Archaeal community composition as key driver of H2 consumption rates at the start-up of the biomethanation process.

The performance of hydrogen consumption by various inocula derived from mesophilic anaerobic digestion plants was evaluated under ex situ biomethanation. A panel of 11 mesophilic inocula was operated at a concentration of 15 gVS.L-1 at a temperature of 35 °C in batch system with two successive injections of H2:CO2 (4:1 mol:mol). Hydrogen consumption and methane production rates were monitored from 44 h to 72 h. Hydrogen consumption kinetics varies significantly based on the inoculum origin, with no accumulation of volatile fatty acids. Microbial community analyses revealed that microbial indicators such as the increase in Methanosarcina sp. abundance and the increase of the Archaea/Bacteria ratio were associated to high initial hydrogen consumption rates. The improvement in the hydrogen consumption rate between the two injections was correlated with the enrichment in hydrogenotrophic methanogens. This work provides new insights into the early response of microbial communities to hydrogen injection and on the microbial structures that may favor their adaptation to the biomethanation process.

Improving the storage of cover crops by co-ensiling with different waste types: Effect on fermentation and effluent production.

Cover crops harvested at a low maturity stage generally have a high moisture content, which may generate energy losses during silage storage via effluent production and undesirable fermentations. This paper investigates the use of different waste types as absorbent co-substrates to be added before ensiling. The relation between the absorbent water holding capacity and silage effluent volume was first studied to find an effective parameter to prevent effluent production. Effluent retention was found to be proportional to the absorbent loading and water holding capacity (r2 = 0.98) and up to 90 % of effluent production was avoided when compared to control (295 l.t-1). The impact of different co-substrates (including bio-waste and manures) on overall ensiling performances was then investigated at an optimized absorbent loading. All co-substrates allowed a total effluent retention while a 76 l.t-1 effluent volume was reported for the control. The silage fermentation was modified or mostly unchanged depending on the co-substrate chemical and microbial properties and different metabolic pathways were observed (e.g. homolactic or butyric fermentation). In most conditions, the methane potential of the crop was efficiently preserved over a storage of 60 days. Co-ensiling was shown to be a relevant silage preparation method for biogas production.

Conditions for efficient alkaline storage of cover crops for biomethane production.

An innovative process aiming to combine storage and alkali pretreatment of cover crops was investigated using lime as a low cost and environmental friendly reactant. Different lime loadings and Total Solid concentrations (TS) allowed to highlight the abiotic mechanisms of deacetylation during the early stages of the process. Long-term storage experiments of rye and sunflower cover crops at 100 g.kgTS-1 lime loading allowed to evaluate the fermentation kinetics and to compare performances in dry and wet conditions to classical silage storage. The dry condition allowed an efficient alkaline storage and up to a 15.7% Biochemical Methane Potential (BMP) increase, while the wet condition underwent a succession of fermentations with a high butyric acid accumulation and H2 production, leading to a 13% BMP loss. Silage experiments allowed an efficient preservation of the BMP, with no significant variation over the 6-month storage duration.

Microbial community redundance in biomethanation systems lead to faster recovery of methane production rates after starvation.

The Power-to-Gas concept corresponds to the use of the electric energy surplus to produce H2 by water electrolysis, that can be further converted to methane by biomethanation. However, the fluctuant production of renewable energy sources can lead to discontinuous H2 injections into the reactors, that may interfere with the adaptation of the microbial community to high H2 partial pressures. In this study, the response of the microbial community to H2 and organic feed starvation was evaluated in in-situ and ex-situ biomethanation. The fed-batch reactors were fed with acetate or glucose and H2, and one or four weeks of starvation periods were investigated. Methane productivity was mostly affected by the four-week starvation period. However, both in-situ and ex-situ biomethanation reactors recovered their methane production rate after starvation within approximately one-week of normal operation, while the anaerobic digestion (AD) reactors did not recover their performances even after 3 weeks of normal operation. The recovery failure of the AD reactors was probably related to a slow growth of the syntrophic and methanogen microorganisms, that led to a VFA accumulation. On the contrary, the faster recovery of both biomethanation reactors was related to the replacement of Methanoculleus sp. by Methanobacterium sp., restoring the methane production in the in-situ and ex-situ biomethanation reactors. This study has shown that biomethanation processes can respond favourably to the intermittent H2 addition without compromising their CH4 production performance.

Long term alkaline storage and pretreatment process of cover crops for anaerobic digestion.

