- Email: [email protected]
Application of immobilized upflow anaerobic sludge blanket reactor using Clostridium LS2 for enhanced biohydrogen production and treatment efficiency of palm oil mill effluent
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/he
Application of immobilized upflow anaerobic sludge blanket reactor using Clostridium LS2 for enhanced biohydrogen production and treatment efficiency of palm oil mill effluent Lakhveer Singh a, Zularisam A. Wahid b,*, Muhammad Faisal Siddiqui c, Anwar Ahmad d, Mohd Hasbi Ab. Rahim a, Mimi Sakinah c a
Faculty of Industrial Sciences & Technology, Universiti Malaysia Pahang (UMP), Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia b Faculty of Civil Engineering and Earth Resources, Universiti Malaysia Pahang (UMP), Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia c Faculty of Chemical and Natural Resource Engineering, Universiti Malaysia Pahang (UMP), Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia d Department of Civil Engineering, King Saud University (KSU), PO Box 800, Riyadh 11421, Saudi Arabia
article info
abstract
Article history:
Polyethylene glycol (PEG) gel was used to immobilize hydrogen producing Clostridium LS2
Received 12 June 2012
bacteria for hydrogen production in an upflow anaerobic sludge blanket (UASB) reactor.
Received in revised form
The UASB reactor with a PEG-immobilized cell packing ratio of 10% weight to volume ratio
21 November 2012
(w/v) was optimal for dark hydrogen production. The performance of the UASB reactor
Accepted 2 December 2012
fed with palm oil mill effluent (POME) as a carbon source was examined under various
Available online 29 December 2012
hydraulic retention time (HRT) and POME concentration. The best volumetric hydrogen production rate of 365 mL H2/L/h (or 16.2 mmol/L/h) with a hydrogen yield of
Keywords:
0.38 L H2/g CODadded was obtained at POME concentration of 30 g COD/L and HRT of 16 h.
Hydrogen
The average hydrogen content of biogas and COD reduction were 68% and 65%, respec-
Polyethylene glycol (PEG)
tively. The primary soluble metabolites were butyric acid and acetic acid with smaller
Immobilization
quantities of other volatile fatty acid and alcohols formed during hydrogen fermentation.
Upflow anaerobic sludge blanket
More importantly, the feasibility of PEG-immobilized cell UASB reactor for the enhance-
(UASB) reactor
ment of the dark-hydrogen production and treatment of wastewater is demonstrated.
Palm oil mill effluent (POME)
Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
The joint challenges of environmental crises and dwindling fossil fuel supplies are driving intensive research focus in
alternative energy production. Hydrogen is widely regarded as one of the most potential future energy vector, capable of assisting in issues of environmental emissions, energy security and versatility as fuel [1]. Despite the “green” nature of hydrogen
* Corresponding author. Tel.: þ60 95493002. E-mail addresses: [email protected], [email protected] (Z.A. Wahid). 0360-3199/$ e see front matter Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijhydene.2012.12.004
2222
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
Abbreviations PEG POME UASB COD HRT w/v SEM TN VSS TS
polyethylene glycol palm oil mill effluent upflow anaerobic sludge blanket chemical oxygen demand hydraulic retention times weight to volume scanning electron microscope total nitrogen volatile suspended solids total solid
as a fuel, it is still usually produced via steam reforming of natural gas, petroleum hydrocarbon, nonrenewable materials and other hydrogenation reactions, which makes hydrogen production environmentally unfriendly and expansive [2]. Among the established hydrogen production technologies, anaerobic fermentation has received considerable attention due to its potential as an inexhaustible, carbon-neutral and cost-effective fuel but has yet to reach a scale large enough for consideration in replacing a major portion of the hydrogen supply [3,4]. A wide variety of organic wastes or renewable substrates could be used as carbon source for biological hydrogen generation [5,6]. Moreover, fermentative hydrogen production using wastewater and organic waste as substrates achieves both energy recovery and bioremediation. In Malaysia, the estimated annual production of palm oil mill effluent (POME) is about 50 million tons. POME is an important renewable biomass energy source that can be harmful to the environment if untreated POME is discharged directly to the surroundings, due to high values of COD and biochemical oxygen demand (BOD) that it generates [7,8]. The high nutrient content in POME makes it an ideal fermentation medium for anaerobic treatment processes [9]. Previous reports have utilized POME as a substrate for hydrogen production with defined microflora, or mixed cultures of POME sludge under continuous mode operation [10e14]. Yusoff et al. [12] found that the maximum biohydrogen yield was 1054 mL/ L-POME and the maximum hydrogen production rate was 44 N mL/h/L POME at a pH of 5.5, a temperature of 22e26 C, a substrate concentration of 50e60 g/L COD-POME, and a HRT of 48 h in a 50-L continuously stirred tank reactor. Badiei et al. [13] evaluated hydrogen production from POME in an anaerobic sequencing batch reactor using enriched mixed microflora from POME sludge, and obtained maximum hydrogen production rate of 6.7 LH2/L/d with a total COD removal of more than 37% at a hydraulic retention time (HRT) of 3 d, an organic loading rate (OLR) of 6.6 g COD/L/d, a pH of 6.8, and a temperature of 37 C. Prasertsan et al. [14] obtained a maximum hydrogen production rate of 9.1 L H2/L/d along with a COD removal of 57% at a controlled pH 5.5, a temperature of 60 C, and 60 g COD/L-POME/d OLR and optimum values of 48 h HRT in the anaerobic sequencing batch reactor. However, these studies on biohydrogen production from POME only focused on suspended cell-systems, which are usually ineffective or difficult to handle in continuous operation
GC BOD TP TN VFAs HBu HAc BuOH EtOH SMPs TVFA HPr
Gas chromatograph biochemical oxygen demand total phosphorus total nitrogen volatile fatty acids butyric acid acetic acid butanol ethanol soluble microbial products total volatile fatty acid propionic acid
because washout of bacteria with effluents may occur from the reactor at shorter HRTs [13] and result in a low biomass concentration in the bioreactor. Recycling of biomass is considered necessary to maintain a sufficient cell concentration in the reactor to maximize hydrogen production [14]. Another potential approach to enhance hydrogen production is to use an immobilized cell system [15]. Immobilized cells offer distinct advantages over suspended cells, because they are resistant to cell wash-out during continuous operation and can maintain a higher cell density that increases hydrogen production. Immobilized cells have been successfully used for continuous hydrogen production in a bioreactor [16e19]. Many different methods have been employed for biomass immobilization including entrapment in polymeric gel and adsorption to solid surfaces, granules and biofilms [20e23]. The primary difficulty of the immobilized system is leak-out. However, it allows better biomass retention at low HRTs and creates a local anaerobic environment, which is well-suited to fermentative hydrogen production. In this study, a new cell immobilization system that employs polyethylene glycol (PEG) gel was employed to immobilize Clostridium LS2 for hydrogen production and treatment of POME in a UASB reactor. The effect of packing ratio of immobilized cells, HRT and POME concentration on the hydrogen producing and treatment capability of the immobilized UASB reactor was examined. The information obtained here is expected to be useful in future development of effective immobilized UASB reactor hydrogen producing systems for large scale application.
2.
Materials and methods
2.1.
Fermentation medium
Raw POME was collected from the final discharge point of an oil palm mill, Lepar Hilir Pahang, Malaysia, and was used as the substrate for hydrogen production. Raw POME has brown color, temperature of 65e75 C, pH 4.0e4.5 and water content of 85e90%. POME was preserved at 4 C in a 25-L container to prevent biodegradation and acidification before use. Prior to being fed into the reactor, the POME was diluted to a required COD concentration for fermentation. The characteristics of the POME wastewater are given in Table 1.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
Table 1 e Characteristics of POME. Parameter Biochemical oxygen demand (BOD) Chemical oxygen demand (COD) pH Total carbohydrate Total nitrogen (TN) Ammoniumenitrogen Total phosphorus (TP) Phosphorus Oil Total solid (TS) Volatile suspended solids (VSS)
Concentration 23,100e55,200 55,100e86,300 4.0e5.0 16,200e20,000 820e910 25e35 95e120 14e20 2000e2500 30,000e42,000 8400e12,000
All values are in mg/L except pH.
2223
20 mL of inoculum was approximately 2.2 g. To start polymerization, 0.25% (w/v) potassium persulfate initiator was mixed in and the mixture allowed to stand for about 80 min to promote bead formation. The immobilized-cell polymer was cut into 3-mm bead particles (density 1.40 g/cm3). The biomass content of the immobilized bead was w10 mg cell/g bead. Prior to use, the immobilized cells were stored in physiological saline solution for 2 h and then washed thoroughly with distilled water. The scanning electron microscopy (SEM) profiles of immobilized beads are shown in Fig. 1.
