- Email: [email protected]
Saponification of fatty slaughterhouse wastes for enhancing anaerobic biodegradability
Bioresource Technology 100 (2009) 3695–3700
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Saponification of fatty slaughterhouse wastes for enhancing anaerobic biodegradability Audrey Battimelli *, Hélène Carrère, Jean-Philippe Delgenès INRA, UR50, Laboratoire de Biotechnologie de l’Environnement, Avenue des Etangs, Narbonne F-11100, France
a r t i c l e
i n f o
Article history: Received 8 September 2008 Received in revised form 8 December 2008 Accepted 15 December 2008 Available online 5 February 2009 Keywords: Anaerobic digestion Biogas Long-chain fatty acid Thermochemical pretreatment
a b s t r a c t The thermochemical pretreatment by saponification of two kinds of fatty slaughterhouse waste – aeroflotation fats and flesh fats from animal carcasses – was studied in order to improve the waste’s anaerobic degradation. The effect of an easily biodegradable compound, ethanol, on raw waste biodegradation was also examined. The aims of the study were to enhance the methanisation of fatty waste and also to show a link between biodegradability and bio-availability. The anaerobic digestion of raw waste, saponified waste and waste with a co-substrate was carried out in batch mode under mesophilic and thermophilic conditions. The results showed little increase in the total volume of biogas, indicating a good biodegradability of the raw wastes. Mean biogas volume reached 1200 mL/g VS which represented more than 90% of the maximal theoretical biogas potential. Raw fatty wastes were slowly biodegraded whereas pretreated wastes showed improved initial reaction kinetics, indicating a better initial bio-availability, particularly for mesophilic runs. The effects observed for raw wastes with ethanol as co-substrate depended on the process temperature: in mesophilic conditions, an initial improvement was observed whereas in thermophilic conditions a significant decrease in biodegradability was observed. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Large amounts of fatty waste are produced by many agro-industries, particularly during food processing. More stringent European Regulations (2002) concerning slaughterhouse wastes and byproducts led to a drastic limitation for their management: re-use, transformation or disposal. Incineration remained one of the few available options for the disposal of fatty wastes, resulting in increased costs. Methanisation could well be a suitable alternative to waste incineration because it leads to the reduction of organic matter content and to the production of a biogas, mainly constituted of methane which is a renewable source of energy. During anaerobic digestion, lipids are initially hydrolysed by extracellular lipases produced by micro-organisms. Long-chain fatty acids (LCFA) and glycerol are released into the mixture. Glycerol is easily biodegraded into volatile fatty acids (VFA) which are then converted into biogas. LCFA are firstly adsorbed onto the surface of the micro-organisms, then transferred into the microbial cells and finally degraded via beta-oxidation, as reported by Li et al. (2002). This last step leads to the production of acetate and hydrogen which are then converted into biogas composed, in the main, of CH4 and CO2.
* Corresponding author. Tel.: +33 (0) 468 425 153; fax: +33 (0) 468 425 160. E-mail address: [email protected] (A. Battimelli). 0960-8524/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2008.12.029
However, direct methanisation of pure fats is difficult to achieve because pure fats are insoluble, less dense than water and with slow biodegradation. The limiting step in this process is assumed to be the physical mass transfer from solid to liquid phase and/or the biological step of LCFA degradation. In addition, some longchain fatty acids are believed to inhibit certain methanogenic micro-organisms, as reported by Rinzema et al. (1994). A batch study carried out by Cirne et al. (2007) showed that the inhibition caused by lipid concentration was linked to their hydrolysis rate but it was also reversible. Saponification is the hydrolysis reaction between a lipid (ester) and an alkali, resulting in LCFA salts and the release of glycerol. Few studies have been done concerning the thermochemical pretreatment of fatty waste prior to anaerobic digestion. Mouneimne et al. (2003) studied fatty waste saponification for the enhancement of VFA production. More recently López Torres and Espinosa Lloréns (2008) tried out the addition of Ca(OH)2 for the anaerobic digestion of the organic fraction of solid municipal waste. This treatment led to the formation of insoluble LCFA salts. The conversion into soluble soaps of the lipids and free LCFA that make-up insoluble fat, oil and grease wastes should improve the contact between the substrate and micro-organisms, thereby enhancing their anaerobic biodegradability. Thus the present work aimed to study the anaerobic biodegradability of fatty slaughterhouse wastes – a solid and a suspended waste – using batch reactors. Four substrates were assessed: ethanol, raw waste, saponified
3696
A. Battimelli et al. / Bioresource Technology 100 (2009) 3695–3700
Table 1 Waste characteristics. Fatty waste characteristics
Appearance Total COD (g/g) HEM (%) TS; VS (g/kg) Organic part (VS/TS)
Aeroflotation fats
Flesh fat from cattle carcasses
High load suspension, brown, viscous, containing hair and flesh 0.41 15 155; 138 89%
White, solid, granular after sieving (5 mm) 1.68 100 935; 934 100%
waste and a mixture of fatty waste with ethanol as co-substrate. Biodegradability was determined in both mesophilic and thermophilic anaerobic conditions. The study also sought to identify a link between bio-availability and biodegradability.
In this study, the experimental biogas potentials were determined for raw waste, pretreated waste and for the raw waste with ethanol, an easily biodegradable co-substrate. Experimental biogas potentials were expressed as mL/g VS. Using the Buswell equation (1), theoretical biogas potentials were calculated as follows:
2. Methodology
a b n a b Cn Ha Ob þ n H2 O ! þ CH4 4 2 2 8 4 n a b þ þ CO2 2 8 4
2.1. Substrate Fatty samples were collected in French slaughterhouses and the sample characteristics are given in Table 1: visual aspect, total chemical oxygen demand (COD), hexane extractible matter (HEM), total solids (TS) and volatile solids (VS). Prior to storage at 20 °C, samples were homogenised. For aeroflotation fats, a light mixing was done; flesh fats were sieved with a 5 mm mesh, which permitted the removal of the membranes containing fats. Ethanol was used for both the biomass activity and the co-substrate tests (95% v/v for mesophilic and 99.8% v/v for thermophilic).
