Accepted Manuscript

Anaerobic digestion of marine micro-algae in different salinity levels

Alexis Mottet, Frédéric Habouzit, Jean Philippe Steyer

PII: S0960-8524(14)00226-0DOI: http://dx.doi.org/10.1016/j.biortech.2014.02.055Reference: BITE 13053

To appear in: Bioresource Technology

Received Date: 17 December 2013Revised Date: 12 February 2014Accepted Date: 14 February 2014

Please cite this article as: Mottet, A., Habouzit, F., Steyer, J.P., Anaerobic digestion of marine micro-algae indifferent salinity levels, Bioresource Technology (2014), doi: http://dx.doi.org/10.1016/j.biortech.2014.02.055

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

1

*corresponding author: frederic.habouzit@supagro.inra.fr

Phone : +33 468 425 187 Fax : +33 468 425 160

Anaerobic digestion of marine micro-algae in different 1

salinity levels 2

Alexis Mottet1, Frédéric Habouzit1,* and Jean Philippe Steyer1 3

1 INRA, UR50, Laboratoire de Biotechnologie de l’Environnement, Avenue des Etangs, Narbonne, F-4

11100, France. 5

ABSTRACT 6

In the context of biofuel production from marine microalgae, anaerobic digestion has 7

the potential to make the process more sustainable and to increase energy efficiency. 8

However, the use of salt-containing microalgae organic residues entails the presence of 9

salts which inhibits methanogenesis. The search for suitable anaerobic microbial 10

consortium adapted to saline conditions can boost the anaerobic conversion into 11

methane. The anaerobic digestion performance of three different anaerobic microbial 12

consortia was assessed in batch tests at different salinities between 15 and 150g.L-1 and 13

for three successive substrate additions. After an acclimation period, the methane (CH4) 14

yield of the halophilic methanogens at 35g.L-1 of salinity was close to the reference 15

value without salt addition. Above 75g.L-1

of salinity, methanogenesis was considerably 16

slowed down. The results underline that methane production from halophilic sediment 17

can be envisaged and promoted for practical application at a seawater concentration. 18

Keywords: microalgae, anaerobic digestion, salinity, acclimatization. 19

2

1 Introduction 20

The use of microalgae appears as promising as an alternative to fossil fuels and as 21

applications for high-value products (Cheng and Timilsina, 2011). However, water and 22

nutrient supplies and, also, energy consumption have been pinpointed as key issues in the 23

biofuel production systems using microalgae (Subhadra and Edwards, 2011). 24

Some species of microalgae can be grown in saline water and can make the system more 25

environmentally friendly by greatly reducing the freshwater requirement (Schenk et al., 26

2008). Combining with anaerobic digestion, microalgae culture system can become: (i) more 27

sustainable by taking advantage of the digestate as nutritive medium for microalgae culture; 28

(ii) more energy-efficient by using the methane (CH4) as biofuel for electricity production 29

(Collet et al., 2011). 30

Sialve et al. (2009) have shown that the anaerobic digestion of microalgae and of the organic 31

residues from oil extraction can be successfully carried out. Moreover, Lakaniemi et al. 32

(2012) have shown that the anaerobic digestion of microalgal biomass leads to 14.4 kJ.g-1

dry 33

wt and presents a higher energy yield compared to butanol and H2 fermentations. 34

Unfortunately, anaerobic digestion of halophilic microalgae, such as Dunaliella tertiolecta 35

that grows in conditions of high salinity, is greatly inhibited by the presence of sodium (Na+) 36

and sulphate (SO42-) (Lakaniemi et al., 2011). Although a low Na+ concentration (100 – 200 37

mg.L-1) is essential for the growth of methanogens (Dimroth and Thomer, 1989), high 38

concentrations indeed cause cell plasmolysis and cell death due to the high external osmotic 39

pressure and the inhibition of reaction pathways in anaerobic digestion. The presence of SO42-

40

induces the growth of sulphate-reducing bacteria (SRB) since they use it as a terminal 41

electron acceptor. Consequently, SRB dominate methanogens in using their common 42

substrates e.g. hydrogen and acetate (Muyzer and Stams, 2008). 43

3

Several strategies have been developed to improve anaerobic degradation performance at high 44

saline concentrations. These include the utilization of halophilic microorganisms (Kapdan and 45

Erten, 2007), a long biomass acclimation period for non-halophilic inocula to adapt to salinity 46

(Kimata-Kino et al., 2011) and the use of compatible solutes (Oh et al., 2008). Compatible 47

solutes seem more effective in sudden variations in feed salinity (Vyrides and Stuckey, 2009) 48

whereas, for treating a high-salinity substrate, adapted inocula are more effective in the long 49

term (Yerkes et al., 1997). Therefore, it is possible to use halophilic microorganisms from 50

saline environments or halo-tolerant microorganisms obtained by a long acclimation period. 51

The entire physiology of halophilic microorganisms are adapted to high conditions of salinity. 52

Two mechanisms are involved in order to balance out the external osmotic pressure: (i) 53

accumulation of inorganic ions in the cytoplasm (“salt-in” strategy), usually used by 54

halobacteria of the Haloanaerobiales order, a strategy that requires far-reaching adaptation of 55

intracellular “machinery” and so limits growth to only certain levels of salinity; and (ii) 56

accumulation of compatible solutes (“compatible-solute” strategy) that do not interfere with 57

the central metabolism of the cell, a solution used by the majority of living cells, including 58

most Archaea (Oren, 1999; Müller et al., 2005). Moreover, in hypersaline environments, 59

microalgae, invertebrates and prokaryotes are the major sources of oxidizable compounds. 60

Thus, halophilic microorganisms are able to grow and metabolize a wide range of substrates. 61

