- Email: [email protected]
Biomethanization of sugar beet byproduct by semi-continuous single digestion and co-digestion with cow manure
Accepted Manuscript Biomethanization of sugar beet byproduct by semi-continuous single digestion and co-digestion with cow manure Kaoutar Aboudi, Carlos José Álvarez-Gallego, Luis Isidoro Romero-García PII: DOI: Reference:
S0960-8524(15)01454-6 http://dx.doi.org/10.1016/j.biortech.2015.10.051 BITE 15676
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
1 September 2015 6 October 2015 7 October 2015
Please cite this article as: Aboudi, K., Álvarez-Gallego, C.J., Romero-García, L.I., Biomethanization of sugar beet byproduct by semi-continuous single digestion and co-digestion with cow manure, Bioresource Technology (2015), doi: http://dx.doi.org/10.1016/j.biortech.2015.10.051
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Biomethanization of sugar beet byproduct by semi-continuous single digestion and co-digestion with cow manure Kaoutar Aboudi*, Carlos José Álvarez-Gallego and Luis Isidoro Romero-García Department of Chemical Engineering and Food Technology. Faculty of Sciences. Agrifood Campus of International Excellence (CeiA3). University of Cádiz. 11510, Puerto Real. Cadiz. Spain. ([email protected]; [email protected]; [email protected]) *
Corresponding author: Kaoutar Aboudi
Email: [email protected] Tel.+34956016474 Fax.+34956016411
1
Abstract Dried pellet of exhausted sugar beet cossettes were digested alone and combined with cow manure as co-substrate in a mesophilic semi-continuous anaerobic system. In single digestion assay, the stable biogas production and stable reactor operation was observed at the hydraulic retention time (HRT) of 20 days (OLR: 3.26 gVS/Lreactor d) which was the minimum HRT tolerated by the system. However, co-digestion with cow manure allowed to decrease the HRT until 15 days (OLR: 4.97 gVS/Lreactor d) with 32% higher biogas generation and efficient reactor operation. Propionic acid was the predominant VFA observed during single digestion assay failure, while acetic acid accumulation was observed in the co-digestion assay. In both single and co-digestion assays, the recovery of digesters was possible by ceasing the feeding and re-inoculation with a well-adapted inoculum.
Keywords Semi-continuous anaerobic digestion; Sugar beet byproduct; Cow manure; Codigestion; Biogas production
2
1. Introduction Energy recovery is a good option to optimize the use of organic wastes and agroindustrial byproducts. The research interest in bio-conversion of organic biomass into renewable energy has been increased markedly in recent years mainly due to reducing fossil fuels reserves, hike in fossil fuels prices and climate change (building up of greenhouse gases in atmosphere) issues (Pucker et al., 2013). Agro-industrial wastes, mainly composed of lignocellulosic material, are derived from the processing of a particular crop or livestock product. Agro-industrial byproducts have shown to be a potential biomass to produce clean energy using anaerobic digestion (AD) process (Anwar et al., 2014; Mata-Alvarez et al., 2014; Ward et al., 2008). Sugar beet byproducts (SBB) generated from sugar beet processing industries is one of the hugely produced and underutilized agro-industrial wastes. SBB are usually destined to use as animal feedstock, however, these byproducts can be utilized to produce clean energy and to offset the high costs of the energy required for extraction and production processes in the industry (Aboudi et al., 2015a; Brooks et al., 2008; Hutnan et al., 2000). Nevertheless, lignocellulosic material degradation could limit the AD process performance due to the complexity of this material (Palmqvist and Hahn-Hägerdal, 2000; Yang et al., 2015). Hence, pretreatment or co-digestion (Aco-D) methods can be used to enhance the anaerobic digestibility of lignocellulosic wastes (Mata-Alvarez et al., 2014; Ward et al., 2008; Yang et al., 2015). Aco-D of carbohydrates rich lignocellulosic wastes with animal manures have demonstrated to be a good option to treat both wastes by balancing the nutrient contents of each type of substrate (Comino et al., 2010; Mata-Alvarez et al., 2014; Sanaei-Moghadam et al., 2014). It was reported
3
that manure can provide high buffering capacity to anaerobic systems, while carbohydrates provide high organic biodegradable material (Comino et al., 2010; Wang et al., 2012). In this research, anaerobic digestion of sugar beet byproduct alone (single digestion) and with cow manure (co-digestion) was carried out in a semi-continuous feeding system. The effect of single digestion and co-digestion was studied with special reference to minimize the hydraulic retention time (HRT) and maximize the organic loading rate (OLR). In previous study, anaerobic co-digestion of SBB with pig manure (PM) was carried out successfully (Aboudi et al., 2015b). The present attempt to use cow manure was mainly due to the difference in the characteristics of cow and pig manures. Differences in both the physico-chemical characteristics and the composition of the microbiota of the gastrointestinal ecosystem have shown to occur between ruminants (cows) and mono-gastric (pigs) animals (McGillivery and Cranwell, 1992; Savage, 1977). There have been several studies about co-digestion of cow manure with crop residues, few of them were using SBB (tops, leaves or wet pulp). However, these codigestion studies were mostly carried out in batch and not with ESBC-DP. Moreover, there is little information available of the effect of OLR and HRT on co-digestion of ESBC-DP with manures in a semi-continuous experiment. This study investigated the start-up and long term operation of mesophilic semi-continuous anaerobic digesters treating SBB alone and with cow manure as co-substrate. 2. Material and Methods 2.1. Substrates origin and characteristics
4
The dried pellet of exhausted sugar beet cossettes (ESBC-DP) used as SBB in this study were collected from a sugar beet processing factory situated at Jerez de la Frontera, at southern of Spain. ESBC-DP was stored at 4ºC to avoid its degradation. The ESBC-DP was composed of 85% of pulp and 15% of molasses and have a total solids content of 80-90%. Fresh cow manure (CM) was collected from a farm facility at El Puerto de Santa María, in the same county. The CM was collected periodically and stored at 4ºC. The physico-chemical characteristics of ESBC-DP and CM are summarized in Table 1. 2.2. Anaerobic semi-continuous digesters design and set-up Three semi-continuous stirred tank reactors (SSTR) made of stainless steel having a working volume of 10L were used in this study. The reactors were operated at mesophilic temperature (35ºC) which it was maintained by a thermostatic recirculating bath (Aboudi et al., 2015b). The reactors R1 and R2 were used for anaerobic single digestion of ESBC-DP and the reactor R3 was used for co-digestion of ESBC-DP with CM. The reactor R1 was started up by adding 5L inoculum, which was collected from a mesophilic anaerobic digester treating organic fraction of municipal solid waste (OFMSW) (Fernández Rodríguez et al., 2012). Subsequently, a daily feeding strategy with pellets of ESBC at HRT of 20 days and total solid content of 8% was adopted (Hutnan et al., 2000). Once the total working volume of the reactor (10L) was achieved, measurement of biogas production and composition, volatile fatty acids (VFA), pH and other parameters evolution was carried out. Moreover, reactor R2 was started by adding 10L of the effluent collected from the reactor R1 during a steady state condition. In this case, the feeding regime was started with HRT of 20 days and keeps running with successive and progressive decrease in HRT.
