� Improved efficiency of anaerobic digestion of sewage sludge.� Improved biogas production and organic matter removal.� Low solid retention times in the system.

a r t i c l e i n f o

Article history:Received 26 February 2013Received in revised form 26 April 2013Accepted 6 May 2013Available online 22 May 2013

Keywords:Anaerobic co-digestionContinuously stirred tank reactorsMethane yieldSewage sludgeSugar beet pulp

a b s t r a c t

This study examine the effect of an increased organic loading rate on the efficiency of the stirred tankreactor treating sewage sludge and sugar beet pellets and to report on its steady-state performance.The digester was subjected to a program of steady-state operation over a range of hydraulic retentiontimes (HRTs) of 30 to 6 days and organic loading rates (OLRs) of up to 1.7 kgCOD/m3 d to evaluate itstreatment capacity.

The COD removal efficiency was found to be 84.23% COD in the digester when treating mixture sewagesludge/lixiviation of sugar beet pulp at 1.27 kgCOD/m3 d (10-days SRT). The volumetric methane levelproduced in the digester reached 0.7 m3CH4/m3 d and the methane yield was 0.64 m3CH4/kgCODremoval.

Therefore, anaerobic co-digestion of sewage sludge and lixiviation of sugar beet pulp improve the bio-gas productivity and the organic matter removal in addition to lowering solids retention times in thesystem.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Sludge production in Spain has grown constantly over the lastfew years. The increasing production of sludge from householdand/or urban wastewater treatment is leading to problems arisingfrom its correct management, treatment and, above all, elimina-tion. With the aim of solving these problems, the National WaterQuality Plan (2007–2015) is contemplating a raft of measures toensure full compliance with Directive 91/271/EEC (1991). This Plananticipates the construction of new wastewater treatment plantsand the correct exploitation, maintenance and management ofexisting installations.

The National Sewage Sludge Plan established 2007 as the dead-line for recovering 80% of wastewater treatment plant (WWTP)sludge and reducing the sludge sent to landfill by 20%. The increasein WWTP sludge production due to application of both Directive

91/271/EEC (1991) and the National Sewage Sludge Plan bringswith it the need to manage this sludge correctly. One of the solu-tions to achieve environmentally friendly sludge management isto use this sludge as agricultural fertiliser.

Excessive sludge generation is the main problem facingWWTPs. Approximately 0.5–2% of the treated water in a WWTPbecomes sludge that needs to be treated before final disposal inthe environment. Furthermore, production of the beet crop has ri-sen in recent years, justified by its high energy productivity (3.5–4.5 t bioethanol/Ha). This crop generates high quantities of waste,such as sugar beet pulp, which has to be treated. In this regard,the anaerobic co-digestion of organic waste is a successfulmethodology for the joint treatment of waste from differentsources and therefore, suitable for treating the waste producedby WWTPs.

The joint digestion of sewage sludge and vegetable waste fromenergy crops has the advantages of shared treatment facilities,combined management methodologies, a reduction in the invest-ment and operating costs, and a decrease in the temporal varia-tions of the composition and production of each waste residue.

656 R. Montañés et al. / Bioresource Technology 142 (2013) 655–662

Furthermore, co-digestion of organic waste from different originshas been proven to be as successful as both mesophilic and ther-mophilic regimens. Hence, these approaches can be applied tothe co-digestion of these wastes (Gómez at al., 2006; Demireland Scherer, 2008).

Most renewable energies, and in particular, agricultural bio-mass waste, are energy sources scattered around a given country.The agro-industrial biomass waste generated is concentrated andits management in the same places from where it originates re-quires sufficient quantities of waste for energy recovery to be tech-nically and economically viable. Therefore, large plants are neededto centralise collection from urban WWTPs, which are widespreadat the country and regional levels.