However, the persistence of methanogens in hypersaline environments is related to the 62

presence of non-competitive substrates since H2 and acetate are used mainly via sulfate 63

reduction (Ollivier et al., 1994). This was confirmed by the study of Nolla-Ardévol et al. 64

(2012) who showed that under such conditions the highest CH4 production was obtained with 65

methanol. 66

The use of halophilic microorganisms has already been reported for wastewater treatment. 67

Aspé et al. (1997) showed that a marine sediment adapted better and faster than an inoculum 68

4

from pig manure for the anaerobic treatment of fishery wastewater. Lefebvre et al. (2004) 69

obtained high removal of soluble COD and soluble NTK with a halophilic sediment in a 70

bench-scale sequencing batch reactor. Kapdan and Erten (2007) used Halanaerobium 71

lacusrosei in an upflow anaerobic sludge blanket (UASB) to treat synthetic wastewater with 72

glucose as the carbon source and a COD removal of 70 % was obtained with salinity levels 73

up to 30 g.L-1

. 74

Recently, Migliore et al. (2012) have evaluated the intrinsic anaerobic degradation potential 75

of the Orbetello lagoon ecosystem for usefully converting macro-algae biomass into CH4. The 76

microbial consortium was well adapted to macroalgal tissue composition and to the salts and 77

toxic components present in water and sediments. They improved anaerobic degradation 78

activity without any lag phase in CH4 production. During the degradation, acetate was the 79

main intermediate metabolite, showing that acetoclastic methanogens from marine sediments 80

can be active in such operating conditions. 81

The duration of the acclimatiion period greatly affects the degree of saline tolerance to saline 82

conditions of the halotolerant anaerobic microbial consortia involved. As reported by Chen et 83

al. (2008), the toxic concentration resulting in a 50% reduction (IC50) in cumulative methane 84

production over a fixed period of exposure time varies from 5.6 to 53 gNa+.L-1. However, the 85

conclusions of authors in the literature differ. Kimata-Kino et al. (2011) used an innovative 86

strategy for acclimate UASB granules by combining a sudden abrupt increasing salinity 87

followed by a gradual rise. The CH4 production was reduced by only 13 % compared to the 88

control at a salinity of 32 g.L-1

. Jeison et al. (2008) showed biomass acclimation for CH4 89

production in an anaerobic membrane reactor with a IC50 value of 25 gNa+.L

-1. However, they 90

noted that small and weak granules were observed which may negatively affect long-term 91

biomass retention in UASB reactors treating very saline wastewater. Luo et al. (2012) 92

observed a decrease in biogas production during biomass acclimation in an anaerobic 93

5

sequencing batch reactor and total inhibition at salinities higher than 8.7 g.L-1

. Sea salt was 94

used to increase salinity since ionic composition of the sea water reduces Na+ inhibition and 95

can enhance the feasibility of methanogenesis (Yerkes et al. 1997). 96

To clarify the different results found in the literature, the two-fold objective of the present 97

study was to evaluate the anaerobic degradation potential of three different anaerobic 98

microbial consortia with different initial tolerances towards salinity in treating species of 99

marine microalgae and to determine the impact of salinity and of successive additions of 100

substrate on the maximum CH4 and H2S yields. 101

2 Materials and methods 102

2.1 Inoculum (anaerobic microbial consortia) 103

Depending on their origin, 3 types of inocula were used. 104

The first one, called” adapted inoculum” in this work, is an halophilic sediment obtained 105

from the feed channels bottom soil of salterns in Gruissan (France) (43°5’N, 3°4’E). The 106

sediment sample (around 5 kg) was collected at a depth of 40 cm below the water surface in 107

April. On that date the salinity of the sediment had a value of 33g/L (see Table1). The 108

seawater in the saltern feed channel undergoes evaporation during the summer season and 109

dilution when it rains. Even if the salinity varies during the year, the population is well 110

adapted to the sea water salinity. In order to removal pebbles, shells and other particles, the 111

inoculum was sieved at 8 mm. 112

The second inoculum, called ”acclimated inoculum” in this work, is one of the two industrial 113

inocula. It came from a UASB digester treating wastewaters from a sugar syrup 114

decolorization process which had a saline concentration of around 10 g.L-1

(See Table 1). 115

The last inoculum was called “non-acclimated inoculum” because it was obtained from a 116

digester treating sewage sludge with a low sodium concentration. 117

6

The inocula were introduced into batch reactors with a working volume of 2.5 L at 35°C. The 118

volatile solid (VS) concentration was adjusted by dilution with feed channel seawater for the 119

halophilicsediment and tap water for the industrial inocula. The characteristics of the inocula 120

are shown in Table 1. Ethanol was first fed out into each reactor in order to evaluate the 121

methanogenic activity of the microorganisms. Once this yield had reached 60 % methane in 122

the biogas the inocula were used for the anaerobic degradation batch tests. 123

2.2 Experimental setup 124

A marine microalgae, Dunaliella salina, was used as a saline substrate model. The culture 125

was produced in batch conditions at a salinity of 120 g.L-1

by the GREENSEA company 126

(Mèze, France). Its main physico-chemical characteristics and ionic composition after 127

harvesting and centrifugation are presented in Table 1. 128

Anaerobic degradation batch tests were carried out in 500 mL serum bottles with a working 129

volume of 400 mL at 35°C and in triplicate for each condition. The degradation performance 130

for the three inocula was evaluated at four levels of salinity: 15, 35, 75 and 150 g.L-1

