Professional Documents
Culture Documents
Available at www.sciencedirect.com
Article history: The effects of C/N and C/P ratio on hydrogen yield and specific production rate were
Received 28 November 2007 investigated in dark fermentation of wheat powder solution (WPS) by external addition of N
Received in revised form and P since the wheat powder was N and P deficient. A BoxWilson statistical experiment
30 January 2008 design approach was used. C/N ratio was varied between 20 and 200 while C/P ratio was
Accepted 31 January 2008 between 50 and 1000. A quadratic response function was used to correlate the objective
Available online 12 March 2008 functions with the independent variables and the response function coefficients were
determined by regression analysis. Hydrogen yield increased with increasing C/N and C/P
Keywords:
ratios. Low nitrogen concentrations required low phosphorous contents for high hydrogen
Biohydrogen
yields. A C/N ratio of 200 and C/P ratio of 1000 yielded the highest H2 yield of
BoxWilson experiment design
281 ml H2 g1 starch at STP. SHPR also increased with increasing C/N and C/P ratios. Again
Dark fermentation
the highest SHPR (98 ml H2 g1 biomass h1 at STP) was obtained at C/N and C/P ratios of 200
Wheat powder solution (WPS)
and 1000, respectively. The C/N/P ratio maximizing the yield and formation rate of
C/N and C/P ratio
hydrogen was 100/0.5/0.1 w w1 w1 .
& 2008 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights
reserved.
hydrolysis. In other words 360 g monosachharide (glucose) is Table 1 Experimental conditions of the BoxWilson
obtained from 342 g of starch by hydrolysis. Total carbon statistical design
content of starch was 40% of the starch concentration.
The samples were centrifuged at 7000g to remove solids Axial X1 X2 Factorial X1 X2
from the liquid media and total sugar and TVFA analyses points points
were carried out in clear supernatants. Total sugar content
A1 20 525 F1 173.6 860.7
was determined by the acidphenol spectrometric method
A2 200 525 F2 173.6 189.31
[33]. TVFA analyses were carried out by using analytical kits A3 110 50 F3 46.4 860.69
(Spectroquant, 1.01763. 0001, Merck, Darmstadt, Germany) A4 110 1000 F4 46.4 189.31
and a PC spectrometer (WTW Photolab S12). Individual VFAs Center point 110 525
in the fermentation effluent were determined by using a high
performance liquid chromatograph (HPLC, Agilent 1100) with
a prevail organic acid column of 5 mm 150 mm 4:6 mm and
a UV detector of 220 nm. The carrying medium was 25 mM
1
KH2 PO4 at pH 2.5 with a flow rate of 1 ml min . (Y 1 , mg H2 g1 starch) and the specific rate of H2 formation
Hydrogen gas was sampled from the head space of the mg H2 g1 biomass h1 were considered as the objective
bottles by using gas-tight glass syringes and hydrogen functions. N and P are the externally added nutrients exclud-
concentration was determined by using a gas chromatograph ing the N and P content of the wheat. The C/N and C/P ratios
(Agilent 6890). The GC column was Alltech, Hayesep D 80/100 were varied between X1 202200 and X2 5021000, respec-
6 in 1=8 in 085 in. Nitrogen gas was used as carrier with a tively. The experiments consisted of four axial and four
flow rate of 30 ml min1 and the head pressure was 22 psi. factorial and a center point totaling nine experiments at
Temperatures of the oven, injection, detector, and filament different C/N and C/P ratios. The center point was repeated
were 35, 120, 120, 140 1C, respectively. The amount of total gas three times to test the reproducibility of the experiments.
produced was determined by water displacement method Table 1 summarizes the list of experiments conducted by using
everyday using sulfuric acid and NaCl containing solution. the BoxWilson statistical experiment design. A control
The cumulative hydrogen gas production was determined by experiment with no inoculation and another experiment with
using the following equation: no external N and P addition were performed along with the
VH2 ;i VH2 ;i1 VW CH2 ;i VG;i CH2 ;i VG;i1 CH2 ;i1 , (1) BoxWilson design experiments.
