Bioresource Technology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system M. Burkhardt ⇑, T. Koschack, G. Busch Faculty of Environmental Science and Process Engineering, Brandenburg University of Technology Cottbus-Senftenberg, Department of Waste Management, Siemens Halske Ring 8, 03046 Cottbus, Germany
h i g h l i g h t s Higher H2 loading rate than known so far in biocatalyzed systems are possible. There is no gas mixing or recirculation necessary. The quality of produced biogas is very high ðcCH4 ¼ 98 vol%Þ. The conversion rate is depending on H2-flow rate and trickle flow rate. Biogas can be used as carbon dioxide source also.
a r t i c l e
i n f o
Article history: Received 6 June 2014 Received in revised form 4 August 2014 Accepted 5 August 2014 Available online xxxx Keywords: Methanation Methane enrichment Power to gas Hydrogen Anaerobic digestion
a b s t r a c t A new type of anaerobic trickle-bed reactor was used for biocatalytic methanation of hydrogen and carbon dioxide under mesophilic temperatures and ambient pressure in a continuous process. The conversion of gaseous substrates through immobilized hydrogenotrophic methanogenic archaea in a biofilm is a unique feature of this type of reactor. Due to the formation of a three-phase system on the carrier surface and operation as a plug flow reactor without gas recirculation, a complete reaction could be observed. With a methane concentration higher than cCH4 ¼ 98%, the product gas exhibits a very high quality. A specific methane production of PCH4 ¼ 1:49 Nm3 =ðm3SV dÞ was achieved at a hydraulic loading rate of LRH2 ¼ 6:0 Nm3 =ðm3SV dÞ. The relation between trickle flow through the reactor and productivity could be shown. An application for methane enrichment in combination with biogas facilities as a source of carbon dioxide has also been positively proven. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction The development of renewable energies, especially in the wind and solar energy sectors, is essential in order to obtain a shift away from reliance on fossil fuels. Great strides have already been made in Germany (DLR, 2012). According to predictions made by the German Federal Environment Ministry in the year 2011, the portion of energy coming from wind and solar power will more than triple by 2050. The federal government has set the goal of raising the portion of energy from renewable sources for gross electricity consumption to 35% by 2020 and 80% by 2050. Notably, the development of offshore wind performance will be accelerated to 25 GW by 2030 and the network infrastructure (north–south route) will be extended. A full use of this potential is already pushing the threshold since energy use and temporary buffer storage are both ⇑ Corresponding author. Tel.: +49 355694328. E-mail address: [email protected] (M. Burkhardt).
limited in times of high energy production. Due to inadequate development of the electricity network, energy input and distribution are limited. Wind and solar units that have been installed need to be connected to the network. An interesting solution appears to be the power-to-gas strategy for the storage and subsequent repowering of energy. Alongside thermal (hot water cylinder), electrical (superconductor) and mechanical (flywheel) storage, electrochemical storage has proven itself to be particularly beneficial. An alternative to storage in an accumulator is the possibility of conversion to hydrogen. The gaseous energy source can be won through the electrolysis of water. However, until now only minimal utilizations and technological thresholds have been available for storage and use. Possible solutions include direct H2 power reconversion, use in automobiles run on H2, input of H2 into the natural gas grid or the generation of CH4 through the Sabatier process. But, these applications are only developed to a small degree. Consequently, their application is limited and they are accompanied by energy loss, cable and pipe-
http://dx.doi.org/10.1016/j.biortech.2014.08.023 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Burkhardt, M., et al. Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.08.023
2
M. Burkhardt et al. / Bioresource Technology xxx (2014) xxx–xxx
line damage, and significant costs, causing them to not yet be economically advantageous. An alternative is biologically catalyzed methanation, which can take place at more moderate temperatures than physico-chemical processes and has a higher resistance to contaminations, such as H2S, organic acids and NH4. In addition to hydrogen, carbon dioxide is needed as a reaction partner in methanation. Sources can be CO2-rich exhaust gas or biogas, which usually contains 30% to 50% carbon dioxide. In these systems the gases are typically fed into a liquid tensed fermenter (Krajete, 2012). But, bubbles often accumulate and ascend in the reactor. The input gases are lead out of the reactor within minutes according to the rate of bubble ascension. Consequently, the contact time, or reaction time, is not sufficient and the achievement potential of the system is limited. A methane productivity of P CH4 ¼ 0:11 Nm3 =ðm3 dÞ could be achieved so far, which is very low. A higher distribution of the bubbling gases in the liquid was possible by using a hollow fiber membrane in a continuously stirred tank reactor (CSTR), which resulted in a very high methane concentration of cCH4 ¼ 99% (Wang et al., 2013). The productivity of the simultaneous biomethanation of sewage sludge and H2-rich coke oven gas could be increased from PCH4 ¼ 0:28 m3 =ðm3reactor dÞ up to P CH4 ¼ 0:65 m3 =ðm3reactor dÞ, which is a very interesting application. Thermophilic methanation has also been conducted in a fixed bed reactor by using pure culture Methanobacterium thermoautotrophicum (Jee et al., 1988). The cost of obtaining the very high productivity of PCH4 ¼ 5:3—10:2 m3 =ðm3 hÞ was an incomplete conversion and low methane concentrations of cCH4 ¼ 15:5—87% in the reactor output. In an attempt to increase the methane concentration of the product, a further experiment passed the inflowing gas through a liquid tensed fixed bed reactor (Bugante et al., 1989). The gas was recirculated with an optimal recirculation speed of Qg = 40 l/h. Nonetheless, while running as a batch process, the final productivity of PCH4 ¼ 0:025 m3 =ðm3reactor dÞ was very low due to the necessity of recirculation. Alternatively, an anaerobic trickle bed reactor was investigated for the methanation of gaseous substrates for the recovery of methane (Burkhardt and Busch, 2013). While running in batch mode and recirculating the product gas, the biocatalyzed methanation of carbon dioxide and hydrogen demonstrated a higher productivity of P CH4 ¼ 1:17 m3 =ðm3reactor dÞ. By using a fermenter equipped with an impeller and a porous Teflon or porous glass, a very high productivity of PCH4 ¼ 288 m3 =ðm3reactor dÞ with cCH4 ¼ 96% could be observed. The result could be reached through a high energy input by the impeller, with it running at 1200 rpm (Peillex et al., 1990).
