Savvas, S, Donnelly, J, Patterson, T, Dinsdale, R & Esteves, S 2016, 'Closed nutrient recycling via microbial catabolism in an eco-engineered self regenerating mixed anaerobic microbiome for hydrogenotrophic methanogenesis' Bioresource Technology, vol 227, pp. 93-101. DOI: 10.1016/j.biortech.2016. 12.052

This is an Accepted Manuscript of an article published by Elsevier in Bioresource Technology on 18/12/2016, available online: http://dx.doi.org/10.1016/j.biortech.2016.12.052

© 2016. This accepted manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/

1

Closed Nutrient Recycling via Microbial Catabolism in an Eco-Engineered Self

Regenerating Mixed Anaerobic Microbiome for Hydrogenotrophic Methanogenesis

Savvas Savvasa,b*, Joanne Donnellya,b, Tim P. Pattersona,b, Richard Dinsdaleb and Sandra. R.

Estevesa,b

aWales Centre of Excellence for Anaerobic Digestion, bSustainable Environment Research

Centre, Faculty of Computing, Engineering and Science, University of South Wales,

Pontypridd CF37 1DL, Wales UK.1

A novel eco-engineered mixed anaerobic culture was successfully demonstrated for the first

time to be capable of continuous regeneration in nutrient limiting conditions. Microbial

catabolism has been found to support a closed system of nutrients able to enrich a culture of

lithotrophic methanogens and provide microbial cell recycling. After enrichment, the

hydrogenotrophic species was the dominating methanogens while a bacterial substratum

was responsible for the redistribution of nutrients. q-PCR results indicated that 7% of the

total population was responsible for the direct conversion of the gases. The efficiency of

H2/CO2 conversion to CH4 reached 100% at a gassing rate of above 60 v/v/d. The pH of the

1 Corresponding author: Savvas Savvas (savvas.savvas@southwales.ac.uk)

2

culture media was effectively sustained at optimal levels (pH 7-8) through a buffering

system created by the dissolved CO2. The novel approach can reduce the process

nutrient/metal requirement and enhance the environmental and financial performance of

hydrogenotrophic methanogenesis for renewable energy storage.

Keywords: methanation; biocatalyst; nutrient recycling; CO2;

Highlights

A methanation biocatalyst was observed to be capable of self-regeneration

Self-regeneration was accomplished by creating a closed nutrient ecosystem

pH of the media was controlled solely by the amount of CO2 entering the reactor

1. Introduction

Despite an increase in the number of mitigation policies, global anthropogenic CO2

emissions are rising (Pachauri et al., 2014). At the same time, due to the recent green

intergovernmental agenda, by 2035 renewable energy sources are expected to be generating

more than 25% of the world’s electricity (Global Wind Energy Council (GWEC), 2014).

However, the fluctuating manner of energy production from these sources calls for

solutions which are capable of equalizing energy production to energy demand. When

power generation exceeds network capacity, curtailment (i.e. reducing renewable electricity

addition into the network) is often the only option. By 2030 due to a significant increase in

wind penetration, curtailment in northern Europe is expected to reach 9.3 TWh (Lew et al.,

2013). In the UK alone, due to the planned increase in onshore and offshore wind capacity,

3

curtailment could reach 2.8 TWh/a by 2020 and 50-100 TWh/a by 2050 (Qadrdan et al.,

2015).

The environmental and economic losses from this practice could be avoided if renewable

energy generation was coupled with the appropriate grid scale energy storage systems.

However, the flexibility and capacity of current energy storage technologies have major

limitations. Batteries need an increase in life cycle and depth of discharge accompanied

with a reduction in production costs (Barnhart et al., 2013). Super capacitors,

superconducting coils and flywheels have very short discharge periods, which makes them

suitable only as emergency UPS units (Gonzalez et al., 2004). Pumped hydro and

compressed air storage are limited by geographical factors (Denholm et al., 2010).

A novel way of tackling renewable energy storage is presented by the Power-to-Methane

(PtM) concept (Götz et al., 2016). This entails the conversion of renewable electricity into

H2 via water electrolysis, followed by the hydrogenation of CO2 to CH4, according to

Equation 1.

4H2 + CO2 → CH4 + 2H2O (ΔGo = -131 kJ/mol) (1)

The storage of renewable electricity as methane not only offers a way of balancing

generation with demand but could also be a means of increasing the supply of low carbon

methane to gas networks. The high capacity natural gas infrastructure of many countries

offers a supreme opportunity for the expansion of on-shore and off-shore renewable energy

generation with much higher energy return on investment (EROI) ratios due to the

reduction of curtailment. Additionally, areas which currently experience restricted

electricity grid availability can benefit from the PtM technology as it can offer an

4

alternative to the commissioning of new electricity transmission infrastructure. Renewable

electricity could then be converted to heat or to a transport fuel.