2.4. UASB reactor setup and operation for hydrogen production
The hydrogen producing strain Clostridium LS2 was used as the inoculum in this study. Clostridium LS2 was isolated from POME sludge taken from an anaerobic digester in the palm oil mill of Lepar Hilir Pahang, Malaysia. The sludge sample was heat-treated at 90 C for 30 min to inhibit methanogenic activity [24]. The heat-treated sludge was diluted with distilled water, and then cultured in medium under anaerobic conditions using an anaerobic jar at 36 C. The composition of the culture medium was (per L): 0.5 g of starch, 10 g of meat extract, 4 g of glucose, 2.5 g of yeast extract, 2.0 g of peptone, 0.5 g of agar, 0.5 g of L-cysteine$HCl$H2O, 3 g of NaCl, pH 6.5 (adjusted with 1 M NaOH). The Clostridium strain was obtained from a pure culture and designated as LS2. Biochemical identification by the rapid ANA II microtest system indicated that the strain belonged to the genus Clostridium. The isolated strain was stored in sterile 15% (v/v) glycerol solution at 30 C before being subjected to immobilization.
A stainless steel laboratory-scale UASB reactor (5126 cm3) with 5 L working volume was used in this study (Fig. 2). The UASB reactor was operated at temperature 37 C and temperature was maintained by hot water circulation through the water jacket. Throughout the experiments, the pH of the reaction medium in the reactor was adjusted around 5.5 by adding 1 M NaOH or 1 M HCl solutions. The medium was fed into the UASB reactor together with PEG-immobilized cells. Sampling points were introduced at appropriate heights in the reactor. A gaseliquid separator was introduced at the top of the reactor, where the biogas and soluble microbial product (SMP) collected separately. The reactor was purged with nitrogen gas for 10 min to create anaerobic conditions. The UASB reactor was loaded with an appropriate amount of PEG-immobilized cells to obtain a final solution of 4e16% (w/v). The UASB reactor was operated on batch mode for 6 h before being switched to continuous mode at a HRT of 32e8 h and a POME concentration of 10e40 g COD/L. The quantity and composition of the biogas, COD removal efficiency (%), POME conversion efficiency (%), volatile fatty acids (VFAs), pH, and temperature were monitored at designated time intervals.
2.3.
2.5.
2.2.
Microorganisms
Immobilization of cells in PEG
Clostridium LS2 cells were immobilized by entrapment in a PEG prepolymer. The cells of anaerobic bacteria were harvested by centrifugation at 8000 rpm for 12 min. A culture sample was heated at 80 C for 5 min prior to use for immobilization. First, 10% (w/v) PEG and 0.6% (w/v) N,N0 -methylenebisacrylamide (MBA) crosslinker were dissolved in a water. The resulting mixture and 20 mL of inoculum (in exponential growth phase) were quickly mixed in a beaker. The dry weight of cells in
Assay methods
Hydrogen gas was measured by a gas chromatograph (GC 8500 Perkin Elmer) equipped with a thermal conductivity detector and a 2-m stainless-steel SS350A column packed with a molecular sieve (80/100 mesh) using nitrogen as a carrier gas at a flow rate of 25 mL/min. VFAs contents of filtered sample (0.2 mm) were determined by gas chromatography with a FID detector (model 6890N, Agilent Inc., glass 2 m 2 mm packed column Carbopack B-DA 80/120% CW 20 M, N2 carrier at kPa,
Fig. 1 e SEM images of the PEG immobilized cell bead: (a) size and shape of the immobilized beads; (b) peripheral surface of the immobilized beads. Scale bar: 10 mm; (c) cross-sectional image of the immobilized beads. Scale bar: 2 mm.
2224
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
Fig. 2 e Experimental setup of immobilized cell UASB reactor: PHT e POME holding tank; PP e peristaltic pump; FM e flow meter; MV e manual valve; M e mixture; IB e immobilized beads; SV e sampling valve; BPT e biogas purification tank; WT e water tank; TS e temperature sensor; HP e heating probe; P e pump.
170 C). For alcohol analysis, 1 cm3 of sample acidified with 0.003 cm3 25% H2SO4, was analyzed using GCeFID and capillary column of the same model and type, respectively. COD, total nitrogen (TN), total solid (TS), VSS, and pH were carried
out using standard methods [25]. The dry weight of immobilized cell in immobilized beads was assessed by measuring the difference in dry weight between the biomass-associated beads and the beads alone.
Table 2 e Effect of various packing ratios of PEG-immobilized cells in UASB reactor on hydrogen production (POME concentration: 20 g COD/L). Packing ratio of PEG-immobilized cells (w/v)%
HPRa (mL H2/L/h)
HYb (L H2/g CODadded)
H2 content in biogas (%)
COD removal (%)
POME conversion (%)
32 24 16 8
4
292 305 310 317
0.275 0.260 0.259 0.251
65 66 64 66
59 60 61 60
93 94 85 69
32 24 16 8
8
302 310 323 331
0.313 0.299 0.283 0.272
65 66 67 66
58 60 61 60
94 92 82 68
32 24 16 8
12
325 337 342 345
0.350 0.341 0.327 0.312
67 68 68 67
62 61 62 62
94 91 84 70
32 24 16 8
16
294 299 276 268
0.319 0.283 0.246 0.235
62 60 61 59
58 57 57 56
93 92 82 68
HRT
a Hydrogen production rate. b Hydrogen yield.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
2225
2.6. Scanning electron microscopy and biochemical analysis The immobilized cells were gently washed with distilled water. Immobilized cells were then fixed with 2% glutaraldehyde and left for 2 h. The fixed immobilized cells were dehydrated by successive passages through 50%, 60%, 70% and 90% ethanol for 20 min and then dried. The dried samples were covered with a layer of gold under vacuum prior to being subjected to SEM (Zeiss EVO50, Germany). The rapid ANA II microtests (Remel) for anaerobic bacteria were utilized for biochemical identification [26].
3.
Results and discussion
3.1. Effect of packing ratio of the PEG-immobilized cell concentrations on hydrogen production To determine the biomass loading for higher hydrogen production and stabilize the operations in the UASB reactor, a suitable amount of the PEG-immobilized cells was added in the reactor for hydrogen production. To determine the superior packing ratio of the PEG-immobilized cells in UASB reactor, a packing ratio of 4, 8, 12, and 16% (w/v) was carried out. During the experiments, reactor was fed under a constant POME concentration of 30 g COD/L and a HRT of 8, 16, 24 and 32 h. The results are presented in Table 2. Hydrogen production rate was increased with a decrease in HRT from 32 to 8 h, when packing ratio of immobilized cells 4% and 12% was used. On other hand, hydrogen yield showed an opposite trend as it decreased when HRT was shortened. The hydrogen production rate trend is due primarily to the increase of organic loading rate at a shorter HRT, thereby allowing an increase in production rate. The latter trend could be a result of an abrupt increase in hydrogen partial pressure due to vigorous hydrogen production at a short HRT, leading to a lower hydrogen yield. At a packing ratio of 12%, the hydrogen production rate and hydrogen yield reached maximum of 345 mL H2/L-POME h and 0.35 L H2/g CODadded, respectively, which were higher compared to those attained at 4% and 8%. This suggested that increases in biomass loading increase the hydrogen production rate. However, the hydrogen production rate and yield decreased, when the packing ratio of immobilized-cell beads in the reactor increased from 12 to 16%. The lower hydrogen production performance at packing ratio of 16% might have been caused by the use of a large amount of substrate for growth, thus directing substrate utilization away from hydrogen fermentation. The hydrogen content in biogas and COD removal remained stable at 68% and 62% when the reactor was loaded with 4% to 12% of immobilized cells. In contrast, the hydrogen percentage and COD removal efficiency were negatively impacted when the loading amount of immobilized-cell biomass increased from 12 to 16%. This could have been caused by reduced bead movement or contact between microflora and substrates at high immobilized-cell biomass levels. Consequently, the overall performance of the anaerobic digester was impaired [27,28]. These results suggest that
Fig. 3 e Effect of different HRTs on hydrogen production rate, hydrogen yield, hydrogen content, COD removal and POME conversion.
there is a critical amount of immobilized cells in the UASB reactor for successful anaerobic hydrogen production. The conversion of POME was higher than 90% for HRT greater than 16 h, whereas it decreased to 68e70% when the HRT was shortened to 8 h, due probably to the insufficient retention time for completely assimilating the carbon substrate at HRT of 8 h.
