- Email: [email protected]
A combination anaerobic digestion scheme for biogas production from dairy effluent—CSTR and ABR, and biogas upgrading
Biomass and Bioenergy xxx (2017) 1e7
Contents lists available at ScienceDirect
Biomass and Bioenergy journal homepage: http://www.elsevier.com/locate/biombioe
Research paper
A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading Lars Jürgensen a, *, Ehiaze Augustine Ehimen c, Jens Born b, Jens Bo Holm-Nielsen a a
Aalborg University Esbjerg, Niels Bohrsvej 8, 6700 Esbjerg, Denmark Flensburg University of Applied Science, Kanzleistr. 91, 24937 Flensburg, Germany c Future Analytics Consulting, 23 Fitzwilliam Square S, Dublin 2, D02 RV08, Ireland b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 August 2016 Received in revised form 25 April 2017 Accepted 26 April 2017 Available online xxx
Anaerobic digestion of low-strength dairy waste water was used for the production of biogas which is aimed at serving as a concentrated carbon dioxide (CO2) source for further methanation. Using hydrogen (which can be produced via electrolysis using renewably sourced electricity), the CO2 fraction of the produced biogas can be used as a mechanism to store surplus electricity by the Sabatier process, which converts the CO2 fractions to methane (CH4), i. e. synthetic natural gas. This study investigates the use a combined reactor scheme for the anaerobic digestion of dairy waste water, and the further upgrading of the biogas products from the process. A combination pilot scale process was established with a 90 d start-up time using a 1 m3 continuous stirred tank reactor (CSTR) and a 0.2 m3 baffled reactor (ABR) in series. The system was fed at constant retention time in the ABR of 1.6 d and with varying substrate organic loading rates between 1.25 and 4.50 kg m3 d1. The average chemical oxygen demand (COD) removal was 82% with a biogas yield of 0.26 m3 kg1. The use of the derived biogas for the Sabatier process to convert hydrogen into CH4 showed no disadvantages compared to synthetic gas mixtures. The combination of CSTR and ABR overcame the individual disadvantages of both reactor types. The investigated anaerobic digestion system can be further optimized and adopted to replace conventional waste water treatment systems. © 2017 Published by Elsevier Ltd.
Keywords: Anaerobic digestion Dairy waste water Two-stage fermentation Biogas Upgrading
1. Introduction Anaerobic baffled reactor (ABR) systems have been widely applied as high rate digesters especially in developing countries where they have been used to meet on-site sanitation goals [1]. Numerous research papers have been published in the last two decades which have reported the potential advantages of the use of ABR as an excellent anaerobic digestion system for low- and highstrength waste water [2e6] and for the treatment of complex organic waste streams [7,8]. The simplicity and inexpensiveness of the ABR system, coupled with its non-requirement of moving parts or mechanical mixing for the anaerobic fermentation process has also strengthened the potential benefits associated with the use of this anaerobic digestion reactor system [9]. Due to the construction of the ABR (as a series of up-flow and down-flow sections), this reactor type enables an internal phase
* Corresponding author. E-mail address: [email protected] (L. Jürgensen).
separation, i. e. different process parameters can be maintained and manipulated in the different reactor compartments. This selective process differentiation in the reactor compartments could enable hydrogen production at low residence times, and improved overall methane yields from the anaerobic digestion process as reported in Refs. [10e12]. The industrial applications of the ABR scheme have however been mainly limited to waste water treatment [1] and have not yet been expanded for use in the digestion of other organic residue streams i. e. animal excreta and industrial organic residues (such as dairy and food processing waste). Dairy waste water effluent is generally regarded to contain a high organic load and is considered a useful substrate for anaerobic processes aimed at methane production [13,14]. The effluents from dairy processing plants (producing fresh milk and yogurt) have however been observed to be characterized by having a low chemical oxygen demand (COD). Typical COD values of the effluent in milk processing dairies are 2e10 kg m3, with up to 70 kg m3 reported for whey waste water [9,15]. Owing to such low COD values, the effluents of milk processing dairy plants can be considered as unsuitable for treatment in a conventional
http://dx.doi.org/10.1016/j.biombioe.2017.04.007 0961-9534/© 2017 Published by Elsevier Ltd.
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
2
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
continuous stirred tank reactor (CSTR)-type biogas plant [16]. Effluents which are considered as ”low-strength”are therefore often treated in high rate anaerobic digesters like upflow anaerobic sludge blanket (UASB) or biofilm reactors. These types of reactors use different hydraulic (HRT) and solid retention time (SRT) distributions, which favour the slow growing methanogenic archea thus improving methanogenesis [17]. These reactor designs are however limited when dealing with substrates which contain high amounts of fats or solids. This is mainly due to transfer restrictions brought about by the absence or lack of stirring or mechanical mixing components. Solids with densities higher than that of water usually sink and accumulate in the first chamber, with the fat contents of usually resulting in the formation of floating layers in the digester [1]. In both cases, organic components become inaccessible to the microbial community. Limitations to the use of currently available, cost effective anaerobic digestion treatment routes for low strength dairy waste water but with a high fat and solids content can therefore be observed and needs to be addressed to ensure the proper utilization of this valuable resource and to prevent potential environmental degradation processes (i. e. euthrophication of water bodies) associated with the nontreatment and disposal of such residues. This study investigates the combination of the CSTR and ABR schemes (in series) as a useful means of overcoming the disadvantages of both systems, with the goal of facilitating the use of this stream as a useful carbon source for the methanation process. The paper will not only highlight the methanogenic processes related to the bacterial flora of the digestion process i. e. eventually leading to the production of biogas (a mixture of methane (CH4), carbon dioxide (CO2) and other trace gases such as hydrogen sulphide (H2S) and ammonia), but will also look at the incoporation of an additional thermochemical process (Sabatier reaction) into the anaerobic digestion system with the overall goal of optimizing the overall methane (and hence energy) yields from the preceding digestion process. The methanation of CO2, also known as Sabatier process, is a thermochemical reaction at increased pressure and temperature using Nickel or Ruthenium catalyst. Previous research has demonstrated the use of this process for the conversion of the CO2 fractions of anaerobically derived biogas to CH4 With regards to the efficiency of the Sabatier process when used for biogas CO2 conversion, significant poisoning of the reaction catalyst (Nickel) was demonstrated to occur in the presence of small amounts of hydrogen sulphide (H2S [18]. A reduction of the biogas H2S levels (i. e. via desulphurisation processes) is therefore essential before the biogas can be applied for the thermochemical methanation process. Additionally, the influence of other trace substances, like ammonia and siloxanes contained in the anaerobically derived biogas on the Sabatier process and the reaction catalysts are unknown. And have not been extensively identified [19] due to the difficulty related to analysing such trace components using conventional biogas analysis systems [20]. The usage of actual biogas mixtures is therefore important to take into account, since it would influence the feasibility of such anerobic-thermochemical combination systems in practice. This is especially important since the use of, and integration with dairy waste water digestion was found to be lacking in the literature. With the presence of contaminants i. e. cleaning agents and disinfectants used for hygienic food processing plants potentially negatively affecting the anaerobic and thermochemical process, this study provides a useful assessment of their influence on both procesess. This study aims to highlight the importance of local residue streams from industrial processes for both, energy generation and as a carbon source. In a future of 100% renewable energy scenario,