Hydrogenotrophic methanogens belong to the domain of Archaea and obtain their energy

by synthesizing CH4 from H2 and CO2. These have typically been an integral part of the

diverse microbial community within anaerobic digesters used for the conversion of

complex organic matter to CH4. However, they can be separately cultivated, thus creating a

single culture system. Such systems have been shown to successfully produce CH4 by using

CO2 as their sole carbon source and H2 as the electron donor, with efficiencies close to

100% (Burkhardt et al., 2015; Lee et al., 2012). Presently, such biological methanation

(biomethanation) has been primarily investigated at laboratory scale (Ako et al., 2008;

Burkhardt et al., 2015; Lee et al., 2012; Luo and Angelidaki, 2012; Martin et al., 2013;

Peillex et al., 1990; Rittmann et al., 2012; Schill et al., 1996; Seifert et al., 2014; Zhang and

Maekawa, 1993) with a small number of pilot projects, mainly in Germany, Denmark and

Austria (DENA, 2015; Götz et al., 2016).

Most knowledge regarding the biochemistry of methane formation from hydrogenotrophic

archaea has been derived by the study of two thermophilic strains, Methanothermobacter

thermautotrophicus and Methanothermobacter marburgensis. Due to their high doubling

times (< 5h and < 2h respectively), they were the first hydrogenotrophic methanogens to be

isolated and cultivated in high enough concentrations that allowed the purification of

enzymes/coenzymes involved in the reduction of CO2 to CH4 (Kaster et al., 2011). They

have fully decoded sequences, and have been the preferred options in pure culture

methanogenic reactors (Martin et al., 2013; Peillex et al., 1990; Rittmann et al., 2012; Schill

et al., 1996; Seifert et al., 2014).

5

Mixed thermophilic and mesophilic cultures have also been sourced from digested sewage

sludge or manure (Ako et al., 2008; Burkhardt et al., 2015; Lee et al., 2012; Zhang and

Maekawa, 1993). In these instances, enrichment of lithotrophic strains takes place by

continuous supply of CO2 and H2 while depriving the culture of an organic load, of solid or

liquid substrate. Mixed cultures have the advantage of being inexpensive and robust.

However, multiple strains with diverse growth rates and needs can result in difficulties in

achieving optimum conditions and performance (Ju et al., 2008; Luo and Angelidaki,

2012). So far, the reported CH4 productivity rates of mixed enriched culture reactors also

appear to be significantly lower (10.1 v/v/d with 91% v/v CH4 in the effluent gas) (Luo and

Angelidaki, 2012) than the ones achieved by monoculture reactors (511 v/v/d with 60% v/v

CH4 in the effluent gas) (Seifert et al., 2014) and (288 v/v/d with 96% CH4% in the effluent

gas) (Peillex et al., 1990), although some recent studies using pure cultures have only

achieved outputs significantly lower than this (9.84 v/v/d with 96% v/v CH4 in the effluent

gas and 65.6 v/v/d with 34% v/v CH4 in the effluent gas) (Martin et al., 2013).

Kinetic studies for pure and mixed enriched cultures in chemostats show that growth and

methane formation are dependent on the availability of hydrogen in its dissolved state as

well as the availability of a number of nutrients in the culture medium (Ako et al., 2008; De

Poorter et al., 2007; Schill et al., 1996; Zhang and Maekawa, 1993). Nutrient availability

has been regulated by the continuous dosing with a defined nutrient medium (liquid media

dilution rate); however, this requirement for nutrient addition can limit the practical and

economic viability of the process. In order for a steady state to be achieved growth rates

need to be equalized to biomass washout rates. However, growth and methanogenesis

appear to be closely coupled under conditions where the rate of hydrogen consumption

6

does not allow for a pool of excess hydrogen to be formed in the media (De Poorter et al.,

2007). For a system that aims for the complete conversion of input gases this means that as

conversion rates increase so do growth rates and consequently the required biomass

washout rates for a steady state to be achieved. This unavoidably leads to high nutrient

dosing requirements.

To give an example of magnitude, CH4 productivities of 288 v/v/d have been reported

(Peillex et al., 1990) with continuous CSTR reactors where the liquid medium was renewed

at a rate of 3.6 reactor volumes d-1 in order for the system to reach a steady state in terms of

biomass concentration. At a commercial scale this would add considerably to the running

costs of the system as it would increase both the degree of control required for stable

operation and the expenses for consumables. A high environmental cost could also be

incurred as effluents rich in microbial biomass and some heavy metals are likely to require

treatment prior to disposal. As a consequence, the elimination or at least the reduction of

the addition of chemical species would be desirable.

In this study an approach was examined where the biomethanation reaction was allowed to

proceed continuously in terms of gas conversion (chemolithotrophy) while the initial

inoculum (obtained from a full-scale digester) in the reactor was starved of any additional

micro or macro nutrients over a period of 6 months of operation. The aim of the study was

to investigate the extent to which the mixed microbial culture catabolism could redistribute

the initial pool of nutrients in the medium and if such a mechanism could exist in parallel

with the process of hydrogenotrophic methanogenesis. The study also introduces for the

first time a pH control mechanism which solely depends on the amount of CO2 entering the

methanation reactor thus substituting the use of pH buffering solutions.

