Bio-hydrogen production by co-digestion of domestic wastewater and biodiesel industry effluent

The increasing water crisis makes fresh water a valuable resource, which must be used wisely. However, with growing population and inefficient waste treatment systems, the amount of wastewater dispelled in rivers is increasing abominably. Utilizing this freely available waste-water along with biodiesel industry waste- crude glycerol for bio-hydrogen production is being reported here. The bacterial cultures of Bacillus thuringiensis strain EGU45 and Bacillus amyloliquefaciens strain CD16 produced2.4–3.0 L H2/day/L feed during a 60 days continuous culture system at hydraulic retention time of 2 days. An average H2 yield of 100–120 L/L CG was reported by the two strains. Recycling of the effluent by up to 25% resulted in up to 94% H2 production compared to control.


Introduction
Availability of clean water is a worldwide crisis. Despite our earth surface being covered with 70% of water, only 2% is a freshwater, of which 3/4th is frozen and unavailable for human consumption [1,2]. Thus, billions of people live with severe water scarcity and poor sanitation. The small amount of available fresh water faces an allocation and competition in agricultural, industrial and municipal sectors. As a result, allocating this sparsely available fresh water to bioenergy production is a very costly affair [3]. During bioenergy production, the substrate occupies 10% of the medium while the rest is water. This water used in most of the studies is distilled and the medium is sterilized [4][5][6]. Since most of the population struggles for fresh water for their daily basic needs, it would be unethical to divert it towards increasing energy demands. The possible solution would be to use wastewater that is generated from domestic and industrial sources. As per the 2016 report published by International Institute of Health and Hygiene, in metro cities like New Delhi (India) about 6.1×10 4 million liters (ML) of wastewater is generated every day. The treatment capacity is around 50% only (http://www. sulabhenvis.nic.in/Database/STST_wastewater_2090.aspx). The rest of the wastewater is drained into the rivers or can meet a less dreadful fate if utilized e.g., for bioenergy production. a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 The surplus availability of wastewater makes it a timeless resource for the researchers struggling with cheap and steady bioenergy generation [7].
Bioenergy being a sustainable alternative to fossil fuels has attracted a huge worldwide support. Hydrogen (H 2 ), Methane (CH 4 ), ethanol, bio-diesel are amongst the most widely studied bio-fuels. H 2 however has gained immense favors owing to its high calorific value and cleaner combustion [8,9]. Most extensively studied technique for biological H 2 production is dark fermentation and is most likely to be commercialized in near future [10]. A variety of organic wastes have been used successfully as substrate for H 2 production. This substrate is mostly present along with some minerals in distilled water [11]. The challenge is thus to replace the valuable clean water with readily available domestic wastewater for H 2 production. H 2 production from various industrial wastewaters such as cassava starch processing wastewater, brown sugar wastewater, paperboard mill wastewater, ethanol wastewater, etc. have been reported [12][13][14][15][16][17][18]. Sugar rich wastewaters such as molasses wastewater, sugarbeet wastewater and sugarcane vinasse have the ability to produce high H 2 yields of around 3.2 mol/mol substrate. Starchy wastewaters generally result in relatively lower H 2 yields of around 1.9 mol/mol substrate [17,19,20]. Wastewaters from biodiesel industry, which are rich in glycerol also have a potential to produce bioenergy [21,22]. Use of crude glycerol (CG) as feed prepared in distilled water resulted in 165 L H 2 /L CG by an immobilized biofilm forming bacteria B. amyloliquefaciens [23]. The use of different industrial wastewaters as medium may not be available throughout the year and may thus hinder the continuity of bio-H 2 production. In contrast, domestic wastewater which is generated everyday throughout the globe may be a better option to counter this problem. Therefore, in the present study we have used freely available domestic wastewater as the medium and biodiesel industry waste-CG as the substrate for an economic bioenergy generation. To further improve the production efficiency, recycling of the effluent is also reported.

Organisms and growth conditions
Hydrogen producers. Bacillus amyloliquefaciensstrain CD16 (KX348272) and B. thuringiensis strain EGU45 (DQ508971) were isolated in our laboratory [23]. These were grown on Himedia nutrient broth (NB) at 37˚C with stirring at 200 rpm for 16 h. The media also contained CG (2%, v v -1 ) and the cultures were thus adapted to the substrate for 5 cycles. The cultures were then used as inoculum at the rate of 10 μg cellular protein mL -1 [6].