5
Five operational periods were tested for reactor R2 corresponding to change in HRT and recovery stages (Table 2). The recovery periods included two extended nonfed stages and the subsequent inoculation with active inoculum. For the co-digestion reactor R3, the start-up was initiated by using 10L of the effluent collected from R2 in stable conditions. The reactor R3 was fed with a mixture of ESBC-DP and CM as co-substrate. The total solids (TS) content and carbon to nitrogen ratio (C/N) of mixture of ESBC-DP and CM were adjusted to 8% and 18.5, respectively, as have been reported in a previous study (Aboudi et al., 2015b). 2.3. Analytical methods Analytical parameters were analyzed according to the Standard Methods (APHA, 2005) and the particular method for each parameter is indicated in parenthesis. Total solids (WPFC-2540-B), volatile solids (VS) (WPFC-2540-E), dissolved organic carbon (DOC) (WPFC-5310-B), total and soluble chemical oxygen demand (tCOD, sCOD) (WPFC-5220-C), pH (WPFC-4500-H+), alkalinity (WPCF-2330) and ammonium (NH4+-N) (WPFC-4500-NH3C). Samples for the sCOD and DOC analysis were lixiviated with deionized water for 2 h and filtered through a 0.47 µm glass microfiber filter. For VFAs analysis, samples from lixiviation and 0.47 µm filtration, were filtrated again through a 0.22 µm Teflon filter (Álvarez-Gallego, 2005) and analyzed by a gas chromatograph (Shimadzu GC-2010) equipped with a flame ionization detector (FID) and capillary column filled with Nukol® (diameter of 0.25 µm and 30 m length). The dissolved organic carbon (DOC) was determined by using an Analytic-Jena multi N/C 3100 carbon analyzer with chemiluminescence detector (CLD) according to combustion-infrared method (5310-B) of Standard Methods (APHA, 2005). The oxidizing gas was oxygen at a pressure of 4-6 bars. Volume of biogas produced was collected in a 40L Tedlar gas bag (SKC®) and was daily measured by a high precision 6
drum-type gas meter (Ritter® TG5). The composition of the biogas generated (including CH4, CO2, and H2) was determined by using a gas chromatograph (Shimadzu® GC-2014) with a stainless steel column packed with Carbosieve® SII (diameter of 3.2 mm and 3.0 m length) and thermal conductivity detector (TCD). The injected sample volume was 1 mL, and the operational conditions were as follows: 7 min at 55 °C; ramped at 27 °C/min until 150 °C; detector temperature, 255 °C; injector temperature, 100 °C. Helium was used as carrier gas with a flow rate of 30 mL/min. 3. Results and discussion 3.1. Substrates and feedstock characteristics The physico-chemical characteristics of the substrates and average values of the single and co-digestion feedstocks are presented in Table 1. ESBC-DP and CM contained 89.5 % and 76.8 % of organic matter (OM) (measured as VS), and a tCOD value of 146 g/kg and 75 g/kg, respectively. The high organic content makes ESBC-DP an attractive substrate for biogas production. Nevertheless, ESBC-DP had lower amount of nitrogen and alkalinity than CM (17 and 12.3 fold lower, respectively). Thus, the use of CM as co-substrate could counterbalance the nutrients deficiency of the lignocelluosic-type waste (ESBC-DP) offering sufficient buffer capacity to the anaerobic system (Wang et al., 2012). It was reported that manure addition is more cost effective for nutritional regulation than the use of chemicals reagents (Comino et al., 2010). The C/N ratio of ESBC-DP was found higher in comparison with CM. As reported in the literature, the suitable C/N ratio for a proper operation of AD process is ranged of 20-30 (Li et al., 2011). However, other authors suggested lower values for lignocellulosic-type wastes (Wang et al., 2012). In the present study, C/N ratio of 18.5 was established for the co-digestion assay of ESBC7
DP and CM as the best mixture ratio reported in a previous study using similar wastes in batch regime (Aboudi et al., 2015a). 3.2. Start-up and operation of the semi-continuous single digestion reactors: R1 and R2 The first part of this research was aimed to evaluate the start-up, operation and optimization of the anaerobic digestion of ESBC-DP in a SSTR by studying a decreasing regime of HRTs (corresponding to an increasing regime of OLRs) to find out the best operation conditions treating ESBC-DP as a sole substrate. The semicontinuous reactor R1 was operating at a fixed HRT of 20 days for a long term period of more than one year, which allowed to achieve steady state conditions and adapted biomass for ESBC-DP treatment. In the second semi-continuous reactor R2, it was attempted to decrease the HRT of 20 days gradually by testing a series of HRT to reach the maximum organic loading rate tolerable by the system to treat ESBC-DP individually (Table 2). 3.2.1. Biogas production and VFA evolution during operation of ESBC-DP single digestion: R1 and R2 Fig. 1-A shows the daily biogas production observed in R1 and R2 during the mesophilic semi-continuous assays. The reactor R1 has shown a stable biogas production of 1.50±0.21 L/Lreactor d with 55.2±4.8 % of methane content during the whole operational period at HRT of 20 days (named period 0). Biogas production from R2 has shown different trend pursuant to HRT changes. At the start-up period of about 60 days (period I), R2 showed similar behavior to R1 with stable daily biogas production around 1.54±0.18 L/Lreactor d. The decrease of HRT to 15 days does not show any disturbances at the beginning and biogas production 8
has even risen to 1.88±0.09 L/Lreactor d according to the OLR increase. However, there was a significant decline in biogas production after about 60 days of operation (period II) (Fig. 1-A). A gradual accumulation of VFAs was observed (Fig. 1-B) which may be the reason of biogas production decrease as a result of partial methanogenic activity inhibition. Hence, in order to prevent a complete inhibition of the anaerobic activity due to acidification, it was decided to stop feeding to the reactor with the aim to give microorganism enough time to degrade the accumulated VFAs in the medium. However, no significant drop in VFAs values was observed even after 20 days, and biogas production was gradually diminishing. The process distortion during this period was also corroborated by the high acidity/alkalinity ratio of 2.47 (Table 3) which was not in the admissible range (0.4-0.8), as reported in the literature for a stable AD process (Chen et al., 2008). Overloading the reactor affected the buffer capacity of the system which was not able to neutralize the VFAs released. Thus, the digester was reinoculated with an active effluent coming from the stable reactor R1. This strategy effectively has permitted to decrease the VFAs accumulated and to recover the system. After this failure episode the feeding was restarted again at a HRT of 20 days before decreasing to 18 days (period IV) with the aim to investigate if the gradual decrease in HRT will allow to avoid the acidification and system failure. AD of easily biodegradable material allows to increase the OLR rapidly, however, when the deal is with a complex lignocellulosic substrate, this strategy must be avoided and a gradual and slow OLR increase should be adopted in order to provide enough time to microorganism for adapting to the new OLR. After 20 days of operation (HRT: 18 days), biogas production dropped again and an increase in VFAs concentration was observed (Fig. 1-A and B). During this period,
9
the feeding was stopped again to digester in order to recover the reactor at the previous HRT of 20 days (period V). The main VFAs produced in R2 were propionic (HPr) and acetic acid (HAc). The higher values of HPr acid were detected during system failure at HRT of 15 days (period II). Fig. 2-A shows the maximum concentrations of both HAc and HPr during the study periods (including recovery stages). Moreover, Fig. 2-B shows the maximum and average values of the ratio between propionic and acetic acids (HPr/HAc). It has been reported that the ratio HPr/HAc could give an earlier indication of digester failure (Chen et al., 2008; Marchaim and Krause, 1993). The HPr/HAc values above 1.4 were considered as the threshold tolerated by an AD system without inhibition (Marchaim and Krause, 1993). On the other hand, propionic acid was widely considered as a strong inhibitor of anaerobic activities and it is difficult to remove from the system (Chen et al., 2008; Pullammanappallil et al., 2001). In the reactor R2, the maximum HPr/HAc ratios were observed at HRT of 15 days (period II) and during recovery period (above 40). However, this ratio was dropped to admissible value after consumption of VFAs (Marchaim and Krause, 1993). Methanogenic archaea are the most vulnerable to different environmental and operational factors in an AD system including organic overloading of the reactor which happened in this study (Chen et al, 2008). The use of an active inoculum with adapted biomass to the substrate allowed to degrade the VFAs accumulated and avoided the irreversible failure of the reactor R2. The role of the inoculum source and quality selection was reported to be crucial in AD processes (De Vrieze et al., 2015). The capacity of the inoculum used to recover the system was likely attributed to the
10
abundance and activity of methanogens and the adapted consortium of microorganisms to degrade ESBC-DP. Aboudi et al. (2015a) studied the batch anaerobic digestion of SBB and deduced that ESBC-DP released high VFAs concentrations in the medium, which delays the biogas production. Alkaya and Demirer (2011) also observed the system acidification when studying AD of sugar beet pulp in a SSTR with the aim to maximize VFA production from sugar beet wet pulp and sugar beet wastewater. The authors observed that the increase in sugar beet wet pulp amount in the feeding leads to the increase in the amount of acidification products (VFAs), which ultimately led to pH drop and inhibition of methanogenic activities. From the results of the current research and the above mentioned references, it could be deduced that SBB anaerobic digestion may cause the acidification of the reactor if not properly digested (i.e. as a sole substrate or at higher OLR). 3.2.2. Comparison of the single digestion performance parameters testing different HRTs Table 3 summarizes the performance of semi-continuous digestion reactors R1 and R2. The findings revealed that the specific methane production (SMP) of R2 at HRT of 15 days (period II) was very low, since the biogas production dropped drastically during this period. At 18 days HRT (period IV), the SMP was also found lower (only 122.8 mLCH4/gVSadded), due to the system failure at this particular HRT. Otherwise, SMP was observed similar in R1 and R2 at 20 days HRT. The methane content of the digesters R1 and R2 was above 50 % during whole periods, which was similar to the CH4 percentage reported by other authors studying AD of SBB (Brooks et al., 2008).
