Fermentative Hydrogen Production in Two Upflow Reactors
Chenlin Li1 and Herbert H. P. Fang2*
1
PhD student and 2 Chair Professor
Department of Civil Engineering, The University of Hong Kong
Pokfulam Road, Hong Kong SAR, China
ABSTRACT: Fermentative hydrogen from a synthetic wastewater containing 10 g/L of sucrose was
studied in two upflow reactors at 26ºC for 400 days. One reactor was filled with packing (RP) and the other
was packing free (RF). The effect of hydraulic retention time (HRT) from 2 h to 24 h was investigated.
Results showed that under steady state, the hydrogen production rate significantly increased from 0.63
L/L/d to 5.35 L/L/d in the RF when HRT decreased from 24 h to 2 h, the corresponding rates were 0.56
L/L/d to 6.17 L/L/d for the RP. In the RF, the hydrogen yield increased from 0.96 mol/mol-sucrose at 24 h
of HRT to the maximum of 1.10 mol/mol-sucrose at 8 h of HRT, and then decreased to 0.68
mol/mol-sucrose at 2 h. In the RP, the yield increased from 0.86 mol/mol-sucrose at 24 h of HRT to the
maximum of 1.22 mol/mol-sucrose at 14 h of HRT, and then decreased to 0.78 mol/mol-sucrose at 2 h.
Overall, the reactor with packing was more effective than the one free of packing.
Key Words: fermentation, granulation, hydrogen, packed-bed, packing, reactor, upflow.
1.
INTRODUCTION
Hydrogen is a promising alternative to fossil fuel. It produces water instead of greenhouse gases upon
combustion. Hydrogen is commonly produced by either electrolytic, thermochemical, or radiolytic process,
all of which are energy intensive. On the other hand, it has been demonstrated that hydrogen can also be
produced sustainably from organic waste and wastewater by fermentation bacteria. Most studies of
bio-hydrogen production were conducted using the continuous stirred tank reactor (CSTR). The reactor
under complete mixing condition facilitates the release of biogas and the effective pH control. CSTR has
been used in treating synthetic wastewaters containing glucose [1], sucrose [2], starch [3], cellulose [4], as
well as actual sugar manufacturing wastewater [5] and bean curd manufacturing waste [6]. However, it is
unable to maintain high levels of biomass due to the low sludge retention time. To retain high biomass in
reactors, various techniques have been developed for hydrogen fermentation, including sludge
immobilization [7], utilization of upflow reactor [8] and membrane bioreactor [9]. Among these techniques,
upflow reactors have been extensively applied in the anaerobic wastewater treatment system with high
efficiency. In the upflow reactor, sludge may agglutinate into granules, resulting in the increase of biomass
concentration and the reduction of sludge washout simultaneously.
This study was thus conducted to investigate the dark fermentative hydrogen production from a
sucrose-containing wastewater in two upflow reactors: one filled with packing (RP) and the other packing
free (RF). The effects of HRT on hydrogen production were investigated, and the porosity and settling
velocity of granules were characterized.
2.
MATERIALS AND METHODS
The seed sludge of mixed cultures was taken from a complete-mix fermentative reactor producing
hydrogen from sucrose at 26 ºC and pH 5.5. The reactor, which had been operated for 1 year, contained 1.0
g-VSS/L and produced a biogas comprising 50% hydrogen. Degrading each gram of sucrose produced 0.16
L of hydrogen.
