Volume 2 Issue 10 - January 4, 2008
Continuous hydrogen production by anaerobic mixed microflora using a hollow-fiber microfiltration membrane bioreactor
Kuo-Shing Lee1, Ping-Jei Lin2, Kai Fangchiang2 and Jo-Shu Chang3*

1Department of Safety Health and Environmental Engineering, Central Taiwan University of Science and Technology, Taichung, Taiwan
2Department of Chemical Engineering, Feng Chia University, Taichung, Taiwan
3Department of Chemical Engineering, National Cheng Kung University, Tainan, Taiwan
*E-mail: changjs@mail.ncku.edu.tw

Paper published in International Journal of Hydrogen Energy 32, 950-957 (2007)

The global shortage of fossil fuel and the severe environmental pollution derived from using fossil fuels have forced the developed countries in the world to search for alternative or new energy sources. Hydrogen has emerged as one of the most promising new energy carriers because it is clean, recyclable, efficient, and can be used in fuel cells to generate electricity. At present, hydrogen is mainly produced from fossil fuels (Das and Veziroglu, 2001; Rosen and Scott, 1998) via chemical or thermochemical methods, while a cost-effective and pollution-free means (such as biological methods) for hydrogen production is still in great demand. Hydrogen can be generated through natural biochemical pathways, such as photosynthesis and fermentation. In green algae or cyanobacteria, hydrogen is formed through photolysis of water, followed by an electron transfer process catalyzed by hydrogenase or nitrogenase. Hydrogen can also be generated via fermentative conversion of organic substrates through a metabolic route that is either light-dependent (photofermentation; e.g., photosynthetic bacteria) or light-independent (dark fermentation; e.g., anaerobic bacteria). Among the biohydrogenation processes, dark fermentation seems more feasible for practical applications since it is able to convert various waste organic substances into H2 and attain higher H2 production rates than the rest of biohydrogen systems.

Continuous dark fermentative H2 production is usually conducted via a continuous flow stirred tank reactor (CSTR) because it is easy to operate and can provide a good substrate-biomass contact by vigorous mixing.  However, poor cell retention in CSTR occurs when it is operated at a high dilution rate (i.e., low hydraulic retention time, HRT), resulting in poor H2-producing efficiency due to cell washout. To upgrade bioH2 production performance, it is required to develop H2 production systems able to retain sufficient active H2 producing cells in the reactor against high organic loading and hydraulic pressure.  The strategies to enhance cell retention for an increased H2 production include matrix entrapment of cells, self-flocculation of cells, biofilm formation, or cell granulation. Moreover, membrane restriction of cells is also feasible way of improving cell retention.

Compared with other cell retention approaches (e.g., cell immobilization), utilizing membrane separation to retain biomass would not cause mass transfer limitation and is thus of great interest in practical application. Recent advances in membrane technology allow reduction of the cost of membrane and also provide solutions to handle the problem of membrane fouling, such as addition of coagulants, crossflushing, imposing a pulsed electric field, backwashing, and rapid backpulsing. Therefore, in this work, we utilized membrane bioreactors (MBRs) to increase biomass retention to improve H2 production efficiency from organic substrates.  In fact, we demonstrated one of the early attempts of using MBR for fermentative H2 production.

Figure 1        Schematic description of the membrane bioreactor (MBR) system.
A membrane bioreactor (MBR) fabricated by connecting a hollow-fiber microfiltration membrane module with a continuous flow stirred tank reactor (CSTR) was used to enhance H2 production through high-dilution rate operations.  The hollow fiber membrane module had a pore size of 0.2 μm, a porosity of 70%, and an effective filtration area of 0.1 m2.  The continuous culture was started up at a hydraulic retention time (HRT) of 8-12 h without using the microfiltration membrane.  The fermentation was carried out at a temperature of 35˚C and an agitation rate of 150 rpm.  The pH was set at 6.7 initially and maintained within the range of 6.2-6.8 throughout the experiments.  After a steady state was reached, the HRT was shifted down until washout of cells was observed.  While cell washout occurred, the liquid effluent of the bioreactor was bypassed to the microfiltration membrane module, allowing recycle of the cells into the fermentor.
ure 2Fig. 2        The effect of hydraulic retention time (HRT) on (a) H2 production rate, (b) biomass concentration, and (c) H2 yield during CSTR and MBR operations using glucose, sucrose and fructose (20 g COD/l) as the sole carbon source.

Three different carbon substrates (glucose, sucrose, and fructose) were examined for their effectiveness in H2 production with a mixed microflora obtained from a municipal sewage treatment plant located in central Taiwan.  In CSTR operation, cell washout occurred at a hydraulic retention time (HRT) of 2-4 h.  Using MBR could avoid cell washout, leading to a substantial increase in both H2 production rate (HPR) and biomass concentration.  The MBR system was very effective in retaining biomass within the reactor as the system can be stably operated at an extremely low HRT of 1 h with an optimal steady-state HPR of 1.48, 2.07, and 2.75 l/h/l, respectively, for using glucose, sucrose, and fructose as the sole carbon source.  Meanwhile, despite operation at a high dilution rate (i.e., HRT=1 h), the H2 yield (HY) could be maintained at a high level of 1.27, 1.39, and 1.36 mol H2/mol hexose, for glucose, sucrose, and fructose, respectively.

Figure 3        Comparison of experimental results and estimated values for (a) H2 production rate and (b) H2 yield. (HPRexpt and HYexpt denote molar H2 production rate and H2 yield obtained from experimental data; HPRest and HYest denote molar H2 production rate and H2 yield estimated from stoichiometric correlation between H2 production and formation of butyrate, acetate, and propionate)
The HPR tended to decrease in the order of fructose > sucrose > glucose. Thus, fructose seems to be the most efficient carbon substrate for H2 production with the H2-producing mixed culture used in this work.  Butyrate (HBu) and acetate (HAc) (especially, HBu) were the major soluble metabolites in all cases, contributing to 70-85% of total soluble microbial product.  The experimental results of HPR and HY can be estimated by production of the two major metabolites (HBu and HAc) and a minor but H2-consuming metabolite (HPr) based on stoichiometric correlation between soluble metabolites formation and H2 production. The estimated values are in good agreement with the experimental results, indicating good consistency between H2 production and metabolite formation.

This study demonstrated the feasibility of using MBR system for the enhancement of H2 production.  Switching CSTR to MBR mode at a HRT of 2-4 h effectively avoid cell-washout, leading to a dramtic increase in both H2 production rate (HPR) and biomass concentration.  The proposed MBR system can be operated at an extremely low HRT (i.e., HRT= 1 h), giving a high HPR with a satisfactory level of H2 yield (HY).  Therefore, it is evident that the MBR system was effective in fermentative H2 production from the three carbohydrates tested.  The HPR appeared to be carbon substrate-dependent as fructose gave the highest HPR.  The outcome of this work seems to substantiate the concept that using the MBR system to retain active biomass at high organic loading rate might improve the performance of H2 production.  However, the effect of cell aging during the prolonged retention of cells in the MBR system should be further investigated. Meanwhile, the commercial viability of using MBR system for H2 production may also require more comprehensive assessment.
< previousnext >
Copyright National Cheng Kung University