2226
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
Table 3 e Soluble metabolites production during anaerobic hydrogen fermentation in the immobilized UASB reactor at different HRT and POME concentration. HRT (h)
HBu (%) HAc (%) HPr (%) BuOH (%) EtOH (%) HBu/HAc TVFA (g/L) SMP (g/L) HYb POME HPRa concentration (mL H2/L/h (L H2/g CODadded) (g COD/L)
32 24 16 8
20
225 278 350 347
0.19 0.28 0.35 0.14
36 53 72 79
15 20 25 29
15 17 11 14
22 18 17 12
12 7 2 5
2.4 2.6 2.9 2.7
2.2 2.6 3.2 2.8
0.02 0.05 0.01 0.04
2.9 3.3 3.4 3.2
0.04 0.04 0.01 0.03
16
10 20 30 40
185 299 365 268
0.17 0.24 0.38 0.26
32 66 84 76
17 23 27 32
13 15 9 17
16 20 15 23
7 4 0.9 11
1.9 2.8 3.1 2.2
1.7 2.4 2.9 2.2
0.01 0.03 0.03 0.02
2.5 3.1 3.2 3.4
0.02 0.02 0.04 0.06
HBu ¼ butyric acid; HAc ¼ acetic acid; HPr ¼ propionic acid; BuOH ¼ butanol; EtOH ¼ ethanol. TVFA ¼ HAc þ HPr þ HBu. SMP ¼ TVFA þ ethanol þ butanol. a Hydrogen production rate. b Hydrogen yield.
3.2.
Effect of HRT on hydrogen production
The effect of HRT on hydrogen production was evaluated as the reactor was operated at different HRTs (32e8 h) but at a constant pH of 5.5, a temperature of 37 C, and POME concentration of 20 g COD/L. Based on the foregoing results, the loading of PEG-immobilized cells (i.e., a packing ratio of 12% (w/v)) was used in UASB reactor operation. After 6 h of batch operation, the continuous operation was started at a stepwise decreasing HRT from 32 h to 8 h (Fig. 3). The hydrogen production rate increased from 215 mL H2/L/h (or 9.5 mmol/L/h) to 350 mL H2/L/h (or 15.6 mmol/L/h). Hydrogen yield increased from 0.16 L H2/g CODadded to 0.36 L H2/g CODadded as the HRT was decreasing from 32 to 16 h. Furthermore, the hydrogen production rate was essentially constant when the HRT decreased from 16 to 8 h. These results suggest that the UASB reactor containing immobilized cells could maintain a high cell concentration even at the lower HRT of 8 h. This indicates that the immobilized-cell UASB reactor was protected from cell washout and could be operated at a lower HRT during operation for continuous hydrogen production. By contrast, previous reports showed a decrease in hydrogen production rate in suspended-cell systems caused by cell washout at low HRT [13]. The hydrogen yield in general decreased as the HRT decreased. The hydrogen yield values were within the range of, but the hydrogen yield significantly decreases from 0.36 L H2/g CODadded to 0.15 L H2/g CODadded when the HRT was shorter to 8 h. At a longer HRT of 32 h, both hydrogen production rate and yield were lower. These results suggest that bacteria are inhibited by low substrate concentrations as a result of a shift in the bacterial population to nonhydrogen producing strains [29]. The hydrogen content in the biogas and the COD removal percentage was maintained in the range of 68e70% and 64e66% at all HRTs of 32e8 h during study-state operation. At all range of HRT, a high POME conversion of over 91.4% was attained except for the runs at a lower HRT of 8 h. This indicates efficient utilization of the substrate by the hydrogen producing bacterial population in the reactor. The lower POME conversion at HRT of 8 h may be due to operation at a lower HRT (or a high food to
microorganism (F/M ratio)) that exceeded the maximum capacity that those hydrogen producers were able to handle efficiently. Over the whole range of HRT (32e8 h), butyric acid and acetic acid increased from 31 to 62% and 18e29% of total SMP, respectively (Table 3). The nearly stable hydrogen content and COD removal performance suggest that the immobilized hydrogen producing cells had high operation stability regardless of changes HRT.
3.3. Effect of influent POME concentration on hydrogen production To evaluate the effect of influent POME concentration on anaerobic hydrogen production, a test was conducted at 16 h HRT and varying POME concentration, from 10 to 40 g COD/L (Fig. 4). The result clearly showed that a POME concentration of 30 g COD/L demonstrated the maximum hydrogen production rates of 365 mL H2/L/h (or 16.2 mmol/L/h) and hydrogen yield of 0.38 L H2/g CODadded. The hydrogen production rate and hydrogen yield both increased with feed concentration increase from 10 to 30 g COD/L, while they decreased significantly when the feed concentration was elevated further to 40 g COD/L. One explanation for low hydrogen production is that substrate inhibition may occur when the POME concentration was higher than 30 g COD/L, or end-product inhibition, such as that exerted by organic acids. Higher hydrogen production ability at HRT of 16 h with a feed concentration of 30 g COD/L possibly correlates to existing suitable condition to activate spore formed bacteria and recover hydrogen-producing bacterial population, which can utilize the POME more efficiently for hydrogen production. Hydrogen composition decreased from 67 to 43% with an increase of the feed concentration from 30 to 40 g COD/L. This seems to suggest that some non-hydrogen producing bacteria started fermenting at a high carbon substrate (POME) concentration and that this led to the conversion of carbon substrate to CO2 without hydrogen production [30]. The COD removal efficiency gradually increased with stepwise increases of feed concentration from 10 to 40 g COD/L, indicating that the removal percentage of COD becomes higher with increasing
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
2227
bacteria resembling Clostridium species were firmly attached to the interior part and surface of the beads (Fig. 1). The PEG-immobilized beads also had a porous structure that facilitated the transfer of substrates and nutrients, thereby ensuring the growth of cells for hydrogen production. We conclude that the UASB reactor containing PEG-immobilized cells is very efficient for anaerobic hydrogen production and treatment of high-strength wastewater at short HRT and high substrates concentration, which could result in significant economic benefits.
3.4.
Soluble metabolites formation
Result of total volatile fatty acids (TVFAs) including soluble microbial products (SMPs) that are produced, and alcohols at various HRT and feed concentration are presented in Table 3. The concentration of TVFAs and their relative proportions have been successfully used as indicators of anaerobic hydrogen production [31]. In this study, HBu and HAc were the major soluble metabolites and accounting more than 75e85% of total SMP, the cultures are appeared to be carried out metabolic pathways in favor of hydrogen production. In contrast, the production of propionic acid (HPr) and ethanol, which are not unfavorable for hydrogen production [32], was relatively insignificant (less than 17% of SMP) throughout the operations. However, butanol (BuOH) was the major species among the acidic solvents formed throughout fermentation in the UASB reactor. This may be due to the fact that POME contains 750 mg/L of BuOH [10]. The HBu/HAc ratio ranges from 2.0 to 3.2 and has been used as an indicator for dark fermentative hydrogen production [33]. The HBu/HAc ratios in the present study varied from 1.9 to 3.1 at various HRT (32e8 h) and feed concentration (10e40 g COD/L) values in the immobilized-cell UASB reactor (Table 3). The highest hydrogen production performance occurred at 16 h HRT and a substrate concentration of 30 g COD/L, resulting in a HBu/HAc ratio of 3.1 (Table 3). The high HBu/SMP ratio and the abundance of TVFAs in total SMP suggest that hydrogen production with the immobilized Clostridium LS2 was directed by acidogenic pathways and was essentially butyrate-type fermentation. The high butyrate concentrations are likely to have been generated by Clostridium species, because these bacteria engage in butyrate-type fermentation [34,35].