A RF and a RP were used in this study. Both reactors had an effective volume of 2.8 L in volume with a 2.0
L built-in gas-liquid-solid separator at the top. The RP was packed with 120 plastic Flexiring (Koch), which
was 25mm in both height and diameter with surface to volume ratio of 235 m2/m3. The two reactors were
operated at 26ºC. The feed solution in both reactors was composed of 10 g/L of sucrose, plus the following
nutrients (in mg/L): NaHCO3 1250; NH4Cl 2500; KH2PO4 250; K2HPO4250; CaCl2 500; NiSO4 32;
MgSO4·7H2O 320; FeCl3 20; Na2BO4H2O 7.2; Na2MoO4·2H2O 14.4; ZnCl2 23; CoCl2·6H2O 21;
CuCl2·H2O 10; MnCl2·4H2O 30. The initial COD of substrate is about 12,000 mg COD/L. The RF was
operated at the HRT of 24, 20, 16, 12, 10, 8, 6, 4, 2 h, whereas the RP was operated at the HRT of 30, 24,
18, 14, 10, 8, 6, 4, 2 h. The initial pH of the feeding solution was controlled at 7.8, however, the effluent pH
decreased to pH 3.8-5.0 with the decrease of HRT. At each HRT, the reactors were operated for 5 weeks to
establish steady-state condition, judging from the constant sucrose degradation, hydrogen production and
effluent quality, before lowering the HRT.
The amount of biogas produced was recorded daily by the water replacement method. The composition of
biogas and concentrations of volatile fatty acids (VFA) and alcohols in the effluent were determined twice a
week as described previously [1]. Sucrose concentration was measured using the anthrone-sulfuric acid
method [10], the COD and VSS were measured according to the Standard Methods [11]. Granular sludge
was sampled from RP at 4 h of HRT to determine the average porosity by centrifugation. The sludge was
firstly filtered and then centrifuged sequentially from 4,000 up to 40,000 G.. The average porosity of the
granules was calculated as the volume ratio of the centrifuged water and the filtered granular sludge. The
settling velocities of individual granules were measured in a glass cylinder (185 mm height, 25 mm ID).
3.
RESULTS AND DISCUSSION
Sucrose conversion (%) .
100
95
90
a
85
45 0
5
10
15
20
25
30
40
35
30
b
Percentage (%) .
Percentage (%)
. Sucrose conversion (%) .
3.1 Production of hydrogen in RF
5
10
15
20
25
90
a
85
45
0
5
10
15
20
25
30
6.0
4.0
2.0
c
35
30
b
5
10
15
20
25
30
1.2
0.9
d
0.6
0
5
10
15
20
25
5
10
15
20
25
30
35
6.0
4.0
2.0
c
0.0
1.5 0
35
40
8.0 0
30
Rate (L/L/d)
8.0 0
30
Yield (mol/mol-sucrose) .
.
95
25
25
Yield (mol/mol-sucrose) . Rate (L/L/d)
100
0.0
1.5 0
5
10
15
20
25
30
35
1.2
0.9
d
0.6
0
5
10
15
20
25
30
35
HRT (h)
HRT (h)
Fig. 1 Effect of HRT on (a) sucrose conversion;
(b) hydrogen percentage; (c) hydrogen
production rate and (d) hydrogen yield in RF
Fig. 2 Effect of HRT on (a) sucrose conversion;
(b) hydrogen percentage; (c) hydrogen
production rate and (d) hydrogen yield in RP
Fig. 1 illustrates the HRT effect on: (a) sucrose conversion, (b) hydrogen percentage, (c) hydrogen
production rate, and (d) hydrogen yield in the RF. The effect of HRT in the range of 24 to 2 h on hydrogen
production was evaluated. Fig. 1(a) illustrates that sucrose degradation increased from 88.6±1.2% at HRT 2
h to 98.3±0.9% at HRT 24 h. Fig. 1(b) shows that the hydrogen percentage was not sensitive to HRT in the
range of 2-12 h, and declined from 38.0±3.1% to 32.1±2.7% as the HRT was further increased from 12 h to
24 h. The biogas was free of methane. Fig. 1(c) illustrates that the hydrogen production rate significantly
decreased from 5.35±0.62 L-H2/L/d to 0.63±0.02 L-H2/L/d when HRT increased from 2 h to 24 h. The
hydrogen yield increased from 0.68±0.08 mol-H2/mol-sucrose at 2 h of HRT to 1.10±0.05
mol-H2/mol-sucrose at 8 h, and then decreased to 0.96±0.04 mol-H2/mol-sucrose at 24 h, as illustrated in
Fig. 1(d).