Co-digestion is the simultaneous anaerobic digestion of a mix-ture of two or more substrates. This technology is an attractive op-tion to improve the yields of the anaerobic digestion of wastes dueto the positive synergisms established in the digestion medium, afact that increases the economic viability of the biogas plants(Mata-Alvarez et al., 2000). The main advantage of this technol-ogy-based system is an improved methane yield created by thesupply of additional nutrients to the mixture. Moreover, co-diges-tion could lead to the following benefits (Alatriste-Mondragónet al., 2006; Mata-Alvarez et al., 2000): (1) dilution of inhibitoryand/or toxic compounds; (2) increase in the organic content insidethe digester, eliciting better utilisation of the digester volume; (3)enhancement of the digestate stabilisation; (4) procurement of therequired moisture contents in the digester feed, with an easierhandling of blended wastes; (5) large reduction of the emissionof greenhouse gases into the atmosphere; and (6) economic advan-tages from the sharing of equipment and costs. However, somedrawbacks exist as well: (1) the high cost of waste transfer fromthe co-substrate generation point to the anaerobic plant; (2) therisk of spreading poisonous substances from the industrial or mu-nicipal waste; and (3) the harmonisation of the different policiesregarding the waste generators. What is more, co-digestionchanges digestion behaviour and the quality of the digestate. Fur-thermore, the addition of unknown co-substrates should be pre-vented. To improve the co-digestion process and detect theamounts of inhibitory or toxic compounds, which can lead to a pro-cess breakdown or reduced methane production, it is necessary tocarry out several laboratory experiments such as biodegradabilitytests and/or laboratory-scale digester assessments.

Pretreatment with sugar dried pellets yields a homogeneous li-quid effluent with high organic load, suitable for use in batch typeco-digestion assays. When mixed with sewage sludge, it producesa final effluent with good characteristics for anaerobic co-digestion.

The feasibility of co-digestion of two or more organicwaste streams (e.g., organic fraction from municipal solid waste,sewage sludge or biosolids, animal waste, and agricultural solidwaste, among others) has been demonstrated at both the labora-tory-scale (Poggi-Varaldo and Oleszkiewicz, 1992; Stenstromet al., 1983) and the full-scale level (Angelidaki and Ahring,1994; Cecchi et al., 1988).

Table 1Main characteristics of the test conditions.

TS (kg TS/m3) VS (kg VS/m3) COD (kg/m3) OLR (kg CO

30-days SRT 33 22 9.9 0.520-days SRT 28 19 10.1 0.6715-days SRT 25 18 8.5 0.8510-days SRT 25.2 17 10.1 1.36-days SRT 25.3 22 10.2 1.7

This study aimed to examine the effect of an increased organicloading rate on the efficiency of the stirred tank reactor treatingsewage sludge and vegetable waste (sugar beet pellets) and to re-port on its steady-state performance. The digester was subjectedto a program of steady-state operation over a range of hydraulicretention times (HRTs) of –six to 30 days and organic loadingrates (OLRs) of up to 1.7 kgCOD/m3�d to evaluate its treatmentcapacity. Five different experimental conditions corresponding tothe solid retention times (SRTs) of 30, 20, 15, 10 and 6 days weretested.

2. Methods

2.1. Experimental equipment

In this study, a semi-continuous laboratory-scale stirred tankreactor, operating at the mesophilic range, was used. The equip-ment consisted of a reactor with a stainless steel vessel that wasagitated and heated, and with a total volume of 6 L and a workingvolume of 5 L. The digester featured a lid that allowed it to besealed to maintain anaerobic conditions within the digester. Thestainless steel lid had several openings (for the output of biogas,insertion of a pH probe, insertion of a temperature probe, two in-puts to correct the pH balance, a power input and an agitation sys-tem). The bottom of the digester had a release valve used forsampling the material inside the digester, which was made possi-ble by the sealing system between the vessel and the cap. Theassembly included an agitator (operating at 17 revolutions perminute) that homogenised the waste using stainless steel bladescrapers. To maintain the operating temperature, the digesterwas heated by recirculating water through a thermostatic jacket.Biogas was collected in 10-L Tedlar bags, and a special syringewas used for sampling gases.

2.2. Operational conditions

The digester was initially loaded with a mixture of inoculumand substrate, resulting in a final concentration of 20% w/w of inoc-ulum, which is considered optimum for biogas production. Theinoculum came from a full-scale mesophilic digester for the treat-ment of waste sludge from a WWTP. Once the inoculum was mixedwith the substrate, a mixture of sewage sludge and the lixiviationof sugar beet pulp, the system remained unfed for a period of oneweek to acclimatise the inoculum to the waste at the selected tem-perature (35 �C).

The digester was fed a mixture of sewage sludge and sugar beetpulp, diluted in water to give a final concentration of 10% w/w.Based on the information found in the literature and previousexperience of our group (Fernández et al., 2012), SRTs of 6, 10,15, 20 and 30 days were selected for study until process break-down. Each condition was maintained for an operational periodlasting three times the duration of the HRT to ensure that stea-dy-state conditions were reached.