. The 131

organic loading rate (O.L.R.) was 0.5 gVSsubstrate.gVSSinoculum-1 and the initial volatile organic 132

matter concentration in each serum bottle was fixed ot 3 gVS.L-1. Therefore, the salinity was 133

adjusted for each level of salinity by adding ions in crystalline form. For each salinity level, 134

the proportion of main ions of seawater (Na+, SO42-, Mg2+, K+ and Ca2+) was maintained. The 135

quantity of each salt, introduced to reach the fixed salinity level, was calculated from the 136

balance between the quantity introduced by the substrate and by the inoculum in the serum 137

bottle and the proportion of each ion in seawater. The salinity and the composition of the 138

seawater used in this work was: 35 g.L-1

; 19.35 gCl- (55.1%), 10.78 gNa

+ (30.7%), 2.71 139

gSO42-

(7.7%), 1.28 gMg2+

(3.7%), 0.41 gCa2+

(1.2%) and 0.4 gK+ (1.1%) per liter, 140

respectively. 141

7

Three successive substrate additions (Feed 1, Feed 2 and Feed 3) were carried out in each 142

serum bottle in order to assess acclimation effects (a repetition corresponds to one substrate 143

addition). The anaerobic degradation batch tests were monitored by measuring the biogas 144

volume until it reached an insignificant level, leading to a maximum duration of 60 days for 145

each serum bottle. An experiment without any addition of salts was performed in order to 146

determine the specific CH4 production from the substrate. 147

The volume of biogas produced was measured by liquid (water, pH = 2, NaCl 10%) 148

displacement. The anaerobic degradation performance was assessed on the basis of methane 149

and hydrogen sulfide yields and volatile fatty acid (VFA) concentrations at the end of each 150

run. The yield for each gas was calculated by normalizing the produced biogas volume by the 151

gram of VSsubstrate introduced in the serum bottle. 152

153

2.3 Analytical methods 154

Total solids (TS), inorganic solids (IS) and volatile suspended solids (VSS) were measured in 155

accordance with standard methods (Federation, 1999). The soluble fraction was separated by 156

centrifugation (10min, 11 000 g, 5°C). VFA (acetate, propionate, iso-butyrate, butyrate, iso-157

valerate and valerate) were measured in the soluble phase using a gas chromatograph (GC-158

800 Fisons Instrument) equipped with a flame ionisation detector. Sodium, sulphate and 159

ammonium concentrations were analysed in the soluble phase by an ion chromatography 160

system (DIONEX 100) using conductivity detection. Conductivity and salinity were 161

determined using the Multi 340i handheld meter (WTW) equipped with a conductivity 162

measuring cell (TetraCon® 325). The headspace gas composition was determined using a gas 163

chromatograph (Perkin Clarus 480) equipped with two different capillary columns (RtUBond 164

column and RtMolsieve 5A column) and a thermal conductivity detector. 165

8

2.4 Statistical analysis 166

A three-way ANOVA was used to examine the significance of the factors i.e. salinity (15, 35, 167

75 and 150 g.L-1

), the initial acclimation factor for sodium (non-acclimated, acclimated and 168

adapted) and the successive substrate additions (Feed 1, Feed 2 and Feed 3). Data were 169

considered to be significantly if p < 0.05. Data analysis was performed using the Rcmdr 170

library in R software version 2.15.1.1 (Comprehensive R Archive Network, 171

Wirtschaftsuniversität, Vienna, Austria) 172

3 Results and discussion 173

3.1 Anaerobic digestion performance of concentrated Dunaliella salina in suspension 174

without salt addition 175

The anaerobic digestion performance the degradation of a marine microalgae Dunaliella 176

salina was determined in normal conditions, i.e. without any addition of salt, in order to 177

evaluate the maximum conversion of organic matter into CH4. An acclimated inoculum was 178

used for this experiment. The acclimated inoculum described in section 2.1 was used for 179

anaerobic batch experiments in our laboratory with an O.L.R. of 0.5 gVSsubstrate.gVSinoculum-1. 180

The salinity and sodium concentrations were 5.5 g.L-1 and 1.5 g.L-1, respectively, in the batch 181

tests. This sodium level is not considered to be inhibiting (Chen et al., 2008). Throughout the 182

batch experiments, H2S content in the head space was less than 0.01%. The specific CH4 183

production was 359.8 ± 7.8 mLCH4.gVS-1

. This value is close to the performance obtained by 184

Mussgnug et al. (2010) with a yield of 323.0 ± 16.0 mLCH4.gVS-1

. The modeled parameters λ 185

and Rmax using the Gompertz equation (Xia et al., 2012) were, respectively, 0.13 ± 0.01 d and 186

59.2 ± 4.1 mLCH4.gVS-1

.d-1

. These results confirmed that Dunaliella salina could be a 187

suitable substrate for anaerobic digestion since a rapid growth of methanogens was observed 188

and a high CH4 yield obtained. 189

9

However, if a digester is continuously fed with a substrate containing a high concentration of 190

salts, the salinity at steady state in the digester will soon equal the salinity of the feeding 191

media. It is therefore necessary to assess the anaerobic digestion performance in conditions 192

with high salinity. 193

3.2 Influence of salinity and acclimation time on CH4 production 194

Three anaerobic microbial consortia, each with a different initial tolerance of Na+ 195

concentration were used to evaluate their potential for converting organic matter into CH4 at 196

various levels of salinity between 15-150 g.L-1. Substrate addition was repeated three times 197

for each serum bottle in order to take into account for the microbial acclimation to the 198

substrate salinity and the operating conditions (Buffiere et al., 2006). 199

The specific CH4 and H2S production for the three inocula as a function of salinity and 200

successive substrate additions are presented in Figures 1 and 2, respectively as a function of 201

salinity and successive substrate additions. The dotted line represents the 95 % confidence 202

interval. Different patterns can be observed for each inocula with, on average, a decrease of 203

the CH4 yield and an increase of the H2S highlighting two distinct ranges of salinity: 15 and 204

35 g.L-1 as low salinity and 75 and 150 g.L-1 as high salinity. 205

3.2.1 Performance in conditions of low salinity (15 and 35 g.L-1) 206

After one addition of substrate, the highest CH4 yields were obtained at a salinity of 15 g.L-1