2. Methods The trickle-bed reactor varies from all of the other known reactors for biogas production, such as impounded or classic fixed-bed reactors, in that the reaction chamber is not filled with liquid. It contains packed bed as a surface on which microorganisms can immobilize. They are surrounded on all sides by a gas phase and are merely sprinkled with a limited amount of liquid. In this way, a three phase system exists (biofilm–liquid phase–gas phase) (Decressin, 1998). Material transport in the biofilm is significantly improved as a result of the higher concentration gradient as a driving force for the mass transfer and short diffusion paths. The thickness of the liquid phase has a decisive influence on the concentration gradient of the gases released and transported to the biofilm. This advantage increases the methanation and final productivity of the whole system. The concentration gradient is a function of the diffusion coefficient, film thickness d and material conversion rate. One needs to be aware that the concentration of hydrogen and carbon dioxide in the water film and biofilm differ
from one another due to varying solubility and diffusion coefficients. As the case may be, these basic conditions may lead to limiting effects. Continuous reactor operation took place in a V = 88 l reactor with the proportions of length l:diameter d = 5:1. The reactor was gas-tight, filled up to a volume of VSV = 61 l with packing material and heated up to 37 ± 0.5 °C with a double jacket and a 20 l heat bath (LAUDA). The packing material, Bioflow 40 (RAUSCHERT), offers a high specific surface area of a = 305 m2/m3. The inoculum was not an isolated strain of methanogenic archaea. The aim was to investigate a complex population diversity fixed in a biofilm. To reach that variety of methanogens, anaerobic sludge of a local sewage plant was used as the inoculum. Therefore, archaeal species like Methanococcus spp., Methanobacterius spp., Methanomicrobium and also Methanosaeta spp. could be confirmed. They were immobilized in a biofilm on the carrier material. To ensure a moist environment and the provision of nutrients and trace elements for the microorganisms, a processing liquid was circulated in the reactor. Recirculation took place constantly at Q RV ¼ 6:15 m3 =ðm3sv dÞ. The processing liquid was held available in the floor of the reactor in the reactor sump. Nutrient and trace element concentrations were set to Al 16.95 mg/l, B 3.63 mg/l, Ca 48.7 mg/l, Co 0.5 mg/l, Cu 0.3 mg/l, Fe 21.6 mg/l, K 303.6 mg/l, Mg 28.5 mg/l, Mn 0.57 mg/l, Mo 0.27 mg/l, Na 132.2 mg/l and Ni 3.96 mg/l. The pH value ranged from 7.2 to 7.4. The feeding in of hydrogen and carbon dioxide gases took place continuously underneath the packing bed through a H2-generator via electrolysis and conventional CO2-pressure cylinders. According to Eq. (1), in doing so the stoichiometric relationship between hydrogen and carbon dioxide needed to remain at 4:1. Mass flow controllers (BROOKS INSTRUMENT) were installed to proportion the substrate feeding, which meant that the rate of flow and load could be varied. The gaseous substrates flowed through the reactor from bottom to top and were converted by methanogenic microorganisms. In the process a plug flow was formed. The calculated average retention time was reduced to a value of s = 4 h. The reactor product, methane, was lead out through the head of the reactor. Determination of the volume took place in the drum gas counter TG05 (RITTER) with storage in a gas bag (TESSERAUX). Determination of the concentrations of methane, hydrogen, carbon dioxide and oxygen took place nearly continuously by way of the SSM6000 LT (PRONOVA). At the same time the gas temperature T CH4 and the gas pressure P CH4 were recorded. Methane production for various loading rates, LRH2 ¼ 0:4—6:0 Nm3 =ðm3SV dÞ, was investigated. The second reaction product, water, was collected and taken away by way of the reactor sump. To avoid losses of nutrients and trace elements, the nutrient solution was added again to keep the required concentration constant. In further observations the flow rate was varied between Q RV ¼ 3:35 and 13:5 m3 =ðm3sv dÞ to investigate the influence of a liquid layer on mass transfer into the biofilm and final methane production. By changing pure CO2 to biogas, mainly consisting of CO2 (23 ± 3 vol%) and CH4 (78 ± 2 vol%), biogas could also be observed as a carbon source and suitable for biological methane enrichment. The Biogas was produced continuously and pumped in stoichiometric relation to H2 flow into the trickle bed reactor. Gas purification and dewatering was not done.
