Methane potential and anaerobic treatment feasibility of Sargassum muticum

Bioresource Technology 189 (2015) 53–61

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Methane potential and anaerobic treatment feasibility of Sargassum muticum M. Soto a,⇑, M.A. Vázquez a, A. de Vega a, J.M. Vilariño b, G. Fernández b, M.E.S. de Vicente a a b

Dept. of Physical Chemistry and Chemical Engineering I, University of A Coruña, Rúa da Fraga n° 10, 15008 A Coruña, Galiza, Spain INVESGA, S.L. Rúa Perseo n° 9, 15179 Oleiros, A Coruña, Spain

h i g h l i g h t s  Methane potential of the algae S. muticum ranged from 166 to 208 mL CH4/gVS.  Alga grinding was not necessary for continuous digestion.  Accumulation of toxic compounds was not observed at least up to 80 gTS/L.  Up to 0.26 L CH4/L d were obtained at loading rates of 3.2 gTS/L d.  Accumulated non-biodegradable solids is the main efficiency limit of digesters.

a r t i c l e

i n f o

Article history: Received 14 February 2015 Received in revised form 13 March 2015 Accepted 14 March 2015 Available online 19 March 2015 Keywords: Sargassum muticum Methane potential Methanogenic toxicity Continuous treatment feasibility

a b s t r a c t The aim of this research was to study the feasibility of anaerobic digestion of the alga Sargassum muticum with special attention to its biodegradability, potential toxicity caused by its salt content, alga components and intermediate process compounds, and potential limitations to continuous treatment. Specific methane potential (SMP) for three samples of S. muticum collected from the Galician coast (Northwest Spain) at different seasons ranged from 166 to 208 mL CH4/gVS while accumulation of toxic compounds was not observed at alga concentrations of up to 100 gTS/L, except for one of the samples in which inhibition started at 80–100 gTS/L. Continuous digestion is feasible at alga concentration up to 100 gTS/L with methane production rates ranging from 0.14 to 0.26 L CH4/L d at organic loading rates of 3.2 gTS/L d, but SMP dropped to 113–159 mL CH4/gVS. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Today most of the naturally produced and harvested algal biomass is an unused resource, in some cases incorporated into compost and spread on fields as enriching agents, but it is mainly dumped or left stranded to decompose on the coasts creating environmental problems (Bruhn et al., 2011; Nielsen and Heiske, 2011). Anaerobic digestion may be economical for methane production from algal biomass generated either during wastewater treatment or harvested from eutrophic water bodies. Methane conversion of whole algae or combined biodiesel and biogas production produces the highest energy output (Bohutskyi et al., 2014). Furthermore, after anaerobic digestion, digested seaweed can be used as a fertilizer, soil conditioner, sorbent or in other environmental applications. ⇑ Corresponding author. Tel.: +34 981 167 050; fax: +34 981 167 065. E-mail address: [email protected] (M. Soto). http://dx.doi.org/10.1016/j.biortech.2015.03.074 0960-8524/Ó 2015 Elsevier Ltd. All rights reserved.

The genus Sargassum is widely distributed and common in the Indo-Pacific region with 335 species and many ethnobotanical uses (Marquez et al., 2014). Although Sargassum muticum is a naturally growing algal on the Asian coast, it is an aggressive invasive species on European coasts, and harvesting has been suggested as a control strategy. Its uses range from aquaculture to alginate, antioxidant production or environmental uses as heavy metal sorption (Lodeiro et al., 2004; González-López et al., 2012; Marquez et al., 2014). Biogas production was another potential use of S. muticum biomass as earlier reported by Bird et al. (1990) and Chynoweth (2005). Seaweed species as Sargassum spp., Turbinaria spp., Hydroclathrus spp., Caulerpa spp., and Ulva spp. allow sustainable biomass supply for methane fermentation in the Philippines, even in household biogas digesters of rural communities (Marquez et al., 2014). The anaerobic digestion of marine macroalgae biomass could meet two currently important needs, the mitigation of eutrophic effects and the production of renewable energy. Because of the

54

M. Soto et al. / Bioresource Technology 189 (2015) 53–61

abundance of seaweed biomass, harvesting and conversion can be highly desirable and convenient, chiefly for countries with long coastlines or eutrophic environments (Migliore et al., 2012). Langlois et al. (2012) performed a life cycle assessment for methane as a biofuel from the anaerobic digestion of seaweeds in European countries and concluded that seaweed could become competitive with terrestrial feedstock for biofuel production in the near future. The main benefits were obtained for greenhouse gas emissions, fossil fuel depletion, ozone depletion and marine eutrophication prevention as compared to natural gas as a fossil fuel reference. Continuous algae digestion was carried out in one step CSTR using algae as the sole substrate (Briand and Morand, 1997; Hinks et al., 2013) or by co-digestion (Nielsen and Heiske, 2011; Schwede et al., 2013a) with farm waste or other wastes mainly at mesophilic temperatures but also at psycrophilic and thermophilic temperatures (Zamalloa et al., 2012; Kinnunen et al., 2014). Two-step anaerobic digestion systems have also been experimented with aiming to optimize the digestion process through a first step of hydrolysis followed by a second step consisting of a high rate methanogenic digester (Vergara-Fernández et al., 2008; Nkemka and Murto, 2010). Several key obstacles to biogas production still remain, such as low biodegradability of algae biomass, potential toxicity caused by ammonia release, and potential toxicity caused by the presence of sodium and sulfate (or sulfide from sulfate reduction) for marine species (Briand and Morand, 1997; Sialve et al., 2009; Nkemka and Murto, 2010; Nielsen and Heiske, 2011; Bohutskyi et al., 2014; Kinnunen et al., 2014). Other inhibiting phenomena may be those caused by heavy metals, tannins, furanic and phenolic compounds and polyphenols (Nkemka and Murto, 2010; Jard et al., 2013; Monlau et al., 2014). Sodium is known to be the inhibitory component of salt in the anaerobic process, when inhibitory concentrations vary from 2.3 to 27 g/L (Feijoo et al., 1995), important adaptation and antagonism effects being reported. It was also shown that ammonia inhibition started at 1.3 gNH3-N/L for non-adapted sludge but at higher concentrations of 3–4 g/L for adapted sludge, and even up to 7 gNH3N/L are tolerated depending on the sludge pre-adaptation and pH (Soto et al., 1991). Furthermore, sulfur inhibition on methanogenic bacteria was reported for digestion of substrates having sea salt levels (Soto et al., 1991). Although ammonium and sulfur inhibition during anaerobic digestion of algae biomass was reported (Tartakovsky et al., 2013; Kinnunen et al., 2014), more concern exists about sodium toxicity (Lakaniemi et al., 2011; Schwede et al., 2013b; Santos et al., 2014). Previously published results reported a low specific methane potential (SMP) of Sargassum ssp. ranging from 120 to 190 mL CH4/gVSS (Bird et al., 1990; Chynoweth et al., 2001) while Jard et al. (2013) found the lowest methane potential (130 mL CH4/ gVS) for S. muticum among ten macroalgae studied. Low carbohydrates content, high content of insoluble fibers (which are difficult to degrade) and high levels of polyphenols (which are potential inhibitors in anaerobic digestion) were cited as the main reasons for low methane yield from Sargassum biomass. Available studies are scarce, and methane yields were usually obtained from batch assays, which could not discern inhibition situations (Jard et al., 2013). The objectives of the present work were (i) to obtain the specific methane potential for three samples of S. muticum collected from the Galician coast (Northwest Spain) at different seasons, (ii) to identify the conditions for absence of process inhibition, and (iii) to assess the operation conditions of continuous digesters treating S. muticum biomass. For these purposes, batch and semi-continuous assays at different alga concentrations and organic loading rates were used.

