- Email: [email protected]
Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach
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 x x x ( 2 0 1 8 ) 1 e7
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach Ebru C¸okay* Department of Environmental Engineering, Dokuz Eylul University, Buca, Izmir, Turkey
article info
abstract
Article history:
Hydrogen gas production was investigated by electrohydrolysis of food waste due to its
Received 11 August 2017
high organic content. Different voltages generated by DC power supply were applied to
Received in revised form
food waste in order to produce hydrogen gas. Effects of the DC voltage, reaction time and
12 January 2018
initial solid content on cumulative hydrogen gas production, hydrogen gas content in the
Accepted 15 January 2018
gas phase and total organic carbon (TOC) removal were investigated by using a Box-
Available online xxx
Behnken statistical experiment design approach. The most suitable voltage/reaction time/solid content values resulting in the highest hydrogen gas content (99%), the highest
Keywords:
cumulative hydrogen gas formation (7000 mL) and total organic carbon removal (33%) were
Hydrogen gas production
determined as 5 V/75 h/20%. The results indicated that food wastes constitute a good
TOC removal
source for H2 gas production by electrohydrolysis. Hydrogen gas produced by electro-
Electrohydrolysis
hydrolysis of food waste can be directly used in fuel cells due to its high putrity.
Food wastes
© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Box-Behnken statistical approach
Introduction In recent years, clean and high energy fuels have been searched and investigation on clean energy has gained significant attention because of energy limitations and green house gas emissions caused by the use of fossil fuels. Due to its high energy content (122 kJ g1), hydrogen gas is an effective and clean energy source with no green house gas emissions. When H2 gas is used as a fuel only water vapor is produced without any other emissions. Besides H2 gas is an important electron carrier which can be used in fuel cells for production of electrical energy [1]. Easy transportation in form of metal hydrides is another advantage of H2 to be used in motor vehicles for electrical power generation. Hydrogen gas is not readily available in nature and is produced by energy intensive, costly methods such as steam
reforming of hydrocarbons or electrolysis of water [2]. The photocatalytic hydrogen production from glucose degradation under visible light were also investigated [3]. As an alternative for these processes, hydrogen gas production by fermentation (dark and photo fermentation) has been considered as a viable approach. Renewable organic wastes or materials containing carbohydrates can be used for H2 gas production by fermentation. However, some obstacles in bio-hydrogen production were observed such as low hydrogen gas content and formation rates because of slow bacterial metabolisms in fermentation process. In order to overcome negative effect of photo/ dark fermentation processes, different hydrogen gas production methods were investigated. Electrohydrolysis of organic wastes may be used for effective hydrogen gas production [4e7]. Based on literature reports electrohydrolysis may be used as a pretreatment at the hydrolysis stage in anaerobic digestion of organic wastes such as lignocellulosics to
* Corresponding author. E-mail address: [email protected]. https://doi.org/10.1016/j.ijhydene.2018.01.079 0360-3199/© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
2
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 x x x ( 2 0 1 8 ) 1 e7
improve the rate and extent of hydrolysis [8,9]. However, in this study, electrohydrolysis was directly applied to organic waste to produce hydrogen gas but not for the pretreatment. Electrohydrolysis is the chemical process to decompose the organic matter by breaking the bonds between polymers provoked by application of current through electrodes. When electrodes connected to direct current, one electrode becomes positively charged and another electrode becomes negatively charged. This starts the movement of electrolyte towards electrodes i.e., positive ions moves to cathode and negative ions to anode. Several studies on hydrogen gas production by application of direct current to organic wastes and effluents of fermentation were reported [10e12]. Microbial electrolysis cells were also used for H2 gas production from organic wastes [13e16]. In one study, the effects of nutrient ratios on hydrogen gas production by batch dark fermentation of waste peach pulp was investigated using statistical experiment design. The effluent of dark fermentation experiments were used for H2 gas production by electrohydrolysis [17]. In our previous studies, raw organic wastes such as leachate, olive mill wastewater, anaerobic sludge were used to produce hydrogen gas by electrohydrolysis [4,5,18,19]. In some studies, PVC panels were used as electrical energy source for H2 gas production by electrohydrolysis from organic wastes [6]. Different electrodes were tested in electrohydrolysis studies and aluminum electrodes were found as the most effective one due to its high electrical conductivity [7]. Based on literature reports, hydrogen gas production from food wastes by electrohydrolysis was not investigated. Electrohydrolysis of food waste with high organic content to produce hydrogen gas is a novel and promising method. A Box-Behnken statistical experiment design approach was used in this study to investigate the effects of important operating variables (applied voltage, solids content of the waste and operation time) on the rate and extent of hydrogen gas production and also on total organic carbon (TOC) removal from the food waste.
Design of experiments The effects of variables on the response are a complicated and observed by means of the mostly known approach which is altering one factor at a time for multivariable systems. However this technic is not usable for estimation of responses. So, some experimental statistical design should be used for optimization of the reaction conditions. For this aim, several significant parameters are determined by response surface methodology. This method as called as RSM. This design contains 3-level factorial design, central composite design (CCD) [20,21], Box-Behnken design [22] and D-optimal design [23]. Between all the RSM designs, Box-Behnken design needs fewer runs than the other RSM designs. This design allows and shows to efficiency at intermediate levels not experimentally studied [24,25]. The mathematical relationship which is offered by BoxBehnken design application between the dependent variables (Y) and the independent variables (X) can be approximated by a (second-order) polynomial equation as follows:
Y ¼ b0 þ b1 X1 þ b2 X2 þ b3 X3 þ b12 X1 X2 þ b13 X1 X3 þ b23 X2 X3 þ b11 X21 þ b22 X22 þ b33 X23
(2)
This approach was selected to predict a potential response function. A total of 15 experiments are demanded to determine 9 coefficients of the quadratic equation. This regression model includes one block term, three linear terms, three quadratic terms, and three interaction terms. The main aim of this study is to observe and interpret the effects of the applied voltage, reaction time and initial solid content on percent hydrogen gas production, cumulative hydrogen gas production and percent total organic carbon removals from food wastes by electrohydrolysis method which is arranged by Box-Behnken design approach.
Materials and methods A power supplier and two aluminum electrodes were used as shown in Fig. 1. Food waste obtained from the university kitchen was filled into bottles. Aluminum electrodes were immersed inside the bottles and connection of power supply to the system was done by wires. The bottles were closed very tightly using silicone rubber stoppers and screw caps. In addition silicon was used to seal the bottles and to eliminate any gas leakage from the bottles. Aluminum electrodes were pure aluminum with diameters of 4 mm, lengths of 20 cm and weight of 30 g each. Distance between the electrodes was 2 cm. Experiments were performed at room temperature using 300 mL food waste. Water content of food waste was 75%. Food waste was homogenized by using a mixer. Water added in food wastes to adjust the solid content. Solid content value in food waste was arranged by different kind of foods. The raw food waste contained 43 mg/g TOC, 23.5 g/L total solids with a pH of 4.67 and ORP of 66 mV. The voltage which is applied to system was nearly kept stable throughout experiment. However, during experiments the current intensities (mA) altered due to variations in the reaction mechanism and changes in the waste composition. These changes caused the corrosion on the electrodes. So, current intensities were continuously reported by the power supply unit and were also recorded by using a multimeter
Fig. 1 e The experimental set-up.