The aim of this work was to study an innovative alkaline process on two cover crops. CaO load of 60 g.kgTS-1 was implemented to combine the functions of storage and pretreatment. Lab-scale reactors were monitored for 180 days to assess the effect of this process on the physico-chemical properties of the biomass. From the first days, pH was not maintained in an alkaline zone and microbial fermentation activity was observed with the degradation of available carbohydrates and production of metabolites, CO2 and H2. High butyric acid accumulation was observed and mass losses of 18.1% and 9.0% of initial VS occurred for oat and rye, respectively. However, no methane potential loss was recorded in the short and long term and the crops were efficiently preserved. The pretreatment had no major impact on fiber solubilization, and no increase in BMP was obtained, which was attributed to the short duration of the alkaline conditions.

Impact of microbial inoculum storage on dark fermentative H2 production.

Complex organic substrates represent an important and relevant feedstock for producing hydrogen by Dark Fermentation (DF). Usually, an external microbial inoculum originated from various natural environments is added to seed the DF reactors. However, H2 yields are significantly impacted by the inoculum origin and the storage conditions as microbial community composition can fluctuate. This study aims to determine how the type and time of inoculum storage can impact the DF performances. Biochemical Hydrogen Potential tests were carried out using three substrates (glucose, the organic fraction of municipal solid waste, and food waste), inocula of three different origins, different storage conditions (freezing or freeze-drying) and duration. As a result, H2 production from glucose with the differently stored inocula was significantly impacted (positively or negatively) and was inoculum-origin-dependent. For complex substrates, hydrogen yields with the stored inocula were not statistically different from the fresh inocula, offering the possibility to store an inoculum.

Comparison of the mesophilic and thermophilic anaerobic digestion of spent cow bedding in leach-bed reactors.

Anaerobic digestion of spent cow bedding in batch leach-bed reactors (LBRs) was compared in mesophilic and thermophilic conditions for the first time. Results show that the use of thermophilic conditions enhanced only the degradation kinetics of easily-degradable matter during the first days of the digestion, whereas similar methane yields (80% of the Biomethane Potential) were reached after 42days at both temperatures. Therefore, the anaerobic digestion in LBRs of spent cow bedding, a substrate rich in slowly-degradable compounds, was not improved in term of methane production considering the overall digestion time. Moreover, the high initial biogas production rate in thermophilic reactors was found to significantly reduce the energetic performance of the cogeneration unit at industrial scale, leading to a 5.9% decrease in the annual electricity production when compared to a mesophilic one.

Leachate flush strategies for managing volatile fatty acids accumulation in leach-bed reactors.

In anaerobic leach-bed reactors (LBRs) co-digesting an easily- and a slowly-degradable substrate, the importance of the leachate flush both on extracting volatile fatty acids (VFAs) at the beginning of newly-started batches and on their consumption in mature reactors was tested. Regarding VFA extraction three leachate flush-rate conditions were studied: 0.5, 1 and 2Lkg-1TSd-1. Results showed that increasing the leachate flush-rate during the acidification phase is essential to increase degradation kinetics. After this initial phase, leachate injection is less important and the flush-rate could be reduced. The injection in mature reactors of leachate with an acetic acid concentration of 5 or 10gL-1 showed that for an optimized VFA consumption in LBRs, VFAs should be provided straight after the methane production peak in order to profit from a higher methanogenic activity, and every 6-7h to maintain a high biogas production rate.

Mesophilic anaerobic digestion of several types of spent livestock bedding in a batch leach-bed reactor: substrate characterization and process performance.

Spent animal bedding is a valuable resource for green energy production in rural areas. The properties of six types of spent bedding collected from deep-litter stables, housing either sheeps, goats, horses or cows, were compared and their anaerobic digestion in a batch Leach-Bed Reactor (LBR) was assessed. Spent horse bedding, when compared to all the other types, appeared to differ the most due to a greater amount of straw added to the litter and a more frequent litter change. Total solids content appeared to vary significantly from one bedding type to another, with consequent impact on the methane produced from the raw substrate. However, all the types of spent bedding had similar VS/TS (82.3-88.9)%, a C/N well-suited to anaerobic digestion (20-28, except that of the horse, 42) and their BMPs were in a narrow range (192-239NmLCH4/gVS). The anaerobic digestion in each LBR was stable and the pH always remained higher than 6.6 regardless of the type of bedding. In contrast to all the other substrates, spent goat bedding showed a stronger acidification resulting in a methane production lag phase. Finally, spent bedding of different origins reached, on average, (89±11)% of their BMP after 60days of operation. This means that this waste is well-suited for treatment in LBRs and that this is a promising process to recover energy from dry agricultural waste.

Conservation of acquired morphology and community structure in aged biofilms after facing environmental stress.