D/m3digester d) OLR (kg VS/m3digester�d) F:M (kg COD/kg VS d)

1.1�(10�3) 0.0251.2�(10�3) 0.0341.8�(10�3) 0.0432.1�(10�3) 0.0735�(10�3) 0.085

Fig. 1. Pretreatment of sugar beet pulp pellets.

R. Montañés et al. / Bioresource Technology 142 (2013) 655–662 657

2.3. Waste characterisation

To characterise the waste and inoculum, as well as monitor theeffluent from the process, the following parameters were deter-mined: pH, volatile fatty acids (VFAs), chemical oxygen demand(COD), total solids (TS) and volatile solids (VS). These analyses wereconducted in accordance with standardised methods (APHA,AWWA, WPCF, 1992) adapted for waste with low solid content(Álvarez Gallego, 2005) and based on the previous leaching of thewaste in an aqueous medium. Biogas volume and composition weredetermined using chromatographic methods (Appels et al., 2008).

Table 1 shows the characteristics of the substrate that was usedin terms of the organic matter content. Pellets were subjected tobiological pretreatment before the co-digestion process in orderto promote hydrolysis and solubilisation of the organic matterand therefore, improve anaerobic digestion in the generation ofbiogas and possible final residue agronomic valorisation.

Pretreatment was divided into different steps, as can be ob-served in Fig. 1.

The F:M ratio has been used as a parameter for treatment per-formance evaluation of AFB reactor (Pérez et al., 1997). This term iscommonly used in practice as a design and control parameter andclosely related to both the specific utilization rate, and the processefficiency is known as the food-to-microorganism (F:M) ratio. TheF:M ratio is defined as follows:F : M ¼ QðS0=XvÞ

where Q is the volumetric rate feed (L/d), S0 is the feed concen-tration (kgCOD/m3

digester, Xv the suspended biomass (Shied et al.,1981) in terms of kg VS/m3. The F:M ratio could be used as aparameter for treatment performance evaluation of SCTR. Thisequation implies that the observed COD removal in the digesteris attributable to suspended cells. F:M ratios observed in this inves-tigation were between 0.025 and 0.085 kgCOD/kgVS d.

Later, mesophilic inoculum with 2.55% of TS was added, untilthe desired conditions were achieved. Table 1 shows the OLR, TS,VS, COD and F:M ratio of feeding into the digester at each stageof the study, which is expressed in terms of COD and VS. The char-acteristics of the inoculum used in the start-up process are the fol-lowing: values of pH, density and alkalinity of 7.38, 1.02 kg/L and2.48 kg CaCO3/m3 respectively. In terms of organic matter theinoculum has low percentages of volatiles and total solids 1.8and 2.55% respectively and a amount of COD of 1.17 kg/m3.

3. Results and discussion

This section discusses the evolution of the main variables dur-ing the semi-continuous mesophilic anaerobic digestion process,

Table 2Percentage of COD removal under the conditions studied.

COD influent (kg/m3) COD effluent (kg/m3) % COD removal

30-days SRT 9.9 2.21 78.120-days SRT 10.1 2.5 75.015-days

SRT8.5 2 76.9

10-days SRT 10.1 1.6 84.26-days SRT 10.2 5.7 43.6

such as pH, total acidity, dissolved organic carbon (COD), and bio-gas production and composition. Discussions are based on thecomparison of the system performance for five different SRT condi-tions (and consequently different OLR) tested in the mesophilicanaerobic co-digestion of a mixture of sewage sludge and lixivia-tion of sugar beet pulp. These conditions were: 30-days SRT, 20-days SRT, 15-days SRT, 10-days SRT and 6-days SRT (correspondingto a range of OLR 0.5–1.7 kg COD/m3

digester d, respectively). In thefigures used in the following discussion, vertical lines were in-cluded to indicate the different SRTs in the system.

3.1. Physicochemical parameters

3.1.1. pHThe primary objective of this study was to investigate and

determine the optimum operational (SRT and OLR) conditions dur-ing long-term fermentation of a mixture of sewage sludge and su-gar beet lixiviation into methane, without any adjustment of thepH.pH is a fundamental parameter regulating anaerobic degrada-tion (Appels et al., 2008). Fig. 2 shows the changes in pH duringthe semi-continuous mesophilic experiments conducted in thisstudy. The digester was run without adjustment of the pH insidethe reactor. The pH of the effluent ranged from 7.41 to 7.61, exceptfor 6-days SRT, when the pH decreased because of the high level ofVFA. These values indicated that the digester was working withinthe optimum range for methane formation and was not underinhibitory conditions.

