Acetic acid (CH3COOH) and ethanol (C2H5OH) are the two major products from syngas fermentation. Equations (1)-(4) show the four basic reactions producing acetic acid and ethanol. In this case, the gaseous substrates CO and H2 follow the acetyl-CoA pathway to produce acetic acid and ethanol under strict anaerobic conditions.

From Equation (1), it is clear that about one third of the carbon from CO is utilized in the product yield. The overall ethanol production, combining Equations (1) and (2), reveals that two thirds of the carbon from CO is converted to ethanol. During the acetyl-CoA pathway, hydrogen provides the required reducing equivalents and electrons when hydrogenase enzyme is present in the fermentation media (Equation (5)).

H2 ! 2H+ + 2e — (5)

If the hydrogenase enzyme is inhibited or hydrogen is not present in the fermentation broth, the required electrons are obtained from CO in the presence of CODH enzyme. In other words, CO is used in supplying electrons, rather than in the biofuel production. This obvi­ously results in a drastic reduction in biofuel yields. It is therefore, important to maintain healthy concentrations of both hydrogen and CO during syngas fermentation.

Syngas-fermenting microorganisms are critical in biofuel production. Under optimum growth conditions, most of the known syngas-fermenting microbes tend to produce more acetate than alcohol products (e. g., ethanol, butanol, etc.). Vega et al. (1989) reported acetate to ethanol product ratio of 20:1. In order to improve the product formation from acidogenesis to solventogenesis, researchers investigated nutrient limitations, pH shifts, reducing agent addition (Klasson et al., 1992), and hydrogen addition.

Pyrolysis is the fundamental chemical reaction process that is the precursor of both the gasification and combustion of solid fuels, and is simply defined as the chemical changes occurring when heat is applied to a material in the absence of oxygen. The products of biomass pyrolysis include water, charcoal (carbonaceous solid), pyrolysis oils or tars, and permanent gases including methane, hydrogen, carbon monoxide, and carbon dioxide. The nature of the changes in pyrolysis depends on the material being pyrolyzed, the final temperature of the pyrolysis process, and the rate at which it is heated up. The pyrolysis pro­cess is a mildly endothermic reaction. The heat of vaporization of pure water is 2.26 KJ g-1 at 100 °C, while the chemical energy content of wood is only about 18.6 KJ g-1. Most of the energy obtained from biomass goes in moisture removal. This reinforces the facts that lower the moisture content, greater is the energy obtained.

As typical lignocellulosic biomass materials such as wood, straws, and stalks are poor heat conductors, management of the rate of heating requires that the size of the particles being heated be quite small. Otherwise, in massive materials such as logs, the heating rate is very slow, and this determines the yield of pyrolysis products. Depending on the thermal environment and the final temperature, pyrolysis will yield mainly char at low temperatures, <450 °C, when the heating rate is quite slow, and mainly gases at high temperatures, >800 °C, with rapid heating rates. An intermediate temperature and under relatively high heating rates, the main product is a liquid bio-oil, a relatively recent discovery, which is just being turned to commercial applications. There are 3 stages in the pyrolysis process: The first stage, prepyrolysis, occurs between 120 and 200 °C with a slight observed weight loss, when some internal rearrangements, such as bond breakage, the appearance of free radicals, and the formation of carbonyl groups take place, with a corresponding release of small amounts of water (H2O), carbon monoxide (CO), and CO2. The second stage is the main pyrolysis pro­cess, during which solid decomposition occurs, accompanied by a significant weight loss from the initially fed biomass. The last stage is the continuous char devolatilization caused by the further cleavage of C—H and C—O bonds.

In reacting chemical systems, the term severity is used to capture the idea that both the duration of heating and the final temperature influence the chemical products of pyrolysis. Very-low-severity treatments of short duration to a maximum temperature of about 250 °C are sometimes called torrefaction and result in a product that has lost some H2O and CO2 from pyrolysis while retaining almost all of the heat value. Traditional charcoaling is a medium-severity process, while the production of bio-oils is a short-duration high-severity process, which, if the duration at high temperature is maintained, will go all the way to gas and soot.

