Accepted Manuscript Review Anaerobic digestion of microalgal biomass: challenges, opportunities and research needs Cristina Gonzalez-Fernandez, Bruno Sialve, Beatriz Molinuevo-Salces PII: DOI: Reference:

S0960-8524(15)01374-7 http://dx.doi.org/10.1016/j.biortech.2015.09.095 BITE 15604

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

7 September 2015 21 September 2015 22 September 2015

Please cite this article as: Gonzalez-Fernandez, C., Sialve, B., Molinuevo-Salces, B., Anaerobic digestion of microalgal biomass: challenges, opportunities and research needs, Bioresource Technology (2015), doi: http:// dx.doi.org/10.1016/j.biortech.2015.09.095

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Anaerobic digestion of microalgal biomass: challenges, opportunities and research needs Cristina Gonzalez-Fernandez1, Bruno Sialve2, Beatriz Molinuevo-Salces1 1 2

IMDEA Energy, Avda. Ramón de la Sagra 3, 28935 Móstoles, Madrid (Spain) INRA, Laboratoire de Biotechnologie de l’Environnement, Avenue des Etangs, Narbonne F-11100, France

* Corresponding author: E-mail address: [email protected] Tel.: +34 917371127

Abstract Integration of anaerobic digestion (AD) with microalgae processes has become a key topic to support economic and environmental development of this resource. Compared with other substrates, microalgae can be produced close to the plant without the need for arable lands and be fully integrated within a biorefinery. As a limiting step, anaerobic hydrolysis appears to be one of the most challenging steps to reach a positive economic balance and to completely exploit the potential of microalgae for biogas and fertilizers production. This review covers recent investigations dealing with microalgae AD and highlights research opportunities and needs to support the development of this resource. Novel approaches to increase hydrolysis rate, the importance of the reactor design and the noteworthiness of the microbial anaerobic community are addressed. Finally, the integration of AD with microalgae processes and the potential of the carboxylate platform for chemicals and biofuels production are reviewed.

Keywords: methane; anaerobic digestion; microalgae; carboxylate; cyanobacteria

1-

Introduction

AD is a well-known technology which converts complex organic matter into methane and carbon dioxide. Besides the production of renewable energy, the advantages of AD include 1

the reduction of greenhouse gas emissions, odor and pathogens and the production of a liquid digestate with fertiliser capacity. Four biological processes are involved in anaerobic digestion, namely hydrolysis, acidogenesis, acetogenesis and methanogenesis (Fig. 1). The hydrolysis of complex organic matter (carbohydrates, proteins and lipids) is carried out by extracellular enzymes that are excreted by different bacteria. These hydrolyzed molecules include amino acids, sugars, long chain fatty acids (LCFA) and alcohols. In the following stages, these molecules are converted into volatile fatty acids (VFA), hydrogen and carbon dioxide during acidogenesis and acetogenesis. The degradation of LCFA into acetate and hydrogen is carried out by obligate hydrogen producing acetogenic bacteria. Finally, methanogenesis is carried out by the methanogenic archaea. Archaeas grow very slowly and are less resilient to stress than bacteria. Acetate, hydrogen and carbon dioxide are the main substrates for methanogenic microorganisms, although some species are also able to metabolize formol, methanol or butyrate. Acetoclastic methanogenesis is generally the preferred catabolic pathway, representing 70% of the produced methane. However, the proportion of hydrogenotropic microorganisms becomes higher under stressful conditions (organic loading increase, inhibitors (ammonium or LCFA) accumulation or temperature variations, Carballa et al., 2015).

Figure 1 near here

Methanogenesis is the limiting phase in AD of easily degradable substrates. The slow growth rate of archaea often results in VFA accumulation and a consequent inhibition of methanogenic microorganism’s activity. Contrary, when working with particulate substrates, as microalgae, the hydrolysis phase determines the successful production of methane (Gonzalez-Fernandez et al., 2012).

2

The economic feasibility of biogas production from microalgae is dependent on the production cost of two technologies, the microalgae cultivation and the anaerobic digestion process. One of the main drawbacks from an environmental point of view for the industrial cultivation of microalgae is the high amount of fertilizer required to perform photosynthesis (Lardon et al., 2009). However, these nutrients can be provided by using nutrient-rich wastewaters as culture broth (Markou et al., 2012). This strategy would merge both technologies into a biorefinery concept by producing sustainable energy while contributing to wastewater bioremediation. Life cycle assessment (LCA) and net energy ratio (NER) have been used to study the techno-economical parameters of biogas production from microalgae. According to these parameters, the raceway reactor appears to be the most adequate bioreactor to produce microalgae in terms of energy, environmental sustainability and process efficiency (Bohutskyi and Bouwer, 2013). Nevertheless, biogas production feasibility using microalgae as substrates is still not clear and different strategies have been suggested. According to Collet et al. (2011), mixing and pumping are the most energy-consuming phases. In order to reduce the impacts, they proposed different solutions such as alternative ways of energy production in the facility (e.g. solar panels or wind turbines) or increasing the methane yield by employing pretreatments, codigestion or microalgae strain selection. Finally, the authors also concluded that the combination of lipid extraction with biogas production would be optimum from both environmental and economic points of view. Contrary, Quinn et al. (2014) evidenced the negative effect of lipid removal from Nannochloropsis salina on methane production. These authors pointed out the overestimation of AD performance from lipid-extracted microalgae in previous LCA studies, which were based on assumptions in favor of microalgae-based biofuels production. This study illustrated the importance of considering the effects of the removal of lipids from microalgae for modeling of anaerobic digestion performance in the LCA of microalgae biofuels since these

3

compounds are energy rich and extracted biomass ultimately provided lower methane yield than raw biomass.

This article studies different aspects of microalgae AD. The current pretreatments as well as novel approaches to increase hydrolysis rate in AD of microalgae are firstly reviewed. Then, the importance of the reactor design and bioprocess parameters on the methane production is addressed. After that, detailed information about the microbial anaerobic community is reported. Finally, the integration of AD with other microalgae processes and alternative products that could be obtained using AD is presented.

2-

Biomass special features and pretreatments prior to anaerobic digestion

Fresh green microalgae have been repeatedly reported to exhibit a hard cell wall that hampers an efficient anaerobic digestion (Gonzalez-Fernandez et al., 2012; Passos et al., 2014). Therefore, many pretreatment methods have been investigated for microalgae cell wall disruption or solubilisation before undergoing AD. Most of the tested pretreatments are the ones previously employed for activated sludge pretreatment. Despite the claimed similarities, the experiments conducted have shown that activated sludge and microalgae biomass are responding differently to same pretreatments (Gonzalez-Fernandez et al., 2012; Mahdy et al., 2015a). Concerning the different pretreatments employed until now, their effects in terms of biomass solubilisation and methane yield enhancement can be found elsewhere (Passos et al., 2014). In this review, pretreatments were divided into four categories namely thermal, mechanical, chemical and biological methods. The key point of these pretreatments is the energy balance involved in the process. Mechanical pretreatments have been pointed out as efficient methods for increasing biogas yield; nevertheless, the energy consumption renders

4

these methodologies economically unfeasible. On the other hand, thermal treatments have also provided promising results in terms of biogas production enhancement but it seems imperative to work with highly concentrated biomass to reach a positive energy production (Mendez et al., 2014; Passos et al., 2013). At this point, it has to be stressed out that the effectiveness of these pretreatments has been reported in several studies conducted in batch mode while the performance of CSTRs (continuous stirred tank reactor) has evidenced controversial results. In this context, several authors have noticed an increase in methane yield when comparing CSTRs fed with raw and thermally pretreated biomass but the absolute values were lower than the ones attained in batch digestion mode (Schwede et al., 2013; González-Fernández et al., 2013; Mendez et al., 2015a). This conclusion stands for different microalgae strains (Chlorella vulgaris, Scenedesmus obliquus, Nannochloropsis salina). In this sense, the scarce information available comparing digestion modes seems to merge to the conclusion of lower methane yield obtained in semicontinuos mode when compared to batch mode. Imbalances in the microbial population led to diminished methane yields. These imbalances can be caused by the presence of inhibitors released during thermal pretreatement. Chemical pretreatments applied to microalgae biomass displayed higher organic matter solubilisation than thermal pretreatments while biogas production was not enhanced accordingly (Mendez et al., 2013). The reason for such a fact was attributed to the polymerization of available reducing sugars and amino acids which led to the formation of complex molecules (Maillard reaction).

