Waste Management xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Waste Management

journal homepage: www.elsevier .com/locate /wasman

Application of Anaerobic Digestion Model No. 1 for simulating anaerobicmesophilic sludge digestion

http://dx.doi.org/10.1016/j.wasman.2014.10.0130956-053X/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail addresses: carllosmendez@gmail.com (C. Mendes), karlaesquerre@ufba.br

(K. Esquerre), lmqueiroz@ufba.br (L. Matos Queiroz).

Please cite this article in press as: Mendes, C., et al. Application of Anaerobic Digestion Model No. 1 for simulating anaerobic mesophilic sludge digWaste Management (2014), http://dx.doi.org/10.1016/j.wasman.2014.10.013

Carlos Mendes ⇑, Karla Esquerre, Luciano Matos QueirozFederal University of Bahia, Salvador-Ba, Brazil

a r t i c l e i n f o

Article history:Received 20 May 2014Accepted 13 October 2014Available online xxxx

Keywords:Anaerobic Digestion Model No. 1SimulationSewage sludgeHydraulic retention time

a b s t r a c t

Improving anaerobic digestion of sewage sludge by monitoring common indicators such as volatile fattyacids (VFAs), gas composition and pH is a suitable solution for better sludge management. Modeling is animportant tool to assess and to predict process performance. The present study focuses on the applicationof the Anaerobic Digestion Model No. 1 (ADM1) to simulate the dynamic behavior of a reactor fed withsewage sludge under mesophilic conditions. Parametric sensitivity analysis is used to select the most sen-sitive ADM1 parameters for estimation using a numerical procedure while other parameters are appliedwithout any modification to the original values presented in the ADM1 report. The results indicate thatthe ADM1 model after parameter estimation was able to predict the experimental results of effluent ace-tate, propionate, composites and biogas flows and pH with reasonable accuracy. The simulation of theeffect of organic shock loading clearly showed that an organic shock loading rate above of 35 kg/m3

day affects the performance of the reactor. The results demonstrate that simulations can be helpful tosupport decisions on predicting the anaerobic digestion process of sewage sludge.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction and controlling the anaerobic digestion processes. They improve

Sludge is generated as a residue from the treatment of munici-pal wastewater. It contains mainly water, organic matter, nutri-ents, living organisms and some non-biodegradable particles(Wawrzynczyk, 2007). Anaerobic digestion has been proven to beone of the most efficient process technologies for stabilizing sludgeand furthermore it produces methane as a form of fuel which mayreduce treatment cost (Wang et al., 2013). A mesophilic tempera-ture regime, 25–40 �C, can be more attractive than a thermophilictemperature regime, mainly because of the lower energy require-ments. Nevertheless, the heat required by thermophilic processescan be balanced with faster kinetics and increased methane gener-ation, resulting in a similar (or higher) net energy outcome.

Anaerobic digestion of sludge may be highly affected by the spe-cific characteristics of the feed substrate, operational parameterssuch as hydraulic retention time (HRT) and organic loading rate, andenvironmental factors such as pH, presence of inhibitory or toxicsubstances, temperature and reactor configuration. Volatile fattyacids (VFAs), gas composition and pH areuseful indicators for monitor-ing gradual changes in sludge treatment processes (Wang et al., 1997).

Modeling is an important tool for assessing aspects of bioprocess-es, process design, determination of optimal operating conditions

the understanding of the system, the formulation and validation ofhypothesis and are able to predict the system’s performance(Donoso-Bravo et al., 2011). Fundamental models describing anaer-obic digestion process have been developed in recent decades(Pavlostathis and Gossett, 1986; Siegrist et al., 1993; Kalyuzhnyi,1997; Angelidaki et al., 1999; Siegrist et al., 2002; Dalmau et al.,2010; Gavala et al., 2003; Mu et al., 2008; Tartakovsky et al., 2008;Lee et al., 2009; Mairet et al., 2011; Muha et al., 2012).

A specialist group working on anaerobic digestion for the Inter-national Water Association have been trying to condense all thisknowledge into a model known as Anaerobic Digestion Model N1(ADM1) which has been successfully applied to simulate the anaer-obic digestion of sewage sludge (Batstone et al., 2002; Blumensaatand Keller, 2005; Galí et al., 2009; Mairet et al., 2011). The aim ofthis research is to apply the ADM1 model to simulate the dynamicbehavior of a reactor fed with sludge and simulate the applicationof organic shock loads under mesophilic conditions.