11
The highest VS removal was observed at 20 days HRT (periods 0, I, III and IV), while as the lowest one was obtained at HRT of 15 days (period II). During this period, the microbial activity was dropped and the OM was slowly or even not degraded due to the system acidification. HRT of 20 days was found to be the optimum HRT for treating ESBC-DP as a sole substrate under the studied conditions. AD of agro-industrial byproducts and crop residues individually usually implies the operation at higher HRT (Brooks et al., 2008; Demirel and Scherer, 2008; Moraes et al., 2015). Demirel and Scherer (2008) studied the AD of sugar beet silage using SSTR and reported that the highest methane generation was obtained at HRT of 25 days. In an earlier work of Labat et al. (1984), HRT of 17 days was found to be the best for anaerobic digestion of the sugar beet pulp (SBP) after a pretreatment of enzymatic hydrolysis. 3.3. Semi-continuous co-digestion reactor of ESBC-DP and CM (R3) In the second stage of this study, the effect of ESBC-DP and CM co-digestion on biogas production and stability of reactor was studied. At this stage, a decreasing sequence in HRTs i.e. 20, 18, 15 and 12 days corresponding to OLRs of 3.73, 4.14, 4.97 and 6.21 gVS/Lreactor d were tested and compared with each other and with results from single digestion assay (Table 2). 3.3.1. Biogas generation from the semi-continuous co-digestion reactor of ESBCDP and CM The daily biogas production from the SSTR fed with ESBC-DP and CM at different HRT tested is illustrated in Fig. 3-A. During the start-up period, some fluctuations in biogas production were observed as a consequence of biomass adaptation to the new cosubstrate (CM). However, a stable biogas production of about 1.61±0.24 L/Lreactor d was 12
observed after 20 days of operation. The biogas production was slightly higher than it was measured at the same HRT in single digestion assay. Upon decreasing the HRT to 18 days and 15 days (periods II and III, respectively), the biogas production was increased to 1.81±0.15 L/Lreactor d and 2.08±0.26 L/Lreactor d, respectively. However, once decreasing the HRT to 12 days (period IV), the biogas production started to decrease gradually accompanied by a pH drop and VFAs increase in the digester (Fig. 4-A). Although, the pH was adjusted by adding an alkali (NaOH-8M), the pH could not remain in the appropriate range for AD process and continued decreasing, which ultimately halted the methanogenic activity due to overloading of the reactor. At this stage, the feeding to reactor was ceased for several days to avoid total failure of the reactor, followed by inoculation with 2L of effluent coming from the SSTR R2, which was adapted for VFAs degradation. This practice allowed to recover the reactor and the accumulated VFAs were degraded (Fig. 4-A). Subsequently, the feeding was restarted at the previous HRT of 15 days and the system was operated under these conditions for a long period to ensure the stability of the digester. Umetsu et al. (2006) studied thermophilic Aco-D of SBB (tops and roots) with dairy manure at HRT of 20 days. They reported that the highest daily biogas production of 2.46 L/Lreactor d with 57.9 % of CH4 concentration was obtained from dairy manure and 40% beet top which was a higher production compared to the findings of the present study at the same HRT of 20 days. Fang et al. (2011) studied Aco-D of SBB (including wet pulp) with and without cow manure in both batch and SSTR conditions. SBP was found to be a good substrate for biogas production and the SMP of 280 mLCH4/gVSadded was obtained in a thermophilic continuously operated reactor, codigesting 50% of SBP with cow manure. The authors used CM to dilute toxic components contained in SBB and considering that CM provide the necessary buffering 13
capacity and trace element to the AD system. Aboudi et al. (2015b) studied Aco-D of ESBC with PM using a mesophilic SSTR. They reported that it was possible to decrease the HRT until 6 days (OLR of 11.2 gVS/Lreactor d) without any inhibition and obtaining high methane production rate (MPR) of 2.9 LCH4/Lreactor d, which was 2.4 fold higher than those obtained in the present study using CM as co-substrate and 3.5 fold higher than those obtained from single digestion assay of ESBC-DP. The results obtained by co-digestion with PM were better than those obtained in the present work using CM as co-substrate. This may be attributed to the differences in the manures used (PM and CM). PM provided more buffering capacity than CM, and this was likely the reason of the possibility to increase OLR in PM co-digestion assay. The physico-chemical characteristics of CM and those of the PM used in the previous study (Aboudi et al., 2015b) showed that PM has higher alkalinity and nitrogen content than CM. On the other hand, probably, the variety of microorganisms provided by each manure may be different (Savage et al., 1977). It was expected that co-digestion with CM would enhance the AD process more than that with PM due to the characteristics of the digestive tract of ruminants (cows) as herbivores are used to degrade vegetal fiber, however it was shown -at least in semi-continuous system and the studied conditions-, that alkalinity supply from the manure type, played the prevalent role in the process improvement. It should be noted that the used manure was stored at 4ºC when it was collected previously to use it in the feed, which likely would affect the autochthonous flora of the substrate. The data for the current co-digestion assay may be also compared with other works using CM as the main substrate and different crop co-substrates. Lehtomaki et al., (2007) examined the AcoD of CM with three different crop residues including sugar beet tops (SBT). A SMP of 229 mlCH4/gVSadded was obtained from mixture of 40% 14
SBT with CM, which was 65% higher than the methane yield from CM individually. However, the SMP they obtained was lower than that found in the present work at the same HRT of 20 days but lowers OLR (2 gVS/Lreactor d). Moreover, and contrastingly to this research, a decreasing in HRT from 20 to 16 days led to 26% decrease in SMP. Authors reported that in their studied conditions, shorter HRTs than 20 days were not sufficient for OM degradation and a high undegraded OM amount was found in the digestate. In an earlier study of Hills, (1980), Aco-D of CM with barley straw (80% of CM in the mixture) was carried out in a SSTR and a decreasing regime of HRT from 25 days to 10 days was tested. The highest SMP of 170 mlCH4/gVSadded was obtained at 20 days-HRT while methane productions decreased with OLR increase and distortions of the system were observed at lower HRTs. In the present study, we reach to decrease HRT successfully until 15 days with higher methane production and system performance. The biogas composition (H2, CH4 and CO2) during the SSTR co-digestion process is shown in Fig. 3-B. As can be seen, the low H2 production for about 3 days during the adaptation period (start-up), was due to the acclimation of the microorganisms to the new co-substrate, however, the pH correction was done using an alkali (NaOH-8M). Afterwards, higher CH4 content was observed in biogas in comparison with CO2 content at HRTs of 20 days and 18 days (periods I and II, respectively). The average CH4 concentrations of 57.7± 1.4 % and 55.5±1.1 % were observed for HRT of 20 days and 18 days, respectively, while CO2 concentration did not rise more than 45%. At HRT of 15 days (period III), the CH4 concentration was also higher but the concentrations of both CH4 and CO2 were closer. This fact could indicate that even the digester was operating conveniently at the HRT of 15 days, the system was closer to the OLR limit tolerated by the digester. This fact was confirmed when 15
CO2 concentrations increased significantly and exceeded the CH4 concentrations due to acidification and high liberation of CO2 upon decreasing the HRT to 12 days (period IV). 3.3.2. pH and VFAs evolution in the co-digestion reactor (R3) Acidogens are fast growers and very little sensitive to the environment conditions variations, while methanogens are inherently slow growers, and are inhibited at acidic pH values (Chen et al., 2008; Pullammanappallil et al., 2001). Thus, to achieve overall high rates, both the production and consumption of acids should be balanced and an imbalance would cause a slowing overall degradation of waste (Ahring et al., 1995; Demirel and Scherer, 2008). In the present study and during co-digestion of ESBC-DP and CM (R3), an imbalance between VFAs production and degradation occurred leading to acidification phenomenon that inhibited the process at HRT of 12 days (period IV). Thus, on 204th day of assay, the VFAs started to accumulate gradually from 357 mg/L to 5296 mg/L, and the best strategy was to stop feeding the reactor during this period. The main VFAs produced during the failure period were acetic and propionic acids with higher values for HAc than HPr (Fig. 4-B). Acetic acid accumulation indicates that acetoclastic methanogenesis was stressed (Ahring et al., 1995), which was traduced by the decrease in methane production during this period. For the rest of the stage, VFAs concentrations were very low, indicating that acidogens, acetogens and methanogens activities were balanced (Ahring et al., 1995; Vavilin et al., 1997). From other hand, it was found that the system failure was reversible and the recovery period was relatively short, mainly due to the predominance of HAc, which is
16
more easy to degrade than other organic acids especially HPr (Ahring et al., 1995; Marchaim and Krause, 1993). The two ratios HPr/HAc and acidity/alkalinity related to the system stability for each HRT tested are shown in Table 4. The HPr/HAc ratio was found in the appropriate range at all HRTs tested (Ahring et al., 1995; Marchaim and Krause, 1993). HAc concentration was higher than HPr which permitted the quick recovery of the digester as commented before. Nevertheless, the acidity/alkalinity ratio was higher than admissible value of 0.4 (Chen et al., 2008) at HRT of 12 days (period IV). 3.3.3. Organic matter evolution and removal Organic matter removal (VS degradation) is shown in Table 4. The findings revealed that the highest VS removal was obtained at HRT of 18 days. However, the VS removal was noted around 54 % at HRTs of 20 days and 15 days. The VS removal was 34.3% at the critical HRT of 12 days due to system failure. The findings suggested that 12 days-HRT was the threshold to carry out the anaerobic co-digestion of ESBC-DP with CM under stable conditions. Conclusions It has been demonstrated that AD of ESBC-DP in semi-continuous regime was feasible at HRT of 20 days (OLR: 3.26 gVS/Lreactor d) and propionic acid accumulation was observed when this limit is overtaken. However, co-digestion of ESBC-DP with CM allowed to decrease the HRT to 15 days (OLR: 4.97 gVS/Lreactor d) with high methane generation and without any process inhibition. HRT of 12 days was found critical for co-digestion assay leading to pH drop accompanied by VFAs accumulation. On the other hand, the digester recovery after failure was possible due to the well-adapted inoculum coming from the single digestion reactor. 17