Overall, the outcomes of these pretreatments in fresh green microalgal biomass have been more or less successful depending on the pretreatment methodology employed and the microalgae strain targeted. This variability is mainly due to the macromolecular distribution profile and cell wall characteristics of each microalga. Therefore, the last two years have been

5

devoted to test different enzymes in order to fully understand which polymer is conferring this anaerobic resilience to microalgae biomass. These biocatalysts are a promising alternative to energy-consuming pretreatments since they involve mild conditions and minimal formation of toxic byproducts. The main drawback of this type of pretreatment is the enzyme production costs. Nevertheless, this could be overcome by in situ enzyme production. In this manner, the enzymatic hydrolysis could be implemented by: i) two-stage anaerobic digestion system (consisting of a hydrolytic-acidogenic stage followed by a methanogenic stage), ii) sludge bioaugmentation (addition of bacterial/fungal cultures required to speed up the rate of degradation), or iii) genetic modification of anaerobic microorganisms. Moreover, the action of some lytic enzymes involved in cell division and programmed cell death have been proven useful in performing cell lysis. In this sense, a recent review covers the use of intrinsic cell enzymes occurring in natural events to induce microalgae autolysis (Demuez et al., 2015).

As mentioned above, the use of biocatalysts can give crucial information about the polymers conferring high anaerobic resistance. Microalgae are closely related to plants, therefore a cellulosic/hemicellulosic cell wall has been repeatedly claimed (Ververis et al., 2007; Fu et al., 2010). Nevertheless, recent investigations have pointed out that this might be doubtful. According to Kim et al., (2014), only pectinase had a significant effect on polysaccharides cell wall degradation of C. vulgaris. It seems likely that C. vulgaris does not have a cellulose rigid cell wall but rather uronic acids and aminosugars are forming the polysaccharide cell wall matrix (Gerken et al., 2013; Mahdy et al., 2015b). In the biogas context, the effect of carbohydrases and proteases was compared in common green fresh microalgae (Scenedesmus obliquus and Chlorella vulgaris, Mahdy et al., 2015b). Out of the tested carbohydrases, the enzymatic cocktail including β-glucanase, xylanase, cellulase and hemicellulase was the most efficient in terms of carbohydrate solubilisation for both microalgae. In spite of the high

6

carbohydrate solubilisation achieved when using carbohydrases, digestion assays revealed that the biomass hydrolyzed with protease reached the highest methane yields. Regardless the different macromolecular profile of both microalgae, protease hydrolysis before AD enhanced methane yield by 1.72-fold and 1.53-fold for C. vulgaris and S. obliquus, respectively. The addition of proteases at dosage of 0.585 AU g dry weight-1 has evidenced complete protein solubilisation (Mahdy et al., 2014). The benefit of proteases for increasing biogas production has been also supported by Ometto et al., (2014). In that study, authors screened several enzymatic cocktails and concluded that the mixture of esterase and protease was the most effective catalysts for three different algae (Scenedesmus obliquus, Chlorella sorokiniana and Arthrospira maxima) while the effect was strain specific. The low hydrolysis rate for proteins, together with the fact that protease action provided the highest methane yield of different microalgae biomass led to the conclusion that cell proteins are directly linked to the low anaerobic biodegradability of this substrate. In this manner, it can be concluded that microalgae proteins deserve further investigation in order to fully elucidate their effect on AD (Mahdy et al., 2015b).

As alternatives to fresh green microalgae, the use of saline strains, cyanobacteria or microalgae-bacteria flocs have been lately studied. With regard to the halophytic strains, available literature concluded that methane production decreases concomitantly with increasing salinity. In this sense, it was suggested that the use of adapted inoculum is of paramount importance to accomplish an optimum digestion. An adapted inoculum contains methanogens more tolerant to saline concentration. This is the case reported by Mottet et al., (2014) who assessed different anaerobic inoculum for the digestion of Dunaliella salina. This study reported an adaptation of three inocula at low salinity (15 g L-1) while at higher values ranging 35 g L-1, the non-acclimated inoculum accumulated VFAs (acetate and propionate)

7

due to the reduced activity of methanogens. On the other hand, when working with two inocula exhibiting salinity concentration of 33 and 10 g L-1 when collected, the reactors were efficiently performing at salinity concentration of 35 g L-1. When increasing further the salinity concentration up to 75 g L-1, methane yield dropped along with time and halophilic methanogens were not able to work properly, not even after a long exposure time. Furthermore, these authors reported an out-competition of methanogens by sulfate reducer microorganisms at high salinity concentration. Opposite to this work, Ward et al., (2015a) described an efficient inoculum adaptation able to work at salinity concentration of 70 g L-1. The inoculum used at the start-up of the reactor was piggery effluents with an initial salinity of 16 g L-1. Due to the ability of bacterial communities to adapt to the different environmental conditions, these authors claimed a positive gas production when digesting pretreated Tetraselmis sp. However when taking a closer look to the data provided, the methane content in the biogas was rather low (50%). As a matter of fact, when comparing the methane yield attained in studies dealing with raw and pretreated saline microalgae (0.01-0.25 L CH4 g (volatile solids) VSin-1) with that of fresh microalgae (0.3-0.8 L CH4 g VSin-1) (Ward et al., 2015a; Ward and Lewis, 2015b; Sialve et al., 2009; Gonzalez-Fernandez et al., 2012), the results evidenced lower anaerobic biodegradabilities of halophytic microalgae strains. Most of these investigations highlighted the sodium concentration as responsible for those low methane yields.

Cyanobacteria are also potential candidates for biofuels production due to its capability to thrive in wastewaters and its filamentous form, making the harvesting step much easier than unicellular microalgae. The most studied cyanobacterium is Spirulina maxima and its potential for biogas production was firstly assessed in 1982 (Samson and Leduy, 1982). Methane yield reported at mesophilic range averaged 0.26 L g VSin-1 at an OLR of 0.97 g VS

8

L-1d-1 and HRT of 33 d. The main difference between green microalgae and cyanobacteria is that the latter ones lack of a hard cell wall which ultimately results in faster methane productivity since anaerobic hydrolysis is not hampered. Cyanobacteria may play a crucial role to circumvent the major drawback of microalgae; the hard cell wall. Cyanobacteria present a cell envelope typical of gram-negative bacteria. When compared to microalgal biomass, the use of cyanobacteria presented promising results without the need of any pretreatment before digestion. In this context, the study conducted by Mendez et al., (2015b) displayed a comparison of four cyanobacteria (Aphanizomenon ovalisporum, Anabaena planctonica, Borzia trilocularis and Synechocystis sp.) and one common robust microalga (Chlorella vulgaris) in terms of biomass productivity, biochemical characterization and methane production. Methane yields ranged 0.25-0.38 L CH4 g VSin-1 for the cyanobacteria strains while Chlorella vulgaris provided 0.24 L CH4 g VSin-1. The low growth rate of some cyanobacteria could be overcome by its anaerobic digestibility. Overall, balances showed that the cyanobacteria A. ovalisporum and A. planctonica were particularly good substrates for anaerobic digestion.

Similarly to green microalgae, cyanobacteria tend also to allocate the uptaken nutrients mainly as proteins and thus, these substrates are characterized by low C/N ratio (Mahdy et al., 2015a). The low C/N ratio could cause process instabilities because of the formation of ammonia. One strategy to increase this ratio and therefore optimize AD would be the use of carbon rich cosubstrates (Mahdy et al., 2015a; Park and Li, 2012). However, when these carbon-rich substrates are not available for codigestion, biomass macromolecular profile can be manipulated during growth. For instance, protein rich Arthrospira platensis shifted towards carbohydrate rich biomass by culturing this cyanobacterium under different amounts of phosphorus (Markou et al., 2012). Markou et al., (2013) demonstrated that the use of

9

carbohydrate-enriched Arthrospira platensis could improve the methane yield of this substrate. More specifically, the highest methane yield in bioreactors fed with 60% carbohydrates biomass was 0.20 ± 0.01 L CH4 g (chemical oxygen demand) CODin-1 while the lowest methane yield in bioreactors fed with 20% carbohydrates biomass was 0.12 ± 0.01 L CH4 g CODin-1. The main drawback of cyanobacteria is their ability to produce metabolites that include hepatotoxins, neurotoxins, and dermatotoxic compounds that can even pose adverse health risks to humans. However, not all cyanobacteria release toxins and moreover, the release of these toxins is highly influenced by biotic and abiotic environmental factors. These characteristics, together with their ability to grow at high pHs (Markou et al., 2014), can be beneficial to avoid culture contamination. Contamination may lead to productivity decrease or even the collapse of the entire culture. In this context, the role of cyanobacteria as a potential substrate for AD should not be neglected.