2. Materials and methods

2.1. Experimental data

The investigation was carried out using a set of data derivedfrom experiments conducted by Siegrist et al. (2002). Two anaero-bic rectors in lab-scale (28 L liquid volume, 5 L headspace) were

estion.

Table 2Model inputs used for simulations.

Variable Component Value Unit

Ssu Monosaccharides 2.8 kg COD m�3

Saa Aminoacids 4.2 kg COD m�3

Sfa L.C. fatty acids 6.3 kg COD m�3

Sic Inorganic carbon 0.005 kmole C m�3

Sin Inorganic nitrogen 0.003571 kmole N m�3

Si Soluble inerts 0.7 kg COD m�3

Scat Cations 0.04 kmole m�3

San Anions 0.003571 kmole m�3

Xc Composites 10 kg COD m�3

Xi Particulate inerts 18 kg COD m�3

2 C. Mendes et al. / Waste Management xxx (2014) xxx–xxx

maintained at 35 �C and continuously fed with a mixture of pri-mary, secondary, and tertiary sludge from a municipal treatmentplant mixed with acetate and propionate. The reactor was operatedwith a hydraulic and solids retention time of 17.5 days. Thehydraulic and solids retention time was reduced to 8.7 on day 16and then to 5.3 on day 37. Table 1 shows the composition of thewastewater sludge described in the study of Siegrist et al. (2002).

2.2. Input characterization: COD fractioning

Fractionation of organic matter measured as total COD of thesewage sludge (46.7 kg m�3) was divided into 40% inert particulateCOD, 30% particulate degradable COD, and 30% hydrolysis productswhich were divided into 9% amino acids, 6% sugars, 13.5% LCFA,1.5% inert soluble COD. The inputs applied to the variables of theimplemented model are described in Table 2.

2.3. Implementation of ADM1 in Matlab/Simulink

ADM1 has been implemented in many software packagesWEST, GPS-X, SIMBA and Aquasim (Gernaey et al., 2004). In thepresent research, the set of ordinary differential equations ofADM1 was coded and implemented using Matlab/Simulink� andintegrated with the ODE15s solvers which solve stiff ODE systems.ADM1 is a highly complex model and its implementation requiressimultaneously solving 32 differential equations in the liquidphase and 3 differential equations in the gas phase. There are 24equations based on biochemical processes in the liquid phase,two for cation and anion states and 6 for acid–base pairs.

The Matlab/Simulink was chosen because of its flexibility forfurther structural modifications (ADM1 implement extensions) toenable integration of the model with other blocks containing otherelements (e.g., controllers) and to monitor the evolution of eachvariable in the course of a simulation in real time throughout a out-put block (display). When a simulation is complete, its outputmatrix can be exported to a spreadsheet.

2.4. Sensitivity analysis and parameter estimation

A parametric sensitivity analysis was performed to define themost sensitive ADM1 parameters during the production and con-sumption of acetate, propionate and biogas flow. The parametersevaluated were hydrolysis constant rates for disintegration of com-posites, carbohydrates, proteins and lipids (kdis; khyd;ch, khyd;pr andkhyd;li, respectively), Monod maximum specific substrate uptakerate ðkmÞ, half-saturation constant (KS) and yield of biomassproduction ðYÞ. The km;KS and Y values underwent a sensitivityanalyses for each of the seven different microbial communities.Although Jeong et al. (2005) mentioned that yield of product onsubstrate ðf Þ values showed high sensitivities to components ofthe ADM1 analyzed, in this study these parameters were negligibledue to their small variations. Lee et al. (2009) also reported thatwith regard to sensitivity, f is negligible. According to Jeong et al.(2005) the methane concentration was set as the focused variable

Table 1Composition of the wastewater sludge (Siegrist et al., 2002).

Parameter Unit Values

Total COD kg m�3 40Soluble COD kg m�3 4.5Acetate kg m�3 1.5Propionate kg m�3 0.7NH4–N kg m�3 0.4TKN kg m�3 1.6pH – 6.8

Please cite this article in press as: Mendes, C., et al. Application of Anaerobic DWaste Management (2014), http://dx.doi.org/10.1016/j.wasman.2014.10.013

to measure sensitivity methane production and the sensitivityindex (SI) is defined as follows:

SI ¼PjCSTDðtÞ � CSENSðtÞj

N; ð1Þ

where CSTDðtÞ and CSENSðtÞ are the simulation results with the sug-gested parameter values and the parameters with a relative changeof target parameter ðm3 d�1Þ, respectively, for each given time (t),and N is the number of data. The ranges tested were defined asthose suggested by Batstone et al. (2002), varying within 30%,100% and 300% depending on the parameter. Whereas the originalparameter has a value of 100%, the generated values range between:70% and 130% in condition 1 (varying within 30%), 0% and 200% incondition 2 (varying within 100%), and �200% to 400% in condition3 (varying within 300%). Since the values of 0% and �200% in con-ditions 2 and 3 are not physically acceptable, we used a minimumvalue in these cases, 10%.