electricity will be an important primary energy carrier [21]. To satisfy the needs of the transportation and industry sector, it is expected that hydrogen will be produced in decentralized units using renewable energy sources like wind and solar [22] and converted into fuels with better properties like CH4 and methanol [23]. This paper initially considers methods to improve the biogas yields from a currently underutilised dairy effluent water, and further assess the upgrading of the energy value (through an increase in the methane outputs) of the anaerobically derived biogas through the integration of the Sabatier process. The work therefore aims to demonstrate at a lab scale concepts of decentralised biogas production and upgrading with the intention to facilitate matching the biogas quality to natural gas grid standards as previously reported in the literature [24e26]. In addressing those issues the paper will also highlight the use of biogas production from a wide range of substrates. Such substrates could the be a local carbon source for the storage of renewable generated hydrogen. 2. Materials and methods The experimental setup used in this study is shown in Fig. 1. It includes sections for: biogas generation, conditioning, and upgrading. All values for feed strength, specific gas production and organic loading rate (OLR) are based on COD. 2.1. Anaerobic digestion using the CSTR-ABR combination system 2.1.1. Substrate The effluent water of a local dairy plant (Osterhusumer Meierei Witzwort EG, North Frisia, Germany) was used as primary substrate for the AD process. It was delivered once a week and stored at ambient temperatures in a 1 m3 intermediate bulk container. From here, it was pumped once a day into the well stirred 0.25 m3 feed tank. At the dairy plant production site, the effluent was collected from a buffer tank, which is used to feed the dairy's conventional (aerobic) waste water treatment plant and to avoid shock loading events to it. The dairy waste water initially contained varying amounts of NaOH and H3PO4, used in the plant cleaning processes. These were however neutralised before this tank. The phosphate content resulted in a high acid base capacity of the dairy waste water mixture. The chemical oxygen demand (COD) of the waste water sample was analysed after delivery to the experiment site. The COD levels of the substrate were observed to usually fluctuate between 2.5 and 5.0 kg m3 during the reactor operation. More extreme fluctuations for the substrate COD were however periodically recorded as indicated in Table 1 which shows the dairy waste water characteristics recorded during one year of operation. The levels of the dairy waste water COD, nitrate, and phosphorus levels were determined using Hach Lange cuvette tests. 2.1.2. Anaerobic digesters A 1 m3 CSTR and a 0.2 m3 ABR were used in series for this study. The CSTR was temperature controlled (operating temperature was 38 C) using a Pt-100 sensor, a micro-controller and four 1 kW heating rods, which were inserted from the top of the vessel into the liquid. The temperature probe was also equipped with a pH sensor. The whole steel vessel was properly insulated to maintain the temperature profile and prevent temperature loss from the reactor. A level drain was used to transfer the effluent to the ABR system by gravity. Each of the four chambers was controlled individually using Pt-100 sensors and heating foils fixed to the external reactor walls. The downflow-to-upflow ratio was 1:2. A siphon was used to drain the effluent to the sink. Each chamber was equipped with
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
3
Fig. 1. Experimental setup containing a two-step anaerobic process (1 m3 CSTR followed by a 200 l-ABR), gas storage, drying, desulfurization and compression. On the right hand side, a heterogeneous catalytic reactor is used to convert the CO2 fraction of the compressed biogas to CH4 using H2 from bottles.
Table 1 Characteristics of the dairy waste water recorded during one year of operation.
average min max
pH
COD in kg m3
NO3-N in g m3
total P in g m3
total N in g m3
8.2 7.5 10
3.0 0.9 6.4
0.48 0.30 0.83
26.3 8.6 52.7
172 99 300
three sample valves located in the bottom, middle, and top of each up-flow section. The outlets for channeling the produced biogas were located at the top of the CSTR and of each chamber of the ABR, and were connected to a common pipe. The common pipe was connected to a dip tube, ensuring an increased pressure of 10e20 kPa in the reactor headspace. A Ritter drum counter was used to measure the volume flow of the produced gas.
2.1.3. Anaerobic digester operation 0.4 m3 and 0.1 m3 sewage sludge from the municipal waste water treatment plant were used as the initial inoculants both for the CSTR and the ABR. After inoculation, the system was left to stand at 38 C for one week. A constant flow of 0.130 m3 d1 was fed to the process resulting in OLR between 1.3 and 4.3 kg m3 d1, according to the varying feed strength. The OLR was calculated on the basis of the ABR volume. Temperature, pH, and gas flows were recorded daily. The substrate COD, the effluent from the CSTR (i. e. the influent to the ABR) and the effluent of the ABR were analysed weekly using HACH Lange cuvette tests. The biogas potential was determined by standardised batch test according to DIN 38414. The COD removal efficiency was calculated as
CODðtÞout X ¼ 1 Pt ; COD ðt Þ t2t
in
2ttot
i
where ttot is the nominal residence time in the total system. Taking into account, that the feed strength changed every week, an average COD value over 2ttot ¼ 18:7d was used (presented in the denominator). The retention time of the ABR was 1.7 d and the retention time of the CSTR was 7.6 d). The gas composition of the whole system was analysed weekly and once a month. The gas composition from each reactor and chamber was also analysed.
2.2. Gas storage and compression The produced gas was delivered to two 90 l gas storage bags (Ritter). From there it was pressurised once a day to the feed-bottle of the methanation unit via a membrane compressor with 11 bar outlet pressure. Before the compressor, a gas cooler was used to remove the main water vapor content (condensate) and a filter filled with doped charcoal [27] was used as desulphurization chamber to remove the sulphur fraction of the biogas. 2.3. Biogas upgrading A tubular reactor 10 mm in diameter was used to convert the CO2 fraction of the biogas to methane. Commercially available Ni catalyst on silica-alumina support delivered by ALFA AESAR (catalogue number MFCD00011137) was used as the methanation catalyst. The catalyst used contained 66% Ni to provoke carbon formation within a shorter time [28] and therefore improve the reaction stability under biogas conditions. The catalyst was pelleted and sieved using two sieves (with mesh opening sizes of 500 mm and 250 mm). 5$103 kg of the fraction obtained on the 250 mm sieve were used for the experiments. A thermocouple was inserted in the catalytic bed to control the reaction temperature. The reactor was heated by an electrical furnace to 250 C. The system pressure was set to 600 kPa. Mass flow controllers were used to deliver the compressed biogas and hydrogen (or other gases for simulated biogas/gas mixtures) to the reactor. 1.5$103 m3 h1 biogas was mixed with a certain amount (depending on the actual biogas composition) of H2 to ensure a stoichiometric CO2:H2 mole ratio of 1:4. The biogas and the product gas composition were analysed using a the Biogasanalyzer VISIT 03 produced by Messtechnik EHEIM GmbH. Any deactivation of the catalyst would therefore be recognized due to reduced conversion observed for the upgrading process. 3. Results and discussion 3.1. pH profile and phase separation Assessing the pH profile for the process, the anaerobic digestion process can be divided into two phases: start up phase, and the steady-state operation phase (see Fig. 2). The term ”steady-state” as used here, means that the phase separation was already established and stable. The concentrations of the substrate however varied according to the waste water produced by the dairy plant.
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
4
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
Fig. 2. pH profile during the digestion of dairy waste water using a CSTR and an ABR in series: after 100 d, a stable and static pH profile was established.