In some cases, contamination cannot be avoided; such is the case of culturing photosynthetic microorganisms for wastewater bioremediation. Microalgae and heterotrophic bacteria have shown beneficial interaction by establishing a cycle of oxygen/carbon dioxide production and usage thereof. Thus, when irradiated with light, microalgae produce the oxygen needed by aerobic bacteria to mineralize the organic pollutants, at the same time that they consume the ammonium and phosphorous released during organic matter mineralization. In turn, microalgae use the carbon dioxide released by heterotrophic bacteria for their metabolism. Even though there is a lot of research dealing with microalgae-bacteria systems for wastewater bioremediation, the investigation devoted to the use of this generated biomass is scarce. For instance, Passos et al., (2013) reported biogas yield of untreated microalgaebacteria biomass of 0.17 L CH4 g VSin-1 while Wieczorek et al., (2015) determined higher values ranging 0.24-0.30 L CH4 g VSin-1 depending on the substrate/inoculum ratio employed

10

for the digestion. These discrepancies can be attributed to the biomass composition (prevailing microalgae strain or amount of microalgae and bacteria forming part of this biomass). Mahdy et al., (2015a) evaluated the feasibility of using microalgae biomass as feedstock for AD together with primary and secondary sludge. The combination of microalgae biomass and secondary sludge (aerobic bacteria) provided neutral outcomes which meant that no synergies were detected when combining both substrates. The combination of C. vulgaris and activated sludge at 75%/25%, 50%/50% and 25%/75% provided 0.107, 0.091 and 0.092 L CH4 g CODin−1, respectively. Thermal pretreatment at 120ºC improved methane yields by approximately 50%. Interestingly, mild temperature pretreatment was more efficient for microalgae biomass than for secondary sludge biogas production (increasing 1.15 and 1.6fold methane yield, respectively). Likewise, the effect of seasonal variability of microalgaebacteria flocs grown in outdoors raceway ponds on the methane yield achievable by this biomass has been investigated by van den Hende et al., (2015). Methane yields fluctuated according to the harvested dates. Authors attributed those changes to the organic content (VS) of biomass total solids (TS). In this sense, the VS content of the flocs in summer were lower than in autumn or spring due to the formation of salts (calcium carbonate). These fluctuations were also observed for the COD/VS ratio of the collected biomass. This study concluded that the harvesting season affected biogas yields due to the different energy content of the biomass produced along the year. In summary, most of these studies highlighted the potential of combining microalgae and bacteria for wastewater bioremediation but the need of an effective pretreatment prior to AD was also pointed out.

3-

Reactor Configuration and Bioprocess parameters affecting methane production

3.1. Anaerobic digestion reactor: the importance of the configuration to set up hydraulic retention time (HRT), solid retention time (SRT) and organic loading rate (OLR)

11

Different anaerobic technologies are available in the marketplace. The reactor configuration should be specifically designed to the substrate fed in the digester. In order to reduce capital costs, reactors should be designed to achieve maximum methane production at the lowest HRT and highest OLR.

Goeluke et al., (1957) published the first study on AD of algae biomass. Coupling microalgae biomass (Chlorella sp. and Scenedesmus sp.) production with AD resulted in 0.17–0.32 L CH4 g VSin–1. The low conversion efficiency was attributed to ammonia-mediated inhibition and cell wall resistance to anaerobic bacterial attack. In this sense, during the last five years, most of the studies conducted were devoted to enhance methane production by using pretreatments for cell wall hydrolysis or disruption. The efficiency of these pretreatments has been mainly tested in batch mode assays for an easier and faster comparison to raw biomass (Mendez et al., 2013; Passos and Ferrer, 2015). Data obtained in batch assays can provide guidance, but assessing the benefits of pretreatments in semicontinuously fed reactors is highly required in order to study in-depth the performance of anaerobic microorganisms fed with pretreated microalgae biomass. As a matter of fact, only few investigations have moved forward to CSTR (Passos and Ferrer, 2015; Gonzalez-Fernandez et al., 2013; Schwede et al., 2013). The performance of CSTRs fed with microalgae is still very limited but probably this reactor configuration is the most used for the digestion of this substrate. This type of reactor entails suspended bacteria growth. Typical OLRs are between 1-6 g COD L-1 d-1 with HRT varying between 10-30 days. One of the first studies working with CSTR dealt with the digestion of Chlorella vulgaris (Ras et al., 2011). The methane yield reported in this study ranged 0.11 L CH4 g CODin-1 at HRT of 16 d and OLR of 1 g COD L-1 d-1. When digesting the same microalgae strain at HRT of 15 days and 1.5 g COD L-1 d-1, Mahdy et al., (2015c) obtained lower methane yield 0.05 L CH4 g COD in-1. Even lower values were reported by

12

Gonzalez-Fernandez et al., (2013) who digested Scenedesmus sp. When feeding at an OLR of 1 g tCOD L−1 d−1 and HRT of 15 d, the methane yield was around 0.03 L CH4 g COD in-1 and thus, 90% total COD fed to the reactor was still present in the effluent after AD. Likewise, a CSTR operated at HRT of 20 days with an average OLR of 2.3 g COD L-1d-1 fed with Oocystis sp. resulted in methane yield of 0.08 L CH4 g CODin-1 (Passos et al., 2015). At this point it should be highlighted that these CSTRs were operated under similar HRT; however it was shown that the different microalgae species exhibited different behaviour towards AD. One possible reason of this variability is the biomass composition and the different cell wall hardness among microalgae strains. Regardless the strain, almost all raw microalgae biomass supported low methane yields. In order to cope with that, one strategy is the use of pretreatments (as mentioned before) to optimize the methane yield but another strategy would be the use of a reactor configuration more suitable for this type of substrates. In this sense, decoupling HRT from SRT by biomass immobilisation, membrane retention or spatial separation of AD reactions may be potential configuration to improve biogas yields.

Attached-growth processes utilize either fixed film or carriers to support bacterial growth. For instance, Upflow Anaerobic Sludge Blanket (UASB) is a three phase reaction enabling the reactor to separate gas, water and anaerobic sludge mixtures. The advantage of this system includes the ability to keep large amounts of the biomass for degradation. The development of granular sludge is the key factor for the successful operation of UASB reactors which prevent sludge wash out. Additionally, this reactor configuration allows treating high OLR (5-30 g COD L-1 d-1) at shorter HRT (7-72 hours). Even though CSTRs are considered the best configuration for microalgal biomass, the use of UASB reactors has reported similar results when digesting Scenedesmus sp. (Tartakovsky et al., 2015). This later investigation reached methane yield of 0.22 L CH4 g VSin-1 when the UASB reactor was operated at HRT of 3.8

13

days and OLR of 2.25 g VS L-1 d-1. This methane yield was similar to the value achieved when digesting the same microalgae strain in CSTR operated at 16 days HRT and OLR of 0.65 g VS L-1 d-1 (Tartakovsky et al., 2013). In this manner, authors claimed the high efficiency of UASB reactor since they reached similar methane yield at lower HRT and higher OLR. Additionally, these authors highlighted the benefit of using this reactor configuration which requires a low influent microalgae concentration and thus avoiding the need of highly efficient harvesting systems. Another investigation working with a UASB-like reactor reported lower methane yield (0.11 L CH4 g VSin-1) when operated at HRT of 2.2 days and OLR at 2.7 g VS L-1 d-1 fed with Scenedesmus sp. (Zamalloa et al.., 2012a). This investigation also supported the possibility of shorten the HRT and increase the OLR. However, authors also determined low conversion efficiencies (approx. 20%) which evidenced the need of employing pretreatments before AD. Another alternative reactor configuration tested for microalgae digestion is the anaerobic membrane bioreactor (AnMBR). By using a membrane to separate the solids from the sludge suspensions, long SRT (up to 50 days) can be used by decreasing biomass waste and recovering nutrients in the liquid phase as potential fertilizer. Under this reactor configuration, similar methane yields to the ones obtained in UASB-like reactors were attained when digesting Scenedesmus sp. (SRT of 20 d and OLR of 10 g COD L-1 d-1, Zamalloa et al., 2012b). Nevertheless, the additional benefit of this reactor configuration is the effluent free of suspended solids. This clean effluent is achieved by the use of membranes, however one of the main disadvantages of this reactor configuration concerns membrane fouling and additional cost. Under this configuration, the OLR increase from 1 to 6 g COD L-1 d-1 was tested. The results evidenced that at the highest OLR, a high transmembrane pressure of 400 mbar was reached when working with membrane flux of 1.0±0.2 L m-2 h-1. In order to decrease this pressure and attain a stable performance, it was required to decrease the flux to 0.8 L m-2 h-1 and therefore, the SRT was set at 8 d (Zamalloa et al., 2012b). Similarly to these alternative reactors, anaerobic