The following quadratic response function was used to
where VH2 ;i and VH2 ;i1 are the volumes of cumulative correlate the independent variables, X1 and X2 with the
hydrogen (ml) calculated after the ith and the previous objective function, Y.
measurement; VW is the total gas volume measured by the
water displacement method (ml); CH2 ;i is the concentration of Y b0 b1 X1 b2 X2 b12 X1 X2 b11 X21 b22 X22 . (2)
H2 gas in the total gas measured by the water displacement The experimental data obtained from the experiments
method (%); VG;i and VG;i1 are the volumes of the gas in the presented in Table 1 was used for determination of the
head space of the bottle for the ith and the previous coefficients in Eq. (2) using a STATISTICA 5 program for
measurement (ml); CH2 ;i and CH2 ;i1 are the percent H2 in the regression analysis.
head space of the bottle for the ith and the previous
measurement. The amount of released hydrogen gas and in
the head space of the bottle were measured independently 3. Results and discussion
and added up to determine cumulative H2 formation for every
period of sampling. The results of the BoxWilson design experiments are
Biomass concentration in the inoculum was determined by presented in Table 2 along with the predicted values from
filtering 20 ml sample through a 0:45 mm milipore filter, drying the response functions. The coefficients of the response
at 105 1C and determining the constant dry weight [34]. pH functions determined by regression analysis are summarized
and ORP of the fermentation medium were monitored by in Table 3 and were used to predict the response function
using a pH meter and ORP meter with relevant probes (WTW, values presented in Table 2. Hydrogen yield predictions were
Germany). pH of the medium decreased from an initial value in good agreement with the experimental results R2 0:86
of 7.0 to nearly 4.5 in early stages of fermentation due to VFA as shown in Table 2. A control experiment was conducted
production. Medium pH was adjusted to 7.0 by addition of without any N and P addition. The results of this experiment
10 M NaOH twice a day. pH was maintained between 6 and 7 are also presented in Table 2 (No NP addition).
by manual pH control. ORP values varied between 100 and Fig. 1 depicts variation of cumulative hydrogen (ml), starch,
300 mV, in general. glucose and TVFA concentrations mg L1 with time for the
experiment F1 where C/N and C/P ratios were 173 and 860,
2.4. BoxWilson statistical experiment design respectively since the hydrogen yield and the rate were the
highest for this experiment. Starch concentration decreased
A BoxWilson statistical experiment design method was used from an initial value of nearly 18 to 2 g L1 after 150 h of
to determine the effects of C/N and C/P ratios on hydrogen fermentation indicating effective starch hydrolysis by the
yield and production rate. C/NX1 and C/PX2 ratios were anaerobic sludge bacteria. Cumulative hydrogen volume
considered as independent variables; the hydrogen yield increased rather fast within the first 100 h of fermentation
ARTICLE IN PRESS
1816 I N T E R N AT 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 33 (2008) 1813 1819
Table 3 Coefficients of the response functions for different C/N and C/P ratios
Y 1 , Yield 21.806 0.186 0.026 1:58 104 6:26 104 9:0 106 0.86
Y 2 , SHPR 6.034 0.061 0.006 4:30 10 5
2:45 10 4
2:5 106 0.78
3000 15000 The first step is the hydrolysis of starch to carbohydrates and
2500 12500 the second step is fermentation of carbohydrates to VFA and
H2 . Glucose and maltose constitute total sugars as inter-
2000 10000
mediate fermentation products and are converted to VFA and
1500 7500 H2 simultaneously. Since no sugar accumulation was ob-
1000 5000 served during fermentation, microbial hydrolysis of starch
was the slowest or the rate limiting step in this conversion
500 2500
scheme. Sugars were converted to VFA and H2 much faster
0 0 than they were produced.