4H2 þ CO2 ! CH4 þ 2H2 O DHoR ¼ 167 KJ=mol
ð1Þ
Taking into consideration the trickle-bed volume VSV, Eq. (1) can be used to determine and depict the theoretical methane production (Eq. (2)). The methane productivity P was determined from the experimental methane volumetric flow under standard conditions and the methane concentration (Eq. (3)). P was used to assess the reactor performance based on the investigated parameters.
Please cite this article in press as: Burkhardt, M., et al. Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.08.023
3
M. Burkhardt et al. / Bioresource Technology xxx (2014) xxx–xxx nH
Ptheo ¼
Q H2 nCH2
4
V SV
"
Nm3CH4
#
m3SV d
amount of CO2 being added, a full conversion could again be attained. Fig. 2 illustrates the progressive increase in the loading rate LR over the investigated period. The H2 loading rate was successively raised to a value of LRH2 ¼ 4:83 Nm3 =ðm3SV dÞ. The experimental methane production P CH4 correlates to the theoretical value Ptheo for nearly all of the loading rates. Even with an H2 loading rate of LRH2 ¼ 4:83 Nm3 =ðm3SV dÞ, a methane productivity of PCH4 ¼ 1:2 Nm34 =ðm3SV dÞ could be achieved. Up to this H2 load the limit of the trickle-bed reactors achievement potential could not be determined. The degree of H2 degradation was nearly always near to gH2 ¼ 100%. Beside methane production, the substrate conversion at a loading rate of LRH2 ¼ 4:02 Nm3 =ðm3SV dÞ results in a minor water yield of 1:6 Nlðm3SV dÞ. By using Eq. (1) the theoretical water production according to the ideal molar ratio is calculated with V H2 O;theo ¼ 1:6 Nl=ðm3SV dÞ and is similar, obviously. The results of our experiment in a bioreactor have not been seen by other workers at similar scales (Jee et al., 1988). Compared to Burkhardt and Busch and Bugante et al., a continuous process was possible in our experiment and no gas recirculation was required. The formation of a plug flow in the fixed bed allowed raised loading rates in a continuous process. As demonstrated, the trickle-bed process has proven itself to have a 13 times higher achievement potential than other known biocatalytic processes (Krajete, 2012). One reason for this is the higher retention time that allows for a longer reaction time, which was sought after in
ð2Þ
" # 3 Q product cCH4 NmCH4 3 P¼ V SV mSV d
ð3Þ
The procedure described is already under review to receive a patent (Burkhardt, 2013).
3. Results and discussion At the beginning of the experiment air was expelled from the reactor so that a constant atmosphere was reached by the 20th day and analysis could begin. Neither oxygen nor nitrogen could be detected any longer (Fig. 1). At this point, a methane concentration of up to cCH4 ¼ 98% could be ascertained from the discharge of the bioreactor. The feed materials, CO2 and H2, were almost completely converted. The notable fluctuations can be attributed to small deviations in the stoichiometric relationship of the gases to each other, which needed an adjustment after an increase in the loading rate. Thereby it was determined that a H2–CO2 relationship of 3.76:1 is needed for a complete conversion. This excess of CO2 can most likely be explained by an increased need for C through the biomass build-up. The recognizable demodification of the 100 90 80
ci [%]
70 60 50 40 30 20 10 0 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 carbon dioxide hydrogen
oxygen methane
time t [d]
Fig. 1. Biogas composition during continuous operation of the anaerobic trickle-bed reactor.
8
1,4
7
1,2 1,0
5 0,8 4 0,6 3 0,4
2
P [Nm³CH4 /(m³SV⋅d)]
LRH2 [Nm 3H2/(m3SV⋅d)]
6
0,2
1 0 20
30
40
50
60
70
80
90
100
0,0 110 time t [d]
loading rate LR
methane productivity P
theoretical methane productivity Ptheo
Fig. 2. H2 loading rate LR and the development of the experimental methane productivity P compared to the theoretical methane productivity Ptheo.