14

baffled reactors also provide unique design characteristics including granulation promotion, long sludge retention time and phase separation. Thus, this configuration allows partial separation between the various AD phases. This spatial separation mediates different bacterial population in the different compartments while it also protects against toxic materials or environmental parameters changes. This type of reactor was tested for the digestion of bluegreen algae (Yu et al., 2014). This investigation evidenced 75-80% COD removal efficiency when operating at OLR of 1.5 g COD L-1 d-1 and HRT of 5 d. Most of the influent COD was removed in the first compartments by hydrolysis-acidification whereas methanogenesis occurred in the latter compartments. Likewise, the biogas composition was rich in methane (60%) in the final compartments while the first ones presented 45% methane content. In this sense, the configuration of this reactor mediated different bacterial population to dominate in each compartment which ultimately resulted in an efficient reactor performance. Under these configurations supporting spatiotemporal reactions separation, the two stage anaerobic digestion could also be a potential option for improving methane yields by employing extended SRT to accommodate the cultivation of slow growing methanogenic archaeas. Even though this configuration has been tested for other substrates, no remarkable data is available in the context of microalgae digestion.

Overall, more research should focus on semicontinuos digestion mode of microalgae for methane production purposes. Until now, only few investigations have tested the performance of different reactor configurations probably due to the low methane yields attained. In this sense, a big research effort was focused on elucidating the most promising pretreatment methods to increase these values. At this point, the performance of different reactor configuration should be reevaluated by feeding pretreated biomass. With the knowledge

15

gained on biomass pretreatments, the options of using novel configurations rather than the traditional CSTRs should be reassessed.

3.2. Anaerobic digestion temperature

The temperature at which the digestion is conducted can have two effects, namely the enhancement of the enzymatic activity of degrading microorganisms and the diminishment of photosynthetic activity of the microalgae when using glass lab-scale reactors. In this manner, microalgae are optimally growing under mesophilic conditions (25–35°C) while under thermophilic conditions (55°C) their activity is reduced. Most of the investigation conducted until now was at mesophilic range and thus, the survival of microalgae under these temperatures should be handled with care. Methanogenic microorganisms are extremely sensitive to temperature changes. There are three temperature ranges at which AD might take place: (i) psychrophilic (5-20°C); (ii) mesophilic (25-45°C); and (iii) thermophilic (45-65°C). When mesophilic range is compared to thermophilic digestion, this latter range enhances the efficiency of enzymatic hydrolysis and the growth rate of methanogens, thereby HRT can be reduced. Nevertheless, the disadvantage of this higher temperature range is related to the high ammonia and VFA levels that are associated with process failure due to possible inhibition of methanogens by these compounds. This is the case for instance of Kinnunen et al., (2014) who compared the anaerobic degradation of Nannochloropsis sp. at thermophilic and mesophilic range. Their study evidenced 48% more methane in continuous thermophilic digestion than at mesophilic range. Nevertheless, this was only true when working at low OLR whereas increasing OLR to 2 kg VS m-3 d-1 resulted in process inhibition by high ammonia concentration. Likewise, the

16

digestion of Scenedesmus obliquus in thermophilic range increased the methane yield by 24% when compared to the digestion in mesophilic range (Zamalloa et al., 2012a). Opposite, this latter investigation also reported no enhancement when digesting the marine alga Phaeodactylum tricornutum in these two temperature ranges. Authors attributed this fact to the osmotic shock occurring when digesting this microalgae, regardless the digestion temperature. It can be concluded that the benefits of using thermophilic digestion over mesophilic digestion is specie specific. Overall, thermophilic digesters can help decreasing retention time and enhancing gas production, but they need high heat input and are more sensitive to inhibitors and disturbances.

4-

Microbial anaerobic community

Very limited information is available on microbial communities of anaerobic digesters fed with microalgae. It is true that this substrate is quite recent but this info would bring some light on the performance optimization of these digesters. AD is a well consolidated bioprocess; however its operation is mostly based on process parameters while the analysis of the anaerobic microbiome is underestimated. There are two main key players in AD, namely bacteria and archaea (Fig. 1). Bacteria degrade organic matter into VFAs, carbon dioxide and hydrogen while archaea is in charge of the conversion of bacterial products into methane and carbon dioxide (biogas). Changing HRT, digestion temperature, OLR, addition of micronutrients or codigestion of a mixture of substrates are some strategies to counterbalance the negative effects of common inhibitors. When anaerobic digestion is inhibited by an unknown inhibitor (Mendez et al., 2015a), the determination of the anaerobic population could be used as a microbial indicator to elucidate the hampered digestion stage.

17

One of the main reasons for the lack of information regarding the anaerobic microbiome of digesters fed with microalgae is related to the inappropriate separation of nucleic acids of eukaryotic and prokaryotic microorganisms. DNA co-amplification may preclude the analysis (Bakke et al., 2011). Most of the investigations concerning the microbial community of anaerobic digesters are performed by using PCR-DGGE (polymerase chain reactiondenaturing gradient gel electrophoresis). Alternative molecular techniques such as pyrosequencing have opened up new possibilities providing a more complete inventory of the constituents of the microbial populations. Nevertheless, when both techniques are compared similar results are obtained and therefore DGGE – based analysis was pointed out to provide robust and cost effective results (Delgado et al., 2013).

Out of the little information available in literature, most of the investigations are dealing with the archaeas domain. This fact is quite surprising since this domain would be relevant if methanogenesis would be the rate limiting step while in the case of microalgae as substrates for AD, the hydrolysis stage is the bottleneck for an efficient methane production. In the context of archaea, Methanosaeta and Methanosarcina are acetate utilizing methanogens. While the first one is an obligated acetotroph, the latter one is able to produce methane through three metabolic pathways, namely acetoclastic, carbon dioxide reduction with hydrogen (hydrogenotrophic) and formate and methylotrophic.

Methanosaeta in the order Methanosarcinales was the dominant archaea identified in microalgae fed digesters. This archaea was identified when the reactor was operated at mesophilic conditions (Ellis et al., 2012). Unfortunately, this investigation did not mention the microalgae strains digested since it was collected from a wastewater lagoon and probably a mixture of robust microalgae were present in this ecosystem. Interestingly, this investigation 18

also highlighted the importance of using the proper primers. Despite of the used primers, 14% of the analyzed sequences could not be assigned taxonomically (Ellis et al., 2012). In accordance to this investigation, Zamalloa et al., (2012b) also reported the dominance of Methanosaeta sp. when digesting Phaeodactylum tricornutum at mesophilic conditions in an anaerobic membrane bioreactor. Together with Methanosaeta sp., Methanoregula, Methanoculleus and Thermogymnomonas were identified in an anaerobic baffled reactor fed with blue-green algae operated at 30ºC (Yu et al., 2014). When digesting Tetraselmis sp. (halophytic microalgae), the archaea exhibiting higher similarities (95%) with BLAST database included Methanospirillum stamsii and Methanogenium marinum (Ward et al., 2015a). At this point it should be stressed that even though Methanosaeta sp. was the predominant archaea, the presence of other species cannot be neglected. As a matter of fact, archaea are more dependent on environmental conditions than anaerobic bacteria in charge of acidogenesis and hydrolysis stages of the AD process. In this sense, the different digestion conditions including substrate characteristics, salinity or reactor configuration mediated the presence of some other archaea together with Methanosaeta sp.

The hydrogenotrophic pathway was reported to be the main methane producer within a microbial community digesting Spirulina sp. at extreme alkaline conditions (pH 10, NollaArdèvol et al., 2015). In this case, the methanogenic community was dominated by Methanocalculus sp. This hydrogenotrophic methanogen is able to use carbon dioxide and formate for the production of methane. This archaea played a key role in the AD by providing atypically high methane content in the biogas (92-96%, Nolla-Ardèvol et al., 2015). At this point, it should be noticed that this investigation reported really high methane content but at the expenses of a really low biogas production which ultimately resulted in low anaerobic

19

biodegradability (2-11%). Moreover, the authors claimed an accumulation of VFA due to the lack of acetoclastic methanogens. This fact mediated a markedly inefficient AD.