2. Methods 2.1. Algae samples Three samples of S. muticum (SM1, SM2 and SM3) were collected on the coast of A Coruña (Northwest Spain) at different seasons of the year. Algae samples were air dried (about 15% of moisture content) and conserved at ambient temperature. Before using them for anaerobic digestion experiments, algae samples were oven dried at 105 °C until they were a constant weight. Two of the algae samples were finely ground (SM1 and SM2, particle size below 3 mm), while the other (SM3) was only chopped to about 10 mm size. 2.2. Anaerobic batch assays for SMP determination Batch assays were carried out in glass bottles of 500 mL while the liquid assay volume was 450 mL. A plastic tube connected the assay bottle to an inverted Marriotte flask of 250 mL, which contained an alkaline solution (2.5% of NaOH), so CO2 was absorbed into the alkaline solution and the generated volume of CH4 was measured as the displaced liquid volume (Soto et al., 1993). 75 mL of an anaerobic sludge coming from a full scale digester treating sea fish canning wastewater was used as inoculum for all assays, being the concentration into the assays 2.7 gVSS/L. Two series of anaerobic assays with different initial algae concentration of 5 and 10 gTS/L as substrate were carried out. All assays were carried out in duplicate, including blank and methanogenic activity control assays. A volatile fatty acid (VFA) mixture (acetic acid, 2 g/ L; propionic, 0.5 g/L and n-butyric, 0.5 g/L) was used as substrate for the control assays, while blank assays contained inoculum but no substrate. 100 mg/L of Na2S9H2O and Na2CO3 at a ratio of 1 g/gTS of alga were added to each assay at the beginning, in order to obtain an anoxic medium and sufficient buffer capacity, respectively. Diluted HCl and NaOH solutions were used to regulate initial pH to the range of 7.0–7.1. Macro and micro nutrients were added only at the beginning of the assays (feed 1) in the amounts usually indicated for anaerobic digestion batch assays (i.e. 1 mL of each macro and micro nutrient stock solutions per L of assay medium, the composition detailed by Ferreiro and Soto, 2003), except for the methanogenic activity control assays, in which nutrients were added again in the 5th feed. SMP of each substrate sample was obtained from the final cumulative methane production after subtracting the blank value, and dividing by the amount of dry alga fed. 2.3. Semi-continuous anaerobic digestion experiments Through successive feeds of alga as substrate, batch assays were converted to semi-continuous digesters. This procedure aimed to provide operating conditions close to those of continuous treatment digesters, by increasing the effective concentration step by step. During feeds 2–4, dry algae substrate was added at the same amount of 5.0 gTS/L (C1) and 10 gTS/L (C2), while for feeds 5–7 the amounts added were 10 gST/L (C1+) and 20 gTS/L (C2+), as indicated in Table 1. Following this procedure, the average organic loading rate (OLR) reached during the second part of the study (feeds 5–7) was in the range of operation of semi-continuous and continuous reactors. Before adding the substrate of a new feed, the pH in the media was regulated to the range of 7.0–7.1 by adding diluted HCl solution. Semi-continuous algae digestion was evaluated through maximum and average volumetric methane production rate (MPR) as well as substrate (TSS and VSS) removal and VFA and soluble COD accumulation.

55

M. Soto et al. / Bioresource Technology 189 (2015) 53–61 Table 1 Characteristics of batch and semi-continuous anaerobic assays: TS and VS of alga added through several feeds and organic loading rates obtained. Added alga

Accumulated alga concentration

gTS/L

gTS/L

C1

C2

gVS/L

C1

C2

SM1

SM1

SM2

SM2

SM3

SM3

C1 5.0

C2 10.00

C1 3.31

C2 6.62

C1 2.38

C2 4.77

C1 3.31

C2 6.61

C1 6.62 9.92 13.23

C2 13.23 19.85 26.46

C1 4.77 7.15 9.53

C2 9.53 14.30 19.07

C1 6.61 9.92 13.22

C2 13.22 19.83 26.44

Semi-continuous experiments (increasing substrate concentration up to 100 gTS/L) C1+ C2+ C1+ C2+ C1+ Feed 5 10.0 20.0 30.0 60.0 19.85 Feed 6 10.0 20.0 40.0 80.0 26.46 Feed 7 10.0 20.0 50.0 100.0 33.08

C2+ 39.69 52.92 66.15

C1+ 14.30 19.07 23.84

C2+ 28.60 38.14 47.67

C1+ 19.83 26.44 33.06

C2+ 39.67 52.89 66.11

Batch assays for SMP determination C1 C2 Feed 1 5.0 10.0

Semi-continuous experiments (increasing substrate concentration up to 40 gTS/L) C1 C2 C1 C2 Feed 2 5.0 10.0 10.0 20.0 Feed 3 5.0 10.0 15.0 30.0 Feed 4 5.0 10.0 20.0 40.0

Average OLR (gVS/L d) Period (d) 30 27 20

Feed 1 Feeds 2–4 Feeds 5–7

SM1-C1 0.11 0.39 1.05

SM1-C2 0.22 0.79 2.11

SM2-C1 0.08 0.28 0.76

SM2-C2 0.16 0.57 1.52



Specific methanogenic activity (SMA) was determined for the successive feeds from the slope of cumulative methane production curves during the exponential phase of methane production after each feed (Soto et al., 1993). Although maximum SMA values cannot be obtained for complex substrates as alga, this procedure provides SMA values at similar readily biodegradable substrate concentrations (added at each feed and removed by digestion before the following feed) but at increasing concentrations of other substrate components such as salts and refractory organic components. In order to obtain an indicator of the potential inhibitory effect of accumulated concentrations of alga components, SMA values obtained at each feed assays were correlated to the overall amount of methane produced. Following a theoretical foundation of this approach to be applied in successive feed batch assays is provided. Anaerobic inoculums are usually characterized by their SMA and VSS concentration. However, their methanogenic activity is due to the presence of active methanogenic microorganisms (concentration Xpm) which constitutes only a small fraction of the overall VSS of the considered sludge. From this concept the following expression applies:

ð1Þ

where Xpm is the concentration of pure methanogenic microorganisms and SMApm their theoretical specific methanogenic activity. On the other hand, the biomass to product yield (Yxp) is defined as the ratio of microorganisms growth rate (rX) and product generation rate (rP). For batch operation with initial concentration of pure microorganisms X°pm and considering Yxp to be constant, Yxp can be expressed as:

Y xp ¼ rX =r P ¼ ðX pm  X



pm Þ=ðP



P Þ

Then, after a time of batch digestion, Xpm can be related to the amount of product P (methane) generated (initial amount P° = 0):

X pm ¼ X



pm

þ Y xp  P

ð2Þ

Combining Eqs. (1) and (2), we obtain:

MPR ¼ SMApm  X



pm

þ SMApm  Y xp  P

SM3-C2 0.22 0.79 2.11

and SMA referred to the initial VSS inoculum concentration will be:

2.4. Methanogenic toxicity assessment

MPR ¼ SMA  ½VSS ¼ SMApm  X pm

SM3-C1 0.11 0.39 1.05

SMA ¼ ðSMApm =½VSS Þ  X





pm

þ ðSMApm =½VSS Þ  Y xp  P

ð3Þ

In Eq. (3) SMApm and Yxp can be considered to be constants for a defined system (i.e. for an inoculum, substrate and defined conditions) as well as [VSS]° and X°pm. Thus, this equation can be expressed as:

SMA ¼ a þ b  P

ð4Þ

stating that the specific methanogenic activity (referred to the initial VSS inoculum concentration) must increase in a linear fashion with the amount of methane (P) generated. For applying Eq. (4) to successive feed batch assays, SMP must be measured during zero order substrate conditions (S  Ks), or alternately at the same substrate conditions, while all the methane generated in each feed must be accounted for P. So, as in conventional methanogenic activity test, SMP is obtained from the maximum slope of cumulative methane generation curves at each feed, while P is the total methane production accumulated since the beginning of the corresponding feed. In these conditions, a linear relationship between SMP (or MPR) and P indicate that the microorganisms initially present in the inoculum maintain a sustainable growth and inhibitory conditions are absent. On the other hand, inhibitory or toxic conditions to methanogenic microorganisms will lead to a reduction of the slope b (Eq. (4)) or even to lower SMP values instead of a higher P value. 2.5. Analysis and monitoring Methane production was determined daily or several times a day as indicated above. From feeds 2–7, pH in the digestion media was determined in the period between 1 and 3 days after the feed looking for possible media acidification. In that period, digestion media samples were obtained in order to analyze the concentration of VFA. Total and volatile suspended solids (TSS, VSS) and soluble chemical oxygen demand (CODs) were determined at the end of feeds 4 and 7. Analyses were carried out as proposed by Standard Methods, 1995. For the determination of TSS and VSS, glass fiber filters (Whatman, GF/C, 4.7 cm diameter, 1.2 lm of pore size) were used. COD was analyzed by the semi-micro determination method,

56

M. Soto et al. / Bioresource Technology 189 (2015) 53–61

titrating the excess of a standard potassium dichromate solution with a standard ferrous ammonium sulfate (FAS). In order to avoid chloride interference in COD results, HgSO4 was dosed as indicated by Soto et al. (1989). An EA1108 elemental analyzer (Carlo Erba Instruments) equipped with an AS200 autosampler was used for nitrogen, carbon and organic carbon determination. For nitrogen and carbon analysis samples were weighed in tin capsules. For organic carbon analysis, samples were weighed in silver capsules and carbonate was removed by addition of HCl before the analysis. Samples were oxidized by combustion at 1020 °C using an oxidation catalyst containing Cr2O3 and AgCO3O4. After combustion, the resulting gases passed through a reduction reactor containing elemental copper at 650 °C. After removal of water, the evolved N2 and CO2 were separated by gas chromatography and detected by thermal conductivity. Metal concentrations were determined using inductively coupled plasma mass spectrometry (ICP-MS Element XR or Element2 from Thermo Electron). Solid samples were air-dried and the solids were finely shredded and mixed. 0.5 g of solids sample was subjected to acid digestion (10 mL HNO3 cc + 1 mL H2O2 cc) in a closed microwave apparatus (ETHOS PLUS de Millestone). The oven temperature was programmed to reach 175 °C in 5.5 min, being held for 4.5 min. After digestion, samples were replenished with MilliQ deionized water to 50 mL and then filtered through a Millex HN de 0.45 lm filters. 3. Results and discussion

3.2. Assay conditions and evolution

3.1. Characteristics of S. muticum samples The characteristics of the three samples of S. muticum used in the present study are shown in Table 2. SM2 presented the lower content in organic matter and carbon (48%VS in SM2 versus 66%VS in SM1 and SM3, and 24%C in SM2 versus 25–28% in SM1 and SM3), but a higher content in nitrogen (2.2%N in SM2 versus 1.7–1.8 in SM1 and SM3) and also a higher content in some metals such as Na and Cd. However, the COD content was higher in SM2 (0.56 mgCOD/gTS) than in SM2 and SM3 (0.52–0.53 mgCOD/gTS), Table 2 Characteristics of the three samples of S. muticum (SM1, SM2, SM3) used in the present study.

VSS (%TS) COD (g/gTS) Electrical conductivity (mS/cm) pH Na (%TS) Mg (%TS) P (%TS) Ca (%TS) K (%TS) N (%TS) C (%TS) COT (%TS) C/N(COT) Al (mg/kg) Fe (mg/kg) Cd (mg/kg) Pb (mg/kg) Hg (mg/kg) Cr (mg/kg) Co (mg/kg) Mo (mg/kg) Mn (mg/kg) Ni (mg/kg) Cu (mg/kg) Zn (mg/kg)

the difference being higher for the VS based COD, which resulted in 1.17 mgCOD/gVS for SM2 versus 0.77–0.79 mgCOD/gVS for SM1 and SM3. Then, although SM2 presented the lowest VS content, the organic matter present in SM2 had a higher energy content (COD content) which may be due to the higher protein content and probably to a high fat content (not analyzed in this study). SM2 also showed higher electrical conductivity, higher pH and lower C/N ratio. Differences between SM1 and SM3 are related to the amounts of some metals and particularly to the amount of organic carbon, higher in SM3 (26%TS) in comparison to both SM1 and SM2 (20%TS). C/N ratio was therefore higher for SM3 (16) than for SM1 (11) and SM2 (9). These results show that both the salt content and the organic composition of S. muticum varied extensively dependant on the season of harvesting. The characteristics of the alga samples used are in the ranges of reported values for brown algae and Sargassum species. Composition of some Sargassum spp. was reported to be in the range of 0.8–1.6% lipids, 40–66% carbohydrates, 10–13% proteins (Monlau et al., 2014; Marquez et al., 2014). Sargassum spp. also contains about 10% fibers, 26% ash, 17–35% alginic acid and variable amounts of mannitol, fucose and xylose (Marquez et al., 2014). Elemental composition of different Sargassum spp. may vary extensively, with reported values of 12–40%C, 0.6–2.0%N, or C/N ratio from 12 to 22 (Marquez et al., 2014). For S. muticum, Jard et al. (2013) reported 531 g/kgTS of fiber and 1.98 g/kgTS of polyphenols. S. muticum is also characterized by a low content of total sugars (166 g/kgTS), total alginates, proteins (84 g/kgTS), low nitrogen content and a higher C/N ratio of 20.