ð1Þ
This calculation requires the knowledge of the make-up of organic matter, and assumes a complete conversion of such matter into biogas. Having ascertained the composition of the fatty waste and calculated both the LCFA and the glycerol biogas potentials, the theoretical waste biogas potential was previously estimated by Battimelli et al. (2008) as 1341 mL/g VS for the same aeroflotation fatty waste and 1346 mL/g VS for the same carcass fatty waste. 2.4. Analysis
2.2. Pretreatment conditions Saponification pretreatment was performed at 60 °C for 30 min with mixing. The alkali used for soap production was sodium hydroxide at 0.3 wt.% concentration. The alkali dose was adjusted to the equivalent of 0.04 mol/g COD, determined in order to ensure an excess of OH. The amount of hydroxyl was more than 10 times higher than the stoichiometric amount needed for a theoretical waste constituted only of palmitic acid (one of the lightest LCFA constituting fatty waste). After obtaining soaps, the required excess of alkali was checked for by measuring the pH which was in fact around 12. 2.3. Biodegradation Experiments were performed using 500 mL glass flasks. Anaerobic sludges were sampled from two pilot reactors: one, mesophilic, treating wine effluent and the second, thermophilic, treating household waste. Incubation was carried out under agitation (150 rpm) in thermostatic chambers, at 35 °C for mesophilic and 55 °C for thermophilic runs. Biogas was periodically removed from flasks and measured using a water column connected to a needle. In order to avoid the dissolving of CO2 in the water during biogas measurement, the pH was regulated at two with HCl and the solution saturated with NaCl at 2% w/v. Gas volumes were calculated at standard temperature and pressure (STP). Three steps were followed for biodegradation tests: – the sludge was starved to obtain complete removal of residual biodegradable COD, – then, ethanol injection in order to determine sludge activity involving an easily biodegradable compound, and – two or three successive fatty waste additions in order to determine biodegradability.
Fatty wastes were conditioned prior to analysis as described by Bridoux et al. (1994). Analytical measurement of chemical oxygen demand (COD), total solids (TS), volatile solids (VS), pH and alkalinity followed APHA Standardised Methods (1992). The soluble fraction was obtained after centrifugation (15,000 rpm, 15 min, 5 °C) and filtration (at 0.45 lm). Hexane extractable matter (HEM) measurements were performed in accordance with the method described by Canler (2001). Three consecutive extractions were required for a steady value. The fatty waste hydrolysis rate was estimated by pH measurement. All measurements were done in triplicate, low variation coefficient (less than 5%), calculated as the ratio between mean value and standard deviation, ensured validation of the analyses. 3. Results The biogas potential of both types of raw wastes and their respective soaps and of the raw wastes with an easily biodegradable co-substrate was determined under mesophilic and thermophilic conditions. 3.1. Mesophilic fatty waste biodegradability 3.1.1. Inoculum anaerobic activity The mesophilic sludge had TS = 16.0 g/L and VS = 8.9 g/L. After the initial period of sludge starvation, the pH was stable at 8.2, residual soluble COD was 0.48 g/L and alkalinity 2.7 g CaCO3/L. At the end of this period, endogenous sludge biogas production was stable at 3 mL/d and could thus be disregarded during the biodegradation of ethanol, raw wastes and soaps. After injection of the ethanol, the mean pH value decreased slightly in each flask to 7.5. An assessment of sludge activity after ethanol addition (0.47 g COD) in each flask revealed a mean biogas production of
3697
A. Battimelli et al. / Bioresource Technology 100 (2009) 3695–3700
419 ± 5 mL/g COD, which represented 90% biodegradation 10 days after injection. This result showed that inoculum biogas activity was good and was similar in each flask.
seems to indicate that even when hydrolysed LCFA continue to be hard to degrade biologically. 3.1.3. Co-substrate impact on raw waste mesophilic biodegradability The same amounts of waste and ethanol were added to the flasks as in the individual waste biodegradation tests: 0.54 g COD for aeroflotation, 0.59 g COD for carcass and 0.47 g COD for ethanol. Thus total COD was twice as high as in previous assays but the initial loading rate remained low at 0.2 g COD/g VS. Biogas production for raw wastes with a co-substrate is shown in Fig. 2. As shown by the curves in Fig. 2, the two mixtures of raw waste plus ethanol and ethanol alone display a similar pattern of biogas production. The initial production rate of both mixtures was high and close to the ethanol biodegradation rate. Subsequently, the curves for both wastes with the co-substrate were similar and they were lower than those for individual waste. Final biodegradation rates were nearly the same for wastes with or without ethanol: in the case of aeroflotation, 91% and 87%, respectively, and for carcass waste 88% and 84%, respectively. Since no significant enhancement of biodegradability was observed with ethanol as cosubstrate, this could well be due to initial good biodegradability of the substrates. The calculation of the ratio between the actual volume of the different mixture and the volume obtained by totalising separate waste and ethanol values, gives a better description of the effect of the co-substrate. An improvement of biodegradation is observed for both mixtures of raw waste and ethanol during the initial six days. At the end of the biodegradation, the mixtures behaved like the sum of the individual substrates, which corresponded to the slow biodegradation of LCFA. The initial improvement may be attributed to a faster degradation of ethanol or glycerol but interpretation is limited by the small initial volumes.