For 30-days SRT, the system regulated itself at a pH of 7.61. For20-days SRT, the pH stabilised at approximately 7.49, the optimumpH for the activity of methanogenic microorganisms. Regarding the15-days SRT, an initial decrease in pH was observed during the firstdays until 7.17, as a result of an increasing OLR. Subsequently, thepH stabilised at 7.41, without the addition of an external agent. For10-days and 6-days SRT, there was also an initial decrease in pHuntil 6.7 and 6.1, respectively, because of the high OLR. After afew days, the pH stabilised at 7.21 for the 10-days SRT condition,but not for 6-days SRT. The sharp decline in pH in the first daysof each SRT condition was related to the destabilisation of the sys-tem as a consequence of the increase in the OLR.

When the added organic load is increased, the acidogenicmicroorganisms respond quickly, given their high specific growthrate, and generate more VFA. However, methanogenic Archaea(with lower growth rates) require more time to reach the popula-tion size necessary to degrade the excessive VFA. During this pro-cess, the pH decreases as a result of the accumulation of VFAs inthe digester (Fernández et al., 2012).

The decrease in pH is more important when working with high-er OLRs, as the imbalance between the activities of the differentmicroorganism groups is more pronounced. Therefore, the initialdecrease in pH in the transition from 20- to 15-days SRT was espe-cially high.

3.1.2. Alkalinity and VFAThe VFAs produced during anaerobic digestion tend to reduce

the pH. This reduction is normally counteracted by the activity ofthe methanogens, which also produce alkalinity in the form of

Fig. 2. pH temporal evolution in the tests.

658 R. Montañés et al. / Bioresource Technology 142 (2013) 655–662

carbon dioxide, ammonia and bicarbonate. The system pH is con-trolled by the CO2 concentration in the gas phase and the HCO3-alkalinity of the liquid phase. If the CO2 concentration in the gasphase remains constant, the addition of HCO3-alkalinity will in-crease the digester pH (Turovskiy and Mathai, 2006).

Fig. 3 shows the relationship between total acidity and alkalin-ity and the evolution of individual volatile fatty acids. It is clear

(b)

(a)

Fig. 3. Temporal evolution of: (a) total acidity (mgAcH/L) against alkalinity (

that the digester was operating with a good buffering capacity,indicated by the low VFA values of less than 200 mg AcH/L com-pared to those in an alkalinity of 2000–3600 mg CaCO3/L (exceptin 6-days SRT). The fluctuation in VFA levels and alkalinity overtime did not affect methanogenic activity as the methane concen-tration in the gas did not drop, demonstrating that the acetogensand methanogens could cope with the fluctuations in the VFA

mgCaCO3/L) and (b) individual volatile acid (as mg/L) levels in the tests.

R. Montañés et al. / Bioresource Technology 142 (2013) 655–662 659

and alkalinity in the digester. This indicates that the conditionswere stable and that the possibility of methanogen inhibitionwas low.

The total acidity representing the total amount of VFAs, ex-pressed as acetic acid, exhibited stable daily values in the effluentfrom the mesophilic digester. For 30-days SRT, the value of totalacidity was 67.5. When the SRT changed to 20 days, a decrease inthe average of total acidity was observed, with a value of46.3 mg AcH/L. Total acidity remained almost constant for the15-days SRT at 53 mg AcH/L. Finally, for the 10-days SRT, VFA con-centration increased to 174 mg AcH/L, due to the higher organicloading rate of the system. For 6-days SRT, the value of4007 mg AcH/L indicated destabilisation of the reactor.

High concentrations of volatile fatty acids tend to occur fre-quently in the effluent from anaerobic digesters due to organiccauses such as overloading of the digester, entrance of toxic com-pounds and changes in the temperature. As shown in Fig. 3a, an in-creased VFA content matched with the start of each SRT, explainingthat the increase in the ORL led to the initial accumulation of VFAs,which were subsequently consumed by the microorganisms. Thistrend illustrates the initial destabilisation caused by the reductionin the SRT, as discussed above.

