Elsevier

Applied Energy

Volume 202, 15 September 2017, Pages 238-247
Applied Energy

Biological methanation of CO2 in a novel biofilm plug-flow reactor: A high rate and low parasitic energy process

https://doi.org/10.1016/j.apenergy.2017.05.134Get rights and content

Highlights

  • A novel bio-methanation reactor was designed and evaluated.

  • A biofilm consisting of mixed anaerobic consortia served as the biocatalyst.

  • High rate methanogenesis was observed without gas-liquid agitation.

  • Gas conversion was successfully de-coupled from energy consumption.

Abstract

The performance of a novel biofilm plug flow reactor containing a mixed anaerobic microbial culture was investigated for the conversion of CO2/H2 to CH4. Unlike conventional gas-liquid contactors that depend on agitation, gas diffusion was decoupled from power consumption for mixing by increasing the gas phase inside the reaction space whilst increasing the gas residence time. The mixed mesophilic culture exhibited good biofilm formation and metabolic activity. Within 82 days of operation, 99% and 90% CH4 conversion efficiencies were achieved at total gas throughputs of 100 and 150 v/v/d, respectively. At a gas input rate of 230 v/v/d, methane evolution rates reached 40 v/v/d, which are the highest to date achieved by fixed film biomethanation systems. Significant gas transfer related parasitic energy savings can be achieved when using the novel plug flow design as compared to a CSTR. The results and modelling parameters of the study can aid the development of high rate, low parasitic energy biological methanation technologies for biogas upgrading and renewable power conversion and storage systems. The study has also established a reactor system which has the potential of accelerating biotechnology developments and deployment of other novel C1 gas routes to low carbon products.

Introduction

Hydrogenotrophic methanogenesis has recently gained significant attention due to its potential as a CO2 utilization process [1], [2], [3]. The low temperature and pressure operating conditions required as well as the inexpensiveness of the microbial catalyst make the biological synthesis of CH4 from CO2 an attractive alternative to conventional biogas upgrading methods [4], by increasing both quantity and quality of the end product. Furthermore, the conversion of excess renewable electricity to methane (Power to Methane, PtM) has the potential to resolve many of the current inadequacies of existing power back up technologies, such as storage capacity, cost and geographical limitations [5] and support renewable energy developments in terms of minimizing power curtailment and facilitating wider deployment of renewable energy generation in regions with limited electricity grid availability. Additionally, in comparison to H2, green CH4 can be readily added to the gas grid [1] since it does not alter the characteristics of the existing fuel.

Methanogens have been typically cultivated in homogenous aqueous solutions due to the ease of control over the entire population. Because of this, continuously stirred tank reactors (CSTRs) have over the years become the standard in anaerobic processes such as the digestion of organic waste as, if designed and operated appropriately, they provide even distribution of temperature and soluble and insoluble constituents as well as high levels of consistency in sampling. The anaerobic digestion of organic feedstock relies considerably on the acetoclastic methanogenic route [6], [7], [8], which is the terminal stage in a series of solid/liquid substrate conversions. However, in the case of ex-situ hydrogenotrophic methanogenesis the very presence of the aqueous solution hinders gas transferability. Hydrogen is a gas with a particularly low solubility coefficient (1.35 × 10−5 v/v in water, 35 °C), meaning that its rate of diffusion into the liquid media is considered to be the main limiting factor for the process [9].

A generally accepted model describing the rate of diffusion of a single gas in a liquid is given by Eq. (1):dq/dt=kLα(Cg-Cl)where (kL) is the linear mass transfer coefficient, (α) is the specific surface area of contact between the gas and the liquid and Cg-Cl is the concentration difference between the gas phase and the liquid phase.

The kLα factor depends on a large number of parameters, some of which cannot be altered (e.g. the molecular weight of the gas), or are difficult to control, especially in biological reactors, (e.g. the rheological behaviour of the liquid). Theoretical models that predict diffusivity gradients in stirred tank bioreactors [10], [11], [12] rely on simplified versions of the actual systems (single gas - single liquid matrix). Hydrogenotrophic methanogenesis however, creates a much more complex multi-component system (gas – liquid – solid matrices) with continuous changes in all three phases due to the biological reactions involved. The process designer is therefore typically left with just three controllable options; pressure, gas hold-up and gas-liquid surface area of contact.

Higher pressure generally results in better gas diffusion since it increases the concentration gradient between the gas phase and the liquid phase. Furthermore, there is evidence that several hydrogenotrophic species are not only resistant to high pressure conditions (>100 atm) but also exhibit improved growth and methanogenesis rates [13], [14].