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
3
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 x x x ( 2 0 1 8 ) 1 e7
Table 1 e Independent variable levels in statistical design. Variable Reaction Time (h) Voltage, (V) Solid Content, (%)
Symbol
Low (1)
Center (0)
High (þ1)
X1 X2 X3
25 1 10
50 3 25
75 5 40
which is connected to the system in series. Hydrogen gas produced from waste was stored in the head space of the bottles. Hydrogen gas taken from the bottles was analyzed by a gas chromatograph (GC, HP Agilent 6890) to determine the percentage of hydrogen gas. Nitrogen gas was used as carrier with a flow rate of 30 mL min1 and the head pressure was 22 psi. Temperatures of the injection, oven, detector, and filament were 35 C, 120 C, 120 C, 140 C, respectively. Food wastes were taken from the bottles everyday and total organic carbon (TOC) analyses was done using A Shimadzu TOC Analyzer (Shimadzu-SSM 550 model). The volume of hydrogen gas in the head space of bottles was measured by water displacement method using acid solution of acid and 10% NaCl. Cumulative hydrogen gas was calculated according to the following equation: VH2;i ¼ VH2;i1 þ Vw CH2;i þ VG;i þ VG;i CH2;i VG;i1 CH2;i1
mentioned as effective and useful way for optimization of the three variable response functions, predicting the response of the fitted model and checking the sufficiency of the design model by the ANOVA tests. The reaction time (X1), direct current voltages (X2), solid content (X3) as the independent and percent hydrogen gas production (Y1), cumulative hydrogen production (Y2) and percent TOC removals (Y3) as the dependent variables were selected in statistical design for this purpose. The low, center and high levels of each variable organized by statical approach as 1, 0, and þ1 respectively are shown in Table 1. While reaction time (X1) changed between 25 and 75 h, voltages (X2) altered between 1 and 5 V. All these selections were executed for observing an effective hydrogen gas production. The initial solid content of waste (X3) was ranged from 10% to 40%. The effects of independent variables (reaction time, voltage and solid content) on dependent variables or objective functions (hydrogen gas percentage, cumulative hydrogen gas production and total organic carbon removal) were evaluated. The experimental conditions which are performed according to Box-Behnken design for hydrogen gas production from food waste are given in Table 2. Obtained results at the pre-determined runs planned by Box-Behnken are also presented in Table 2. After experiments, predicted results are determined by Box-Behnken statistical approach at the predetermined experimental runs. In order to compare observed and predicted results, all experimental and predicted results are presented in Table 3. The effects of each parameter on hydrogen gas production and total organic matter removal by means of regression
(1)
where; VH2,ie1 and VH2,ie are cumulative H2 volumes (mL) calculated for (ie1)th and (ith) condition, VW is the total amount of gas (mL) measured in water displacement, CH2;i and CH2;i are measured H2 percentage at (ith) and (ie1)th condition, VG,i and VG,ie1 are the headspace volumes at (ith) and (ie1) the condition.
Results and discussion The effects of the independent variables on dependent variables in another saying the response functions were evaluated by Box-Behnken design. This design has main effects, interaction effects, and quadratic effects [22]. This approach was
Table 2 e Obtained results at the pre-determined experimental runs arranged by Box-Behnken statistical approach. Run No
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Variable levels
Experimental results
Reaction Time (h)
Voltage (V)
Solid Content (%)
HGP (%)
CHGV (mL)
TOC Removal (%)
50 75 75 50 75 75 50 25 25 50 25 25 50 50 50
1 5 1 5 3 3 5 3 1 1 5 3 3 3 3
40 25 25 40 10 40 10 10 25 10 25 40 25 25 25
26 100 32.8 94.1 99 98.5 99 75.8 14 16.4 79.4 55.7 90.8 90 92
63.2 6411 218 1302 4250 1792 4255 985 34 76 1032 446 2204 1873 2203
3.83 32.8 8.63 7.89 22.25 11.2 16.45 7.8 0.9 1.8 2.32 5.58 7.72 7.72 7.1
HGP: Hydrogen Gas Percentage, CHGV: Cumulative Hydrogen Gas Volume; TOC: Total Organic Carbon.
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
4
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 x x x ( 2 0 1 8 ) 1 e7
Table 3 e Observed and predicted results at the pre-determined experimental runs. Y1, HGP (%)
Run No
Y2, CHGV (mL)
Y3, TOC Removal (%)
Obs.results
Pred. results
Obs.results
Pred. results
Obs.results
Pred. results
26 100 32.8 94.1 99 98.5 99 75.8 14 16.4 79.4 55.7 90.8 90 92
25.41 100 34.05 88.74 93.39 99.34 100 76.46 7.98 21.76 78.15 62.31 90.93 90.93 90.93
63.2 6411 218 1302 4250 1792 4255 985 34 76 1032 446 2204 1873 2203
162.35 6070.35 320.65 1519.75 4365.10 1914.90 4480.55 862.10 374.65 141.75 929.35 330.90 2093.33 2093.33 2093.33
3.83 32.8 8.63 7.89 22.25 11.2 16.45 7.8 0.9 1.8 2.32 5.58 7.72 7.72 7.1
2.13 29.67 7.22 7.90 23.68 14.31 18.16 4.69 4.03 1.78 3.73 4.15 7.51 7.51 7.51
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
HGP: Hydrogen Gas Percentage, CHGV: Cumulative Hydrogen Gas Volume; TOC: Total Organic Removal.
model can be shown in Eqs. (3)e(5). It can be said that all parameters had an increasing effect on response function such as hydrogen gas percentage, cumulative hydrogen gas formation and TOC Removal. Voltage was the most effective on hydrogen gas production because of its highest coefficient followed by other independent variables such as reaction time and solid content. Analysis of variance (ANOVA) results are presented in Tables 4e6. ANOVA test indicates that the predictability of the model is at 95% confidence interval. The experimental data are compatible with response functions. Coefficient of determination (R2) was larger than 0.99. In addition, the F value is much greater than 3.70, so this means that the treatment is significant. The p-values for the linear terms in the ANOVA, means that independent variables had significant effect (p < 0.05). Independent variables had significant effect on hydrogen gas production and organic matter removal because of smaller than p value limit. However, solid waste did not have significant effect on dependent variables. Coefficients which were determined for percent hydrogen gas production (Y1), cumulative hydrogen gas volume (Y2) and
percent total organic carbon removal (Y3) are presented in following equations: Y1 ¼174;64375þ2;36817X1 þ51;23125X2 þ2;24694X3 9*103 X1 X2 0:0134X1 X3 þ0:12917X2 X3 8:5066*103 X21 7:26667X22 0:012185X23 R2 ¼0;9852
(3)
Y2 ¼1535:506144:97433X1 þ1887:98750X2 þ153:81278X3 þ25:97500X1 X2 þ1:27933X1 X3 þ24:50167X2 X3 0:21969X21 76:72292X22 1:61063X23 R2 ¼0;9881
(4)
Y3 ¼þ52:957781:12044X1 9:12210X2 0:57016X3 þ0:11375X1 X2 þ5:88569*103 X1 X3 þ0:088332X2 X3 þ6:29172*103 X1 0:070319X22 þ1:17293*103 X23 R2 ¼0;9474
(5)
Table 4 e ANOVA test for hydrogen gas percentage (%) Y1.
Table 5 e ANOVA test for cumulative hydrogen gas volume (mL) Y2.