Fig. 2 shows the pH values of the CSTR and each chamber of the ABR (indicated in the figure legend as Comp. 1 to 4. As observed, during the first 70 d, the pH levels along the reactor remained relatively constant. After day 70, a well developed pH profile was seen to have been established. The pH values in the CSTR dropped to values of 6.5, while a pH increase along the ABR can be seen. The effluent of the fourth and last chamber left the reactor at a pH of 7.2e7.4. 3.2. Reactor start-up 3.2.1. CSTR performance during start up During the first 60 d of operation, 90% of the total biogas was produced in the CSTR, while the ABR accounted for no (or only minor) contributions to the COD removal. A wash-out of the active sludge was also observed with the samples taken from the CSTR containing less sludge with time. This pseudo-steady state was observed to be altered beyond t ¼ 70 d. Fig. 3 shows the pH values (as also seen from 2) and the COD content of the CSTR effluent during the whole reactor operation. During the first 32 d, the pH values increased from 6.8 to 7.1 and was observed to remain constant between 7.0 and 7.1 for a further 30 d. Until 62 d, the COD removal in the CSTR was high, resulting in effluent COD values of 350 kg m3 (i. e. the influent to the ABR), which corresponds to a COD removal of 88%. While the pH value
remained over 7.0 for further 12 d, the COD increased rapidly, indicating an acidification of the reactor. In a single-step anaerobic digestion process, this would indicate the imminent failure of the fermentation process. However, using the investigated combination process, when the pH dropped below 7.0, the COD removal efficiency was already under 65%. From day 74 to day 94, the pH dropped to 6.3 and remained constant for the following 100 d at 6.5. The COD of the CSTR effluent increased to values between 2.5 and 3.5 kg m3, corresponding to an average COD removal efficiency of 28% (see Fig. 4). The fraction of the biogas produced by the CSTR decreased to 10% of the total biogas production. 3.2.2. Gas production and COD removal The gas production increased during the first 20 d up to >0.150 m3 d1, which can be explained by high amount of active sludge available after inoculation. This sludge was washed out due to the comparatively low retention time in the CSTR of 7.7 d. This washout resulted in a decrease of the biogas production to 0.110e0.120 m3 d1 during the later start up stages. The overall COD removal (i. e. by the combined CSTR and ABR system) from the beginning of the digestion period was observed to be high. Before phase separation (i. e. decrease of pH in the CSTR), the average COD removal was 91%. With the drop in the pH values in the CSTR, the overall COD removal consequently decreased to 59%, later increasing within one week back to >79%. Fig. 4(b) shows the COD removal efficiencies for CSTR and the whole system. The COD removal by the ABR was observed to be decreasing, while the overall COD removal remained at a high level, indicating that the ABR was working well. 3.3. Steady-state operation As mentioned above, steady-state operation was established between day 90 and 190 as seen in Fig. 2.
Fig. 3. Start up of the CSTR: Beyond t ¼ 74 d, the pH dropped significant, indicating the acidification of the hydrolysis stage. Simultaneously, the effluent from the CSTR increased in COD by inhibition of methanogenes.
3.3.1. COD removal and gas yield The overall COD removal using the combined system was observed to be high with an average of 82%. Fig. 4(a) and (b) shows the COD of the substrate and the ABR effluent. After phase separation, the COD removal was seen to decrease slightly, but it overall maintained a high level, even when the feed concentration changed rapidly. The biogas production was 122 l d1. As expected, the biogas production increased and decreased according to the OLR. The
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
5
Fig. 4. System performance during start up and steady state operation: (a) CODs, (b) OLR and COD removal efficiencies, (c) total gas production and (d) gas yield based on the removed COD. The variations in OLR are caused by variation of the dairy waste water strength.
average biogas yield was 0.264 m3 kg1, based on the removed COD. Fig. 5 shows the biogas potential of the dairy waste water. Additionally, the fat fraction was separated by flocculation to demonstrate the waste water components which contributes to the gas yield. Here, the biogas potential was measured to be 0.58 m3 kg1 based on the COD of the sample used for this batch test, which was high in COD (5.6 kg m3). This difference can be explained by the high amount of fat which comes with waste water loaded with a high COD, resulting in high gas yields due to the fat degradation. Furthermore, the high amount of water passing the reactor, corresponding to the low retention time, will result in less CO2 released to the gas phase. Thus, the measured amount of biogas in
the continuous experiments using a high rate anaerobic digester will expectedly be lower compared to the batch test.
3.3.2. Biogas composition The biogas composition was high in CH4 during the whole fermentation. The average methane volume fraction was 76% ± 2% with minor amounts of hydrogen released from the first reactor. Such high methane concentrations have been previously reported as typically following the digestion of dairy waste water [15]. The gas released from the CSTR contained up to 10% hydrogen due to hydrolysis of fats, which is in good agreement with other studies conducted on the anaerobic production of drogen
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
6
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
Fig. 5. Biogas potential batch test results of the waste water (given in m3/kg based on organic dry matter (oDM) and COD), the fat fraction separated by flocculation and the inoculum as reference.
production from fats [29]. However, the gas production from the CSTR was only approximately 10% of the total biogas production from the combination system. Thus, the amount of hydrogen (produced via the hydrolysis of fats and from the dairy waste water) in the total biogas stream was considered to be negligible. 3.3.3. Shock-load stability During the operation, several shock-load events can be seen from Fig. 4: On day 110, 139, and 179, the COD values increased to from <3.0 to >5.5 kg m3. As explained before, the shock loadings are caused by variations in the dairy waste water strength which varied due to the dairy processes. The COD of the CSTR's effluent was influenced only slightly and also the COD of the effluent of the ABR. It becomes clear that the CSTR acts as (i) a hydrolysis stage and (ii) as a buffer tank. This is since the kinetics of the hydrolysis reactions are quite fast. However the use of a CSTR, with 7.6 d retention time in this study leads to a considerable reduction in the shock loading to the ABR. However, an optimum size of CSTR:ABR has to be found but that optimization was not part of this study. 3.4. Biogas upgrading Over 30 d, the catalytic system remained stable showing no deactivation of the catalyst. This proves that real biogas emanating from the dairy waste water digestion is suitable as feed gas for the Sabatier process. As previous experiments indicated [19], the ammonia content of the biogas did not affect the conversion reaction of the Sabatier process in this study. To establish such an upgrading process using biogas from anaerobic digestion, no special pretreatment technologies for ammonia removal from the feed biogas are necessary. With regards to the removal of the harmful H2S biogas constituents, this can be achieved using doped charcoal (e. g. potassium iodide), which is commercially available [30], so its adaption and integration in practice should be easy. In a practical local context, anaerobic digestion systems could therefore utilize waste streams from the food industry to supply a
