Investigation of kinetics of anaerobic digestion of Canary grass ...

Chemical Papers 66 (6) 550–555 (2012) DOI: 10.2478/s11696-012-0136-4

ORIGINAL PAPER

Investigation of kinetics of anaerobic digestion of Canary grass‡ Anna Kacprzak*, Liliana Krzystek, Katarzyna Pa´ zdzior, Stanis
law Ledakowicz Department of Bioprocess Engineering, Faculty of Process and Environmental Engineering, Technical University of Lodz, Wolczanska 213, 90-924 Lodz, Poland Retrieved 31 May 2011; Revised 21 November 2011; Accepted 26 November 2011

The most promising technology to generate energy is the production of biogas from agricultural feedstock. This technology allows obtaining high biogas yields per hectare. Biogas of high methane content can be produced from various kinds of energy crops, animal manure, and other organic waste. This paper presents results of both biochemical methane potential (BMP) assay of Canary grass (Phalaris canariensis) and its anaerobic co-digestion with cheese whey and glycerin fraction in a large laboratory scale as well as the kinetics of both co-digestion and digestion of Canary grass alone. The suggested mathematical description of the anaerobic co-digestion process gave satisfying conformity with the experimental data. c 2012 Institute of Chemistry, Slovak Academy of Sciences
Keywords: biochemical methane potential, Canary grass, biogas

Introduction An outstanding advantage of the conversion of biomass and various organic wastes to renewable energy is the reduction of carbon dioxide emissions and the resulting environment protection (Demirel & Scherer, 2009). During the last few decades, anaerobic digestion of organic matter has been presented as a suitable technology for the treatment of organic wastes and the production of biogas (Fernández et al., 2005). In anaerobic digestion, co-digestion is the term used to describe combined treatment of several wastes with complementary characteristics, which is one of the main advantages of the anaerobic technology. There is abundant literature on the utilization of co-digestion, such as co-digestion of organic fraction of municipal solid wastes (OFMSW) and agricultural residues, organic solids wastes or more specific wastes (Fernández et al., 2005). Energy crops, i.e. plants grown specifically for the purpose of producing energy, are a carbon-neutral source of domestic renewable energy. The most important parameter in choosing crops for methane pro-

duction is the net energy per hectare, which is defined mainly by the biomass yield and convertibility of the biomass to methane (Lehtomäki et al., 2008). The biochemical methane potential (BMP) is a simple method to measure anaerobic digestibility of a given substrate. It allows comparing potential biogas production of different substrates; moreover, it is inexpensive and provides repeatable results. The benefits of using biochemical methane potential assays are as follows: 1) determination of the concentration of organics in wastewater that can be anaerobically converted to CH4 , 2) evaluation of potential efficiency of the anaerobic process, 3) measurement of the residual organic material amenable to further anaerobic treatment, and 4) testing for non-biodegradables remaining after treatment (Moody et al., 2009). Literature related to BMP assays for agricultural residues proves that this method has been widely implemented. Labatut and Scott (2008) used BMP to check which available food residues could be co-digested with dairy manure and at what ratio the substrates should be mixed to improve the economic efficiency of the on-farm digester. Similarly, the BMP assay was used by Lovanh et al.

*Corresponding author, e-mail: [email protected] ‡ Presented at 38th International Conference of the Slovak Society of Chemical Engineering, Tatranské Matliare, 23–27 May 2011.