Source
Sum of squares
Df
Mean Square
F ratio
P value
Source
Model X1(Reaction Time) X2 (Voltage) X3 (Solid Content) X1X2 X1X3 X2X3 X21 X22 X23 Residual Lack of Fit Pure Error Total (corr)
14909.19 1455.30 10103.31 33.62 0.81 101.00 60.06 104.37 3119.52 27.75 224.28 222.26 2.03 15133.47
9 1 1 1 1 1 1 1 1 1 5 3 2 14
1656.58 1455.30 10103.31 33.62 0.81 101.00 60.06 104.37 3119.52 27.75 44.86 74.09 1.01
36.93 32.44 225.23 0.75 0.018 2.25 1.34 2.33 69.54 0.62
0.0005 0.0023 <0.0001 0.4262 0.8983 0.1938 0.2995 0.1877 0.0004 0.4671
73.11
0.0135
Model X1(Reaction Time) X2 (Voltage) X3 (Solid Content) X1X2 X1X3 X2X3 X21 X22 X23 Residual Lack of Fit Pure Error Total (corr)
Sum of squares
Df
Mean Square
F ratio
P value
4.799Eþ007 1.294Eþ007 1.987Eþ007 4.444Eþ006 6.747Eþ006 9.206Eþ005 2.161Eþ006 69613.21 3.478Eþ005 4.849Eþ005 5.793Eþ005 5.064Eþ005 72820.67 4.857Eþ007
9 1 1 1 1 1 1 1 1 1 5 3 2 14
5.332Eþ006 1.294Eþ007 1.987Eþ007 4.444Eþ006 6.747Eþ006 9.206Eþ005 2.161Eþ006 69613.21 3.478Eþ005 4.849Eþ005 1.159Eþ005 1.688Eþ005 36410.33
46.02 111.68 171.54 38.36 58.24 7.95 18.65 0.60 3.00 4.19
0.0003 0.0001 <0.0001 0.0016 0.0006 0.0372 0.0076 0.4733 0.1437 0.0961
4.64
0.1825
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
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 x x x ( 2 0 1 8 ) 1 e7
5
Table 6 e ANOVA test for total organic carbon removal (%) Y3. Source Model X1(Reaction Time) X2 (Voltage) X3 (Solid Content) X1X2 X1X3 X2X3 X21 X22 X23 Residual Lack of Fit Pure Error Total (corr)
Sum of squares
Df
Mean Square
F ratio
P value
954.27 424.62 245.27 49.04 129.40 19.49 28.09 57.09 0.29 0.26 52.96 52.71 0.25 1007.23
9 1 1 1 1 1 1 1 1 1 5 3 2 14
106.03 424.62 245.27 49.04 129.40 19.49 28.09 57.09 0.29 0.26 10.59 17.57 0.13
10.01 40.09 23.15 4.63 12.22 1.84 2.65 5.39 0.028 0.024
0.0103 0.0014 0.0048 0.0841 0.0174 0.2330 0.1644 0.0679 0.8746 0.8823
138.75
0.0072
Hydrogen gas evolution Percent hydrogen gas production for different voltages is depicted in Fig. 2. Applied voltage was changed between 1 V and 5 V. Percent hydrogen gas increased up to 4 V and then slightly decreased with increasing voltages over 4 V at the end of 75 h reaction time. Percent hydrogen gas also changed throught experiments and reached nearly pure H2 gas (99% H2) at 4 V and 5 V. The lowest voltage resulting in the highest H2 content (99%) was 4 V. Cumulative hydrogen gas production with applied voltage is presented in Fig. 3. The highest cumulative hydrogen gas (6410 mL) was achieved by 5 V. Moreover, the most suitable direct current voltage maximizing cumulative hydrogen gas volume (6410 mL), hydrogen gas production rate (2050 mL d1) and percent hydrogen gas (99%) in the gas phase was 5 V. Nearly pure (99%) hydrogen gas produced by electrohydrolysis may be directly used in fuel cells.
Fig. 3 e Variation of cumulative hydrogen gas volume at different voltages and reaction times (Solid Content ¼ 25%).
Control experiments which include water electrolysis and natural decomposition of food waste were also performed to evaluate hydrogen gas percentage and the cumulative hydrogen gas evolution. According to evaluation of control water experiments at different voltages, maximum cumulative hydrogen gas volume (50 mL) and percent hydrogen gas (5%) in the gas phase was obtained at maximum voltage (5 V) during 75 h reaction time. At these experimental conditions, food waste produced maximum cumulative hydrogen production as 6410 mL by means of electrohydrolysis of food waste. So, hydrogen gas production in water electrolysis was less than 1% of cumulative hydrogen gas production for maximum voltage in experiments. It can be said that cumulative hydrogen gas volume in experiments didn't affected by water electrolysis. In addition, natural decomposition of food waste was not yield hydrogen gas volume and hydrogen gas percentage within 3 days with 5 V. It can be also said that natural decomposition of food waste did not affect the cumulative hydrogen gas volume. Therefore, hydrogen gas evolution was mainly achieved by electrohydrolysis of food waste according to applied voltage. Protons (Hþ ions) arised from organic waste decomposition reacted with electrons obtained by Al ionization from anode to yield hydrogen gas as expressed as Eqs. (6)e(8). The amount of electrons achieved at low voltages (<2 V) was not enough to neutralize protons arised from food waste decomposition. In other words, hydrogen gas production was restricted by available electrons at low voltages producing low volumes of hydrogen gas in spite of the existence of protons. Al (III) ion released from the electrode ionization was also low concentration in medium at low voltages. Al /AlðIIIÞ þ 3e o
(6)
Food waste Organic compounds/nHþ
þ decomposed organics nHþ þ n e /n=2 H2 ðgÞ
Fig. 2 e Variation of percent hydrogen gas production with reaction time at different applied voltages (Solid content ¼ 25%).
(7) (8)
Decomposed organic compounds such as volatile fatty acids were generally yielded from decomposition of food waste. When organic compounds in food waste were degraded into other organic matters, protons were arising in the media. Electrons and Al (III) ions were also released to the
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
6
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 x x x ( 2 0 1 8 ) 1 e7
medium after application of voltage. The arised protons and electrons were balanced with 5 V voltages producing the highest cumulative hydrogen gas production. The protons released from decomposition of food waste and Al (III) ions released from anode ionization competed for the electrons provided by DC current. Part of the electrons was reacted with Al (III) ions and part of them produced hydrogen by complexing with protons (Hþ). Al (III) ions were come out the organic waste forming Al salts. Finally, anode lost weight due to occuration of Al (III) and electron in reactions. Al (III) was accumulated on cathode in form of pure Alo and cathode weight slightly increased. The weight loss of aluminum electrodes during the experiment was observed as 1 g at 5 V and 75 h reaction time. The weight loss of aluminum electrodes was observed as 0.232 g and 0.096 g at 3 V and 1 V respectively, during 75 h reaction time. The weight loss of aluminum is very small than the weight of aluminum. This observation (only 0.096e1 g aluminum corrosion into solution) is good for hydrogen gas production. Initial pH increased with time and achieved a higher level at the end of reaction due to (Hþ) ion removal from the media by electrohydrolysis. The final pHs were around 4.5. Initial conductivities also ascended with arising decomposed organic compounds concentration. Morever, conductivities also ascended a higher level at the end of reaction due to Al(III) release to the reaction medium throughout electrohydrolysis.
Total organic carbon (TOC) removal Total organic carbon removals were also evaluated at predetermined runs arranged by Box-Behnken statistical approach by electrohydrolysis of food waste. TOC concentrations decreased from initial 43 mg/g to nearly 25.8 mg/g after electrohydrolysis application on food waste. As can be shown in Fig. 4, organic matter in food waste was decomposed by the microorganisms and electrohydrolysis and the released organic compounds (mainly VFAs) and CO2 resulting in remarkable total organic carbon removal. In other words, the applied voltage and natural decomposition of food waste was caused significant TOC removal due to high organic matter. However, only 2% TOC removal was observed with anaerobic decomposition of food waste in control experiments.