0 50 100 150 200 250 300 350
The response functions with determined coefficients were
Time (h)
used to predict the variations of hydrogen yield and formation
Fig. 1 Variation of cumulative hydrogen E, starch (), rate with the C/N and C/P ratio. Variations of hydrogen yield
TVFA m, and total sugar concentrations with time for mg H2 g1 starch with C/P ratio at different C/N ratios are
the experiment F1 (C=N 173, C=P 860). depicted in Fig. 2. At low C/N ratios o46 indicating excess
nitrogen (high N/C ratio), hydrogen yield increased with
decreasing C/P or increasing P/C ratio indicating high
phosphorous requirements at high nitrogen concentrations.
and reached nearly 3650 ml (30 1C, 1 atm) at the end of 200 h of At high C/N ratios 4110 indicating low nitrogen contents
incubation which is equivalent to 3590 ml at STP (25 1C, 1 atm). (low N/C ratios), hydrogen yield increased with increasing C/P
Sugar was an intermediary product in starch fermentation to or decreasing P/C ratio indicating low phosphorous require-
organic acids and hydrogen gas. Therefore, total sugar ments at low nitrogen contents. At a high C/P ratio of 950 (low
concentration increased from an initial level of 2 g L1 to P content) hydrogen yield increased with increasing C/N or
nearly 9 g L1 within 50 h and then decreased to less than decreasing N/C ratio indicating low nitrogen requirements at
1 g L1 due to sugar conversion to VFA and hydrogen gas. low phosphorous contents. N and P requirements were
Sugar concentration remained almost constant at 0:4 g L1 for proportional with each other and also with the carbon or
the rest of the fermentation period. The TVFA concentration starch content of the wheat due to prevailing stoichiometry of
increased steadily from 2 to 11 g L1 at the end of 300 h of the anaerobic metabolism. The highest hydrogen yield
fermentation. HPLC analysis of VFA indicated that the major (23 mg H2 g1 starch or 281 ml H2 g1 starch at STP) was ob-
organic acids were acetic, butyric, propionic and lactic acids. tained at the highest C/N and C/P ratios of 200 and 1000
ARTICLE IN PRESS
I N T E R N AT 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 33 (2008) 1813 1819 1817
200 requiring low P for low N contents for high hydrogen yield due
20
to prevailing stoichiometry of anaerobic metabolism. Within
173 the range of independent variables the highest hydrogen yield
15 20
46 (23 mg H2 g1 starch or 281 ml H2 g1 starch at STP) was ob-
10 tained with the highest C/N and C/P ratios of 200 and 1000 or
110 the lowest N/C and P/C ratios of 0.005 and 0.001. N/P ratio
5 maximizing the hydrogen yield was 5. The hydrogen yield
C/N obtained with no nitrogen and phosphorous addition was
0 6:09 mg H2 g1 starch or 74:4 ml H2 g1 starch at STP which was
50 200 350 500 650 800 950 considerably lower than the maximum yield obtained at
C/P ratio C=N 200 and C=P 1000. The N and P content of the WP was
Fig. 2 Variation of hydrogen yield with the C/P ratio at not sufficient and external addition of N and P was required
different C/N ratios. for effective hydrogen production by dark fermentation of
WPS.
The specific hydrogen production rate (SHPR) is another
30 important parameter indicating effectiveness of the fermen-
tation. SHPR was calculated as the mg or ml hydrogen
25
produced per gram dry weight of initial biomass per unit
Yield (mgH2.g-1starch)
8
with C/N ratio at different C/P ratios. At low C/P ratios (o525,
7
excess P), hydrogen yield decreased with increasing C/N ratio
(decreasing N/C ratio) indicating limitations by the nitrogen 6 200
source at high phosphorous contents. However, at high C/P 5
173
ratios 4525 or low P contents hydrogen yield increased with 4
increasing C/N ratio (decreasing N/C ratio) steadily indicating 3 20
46 110
low nitrogen requirements at low phosphorous concentra- 2
1 C/N
tions. Similarly, at a low C/N ratio of 45 indicating presence of
excess nitrogen, hydrogen yield increased with decreasing 0
50 200 350 500 650 800 950
C/P ratio or increasing P/C ratio indicating the requirement for
C/P ratio
high phosphorous contents. At high C/N ratios of 195 and
above indicating low nitrogen content, hydrogen yield in- Fig. 4 Variation of SHPR with the C/P ratio at different C/N
creased with increasing C/P or decreasing P/C ratio indicating ratios.