Depending on the reaction temperature and residence time, pyrolysis can be divided into fast pyrolysis, intermediate pyrolysis, and slow pyrolysis. Typically, fast pyrolysis has an extremely short residence time (~1 s); the reaction temperature is approximately 100 °C higher than that of slow pyrolysis (e. g. ^500 °C vs. ^400 °C). Short reaction times com­bined with an elevated temperature generally result in a higher yield of liquid product. A conventional moderate or slow pyrolysis process, with a relatively long vapor residence time and low heating rate, has been used to produce charcoal for thousands of years (Zhang et al., 2010).

Among the short residence-time processes (0.5-5 s) under development are vacuum pyrolysis at about 300-400 °C and 0.3 atm (U. of Sherbrooke, Canada), flash pyrolysis at about 500-650 °C and 1 atm (U. of Waterloo, Canada), hydropyrolysis in an atmosphere of hydrogen at about 500-600 °C and 5-6 atm (HYFLEX TM, IGT), and flash pyrolysis in atmospheres of hydrogen or methane at 600-1000 °C and 1-70 atm (Brookhaven National Laboratory). An interesting report of a relatively long residence time (10-15 min heat up, several hours at temperature) pyrolysis study at reduced pressures of 0.0004-0.004 atm and temperatures of 250-320 °C of wild cherry wood seems to contrast with the results of several reports on flash pyrolysis (http://journeytoforever. org/biofuel_library/liquefaction. html).

In microbial syngas fermentation, the gaseous substrates, such as CO and H2, require trans­port from gas phase to the cell surface (Vega et al., 1990). In that case, the gaseous substrate is first absorbed at the gas-liquid interface and then diffused through the culture media to the cells. Microbes consume the diffused substrates as their carbon and energy sources and pro­duce the metabolites such as biofuel and other byproducts.

There are several intermediate steps involved in transporting substrate gases into the microbial cells. These steps include the diffusion through the bulk gas to the gas-liquid interface, moving across the gas-liquid interface, transport into the bulk liquid surrounding the microbial cells, and the diffusive transport through the liquid-solid boundary. Out of these, the gas-liquid inter­face mass transfer is the major resistance for gaseous substrate diffusion (Klasson et al., 1992).

Poor solubility of a gaseous substrate in the culture media results in low substrate uptake by microbes and, thus, leads to low productivity. The volumetric mass transfer coefficient (kLa) is often used to quantify the solubility of a gas in the liquid phase. Klasson et al. (1992) proposed the following equation (Equation (10)) to calculate the mass transfer coeffi­cient (kL) in the liquid phase.

where N(?(mol) is the molar substrate transferred from the gas phase, VL (L) is the volume of the reactor, PG and Pi (atm) are the partial pressures of the gaseous substrate in gas and the liquid phase, H (Latm/mol) is Henry’s law constant, and a (m2/L) is the gas-liquid interfacial area for unit volume.

The difference in the partial pressures of the gaseous substrate (PG — P| ) is the driving force for mass transfer and thus controls the solubility of the substrate. High-pressure opera­tion improves the solubility of the gas in aqueous phase. However, at higher concentrations of gaseous substrates, especially CO, anaerobic microorganisms are inhibited. Therefore, the determination of a correlation between the substrate diffusion and the specific substrate uptake rate (qs) (h-1) is important in order to evaluate the process kinetics (Equation (11)).

where qm (h-1), W (atm) and Kp(atm) are empirical constants. Furthermore, Qs (mol/L h), the substrate uptake rate, can be written as Qs = qsX, where X (mol/L) is the microbial cell concentration. By comparing Equations (10) and (11), it is evident that the difference between the partial pressures of the substrate gases and the cell concentration of the reactor is directly proportional (Vega et al., 1990).

Different approaches such as high gas and liquid flow rates, large specific gas-liquid interfacial areas, increased pressure, different reactor configurations (Munasinghe and Khanal, 2010 (b)), innovative impeller designs, modified fluid flow patterns, varying mixing times and speeds, and the use of microbubble dispersers have been examined to enhance gas solubility in the liquid phase. Many of these approaches increase the agitator’s power input to volume ratio which facilitates bubble breakup, and increases the interfacial surface area available for mass transfer. This approach, however, is not economically attrac­tive for commercial syngas fermentation due to high energy costs. Additionally, higher power inputs can also damage the sensitive microorganisms in the culture media. In order to achieve energy efficient mass transfer, alternative bioreactor configurations such as trick­ling beds and airlift reactors were examined for syngas fermentation (Bredwell et al., 1999; Munasinghe and Khanal, 2010 (b)).

