- Email: [email protected]
Optimization of biohydrogen production from sugarcane bagasse by mixed cultures using a statistical method
Accepted Manuscript Optimization of biohydrogen production from sugarcane bagasse by mixed cultures using a statistical method Suksaman Sangyoka, Alissara Reungsang, Chiu-Yue Lin PII:
S2468-2039(16)30032-2
DOI:
10.1016/j.serj.2016.05.001
Reference:
SERJ 32
To appear in:
Sustainable Environment Research
Received Date: 18 August 2015 Revised Date:
16 November 2015
Accepted Date: 15 February 2016
Please cite this article as: Sangyoka S, Reungsang A, Lin C-Y, Optimization of biohydrogen production from sugarcane bagasse by mixed cultures using a statistical method, Sustainable Environment Research (2016), doi: 10.1016/j.serj.2016.05.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Received 18 August 2015 Received in revised form 16 November 2015 Accepted 15 February 2016
statistical method
RI PT
Optimization of biohydrogen production from sugarcane bagasse by mixed cultures using a
SC
Suksaman Sangyoka,1,* Alissara Reungsang2,3 and Chiu-Yue Lin4,5,6
1
AC C
EP
TE D
M AN U
Faculty of Science and Technology Pibulsongkram Rajabhat University Phitsanulok 65000, Thailand 2 Department of Biotechnology Khon Kaen University Khon Kaen 40002, Thailand 3 Fermentation Research Center for Value Added of Agricultural Products Khon Kaen University Khon Kaen 40002, Thailand 4 Department of Environmental Engineering and Science Feng Chia University Taichung 40724, Taiwan 5 Master’s Program of Green Energy Science and Technology Feng Chia University Taichung 40724, Taiwan 6 Green Energy Development Center Feng Chia University Taichung 40724, Taiwan
Key Words: Biohydrogen, central composite design (CCD), response surface methodology (RSM), sugarcane bagasse, hydrolysate
*Correspondence author Email: [email protected]
1
ACCEPTED MANUSCRIPT
ABSTRACT The aim of this study is to determine the optimum condition for biohydrogen production from sugarcane bagasse (SCB) hydrolysate using a central composite design (CCD) and response
RI PT
surface methodology (RSM). SCB was hydrolyzed with 0.5% (v/v) sulfuric acid at 121 °C, 0.15 MPa for 60 min in an autoclave at a solid to liquid ratio of 1:15 (g:mL). Heat-treated bacterium obtained from a hydrogen producing fermentor was used as the inoculum. The interaction of
SC
three factors, i.e., substrate concentration, substrate:buffer ratio and inoculum:substrate ratio on hydrogen production potential (P) were investigated. The results indicated that the substrate
M AN U
concentration, substrate:buffer ratio and inoculum:substrate ratio had a significant influence effect on P. An optimal condition was found at substrate concentration of 22.77 g-toal sugar L-1, 4.31 substrate:buffer ratio, and 0.31 inoculum:substrate ratio resulted in a maximum P of 6,980 mL H2 L-1. The confirmation experiment results indicated that optimum P was statistically
TE D
significant, from the predicted value obtained by RSM which suggests that RSM could be efficiently used to optimize a biohydrogen production from SCB hydrolysate using mixed cultures. These results indicates that the SCB hemicellulose hydrolysate is suitable as a
EP
fermentation media for producing biohydrogen. This approach will add value to SCB by
AC C
converting agricultural waste into a safe and clean form of energy. 1. Introduction
Our energy supply comes mainly from fossil fuels for transportation and industrialization
resulting in not only environmental pollution, but also economic and political problems owing to their limited reserves and uneven distribution. Nowadays, alternative energy can come from natural resources (wind, sunlight, geothermal power and biomass) which are inexhaustible. Using these resources to supply our energy needs supports sustainable development, while also
2
ACCEPTED MANUSCRIPT
lowering greenhouse gas emissions. The different characteristics of an energy resource can be evaluated in terms of sustainability and their ability to replace traditional fuels at different levels including power generation, heating, as a transport fuel and for rural energy. In developing
RI PT
countries, biomass represents the major source of renewable energy because of its commonly used as the local energy supply. Bioenergy is a fuel derived from a biological source (biomass) and is also referred to as biofuel. Biomass is defined as any organic material coming from any
SC
form of life or its derived metabolic production. Biodiesel and bioethanol are biofuels that currently are the only alternative energy source able to replace transportation fuel in vehicle
M AN U
engines without involving major modifications. Previous researchs study biofuel from biomass such as agricultural waste [1], food waste [2] and microalgae [3-5].
On the other hand, some scientists predict hydrogen as the “fuel of the future”, and the worldwide trend is the shift away from non-renewable fuels towards the establishment of
TE D
renewable hydrogen energy. Many companies are working to develop technologies that might efficiently exploit the potential of hydrogen energy for use in motor vehicles. As of November 2013 there are demonstration fleets of hydrogen fuel cell vehicles undergoing field testing
EP
including; the Chevrolet Equinox Fuel Cell, Honda FCX Clarity, Hyundai ix35 FCEV and Mercedes-Benz B-Class F-Cell [6]. Hydrogen is considered to be an ideal energy source for the
AC C
future due to its cleanliness and high energy yield of 142 kJ g-1, which is 2.75 times greater than that of any hydrocarbon fuel. Hydrogen production technologies can be classified as physicalchemical and biological methods. Physical-chemical methods are very energy intensive and result in the emission of harmful greenhouse gases (CO, CO2 and CH4) that have an impact on global warming. In contrast, biological methods for biohydrogen production are more environmentally friendly, lower in energy consumption and have a cheaper substrate cost
3
ACCEPTED MANUSCRIPT
because hydrogen can be produced from raw materials such as organic waste, agricultural waste [7,8], industrial wastewater [9] and municipal waste [10]. These substrates are readily available and have been widely used in anaerobic fermentation to produce hydrogen gas in
RI PT
environmentally friendly ways [11].
Biohydrogen from cellulose is a high value-added product, since cellulosic biomass is the most abundant renewable resource on earth. In addition, biohydrogen can be produced from non
SC
food crops of inedible waste products and does not divert food away from the animal or human food chain. However, pretreatment is an important tool for the cellulosic bioconversion process
M AN U
because it has great potential for improving the efficiency of the anaerobic process. Thailand is an agricultural country and sugarcane is one of the important industrial crops. It can be cultivated in all parts of Thailand, except in the South, with a cultivation area of more than 960,000 ha. Approximately 48 Mt of sugarcane are produced each year [12]. Sugarcane bagasses (SCB) is a
TE D
waste left after the sugarcane extraction process. Since the bagasse accounts for approximately 25% of sugarcane mass, about 12 Mt of SCB are produced annually. The most common use for SCB is for energy production by combustion [13] with can cause environmental problems
EP
because of the emissions of CO2. SCB consists of three main fractions, i.e., cellulose, hemicelluloses and lignin of which 30-35% is hemicelluloses [14]. The generation of hydrogen
AC C
from SCB using anaerobic fermentation usually requires substrate pretreatment procedures. Diluted acid treatment of hemicelluloses fraction in SCB yields a solution containing mainly glucose and xylose with a small amount of arabinose [14,15]. Since bonds in cellulose are stronger than in hemicelluloses, a solid waste of cellulose and lignin is obtained in the diluted acid hydrolysis of SCB [15]. Due to its composition, hydrolysate of SCB is a very attractive raw material for the production of hydrogen. An important consideration in hydrogen production
4
ACCEPTED MANUSCRIPT
from SCB hydrolysate though, is that during hydrolysis several inhibitory compounds are formed [16]. Several factors affect biohydrogen production in addition to nature of the microbial flora
RI PT
[17] such as temperature, pH, mineral medium formulation, the type of inoculums, the profile of organic acids produced and the type and concentration of substrate. Particularly, pH is a key parameter in biological processes as it affects enzyme activities, metabolite transporters, and the
SC
microbial community. Therefore the production media formulation must include buffering compounds such as sodium or ammonium biocarbonate to reduce pH variations during
M AN U
cultivations. Teli et al. [18] reported the buffering capacity, strongly affects the biohydrogen yield. However, high concentrations of a buffer may have an inhibitory effect on anaerobic fermentation. An optimal substrate:buffer ratio of 2-2.5 was found which maintained the pH around 5-5.5. Therefore, the optimization of fermentation conditions are importance for
TE D
biohydrogen production.
In conventional multifactor experiments, optimization is usually carried out by varying a single factor while keeping all other factors fixed at a specific set of conditions. This is not only
EP
time-consuming, but it is usually impossible to reach the true optimum because the process ignores the interactions among the other variables. As a result, response surface methodology
AC C
(RSM) has been proposed to determine the influences of individual factors and their interactive influences. RSM is a statistical technique for designing experiments, building models, evaluating the effects of several factors, and searching for optimum conditions [19]. Recent studies have looked at the optimization on hydrogen production [1,20]. However, the statistical optimization on biohydrogen production from cellulosic hydrolysate is still lacking in the literature.
5
ACCEPTED MANUSCRIPT
The present study investigats the effects of substrate concentration, substrate:buffer ratio and inoculum:substrate ratio on hydrogen production from SCB hydrolysate. A central composite experimental design was employed in planning the experiment in order to find out
polynomial quadratic equation. 2. Material and methods
SC
2.1. Sugarcane bagasse hydrolysate
RI PT
which experiment variables affect hydrogen production potential by using RSM and a predictive
SCB used in this study was obtained from a local sugar processing plant (Thai Roong
M AN U
Reuang sugar industry, Phitsanulok, Thailand). The SCB was air dried, milled and sieved through a 0.5 mm screen before being stored at room temperature prior to usage. The composition (w/v) of the SCB was 38.1% cellulose, 21.2% hemicelluloses, 8.3% lignin, 2.5% ash and 40.1% of other components.
TE D
Acid hydrolysis of the SCB hemicellulose fraction was conducted at 121 °C, 0.15 MPa for 60 min with 0.5% (v/v) sulfuric acid in an autoclave at a solid to liquid ratio of 1:15 (g:mL). After hydrolysis, a solid residue was separated by filtration through a thin layer cloth. The pH of
EP
the hydrolysate was adjusted to 10 with Ca(OH)2. The resulting precipitate was removed by centrifugation (1500 rpm, 15 min) and then re-acidified to pH 7, followed by further
AC C
centrifugation. The detailed procedures are delineated in a previous study [1]. The supernatant was concentrated until a total-sugar content of 200 mg L-1 was achieved and kept at 4 °C prior to use. Total sugar was determined by the phenol sulfuric acid method [21]. 2.2. Seed microorganisms
The seed microorganisms in this study were taken from a 200 L hydrogen pilot plant, with a volatile suspended solid (VSS) of 3.23 g L-1. This reactor was continuously fed with 20
6
ACCEPTED MANUSCRIPT
mg-total sugar L-1 and maintained at pH 6. Prior to use, the seed sludge was first washed with 0.75% w/v of NaCl2 and heated at 94 °C for 60 min to inhibit the bioactivity of the hydrogen consumers and to harvest spore-forming anaerobic bacteria.