The ADM1 parameters selected by the sensitivity analysis wereestimated by minimizing the difference between the measure-ments and model outputs. Measurable process variables includedflow biogas, pH, propionate, and acetate concentrations. Parame-ters were optimized only with measured experiment data of themethane concentration. The following objective function ðFobjÞwas used for parameter estimation:

FobjðHÞ ¼minXn

t¼1

ðyexpðtÞ � ysimðt;HÞÞ2 ð2Þ

yexp are the collected measurements, ysim are the model predictedoutputs, H represents the parameters to be determined and n isthe number of measurements. Minimizing the objective functionis an important issue for prediction purposes or process stability(Batstone et al., 2003). Optimization was made with the fminsearchfunction in the Matlab� toolset.

Fig. 1 describes the method used to fit the ADM1 model to theset of experimental results. In the first step, the sensitivity analysiswas performed using the set of initial values of all the modelparameters. Then simulations were undertaken to fit the modeloutputs to the experimental results data by changing ADM1parameters selected by the sensitivity analysis until the bestparameters values are found. The other parameters showed lowsensitivity in the model outputs and therefore they were usedwithout any modification.

2.5. Effect of organic shock loads on the stability of the anaerobicreactor

Model simulations were used to investigate the effect of shockloads on the composition of biogas, pH and VFA production undercontinuous conditions over 100 days. The liquid and gas volume ofthe reactor were the same as those used in the parameter estima-tion. The composition of the simulated influent was same as that

igestion Model No. 1 for simulating anaerobic mesophilic sludge digestion.

Fig. 1. Estimation procedure to identify most sensitive parameters.

C. Mendes et al. / Waste Management xxx (2014) xxx–xxx 3

used in the parameter estimation, see Table 2 with a flow rate of1.5 m3 day�1. The temperature and the HRT of the system wereassumed to be 35 �C and 18 days, respectively.

3. Results and discussion

3.1. Sensitivity analysis and parameter estimation

The results of the sensitive analysis are shown in Fig. 2. The sen-sitivities of the hydrolysis constants, kdis;Y , and KS displayed signif-icant range variation and asymmetric characteristics, as shown inFig. 2d.

High sensitivity values were also observed for km;ac , km;h2; km;su

and km;aa at ranges below (Fig. 2b). Low values of km;ac caused accu-mulation of acetic acid, lowering the pH and inhibiting methano-genesis, which resulted in greater sensitivity. Higher hydrogenconcentrations ðkm;h2Þ are quite important as they may inhibit LCFAand organic acid degradation. The same is true for km;fa whichshowed moderate sensitivity. KS and Y presented very low sensitiv-ities (Fig. 2d) with values below 1.5 m3 d�1. Jeong et al. (2005) alsoreported that, with regard to sensitivity, Y and KS are not so impor-tant, except for KS;ac sensitivity to acetate and methane concentra-tions. Based on the sensitivity results, we decided to use km;ac; km;h2,km;su and km;aa for parameter estimation.

Table 3 shows the initial values of the kinetic parameters andtheir respective optimized values. Maximum specific substrateuptake rate of acetate ðkm;acÞ, hydrogen ðkm;h2Þ, monosaccharides

Please cite this article in press as: Mendes, C., et al. Application of Anaerobic DiWaste Management (2014), http://dx.doi.org/10.1016/j.wasman.2014.10.013

ðkm;suÞ and amino acids ðkm;aaÞ, were considered as having the great-est impact on experimental data and they were estimated in com-parison with the experimental data.