Please cite this article in press as: Burkhardt, M., et al. Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.08.023
4
M. Burkhardt et al. / Bioresource Technology xxx (2014) xxx–xxx
this case. Furthermore, due to the development of the three phase system in the trickle-bed reactor, there is an increased amount of substrate transport in the biofilm, which is a result of raised concentration gradients on the concentration boundary layer Lc. Processes merely using membrane techniques and fast rotating stirrers allow for higher productivities than in our current research (Peillex et al., 1990), but they are limited to very small reactor volumes due to the very high energy consumption and technological boundaries of using impellers at larger scales. The input of substrate was suspended from the 86th to the 89th day of the experiment ðLRH2 ¼ 0 Nm3 =ðm3SV dÞÞ. After starting operations again on the 89th day (LRH2 ¼ 4:02 Nm3 =ðm3SV dÞ), the same achievement potential as before the break in operations could be obtained. The methane concentration of the product gas was again at cCH4 > 90%. This proved the ability to conduct flexible reactor operations. The methane production is considered very controllable and flexible since the achievement potential was immediately obtainable again after a suspension of substrate input. Therefore, reactor operation can be carried out independently from substrate availability. It is possible to react on short notice to fluctuations of need and availability of product gases as well as input substrate. By changing the liquid recirculation rate QRV between 3.3 and 9:2 m3 =ðm3SV dÞ the influence of liquid phase thickness on the mass transfer into the biofilm was investigated. The entire liquid distribution in the trickle bed must be maintained. A correlation of the increasing methane concentration (and increasing hydrogen conversion) by a decreasing liquid recirculation rate was observed. Therefore higher H2 loading rates of LRH2 ¼ 6:00 Nm3 =ðm3SV dÞ could be run at lower liquid recirculation rates QRV. It results in a higher methane productivity of PCH4 ¼ 1:49 Nm3 =ðm3SV dÞ. Instead of feeding pure carbon dioxide, biogas can be used as a CO2 source. The Biogas with a methane concentration of 77% was produced in a double stage process which is a typical average. Due to passing it into the trickle bed reactor and an additional H2 flow (LRH2 ¼ 0:51 Nm3 =ðm3SV dÞ) according to the essential stoichiometrics, a methane enrichment up to cCH4 ¼ 96% could be observed. The amount of CO2 was reduced nearly completely. The methane production increased by 23%. Further experiments of increasing LRH2 will show the limits of the reactor performance. The process thus suits itself very well for application in the power-to-gas strategy since a storable and usable methane gas can be produced after recovering electrolyzed hydrogen. Considering that the methane is nearly the quality of natural gas, it is possible to feed it into the natural gas grid with only minimal filtration necessary. Alternatively, it constitutes a sought-after, basic material for synthesis. Using biogas as a source of carbon dioxide and maintaining consideration of the appropriate hydrogen input, methane enrichment can take place in a reactor. The methane pro-
duction can be increased by 23%. Traces of H2S, NH4 or water vapor are unproblematic for the process, although if necessary they can be removed before the product is put into storage. Consequently, the trickle-bed reactor is applicable as downstream methane enrichment step in which no by-products accumulate. 4. Conclusion By means of the anaerobic trickle-bed process, hydrogen and carbon dioxide were converted biocatalytically to methane. A concentration of up to 98% could be achieved. A methane productivity of P CH4 ¼ 1:49 Nm3 =ðm3SV dÞ for a H2 loading rate of LRH2 ¼ 6:00 Nm3 =ðm3SV dÞ and a liquid recirculation rate of Q RV ¼ 3:1 m3 =ðm3SV dÞ could be established. While running the process continuously, there was no necessity for gas recirculation or other gas mixing. The present technology can be used for coupling of various renewable energy sources, to convert electrical into chemical bound energy and to store methane into the natural gas grid. Acknowledgements The authors wish to thank Mr. U. Klopsch, Mr. R. Kabelitz, Mr. T. Koschack and Mr. E. Sueß for their help with analytical and technical issues. References Bugante, E.C., Shimomura, Y., Tanaka, T., Taniguchi, M., Oi, A., 1989. Methane production from hydrogen and carbon dioxide and monoxide in a column bioreactor of thermophilic methanogens by gas recirculation. J. Ferment. Bioeng. 67 (6), 419–442. Burkhardt, M., Busch, G., 2013. Methanation of hydrogen and carbon dioxide. Appl. Energy 111, 74–79. Burkhardt, M., Busch, G., Großmann, J., 2013. DE 102013209734.4, Verfahren und Vorrichtung für die Methanisierung von Gasen mittels Rieselbettreaktoren. Patent. Decressin, O., 1998. Biofilter-Modellbildung, Verifikation und Simulation. FITVerlag, Paderborn. DLR, 2012. . Langfristszenarien und Strategien für den Ausbau erneuerbarer Energien in Deutschland bei Berücksichtigung der Entwicklung in Europa und global. Jee, H., Nishio, N., Ngai, S., 1988. Continuous CH4 production from H2 and CO2 by Methanobacterium thermoautotrophicum in a fixed bed reactor. J. Ferment. Technol. 66 (2), 235–238. Krajete, A., 2012. WO 2012/110256 A1, 2012. Method of converting carbon dioxide and hydrogen to methane by microorganisms, Patent. Peillex, J.P., Fardeau, M.L., Belaich, J.P., 1990. Growth of Methanobacterium thermoautotrophicum on H2–CO2: high CH4 productivities in continuous culture. Biomass 21 (4), 315–321. Wang, W., Xie, L., Luo, G., Zhou, Q., Angelidaki, I., 2013. Performance and microbial community analysis of the anaerobic reactor with coke oven gas biomethanation and in situ biogas upgrading. Bioresour. Technol. 146, 234–239.