RI PT
2.3. Experimental design
In order to assess the optimization of factors affecting hydrogen production from SCB hydrolysate, a central composite experimental design (CCD) and RSM were conducted to
SC
evaluate the key variables influencing hydrogen production potential from three factors; substrate concentration (X1), substrate:buffer ratio (X2) and inoculum:substrate ratio (X3). RSM
M AN U
is an empirical statistical technique employed for multiple regression analysis using quantitative data obtained from properly designed experiments to simultaneously solve multivariate equations. The graphical representations of these equations are called response surface, which can be used to describe the individual and cumulative effects of the test variables on the
TE D
production of biogas. It can also be used to determine the mutual interactions between the test variables and their effect on the production of biogas [22]. The range and the levels of variables employed in this study are listed in Table 1. The center values (zero level) chosen for
EP
experimental design were: substrate concentration = 20 g-total sugar L-1, substrate:buffer ratio = 4, and inoculum:substrate ratio = 0.2. In developing the regression equation, the test variables
AC C
were coded according to the equation:
X − X i* X i i ∆X i
(1)
Where xi is the coded value of the ith test variable, Xi is an uncoded value of the ith test variable, Xi* is an uncoded value of the ith test variable at the center point and ∆Xi is the step size in varing the value. 7
ACCEPTED MANUSCRIPT
A total of twenty experimental runs (Table 1), in triplicate were conducted [22]. The value of hydrogen production potential was obtained as the response of the experiments. To predict the optimal condition, the quadratic polynomial equation was fitted to correlate between
following equation:
3
3
i =1
i =1
3
3
(2)
M AN U
i =1 j =1
SC
Y = A0 + ∑ Ai X i + ∑ Aii X i2 + ∑∑ Aij X i X j
RI PT
variables, the response (i.e., hydrogen production potential) and estimated response as the
Where Xi are the input variables which influence the response variable Y, A0 is the offset term, Ai the linear effect, Aii the squared effect and Aij is the interaction effect. The input values of X1, X2 and X3 corresponding to the maximum value of Y were solved by setting the partial
2.4. Hydrogen production
TE D
derivatives of the functions to zero.
Hydrogen production experiments were performed in 200 mL serum bottles with a
EP
working volume of 100 mL containing 1.5 mL of 3.75% (w/v) L-cysteine as a reducing agent, various amounts of substrate concentration, NaHCO3 as a buffer, and seed microorganisms. The
AC C
concentration of the substrate ranged from 10 to 30 g-total sugar L-1. The ratio of substrate to buffer and the ratio of inoculum to substrate were calculated
from the total sugar. The fermentation broth contained different concentrations according to the parameters in the design (Table 1). Each bottle was supplemented with nutrients stock solution (Endo formulation) at a dosage of 0.5 mL L-1. The nutrient stock solution contains following in the mg L-1; 5240 NH4HCO3, 125 K2HPO4, 100 MgCl2· 6H2O, 15 MnSO4· 6H2O, 25
8
ACCEPTED MANUSCRIPT
FeSO4· 7H2O, 5 CuSO4· 5H2O, 0.125 CoCl2· 5H2O which is modified from Lin and Lay [17]. The solution is then filled to 100 mL with distilled water and the pH was adjusted to 6 using either 1
RI PT
N HCl or 1 N NaOH. Subsequently, the bottles were tightly sealed using rubber septa and an aluminum cap. After replacement of the gas phase with argon to create an anaerobic condition, the serum bottle was incubated at 37 °C and put in an orbital shaker set at 150 rpm. All treatments were carried out in triplicate.
SC
2.5. Analytical method
Biogas volume was determined by a gas tight syringe at room temperature (20 °C) and
M AN U
pressure (760 mm Hg). The cumulative hydrogen gas production was determined by using the following equation [23].
(3)
TE D
VH ,i = VH ,i −1 + C H ,i (VG ,i − VG ,i −1 ) + VH (C H ,i − C H ,i −1 )
Where VH,i and VH,i-1 are the cumulative hydrogen gas volumes at the current (i) and previous time interval (i-1), respectively; VG,I and VG,i-1 the total biogas volume at the current and
EP
previous time intervals; CH,i and CH,i-1 is the fraction of hydrogen gas in the headspace at the
AC C
current and previous time intervals; and VH is the volume of headspace of vials (100 mL). The removed biogas volume for analysis was also taken into account during the hydrogen mass balance.
The composition of the product gas was measured with a CHINA Gas Chromatography 8700T equipped with a thermal conductivity detector and stainless steel column packed with Porapak Q. Argon was used as the carrier gas. The temperature of the injector, column and detector were all 50 °C.
9
ACCEPTED MANUSCRIPT
The concentration of organic acids and alcohols were determined with a flame ionization detector and glass column packed FON. The injector, column and detector temperature used were at 175, 145 and 175 °C respectively, with nitrogen as the carrier gas. The same methods
RI PT
were used in the previous studies [24,25].
pH, Total sugar, suspended solids (SS), and VSS were determined according to Standard Methods [26].
SC
2.6. Kinetic modeling
The cumulative volume of hydrogen production in the batch experiment followed the
M AN U
modified Gompertz equation:
(4)
TE D
R e H (t ) = P exp− exp m (λ − t ) + 1 P
Where H is the cumulative hydrogen production (mL); λ the lag time (h), P the hydrogen production potential (mL), and Rm the maximum hydrogen production rate (mL h-1).
3.1.
EP
3. Results and discussion
Overall performance in a typical experiment
AC C
The present study aims at understanding the effects of important process variables i.e. substrate concentration, substrate:buffer ratio and inoculum:substrate ratio for fermentative hydrogen production from SCB hydrolysate, and is partly based on our previous experiment [1]. The cumulative hydrogen production was measured and calculated using Eqs. 3 and 4 with the modified Gompertz equation, and P, Rm and λ were calculated. The regression coefficients were greater than 0.90 for all trials, indicating that the parameters were statistically significant.
10
ACCEPTED MANUSCRIPT
Cumulative hydrogen production reached the maximum values within 4 d. Methane was not observed in any bottles and λ for hydrogen production was less than 0.37 d, which was shorter than previously reported values (0.78-4 d) for batch tests with heat-treated inoculums
RI PT
[27]. Organic acid productions for all batch reactors included mainly n-butyrate, acetate and propionate are shown in Table 2. Among organic acid productions, n-butyrate was produced almost simultaneously with hydrogen production. The butyrate to acetate ratio (HBu/HAc) is
SC
often used to as an indicator of the extent of biohydrogen production. The range of HBu/HAc varied between 1.0-2.5. Previous studies indicated that efficient hydrogen occurs for HBu/HAc
M AN U
ratios between 2.6 and 4.0 [7]. This suggests that the metabolite composition is greatly dependent on the type of carbon substrate and bacterial population used. Our results suggest that SCB hydrolysate is a potential source of renewable biomass to produced hydrogen and inhibitory compounds are not formed during SCB hydrolysis.
Optimization Study for Maximum Hydrogen Production Rate
TE D
3.2.
The optimal values of the three factors i.e. substrate concentration, substrate:buffer ratio and inoculums:substrate ratio and their interactions on hydrogen production were further
EP
explored by CCD and RSM.
The statistical model was developed by applying multiple regression analysis on the
AC C
experimental data and the following second-order polynomial equation was established to calculate the hydrogen production:
HP = +6961.53 + 109.35*X1 + 190.88*X2 + 77.45*X3 + 137.75 *X1*X2 + 41.25*X1*X3 + 12.75*X2*X3 - 887.52 *X12 - 645.16*X22 - 414.29*X32
11
(5)
ACCEPTED MANUSCRIPT
X1, X2 and X3 are the coded values of substrate concentration, substrate:buffer ratio and inoculums:substrate ratio, respectively. Table 3 shows the results of the statistical model for the
RI PT
F-test, and the analysis of variance (ANOVA) of the response surface quadratic model. The ANOVA of quadratic regression model demonstrates that the model is strongly statistically significant (P < 0.0001). R2 was found to be 0.995, which means that the model could explain
SC
99.5% of the total variations in the system. In regression, the R2 coefficient is a statistical measure of how well the regression line approximates the real data points. An R2 of 1.0 indicates
M AN U
that the regression line perfectly fits the data [28]. The value of the regression coefficient (R2 = 0.995) suggests that the regression model was an accurate representation of the experimental data. The result shows that the quadratic model in terms of substrate concentration (X1), substrate:buffer ratio (X2), inoculums:substrate ratio (X3) are significant influences on hydrogen
TE D
production potential (P-value < 0.05).
The results also demonstrate there is a significant interaction between the substrate concentration and substrate:buffer ratio (X1X2) (P-value = 0.0019) while the interaction between
EP
the substrate concentration and inoculums:substrate ratio (X1X3) (P-value = 0.2381), the substrate:buffer ratio and inoculums:substrate ratio (X2X3) (P-value = 0.7063) are not significant
3.3.
AC C
(Table 3).
Effects of substrate concentration, substrate:buffer ratio and inoculums:substrate ratio The three-dimensional response surface and two-dimensional contour lines as shown in
Fig. 1a and 1b are based on Eq. 5 with one variable kept constant at its optimum level and varying the other two variables within the experimental range. The response surface of hydrogen production potential (Fig. 1a) shows a clear peak, indicating that optimum conditions fell inside
12
ACCEPTED MANUSCRIPT
the design boundary. The effect of the substrate concentration was statistically significant (P = 0.0015). hydrogen production potential increased with the increase in substrate concentration from 10 to 22.77 g-total sugar L-1 and then hydrogen production potential decreased when
RI PT
substrate concentration was greater than 22.77 g-total sugar L-1. This is expected because more carbon is available at a higher total sugar concentration. Hydrogen production decreased at a total sugar concentration of 30 g L-1 which is probably due to substrate inhibition.
SC
Previous work has shown the effects of substrate concentration on hydrogen production. Initial substrate concentration plays an important role on the yield and production rate of
M AN U
hydrogen [29]. Low initial substrate concentration results in a low rate of fermentation. Fermentation time increases as substrate concentration increases [30]. Further evidence can also be found that substrate concentration influences the hydrogen production. The increase in substrate concentration could increase hydrogen production up to a certain level. Also, excessive
TE D
substrate concentration can cause a buildup of cell concentration and volatile fatty acids (VFAs) in the system leading to a decline of pH in the reactor which could inhibit hydrogen production [31]. In addition, an increase in substrate concentration could lead to buildup partial pressure in
EP
the fermentation system. The accumulation of partial pressure in the headspace of reactor could cause the hydrogen production to be switched to solvent production resulting in reduced
AC C
hydrogen production [32]. An inhibitory effect of high substrate concentration generally occurs in anaerobic digeston processes, depending on the types of substrates and microorganisms. Chen et al. [25] reported that hydrogen production from sucrose by Clostridium butyricum CGS5 showed the best performance at the initial sucrose concentration of 20 g-COD L-1, while the fermentation process was inhibited at the initial sucrose concentration of 30 g-COD L-1. This agrees with Oh et al. [32] who reported biohydrogen generated by Citrobacter sp. Y19 with
13
ACCEPTED MANUSCRIPT
glucose as a carbon source showed the yield of hydrogen production gradually decreased with increasing glucose concentration at the levels higher than 20 g L-1. Our previous results show that the optimum substrate concentration for biohydrogen production from the fermentation of
RI PT
SCB hydrolysate by C. butyricum was 20 g-COD L-1 [1]. In the present study, hydrogen production potential reached the peak value of 6,980 mL L-1 at 22.77 g-total sugar L-1 initial substrate concentration.