ARTICLE IN PRESS
1818 I N T E R N AT 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 33 (2008) 1813 1819
nitrogen and phosphorous requirements for maximum rate was found to be 200. The external nitrogen requirement in
of hydrogen formation. The C/N/P ratio yielding the highest our case was lower making the process economically more
SHPR was 100/0.5/0.1. The SHPR obtained with no nitrogen attractive.
and phosphorous addition was 2:36 mg H2 g1 biomass h1 or Theoretical, hydrogen yields are 4 mol H2 mol1 glucose and
28:8 ml H2 g1 biomass h1 which was considerably lower than 2 mol of H2 mol1 glucose when acetic (2 mol) or butyric
that of the maximum SHPR indicating the need for external (1 mol) acids were the only end products from one mole of
nitrogen and phosphorous addition to yield C=N=P glucose, respectively. Formation of propionic acid (2 mol)
100=0:5=0:1 in the fermentation medium. consumes 2 moles of hydrogen per mole of glucose. That is
Fig. 5 depicts variation of SHPR with the C/N ratio at propionic acid formation is a hydrogen consuming reaction
different C/P ratios. At low C/P ratios (o525, high P content), and should be avoided. Lactic acid and ethanol formations do
SHPR decreased with increasing C/N ratio up to C=N 95 not yield hydrogen gas.
indicating limitations by the available nitrogen when phos- Since a mixture of acetic, butyric, propionic and lactic acids
phorous was in excess. However, at high C/N ratios above 95, were produced in this study, the highest hydrogen yield was
the SHPR increased steadily with increasing C/N ratio estimated to be 2 mol H2 mol1 sugar. At the experimental
indicating low requirement for nitrogen source. Similarly, at point of F1 with the highest hydrogen yield, 9072 mg L1 TVFA
high C/N ratios above 95 (low nitrogen content), the SHPR was produced at the end of the fermentation which is nearly
increased with increasing C/P ratio due to low phosphorous 50% of the initial starch. TVFAS consisted of acetic, butyric,
requirements by the organisms. The highest SHPR lactic and propionic acids with the concentrations of
(8:0 mg H2 g1 biomass h1 or 98 ml H2 g1 biomass h1 at STP) 3755 mg L1 (41%), 1627 mg L1 (18%), 1130 mg L1 (12.5%) and
was obtained at the highest C/N and C/P ratios of 200 and 1520 mg L1 (17%), respectively.
1000, respectively. The optimal C/N/P ratio maximizing the
SHPR was found to be 100/0.5/0.1. Low nitrogen and phos-
phorous requirements are due to low growth yields in
anaerobic organisms. 4. Conclusions
The highest hydrogen yield obtained in this study
(281 ml H2 g1 starch at STP) corresponds to a hydrogen yield Hydrogen production by dark fermentation of WPS using heat
of nearly 2 mol H2 mol1 glucose which is comparable with the treated anaerobic acidogenic sludge was investigated at
literature studies reporting hydrogen production from starch different initial C/N and C/P ratios in order to determine the
by dark fermentation [12,14,17]. Liu and Shen [13] studied N and P requirements of the organisms for effective hydrogen
hydrogen production from starch by dark fermentation and production.
obtained a maximum hydrogen yield of 194 ml H2 g1 starch A BoxWilson statistical experiment design was used by
and hydrogen formation rate of 9:8 ml H2 g1 biomass h1 at considering C/N (20200) and C/P (501000) ratio as independent
STP with a starch concentration of 15 g L1 . Our results were variables while hydrogen formation yield and the rate
considerably higher than those reported by Liu and Shen due were the objective functions. The experimental data was
to external addition of nitrogen and phosphorous and using correlated with a quadratic response function and the coeffi-
the most suitable C/N and C/P ratios to meet nutritional cients were determined by regression analysis. The response
requirements of the organisms. In a study by Lin and Lay [29] function predictions were in good agreement with the
the optimal C/N ratio was reported 47 for biohydrogen experimental data for hydrogen yield. The hydrogen yield
formation from sucrose solution. Since part of the nitrogen increased with increasing C/N and C/P ratio yielding the
was provided by the wheat in our study, the optimal C/N ratio maximum yield (23 mg H2 g1 starch or 281 ml H2 g1 starch at
STP) at C=N 200 and C=P 1000. The SHPR also increased
with increasing C/N and C/P ratios yielding the highest SHPR
(8:0 mg H2 g1 biomass h1 or 98 ml H2 g1 biomass h1 at STP)
at C=N 200 and C=P 1000. Low nitrogen contents required
10 low phosphorous for maximum hydrogen yields and formation
9 rates due to prevailing stoichiometry of the anaerobic
metabolism. High nitrogen and phosphorous concentrations
SHPR (mgH2.g-1cells.h-1)
8
inhibited hydrogen formation by dark fermentation probably
7 1000
by shifting the metabolic pathway. The C/N/P ratio yielding
6 860
the highest hydrogen yield 23 mg H2 g1 starch and formation
5
525 rate 8:0 mg H2 g1 biomass h1 was C=N=P 100=0:5=0:1.