Please cite this article in press as: Burkhardt, M., et al. Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.08.023
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system M. Burkhardt ⇑, T. Koschack, G. Busch Faculty of Environmental Science and Process Engineering, Brandenburg University of Technology Cottbus-Senftenberg, Department of Waste Management, Siemens Halske Ring 8, 03046 Cottbus, Germany
h i g h l i g h t s Higher H2 loading rate than known so far in biocatalyzed systems are possible. There is no gas mixing or recirculation necessary. The quality of produced biogas is very high ðcCH4 ¼ 98 vol%Þ. The conversion rate is depending on H2-flow rate and trickle flow rate. Biogas can be used as carbon dioxide source also.
a r t i c l e
i n f o
Article history: Received 6 June 2014 Received in revised form 4 August 2014 Accepted 5 August 2014 Available online xxxx Keywords: Methanation Methane enrichment Power to gas Hydrogen Anaerobic digestion
a b s t r a c t A new type of anaerobic trickle-bed reactor was used for biocatalytic methanation of hydrogen and carbon dioxide under mesophilic temperatures and ambient pressure in a continuous process. The conversion of gaseous substrates through immobilized hydrogenotrophic methanogenic archaea in a biofilm is a unique feature of this type of reactor. Due to the formation of a three-phase system on the carrier surface and operation as a plug flow reactor without gas recirculation, a complete reaction could be observed. With a methane concentration higher than cCH4 ¼ 98%, the product gas exhibits a very high quality. A specific methane production of PCH4 ¼ 1:49 Nm3 =ðm3SV dÞ was achieved at a hydraulic loading rate of LRH2 ¼ 6:0 Nm3 =ðm3SV dÞ. The relation between trickle flow through the reactor and productivity could be shown. An application for methane enrichment in combination with biogas facilities as a source of carbon dioxide has also been positively proven. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction The development of renewable energies, especially in the wind and solar energy sectors, is essential in order to obtain a shift away from reliance on fossil fuels. Great strides have already been made in Germany (DLR, 2012). According to predictions made by the German Federal Environment Ministry in the year 2011, the portion of energy coming from wind and solar power will more than triple by 2050. The federal government has set the goal of raising the portion of energy from renewable sources for gross electricity consumption to 35% by 2020 and 80% by 2050. Notably, the development of offshore wind performance will be accelerated to 25 GW by 2030 and the network infrastructure (north–south route) will be extended. A full use of this potential is already pushing the threshold since energy use and temporary buffer storage are both ⇑ Corresponding author. Tel.: +49 355694328. E-mail address: [email protected] (M. Burkhardt).
limited in times of high energy production. Due to inadequate development of the electricity network, energy input and distribution are limited. Wind and solar units that have been installed need to be connected to the network. An interesting solution appears to be the power-to-gas strategy for the storage and subsequent repowering of energy. Alongside thermal (hot water cylinder), electrical (superconductor) and mechanical (flywheel) storage, electrochemical storage has proven itself to be particularly beneficial. An alternative to storage in an accumulator is the possibility of conversion to hydrogen. The gaseous energy source can be won through the electrolysis of water. However, until now only minimal utilizations and technological thresholds have been available for storage and use. Possible solutions include direct H2 power reconversion, use in automobiles run on H2, input of H2 into the natural gas grid or the generation of CH4 through the Sabatier process. But, these applications are only developed to a small degree. Consequently, their application is limited and they are accompanied by energy loss, cable and pipe-
http://dx.doi.org/10.1016/j.biortech.2014.08.023 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Burkhardt, M., et al. Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.08.023
2
M. Burkhardt et al. / Bioresource Technology xxx (2014) xxx–xxx
line damage, and significant costs, causing them to not yet be economically advantageous. An alternative is biologically catalyzed methanation, which can take place at more moderate temperatures than physico-chemical processes and has a higher resistance to contaminations, such as H2S, organic acids and NH4. In addition to hydrogen, carbon dioxide is needed as a reaction partner in methanation. Sources can be CO2-rich exhaust gas or biogas, which usually contains 30% to 50% carbon dioxide. In these systems the gases are typically fed into a liquid tensed fermenter (Krajete, 2012). But, bubbles often accumulate and ascend in the reactor. The input gases are lead out of the reactor within minutes according to the rate of bubble ascension. Consequently, the contact time, or reaction time, is not sufficient and the achievement potential of the system is limited. A methane productivity of P CH4 ¼ 0:11 Nm3 =ðm3 dÞ could be achieved so far, which is very low. A higher distribution of the bubbling gases in the liquid was possible by using a hollow fiber membrane in a continuously stirred tank reactor (CSTR), which resulted in a very high methane concentration of cCH4 ¼ 99% (Wang et al., 2013). The productivity of the simultaneous biomethanation of sewage sludge and H2-rich coke oven gas could be increased from PCH4 ¼ 0:28 m3 =ðm3reactor dÞ up to P CH4 ¼ 0:65 m3 =ðm3reactor dÞ, which is a very interesting application. Thermophilic methanation has also been conducted in a fixed bed reactor by using pure culture Methanobacterium thermoautotrophicum (Jee et al., 1988). The cost of obtaining the very high productivity of PCH4 ¼ 5:3—10:2 m3 =ðm3 hÞ was an incomplete conversion and low methane concentrations of cCH4 ¼ 15:5—87% in the reactor output. In an attempt to increase the methane concentration of the product, a further experiment passed the inflowing gas through a liquid tensed fixed bed reactor (Bugante et al., 1989). The gas was recirculated with an optimal recirculation speed of Qg = 40 l/h. Nonetheless, while running as a batch process, the final productivity of PCH4 ¼ 0:025 m3 =ðm3reactor dÞ was very low due to the necessity of recirculation. Alternatively, an anaerobic trickle bed reactor was investigated for the methanation of gaseous substrates for the recovery of methane (Burkhardt and Busch, 2013). While running in batch mode and recirculating the product gas, the biocatalyzed methanation of carbon dioxide and hydrogen demonstrated a higher productivity of P CH4 ¼ 1:17 m3 =ðm3reactor dÞ. By using a fermenter equipped with an impeller and a porous Teflon or porous glass, a very high productivity of PCH4 ¼ 288 m3 =ðm3reactor dÞ with cCH4 ¼ 96% could be observed. The result could be reached through a high energy input by the impeller, with it running at 1200 rpm (Peillex et al., 1990).