7

2. Materials and Methods

2.1. Inoculum

Anaerobically digested mesophilic sewage sludge collected from Cog Moors Wastewater

Treatment Plant in Cardiff, South Wales, UK was used as the inoculum. The plant operates

a conventional digestion process and treats approximately 90% of secondary and 10% of

primary sludge. The plant runs typically at a temperature of 35±2oC. Prior to use, the sludge

was filtered through a 125 μm stainless steel sieve to approximately 17 g/L TS and 9.5 g/L

VS. No pH/redox buffering agents or nutrient solutions were used throughout the

experiment.

2.2. Reactor set up and operating conditions

The main body of the reactor comprised a 110 cm tall glass cylinder with a working volume

of 1.5 L. A centrifugal pump (NewJet 1200, Newa Tecno Industria Srl, Italy) was

connected to a port at the side of the reactor to recirculate the liquid media at a constant

flowrate (6 L/min) by taking the liquid from the bottom side of the reactor and

reintroducing it at the top as shown in Figure 1. A gas inlet was attached to the bottom of

the glass cylinder so that the gas was directly drawn into the pump as soon as it entered the

reactor. As such, the pump served as a gas - liquid mixer and contributed to the breaking up

of the gas bubbles as these passed through the pump impeller (Esteves et al., 2015). Several

other ports served as data collection points for continuous pH, temperature and reaction

efficiency measurements and for sampling and analysis purposes. The gas feedstock was a

mixture of CO2 and H2, both supplied from compressed gas cylinders. The gas flowrate and

composition were continuously monitored by a series of gas sensors and flow meters

(bespoke manufactured tip meters) and adjusted according to experimental requirements.

8

The gas flow exiting the reactor was also continuously analysed by the same methodology.

The temperature of the liquid media was maintained at 37±0.5oC throughout the experiment

with the use of an external heating element connected to a temperature control unit (Elitech

STC-1000, Elitech, UK).

2.3. Analytical methods

Gas composition was determined in real time by infra-red sensors (Premier Series 0-100%

Vol CO2/CH4 Voltage output 0.4-2.0V, Dynament Ltd) and by in-line hydrogen solid-state

sensors (H2Scan HY-OPTIMA 740, 0-100% Vol H2, 4-20mA output). Gas composition

was also periodically analysed with a gas chromatograph (Varian Inc., CP-4900) equipped

with two columns, one for CO2 (Porapack Q, Varian – 10 m x 0.15 mm) and one for CH4,

H2, N2 and O2 (Molsieve 5A Plot, Varian – 10 m x 0.32 mm). The carrier gas used was Ar.

Gas flow rates were measured by custom made tip-meters and logged in LabVIEWTM.

Volatile Fatty Acids (VFAs) were determined according to (Cruwys et al., 2002) using a

head space autosampler gas chromatograph (Perkin Elmer, AutosystemXL) equipped with

a flame ionization detector and a Supelco Ltd. column (30 m x 0.32 mm). The carrier gas

was N2. Metal analysis was carried out by ICP-OES (Inductively coupled plasma optical

emission spectrometry) on samples prepared by acid digestion according to (EPA, 1996).

pH was determined in real time with the use of a pH electrode HI-1001 (Hannah

Instruments, UK) connected to the main body of the reactor. The electrode was connected

to a BL-931700-1 pH controller (Hannah Instruments, UK). Total Solids (TS) and Volatile

Solids (VS) of the reactor matrix/effluent were measured according to (APHA, 2012).

2.4. Data acquisition

9

Data on temperature, pH, gas composition and gas flow were collected in real time with the

use of individually dedicated sensors connected to a data acquisition device (DAQ USB

6002, National InstrumentsTM UK). The software used for data logging was LabVIEW

(National InstrumentsTM UK). An interface unit was built and used as a server for signal and

power distribution. The data logging frequency was set to 0.3 Hz. Recalibration of the gas

and pH sensors took place once a month according to the manufacturer’s guidelines.

2.5. Microbial profiling

A PowerSoil DNA Isolation kit (Mo Bio Laboratories Inc., USA) was used for DNA

extraction. After purification, DNA concentration was measured with a spectrophotometer

based on absorbance at 260 nm (NanoDrop 1000, Thermo Scientific, UK). Bacterial rDNA

standards were used for the quantification of total bacteria according to the method

described in (Suzuki et al., 2000). Hydrogenotrophic and acetoclastic methanogens were

quantified by using the method defined in (Yu et al., 2005). The DNA for positive controls

was extracted from pure cultures supplied by the Leibniz Institute (DSMZ). The species

Halorubrum saccharovorum (DSM 1137) was used as control for total bacteria. In order to

cover most of the methanogenic populations typically present in AD systems, five different

order and family levels were investigated by using the following species: Methanosaeta

concilii (DSM 6752), Methanosarcina barkeri (DSM 800), Methanobacterium bryantii

(DSM 863), Methanomicrobium mobile (DSM 1539), Methanococcus voltae (DSM 1537).