highly concentrated CO2 source to the Sabatier process. In North Frisa, Germany, the county where the dairy is located (which provided the study dairy waste water substrate), 1.6 GW wind power is installed [31], resulting in overloaded electrical grids [32]. To avoid a breakdown of the electrical grid, wind times are curtailed up to 1660 h/a [33]. Instead of curtailing the electricity generation from wind and solar to avoid overloads, the surplus could be used to produce hydrogen via water electrolysis. This renewable hydrogen can then be applied for biogas upgrading, which in turns serves as a scheme for (chemical) energy storage. 4. Conclusions As demonstrated by the results of this study, anaerobic digestion can be applied as a useful scheme to convert food industry wastes streams (i. e. dairy industry wastes) which are currently underutilised in energy generation due to the issues associated with their use in conventional processes. The combination of CSTR and ABR (in series) could be a useful solution to overcome the disadvantages of both reactors. This is true in particular because high-rate anaerobic digesters are usually unable to handle the characteristically high particles and fats contents contained in such waste water streams. This problem is therefore improved with the use of the proposed combination scheme. Anaerobic digestion facilities provide a useful option for electricity storage via its use for the upgrading of the biogas to methane (according to natural gas grid standards) which can be conveniently stored and used in other applications. The excess electricity is therefore used for the electrolysis process for hydrogen production which is consumed in the Sabatier process. The use of biogas derived from dairy effluent water as a digestion substrate shows no disadvantages in the Sabatier process compared to synthetic mixtures. Acknowledgment The authors will like to acknowledge the European INTERREG
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
Region Fund that facilitated the scientific work presented in this paper through the funding of the project Large Scale Bio Energy Lab II. The authors will also like to sincerely thank all the students and staff who worked with us during the project period. References [1] K. Foxon, S. Pillay, T. Lalbahadur, N. Rodda, F. Holder, C. Buckley, The anaerobic baffled reactor (abr): an appropriate technology for on-site sanitation, Water SA 30 (5) (2007) 44e50. [2] W.P. Barber, D.C. Stuckey, The use of the anaerobic baffled reactor (abr) for wastewater treatment: a review, Water Res. 33 (7) (1999) 1559e1578, http:// dx.doi.org/10.1016/S0043-1354(98)00371-6. http://www.sciencedirect.com/ science/article/pii/S0043135498003716. [3] G. Gopala Krishna, P. Kumar, P. Kumar, Treatment of low-strength soluble wastewater using an anaerobic baffled reactor (abr), J. Environ. Manag. 90 (1) (2009) 166e176. [4] S. Bodkhe, A modified anaerobic baffled reactor for municipal wastewater treatment, J. Environ. Manag. 90 (8) (2009) 2488e2493. [5] Gopala Krishna, G.V.T.,P. Kumar, P. Kumar, Treatment of low strength complex wastewater using an anaerobic baffled reactor (abr), Bioresour. Technol. 99 (17) (2008) 8193e8200, http://dx.doi.org/10.1016/j.biortech.2008.03.016. \url {18455391}. [6] G.D. Ji, T.H. Sun, J.R. Ni, J.J. Tong, Anaerobic baffled reactor (abr) for treating heavy oil produced water with high concentrations of salt and poor nutrient, Bioresour. Technol. 100 (3) (2009) 1108e1114, http://dx.doi.org/10.1016/ j.biortech.2008.08.015. \url{18793845}. [7] J. Zhang, Y. Wei, W. Xiao, Z. Zhou, X. Yan, Performance and spatial community succession of an anaerobic baffled reactor treating acetone-butanol-ethanol fermentation wastewater, Bioresour. Technol. 102 (16) (2011) 7407e7414, http://dx.doi.org/10.1016/j.biortech.2011.05.035. \url{21664129}. [8] A. Bayrakdar, E. Sahinkaya, M. Gungor, S. Uyanik, A.D. Atasoy, Performance of sulfidogenic anaerobic baffled reactor (abr) treating acidic and zinc-containing wastewater, Bioresour. Technol. 100 (19) (2009) 4354e4360, http:// dx.doi.org/10.1016/j.biortech.2009.04.028. [9] M. Nasr, A. Khalil, A. Abdel-Kader, W. Elbarki, M. Moustafa, Comparative performance between conventional and hybrid anaerobic baffled reactors for treating dairy wastewater, Int. J. 5 (4). [10] A. Ahamed, C.-L. Chen, R. Rajagopal, D. Wu, Y. Mao, I. Ho, J. Lim, J.-Y. Wang, Multi-phased anaerobic baffled reactor treating food waste, Bioresour. Technol. 182 (2015) 239e244. [11] A. Tawfik, M. El-Qelish, Continuous hydrogen production from co-digestion of municipal food waste and kitchen wastewater in mesophilic anaerobic baffled reactor, Bioresour. Technol. 114 (2012) 270e274. [12] P. Thanwised, W. Wirojanagud, A. Reungsang, Effect of hydraulic retention time on hydrogen production and chemical oxygen demand removal from tapioca wastewater using anaerobic mixed cultures in anaerobic baffled reactor (abr), Int. J. Hydrogen Energy 37 (20) (2012) 15503e15510. [13] S.V. Mohan, V.L. Babu, P. Sarma, Anaerobic biohydrogen production from dairy wastewater treatment in sequencing batch reactor (ansbr): effect of organic loading rate, Enzyme Microb. Technol. 41 (4) (2007) 506e515. [14] B. Sarkar, P. Chakrabarti, A. Vijaykumar, V. Kale, Wastewater treatment in dairy industries at possibility of reuse, Desalination 195 (1) (2006) 141e152. [15] B. Demirel, O. Yenigun, T.T. Onay, Anaerobic treatment of dairy wastewaters: a review, Process Biochem. 40 (8) (2005) 2583e2595, http://dx.doi.org/
7
10.1016/j.procbio.2004.12.015. € rog lu, B. Ozkaya, M. Cakmakci, A review on anaerobic [16] D. Karadag, O.E. Ko biofilm reactors for the treatment of dairy industry wastewater, Process Biochem. 50 (2) (2015) 262e271. [17] L. Seghezzo, G. Zeeman, J.B. van Lier, H. Hamelers, G. Lettinga, A review: the anaerobic treatment of sewage in uasb and egsb reactors, Bioresour. Technol. 65 (3) (1998) 175e190. [18] M. Twigg, M. Twigg, Catalyst Handbook, CSIRO, 1989. [19] L. Jürgensen, E.A. Ehimen, J. Born, J.B. Holm-Nielsen, D. Rooney, Influence of trace substances on methanation catalysts used in dynamic biogas upgrading, Bioresour. Technol. 178 (2015) 319e322. [20] R. Dewil, L. Appels, J. Baeyens, Energy use of biogas hampered by the presence of siloxanes, Energy Convers. Manag. 47 (13e14) (2006) 1711e1722, http:// dx.doi.org/10.1016/j.enconman.2005.10.016. [21] U. Bossel, Does a hydrogen economy make sense? Proc. IEEE 94 (10) (2006) 1826e1837, http://dx.doi.org/10.1109/JPROC.2006.883715. [22] R.R. Dickinson, D.L. Battye, V.M. Linton, P.J. Ashman, Nathan, J. Graham, Alternative carriers for remote renewable energy sources using existing cng infrastructure, Int. J. Hydrogen Energy 35 (3) (2010) 1321e1329, http:// dx.doi.org/10.1016/j.ijhydene.2009.11.052. [23] G.A. Olah, A. Goeppert, Prakash, G.K. Surya, Chemical recycling of carbon dioxide to methanol and dimethyl ether: from greenhouse gas to renewable, environmentally carbon neutral fuels and synthetic hydrocarbons, J. Org. Chem. 74 (2) (2009) 487e498, http://dx.doi.org/10.1021/jo801260f. [24] E.J. Hengeveld, W. van Gemert, J. Bekkering, A.A. Broekhuis, When does decentralized production of biogas and centralized upgrading and injection into the natural gas grid make sense? Biomass Bioenergy 67 (2014) 363e371, http://dx.doi.org/10.1016/j.biombioe.2014.05.017. [25] E.J. Hengeveld, J. Bekkering, W. van Gemert, A.A. Broekhuis, Biogas infrastructures from farm to regional scale, prospects of biogas transport grids, Biomass Bioenergy 86 (2016) 43e52, http://dx.doi.org/10.1016/ j.biombioe.2016.01.005. Antonio, Assessment of membrane [26] G. Valenti, A. Arcidiacono, Nieto Ruiz, Jose plants for biogas upgrading to biomethane at zero methane emission, Biomass Bioenergy 85 (2016) 35e47, http://dx.doi.org/10.1016/ j.biombioe.2015.11.020. [27] S. Rossow, Entwicklng einer speziellen aktivkohle für den einsatz zur biogasentschwefelung und untersuchung der leistungsf€ ahigkeit im labor und im praktischen einsatz, (Dissertation). [28] W.-S. Dong, H.-S. Roh, K.-W. Jun, S.-E. Park, Y.-S. Oh, Methane reforming over ni/ce-zro 2 catalysts: effect of nickel content, Appl. Catal. A General 226 (1) (2002) 63e72. [29] M. Monti, A. Scoma, G. Martinez, L. Bertin, F. Fava, Uncoupled hydrogen and volatile fatty acids generation in a two-step biotechnological anaerobic process fed with actual site wastewater, New Biotechnol. 32 (3) (2015) 341e346, http://dx.doi.org/10.1016/j.nbt.2014.08.002. \url{25174889}. [30] Philipp Adler, et al., Leitfaden Biogasaufbereitung und -einspeisung, FNR, Federal Ministry of Food and Agriculture Germany. [31] T. Engel, Die karte der erneuerbaren energien, Database provided by the German Association of Solar Energy, energymap.info, 2015. [32] L. Jürgensen, E.A. Ehimen, J. Born, J.B. Holm-Nielsen, Utilization of surplus electricity from wind power for dynamic biogas upgrading: Northern Germany case study, Biomass Bioenergy 66 (2014) 126e132. [33] L. Jürgensen, E. A. Ehimen, J. Born, J. B. Holm-Nielsen, Use of the Sabatier process for dynamic biogas upgrading in northern Germany, 22nd European Biomass Conference and Exhibition, Hamburg (2014) 881e886.