The influence of growth history on biofilm morphology and microbial community structure is poorly studied despite its important role for biofilm development. Here, biofilms were exposed to a change in hydrodynamic conditions at different growth stages and we observed how biofilm age affected the change in morphology and bacterial community structure. Biofilms were developed in two bubble column reactors, one operated under constant shear stress and one under variable shear stress. Biofilms were transferred from one reactor to the other at different stages in their development by withdrawing and inserting the support medium from one reactor to the other. The developments of morphology and microbial community structure were followed by image analysis and molecular tools. When transferred early in biofilm development, biofilms adapted to the new hydrodynamic conditions and adopted features of the biofilm already developed in the receiving reactor. Biofilms transferred at a late state of biofilm development continued their initial trajectories of morphology and community development even in a new environment. These biofilms did not immediately adapt to their new environment and kept features acquired during their early growth phase, a property we called memory effect.

Dynamic observation of the biodegradation of lignocellulosic tissue under solid-state anaerobic conditions.

The solid-state anaerobic digestion (SS-AD) of wheat straw was characterized under low inoculated batch tests during 244 days. High levels of degradation of the cellulose (52%±1) and hemicelluloses (55%±2) were observed at the final stages and associated to a methane yield of 204±16 NmL gTS(-1). Ultrastructural observations, using transmission electronic microscopy, indicated that microorganisms degraded wheat straw from the central to the outer tissue (i.e. parenchyma to epidermis), depending on cell chemical, physical accessibility and the degree of lignification. Furthermore, major degradation of sclerenchyma secondary walls was observed. The bioaccessibility of lignocellulosic structures of wheat straw is mainly limited by the external waxy layer (cuticle), tertiary cell walls, high silica content and access to the cell lumen.

Biomass hydrolysis inhibition at high hydrogen partial pressure in solid-state anaerobic digestion.

In solid-state anaerobic digestion, so-called ss-AD, biogas production is inhibited at high total solids contents. Such inhibition is likely caused by a slow diffusion of dissolved reaction intermediates that locally accumulate. In this study, we investigated the effect of H2 and CO2 partial pressure on ss-AD. Partial pressure of H2 and/or CO2 was artificially fixed, from 0 to 1 557mbars for H2 and from 0 to 427mbars for CO2. High partial pressure of H2 showed a significant effect on methanogenesis, while CO2 had no impact. At high [Formula: see text] , the overall substrate degradation decreased with no accumulation of metabolites from acidogenic bacteria, indicating that the hydrolytic activity was specifically impacted. Interestingly, such inhibition did not occur when CO2 was added with H2. This result suggests that CO2 gas transfer is probably a key factor in ss-AD from biomass.

Substrate milling pretreatment as a key parameter for Solid-State Anaerobic Digestion optimization.

The effect of milling pretreatment on performances of Solid-State Anaerobic Digestion (SS-AD) of raw lignocellulosic residue is still controverted. Three batch reactors treating different straw particle sizes (milled 0.25 mm, 1 mm and 10 mm) were followed during 62 days (6 sampling dates). Although a fine milling improves substrate accessibility and conversion rate (up to 30% compared to coarse milling), it also increases the risk of media acidification because of rapid and high acids production during fermentation of the substrate soluble fraction. Meanwhile, a gradual adaptation of microbial communities, were observed according to both reaction progress and methanogenic performances. The study concluded that particle size reduction affected strongly the performances of the reaction due to an increase of substrate bioaccessibility. An optimization of SS-AD processes thanks to particle size reduction could therefore be applied at farm or industrial scale only if a specific management of the soluble compounds is established.

An automated method for the quantification of moving predators such as rotifers in biofilms by image analysis.

In natural environments as well as in industrial processes, microorganisms form biofilms. Eukaryotic microorganisms, like metazoans and protozoans, can shape the microbial communities because of their grazing activity. However, their influence on biofilm structure is often neglected because of the lack of appropriate methods to quantify their presence. In the present work, a method has been developed to quantify moving population of rotifers within a biofilm. We developed an automated approach to characterize the rotifer population density. Two time lapse images are recorded per biofilm location at an interval of 1s. By subtracting the two images from each other, rotifer displacements that occurred between the two images acquisition can be quantified. A comparison of the image analysis approach with manually counted rotifers showed a correlation of R(2)=0.90, validating the automated method. We verified our method with two biofilms of different superficial and community structures and measured rotifer densities of up to 1700 per cm(2). The method can be adapted for other types of moving organisms in biofilms like nematodes and ciliates.

Dynamic effect of total solid content, low substrate/inoculum ratio and particle size on solid-state anaerobic digestion.

Among all the process parameters of solid-state anaerobic digestion (SS-AD), total solid content (TS), inoculation (S/X ratio) and size of the organic solid particles can be optimized to improve methane yield and process stability. To evaluate the effects of each parameter and their interactions on methane production, a three level Box-Behnken experimental design was implemented in SS-AD batch tests degrading wheat straw by adjusting: TS content from 15% to 25%, S/X ratio (in volatile solids) between 28 and 47 and particle size with a mean diameter ranging from 0.1 to 1.4mm. A dynamic analysis of the methane production indicates that the S/X ratio has only an effect during the start-up phase of the SS-AD. During the growing phase, TS content becomes the main parameter governing the methane production and its strong interaction with the particle size suggests the important role of water compartmentation on SS-AD.