In the RF, the effluent COD decreased from 11,000 mg COD/L at 24 h to 10,433 mg COD/L at 10 h, and
then increased to 10969 mg COD/L at 2 h. The COD removal efficiencies were in the range of 8.3-13.1% at
various HRT, in which hydrogen accounted for about 2.9-4.9%, and the other were estimated to be
converted into biomass. VFA and alcohol analysis shows that butyrate, acetate and caproate were the three
most abundant species in the effluent. Decreasing HRT from 24 to 2 h resulted in a decrease of butyrate in
effluent from 30.2% to 20.8%. Acetate accounted for 13.6% of total COD in effluent at HRT 24 h and
increased to 25.8% at HRT 2 h. Caproate increased from 7.5% to 22.4% with the decrease of HRT from 24
to 2 h. Other constituents in the effluent included methanol (2.4-10.6%), ethanol (4.2-14.4%), valerate
(4.3-9.3%), plus propionate, propanol and unconverted sucrose. The overall COD balance in the effluent
was 80.8-101.2%.
3.2 Production of hydrogen in RP
Fig. 2 illustrates the HRT effect over the range of 2-30 h on hydrogen production: (a) sucrose conversion,
(b) hydrogen percentage, (c) hydrogen production rate, and (d) hydrogen yield in the RP. Fig. 2(a) shows
that sucrose degradation increased from 89.2±0.8% at HRT 2 h to 98.8±0.9% at 24 h, and to 99.1±0.8% at
HRT 30 h. Fig. 2(b) shows that hydrogen content in biogas was not sensitive to HRT at 2-8 h of HRT, but
declined from 40.1±2.2% to 33.3±2.1% and 28.5±1.9% as the HRT was further increased from 8 h to 24 h
and 30 h, respectively. The biogas was free of methane. Fig. 2(c) illustrates that the hydrogen production
rate significantly decreased from 6.17±0.39 L/L/d to 0.56±0.04 L/L/d and 0.46±0.04 L/L/d when HRT
increased from 2 h to 24 h and 30 h, respectively. The hydrogen yield increased from 0.78±0.07
mol/mol-sucrose at 2 h of HRT to 1.22±0.13 mol/mol-sucrose at 14 h, and then decreased to 0.86±0.08
mol/mol-sucrose and 0.88±0.08 mol/mol-sucrose at 24 and 30 h, respectively, as illustrated in Fig. 2(d).
In the RP, the effluent COD decreased from 11,040 mg COD/L at 30 h to 10,432 mg COD/L at 14 h, and
then increased to 10,664 mg COD/L at 2 h. The COD removal efficiencies at various HRT were 8.0-13.1%,
of which hydrogen accounted for about 3.4-5.5%. Results show that butyrate, acetate and caproate were the
three most abundant byproducts in the effluent, similar to the product distribution in the packing-free
reactor. Decreasing HRT from 30 to 2 h resulted in an increase of acetate in effluent from 16.6% to 29.6%.
Butyrate accounted for 34.4% of total COD in effluent at HRT 30h and decreased to 25.1% at HRT 2 h.
Caproate increased from 6.8% to 14.2% with the decrease of HRT from 30 to 8 h, and then decreased to
8.7% at 2 h. Other constituents in the effluent included methanol (4.2-9.6%), ethanol (5.4-15.5%), plus
propionate, valerate, propanol and unconverted sucrose. The COD balance in the effluent was 82.6-98.6%.
3.3 Physical characteristics of hydrogen-producing granules
In both reactors, sludge agglutinated into granules after 2 months operation, resulting in the increase of
biomass concentration and the reduction of sludge washout. After 400 days operation, the average VSS
concentration increased up to 50 g/L, which indicate that upflow reactor is a promising system to establish
hydrogen production systems with high biomass. In the RP, visible 1.0-3.0 mm in diameter granules was
observed. In the RF, the average granular diameter was 1.0 mm. As shown in Fig. 3, effect of centrifugation
speed on granular average porosity was investigated. The granular porosity increased from 24% to 52%
with the centrifugation speed from 4,000 to 30,000 G and then leveled off from 35,000G to 40,000G, with
the porosity at 54%. The size of the granules ranged from 1.0 to 3.5 mm and their settling velocity ranged
from 10 to 37 mm/s. As shown in Fig. 4, the settling velocity increased significantly with the granular size.