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
Contents lists available at ScienceDirect
Biomass and Bioenergy journal homepage: http://www.elsevier.com/locate/biombioe
Research paper
A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading Lars Jürgensen a, *, Ehiaze Augustine Ehimen c, Jens Born b, Jens Bo Holm-Nielsen a a
Aalborg University Esbjerg, Niels Bohrsvej 8, 6700 Esbjerg, Denmark Flensburg University of Applied Science, Kanzleistr. 91, 24937 Flensburg, Germany c Future Analytics Consulting, 23 Fitzwilliam Square S, Dublin 2, D02 RV08, Ireland b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 August 2016 Received in revised form 25 April 2017 Accepted 26 April 2017 Available online xxx
Anaerobic digestion of low-strength dairy waste water was used for the production of biogas which is aimed at serving as a concentrated carbon dioxide (CO2) source for further methanation. Using hydrogen (which can be produced via electrolysis using renewably sourced electricity), the CO2 fraction of the produced biogas can be used as a mechanism to store surplus electricity by the Sabatier process, which converts the CO2 fractions to methane (CH4), i. e. synthetic natural gas. This study investigates the use a combined reactor scheme for the anaerobic digestion of dairy waste water, and the further upgrading of the biogas products from the process. A combination pilot scale process was established with a 90 d start-up time using a 1 m3 continuous stirred tank reactor (CSTR) and a 0.2 m3 baffled reactor (ABR) in series. The system was fed at constant retention time in the ABR of 1.6 d and with varying substrate organic loading rates between 1.25 and 4.50 kg m3 d1. The average chemical oxygen demand (COD) removal was 82% with a biogas yield of 0.26 m3 kg1. The use of the derived biogas for the Sabatier process to convert hydrogen into CH4 showed no disadvantages compared to synthetic gas mixtures. The combination of CSTR and ABR overcame the individual disadvantages of both reactor types. The investigated anaerobic digestion system can be further optimized and adopted to replace conventional waste water treatment systems. © 2017 Published by Elsevier Ltd.
Keywords: Anaerobic digestion Dairy waste water Two-stage fermentation Biogas Upgrading
1. Introduction Anaerobic baffled reactor (ABR) systems have been widely applied as high rate digesters especially in developing countries where they have been used to meet on-site sanitation goals [1]. Numerous research papers have been published in the last two decades which have reported the potential advantages of the use of ABR as an excellent anaerobic digestion system for low- and highstrength waste water [2e6] and for the treatment of complex organic waste streams [7,8]. The simplicity and inexpensiveness of the ABR system, coupled with its non-requirement of moving parts or mechanical mixing for the anaerobic fermentation process has also strengthened the potential benefits associated with the use of this anaerobic digestion reactor system [9]. Due to the construction of the ABR (as a series of up-flow and down-flow sections), this reactor type enables an internal phase
* Corresponding author. E-mail address: [email protected] (L. Jürgensen).
separation, i. e. different process parameters can be maintained and manipulated in the different reactor compartments. This selective process differentiation in the reactor compartments could enable hydrogen production at low residence times, and improved overall methane yields from the anaerobic digestion process as reported in Refs. [10e12]. The industrial applications of the ABR scheme have however been mainly limited to waste water treatment [1] and have not yet been expanded for use in the digestion of other organic residue streams i. e. animal excreta and industrial organic residues (such as dairy and food processing waste). Dairy waste water effluent is generally regarded to contain a high organic load and is considered a useful substrate for anaerobic processes aimed at methane production [13,14]. The effluents from dairy processing plants (producing fresh milk and yogurt) have however been observed to be characterized by having a low chemical oxygen demand (COD). Typical COD values of the effluent in milk processing dairies are 2e10 kg m3, with up to 70 kg m3 reported for whey waste water [9,15]. Owing to such low COD values, the effluents of milk processing dairy plants can be considered as unsuitable for treatment in a conventional
http://dx.doi.org/10.1016/j.biombioe.2017.04.007 0961-9534/© 2017 Published by Elsevier Ltd.
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
2
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
continuous stirred tank reactor (CSTR)-type biogas plant [16]. Effluents which are considered as ”low-strength”are therefore often treated in high rate anaerobic digesters like upflow anaerobic sludge blanket (UASB) or biofilm reactors. These types of reactors use different hydraulic (HRT) and solid retention time (SRT) distributions, which favour the slow growing methanogenic archea thus improving methanogenesis [17]. These reactor designs are however limited when dealing with substrates which contain high amounts of fats or solids. This is mainly due to transfer restrictions brought about by the absence or lack of stirring or mechanical mixing components. Solids with densities higher than that of water usually sink and accumulate in the first chamber, with the fat contents of usually resulting in the formation of floating layers in the digester [1]. In both cases, organic components become inaccessible to the microbial community. Limitations to the use of currently available, cost effective anaerobic digestion treatment routes for low strength dairy waste water but with a high fat and solids content can therefore be observed and needs to be addressed to ensure the proper utilization of this valuable resource and to prevent potential environmental degradation processes (i. e. euthrophication of water bodies) associated with the nontreatment and disposal of such residues. This study investigates the combination of the CSTR and ABR schemes (in series) as a useful means of overcoming the disadvantages of both systems, with the goal of facilitating the use of this stream as a useful carbon source for the methanation process. The paper will not only highlight the methanogenic processes related to the bacterial flora of the digestion process i. e. eventually leading to the production of biogas (a mixture of methane (CH4), carbon dioxide (CO2) and other trace gases such as hydrogen sulphide (H2S) and ammonia), but will also look at the incoporation of an additional thermochemical process (Sabatier reaction) into the anaerobic digestion system with the overall goal of optimizing the overall methane (and hence energy) yields from the preceding digestion process. The methanation of CO2, also known as Sabatier process, is a thermochemical reaction at increased pressure and temperature using Nickel or Ruthenium catalyst. Previous research has demonstrated the use of this process for the conversion of the CO2 fractions of anaerobically derived biogas to CH4 With regards to the efficiency of the Sabatier process when used for biogas CO2 conversion, significant poisoning of the reaction catalyst (Nickel) was demonstrated to occur in the presence of small amounts of hydrogen sulphide (H2S [18]. A reduction of the biogas H2S levels (i. e. via desulphurisation processes) is therefore essential before the biogas can be applied for the thermochemical methanation process. Additionally, the influence of other trace substances, like ammonia and siloxanes contained in the anaerobically derived biogas on the Sabatier process and the reaction catalysts are unknown. And have not been extensively identified [19] due to the difficulty related to analysing such trace components using conventional biogas analysis systems [20]. The usage of actual biogas mixtures is therefore important to take into account, since it would influence the feasibility of such anerobic-thermochemical combination systems in practice. This is especially important since the use of, and integration with dairy waste water digestion was found to be lacking in the literature. With the presence of contaminants i. e. cleaning agents and disinfectants used for hygienic food processing plants potentially negatively affecting the anaerobic and thermochemical process, this study provides a useful assessment of their influence on both procesess. This study aims to highlight the importance of local residue streams from industrial processes for both, energy generation and as a carbon source. In a future of 100% renewable energy scenario,
electricity will be an important primary energy carrier [21]. To satisfy the needs of the transportation and industry sector, it is expected that hydrogen will be produced in decentralized units using renewable energy sources like wind and solar [22] and converted into fuels with better properties like CH4 and methanol [23]. This paper initially considers methods to improve the biogas yields from a currently underutilised dairy effluent water, and further assess the upgrading of the energy value (through an increase in the methane outputs) of the anaerobically derived biogas through the integration of the Sabatier process. The work therefore aims to demonstrate at a lab scale concepts of decentralised biogas production and upgrading with the intention to facilitate matching the biogas quality to natural gas grid standards as previously reported in the literature [24e26]. In addressing those issues the paper will also highlight the use of biogas production from a wide range of substrates. Such substrates could the be a local carbon source for the storage of renewable generated hydrogen. 2. Materials and methods The experimental setup used in this study is shown in Fig. 1. It includes sections for: biogas generation, conditioning, and upgrading. All values for feed strength, specific gas production and organic loading rate (OLR) are based on COD. 2.1. Anaerobic digestion using the CSTR-ABR combination system 2.1.1. Substrate The effluent water of a local dairy plant (Osterhusumer Meierei Witzwort EG, North Frisia, Germany) was used as primary substrate for the AD process. It was delivered once a week and stored at ambient temperatures in a 1 m3 intermediate bulk container. From here, it was pumped once a day into the well stirred 0.25 m3 feed tank. At the dairy plant production site, the effluent was collected from a buffer tank, which is used to feed the dairy's conventional (aerobic) waste water treatment plant and to avoid shock loading events to it. The dairy waste water initially contained varying amounts of NaOH and H3PO4, used in the plant cleaning processes. These were however neutralised before this tank. The phosphate content resulted in a high acid base capacity of the dairy waste water mixture. The chemical oxygen demand (COD) of the waste water sample was analysed after delivery to the experiment site. The COD levels of the substrate were observed to usually fluctuate between 2.5 and 5.0 kg m3 during the reactor operation. More extreme fluctuations for the substrate COD were however periodically recorded as indicated in Table 1 which shows the dairy waste water characteristics recorded during one year of operation. The levels of the dairy waste water COD, nitrate, and phosphorus levels were determined using Hach Lange cuvette tests. 2.1.2. Anaerobic digesters A 1 m3 CSTR and a 0.2 m3 ABR were used in series for this study. The CSTR was temperature controlled (operating temperature was 38 C) using a Pt-100 sensor, a micro-controller and four 1 kW heating rods, which were inserted from the top of the vessel into the liquid. The temperature probe was also equipped with a pH sensor. The whole steel vessel was properly insulated to maintain the temperature profile and prevent temperature loss from the reactor. A level drain was used to transfer the effluent to the ABR system by gravity. Each of the four chambers was controlled individually using Pt-100 sensors and heating foils fixed to the external reactor walls. The downflow-to-upflow ratio was 1:2. A siphon was used to drain the effluent to the sink. Each chamber was equipped with
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
3
Fig. 1. Experimental setup containing a two-step anaerobic process (1 m3 CSTR followed by a 200 l-ABR), gas storage, drying, desulfurization and compression. On the right hand side, a heterogeneous catalytic reactor is used to convert the CO2 fraction of the compressed biogas to CH4 using H2 from bottles.