Fig. 3 shows a comparison of simulation and experimentalresults for anaerobic digestion of acetate and propionate. We canconclude that the concentrations of acetate and propionate werewell predicted by the ADM1 model. Moreover, the simulationresults with optimized parameters showed good agreement withthe experimental data. Furthermore, acetate and propionate failureat a HRT of 5.3 days was well justified by the ADM1 model but itwas predicted with some deviations between the simulationresults without optimization and measurements. These discrepan-cies between the simulations and measurements observed may bedue to the correlation of some sensitive parameters such as hydro-lysis constants with feed concentration and HRT as mentioned byGavala et al. (2003) and to some potential weaknesses in theADM1 model as addressed by many researchers such as Parker(2005), Kleerebezem and van Loosdrecht (2006) and Batstone (2006).

The model calibration results for effluent biogas flow and com-position, pH, using the optimized parameter set are presented inFigs. 4 and 5. Effluent pH value, biogas flow and methane, car-bon-dioxide and hydrogen composition were predicted with highaccuracy for all HRTs.

As can be seen in Figs. 3 and 4, from day 11 there was anincrease in the concentrations of acetate, propionate and biogasflow. This was due to a reduction in the hydraulic retention time(HRT). In this case, the reduction in HRT did not cause process

gestion Model No. 1 for simulating anaerobic mesophilic sludge digestion.

(a) (b)

(c) (d)

Fig. 2. Sensitivity indices (SI) calculated for different ranges of (a) hydrolysis constant rates; (b) Monod maximum specific uptake rates; (c) half-saturation constants; and (d)yields of biomass on substrate.

Table 3Initial and estimated values of kinetic parameters.

Kinetic parameters Units Initial value Estimated value

km;ac day�1 8.00 13.80km;h2 day�1 35.00 26.01km;su day�1 30.00 20.22km;aa day�1 50.00 41.12

4 C. Mendes et al. / Waste Management xxx (2014) xxx–xxx

inhibition and the methane concentration in the biogas was notsignificantly affected. The pH value remains within the range(6.5–8.5) recommended in the literature for anaerobic bacteria(the methanogens) (Kalyuzhnyi, 1997).

The performance of the microbial community growing in ananaerobic digestion system is determined primarily by biologicalsolids retention time (SRT). For a CSTR system without recycling,SRT equals HRT, as is the case in this study. The reactions of theanaerobic digestion process are directly related to the SRT. Thereis a minimum SRT for each reaction. If the SRT is less than the min-imum SRT, the microorganisms cannot grow rapidly enough andthe digestion process will fail eventually. Metcalf and Eddy(2007) reported the typical minimum value and design of SRT as4 and 10 days, respectively, when the reactor is operated in meso-philic conditions. After the simulation, we therefore concluded thatthe optimum HRT for such a system is in the range of 6–9 days.

3.2. Effect of organic shock loads on the stability of the anaerobicreactor

The results of model simulation for a organic shock loadingare shown in Fig. 6. The organic loading rate increased from 5 to

Please cite this article in press as: Mendes, C., et al. Application of Anaerobic DWaste Management (2014), http://dx.doi.org/10.1016/j.wasman.2014.10.013

35 kg/m3 day on day 30, then to 70 kg/m3 day on day 60. Theorganic loading rate was chosen to be above the value range min-imum suggested in the literature (2–32 kg/m3 day) (Metcalf andEddy, 2007).

Model simulations revealed that the accumulation of carbondioxide results in an increase in the dissolved carbonic acid con-centration, thereby increasing the production of H+. The accumula-tion of hydrogen results in a greater production of propionate andbutyrate whose degradation is inhibited by the presence of a highconcentration of acetate (Fig. 6c). Although ADM1 reported a lowvalue of KS for hydrogenotrophic methanogenic Archaea comparedto other organisms of the microbial consortium, the results inFig. 6b show that these microorganisms reduce carbon dioxidebelow the maximum working capacity. According to Giraldo-Gomez et al. (1992) this occurs because the concentration of dis-solved hydrogen is available and generally low due to mass trans-fer limitation.

Inhibition of the syntrophic bacteria by the accumulation of theproducts formed by fermentative bacteria will lead to a continuedaccumulation of VFA which contributes to the consumption ofalkalinity and a pH decrease (Fig. 6a). The pH decrease affects thegrowth of methanogenic microorganisms and consequentlyreduces the composition of methane in the biogas, as shown inFig. 6b.

4. Conclusions

The ADM1 model is a powerful tool for predicting the dynamicbehavior of anaerobic digesters treating sewage sludge under

igestion Model No. 1 for simulating anaerobic mesophilic sludge digestion.

Fig. 3. Comparison between simulations and experimental data with and without optimization: (a) acetate and (b) propionate.