Forcing the microbial systems towards the domination of hydrogenotrophic methanogens can also be positive in the digestion of microalgae biomass. Microalgae biomass is typically described as a protein rich substrate. During AD, proteins are degraded into aminoacids and further converted into ammonium in a process called nitrogen mineralization. High nitrogen mineralization can mediate anaerobes inhibition by high ammonium/ammonia concentration (Mahdy et al., 2015c). Archaea have been pointed out as the anaerobic microorganisms most affected by these compounds (Yenigün and Demirel, 2013). More specifically, AD of protein rich substrates relies on syntrophic acetate oxidation as the dominant acetate-consuming process due to the inhibition of acetoclastic methanogenesis (Fotidis et al., 2013a). One potential strategy to tackle with ammonium inhibition when digesting protein rich substrates is the addition of specific microorganisms to this biological system, the so-called bioaugmentation. With the aim of coping with ammonium inhibition (5 g NH4+ L-1), anaerobic sludge bioaugmentation with Methanoculleus bourgensis resulted in 31.3% increase in methane production. This enhancement was mainly attributed to the hydrogenotrophic methanogens developed in the mesophilic reactor which in fact revealed a 5-fold increase in relative abundance of Methanoculleus sp. after bioaugmentation of a CSTR (Fotidis et al., 2013b). At this point it should be stressed out that this bioaugmentation does not always end successfully. This is the case for instance of a UASB reactor operated under mesophilic conditions where the anaerobic sludge was bioaugmented with Clostridium ultunense in association with Methanoculleus spp. (Fotidis et al., 2013b). The failure of this bioaugmentation was caused by the slow growth rate of the microorganisms added. Therefore, the main challenge lies in the appropriate selection of the microorganisms to avoid the wash

20

out from the reactor. Unfortunately, this strategy has not been applied yet to anaerobic digesters fed with microalgae however, once identified the main microalgae polymer hampering the hydrolytic stage, sludge bioaugmentation could be a powerful tool to enhance methane yields.

With regard to anaerobic bacteria, the main phyla identified include Proteobacteria, Firmicutes, Bacteriodetes and Chloroflexi (Carballa et al., 2015). The predominance of one or other phylum is mainly depending on the digested substrate, temperature, organic loading rate and unexpected chemicals concurrent with the substrate fed. The metabolic ability of Chloroflexi involves the degradation of carbohydrates while that of Proteobacteria and Bacteriodetes is more related to the degradation of proteins. Concerning specifically the anaerobic bacteria identified in methane producer digesters, different DGGE profiles were attained depending on the microalgal biomass digested. As previously mentioned for archaeas; amplified bacterial 16S rDNA also provided an important percentage of uncultured bacteria with no species-level information. When Chlorella vulgaris was digested, the matches corresponded to Petrimonas spp. and Bilophila wadsworthia while when digesting Dunaliella tertiolecta, the cultures identified corresponded to Wolinella succinogenes, Oceanibulbus indolifex and Syntrophobacter sp. (Lakaniemi et al., 2011). Another investigation dealing with the bacterial community in anaerobic fermenters for VFAs production using a microalgae mixture as substrates (Desmodesmus sp., Scenedesmus sp. and Chlamydomonas sp.) revealed a decreased bacterial diversity at increasing digestion temperatures (Cho et al., 2015). In addition to that, digestion at 35ºC exhibited a great population of Pseudomonas sp. and Proteobacterium while at 55ºC, the population shifted mainly to Bacillus pumilus and Proteobacterium.

21

Opposite to the microbial communities normally identified at neutral pHs, the relative abundance of these microbial groups changed towards a prevailing dominance of Bacteroidetes and Halanaerobiales in alkaline digestion systems (Nolla-Ardèvol et al., 2015). This change could be attributed to the fact that the anaerobic inoculum used in this investigation was collected from soda lake sediments. Even though an adapted inoculum to alkaline conditions was used, the microbial population was not really suitable for an efficient AD. The anaerobic sludge used as inoculum in most of the anaerobic digester devoted to microalgae biomass digestion is mostly collected in wastewater treatment plants. In this sense, the microbial population of this seed is adapted to digest sewage sludge rather than microalgae biomass. Bioaugmentation of anaerobic sludge with microorganisms particularly suitable for the digestion of specific algae may be a potential strategy to improve the AD of these substrates. In the case of microalgae in which the hydrolytic stage has been pointed out repeatedly as the main bottleneck, the development of a stronger hydrolytic anaerobic consortium may be of paramount importance. As mentioned above, bioaugmentation does not always result in successful digestion and therefore natural acclimation of anaerobic microorganisms to the digested substrate may also provide an alternative to enhance methane yields. The main drawback would be the time required for natural acclimation while the bioaugmentation would shorten this time by directly adding the appropriate microorganisms.

Overall, little information related to the anaerobic microbiome digesting microalgae is available. Research should focus in understanding the relationship between the anaerobic population and the anaerobic process. Process disturbances caused by operational or environmental perturbation results in microbial imbalances. By controlling the anaerobic population, valuable insights into the complex processes occurring within the digester could be predicted. This could be used as a microbial indicator to detect digester instability and

22

facilitate digester optimization rather than the traditional procedure of recovering deteriorated systems (Carballa et al., 2015).

5-

New opportunities for AD integration with microalgae processes

5.1. Opportunities for implementing AD in microalgae based processes

The integration of AD with microalgae processes can lead to several configurations as shown in Fig. 2. For instance, CO2 addition to microalgae cultures may enhance significantly the growth yields as the cells are placed in non-limiting carbon conditions. Assuming an average carbon content of 47% in the dry biomass, 1.72 g of CO2 is required stoichiometrically to obtain each gram of biomass. However, carbon supply represents a significant part of the production costs. CO2 rich flue gas from biogas energetic valorisation can be used as inorganic carbon substrate for the microalgae and also the biogas can be injected directly in the culture. In this last case, desorption of oxygen can be an issue for further energetic valorization. Indeed, these authors proposed a two-steps process in order to keep low level of oxygen in the biogas. Low cost CO2 availability and spatial closeness to the microalgae cultivation system are key conditions from an economical point of view. The optimal energetic valorisation of biogas requires the elimination of trace compounds such as hydrogen sulfide, siloxane, ammonia and organic volatiles molecules. This process favours the relative good quality of the exhausted gas produced after combustion compared with other industrial activities (Van den Hende et al., 2012) and make it suitable as inorganic carbon source for microalgae and cyanobacteria. In particular for large scale microalgae cultivation, integration of AD is an opportunity to provide nutrients, inorganic carbon to the cells and energy as heat

23

and electricity to the cultivation and downstream processes (Fig. 2-A) with economic and environmental savings.

Likewise, in addition to CO2 supply, some other nutrients such as nitrogen and phosphorus are also required during microalgae cultivation. As pointed out by Lardon et al. (2009), industrial fertilisers from fossil fuels present a major environmental impact in the perspective of mass production of microalgae. The authors proposed the use of digestate to substitute the need for macro and microelements. It is worth to mention that at industrial scale, the composition of digestate makes it suitable as a fertiliser for agriculture (Möller and Müller, 2012). The digestate can be originated from an external activity to the microalgal process (Franchino et al., 2013) or the co-digestion of different types of organic matter (Bjornsson et al., 2013) (Fig. 2-B).

The dark fermentation of organic matter can be considered as a “simplified” AD for which the low pH condition inhibits the methanogenic activity. The biogas is then composed by hydrogen and carbon dioxide while VFAs are accumulated in the digestate. In such medium, these organics compounds can be successfully used as carbon source for heterotrophic growth of microalgae (Fig. 2-C, Mohan and Devi, 2012).

Fig. 2 near here

Depending on the market opportunities of the microalgae grown in digestate, the sanitary quality of the digestate has obviously to be taken into consideration (Markou et al., 2014).

24

Heavy metals, pathogens or trace organic compounds can be found in anaerobic digesters in various concentrations depending on the organic wastes quality and operating conditions. Therefore, an efficient sanitation of the digestate or the restriction of the use to biofuels or any non-nutritional application is essential unless product safety can be guaranteed.

Some other integration opportunities include the conversion of organic nitrogen into soluble ammonium for the production of sustainable mineral ammonia fertilisers (Razon, 2012, Fig. 2-D). This approach is highly dependent on the digested microalgae which should be a protein rich substrate. In this sense, nitrogen-fixing cyanobacteria are probably more suitable than microalgae. Then, the harvested biomass can be converted into biogas and the digestate transformed into ammonium sulfate by a chemical process. Another approach entails the fact that chemical energy can be converted in electric energy by the catalytic activity of microorganisms in a microbial fuel cell device (Fig. 2-E, De Schamphelaire and Verstraete 2009). Before the introduction of the culture, the anode of a microbial fuel cell was able to consume part of the organic matter contained in the digestate, while the culture is recirculated in the cathode chamber in order to provide oxygen as electron acceptor. The process studied at lab scale couldn’t work continuously due to the bad performances of the digester.