Real-time PCR was conducted on a Roche LightCycler nano by using TaqMan

primers/probe sets (Life Technologies, Thermo Scientific, U.K) targeting 16S rRNA gene

sequences. Calibration curves were produced by using known amounts of oligonucleotides

10

(Table 1) that contained complementary sequences to the primers and probes. All real-time

PCR samples were analysed in triplicate.

3. Results and discussion

3.1. Hydrogenotrophic methanogenic capacity and regulation of pH

During operation the reactor was subjected to a series of different gas feeding rates

according to the observed methanogenic capacity of the culture at various H2/CO2 ratios.

The aim was to achieve almost complete conversion of CO2 to CH4. The reason for this was

the fact that in the absence of additional chemical pH buffering agents, any amount of

unconverted CO2 was observed to lower the pH to sub-optimal levels (< 7) due to the

formation of H2CO3. Conversely, it was also observed that by decreasing the amount of

CO2 in the feeding gas to levels below the level required for its complete utilization resulted

in an increase of the pH to sub-optimally high levels (> 8). Figure 2 shows the volumetric

percentage of CH4 in the effluent gas relative to the percentage of CO2 in the feeding gas in

the course of 180 days. Figure 3 shows the volumetric percentage of CH4 in the effluent gas

in relation to the applied gas feeding rates during the same period as well as the pH of the

media.

Due to the high data logging frequency (0.3 Hz) the data points representing the pH and the

volumetric percentage of gases in Figures 2 and 3 had to be averaged to 1 hour periods.

Also any non-operational periods have been omitted from the graphs. From Figure 2 it can

be observed that for almost half of the operational period (up to day 65), the percentage of

CO2 entering the reactors was not stable. This is due to the gas mixing and delivery system

being sub-optimised and regularly drifting from the chosen set-point as well as due to the

effort to control and stabilize the pH of the culture by regulating the amount of CO2 in the

11

gas feed (H2/CO2 ratio), which was accomplished through a degree of adjustments based

initially on trial and error. Figure 2 reveals how strongly the percentage of CH4 in the

effluent gas depended on the percentage of CO2 in the influent gas since many of the peaks

and troughs of the two curves are mirroring each other. After a certain degree of stability

has been established regarding the feeding gas composition (after day 65), it can also be

seen that the H2/CO2 ratio does not follow the stoichiometry described in equation 1. This

can be attributed to the fact that during hydrogenotrophic methanogenesis an extra amount

of CO2 is actually directed towards the anabolic needs of the microbes and therefore it has

to be added on top of the amount used for methanogenesis.

Regarding the marked disturbances in Figure 2, numbers 1, 3, 6 and 8 were due to known

technical reasons whereas numbers 2, 4 and 7 were related to large drifts from the optimum

H2/CO2 ratio. Specifically, disturbance number 3 was produced by a pump failure and

resulted in the introduction of an undetermined amount of oxygen in the reactor for a

couple of hours. Disturbance number 5 was due to a 45 day fasting period with no gas input

into the system (this period of non-feeding is omitted from the graph). Disturbance number

9 was related to an oxygenation experiment; the effects of fasting and oxygenation on the

microbiome of the media are not discussed here. Smaller non-marked disturbances indicate

sampling and routine equipment maintenance.

From Figure 3 it can be seen that the pH of the media could be regulated within an

acceptable range (pH 7-8) throughout the experiment despite the stepwise increase of gas

feeding rates (from 27.9 to 60.5 v/v/d). This was possible due to the gradual increase in the

hydrogenotrophic activity of the culture.

12

As the enrichment of H2/CO2 consuming groups took place, the higher rates of CO2

consumption allowed for higher rates of CO2 injection into the system without this affecting

the pH of the media. A degree of volatilization of nitrogen (via NH3) may have also

affected the impact of CO2 injection on the pH.

Figure 4 indicates that the pH of the media could be finely tuned by controlling the levels

of CO2 entering the system in relation to the data obtained at the time in terms of

conversion efficiency and pH. The figure displays the course of 3 days of operation during

which, by finding an optimum for the CO2/H2 ratio of the gas entering the system. pH was

stabilized at just below 7.2 whereas conversion efficiency was close to 100%. It can also be

seen that during this period CO2 conversion was complete as there was no detection of CO2

exiting the reactor.

The diffusion rate of CO2 into the media was directly linked to two measurable parameters:

the pH and the amount of CO2 in the exhaust gas. Since the values of both these parameters

were continuously registered, they could potentially be used as reference for a system that

automatically controls the H2/CO2 ratio of the feeding gas as well as the gassing rate. By

simultaneously aiming for a certain pH range and for complete conversion of CO2 to CH4,

the H2/CO2 ratio and gassing rate could be continuously adjusted by active control of the

gas feeding devices (e.g. individual mass flow controllers). This way, and without human

interference, the system would be able to always detect the methanogenic capacity of the

culture in order to achieve the highest conversion rates and efficiency.