Hydrogen production
Immobilization of cells on lignocellulosic support material. Coconut coir (CC) was dried and packed in PVC tubes to prepare cartridges (3 x 2 cm) containing 3 g coir each, as reported earlier [6]. These cartridges were used as support material for bacterial immobilization. Aspirator bottles (1.2 L) with working volume of 1.0 L were used to perform the experiments. In order to allow the growth of biofilm on cartridges, casein enzyme hydrolyzate (CEH) was used as a biofilm forming media [21]. In bottles containing different amounts of cartridges (5-15%, vv -1 ), 100 mL of CEH was added. Each cartridge occupied~10 mL of the 1.0 L working volume used. Thus, 5, 10 and 15 cartridges were used for 5%, 10% and 15% CC reactors. Free floating (FF) cultures were used as controls. Inoculation with CG acclimatized cultures was done at the rate of 10 μg cell protein mL -1 . Anaerobic conditions were maintained by flushing the reactors with argon gas. The bottles were incubated at 37˚C for 24 h without shaking, to allow biofilm formation. Domestic waste water diluted with tap water in the ratio of 3:1 was used as H 2 producing media (DWW). Minimal salts (KH 2 PO 4 -1.5 g/L, NaCl-0.25 g/L, NH 4 Cl-0.5 g/L, MgSO 4 (1.0 M)-0.5 mL/L, CaCl 2 (1.0 M)-0.5 mL/L) were added to the DWW media. After 24 h biofilm growth on cartridges in aspirator bottles, DWW containing CG (2%, v v -1 ) was used to complete the remaining working volume.
Batch culture. The aspirator bottles containing hydrogen producers immobilized on CC (5-15%, vv -1 ) in CG supplemented DWW media were made air tight using glass stoppers after adjusting the pH to 7.0. The pH was adjusted using NaOH (2.0 N) or HCl (2.0 N) after which argon flushing was given to maintain anaerobic environment inside the aspirator bottles. These were incubated at 37˚C. A provision for gas outlet and liquid sampling was provided in the bottles. On a daily basis, the gas evolved was collected and analyzed. Adjustment of pH and argon flushing was also done on a daily basis until the gas production ceased. After this the fermentation was switched to continuous mode.
Continuous culture. For the continuous culture digestion, a hydraulic retention time (HRT) of 2 days was used. On a daily basis, 500 mL of the effluent was removed from each reactor and was replenished with fresh DWWTW media containing CG (2%, v v -1 ). Adjustment of pH and argon flushing was done daily, and the evolved gas was analyzed. Incubation was done at 37˚C and the process was continued for 60 days to obtain steady gas production. The experiments were performed in triplicates.
Recycling of effluent. During the continuous H 2 production effluent was generated daily. This effluent was further used for H 2 production by mixing it with fresh DWW medium indifferent ratios: (i) 1:3, (ii) 1:1, and (iii) 3:1. The gas production was compared with the controls and the process was continued for an additional 60 days. The support material used in all these reactors contained 15%(v v -1 ) CC.

Analytical methods
Gas analysis. Water displacement method was used to determine the volume of biogas produced. The composition of gas was analyzed using gas chromatography (Nucon GC5765, India) equipped with molecular sieve and Porapak-Q columns (1.8 m long and 2 mm inner diameter) and a thermal conductivity detector, as reported earlier [6,24]. For the daily fed culture experiments, H 2 yields were calculated on the glycerol fed basis.
Glycerol estimation. The amount of residual glycerol in the fermented medium was estimated by taking 1 mL of the sample. It was centrifuged at 10,000 g for 5 min. Supernatant (1μL) was injected into Gas chromatograph (Nucon GC5765, India) and analyzed under standard conditions as described earlier [6].

Results and discussion
The effectiveness of biofilm forming B. amyloliquefaciens strain CD16 for a high and steady continuous H 2 production has been reported [23]. However, the medium used in these studies is sterile distilled water. This increases the production cost for bio-energy generation. The medium thus used in present study is unsterile domestic waste diluted with tap water.