Polyhydroxyalkanoates (PHA) define a class of polymers that are synthetized by numerous bacterial species. These molecules are precursors for the production of biodegradable plastics and therefore meet a high environmental concern and an economic opportunity. Van der Hende (2012) has proposed a potential strategy of producing PHA by using the capability of microalgae to substitute carbon dioxide in biogas by oxygen (Fig. 2-F). In its conceptual process, biogas is provided to a microalgal culture and the methane enriched exhausted gas act as a carbon substrate for methane oxidizing bacteria (PHA producers). 25

Finally, hydrothermal liquefaction (HTL) is a process able that can convert microalgae into crude oil with a calorific value close to fossil oil. The reaction produces also an aqueous phase with high content of organic matter and nutrients and a gas phase mainly composed by CO2 and H2. As oil extraction from microalgae is a key economic challenge, this technology has been intensively evaluated as an alternative to the conventional solvent extraction processes. Indeed, most of the methods for lipid recovery require dry extraction while water elimination from the harvested microalgae is a major environmental and energetic cost. In the case of HTL, this process can be performed with wet biomass. HTL requires high pressure (10-25 MPa) and temperature (280 to 370°C). When used with microalgae, the nutrient rich aqueous phase can be recycled as fertiliser and the gas phase as carbon source. Tommaso et al. (2015) have evaluated the combination of both technologies (HTL+AD) for energetic conversion of a mixed-culture algal biomass (Fig. 2-G). The process appears to be feasible, but the conversion rates of organic matter into biogas remained low.

5.2. The carboxylate platform

Carboxylates are intermediate chemicals obtained during AD. By inhibiting methanogens, organic matter is converted to VFAs and not further gasified into biogas. This strategy of using methane-inhibited anaerobic digester entails major benefits such as the lack of sensitive methanogens and the avoidance of carbon dioxide production which ultimately results in greater carbon recovery for end-products. VFAs are produced during the first three steps of AD: hydrolysis, acidogenesis and acetogenesis (Fig. 1). Products of these steps include amino acids, sugars, alcohols, hydrogen, carbon dioxide and VFAs. The most common VFAs encountered during AD are acetate, propionate and butyrate while also caproate and valerate can be present. The prevailing VFA depends on the type of bacterium that dominates the 26

fermentation, the substrate and the operational conditions of the fermentation (Wang et al., 2014; Cysneiros et al., 2012). Likewise, this VFA prevalence rules the end-product that could be targeted for an efficient production. For instance, when the substrate is rich in proteins, such is the case of many microalgae; propionate and butyrate are mainly formed (Nagase and Matsuo, 1982).

These VFAs can be chemically or biologically transformed into a wide variety of building blocks for the production of fuels or chemicals. While the chemical route requires large amounts of chemicals and energy inputs, the biological route entails mild conditions. With regard to polymers, PHAs for the production of bioplastics was produced by combining anaerobic acidogen fermentation and aerobic VFA conversion (Chen et al., 2013). This second conversion was performed by activated sludge to obtain mixed cultures able to store PHAs at high rates and yields. Moreover, the composition of VFAs determines the composition of the main products. For instance, acetate and butyrate usually result in formation of hydroxybutyrate monomers whereas propionate leads to increased concentration of hydroxyvalerate monomers (Bengtsson et al., 2008). Similarly, feeding VFA mixed liquor to different bacterial strains has also been proven successful in bioflocculant production (Zhao et al., 2012).

In the context of energy carriers, conversion of carboxylates into their corresponding alcohols and further conversion into alkanes via reduction has also been pointed out as a potential approach to get alternative products through AD (Agler et al., 2011). Likewise, VFAs also serve as potential substrates for the production of hydrogen via photofermentation by mix microbial cultures (Singhania et al., 2013) or carbon source of oleaginous microorganisms.

27

For instance, Fontanille et al., (2012) reported similar lipids composition to that of vegetable oils when VFAs were fed to Yarrowia lipolytica.

All these are potential strategies that could be followed instead of producing biogas during AD; however the process gets more complicated. Opposite to biogas that just flows out the culture broth, when producing something different than a gas, a separation of the targeted product from the medium can be challenging. In most of the cases, since biological systems are used, the separation of VFAs out of the culture is required to avoid any other byproduct that could hamper the end-product generation. According to Agler et al., (2011), this product separation together with the ceased activity of methanogens is the main barriers for largescale liquid fuel and chemicals production using the carboxylate platform. Since chemicals with higher carbon chain are generally energetically superior and easier to separate, the elongation of VFAs is regarded as a promising approach. As outlined by Spirito et al., (2014), there are three microbial pathways to elongate the carbon chain, namely homoacetogenesis (combination of two carbon dioxide molecules into acetate), succinate formation to elongate glycerol and reverse β-oxidation to elongate short chain carboxylates with two carbons. Overall, the production of alternative products entails at least two-stage bioconversion processes. Even though this adds complexity to the systems, it also favors high product yields and quality while it also increases the economical revenue.

6. Conclusions

Microalgae AD is a promising technology for sustainable energy production. A detailed study of microalgae cell-wall composition and low-cost alternatives to degrade it are essential to 28

increase the anaerobic hydrolysis rate. A deeper knowledge on microbial population, especially on hydrolytic bacteria, is of major importance to control digester instabilities. Research is needed in continuous operation focusing on: 1) novel reactor designs that ensure low HRT and high OLR and 2) new approaches for integrating microalgae culture and AD. In addition to methane, it should be kept in mind that a wide range of fermentation products can be obtained via AD.

Acknowledgements Authors want to thank the Spanish Ministry of Economy and Competitiveness for financial support to this project (WW-ALGAS, ENE2013-45416-R and RYC-2014-16823). Beatriz Molinuevo acknowledges the People Programme (Marie Curie Actions) of the European Union's Seventh Framework Programme (FP7/2007-2013) under REA grant agreement n° 291803.

References 1. Agler, M. T., Wrenn, B. A., Zinder, S. H., Angenent, L. T., 2011. Waste to bioproduct conversion with undefined mixed cultures: the carboxylate platform. Trends. Biotechnol. 29, 70-78. 2. Bakke, I., De Schryver, P., Boon, N., Vadsteina, O., 2011. PCR-based community structure studies of Bacteria associated with eukaryotic organisms: A simple PCR strategy to avoid co amplification of eukaryotic DNA. J. Microbiol. Methods 84, 349-351. 3. Bengtsson, S., Hallquist, J., Werker, A., Welander, T., 2008. Acidogenic fermentation of industrial wastewaters: effects of chemostat retention time and pH on volatile fatty acids production. Biochem. Eng. J., 40, 492-499. 29

4. Bjornsson, W. J., Nicol, R. W., Dickinson, K. E., McGinn, P. J., 2013. Anaerobic digestates are useful nutrient sources for microalgae cultivation: functional coupling of energy and biomass production. J. Appl. Phycol. 25, 1523-1528. 5. Bohutskyi, P., Bouwer, E., 2013. Biogas production from algae and cyanobacteria through anaerobic digestion: a review, analysis, and research needs. In Advanced Biofuels and Bioproducts (pp. 873-975). Springer New York. 6. Carballa, M., Regueiro, L., Lema, J.M., 2015. Microbial management of anaerobic digestion: exploiting the microbiome-functionality nexus. Curr. Opin. Biotechnol. 33, 103111. 7. Chen, H., Meng, H. J., Nie, Z. C., Zhang, M. M., 2013. Polyhydroxyalkanoate production from fermented volatile fatty acids: effect of pH and feeding regimes. Bioresour. Technol. 128, 533-538. 8. Cho, H. U., Kim, Y. M., Choi, Y. N., Kim, H. G., Park, J. M., 2015. Influence of temperature on volatile fatty acid production and microbial community structure during anaerobic fermentation of microalgae. Bioresour. Technol. 191, 475-480. 9. Collet, P., Arnaud, H., Laurent, L., Monique, R., Romy-Alice, G., Steyer, J. P., 2011. Lifecycle assessment of microalgae culture coupled to biogas production. Bioresour. Technol. 102, 207-214. 10. Cysneiros, D., Banks, C. J., Heaven, S., Karatzas, K. A. G., 2012. The effect of pH control and ‘hydraulic flush’ on hydrolysis and Volatile Fatty Acids (VFA) production and profile in anaerobic leach bed reactors digesting a high solids content substrate. Bioresour. Technol. 123, 263-271. 11. Delgado, S., Rachid, C.T.C.C., Fernández, E., Rychlik, T., Alegría, A., Peixoto, R. S., et al., 2013. Diversity of thermophilic bacteria in raw, pasteurized and selectively-cultured milk, as assessed by culturing, PCR-DGGE and pyrosequencing. Food Microbiol. 36, 103111.