Acknowledgements This work was financially supported by the Spanish Ministry of Economy and Finance (Project CTM2013-43938-R) and the MICINN (Project UNCA08-1E-035) and the European Regional Development Funds (ERDF). Authors also acknowledge support from the University of Cádiz in Spain for the PhD scholarship (UCA-2010063PU/EPIF-FPI-A/BC). Authors thank the Agrifood Campus of International Excellence (Ceia3) and the company (AB-sugars group) for providing the ESBC-DP used in this research. References Aboudi, K., Álvarez-Gallego, C.J., Romero-García, L.I. 2015a. Improvement of exhausted sugar beet cossettes anaerobic digestion process by Co-digestion with pig manure. Energy and Fuels, 29(2), 754-762. Aboudi, K., Álvarez-Gallego, C.J., Romero-García, L.I. 2015b. Semi-continuous anaerobic co-digestion of sugar beet byproduct and pig manure: Effect of the organic loading rate (OLR) on process performance. Bioresource Technology, 194, 283-290. Ahring, B.K., Sandberg, M., Angelidaki, I. 1995. Volatile fatty acids as indicators of process imbalance in anaerobic digesters. Applied Microbiology and Biotechnology, 43(3), 559-565. Alkaya, E., Demirer, G.N. 2011. Anaerobic acidification of sugar-beet processing wastes: Effect of operational parameters. Biomass and Bioenergy, 35(1), 32-39. Álvarez-Gallego, C. 2005. Testing different procedures for the start-up of a dry anaerobic co-digestion process of OFMSW and sewage sludge at thermophilic range. in: PhD Thesis, University of Cádiz. Spain. Anwar, Z., Gulfraz, M., Irshad, M. 2014. Agro-industrial lignocellulosic biomass a key to unlock the future bio-energy: A brief review. Journal of Radiation Research and Applied Sciences, 7(2), 163-173. APHA, A., Standards Methods for the Examination of Water and Wastewater,. 2005. 20th edition, American Public Health Association Washington DC, USA
18
Brooks, L., Parravicini, V., Svardal, K., Kroiss, H., Prendl, L. 2008. Biogas from sugar beet press pulp as substitute of fossil fuel in sugar beet factories. Water Science and Technology, 58(7), 1497-1504. Chen, Y., Cheng, J.J., Creamer, K.S. 2008. Inhibition of anaerobic digestion process: A review. Bioresource Technology, 99(10), 4044-4064. Comino, E., Rosso, M., Riggio, V. 2010. Investigation of increasing organic loading rate in the co-digestion of energy crops and cow manure mix. Bioresource Technology, 101(9), 3013-3019. Demirel, B., Scherer, P. 2008. Production of methane from sugar beet silage without manure addition by a single-stage anaerobic digestion process. Biomass and Bioenergy, 32(3), 203-209. De Vrieze, J., Raport, L., Willems, B., Verbrugge, S., Volcke, E., Meers, E., Angenent, L.T., Boon, N. 2015. Inoculum selection influences the biochemical methane potential of agro-industrial substrates. Microbial Biotechnology, 8(5), 776-786. Fang, C., Boe, K., Angelidaki, I. 2011. Anaerobic co-digestion of by-products from sugar production with cow manure. Water Research, 45(11), 3473-3480. Fernández Rodríguez, J., Pérez, M., Romero, L.I. 2012. Mesophilic anaerobic digestion of the organic fraction of municipal solid waste: Optimisation of the semicontinuous process. Chemical Engineering Journal, 193–194(0), 10-15. Hill, D.J. 1980. Biogas from a high solids combination of dairy manure and barley Straw. Transactions of ASAE, 23(6), 1500-1504. Hutnan, M., Drtil, M., Mrafkova, L. 2000. Anaerobic biodegradation of sugar beet pulp. Biodegradation, 11(4), 203-211. Labat, M., Garcia, J.L., Meyer, F., Deschamps, F. 1984. Anaerobic digestion of sugar beet pulps. Biotechnology Letters, 6(6), 379-384. Lehtomäki, A., Huttunen, S., Rintala, J.A. 2007. Laboratory investigations on codigestion of energy crops and crop residues with cow manure for methane production: Effect of crop to manure ratio. Resources, Conservation and Recycling, 51(3), 591-609. Li, Y., Park, S.Y., Zhu, J. 2011. Solid-state anaerobic digestion for methane production from organic waste. Renewable and Sustainable Energy Reviews, 15(1), 821826. Marchaim, U., Krause, C. 1993. Propionic to acetic acid ratios in overloaded anaerobic digestion. BioresourceTechnology, 43(3), 195-203. 19
Mata-Alvarez, J., Dosta, J., Romero-Güiza, M.S., Fonoll, X., Peces, M., Astals, S. 2014. A critical review on anaerobic co-digestion achievements between 2010 and 2013. Renewable and Sustainable Energy Reviews, 36(0), 412-427. Moraes, B.S., Triolo, J.M., Lecona, V.P., Zaiat, M., Sommer, S.G. 2015. Biogas production within the bioethanol production chain: Use of co-substrates for anaerobic digestion of sugar beet vinasse. Bioresource Technology, 190, 227234. McGillervy, D.J., Cranwell, P. D. 1992. Anaerobic microflora associated with the pars oesophagea of the pig. Research in Veterinary Science, 53(1), 110-115. Palmqvist, E., Hahn-Hägerdal, B. 2000. Fermentation of lignocellulosic hydrolysates. II: Inhibitors and mechanisms of inhibition. Bioresource Technology, 74(1), 2533. Pucker, J., Jungmeier, G., Siegl, S., Pötsch, E.M. 2013. Anaerobic digestion of agricultural and other substrates-implications for greenhouse gas emissions. Animal, 7(2), 283-291. Pullammanappallil, P. C., Chynoweth, D. P., Lyberatos, G., Svoronos, S. A. 2001. Stable performance of anaerobic digestion in the presence of a high concentration of propionic acid. Bioresource Technology, 78(2), 165-169. Sanaei-Moghadam, A., Abbaspour-Fard, M.H., Aghel, H., Aghkhani, M.H., AbediniTorghabeh, J. 2014. Enhancement of biogas production by co-digestion of potato pulp with cow manure in a CSTR system. Applied Biochemistry and Biotechnology, 173(7), 1858-1869. Savage, D.C. 1977. Microbial Ecology of the Gastrointestinal Tract. Annual Review of Microbiology, 31(1), 107-133. Umetsu, K., Yamazaki, S., Kishimoto, T., Takahashi, J., Shibata, Y., Zhang, C., Misaki, T., Hamamoto, O., Ihara, I., Komiyama, M. 2006. Anaerobic co-digestion of dairy manure and sugar beets. International Congress Series, 1293(0), 307-310. Vavilin, V.A., Rytov, S.V., Lokshina, L.Y. 1997. A balance between hydrolysis and methanogenesis during the anaerobic digestion of organic matter. Microbiology, 66(6), 712-717. Wang, X., Yang, G., Feng, Y., Ren, G., Han, X. 2012. Optimizing feeding composition and carbon-nitrogen ratios for improved methane yield during anaerobic codigestion of dairy, chicken manure and wheat straw. Bioresource Technology, 120, 78-83. 20
Ward, A.J., Hobbs, P.J., Holliman, P.J., Jones, D.L. 2008. Optimisation of the anaerobic digestion of agricultural resources. Bioresource Technology, 99(17), 7928-7940. Yang, L., Xu, F., Ge, X., Li, Y. 2015. Challenges and strategies for solid-state anaerobic digestion of lignocellulosic biomass. Renewable and Sustainable Energy Reviews, 44(0), 824-834.
21
Figure captions Fig.1. (A) Daily biogas production and (B) TVFA evolution in the reactors R1 and R2. *Black lines: stages separation; Red lines: stop feeding; Blue lines: inoculation. Fig.2. (A) Maximum main VFAs produced and (B) Maximum and average values of HPr/HAc ratio in the reactors R1 and R2 Fig.3. (A) Daily biogas production and (B) Biogas composition in the co-digestion reactor R3. *Black lines: stages separation; Red lines: stop feeding; Blue lines: inoculation. Fig.4. (A) pH and TVFA evolutions and in the co-digestion reactor R3 and (B) Detailed VFAs evolution during failure period. *Black lines: stages separation; Red lines: stop feeding; Blue lines: inoculation.
22
Fig.1 (A) and (B)
23
Fig.2 (A) and (B)
24
Fig.3 (A) and (B)
25
Fig.4 (A) and (B)
26
Table 1. Physico-chemical characteristics of substrates (ESBC-DP and CM) and feedstocks (single digestion and co-digestion)
Parameters
ESBC-DP
CM
Feedstock (single digestion)
Feedstock (co-digestion)
pH
5.8 ± 0.7
6.3 ± 0.6
5.3 ± 1.1
6.5 ± 0.2
TS (g/kg)
873.3 ± 1.0
211.1 ± 32.8
74.0 ± 3.4
78.4 ± 5.8
VS (%TS)
89.5 ± 0.7
76.8 ± 11.3
52.1 ± 7.6
74.5 ± 5.3
tCOD (gO2/kg)*
146.0 ± 12.0
75.0 ± 4.0
73.0 ± 6.0
69.0 ± 8.0
sCOD
61.9 ± 8.0
17.2 ± 1.0
14.2 ± 3.3
14.5 ± 1.6
DOC (gC/kg)*
38.8 ± 7.5
5.5 ± 0.5
11.4 ± 3.6
4.8 ± 1.9
TVFA (gHAc/kg)*
1.8 ± 1.2
4.6 ± 0.5
1.5 ± 1. 2
0.6 ± 0.4
(gO2/kg)*
Alkalinity (gCaCO3/kg)*
3.1 ± 1.2
38.3 ± 11.1
2.8 ± 1.4
52.6 ± 4.7
NH4-N (gN/kg)*
0.2 ± 0.1
2.4 ± 0.6
0.3 ± 0.2
0.8 ± 0.4
TKN (gN/kg)*
12.8 ± 1.3
32.5 ± 2.8
10.3 ± 1.2
22.5 ± 3.5
C/N ratio
40.8 ± 1.0
13.7 ± 1.7
29.5 ± 1.2
18.7 ± 0.9
* g/kg (wet basis)
27
Table 2. HRTs and OLRs applied to reactors R1, R2 and R3 Reactor
Period
HRT
OLR (gVS/Lreactor d)
R1
0
20
2.95±1.35
R2
I
20
3.26±1.02
R3
II
15
4.35±0.23
III
20
3.26±1.02
IV
18
3.63±0.25
V
20
3.26±1.02
I
20
3.73±0.58
II
18
4.14±0.74
III
15
4.97±0.94
IV
12
6.21±1.01
V
15
4.97±0.94
28
Table 3. Summary of the performance parameters of the semi-continuous single digestion reactors R1 and R2
Reactor R2
R1
Period
HRT
MPR
(Days) (LCH4/Lreactor d)
SMP
CH4
HPr/HAc
(mLCH4/gVSadded)
(%)
ratio
Acidity/alkalinity VS removal (%) ratio
I
20
0.84±0.48
225.71±4.49
54.48±2.47
0.10±0.03
0.12±0.04
62.87±5.12
II
15
0.51±0.62
86.43±1.44
53.61±1.66
2.18±0.45
2.47±0.11
33.48±3.26
III
20
0.83±0.16
215.91±2.28
54.85±2.63
0.53±0.34
0.09±0.02
54.45±7.42
IV
18
0.47±0.08
122.86±1.18
53.45±2.26
3.74±1.81
0.65±0.06
41.72±4.68
V
20
0.85±0.21
208.41±2.14
53.65±1.59
0.54±0.22
0.14±0.02
57.48±2.52
0
20
0.83±0.09
214.64±6.12
55.22±4.76
0.94±0.35
0.18±0.11
57.63±14.75
29
Table 4. Summary of the performance parameters of the semi-continuous co-digestion reactor R3
Period
HRT (Days)
MPR (LCH4/Lreactor d)
SMP HPr/HAc (mLCH4/gVSadded) ratio
Acidity/alkalinity VS removal ratio (%)
I
20
0.88±0.09
242.59±4.26
0.78±0.58
0.02±0.01
54.82±1.45
II
18
1.09±0.06
252.82±10.02
0.25±0.04
0.03±0.02
62.63±1.87
III
15
1.23±0.11
313.98±15.47
0.13±0.06
0.02±0.01
53.41±5.12
IV
12
0.34±0.17
64.81±2.43
0.61±0.08
0.83±0.42
34.32±7.19
V
15
1.30±0.11
295.36±7.52
0.49±0.31
0.02±0.03
52.75±4.11
30
Highlights
Individual digestion of sugar beet byproduct is feasible in semi-continuous system
20-days hydraulic retention time was the threshold for the single digestion system
Co-digestion with cow manure allowed to reduce hydraulic retention time to 15days
Volatile fatty acids were accumulated during reactors failure
The source and quality of the inoculum used has permitted to recover the systems
31
S0960-8524(15)01454-6 http://dx.doi.org/10.1016/j.biortech.2015.10.051 BITE 15676
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
1 September 2015 6 October 2015 7 October 2015
Please cite this article as: Aboudi, K., Álvarez-Gallego, C.J., Romero-García, L.I., Biomethanization of sugar beet byproduct by semi-continuous single digestion and co-digestion with cow manure, Bioresource Technology (2015), doi: http://dx.doi.org/10.1016/j.biortech.2015.10.051
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Biomethanization of sugar beet byproduct by semi-continuous single digestion and co-digestion with cow manure Kaoutar Aboudi*, Carlos José Álvarez-Gallego and Luis Isidoro Romero-García Department of Chemical Engineering and Food Technology. Faculty of Sciences. Agrifood Campus of International Excellence (CeiA3). University of Cádiz. 11510, Puerto Real. Cadiz. Spain. ([email protected]; [email protected]; [email protected]) *
Corresponding author: Kaoutar Aboudi
Email: [email protected] Tel.+34956016474 Fax.+34956016411
1
Abstract Dried pellet of exhausted sugar beet cossettes were digested alone and combined with cow manure as co-substrate in a mesophilic semi-continuous anaerobic system. In single digestion assay, the stable biogas production and stable reactor operation was observed at the hydraulic retention time (HRT) of 20 days (OLR: 3.26 gVS/Lreactor d) which was the minimum HRT tolerated by the system. However, co-digestion with cow manure allowed to decrease the HRT until 15 days (OLR: 4.97 gVS/Lreactor d) with 32% higher biogas generation and efficient reactor operation. Propionic acid was the predominant VFA observed during single digestion assay failure, while acetic acid accumulation was observed in the co-digestion assay. In both single and co-digestion assays, the recovery of digesters was possible by ceasing the feeding and re-inoculation with a well-adapted inoculum.