A. Kacprzak et al./Chemical Papers 66 (6) 550–555 (2012)

(2008) to verify how swine manure with poultry litter improved methane production rates. Kirk and Bickert (2004) used BMPs to examine manure slurry from multiple points in a dairy manure treatment system in order to determine the optimal location of the digester within the treatment system for maximum gas production and pathogen reduction (Moody et al., 2009). Angelidaki et al. (2009) presented some guidelines for biomethane potential assays and advices to researchers involved in BMP analysis in order to make an attempt to unify the use of units and techniques in future studies. One of the most important parameters determining the correct course of the methane fermentation process is the carbon to nitrogen ratio. Suitable C : N ratio is necessary to achieve high efficiency of biogas production. It is due to the fact that the microorganisms in the fermentation process utilize carbon 25–30 times faster than nitrogen (Santos et al., 2004). Addition of a considerable source of easily degradable carbon improves the nutrient balance and what is more, co-digestion results in an increase of biogas yields and of the quality of fertilizer (Kacprzak et al., 2010; Sosnowski et al., 2008). The most important advantages of co-digestion include dilution of toxic substances contained in the digested medium, fairly high degree of digestibility, higher production of biogas as well as increased amounts of biodegradable organic matter (Sosnowski et al., 2003). In literature, many examples of successfully conducted co-fermentation processes of different substrates have been presented so far (Sosnowski et al., 2008). Anaerobic co-digestion of OFMSW and fats of animal and vegetable origin was examined by Fernández et al. (2005). Lehtomäki et al. (2007) studied co-digestion of energy crops and crop residues with cow manure. Bouallagui et al. (2009) attempted the improvement of fruit and vegetable waste anaerobic digestion performance and stability with co-substrate addition. Although the knowledge on anaerobic co-fermentation widens, further research is needed in order to determine the impact of various combinations of substrates on the stability of the process (Murto et al., 2004; Gelegenis et al., 2007). This is why the presented investigations were aimed not only at the extension of the anaerobic co-digestion potential, but also at the optimization of the process. Kinetic models are reported to be a useful tool to optimize co-digestion processes (Gavala et al., 2003). A model can be defined as a set of relationships between the variables of interest in the system being investigated, which can be expressed in the form of equations (Sosnowski et al., 2008; Kayode Coker, 2001). This paper presents results of both the BMP assay of Canary grass (Phalaris canariensis) and its co-digestion with cheese whey and glycerin fraction in a large laboratory scale as well as the mathematical modeling of such quasi-continuous

551

anaerobic co-digestion and batch anaerobic digestion processes.

Experimental BMP assays of Canary grass from the first and second harvest were determined in batch experiments in 1 L glass bottles (liquid volume of 0.5 L) incubated statically at (37 ± 1) ◦C. The crop material was used in the form of silage. The initial loading rates of volatile solids (VS) for Canary grass I and II were: 5.09 kg m−3 and 4.96 kg m−3 , respectively. Inoculum (sludge after anaerobic digestion) and substrate were added into the bottles, and distilled water was refilled to produce the liquid volume of 0.5 L. For the Canary grass from the first harvest, inoculum from anaerobic chambers of a municipal wastewater treatment plant was used; while for the second harvest, inoculum from a working anaerobic digester operating in the quasi-continuous mode was used. The contents of the bottles were flushed with N2 –CO2 gas for 5 min and the bottles were then sealed with rubber stoppers. Daily methane production from each digester was measured by the water displacement method. Biogas production is given in norm litres (NL) per kg of VS (L kg−1 ). The content of methane and carbon dioxide was determined using a gas analyzer LMS GAS DATA (GFM 410, OMC ENVAG, Poland). The bottles were shaken at 80 min−1 . The reactors ran until no further methane production could be detected. The process of co-fermentation of Canary grass and cheese whey together with waste glycerin fraction was carried out in a 25 L bioreactor operated mesophically in the quasi-continuous mode. Inoculum (sludge after anaerobic digestion) came from anaerobic digestion chambers of the Municipal Wastewater Treatment Plant in Lodz, Poland. The mixture of substrates (Canary grass, whey, glycerin fraction) in specified weight ratio was fed to the digester once a day, after withdrawing the same amount of the fermentation broth. Characteristics of all substrates are shown in Table 1. Every three days, the ratio was increased until the highest yield of biogas and methane content were obtained. When the steady state was reached, samples from the bioreactor were taken every 1 h during day and night (24 h period between feeding) and they were analyzed. Based on these experimental data, a kinetic model was proposed. Mixed samples drawn from the bioreactor were ¨ measured to determine volatile fatty acids (BUCHI B-324, Switzerland) and chemical oxygen demand (COD) on centrifuged samples (Hach-Lange D5000, Poland). Biogas flow rate (flow meter Ritter TG01PVC-PVC, Germany), pH (pH-meter electrode, Mettler Toledo, Poland), and biogas content (gas content

552

A. Kacprzak et al./Chemical Papers 66 (6) 550–555 (2012)

Table 1. Characteristics of substrates used in the experiments; VFA = volatile fatty acids, acetic acid in this case Parameter

Canary grass I

Canary grass II

Whey

Glycerin

4.83 18.6 88.5 – – –

4.18 29.5 83.8 – – –

4.48 6.1 5.26 66.7 2220 21 : 1

7.23 11 7.2 – 1296 25 : 1

pH Total solids/% Volatile solids/% COD/(g L−1 ) VFA/(mg L−1 ) C to N mole ratio

Table 2. Yield of biogas production Biogas Source

Wet mass

Dry mass

Methane Volatile solid

Wet mass

Dry mass

Volatile solid

295 158

333 189

L kg−1 Canary grass I Canary grass II

102 66

551 223

622 266

analyzer LMS GAS DATA GFM 410, OMC ENVAG, Poland) were measured continuously.