30

12. de Schamphelaire , L., Verstraete ,W., 2009. Revival of the biological sunlight-to-biogas energy conversion system. Biotechnol Bioeng. 103, 296-304. 13. Demuez, M., Mahdy, A., Tomas-Pejo, E., Gonzalez-Fernandez, C., Ballesteros, M., 2015. Enzymatic cell disruption of microalgae biomass in biorefinery processes. Biotechnol. Bioeng. 112, 1955-1966. 14. Ellis, J. T., Tramp, C., Sims, R. C., Miller, C. D., 2012. Characterization of a methanogenic community within an algal fed anaerobic digester. ISRN microbiology. 15. Fontanille, P., Kumar, V., Christophe, G., Nouaille, R., Larroche, C., 2012. Bioconversion of volatile fatty acids into lipids by the oleaginous yeast Yarrowia lipolytica. Bioresour. Technol. 114, 443-449. 16. Franchino, M., Comino, E., Bona, F., Riggio,V. A., 2013. Growth of three microalgae strains and nutrient removal from an agro-zootechnical digestate. Chemosphere, 92, 738-744. 17. Fotidis, I. A., Karakashev, D., Kotsopoulos, T. A., Martzopoulos, G. G., Angelidaki, I. 2013a. Effect of ammonium and acetate on methanogenic pathway and methanogenic community composition. FEMS Microbial. Ecol. 83, 38-48. 18. Fotidis, I.A., Karakashev, D., Angelidaki, I., 2013b. Bioaugmentation with an acetateoxidising consortium as a tool to tackle ammonia inhibition of anaerobic digestion. Bioresour. Technol. 146, 57-62. 19. Fu, C. C., Hung, T. C., Chen, J. Y., Su, C. H., Wu, W. T., 2010. Hydrolysis of microalgae cell walls for production of reducing sugar and lipid extraction. Bioresour. Technol. 101, 8750-8754 20. Gerken, H.G., Donohoe, B., Knoshaug, E.P., 2013. Enzymatic cell wall degradation of Chlorella vulgaris and other microalgae for biofuels production. Planta 237,239-253. 21. Golueke, C.G., Oswald, W.J., Gotaas, H.B., 1957. Anaerobic digestion of algae. Appl. Microbiol. 592, 47–55.

31

22. González-Fernández, C., Sialve, B., Bernet, N., Steyer, J. P., 2012. Impact of microalgae characteristics on their conversion to biofuel. Part II: Focus on biomethane production. Biofuels, Bioprod. Bioref. 6, 205–218. 23. González-Fernández, C., Sialve, B., Bernet, N., Steyer, J. P., 2013. Effect of organic loading rate on anaerobic digestion of thermally pretreated Scenedesmus sp. biomass. Bioresour. Technol. 129, 219-223. 24. Gonzalez-Fernandez, C., Timmers, R. A., Ruiz, B., Molinuevo-Salces, B., 2014. Ultrasound-enhanced biogas production from different substrates. In Production of Biofuels and Chemicals with Ultrasound, pp. 209-242. Springer Netherlands. 25. Kim, K. H., Choi, I. S., Kim, H. M., Wic, S. G., Bae, H., 2014. Bioethanol production from the nutrient stress-induced microalga Chlorella vulgaris by enzymatic hydrolysis and immobilized yeast fermentation. Bioresour. Technol. 153, 47-54. 26. Kinnunen, H.V., Koskinen, P.E.P., Rintala, J., 2014. Mesophilic and thermophilic anaerobic laboratory-scale digestion of Nannochloropsis microalga residues. Bioresour. Technol. 155, 314-322. 27. Lakaniemi, A. M., Hulatt, C. J., Thomas, D. N., Tuovinen, O. H., Puhakka, J. A. 2011. Biogenic hydrogen and methane production from Chlorella vulgaris and Dunaliella tertiolecta biomass. Biotechnol. Biofuels, 4, 34. 28. Lardon, L., Hélias, A., Sialve, B., Steyer, J. P., Bernard, O., 2009. Life-cycle assessment of biodiesel production from microalgae. Environ. Sci. Technol. 43, 6475-6481. 29. Mahdy, A., Mendez, L., Ballesteros, M., González-Fernández, C., 2014. Enhanced methane production of Chlorella vulgaris and Chlamydomonas reindhardtii by hydrolytic enzymes addition. Energ. Convers. Manage. 85, 551-557. 30. Mahdy, A., Mendez, L., Ballesteros, M., González-Fernández, C., 2015a. Algaculture integration in conventional wastewater treatment plants: anaerobic digestion comparison of primary and secondary sludge with microalgae biomass. Bioresour. Technol.12, 236-244.

32

31. Mahdy, A., Mendez, L., Tomás-Pejó, E., del Mar Morales, M., Ballesteros, M. and González-Fernández, C., 2015b. Influence of the enzymatic hydrolysis on the biochemical methane potential of Chlorella vulgaris and Scenedesmus sp. J. Chem. Technol. Biotechnol. DOI: 10.1002/jctb.4722 32. Mahdy, A., Mendez, L., Ballesteros, M., González-Fernández, C., 2015c. Protease pretreated Chlorella vulgaris biomass bioconversion to methane via semi-continuous anaerobic digestión. Fuel 158, 3-41. 33. Markou, G., 2012. Alteration of the biomass composition of Arthrospira (Spirulina) platensis under various amounts of limited phosphorus. Bioresour. Technol. 116, 533-535. 34. Markou, G., Angelidaki, I., Georgakakis, D., 2013. Carbohydrate-enriched cyanobacterial biomass as feedstock for bio-methane production through anaerobic digestion. Fuel 111, 872879. 35. Markou, G., Vandamme, D., Muylaert, K., 2014. Ammonia inhibition on Arthrospira platensis in relation to the initial biomass density and pH. Bioresour. Technol 166, 259-265. 36. Mendez, L., Mahdy, A., Timmers, R. A., Ballesteros, M., González-Fernández, C., 2013. Enhancing methane production of Chlorella vulgaris via thermochemical pretreatments. Bioresour. Technol. 149, 136-141. 37. Mendez, L., Mahdy, A., Ballesteros, M., González-Fernández, C., 2014. Effect of high pressure thermal pretreatment on Chlorella vulgaris biomass: Organic matter solubilisation and biochemical methane potential. Fuel 117, 674–679 38. Mendez, L., Mahdy, A., Ballesteros, M., González-Fernández, C., 2015a. Biomethane production using fresh and thermally pretreated Chlorella vulgaris biomass: a comparison of batch and semi-continuous feeding mode. Accepted Ecol. Eng. 39. Mendez, L., Mahdy, A., Ballesteros, M., González-Fernández, C., 2015b. Chlorella vulgaris vs cyanobacterial biomasses: Comparison in terms of biomass productivity and biogas yield. Energ. Convers. Manag. 92, 137-142.

33

40. Mohan, S. V., Devi, M. P., 2012. Fatty acid rich effluent from acidogenic biohydrogen reactor as substrate for lipid accumulation in heterotrophic microalgae with simultaneous treatment. Bioresor. Technol. 123, 627-635. 41. Möller, K., Müller, T., 2012. Effects of anaerobic digestion on digestate nutrient availability and crop growth: A review. Eng. Life Sci.12, 242-257. 42. Mottet, A., Habouzit, Steyer, J.P., 2014. Anaerobic digestion of marine micro-algae in different salinity levels, Bioresour. Technol. 158,300-306. 43. Nagase, M., Matsuo, T., 1982. Interactions between amino-acid-degrading bacteria and methanogenic bacteria in anaerobic digestion. Biotechnol. Bioeng. 24, 2227-2239. 44. Nolla-Ardèvol, V., Strous, M., Tegetmeyer, H. E., 2015. Anaerobic digestion of the microalga Spirulina at extreme alkaline conditions: biogas production, metagenome, and metatranscriptome. Frontiers in microbiology 6. 45. Ometto, F., Quiroga, G., Pšenicka, P., Whitton, R., Jefferson, B., Villa, R., 2014. Impacts of microalgae pre-treatments for improved anaerobic digestion: Thermal pretreatment, thermal hydrolysis, ultrasound and enzymatic hydrolysis. Water Res. 65, 350-361. 46. Park, S., Li, Y., 2012. Evaluation of methane production and macronutrient degradation in the anaerobic co-digestion of algae biomass residue and lipid waste. Bioresour. Technol. 11, 42-48. 47. Passos, F., Solé, M., García, J., Ferrer, I., 2013. Biogas production from microalgae grown in wastewater: Effect of microwave pretreatment. Appl. Energy108, 168-175. 48. Passos, F., Uggetti, E., Carrère, E., Ferrer, I., 2014. Pretreatment of microalgae to improve biogas production: A review. Bioresour. Technol. 172, 403-412. 49. Passos, F., Ferrer, I., 2015. Influence of hydrothermal pretreatment on microalgal biomass anaerobic digestion and bioenergy production. Water Res. 68, 364 -373.