Table 7 e Comparison of hydrogen gas production from food waste with other organic wastes. Waste
Voltage
HGP (%)
CHGV (mL/d)
HGPR (mL/ d)
2 3
94 95
2775 3020
687 614
[4] [5]
4 5
99 99
5000 7000
1277 2240
[19] This study
Sludge Olive Oil Mill Leachate Food waste
References
HGP: Hydrogen Gas Percentage; CHGV: Cumulative Hydrogen Gas Volume; HGPR: Hydrogen Gas Production Rate.
Microbial TOC removal is very small in experiments. The yield of hydrogen gas production (mL H2 g1 TOC) is a significant parameter pointing the influence of hydrogen gas production per unit TOC removal. According to results, the highest yield of hydrogen production was 493 mL H2 mg1 TOC at 5 V direct current voltage. Electrohydrolysis of food waste for production of hydrogen gas was not reported in literature. In our previous studies, different organic wastes and wastewaters were used to produce hydrogen gas using electrohydrolysis [4e7,19]. The results of this study are compared with our previous studies using other organic wastes for H2 gas production by electrohydrolysis in Table 7. Hydrogen gas content (99%), cumulative hydrogen gas volume (7000 mL), hydrogen gas production rate (2240 mL d1) and hydrogen yield (493 mL H2 g1 TOC) obtained in this study are higher than the other studies listed in Table 7 indicating high potential of food waste as raw material for H2 gas production by electrohydrolysis. The most suitable voltage/reaction time/solid content values resulting in the highest hydrogen gas content (99%), the highest cumulative hydrogen gas volume (7000 mL) and TOC removal (33%) were found to be 5 V/75 h/20%. Validation experiment at the most suitable conditions was performed and almost the same results were obtained as predicted by the response function.
Conclusions
Fig. 4 e Variation of total organic carbon removal with different reaction times at applied voltages (Solid Content ¼ 25%).
A Box-Behnken statistical approach was used to investigate H2 gas production by electrohydrolysis of food wastes and to determine the most suitable reaction time, voltage and solid content maximizing the hydrogen gas content, cumulative hydrogen gas volume and TOC removal. According to experimental results and ANOVA test, reaction time and applied voltage were determined to have significant effects on H2 gas production and TOC removal. The optimal voltage/reaction time/solid contents resulting in the highest hydrogen gas content (99%), the highest cumulative hydrogen gas volume (7000 mL) and TOC removal (33%) were 5 V/75 h/20%. Hydrogen gas production rate of 2240 mL d1 and hydrogen yield of 493 mL H2 g1 TOC were achieved under these experimental conditions. Electrohydrolysis was found to be an effective method of producing nearly pure (99%) hydrogen gas from food wastes.
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
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 x x x ( 2 0 1 8 ) 1 e7
references
[1] Winter CJ. Into the hydrogen energy economy-milestones. Int J Hydrogen Energy 2005;30:681e5. [2] Das D, Veziroglu TN. Hydrogen production by biological processes: survey of literature. Int J Hydrogen Energy 2001;26:13e28. [3] Iervolino G, Vaiano V, Sannino D, Rizzo L, Galluzzi A, Polichetti M, et al. Hydrogen production from glucose degradation in water and wastewater treated by Ru-LaFeO3/ Fe2O3 magnetic particles photocatalysis and heterogeneous photo-Fenton. Int J Hydrogen Energy 2018. in press. [4] Kargi F, Catalkaya Ebru C, Uzuncar S. Hydrogen gas production from waste anaerobic sludge by electrohydrolysis: effects of applied DC voltage. Int J Hydrogen Energy 2011;36:2049e56. [5] Kargi F, Catalkaya EC. Hydrogen gas production from olive mill wastewater by electrohydrolysis with simultaneous COD removal. Int J Hydrogen Energy 2011;36:3457e64. [6] Eker S, Kargi F. Hydrogen gas production from electrohydrolysis of industrial wastewater organics by using photo-voltaic cells (PVC). Int J Hydrogen Energy 2010;35:12761e6. [7] Kargi F. Comparison of different electrodes in hydrogen gas production from electrohydrolysis of wastewater organics using photo-voltaic cells (PVC). Int J Hydrogen Energy 2011;36:3450e6. [8] Elliott A, Mahmood T. Comparison of mechanical pretreatment methods for the enhancement of anaerobic digestion of pulp and paper waste activated sludge. Water Environ Res 2012;84(6):497e505. [9] Wood N, Tran H, Master E. Pretreatment of pulp mill secondary sludge for high-rate anaerobic conversion to biogas. Bioresour Technol 2009;100(23):5729e35. [10] Veluchamy C, Kalamdhad AS. Biochemical methane potential test for pulp and paper mill sludge with different food/microorganisms ratios and its kinetics. Int Biodeterior Biodegrad 2017;117:197e204. [11] Day DL, Steinberg MP. Electrochemical conversion of organic material. US Patent No. 4200505. April 1980. [12] Roychowdhury S. Process for production of hydrogen from anaerobically decomposed organic materials. US Patent No.6090266. July 2000.
7
[13] Paterek JR. Method and apparatus for hydrogen production from organic wastes and manure. US Patent No. 6887692. May 2005. [14] Liu H, Grot S, Logan BE. Electrochemically assisted microbial production of hydrogen from acetate. Environ Sci Technol 2005;39:4317e20. [15] Cheng S, Logan BE. Sustainable and efficient biohydrogen production via electro hydrogenesis. Proc Natl Acad Sci USA 2007;104:18871e3. [16] Ditzig J, Liu H, Logan BE. Production of hydrogen from domestic wastewater using a bioelectrochemically assisted microbial reactor (BEAMR). Int J Hydrogen Energy 2007;32:2296e304. [17] Argun H, Siaka D. Hydrogen gas production from waste peach pulp by dark fermentation and electrohydrolysis. Int J Hydrogen Energy 2016;41:1156e576. [18] Call D, Logan BE. Hydrogen production in a single chamber microbial electrolysis cell lacking a membrane. Environ Sci Technol 2008;42:3401e6. [19] Kargi F, Catalkaya EC. Electrohydrolysis of landfill leachate organics for hydrogen gas production and COD removal. Int J Hydrogen Energy 2011;36:8252e60. [20] Boza A, De la Cruz Y, Jordan G, Jauregui-Haza U, Aleman A, Caraballo I. Statistical optimization of a sustained-release matrix tablet of lobenzarit disodium. Drug Dev Ind Pharm 2000;26:1303e7. [21] Box GEP, Wilson KB. On the experimental attainment of optimum multifactorial conditions. Roy Stat Soc 1951;13:1e12. [22] Singh SK, Dodge J, Durrani MJ, Khan MA. Optimization and characterization of controlled release pellets coated with experimental latex: I. Anionic drug. Int J Pharm 1995;125:243e55. [23] Sanchez-Lafuente C, Furlanetto S, Fernandez-Arevalo M. Didanosine extended-release matrix tablets: optimization of formulation variables using statistical experimental design. Int J Pharm 2002;237:107e18. [24] Hamed E, Sakr A. Application of multiple response optimization technique to extended release formulations design. J Contr Release 2001;73:329e38. [25] Abbasi AF, Ahmad M, Wasim M. Optimization of concrete mix proportioning using reduced factorial experimental technique. ACI Mater J 1987:55e63 (JanuaryeFebruary).