3.2. Enrichment of the hydrogenotrophic methanogenic population

As the only input of external agents into the system after inoculation was a stream of

H2/CO2 mixture, this created a closed ecosystem in terms of organic material (sugars, lipids

13

and amino-acids) essential for the survival of a wide range of bacterial species. In a

chemostat arrangement this would have resulted in the decline of all microbial activity

apart from the one relying on the conversion of the two supplied gases. However, since

there was no washout of biomass during the experimental period (apart from sampling and

to keep a constant liquid media volume due to the generation of water) dead biomass in the

reactor appeared to have been continuously recycled as feed. This hypothesis is supported

by three lines of evidence.

Firstly, the gradual increase in gas conversion rates (from 27.9 to 60.5 v/v/d) and efficiency

as shown in Figure 3 indicates that there was no decrease in metabolic activity as would

have been expected in the case of a non-dividing, ageing population. A non-dividing

population would have been the result of nutrient limiting conditions as dictated by batch

culture kinetics (Shuler and Kargi, 2002). Since no further micro or macro nutrients where

supplied to the system after start-up, a mechanism for the recycling of the initial pool of

these must have taken place.

Figure 5 supplies the second line of evidence. Samples taken on days 1 and 171 and

analysed by q-PCR show that the relative quantities of bacteria and archaea have stayed

comparatively unaffected while there was almost complete displacement of the acetotrophic

methanogenic species by their hydrogenotrophic counterparts. More specifically, the gene

copy numbers of total bacteria increased from 2.6 x 109 to 6.8 x 109 and the gene copy

numbers for hydrogenotrophic methanogens from 4.6 x 105 to 4.8 x 108 while the gene

copy numbers of acetotrophic methanogens declined from 1.7 x 108 to 1.5 x 107. It must be

noted that although the hydrogenotrophic methanogens detected included the orders of

Methanobacteriales and Methanomicrobiales and the family of Methanosarcinaceae, on

14

day 171 the order of Methanobacteriales dominated the hydrogenotrophic population by

100%.

The initial inoculum contained all the necessary microbes involved in the stages of

hydrolysis, acidogenesis, acetogenesis, and acetoclasis together with a range of

hydrogenotrophic methanogens. Under non-restricted conditions in terms of carbon and

energy but restricted in terms of micro and macronutrients, the mixed populations created

syntrophic relationships that promoted recycling of their cellular material, thus leading to a

self-regenerative system where decaying biomass became utilized to yield new biomass.

In this closed system a significant imbalance factor came from the metabolically produced

water as shown in the following, (equation 1) hydrogenotrophic methanogenesis reaction.

Due to the high rate of methanogenesis when compared to the digestion of biomass any

other possible routes of metabolically produced water can be ignored.

Equation 1 states that for the generation of every mole of CH4, 2 moles (≈36 ml) of water

are produced that result in a reduction of the concentration of nutrients over time through

dilution. The effect of dilution through metabolically produced water can be seen in Figure

6 which displays the concentration of total and volatile solids throughout the experiment.

The amount of undigested solids can be calculated from equation 2:

Mineral content = Total solids – Volatile solids (2)

After 185 days of operation it can be safely assumed that the concentration of biomass in

g/L in the media was represented by the VS which experienced an overall reduction of

35.2%. The mineral content experienced a reduction of 76.4%. The difference in the two

values can be explained by the fact that although the concentration of undigested solids was

influenced by the production of water, the concentration of the microbial mass in the

15

reactor was dictated by the amount of available nutrients which appeared to be adequate at

a wide range of dilutions. This is also reflected in Figure 6 which shows that despite the

dilution factor the density of the microbial population experienced an increase of 167%. At

the beginning of the experiment the inoculum contained a certain amount of organics that

must have been gradually taken up by the microbes which justifies the reduction in VS with

the concurrent increase in gene copy numbers.

Table 2 displays the concentration of 16 elements in the media that are considered essential

to methanogenic cells (Angelidaki and Sanders, 2004; Demirel and Scherer, 2011; Glass

and Orphan, 2012; Zhang and Gladyshev, 2010). Concentration was measured with ICP-

OES and the values have been adjusted to represent their concentration per 10 g of TS. The

values show that apart from the dilution factor there was no loss of these elements during

operation.

The third line of evidence for the recycling of dead biomass material comes from the fact

that the concentration of VFAs in the media throughout the duration of the experiment was

always below 100 mg/L (data not shown). In AD systems several intermediates play a

crucial role for the mineralization of organic compounds to CH4 and CO2. Among them,

propionate and butyrate are considered as rate limiting because acetotrophic

methanogenesis depends on their successful oxidation which is thermodynamically

unfavorable (Amani et al., 2010). The degradation of these intermediates depends on the

establishment of a syntrophic relation between acetogenic bacteria and H2/formate

scavengers which are part of the hydrogenotrophic population. Accumulation of propionate

or butyrate is therefore often a sign of disruption of this syntrophy which can be caused by

various factors such as hydrogen partial pressure, pH and even the accumulation of VFAs

16

as a form of self-inhibition (Li et al., 2012). In the case of acetate, its oxidation is exergonic

and accumulation is most often linked directly to underperforming acetoclastic species

(Amani et al., 2010).