2. Methods The trickle-bed reactor varies from all of the other known reactors for biogas production, such as impounded or classic fixed-bed reactors, in that the reaction chamber is not filled with liquid. It contains packed bed as a surface on which microorganisms can immobilize. They are surrounded on all sides by a gas phase and are merely sprinkled with a limited amount of liquid. In this way, a three phase system exists (biofilm–liquid phase–gas phase) (Decressin, 1998). Material transport in the biofilm is significantly improved as a result of the higher concentration gradient as a driving force for the mass transfer and short diffusion paths. The thickness of the liquid phase has a decisive influence on the concentration gradient of the gases released and transported to the biofilm. This advantage increases the methanation and final productivity of the whole system. The concentration gradient is a function of the diffusion coefficient, film thickness d and material conversion rate. One needs to be aware that the concentration of hydrogen and carbon dioxide in the water film and biofilm differ
from one another due to varying solubility and diffusion coefficients. As the case may be, these basic conditions may lead to limiting effects. Continuous reactor operation took place in a V = 88 l reactor with the proportions of length l:diameter d = 5:1. The reactor was gas-tight, filled up to a volume of VSV = 61 l with packing material and heated up to 37 ± 0.5 °C with a double jacket and a 20 l heat bath (LAUDA). The packing material, Bioflow 40 (RAUSCHERT), offers a high specific surface area of a = 305 m2/m3. The inoculum was not an isolated strain of methanogenic archaea. The aim was to investigate a complex population diversity fixed in a biofilm. To reach that variety of methanogens, anaerobic sludge of a local sewage plant was used as the inoculum. Therefore, archaeal species like Methanococcus spp., Methanobacterius spp., Methanomicrobium and also Methanosaeta spp. could be confirmed. They were immobilized in a biofilm on the carrier material. To ensure a moist environment and the provision of nutrients and trace elements for the microorganisms, a processing liquid was circulated in the reactor. Recirculation took place constantly at Q RV ¼ 6:15 m3 =ðm3sv dÞ. The processing liquid was held available in the floor of the reactor in the reactor sump. Nutrient and trace element concentrations were set to Al 16.95 mg/l, B 3.63 mg/l, Ca 48.7 mg/l, Co 0.5 mg/l, Cu 0.3 mg/l, Fe 21.6 mg/l, K 303.6 mg/l, Mg 28.5 mg/l, Mn 0.57 mg/l, Mo 0.27 mg/l, Na 132.2 mg/l and Ni 3.96 mg/l. The pH value ranged from 7.2 to 7.4. The feeding in of hydrogen and carbon dioxide gases took place continuously underneath the packing bed through a H2-generator via electrolysis and conventional CO2-pressure cylinders. According to Eq. (1), in doing so the stoichiometric relationship between hydrogen and carbon dioxide needed to remain at 4:1. Mass flow controllers (BROOKS INSTRUMENT) were installed to proportion the substrate feeding, which meant that the rate of flow and load could be varied. The gaseous substrates flowed through the reactor from bottom to top and were converted by methanogenic microorganisms. In the process a plug flow was formed. The calculated average retention time was reduced to a value of s = 4 h. The reactor product, methane, was lead out through the head of the reactor. Determination of the volume took place in the drum gas counter TG05 (RITTER) with storage in a gas bag (TESSERAUX). Determination of the concentrations of methane, hydrogen, carbon dioxide and oxygen took place nearly continuously by way of the SSM6000 LT (PRONOVA). At the same time the gas temperature T CH4 and the gas pressure P CH4 were recorded. Methane production for various loading rates, LRH2 ¼ 0:4—6:0 Nm3 =ðm3SV dÞ, was investigated. The second reaction product, water, was collected and taken away by way of the reactor sump. To avoid losses of nutrients and trace elements, the nutrient solution was added again to keep the required concentration constant. In further observations the flow rate was varied between Q RV ¼ 3:35 and 13:5 m3 =ðm3sv dÞ to investigate the influence of a liquid layer on mass transfer into the biofilm and final methane production. By changing pure CO2 to biogas, mainly consisting of CO2 (23 ± 3 vol%) and CH4 (78 ± 2 vol%), biogas could also be observed as a carbon source and suitable for biological methane enrichment. The Biogas was produced continuously and pumped in stoichiometric relation to H2 flow into the trickle bed reactor. Gas purification and dewatering was not done.