SM1

SM2

SM3

66.15 0.512 33.59 6.89 4.04 1.54 0.34 5.13 8.9 1.82 25.48 20.40 11.2 754 459 1.26 2.31 <0.03 1.54 0.93 0.72 75 2.18 2.06 73.2

47.67 0.559 42.95 8.92 4.54 1.87 0.42 5.83 10.6 2.16 23.71 19.88 9.2 777 724 1.51 2.7 <0.03 1.41 8.23 0.64 96.9 8.23 2.5 85.5

66.11 0.524 37.67 6.31 2.95 2.65 0.31 2.44 8.07 1.69 28.17 26.49 15.7 746 544 0.86 1.4 <0.03 0.75 1.97 0.32 28.1 1.97 1.74 14.3

Batch tests were often conducted to evaluate the methane production from raw seaweed and derived materials. Biodegradation rates and the required time to final methane yield varied considerably according to the nature of the inoculum, mainly with its specific methanogenic activity and the ratio of substrate to inoculum used. While lower inoculum amounts will require longer assay times, higher substrate concentration could create toxic conditions for methane production. The initial substrate to inoculum ratio (dry weight) used in the present study was 1.25 (C1) and 2.5 (C2). The evolution of methane production in SMP batch assays during feed 1 of the three alga samples is shown in Fig. 1, including methane production in blank assays (without substrate) and control assays for methanogenic activity of the inoculum (VFA as substrate). The inoculum was previously kept at ambient temperature in order to remove any biodegradable organic matter which it may contain. In this way, the methane production obtained from the inoculum was only about 7 mL/gVS. This amount was subtracted from the cumulative methane production of the different assays in order to calculate the SMP of each substrate sample. A lag phase of only 2–3 days was observed for SM1 and VFA assays, while SM2 and SM3 showed the higher methane production rate since the starting of the assays. The maximum specific methanogenic activity (SMA) for the inoculum, obtained from the VFA control assays, resulted in 0.11 gCOD-CH4/gVSS d. The maximum SMA for the different alga assays was lower, ranging from 26% to 77% of that in the control assay. Replicate assays showed very similar methane production curves and final cumulative methane production values, which gave the SMP of each sample. Coefficients of variation for final cumulative methane production values are in the range of 0.1–7.7% (average 3.1 ± 2.4% for all assays), indicating accurate measurements of SMP. In order to simplify data presentation, in the following sections only average values of SMA and SMP from replicated assays will be used.

57

M. Soto et al. / Bioresource Technology 189 (2015) 53–61

700

400

SM1

500 400

SM2

350

C1 C1' C2 C2'

Methane (mL)

Methane (mL)

600

300 200

C1 C1' C2 C2'

300 250 200 150 100

100

50 0

0 0

5

10

15

20

25

0

30

5

10

600

20

25

30

700

Blank (B) and Control (C)

SM3 500

600

C1 C1' C2 C2'

400

Methane (mL)

Methane (mL)

15

Time(d)

Time (d)

300 200

B B' C C'

500 400 300 200

100

100

0

0 0

5

10

15

20

25

30

0

5

Time (d)

10

15

20

25

30

Time (d)

Fig. 1. Curves of methane production in SMP batch assays (replicates at concentrations C1 and C2) during feed 1 of the three alga samples (SM1, SM2 and SM3), including blank assays (without substrate) and control assay for methanogenic activity of the inoculum (VFA as substrate).

The pH was regulated at the beginning of each feed to the range of 7.0–7.1. Measured pH values during the course of each feed were always higher than the initial pH, even during the periods of higher methane production rates (corresponding to the first hours and days after feeding), and ranged from 7.2 to 7.4. pH values were slightly higher for C2 assays in comparison with C1 assays, and for SM2 in comparison with SM1 and SM3 alga samples. Accumulated substrate concentration reached 50 and 100 gTS/L in assays C1 and C2, respectively (Table 1). Thus, maximum sodium concentrations obtained from the added algae were 4.04, 4.54 and 2.95 g/L for assays C2 (feed 7) of SM1, SM2 and SM3, respectively. Additional sodium came from NaHCO3 added as buffer at the first feed, so total sodium concentration at the last feed resulted in 5.4, 6.0 and 4.4 gNa/L. As indicated above, these sodium concentrations could create methanogenic toxicity for non-adapted sludge. Sodium concentration in VFA control assays came from VFA neutralization and resulted in 1 gNa/L added at each feed, therefore reaching 7 g/L at the last feed. Furthermore, maximum nitrogen concentrations from the added algae were 1.82, 2.16 and 1.69 gN/L for assays C2 (feed 7) of SM1, SM2 and SM3, respectively. Maximum sodium and nitrogen concentrations for the C1 assays in feed 7 were half of those indicated above. As algae substrate was the unique source of nitrogen, indicated concentrations were considered the upper limit of ammonium nitrogen concentrations, thus ammonium toxicity was not expected during the performed assays. As the successive feeds were applied, the concentration of both soluble and suspended matter in the digesters increased. As indicated, the accumulated alga concentration at the 7th feed in the more concentrated assays (C2) was 100 gTS/L or 66 gVS/L for SM1 and SM3 and 48 gVS/L for SM2 (Table 1). Part of the added algae solids were hydrolyzed and solubilized during digestion, while the other fraction remained in the digesters thus increasing the amount of digested sludge in the digester. At the end of the experiments, the sludge build up in C2 digesters reached 35–47 gTSS/L and 28–30 gVSS/L, occupying more than 80% of the

liquid volume, then imposing the end of the experiments. Furthermore, this fact means that the highest influent concentration that can be applied to batch and semi-continuous digesters was estimated to be about 100 gTS/L. Probably for the same reason reported influent concentrations during anaerobic digestion in completely mixed continuous and semi-continuous digesters were below this limit, usually in the range of 10–100 gTS/L (Nielsen and Heiske, 2011; Hinks et al., 2013; Kinnunen et al., 2014). 3.3. Influence of substrate concentration and OLR on methane yield Results obtained for methane yield are presented in Fig. 2. Cumulative methane profile for SM2 at C1 and C2 concentrations (Fig. 1) indicate that a process of progressive adaptation of the inoculum had culminated before the end of the assays for C1 but not for C2, which may be the reason for the lower SMP of SM2 at C2 concentration. For all the other assays, the cumulative methane

Methane yield (mL CH4/gVS

Feed 1

250

Feeds 2-4 Feeds 5-7

200 150 100 50 0 SM1-C1

SM1-C2

SM2-C1

SM2-C2

SM3-C1

SM3-C2

Fig. 2. Specific methane potential (SMP) determined in batch assays (feed 1) and in successive feed assays (semi-continuous digestion assays).