Keywords Semi-continuous anaerobic digestion; Sugar beet byproduct; Cow manure; Codigestion; Biogas production
2
1. Introduction Energy recovery is a good option to optimize the use of organic wastes and agroindustrial byproducts. The research interest in bio-conversion of organic biomass into renewable energy has been increased markedly in recent years mainly due to reducing fossil fuels reserves, hike in fossil fuels prices and climate change (building up of greenhouse gases in atmosphere) issues (Pucker et al., 2013). Agro-industrial wastes, mainly composed of lignocellulosic material, are derived from the processing of a particular crop or livestock product. Agro-industrial byproducts have shown to be a potential biomass to produce clean energy using anaerobic digestion (AD) process (Anwar et al., 2014; Mata-Alvarez et al., 2014; Ward et al., 2008). Sugar beet byproducts (SBB) generated from sugar beet processing industries is one of the hugely produced and underutilized agro-industrial wastes. SBB are usually destined to use as animal feedstock, however, these byproducts can be utilized to produce clean energy and to offset the high costs of the energy required for extraction and production processes in the industry (Aboudi et al., 2015a; Brooks et al., 2008; Hutnan et al., 2000). Nevertheless, lignocellulosic material degradation could limit the AD process performance due to the complexity of this material (Palmqvist and Hahn-Hägerdal, 2000; Yang et al., 2015). Hence, pretreatment or co-digestion (Aco-D) methods can be used to enhance the anaerobic digestibility of lignocellulosic wastes (Mata-Alvarez et al., 2014; Ward et al., 2008; Yang et al., 2015). Aco-D of carbohydrates rich lignocellulosic wastes with animal manures have demonstrated to be a good option to treat both wastes by balancing the nutrient contents of each type of substrate (Comino et al., 2010; Mata-Alvarez et al., 2014; Sanaei-Moghadam et al., 2014). It was reported
3
that manure can provide high buffering capacity to anaerobic systems, while carbohydrates provide high organic biodegradable material (Comino et al., 2010; Wang et al., 2012). In this research, anaerobic digestion of sugar beet byproduct alone (single digestion) and with cow manure (co-digestion) was carried out in a semi-continuous feeding system. The effect of single digestion and co-digestion was studied with special reference to minimize the hydraulic retention time (HRT) and maximize the organic loading rate (OLR). In previous study, anaerobic co-digestion of SBB with pig manure (PM) was carried out successfully (Aboudi et al., 2015b). The present attempt to use cow manure was mainly due to the difference in the characteristics of cow and pig manures. Differences in both the physico-chemical characteristics and the composition of the microbiota of the gastrointestinal ecosystem have shown to occur between ruminants (cows) and mono-gastric (pigs) animals (McGillivery and Cranwell, 1992; Savage, 1977). There have been several studies about co-digestion of cow manure with crop residues, few of them were using SBB (tops, leaves or wet pulp). However, these codigestion studies were mostly carried out in batch and not with ESBC-DP. Moreover, there is little information available of the effect of OLR and HRT on co-digestion of ESBC-DP with manures in a semi-continuous experiment. This study investigated the start-up and long term operation of mesophilic semi-continuous anaerobic digesters treating SBB alone and with cow manure as co-substrate. 2. Material and Methods 2.1. Substrates origin and characteristics
4
The dried pellet of exhausted sugar beet cossettes (ESBC-DP) used as SBB in this study were collected from a sugar beet processing factory situated at Jerez de la Frontera, at southern of Spain. ESBC-DP was stored at 4ºC to avoid its degradation. The ESBC-DP was composed of 85% of pulp and 15% of molasses and have a total solids content of 80-90%. Fresh cow manure (CM) was collected from a farm facility at El Puerto de Santa María, in the same county. The CM was collected periodically and stored at 4ºC. The physico-chemical characteristics of ESBC-DP and CM are summarized in Table 1. 2.2. Anaerobic semi-continuous digesters design and set-up Three semi-continuous stirred tank reactors (SSTR) made of stainless steel having a working volume of 10L were used in this study. The reactors were operated at mesophilic temperature (35ºC) which it was maintained by a thermostatic recirculating bath (Aboudi et al., 2015b). The reactors R1 and R2 were used for anaerobic single digestion of ESBC-DP and the reactor R3 was used for co-digestion of ESBC-DP with CM. The reactor R1 was started up by adding 5L inoculum, which was collected from a mesophilic anaerobic digester treating organic fraction of municipal solid waste (OFMSW) (Fernández Rodríguez et al., 2012). Subsequently, a daily feeding strategy with pellets of ESBC at HRT of 20 days and total solid content of 8% was adopted (Hutnan et al., 2000). Once the total working volume of the reactor (10L) was achieved, measurement of biogas production and composition, volatile fatty acids (VFA), pH and other parameters evolution was carried out. Moreover, reactor R2 was started by adding 10L of the effluent collected from the reactor R1 during a steady state condition. In this case, the feeding regime was started with HRT of 20 days and keeps running with successive and progressive decrease in HRT.
5
Five operational periods were tested for reactor R2 corresponding to change in HRT and recovery stages (Table 2). The recovery periods included two extended nonfed stages and the subsequent inoculation with active inoculum. For the co-digestion reactor R3, the start-up was initiated by using 10L of the effluent collected from R2 in stable conditions. The reactor R3 was fed with a mixture of ESBC-DP and CM as co-substrate. The total solids (TS) content and carbon to nitrogen ratio (C/N) of mixture of ESBC-DP and CM were adjusted to 8% and 18.5, respectively, as have been reported in a previous study (Aboudi et al., 2015b). 2.3. Analytical methods Analytical parameters were analyzed according to the Standard Methods (APHA, 2005) and the particular method for each parameter is indicated in parenthesis. Total solids (WPFC-2540-B), volatile solids (VS) (WPFC-2540-E), dissolved organic carbon (DOC) (WPFC-5310-B), total and soluble chemical oxygen demand (tCOD, sCOD) (WPFC-5220-C), pH (WPFC-4500-H+), alkalinity (WPCF-2330) and ammonium (NH4+-N) (WPFC-4500-NH3C). Samples for the sCOD and DOC analysis were lixiviated with deionized water for 2 h and filtered through a 0.47 µm glass microfiber filter. For VFAs analysis, samples from lixiviation and 0.47 µm filtration, were filtrated again through a 0.22 µm Teflon filter (Álvarez-Gallego, 2005) and analyzed by a gas chromatograph (Shimadzu GC-2010) equipped with a flame ionization detector (FID) and capillary column filled with Nukol® (diameter of 0.25 µm and 30 m length). The dissolved organic carbon (DOC) was determined by using an Analytic-Jena multi N/C 3100 carbon analyzer with chemiluminescence detector (CLD) according to combustion-infrared method (5310-B) of Standard Methods (APHA, 2005). The oxidizing gas was oxygen at a pressure of 4-6 bars. Volume of biogas produced was collected in a 40L Tedlar gas bag (SKC®) and was daily measured by a high precision 6
drum-type gas meter (Ritter® TG5). The composition of the biogas generated (including CH4, CO2, and H2) was determined by using a gas chromatograph (Shimadzu® GC-2014) with a stainless steel column packed with Carbosieve® SII (diameter of 3.2 mm and 3.0 m length) and thermal conductivity detector (TCD). The injected sample volume was 1 mL, and the operational conditions were as follows: 7 min at 55 °C; ramped at 27 °C/min until 150 °C; detector temperature, 255 °C; injector temperature, 100 °C. Helium was used as carrier gas with a flow rate of 30 mL/min. 3. Results and discussion 3.1. Substrates and feedstock characteristics The physico-chemical characteristics of the substrates and average values of the single and co-digestion feedstocks are presented in Table 1. ESBC-DP and CM contained 89.5 % and 76.8 % of organic matter (OM) (measured as VS), and a tCOD value of 146 g/kg and 75 g/kg, respectively. The high organic content makes ESBC-DP an attractive substrate for biogas production. Nevertheless, ESBC-DP had lower amount of nitrogen and alkalinity than CM (17 and 12.3 fold lower, respectively). Thus, the use of CM as co-substrate could counterbalance the nutrients deficiency of the lignocelluosic-type waste (ESBC-DP) offering sufficient buffer capacity to the anaerobic system (Wang et al., 2012). It was reported that manure addition is more cost effective for nutritional regulation than the use of chemicals reagents (Comino et al., 2010). The C/N ratio of ESBC-DP was found higher in comparison with CM. As reported in the literature, the suitable C/N ratio for a proper operation of AD process is ranged of 20-30 (Li et al., 2011). However, other authors suggested lower values for lignocellulosic-type wastes (Wang et al., 2012). In the present study, C/N ratio of 18.5 was established for the co-digestion assay of ESBC7