4H2 þ CO2 ! CH4 þ 2H2 O DHoR ¼ 167 KJ=mol
ð1Þ
Taking into consideration the trickle-bed volume VSV, Eq. (1) can be used to determine and depict the theoretical methane production (Eq. (2)). The methane productivity P was determined from the experimental methane volumetric flow under standard conditions and the methane concentration (Eq. (3)). P was used to assess the reactor performance based on the investigated parameters.
Please cite this article in press as: Burkhardt, M., et al. Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.08.023
3
M. Burkhardt et al. / Bioresource Technology xxx (2014) xxx–xxx nH
Ptheo ¼
Q H2 nCH2
4
V SV
"
Nm3CH4
#
m3SV d
amount of CO2 being added, a full conversion could again be attained. Fig. 2 illustrates the progressive increase in the loading rate LR over the investigated period. The H2 loading rate was successively raised to a value of LRH2 ¼ 4:83 Nm3 =ðm3SV dÞ. The experimental methane production P CH4 correlates to the theoretical value Ptheo for nearly all of the loading rates. Even with an H2 loading rate of LRH2 ¼ 4:83 Nm3 =ðm3SV dÞ, a methane productivity of PCH4 ¼ 1:2 Nm34 =ðm3SV dÞ could be achieved. Up to this H2 load the limit of the trickle-bed reactors achievement potential could not be determined. The degree of H2 degradation was nearly always near to gH2 ¼ 100%. Beside methane production, the substrate conversion at a loading rate of LRH2 ¼ 4:02 Nm3 =ðm3SV dÞ results in a minor water yield of 1:6 Nlðm3SV dÞ. By using Eq. (1) the theoretical water production according to the ideal molar ratio is calculated with V H2 O;theo ¼ 1:6 Nl=ðm3SV dÞ and is similar, obviously. The results of our experiment in a bioreactor have not been seen by other workers at similar scales (Jee et al., 1988). Compared to Burkhardt and Busch and Bugante et al., a continuous process was possible in our experiment and no gas recirculation was required. The formation of a plug flow in the fixed bed allowed raised loading rates in a continuous process. As demonstrated, the trickle-bed process has proven itself to have a 13 times higher achievement potential than other known biocatalytic processes (Krajete, 2012). One reason for this is the higher retention time that allows for a longer reaction time, which was sought after in
ð2Þ
" # 3 Q product cCH4 NmCH4 3 P¼ V SV mSV d
ð3Þ
The procedure described is already under review to receive a patent (Burkhardt, 2013).
3. Results and discussion At the beginning of the experiment air was expelled from the reactor so that a constant atmosphere was reached by the 20th day and analysis could begin. Neither oxygen nor nitrogen could be detected any longer (Fig. 1). At this point, a methane concentration of up to cCH4 ¼ 98% could be ascertained from the discharge of the bioreactor. The feed materials, CO2 and H2, were almost completely converted. The notable fluctuations can be attributed to small deviations in the stoichiometric relationship of the gases to each other, which needed an adjustment after an increase in the loading rate. Thereby it was determined that a H2–CO2 relationship of 3.76:1 is needed for a complete conversion. This excess of CO2 can most likely be explained by an increased need for C through the biomass build-up. The recognizable demodification of the 100 90 80
ci [%]
70 60 50 40 30 20 10 0 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 carbon dioxide hydrogen
oxygen methane
time t [d]
Fig. 1. Biogas composition during continuous operation of the anaerobic trickle-bed reactor.
8
1,4
7
1,2 1,0
5 0,8 4 0,6 3 0,4
2
P [Nm³CH4 /(m³SV⋅d)]
LRH2 [Nm 3H2/(m3SV⋅d)]
6
0,2
1 0 20
30
40
50
60
70
80
90
100
0,0 110 time t [d]
loading rate LR
methane productivity P
theoretical methane productivity Ptheo
Fig. 2. H2 loading rate LR and the development of the experimental methane productivity P compared to the theoretical methane productivity Ptheo.