58

M. Soto et al. / Bioresource Technology 189 (2015) 53–61

profiles suggest that a plateau of methane production was reached at the end of the assays (30 d of digestion). About 75% of the plateau methane generation corresponded to the quickly degradation step and the remaining 25% to the final slow degradation step. Specific methane potential (SMP) as determined in feed 1 ranged from 150 to 210 mL CH4/gVS. Relative values for the three alga samples changed depending on the substrate concentration. Lower values were obtained for SM2 at the higher concentration and for SM3 at the lower concentration. These results indicated that substrate to inoculum ratio had no influence on SMP from sample SM1, while opposite effects were found for samples SM2 and SM3, suggesting that the observed effects were probably due to the nature and concentration of the substrate itself. These results agree with those reported by Costa et al. (2012), who found no significant differences for the SMP from assays at 1%, 2.5% and 5% TS of substrate (substrate VS to inoculum VS ratio ranging from 1 to 7). Conversely, Nkemka and Murto (2010) found slightly lower methane yield at a higher substrate to inoculum ratio. Average SMP for each alga sample (first feed in batch assays) resulted of 208, 175 and 166 mL CH4/gVS for SM1, SM2 and SM3, respectively. Considering the similar characteristics of samples SM1 and SM3 (similar VSS and COD content), these results suggest that fine grinding of the alga (SM1 versus SM3) increased SMP during batch digestion by about 25%. On the other hand, the lower SMP for SM2 (16% lower than SMP for SM1) could be due to the chemical composition of this alga sample, according to the results of biodegradability of both VS and soluble COD, which are presented below. Fig. 2 also shows the evolution of the SMP through the successive feeds carried out, with average values of SMP for feeds 2–4 and 5–7. Ground algae (SM1 and SM2) suffered a significant decrease (p < 0.05) of SMP as the OLR increased by a factor of 10, from the low values typical of batch assays to values typical of semi or continuous digesters (Table 1). For these two samples, SMP showed a strong reduction of 35% from feed 1 to the average of feeds 5–7. For example, the SMP of 208 mL CH4/gVS for SM1 given by the batch assays (feed 1) would be reduced to 164 mL CH4/gVS at 0.6 gVS/ L d and to 135 mL CH4/gVS at 1.6 gVS/L d, while for SM2 the methane yield decreased from 175 to 156 and 113 mL CH4/gVS, respectively. On the other hand, the unground alga SM3 showed a constant SMP instead of the increasing OLR (p > 0.1), reaching an average value of 159 mL CH4/gVS. These results suggest that alga grounded will not be necessary if a long solids retention time is applied, as slowly hydrolysable substrate fractions could reach a complete conversion in those conditions. However, as found by Briand and Morand (1997), at low solid retention times the hydrolysis rate could be increased, by reducing the particle size through grinding. Finally, substrate concentration did not show significant influence on methane yield of SM1 and SM3 (p > 0.1), while a lower SMP was obtained for SM2 at C2 concentrations as compared to C1 concentrations (p = 0.05). Overall SMP values for all the feeds resulted in 159 and 144 mL CH4/gVS for SM1 at C1 and C2 concentrations, respectively. Similar results of 159 (C1) and 153 (C2) mL CH4/gVS were obtained for SM3, while SM2 reached lower overall values of 144 (C1) and 120 (C2) mL CH4/gVS. COD content of alga samples is shown in Table 2, allowing the expression of recovered methane as a percentage of theoretical methane potential, which in these operational conditions is 405 mL CH4/gCOD. While 66% (SM1), 52% (SM3) and 37% (SM2) of theoretical methane potential was obtained during batch assays (first feed), the recovered methane potential through the overall digestion period dropped to 48% for SM1, 49% for SM3 and 28% for SM2. These values are in the range of previous published values for seaweeds (Gurung et al., 2012; Kinnunen et al., 2014). The higher concentration in fiber and the presence of polyphenols in

Sargassum spp. together with a low content of sugars and fats, lead to low methane production from this brown seaweed (Briand and Morand, 1997; Jard et al., 2013). 3.4. Solids removal and COD solubilization Suspended solids removal was determined at the end of feed 4. Results (Fig. 3) indicate a higher liquefaction of SM3 organic matter with an average 58.3% VSS removal, followed by alga SM1 (56.2% VSS) and a lower biodegradability of SM2 (41.2% VSS). Total solids removal followed a similar pattern, although showing higher values (66.2, 61.3 and 55.1%TSS removal for algae SM3, SM1 and SM2, respectively). Significant differences existed for solid removals of the different algae samples (p < 0.01). On the other hand, while TSS liquefaction was slightly lower at higher concentrations (average TSS removals of 62.6% and 59.2% at C1 and C2, respectively, p < 0.05), removal of VSS did not show significant differences (52.5% and 51.3% at C1 and C2, respectively, p > 0.1). The higher values of TSS removal in comparison to VSS removal might be due to the solubilization of salts contained in the algae. In fact, the bigger difference between %TSS and %VSS removals corresponded to SM2, which showed a higher saline content (Table 2). The liquefaction of alga organic matter led to the generation of soluble substrate, which in part was no further biodegradable in anaerobic conditions and accumulated in the assay medium. Soluble chemical oxygen demand (CODs) concentration in the assays was determined at the end of feeds 4 and 7 (Table 3). CODs increased feed after feed in an amount directly proportional to the amount of alga feed. The ratio of generated CODs to VS feed did not show clear differences between assays with different alga concentration (C1 and C2), but increased from the end of feed 4 (67, 163 and 54 mgCODs/gVS feed, as averages for algae SM1, SM2 and SM3, respectively) to the end of the feed 7 (82, 190 and 63 mgCODs/gVS feed, respectively). Thus, a 16–23% increase of CODs not converted to methane was registered when the OLR increased from 0.28–0.79 gVS/L d to 0.76–2.11 gVS/L d. For the overall process, the accumulated CODs reached 11%, 16% and 9% of the total COD feed, for alga samples SM1, SM2 and SM3, respectively. However, larger differences in CODs accumulation were observed between algae samples, with higher values for SM2, which doubled or even triplicated the values of the other two samples. As SM2 showed lower percentage of VS removal, differences were even higher for the ratio CODs/VS removed. Average ratios of CODs/VS removed were 119, 395 and 93 mgCODs/gVS removed for algae SM1, SM2 and SM3, respectively. SM2 also showed a higher ratio of methane generation to removed solids, with an

Suspended solids removal (%) 80

TSS

VSS

70 60 50 40 30 20 10 0 SM1-C1

SM1-C2

SM2-C1

SM2-C2

SM3-C1

Fig. 3. Suspended solids removal at the end of feed 4.