207

with a mean value of 204.3 mLCH4.gVS-1

, showing no significant difference between the 208

three inocula (Figure 1). This value represents 57% of the specific CH4 production obtained in 209

normal conditions. At 35 g.L-1

, the CH4 yield values remained similar for the adapted and 210

acclimated inocula. A significant drop of 42% was observed for the non-acclimated inoculum 211

when the salinity increased from 15 to 35 g.L-1. 212

10

At 15 g.L-1

, the inhibiting of salinity on the methanogens had the same effect on all three 213

inocula since their CH4 yields were similar. With salinity at 35 g.L-1

, the inoculum unused to 214

treating a saline influent suffered considerable inhibition in its CH4 production whereas the 215

performances of the acclimated and adapted inocula were similarly good. Therefore, the 216

methanogens used for these experiments had a tolerance for toxic sodium concentration in 217

relation to their initial acclimatization to sodium concentration. 218

After the second addition of substrate, the acclimated and adapted inocula produced CH4 219

yields significantly higher than for the first addition at 15 g.L-1: the increase was 45% for both 220

inocula. For the non-acclimated inoculum, in contrast, no improvement was observed. At 35 221

g.L-1 of salinity, a slight decrease in CH4 production was observed for the acclimated 222

inoculum while the performance remained similar for the other inocula compared to the first 223

substrate addition. These results show that methanogens have a capacity of acclimatization to 224

conditions of very low salinity and this capacity is independent to their initial tolerance Na+. 225

After a long period of acclimation to the specific process conditions, high CH4 yields were 226

obtained at 15 g.L-1

of salinity for the three inocula, with a mean value of 346.8 ± 18.8 227

mLCH4.gVS-1

which is close to the CH4 yield with no added salts. These results show that 228

methanogens, no matter their origin, had a capacity of acclimation to conditions of very low 229

salinity (15 g.L-1) and this capacity is independent to their initial tolerance Na+. At a salinity 230

of 35 g.L-1, the CH4 yields increased significantly compared to the initial performances: 178.2 231

± 20.2 mLCH4.gVS-1 for the non-acclimated inoculum, 266.8 ± 18.2 mLCH4.gVS-1 for the 232

acclimated inoculum and 325.1 ± 20.0 mLCH4.gVS-1

for the adapted inoculum. 233

A salinity of 15 g.L-1

corresponds to a sodium concentration of 4.6 g.L-1

which is below the 234

minimum toxic concentration reordered by Chen et al. (2008). Therefore, with very low 235

salinity, total acclimation can be achieved for each inoculum. At a salinity of 35 g.L-1

, the 236

11

methanogens of acclimated or adapted inocula are able to remain a highly activity and achieve 237

a promising performance after a period of acclimation. 238

Figure 2 shows the three-way ANOVA results for the three inocula representing the specific 239

H2S production as a function of salinity and successive substrate additions. At low saline 240

conditions of 15 and 35 g.L-1

, the H2S yields after the first substrate addition were very low 241

for despite an initial sulfate concentration of 1.2 and 2.7 g.L-1

respectively. After the 242

acclimation period of three successive substrate additions, the H2S yields did not show 243

significant increases for the three inocula at 15 g.L-1 of salinity. At 35 g.L-1, an increase in the 244

H2S yield was observed which suggests an acclimation of the SRB to seawater concentration. 245

The non-acclimated inoculum had a H2S production significantly lower than that of the 246

acclimated and adapted inocula which did not differ significantly between 15 and 35 g.L-1

of 247

salinity. However, in conditions of low salinity, it can be seen from the ratio of the methane 248

volume and the total biogas produced that for each inoculum at 15 and 35 g.L-1

, 88 % or more 249

of the biogas was produced via methanogenesis. 250

During the anaerobic digestion of microalgae at a salinity of 35 g.L-1

, which was equivalent to 251

the seawater concentration on composition in this work, the methanogenesis was thus the 252

main degradation pathway despite a low COD/sulphate ratio of around 0.88 which could well 253

promote sulphate reduction, as reported by Visser (1995). This confirms the conclusions of 254

Colleran and Pender (2002) concluded that SRB have a lower affinity for acetate than do 255

acetoclastic methanogens. Thus, in our operating conditions, the conversion of organic matter 256

into CH4 was carried out mainly to acetoclastic methanogenesis. 257

In Table 2, it can be seen that in conditions of low salinity, the acetate and propionate 258

concentrations at the end of each feedwere equal to 0 in the case of the acclimated and 259

adapted inocula. Other volatile fatty acid (e.g. butyrate, valerate) were never detected. The 260

non–acclimated inoculum showed a slight accumulation of acetate and propionate at 35 g.L-1

261

12

of salinity after the three substrate additions, with values of 0.11 ± 0.04 gCOD.L-1

and 0.17 ± 262

0.11 gCOD.L-1

, respectively. Clearly, acetogenic bacteria from an acclimated or adapted to 263

saline conditions are less sensitive than non-acclimated bacteria. 264

Up to a saline concentration of 35 g.L-1

, equivalent to a seawater concentration and 265

composition, the adapted inoculum showed a faster acclimation to process conditions than the 266

industrial inocula and obtained the most promising performance in terms of CH4 production 267

after an acclimation time. A CH4 yield of 345.8 ± 7.1 mLCH4.gVS-1 for the adapted inoculum 268

was obtained. The CH4 yield was similar to the performance in normal conditions indicating 269

low inhibition of methanogens by salinity. The acclimation period is therefore an important 270

factor in the improvement of methanogenic activity, as it has already been reported by Feijoo 271

et al. (1995) who obtained a better performance from an anaerobic filter treating highly saline 272

wastewater for 2 years than with a digester treating conventional wastewater. Mendéz et al. 273