Fig. 4. Comparison between simulations and experimental data after parameter estimation: (a) biogas production and (b) biogas composition.

C. Mendes et al. / Waste Management xxx (2014) xxx–xxx 5

mesophilic conditions. This study evaluates the application of theAnaerobic Digestion Model No. 1 (ADM1) developed by the IWAtask group for mathematical modeling of anaerobic processes.

A sensitivity analysis showed that among all the kinetic param-eters, the Monod maximum specific substrate uptake rate affectsbiogas production significantly, which was related to all the pro-cesses used in the model. Important parameters includingkm;ac; km;h2; km;su and km;aa were estimated using iterative methods,which optimized the parameters with experimental results. With

Please cite this article in press as: Mendes, C., et al. Application of Anaerobic DiWaste Management (2014), http://dx.doi.org/10.1016/j.wasman.2014.10.013

regard to the model output for acetate and propionate concentra-tions, the simulation results showed good agreement with theexperimental data after parameter optimization. Despite in prac-tice for CSTR sludge digestion, values for the SRTs range from 10to 20 days are reasonable considering that the optimum HRT forsuch a system is in the range of 6–9 days where there was a greaterproduction of methane in the reactor.

The simulation results obtained to evaluate the effect of organicshock loads on the anaerobic reactor using the ADM1 clearly show

gestion Model No. 1 for simulating anaerobic mesophilic sludge digestion.

Fig. 5. Comparison between simulations and experimental data of the pH after parameter estimation.

Fig. 6. Model simulations for a continuous flow system with shock loading of5 kg/m3 day experiencing a step increase in organic loading rate to 35 kg/m3 day onday 30 and to 70 kg/m3 day on day 60. (a): pH and (b) Biogas composite (c) AGVs(18 days HRT).

6 C. Mendes et al. / Waste Management xxx (2014) xxx–xxx

that an organic loading rate above 35 kg/m3 day causes processinstability and even failure when fermentation processes exceedmethanogenesis, leading to the accumulation of VFAs, falling pH,and a reduction in methane production.

Acknowledgements

We would like to thank the Coordination for the Improvementof Higher Education Personnel – CAPES, for its financial support forthis research. Special thanks goes to the lecturer Márcio Martins forour fruitful discussions about modeling aspects.

Please cite this article in press as: Mendes, C., et al. Application of Anaerobic DWaste Management (2014), http://dx.doi.org/10.1016/j.wasman.2014.10.013

References

Angelidaki, I., Ellegaard, L., Ahring, B.K., 1999. A comprehensive model of anaerobicbioconversion of complex substrates to biogas. Biotechnol. Bioeng. 63, 363–372.

Batstone, D.J., 2006. Mathematical modelling of anaerobic reactors treatingdomestic wastewater: rational criteria for model use. Rev. Environ. Sci. BioTechnol. 5, 57–71.

Batstone, D.J., Keller, J., Angelidaki, I., Kalyuzhny, S., Pavlostathis, S., Rozzi, A.,Sanders, W., Siegrist, H., Vavilin, V., 2002. Anaerobic digestion model No.1.Scientific and Technical Report, vol. 13. IWA Publishing, London.

Batstone, D.J., Pind, P.F., Angelidaki, I., 2003. Kinetics of thermophilic, anaerobicoxidation of straight and branched chain butyrate and valerate. Biotechnol.Bioeng. 84, 195–204.

Blumensaat, F., Keller, J., 2005. Modelling of two-stage anaerobic digestion using theIWA anaerobic digestion model no. 1 (adm1). Water Res. 39, 171–183.

Dalmau, J., Comas, J., Rodríguez-Roda, I., Pagilla, K., Steyer, J.P., 2010. Modeldevelopment and simulation for predicting risk of foaming in anaerobicdigestion systems. Bioresour. Technol. 101, 4306–4314.

Donoso-Bravo, A., Mailier, J., Martin, C., Rodríguez, J., Aceves-Lara, C.A., Wouwer,A.V., 2011. Model selection, identification and validation in anaerobic digestion:a review. Water Res. 45, 5347–5364.

Galí, A., Benabdallah, T., Astals, S., Mata-Alvarez, J., 2009. Modified version of adm1model for agro-waste application. Bioresour. Technol. 100, 2783–2790.

Gavala, H.N., Angelidaki, I., Ahring, B.K., 2003. Kinetics and modeling of anaerobicdigestion process. Advan. Biochem. Eng. Biotechnol. 81, 57–93.