4. Fig. 4 e Effect of POME concentration on hydrogen production rate, hydrogen yield, hydrogen content, COD removal and POME conversion.
substrate concentration. The POME conversion rate gradually increases upon increasing the POME concentration from 10 to 40 g COD/L. This is mainly attributed to the enhancing POME transport from medium phase into the gel beads with the increasing influent feed concentration. HBu and HAc were the most abundant products, with contents in the range of 36e87% and 17e32% of total SMP, respectively (Table 3). SEM analysis showed that rod-shaped
Conclusions
The PEG-immobilized cells appeared to serve as effective hydrogen producing Clostridium LS2 in continuous hydrogen production and the treatment of POME in UASB reactor operations, under different HRT and POME concentration. The hydrogen production rate tended to increase as the HRT was shortened at constant substrate concentration of 20 g COD/L. On other hand, the hydrogen yield did not have a common trend against HRT and POME concentration, where it usually decreased at high POME concentration and lower HRT. The UASB reactor loaded with PEG-immobilized cells generated an optimal hydrogen production rate of 365 mL H2/L/h (or 16.2 mmol/L/h) and a hydrogen yield of 0.38 L H2/g CODadded, with an effluent containing mainly butyric acid and
2228
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
acetic acid when operated at a HRT of 12 h and a POME concentration of 30 g COD/L. Additionally, the maximum COD removal efficiency in the reactor was 66%. Microscopic examination clearly showed that the bacteria covered the surface of beads and were present in the core of the beads. Considering that, the use of PEG-immobilized Clostridium LS2 containing UASB reactor might be practically and economically attractive for industrial scale hydrogen production from real wastewater.
Acknowledgments The study was funded by Universiti Malaysia Pahang (UMP) under (grant no. GRS-110332) and the Research Center, College of Engineering, King Saud University, KSA. We are thankful to the anonymous reviewers for their valuable comments.
[13]
[14]
[15]
[16]
[17]
[18]
references [19] [1] Pakarinen O, Lehtomaki A, Rintala J. Batch dark fermentative hydrogen production from grass silage: the effect of inoculums, pH, temperature and VS ration. Int J Hydrogen Energy 2008;33:594e601. [2] Ginkel SV, Sung S. Biohydrogen production as a function of pH and substrate concentration. Environ Sci Technol 2001;35: 4726e30. [3] Benemann J. Hydrogen biotechnology: progress and prospect. Nat Biotechnol 1996;14:1101e3. [4] Brentner LB, Peccia J, Zimmerman JB. Challenges in developing biohydrogen as a sustainable energy source: implications for a research agenda. Environ Sci Technol 2010;44:2243e54. [5] Hallenbeck PC. Fermentative hydrogen production: principles, progress, and prognosis. Int J Hydrogen Energy 2009;34(17):7379e89. [6] Lalaurette E, Thammannagowda S, Mohagheghi A, Maness PC, Logan BE. Hydrogen production from cellulose in a two stage process combining fermentation and electrohydrogenesis. Int J Hydrogen Energy 2009;34(15):6201e10. [7] Wu TY, Mohammad AW, Md Jahim J, Anuar N. A holistic approach to managing palm oil mill effluent (POME): biotechnological advances in the sustainable reuse of POME. Biotechnol Adv 2009;27:40e52. [8] Lam MK, Lee KT. Renewable and sustainable bioenergies production from palm oil mill effluent (POME): winewin strategies toward better environmental protection. Biotechnol Adv 2011;29:124e41. [9] Atif AAY, Fakhru’I-Razia A, Ngan MA, Morimoto M, Iyuke SE, Veziroglu NT. Fed batch production of hydrogen from palm oil mill effluent using anaerobic microflora. Int J Hydrogen Energy 2005;30:1393e7. [10] O-Thong S, Prasertsan P, Intrasungkha N, Dhamwichukorn S, Birkeland NK. Improvement of biohydrogen production and pollution reduction from palm oil mill effluent with nutrient supplementation at thermophilic condition using an anaerobic sequencing batch reactor. Enzyme Microb Technol 2007;41:583e90. [11] Ismail I, Hassan MA, Rahman NAA, Soon CS. Thermophilic biohydrogen production from palm oil mill effluent (POME) using suspended mixed culture. Biomass Bioenergy 2010;34:42e7. [12] Yusoff MZM, Rahman NAA, Abd-Aziz S, Ling CM, Hassan MA, Shirai Y. The effect of hydraulic retention time and volatile fatty acids on biohydrogen production from palm oil mill
[20]
[21]
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30] [31]
[32]
effluent under non-sterile condition. Austral J Basic Appl Sci 2010;4(4):577e87. Badiei M, Md Jahim J, Anuar N, Abdullah SRS. Effect of hydraulic retention time on biohydrogen production from palm oil mill effluent in anaerobic sequencing batch reactor. Int J Hydrogen Energy 2011;36:5912e9. Prasertsan P, O-Thong S, Birkeland NK. Optimization and microbial community analysis for production of biohydrogen from palm oil mill effluent by thermophilic fermentative process. Int J Hydrogen Energy 2009;34:7448e59. Hu B, Chen S. Pretreatment of methanogenic granules for immobilized hydrogen fermentation. Int J Hydrogen Energy 2007;32:3266e73. Kumar N, Das D. Continuous hydrogen production by Enterobacter cloacae IIT-BT 08, using lignocellulosic materials as solid matrices. Enzyme Microb Technol 2001;29:280e7. Peixoto G, Saavedra NK, Varesche MBA, Zaiat M. Hydrogen production from soft-drink wastewater in an upflow anaerobic packed-bed reactor. Int J Hydrogen Energy 2011;36:8953e66. Keskin T, Aksoyek E, Azbar N. Comparative analysis of thermophilic immobilized biohydrogen production using packed materials of ceramic ring and pumice stone. Int J Hydrogen Energy 2011;36:15160e7. Chu CY, Wu SY, Hsieh PC, Lin CY. Biohydrogen production from immobilized cells and suspended sludge systems with condensed molasses fermentation soluble. Int J Hydrogen Energy 2011;36:14078e85. Zhang ZP, Show KY, Tay JH, Liang DT, Lee DJ. Enhanced continuous biohydrogen production by immobilized anaerobic microflora. Energy Fuels 2008;22:87e92. Wu SY, Lin CN, Chang JS, Chang JS. Biohydrogen production with anaerobic sludge immobilized by ethylene-vinyl acetate copolymer. Int J Hydrogen Energy 2005;30:1375e81. Fang HHP, Liu H, Zhang T. Characterization of a hydrogen producing granular sludge. Biotechnol Bioeng 2002;78:44e52. Palazzi E, Fabiano B, Perego P. Process development of continuous hydrogen production by Enterobacter aerogenes in a packed column reactor. Bioprocess Eng 2000;22:205e13. Kim DH, Han SK, Kim SH, Shin HS. Effect of gas sparging on continuous fermentative hydrogen production. Int J Hydrogen Energy 2006;31:2158e69. APHA. Standard methods for examination of water and wastewater. 19th ed. Washington, DC, USA: American Public Health Association; 1998. p. 1220. Lo YC, Saratale GD, Chen WM, Bai MD, Chang JS. Isolation of cellulose-hydrolytic bacteria and applications of the cellulolytic enzymes for cellulosic biohydrogen production. Enzyme Microb Technol 2009;44:417e25. Karim K, Hoffmann R, Klasson KT, Al-Dahhan MH. Anaerobic digestion of animal waste: effect of mode of mixing. Water Res 2005;39:3597e606. Latif MA, Ghufran R, Wahid ZA, Ahmad A. Integrated application of upflow anaerobic sludge blanket reactor for the treatment of wastewater. Water Res 2011;45:4683e99. Chen Wen-Hsing, Sung Shihwu, Chen Shen-Yi. Biological hydrogen production in an anaerobic sequencing batch reactor: pH and cyclic duration effects. Int J Hydrogen Energy 2008;34:1e8. Chang LS, Lee KS, Lin PJ. Biohydrogen production with fixedbed bioreactors. Int J Hydrogen Energy 2002;27:1167e74. Chu Y, Wei Y, Yuan X, Shi X. Bioconversion of wheat stalk to hydrogen by dark fermentation: effect of different mixed microflora on hydrogen yield and cellulose solubilisation. Bioresour Technol 2011;102:3805e909. Yan RT, Zhu CX, Golemboski C, Chen JS. Expression of solvent-forming enzymes and onset of solvent production in batch culture of Clostridium butyricum. Appl Environ Microbiol 1988;54:642e8.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
[33] Oh YK, Kim SH, Kim MS, Park S. Thermophilic biohydrogen production from glucose with trickling biofilter. Biotechnol Bioeng 2004;88:690e8. [34] Chen WM, Tseng ZJ, Lee KS, Chang JS. Fermentative hydrogen production with Clostridium butyricum CGS5
2229
isolated from anaerobic sewage sludge. Int J Hydrogen Energy 2005;30:1063e70. [35] Dinopolou G, Rudd T, Lester JN. Anaerobic acidogenesis if complex wastewater: I. The influence of operational parameters on reactor performance. Biotechnol Bioeng 1988;31:958e68.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/he
Application of immobilized upflow anaerobic sludge blanket reactor using Clostridium LS2 for enhanced biohydrogen production and treatment efficiency of palm oil mill effluent Lakhveer Singh a, Zularisam A. Wahid b,*, Muhammad Faisal Siddiqui c, Anwar Ahmad d, Mohd Hasbi Ab. Rahim a, Mimi Sakinah c a
Faculty of Industrial Sciences & Technology, Universiti Malaysia Pahang (UMP), Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia b Faculty of Civil Engineering and Earth Resources, Universiti Malaysia Pahang (UMP), Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia c Faculty of Chemical and Natural Resource Engineering, Universiti Malaysia Pahang (UMP), Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia d Department of Civil Engineering, King Saud University (KSU), PO Box 800, Riyadh 11421, Saudi Arabia
article info
abstract
Article history:
Polyethylene glycol (PEG) gel was used to immobilize hydrogen producing Clostridium LS2
Received 12 June 2012
bacteria for hydrogen production in an upflow anaerobic sludge blanket (UASB) reactor.