The degradation of propionic acid has been reported to be dependent on the successful

removal of its products, most notably H2 although formate and acetate might as well play a

role (De Bok et al., 2004). In the present case, the absence of accumulated intermediates

points towards their continuous and successful utilization even at high H2 partial pressures.

The continuous gassing with H2 did not appear to inhibit the oxidation of propionate or

butyrate due to its immediate consumption by the hydrogenotrophic population. The low

acetate levels also indicate the presence of uninhibited aceticlastic methanogens at high

enough numbers for its complete utilization.

Nevertheless, the gradual weakening of the culture through consecutive mutations, loss of

volatile nutrients (e.g. H2S, NH3) through evaporation or the accumulation of undigested

intermediates could cause imbalance and possible failure. Therefore, longer operational

periods need to be assessed in order for any potential impediments to be identified. Another

issue that has to be addressed is the water that is produced during methanogenesis. In the

case of a closed system, dilution is unavoidable and will eventually lead to nutrient scarcity.

Consequently, methods for the removal of water should be devised. However, these ought

not to be energy intensive or disrupt the methanation process. Biofilms will likely aid in

this direction by separating the culture from any liquid processing route.

Stable isotope labeling could also provide a clearer image of the pathways involved in the

process via the detection of the carbon entering the system with the gas in certain microbial

groups and their metabolites. Since the only constantly renewed carbon source is CO2, the

17

substitution of 12C with its heavier isotope 13C intermittently and for short time periods

could help identify how, at what rate and by which microbial groups it is utilized thus

giving us their individual growth and death rates. The quantification of 13C in the biomass

and its metabolic products versus time would offer a better understanding of how this

mixed, partially closed ecosystem works and an insight on how it can be improved.

4. Conclusions

The present study shows that an eco-engineered mixed culture biocatalyst that promotes the

gaseous conversion of CO2 and H2 to CH4 is capable of regeneration relying on catabolism.

This is the first time that microbial recycling has been allowed to evolve in the process of

ex-situ biomethanation. This is also the first time that the different groups of a mixed

microbial population involved in ex-situ hydrogenotrophic methanogenesis have been

quantified. The avoidance of the continuous addition of nutrients and pH buffering agents is

expected to bring a significant reduction in the running costs of future commercial

biomethanation units.

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), and funding provided by Innovate UK / BBSRC for an

Industrial Biotechnology Catalyst program early stage feasibility study (Project Ref:

P132133).

18

Table 1 Oligonucleotide sequences used as calibration standards.

Microbial target group

Sequence

Total bacteria CGGTGAATACGTTCYCGGGACTTGTACACACCGCCCGTCTCAAGTCGTAACAAGGTAWCC

Methanosaetaceae TAATCCTYGARGGACCACCAGTACGGCAAGGGACGAAAGCTAGGACGTKGTYGGTGCCGTAGG

Methanosarcinaceae GAAACCGYGATAAGGGGAGTTTAGCAAGGGCCGGGCAAACCGTAAACGATGYTCGCTA

Methanococcales TAAGGGCTGGGCAAGTACTAGCGGTGRAATGYGTTGATCCGTTAAACTYTGCGRACTAGGTG

Methanomicrobiales ATCGRTACGGGTTGTGGGACTYCGACAGTGAGGRACGAAAGCTGGTGTAAACDATGYGCGTTAGGTG

Methanobacteriales CGWAGGGAAGCTGTTAAGTGTAGCACCACAACGCGTGGAACAAGGAGTGGACGACGGTA

Table 2 Concentration of 16 elements in the liquid media at three different dates

mg/10 g TS

Al B Ca Co Cu Fe K Mg

day 39 115.61 0.43 257.54 0.09 2.69 188.26 76.83 59.26 day 67 131.69 0.08 252.04 0.09 3.06 176.52 78.28 63.82 day 133 117.45 0.37 244.85 0.10 2.83 179.46 89.35 59.98 Mn Mo Na Ni P S Se Zn

day 39 4.69 0.93 49.52 3.78 148.73 102.97 <0.02 7.59 day 67 5.64 1.23 43.24 4.16 155.27 123.79 <0.02 8.73 day 133 4.95 1.42 56.84 5.44 148.38 109.21 <0.02 7.48

19

Fig. 1 Schematic of the methanation reactor connected to the gas supply and the real time

data logging system.

Fig. 2 Volumetric percentage of CH4 in the effluent gas compared to the volumetric percentage of CO2 in the feeding gas; full monitoring was available from day 10.

20

Fig. 3 Volumetric percentage of CH4 in the effluent gas related to the gassing rate with the H2/CO2 mix and the pH of the liquid media.

Fig. 4 Stabilisation of the pH of the media by manual control of the CO2 entering the reactor.