(1995) also reported an increase of 42% in the toxic concentration that causes a 90% fall in 274

cumulative methane production from a sludge with only one day acclimation compared to a 275

sludge with 719 days of acclimation. At seawater concentrations, we obtained a CH4 yield of 276

345.8 ± 7.1 mLCH4.gVS-1

for the adapted inoculum. This is similar to the performance in 277

normal conditions indicating low inhibition of methanogens by salinity. 278

3.2.2 Performance in conditions of high salinity (75 and 150 g.L-1) 279

In the case of the adapted inoculum, a CH4 yield of 174.3 ± 5.9 mLCH4.gVS-1

was obtained at 280

75 g.L-1

, followed by a sharp drop at 150 g.L-1

(Figure 1). Through the successive substrate 281

additions, specific CH4 production decreased significantly at 75 g.L-1

of salinity until reaching 282

a very low value of 10.4 ± 3.7 mLCH4.gVS-1

which is similar to the performance at 150 g.L-1

. 283

Therefore, it can be conclude that halophilic methanogens become sensitive to high Na+ 284

concentrations after long exposure. 285

13

After the first substrate addition the two industrial inocula showed very poor performance in 286

terms of CH4 production at 75 and 150 g.L-1

(Figure 1). For the ANOVA calculation, one 287

datum and two data were removed for the acclimated inoculum at 75 g.L-1

during the second 288

and third substrate additions respectively to have normality of the residuals. Indeed, as 289

presented in Figure3, CH4 yields have a high standard deviation of 107.2 ± 166.1 290

mLCH4.gVS-1

(Feed 2) and 253.0 ± 129.2 mLCH4.gVS-1

(feed 3). This variation is explained 291

by the uptake of acetate accumulated over the previous batch tests.Indeed, in the case of the 292

serum bottle A, the largeCH4 production was linked to the acetate uptake accumulated over 293

the first substrate addition. For serum bottles B and C, the CH4 was produced from the acetate 294

uptake accumulated over the first and second substrate additions. The non-acclimated 295

inoculum also showed a similar trend. In order to determine a more realistic estimate of the 296

methane potential of both inocula, it was thus necessary to calculate the CH4 yield based on 297

the total CH4 produced and the total organic matter introduced through the three substrate 298

additions. For the non-acclimated and acclimated inocula, we obtained CH4 yields of 122.4 ± 299

5.5 mLCH4.gVS-1

and 136.4 ± 31.8 mLCH4.gVS-1

were obtained, which represents 34.0 % 300

and 37.9 % of the methane potential in normal conditions. Therefore, methanogens of 301

industrial origin revealed mechanisms for physiological acclimation to sodium concentrations 302

as high as 23 g.L-1 (i.e., a total salinity of 75 g.L-1) but this value corresponds to a threshold 303

above which methanogenesis is greatly slowed down. 304

The mechanism used by the inocula of industrial origin and the adapted inoculum to balance 305

their external osmotic pressure seems different in conditions of high salinity.As reported by 306

Oren (1999), halophilic microorganisms usually use the “salt-in-cytoplasm” strategy to 307

survive under osmotic stress involving the adaptation of intracellular enzymes. However, this 308

evolution limits the growth to certain levels of salinity. This observation could explain the 309

failure of halophilic methanogens to acclimate to a salinity above 75 g.L-1

. For the 310

14

methanogens of industrial origin, it seems that less energy was derived for methanogens 311

growth through the degradation of acetate into CH4 since the observed production rate was 312

slower. Thus, the “compatible solutes” strategy seems predominant, allowing acclimation to 313

high levels of salinity by accumulation of solutes without interference with the central 314

metabolism of the cell. This mechanism can be explained by the energetic cost of 315

haloadaptation (Oren, 2001). Micro-organisms indeed spend a lot of energy to maintain 316

gradient of Na+ and K+ concentrations across their cytoplasmic membrane. 317

In high saline conditions, the adapted inoculum reachedhigher H2S production than the 318

inocula of industrial origin throughout the three feeds (Figure 2).The halophilic SRB showed 319

the best acclimation to process conditions (Figure 2). The sulphate reducer activity became 320

dominant and out-competed methanogens in conditions of high salinity since the H2S 321

produceded represent 85 % and 95% of the total biogas produced at, respectively, 75 and 150 322

g.L-1

after the third substrate addition. In marine environments, SRB out-compete 323

methanogens for common substrates such as H2 and acetate that are mainly used via sulfato-324

reduction (Ollivier et al., 1994). This can explain the greater activity of SRB in the case of the 325

adapted inoculum. 326

An acclimation of industrial-origin SRB in high saline conditions was also noted. Maximum 327

H2S yields were obtained at the extreme salinity of 75 and 75-150 g.L-1 for, respectively, the 328

non-acclimated and the acclimated inocula. In addition to sodium inhibition, competition 329

between methanogenesis and sulfato-reduction and sulfide inhibition were also present 330

An interesting additional observation was made at a salinity of 150 g.L-1

. In these conditions, 331

an accumulation of acetate was recorded for the three inocula during the successive substrate 332

additions (Table 2). At the end of the third anaerobic degradation tests for the non-acclimated, 333

acclimated and adapted inocula, 27.7%, 33.1% and 21.2%, respectively, of the total COD 334

introduced during the successive feeds was converted into acetate. Inocula of industrial origin 335

15

showed better performance than the adapated inoculum. This confirms that the salinity stress 336

involved a specific selection of salt-tolerant microorganisms which possess effective 337

mechanisms for compensating for the high external osmotic pressure within the cytoplasm. 338