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach Ebru C¸okay* Department of Environmental Engineering, Dokuz Eylul University, Buca, Izmir, Turkey
article info
abstract
Article history:
Hydrogen gas production was investigated by electrohydrolysis of food waste due to its
Received 11 August 2017
high organic content. Different voltages generated by DC power supply were applied to
Received in revised form
food waste in order to produce hydrogen gas. Effects of the DC voltage, reaction time and
12 January 2018
initial solid content on cumulative hydrogen gas production, hydrogen gas content in the
Accepted 15 January 2018
gas phase and total organic carbon (TOC) removal were investigated by using a Box-
Available online xxx
Behnken statistical experiment design approach. The most suitable voltage/reaction time/solid content values resulting in the highest hydrogen gas content (99%), the highest
Keywords:
cumulative hydrogen gas formation (7000 mL) and total organic carbon removal (33%) were
Hydrogen gas production
determined as 5 V/75 h/20%. The results indicated that food wastes constitute a good
TOC removal
source for H2 gas production by electrohydrolysis. Hydrogen gas produced by electro-
Electrohydrolysis
hydrolysis of food waste can be directly used in fuel cells due to its high putrity.
Food wastes
© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Box-Behnken statistical approach
Introduction In recent years, clean and high energy fuels have been searched and investigation on clean energy has gained significant attention because of energy limitations and green house gas emissions caused by the use of fossil fuels. Due to its high energy content (122 kJ g1), hydrogen gas is an effective and clean energy source with no green house gas emissions. When H2 gas is used as a fuel only water vapor is produced without any other emissions. Besides H2 gas is an important electron carrier which can be used in fuel cells for production of electrical energy [1]. Easy transportation in form of metal hydrides is another advantage of H2 to be used in motor vehicles for electrical power generation. Hydrogen gas is not readily available in nature and is produced by energy intensive, costly methods such as steam
reforming of hydrocarbons or electrolysis of water [2]. The photocatalytic hydrogen production from glucose degradation under visible light were also investigated [3]. As an alternative for these processes, hydrogen gas production by fermentation (dark and photo fermentation) has been considered as a viable approach. Renewable organic wastes or materials containing carbohydrates can be used for H2 gas production by fermentation. However, some obstacles in bio-hydrogen production were observed such as low hydrogen gas content and formation rates because of slow bacterial metabolisms in fermentation process. In order to overcome negative effect of photo/ dark fermentation processes, different hydrogen gas production methods were investigated. Electrohydrolysis of organic wastes may be used for effective hydrogen gas production [4e7]. Based on literature reports electrohydrolysis may be used as a pretreatment at the hydrolysis stage in anaerobic digestion of organic wastes such as lignocellulosics to
* Corresponding author. E-mail address: [email protected]. https://doi.org/10.1016/j.ijhydene.2018.01.079 0360-3199/© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
2
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 x x x ( 2 0 1 8 ) 1 e7
improve the rate and extent of hydrolysis [8,9]. However, in this study, electrohydrolysis was directly applied to organic waste to produce hydrogen gas but not for the pretreatment. Electrohydrolysis is the chemical process to decompose the organic matter by breaking the bonds between polymers provoked by application of current through electrodes. When electrodes connected to direct current, one electrode becomes positively charged and another electrode becomes negatively charged. This starts the movement of electrolyte towards electrodes i.e., positive ions moves to cathode and negative ions to anode. Several studies on hydrogen gas production by application of direct current to organic wastes and effluents of fermentation were reported [10e12]. Microbial electrolysis cells were also used for H2 gas production from organic wastes [13e16]. In one study, the effects of nutrient ratios on hydrogen gas production by batch dark fermentation of waste peach pulp was investigated using statistical experiment design. The effluent of dark fermentation experiments were used for H2 gas production by electrohydrolysis [17]. In our previous studies, raw organic wastes such as leachate, olive mill wastewater, anaerobic sludge were used to produce hydrogen gas by electrohydrolysis [4,5,18,19]. In some studies, PVC panels were used as electrical energy source for H2 gas production by electrohydrolysis from organic wastes [6]. Different electrodes were tested in electrohydrolysis studies and aluminum electrodes were found as the most effective one due to its high electrical conductivity [7]. Based on literature reports, hydrogen gas production from food wastes by electrohydrolysis was not investigated. Electrohydrolysis of food waste with high organic content to produce hydrogen gas is a novel and promising method. A Box-Behnken statistical experiment design approach was used in this study to investigate the effects of important operating variables (applied voltage, solids content of the waste and operation time) on the rate and extent of hydrogen gas production and also on total organic carbon (TOC) removal from the food waste.
Design of experiments The effects of variables on the response are a complicated and observed by means of the mostly known approach which is altering one factor at a time for multivariable systems. However this technic is not usable for estimation of responses. So, some experimental statistical design should be used for optimization of the reaction conditions. For this aim, several significant parameters are determined by response surface methodology. This method as called as RSM. This design contains 3-level factorial design, central composite design (CCD) [20,21], Box-Behnken design [22] and D-optimal design [23]. Between all the RSM designs, Box-Behnken design needs fewer runs than the other RSM designs. This design allows and shows to efficiency at intermediate levels not experimentally studied [24,25]. The mathematical relationship which is offered by BoxBehnken design application between the dependent variables (Y) and the independent variables (X) can be approximated by a (second-order) polynomial equation as follows:
Y ¼ b0 þ b1 X1 þ b2 X2 þ b3 X3 þ b12 X1 X2 þ b13 X1 X3 þ b23 X2 X3 þ b11 X21 þ b22 X22 þ b33 X23
(2)
This approach was selected to predict a potential response function. A total of 15 experiments are demanded to determine 9 coefficients of the quadratic equation. This regression model includes one block term, three linear terms, three quadratic terms, and three interaction terms. The main aim of this study is to observe and interpret the effects of the applied voltage, reaction time and initial solid content on percent hydrogen gas production, cumulative hydrogen gas production and percent total organic carbon removals from food wastes by electrohydrolysis method which is arranged by Box-Behnken design approach.
Materials and methods A power supplier and two aluminum electrodes were used as shown in Fig. 1. Food waste obtained from the university kitchen was filled into bottles. Aluminum electrodes were immersed inside the bottles and connection of power supply to the system was done by wires. The bottles were closed very tightly using silicone rubber stoppers and screw caps. In addition silicon was used to seal the bottles and to eliminate any gas leakage from the bottles. Aluminum electrodes were pure aluminum with diameters of 4 mm, lengths of 20 cm and weight of 30 g each. Distance between the electrodes was 2 cm. Experiments were performed at room temperature using 300 mL food waste. Water content of food waste was 75%. Food waste was homogenized by using a mixer. Water added in food wastes to adjust the solid content. Solid content value in food waste was arranged by different kind of foods. The raw food waste contained 43 mg/g TOC, 23.5 g/L total solids with a pH of 4.67 and ORP of 66 mV. The voltage which is applied to system was nearly kept stable throughout experiment. However, during experiments the current intensities (mA) altered due to variations in the reaction mechanism and changes in the waste composition. These changes caused the corrosion on the electrodes. So, current intensities were continuously reported by the power supply unit and were also recorded by using a multimeter
Fig. 1 e The experimental set-up.