21

Fig 5 Gene copy numbers per ml of sample and relative quantities of bacteria and archaea at start-up and after 171 days of operation. 1Acetoclastic methanogens cover the family of Methanosaetaceae. 2Hydrogenotrophic methanogens cover the orders of Methanobacteriales (MBT) and Methanomicrobiales (MMB) and the family of Methanosarcinaceae (MSC).

Fig. 6 Concentration of Total Solids and Volatile Solids in the reactor during operation; comparison with the density of the microbial population on days 1 and 171.

22

References

1. Ako, O.Y., Kitamura, Y., Intabon, K., Satake, T., 2008. Steady state characteristics of

acclimated hydrogenotrophic methanogens on inorganic substrate in continuous

chemostat reactors. Bioresource Technology 99, 6305–6310.

doi:10.1016/j.biortech.2007.12.016

2. Amani, T., Nosrati, M., Mousavi, S.M., Kermanshahi, R.K., 2010. Study of syntrophic

anaerobic digestion of volatile fatty acids using enriched cultures at mesophilic

conditions. International Journal of Environmental Science & Technology 8, 83–96.

doi:10.1007/BF03326198

3. Angelidaki, I., Sanders, W., 2004. Assessment of the anaerobic biodegradability of

macropollutants. Reviews in Environmental Science and Biotechnology 3, 117–129.

doi:10.1007/s11157-004-2502-3

4. APHA, 2012. Standard Methods for the Examination of Water and Wastewater, 22nd

ed. American Public Health Association, American Water Works Association, Water

Environment Federation, Washington, D.C.

5. Barnhart, C.J., Dale, M., Brandt, A.R., Benson, S.M., 2013. The energetic implications

of curtailing versus storing solar- and wind-generated electricity. Energy &

Environmental Science 6, 2804–2810. doi:10.1039/c3ee41973h

6. Burkhardt, M., Koschack, T., Busch, G., 2015. Biocatalytic methanation of hydrogen

and carbon dioxide in an anaerobic three-phase system. Bioresource Technology 178,

330–333. doi:10.1016/j.biortech.2014.08.023

7. Cruwys, J.., Dinsdale, R.., Hawkes, F.., Hawkes, D.., 2002. Development of a static

headspace gas chromatographic procedure for the routine analysis of volatile fatty

23

acids in wastewaters. Journal of Chromatography A 945, 195–209. doi:10.1016/S0021-

9673(01)01514-X

8. De Bok, F.A.M., Plugge, C.M., Stams, A.J.M., 2004. Interspecies electron transfer in

methanogenic propionate degrading consortia. Water Research 38, 1368–1375.

doi:10.1016/j.watres.2003.11.028

9. De Poorter, L.M.I., Geerts, W.J., Keltjens, J.T., 2007. Coupling of

Methanothermobacter thermautotrophicus methane formation and growth in fed-batch

and continuous cultures under different H2 gassing regimens. Applied and

Environmental Microbiology 73, 740–749. doi:10.1128/AEM.01885-06

10. Demirel, B., Scherer, P., 2011. Trace element requirements of agricultural biogas

digesters during biological conversion of renewable biomass to methane. Biomass and

Bioenergy 35, 992–998. doi:10.1016/j.biombioe.2010.12.022

11. DENA, 2015. Power to Gas system solution. Opportunities, challenges and parameters

on the way to marketability. Danish Energy Agency.

12. Denholm, P., Ela, E., Kirby, B., Milligan, M., 2010. The Role of Energy Storage with

Renewable Electricity Generation The Role of Energy Storage with Renewable

Electricity Generation. National Renewable Energy Laboratory Publication (NREL) 1–

61. doi:69

13. EPA, 1996. Acid Digestion of Sediments, Sludges and Soils [WWW Document]. U.S

Environmental Protection Agency. URL

http://www3.epa.gov/epawaste/hazard/testmethods/sw846/pdfs/3050b.pdf

14. Esteves, S., Dinsdale, R., Patterson, T., Savvas, S., 2015. Microbial Processing of

Gasses. PCT/GB2015/054176.

24

15. Glass, J.B., Orphan, V.J., 2012. Trace metal requirements for microbial enzymes

involved in the production and consumption of methane and nitrous oxide. Frontiers in

Microbiology 3, 1–20. doi:10.3389/fmicb.2012.00061

16. Global Wind Energy Council (GWEC), 2014. Global Wind Report Annual Market

Update 2013. Brussels, Belgium.

17. Gonzalez, A., Gallachoir, B., McKeogh, E., 2004. Study of Electricity Storage

Technologies and Their Potential to Address Wind Energy Intermittency in Ireland -

Final Report. UCC Sustainable Energy Research Group.

18. Götz, M., Lefebvre, J., Mörs, F., McDaniel Koch, A., Graf, F., Bajohr, S., Reimert, R.,

Kolb, T., 2016. Renewable Power-to-Gas: A technological and economic review.