Gas hold-up and interfacial area of contact are interlinked and in liquid flooded reactor configurations are maximized with the reduction of bubble size, which is typically achieved through intensive mixing. CSTRs have been shown to be efficient at high angular velocities (>1200 rpm) [15], [16]. However, intensive mixing also has a profound negative effect on the energy balance of the system. Power dissipation is directly linked to the gas-liquid mass transfer in CSTRs. With increasing power dissipation the bubble diameter decreases which, in turn, increases the interfacial surface area [11], [17]. Bubbles break apart because the surface tension forces are overcome by a higher power density. Therefore CSTR systems depend on high impeller rotational speeds in order to increase gas diffusivity.

Due to the turbulent regime in stirred gas-liquid contactors the power drawn by the impeller is typically given by Eq. (2):P=PoρN3D5where (Po) is the power number and depends on the structural characteristics of the mixing system like the geometry of the impeller and the vessel, (ρ) is the density of the fluid, (N) is the angular velocity of the impeller and (D) is the diameter of the impeller. From Eq. (2) it is evident that power consumption increases exponentially with increasing agitation rates [18].

The microbial cultures that are used as catalysts in biomethanation may also be influenced by the shear forces created by the impeller. Although there is no reference of inhibition due to shear in pure culture biomethanation systems, cell damage cannot be excluded. In mixed culture systems there is also the added possibility of reduced cell to cell interaction. For example, syntrophic relationships between different groups have been found to be hindered by shear in CSTRs treating animal manure [19]. Experiments with lab-scale CSTRs treating sewage sludge also indicate that there is a mixing speed threshold above which biogas production declines significantly [20].

The major inefficiency in the energy balance of any biomethanation reactor that is flooded with an aqueous solution occurs because most of the working volume is occupied by water which acts as a barrier between the microbes and their gaseous feed. If the feeding gas could be in contact with the culture without the need for intense mixing, then energy consumption would not necessarily be linked to the gas-liquid mass transfer factor, therefore delivering a far more energy efficient process. The ability of mixed bacterial and methanogenic cultures to form colonies attached to various materials could be utilised to achieve such a system. The microbes would still need to be wetted and in contact with essential nutrients in solution but this could be achieved by reducing the volume of the solution that surrounds the biomass to a minimum level compared with flooded systems.

Eq. (1) is a derivative of Fick’s first law of diffusion which states that the molar flux of a species (i) is proportional to its concentration gradient across the interface between the gas phase and the liquid phase (two film model) as shown by Eq. (3):ji=-Diciwhere Di is the diffusion coefficient and Ci is the concentration of the species

For transfer along a vector z,ci=dcidzwhich means that the rate of transfer is inversely proportional to the separation distance of the two media. A thin liquid layer means that diffused gases will get converted closer to the interface, thus increasing the concentration gradient and therefore diffusivity.

Novel gas delivery systems that rely on the formation of biofilm after inoculation with mixed methanogenic cultures have been used as candidates to increase biomass while reducing the energy consumption of biomethanation. Fixed bed [21], [22], trickling bed [23], [24] and hollow fibre membrane reactors [25] have been studied in the past with high gas conversion efficiencies (>98% of CH4 in the effluent gas). However, the volume of gas per working volume of reactor (v/v) conversion throughputs of these systems have been reported to be at significantly lower levels (≈1–6 v/v/d) to those of intense mixing systems (≈18–28 v/v/d) that used similar mixed methanogenic inocula [26], [27]. In the case of flooded fixed film reactors (fixed bed and hollow fibre reactors), the disparity in the conversion rates can be explained by the fact that the attached microbes were submerged in a liquid media and the presence of attachment media prevented thorough gas-liquid mixing leading to low gas dilution rates and localized conversion. In the case of the trickling bed reactors the low conversion rates could be linked to the geometry of the rector vessels which did not allow for enough contact time between the biofilm and the gas phase.

In the last decade, the upsurge in the number of studies on ex-situ biomethanation is a result of an increased interest in PtG technologies for the valorisation of CO2 and as an aid to a more sustainable deployment of intermittent renewable energy sources [28], [29]. However, there is presently a gap in literature regarding the parasitic energy losses that accompany the process, especially in systems that rely on agitation in order to increase gas-to-liquid mass transfer. In a process that already suffers from the energy losses that accompany the conversion of electricity to hydrogen, steps need to be taken in order for any further energy conversion losses to be minimized. It is envisaged that for such minimization to occur, biomethanation efficiency needs to be de-coupled from agitation through a more efficient way of biofilm utilization.

In biofilm reactors where the availability of the gas phase and the liquid phase are increased and reduced respectively, it is suggested that the geometry of the reactor should achieve two things: (i) reduce the thickness of the liquid layer around the biofilm to the bare essential minimum, and (ii) maximize the gas residence time by increasing the distance the gas molecules need to travel before they lose contact with the biofilm. In the present study, the above hypothesis was tested with the use of a mixed anaerobic culture. A novel reactor consisting of a single tube filled with microbial attachment media was constructed and evaluated for its ability to convert H2 and CO2 input gases to CH4 with lower energy inputs than for CSTRs or other liquid media flooded reactor designs.