55 47

Table 3. Summary of kinetic parameters Parameter

Results and discussion Biochemical methane potential (BMP) assays

P/(L kg−1 ) Rm /(L d−1 ) λ/d−1 R2

Canary Grass I

Canary Grass II

648.44 15.64 14.67 0.9926

223.21 18.47 1.75 0.9995

Data obtained from batch anaerobic processes were used to determine kinetics of the methane production. In Table 2, the yields of biogas and methane production are presented. The methane yields from the first and second harvest were 0.333 m3 kg−1 and 0.189 m3 kg−1 , respectively, and they are comparable to the corresponding results (0.17–0.49 m3 kg−1 ) obtained by Lechtomäki et al. (2008). The data of cumulative methane production obtained from the experiments were fitted to the modified Gompertz equation that describes cumulative methane production in batch assays assuming that CH4 production is a function of bacterial growth, i.e. 
 Rm e (λ − t) + 1 M = P exp − exp P

(1)

where M is the cumulative methane production (L kg−1 ), P is the methane production potential (L kg−1 ), Rm the maximum methane production rate (norm conditions: 273 K and 101.3 kPa) (L d−1 ), λ the duration of the lag phase, and t is the duration of the assay at which cumulative methane production M is calculated (Lay et al., 1997). Fitting the kinetic model to the experimental data and the estimation of kinetic parameters was carried out in the Origin software, (Microcal Software Inc., USA). Table 3 shows the values of kinetic parameters obtained by fitting the model to experimental

Fig. 1. Fitting of experimental data ( ) to the kinetic model (–) for Canary grass I.

data with their corresponding standard deviations. The high values obtained for R2 demonstrates the suitability of the proposed model for realistic estimation of the anaerobic digestibility of the given substrate. Figs. 1 and 2 show fitting of the kinetic model to experimental data for Canary grass I and II, respectively. It can be noticed that the duration of the lag phase is much longer for the Canary grass from the first harvest. It can be explained by the fact that

553

A. Kacprzak et al./Chemical Papers 66 (6) 550–555 (2012)

uniformity of the system. Moreover, the model was not segregated as biomass was treated generally, the model did not differentiate particular groups of microorganisms. This generalization can be justified by the fact that as the inoculum, digested sludge containing mixed bacterial culture was applied. Generally, it was assumed that the substrate undergoes the following changes: S (solid) → A (liquid) → G (gas)

Fig. 2. Fitting of experimental data ( ) to the kinetic model (–) for Canary grass II.

the inoculum used in case of the second harvest was adapted to the substrate (as it was taken from the bioreactor fed with the same substrate). As a result, the process proceeded faster and the maximum production potential rate was higher (18.47 L d−1 ) compared to that of the adapted sludge (15.64 L d−1 ). Co-digestion process Experiments on co-digestion of Canary grass I with cheese whey and glycerin fraction were performed. Based on the experimental data obtained on the way of the co-digestion processes, a kinetic model describing the changes of organic carbon was proposed. Fitting kinetic models and the estimation of kinetic parameters were made using the optimization procedure which combines the Gauss Newton method and the quasi-Newton, DFNLP (Schittkowski, 2001), built-in program EasyFit (Schittkowski, Germany). This procedure requires the specification of boundary conditions, initial ones. As the initial conditions, experimental data determined at the start of the processes were assumed. The selected optimization method consisted in finding the minimum deviation calculated as the sum of squared deviations from the experimental data from the function divided by the sum of the squares of the experimental data, Eq. (2). m n  

R=

(yji − fj (xi , k))

j=1 i=1 m n  

2

(2) 2 yji

j=1 i=1

Other calculations were performed using the Microsoft Excel 2010 program. Due to the fact that biological processes are characterized by high complexity, developing a mathematical model accurately describing the real process is extremely difficult. Therefore, simplifications are necessary. One of the simplifying assumptions concerns the

(3)