Received in revised form
The UASB reactor with a PEG-immobilized cell packing ratio of 10% weight to volume ratio
21 November 2012
(w/v) was optimal for dark hydrogen production. The performance of the UASB reactor
Accepted 2 December 2012
fed with palm oil mill effluent (POME) as a carbon source was examined under various
Available online 29 December 2012
hydraulic retention time (HRT) and POME concentration. The best volumetric hydrogen production rate of 365 mL H2/L/h (or 16.2 mmol/L/h) with a hydrogen yield of
Keywords:
0.38 L H2/g CODadded was obtained at POME concentration of 30 g COD/L and HRT of 16 h.
Hydrogen
The average hydrogen content of biogas and COD reduction were 68% and 65%, respec-
Polyethylene glycol (PEG)
tively. The primary soluble metabolites were butyric acid and acetic acid with smaller
Immobilization
quantities of other volatile fatty acid and alcohols formed during hydrogen fermentation.
Upflow anaerobic sludge blanket
More importantly, the feasibility of PEG-immobilized cell UASB reactor for the enhance-
(UASB) reactor
ment of the dark-hydrogen production and treatment of wastewater is demonstrated.
Palm oil mill effluent (POME)
Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
The joint challenges of environmental crises and dwindling fossil fuel supplies are driving intensive research focus in
alternative energy production. Hydrogen is widely regarded as one of the most potential future energy vector, capable of assisting in issues of environmental emissions, energy security and versatility as fuel [1]. Despite the “green” nature of hydrogen
* Corresponding author. Tel.: þ60 95493002. E-mail addresses: [email protected], [email protected] (Z.A. Wahid). 0360-3199/$ e see front matter Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijhydene.2012.12.004
2222
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
Abbreviations PEG POME UASB COD HRT w/v SEM TN VSS TS
polyethylene glycol palm oil mill effluent upflow anaerobic sludge blanket chemical oxygen demand hydraulic retention times weight to volume scanning electron microscope total nitrogen volatile suspended solids total solid
as a fuel, it is still usually produced via steam reforming of natural gas, petroleum hydrocarbon, nonrenewable materials and other hydrogenation reactions, which makes hydrogen production environmentally unfriendly and expansive [2]. Among the established hydrogen production technologies, anaerobic fermentation has received considerable attention due to its potential as an inexhaustible, carbon-neutral and cost-effective fuel but has yet to reach a scale large enough for consideration in replacing a major portion of the hydrogen supply [3,4]. A wide variety of organic wastes or renewable substrates could be used as carbon source for biological hydrogen generation [5,6]. Moreover, fermentative hydrogen production using wastewater and organic waste as substrates achieves both energy recovery and bioremediation. In Malaysia, the estimated annual production of palm oil mill effluent (POME) is about 50 million tons. POME is an important renewable biomass energy source that can be harmful to the environment if untreated POME is discharged directly to the surroundings, due to high values of COD and biochemical oxygen demand (BOD) that it generates [7,8]. The high nutrient content in POME makes it an ideal fermentation medium for anaerobic treatment processes [9]. Previous reports have utilized POME as a substrate for hydrogen production with defined microflora, or mixed cultures of POME sludge under continuous mode operation [10e14]. Yusoff et al. [12] found that the maximum biohydrogen yield was 1054 mL/ L-POME and the maximum hydrogen production rate was 44 N mL/h/L POME at a pH of 5.5, a temperature of 22e26 C, a substrate concentration of 50e60 g/L COD-POME, and a HRT of 48 h in a 50-L continuously stirred tank reactor. Badiei et al. [13] evaluated hydrogen production from POME in an anaerobic sequencing batch reactor using enriched mixed microflora from POME sludge, and obtained maximum hydrogen production rate of 6.7 LH2/L/d with a total COD removal of more than 37% at a hydraulic retention time (HRT) of 3 d, an organic loading rate (OLR) of 6.6 g COD/L/d, a pH of 6.8, and a temperature of 37 C. Prasertsan et al. [14] obtained a maximum hydrogen production rate of 9.1 L H2/L/d along with a COD removal of 57% at a controlled pH 5.5, a temperature of 60 C, and 60 g COD/L-POME/d OLR and optimum values of 48 h HRT in the anaerobic sequencing batch reactor. However, these studies on biohydrogen production from POME only focused on suspended cell-systems, which are usually ineffective or difficult to handle in continuous operation
GC BOD TP TN VFAs HBu HAc BuOH EtOH SMPs TVFA HPr
Gas chromatograph biochemical oxygen demand total phosphorus total nitrogen volatile fatty acids butyric acid acetic acid butanol ethanol soluble microbial products total volatile fatty acid propionic acid
because washout of bacteria with effluents may occur from the reactor at shorter HRTs [13] and result in a low biomass concentration in the bioreactor. Recycling of biomass is considered necessary to maintain a sufficient cell concentration in the reactor to maximize hydrogen production [14]. Another potential approach to enhance hydrogen production is to use an immobilized cell system [15]. Immobilized cells offer distinct advantages over suspended cells, because they are resistant to cell wash-out during continuous operation and can maintain a higher cell density that increases hydrogen production. Immobilized cells have been successfully used for continuous hydrogen production in a bioreactor [16e19]. Many different methods have been employed for biomass immobilization including entrapment in polymeric gel and adsorption to solid surfaces, granules and biofilms [20e23]. The primary difficulty of the immobilized system is leak-out. However, it allows better biomass retention at low HRTs and creates a local anaerobic environment, which is well-suited to fermentative hydrogen production. In this study, a new cell immobilization system that employs polyethylene glycol (PEG) gel was employed to immobilize Clostridium LS2 for hydrogen production and treatment of POME in a UASB reactor. The effect of packing ratio of immobilized cells, HRT and POME concentration on the hydrogen producing and treatment capability of the immobilized UASB reactor was examined. The information obtained here is expected to be useful in future development of effective immobilized UASB reactor hydrogen producing systems for large scale application.
2.
Materials and methods
2.1.
Fermentation medium
Raw POME was collected from the final discharge point of an oil palm mill, Lepar Hilir Pahang, Malaysia, and was used as the substrate for hydrogen production. Raw POME has brown color, temperature of 65e75 C, pH 4.0e4.5 and water content of 85e90%. POME was preserved at 4 C in a 25-L container to prevent biodegradation and acidification before use. Prior to being fed into the reactor, the POME was diluted to a required COD concentration for fermentation. The characteristics of the POME wastewater are given in Table 1.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
Table 1 e Characteristics of POME. Parameter Biochemical oxygen demand (BOD) Chemical oxygen demand (COD) pH Total carbohydrate Total nitrogen (TN) Ammoniumenitrogen Total phosphorus (TP) Phosphorus Oil Total solid (TS) Volatile suspended solids (VSS)
Concentration 23,100e55,200 55,100e86,300 4.0e5.0 16,200e20,000 820e910 25e35 95e120 14e20 2000e2500 30,000e42,000 8400e12,000
All values are in mg/L except pH.
2223
20 mL of inoculum was approximately 2.2 g. To start polymerization, 0.25% (w/v) potassium persulfate initiator was mixed in and the mixture allowed to stand for about 80 min to promote bead formation. The immobilized-cell polymer was cut into 3-mm bead particles (density 1.40 g/cm3). The biomass content of the immobilized bead was w10 mg cell/g bead. Prior to use, the immobilized cells were stored in physiological saline solution for 2 h and then washed thoroughly with distilled water. The scanning electron microscopy (SEM) profiles of immobilized beads are shown in Fig. 1.