Table 1 Characteristics of the dairy waste water recorded during one year of operation.
average min max
pH
COD in kg m3
NO3-N in g m3
total P in g m3
total N in g m3
8.2 7.5 10
3.0 0.9 6.4
0.48 0.30 0.83
26.3 8.6 52.7
172 99 300
three sample valves located in the bottom, middle, and top of each up-flow section. The outlets for channeling the produced biogas were located at the top of the CSTR and of each chamber of the ABR, and were connected to a common pipe. The common pipe was connected to a dip tube, ensuring an increased pressure of 10e20 kPa in the reactor headspace. A Ritter drum counter was used to measure the volume flow of the produced gas.
2.1.3. Anaerobic digester operation 0.4 m3 and 0.1 m3 sewage sludge from the municipal waste water treatment plant were used as the initial inoculants both for the CSTR and the ABR. After inoculation, the system was left to stand at 38 C for one week. A constant flow of 0.130 m3 d1 was fed to the process resulting in OLR between 1.3 and 4.3 kg m3 d1, according to the varying feed strength. The OLR was calculated on the basis of the ABR volume. Temperature, pH, and gas flows were recorded daily. The substrate COD, the effluent from the CSTR (i. e. the influent to the ABR) and the effluent of the ABR were analysed weekly using HACH Lange cuvette tests. The biogas potential was determined by standardised batch test according to DIN 38414. The COD removal efficiency was calculated as
CODðtÞout X ¼ 1 Pt ; COD ðt Þ t2t
in
2ttot
i
where ttot is the nominal residence time in the total system. Taking into account, that the feed strength changed every week, an average COD value over 2ttot ¼ 18:7d was used (presented in the denominator). The retention time of the ABR was 1.7 d and the retention time of the CSTR was 7.6 d). The gas composition of the whole system was analysed weekly and once a month. The gas composition from each reactor and chamber was also analysed.
2.2. Gas storage and compression The produced gas was delivered to two 90 l gas storage bags (Ritter). From there it was pressurised once a day to the feed-bottle of the methanation unit via a membrane compressor with 11 bar outlet pressure. Before the compressor, a gas cooler was used to remove the main water vapor content (condensate) and a filter filled with doped charcoal [27] was used as desulphurization chamber to remove the sulphur fraction of the biogas. 2.3. Biogas upgrading A tubular reactor 10 mm in diameter was used to convert the CO2 fraction of the biogas to methane. Commercially available Ni catalyst on silica-alumina support delivered by ALFA AESAR (catalogue number MFCD00011137) was used as the methanation catalyst. The catalyst used contained 66% Ni to provoke carbon formation within a shorter time [28] and therefore improve the reaction stability under biogas conditions. The catalyst was pelleted and sieved using two sieves (with mesh opening sizes of 500 mm and 250 mm). 5$103 kg of the fraction obtained on the 250 mm sieve were used for the experiments. A thermocouple was inserted in the catalytic bed to control the reaction temperature. The reactor was heated by an electrical furnace to 250 C. The system pressure was set to 600 kPa. Mass flow controllers were used to deliver the compressed biogas and hydrogen (or other gases for simulated biogas/gas mixtures) to the reactor. 1.5$103 m3 h1 biogas was mixed with a certain amount (depending on the actual biogas composition) of H2 to ensure a stoichiometric CO2:H2 mole ratio of 1:4. The biogas and the product gas composition were analysed using a the Biogasanalyzer VISIT 03 produced by Messtechnik EHEIM GmbH. Any deactivation of the catalyst would therefore be recognized due to reduced conversion observed for the upgrading process. 3. Results and discussion 3.1. pH profile and phase separation Assessing the pH profile for the process, the anaerobic digestion process can be divided into two phases: start up phase, and the steady-state operation phase (see Fig. 2). The term ”steady-state” as used here, means that the phase separation was already established and stable. The concentrations of the substrate however varied according to the waste water produced by the dairy plant.
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
4
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
Fig. 2. pH profile during the digestion of dairy waste water using a CSTR and an ABR in series: after 100 d, a stable and static pH profile was established.
Fig. 2 shows the pH values of the CSTR and each chamber of the ABR (indicated in the figure legend as Comp. 1 to 4. As observed, during the first 70 d, the pH levels along the reactor remained relatively constant. After day 70, a well developed pH profile was seen to have been established. The pH values in the CSTR dropped to values of 6.5, while a pH increase along the ABR can be seen. The effluent of the fourth and last chamber left the reactor at a pH of 7.2e7.4. 3.2. Reactor start-up 3.2.1. CSTR performance during start up During the first 60 d of operation, 90% of the total biogas was produced in the CSTR, while the ABR accounted for no (or only minor) contributions to the COD removal. A wash-out of the active sludge was also observed with the samples taken from the CSTR containing less sludge with time. This pseudo-steady state was observed to be altered beyond t ¼ 70 d. Fig. 3 shows the pH values (as also seen from 2) and the COD content of the CSTR effluent during the whole reactor operation. During the first 32 d, the pH values increased from 6.8 to 7.1 and was observed to remain constant between 7.0 and 7.1 for a further 30 d. Until 62 d, the COD removal in the CSTR was high, resulting in effluent COD values of 350 kg m3 (i. e. the influent to the ABR), which corresponds to a COD removal of 88%. While the pH value
remained over 7.0 for further 12 d, the COD increased rapidly, indicating an acidification of the reactor. In a single-step anaerobic digestion process, this would indicate the imminent failure of the fermentation process. However, using the investigated combination process, when the pH dropped below 7.0, the COD removal efficiency was already under 65%. From day 74 to day 94, the pH dropped to 6.3 and remained constant for the following 100 d at 6.5. The COD of the CSTR effluent increased to values between 2.5 and 3.5 kg m3, corresponding to an average COD removal efficiency of 28% (see Fig. 4). The fraction of the biogas produced by the CSTR decreased to 10% of the total biogas production. 3.2.2. Gas production and COD removal The gas production increased during the first 20 d up to >0.150 m3 d1, which can be explained by high amount of active sludge available after inoculation. This sludge was washed out due to the comparatively low retention time in the CSTR of 7.7 d. This washout resulted in a decrease of the biogas production to 0.110e0.120 m3 d1 during the later start up stages. The overall COD removal (i. e. by the combined CSTR and ABR system) from the beginning of the digestion period was observed to be high. Before phase separation (i. e. decrease of pH in the CSTR), the average COD removal was 91%. With the drop in the pH values in the CSTR, the overall COD removal consequently decreased to 59%, later increasing within one week back to >79%. Fig. 4(b) shows the COD removal efficiencies for CSTR and the whole system. The COD removal by the ABR was observed to be decreasing, while the overall COD removal remained at a high level, indicating that the ABR was working well. 3.3. Steady-state operation As mentioned above, steady-state operation was established between day 90 and 190 as seen in Fig. 2.