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
3
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 x x x ( 2 0 1 8 ) 1 e7
Table 1 e Independent variable levels in statistical design. Variable Reaction Time (h) Voltage, (V) Solid Content, (%)
Symbol
Low (1)
Center (0)
High (þ1)
X1 X2 X3
25 1 10
50 3 25
75 5 40
which is connected to the system in series. Hydrogen gas produced from waste was stored in the head space of the bottles. Hydrogen gas taken from the bottles was analyzed by a gas chromatograph (GC, HP Agilent 6890) to determine the percentage of hydrogen gas. Nitrogen gas was used as carrier with a flow rate of 30 mL min1 and the head pressure was 22 psi. Temperatures of the injection, oven, detector, and filament were 35 C, 120 C, 120 C, 140 C, respectively. Food wastes were taken from the bottles everyday and total organic carbon (TOC) analyses was done using A Shimadzu TOC Analyzer (Shimadzu-SSM 550 model). The volume of hydrogen gas in the head space of bottles was measured by water displacement method using acid solution of acid and 10% NaCl. Cumulative hydrogen gas was calculated according to the following equation: VH2;i ¼ VH2;i1 þ Vw CH2;i þ VG;i þ VG;i CH2;i VG;i1 CH2;i1
mentioned as effective and useful way for optimization of the three variable response functions, predicting the response of the fitted model and checking the sufficiency of the design model by the ANOVA tests. The reaction time (X1), direct current voltages (X2), solid content (X3) as the independent and percent hydrogen gas production (Y1), cumulative hydrogen production (Y2) and percent TOC removals (Y3) as the dependent variables were selected in statistical design for this purpose. The low, center and high levels of each variable organized by statical approach as 1, 0, and þ1 respectively are shown in Table 1. While reaction time (X1) changed between 25 and 75 h, voltages (X2) altered between 1 and 5 V. All these selections were executed for observing an effective hydrogen gas production. The initial solid content of waste (X3) was ranged from 10% to 40%. The effects of independent variables (reaction time, voltage and solid content) on dependent variables or objective functions (hydrogen gas percentage, cumulative hydrogen gas production and total organic carbon removal) were evaluated. The experimental conditions which are performed according to Box-Behnken design for hydrogen gas production from food waste are given in Table 2. Obtained results at the pre-determined runs planned by Box-Behnken are also presented in Table 2. After experiments, predicted results are determined by Box-Behnken statistical approach at the predetermined experimental runs. In order to compare observed and predicted results, all experimental and predicted results are presented in Table 3. The effects of each parameter on hydrogen gas production and total organic matter removal by means of regression
(1)
where; VH2,ie1 and VH2,ie are cumulative H2 volumes (mL) calculated for (ie1)th and (ith) condition, VW is the total amount of gas (mL) measured in water displacement, CH2;i and CH2;i are measured H2 percentage at (ith) and (ie1)th condition, VG,i and VG,ie1 are the headspace volumes at (ith) and (ie1) the condition.
Results and discussion The effects of the independent variables on dependent variables in another saying the response functions were evaluated by Box-Behnken design. This design has main effects, interaction effects, and quadratic effects [22]. This approach was
Table 2 e Obtained results at the pre-determined experimental runs arranged by Box-Behnken statistical approach. Run No
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Variable levels
Experimental results
Reaction Time (h)
Voltage (V)
Solid Content (%)
HGP (%)
CHGV (mL)
TOC Removal (%)
50 75 75 50 75 75 50 25 25 50 25 25 50 50 50
1 5 1 5 3 3 5 3 1 1 5 3 3 3 3
40 25 25 40 10 40 10 10 25 10 25 40 25 25 25
26 100 32.8 94.1 99 98.5 99 75.8 14 16.4 79.4 55.7 90.8 90 92
63.2 6411 218 1302 4250 1792 4255 985 34 76 1032 446 2204 1873 2203
3.83 32.8 8.63 7.89 22.25 11.2 16.45 7.8 0.9 1.8 2.32 5.58 7.72 7.72 7.1
HGP: Hydrogen Gas Percentage, CHGV: Cumulative Hydrogen Gas Volume; TOC: Total Organic Carbon.
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
4
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 x x x ( 2 0 1 8 ) 1 e7
Table 3 e Observed and predicted results at the pre-determined experimental runs. Y1, HGP (%)
Run No
Y2, CHGV (mL)
Y3, TOC Removal (%)
Obs.results
Pred. results
Obs.results
Pred. results
Obs.results
Pred. results
26 100 32.8 94.1 99 98.5 99 75.8 14 16.4 79.4 55.7 90.8 90 92
25.41 100 34.05 88.74 93.39 99.34 100 76.46 7.98 21.76 78.15 62.31 90.93 90.93 90.93
63.2 6411 218 1302 4250 1792 4255 985 34 76 1032 446 2204 1873 2203
162.35 6070.35 320.65 1519.75 4365.10 1914.90 4480.55 862.10 374.65 141.75 929.35 330.90 2093.33 2093.33 2093.33
3.83 32.8 8.63 7.89 22.25 11.2 16.45 7.8 0.9 1.8 2.32 5.58 7.72 7.72 7.1
2.13 29.67 7.22 7.90 23.68 14.31 18.16 4.69 4.03 1.78 3.73 4.15 7.51 7.51 7.51
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
HGP: Hydrogen Gas Percentage, CHGV: Cumulative Hydrogen Gas Volume; TOC: Total Organic Removal.
model can be shown in Eqs. (3)e(5). It can be said that all parameters had an increasing effect on response function such as hydrogen gas percentage, cumulative hydrogen gas formation and TOC Removal. Voltage was the most effective on hydrogen gas production because of its highest coefficient followed by other independent variables such as reaction time and solid content. Analysis of variance (ANOVA) results are presented in Tables 4e6. ANOVA test indicates that the predictability of the model is at 95% confidence interval. The experimental data are compatible with response functions. Coefficient of determination (R2) was larger than 0.99. In addition, the F value is much greater than 3.70, so this means that the treatment is significant. The p-values for the linear terms in the ANOVA, means that independent variables had significant effect (p < 0.05). Independent variables had significant effect on hydrogen gas production and organic matter removal because of smaller than p value limit. However, solid waste did not have significant effect on dependent variables. Coefficients which were determined for percent hydrogen gas production (Y1), cumulative hydrogen gas volume (Y2) and
percent total organic carbon removal (Y3) are presented in following equations: Y1 ¼174;64375þ2;36817X1 þ51;23125X2 þ2;24694X3 9*103 X1 X2 0:0134X1 X3 þ0:12917X2 X3 8:5066*103 X21 7:26667X22 0:012185X23 R2 ¼0;9852
(3)
Y2 ¼1535:506144:97433X1 þ1887:98750X2 þ153:81278X3 þ25:97500X1 X2 þ1:27933X1 X3 þ24:50167X2 X3 0:21969X21 76:72292X22 1:61063X23 R2 ¼0;9881
(4)
Y3 ¼þ52:957781:12044X1 9:12210X2 0:57016X3 þ0:11375X1 X2 þ5:88569*103 X1 X3 þ0:088332X2 X3 þ6:29172*103 X1 0:070319X22 þ1:17293*103 X23 R2 ¼0;9474
(5)
Table 4 e ANOVA test for hydrogen gas percentage (%) Y1.
Table 5 e ANOVA test for cumulative hydrogen gas volume (mL) Y2.