Cell immobilization
Several support materials have been widely used for bacterial immobilization. These may include activated carbon, alginate gel, polyester fiber, porous glass beads, egg shells and lignocellulosic wastes such as banana leaves, groundnut shells, coconut coir, bamboo stem, etc. [25][26][27][28]. Apart from these, biofilms as natural cell entrapment strategy has also gained importance. Biofilms although have been extensively utilized for bioremediation is also gaining interest with bioenergy production [29,30]. With the availability of medium that can screen biofilm formers, biofilm forming H 2 producers have been isolated [21,23]. One of these biofilm formers has been utilized for H 2 production using wastewater in the present study. After a 24 h incubation, in the reactors inoculated with B. amyloliquefaciens strain CD16, biofilm formation was observed on CC cartridges. While in case of B. thuringiensis strain EGU45 no biofilm was formed. The cells immobilized in biofilm are resistant to environmental stresses and thus, may provide a better robust environment for gas production.

Batch culture H 2 production
The total biogas produced during 5 days batch fermentation by B. thuringiensis strain EGU45 ranged from 2.0 L to 3.1 L. The biogas constituted a mixture of H 2 and CO 2 . The H 2 in the produced gas constituted 56.2-70.2%. With B. amyloliquefaciens strain CD16, 2.4 L-3.3 L biogas was produced which consisted of 58.3-60.0% H 2 ( Table 1). When comparing the H 2 yield, 55 L H 2 /L CG to 110 L H 2 /L CG was produced by B. thuringiensis strain EGU45 while with B. amyloliquefaciens strain CD16, 70 L H 2 /L CG-100 L H 2 /L CG was obtained. The results obtained with DWWTW were strikingly similar to that obtained with sterile M-9 medium [23]. This shows that there are no deteriorating effects of using unsterile waste water as medium for biogas production under batch conditions.

Continuous culture H 2 production
For an economical large-scale bioenergy production, continuous culture fermentation is required. However, despite being an ideal system for higher product yields, cell washout is a major concern of this mode of fermentation. To deal with the problem, a number of bacterial support materials have been utilized. Recently, 1.18-fold increase in H 2 production by using biofilm immobilized on lignocellulosic wastes has been reported [23].
Similar strategy when applied to prevent cell washout from waste water medium in the current work resulted in encouragingly higher H 2 yields during 60 days continuous fermentation. Without any cell immobilization, i.e. FF conditions, the biogas production showed a significant decline with both the strains (Figs 1 and 2, and S1 and S2 Tables). In case of B. thuringiensis strain EGU45, from an average of 0.6 L H 2 / 0.5 L feed/day during initial 10 days of fermentation, the production declined to 0.06 L H 2 / 0.5 L feed/day at the end of 30 days of fermentation. The gas production thereafter became so low that the reactors had to be terminated ( Table 2). Similar was the case observed with biofilm forming B. amyloliquefaciens strain CD16, where the production declined from 0.8 L H 2 / 0.5 L feed/day to 0.07 L H 2 / 0.5 L feed/ day during 30 days of fermentation and ceased thereafter ( Table 2). Effect of support material on biogas production could be clearly seen with its increasing quantity from 5-15% CC (vv -1 ). At 5% CC, non-biofilm former B. thuringiensis strain EGU45 produced 0.7 L H 2 / 0.5 L feed/day for 30 days after which it reduced to 0.4 L H 2 / 0.5 L feed/day and continued till 60 days maintaining a stable yield of 0.16-0.18 mol H 2 / mol CG. On increasing the support material to 10% CC, 0.8 L H 2 / 0.5 L feed/day was observed during initial 10 days of continuous culture. It increased thereafter and maintained a stable value of around 0.95-1.2 L H 2 / 0.5 L feed/day during 60 days fermentation. The average H 2 production with 10% CC was 2.28-fold higher than with 5% CC ( Table 3). On increasing the support material to 15% CC, a higher and stable H 2 of 1.2-1.3 L/ 0.5 L feed/day was produced during 60 days continuous fermentation. This corresponds to 0.48-0.53 mol H 2 /mol CG which was 1.23-fold higher than with 10%CC. Similar effect of increasing gas production with support material was also seen with biofilm former B. amyloliquefaciens strain CD16. At 5% CC, 1.0-1.2 L H 2 / 0.5 L feed/day was produced during 0-30 days of fermentation which achieved a steady 0.7-0.8 L H 2 / 0.5 L feed/day during 31-60 days of fermentation. At 10% CC, the gas production maintained a steady biogas production constituting 61.1-62.9% H 2 throughout 60 days of fermentation. The amount of H 2 varied from 1.1-1.3 L/0.5 L feed/day. This was 1.55-fold higher  than with 5% CC. On increasing the support material to 15% CC, 1.2-1.5 L H 2 / 0.5 L feed/day was produced during 60 days of fermentation, which was 1.23-fold higher than with 10% CC.
On comparing the H 2 producing abilities of two strains, without any support material both the strains produced an abysmal low H 2 (Figs 1 and 2). Immobilizing B. thuringiensis strain EGU45 on CC increased the H 2 production by 4-to 10-fold. With biofilm forming B. amyloliquefaciens strain CD16, the increase in H 2 production with CC was 3-to 5-fold. Biofilm forming strain at 15% CC resulted in 1.17 times more H 2 as compared to non-biofilm forming strain ( Table 3).