Fig. 3. Start up of the CSTR: Beyond t ¼ 74 d, the pH dropped significant, indicating the acidification of the hydrolysis stage. Simultaneously, the effluent from the CSTR increased in COD by inhibition of methanogenes.
3.3.1. COD removal and gas yield The overall COD removal using the combined system was observed to be high with an average of 82%. Fig. 4(a) and (b) shows the COD of the substrate and the ABR effluent. After phase separation, the COD removal was seen to decrease slightly, but it overall maintained a high level, even when the feed concentration changed rapidly. The biogas production was 122 l d1. As expected, the biogas production increased and decreased according to the OLR. The
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
5
Fig. 4. System performance during start up and steady state operation: (a) CODs, (b) OLR and COD removal efficiencies, (c) total gas production and (d) gas yield based on the removed COD. The variations in OLR are caused by variation of the dairy waste water strength.
average biogas yield was 0.264 m3 kg1, based on the removed COD. Fig. 5 shows the biogas potential of the dairy waste water. Additionally, the fat fraction was separated by flocculation to demonstrate the waste water components which contributes to the gas yield. Here, the biogas potential was measured to be 0.58 m3 kg1 based on the COD of the sample used for this batch test, which was high in COD (5.6 kg m3). This difference can be explained by the high amount of fat which comes with waste water loaded with a high COD, resulting in high gas yields due to the fat degradation. Furthermore, the high amount of water passing the reactor, corresponding to the low retention time, will result in less CO2 released to the gas phase. Thus, the measured amount of biogas in
the continuous experiments using a high rate anaerobic digester will expectedly be lower compared to the batch test.
3.3.2. Biogas composition The biogas composition was high in CH4 during the whole fermentation. The average methane volume fraction was 76% ± 2% with minor amounts of hydrogen released from the first reactor. Such high methane concentrations have been previously reported as typically following the digestion of dairy waste water [15]. The gas released from the CSTR contained up to 10% hydrogen due to hydrolysis of fats, which is in good agreement with other studies conducted on the anaerobic production of drogen
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
6
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
Fig. 5. Biogas potential batch test results of the waste water (given in m3/kg based on organic dry matter (oDM) and COD), the fat fraction separated by flocculation and the inoculum as reference.
production from fats [29]. However, the gas production from the CSTR was only approximately 10% of the total biogas production from the combination system. Thus, the amount of hydrogen (produced via the hydrolysis of fats and from the dairy waste water) in the total biogas stream was considered to be negligible. 3.3.3. Shock-load stability During the operation, several shock-load events can be seen from Fig. 4: On day 110, 139, and 179, the COD values increased to from <3.0 to >5.5 kg m3. As explained before, the shock loadings are caused by variations in the dairy waste water strength which varied due to the dairy processes. The COD of the CSTR's effluent was influenced only slightly and also the COD of the effluent of the ABR. It becomes clear that the CSTR acts as (i) a hydrolysis stage and (ii) as a buffer tank. This is since the kinetics of the hydrolysis reactions are quite fast. However the use of a CSTR, with 7.6 d retention time in this study leads to a considerable reduction in the shock loading to the ABR. However, an optimum size of CSTR:ABR has to be found but that optimization was not part of this study. 3.4. Biogas upgrading Over 30 d, the catalytic system remained stable showing no deactivation of the catalyst. This proves that real biogas emanating from the dairy waste water digestion is suitable as feed gas for the Sabatier process. As previous experiments indicated [19], the ammonia content of the biogas did not affect the conversion reaction of the Sabatier process in this study. To establish such an upgrading process using biogas from anaerobic digestion, no special pretreatment technologies for ammonia removal from the feed biogas are necessary. With regards to the removal of the harmful H2S biogas constituents, this can be achieved using doped charcoal (e. g. potassium iodide), which is commercially available [30], so its adaption and integration in practice should be easy. In a practical local context, anaerobic digestion systems could therefore utilize waste streams from the food industry to supply a
highly concentrated CO2 source to the Sabatier process. In North Frisa, Germany, the county where the dairy is located (which provided the study dairy waste water substrate), 1.6 GW wind power is installed [31], resulting in overloaded electrical grids [32]. To avoid a breakdown of the electrical grid, wind times are curtailed up to 1660 h/a [33]. Instead of curtailing the electricity generation from wind and solar to avoid overloads, the surplus could be used to produce hydrogen via water electrolysis. This renewable hydrogen can then be applied for biogas upgrading, which in turns serves as a scheme for (chemical) energy storage. 4. Conclusions As demonstrated by the results of this study, anaerobic digestion can be applied as a useful scheme to convert food industry wastes streams (i. e. dairy industry wastes) which are currently underutilised in energy generation due to the issues associated with their use in conventional processes. The combination of CSTR and ABR (in series) could be a useful solution to overcome the disadvantages of both reactors. This is true in particular because high-rate anaerobic digesters are usually unable to handle the characteristically high particles and fats contents contained in such waste water streams. This problem is therefore improved with the use of the proposed combination scheme. Anaerobic digestion facilities provide a useful option for electricity storage via its use for the upgrading of the biogas to methane (according to natural gas grid standards) which can be conveniently stored and used in other applications. The excess electricity is therefore used for the electrolysis process for hydrogen production which is consumed in the Sabatier process. The use of biogas derived from dairy effluent water as a digestion substrate shows no disadvantages in the Sabatier process compared to synthetic mixtures. Acknowledgment The authors will like to acknowledge the European INTERREG
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007
L. Jürgensen et al. / Biomass and Bioenergy xxx (2017) 1e7