2.4. UASB reactor setup and operation for hydrogen production
The hydrogen producing strain Clostridium LS2 was used as the inoculum in this study. Clostridium LS2 was isolated from POME sludge taken from an anaerobic digester in the palm oil mill of Lepar Hilir Pahang, Malaysia. The sludge sample was heat-treated at 90 C for 30 min to inhibit methanogenic activity [24]. The heat-treated sludge was diluted with distilled water, and then cultured in medium under anaerobic conditions using an anaerobic jar at 36 C. The composition of the culture medium was (per L): 0.5 g of starch, 10 g of meat extract, 4 g of glucose, 2.5 g of yeast extract, 2.0 g of peptone, 0.5 g of agar, 0.5 g of L-cysteine$HCl$H2O, 3 g of NaCl, pH 6.5 (adjusted with 1 M NaOH). The Clostridium strain was obtained from a pure culture and designated as LS2. Biochemical identification by the rapid ANA II microtest system indicated that the strain belonged to the genus Clostridium. The isolated strain was stored in sterile 15% (v/v) glycerol solution at 30 C before being subjected to immobilization.
A stainless steel laboratory-scale UASB reactor (5126 cm3) with 5 L working volume was used in this study (Fig. 2). The UASB reactor was operated at temperature 37 C and temperature was maintained by hot water circulation through the water jacket. Throughout the experiments, the pH of the reaction medium in the reactor was adjusted around 5.5 by adding 1 M NaOH or 1 M HCl solutions. The medium was fed into the UASB reactor together with PEG-immobilized cells. Sampling points were introduced at appropriate heights in the reactor. A gaseliquid separator was introduced at the top of the reactor, where the biogas and soluble microbial product (SMP) collected separately. The reactor was purged with nitrogen gas for 10 min to create anaerobic conditions. The UASB reactor was loaded with an appropriate amount of PEG-immobilized cells to obtain a final solution of 4e16% (w/v). The UASB reactor was operated on batch mode for 6 h before being switched to continuous mode at a HRT of 32e8 h and a POME concentration of 10e40 g COD/L. The quantity and composition of the biogas, COD removal efficiency (%), POME conversion efficiency (%), volatile fatty acids (VFAs), pH, and temperature were monitored at designated time intervals.
2.3.
2.5.
2.2.
Microorganisms
Immobilization of cells in PEG
Clostridium LS2 cells were immobilized by entrapment in a PEG prepolymer. The cells of anaerobic bacteria were harvested by centrifugation at 8000 rpm for 12 min. A culture sample was heated at 80 C for 5 min prior to use for immobilization. First, 10% (w/v) PEG and 0.6% (w/v) N,N0 -methylenebisacrylamide (MBA) crosslinker were dissolved in a water. The resulting mixture and 20 mL of inoculum (in exponential growth phase) were quickly mixed in a beaker. The dry weight of cells in
Assay methods
Hydrogen gas was measured by a gas chromatograph (GC 8500 Perkin Elmer) equipped with a thermal conductivity detector and a 2-m stainless-steel SS350A column packed with a molecular sieve (80/100 mesh) using nitrogen as a carrier gas at a flow rate of 25 mL/min. VFAs contents of filtered sample (0.2 mm) were determined by gas chromatography with a FID detector (model 6890N, Agilent Inc., glass 2 m 2 mm packed column Carbopack B-DA 80/120% CW 20 M, N2 carrier at kPa,
Fig. 1 e SEM images of the PEG immobilized cell bead: (a) size and shape of the immobilized beads; (b) peripheral surface of the immobilized beads. Scale bar: 10 mm; (c) cross-sectional image of the immobilized beads. Scale bar: 2 mm.
2224
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
Fig. 2 e Experimental setup of immobilized cell UASB reactor: PHT e POME holding tank; PP e peristaltic pump; FM e flow meter; MV e manual valve; M e mixture; IB e immobilized beads; SV e sampling valve; BPT e biogas purification tank; WT e water tank; TS e temperature sensor; HP e heating probe; P e pump.
170 C). For alcohol analysis, 1 cm3 of sample acidified with 0.003 cm3 25% H2SO4, was analyzed using GCeFID and capillary column of the same model and type, respectively. COD, total nitrogen (TN), total solid (TS), VSS, and pH were carried
out using standard methods [25]. The dry weight of immobilized cell in immobilized beads was assessed by measuring the difference in dry weight between the biomass-associated beads and the beads alone.
Table 2 e Effect of various packing ratios of PEG-immobilized cells in UASB reactor on hydrogen production (POME concentration: 20 g COD/L). Packing ratio of PEG-immobilized cells (w/v)%
HPRa (mL H2/L/h)
HYb (L H2/g CODadded)
H2 content in biogas (%)
COD removal (%)
POME conversion (%)
32 24 16 8
4
292 305 310 317
0.275 0.260 0.259 0.251
65 66 64 66
59 60 61 60
93 94 85 69
32 24 16 8
8
302 310 323 331
0.313 0.299 0.283 0.272
65 66 67 66
58 60 61 60
94 92 82 68
32 24 16 8
12
325 337 342 345
0.350 0.341 0.327 0.312
67 68 68 67
62 61 62 62
94 91 84 70
32 24 16 8
16
294 299 276 268
0.319 0.283 0.246 0.235
62 60 61 59
58 57 57 56
93 92 82 68
HRT
a Hydrogen production rate. b Hydrogen yield.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
2225
2.6. Scanning electron microscopy and biochemical analysis The immobilized cells were gently washed with distilled water. Immobilized cells were then fixed with 2% glutaraldehyde and left for 2 h. The fixed immobilized cells were dehydrated by successive passages through 50%, 60%, 70% and 90% ethanol for 20 min and then dried. The dried samples were covered with a layer of gold under vacuum prior to being subjected to SEM (Zeiss EVO50, Germany). The rapid ANA II microtests (Remel) for anaerobic bacteria were utilized for biochemical identification [26].
3.
Results and discussion
3.1. Effect of packing ratio of the PEG-immobilized cell concentrations on hydrogen production To determine the biomass loading for higher hydrogen production and stabilize the operations in the UASB reactor, a suitable amount of the PEG-immobilized cells was added in the reactor for hydrogen production. To determine the superior packing ratio of the PEG-immobilized cells in UASB reactor, a packing ratio of 4, 8, 12, and 16% (w/v) was carried out. During the experiments, reactor was fed under a constant POME concentration of 30 g COD/L and a HRT of 8, 16, 24 and 32 h. The results are presented in Table 2. Hydrogen production rate was increased with a decrease in HRT from 32 to 8 h, when packing ratio of immobilized cells 4% and 12% was used. On other hand, hydrogen yield showed an opposite trend as it decreased when HRT was shortened. The hydrogen production rate trend is due primarily to the increase of organic loading rate at a shorter HRT, thereby allowing an increase in production rate. The latter trend could be a result of an abrupt increase in hydrogen partial pressure due to vigorous hydrogen production at a short HRT, leading to a lower hydrogen yield. At a packing ratio of 12%, the hydrogen production rate and hydrogen yield reached maximum of 345 mL H2/L-POME h and 0.35 L H2/g CODadded, respectively, which were higher compared to those attained at 4% and 8%. This suggested that increases in biomass loading increase the hydrogen production rate. However, the hydrogen production rate and yield decreased, when the packing ratio of immobilized-cell beads in the reactor increased from 12 to 16%. The lower hydrogen production performance at packing ratio of 16% might have been caused by the use of a large amount of substrate for growth, thus directing substrate utilization away from hydrogen fermentation. The hydrogen content in biogas and COD removal remained stable at 68% and 62% when the reactor was loaded with 4% to 12% of immobilized cells. In contrast, the hydrogen percentage and COD removal efficiency were negatively impacted when the loading amount of immobilized-cell biomass increased from 12 to 16%. This could have been caused by reduced bead movement or contact between microflora and substrates at high immobilized-cell biomass levels. Consequently, the overall performance of the anaerobic digester was impaired [27,28]. These results suggest that
Fig. 3 e Effect of different HRTs on hydrogen production rate, hydrogen yield, hydrogen content, COD removal and POME conversion.
there is a critical amount of immobilized cells in the UASB reactor for successful anaerobic hydrogen production. The conversion of POME was higher than 90% for HRT greater than 16 h, whereas it decreased to 68e70% when the HRT was shortened to 8 h, due probably to the insufficient retention time for completely assimilating the carbon substrate at HRT of 8 h.