Yang et al. (2001) reported kLa values of 54 and 119 h-1 for H2 and CO gases, respectively, in a slurry bubble column operated at temperature of 20 °C and pressure of 10 bar. In a separate study, Bredwell et al. (1999) reported a maximum kLa of 190 and 75 h-1 for H2 gas in a stirred — tank reactor at a mixing speed of 300 rpm with and without microbubble sparging, respec­tively. The authors used a mixed culture of sulfate-reducing bacteria (SRB) in their study. Munasinghe and Khanal (2010 (b)) reported kLa values for CO, ranging from 0.4 to 91 L/h for eight different reactor configurations including a submerged composite hollow fiber membrane (CHFM) reactor. Some of the reported values of kLa for different reactor configurations under various hydrodynamic conditions are shown in Table 2.

Hydrothermal liquefaction is the conversion of solid biomass into gaseous and/or liquid products in the presence of water. Liquefaction consists of the catalytic thermal decomposi­tion of large molecules to unstable shorter species that polymerize again into a bio-oil.

Biomass is mixed with water and basic catalysts like sodium carbonate, and the process is carried out at lower temperatures than pyrolysis (252-472 °C) but higher pressures (50-150 atm) and longer residence times (5-30 min.). These factors combine to make lique­faction a more expensive process; however, the liquid product obtained contains less oxygen (12-14%) than the bio-oil produced by pyrolysis and typically requires less extensive processing (Elliott, 2007; Inoue et al., 1999; Karagoz et al., 2006; Kruse et al., 2003; Minowa et al., 1997).

Microorganisms can be categorized into two major types depending on their optimum growth temperature, namely, mesophilic and thermophilic. In general, optimum growth tem­perature for mesophilic microorganisms varies from 37 to 40 °C, whereas for thermophiles, the temperature range varies from 55 to 80 °C (Munasinghe and Khanal, 2010 (a)). Mesophilic microorganisms, for example, Clostridium aceticum, Acetobacterium woodii, C. carboxydivorans,

and C. Ijungdahlii, have been dominated in syngas fermentation with higher syngas to biofuel conversion efficiencies compared to the thermophilic counterpart (Henstra et al., 2007; Younesi et al., 2005).

C. Ijungdahlii, one of the most widely used homoacetogenic microorganism, has an opti­mum growth temperature between 37 and 40 °C, and a pH of 5.8-6.0 (Tanner et al., 1993). However, C. Ijungdahlii was reported producing an ethanol concentration as high as 48 g/L in a continuous stirred-tank reactor (CSTR) at a low pH of 4.0-4.5 in nutrient-limited culture media (Klasson et al., 1993). C. carboxydivorans (earlier referred to as bacterium strain P7) is another mesophilic organism which has an ability to grow on a gas mixture consists of CO, CO2, H2, and N2, and produces mainly ethanol and acetic acid. The ethanol yield obtained was 0.16% (by weight) during a 10-day fermentation experiment at pH of 5.75 and tempera­ture of 37 °C in a bubble column reactor (Rajagopalan et al., 2002). Later, the authors com­pared the ethanol yields of C. carboxydivorans and C. ljungdahlii and found out that the results were similar. In a separate study, Heiskanen et al. (2007) used B. methylotrophicum at 37 °C and a pH of 6.0-6.9 to convert syngas to acetic acid. The authors claimed a maximum acetic acid concentration of 1.3 g/L at a gas mixture of 40% H2,35% CO and 25% CO2 after 144 h of fermentation.

The major advantage of thermophilic microbes in syngas fermentation is their capability of fermenting syngas at relatively high temperatures (around 60 ° C) with higher conversion rates. Though the high temperature reduces the solubility of the component gases in the fer­mentation broth, benefit in product recovery improves the overall cost effectiveness of the process (Henstra et al., 2007). There were several attempts to utilize extreme thermophiles (optimum growth temperature >70 °C) in producing organic solvents. During the last decade, researchers were able to isolate several thermophilic microorganisms which were able to grow on CO as a substrate (Henstra et al., 2007). Desulfotomaculum carboxydivorans, Carboxydocella sporoproducens, Moorella thermoacetica, and M. thermoautotrophica are some examples of thermophiles with optimum temperature ranges of 55-58 °C.