Please cite this article in press as: Burkhardt, M., et al. Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.08.023
4
M. Burkhardt et al. / Bioresource Technology xxx (2014) xxx–xxx
this case. Furthermore, due to the development of the three phase system in the trickle-bed reactor, there is an increased amount of substrate transport in the biofilm, which is a result of raised concentration gradients on the concentration boundary layer Lc. Processes merely using membrane techniques and fast rotating stirrers allow for higher productivities than in our current research (Peillex et al., 1990), but they are limited to very small reactor volumes due to the very high energy consumption and technological boundaries of using impellers at larger scales. The input of substrate was suspended from the 86th to the 89th day of the experiment ðLRH2 ¼ 0 Nm3 =ðm3SV dÞÞ. After starting operations again on the 89th day (LRH2 ¼ 4:02 Nm3 =ðm3SV dÞ), the same achievement potential as before the break in operations could be obtained. The methane concentration of the product gas was again at cCH4 > 90%. This proved the ability to conduct flexible reactor operations. The methane production is considered very controllable and flexible since the achievement potential was immediately obtainable again after a suspension of substrate input. Therefore, reactor operation can be carried out independently from substrate availability. It is possible to react on short notice to fluctuations of need and availability of product gases as well as input substrate. By changing the liquid recirculation rate QRV between 3.3 and 9:2 m3 =ðm3SV dÞ the influence of liquid phase thickness on the mass transfer into the biofilm was investigated. The entire liquid distribution in the trickle bed must be maintained. A correlation of the increasing methane concentration (and increasing hydrogen conversion) by a decreasing liquid recirculation rate was observed. Therefore higher H2 loading rates of LRH2 ¼ 6:00 Nm3 =ðm3SV dÞ could be run at lower liquid recirculation rates QRV. It results in a higher methane productivity of PCH4 ¼ 1:49 Nm3 =ðm3SV dÞ. Instead of feeding pure carbon dioxide, biogas can be used as a CO2 source. The Biogas with a methane concentration of 77% was produced in a double stage process which is a typical average. Due to passing it into the trickle bed reactor and an additional H2 flow (LRH2 ¼ 0:51 Nm3 =ðm3SV dÞ) according to the essential stoichiometrics, a methane enrichment up to cCH4 ¼ 96% could be observed. The amount of CO2 was reduced nearly completely. The methane production increased by 23%. Further experiments of increasing LRH2 will show the limits of the reactor performance. The process thus suits itself very well for application in the power-to-gas strategy since a storable and usable methane gas can be produced after recovering electrolyzed hydrogen. Considering that the methane is nearly the quality of natural gas, it is possible to feed it into the natural gas grid with only minimal filtration necessary. Alternatively, it constitutes a sought-after, basic material for synthesis. Using biogas as a source of carbon dioxide and maintaining consideration of the appropriate hydrogen input, methane enrichment can take place in a reactor. The methane pro-
duction can be increased by 23%. Traces of H2S, NH4 or water vapor are unproblematic for the process, although if necessary they can be removed before the product is put into storage. Consequently, the trickle-bed reactor is applicable as downstream methane enrichment step in which no by-products accumulate. 4. Conclusion By means of the anaerobic trickle-bed process, hydrogen and carbon dioxide were converted biocatalytically to methane. A concentration of up to 98% could be achieved. A methane productivity of P CH4 ¼ 1:49 Nm3 =ðm3SV dÞ for a H2 loading rate of LRH2 ¼ 6:00 Nm3 =ðm3SV dÞ and a liquid recirculation rate of Q RV ¼ 3:1 m3 =ðm3SV dÞ could be established. While running the process continuously, there was no necessity for gas recirculation or other gas mixing. The present technology can be used for coupling of various renewable energy sources, to convert electrical into chemical bound energy and to store methane into the natural gas grid. Acknowledgements The authors wish to thank Mr. U. Klopsch, Mr. R. Kabelitz, Mr. T. Koschack and Mr. E. Sueß for their help with analytical and technical issues. References Bugante, E.C., Shimomura, Y., Tanaka, T., Taniguchi, M., Oi, A., 1989. Methane production from hydrogen and carbon dioxide and monoxide in a column bioreactor of thermophilic methanogens by gas recirculation. J. Ferment. Bioeng. 67 (6), 419–442. Burkhardt, M., Busch, G., 2013. Methanation of hydrogen and carbon dioxide. Appl. Energy 111, 74–79. Burkhardt, M., Busch, G., Großmann, J., 2013. DE 102013209734.4, Verfahren und Vorrichtung für die Methanisierung von Gasen mittels Rieselbettreaktoren. Patent. Decressin, O., 1998. Biofilter-Modellbildung, Verifikation und Simulation. FITVerlag, Paderborn. DLR, 2012. . Langfristszenarien und Strategien für den Ausbau erneuerbarer Energien in Deutschland bei Berücksichtigung der Entwicklung in Europa und global. Jee, H., Nishio, N., Ngai, S., 1988. Continuous CH4 production from H2 and CO2 by Methanobacterium thermoautotrophicum in a fixed bed reactor. J. Ferment. Technol. 66 (2), 235–238. Krajete, A., 2012. WO 2012/110256 A1, 2012. Method of converting carbon dioxide and hydrogen to methane by microorganisms, Patent. Peillex, J.P., Fardeau, M.L., Belaich, J.P., 1990. Growth of Methanobacterium thermoautotrophicum on H2–CO2: high CH4 productivities in continuous culture. Biomass 21 (4), 315–321. Wang, W., Xie, L., Luo, G., Zhou, Q., Angelidaki, I., 2013. Performance and microbial community analysis of the anaerobic reactor with coke oven gas biomethanation and in situ biogas upgrading. Bioresour. Technol. 146, 234–239.
Please cite this article in press as: Burkhardt, M., et al. Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.08.023