Acetate has been described as the least toxic fatty acid (Iannottiet al., 1978), while an increase in propionate concentration hasbeen shown to be associated with system failure (Hobson andShaw, 1976). Propionate is even more inhibitory than butyrate.Methanogenic populations are inhibited by a propionate concen-tration of about 3000 mg/L (Riau et al., 2010). In this study, propi-onic acid values did not exceed this value and therefore, did notaffect the methanogenic populations.

However, at the end of the 30-, 20- and 15-days SRT, the aver-age acidity values were close to 20 mg AcH/L (Fig. 3b). The changeto 20- and 15-days SRT led to an accumulation of the acids duringthe first days. At the end of this stage, the level of VFA decreasedconsiderably. Therefore, these SRTs seem to be times commitmentto the stability of the system. SRTs of 30, 20 and 15 days gave themost stable conditions, although they did not elicit the highest per-centage of COD removal. A low pH stimulates acidogenic activity(VFA-production) and makes methanogenic activity difficult (VFAconsumption). This could explain the observed VFA evolution forthe 6-days SRT.

During the whole operating time of the digester, alkalinity re-mained constant at around 3000 mg CaCO3/L, except for 6-daysSRT because of the reduction in pH. As can be observed, the totalacidity/alkalinity ratio (Fig. 4) showed an average value of 0.012.The balance is essential for a proper operation and optimal degra-dation of VFA. The ratio of total acidity/alkalinity of between 0.0and 0.1 is desirable for a strong system. A ratio of less than 0.5

Fig. 4. Evolution in the ratio of

inhibits the formation of methanogens. Ratios between 0.1 and0.5 will produce imbalances in the operational process (de la RubiaRomero, 2003). In general, both VFA and alkalinity concentrationsshould be between 50 and 300 mg/l and 1500 and 3000 mg/l,respectively. Moreover, alkalinity should be between 2500 and5000 to avoid a drop in the pH value.

3.1.3. Chemical oxygen demand (COD)Fig. 5 shows the COD evolution in the mesophilic digester for

different SRTs. The start-up of the process (for 30-days SRT)showed good adaptation of the inoculum to the waste. This wasachieved immediately and later, the trend displayed lower varia-tions. For the 30-, 20- and 15-days SRTs, COD reached approxi-mately the same values in the stable phase of the operation. Forthe 10-days SRT, COD reached a small value. However, at thebeginning of the new ORL, the COD concentration reached a peak,before decreasing in the stable phase. The changes in the COD inthe different tests suggested that the digester needed to adapt tothe new conditions of increased organic loading.

The best results for COD concentration, in terms of the qualityof the effluent, were obtained with a SRT of 10 days. Thus, as canbe seen in Table 2, the COD removal percentage was the highestfor the 10-days SRT.

Based on the analysis of this parameter, we confirmed that forthe 6-days SRT, the digester was not degrading organic matterand could not cope with the OLR , consequently leading to a dropin pH in the system and an accumulation of VFAs, as seen in theprevious sections.

The influence of the retention time on breakdown efficiency hasbeen mostly studied at the laboratory scale (STORA, 1985). The ob-tained relationship between gas production and retention time in a(semi-)CSTR indicates that (i) retention times shorter than 5 daysare insufficient for stable digestion; VFA concentrations increasedue to a washout of methanogens, (ii) VFA concentrations are stillrelatively high for an SRT of 5–8 days and hence, there is an incom-plete breakdown of compounds, especially of lipids, (iii) stabledigestion is obtained after 8–10 days when VFA concentrationsare low, initiating the breakdown of lipids, and (iv) the breakdowncurve stabilises for a SRT of 10 days, when all sludge compoundsare significantly reduced (Chen et al., 2008). Therefore, the SRT isa fundamental design and operating parameter for all anaerobicprocesses.

Overall, the purifying efficiency remained approximately con-stant for 30-, 20- and 15-days SRTs, with an average value of 75%removal of the COD. For 10-days SRT, we obtained a high % COD re-moval and subsequently, higher productivity, as discussed in thenext section.

total acidity to alkalinity.

Fig. 5. Evolution of organic matter (as kgO2/m3) in the tests.

Table 3Productivities under the conditions studied.