It was assumed that the hydrolysis process is a system of pseudo-first order reactions Eq. (4), and the process of methane production takes place with the participation of methane bacteria. During the first trials of fitting of the model to experimental data it was found that the saturation constant reached very high values. Therefore, an attempt to describe the production of biogas as a pseudo-first order reaction was made. As a result, biogas production was described by Eq. (5) and changes in the concentration of volatile organic acids were described by Eq. (6). All values were converted to mg dm−3 . In the course of the process, the organic matter content in the fermented plant material was not determined; however, changes its carbon content were estimated by Eq. (7). Co-fermentation process was described by the following system of differential equations: dS = −kS S dt

(4)

dG = kA A dt

(5)

dA = kS S − kA A dt

(6)

S = So − A − G

(7)

where k is the first order reaction constant (d−1 ), S is the carbon content in the fermented plant material (mg L−1 ), A is the carbon content in VFA (mg L−1 ), G is the carbon content in the biogas produced (mg L−1 ). This model describes the experimental data with sufficient precision; the sum of squared deviations of experimental data from the value of the function being an effect of mathematical simulation was lower than 0.02. Fig. 3 also confirms the good agreement of numerical simulation results with experimental data, which means that the model is suitable for the description of the increase in the concentrations of VFA and its use for biogas production. VFA concentrations obtained by the model were comparable to the values reached by Lei et al. (2010). In Table 4, the values of kinetic constants obtained from the adjustment of the model to the experimental data are presented. The proposed kinetic model defines the experimental data in a satisfactory way.

554

A. Kacprzak et al./Chemical Papers 66 (6) 550–555 (2012)

References

Fig. 3. Adjustment of the kinetic model with the experimental data (VFA experimental data (
) and model data (—), biogas experimental data ( ), and model data (– – –).

Table 4. Kinetic constants Constant

Process 1

Process 2

kS /(d−1 ) kA /(d−1 ) Residuals

1.2 0.03 0.0150

1.2 0.04 0.0079

Conclusions Numerous advantages of the anaerobic digestion process have brought about increased interest in the application of this technology. However, costs of investment in biogas plants are high and thus, precise planning and design are necessary to optimize the production of biogas. Biochemical methane potential assays (BMPs) provide realistic estimation of anaerobic digestibility of a given substrate and the potential biogas production. Obtained BMP values for the Canary grass varied in the range of 0.19–0.33 m3 kg−1 . The conducted experiments confirmed the necessity of sludge adaptation. Experiment conducted with adapted sludge was characterized by a higher maximum production potential rate (18.47 L d−1 ) than with un-adapted sludge (15.64 L d−1 ). The co-digestion process of cheese whey, glycerin and Canary grass was described by a simple kinetic model. The process was divided into two steps; each step was characterized by a pseudo-first order reaction. The model was validated by the real anaerobic co-digestion process. The obtained residuals were lower than 0.02, confirming that the model defines the experimental data in a satisfactory way. Acknowledgements. The work was supported by the European Union under the Operational Program Innovative Economy 2007–2013: POIG.01.03.01-00-132/08 (European Regional Development Fund).