These results highlight the strong acclimation potential of hydrolytic, acidogenesis and 339

acetogenesis bacteria and are consistent with Lim et al. (2008) who found that salinity has no 340

significant effect on hydrolysis and on acid-forming bacteria. 341

3.3 Adapted inoculum vs inocula of industrial origin 342

Despite a context of extensive ongoing global development of marine microalgal systems for 343

producing lipids, high-value products and food additives, few studies have beenwere devoted 344

to hypersaline conditions (Doan et al., 2012). The strategy presented in this study was to use a 345

suitable anaerobic inoculum to boost the performance of anaerobic digestion and energy 346

recovery despite high salt concentrations. Below a saline concentration of 35 g.L-1

, the 347

adapted inoculum, coming from a natural environment, showed good performance in terms of 348

CH4 production and its great capacity to acclimate. Above 75 g.L-1

of salinity, the CH4 yields 349

dropped and the impact became more pronounced with each substrate addition (i.e. with the 350

increase of exposure time). This threshold concentration of 35 g.L-1 corresponds to saline 351

conditions in the natural environment. Natural salinity can change during the seasons and this 352

implies a versatile microbial consortium able to adapt to variations in salt levels. Beyond 75 353

g.L-1

, methanogens require more energy to maintain homeostasis than anabolism and this 354

could explain the low CH4 productions (Oren, 1999). Therefore the halophilic methanogens, 355

used in this work, possess mechanisms which efficiently counterbalance the osmotic pressure 356

and lead to promising CH4 yields at a salinity of 35 g.L-1

, equivalent to seawater concentration 357

and composition. 358

In conditions of high salinity, inocula of industrial origin showed a better capacity to 359

acclimation than the adapted inoculum but this performance in terms of CH4 production was 360

16

not sufficient for a full-scale development. The microbial communities in these inocula may 361

have specific mechanisms such as a “compatible solutes” strategy to counterbalance within 362

the cytoplasm the high external osmotic pressure. However, the high saline conditions of 75 363

g.L-1

led to a strong selective pressure on methanogens of industrial origin. Such acclimation 364

may also result from internal changes in the predominant methanogen species or from a shift 365

in the methanogenic population with the selection of salt-tolerant microorganisms. 366

4 Conclusion 367

In order to develop a suitable strategy for converting organic residues from species of marine 368

microalgae, we evaluated anaerobic degradation performance of different anaerobic microbial 369

consortia over a large range of salinity were evaluated. 370

The results have shown that a halophilic sediment was able to produce CH4 efficiently up to 371

salinity conditions of 35 g.L-1

. After an acclimation period involving the three successive 372

additions of substrate, the performance was very close to the CH4 production in normal 373

condition at 345.8 ± 7.1 mLCH4.gVS-1

. 374

In conditions of high salinity (75 g.L-1

and above), only methanogens of industrial origin 375

showed a capacity of acclimation. 376

5 Acknowledgment 377

The authors are grateful for the financial support of the FUI Salinalgue project coordinated by 378

La Compagnie du Vent (Montpellier, France). 379

6 References 380

Aspé E., Marti M.C., Roeckel M., 1997. Anaerobic treatment of fishery wastewater using a 381

marine sediment inoculum. Water Res. 31(9):2147–2160. 382

17

Buffiere P., Loisel D., Bernet N., Delgenes J-P., 2006. Towards new indicators for the 383

prediction of solid waste anaerobic digestion properties. Water Sci. Technol. 53(8):233-384

241. 385

Chen Y., Cheng J.J., Creamer K.S., 2008. Inhibition of anaerobic digestion process: a review. 386

Bioresource Technol. 99:4044–4064. 387

Cheng J.J., Timilsina G.R., 2011. Status and barriers of advanced biofuel technologies: A 388

review. Renew. Energ. 36(12):3541–3549. 389

Colleran E., Pender S., 2002. Mesophilic and thermophilic anaerobic digestion of sulphate-390

containing wastewaters. Water Sci. Technol. 45(10):231–5. 391

Collet P., Hélias A., Lardon L., Ras M., Goy R-A., Steyer J-P., 2011. Life-cycle assessment 392

of microalgae culture coupled to biogas production. Bioresource Technol. 102(1):207–393

214. 394

Dimroth P., Thomer A., 1989. A primary respiratory Na+ pump of an anaerobic bacterium: 395

the Na+-dependent NADH:quinone oxidoreductase of Klebsiella pneumoniae. Arch. 396

Microbiol. 151(5):439–444. 397

Doan Q.C., Moheimani N.R., Mastrangelo A.J., Lewis D.M., 2012. Microalgal biomass for 398

bioethanol fermentation: Implications for hypersaline systems with an industrial focus. 399

Biomass Bioenerg. 46:79-88. 400

Federation W.E., 1999. Standard Methods for the Examination of Water and Wastewater 401

Standard Methods for the Examination of Water and Wastewater. 402

18

Feijoo G., Soto M., Méndez R., Lema J.M., 1995. Sodium inhibition in the anaerobic 403

digestion process: Antagonism and adaptation phenomena. Enzyme Microb. Tech. 404

17(2):180–188. 405

Jeison D., Kremer B., Van Lier J.B., 2008. Application of membrane enhanced biomass 406

retention to the anaerobic treatment of acidified wastewaters under extreme saline 407

conditions. Sep. Purif. Technol. 64(2):198–205. 408

Kapdan I.K., Erten B., 2007. Anaerobic treatment of saline wastewater by Halanaerobium 409

lacusrosei. Process Biochem. 42(3):449–453. 410

Kimata-Kino N., Ikeda S., Kurosawa N., Toda T., 2011. Saline adaptation of granules in 411

mesophilic UASB reactors. Int Biodeterior. Biodegrad. 65(1):65–72. 412

Lakaniemi A.M., Hulatt C.J., Thomas D.N., Tuovinen O.H., Puhakka JA., 2011. Biogenic 413

hydrogen and methane production from Chlorella vulgaris and Dunaliella tertiolecta 414

biomass. Biotechnol. Biofuels 4:34. 415

Lakaniemi A.M., Tuovinen O.H., Puhakka J., 2012. Anaerobic conversion of microalgal 416

biomass to sustainable energy carriers – a review. Bioresource Technol. 135:222-231. 417