Source
Sum of squares
Df
Mean Square
F ratio
P value
Source
Model X1(Reaction Time) X2 (Voltage) X3 (Solid Content) X1X2 X1X3 X2X3 X21 X22 X23 Residual Lack of Fit Pure Error Total (corr)
14909.19 1455.30 10103.31 33.62 0.81 101.00 60.06 104.37 3119.52 27.75 224.28 222.26 2.03 15133.47
9 1 1 1 1 1 1 1 1 1 5 3 2 14
1656.58 1455.30 10103.31 33.62 0.81 101.00 60.06 104.37 3119.52 27.75 44.86 74.09 1.01
36.93 32.44 225.23 0.75 0.018 2.25 1.34 2.33 69.54 0.62
0.0005 0.0023 <0.0001 0.4262 0.8983 0.1938 0.2995 0.1877 0.0004 0.4671
73.11
0.0135
Model X1(Reaction Time) X2 (Voltage) X3 (Solid Content) X1X2 X1X3 X2X3 X21 X22 X23 Residual Lack of Fit Pure Error Total (corr)
Sum of squares
Df
Mean Square
F ratio
P value
4.799Eþ007 1.294Eþ007 1.987Eþ007 4.444Eþ006 6.747Eþ006 9.206Eþ005 2.161Eþ006 69613.21 3.478Eþ005 4.849Eþ005 5.793Eþ005 5.064Eþ005 72820.67 4.857Eþ007
9 1 1 1 1 1 1 1 1 1 5 3 2 14
5.332Eþ006 1.294Eþ007 1.987Eþ007 4.444Eþ006 6.747Eþ006 9.206Eþ005 2.161Eþ006 69613.21 3.478Eþ005 4.849Eþ005 1.159Eþ005 1.688Eþ005 36410.33
46.02 111.68 171.54 38.36 58.24 7.95 18.65 0.60 3.00 4.19
0.0003 0.0001 <0.0001 0.0016 0.0006 0.0372 0.0076 0.4733 0.1437 0.0961
4.64
0.1825
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
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 x x x ( 2 0 1 8 ) 1 e7
5
Table 6 e ANOVA test for total organic carbon removal (%) Y3. Source Model X1(Reaction Time) X2 (Voltage) X3 (Solid Content) X1X2 X1X3 X2X3 X21 X22 X23 Residual Lack of Fit Pure Error Total (corr)
Sum of squares
Df
Mean Square
F ratio
P value
954.27 424.62 245.27 49.04 129.40 19.49 28.09 57.09 0.29 0.26 52.96 52.71 0.25 1007.23
9 1 1 1 1 1 1 1 1 1 5 3 2 14
106.03 424.62 245.27 49.04 129.40 19.49 28.09 57.09 0.29 0.26 10.59 17.57 0.13
10.01 40.09 23.15 4.63 12.22 1.84 2.65 5.39 0.028 0.024
0.0103 0.0014 0.0048 0.0841 0.0174 0.2330 0.1644 0.0679 0.8746 0.8823
138.75
0.0072
Hydrogen gas evolution Percent hydrogen gas production for different voltages is depicted in Fig. 2. Applied voltage was changed between 1 V and 5 V. Percent hydrogen gas increased up to 4 V and then slightly decreased with increasing voltages over 4 V at the end of 75 h reaction time. Percent hydrogen gas also changed throught experiments and reached nearly pure H2 gas (99% H2) at 4 V and 5 V. The lowest voltage resulting in the highest H2 content (99%) was 4 V. Cumulative hydrogen gas production with applied voltage is presented in Fig. 3. The highest cumulative hydrogen gas (6410 mL) was achieved by 5 V. Moreover, the most suitable direct current voltage maximizing cumulative hydrogen gas volume (6410 mL), hydrogen gas production rate (2050 mL d1) and percent hydrogen gas (99%) in the gas phase was 5 V. Nearly pure (99%) hydrogen gas produced by electrohydrolysis may be directly used in fuel cells.
Fig. 3 e Variation of cumulative hydrogen gas volume at different voltages and reaction times (Solid Content ¼ 25%).
Control experiments which include water electrolysis and natural decomposition of food waste were also performed to evaluate hydrogen gas percentage and the cumulative hydrogen gas evolution. According to evaluation of control water experiments at different voltages, maximum cumulative hydrogen gas volume (50 mL) and percent hydrogen gas (5%) in the gas phase was obtained at maximum voltage (5 V) during 75 h reaction time. At these experimental conditions, food waste produced maximum cumulative hydrogen production as 6410 mL by means of electrohydrolysis of food waste. So, hydrogen gas production in water electrolysis was less than 1% of cumulative hydrogen gas production for maximum voltage in experiments. It can be said that cumulative hydrogen gas volume in experiments didn't affected by water electrolysis. In addition, natural decomposition of food waste was not yield hydrogen gas volume and hydrogen gas percentage within 3 days with 5 V. It can be also said that natural decomposition of food waste did not affect the cumulative hydrogen gas volume. Therefore, hydrogen gas evolution was mainly achieved by electrohydrolysis of food waste according to applied voltage. Protons (Hþ ions) arised from organic waste decomposition reacted with electrons obtained by Al ionization from anode to yield hydrogen gas as expressed as Eqs. (6)e(8). The amount of electrons achieved at low voltages (<2 V) was not enough to neutralize protons arised from food waste decomposition. In other words, hydrogen gas production was restricted by available electrons at low voltages producing low volumes of hydrogen gas in spite of the existence of protons. Al (III) ion released from the electrode ionization was also low concentration in medium at low voltages. Al /AlðIIIÞ þ 3e o
(6)
Food waste Organic compounds/nHþ
þ decomposed organics nHþ þ n e /n=2 H2 ðgÞ
Fig. 2 e Variation of percent hydrogen gas production with reaction time at different applied voltages (Solid content ¼ 25%).
(7) (8)
Decomposed organic compounds such as volatile fatty acids were generally yielded from decomposition of food waste. When organic compounds in food waste were degraded into other organic matters, protons were arising in the media. Electrons and Al (III) ions were also released to the
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
6
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 x x x ( 2 0 1 8 ) 1 e7
medium after application of voltage. The arised protons and electrons were balanced with 5 V voltages producing the highest cumulative hydrogen gas production. The protons released from decomposition of food waste and Al (III) ions released from anode ionization competed for the electrons provided by DC current. Part of the electrons was reacted with Al (III) ions and part of them produced hydrogen by complexing with protons (Hþ). Al (III) ions were come out the organic waste forming Al salts. Finally, anode lost weight due to occuration of Al (III) and electron in reactions. Al (III) was accumulated on cathode in form of pure Alo and cathode weight slightly increased. The weight loss of aluminum electrodes during the experiment was observed as 1 g at 5 V and 75 h reaction time. The weight loss of aluminum electrodes was observed as 0.232 g and 0.096 g at 3 V and 1 V respectively, during 75 h reaction time. The weight loss of aluminum is very small than the weight of aluminum. This observation (only 0.096e1 g aluminum corrosion into solution) is good for hydrogen gas production. Initial pH increased with time and achieved a higher level at the end of reaction due to (Hþ) ion removal from the media by electrohydrolysis. The final pHs were around 4.5. Initial conductivities also ascended with arising decomposed organic compounds concentration. Morever, conductivities also ascended a higher level at the end of reaction due to Al(III) release to the reaction medium throughout electrohydrolysis.
Total organic carbon (TOC) removal Total organic carbon removals were also evaluated at predetermined runs arranged by Box-Behnken statistical approach by electrohydrolysis of food waste. TOC concentrations decreased from initial 43 mg/g to nearly 25.8 mg/g after electrohydrolysis application on food waste. As can be shown in Fig. 4, organic matter in food waste was decomposed by the microorganisms and electrohydrolysis and the released organic compounds (mainly VFAs) and CO2 resulting in remarkable total organic carbon removal. In other words, the applied voltage and natural decomposition of food waste was caused significant TOC removal due to high organic matter. However, only 2% TOC removal was observed with anaerobic decomposition of food waste in control experiments.