3.1.2. Fatty waste mesophilic biodegradability The loading rate at the beginning of each batch was low, around 0.1 g COD/g VS, as the amount of waste was 0.54 g COD for aeroflotation fat and 0.59 g COD for carcass fat. For the second batch, Fig. 1 presents the biogas production, after 18 days of biodegradation at 35 °C, of each raw waste, their respective soaps and ethanol. A slow biodegradation was observed for all greasy wastes. When compared to ethanol, which is an easily biodegradable compound, initial production (day 2) was five times lower for grease and soaps. The biogas production for both soaps was slightly higher than for that of raw waste. Final biogas production for raw waste and pretreated wastes is given in Table 2. All final volumes for both wastes, with and without pretreatment, display very similar biogas production, with an average volume of 1200 mL/g VS ± 6%. The experimental values, when compared to theoretical biogas potentials calculated using Eq. (1), corresponded to about 90%. This clearly indicates that all wastes were well-degraded in batch reactors after 18 days. The biodegradation of aeroflotation fat was slightly better than for carcass fat. Similarly, their respective soaps had very slightly better biodegradation compared to the raw wastes. The calculation of ratios between soap and raw waste biogas production during the experiment gives an estimation of the pretreatment impact on kinetic. For both wastes, biogas production was improved by the pretreatment (except day 1 for carcass waste) as the ratio is higher than one. The biggest improvement was observed during the initial period, notably on day 1 for saponified aeroflotation waste when biogas production was four times higher than for the raw waste. The ratios of biogas production were close to one, as final volumes were equivalent. The aim of the pretreatment was the conversion of solid fats into more soluble soaps so as to obtain better contact between the substrate and anaerobic micro-organisms. Assuming that biological hydrolysis is slower than chemical hydrolysis, waste saponification should have enhanced the biodegradation rate. Ratio calculations seem to confirm this assumption and highlight an initial advantage of waste saponification prior to anaerobic digestion. The fact that during the end period soap kinetics remained slow
3.2. Thermophilic fatty waste biodegradability 3.2.1. Inoculum anaerobic activity Initial thermophilic sludge starvation led to a lower pH, down from 8.0 to 7.5, residual soluble COD was 0.90 g/L, TS and VS were, respectively, 3.7 and 2.7 g/L, total alkalinity was 5.0 g CaCO3/L. At the end of this period, endogenous sludge biogas production was stable at 3 mL/d, and could thus be disregarded during the biodegradation of ethanol, raw wastes and soaps. Thermophilic sludge activity assessment based on two successive additions of ethanol (0.47 g COD) in each flask led to a mean biogas production of
100% 90%
biogas production (%)
80% 70% 60% 50% 40% 30% 20% 10% 0% 0
2
4
6
8
10
12
14
16
duration (days) Raw Aerofl.
Raw Carcass
Soap Aeroflot.
Soap Carcass
Ethanol
Fig. 1. Biogas production in mesophilic conditions (% of theoretical production).
18
3698
A. Battimelli et al. / Bioresource Technology 100 (2009) 3695–3700
Table 2 Biogas volume for raw and saponified wastes at mesophilic conditions. Mesophilic anaerobic conditions
Final biogas volume STP (mL/g VS)
Aeroflotation fat Carcass fat
Experimental/theoretical (%)
Raw waste
Soap waste
Raw waste
Soap waste
1222 1126
1284 1166
91 84
96 87
100% 90%
% (Vcumul./Vtheor.)
80% 70% 60% 50% 40% 30% 20% 10% 0% 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
duration (days) Raw Aerofl.
Raw Carcass
Aeroflot. + Eth
Carcass + Eth
Ethanol
Fig. 2. Comparison of biogas volumes generated by raw waste alone and by raw waste with ethanol co-substrate under mesophilic conditions.
462 ± 5 mL/g COD, representing 99% biodegradation after 12 days of incubation. This showed that inoculum biogas activity was correct and similar in each flask.
than from the same raw waste, whereas no significant difference occurred with aeroflotation fats. The final volumes after 15 days of degradation are given in Table 3. All final volumes for both wastes, with and without pretreatment, highlight very similar biogas production levels, with an average volume of 1214 mL/g VS ± 7%. When compared to theoretical biogas potentials calculated using Eq. (1), the experimental values turned out to be close, around 91%. This clearly indicates that all wastes were well-biodegraded in batch reactors after 15 days of treatment. Carcass fat biodegradation was significantly higher than that of aeroflotation fat. The ratios between soap biogas production and raw wastes during the experiment are given in Fig. 4.
3.2.2. Thermophilic fatty waste biodegradability Initial waste loading was 0.54 g COD for aeroflotation fat and 0.59 g COD for carcass fat. The loading rate at the start-up of each batch was low, around 0.4 g COD/g VS. Biogas production for each flask of the third batch is shown in Fig. 3. All fatty waste was slowly biodegraded at the thermophilic temperature. No clear initial lag period was observed but during the early days volumes for waste were very low compared to ethanol. The biogas production from carcass soap was slightly higher 100% 90%
biogas production (%)
80% 70% 60% 50% 40% 30% 20% 10% 0% 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
duration (days) Raw Aerofl.
Raw Carcass
Soap Aeroflot.
Soap Carcass
Ethanol
Fig. 3. Biogas production under thermophilic conditions (% of theoretical production).
15
3699
A. Battimelli et al. / Bioresource Technology 100 (2009) 3695–3700 Table 3 Biogas volume for raw and saponified wastes under thermophilic conditions. Thermophilic anaerobic conditions
Final biogas volume STP (mL/g VS)
Aeroflotation fat Carcass fat
Experimental/theoretical (%)
Raw waste
Soap waste
Raw waste
Soap waste
1161 1233
1140 1324
87 92
85 99
Ratio V Soap / V Waste
2
1.5
1
0.5
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
duration (days)
Aeroflotation
Carcass
Fig. 4. Comparison of biogas volumes generated by saponified and raw waste during thermophilic anaerobic digestion.