Section snippets

Inoculum

The inoculum used was anaerobically digested sewage sludge collected from Cog Moors Wastewater treatment Plant in Cardiff, South Wales. Prior to use the sludge was filtered through a 125 μm stainless steel sieve and then left to settle for 48 h at room temperature. The reactor was then filled with approximately 0.75 L of the supernatant at a concentration of 4.6 g/L TS, 3.7 g/L VS.

Nutrient solution

A nutrient solution was created from the same batch of digested sewage sludge as the inoculum, centrifuged and filtered

First stage of operation – Vertical arrangement

In the first weeks it became clear that the two phases (i.e. up-flow/down-flow) exhibited totally different physical/biological characteristics. In the up-flow (flooded) compartments the feeding gas was observed to be rising rapidly as fairly large bubbles (cm range) whereas in the trickling compartments it remained in contact with the whole volume of microbes for longer periods. This was evident by the difference in biofilm formation between the two compartments as shown in Fig. 2a and b.

After

Theoretical process power requirements

With regards to energy input requirements, the microbial factor renders a direct comparison between different biomethanation reactor designs impractical. This is because the conditions applied to the culture (method and intensity of agitation, the presence or not, as well as the type of microbial attachment media) have a direct impact on the culture itself and thus its metabolic activity. Therefore, the biological, physical and chemical parameters responsible for this energy conversion process

Conclusions

The de-coupling of energy input from gas-to-liquid mass transfer was demonstrated in a prototype biofilm plug flow biomethanation reactor. The study showed that it is possible to obtain high biomethanation conversion rates and efficiencies by changing the way a mixed microbial culture is utilized, with the specific aim of reducing the liquid volume in the reactor while increasing the gas residence time. The novelty of the present design (in horizontal mode) relies on the adhesive properties of

Acknowledgements

This research was supported by the University of South Wales, UK, through the award of a Centenary Postgraduate Scholarship. The authors also acknowledge the European Regional Development Funding (ERDF) support provided by the Welsh Government A4B scheme for the Knowledge Transfer Centre for Advanced Anaerobic Processes and Biogas Systems (Project Ref: HE 14 15 1009).

References (47)

  • M. Martín et al.

    Bubbling process in stirred tank reactors I: Agitator effect on bubble size, formation and rising

    Chem Eng Sci

    (2008)
  • A. Alitalo et al.

    Biocatalytic methanation of hydrogen and carbon dioxide in a fixed bed bioreactor

    Biores Technol

    (2015)
  • M. Burkhardt et al.

    Biocatalytic methanation of hydrogen and carbon dioxide in an anaerobic three-phase system

    Biores Technol

    (2015)
  • D.-H. Ju et al.

    Effects of pH conditions on the biological conversion of carbon dioxide to methane in a hollow-fiber membrane biofilm reactor (Hf–MBfR)

    Desalination

    (2008)
  • X. Zhang et al.

    Life cycle assessment of power-to-gas: approaches, system variations and their environmental implications

    Appl Energy

    (2017)
  • J. Cruwys et al.

    Development of a static headspace gas chromatographic procedure for the routine analysis of volatile fatty acids in wastewaters

    J Chromatogr A

    (2002)
  • R. Conrad et al.

    Anaerobic conversion of carbon dioxide to methane, acetate and propionate on washed rice roots

    FEMS Microbiol Ecol

    (1999)
  • S. Savvas et al.

    Closed nutrient recycling via microbial catabolism in an eco-engineered self regenerating mixed anaerobic microbiome for hydrogenotrophic methanogenesis

    Biores Technol

    (2017)
  • R. Ye et al.

    Homoacetogenesis: a potentially underappreciated carbon pathway in peatlands

    Soil Biol Biochem

    (2014)
  • J. Wang et al.

    Trophic link between syntrophic acetogens and homoacetogens during the anaerobic acidogenic fermentation of sewage sludge

    Biochem Eng J

    (2013)
  • O.R. Kotsyurbenko et al.

    Competition between homoacetogenic bacteria and methanogenic archaea for hydrogen at low temperature

    FEMS Microbiol Ecol

    (2001)
  • V. Siriwongrungson et al.

    Homoacetogenesis as the alternative pathway for H2 sink during thermophilic anaerobic degradation of butyrate under suppressed methanogenesis

    Water Res

    (2007)
  • A.K. Haydock et al.

    Continuous culture of Methanococcus maripaludis under defined nutrient conditions

    FEMS Microbiol Lett

    (2004)
  • Cited by (76)

    View all citing articles on Scopus
    View full text