DP and CM as the best mixture ratio reported in a previous study using similar wastes in batch regime (Aboudi et al., 2015a). 3.2. Start-up and operation of the semi-continuous single digestion reactors: R1 and R2 The first part of this research was aimed to evaluate the start-up, operation and optimization of the anaerobic digestion of ESBC-DP in a SSTR by studying a decreasing regime of HRTs (corresponding to an increasing regime of OLRs) to find out the best operation conditions treating ESBC-DP as a sole substrate. The semicontinuous reactor R1 was operating at a fixed HRT of 20 days for a long term period of more than one year, which allowed to achieve steady state conditions and adapted biomass for ESBC-DP treatment. In the second semi-continuous reactor R2, it was attempted to decrease the HRT of 20 days gradually by testing a series of HRT to reach the maximum organic loading rate tolerable by the system to treat ESBC-DP individually (Table 2). 3.2.1. Biogas production and VFA evolution during operation of ESBC-DP single digestion: R1 and R2 Fig. 1-A shows the daily biogas production observed in R1 and R2 during the mesophilic semi-continuous assays. The reactor R1 has shown a stable biogas production of 1.50±0.21 L/Lreactor d with 55.2±4.8 % of methane content during the whole operational period at HRT of 20 days (named period 0). Biogas production from R2 has shown different trend pursuant to HRT changes. At the start-up period of about 60 days (period I), R2 showed similar behavior to R1 with stable daily biogas production around 1.54±0.18 L/Lreactor d. The decrease of HRT to 15 days does not show any disturbances at the beginning and biogas production 8
has even risen to 1.88±0.09 L/Lreactor d according to the OLR increase. However, there was a significant decline in biogas production after about 60 days of operation (period II) (Fig. 1-A). A gradual accumulation of VFAs was observed (Fig. 1-B) which may be the reason of biogas production decrease as a result of partial methanogenic activity inhibition. Hence, in order to prevent a complete inhibition of the anaerobic activity due to acidification, it was decided to stop feeding to the reactor with the aim to give microorganism enough time to degrade the accumulated VFAs in the medium. However, no significant drop in VFAs values was observed even after 20 days, and biogas production was gradually diminishing. The process distortion during this period was also corroborated by the high acidity/alkalinity ratio of 2.47 (Table 3) which was not in the admissible range (0.4-0.8), as reported in the literature for a stable AD process (Chen et al., 2008). Overloading the reactor affected the buffer capacity of the system which was not able to neutralize the VFAs released. Thus, the digester was reinoculated with an active effluent coming from the stable reactor R1. This strategy effectively has permitted to decrease the VFAs accumulated and to recover the system. After this failure episode the feeding was restarted again at a HRT of 20 days before decreasing to 18 days (period IV) with the aim to investigate if the gradual decrease in HRT will allow to avoid the acidification and system failure. AD of easily biodegradable material allows to increase the OLR rapidly, however, when the deal is with a complex lignocellulosic substrate, this strategy must be avoided and a gradual and slow OLR increase should be adopted in order to provide enough time to microorganism for adapting to the new OLR. After 20 days of operation (HRT: 18 days), biogas production dropped again and an increase in VFAs concentration was observed (Fig. 1-A and B). During this period,
9
the feeding was stopped again to digester in order to recover the reactor at the previous HRT of 20 days (period V). The main VFAs produced in R2 were propionic (HPr) and acetic acid (HAc). The higher values of HPr acid were detected during system failure at HRT of 15 days (period II). Fig. 2-A shows the maximum concentrations of both HAc and HPr during the study periods (including recovery stages). Moreover, Fig. 2-B shows the maximum and average values of the ratio between propionic and acetic acids (HPr/HAc). It has been reported that the ratio HPr/HAc could give an earlier indication of digester failure (Chen et al., 2008; Marchaim and Krause, 1993). The HPr/HAc values above 1.4 were considered as the threshold tolerated by an AD system without inhibition (Marchaim and Krause, 1993). On the other hand, propionic acid was widely considered as a strong inhibitor of anaerobic activities and it is difficult to remove from the system (Chen et al., 2008; Pullammanappallil et al., 2001). In the reactor R2, the maximum HPr/HAc ratios were observed at HRT of 15 days (period II) and during recovery period (above 40). However, this ratio was dropped to admissible value after consumption of VFAs (Marchaim and Krause, 1993). Methanogenic archaea are the most vulnerable to different environmental and operational factors in an AD system including organic overloading of the reactor which happened in this study (Chen et al, 2008). The use of an active inoculum with adapted biomass to the substrate allowed to degrade the VFAs accumulated and avoided the irreversible failure of the reactor R2. The role of the inoculum source and quality selection was reported to be crucial in AD processes (De Vrieze et al., 2015). The capacity of the inoculum used to recover the system was likely attributed to the
10
abundance and activity of methanogens and the adapted consortium of microorganisms to degrade ESBC-DP. Aboudi et al. (2015a) studied the batch anaerobic digestion of SBB and deduced that ESBC-DP released high VFAs concentrations in the medium, which delays the biogas production. Alkaya and Demirer (2011) also observed the system acidification when studying AD of sugar beet pulp in a SSTR with the aim to maximize VFA production from sugar beet wet pulp and sugar beet wastewater. The authors observed that the increase in sugar beet wet pulp amount in the feeding leads to the increase in the amount of acidification products (VFAs), which ultimately led to pH drop and inhibition of methanogenic activities. From the results of the current research and the above mentioned references, it could be deduced that SBB anaerobic digestion may cause the acidification of the reactor if not properly digested (i.e. as a sole substrate or at higher OLR). 3.2.2. Comparison of the single digestion performance parameters testing different HRTs Table 3 summarizes the performance of semi-continuous digestion reactors R1 and R2. The findings revealed that the specific methane production (SMP) of R2 at HRT of 15 days (period II) was very low, since the biogas production dropped drastically during this period. At 18 days HRT (period IV), the SMP was also found lower (only 122.8 mLCH4/gVSadded), due to the system failure at this particular HRT. Otherwise, SMP was observed similar in R1 and R2 at 20 days HRT. The methane content of the digesters R1 and R2 was above 50 % during whole periods, which was similar to the CH4 percentage reported by other authors studying AD of SBB (Brooks et al., 2008).
11
The highest VS removal was observed at 20 days HRT (periods 0, I, III and IV), while as the lowest one was obtained at HRT of 15 days (period II). During this period, the microbial activity was dropped and the OM was slowly or even not degraded due to the system acidification. HRT of 20 days was found to be the optimum HRT for treating ESBC-DP as a sole substrate under the studied conditions. AD of agro-industrial byproducts and crop residues individually usually implies the operation at higher HRT (Brooks et al., 2008; Demirel and Scherer, 2008; Moraes et al., 2015). Demirel and Scherer (2008) studied the AD of sugar beet silage using SSTR and reported that the highest methane generation was obtained at HRT of 25 days. In an earlier work of Labat et al. (1984), HRT of 17 days was found to be the best for anaerobic digestion of the sugar beet pulp (SBP) after a pretreatment of enzymatic hydrolysis. 3.3. Semi-continuous co-digestion reactor of ESBC-DP and CM (R3) In the second stage of this study, the effect of ESBC-DP and CM co-digestion on biogas production and stability of reactor was studied. At this stage, a decreasing sequence in HRTs i.e. 20, 18, 15 and 12 days corresponding to OLRs of 3.73, 4.14, 4.97 and 6.21 gVS/Lreactor d were tested and compared with each other and with results from single digestion assay (Table 2). 3.3.1. Biogas generation from the semi-continuous co-digestion reactor of ESBCDP and CM The daily biogas production from the SSTR fed with ESBC-DP and CM at different HRT tested is illustrated in Fig. 3-A. During the start-up period, some fluctuations in biogas production were observed as a consequence of biomass adaptation to the new cosubstrate (CM). However, a stable biogas production of about 1.61±0.24 L/Lreactor d was 12