For both wastes, except day 1, volume ratios were relatively close to one, which indicate that the pretreatment had little effect on waste biodegradation. Carcass waste biodegradation was slightly improved after day 3 by saponification whereas aeroflotation waste conversion was hampered by the pretreatment. At the mesophilic temperature, the improvement was recorded for both wastes whereas at the thermophilic temperature it was only observed for carcass waste. This can be explained by the solubilisation of the waste induced by thermochemical pretreatment. At the thermophilic temperature, raw wastes were fluid which might explain the difference in biodegradability enhancement at the mesophilic temperature.
3.2.3. Co-substrate impact on thermophilic biodegradability of raw waste The amounts of substrate were the same as in mesophilic runs. Biogas production for wastes with a co-substrate, raw wastes and ethanol are shown in Fig. 5. As represented in Fig. 5, no significant difference in levels of waste biogas production is evidenced: not even the initial output was improved by the addition of ethanol. The biogas curves for both wastes with a co-substrate were similar and they were lower than those for individual wastes. Final biodegradation rates for aeroflotation wastes with and without ethanol are nearly the same (respectively, 87% and 85%), whereas for carcass wastes
100% 90%
% (Vcumul./Vtheor.)
80% 70% 60% 50% 40% 30% 20% 10% 0% 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
duration (days) Raw Aerofl.
Raw Carcass
Aero + Eth
Carcass + Eth
Ethanol
Fig. 5. Comparison of biogas volumes generated by raw waste alone and by raw waste with ethanol co-substrate, under thermophilic temperature.
3700
A. Battimelli et al. / Bioresource Technology 100 (2009) 3695–3700 1.2
Ratio V cumul (exper./calcul.)
1
0.8
0.6
0.4
0.2
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
time (days) Ratio Aeroflotation
Ratio Carcass
Fig. 6. Ratio of biogas from the mixtures and the sum of individual substrates (thermophilic).
the addition of ethanol resulted in a marked decrease, from 92% to 78%. The calculation of the ratio between the actual volume of the different mixtures and the volume obtained by totalising the separate waste and ethanol values, gives a better description of the effect of the co-substrate. These ratios are presented in Fig. 6. For all wastes, a negative effect of ethanol was observed, particularly at the beginning of the experiment. This initial lag phase but also the final lower biogas production seems to indicate an inhibition of thermophilic biomass by the substrates probably due to inadequate part of active biomass into global volatile solid determination. This assumption is partially confirmed by the curve of ethanol kinetic in Fig. 5 which is rather slow when compared to relative mesophilic one, in Fig. 2. 4. Conclusions The saponification of fatty wastes from slaughterhouses as a mean of improving biogas production was studied in batch reactors. Results showed little enhancement in the total volume produced, pointing to good biodegradability of both aeroflotation and carcass wastes at mesophilic and thermophilic temperatures. However, the pretreatment led to a slight improvement in the initial reaction kinetics, indicating enhanced bio-availability due to chemical hydrolysis, particularly at the mesophilic temperatures. LCFA anaerobic biodegradation kinetics remained slow in all experimental conditions, even after hydrolysis. The effects observed for raw wastes with ethanol as their co-substrate depended on the temperature during the anaerobic process. Anaerobic batch tests applied to fatty waste were suitable for biogas potential determination. Still they remained insufficient for a clear evaluation of LCFA inhibition even after three waste additions. In that case the positive impact of the pretreatment was probably underestimated. Aiming at a better determination of anaerobic digestion of fatty waste, saponification should be assessed using semi-continuous reactors. In this configuration the
cumulative effects, such as grease accumulation, biomass inhibition or adaptation, should be stated.
Acknowledgements The authors would like to acknowledge the help from the French ‘‘Office de l’Elevage” (Livestock Service) and INTERBEV in funding this study.
References American Public Health Association, American Water Works Association, Water Pollution Control Federation, 1992. In: Clesceri, L.S., Greenberg, A.E., Trussel, R.R. (Eds.), Standard Methods for the Examination of Water and Wastewater, 18th ed. Battimelli, A., Torrijos, M., Delgenès, J.P., 2008. Fatty Slaughterhouse Waste Saponification to Increase Biogas Yield. International Water Association, International Symposium on Anaerobic Digestion of Solid Wastes and Energy Crops, Hammamet Tunisia, Proceeding Book, 8B. Bridoux, G., Dhulster, P., Manem, J., 1994. Analyse des graisses dans les stations d’épuration. Techniques Sciences Méthodes 5, 257–262. Canler, J.P., 2001. Performances des systèmes de traitement biologique aérobie des graisses – Graisses issues des dégraisseurs de station d’épuration traitant des effluents à dominances domestique. Cemagref édition, p. 64. Cirne, D.G., Paloumet, X., Bjornsson, L., Alves, M.M., Mattiasson, B., 2007. Anaerobic digestion of lipid-rich waste – effects of lipid concentration. Renewable Energy 32 (6), 965–975. Regulation (EC) No. 1774/2002 of the European Parliament and of the Council of 3 October 2002 laying down health rules concerning animal by-products not intended for human consumption. Official Journal of the European Communities, L.273/1–L273/95. Li, Y.Y., Sasaki, H., Yamashita, K., Seki, K., Kamigochi, I., 2002. High-rate methane fermentation of lipid-rich food wastes by a high-solids co-digestion process. Water Science and Technology 45 (12), 143–150. López Torres, M., Espinosa Lloréns, M.d.C., 2008. Effect of alkaline pretreatment on anaerobic digestion of solid wastes. Waste Management 28 (11), 2229–2234. Mouneimne, A.H., Carrère, H., Bernet, N., Delgenès, J.P., 2003. Effect of saponification on the anaerobic digestion of solid fatty residues. Bioresource Technology 90 (1), 89–94. Rinzema, A., Boone, M., Van Knippenberg, K., Lettinga, G., 1994. Bactericidal effect of long chain fatty acids in anaerobic digestion. Water Environment Research 66 (1), 40–49.