SC
Figure 1c and 1d illustrate the interactive effect of substrate concentration and the inoculums:substrate ratio on hydrogen production potential with its optimums shown at the peak
M AN U
of the response curve (Fig. 1c). Hydrogen production potential increases with increasing substrate concentration and inoculums:substrate ratio at low values but it decreases with further increasing in substrate concentration and inoculums:substrate ratio at high levels. The highest hydrogen production potential (6,980 mL L-1) was obtained when the inoculums:substrate ratio
TE D
was 0.31. The two-dimensional contour of hydrogen production potential with respect to substrate concentration and inoculums:substrate ratio (Fig. 1d), showes an ellipse shaped contour plot (Fig. 1d). However, the interactive effect of the substrate concentration and
EP
inoculums:substrate ratio on hydrogen production potential was not significant (P > 0.05). This phenomenon was attributed to the fact that the ratio of inoculums to substrate can increase the
AC C
hydrogen production rate. However, at a higher inoculum volume, more of the carbon source was devoted for biomass production than hydrogen production [33]. The graphical representation provides a method to visualize the relationship between the
response and experimental values of inoculums:substrate ratio and substrate:buffer ratio on hydrogen production potential for the purpose of finding the optimum conditions (Figs. 1e and 1f with the optimum shown on the top of the surfaces and the smallest ellipse in the contour line.
14
ACCEPTED MANUSCRIPT
The highest hydrogen production potential was obtained at a ratio of substrate to buffer of 4.31. The interaction between substrate:buffer ratio and inoculums:substrate ratio (X2X3) was not significant (P > 0.05). However, the ratio of substrate to buffer was significant (P < 0.05),
RI PT
indicating that the buffer plays an important effect on hydrogen production potential. The formation of hydrogen is accompanied by the production of VFAs or other solvents during the anaerobic digestion process, which causes the pH to drop. Therefore, alkalinity in the form of
SC
carbonate or bicarbonate as a supplement is necessary in the reactor. pH is important as one of the factors controlling the anaerobic biological processes due to its effects on Fe-hydrogenase
M AN U
activity, metabolic pathways, and the duration of the lag phase [34]. A bacterial medium always contains buffering compounds to reduce the pH fluctuations during the initial growth periods. In an anaerobic reactor, the pH value and its stability are important. The potential of a substrate to generate acids can be predicted by its composition. Carbohydrate-rich substrates have a greater
TE D
potential to acidify the media during anaerobic fermentation. As a result, in some situations, system stability is hard to maintain. In hydrogen fermentation, reactors tend to acidify readily and a reduction in the pH may result [35]. Low initial pH values below 5.0 inhibit hydrogen
EP
production [36]. On the other hand, high initial pH values such as 9.0 decrease lag time, but have a lower level of hydrogen production [37]. In these cases, it becomes necessary to either add
AC C
external sources of alkalinity or to adopt adequate operational strategies [38,39]. In biohydrogen production systems, bicarbonate alkalinity plays an important role, as it allows buffering of pH when VFA production becomes excessive. However, biocarbonate interacts with CO2, another major gaseous end product in anaerobic systems. An increase in the biocarbonate concentration in the feed increases the CO2 fraction in the gas phase because of carbonate dissolution [39]. 3.4.
Confirmation experiments and adequacy of the models
15
ACCEPTED MANUSCRIPT
To confirm the validity of the statistical experimental strategies and to gain a better understanding of hydrogen production from SCB hydrolysate by mixed culture, four additional confirmation experiments were conducted. The chosen conditions for substrate concentration,
RI PT
substrate:buffer ratio and inoculums:substrate ratio for hydrogen production potential are listed in Table 4. The data in Table 4 reveas that the optimum hydrogen production potential measured
technique to optimize the biohydrogen production.
4. Conclusions
SC
was close to those estimated using RSM. This confirms that the RSM analysis is a useful
M AN U
The present work focused on optimizing factors affecting hydrogen production from SCB hydrolysate by mixed cultures. The CCD/RSM approach was employed for research planning. The optimal condition for hydrogen production in the culture was at pH 6 and 37 °C. Experimental results showed that buffer:substrate ratio had a significant influence on the
TE D
hydrogen production potential. A maximum hydrogen production potential of 6,980 mL H2 L-1 was estimated at the optimum conditions of substrate concentration 22.77 g-total sugar L-1, 4.31 substrate:buffer ratio, and 0.31 inoculum:substrate ratio. Four additional experiments confirm
EP
that the optimum hydrogen production potential measured was close to the estimated value using the RSM. The experimental results demonstrated that RSM with the CCD was useful for
AC C
optimizing hydrogen production from SCB hydrogen by mixed culture.
Acknowledgements
This work was supported by research funds from The Research Group for Development
of Microbial Hydrogen Production Process from Biomass, the Higher Education Research promotion and National Research University Projects for Thailand office of the Higher Education Commission. We also thank Prof. Chiu-Yue Lin from BioHydrogen Laboratory,
16
ACCEPTED MANUSCRIPT
Department of Environmental Engineering and Science, Feng Chia University for his assistance in the experimental work.
Refferences Pattra S, Sangyoka S, Boonmee M, Reungsang A. Bio-hydrogen production from the
RI PT
[1]
fermentation of sugarcane bagasse hydrolysate by Clostridium butyricum. Int J Hydrogen Energ 2008;33:5256–65.
Sreela-or C, Plangklang P, Imai T, Reungsang A. Co-digestion of food waste and sludge
SC
[2]
Hydrogen Energ 2011;36:14227–37. [3]
M AN U
for hydrogen production by anaerobic mixed cultures: statistical key factors optimization. Int J
Zhang XL, Van S, Tyagi RD, Surampalli RY. Biodiesel production from heterotrophic
microalgae through transesterification and nanotechnology application in the production. Renew Sust Energ Rev 2013;26:216–23.
Huang YT, Su CP. High lipid content and productivity of microalgae cultivating under
TE D
[4]
elevated carbon dioxide. Int J Environ Sci Te 2014;11:703–10. [5]
Komolafe O, Orta SBV, Monje-Ramirez I, Noguez IY, Harvey AP, Ledesma MTO.
EP
Biodiesel production from indigenous microalgae grown in wastewater. Bioresource Technol 2014;154:297–304.
[7]
Berman B. Fuel cells at center stage. The New York Times 2013 Nov 22.
AC C
[6]
Lin CY, Hung CH, Chen CH, Chung WT, Cheng LH. Effects of initial cultivation pH on
fermentative hydrogen production from xylose using natural mixed cultures. Process Biochem 2006;41:1383–90. [8]
Guo XM, Trably E, Latrille E, Carrere H, Steyer JP. Hydrogen production from
agricultural waste by dark fermentation: a review. Int J Hydrogen Energ 2010;35:10660–73.
17
ACCEPTED MANUSCRIPT
[9]
Chong ML, Sabaratnam V, Shirai Y, Hassan MA. Biohydrogen production from biomass
and industrial wastes by dark fermentation. Int J Hydrogen Energ 2009;34:3277–87. [10]
Aguilar MAR, Fdez-Güelfo LA, Álvarez-Gallego CJ, García LIR. Effect of HRT on
RI PT
hydrogen production and organic matter solubilization in acidogenic anaerobic digestion of OFMSW. Chem Eng J 2013;219:443–9. [11]
Singh L, Wahid ZA, Siddiqui MF, Ahmad A, Ab Rahim MH, Sakinah M. Biohydrogen
polyethylene glycol. Process Biochem 2013;48:294–8.
OCSB. Report on Cane and Sugar Production. Bangkok, Thailand: Office of the Cane
M AN U
[12]
and Sugar Board; 2011. [13]
SC
production from palm oil mill effluent using immobilized Clostridium butyricum EB6 in
Neureiter M, Danner H, Thomasser C, Saidi B, Braun R. Dilute-acid hydrolysis of
sugarcane bagasse at varying conditions. Appl Biochem Biotechnol 2002;98:49–58. Gámez S, González-Cabriales JJ, Ramírez JA, Garrote G, Vázquez M. Study of the
TE D
[14]
hydrolysis of sugar cane bagasse using phosphoric acid. J Food Eng 2006;74:78–88. [15]
Aguilar R, Ramirez JA, Garrote G, Vázquez M. Kinetic study of the acid hydrolysis of
[16]
EP
sugar cane bagasse. J Food Eng 2002;55:309–18. Singh SP, Asthana RK, Singh AP. Prospects of sugarcane milling waste utilization for
[17]
AC C
hydrogen production in India. Energ Policy 2007;35:4164–8. Lin CY, Lay CH. A nutrient formulation for fermentative hydrogen production using
anaerobic sewage sludge microflora. Int J Hydrogen Energ 2005;30:285–92. [18]
Teli A, Ficara E, Malpei E. Bio-hydrogen production from cheese whey by dark
fermentation. Chem Eng Trans 2014;37:613–8. [19]
Wang JP, Chen YZ, Ge XW, Yu HQ. Optimization of coagulation-flocculation process
18
ACCEPTED MANUSCRIPT
for a paper-recycling wastewater treatment using response surface methodology. Colloid Surface A 2007;302:204–10. [20]
Pan CM, Fan YT, Xing Y, Hou HW, Zhang ML. Statistical optimization of process
RI PT
parameters on biohydrogen production from glucose by Clostridium sp. Fanp2. Bioresource Technol 2008;99:3146–54. [21]
Saha SK, Brewer CF. Determination of the concentrations of oligosaccharides, complex
SC
type carbohydrates, and glycoproteins using the phenol-sulfuric acid method. Carbohydr Res 1994;254:157–67.
Box GEP, Hunter WG, Hunter JS. Statistics for Experimenters. New York: Wiley; 1978.
[23]
Zheng XJ, Yu HQ. Inhibitory effects of butyrate on biological hydrogen production with
M AN U
[22]
mixed anaerobic cultures. J Environ Manage 2005;74:65–70. [24]
Lin CY, Chang RC. Hydrogen production during the anaerobic acidogenic conversion of
[25]
TE D
glucose. J Chem Technol Biot 1999;74:498–500.
Chen WM, Tseng ZJ, Lee KS, Chang JS. Fermentative hydrogen production with
Clostridium butyricum CGS5 isolated from anaerobic sewage sludge. Int J Hydrogen Energ
[26]
EP
2005;30:1063–70.
APHA. Standard Methods for the Examination of Water and Wastewater. 19th, ed.
[27]
AC C
Washington, DC: American Public Health Association; 1995. Kim SH, Han SK, Shin HS. Feasibility of biohydrogen production by anaerobic co-
digestion of food waste and sewage sludge. Int J Hydrogen Energ 2004;29:1607–16. [28]
Niladevi KN, Sukumaran RK, Jacob N, Anisha GS, Prema P. Optimization of laccase
production from a novel strain – Streptomyces psammoticus using response surface methodology. Microbiol Res 2009;164:105–13.