observed after 20 days of operation. The biogas production was slightly higher than it was measured at the same HRT in single digestion assay. Upon decreasing the HRT to 18 days and 15 days (periods II and III, respectively), the biogas production was increased to 1.81±0.15 L/Lreactor d and 2.08±0.26 L/Lreactor d, respectively. However, once decreasing the HRT to 12 days (period IV), the biogas production started to decrease gradually accompanied by a pH drop and VFAs increase in the digester (Fig. 4-A). Although, the pH was adjusted by adding an alkali (NaOH-8M), the pH could not remain in the appropriate range for AD process and continued decreasing, which ultimately halted the methanogenic activity due to overloading of the reactor. At this stage, the feeding to reactor was ceased for several days to avoid total failure of the reactor, followed by inoculation with 2L of effluent coming from the SSTR R2, which was adapted for VFAs degradation. This practice allowed to recover the reactor and the accumulated VFAs were degraded (Fig. 4-A). Subsequently, the feeding was restarted at the previous HRT of 15 days and the system was operated under these conditions for a long period to ensure the stability of the digester. Umetsu et al. (2006) studied thermophilic Aco-D of SBB (tops and roots) with dairy manure at HRT of 20 days. They reported that the highest daily biogas production of 2.46 L/Lreactor d with 57.9 % of CH4 concentration was obtained from dairy manure and 40% beet top which was a higher production compared to the findings of the present study at the same HRT of 20 days. Fang et al. (2011) studied Aco-D of SBB (including wet pulp) with and without cow manure in both batch and SSTR conditions. SBP was found to be a good substrate for biogas production and the SMP of 280 mLCH4/gVSadded was obtained in a thermophilic continuously operated reactor, codigesting 50% of SBP with cow manure. The authors used CM to dilute toxic components contained in SBB and considering that CM provide the necessary buffering 13
capacity and trace element to the AD system. Aboudi et al. (2015b) studied Aco-D of ESBC with PM using a mesophilic SSTR. They reported that it was possible to decrease the HRT until 6 days (OLR of 11.2 gVS/Lreactor d) without any inhibition and obtaining high methane production rate (MPR) of 2.9 LCH4/Lreactor d, which was 2.4 fold higher than those obtained in the present study using CM as co-substrate and 3.5 fold higher than those obtained from single digestion assay of ESBC-DP. The results obtained by co-digestion with PM were better than those obtained in the present work using CM as co-substrate. This may be attributed to the differences in the manures used (PM and CM). PM provided more buffering capacity than CM, and this was likely the reason of the possibility to increase OLR in PM co-digestion assay. The physico-chemical characteristics of CM and those of the PM used in the previous study (Aboudi et al., 2015b) showed that PM has higher alkalinity and nitrogen content than CM. On the other hand, probably, the variety of microorganisms provided by each manure may be different (Savage et al., 1977). It was expected that co-digestion with CM would enhance the AD process more than that with PM due to the characteristics of the digestive tract of ruminants (cows) as herbivores are used to degrade vegetal fiber, however it was shown -at least in semi-continuous system and the studied conditions-, that alkalinity supply from the manure type, played the prevalent role in the process improvement. It should be noted that the used manure was stored at 4ºC when it was collected previously to use it in the feed, which likely would affect the autochthonous flora of the substrate. The data for the current co-digestion assay may be also compared with other works using CM as the main substrate and different crop co-substrates. Lehtomaki et al., (2007) examined the AcoD of CM with three different crop residues including sugar beet tops (SBT). A SMP of 229 mlCH4/gVSadded was obtained from mixture of 40% 14
SBT with CM, which was 65% higher than the methane yield from CM individually. However, the SMP they obtained was lower than that found in the present work at the same HRT of 20 days but lowers OLR (2 gVS/Lreactor d). Moreover, and contrastingly to this research, a decreasing in HRT from 20 to 16 days led to 26% decrease in SMP. Authors reported that in their studied conditions, shorter HRTs than 20 days were not sufficient for OM degradation and a high undegraded OM amount was found in the digestate. In an earlier study of Hills, (1980), Aco-D of CM with barley straw (80% of CM in the mixture) was carried out in a SSTR and a decreasing regime of HRT from 25 days to 10 days was tested. The highest SMP of 170 mlCH4/gVSadded was obtained at 20 days-HRT while methane productions decreased with OLR increase and distortions of the system were observed at lower HRTs. In the present study, we reach to decrease HRT successfully until 15 days with higher methane production and system performance. The biogas composition (H2, CH4 and CO2) during the SSTR co-digestion process is shown in Fig. 3-B. As can be seen, the low H2 production for about 3 days during the adaptation period (start-up), was due to the acclimation of the microorganisms to the new co-substrate, however, the pH correction was done using an alkali (NaOH-8M). Afterwards, higher CH4 content was observed in biogas in comparison with CO2 content at HRTs of 20 days and 18 days (periods I and II, respectively). The average CH4 concentrations of 57.7± 1.4 % and 55.5±1.1 % were observed for HRT of 20 days and 18 days, respectively, while CO2 concentration did not rise more than 45%. At HRT of 15 days (period III), the CH4 concentration was also higher but the concentrations of both CH4 and CO2 were closer. This fact could indicate that even the digester was operating conveniently at the HRT of 15 days, the system was closer to the OLR limit tolerated by the digester. This fact was confirmed when 15
CO2 concentrations increased significantly and exceeded the CH4 concentrations due to acidification and high liberation of CO2 upon decreasing the HRT to 12 days (period IV). 3.3.2. pH and VFAs evolution in the co-digestion reactor (R3) Acidogens are fast growers and very little sensitive to the environment conditions variations, while methanogens are inherently slow growers, and are inhibited at acidic pH values (Chen et al., 2008; Pullammanappallil et al., 2001). Thus, to achieve overall high rates, both the production and consumption of acids should be balanced and an imbalance would cause a slowing overall degradation of waste (Ahring et al., 1995; Demirel and Scherer, 2008). In the present study and during co-digestion of ESBC-DP and CM (R3), an imbalance between VFAs production and degradation occurred leading to acidification phenomenon that inhibited the process at HRT of 12 days (period IV). Thus, on 204th day of assay, the VFAs started to accumulate gradually from 357 mg/L to 5296 mg/L, and the best strategy was to stop feeding the reactor during this period. The main VFAs produced during the failure period were acetic and propionic acids with higher values for HAc than HPr (Fig. 4-B). Acetic acid accumulation indicates that acetoclastic methanogenesis was stressed (Ahring et al., 1995), which was traduced by the decrease in methane production during this period. For the rest of the stage, VFAs concentrations were very low, indicating that acidogens, acetogens and methanogens activities were balanced (Ahring et al., 1995; Vavilin et al., 1997). From other hand, it was found that the system failure was reversible and the recovery period was relatively short, mainly due to the predominance of HAc, which is
16
more easy to degrade than other organic acids especially HPr (Ahring et al., 1995; Marchaim and Krause, 1993). The two ratios HPr/HAc and acidity/alkalinity related to the system stability for each HRT tested are shown in Table 4. The HPr/HAc ratio was found in the appropriate range at all HRTs tested (Ahring et al., 1995; Marchaim and Krause, 1993). HAc concentration was higher than HPr which permitted the quick recovery of the digester as commented before. Nevertheless, the acidity/alkalinity ratio was higher than admissible value of 0.4 (Chen et al., 2008) at HRT of 12 days (period IV). 3.3.3. Organic matter evolution and removal Organic matter removal (VS degradation) is shown in Table 4. The findings revealed that the highest VS removal was obtained at HRT of 18 days. However, the VS removal was noted around 54 % at HRTs of 20 days and 15 days. The VS removal was 34.3% at the critical HRT of 12 days due to system failure. The findings suggested that 12 days-HRT was the threshold to carry out the anaerobic co-digestion of ESBC-DP with CM under stable conditions. Conclusions It has been demonstrated that AD of ESBC-DP in semi-continuous regime was feasible at HRT of 20 days (OLR: 3.26 gVS/Lreactor d) and propionic acid accumulation was observed when this limit is overtaken. However, co-digestion of ESBC-DP with CM allowed to decrease the HRT to 15 days (OLR: 4.97 gVS/Lreactor d) with high methane generation and without any process inhibition. HRT of 12 days was found critical for co-digestion assay leading to pH drop accompanied by VFAs accumulation. On the other hand, the digester recovery after failure was possible due to the well-adapted inoculum coming from the single digestion reactor. 17