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Saponification of fatty slaughterhouse wastes for enhancing anaerobic biodegradability Audrey Battimelli *, Hélène Carrère, Jean-Philippe Delgenès INRA, UR50, Laboratoire de Biotechnologie de l’Environnement, Avenue des Etangs, Narbonne F-11100, France
a r t i c l e
i n f o
Article history: Received 8 September 2008 Received in revised form 8 December 2008 Accepted 15 December 2008 Available online 5 February 2009 Keywords: Anaerobic digestion Biogas Long-chain fatty acid Thermochemical pretreatment
a b s t r a c t The thermochemical pretreatment by saponification of two kinds of fatty slaughterhouse waste – aeroflotation fats and flesh fats from animal carcasses – was studied in order to improve the waste’s anaerobic degradation. The effect of an easily biodegradable compound, ethanol, on raw waste biodegradation was also examined. The aims of the study were to enhance the methanisation of fatty waste and also to show a link between biodegradability and bio-availability. The anaerobic digestion of raw waste, saponified waste and waste with a co-substrate was carried out in batch mode under mesophilic and thermophilic conditions. The results showed little increase in the total volume of biogas, indicating a good biodegradability of the raw wastes. Mean biogas volume reached 1200 mL/g VS which represented more than 90% of the maximal theoretical biogas potential. Raw fatty wastes were slowly biodegraded whereas pretreated wastes showed improved initial reaction kinetics, indicating a better initial bio-availability, particularly for mesophilic runs. The effects observed for raw wastes with ethanol as co-substrate depended on the process temperature: in mesophilic conditions, an initial improvement was observed whereas in thermophilic conditions a significant decrease in biodegradability was observed. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Large amounts of fatty waste are produced by many agro-industries, particularly during food processing. More stringent European Regulations (2002) concerning slaughterhouse wastes and byproducts led to a drastic limitation for their management: re-use, transformation or disposal. Incineration remained one of the few available options for the disposal of fatty wastes, resulting in increased costs. Methanisation could well be a suitable alternative to waste incineration because it leads to the reduction of organic matter content and to the production of a biogas, mainly constituted of methane which is a renewable source of energy. During anaerobic digestion, lipids are initially hydrolysed by extracellular lipases produced by micro-organisms. Long-chain fatty acids (LCFA) and glycerol are released into the mixture. Glycerol is easily biodegraded into volatile fatty acids (VFA) which are then converted into biogas. LCFA are firstly adsorbed onto the surface of the micro-organisms, then transferred into the microbial cells and finally degraded via beta-oxidation, as reported by Li et al. (2002). This last step leads to the production of acetate and hydrogen which are then converted into biogas composed, in the main, of CH4 and CO2.
* Corresponding author. Tel.: +33 (0) 468 425 153; fax: +33 (0) 468 425 160. E-mail address: [email protected] (A. Battimelli). 0960-8524/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2008.12.029
However, direct methanisation of pure fats is difficult to achieve because pure fats are insoluble, less dense than water and with slow biodegradation. The limiting step in this process is assumed to be the physical mass transfer from solid to liquid phase and/or the biological step of LCFA degradation. In addition, some longchain fatty acids are believed to inhibit certain methanogenic micro-organisms, as reported by Rinzema et al. (1994). A batch study carried out by Cirne et al. (2007) showed that the inhibition caused by lipid concentration was linked to their hydrolysis rate but it was also reversible. Saponification is the hydrolysis reaction between a lipid (ester) and an alkali, resulting in LCFA salts and the release of glycerol. Few studies have been done concerning the thermochemical pretreatment of fatty waste prior to anaerobic digestion. Mouneimne et al. (2003) studied fatty waste saponification for the enhancement of VFA production. More recently López Torres and Espinosa Lloréns (2008) tried out the addition of Ca(OH)2 for the anaerobic digestion of the organic fraction of solid municipal waste. This treatment led to the formation of insoluble LCFA salts. The conversion into soluble soaps of the lipids and free LCFA that make-up insoluble fat, oil and grease wastes should improve the contact between the substrate and micro-organisms, thereby enhancing their anaerobic biodegradability. Thus the present work aimed to study the anaerobic biodegradability of fatty slaughterhouse wastes – a solid and a suspended waste – using batch reactors. Four substrates were assessed: ethanol, raw waste, saponified
3696
A. Battimelli et al. / Bioresource Technology 100 (2009) 3695–3700
Table 1 Waste characteristics. Fatty waste characteristics
Appearance Total COD (g/g) HEM (%) TS; VS (g/kg) Organic part (VS/TS)
Aeroflotation fats
Flesh fat from cattle carcasses
High load suspension, brown, viscous, containing hair and flesh 0.41 15 155; 138 89%
White, solid, granular after sieving (5 mm) 1.68 100 935; 934 100%
waste and a mixture of fatty waste with ethanol as co-substrate. Biodegradability was determined in both mesophilic and thermophilic anaerobic conditions. The study also sought to identify a link between bio-availability and biodegradability.
In this study, the experimental biogas potentials were determined for raw waste, pretreated waste and for the raw waste with ethanol, an easily biodegradable co-substrate. Experimental biogas potentials were expressed as mL/g VS. Using the Buswell equation (1), theoretical biogas potentials were calculated as follows:
2. Methodology
a b n a b Cn Ha Ob þ n H2 O ! þ CH4 4 2 2 8 4 n a b þ þ CO2 2 8 4
2.1. Substrate Fatty samples were collected in French slaughterhouses and the sample characteristics are given in Table 1: visual aspect, total chemical oxygen demand (COD), hexane extractible matter (HEM), total solids (TS) and volatile solids (VS). Prior to storage at 20 °C, samples were homogenised. For aeroflotation fats, a light mixing was done; flesh fats were sieved with a 5 mm mesh, which permitted the removal of the membranes containing fats. Ethanol was used for both the biomass activity and the co-substrate tests (95% v/v for mesophilic and 99.8% v/v for thermophilic).