4 50 The hydrogen yield and formation rate with no nitrogen
3 190
and phosphorous addition were 6:09 mg H2 g1 starch
2 (74:4 ml H2 g1 starch at STP) and 2:36 mg H2 g1 biomass h1
1 C/P
28:8 ml H2 g1 biomass h1 which were considerably lower
0 than the highest yield and the rate. The results indicated low
20 45 70 95 120 145 170 195
nitrogen and phosphorous requirements (N=P=starch
C/N ratio
0:2=0:04=100, w w1 w1 ) for high hydrogen yields and rates by
Fig. 5 Variation of SHPR with the C/N ratio at different C/P dark fermentation of WPS since part of the N and P present in
ratios. WP was also used for fermentation.
ARTICLE IN PRESS
I N T E R N AT 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 33 (2008) 1813 1819 1819
[19] Federov AS, Tsygankov AA, Rao KK, Hall DO. Hydrogen
Acknowledgment photoproduction by Rhodobacter sphaeroides immobilized on
polyurethane foam. Biotechnol Lett 1998;20:10079.
This study was supported by the Scientific and Technological [20] Eroglu I, Aslan K, Gunduz U, Yucel M, Turker L. Substrate
consumption rate for hydrogen production by Rhodobacter
Research Council of Turkey by a Grant number 105M296.
sphaeroides in a column photobioreactor. J Biotechnol
1999;70:10313.
R E F E R E N C E S
[21] Koku H, Eroglu I, Gunduz U, Yucel M, Turker L. Kinetics of
biohydrogen production by the photosynthetic bacterium
Rhodobacter spheroids O.U. 001. Int J Hydrogen Energy
[1] Winter CJ. Into the hydrogen energy economy-milestones. Int 2003;28:3818.
J Hydrogen Energy 2005;30:6815. [22] Shi XY, Yu HQ. Response surface analysis on the effect of cell
[2] United States Department of Energy. An integrated research, concentration and light intensity on hydrogen production by
development and demonstration plan. Hydrogen Pasteur Rhodopseudomonas capsulate. Process Biochem
Plan 2004. 2005;40:247581.
[3] Kapdan IK, Kargi F. Biohydrogen production from waste [23] Shi XY, Yu HQ. Hydrogen production from propionate by
materials. Enzyme Microb Technol 2006;38:56982. Rhodopseudomonas capsulate. Appl Biochem Biotechnol
[4] Levin DB, Pitt L, Love M. Biohydrogen production: prospects 2004;117:14354.
and limitations to practical application. Int J Hydrogen [24] Barbosa MJ, Rocha JMS, Tramper J, Wijffels RH. Acetate as a
Energy 2004;29:17385. carbon source for hydrogen production by photosynthetic
[5] Das D, Veziroglu TN. Hydrogen production by biological bacteria. J Biotechnol 2001;85:2533.
processes: survey of literature. Int J Hydrogen Energy [25] Oh YK, Scol EH, Kim MS, Park S. Photoproduction of hydrogen
2001;26:1328. from acetate by a chemoheterotrophic bacterium Rhodo-
[6] Asada Y, Miyake J. Photobiological hydrogen production. pseudomonas palustris P4. Int J Hydrogen Energy
J Biosci Biotechnol 1999;88:16. 2004;29:111521.
