Anaerobic bioconversion of carbon dioxide to biogas in an upflow anaerobic sludge blanket reactor.

ABSTRACT

The increasing concentration of carbon dioxide (C[O.sub.2])--the most dominant component of greenhouse gases--in the atmosphere has been of growing concern for many years. Many methods focus on C[O.sub.2] capture and storage and there is always the risk of C[O.sub.2] release to the environment. In this study, a new method to convert C[O.sub.2] to biogas with a high content of methane (C[H.sub.4]) in an anaerobic system with a lab-scale upflow anaerobic sludge blanket reactor at 35 [degrees]C was developed. In a series of experiments, the reactor was run with and without C[O.sub.2]-saturated solutions including volatile fatty acids (VFAs) as sources of hydrogen. The concentration of dissolved C[O.sub.2] in the influent solutions was 2.2-6.1 g/L, with corresponding chemical oxygen demand (COD) values of 2.6-8.4 g/L for the solutions. Overall C[O.sub.2] removal values of 2.7-20 g/day (49-88% conversion) were obtained for the organic loading rates (OLR) and C[O.sub.2] loading rates of 8-36 gCOD/L x day and 6-26 gC[O.sub.2]/L x day, respectively with C[H.sub.4] purity of above 70%. Also, VFA and COD removal were in the range of 79-95% and 75-90%, respectively. Methanogenic activities of the cultures with the concentrations measured as volatile suspended solids (VSSs) were 0.12-0.40 L C[H.sub.4]/gVSS x d with the highest value for the system containing acetic acid. This anaerobic method can be applied to reduce C[O.sub.2] emitted to the atmosphere from a wide variety of industrial point sources with a value-added product, C[H.sub.4].

INTRODUCTION

According to the U.S. Environmental Protection Agency (EPA), (1) since the beginning of the industrial revolution, atmospheric concentrations of carbon dioxide (C[O.sub.2]) have increased nearly 30%. The mean annual concentration of C[O.sub.2] in the atmosphere has increased from 315.98 ppmv of dry air in 1959 to approximately 375.64 ppmv in 2003. (2)

There have been many attempts aimed at reducing C[O.sub.2] emissions in the form of C[O.sub.2] sequestration through its injection in the underground waters, e.g., saline waters, aquifers, or deep oceans. (3-5) However, in all of these processes, C[O.sub.2] is transferred from one place to another and there is always the risk of C[O.sub.2] release to the atmosphere again.

C[O.sub.2] can be converted to methane (C[H.sub.4]) either chemically or biologically. Some experiments in microchemical catalytic reactors at 250 [degrees]C have reached 90% C[O.sub.2] conversion. (6) Biological conversion of C[O.sub.2] into sparingly soluble carbonate minerals such as calcite (CaC[O.sub.3]) and siderite (FeC[O.sub.3]) has been studied using Fe(III)-reducing bacteria in conjunction with metal containing fly ash and lime. (7) In this process, fly ash is stabilized into carbonate solid conglomerates that could potentially be useful as fill materials or road construction aggregates. Examples of biological methods to reduce C[O.sub.2] emissions in power plants have been photosynthetic systems with cyanobacteria or microalgae (8,9) and bioelectro methods. (10) In the method developed by Otaguchi et al., (9) C[O.sub.2] was converted to biofuel, [H.sub.2], and ethanol with different cultures. They used a strain of cyanobacteria cultivated on 6% C[O.sub.2] in an air stream and obtained a maximum [H.sub.2] production rate of 0.13 mmol/[dm.sup.3] x hr at 35 [degrees]C.

Lombardi (11) compared three C[O.sub.2] low-emission power cycles using a life cycle assessment procedure. These combined cycles include application of chemical methods, e.g. chemical absorption, synthesis gas treatment, or the use of gas liquefaction units. For these systems, values of 85% C[O.sub.2] removal efficiency have been reported. In another study, efficiency values of 85-99.5% were reported for C[O.sub.2] removal in iron and steel industry (12) using special absorbents like Selexol (dimethylether of polyethylene glycol), in which the captured C[O.sub.2] was then compressed and transferred for storage, e.g., to aquifers or oceans.