ð1Þ
This calculation requires the knowledge of the make-up of organic matter, and assumes a complete conversion of such matter into biogas. Having ascertained the composition of the fatty waste and calculated both the LCFA and the glycerol biogas potentials, the theoretical waste biogas potential was previously estimated by Battimelli et al. (2008) as 1341 mL/g VS for the same aeroflotation fatty waste and 1346 mL/g VS for the same carcass fatty waste. 2.4. Analysis
2.2. Pretreatment conditions Saponification pretreatment was performed at 60 °C for 30 min with mixing. The alkali used for soap production was sodium hydroxide at 0.3 wt.% concentration. The alkali dose was adjusted to the equivalent of 0.04 mol/g COD, determined in order to ensure an excess of OH. The amount of hydroxyl was more than 10 times higher than the stoichiometric amount needed for a theoretical waste constituted only of palmitic acid (one of the lightest LCFA constituting fatty waste). After obtaining soaps, the required excess of alkali was checked for by measuring the pH which was in fact around 12. 2.3. Biodegradation Experiments were performed using 500 mL glass flasks. Anaerobic sludges were sampled from two pilot reactors: one, mesophilic, treating wine effluent and the second, thermophilic, treating household waste. Incubation was carried out under agitation (150 rpm) in thermostatic chambers, at 35 °C for mesophilic and 55 °C for thermophilic runs. Biogas was periodically removed from flasks and measured using a water column connected to a needle. In order to avoid the dissolving of CO2 in the water during biogas measurement, the pH was regulated at two with HCl and the solution saturated with NaCl at 2% w/v. Gas volumes were calculated at standard temperature and pressure (STP). Three steps were followed for biodegradation tests: – the sludge was starved to obtain complete removal of residual biodegradable COD, – then, ethanol injection in order to determine sludge activity involving an easily biodegradable compound, and – two or three successive fatty waste additions in order to determine biodegradability.
Fatty wastes were conditioned prior to analysis as described by Bridoux et al. (1994). Analytical measurement of chemical oxygen demand (COD), total solids (TS), volatile solids (VS), pH and alkalinity followed APHA Standardised Methods (1992). The soluble fraction was obtained after centrifugation (15,000 rpm, 15 min, 5 °C) and filtration (at 0.45 lm). Hexane extractable matter (HEM) measurements were performed in accordance with the method described by Canler (2001). Three consecutive extractions were required for a steady value. The fatty waste hydrolysis rate was estimated by pH measurement. All measurements were done in triplicate, low variation coefficient (less than 5%), calculated as the ratio between mean value and standard deviation, ensured validation of the analyses. 3. Results The biogas potential of both types of raw wastes and their respective soaps and of the raw wastes with an easily biodegradable co-substrate was determined under mesophilic and thermophilic conditions. 3.1. Mesophilic fatty waste biodegradability 3.1.1. Inoculum anaerobic activity The mesophilic sludge had TS = 16.0 g/L and VS = 8.9 g/L. After the initial period of sludge starvation, the pH was stable at 8.2, residual soluble COD was 0.48 g/L and alkalinity 2.7 g CaCO3/L. At the end of this period, endogenous sludge biogas production was stable at 3 mL/d and could thus be disregarded during the biodegradation of ethanol, raw wastes and soaps. After injection of the ethanol, the mean pH value decreased slightly in each flask to 7.5. An assessment of sludge activity after ethanol addition (0.47 g COD) in each flask revealed a mean biogas production of
3697
A. Battimelli et al. / Bioresource Technology 100 (2009) 3695–3700
419 ± 5 mL/g COD, which represented 90% biodegradation 10 days after injection. This result showed that inoculum biogas activity was good and was similar in each flask.
seems to indicate that even when hydrolysed LCFA continue to be hard to degrade biologically. 3.1.3. Co-substrate impact on raw waste mesophilic biodegradability The same amounts of waste and ethanol were added to the flasks as in the individual waste biodegradation tests: 0.54 g COD for aeroflotation, 0.59 g COD for carcass and 0.47 g COD for ethanol. Thus total COD was twice as high as in previous assays but the initial loading rate remained low at 0.2 g COD/g VS. Biogas production for raw wastes with a co-substrate is shown in Fig. 2. As shown by the curves in Fig. 2, the two mixtures of raw waste plus ethanol and ethanol alone display a similar pattern of biogas production. The initial production rate of both mixtures was high and close to the ethanol biodegradation rate. Subsequently, the curves for both wastes with the co-substrate were similar and they were lower than those for individual waste. Final biodegradation rates were nearly the same for wastes with or without ethanol: in the case of aeroflotation, 91% and 87%, respectively, and for carcass waste 88% and 84%, respectively. Since no significant enhancement of biodegradability was observed with ethanol as cosubstrate, this could well be due to initial good biodegradability of the substrates. The calculation of the ratio between the actual volume of the different mixture and the volume obtained by totalising separate waste and ethanol values, gives a better description of the effect of the co-substrate. An improvement of biodegradation is observed for both mixtures of raw waste and ethanol during the initial six days. At the end of the biodegradation, the mixtures behaved like the sum of the individual substrates, which corresponded to the slow biodegradation of LCFA. The initial improvement may be attributed to a faster degradation of ethanol or glycerol but interpretation is limited by the small initial volumes.