30-days SRT 20-days SRT 15-days SRT 10-days SRT 6-days SRT

m3CH4/d 1.15�(10�3) 0.62�(10�3) 1.43�(10�3) 3.42�(10�3) 1.53�(10�3)m3CH4/kg CODremoval 0.50 0.23 0.44 0.64 0.41% CH4 68.90 70.40 68.50 68.80 66.60m3/m3

digester d 0.23 0.13 0.29 0.68 0.31

660 R. Montañés et al. / Bioresource Technology 142 (2013) 655–662

Fig. 6 relate to the dissolved organic carbon with different oper-ating parameters as the F:M ratio, different retention times (SRT)and methane production. Fig. 6a presents observed COD removaldata and m3 CH4/m3

digesterd as a function of F:M ratio. The percent-age of COD removal is not affected by the ratio F:M in the firstthree times tested and has the highest value for the F:M ratio of0.074 kgCOD/kgVS d, which matches the SRT which is more meth-ane generated. Methane production activity as a function of F:Mratio is shown in the same Figure. As would be expected, the volu-metric production rate of methane increased with the F:M ratiountil it reached a maximum for 0.074 kgCOD/kgVS d.

Fig. 6b and c confirm a stable performance system for the firstthree retention times, in relation with the production of methaneand the % COD removal. 10-days SRT is the optimal conditionsfor this test. Also we can say that for 6-days SRT, methane produc-tion falls although the amount of organic matter added is the high-est. So in these conditions the system works with good efficiency.

3.2. Productivity

Methane generation shows the same trend as that of the elim-ination of organic matter. Increased biogas productivity was de-tected for the 10-days SRT, with an average of 0.68 m3 CH4/m3

digester d, while for 30-, 20-, 15 and 6-days SRTs, low productiv-ities were obtained (Table 3). However, the percentage of methanein the biogas stream was similar for all SRTs (66–70%).

Fig. 7 shows the evolution of methane production, expressed asm3/m3

digester d. An increased production of biogas was detected forthe 15- and 10-days SRTs, with an average of 0.29 m3/m3

digester dand 0.68 m3/m3

digester d, respectively. The 20-days SRT resulted inthe lowest average, 0.13 m3/m3

digester d, during the stable phase.This trend was similar to that for the total biogas produced becausemethane represents approximately 66–70% of the biogas. Therewas an initial drop in methane production at the transition from30- to 20-days SRT, 20- to 15-days SRT and 15- to 10-days SRT. Thisevolution coincided with the increased production of acids andwas associated with the peak in soluble organic matter causedby the increase in the organic load of the system.

After a short adaptation period, there was a linear increase inmethane production with time until the stable stage. Thus, a com-parison of the rates of methane production for each SRT tested(represented by the slopes of the linear trends shown in Fig. 7for the stable stage of each condition) can also provide relevantinformation about the process. Table 3 shows the average methaneproduction for each of the tested conditions in m3CH4/d andm3CH4/m3

digester�d.For the 30-days SRT, there was no initial period of acclimatisa-

tion to the mesophilic inoculum. Hence, methane production be-gan with an average rate of production of 0.22 m3CH4/m3

digester d,similar to the methane production at the stable stage.

For the 20-days SRT, the rate was 0.13 m3CH4/m3digester d in the

third time of the SRT. The change to the 20-days SRT was a stableperiod when a rate of 0.33 m3CH4/m3

digester d was reached. Thelowest methane productivity, 0.23 m3CH4/kgCODremoval, wasachieved for the 20-days SRT, followed by the 6-days SRT, with amethane productivity of 0.41 m3CH4/kgCODremoval.

The highest production of methane per gram of COD removal inthe system was obtained with the 10-days SRT (0.64 m3CH4/kgCODremoval), while for the 30- and 15-days SRTs, the values were20% and 30% lower than the value obtained for the 10-days SRT,respectively.

The production of methane varied from 0.7 to 4.4 L CH4/d forco-digestion of primary sludge and fruit and vegetable. Otherauthors have reported methane yields between 0.21 and0.52 m3 CH4/m3

digester d for various types of waste (Gelegeniset al., 2007; Astals et al., 2012).

Comparing mesophilic anaerobic digestion of sewage sludge withthe anaerobic co-digestion of sewage sludge and sugar beet pulp lix-iviation, it can be said that the latter demonstrated improved pro-ductivity and organic matter removal in the system, as well asreaching lower operational SRT under stable conditions. In the liter-ature, a maximumof 0.29 m3CH4/kg CODremoval for a 15-days SRT and72.5% of CODremoval for a 40-days SRT (de la Rubia Romero et al.,2006), and a productivity of 0.30 m3CH4/m3

digester d for a 20-daysSRT test and 0.18 m3biogas/m3

digester d for 21-days SRT (Forster-Car-neiro et al., 2010; Bolzonella et al., 2005) have been described.