Table 7 e Comparison of hydrogen gas production from food waste with other organic wastes. Waste
Voltage
HGP (%)
CHGV (mL/d)
HGPR (mL/ d)
2 3
94 95
2775 3020
687 614
[4] [5]
4 5
99 99
5000 7000
1277 2240
[19] This study
Sludge Olive Oil Mill Leachate Food waste
References
HGP: Hydrogen Gas Percentage; CHGV: Cumulative Hydrogen Gas Volume; HGPR: Hydrogen Gas Production Rate.
Microbial TOC removal is very small in experiments. The yield of hydrogen gas production (mL H2 g1 TOC) is a significant parameter pointing the influence of hydrogen gas production per unit TOC removal. According to results, the highest yield of hydrogen production was 493 mL H2 mg1 TOC at 5 V direct current voltage. Electrohydrolysis of food waste for production of hydrogen gas was not reported in literature. In our previous studies, different organic wastes and wastewaters were used to produce hydrogen gas using electrohydrolysis [4e7,19]. The results of this study are compared with our previous studies using other organic wastes for H2 gas production by electrohydrolysis in Table 7. Hydrogen gas content (99%), cumulative hydrogen gas volume (7000 mL), hydrogen gas production rate (2240 mL d1) and hydrogen yield (493 mL H2 g1 TOC) obtained in this study are higher than the other studies listed in Table 7 indicating high potential of food waste as raw material for H2 gas production by electrohydrolysis. The most suitable voltage/reaction time/solid content values resulting in the highest hydrogen gas content (99%), the highest cumulative hydrogen gas volume (7000 mL) and TOC removal (33%) were found to be 5 V/75 h/20%. Validation experiment at the most suitable conditions was performed and almost the same results were obtained as predicted by the response function.
Conclusions
Fig. 4 e Variation of total organic carbon removal with different reaction times at applied voltages (Solid Content ¼ 25%).
A Box-Behnken statistical approach was used to investigate H2 gas production by electrohydrolysis of food wastes and to determine the most suitable reaction time, voltage and solid content maximizing the hydrogen gas content, cumulative hydrogen gas volume and TOC removal. According to experimental results and ANOVA test, reaction time and applied voltage were determined to have significant effects on H2 gas production and TOC removal. The optimal voltage/reaction time/solid contents resulting in the highest hydrogen gas content (99%), the highest cumulative hydrogen gas volume (7000 mL) and TOC removal (33%) were 5 V/75 h/20%. Hydrogen gas production rate of 2240 mL d1 and hydrogen yield of 493 mL H2 g1 TOC were achieved under these experimental conditions. Electrohydrolysis was found to be an effective method of producing nearly pure (99%) hydrogen gas from food wastes.
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079
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 x x x ( 2 0 1 8 ) 1 e7
references
[1] Winter CJ. Into the hydrogen energy economy-milestones. Int J Hydrogen Energy 2005;30:681e5. [2] Das D, Veziroglu TN. Hydrogen production by biological processes: survey of literature. Int J Hydrogen Energy 2001;26:13e28. [3] Iervolino G, Vaiano V, Sannino D, Rizzo L, Galluzzi A, Polichetti M, et al. Hydrogen production from glucose degradation in water and wastewater treated by Ru-LaFeO3/ Fe2O3 magnetic particles photocatalysis and heterogeneous photo-Fenton. Int J Hydrogen Energy 2018. in press. [4] Kargi F, Catalkaya Ebru C, Uzuncar S. Hydrogen gas production from waste anaerobic sludge by electrohydrolysis: effects of applied DC voltage. Int J Hydrogen Energy 2011;36:2049e56. [5] Kargi F, Catalkaya EC. Hydrogen gas production from olive mill wastewater by electrohydrolysis with simultaneous COD removal. Int J Hydrogen Energy 2011;36:3457e64. [6] Eker S, Kargi F. Hydrogen gas production from electrohydrolysis of industrial wastewater organics by using photo-voltaic cells (PVC). Int J Hydrogen Energy 2010;35:12761e6. [7] Kargi F. Comparison of different electrodes in hydrogen gas production from electrohydrolysis of wastewater organics using photo-voltaic cells (PVC). Int J Hydrogen Energy 2011;36:3450e6. [8] Elliott A, Mahmood T. Comparison of mechanical pretreatment methods for the enhancement of anaerobic digestion of pulp and paper waste activated sludge. Water Environ Res 2012;84(6):497e505. [9] Wood N, Tran H, Master E. Pretreatment of pulp mill secondary sludge for high-rate anaerobic conversion to biogas. Bioresour Technol 2009;100(23):5729e35. [10] Veluchamy C, Kalamdhad AS. Biochemical methane potential test for pulp and paper mill sludge with different food/microorganisms ratios and its kinetics. Int Biodeterior Biodegrad 2017;117:197e204. [11] Day DL, Steinberg MP. Electrochemical conversion of organic material. US Patent No. 4200505. April 1980. [12] Roychowdhury S. Process for production of hydrogen from anaerobically decomposed organic materials. US Patent No.6090266. July 2000.
7
[13] Paterek JR. Method and apparatus for hydrogen production from organic wastes and manure. US Patent No. 6887692. May 2005. [14] Liu H, Grot S, Logan BE. Electrochemically assisted microbial production of hydrogen from acetate. Environ Sci Technol 2005;39:4317e20. [15] Cheng S, Logan BE. Sustainable and efficient biohydrogen production via electro hydrogenesis. Proc Natl Acad Sci USA 2007;104:18871e3. [16] Ditzig J, Liu H, Logan BE. Production of hydrogen from domestic wastewater using a bioelectrochemically assisted microbial reactor (BEAMR). Int J Hydrogen Energy 2007;32:2296e304. [17] Argun H, Siaka D. Hydrogen gas production from waste peach pulp by dark fermentation and electrohydrolysis. Int J Hydrogen Energy 2016;41:1156e576. [18] Call D, Logan BE. Hydrogen production in a single chamber microbial electrolysis cell lacking a membrane. Environ Sci Technol 2008;42:3401e6. [19] Kargi F, Catalkaya EC. Electrohydrolysis of landfill leachate organics for hydrogen gas production and COD removal. Int J Hydrogen Energy 2011;36:8252e60. [20] Boza A, De la Cruz Y, Jordan G, Jauregui-Haza U, Aleman A, Caraballo I. Statistical optimization of a sustained-release matrix tablet of lobenzarit disodium. Drug Dev Ind Pharm 2000;26:1303e7. [21] Box GEP, Wilson KB. On the experimental attainment of optimum multifactorial conditions. Roy Stat Soc 1951;13:1e12. [22] Singh SK, Dodge J, Durrani MJ, Khan MA. Optimization and characterization of controlled release pellets coated with experimental latex: I. Anionic drug. Int J Pharm 1995;125:243e55. [23] Sanchez-Lafuente C, Furlanetto S, Fernandez-Arevalo M. Didanosine extended-release matrix tablets: optimization of formulation variables using statistical experimental design. Int J Pharm 2002;237:107e18. [24] Hamed E, Sakr A. Application of multiple response optimization technique to extended release formulations design. J Contr Release 2001;73:329e38. [25] Abbasi AF, Ahmad M, Wasim M. Optimization of concrete mix proportioning using reduced factorial experimental technique. ACI Mater J 1987:55e63 (JanuaryeFebruary).
Please cite this article in press as: C¸okay E, Hydrogen gas production from food wastes by electrohydrolysis using a statical design approach, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.01.079