3.1.2. Fatty waste mesophilic biodegradability The loading rate at the beginning of each batch was low, around 0.1 g COD/g VS, as the amount of waste was 0.54 g COD for aeroflotation fat and 0.59 g COD for carcass fat. For the second batch, Fig. 1 presents the biogas production, after 18 days of biodegradation at 35 °C, of each raw waste, their respective soaps and ethanol. A slow biodegradation was observed for all greasy wastes. When compared to ethanol, which is an easily biodegradable compound, initial production (day 2) was five times lower for grease and soaps. The biogas production for both soaps was slightly higher than for that of raw waste. Final biogas production for raw waste and pretreated wastes is given in Table 2. All final volumes for both wastes, with and without pretreatment, display very similar biogas production, with an average volume of 1200 mL/g VS ± 6%. The experimental values, when compared to theoretical biogas potentials calculated using Eq. (1), corresponded to about 90%. This clearly indicates that all wastes were well-degraded in batch reactors after 18 days. The biodegradation of aeroflotation fat was slightly better than for carcass fat. Similarly, their respective soaps had very slightly better biodegradation compared to the raw wastes. The calculation of ratios between soap and raw waste biogas production during the experiment gives an estimation of the pretreatment impact on kinetic. For both wastes, biogas production was improved by the pretreatment (except day 1 for carcass waste) as the ratio is higher than one. The biggest improvement was observed during the initial period, notably on day 1 for saponified aeroflotation waste when biogas production was four times higher than for the raw waste. The ratios of biogas production were close to one, as final volumes were equivalent. The aim of the pretreatment was the conversion of solid fats into more soluble soaps so as to obtain better contact between the substrate and anaerobic micro-organisms. Assuming that biological hydrolysis is slower than chemical hydrolysis, waste saponification should have enhanced the biodegradation rate. Ratio calculations seem to confirm this assumption and highlight an initial advantage of waste saponification prior to anaerobic digestion. The fact that during the end period soap kinetics remained slow
3.2. Thermophilic fatty waste biodegradability 3.2.1. Inoculum anaerobic activity Initial thermophilic sludge starvation led to a lower pH, down from 8.0 to 7.5, residual soluble COD was 0.90 g/L, TS and VS were, respectively, 3.7 and 2.7 g/L, total alkalinity was 5.0 g CaCO3/L. At the end of this period, endogenous sludge biogas production was stable at 3 mL/d, and could thus be disregarded during the biodegradation of ethanol, raw wastes and soaps. Thermophilic sludge activity assessment based on two successive additions of ethanol (0.47 g COD) in each flask led to a mean biogas production of
100% 90%
biogas production (%)
80% 70% 60% 50% 40% 30% 20% 10% 0% 0
2
4
6
8
10
12
14
16
duration (days) Raw Aerofl.
Raw Carcass
Soap Aeroflot.
Soap Carcass
Ethanol
Fig. 1. Biogas production in mesophilic conditions (% of theoretical production).
18
3698
A. Battimelli et al. / Bioresource Technology 100 (2009) 3695–3700
Table 2 Biogas volume for raw and saponified wastes at mesophilic conditions. Mesophilic anaerobic conditions
Final biogas volume STP (mL/g VS)
Aeroflotation fat Carcass fat
Experimental/theoretical (%)
Raw waste
Soap waste
Raw waste
Soap waste
1222 1126
1284 1166
91 84
96 87
100% 90%
% (Vcumul./Vtheor.)
80% 70% 60% 50% 40% 30% 20% 10% 0% 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
duration (days) Raw Aerofl.
Raw Carcass
Aeroflot. + Eth
Carcass + Eth
Ethanol
Fig. 2. Comparison of biogas volumes generated by raw waste alone and by raw waste with ethanol co-substrate under mesophilic conditions.
462 ± 5 mL/g COD, representing 99% biodegradation after 12 days of incubation. This showed that inoculum biogas activity was correct and similar in each flask.
than from the same raw waste, whereas no significant difference occurred with aeroflotation fats. The final volumes after 15 days of degradation are given in Table 3. All final volumes for both wastes, with and without pretreatment, highlight very similar biogas production levels, with an average volume of 1214 mL/g VS ± 7%. When compared to theoretical biogas potentials calculated using Eq. (1), the experimental values turned out to be close, around 91%. This clearly indicates that all wastes were well-biodegraded in batch reactors after 15 days of treatment. Carcass fat biodegradation was significantly higher than that of aeroflotation fat. The ratios between soap biogas production and raw wastes during the experiment are given in Fig. 4.
3.2.2. Thermophilic fatty waste biodegradability Initial waste loading was 0.54 g COD for aeroflotation fat and 0.59 g COD for carcass fat. The loading rate at the start-up of each batch was low, around 0.4 g COD/g VS. Biogas production for each flask of the third batch is shown in Fig. 3. All fatty waste was slowly biodegraded at the thermophilic temperature. No clear initial lag period was observed but during the early days volumes for waste were very low compared to ethanol. The biogas production from carcass soap was slightly higher 100% 90%
biogas production (%)
80% 70% 60% 50% 40% 30% 20% 10% 0% 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
duration (days) Raw Aerofl.
Raw Carcass
Soap Aeroflot.
Soap Carcass
Ethanol
Fig. 3. Biogas production under thermophilic conditions (% of theoretical production).
15
3699
A. Battimelli et al. / Bioresource Technology 100 (2009) 3695–3700 Table 3 Biogas volume for raw and saponified wastes under thermophilic conditions. Thermophilic anaerobic conditions
Final biogas volume STP (mL/g VS)
Aeroflotation fat Carcass fat
Experimental/theoretical (%)
Raw waste
Soap waste
Raw waste
Soap waste
1161 1233
1140 1324
87 92
85 99
Ratio V Soap / V Waste
2
1.5
1
0.5
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
duration (days)
Aeroflotation
Carcass
Fig. 4. Comparison of biogas volumes generated by saponified and raw waste during thermophilic anaerobic digestion.