Settling velocity (mm/s) `
Granule porosity (%) .
60
50
40
30
20
0
10000
20000
30000
40000
Centrifugation speed (G)
Fig. 3 Effect of centrifugation speed on granule
porosity
4.
45
35
25
y = 8.9916x + 4.2898
15
R2 = 0.6973
5
0.5
1.5
2.5
3.5
4.5
Granule diameter (mm)
Fig. 4 Effect of granule size on settling velocity
SUMMARY
Overall, packing enhanced the fermentative hydrogen production in upflow reactors. The maximum
hydrogen production rates were 6.17 L/L/d in the reactor filled with packing and 5.35 L/L/d in the reactor
without packing. The maximum hydrogen yields were 1.21 mol/mol-sucrose for the former reactor and
1.10 mol/mol-sucrose for the latter. In both reactors, sludge agglutinated into granules resulting in the
increase of biomass concentration. The granules had an average porosity of 54% and a settling velocity of
10-37 mm/s.
5.
ACKNOWLEDGEMENTS
The authors wish to thank the Hong Kong Research Grant Council for the financial support of this project
(HKU7007/02E) and Chenlin Li wishes to thank HKU for the postgraduate studentship..
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Fang, H.H.P. and Liu H. (2002), “Effect of pH on hydrogen production from glucose by a mixed
culture,” Bioresource Technol, 82(2), 87-93.
Fang, H.H.P., Liu, H. and Zhang, T. (2002), “Characterization of a Hydrogen-Producing Granular
Sludge,” Biotechnol Bioeng, 78, 44-52.
Lay, J.J. (2000), “Modeling and optimization of anaerobic digested sludge converting starch to
hydrogen,” Biotechnol Bioeng 68(3), 269-278.
Ueno, Y., Haruta, S., Ishii, M. and Igarashi, Y. (2001), “Microbial community in anaerobic
hydrogen-producing microflora enriched from sludge compost,” Appl Microbial Biotech, 57(4),
555-562.
Ueno, Y., Otsuka, S. and Morimoto, M. (1996), “Hydrogen production from industrial wastewater by
anaerobic microflora in chemostat culture,” J Ferment Bioeng, 82(2), 194-197.
Noike, T., Ko, I.B., Lee, D.Y. and Yokoyama, S. (2003), “Continuous hydrogen production from
organic municipal wastes,” Proceedings of the 1st NRL International Workshop on Innovative
Anaerobic Technology, Daejeon, Korea. Dec 12, 2003, 53-60.
Wu, S.Y., Lin, C.N., and Chang, J.S. (2003), “Hydrogen production with immobilized sewage sludge
in three-phase fluidized-bed bioreactors,” Biotechnol Prog, 19, 828-832.
Chang, F.Y. and Lin, C.Y. (2004), “Biohydrogen production using an up-flow anaerobic sludge
blanket reactor,” Int J Hydrogen Energy, 29, 33-39
Cheng, S.S., Lin, C.Y., Tseng, I. C., Lee, C.M., Lin, H.I., Lin, M.R., Chen, S.T., Chen, S.D. and Liu,
P.W. (2003), “Biohydrogen production mechanisms and processes application on multiple substrates,”
Proceedings of the 1st NRL International Workshop on Innovative Anaerobic Technology, Daejeon,
Korea. Dec 12, 2003, 33-39.
Gaudy, A.F. (1962), “Colorimetric determination of protein and carbohydrate,” Ind. Water Wastes, 7,
17-22.
APHA, AWWA, WEF, (1992). Standard Methods for the Examination of water and Wastewater, 18th
ed. American Public Health Association, Washington, DC (Chapter 2).