Angelidaki, I., Alves, M., Bolzonella, D., Borzacconi, L., Campos, J. L., Guwy, A. J., Kalyuzhnyi, S., Jenicek, P., & van Lier, J. B. (2009). Defining the biomethane potential (BMP) of solid organic wastes and energy crops: a proposed protocol for batch assays. Water Science and Technology, 59, 927–934. DOI: 10.2166/wst.2009.040. Bouallagui, H., Lahdheb, H., Ben Romdan, E., Rachdi, B., & Hamdi, M. (2009). Improvement of fruit and vegetable waste anaerobic digestion performance and stability with cosubstrates addition. Journal of Environmental Management, 90, 1844–1849. DOI: 10.1016/j.jenvman.2008.12.002. Demirel, B., & Scherer, P. (2009). Bio-methanization of energy crops through mono-digestion for continuous production of renewable biogas. Renewable Energy, 34, 2940–2945. DOI: 10.1016/j.renene.2009.05.013. Fernández, A., Sánchez, A., & Font, X. (2005). Anaerobic co-digestion of a simulated organic fraction of municipal solid wastes and fats of animal and vegetable origin. Biochemical Engineering Journal, 26, 22–28. DOI: 10.1016/j.bej.2005.02.018. Gavala, N. H., Angelidaki, I., & Ahring, B. K. (2003). Kinetics and modeling of anaerobic digestion process. Advances in Biochemical Engineering/Biotechnology, 81, 57–93. DOI: 10.1007/3-540-45839-5 3. Gelegenis, J., Georgakakis, D., Angelidaki, I., Christopoulou, N., & Goumenaki, M. (2007). Optimization of biogas production from olive-oil mill wastewater, by codigesting with diluted poultry-manure. Applied Energy, 84, 646–663. DOI: 10.1016/j.apenergy.2006.12.001. Kacprzak, A., Krzystek, L., & Ledakowicz, S. (2010). Codigestion of agricultural and industrial wastes. Chemical Papers, 64, 127–131. DOI: 10.2478/s11696-009-0108-5. Kayode Coker, A. (2001). Modeling of chemical kinetics and reactor design. Woburn, MA, USA: Gulf Professional Publishing. Kirk, D., & Bickert, W. (2004). The use of biochemical methane potential tests to evaluate the digestibility of processed dairy manure. St. Joseph, MI, USA: American Society of Agricultural and Biological Engineers. (ASABE Paper No. 044066) Labatut, R. A., & Scott, N. R. (2008). Experimental and predicted methane yields from the anaerobic co-digestion of animal manure with complex organic substrates. St. Joseph, MI, USA: American Society of Agricultural and Biological Engineers. (ASABE Paper No. 085087) Lay, J. J., Li, Y. Y., & Noike, T. (1997). The influences of pH and moisture content on the methane production in highsolids sludge digestion. Water Research, 31, 1518–1524. DOI: 10.1016/s0043-1354(96)00413-7. Lehtomäki, A., Huttunen, S., & Rintala, J. A. (2007). Laboratory investigations on co-digestion of energy crops and crop residues with cow manure for methane production: Effect of crop to manure ratio. Resources, Conservation and Recycling, 51, 591–609. DOI: 10.1016/j.resconrec.2006.11.004. Lehtomäki, A., Viinikainen, T. A., & Rintala, J. A. (2008). Screening boreal energy crops and crop residues for methane biofuel production. Biomass and Bioenergy, 32, 541–550. DOI: 10.1016/j.biombioe.2007.11.013. Lei, Z., Chen, J., Zhang, Z., & Sugiura, N. (2010). Methane production from rice straw with acclimated anaerobic sludge: Effect of phosphate supplementation. Bioresource Technology, 101, 4343–4348. DOI: 10.1016/j.biortech.2010.01.083. Lovanh, N., Loughrin, J., Ruiz-Aguilar, G., & Sistani, K. (2008). Evaluation of methane potential from swine slurry with poultry litter amendments. St. Joseph, MI, USA: American Society of Agricultural and Biological Engineers. (ASABE Paper No. 083989)

A. Kacprzak et al./Chemical Papers 66 (6) 550–555 (2012)

Moody, L., Burns, R., Wu-Haan, W., & Spaji´ c, R. (2009). Use of biochemical methane potential (BMP) assays for predicting and enhancing anaerobic digester performance. In 44th Croatian & 4th International Symposium on Agriculture, February 16–20, 2009 (pp. 930–934). Opatija, Croatia: Faculty of Agriculture, University “Josip Juraj Strossmayer” in Osijek. Murto, M., Bj¨ ornsson, L., & Mattiasson, B. (2004). Impact of food industrial waste on anaerobic co-digestion of sewage sludge and pig manure. Journal of Environmental Management, 70, 101–107. DOI: 10.1016/j.jenvman.2003.11.001. Santosh, Y., Sreekrishnan, T. R., Kohli, S., & Rana, V. (2004). Enhancement of biogas production from solid substrates using different techniques—a review. Bioresource Technology, 95, 1–10. DOI: 10.1016/j.biortech.2004.02.010.

555

Schittkowski, K. (2001). Easy-fit [computer software]. Bayreuth, Germany: University of Bayreuth. Sosnowski, P., Klepacz-Smolka, A., Kaczorek, K., & Ledakowicz, S. (2008). Kinetic investigations of methane co-fermentation of sewage sludge and organic fraction of municipal solid waste. Bioresource Technology, 99, 5731–5737. DOI: 10.1016/j.biortech.2007.10.019. Sosnowski, P., Wieczorek, A., & Ledakowicz, S. (2003). Anaerobic co-digestion of sewage sludge and organic fraction of municipal solid wastes. Advances in Environmental Research, 7, 609–616. DOI: 10.1016/s1093-0191(02)00049-7.