19
ACCEPTED MANUSCRIPT
[29]
Fabiano B, Perego P. Thermodynamic study and optimization of hydrogen production by
Enterobacter aerogenes. Int J Hydrogen Energ 2002;27:149–56. [30]
Jo JH, Lee DS, Park D, Choe WS, Park JM. Optimization of key process variables for
RI PT
enhanced hydrogen production by Enterobacter aerogenes using statistical methods. Bioresource Technol 2008;99:2061–6. [31]
Fan YT, Li CL, Lay JJ, Hou HW, Zhang GS. Optimization of initial substrate and pH
SC
levels for germination of sporing hydrogen-producing anaerobes in cow dung compost. Bioresource Technol 2004;91:189–93.
Oh YK, Seol EH, Kim JR, Park S. Fermentative biohydrogen production by a new
M AN U
[32]
chemoheterotrophic bacterium Citrobacter sp. Y19. Int J Hydrogen Energ 2003;28:1353–9. [33]
Kotay SM, Das D. Microbial hydrogen production with Bacillus coagulans IIT-BT S1
isolated from anaerobic sewage sludge. Bioresource Technol 2007;98:1183–90. Lay JJ. Modeling and optimization of anaerobic digested sludge converting starch to
TE D
[34]
hydrogen. Biotechnol Bioeng 2000;68:269–78. [35]
Vijayaraghavan K, Ahmad D, Ibrahim MKB, Herman HNB. Isolation of hydrogen
2006;31:708–20.
Chen CC, Lin CY, Lin MC. Acid-base enrichment enhances anaerobic hydrogen
AC C
[36]
EP
generating microflora from cow dung for seeding anaerobic digester. Int J Hydrogen Energ
production process. Appl Microbiol Biot 2002;58:224–8. [37]
Khanal SK, Chen WH, Li L, Sung SW. Biological hydrogen production: effects of pH
and intermediate products. Int J Hydrogen Energ 2004;29:1123–31. [38]
Ergüder TH, Güven E, Demirer GN. Anaerobic treatment of olive mill wastes in batch
reactors. Process Biochem 2000;36:243–8.
20
ACCEPTED MANUSCRIPT
[39]
Lin CY, Lay CH. Effects of carbonate and phosphate concentrations on hydrogen
AC C
EP
TE D
M AN U
SC
RI PT
production using anaerobic sewage sludge microflora. Int J Hydrogen Energ 2004;29:275–81.
21
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 1. Full factorial central composite design matrix defining substrate concentration, substrate: buffer ratio and inoculums:substrate ratio Substrate:Buffer ratio Inoculum:Substrate ratio Substrate concentration Hydrogen (X1) (X2) (X3) production CO2/H2 (g-total sugar L-1) (g L-1 of total sugar/g L-1 of NaHCO3) (g L-1 of VSS/g L-1 of total sugar) potential (P) Trial ratio Code Actual Actual value Code Actual value Code value (mL) value value Ratio (value) value Ratio (value) 1 -1.000 10.00 -1.000 2.00 (10/5) -1.000 0.20 (2/10) 4827 1.41 ± 0.22 2 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 6852 1.28 ± 0.02 3 1.000 30.00 -1.000 2.00 (30/15) 1.000 0.40 (12/30) 4938 1.24 ± 0.04 4 1.000 30.00 -1.000 2.00 (30/15) -1.000 0.20 (6/30) 4605 1.27 ± 0.04 5 0.000 20.00 -1.682 0.64 (20/31.25) 0.000 0.30 (6/20) 4794 1.29 ± 0.07 6 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 7019 1.17 ± 0.05 7 1.000 30.00 1.000 6.00 (30/5) -1.000 0.20 (6/30) 5229 1.13 ± 0.08 8 1.682 36.82 0.000 4.00 (36.82/9.21) 0.000 0.30 (11.05/36.82) 4721 1.09 ± 0.02 9 0.000 20.00 0.000 4.00 (20/5) 1.682 0.47 (9.4/20) 5836 1.17 ± 0.09 0 0.000 20.00 0.000 4.00 (20/5) -1.682 0.13 (2.6/20) 5791 1.45 ± 0.08 11 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 6934 1.26 ± 0.03 12 -1.000 10.00 1.000 6.00 (10/1.67) 1.000 0.40 (4/10) 5058 1.18 ± 0.01 13 0.000 20.00 1.682 7.35 (20/2.72) 0.000 0.30 (6/20) 5527 1.06 ± 0.06 14 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 6957 2.13 ± 0.13 15 -1.000 10.00 1.000 6.00 (10/1.67) -1.000 0.20 (2/10) 4839 1.08 ± 0.15 16 -1.682 3.18 0.000 4.00 (3.18/0.80) 0.000 0.30 (0.95/3.18) 4229 1.22 ± 0.13 17 1.000 30.00 1.000 6.00 (30/5) 1.000 0.40 (12/30) 5552 1.10 ± 0.90 18 -1.000 10.00 -1.000 2.00 (10/5) 1.000 0.40 (4/10) 4934 1.04 ± 0.03 19 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 7008 0.96 ± 0.02 20 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 6991 1.05 ± 0.05
22
ACCEPTED MANUSCRIPT
Butyric (HBu) (%) 55.4 57.5 52.8 49.9 55.2 59.1 58.7 50.0 60.0 65.1 57.9 58.1 60.6 47.7 57.9 59.8 55.1 49.3 50.2 54.4
pH range 4.05-4.61 4.35-4.36 4.33-4.35 4.56-4.56 5.17-5.19 4.35-4.37 4.06-4.36 4.09-4.29 4.17-4.48 5.19-5.34 4.28-4.42 4.51-4.61 4.40-4.62 6.11-6.19 4.18-4.32 4.37-4.85 4.16-4.32 4.14-4.16 4.08-4.10 4.19-4.31
M AN U
SC
1.53 1.64 1.23 1.16 1.50 1.78 1.58 1.19 1.84 2.52 1.66 1.73 1.71 1.09 2.04 1.49 1.50 1.16 1.01 1.43
RI PT
HBu/HAc
AC C
EP
TE D
Table 2. pH range and VFA distribution in the final product Acetic Propionic Trial VFA (mg L-1) (HAc) (HPr) (%) (%) 1 3924 36.2 8.5 2 4195 35.1 7.4 3 4294 42.9 4.4 4 5582 43.2 6.9 5 5071 36.8 8.0 6 4034 33.2 7.8 7 3517 37.1 4.2 8 4398 42.2 7.9 9 4191 32.7 7.4 10 4209 25.8 9.1 11 4145 35.0 7.2 12 2750 33.6 8.3 13 2478 35.4 4.0 14 2728 43.7 8.6 15 3669 28.5 13.6 16 3041 40.2 0.0 17 3875 36.7 8.2 18 4257 42.6 8.1 19 2464 49.8 0.0 20 3633 37.9 7.8
23
ACCEPTED MANUSCRIPT Table 3. Analysis of variance for quadratic polynomial model Sum of squares
DF
Mean square
F-value
Model X1 X2 X3 X1X2 X1X3 X2X3 X12 X22 X32 R2
1.789E + 007 1.633E + 005 4.976E + 005 81914.06 1.518E + 005 13612.50 1300.50 1.135E + 007 5.998E + 006 2.474E + 006 0.9952
9 1 1 1 1 1 1 1 1 1
1.988E + 006 1.633E + 005 4.976E + 005 81914.06 1.518E + 005 13612.50 1300.50 1.135E + 007 5.998E + 006 2.474E + 006
229.91 18.89 57.54 9.47 17.55 1.57 0.15 1312.66 693.63 286.03
SC M AN U TE D EP AC C 24
P-value Prob > F < 0.0001 0.0015 < 0.0001 0.0117 0.0019 0.2381 0.7063 < 0.0001 < 0.0001 < 0.0001
RI PT
Source
ACCEPTED MANUSCRIPT
RI PT
Hydrogen production potential (mL H2 L-1 substrate) Measure 6054 ± 40 5770 ± 20 6243 ± 58 6928 ± 68
EP
TE D
Bias was calculated using the equation: [(calculated value – experimental value)/calculated value] x 100
AC C
a
M AN U
SC
Table 4. Measured and calculated values of the confirmation experiments Inoculums: Substrate:buffer ratio Substrate substrate ration (g L-1 of total sugar/g L-1 Trial (g-total sugar L-1) (g L-1 of VSS/g L-1 of total of NaHCO3) sugar) Ratio (value) Ratio (value) 21 20 2 (20/10) 0.30 (6/20) 22 20 4 (20/5) 0.20 (4/20) 23 20 6 (20/3.33) 0.30 (6/20) 24 22.77 4.31 (22.71/5.30) 0.31 (7.04/22.71)
25
Calculated 6094 6972 6601 6980
Bias%a
0.70 17.20 5.40 0.75
ACCEPTED MANUSCRIPT 6.00
(b) 6520 5940 5360 4780 4200
6.00
X2: Substrate:Buffer ratio (g L-1 of total sugar/g L-1 of NaHCO3)
3.00
15.00 10.00
6094.75 5799.69
X1: Substrate concentration (g-total sugar L-1)
(c)
5799.69
6389.81
5504.63
15.00
20.00
25.00
30.00
X1: Substrate concentration (g-total sugar L-1)
(d) X3: Inoculum/Substrate ratio (g L-1 of VSS/g L-1 of total sugar)
SC
0.40
7100 6520 5940 5360 4780 4200
0.40
0.35
0.30
0.25
6
6241.37
M AN U
6241.37
5998.97
6726.18 5998.97
30.00
0.35
5756.57
25.00
0.30 X3: Inoculum:Substrate ratio (g L-1 of VSS/g L-1 of total sugar)
20.00 0.25
15.00 0.20
7100 6520 5940 5360
EP
4780
AC C 0.30
3.00 0.20
2.00
(f)
20.00
25.00
30.00
X1: Substrate concentration (g-total sugar L-1)
0.40 6090.81
0.35
0.30
6 6312.97
0.25
6757.28
6090.81
6312.97
5.00
6535.12
5868.66
4.00
0.25
15.00
6.00
0.35
X2: Inoculum:Substrate ratio (g L-1 off VSS/g L-1 of total sugar)
6483.78
X1: Substrate concentration (g-total sugar L-1)
TE D
(e)
10.00
0.20 10.00
X3: Inoculum:Substrate ratio (g L-1 of VSS/g L-1 of total sugar)
Hydrogen production potential (mL)
6684.86
3.00
2.00 10.00
20.00 2.00
Hydrogen production potential (P)
6094.75
5504.63
25.00 4.00
0.40
6
4.00
30.00 5.00
4200
5.00
RI PT
7100
5799.69
X2: Substrate:Buffer ratio (g L-1 of total sugar/g L-1 of NaHCO3)
Hydrogen production potential (mL)
(a)
0.20
X3: Substrate:Buffer ratio (g-total sugar L-1)
2.00
3.00
4.00
5.00
6.00
X2: Substrate:Buffer ratio (g L-1 of total sugar/g L-1 of NaHCO3)
Fig. 1. Response surface and contour plots for the effect on substrate concentration, substrate:buffer ratio and inoculums:substrate ratio on hydrogen production potential. (a) Response surface and (b) contour plots for the effect on substrate concentration and substrate:buffer ratio on hydrogen production potential. (c) Response surface and (d) contour plots for the effect on substrate concentration and inoculums:substrate ratio on hydrogen production potential. (e) Response surface and (f) contour plots for the effect on substrate:buffer and inoculums:substrate ratio on hydrogen production potential.