Region Fund that facilitated the scientific work presented in this paper through the funding of the project Large Scale Bio Energy Lab II. The authors will also like to sincerely thank all the students and staff who worked with us during the project period. References [1] K. Foxon, S. Pillay, T. Lalbahadur, N. Rodda, F. Holder, C. Buckley, The anaerobic baffled reactor (abr): an appropriate technology for on-site sanitation, Water SA 30 (5) (2007) 44e50. [2] W.P. Barber, D.C. Stuckey, The use of the anaerobic baffled reactor (abr) for wastewater treatment: a review, Water Res. 33 (7) (1999) 1559e1578, http:// dx.doi.org/10.1016/S0043-1354(98)00371-6. http://www.sciencedirect.com/ science/article/pii/S0043135498003716. [3] G. Gopala Krishna, P. Kumar, P. Kumar, Treatment of low-strength soluble wastewater using an anaerobic baffled reactor (abr), J. Environ. Manag. 90 (1) (2009) 166e176. [4] S. Bodkhe, A modified anaerobic baffled reactor for municipal wastewater treatment, J. Environ. Manag. 90 (8) (2009) 2488e2493. [5] Gopala Krishna, G.V.T.,P. Kumar, P. Kumar, Treatment of low strength complex wastewater using an anaerobic baffled reactor (abr), Bioresour. Technol. 99 (17) (2008) 8193e8200, http://dx.doi.org/10.1016/j.biortech.2008.03.016. \url {18455391}. [6] G.D. Ji, T.H. Sun, J.R. Ni, J.J. Tong, Anaerobic baffled reactor (abr) for treating heavy oil produced water with high concentrations of salt and poor nutrient, Bioresour. Technol. 100 (3) (2009) 1108e1114, http://dx.doi.org/10.1016/ j.biortech.2008.08.015. \url{18793845}. [7] J. Zhang, Y. Wei, W. Xiao, Z. Zhou, X. Yan, Performance and spatial community succession of an anaerobic baffled reactor treating acetone-butanol-ethanol fermentation wastewater, Bioresour. Technol. 102 (16) (2011) 7407e7414, http://dx.doi.org/10.1016/j.biortech.2011.05.035. \url{21664129}. [8] A. Bayrakdar, E. Sahinkaya, M. Gungor, S. Uyanik, A.D. Atasoy, Performance of sulfidogenic anaerobic baffled reactor (abr) treating acidic and zinc-containing wastewater, Bioresour. Technol. 100 (19) (2009) 4354e4360, http:// dx.doi.org/10.1016/j.biortech.2009.04.028. [9] M. Nasr, A. Khalil, A. Abdel-Kader, W. Elbarki, M. Moustafa, Comparative performance between conventional and hybrid anaerobic baffled reactors for treating dairy wastewater, Int. J. 5 (4). [10] A. Ahamed, C.-L. Chen, R. Rajagopal, D. Wu, Y. Mao, I. Ho, J. Lim, J.-Y. Wang, Multi-phased anaerobic baffled reactor treating food waste, Bioresour. Technol. 182 (2015) 239e244. [11] A. Tawfik, M. El-Qelish, Continuous hydrogen production from co-digestion of municipal food waste and kitchen wastewater in mesophilic anaerobic baffled reactor, Bioresour. Technol. 114 (2012) 270e274. [12] P. Thanwised, W. Wirojanagud, A. Reungsang, Effect of hydraulic retention time on hydrogen production and chemical oxygen demand removal from tapioca wastewater using anaerobic mixed cultures in anaerobic baffled reactor (abr), Int. J. Hydrogen Energy 37 (20) (2012) 15503e15510. [13] S.V. Mohan, V.L. Babu, P. Sarma, Anaerobic biohydrogen production from dairy wastewater treatment in sequencing batch reactor (ansbr): effect of organic loading rate, Enzyme Microb. Technol. 41 (4) (2007) 506e515. [14] B. Sarkar, P. Chakrabarti, A. Vijaykumar, V. Kale, Wastewater treatment in dairy industries at possibility of reuse, Desalination 195 (1) (2006) 141e152. [15] B. Demirel, O. Yenigun, T.T. Onay, Anaerobic treatment of dairy wastewaters: a review, Process Biochem. 40 (8) (2005) 2583e2595, http://dx.doi.org/
7
10.1016/j.procbio.2004.12.015. € rog lu, B. Ozkaya, M. Cakmakci, A review on anaerobic [16] D. Karadag, O.E. Ko biofilm reactors for the treatment of dairy industry wastewater, Process Biochem. 50 (2) (2015) 262e271. [17] L. Seghezzo, G. Zeeman, J.B. van Lier, H. Hamelers, G. Lettinga, A review: the anaerobic treatment of sewage in uasb and egsb reactors, Bioresour. Technol. 65 (3) (1998) 175e190. [18] M. Twigg, M. Twigg, Catalyst Handbook, CSIRO, 1989. [19] L. Jürgensen, E.A. Ehimen, J. Born, J.B. Holm-Nielsen, D. Rooney, Influence of trace substances on methanation catalysts used in dynamic biogas upgrading, Bioresour. Technol. 178 (2015) 319e322. [20] R. Dewil, L. Appels, J. Baeyens, Energy use of biogas hampered by the presence of siloxanes, Energy Convers. Manag. 47 (13e14) (2006) 1711e1722, http:// dx.doi.org/10.1016/j.enconman.2005.10.016. [21] U. Bossel, Does a hydrogen economy make sense? Proc. IEEE 94 (10) (2006) 1826e1837, http://dx.doi.org/10.1109/JPROC.2006.883715. [22] R.R. Dickinson, D.L. Battye, V.M. Linton, P.J. Ashman, Nathan, J. Graham, Alternative carriers for remote renewable energy sources using existing cng infrastructure, Int. J. Hydrogen Energy 35 (3) (2010) 1321e1329, http:// dx.doi.org/10.1016/j.ijhydene.2009.11.052. [23] G.A. Olah, A. Goeppert, Prakash, G.K. Surya, Chemical recycling of carbon dioxide to methanol and dimethyl ether: from greenhouse gas to renewable, environmentally carbon neutral fuels and synthetic hydrocarbons, J. Org. Chem. 74 (2) (2009) 487e498, http://dx.doi.org/10.1021/jo801260f. [24] E.J. Hengeveld, W. van Gemert, J. Bekkering, A.A. Broekhuis, When does decentralized production of biogas and centralized upgrading and injection into the natural gas grid make sense? Biomass Bioenergy 67 (2014) 363e371, http://dx.doi.org/10.1016/j.biombioe.2014.05.017. [25] E.J. Hengeveld, J. Bekkering, W. van Gemert, A.A. Broekhuis, Biogas infrastructures from farm to regional scale, prospects of biogas transport grids, Biomass Bioenergy 86 (2016) 43e52, http://dx.doi.org/10.1016/ j.biombioe.2016.01.005. Antonio, Assessment of membrane [26] G. Valenti, A. Arcidiacono, Nieto Ruiz, Jose plants for biogas upgrading to biomethane at zero methane emission, Biomass Bioenergy 85 (2016) 35e47, http://dx.doi.org/10.1016/ j.biombioe.2015.11.020. [27] S. Rossow, Entwicklng einer speziellen aktivkohle für den einsatz zur biogasentschwefelung und untersuchung der leistungsf€ ahigkeit im labor und im praktischen einsatz, (Dissertation). [28] W.-S. Dong, H.-S. Roh, K.-W. Jun, S.-E. Park, Y.-S. Oh, Methane reforming over ni/ce-zro 2 catalysts: effect of nickel content, Appl. Catal. A General 226 (1) (2002) 63e72. [29] M. Monti, A. Scoma, G. Martinez, L. Bertin, F. Fava, Uncoupled hydrogen and volatile fatty acids generation in a two-step biotechnological anaerobic process fed with actual site wastewater, New Biotechnol. 32 (3) (2015) 341e346, http://dx.doi.org/10.1016/j.nbt.2014.08.002. \url{25174889}. [30] Philipp Adler, et al., Leitfaden Biogasaufbereitung und -einspeisung, FNR, Federal Ministry of Food and Agriculture Germany. [31] T. Engel, Die karte der erneuerbaren energien, Database provided by the German Association of Solar Energy, energymap.info, 2015. [32] L. Jürgensen, E.A. Ehimen, J. Born, J.B. Holm-Nielsen, Utilization of surplus electricity from wind power for dynamic biogas upgrading: Northern Germany case study, Biomass Bioenergy 66 (2014) 126e132. [33] L. Jürgensen, E. A. Ehimen, J. Born, J. B. Holm-Nielsen, Use of the Sabatier process for dynamic biogas upgrading in northern Germany, 22nd European Biomass Conference and Exhibition, Hamburg (2014) 881e886.
Please cite this article in press as: L. Jürgensen, et al., A combination anaerobic digestion scheme for biogas production from dairy effluentdCSTR and ABR, and biogas upgrading, Biomass and Bioenergy (2017), http://dx.doi.org/10.1016/j.biombioe.2017.04.007