(a)

(b)

(c)

Fig. 6. (a) Effect of F:M ratio on methane production activity (m3/m3digester d) and organic removal efficiency (as percentage of initial COD); (b) organic removal efficiency (as

percentage of initial COD) and OLR influent (kgCOD/m3digester d) as a function of SRT; (c) Methane production activity (m3/m3

digester d) and OLR removal (kgCOD/m3digester d) as

a function of SRT.

Fig. 7. Temporal evolution of methane yield (m3 CH4/m3digester d) in the tests.

R. Montañés et al. / Bioresource Technology 142 (2013) 655–662 661

4. Conclusions

According to the results, the best operating conditions for themesophilic anaerobic co-digestion of sewage sludge and sugarbeep pulp lixiviation were achieved with a solid retention timeof 10 days. Moreover, a higher level of methane production was

reached in this condition with respect to the organic loading rateremoval in the system, 0.64 m3 CH4/g CODremoval, compared tothe 0.50, 0.23, 0.44 and 0.41 m3 CH4/g CODremoval for the 30-, 20-,15- and 6-days SRTs, respectively. Also, a 10-days SRT attainedthe highest removal of organic matter, 84.23% COD removal, incontrast to the 78.14%, 75.02%, 76.92% and 43.65% observed for

662 R. Montañés et al. / Bioresource Technology 142 (2013) 655–662

30-, 20-, 15- and 6-days SRTs, respectively. However, the levels ofVFA were high compared with the other SRTs.

Furthermore, the 6-days SRT was not good for anaerobic co-digestion because it could not cope with the ORL, leading to thesystem becoming unstable.

Our results indicate that the waste used in this test can be suc-cessfully converted using anaerobic co-digestion. Further investi-gation into the tech-no-economics is required to make thisprocess economical and scalable.

Acknowledgements

The authors wish to express their gratitude to Junta de And-alucía, especifically to Proyecto de Excelencia with reference P09-TEP-5275, called ‘‘Codigestión anaerobia de lodos de depuradoray residuos de cultivos vegetales energéticos. Estrategias para mej-orar producción de biogas y la valorización agronómica del residuofinal’’.

References

Alatriste-Mondragón, F., Samar, P., Cox, H.H., Ahring, B.K., Iranpour, R., 2006.Anaerobic co-digestion. Water Environ. Res. 78, 607–636.

Álvarez Gallego, CJ., 2005. Testing Different Procedures for the Start Up of a DryAnaerobic Co-digestion Process of OFMSW and Sewage Sludge at ThermophilicRange, University of Cádiz, Doctoral Thesis.

American Public Health Association, American Water Works Association, WaterPollution Control Federation, 1992. Standard Methods for the Examination ofWater and Wastewater, 18th ed. Washington DC, USA.

Angelidaki, I., Ahring, B.K., 1994. Anaerobic thermophilic digestion of manure atdifferent ammonia loads: effect of temperature. Water Res. 28, 727–731.

Appels, L., Baeyens, J., Degreve, J., Dewil, R., 2008. Principles and potential of theanaerobic digestion of waste-activated sludge. Prog. Energy Combust. 34, 755–781.

Astals, S., Nolla-Ardevol, V., Mata-Alvarez, J., 2012. Anaerobic codigestion of pigmanure and crude glycerol of mesophilic conditions. Biogas and digestate.Bioresour. Technol. 110, 63–70.

Bolzonella, D., Pavan, P., Battistoni, P., Cecchi, F., 2005. Mesophilic anaerobicdigestion of waste activated sludge: influence of solid retention time in thewaste water treatment process. Process Biochem. 40, 1453–1460.

Cecchi, F., Traverso, P.G., Perin, G., Vallini, G., 1988. Comparison of codigestionperformance of two differently collected organic fractions of municipal solidwaste with sewage sludges. Environ. Technol. Lett. 9, 391–400.