26
S2468-2039(16)30032-2
DOI:
10.1016/j.serj.2016.05.001
Reference:
SERJ 32
To appear in:
Sustainable Environment Research
Received Date: 18 August 2015 Revised Date:
16 November 2015
Accepted Date: 15 February 2016
Please cite this article as: Sangyoka S, Reungsang A, Lin C-Y, Optimization of biohydrogen production from sugarcane bagasse by mixed cultures using a statistical method, Sustainable Environment Research (2016), doi: 10.1016/j.serj.2016.05.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Received 18 August 2015 Received in revised form 16 November 2015 Accepted 15 February 2016
statistical method
RI PT
Optimization of biohydrogen production from sugarcane bagasse by mixed cultures using a
SC
Suksaman Sangyoka,1,* Alissara Reungsang2,3 and Chiu-Yue Lin4,5,6
1
AC C
EP
TE D
M AN U
Faculty of Science and Technology Pibulsongkram Rajabhat University Phitsanulok 65000, Thailand 2 Department of Biotechnology Khon Kaen University Khon Kaen 40002, Thailand 3 Fermentation Research Center for Value Added of Agricultural Products Khon Kaen University Khon Kaen 40002, Thailand 4 Department of Environmental Engineering and Science Feng Chia University Taichung 40724, Taiwan 5 Master’s Program of Green Energy Science and Technology Feng Chia University Taichung 40724, Taiwan 6 Green Energy Development Center Feng Chia University Taichung 40724, Taiwan
Key Words: Biohydrogen, central composite design (CCD), response surface methodology (RSM), sugarcane bagasse, hydrolysate
*Correspondence author Email: [email protected]
1
ACCEPTED MANUSCRIPT
ABSTRACT The aim of this study is to determine the optimum condition for biohydrogen production from sugarcane bagasse (SCB) hydrolysate using a central composite design (CCD) and response
RI PT
surface methodology (RSM). SCB was hydrolyzed with 0.5% (v/v) sulfuric acid at 121 °C, 0.15 MPa for 60 min in an autoclave at a solid to liquid ratio of 1:15 (g:mL). Heat-treated bacterium obtained from a hydrogen producing fermentor was used as the inoculum. The interaction of
SC
three factors, i.e., substrate concentration, substrate:buffer ratio and inoculum:substrate ratio on hydrogen production potential (P) were investigated. The results indicated that the substrate
M AN U
concentration, substrate:buffer ratio and inoculum:substrate ratio had a significant influence effect on P. An optimal condition was found at substrate concentration of 22.77 g-toal sugar L-1, 4.31 substrate:buffer ratio, and 0.31 inoculum:substrate ratio resulted in a maximum P of 6,980 mL H2 L-1. The confirmation experiment results indicated that optimum P was statistically
TE D
significant, from the predicted value obtained by RSM which suggests that RSM could be efficiently used to optimize a biohydrogen production from SCB hydrolysate using mixed cultures. These results indicates that the SCB hemicellulose hydrolysate is suitable as a
EP
fermentation media for producing biohydrogen. This approach will add value to SCB by
AC C
converting agricultural waste into a safe and clean form of energy. 1. Introduction
Our energy supply comes mainly from fossil fuels for transportation and industrialization
resulting in not only environmental pollution, but also economic and political problems owing to their limited reserves and uneven distribution. Nowadays, alternative energy can come from natural resources (wind, sunlight, geothermal power and biomass) which are inexhaustible. Using these resources to supply our energy needs supports sustainable development, while also
2
ACCEPTED MANUSCRIPT
lowering greenhouse gas emissions. The different characteristics of an energy resource can be evaluated in terms of sustainability and their ability to replace traditional fuels at different levels including power generation, heating, as a transport fuel and for rural energy. In developing
RI PT
countries, biomass represents the major source of renewable energy because of its commonly used as the local energy supply. Bioenergy is a fuel derived from a biological source (biomass) and is also referred to as biofuel. Biomass is defined as any organic material coming from any
SC
form of life or its derived metabolic production. Biodiesel and bioethanol are biofuels that currently are the only alternative energy source able to replace transportation fuel in vehicle
M AN U
engines without involving major modifications. Previous researchs study biofuel from biomass such as agricultural waste [1], food waste [2] and microalgae [3-5].
On the other hand, some scientists predict hydrogen as the “fuel of the future”, and the worldwide trend is the shift away from non-renewable fuels towards the establishment of
TE D
renewable hydrogen energy. Many companies are working to develop technologies that might efficiently exploit the potential of hydrogen energy for use in motor vehicles. As of November 2013 there are demonstration fleets of hydrogen fuel cell vehicles undergoing field testing
EP
including; the Chevrolet Equinox Fuel Cell, Honda FCX Clarity, Hyundai ix35 FCEV and Mercedes-Benz B-Class F-Cell [6]. Hydrogen is considered to be an ideal energy source for the
AC C
future due to its cleanliness and high energy yield of 142 kJ g-1, which is 2.75 times greater than that of any hydrocarbon fuel. Hydrogen production technologies can be classified as physicalchemical and biological methods. Physical-chemical methods are very energy intensive and result in the emission of harmful greenhouse gases (CO, CO2 and CH4) that have an impact on global warming. In contrast, biological methods for biohydrogen production are more environmentally friendly, lower in energy consumption and have a cheaper substrate cost
3
ACCEPTED MANUSCRIPT
because hydrogen can be produced from raw materials such as organic waste, agricultural waste [7,8], industrial wastewater [9] and municipal waste [10]. These substrates are readily available and have been widely used in anaerobic fermentation to produce hydrogen gas in
RI PT
environmentally friendly ways [11].
Biohydrogen from cellulose is a high value-added product, since cellulosic biomass is the most abundant renewable resource on earth. In addition, biohydrogen can be produced from non
SC
food crops of inedible waste products and does not divert food away from the animal or human food chain. However, pretreatment is an important tool for the cellulosic bioconversion process
M AN U
because it has great potential for improving the efficiency of the anaerobic process. Thailand is an agricultural country and sugarcane is one of the important industrial crops. It can be cultivated in all parts of Thailand, except in the South, with a cultivation area of more than 960,000 ha. Approximately 48 Mt of sugarcane are produced each year [12]. Sugarcane bagasses (SCB) is a
TE D
waste left after the sugarcane extraction process. Since the bagasse accounts for approximately 25% of sugarcane mass, about 12 Mt of SCB are produced annually. The most common use for SCB is for energy production by combustion [13] with can cause environmental problems
EP
because of the emissions of CO2. SCB consists of three main fractions, i.e., cellulose, hemicelluloses and lignin of which 30-35% is hemicelluloses [14]. The generation of hydrogen
AC C
from SCB using anaerobic fermentation usually requires substrate pretreatment procedures. Diluted acid treatment of hemicelluloses fraction in SCB yields a solution containing mainly glucose and xylose with a small amount of arabinose [14,15]. Since bonds in cellulose are stronger than in hemicelluloses, a solid waste of cellulose and lignin is obtained in the diluted acid hydrolysis of SCB [15]. Due to its composition, hydrolysate of SCB is a very attractive raw material for the production of hydrogen. An important consideration in hydrogen production
4
ACCEPTED MANUSCRIPT
from SCB hydrolysate though, is that during hydrolysis several inhibitory compounds are formed [16]. Several factors affect biohydrogen production in addition to nature of the microbial flora
RI PT
[17] such as temperature, pH, mineral medium formulation, the type of inoculums, the profile of organic acids produced and the type and concentration of substrate. Particularly, pH is a key parameter in biological processes as it affects enzyme activities, metabolite transporters, and the
SC
microbial community. Therefore the production media formulation must include buffering compounds such as sodium or ammonium biocarbonate to reduce pH variations during
M AN U
cultivations. Teli et al. [18] reported the buffering capacity, strongly affects the biohydrogen yield. However, high concentrations of a buffer may have an inhibitory effect on anaerobic fermentation. An optimal substrate:buffer ratio of 2-2.5 was found which maintained the pH around 5-5.5. Therefore, the optimization of fermentation conditions are importance for
TE D
biohydrogen production.
In conventional multifactor experiments, optimization is usually carried out by varying a single factor while keeping all other factors fixed at a specific set of conditions. This is not only
EP
time-consuming, but it is usually impossible to reach the true optimum because the process ignores the interactions among the other variables. As a result, response surface methodology
AC C
(RSM) has been proposed to determine the influences of individual factors and their interactive influences. RSM is a statistical technique for designing experiments, building models, evaluating the effects of several factors, and searching for optimum conditions [19]. Recent studies have looked at the optimization on hydrogen production [1,20]. However, the statistical optimization on biohydrogen production from cellulosic hydrolysate is still lacking in the literature.
5
ACCEPTED MANUSCRIPT
The present study investigats the effects of substrate concentration, substrate:buffer ratio and inoculum:substrate ratio on hydrogen production from SCB hydrolysate. A central composite experimental design was employed in planning the experiment in order to find out
polynomial quadratic equation. 2. Material and methods
SC
2.1. Sugarcane bagasse hydrolysate
RI PT
which experiment variables affect hydrogen production potential by using RSM and a predictive
SCB used in this study was obtained from a local sugar processing plant (Thai Roong
M AN U
Reuang sugar industry, Phitsanulok, Thailand). The SCB was air dried, milled and sieved through a 0.5 mm screen before being stored at room temperature prior to usage. The composition (w/v) of the SCB was 38.1% cellulose, 21.2% hemicelluloses, 8.3% lignin, 2.5% ash and 40.1% of other components.
TE D
Acid hydrolysis of the SCB hemicellulose fraction was conducted at 121 °C, 0.15 MPa for 60 min with 0.5% (v/v) sulfuric acid in an autoclave at a solid to liquid ratio of 1:15 (g:mL). After hydrolysis, a solid residue was separated by filtration through a thin layer cloth. The pH of
EP
the hydrolysate was adjusted to 10 with Ca(OH)2. The resulting precipitate was removed by centrifugation (1500 rpm, 15 min) and then re-acidified to pH 7, followed by further
AC C
centrifugation. The detailed procedures are delineated in a previous study [1]. The supernatant was concentrated until a total-sugar content of 200 mg L-1 was achieved and kept at 4 °C prior to use. Total sugar was determined by the phenol sulfuric acid method [21]. 2.2. Seed microorganisms
The seed microorganisms in this study were taken from a 200 L hydrogen pilot plant, with a volatile suspended solid (VSS) of 3.23 g L-1. This reactor was continuously fed with 20
6
ACCEPTED MANUSCRIPT
mg-total sugar L-1 and maintained at pH 6. Prior to use, the seed sludge was first washed with 0.75% w/v of NaCl2 and heated at 94 °C for 60 min to inhibit the bioactivity of the hydrogen consumers and to harvest spore-forming anaerobic bacteria.