2226
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
Table 3 e Soluble metabolites production during anaerobic hydrogen fermentation in the immobilized UASB reactor at different HRT and POME concentration. HRT (h)
HBu (%) HAc (%) HPr (%) BuOH (%) EtOH (%) HBu/HAc TVFA (g/L) SMP (g/L) HYb POME HPRa concentration (mL H2/L/h (L H2/g CODadded) (g COD/L)
32 24 16 8
20
225 278 350 347
0.19 0.28 0.35 0.14
36 53 72 79
15 20 25 29
15 17 11 14
22 18 17 12
12 7 2 5
2.4 2.6 2.9 2.7
2.2 2.6 3.2 2.8
0.02 0.05 0.01 0.04
2.9 3.3 3.4 3.2
0.04 0.04 0.01 0.03
16
10 20 30 40
185 299 365 268
0.17 0.24 0.38 0.26
32 66 84 76
17 23 27 32
13 15 9 17
16 20 15 23
7 4 0.9 11
1.9 2.8 3.1 2.2
1.7 2.4 2.9 2.2
0.01 0.03 0.03 0.02
2.5 3.1 3.2 3.4
0.02 0.02 0.04 0.06
HBu ¼ butyric acid; HAc ¼ acetic acid; HPr ¼ propionic acid; BuOH ¼ butanol; EtOH ¼ ethanol. TVFA ¼ HAc þ HPr þ HBu. SMP ¼ TVFA þ ethanol þ butanol. a Hydrogen production rate. b Hydrogen yield.
3.2.
Effect of HRT on hydrogen production
The effect of HRT on hydrogen production was evaluated as the reactor was operated at different HRTs (32e8 h) but at a constant pH of 5.5, a temperature of 37 C, and POME concentration of 20 g COD/L. Based on the foregoing results, the loading of PEG-immobilized cells (i.e., a packing ratio of 12% (w/v)) was used in UASB reactor operation. After 6 h of batch operation, the continuous operation was started at a stepwise decreasing HRT from 32 h to 8 h (Fig. 3). The hydrogen production rate increased from 215 mL H2/L/h (or 9.5 mmol/L/h) to 350 mL H2/L/h (or 15.6 mmol/L/h). Hydrogen yield increased from 0.16 L H2/g CODadded to 0.36 L H2/g CODadded as the HRT was decreasing from 32 to 16 h. Furthermore, the hydrogen production rate was essentially constant when the HRT decreased from 16 to 8 h. These results suggest that the UASB reactor containing immobilized cells could maintain a high cell concentration even at the lower HRT of 8 h. This indicates that the immobilized-cell UASB reactor was protected from cell washout and could be operated at a lower HRT during operation for continuous hydrogen production. By contrast, previous reports showed a decrease in hydrogen production rate in suspended-cell systems caused by cell washout at low HRT [13]. The hydrogen yield in general decreased as the HRT decreased. The hydrogen yield values were within the range of, but the hydrogen yield significantly decreases from 0.36 L H2/g CODadded to 0.15 L H2/g CODadded when the HRT was shorter to 8 h. At a longer HRT of 32 h, both hydrogen production rate and yield were lower. These results suggest that bacteria are inhibited by low substrate concentrations as a result of a shift in the bacterial population to nonhydrogen producing strains [29]. The hydrogen content in the biogas and the COD removal percentage was maintained in the range of 68e70% and 64e66% at all HRTs of 32e8 h during study-state operation. At all range of HRT, a high POME conversion of over 91.4% was attained except for the runs at a lower HRT of 8 h. This indicates efficient utilization of the substrate by the hydrogen producing bacterial population in the reactor. The lower POME conversion at HRT of 8 h may be due to operation at a lower HRT (or a high food to
microorganism (F/M ratio)) that exceeded the maximum capacity that those hydrogen producers were able to handle efficiently. Over the whole range of HRT (32e8 h), butyric acid and acetic acid increased from 31 to 62% and 18e29% of total SMP, respectively (Table 3). The nearly stable hydrogen content and COD removal performance suggest that the immobilized hydrogen producing cells had high operation stability regardless of changes HRT.
3.3. Effect of influent POME concentration on hydrogen production To evaluate the effect of influent POME concentration on anaerobic hydrogen production, a test was conducted at 16 h HRT and varying POME concentration, from 10 to 40 g COD/L (Fig. 4). The result clearly showed that a POME concentration of 30 g COD/L demonstrated the maximum hydrogen production rates of 365 mL H2/L/h (or 16.2 mmol/L/h) and hydrogen yield of 0.38 L H2/g CODadded. The hydrogen production rate and hydrogen yield both increased with feed concentration increase from 10 to 30 g COD/L, while they decreased significantly when the feed concentration was elevated further to 40 g COD/L. One explanation for low hydrogen production is that substrate inhibition may occur when the POME concentration was higher than 30 g COD/L, or end-product inhibition, such as that exerted by organic acids. Higher hydrogen production ability at HRT of 16 h with a feed concentration of 30 g COD/L possibly correlates to existing suitable condition to activate spore formed bacteria and recover hydrogen-producing bacterial population, which can utilize the POME more efficiently for hydrogen production. Hydrogen composition decreased from 67 to 43% with an increase of the feed concentration from 30 to 40 g COD/L. This seems to suggest that some non-hydrogen producing bacteria started fermenting at a high carbon substrate (POME) concentration and that this led to the conversion of carbon substrate to CO2 without hydrogen production [30]. The COD removal efficiency gradually increased with stepwise increases of feed concentration from 10 to 40 g COD/L, indicating that the removal percentage of COD becomes higher with increasing
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
2227
bacteria resembling Clostridium species were firmly attached to the interior part and surface of the beads (Fig. 1). The PEG-immobilized beads also had a porous structure that facilitated the transfer of substrates and nutrients, thereby ensuring the growth of cells for hydrogen production. We conclude that the UASB reactor containing PEG-immobilized cells is very efficient for anaerobic hydrogen production and treatment of high-strength wastewater at short HRT and high substrates concentration, which could result in significant economic benefits.
3.4.
Soluble metabolites formation
Result of total volatile fatty acids (TVFAs) including soluble microbial products (SMPs) that are produced, and alcohols at various HRT and feed concentration are presented in Table 3. The concentration of TVFAs and their relative proportions have been successfully used as indicators of anaerobic hydrogen production [31]. In this study, HBu and HAc were the major soluble metabolites and accounting more than 75e85% of total SMP, the cultures are appeared to be carried out metabolic pathways in favor of hydrogen production. In contrast, the production of propionic acid (HPr) and ethanol, which are not unfavorable for hydrogen production [32], was relatively insignificant (less than 17% of SMP) throughout the operations. However, butanol (BuOH) was the major species among the acidic solvents formed throughout fermentation in the UASB reactor. This may be due to the fact that POME contains 750 mg/L of BuOH [10]. The HBu/HAc ratio ranges from 2.0 to 3.2 and has been used as an indicator for dark fermentative hydrogen production [33]. The HBu/HAc ratios in the present study varied from 1.9 to 3.1 at various HRT (32e8 h) and feed concentration (10e40 g COD/L) values in the immobilized-cell UASB reactor (Table 3). The highest hydrogen production performance occurred at 16 h HRT and a substrate concentration of 30 g COD/L, resulting in a HBu/HAc ratio of 3.1 (Table 3). The high HBu/SMP ratio and the abundance of TVFAs in total SMP suggest that hydrogen production with the immobilized Clostridium LS2 was directed by acidogenic pathways and was essentially butyrate-type fermentation. The high butyrate concentrations are likely to have been generated by Clostridium species, because these bacteria engage in butyrate-type fermentation [34,35].