Lefebvre O., Habouzit F., Bru V., Delgenes J.P., Godon J.J., Moletta R., 2004. Treatment of 418

hypersaline industrial wastewater by a microbial consortium in a sequencing batch 419

reactor. Environ. Technol. 25(5):543–553. 420

Lim Y.G., Niwa C., Nagao N., Toda T., 2008. Solubilization and methanogenesis of blue 421

mussels in saline mesophilic anaerobic biodegradation. Int Biodeterior. Biodegrad. 422

61(3):251–260. 423

19

Luo G., Li P., Tan H., Du J., Liang W., 2012. The start-up and saline adaptation of mesophilic 424

anaerobic sequencing batch reactor treating sludge from recirculating aquaculture 425

systems. Aquacult. Eng. 54:9-15. 426

Mendez R., Lema J.M., Soto M., 1995. Treatment of seafood-processing wastewaters in 427

mesophilic and thermophilic anaerobic filters. Water Environ. Res. 67(1):33-45. 428

Migliore G., Alisi C., Sprocati A.R., Massi E., Ciccoli R., Lenzi M., Wang A., Cremisini C., 429

2012. Anaerobic digestion of macroalgal biomass and sediments sourced from the 430

Orbetello lagoon, Italy. Biomass Bioenergy 42:69–77. 431

Müller V., Spanheimer R., Santos H., 2005. Stress response by solute accumulation in 432

archaea. Current Opin. Microbiol. 8(6):729–736. 433

Mussgnug J.H., Klassen V., Schlüter A., Kruse O., 2010. Microalgae as substrates for 434

fermentative biogas production in a combined biorefinery concept. J. Biotechnol. 435

150(1):51–6. 436

Muyzer G., Stams A.J.M., 2008. The ecology and biotechnology of sulphate-reducing 437

bacteria. Nat. Rev. Microbiol. 6:441–454. 438

Nolla-Ardèvol V., Strous M., Sorokin D.Y., Merkel A.Y., Tegetmeyer H.E., 2012. Activity 439

and diversity of haloalkaliphilic methanogens in Central Asian soda lakes. J. Biotechnol. 440

161(2):167–73. 441

Oh G., Zhang L., Jahng D., 2008. Osmoprotectants enhance methane production from the 442

anaerobic digestion of food wastes containing a high content of salt. J. Chem. Technol. 443

Biotechnol. 83(9):1204–1210. 444

20

Ollivier B., Caumette P., Garcia J.L., Mah R., 1994. Anaerobic bacteria from hypersaline 445

environments. Microbiol. Rev. 58(1):27–38. 446

Oren A., 1999. Bioenergetic Aspects of Halophilism. Microbiol. Mol. Biol. Rev. 63(2):334–447

348. 448

Oren A., 2001. The bioenergetic basis for the decrease in metabolic diversity at increasing salt 449

concentrations : implications for the functioning of salt lake ecosystems. Hydrobiologia 450

466:61–72. 451

Schenk P.M., Thomas-Hall S.R., Stephens E., Marx U.C., Mussgnug J.H., Posten C., Kruse 452

O., Hankamer B., 2008. Second Generation Biofuels: High-Efficiency Microalgae for 453

Biodiesel Production. Bioenergy Res. 1(1):20–43. 454

Sialve B., Bernet N., Bernard O., 2009. Anaerobic digestion of microalgae as a necessary step 455

to make microalgal biodiesel sustainable. Biotechnol. Adv. 27(4):409–416. 456

Subhadra B.G., Edwards M., 2011. Coproduct market analysis and water footprint of 457

simulated commercial algal biorefineries. Appl. Energy 88(10):3515–3523. 458

Visser A., 1995. The anaerobic treatment of sulfate containing wastewater. Thesis. 459

Wageningen Agricultural University, Netherlands. 460

Vyrides I., Stuckey D.C., 2009. Adaptation of anaerobic biomass to saline conditions: Role of 461

compatible solutes and extracellular polysaccharides. Enzyme Microb. Technol. 462

44(1):46–51. 463

21

Xia A., Cheng J., Lin R., Liu J., Zhou J., Cen K., 2012. Sequential generation of hydrogen 464

and methane from glutamic acid through combined photo-fermentation and 465

methanogenesis. Bioresour. Technol. 131:146-151. 466

Yerkes D., Boonyakitsombut S., Speece R., 1997. Antagonism of sodium toxicity by the 467

compatible solute betaine in anaerobic methanogenic systems. Water Sci. Technol. 36(6-468

7):15–24. 469

470

471

472

473

7 Figure captions 474

Figure 1. Three-way ANOVA representing specific methane production after three substrate 475

additions for the three anaerobic inocula at 15, 35, 75 and 150 g.L-1

of salinity (red dotted 476

lines represent the 95 % confidence interval). 477

Figure 2. Three-way ANOVA representing specific hydrogen sulfide production after three 478

substrate additions for the three anaerobic inocula at 15, 35, 75 and 150 g.L-1 of salinity (red 479

dotted lines represent the 95 % confidence interval). 480

Figure 3. Methane yield at 75 g.L-1 of salinity obtained in serum bottles A, B and C (each 481

condition was performed in triplicate) with the acclimated inoculum through the successive 482

substrate additions (the labels indicate the acetate concentration at the end of the batch test 483

expressed in g.L-1

). 484

485

22

8 Table captions 486

Table 1. Main physico-chemical characteristics and ionic composition of inocula and 487

substrate 488

Table 2. Acetate and propionate concentrations for the three successive substrate additions 489

and each inoculum at the end of each batch test. 490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