Chen, Y., Cheng, J.J., Creamer, K., 2008. Inhibition of anaerobic digestion process: areview. Bioresour. Technol. 99, 4044–4064.

de la Rubia Romero, M.A., Pérez, M., Romero, L.I., Sales, D., 2006. Effect of solidsretention time (SRT) on pilot anaerobic thermophilic sludge digestion. ProcessBiochem. 41, 79–86.

de la Rubia Romero, M.A., 2003. Puesta en marcha y optimización de la degradaciónanaerobia termofílica de lodos de EDAR, Tesis doctoral, Universidad de Cádiz.

Demirel, B., Scherer, P., 2008. Production of methane from sugar beet silage withoutmanure addition by a single-stage anaerobic digestion process. BiomassBioenerg. 32 (3), 203–209. http://dx.doi.org/10.1016/j.biombioe.2007.09.011.

Directive 91/271/EEC of 21 May 1991 concerning urban waste-water treatment.

Fernández, J., Pérez, M., Romero, L.I., 2012. Mesophilic anaerobic digestion of theorganic fraction of municipal solid waste: optimisation of the semicontinuousprocess. Chem. Eng. 193–194, 10–15.

Forster-Carneiro, T., Riau, V., Pérez, M., 2010. Mesophilic anaerobic digestion ofsewage sludge to obtain class B biosolids: microbiological methodsdevelopment. Biomass Bioenerg. 34, 1805–1812.

Gelegenis, J., Georgakakis, D., Angelideki, I., Chistopoulaut, N., Goumenaki, M., 2007.Optimization of biogas production from olive-oil mill wastewater bycodigesting with diluted poultry manure. Appl. Energy 84, 646–663.

Gómez, X., Cuetos, M.J., Cara, J., Moron, A., Garcia, A.I., 2006. Anaerobic codigestionof primary sludge and the fruit and vegetable fraction of the municipal solidwaste. Conditions for mixing and evaluation of the organic loading rate.Renewable Energy 31, 2017–2024.

Hobson, P.N., Shaw, B.G., 1976. Inhibition of methane production byMethanobacterium formicicum. Water Res. 10 (10), 849–852.

Iannotti, E.L., Fischer, J.R., Sievers, D.M., 1978. Medium for the enumeration andisolation of bacteria from a swine waste digester. Appl. Environ. Microbiol. 36,555–566.

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.

National Water Quality Plan 2007–2015. http://www.magrama.gob.es/.Pérez, M., Romero, L.I., Sales, D., 1997. Thermophilic degradation of distillery

wastewater in continuous-flow fluidized bed bioreactors. Biotechnol. Prog. 13,33–38.

Poggi-Varaldo, H.M., Oleszkiewicz, J.A., 1992. Anaerobic co-composting ofmunicipal solid waste and waste sludge at high total solid levels. Environ.Technol. 13, 409–421.

Riau, V., De la Rubia, M.A., Pérez, M., 2010. Temperature-phased anaerobic digestion(TPAD) to obtain class A biosolids: a semi-continuous study. Bioresour. Technol.101, 2706–2712.

Shied, W.T., Sutton, P.M., Kos, P., 1981. Predicting reactor biomass concentration in afluidized-bed system. J. Water Pollut. Control Fed. 53, 1574–1584.

Stenstrom, M., Ng, A.S., Bhunia, P.K., Abramson, S.D., 1983. Anaerobic digestion ofmunicipal solid waste. ASCE J. Environ. Eng. 109 (5), 1148–1158.

STORA, Stichting Toegepast Onderzoek Reiniging Afvalwater, 1985. Optimalisatievan de gistingsgasproductie [in Dutch].

Turovskiy, I.S., Mathai, P.K., 2006. Wastewater Sludge Processing. WileyInterscience. John Wiley & Sons, New York.

Glossary and notations

CSTRs: continuously stirred tank reactorsDOC: dissolved organic carbon, expressed as mg/LCOD: chemical oxygen demand expressed as mg O2/LWWTP: urban wastewater treatment plantOLR: fed organic loading rate, expressed as kg COD/m3digester/d or kg VS/m3-

digester/d6-days SRT: 6 days of solid retention time in the digester10-days SRT: 10 days of solid retention time in the digester15-days SRT: 15 days of solid retention time in the digester20-days SRT: 20 days of solid retention time in the digester30-days SRT: 30 days of solid retention time in the digesterSRT: solid retention time, expressed as daysHRT: hydraulic retention timeSSTR: semi-continuously stirred tank digesterTS: total solids, expressed as %VS: volatile solids, expressed as %