Another biological process with the potential to remove C[O.sub.2] is anaerobic treatment. In the final step (methanogenesis), simple organic molecules including short-chain fatty acids along with C[O.sub.2] and hydrogen are converted to biogas. Therefore, it is possible to simulate this step and provide conditions to convert C[O.sub.2] to biogas using methanogens. Methanogenic archaea are obligate anaerobes. In fact, they are the strictest anaerobes discovered. (13) Most methanogens can grow using hydrogen as a source of electrons via hydrogenase. In many methanogenic environments, [H.sub.2] is utilized rapidly even when it is present at very low concentrations. (14)

Developed at Wageningen Agricultural University, The Netherlands, (15) an upflow anaerobic sludge blanket (UASB) reactor uses anaerobic bacteria, especially methanogens, which form self-immobilized granular structures in a "blanket" with good settling properties inside the reactor. These anaerobic bacteria granules form a blanket through which the effluent flows up through the reactor. Because of the high biomass concentrations of these reactors, they can achieve conversions higher than possible by conventional anaerobic processes and can tolerate fluctuations in influent feed, temperature and pH. (16) Moreover, because no support medium is required for the biomass, it decreases the capital cost and minimizes the possibility of plugging. The energy requirement also is small because there is no mechanical mixing within the reactor, no recirculation of sludge, and no recirculation of the effluent. (17)

The main objective of this study was to develop a new biological method to convert gaseous C[O.sub.2] to biogas with a high- C[H.sub.4] content using a wastewater treatment technology. For this purpose, gaseous C[O.sub.2] was first dissolved in a medium. The C[O.sub.2] substrate was then treated anaerobically in a UASB reactor in combination with fatty acids as sources of hydrogen to convert to biogas. Parameters such as dissolved and removed C[O.sub.2] and biogas purity and rate were measured for two sets of experiments: with and without C[O.sub.2] using acetic, propionic, and butyric acids.

MATERIALS AND METHODS

Inoculum

The granulated biomass used in the system was collected from the multiplate anaerobic reactor that was developed by SNC-Lavalin (Montreal, Canada) for the treatment of cheese water at a cheese factory in Chambord, Quebec. (18) The sludge was first kept in the incubator at 35 [degrees]C for temperature acclimation. Subsequently, it was acclimated in the reactor with solutions including acetic acid and sodium acetate.

Materials

A C[O.sub.2] gas tank purchased from Praxair Inc. with an industrial purity of 99% was used as the source of C[O.sub.2] Before making the substrates, the water was purged with nitrogen to minimize dissolved oxygen and chlorine.

Substrate solutions were made based on methanogenic media, (19) and volatile fatty acids (VFAs) composed the main constituents of the solutions. C[O.sub.2] was injected into the influent storage tank using a gas sparger. To achieve the maximum dissolved C[O.sub.2], alkalinity was added using a 1 N KOH solution and a positive pressure of C[O.sub.2] (1.01 x [10.sup.5] Pa) was maintained in the tank to keep the saturation condition.

The necessary components and elements for the cell growth were included in the solutions. Ammonium chloride (crystalline 99.5%) and potassium hydrogen phosphate (100% dry basis), purchased from Fisher Scientific Ltd., served as the source of nutrients, and other inorganic components and trace mineral solutions were also included.

Table 1 shows the amounts of these components per 1 L of solution (all chemicals were supplied from Fisher Scientific Ltd.). The composition of the solution containing trace elements necessary for bacterial growth is shown in Table 2. This solution was made in 1-L quantities and each time a certain amount (10 mL) was added to the main solution according to Table 1. Other components used in the solutions were peptone and yeast extract that contained vitamins, amino acids, and coenzymes.

The amounts of the acids, sodium acetate, peptone, and yeast extract (supplied from Fisher Scientific Ltd.) and the approximate chemical oxygen demand (COD) values of the solutions are shown in Table 3. The adjustments were made for the flow rate and concentration for each loading rate to avoid possible channeling in the reactor and maintaining recommended upflow velocities. (20,21) A 1 N KOH solution (from 86.9% solid KOH) was prepared and added to adjust the pH of the solutions.

ANALYTICAL METHODS

Dissolved C[O.sub.2] concentration in the reactor influent and effluent was measured using the alkalinity and pH values based on the equilibrium relationships among carbonate species, hydrogen, and hydroxyl ions. It was also measured in the influent by measuring the amount of injected C[O.sub.2] using a gas-flow meter and the injection time to reach a constant pH.

For dissolved C[O.sub.2](aq), the equilibrium reactions are as follows:

[H.sub.2]O + C[O.sub.2](aq) [left and right arrow] [H.sup.+] + HC[O.sub.3.sup.-]

[K.sub.1] = [[H.sup.+]][HC[O.sub.3.sup.-]]/[C[O.sub.2](aq)] = 4.47 x [10.sup.-7] M (1)

HC[O.sub.3.sup.-] [left and right arrow] [H.sup.+] + C[O.sub.3.sup.2-]

[K.sub.2] = [[H.sup.+]][C[O.sub.3.sup.2-]]/[HC[O.sub.3.sup.-]] = 4.68 x [10.sup.-11] M (2)

By combining two equations, the following is obtained:

[K.sub.1][K.sub.2] = [[[H.sup.+]][.sup.2]][C[O.sub.3.sup.2-]]/[C[O.sub.2](aq)] = 2.1 x [10.sup.-17] (3)