[7] Ghirardi ML, Zhang L, Lee JW, Flynn T, Seibert M, Greenbaum [26] Yokoi H, Mori S, Hirose J, Hayashi S, Takasaki Y. H2
E, et al. Microalgae: a green source of renewable H2 . Tibtech production from starch by mixed culture of Clostridium
2000;18:50611. butyricum and Rhodobacter sp M-19. Biotechnol Lett
[8] Melis A. Green alga hydrogen production: progress, chal- 1998;20:8959.
lenges and prospects. Int J Hydrogen Energy 2002;27:121728. [27] Yasuo A, Masaru T, Yasuyuki A, Masayo O, Katsuhiro I,
[9] Guan Y, Deng M, Yu X, Zang W. Two stage photo-production Tatsuki W, et al. Hydrogen production by co-cultures of
of hydrogen by marine green algae Platymonas subcordiformis. Lactobacillus and a photosynthetic bacterium Rhodobacter
Biochem Eng J 2004;19:6973. sphaeroides RV. Int J Hydrogen Energy 2006;31:150931.
[10] Pinto FAL, Troshina O, Lindbald P. A brief look at three [28] Yongzhen T, Yang C, Yongqiang W, Yanling H, Zhihua Z. High
decades of research on cyanobacterial hydrogen evolution. hydrogen yield from a two step process of dark and photo-
Int J Hydrogen Energy 2002;27:120915. fermentation of sucrose. Int J Hydrogen Energy
[11] Tsygankov AA, Federov AS, Kosourov SN, Rao KK. Hydrogen 2007;32:2006.
Production by cyanobacteria in an automated outdoor [29] Lin CY, Lay CH. Carbon/nitrogen-ratio effect on fermentative
photobioreactor under anaerobic condition. Biotechnol hydrogen production by mixed microflora. Int J Hydrogen
Bioeng 2002;80:77783. Energy 2004;29:415.
[12] Yokoi H, Saitsu AS, Uchida H, Hirose J, Hayashi S, Takasaki Y. [30] Jo JH, Lee DS, Park D, Choe WS, Park JM. Optimization of key
Microbial hydrogen production from sweet potato starch process variables for enhanced hydrogen production by
residue. J Bioscience Bioeng 2001;91:5863. Enterobacter aerogenes using statistical methods. Biore-
[13] Liu G, Shen J. Effects of culture medium and medium source Technol 2008;99:20616.
conditions on hydrogen production from starch using [31] Lin CY, Lay CH. Effects of carbonate and phosphate
anaerobic bacteria. J Bioscience Bioeng 2004;98:2516. concentrations on hydrogen production using anaerobic
[14] Yokoi H, Tokushige T, Hirose J, Hayashi S, Takasaki Y. H2 sewage sludge microflora. Int J Hydrogen Energy 2004;29:
production from starch by mixed culture of Clostridium 27581.
buytricum and Enterobacter aerogenes. Biotechnol Lett [32] Lee KS, Hsu YF, Lo YC, Lin PJ, Lin CY, Chang JS. Exploring
1998;20:1437. optimal environmental factors for fermentative hydrogen
[15] Liu H, Zhang T, Fang HPP. Thermophilic H2 production from production from starch using mixed anaerobic microflora.
cellulose containing wastewater. Biotechnol Lett Int J Hydrogen Energy 2007 in press, doi:10.1016/j.ijhydene.
2003;25:3659. 2007.10.019.
[16] Lay JJ. Biohydrogen generation by mesophilic anaerobic [33] Dubois M, Gilles KA, Hamilton JK, Rebers PA, Smith F.
fermentation of micro-crystalline cellulose. Biotechnol Colorimetric method for determination of sugars and re-
Bioeng 2001;74:2817. lated substances. Anal Chem 1956;8:35066.
[17] Yokoi H, Maki R, Hirose J, Hayashi S. Microbial production of [34] Greenberg AE, Clesceri LS, Eaton AD. Eds. Standard
hydrogen from starch manufacturing wastes. Biomass Bio- methods for the examination of water and wastewater. 21st
ener 2002;22:38995. edn. American Public Health Association (APHA), 2005,
[18] Han SK, Shin HS. Biohydrogen production by anaerobic Washington, DC, USA.
fermentation of food waste. Int J Hydrogen Energy
2004;29:56977.