23

511

512

513

514

515

Figure 1. Three-way ANOVA representing, after three successive additions of substrate, 516

specific methane production for the three anaerobic inocula at 15, 35, 75 and 150 g.L-1 of 517

salinity (red dotted lines represent the 95 % confidence interval). 518

519

520

521

522

523

524

525

526

0

100

200

300

15 35 75 150 15 35 75 150

0

100

200

300

Non acclimated Acclimated Adapted

Fee

d1

Fee

d2

Fe

ed

3

Salinity (g.L-1)

CH

4yie

ld(m

LC

H4.g

VS

-1)

400

400

400

0

100

200

300

15 35 75 150

25

552

553

Figure 2. Three-way ANOVA representing, after three successive additions of substrate, 554

specific hydrogen sulfide production for the three anaerobic inocula at 15, 35, 75 and 150 g.L-555

1 of salinity (red dotted lines represent the 95 % confidence interval). 556

557

558

559

560

561

562

563

564

565

566

567

15 35 75 150 15 35 75 150

Non acclimated Acclimated Adapted

Feed

1F

ee

d2

Feed

3

Salinity (g.L-1)

H2S

yie

ld(m

LH

2S

.gV

S-1

)

15 35 75 1500

20

40

60

0

20

40

60

0

20

40

60

26

568

569

570

571

572

573

574

575

Figure 3. Methane yield at 75 g.L-1

of salinity obtained in serum bottles A, B and C (each 576

condition was performed in triplicate) with the acclimated inoculum through the successive 577

substrate additions (the labels indicate the acetate concentration at the end of the batch test 578

expressed in g.L-1). 579

580

581

582

583

584

585

586

0,90,8 0,2

0,0

1,7 1,0

0,0

0,0

0,0

0

50

100

150

200

250

300

350

400

450

Feed 1 Feed 2 Feed 3

CH

4yie

ld (

mL

CH

4.g

VS

-1)

Bottle A Bottle B Bottle C

27

587

Table 1. Main physico-chemical characteristics and ionic composition of inocula and 588

substrate 589

590

Adapted

inoculum

Acclimated

inoculum

Non-

acclimated

inoculum

Substrate

Total solids gTS.L-1

259.0 ± 0.5 57.7 ± 0.1 19.2 ± 0.3 169.3 ± 0.2

Inorganic solids gIS.L-1

222.5 ± 0.9 9.3 ± 0.1 7.5 ± 0.1 100.3 ± 0.1

Volatile solids gVS.L-1

18.3 ± 0.7 36.5 ± 0.4 11.6 ± 0.2 69.1 ± 0.2

Volatile

suspended

solids

gVSS.L-1

18.3 ± 0.7 46.9 ± 0.1 10.5 ± 0.1 ND*

Conductivity mS.cm-1

46.6 8.7 7.8 97.0

Salinity g.L-1

33.0 9.8 4.3 104.8

Sodium gNa+.L

-1 9.02 3.85 0.14 38.0

Sulphate gSO42-

.L-1

2.48 0.0 0.0 1.94

Ammonium gNH4+.L

-1 0.02 0.49 1.05 0.49

Density g.cm-3

1.15 ± 0.04 1.01 ± 0.02 1.00 ± 0.01 1.08 ± 0.01 *: Not determined 591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

606

28

Table 2. Acetate and propionate concentrations for the three successive substrate additions 607

and each inoculum at the end of each batch test. 608

609

Non-acclimated inoculum

Salinity C2 concentration

C3 concentration

Mean SD Mean SD

g.L-1

gCOD.L-1

gCOD.L-1

Fee

d 1

15 0.06 0.02 0.00 0.00

35 0.09 0.02 0.14 0.04

75 0.37 0.27 0.04 0.00

150 0.17 0.07 0.00 0.00

Fee

d 2

15 0.03 0.00 0.14 0.01

35 0.08 0.02 0.18 0.01

75 0.09 0.01 0.08 0.03

150 0.77 0.26 0.08 0.05

Fee

d 3

15 0.01 0.02 0.00 0.00

35 0.11 0.04 0.17 0.11

75 0.16 0.06 0.05 0.07

150 2.03 0.27 0.21 0.01

Acclimated inoculum

Fee

d 1

15 0.00 0.00 0.00 0.00 35 0.00 0.00 0.00 0.00

75 0.68 0.40 0.00 0.00 150 0.42 0.13 0.15 0.11

Fee

d 2

15 0.00 0.00 0.00 0.00

35 0.00 0.00 0.01 0.01

75 0.96 0.90 0.00 0.00

150 0.70 0.63 0.18 0.19

Fee

d 3

15 0.00 0.00 0.00 0.00

35 0.00 0.00 0.08 0.07

75 0.01 0.03 0.00 0.00

150 2.42 0.42 0.08 0.05

Adapted inoculum

Fee

d 1

15 0.01 0.01 0.00 0.00

35 0.00 0.00 0.00 0.00

75 0.00 0.00 0.00 0.00

150 0.34 0.23 0.00 0.00

Fee

d 2

15 0.00 0.00 0.00 0.00

35 0.04 0.00 0.00 0.00

75 0.00 0.00 0.00 0.00

150 0.94 0.20 0.00 0.00

Fee

d 3

15 0.00 0.00 0.00 0.00 35 0.00 0.00 0.00 0.00

75 0.17 0.03 0.00 0.00 150 1.55 0.17 0.03 0.01

610

611

612

29

613

� Boost the anaerobic conversion into CH4 of saline microalgae organic residues 614

� Three different anaerobic microbial consortia were tested in high saline condition 615

� Most promising CH4 production for halophilic methanogens at 35 g.L-1 of salinity 616

� Hydrolytic, acidogenic and acetogenic activities at 150 g.L-1

of salinity 617

618

619

620

621

622

623

624

625

626

627

628

629

630

631

632

633