34

50. Quinn, J.C., Hanif, A., Sharvelle, S., Bradley, T.H., 2014. Microalgae to biofuels: Life cycle impacts of methane production of anaerobically digested lipid extracted algae. Bioresour. Technol. 171, 37-43. 51. Ras, M., Lardon, L., Sialve, B., Bernet, N., Steyer, J.P., 2011. Experimental study on a coupled process of production and anaerobic digestion of Chlorella vulgaris. Bioresour. Technol. 102, 200–206. 52. Razon, L.F., 2012. Life cycle energy and greenhouse gas profile of a process for the production of ammonium sulfate from nitrogen-fixing photosynthetic cyanobacteria. Bioresour. Technol. 107, 339–346. 53. Samson, R., Leduy, A., 1982. Biogas production from anaerobic digestion of Spirulina maxima algal biomass. Biotechnol Bioeng. 24, 1919-1924. 54. Schwede, S., Rehman, Z. U., Gerber, M., Theiss, C., Span, R., 2013. Effects of thermal pretreatment on anaerobic digestion of Nannochloropsis salina biomass. Bioresour. Technol. 143, 505-511. 55. Sialve, B., Bernet, N., Bernard, O., 2009. Anaerobic digestion of microalgae as a necessary step to make microalgal biodiesel sustainable. Biotechnol. Adv. 27, 409-416. 56. Singhania, R.R., Patel, A.K., Christophe, G., Fontanille, P., Larroche, C., 2013. Biological upgrading of volatile fatty acids, key intermediates for the valorization of biowaste through dark anaerobic fermentation. Bioresour. Technol. 145, 166-174. 57. Spirito, C.M., Richter, H., Rabaey, K., Stams, A.J.M., Angenent, L.T., 2014. Chain elongation in anaerobic reactor microbiomes to recover resources from waste. Curr. Opin. Biotech. 27, 115-122. 58. Stronach, S.M., Rudd, T., Lester, J.N., 2012. Anaerobic digestion processes in industrial wastewater treatment. Vol. 2. Springer Science & Business Media.

35

59. Tartakovsky, B., Lebrun, F. M., McGinn, P. J., O'Leary, S. J. B., Guiot, S. R., 2013. Methane production from the microalga Scenedesmus sp. AMDD in a continuous anaerobic reactor. Algal Res. 2, 394-400. 60. Tartakovsky, B., Lebrun, F. M., Guiot, S. R., 2015. High-rate biomethane production from microalgal biomass in a UASB reactor. Algal Res. 7, 86-91. 61 Tommaso, G., Chen, W.-T., Li, P., 2015. Chemical characterization and anaerobic biodegradability of hydrothermal liquefaction aqueous products from mixed-culture wastewater algae, 178, 139–146. 62. Van Den Hende, S., Vervaeren, H., Boon, N., 2012. Flue gas compounds and microalgae: (Bio-)chemical interactions leading to biotechnological opportunities. Biotechnol. Adv. 30, 1405–1424. 63. Van Den Hende, S., Laurent, C., Bégué, M., 2015. Anaerobic digestion of microalgal bacterial flocs from a raceway pond treating aquaculture wastewater: Need for a biorefinery. Bioresour. Technol. 196, 184-193. 64. Ververis, C., Georghiou, K., Danielidis, D., Hatzinikolaou, D. G., Santas, P., Santas, R., et al., 2007. Cellulose, hemicelluloses, lignin and ash content of some organic materials and their suitability for use as paper pulp supplements. Bioresour. Technol. 98, 296-301. 65. Wang, K., Yin, J., Shen, D., Li, N., 2014. Anaerobic digestion of food waste for volatile fatty acids (VFAs) production with different types of inoculum: Effect of pH. Bioresour. Technol. 161, 395-401. 66. Ward, A., Ball, A., Lewis, D., 2015a. Halophytic microalgae as a feedstock for anaerobic digestion. Algal Res. 7, 16-23. 67. Ward, A., Lewis, D., 2015b. Pre-treatment options for halophytic microalgae and associated methane production. Bioresour. Technol. 177, 410-413. 68. Wieczorek, N., Kucuker, M. A., Kuchta, K., 2015. Microalgae-bacteria flocs (MaB-Flocs) as a substrate for fermentative biogas production. Bioresour. Technol. 194, 130-136.

36

69. Yenigün, O., Demirel, B., 2013. Ammonia inhibition in anaerobic digestion: a review. Process Biochem. 48, 901-911. 70. Yu, Y., Lu, X., Wu, Y., 2014. Performance of an anaerobic baffled filter reactor in the treatment of algae-laden water and the contribution of granular sludge. Water 6, 122-138. 71. Zamalloa, C., Boon, N., Verstraete, W., 2012a. Anaerobic digestibility of Scenedesmus obliquus and Phaeodactylum tricornutum under mesophilic and thermophilic conditions. Appl. Energy 92, 733-738. 72. Zamalloa, C., De Vrieze, J., Boon, N., Verstraete, W., 2012b. Anaerobic digestibility of marine microalgae Phaeodactylum tricornutum in a lab-scale anaerobic membrane bioreactor. Appl. Microbiol. Biotechnol. 93, 859-869. 73. Zhao, G., Ma, F., Wei, L., Chua, H., 2012. Using rice straw fermentation liquor to produce bioflocculants during an anaerobic dry fermentation process. Bioresour. Technol. 113, 83-88.

37

MICROALGAE HYDROLYSIS

Carbohydrates Cellulase, hemicellulase, xylanase, amylase

Clostridium Acetivibrio Staphylococcus Bacteroides

Proteins

Lipids

Protease

Lipase, phospholipase

Clostridium, Vibrio Peptococcus, Bacillus, Proteus, Bacteroides

Clostridium, Micrococcus, Staphylococcus

Amino acids, sugars

LCFA , alcohols Zymomonas

ACIDOGENESIS Lactobacillus Escherichia, Staphylococcus, Micrococcus, Bacillus, Pseudomonas,Desulfovibrio, Desulfuromonas, Desulfobacter, Selenomonas, Veillonella, Sarcina, Streptococcus

ACETOGENESIS

Clostridium Eubacterium Streptococcus

Clostridium Syntrophomomas

Intermediate products (VFA) Syntrophobacter Syntrophomomas

H2, CO2

Acetate Clostridium Methanotrix Methanosarcina Methanospirillum Methanosaeta

Methanobacterium, Methanocalculus Methanobervibacterium, Methanoplanus, Methanoregula Methanogenius, Methanoculleus

ACETOCLASTIC METHANOGENESIS

HYDROGENOTROPHIC METHANOGENESIS

CH4, CO2

Figure 1. Anaerobic degradation of microalgae and the respective genera of microorganisms involved in each stage. (Adapted from Stronach et al., 2015 and Gonzalez-Fernandez et al., 2014).

38

Digestate liquid phase Biomass

A

Microalgae

B

AD

C

Dark Fermantation

Digestate

D

N2 fixing Cyanobacteria

Biomass

Digestate

Organic Processing

Residu

AD

Organic fertilizer Energy Added value products

Microalgae

Biomass Added value products

Heterotrophic Microalgae

Biomass Added value products Digestate

AD

Stripping

(NH4)2SO4

Digestate

E

Microalgae

Biomass AD

Fuel Cell

Biogas Electricity

CH4+O2

CH4+CO2

F

AD

G

Microalgae

Microalgae

Methanotrophs

PHA

Aqueous

Biomass HTL

phase

AD

Biogas Crude oil

Figure 2. Examples of integration of anaerobic digestion with microalgal processes.

39