SM3-C2

59

M. Soto et al. / Bioresource Technology 189 (2015) 53–61 Table 3 Soluble chemical oxygen demand (CODs) concentration in the assays at the end of feeds 4 and 7. Blank CODs (mg/L) Feed 4 Feed 7

0.17 ± 0.03 0.22 ± 0.01

CODs/VS ratio (mgCODs/gVS feed) Feed 4 Feed 7

SM1

SM2

C1

C2

C1

C2

C1

C2

0.89 ± 0.06 2.96 ± 0.21

2.28 ± 0.23 5.60 ± 0.25

1.75 ± 0.00 5.03 ± 0.57

3.23 ± 0.03 8.75 ± 0.42

0.89 ± 0.45 2.14 ± 0.07

1.62 ± 0.26 4.70 ± 0.46

54.0 83.0

79.4 81.4

165.7 201.7

160.3 179.0

54.1 58.1

54.5 67.8

average value of 359 mL CH4/gVS removed in comparison with 303 and 264 mL CH4/gVS removed for SM1 and SM3. VFA accumulation in assay media was determined during the exponential phase of methane production of feeds 3, 4, 5 and 7 in the higher concentration (C2) assays (Fig. 4). Low concentrations of acetate ranging from 5 to 50 mg/L and higher concentrations of butyrate, from 44 to 253 mg/L were usually present. Propionate appeared always below the detection limit of 5 mg/L, except for the last feed of SM3 at C2 concentration, in which 668 mg/L of propionic acid was found. Expressed as COD, the sum of acetate and butyrate results in about 200 mgCOD/L for feeds 3, 4 and 5, and about 400 mgCOD/L for feed 7. These amounts were equivalent to only 5–10% of COD in assays media, and to 5–15% of methane produced at the end of each feed (or in overall to only 0.5–1% of the COD feed). Thus, as an intermediate of the anaerobic digestion process, these low VFA levels indicate that both the acetogenic and the methanogenic steps were not limiting the overall degradation process. Therefore, the limiting step must be the hydrolytic and/or the acidogenic steps. These results agree with previous observations of several authors who consider hydrolysis to be the limiting step (Costa et al., 2012; Kinnunen et al., 2014). Data for VFA concentration at the end of feeds 4 and 7 revealed a slow removal rate of the accumulated butyrate, lower than the rate of methane generation (data not shown). These results suggest that butyrate was not a main product of fermentation, but butyrate degradation by acetogenic bacteria was inhibited, particularly during SM2 anaerobic fermentation. On the other hand, the high amount of propionate during the 7th feed of SM3 C2 assay was accompanied by a methane production rate lower than expected, indicating that some inhibition effect could exist at this high alga concentration.

through successive feeds tells us of the enrichment of the culture in methanogenic population and the presence or absence of process inhibition. SMA of the inoculum using a VFA mixture as substrate (assay control) increased from 0.11 to about 0.27 gCH4COD/gVSS d (1st to 4th feeds) and going forward remained stable at about 0.24 gCH4-COD/gVSS d (5th to 7th feeds). Algae treatment assays increased their SMA continuously from the 1st to the 7th feed. Lower SMA values for the algae treatment assays were due to a lower amount of substrate available to methanogenic bacteria in comparison to VFA control assays as was proved by the results of VFA analysis indicated above. In fact, except for reduced amounts of butyrate, other VFA components (acetate, propionate) were completely removed after each feed and before the following feed. The overall amount of substrate available to methanogenic bacteria can be assessed from the accumulated amount of methane production. Methanogenic bacteria use the available substrate to grow and to produce methane. If ambient conditions are adequate, methanogenic bacteria will maintain a continuous growth. In this situation the SMA should increase as a function of the available substrate or, on the other hand, as a function of the methane produced, as indicated by Eq. (4) (see Section 2). This means that a relationship between SMA and the accumulated methane production must be found, if the ambient conditions are free of inhibition factors. For the present study, using for each alga sample both C1 and C2 assays data, the following correlations were obtained (SMA in gCH4-COD/gVSS d, VCH4 in mL):

SM1

SMA ¼ 6:94  105  VCH4 þ 0:037

R2 ¼ 0:923

SM2

SMA ¼ 6:78  105  VCH4 þ 0:019

R2 ¼ 0:915

SM3

SMA ¼ 10:22  105  VCH4 þ 0:017 R2 ¼ 0:978 ðexcluded C2—F7thdatumÞ

3.5. Methanogenic activity and toxicity Specific methanogenic activity (SMA), referred to the initial inoculum added to the assays, is shown in Fig. 5. SMA evolution 300

VFA (mg/L)

250 200

Acetate

Butyrate

150 100 50 0 SM1 SM2 SM3 SM1 SM2 SM3 SM1 SM2 SM3 SM1 SM2 SM3 Feed 3

Feed 4

SM3

Feed 5

Feed 7

Fig. 4. VFA accumulation in assay media during the exponential phase of methane production of feeds 3, 4, 5 and 7 in C2 assays (propionate appeared always below detection limit of 5 mg/L, except for SM3-C2 at feed 7, in which 668 mg/L of propionic acid was found).

VFA

SMA ¼ 9:00  105  VCH4 þ 0:059 R2 ¼ 0:959 ðfeeds 1—4Þ

These correlations indicate that while the VFA control enables good growing conditions up to feed 4, the assay treatments for the three algae samples showed adequate growth from the first to the seventh feed, except for the C2–7th feed of SM3. The slope of these correlations is indicative of the relative growth of methanogenic bacteria in each assay. A higher relative growth for SM3 assays in comparison to VFA assay could be due to the presence of different methanotrophic routes, for example hydrogenophile methanogens. The relative growth indicated by these correlations was lower for SM1 and SM2. This lower growth could be due to substrate differences, while possible inhibition caused by biodegradable components must be disregarded as no differences were found for C1 and C2 assays. In any case, close values for the slope of all correlations were obtained, indicating that differences in methanogenic bacteria growth are reduced. Furthermore, accumulation of compounds showing methanogenic toxicity was not observed even when the accumulated concentration of SM1 and SM2 algae reached 100 gTS/L (feed 7),

60

M. Soto et al. / Bioresource Technology 189 (2015) 53–61

Specific methanogenic activity (gCH 4-COD/gVSS·d) 0.40 0.35 0.30

Feed 1 Feed 2

0.25

Feed 3

0.20

Feed 4

0.15

Feed 5

0.10

Feed 6 Feed 7

0.05 0.00 SM1-C1

SM1-C2

SM2-C1

SM2-C2

SM3-C1

SM3-C2

AGV

Fig. 5. Evolution through successive feeds of specific methanogenic activity (SMA) referred to the initial inoculum added to the assays.