RI PT
2.3. Experimental design
In order to assess the optimization of factors affecting hydrogen production from SCB hydrolysate, a central composite experimental design (CCD) and RSM were conducted to
SC
evaluate the key variables influencing hydrogen production potential from three factors; substrate concentration (X1), substrate:buffer ratio (X2) and inoculum:substrate ratio (X3). RSM
M AN U
is an empirical statistical technique employed for multiple regression analysis using quantitative data obtained from properly designed experiments to simultaneously solve multivariate equations. The graphical representations of these equations are called response surface, which can be used to describe the individual and cumulative effects of the test variables on the
TE D
production of biogas. It can also be used to determine the mutual interactions between the test variables and their effect on the production of biogas [22]. The range and the levels of variables employed in this study are listed in Table 1. The center values (zero level) chosen for
EP
experimental design were: substrate concentration = 20 g-total sugar L-1, substrate:buffer ratio = 4, and inoculum:substrate ratio = 0.2. In developing the regression equation, the test variables
AC C
were coded according to the equation:
X − X i* X i i ∆X i
(1)
Where xi is the coded value of the ith test variable, Xi is an uncoded value of the ith test variable, Xi* is an uncoded value of the ith test variable at the center point and ∆Xi is the step size in varing the value. 7
ACCEPTED MANUSCRIPT
A total of twenty experimental runs (Table 1), in triplicate were conducted [22]. The value of hydrogen production potential was obtained as the response of the experiments. To predict the optimal condition, the quadratic polynomial equation was fitted to correlate between
following equation:
3
3
i =1
i =1
3
3
(2)
M AN U
i =1 j =1
SC
Y = A0 + ∑ Ai X i + ∑ Aii X i2 + ∑∑ Aij X i X j
RI PT
variables, the response (i.e., hydrogen production potential) and estimated response as the
Where Xi are the input variables which influence the response variable Y, A0 is the offset term, Ai the linear effect, Aii the squared effect and Aij is the interaction effect. The input values of X1, X2 and X3 corresponding to the maximum value of Y were solved by setting the partial
2.4. Hydrogen production
TE D
derivatives of the functions to zero.
Hydrogen production experiments were performed in 200 mL serum bottles with a
EP
working volume of 100 mL containing 1.5 mL of 3.75% (w/v) L-cysteine as a reducing agent, various amounts of substrate concentration, NaHCO3 as a buffer, and seed microorganisms. The
AC C
concentration of the substrate ranged from 10 to 30 g-total sugar L-1. The ratio of substrate to buffer and the ratio of inoculum to substrate were calculated
from the total sugar. The fermentation broth contained different concentrations according to the parameters in the design (Table 1). Each bottle was supplemented with nutrients stock solution (Endo formulation) at a dosage of 0.5 mL L-1. The nutrient stock solution contains following in the mg L-1; 5240 NH4HCO3, 125 K2HPO4, 100 MgCl2· 6H2O, 15 MnSO4· 6H2O, 25
8
ACCEPTED MANUSCRIPT
FeSO4· 7H2O, 5 CuSO4· 5H2O, 0.125 CoCl2· 5H2O which is modified from Lin and Lay [17]. The solution is then filled to 100 mL with distilled water and the pH was adjusted to 6 using either 1
RI PT
N HCl or 1 N NaOH. Subsequently, the bottles were tightly sealed using rubber septa and an aluminum cap. After replacement of the gas phase with argon to create an anaerobic condition, the serum bottle was incubated at 37 °C and put in an orbital shaker set at 150 rpm. All treatments were carried out in triplicate.
SC
2.5. Analytical method
Biogas volume was determined by a gas tight syringe at room temperature (20 °C) and
M AN U
pressure (760 mm Hg). The cumulative hydrogen gas production was determined by using the following equation [23].
(3)
TE D
VH ,i = VH ,i −1 + C H ,i (VG ,i − VG ,i −1 ) + VH (C H ,i − C H ,i −1 )
Where VH,i and VH,i-1 are the cumulative hydrogen gas volumes at the current (i) and previous time interval (i-1), respectively; VG,I and VG,i-1 the total biogas volume at the current and
EP
previous time intervals; CH,i and CH,i-1 is the fraction of hydrogen gas in the headspace at the
AC C
current and previous time intervals; and VH is the volume of headspace of vials (100 mL). The removed biogas volume for analysis was also taken into account during the hydrogen mass balance.
The composition of the product gas was measured with a CHINA Gas Chromatography 8700T equipped with a thermal conductivity detector and stainless steel column packed with Porapak Q. Argon was used as the carrier gas. The temperature of the injector, column and detector were all 50 °C.
9
ACCEPTED MANUSCRIPT
The concentration of organic acids and alcohols were determined with a flame ionization detector and glass column packed FON. The injector, column and detector temperature used were at 175, 145 and 175 °C respectively, with nitrogen as the carrier gas. The same methods
RI PT
were used in the previous studies [24,25].
pH, Total sugar, suspended solids (SS), and VSS were determined according to Standard Methods [26].
SC
2.6. Kinetic modeling
The cumulative volume of hydrogen production in the batch experiment followed the
M AN U
modified Gompertz equation:
(4)
TE D
R e H (t ) = P exp− exp m (λ − t ) + 1 P
Where H is the cumulative hydrogen production (mL); λ the lag time (h), P the hydrogen production potential (mL), and Rm the maximum hydrogen production rate (mL h-1).
3.1.
EP
3. Results and discussion
Overall performance in a typical experiment
AC C
The present study aims at understanding the effects of important process variables i.e. substrate concentration, substrate:buffer ratio and inoculum:substrate ratio for fermentative hydrogen production from SCB hydrolysate, and is partly based on our previous experiment [1]. The cumulative hydrogen production was measured and calculated using Eqs. 3 and 4 with the modified Gompertz equation, and P, Rm and λ were calculated. The regression coefficients were greater than 0.90 for all trials, indicating that the parameters were statistically significant.
10
ACCEPTED MANUSCRIPT
Cumulative hydrogen production reached the maximum values within 4 d. Methane was not observed in any bottles and λ for hydrogen production was less than 0.37 d, which was shorter than previously reported values (0.78-4 d) for batch tests with heat-treated inoculums
RI PT
[27]. Organic acid productions for all batch reactors included mainly n-butyrate, acetate and propionate are shown in Table 2. Among organic acid productions, n-butyrate was produced almost simultaneously with hydrogen production. The butyrate to acetate ratio (HBu/HAc) is
SC
often used to as an indicator of the extent of biohydrogen production. The range of HBu/HAc varied between 1.0-2.5. Previous studies indicated that efficient hydrogen occurs for HBu/HAc
M AN U
ratios between 2.6 and 4.0 [7]. This suggests that the metabolite composition is greatly dependent on the type of carbon substrate and bacterial population used. Our results suggest that SCB hydrolysate is a potential source of renewable biomass to produced hydrogen and inhibitory compounds are not formed during SCB hydrolysis.
Optimization Study for Maximum Hydrogen Production Rate
TE D
3.2.
The optimal values of the three factors i.e. substrate concentration, substrate:buffer ratio and inoculums:substrate ratio and their interactions on hydrogen production were further
EP
explored by CCD and RSM.
The statistical model was developed by applying multiple regression analysis on the
AC C
experimental data and the following second-order polynomial equation was established to calculate the hydrogen production:
HP = +6961.53 + 109.35*X1 + 190.88*X2 + 77.45*X3 + 137.75 *X1*X2 + 41.25*X1*X3 + 12.75*X2*X3 - 887.52 *X12 - 645.16*X22 - 414.29*X32
11
(5)
ACCEPTED MANUSCRIPT
X1, X2 and X3 are the coded values of substrate concentration, substrate:buffer ratio and inoculums:substrate ratio, respectively. Table 3 shows the results of the statistical model for the
RI PT
F-test, and the analysis of variance (ANOVA) of the response surface quadratic model. The ANOVA of quadratic regression model demonstrates that the model is strongly statistically significant (P < 0.0001). R2 was found to be 0.995, which means that the model could explain
SC
99.5% of the total variations in the system. In regression, the R2 coefficient is a statistical measure of how well the regression line approximates the real data points. An R2 of 1.0 indicates
M AN U
that the regression line perfectly fits the data [28]. The value of the regression coefficient (R2 = 0.995) suggests that the regression model was an accurate representation of the experimental data. The result shows that the quadratic model in terms of substrate concentration (X1), substrate:buffer ratio (X2), inoculums:substrate ratio (X3) are significant influences on hydrogen
TE D
production potential (P-value < 0.05).
The results also demonstrate there is a significant interaction between the substrate concentration and substrate:buffer ratio (X1X2) (P-value = 0.0019) while the interaction between
EP
the substrate concentration and inoculums:substrate ratio (X1X3) (P-value = 0.2381), the substrate:buffer ratio and inoculums:substrate ratio (X2X3) (P-value = 0.7063) are not significant
3.3.
AC C
(Table 3).
Effects of substrate concentration, substrate:buffer ratio and inoculums:substrate ratio The three-dimensional response surface and two-dimensional contour lines as shown in
Fig. 1a and 1b are based on Eq. 5 with one variable kept constant at its optimum level and varying the other two variables within the experimental range. The response surface of hydrogen production potential (Fig. 1a) shows a clear peak, indicating that optimum conditions fell inside
12
ACCEPTED MANUSCRIPT
the design boundary. The effect of the substrate concentration was statistically significant (P = 0.0015). hydrogen production potential increased with the increase in substrate concentration from 10 to 22.77 g-total sugar L-1 and then hydrogen production potential decreased when
RI PT
substrate concentration was greater than 22.77 g-total sugar L-1. This is expected because more carbon is available at a higher total sugar concentration. Hydrogen production decreased at a total sugar concentration of 30 g L-1 which is probably due to substrate inhibition.
SC
Previous work has shown the effects of substrate concentration on hydrogen production. Initial substrate concentration plays an important role on the yield and production rate of
M AN U
hydrogen [29]. Low initial substrate concentration results in a low rate of fermentation. Fermentation time increases as substrate concentration increases [30]. Further evidence can also be found that substrate concentration influences the hydrogen production. The increase in substrate concentration could increase hydrogen production up to a certain level. Also, excessive
TE D
substrate concentration can cause a buildup of cell concentration and volatile fatty acids (VFAs) in the system leading to a decline of pH in the reactor which could inhibit hydrogen production [31]. In addition, an increase in substrate concentration could lead to buildup partial pressure in
EP
the fermentation system. The accumulation of partial pressure in the headspace of reactor could cause the hydrogen production to be switched to solvent production resulting in reduced
AC C
hydrogen production [32]. An inhibitory effect of high substrate concentration generally occurs in anaerobic digeston processes, depending on the types of substrates and microorganisms. Chen et al. [25] reported that hydrogen production from sucrose by Clostridium butyricum CGS5 showed the best performance at the initial sucrose concentration of 20 g-COD L-1, while the fermentation process was inhibited at the initial sucrose concentration of 30 g-COD L-1. This agrees with Oh et al. [32] who reported biohydrogen generated by Citrobacter sp. Y19 with
13
ACCEPTED MANUSCRIPT
glucose as a carbon source showed the yield of hydrogen production gradually decreased with increasing glucose concentration at the levels higher than 20 g L-1. Our previous results show that the optimum substrate concentration for biohydrogen production from the fermentation of
RI PT
SCB hydrolysate by C. butyricum was 20 g-COD L-1 [1]. In the present study, hydrogen production potential reached the peak value of 6,980 mL L-1 at 22.77 g-total sugar L-1 initial substrate concentration.