Renewable Energy 85, 1371–1390. doi:10.1016/j.renene.2015.07.066

19. Ju, D.-H., Shin, J.-H., Lee, H.-K., Kong, S.-H., Kim, J.-I., Sang, B.-I., 2008. Effects of

pH conditions on the biological conversion of carbon dioxide to methane in a hollow-

fiber membrane biofilm reactor (Hf–MBfR). Desalination 234, 409–415.

doi:10.1016/j.desal.2007.09.111

20. Kaster, A.K., Goenrich, M., Seedorf, H., Liesegang, H., Wollherr, A., Gottschalk, G.,

Thauer, R.K., 2011. More than 200 genes required for methane formation from H2 and

CO2 and energy conservation are present in methanothermobacter marburgensis and

methanothermobacter thermautotrophicus. Archaea 2011. doi:10.1155/2011/973848

21. Lee, J.C., Kim, J.H., Chang, W.S., Pak, D., 2012. Biological conversion of CO2 to

CH4 using hydrogenotrophic methanogen in a fixed bed reactor. Journal of Chemical

Technology and Biotechnology 87, 844–847. doi:10.1002/jctb.3787

25

22. Lew, D., Bird, L., Milligan, M., Speer, B., Wang, X., Carlini, E.M., Estanqueiro, A.,

Flynn, D., Gomez-lazaro, E., Menemenlis, N., Orths, A., Pineda, I., Smith, J.C., Soder,

L., Sorensen, P., 2013. Wind and Solar Curtailment Preprint. International Workshop

on Large-Scale Integration of Wind Power Into Power Systems.

23. Li, J., Ban, Q., Zhang, L., Jha, A.K., 2012. Syntrophic propionate degradation in

anaerobic digestion: A review. International Journal of Agriculture and Biology 14,

843–850.

24. Luo, G., Angelidaki, I., 2012. Integrated biogas upgrading and hydrogen utilization in

an anaerobic reactor containing enriched hydrogenotrophic methanogenic culture.

Biotechnology and Bioengineering 109, 2729–2736. doi:10.1002/bit.24557

25. Martin, M.R., Fornero, J.J., Stark, R., Mets, L., Angenent, L.T., 2013. A single-culture

bioprocess of methanothermobacter thermautotrophicus to upgrade digester biogas by

CO2-to-CH4 conversion with H2. Archaea 2013, 11. doi:10.1155/2013/157529

26. Pachauri, R.K., Allen, M.R., Barros, V.R., Broome, J., Cramer, W., Christ, R., 2014.

Climate Change 2014 Synthesis Report.

27. 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, 315–321.

28. Qadrdan, M., Abeysekera, M., Chaudry, M., Wu, J., Jenkins, N., 2015. Role of power-

to-gas in an integrated gas and electricity system in Great Britain. International Journal

of Hydrogen Energy 40, 5763–5775. doi:10.1016/j.ijhydene.2015.03.004

26

29. Rittmann, S., Seifert, A., Herwig, C., 2012. Quantitative analysis of media dilution rate

effects on Methanothermobacter marburgensis grown in continuous culture on H2 and

CO2. Biomass and Bioenergy 36, 293–301. doi:10.1016/j.biombioe.2011.10.038

30. Schill, N., van Gulik, W.M., Voisard, D., von Stockar, U., 1996. Continuous cultures

limited by a gaseous substrate: development of a simple, unstructure mathematical

model and experimental verification with Methanobacterium thermoautotrophicum.

Biotechnol. Bioeng. 51, 645–658.

31. Seifert, A.H., Rittmann, S., Herwig, C., 2014. Analysis of process related factors to

increase volumetric productivity and quality of biomethane with Methanothermobacter

marburgensis. Applied Energy 132, 155–162. doi:10.1016/j.apenergy.2014.07.002

32. Shuler, M., Kargi, F., 2002. Bioprocess Engineering: Basic Concepts. Upper Saddle

River, N.J.: Prentice Hall.

33. Suzuki, M.T., Taylor, L.T., Delong, E.F., Long, E.F.D.E., 2000. Quantitative Analysis

of Small-Subunit rRNA Genes in Mixed Microbial Populations via 5 ′ -Nuclease

Assays. Applied and environmental microbiology 66, 4605–4614.

doi:10.1128/AEM.66.11.4605-4614.2000.Updated

34. Yu, Y., Lee, C., Kim, J., Hwang, S., 2005. Group-specific primer and probe sets to

detect methanogenic communities using quantitative real-time polymerase chain

reaction. Biotechnology and Bioengineering 89, 670–679. doi:10.1002/bit.20347

35. Zhang, Y., Gladyshev, V.N., 2010. General trends in trace element utilization revealed

by comparative genomic analyses of Co, Cu, Mo, Ni, and Se. Journal of Biological

Chemistry 285, 3393–3405. doi:10.1074/jbc.M109.071746

27

36. Zhang, Z.Y., Maekawa, T., 1993. Kinetic Study on Fermentation from CO2 and H2

using the Acclimated-Methanogen in Batch Culture. Biomass and Bioenergy 4, 439–

446.

28