Methane production (mL CH 4/d) 300

SM1 C2

SM2 C2

SM3 C2

250 200 150 100 50 0 0

10

20

30 Time (d)

40

50

60

Fig. 6. Daily evolution of methane production during semi-continuous C2 assays (feeds 2 to 7).

because the linear correlation was maintained. The same behavior was found for alga SM3 until 80 gTS/L, while the assay at 100 gTS/L suffered propionate accumulation and a lower methane generation rate than expected, indicating that methanogenic inhibition started at about this alga concentration. As indicated above, total sodium concentration at the last feeds resulted in 5.4, 6.0 and 4.4 gNa/L for C2 assays of SM1, SM2 and SM3, respectively, while nitrogen concentrations were 1.82, 2.16 and 1.69 gN/L, respectively. Thus, the observed inhibition of propionate degradation during the last feed of SM3 at C2 concentration was not due to ammonium or sodium concentration but is more likely due to the accumulation of refractory or partially biodegradable substrate, such as tannins and phenols, as suggested by several authors (Nkemka and Murto, 2010; Jard et al., 2013; Monlau et al., 2014). On the other hand, inhibition of methane production from VFA assays at 5th to 7th feeds could be related to sodium concentration that reached 5–7 gNa/L. Although these sodium concentrations are not much higher than those in concentrate algae assays, we must note that the sodium effect would be stronger in VFA assays because of the lack of antagonism effects that probably exist in alga assays because of the presence of different sea salts (Feijoo et al., 1995). Sodium inhibition during anaerobic digestion of seaweeds was reported at concentrations of 2.1 gNa+/L (Lakaniemi et al., 2011), 3.6 gNa+/L (Santos et al., 2014), 4.1 gNa+/L (Mottet et al., 2014), 5.0 gNa+/L (Schwede et al., 2013b) or at 20 gVS/L of alga concentration (Nielsen and Heiske, 2011). Hinks et al. (2013) also found methane production inhibition at concentrations of Laminaria hyperborea above 15 gVS/L during continuous treatment, and reported a MPR of only 38% at 22.5 gVS/L in comparison to the

MPR at 15 gVS/L. On the other hand, no toxicity or adaptation was found at concentrations of 2.9 gNa+/L (Kinnunen et al., 2014), at 4.1 gNa+/L or even at 9 gNa+/L for adapted halophilic sludge (Mottet et al., 2014). Our results indicate that inhibition didn’t take place at increasing sodium concentrations of up to 6 gNa/L. This can be due to some degree of inoculum preadaptation or to adaptation during the progressive increase of alga concentration. 3.6. Semi and continuous treatment considerations In the present study, after the start-up during the first feed, a multi feed batch operation was carried out, which can be comparable to a semi-continuous operation mode. OLR progressively increased through the operation time from the second to the seventh feed. Thus, this operation mode combined the characteristics of semi-continuous operation and simultaneous increases of both OLR and no biodegradable substrate concentration. Fig. 6 shows the daily evolution of methane production for the more concentrated C2 alga assays. Average OLR for the period of the 2nd to 4th feeds (0–27 days in Fig. 6) resulted in 0.6 gVS/L d for SM2 and 0.8 gVS/L d for SM1 and SM3, while average MPR resulted in 0.11, 0.07 and 0.11 L CH4/L d for SM1, SM2 and SM3 respectively. Average OLR for the period of 5th–7th (32–52 days in Fig. 6) resulted in 1.5 gVS/L d for SM2 and 2.1 gVS/L d for SM1 and SM3, while average MPR reached 0.22, 0.14 and 0.26 L CH4/L d for SM1, SM2 and SM3 respectively. On the other hand, maximum MPR reached 0.35, 0.21 and 0.37 LCH4/L d for SM1, SM2 and SM3 respectively during the first days after the 7th feed.

M. Soto et al. / Bioresource Technology 189 (2015) 53–61

Most studies of alga digestion were carried out in one step CSTR at HRT ranging from 10 to 60 d, OLR from 0.75 to 3 gVS/L d and reaching MPR of 0.1–0.38 L CH4/L d (Briand and Morand, 1997; Hinks et al., 2013; Kinnunen et al., 2014). Thus, OLR and MPR obtained for S. muticum in the present study were comparable to those of previous studies for other algae specie such as Ulva sp., Laminaria hyperborean or Oil extracted algae cake. As indicated above, the sludge build up in C2 digesters reached 28–30 gVSS/L and occupied nearly all the active volume of these digesters. The sludge in the digesters was composed of the active biomass, including acidogenic and methanogenic bacteria, and refractory alga solids that had not been hydrolyzed. Thus, the accumulation of non-biodegradable solids in digesters constitutes a physical limit to the amount of active biomass in digesters which would be necessary to reach higher MPRs. In fact, although SMA referred to the initial inoculum clearly increased which indicates good growth of methanogens, the actual SMA of the sludge built up in the digesters was lower. We found constant values at the end of feeds 4th and 7th of about 0.03 ± 0.01, 0.02 ± 0.0 and 0.04 ± 0.01 gCH4-COD/gVSS d for SM1, SM2 and SM3 respectively. Similar results were reported by Nielsen and Heiske (2011) for co-digestion of Ulva lactuca with cattle manure. Furthermore, our results indicated that MPR from semi-continuous digesters treating S. muticum was not limited by toxicity but by the accumulation of non-biodegradable alga solids, which impede the accommodation of a larger amount of active methanogenic bacteria. 4. Conclusions Average SMP for three samples of S. muticum collected at different seasons resulted in 166–208 mL CH4/gVS. Fine grinding of the alga increased SMP but only during batch digestion, thus alga grinding was not necessary for continuous digestion. Substrate concentration and substrate to inoculum ratio had no influence on SMP or VSS removal values. Accumulation of compounds showing methanogenic toxicity was not observed at least up to 80 gTS/L. Semi-continuous operation at OLR of 3.2 gTS/L d showed MPR of 0.14–0.26 L CH4/L d while the accumulation of non-biodegradable solids in digesters constitutes a physical limit to higher digester efficiencies. References Briand, X., Morand, P., 1997. Anaerobic digestion of Ulva sp. 1. Relationship between Ulva composition and methanization. J. Appl. Phycol. 9 (6), 511–524. Bird, K.T., Chynoweth, D.P., Jerger, D.E., 1990. Effects of marine algal proximate composition on methane yields. J. Appl. Phycol. 2 (3), 207–213. Bohutskyi, P., Betenbaugh, M.J., Bouwer, E.J., 2014. The effects of alternative pretreatment strategies on anaerobic digestion and methane production from different algal strains. Bioresour. Technol. 155, 366–372. Bruhn, A., Dahl, J., Nielsen, H.B., Nikolaisen, L., Rasmussen, M.B., Markager, S., Olesen, B., Arias, C., Jensen, P.D., 2011. Bioenergy potential of Ulva lactuca: Biomass yield, methane production and combustion. Bioresour. Technol. 102, 2595–2604. Chynoweth, D.P., Owens, J.M., Legrand, R., 2001. Renewable methane from anaerobic digestion of biomass. Renew. Energy 22, 1–8. Chynoweth, D.P., 2005. Renewable biomethane from land and ocean energy crops and organic wastes. HortScience 40, 283–286. Costa, J.C., Gonçalves, P.R., Nobre, A., Alves, M.M., 2012. Biomethanation potential of macroalgae Ulva spp. and Gracilaria spp. and in co-digestion with waste activated sludge. Bioresour. Technol. 114, 320–326. Feijoo, G., Soto, M., Méndez, R., Lema, J.M., 1995. Sodium inhibition in the anaerobic digestion process: antagonism and adaptation phenomena. Enzyme Microb. Technol. 17 (1), 1–9.