SC
Figure 1c and 1d illustrate the interactive effect of substrate concentration and the inoculums:substrate ratio on hydrogen production potential with its optimums shown at the peak
M AN U
of the response curve (Fig. 1c). Hydrogen production potential increases with increasing substrate concentration and inoculums:substrate ratio at low values but it decreases with further increasing in substrate concentration and inoculums:substrate ratio at high levels. The highest hydrogen production potential (6,980 mL L-1) was obtained when the inoculums:substrate ratio
TE D
was 0.31. The two-dimensional contour of hydrogen production potential with respect to substrate concentration and inoculums:substrate ratio (Fig. 1d), showes an ellipse shaped contour plot (Fig. 1d). However, the interactive effect of the substrate concentration and
EP
inoculums:substrate ratio on hydrogen production potential was not significant (P > 0.05). This phenomenon was attributed to the fact that the ratio of inoculums to substrate can increase the
AC C
hydrogen production rate. However, at a higher inoculum volume, more of the carbon source was devoted for biomass production than hydrogen production [33]. The graphical representation provides a method to visualize the relationship between the
response and experimental values of inoculums:substrate ratio and substrate:buffer ratio on hydrogen production potential for the purpose of finding the optimum conditions (Figs. 1e and 1f with the optimum shown on the top of the surfaces and the smallest ellipse in the contour line.
14
ACCEPTED MANUSCRIPT
The highest hydrogen production potential was obtained at a ratio of substrate to buffer of 4.31. The interaction between substrate:buffer ratio and inoculums:substrate ratio (X2X3) was not significant (P > 0.05). However, the ratio of substrate to buffer was significant (P < 0.05),
RI PT
indicating that the buffer plays an important effect on hydrogen production potential. The formation of hydrogen is accompanied by the production of VFAs or other solvents during the anaerobic digestion process, which causes the pH to drop. Therefore, alkalinity in the form of
SC
carbonate or bicarbonate as a supplement is necessary in the reactor. pH is important as one of the factors controlling the anaerobic biological processes due to its effects on Fe-hydrogenase
M AN U
activity, metabolic pathways, and the duration of the lag phase [34]. A bacterial medium always contains buffering compounds to reduce the pH fluctuations during the initial growth periods. In an anaerobic reactor, the pH value and its stability are important. The potential of a substrate to generate acids can be predicted by its composition. Carbohydrate-rich substrates have a greater
TE D
potential to acidify the media during anaerobic fermentation. As a result, in some situations, system stability is hard to maintain. In hydrogen fermentation, reactors tend to acidify readily and a reduction in the pH may result [35]. Low initial pH values below 5.0 inhibit hydrogen
EP
production [36]. On the other hand, high initial pH values such as 9.0 decrease lag time, but have a lower level of hydrogen production [37]. In these cases, it becomes necessary to either add
AC C
external sources of alkalinity or to adopt adequate operational strategies [38,39]. In biohydrogen production systems, bicarbonate alkalinity plays an important role, as it allows buffering of pH when VFA production becomes excessive. However, biocarbonate interacts with CO2, another major gaseous end product in anaerobic systems. An increase in the biocarbonate concentration in the feed increases the CO2 fraction in the gas phase because of carbonate dissolution [39]. 3.4.
Confirmation experiments and adequacy of the models
15
ACCEPTED MANUSCRIPT
To confirm the validity of the statistical experimental strategies and to gain a better understanding of hydrogen production from SCB hydrolysate by mixed culture, four additional confirmation experiments were conducted. The chosen conditions for substrate concentration,
RI PT
substrate:buffer ratio and inoculums:substrate ratio for hydrogen production potential are listed in Table 4. The data in Table 4 reveas that the optimum hydrogen production potential measured
technique to optimize the biohydrogen production.
4. Conclusions
SC
was close to those estimated using RSM. This confirms that the RSM analysis is a useful
M AN U
The present work focused on optimizing factors affecting hydrogen production from SCB hydrolysate by mixed cultures. The CCD/RSM approach was employed for research planning. The optimal condition for hydrogen production in the culture was at pH 6 and 37 °C. Experimental results showed that buffer:substrate ratio had a significant influence on the
TE D
hydrogen production potential. A maximum hydrogen production potential of 6,980 mL H2 L-1 was estimated at the optimum conditions of substrate concentration 22.77 g-total sugar L-1, 4.31 substrate:buffer ratio, and 0.31 inoculum:substrate ratio. Four additional experiments confirm
EP
that the optimum hydrogen production potential measured was close to the estimated value using the RSM. The experimental results demonstrated that RSM with the CCD was useful for
AC C
optimizing hydrogen production from SCB hydrogen by mixed culture.
Acknowledgements
This work was supported by research funds from The Research Group for Development
of Microbial Hydrogen Production Process from Biomass, the Higher Education Research promotion and National Research University Projects for Thailand office of the Higher Education Commission. We also thank Prof. Chiu-Yue Lin from BioHydrogen Laboratory,
16
ACCEPTED MANUSCRIPT
Department of Environmental Engineering and Science, Feng Chia University for his assistance in the experimental work.
Refferences Pattra S, Sangyoka S, Boonmee M, Reungsang A. Bio-hydrogen production from the
RI PT
[1]
fermentation of sugarcane bagasse hydrolysate by Clostridium butyricum. Int J Hydrogen Energ 2008;33:5256–65.
Sreela-or C, Plangklang P, Imai T, Reungsang A. Co-digestion of food waste and sludge
SC
[2]
Hydrogen Energ 2011;36:14227–37. [3]
M AN U
for hydrogen production by anaerobic mixed cultures: statistical key factors optimization. Int J
Zhang XL, Van S, Tyagi RD, Surampalli RY. Biodiesel production from heterotrophic
microalgae through transesterification and nanotechnology application in the production. Renew Sust Energ Rev 2013;26:216–23.
Huang YT, Su CP. High lipid content and productivity of microalgae cultivating under
TE D
[4]
elevated carbon dioxide. Int J Environ Sci Te 2014;11:703–10. [5]
Komolafe O, Orta SBV, Monje-Ramirez I, Noguez IY, Harvey AP, Ledesma MTO.
EP
Biodiesel production from indigenous microalgae grown in wastewater. Bioresource Technol 2014;154:297–304.
[7]
Berman B. Fuel cells at center stage. The New York Times 2013 Nov 22.
AC C
[6]
Lin CY, Hung CH, Chen CH, Chung WT, Cheng LH. Effects of initial cultivation pH on
fermentative hydrogen production from xylose using natural mixed cultures. Process Biochem 2006;41:1383–90. [8]
Guo XM, Trably E, Latrille E, Carrere H, Steyer JP. Hydrogen production from
agricultural waste by dark fermentation: a review. Int J Hydrogen Energ 2010;35:10660–73.
17
ACCEPTED MANUSCRIPT
[9]
Chong ML, Sabaratnam V, Shirai Y, Hassan MA. Biohydrogen production from biomass
and industrial wastes by dark fermentation. Int J Hydrogen Energ 2009;34:3277–87. [10]
Aguilar MAR, Fdez-Güelfo LA, Álvarez-Gallego CJ, García LIR. Effect of HRT on
RI PT
hydrogen production and organic matter solubilization in acidogenic anaerobic digestion of OFMSW. Chem Eng J 2013;219:443–9. [11]
Singh L, Wahid ZA, Siddiqui MF, Ahmad A, Ab Rahim MH, Sakinah M. Biohydrogen
polyethylene glycol. Process Biochem 2013;48:294–8.
OCSB. Report on Cane and Sugar Production. Bangkok, Thailand: Office of the Cane
M AN U
[12]
and Sugar Board; 2011. [13]
SC
production from palm oil mill effluent using immobilized Clostridium butyricum EB6 in
Neureiter M, Danner H, Thomasser C, Saidi B, Braun R. Dilute-acid hydrolysis of
sugarcane bagasse at varying conditions. Appl Biochem Biotechnol 2002;98:49–58. Gámez S, González-Cabriales JJ, Ramírez JA, Garrote G, Vázquez M. Study of the
TE D
[14]
hydrolysis of sugar cane bagasse using phosphoric acid. J Food Eng 2006;74:78–88. [15]
Aguilar R, Ramirez JA, Garrote G, Vázquez M. Kinetic study of the acid hydrolysis of
[16]
EP
sugar cane bagasse. J Food Eng 2002;55:309–18. Singh SP, Asthana RK, Singh AP. Prospects of sugarcane milling waste utilization for
[17]
AC C
hydrogen production in India. Energ Policy 2007;35:4164–8. Lin CY, Lay CH. A nutrient formulation for fermentative hydrogen production using
anaerobic sewage sludge microflora. Int J Hydrogen Energ 2005;30:285–92. [18]
Teli A, Ficara E, Malpei E. Bio-hydrogen production from cheese whey by dark
fermentation. Chem Eng Trans 2014;37:613–8. [19]
Wang JP, Chen YZ, Ge XW, Yu HQ. Optimization of coagulation-flocculation process
18
ACCEPTED MANUSCRIPT
for a paper-recycling wastewater treatment using response surface methodology. Colloid Surface A 2007;302:204–10. [20]
Pan CM, Fan YT, Xing Y, Hou HW, Zhang ML. Statistical optimization of process
RI PT
parameters on biohydrogen production from glucose by Clostridium sp. Fanp2. Bioresource Technol 2008;99:3146–54. [21]
Saha SK, Brewer CF. Determination of the concentrations of oligosaccharides, complex
SC
type carbohydrates, and glycoproteins using the phenol-sulfuric acid method. Carbohydr Res 1994;254:157–67.
Box GEP, Hunter WG, Hunter JS. Statistics for Experimenters. New York: Wiley; 1978.
[23]
Zheng XJ, Yu HQ. Inhibitory effects of butyrate on biological hydrogen production with
M AN U
[22]
mixed anaerobic cultures. J Environ Manage 2005;74:65–70. [24]
Lin CY, Chang RC. Hydrogen production during the anaerobic acidogenic conversion of
[25]
TE D
glucose. J Chem Technol Biot 1999;74:498–500.
Chen WM, Tseng ZJ, Lee KS, Chang JS. Fermentative hydrogen production with
Clostridium butyricum CGS5 isolated from anaerobic sewage sludge. Int J Hydrogen Energ
[26]
EP
2005;30:1063–70.
APHA. Standard Methods for the Examination of Water and Wastewater. 19th, ed.
[27]
AC C
Washington, DC: American Public Health Association; 1995. Kim SH, Han SK, Shin HS. Feasibility of biohydrogen production by anaerobic co-
digestion of food waste and sewage sludge. Int J Hydrogen Energ 2004;29:1607–16. [28]
Niladevi KN, Sukumaran RK, Jacob N, Anisha GS, Prema P. Optimization of laccase
production from a novel strain – Streptomyces psammoticus using response surface methodology. Microbiol Res 2009;164:105–13.
19
ACCEPTED MANUSCRIPT
[29]
Fabiano B, Perego P. Thermodynamic study and optimization of hydrogen production by
Enterobacter aerogenes. Int J Hydrogen Energ 2002;27:149–56. [30]
Jo JH, Lee DS, Park D, Choe WS, Park JM. Optimization of key process variables for
RI PT
enhanced hydrogen production by Enterobacter aerogenes using statistical methods. Bioresource Technol 2008;99:2061–6. [31]
Fan YT, Li CL, Lay JJ, Hou HW, Zhang GS. Optimization of initial substrate and pH
SC
levels for germination of sporing hydrogen-producing anaerobes in cow dung compost. Bioresource Technol 2004;91:189–93.