Acknowledgements This work was financially supported by the Spanish Ministry of Economy and Finance (Project CTM2013-43938-R) and the MICINN (Project UNCA08-1E-035) and the European Regional Development Funds (ERDF). Authors also acknowledge support from the University of Cádiz in Spain for the PhD scholarship (UCA-2010063PU/EPIF-FPI-A/BC). Authors thank the Agrifood Campus of International Excellence (Ceia3) and the company (AB-sugars group) for providing the ESBC-DP used in this research. References Aboudi, K., Álvarez-Gallego, C.J., Romero-García, L.I. 2015a. Improvement of exhausted sugar beet cossettes anaerobic digestion process by Co-digestion with pig manure. Energy and Fuels, 29(2), 754-762. Aboudi, K., Álvarez-Gallego, C.J., Romero-García, L.I. 2015b. Semi-continuous anaerobic co-digestion of sugar beet byproduct and pig manure: Effect of the organic loading rate (OLR) on process performance. Bioresource Technology, 194, 283-290. Ahring, B.K., Sandberg, M., Angelidaki, I. 1995. Volatile fatty acids as indicators of process imbalance in anaerobic digesters. Applied Microbiology and Biotechnology, 43(3), 559-565. Alkaya, E., Demirer, G.N. 2011. Anaerobic acidification of sugar-beet processing wastes: Effect of operational parameters. Biomass and Bioenergy, 35(1), 32-39. Álvarez-Gallego, C. 2005. Testing different procedures for the start-up of a dry anaerobic co-digestion process of OFMSW and sewage sludge at thermophilic range. in: PhD Thesis, University of Cádiz. Spain. Anwar, Z., Gulfraz, M., Irshad, M. 2014. Agro-industrial lignocellulosic biomass a key to unlock the future bio-energy: A brief review. Journal of Radiation Research and Applied Sciences, 7(2), 163-173. APHA, A., Standards Methods for the Examination of Water and Wastewater,. 2005. 20th edition, American Public Health Association Washington DC, USA
18
Brooks, L., Parravicini, V., Svardal, K., Kroiss, H., Prendl, L. 2008. Biogas from sugar beet press pulp as substitute of fossil fuel in sugar beet factories. Water Science and Technology, 58(7), 1497-1504. Chen, Y., Cheng, J.J., Creamer, K.S. 2008. Inhibition of anaerobic digestion process: A review. Bioresource Technology, 99(10), 4044-4064. Comino, E., Rosso, M., Riggio, V. 2010. Investigation of increasing organic loading rate in the co-digestion of energy crops and cow manure mix. Bioresource Technology, 101(9), 3013-3019. Demirel, B., Scherer, P. 2008. Production of methane from sugar beet silage without manure addition by a single-stage anaerobic digestion process. Biomass and Bioenergy, 32(3), 203-209. De Vrieze, J., Raport, L., Willems, B., Verbrugge, S., Volcke, E., Meers, E., Angenent, L.T., Boon, N. 2015. Inoculum selection influences the biochemical methane potential of agro-industrial substrates. Microbial Biotechnology, 8(5), 776-786. Fang, C., Boe, K., Angelidaki, I. 2011. Anaerobic co-digestion of by-products from sugar production with cow manure. Water Research, 45(11), 3473-3480. Fernández Rodríguez, J., Pérez, M., Romero, L.I. 2012. Mesophilic anaerobic digestion of the organic fraction of municipal solid waste: Optimisation of the semicontinuous process. Chemical Engineering Journal, 193–194(0), 10-15. Hill, D.J. 1980. Biogas from a high solids combination of dairy manure and barley Straw. Transactions of ASAE, 23(6), 1500-1504. Hutnan, M., Drtil, M., Mrafkova, L. 2000. Anaerobic biodegradation of sugar beet pulp. Biodegradation, 11(4), 203-211. Labat, M., Garcia, J.L., Meyer, F., Deschamps, F. 1984. Anaerobic digestion of sugar beet pulps. Biotechnology Letters, 6(6), 379-384. Lehtomäki, A., Huttunen, S., Rintala, J.A. 2007. Laboratory investigations on codigestion of energy crops and crop residues with cow manure for methane production: Effect of crop to manure ratio. Resources, Conservation and Recycling, 51(3), 591-609. Li, Y., Park, S.Y., Zhu, J. 2011. Solid-state anaerobic digestion for methane production from organic waste. Renewable and Sustainable Energy Reviews, 15(1), 821826. Marchaim, U., Krause, C. 1993. Propionic to acetic acid ratios in overloaded anaerobic digestion. BioresourceTechnology, 43(3), 195-203. 19
Mata-Alvarez, J., Dosta, J., Romero-Güiza, M.S., Fonoll, X., Peces, M., Astals, S. 2014. A critical review on anaerobic co-digestion achievements between 2010 and 2013. Renewable and Sustainable Energy Reviews, 36(0), 412-427. Moraes, B.S., Triolo, J.M., Lecona, V.P., Zaiat, M., Sommer, S.G. 2015. Biogas production within the bioethanol production chain: Use of co-substrates for anaerobic digestion of sugar beet vinasse. Bioresource Technology, 190, 227234. McGillervy, D.J., Cranwell, P. D. 1992. Anaerobic microflora associated with the pars oesophagea of the pig. Research in Veterinary Science, 53(1), 110-115. Palmqvist, E., Hahn-Hägerdal, B. 2000. Fermentation of lignocellulosic hydrolysates. II: Inhibitors and mechanisms of inhibition. Bioresource Technology, 74(1), 2533. Pucker, J., Jungmeier, G., Siegl, S., Pötsch, E.M. 2013. Anaerobic digestion of agricultural and other substrates-implications for greenhouse gas emissions. Animal, 7(2), 283-291. Pullammanappallil, P. C., Chynoweth, D. P., Lyberatos, G., Svoronos, S. A. 2001. Stable performance of anaerobic digestion in the presence of a high concentration of propionic acid. Bioresource Technology, 78(2), 165-169. Sanaei-Moghadam, A., Abbaspour-Fard, M.H., Aghel, H., Aghkhani, M.H., AbediniTorghabeh, J. 2014. Enhancement of biogas production by co-digestion of potato pulp with cow manure in a CSTR system. Applied Biochemistry and Biotechnology, 173(7), 1858-1869. Savage, D.C. 1977. Microbial Ecology of the Gastrointestinal Tract. Annual Review of Microbiology, 31(1), 107-133. Umetsu, K., Yamazaki, S., Kishimoto, T., Takahashi, J., Shibata, Y., Zhang, C., Misaki, T., Hamamoto, O., Ihara, I., Komiyama, M. 2006. Anaerobic co-digestion of dairy manure and sugar beets. International Congress Series, 1293(0), 307-310. Vavilin, V.A., Rytov, S.V., Lokshina, L.Y. 1997. A balance between hydrolysis and methanogenesis during the anaerobic digestion of organic matter. Microbiology, 66(6), 712-717. Wang, X., Yang, G., Feng, Y., Ren, G., Han, X. 2012. Optimizing feeding composition and carbon-nitrogen ratios for improved methane yield during anaerobic codigestion of dairy, chicken manure and wheat straw. Bioresource Technology, 120, 78-83. 20
Ward, A.J., Hobbs, P.J., Holliman, P.J., Jones, D.L. 2008. Optimisation of the anaerobic digestion of agricultural resources. Bioresource Technology, 99(17), 7928-7940. Yang, L., Xu, F., Ge, X., Li, Y. 2015. Challenges and strategies for solid-state anaerobic digestion of lignocellulosic biomass. Renewable and Sustainable Energy Reviews, 44(0), 824-834.
21
Figure captions Fig.1. (A) Daily biogas production and (B) TVFA evolution in the reactors R1 and R2. *Black lines: stages separation; Red lines: stop feeding; Blue lines: inoculation. Fig.2. (A) Maximum main VFAs produced and (B) Maximum and average values of HPr/HAc ratio in the reactors R1 and R2 Fig.3. (A) Daily biogas production and (B) Biogas composition in the co-digestion reactor R3. *Black lines: stages separation; Red lines: stop feeding; Blue lines: inoculation. Fig.4. (A) pH and TVFA evolutions and in the co-digestion reactor R3 and (B) Detailed VFAs evolution during failure period. *Black lines: stages separation; Red lines: stop feeding; Blue lines: inoculation.
22
Fig.1 (A) and (B)
23
Fig.2 (A) and (B)
24
Fig.3 (A) and (B)
25
Fig.4 (A) and (B)
26
Table 1. Physico-chemical characteristics of substrates (ESBC-DP and CM) and feedstocks (single digestion and co-digestion)
Parameters
ESBC-DP
CM
Feedstock (single digestion)
Feedstock (co-digestion)
pH
5.8 ± 0.7
6.3 ± 0.6
5.3 ± 1.1
6.5 ± 0.2
TS (g/kg)
873.3 ± 1.0
211.1 ± 32.8
74.0 ± 3.4
78.4 ± 5.8
VS (%TS)
89.5 ± 0.7
76.8 ± 11.3
52.1 ± 7.6
74.5 ± 5.3
tCOD (gO2/kg)*
146.0 ± 12.0
75.0 ± 4.0
73.0 ± 6.0
69.0 ± 8.0
sCOD
61.9 ± 8.0
17.2 ± 1.0
14.2 ± 3.3
14.5 ± 1.6
DOC (gC/kg)*
38.8 ± 7.5
5.5 ± 0.5
11.4 ± 3.6
4.8 ± 1.9
TVFA (gHAc/kg)*
1.8 ± 1.2
4.6 ± 0.5
1.5 ± 1. 2
0.6 ± 0.4
(gO2/kg)*
Alkalinity (gCaCO3/kg)*
3.1 ± 1.2
38.3 ± 11.1
2.8 ± 1.4
52.6 ± 4.7
NH4-N (gN/kg)*
0.2 ± 0.1
2.4 ± 0.6
0.3 ± 0.2
0.8 ± 0.4
TKN (gN/kg)*
12.8 ± 1.3
32.5 ± 2.8
10.3 ± 1.2
22.5 ± 3.5
C/N ratio
40.8 ± 1.0
13.7 ± 1.7
29.5 ± 1.2
18.7 ± 0.9
* g/kg (wet basis)
27
Table 2. HRTs and OLRs applied to reactors R1, R2 and R3 Reactor
Period
HRT
OLR (gVS/Lreactor d)
R1
0
20
2.95±1.35
R2
I
20
3.26±1.02
R3
II
15
4.35±0.23
III
20
3.26±1.02
IV
18
3.63±0.25
V
20
3.26±1.02
I
20
3.73±0.58
II
18
4.14±0.74
III
15
4.97±0.94
IV
12
6.21±1.01
V
15
4.97±0.94
28
Table 3. Summary of the performance parameters of the semi-continuous single digestion reactors R1 and R2
Reactor R2
R1
Period
HRT
MPR
(Days) (LCH4/Lreactor d)
SMP
CH4
HPr/HAc
(mLCH4/gVSadded)
(%)
ratio
Acidity/alkalinity VS removal (%) ratio
I
20
0.84±0.48
225.71±4.49
54.48±2.47
0.10±0.03
0.12±0.04
62.87±5.12
II
15
0.51±0.62
86.43±1.44
53.61±1.66
2.18±0.45
2.47±0.11
33.48±3.26
III
20
0.83±0.16
215.91±2.28
54.85±2.63
0.53±0.34
0.09±0.02
54.45±7.42
IV
18
0.47±0.08
122.86±1.18
53.45±2.26
3.74±1.81
0.65±0.06
41.72±4.68
V
20
0.85±0.21
208.41±2.14
53.65±1.59
0.54±0.22
0.14±0.02
57.48±2.52
0
20
0.83±0.09
214.64±6.12
55.22±4.76
0.94±0.35
0.18±0.11
57.63±14.75
29
Table 4. Summary of the performance parameters of the semi-continuous co-digestion reactor R3
Period
HRT (Days)
MPR (LCH4/Lreactor d)
SMP HPr/HAc (mLCH4/gVSadded) ratio
Acidity/alkalinity VS removal ratio (%)
I
20
0.88±0.09
242.59±4.26
0.78±0.58
0.02±0.01
54.82±1.45
II
18
1.09±0.06
252.82±10.02
0.25±0.04
0.03±0.02
62.63±1.87
III
15
1.23±0.11
313.98±15.47
0.13±0.06
0.02±0.01
53.41±5.12
IV
12
0.34±0.17
64.81±2.43
0.61±0.08
0.83±0.42
34.32±7.19
V
15
1.30±0.11
295.36±7.52
0.49±0.31
0.02±0.03
52.75±4.11
30
Highlights
Individual digestion of sugar beet byproduct is feasible in semi-continuous system
20-days hydraulic retention time was the threshold for the single digestion system
Co-digestion with cow manure allowed to reduce hydraulic retention time to 15days
Volatile fatty acids were accumulated during reactors failure
The source and quality of the inoculum used has permitted to recover the systems
31