Effect of effluent recycling
To increase the overall process efficiency and economy, effluent generated from the H 2 production stage was recycled. With, B. thuringiensis strain EGU45, at 75% and 50% effluent recycling a very sharp decline in gas production was recorded (Figs 3 and 4 and S3 and S4  Tables). After an initial production of 0.6-1.0 L H 2 / 0.5 L feed/day in these cases, the H 2 production declined to 0.09-0.3 L H 2 / 0.5 L feed/day and ceased thereafter. However, at 25% effluent recycling a significant difference was not observed with respect to control during 60 days of fermentation. During initial 20 days of recycling, 1.2-1.4 L H 2 / 0.5 L feed/day was observed.
After this a small decline in gas production was observed which maintained an average of 0.9 LH 2 / 0.5 L feed/day till 60 days of recycling ( Table 2). Considering the average gas produced, a 6% drop in H 2 was observed with 25% effluent recycling as compared to controls This corresponded to 0.36 mol H 2 /mol CG as compared to controls which produced 0.41 mol H 2 /mol   Table 3). With B. amyloliquefaciens strain CD16, a similar trend of sharp decline in gas production was observed at 75% and 50% recycling. The gas production declined to 0.2 L H 2 / 0.5 L feed/day and 0.05 L H 2 / 0.5 L feed/day at 75% and 50% recycling respectively within 40 days of recycling and ceased thereafter. However, at 25% recycling the gas production did not show any significant reduction. An average of 1.0 L H 2 / 0.5 L feed/day was produced with 25% recycling which was a 10% decline as compared to controls ( Table 3).The reactors with 15% CC support material were run for 120 days (an additional 60 days during recycling of effluent) for the entire duration of which an average H 2 yield of 100-120 L/L CG was maintained by both the strains (Tables 2 and 3).

Conclusion
Waste generation is an integral part of our routine activities. Under natural environmental conditions, microbes metabolize the organic matter content and release gases into the atmosphere. This contributes significantly to environmental pollution. Another contributor to environmental pollution is the burning of fossil fuels. Efforts to treat biowastes through microbial activity have revealed that this concept can be exploited to produce energy rich gases (H 2 and methane, CH 4 ) through fermentation. A wide range of biowastes have the potential to produce H 2 and CH 4 . A major limitation in the use of H 2 producers is the risk of contamination which emanates from bacteria present in the unsterile biowastes. So to avoid contamination, sterilization become imperative, this obviously results in lowering economic efficiency. Secondly, in most biological processes, the substrate concentrations vary from very low of 0.1% to a maximum 10%. It implies that 90 to 99.9% is water or medium. In this study, we have circumvented almost all the issues related to biological H 2 production: (i) use of unsterile conditions, (ii) use of sewage water as medium, (iii) use crude glycerol, which otherwise cause heavy pollution, (iii) use of a single bacterium with abilities to form biofilm and produce H 2 , (iv) recycling of the effluent for further enhancing the process efficiency, (v) continuous culture conditions enable easy operation (vi) no stirring required, (vii) independent of light, and (viii) used organism i.e. Bacillus, which has been categorized as GRAS (Generally Regarded As Safe) organism [31].  biogas [32]. Using activated sludge from biodiesel industry effluent, 75L H 2 /L glycerol consumed was reported in anaerobic sequencing batch reactors, with H 2 constituting only 33.4% of the total biogas produced [22]. Further, we have observed that on recycling the effluent up to 25%, no drastic changes in gas production were observed. In comparison to the use of sterile distilled water as medium, the use of wastewater did not result in any adverse effect on H 2 production. A similar response on recycling of effluent from H2 production stage was shown in our previous study, where sterile distilled water was used for preparing the slurry [23]. Another interesting aspect of this study is the possibility of further utilization of effluent of H 2 production stage for producing value added products such as polyhydroxyalkanoates and CH 4 [14,18].
Supporting information S1