Oh YK, Seol EH, Kim JR, Park S. Fermentative biohydrogen production by a new
M AN U
[32]
chemoheterotrophic bacterium Citrobacter sp. Y19. Int J Hydrogen Energ 2003;28:1353–9. [33]
Kotay SM, Das D. Microbial hydrogen production with Bacillus coagulans IIT-BT S1
isolated from anaerobic sewage sludge. Bioresource Technol 2007;98:1183–90. Lay JJ. Modeling and optimization of anaerobic digested sludge converting starch to
TE D
[34]
hydrogen. Biotechnol Bioeng 2000;68:269–78. [35]
Vijayaraghavan K, Ahmad D, Ibrahim MKB, Herman HNB. Isolation of hydrogen
2006;31:708–20.
Chen CC, Lin CY, Lin MC. Acid-base enrichment enhances anaerobic hydrogen
AC C
[36]
EP
generating microflora from cow dung for seeding anaerobic digester. Int J Hydrogen Energ
production process. Appl Microbiol Biot 2002;58:224–8. [37]
Khanal SK, Chen WH, Li L, Sung SW. Biological hydrogen production: effects of pH
and intermediate products. Int J Hydrogen Energ 2004;29:1123–31. [38]
Ergüder TH, Güven E, Demirer GN. Anaerobic treatment of olive mill wastes in batch
reactors. Process Biochem 2000;36:243–8.
20
ACCEPTED MANUSCRIPT
[39]
Lin CY, Lay CH. Effects of carbonate and phosphate concentrations on hydrogen
AC C
EP
TE D
M AN U
SC
RI PT
production using anaerobic sewage sludge microflora. Int J Hydrogen Energ 2004;29:275–81.
21
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 1. Full factorial central composite design matrix defining substrate concentration, substrate: buffer ratio and inoculums:substrate ratio Substrate:Buffer ratio Inoculum:Substrate ratio Substrate concentration Hydrogen (X1) (X2) (X3) production CO2/H2 (g-total sugar L-1) (g L-1 of total sugar/g L-1 of NaHCO3) (g L-1 of VSS/g L-1 of total sugar) potential (P) Trial ratio Code Actual Actual value Code Actual value Code value (mL) value value Ratio (value) value Ratio (value) 1 -1.000 10.00 -1.000 2.00 (10/5) -1.000 0.20 (2/10) 4827 1.41 ± 0.22 2 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 6852 1.28 ± 0.02 3 1.000 30.00 -1.000 2.00 (30/15) 1.000 0.40 (12/30) 4938 1.24 ± 0.04 4 1.000 30.00 -1.000 2.00 (30/15) -1.000 0.20 (6/30) 4605 1.27 ± 0.04 5 0.000 20.00 -1.682 0.64 (20/31.25) 0.000 0.30 (6/20) 4794 1.29 ± 0.07 6 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 7019 1.17 ± 0.05 7 1.000 30.00 1.000 6.00 (30/5) -1.000 0.20 (6/30) 5229 1.13 ± 0.08 8 1.682 36.82 0.000 4.00 (36.82/9.21) 0.000 0.30 (11.05/36.82) 4721 1.09 ± 0.02 9 0.000 20.00 0.000 4.00 (20/5) 1.682 0.47 (9.4/20) 5836 1.17 ± 0.09 0 0.000 20.00 0.000 4.00 (20/5) -1.682 0.13 (2.6/20) 5791 1.45 ± 0.08 11 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 6934 1.26 ± 0.03 12 -1.000 10.00 1.000 6.00 (10/1.67) 1.000 0.40 (4/10) 5058 1.18 ± 0.01 13 0.000 20.00 1.682 7.35 (20/2.72) 0.000 0.30 (6/20) 5527 1.06 ± 0.06 14 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 6957 2.13 ± 0.13 15 -1.000 10.00 1.000 6.00 (10/1.67) -1.000 0.20 (2/10) 4839 1.08 ± 0.15 16 -1.682 3.18 0.000 4.00 (3.18/0.80) 0.000 0.30 (0.95/3.18) 4229 1.22 ± 0.13 17 1.000 30.00 1.000 6.00 (30/5) 1.000 0.40 (12/30) 5552 1.10 ± 0.90 18 -1.000 10.00 -1.000 2.00 (10/5) 1.000 0.40 (4/10) 4934 1.04 ± 0.03 19 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 7008 0.96 ± 0.02 20 0.000 20.00 0.000 4.00 (20/5) 0.000 0.30 (6/20) 6991 1.05 ± 0.05
22
ACCEPTED MANUSCRIPT
Butyric (HBu) (%) 55.4 57.5 52.8 49.9 55.2 59.1 58.7 50.0 60.0 65.1 57.9 58.1 60.6 47.7 57.9 59.8 55.1 49.3 50.2 54.4
pH range 4.05-4.61 4.35-4.36 4.33-4.35 4.56-4.56 5.17-5.19 4.35-4.37 4.06-4.36 4.09-4.29 4.17-4.48 5.19-5.34 4.28-4.42 4.51-4.61 4.40-4.62 6.11-6.19 4.18-4.32 4.37-4.85 4.16-4.32 4.14-4.16 4.08-4.10 4.19-4.31
M AN U
SC
1.53 1.64 1.23 1.16 1.50 1.78 1.58 1.19 1.84 2.52 1.66 1.73 1.71 1.09 2.04 1.49 1.50 1.16 1.01 1.43
RI PT
HBu/HAc
AC C
EP
TE D
Table 2. pH range and VFA distribution in the final product Acetic Propionic Trial VFA (mg L-1) (HAc) (HPr) (%) (%) 1 3924 36.2 8.5 2 4195 35.1 7.4 3 4294 42.9 4.4 4 5582 43.2 6.9 5 5071 36.8 8.0 6 4034 33.2 7.8 7 3517 37.1 4.2 8 4398 42.2 7.9 9 4191 32.7 7.4 10 4209 25.8 9.1 11 4145 35.0 7.2 12 2750 33.6 8.3 13 2478 35.4 4.0 14 2728 43.7 8.6 15 3669 28.5 13.6 16 3041 40.2 0.0 17 3875 36.7 8.2 18 4257 42.6 8.1 19 2464 49.8 0.0 20 3633 37.9 7.8
23
ACCEPTED MANUSCRIPT Table 3. Analysis of variance for quadratic polynomial model Sum of squares
DF
Mean square
F-value
Model X1 X2 X3 X1X2 X1X3 X2X3 X12 X22 X32 R2
1.789E + 007 1.633E + 005 4.976E + 005 81914.06 1.518E + 005 13612.50 1300.50 1.135E + 007 5.998E + 006 2.474E + 006 0.9952
9 1 1 1 1 1 1 1 1 1
1.988E + 006 1.633E + 005 4.976E + 005 81914.06 1.518E + 005 13612.50 1300.50 1.135E + 007 5.998E + 006 2.474E + 006
229.91 18.89 57.54 9.47 17.55 1.57 0.15 1312.66 693.63 286.03
SC M AN U TE D EP AC C 24
P-value Prob > F < 0.0001 0.0015 < 0.0001 0.0117 0.0019 0.2381 0.7063 < 0.0001 < 0.0001 < 0.0001
RI PT
Source
ACCEPTED MANUSCRIPT
RI PT
Hydrogen production potential (mL H2 L-1 substrate) Measure 6054 ± 40 5770 ± 20 6243 ± 58 6928 ± 68
EP
TE D
Bias was calculated using the equation: [(calculated value – experimental value)/calculated value] x 100
AC C
a
M AN U
SC
Table 4. Measured and calculated values of the confirmation experiments Inoculums: Substrate:buffer ratio Substrate substrate ration (g L-1 of total sugar/g L-1 Trial (g-total sugar L-1) (g L-1 of VSS/g L-1 of total of NaHCO3) sugar) Ratio (value) Ratio (value) 21 20 2 (20/10) 0.30 (6/20) 22 20 4 (20/5) 0.20 (4/20) 23 20 6 (20/3.33) 0.30 (6/20) 24 22.77 4.31 (22.71/5.30) 0.31 (7.04/22.71)
25
Calculated 6094 6972 6601 6980
Bias%a
0.70 17.20 5.40 0.75
ACCEPTED MANUSCRIPT 6.00
(b) 6520 5940 5360 4780 4200
6.00
X2: Substrate:Buffer ratio (g L-1 of total sugar/g L-1 of NaHCO3)
3.00
15.00 10.00
6094.75 5799.69
X1: Substrate concentration (g-total sugar L-1)
(c)
5799.69
6389.81
5504.63
15.00
20.00
25.00
30.00
X1: Substrate concentration (g-total sugar L-1)
(d) X3: Inoculum/Substrate ratio (g L-1 of VSS/g L-1 of total sugar)
SC
0.40
7100 6520 5940 5360 4780 4200
0.40
0.35
0.30
0.25
6
6241.37
M AN U
6241.37
5998.97
6726.18 5998.97
30.00
0.35
5756.57
25.00
0.30 X3: Inoculum:Substrate ratio (g L-1 of VSS/g L-1 of total sugar)
20.00 0.25
15.00 0.20
7100 6520 5940 5360
EP
4780
AC C 0.30
3.00 0.20
2.00
(f)
20.00
25.00
30.00
X1: Substrate concentration (g-total sugar L-1)
0.40 6090.81
0.35
0.30
6 6312.97
0.25
6757.28
6090.81
6312.97
5.00
6535.12
5868.66
4.00
0.25
15.00
6.00
0.35
X2: Inoculum:Substrate ratio (g L-1 off VSS/g L-1 of total sugar)
6483.78
X1: Substrate concentration (g-total sugar L-1)
TE D
(e)
10.00
0.20 10.00
X3: Inoculum:Substrate ratio (g L-1 of VSS/g L-1 of total sugar)
Hydrogen production potential (mL)
6684.86
3.00
2.00 10.00
20.00 2.00
Hydrogen production potential (P)
6094.75
5504.63
25.00 4.00
0.40
6
4.00
30.00 5.00
4200
5.00
RI PT
7100
5799.69
X2: Substrate:Buffer ratio (g L-1 of total sugar/g L-1 of NaHCO3)
Hydrogen production potential (mL)
(a)
0.20
X3: Substrate:Buffer ratio (g-total sugar L-1)
2.00
3.00
4.00
5.00
6.00
X2: Substrate:Buffer ratio (g L-1 of total sugar/g L-1 of NaHCO3)
Fig. 1. Response surface and contour plots for the effect on substrate concentration, substrate:buffer ratio and inoculums:substrate ratio on hydrogen production potential. (a) Response surface and (b) contour plots for the effect on substrate concentration and substrate:buffer ratio on hydrogen production potential. (c) Response surface and (d) contour plots for the effect on substrate concentration and inoculums:substrate ratio on hydrogen production potential. (e) Response surface and (f) contour plots for the effect on substrate:buffer and inoculums:substrate ratio on hydrogen production potential.
26