Biofilm model calibration and microbial diversity study using Monte Carlo simulations.

Mathematical models are useful tools for studying and exploring biological conversion processes as well as microbial competition in biological treatment processes. A single-species biofilm model was used to describe biofilm reactor operation at three different hydraulic retention times (HRT). The single-species biofilm model was calibrated with sparse experimental data using the Monte Carlo filtering method. This calibrated single-species biofilm model was then extended to a multi-species model considering 10 different heterotrophic bacteria. The aim was to study microbial diversity in bulk phase biomass and biofilm, as well as the competition between suspended and attached biomass. At steady state and independently of the HRT, Monte Carlo simulations resulted only in one unique dominating bacterial species for suspended and attached biomass. The dominating bacterial species was determined by the highest specific substrate affinity (ratio of µ/KS ). At a short HRT of 20 min, the structure of the microbial community in the bulk liquid was determined by biomass detached from the biofilm. At a long HRT of 8 h, both biofilm detachment and microbial growth in the bulk liquid influenced the microbial community distribution.

Heterogeneity and spatial distribution of bacterial background contamination in pulp and process water of a paper mill.

Identifying the source and the distribution of bacterial contaminant communities in water circuits of industrial applications is critical even when the process may not show signs of acute biofouling. The endemic contamination of facilities can cause adverse effects on process runability but may be masked by the observed daily variability. The distribution of background communities of bacterial contaminants may therefore be critical in the development of new site-specific antifouling strategies. In a paper mill as one example for a full-scale production process, bacterial contaminants in process water and pulp suspensions were mapped using molecular fingerprints at representative locations throughout the plant. These ecological data were analyzed in the process-engineering context of pulp and water flow in the facilities. Dispersal limits within the plant environment led to the presence of distinct groups of contaminant communities in the primary units of the plant, despite high flows of water and paper pulp between units. In the paper machine circuit, community profiles were more homogeneous than in the other primary units. The variability between sampled communities in each primary unit was used to identify a possible point source of microbial contamination, in this case a storage silo for reused pulp. Part of the contamination problem in the paper mill is likely related to indirect effects of microbial activity under the local conditions in the silo rather than to the direct presence of accumulated microbial biomass.

Rapid measurement of the yield stress of anaerobically-digested solid waste using slump tests.

The anaerobic digestion of solid waste is usually performed using dry or semi-dry technology. Incoming waste and fermenting digestate are pasty media and thus, at the industrial scale, their suitability for pumping and mixing is a prerequisite at the industrial scale. However, their rheology has been poorly characterised in the literature because there is no suitable experimental system for analysing heterogeneous media composed of coarse particles. We have developed a practical rheometrical test, a "slump test", for the analysis of actual digested solid waste. It makes it possible to estimate yield stress from the final slump height. From the slump behavior, we conclude that digestates behave as visco-elastic materials. The yield stress of different digested waste was measured between 200 and 800Pa. We show that the media containing smaller particles or with higher moisture content are characterised by smaller yield stresses. This study thus demonstrates the impact of the origin of the digestate on the yield stress.

Experimental and modeling investigations of a hybrid upflow anaerobic sludge-filter bed (UASFB) reactor.

A 9.8-L hybrid UASFB reactor, in which the lower half was occupied by a sludge blanket and the upper half by small floating polyethylene media, was evaluated using wine distillery vinasse as substrate. The reactor was operated for a total period of 232 days at 33 + 1 degrees C. Continuous feeding of the reactor was started with an initial OLR of 2.9 g COD/L.d and then it was increased step wise to 19.5 g COD/L.d by increasing the feed COD, while maintaining a constant HRT (1.05 d). The reactor was equipped with a continuous internal recirculation system from top to the bottom at the rate of 9 L/h (upflow velocity = 0.83 m/h) upto day 159 and then it was reduced to about half on day 160 onwards. It was observed that the reduced recirculation rate did not affect the performance of the reactor with an average COD(t) and COD(s) removal efficiencies of 82 and 88%, respectively. A maximum gas production rate of 6.7 L CH(4)/L(reactor).d was achieved for the highest OLR applied. The specific activity analysis depicts that the activity of the attached biomass was more than 2 times higher than that of the granular sludge. The efficiency of liquid mixing was good through out this study. The packing medium had a dual role in the retention of the biomass inside the reactor: i.e. entrapment of biomass within the support and filtration of the granular biomass, preventing it from going out of the reactor. ADM1_10 model simulated well the dynamic evolutions of the main variables in the liquid as well as in the gas phases.