61

Ferreiro, N., Soto, M., 2003. Anaerobic hydrolysis of primary sludge: influence of sludge concentration and temperature. Water Sci. Technol. 47 (12), 239–246. González-López, N., Moure, A., Domínguez, H., 2012. Hydrothermal fractionation of Sargassum muticum biomass. J. Appl. Phycol. 24, 1569–1578. Gurung, A., Van Ginkel, S.W., Kang, W.C., Qambrani, N.A., Oh, S.E., 2012. Evaluation of marine biomass as a source of methane in batch tests: a lab-scale study. Energy 43, 396–401. Hinks, J., Edwards, S., Sallis, P.J., Caldwell, G.S., 2013. The steady state anaerobic digestion of Laminaria hyperborea – effect of hydraulic residence on biogas production and bacterial community composition. Bioresour. Technol. 143, 221–230. Jard, G., Marfaing, H., Carrère, H., Delgenes, J.P., Steyer, J.P., Dumas, C., 2013. French Brittany macroalgae screening: composition and methane potential for potential alternative sources of energy and products. Bioresour. Technol. 144, 492–498. 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. 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. Langlois, J., Sassi, J.-F., Jard, G., Steyer, J.-P., Delgenes, J.-P., Hélias, A., 2012. Life cycle assessment of biomethane from offshore-cultivated seaweed. Biofuel. Bioprod. Bior. 6, 387–404. Lodeiro, P., Cordero, B., Grille, Z., Herrero, R., Sastre de Vicente, M.E., 2004. Physicochemical studies of cadmium(II) biosorption by the invasive alga in Europe, Sargassum muticum. Biotechnol. Bioeng. 88, 237–247. Marquez, P.G.B., Santiañez, W.J.E., Trono Jr., G.C., Montaño, M.N.E., Araki, H., Takeuchi, H., Hasegawa, T., 2014. Seaweed biomass of the Philippines: Sustainable feedstock for biogas production. Renew. Sustain. Energy Rev. 38, 1056–1068. Monlau, F., Sambusiti, C., Barakat, A., Quéméneur, M., Trably, E., Steyer, J.P., Carrère, H., 2014. Do furanic and phenolic compounds of lignocellulosic and algae biomass hydrolyzate inhibit anaerobic mixed cultures? A comprehensive review. Biotechnol. Adv. 32 (5), 934–951. Migliore, G., Alisi, C., Sprocati, A.R., Massi, E., Ciccoli, R., Lenzi, M., Wang, A., Cremisini, C., 2012. Anaerobic digestion of macroalgal biomass and sediments sourced from the Orbetello lagoon, Italy. Biomass. Bioenerg. 42, 69–77. Mottet, A., Habouzit, F., Steyer, J.P., 2014. Anaerobic digestion of marine microalgae in different salinity levels. Bioresour. Technol. 158, 300–306. Nielsen, H.B., Heiske, S., 2011. Anaerobic digestion of macroalgae: methane potentials, pre-treatment, inhibition and co-digestion. Water Sci. Technol. 64 (8), 1723–1729. Nkemka, V.N., Murto, M., 2010. Evaluation of biogas production from seaweed in batch tests and in UASB reactors combined with the removal of heavy metals. J. Environ. Manage. 91, 1573–1579. Santos, N.O., Oliveira, S.M., Alves, L.C., Cammarota, M.C., 2014. Methane production from marine microalgae Isochrysis galbana. Bioresour. Technol. 157, 60–67. Schwede, S., Kowalczyk, A., Gerber, M., Span, R., 2013a. Anaerobic co-digestion of the marine microalga Nannochloropsis salina with energy crops. Bioresour. Technol. 148, 428–435. Schwede, S., Rehman, Z.U., Gerber, M., Theiss, C., Span, R., 2013b. Effects of thermal pretreatment on anaerobic digestion of Nannochloropsis salina biomass. Bioresour. Technol. 143, 505–511. Sialve, B., Bernet, N., Bernard, O., 2009. Anaerobic digestion of microalgae as a necessary step to make microalgal biodiesel sustainable. Biotechnol. Adv. 27 (4), 409–416. Soto, M., Veiga, M.C., Méndez, R., Lema, J.M., 1989. Semi-micro COD determination method for high saline wastewater. Environ. Technol. Lett. 10, 541–548. Soto, M., Méndez, R., Lema, J.M., 1991. Biodegradability and toxicity in the anaerobic treatment of fish canning wastewaters. Environ. Technol. 12, 669–677. Soto, M., Méndez, R., Lema, J.M., 1993. Methanogenic and non-methanogenic activity test. Theoretical basis and experimental set up. Water Res. 27 (8), 1361–1376. Standard Methods for the Examination of Water and Wastewater (1995). 19th. Edition, American Public Health Association/American Water Works Association/ Water Environment Federation, Washington D.C., USA. Tartakovsky, B., Matteau-Lebrun, F., 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 (4), 394–400. Vergara-Fernández, A., Vargas, G., Alarcón, N., Velasco, A., 2008. Evaluation of marine algae as a source of biogas in a two-stage anaerobic reactor system. Biomass Bioenergy 32, 338–344. Zamalloa, C., Boon, N., Verstraete, W., 2012. Anaerobic digestibility of Scenedesmus obliquus and Phaeodactylum tricornutum under mesophilic and thermophilic conditions. Appl. Energy 92, 733–738.