Gernaey, K.V., van Loosdrecht, M.C., Henze, M., Lind, M., Jorgensen, S.B., 2004.Activated sludge wastewater treatment plant modelling and simulation: stateof the art. Environ. Model. Softw. 19, 763–783 (Environmental Sciences andArtificial Intelligence).

Giraldo-Gomez, E., Goodwin, S., Switzenbaum, M.S., 1992. Influence of masstransfer limitations on determination of the half saturation constant forhydrogen uptake in a mixed-culture ch(4)-producing enrichment. Biotechnol.Bioeng. 40, 768–776.

Jeong, H.S., Suh, C.W., Lim, J.L., Lee, S.H., Shin, H.S., 2005. Analysis and application ofadm1 for anaerobic methane production. Bioproc. Biosyst. Eng. 27, 81–89.

Kalyuzhnyi, S., 1997. Batch anaerobic digestion of glucose and its mathematicalmodeling. II. Description, verification and application of model. Bioresour.Technol. 59, 249–258.

Kleerebezem, R., van Loosdrecht, M., 2006. Critical analysis of some conceptsproposed in adm1. Water Sci. Technol. 54, 51–57.

Lee, M.Y., Suh, C.W., Ahn, Y.T., Shin, H.S., 2009. Variation of adm1 by usingtemperature-phased anaerobic digestion (tpad) operation. Bioresour. Technol.100, 2816–2822.

Mairet, F., Bernard, O., Ras, M., Lardon, L., Steyer, J.P., 2011. Modeling anaerobicdigestion of microalgae using adm1. Bioresour. Technol. 102, 6823–6829.

Metcalf, Eddy, 2007. Wastewater Engineering: Treatment Disposal Reuse. NewYork.

Mu, S., Zeng, Y., Wu, P., Lou, S., Tartakovsky, B., 2008. Anaerobic digestion model no.1-based distributed parameter model of an anaerobic reactor: I. Modeldevelopment. Bioresour. Technol. 99, 3665–3675.

Muha, I., Grillo, A., Heisig, M., Schönberg, M., Linke, B., Wittum, G., 2012.Mathematical modeling of process liquid flow and acetoclasticmethanogenesis under mesophilic conditions in a two-phase biogas reactor.Bioresour. Technol. 106, 1–9.

Parker, W.J., 2005. Application of the adm1 model to advanced anaerobic digestion.Bioresour. Technol. 96, 1832–1842.

Pavlostathis, S.G., Gossett, J.M., 1986. A kinetic model for anaerobic digestion ofbiological sludge. Biotechnol. Bioeng. 28, 1519–1530.

igestion Model No. 1 for simulating anaerobic mesophilic sludge digestion.

C. Mendes et al. / Waste Management xxx (2014) xxx–xxx 7

Siegrist, H., Vogt, D., Garcia-Heras, J.L., Gujer, W., 2002. Mathematical model formeso- and thermophilic anaerobic sewage sludge digestion. Environ. Sci.Technol. 36, 1113–1123.

Siegrist, H., Renggli, D., Gujer, W., 1993. Mathematical modelling ofanaerobic mesophilic sewage sludge treatment. Water Sci. Technol. 27,25–36.

Tartakovsky, B., Mu, S., Zeng, Y., Lou, S., Guiot, S., Wu, P., 2008. Anaerobic digestionmodel no. 1-based distributed parameter model of an anaerobic reactor: II.Model validation. Bioresour. Technol. 99, 3676–3684.

Please cite this article in press as: Mendes, C., et al. Application of Anaerobic DiWaste Management (2014), http://dx.doi.org/10.1016/j.wasman.2014.10.013

Wang, L., Aziz, T.N., de los Reyes, F.L., 2013. Determining the limits of anaerobic co-digestion of thickened waste activated sludge with grease interceptor waste.Water Res. 47, 3835–3844.

Wang, Q., Noguchi, C., Kuninobu, M., Hara, Y., Kakimoto, K., Ogawa, H., Kato, Y.,1997. Influence of hydraulic retention tine on anaerobic digestion of pretreatedsludge. Bioresour. Technol. 11, 105–108.

Wawrzynczyk, J., 2007. Enzymatic Treatment of Wastewater Sludge,Sludge Solubilisation, Improvement of Anaerobic Digestion and Extractionof Extra Cellular Polymeric Substances, Ph.D. thesis. Lund University.

gestion Model No. 1 for simulating anaerobic mesophilic sludge digestion.