4. Fig. 4 e Effect of POME concentration on hydrogen production rate, hydrogen yield, hydrogen content, COD removal and POME conversion.
substrate concentration. The POME conversion rate gradually increases upon increasing the POME concentration from 10 to 40 g COD/L. This is mainly attributed to the enhancing POME transport from medium phase into the gel beads with the increasing influent feed concentration. HBu and HAc were the most abundant products, with contents in the range of 36e87% and 17e32% of total SMP, respectively (Table 3). SEM analysis showed that rod-shaped
Conclusions
The PEG-immobilized cells appeared to serve as effective hydrogen producing Clostridium LS2 in continuous hydrogen production and the treatment of POME in UASB reactor operations, under different HRT and POME concentration. The hydrogen production rate tended to increase as the HRT was shortened at constant substrate concentration of 20 g COD/L. On other hand, the hydrogen yield did not have a common trend against HRT and POME concentration, where it usually decreased at high POME concentration and lower HRT. The UASB reactor loaded with PEG-immobilized cells generated an optimal hydrogen production rate of 365 mL H2/L/h (or 16.2 mmol/L/h) and a hydrogen yield of 0.38 L H2/g CODadded, with an effluent containing mainly butyric acid and
2228
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
acetic acid when operated at a HRT of 12 h and a POME concentration of 30 g COD/L. Additionally, the maximum COD removal efficiency in the reactor was 66%. Microscopic examination clearly showed that the bacteria covered the surface of beads and were present in the core of the beads. Considering that, the use of PEG-immobilized Clostridium LS2 containing UASB reactor might be practically and economically attractive for industrial scale hydrogen production from real wastewater.
Acknowledgments The study was funded by Universiti Malaysia Pahang (UMP) under (grant no. GRS-110332) and the Research Center, College of Engineering, King Saud University, KSA. We are thankful to the anonymous reviewers for their valuable comments.
[13]
[14]
[15]
[16]
[17]
[18]
references [19] [1] Pakarinen O, Lehtomaki A, Rintala J. Batch dark fermentative hydrogen production from grass silage: the effect of inoculums, pH, temperature and VS ration. Int J Hydrogen Energy 2008;33:594e601. [2] Ginkel SV, Sung S. Biohydrogen production as a function of pH and substrate concentration. Environ Sci Technol 2001;35: 4726e30. [3] Benemann J. Hydrogen biotechnology: progress and prospect. Nat Biotechnol 1996;14:1101e3. [4] Brentner LB, Peccia J, Zimmerman JB. Challenges in developing biohydrogen as a sustainable energy source: implications for a research agenda. Environ Sci Technol 2010;44:2243e54. [5] Hallenbeck PC. Fermentative hydrogen production: principles, progress, and prognosis. Int J Hydrogen Energy 2009;34(17):7379e89. [6] Lalaurette E, Thammannagowda S, Mohagheghi A, Maness PC, Logan BE. Hydrogen production from cellulose in a two stage process combining fermentation and electrohydrogenesis. Int J Hydrogen Energy 2009;34(15):6201e10. [7] Wu TY, Mohammad AW, Md Jahim J, Anuar N. A holistic approach to managing palm oil mill effluent (POME): biotechnological advances in the sustainable reuse of POME. Biotechnol Adv 2009;27:40e52. [8] Lam MK, Lee KT. Renewable and sustainable bioenergies production from palm oil mill effluent (POME): winewin strategies toward better environmental protection. Biotechnol Adv 2011;29:124e41. [9] Atif AAY, Fakhru’I-Razia A, Ngan MA, Morimoto M, Iyuke SE, Veziroglu NT. Fed batch production of hydrogen from palm oil mill effluent using anaerobic microflora. Int J Hydrogen Energy 2005;30:1393e7. [10] O-Thong S, Prasertsan P, Intrasungkha N, Dhamwichukorn S, Birkeland NK. Improvement of biohydrogen production and pollution reduction from palm oil mill effluent with nutrient supplementation at thermophilic condition using an anaerobic sequencing batch reactor. Enzyme Microb Technol 2007;41:583e90. [11] Ismail I, Hassan MA, Rahman NAA, Soon CS. Thermophilic biohydrogen production from palm oil mill effluent (POME) using suspended mixed culture. Biomass Bioenergy 2010;34:42e7. [12] Yusoff MZM, Rahman NAA, Abd-Aziz S, Ling CM, Hassan MA, Shirai Y. The effect of hydraulic retention time and volatile fatty acids on biohydrogen production from palm oil mill
[20]
[21]
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30] [31]
[32]
effluent under non-sterile condition. Austral J Basic Appl Sci 2010;4(4):577e87. Badiei M, Md Jahim J, Anuar N, Abdullah SRS. Effect of hydraulic retention time on biohydrogen production from palm oil mill effluent in anaerobic sequencing batch reactor. Int J Hydrogen Energy 2011;36:5912e9. Prasertsan P, O-Thong S, Birkeland NK. Optimization and microbial community analysis for production of biohydrogen from palm oil mill effluent by thermophilic fermentative process. Int J Hydrogen Energy 2009;34:7448e59. Hu B, Chen S. Pretreatment of methanogenic granules for immobilized hydrogen fermentation. Int J Hydrogen Energy 2007;32:3266e73. Kumar N, Das D. Continuous hydrogen production by Enterobacter cloacae IIT-BT 08, using lignocellulosic materials as solid matrices. Enzyme Microb Technol 2001;29:280e7. Peixoto G, Saavedra NK, Varesche MBA, Zaiat M. Hydrogen production from soft-drink wastewater in an upflow anaerobic packed-bed reactor. Int J Hydrogen Energy 2011;36:8953e66. Keskin T, Aksoyek E, Azbar N. Comparative analysis of thermophilic immobilized biohydrogen production using packed materials of ceramic ring and pumice stone. Int J Hydrogen Energy 2011;36:15160e7. Chu CY, Wu SY, Hsieh PC, Lin CY. Biohydrogen production from immobilized cells and suspended sludge systems with condensed molasses fermentation soluble. Int J Hydrogen Energy 2011;36:14078e85. Zhang ZP, Show KY, Tay JH, Liang DT, Lee DJ. Enhanced continuous biohydrogen production by immobilized anaerobic microflora. Energy Fuels 2008;22:87e92. Wu SY, Lin CN, Chang JS, Chang JS. Biohydrogen production with anaerobic sludge immobilized by ethylene-vinyl acetate copolymer. Int J Hydrogen Energy 2005;30:1375e81. Fang HHP, Liu H, Zhang T. Characterization of a hydrogen producing granular sludge. Biotechnol Bioeng 2002;78:44e52. Palazzi E, Fabiano B, Perego P. Process development of continuous hydrogen production by Enterobacter aerogenes in a packed column reactor. Bioprocess Eng 2000;22:205e13. Kim DH, Han SK, Kim SH, Shin HS. Effect of gas sparging on continuous fermentative hydrogen production. Int J Hydrogen Energy 2006;31:2158e69. APHA. Standard methods for examination of water and wastewater. 19th ed. Washington, DC, USA: American Public Health Association; 1998. p. 1220. Lo YC, Saratale GD, Chen WM, Bai MD, Chang JS. Isolation of cellulose-hydrolytic bacteria and applications of the cellulolytic enzymes for cellulosic biohydrogen production. Enzyme Microb Technol 2009;44:417e25. Karim K, Hoffmann R, Klasson KT, Al-Dahhan MH. Anaerobic digestion of animal waste: effect of mode of mixing. Water Res 2005;39:3597e606. Latif MA, Ghufran R, Wahid ZA, Ahmad A. Integrated application of upflow anaerobic sludge blanket reactor for the treatment of wastewater. Water Res 2011;45:4683e99. Chen Wen-Hsing, Sung Shihwu, Chen Shen-Yi. Biological hydrogen production in an anaerobic sequencing batch reactor: pH and cyclic duration effects. Int J Hydrogen Energy 2008;34:1e8. Chang LS, Lee KS, Lin PJ. Biohydrogen production with fixedbed bioreactors. Int J Hydrogen Energy 2002;27:1167e74. Chu Y, Wei Y, Yuan X, Shi X. Bioconversion of wheat stalk to hydrogen by dark fermentation: effect of different mixed microflora on hydrogen yield and cellulose solubilisation. Bioresour Technol 2011;102:3805e909. Yan RT, Zhu CX, Golemboski C, Chen JS. Expression of solvent-forming enzymes and onset of solvent production in batch culture of Clostridium butyricum. Appl Environ Microbiol 1988;54:642e8.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 2 2 2 1 e2 2 2 9
[33] Oh YK, Kim SH, Kim MS, Park S. Thermophilic biohydrogen production from glucose with trickling biofilter. Biotechnol Bioeng 2004;88:690e8. [34] Chen WM, Tseng ZJ, Lee KS, Chang JS. Fermentative hydrogen production with Clostridium butyricum CGS5
2229
isolated from anaerobic sewage sludge. Int J Hydrogen Energy 2005;30:1063e70. [35] Dinopolou G, Rudd T, Lester JN. Anaerobic acidogenesis if complex wastewater: I. The influence of operational parameters on reactor performance. Biotechnol Bioeng 1988;31:958e68.