For both wastes, except day 1, volume ratios were relatively close to one, which indicate that the pretreatment had little effect on waste biodegradation. Carcass waste biodegradation was slightly improved after day 3 by saponification whereas aeroflotation waste conversion was hampered by the pretreatment. At the mesophilic temperature, the improvement was recorded for both wastes whereas at the thermophilic temperature it was only observed for carcass waste. This can be explained by the solubilisation of the waste induced by thermochemical pretreatment. At the thermophilic temperature, raw wastes were fluid which might explain the difference in biodegradability enhancement at the mesophilic temperature.
3.2.3. Co-substrate impact on thermophilic biodegradability of raw waste The amounts of substrate were the same as in mesophilic runs. Biogas production for wastes with a co-substrate, raw wastes and ethanol are shown in Fig. 5. As represented in Fig. 5, no significant difference in levels of waste biogas production is evidenced: not even the initial output was improved by the addition of ethanol. The biogas curves for both wastes with a co-substrate were similar and they were lower than those for individual wastes. Final biodegradation rates for aeroflotation wastes with and without ethanol are nearly the same (respectively, 87% and 85%), whereas for carcass wastes
100% 90%
% (Vcumul./Vtheor.)
80% 70% 60% 50% 40% 30% 20% 10% 0% 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
duration (days) Raw Aerofl.
Raw Carcass
Aero + Eth
Carcass + Eth
Ethanol
Fig. 5. Comparison of biogas volumes generated by raw waste alone and by raw waste with ethanol co-substrate, under thermophilic temperature.
3700
A. Battimelli et al. / Bioresource Technology 100 (2009) 3695–3700 1.2
Ratio V cumul (exper./calcul.)
1
0.8
0.6
0.4
0.2
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
time (days) Ratio Aeroflotation
Ratio Carcass
Fig. 6. Ratio of biogas from the mixtures and the sum of individual substrates (thermophilic).
the addition of ethanol resulted in a marked decrease, from 92% to 78%. The calculation of the ratio between the actual volume of the different mixtures and the volume obtained by totalising the separate waste and ethanol values, gives a better description of the effect of the co-substrate. These ratios are presented in Fig. 6. For all wastes, a negative effect of ethanol was observed, particularly at the beginning of the experiment. This initial lag phase but also the final lower biogas production seems to indicate an inhibition of thermophilic biomass by the substrates probably due to inadequate part of active biomass into global volatile solid determination. This assumption is partially confirmed by the curve of ethanol kinetic in Fig. 5 which is rather slow when compared to relative mesophilic one, in Fig. 2. 4. Conclusions The saponification of fatty wastes from slaughterhouses as a mean of improving biogas production was studied in batch reactors. Results showed little enhancement in the total volume produced, pointing to good biodegradability of both aeroflotation and carcass wastes at mesophilic and thermophilic temperatures. However, the pretreatment led to a slight improvement in the initial reaction kinetics, indicating enhanced bio-availability due to chemical hydrolysis, particularly at the mesophilic temperatures. LCFA anaerobic biodegradation kinetics remained slow in all experimental conditions, even after hydrolysis. The effects observed for raw wastes with ethanol as their co-substrate depended on the temperature during the anaerobic process. Anaerobic batch tests applied to fatty waste were suitable for biogas potential determination. Still they remained insufficient for a clear evaluation of LCFA inhibition even after three waste additions. In that case the positive impact of the pretreatment was probably underestimated. Aiming at a better determination of anaerobic digestion of fatty waste, saponification should be assessed using semi-continuous reactors. In this configuration the
cumulative effects, such as grease accumulation, biomass inhibition or adaptation, should be stated.
Acknowledgements The authors would like to acknowledge the help from the French ‘‘Office de l’Elevage” (Livestock Service) and INTERBEV in funding this study.
References American Public Health Association, American Water Works Association, Water Pollution Control Federation, 1992. In: Clesceri, L.S., Greenberg, A.E., Trussel, R.R. (Eds.), Standard Methods for the Examination of Water and Wastewater, 18th ed. Battimelli, A., Torrijos, M., Delgenès, J.P., 2008. Fatty Slaughterhouse Waste Saponification to Increase Biogas Yield. International Water Association, International Symposium on Anaerobic Digestion of Solid Wastes and Energy Crops, Hammamet Tunisia, Proceeding Book, 8B. Bridoux, G., Dhulster, P., Manem, J., 1994. Analyse des graisses dans les stations d’épuration. Techniques Sciences Méthodes 5, 257–262. Canler, J.P., 2001. Performances des systèmes de traitement biologique aérobie des graisses – Graisses issues des dégraisseurs de station d’épuration traitant des effluents à dominances domestique. Cemagref édition, p. 64. Cirne, D.G., Paloumet, X., Bjornsson, L., Alves, M.M., Mattiasson, B., 2007. Anaerobic digestion of lipid-rich waste – effects of lipid concentration. Renewable Energy 32 (6), 965–975. Regulation (EC) No. 1774/2002 of the European Parliament and of the Council of 3 October 2002 laying down health rules concerning animal by-products not intended for human consumption. Official Journal of the European Communities, L.273/1–L273/95. Li, Y.Y., Sasaki, H., Yamashita, K., Seki, K., Kamigochi, I., 2002. High-rate methane fermentation of lipid-rich food wastes by a high-solids co-digestion process. Water Science and Technology 45 (12), 143–150. López Torres, M., Espinosa Lloréns, M.d.C., 2008. Effect of alkaline pretreatment on anaerobic digestion of solid wastes. Waste Management 28 (11), 2229–2234. Mouneimne, A.H., Carrère, H., Bernet, N., Delgenès, J.P., 2003. Effect of saponification on the anaerobic digestion of solid fatty residues. Bioresource Technology 90 (1), 89–94. Rinzema, A., Boone, M., Van Knippenberg, K., Lettinga, G., 1994. Bactericidal effect of long chain fatty acids in anaerobic digestion. Water Environment Research 66 (1), 40–49.