Heating of the lignocellulosic biomass in inert atmosphere for hours to a maximum temperature of 400-500 °C is called slow pyrolysis. The charcoal yield is 35-40% by weight. In general, the yield of liquid products would be less than the fast pyrolysis of biomass. Several types of catalysts can be employed for the pyrolysis of biomass and/or upgradation of the vapors produced from the thermal pyrolysis.

2.1.4 Fast Pyrolysis

The goal of fast pyrolysis is to produce liquid fuel from lignocellulosic biomass that can substitute for fuel oil in any application. The liquid can also be used to produce a range of specialty and commodity chemicals. The essential features of a fast pyrolysis process are very high heating and heat transfer rates, which often require a finely ground biomass feed. Carefully controlled reaction temperature of ca. 500 °C in the vapor phase and residence time of pyrolysis vapors in the reactor less than 1 s; and then quenching (rapid cooling) of the pyrolysis vapors to give the bio-oil product. The main product of fast pyrolysis is bio-oil, which is obtained in yields of up to 80 wt% of dry feed.

Fast pyrolysis is a promising process to produce transportable oil with a high volumetric energy density from bulky and inhomogeneous biomass. There are several applications foreseen for pyrolysis oil. It has been tested as a substitute for fuel oil or diesel in boilers, furnaces, engines, and turbines for heat and power generation and has been considered as a precursor for transportation fuels and chemicals. Water is the most abundant component in pyrolysis oil; typically, it is present in the range of 15-35 wt%. Probably all applications require different specifications with respect to the water content of pyrolysis oil. For fueling into a diesel engine, the water content should be below 30 wt% to decrease emissions of particles and to prevent ignition delay and phase separation. But there should also be a mini­mum amount of water present to limit NOx emissions and to ensure a uniform temperature distribution in the cylinders. For cofeeding pyrolysis oil in a mineral oil refinery, nearly all water and most organically bound oxygen must be removed. Generally, less water is beneficial for the energy density, transportation costs, stability, and acidity of pyrolysis oil. Fast pyrolysis oil possesses many undesirable properties including a high total acid number (TAN ^200), low heating value (^6560 BTU/lb), high oxygen content (^40%), chemical instability, high water content (20%), and incompatibility with petroleum fractions. Inherent low-energy density makes pyrolysis oil expensive to transport, and the high TAN makes it metallurgically incompatible with conventional transport vessels and refinery hydro­conversion equipment, both designed for feeds with TANs less than 2. In addition to these undesirable properties, pyrolysis oil is not miscible with petroleum fractions and if added into existing refinery equipment (hydrotreaters or hydrocrackers) will require a separate pyrolysis-oil feed system. Thus, pyrolysis oil needs effective pretreatment and upgradation before it is used as crude oil replacement. Depending on the reactors used, we have many kinds of fast pyrolysis processes.

Bioreactor configuration is closely related to the effective gas-liquid mass transfer. Thus, reactor design plays an important role in syngas fermentation. High mass transfer rates, high cell densities, low operation and maintenance costs, and easy scale-up are some of the key parameters for designing an efficient bioreactor system. Similarly, the bioreactor size greatly depends on the rate of mass transfer for sparingly soluble gases (Klasson et al., 1992; Vega et al., 1990). Some of the commonly used reactor configurations are discussed in the following section.

Currently, more research is being done on direct and indirect thermal liquefaction methods for biomass and wastes than on the other methods. Direct liquefaction is either reaction of bio­mass components with smaller molecules such as H2 and CO (e. g., Pittsburg Energy Research Centre (PERC) and Lawrence Berkeley Laboratory, Berkeley, USA (LBL) processes) or short­term pyrolytic treatment, sometimes in the presence of gases such as H2. Indirect liquefaction involves successive production of an intermediate, such as synthesis gas or ethylene, and its chemical conversion to liquid fuels, In 1983, after several years of laboratory and pilot-plant work on the PERC and LBL processes, which involve reaction of product oil or water slurries of wood particles with H2 and CO at temperatures up to about 370 °C and pressures up to 272 atm in the presence of sodium carbonate catalyst, researchers concluded that neither process can be commercialized for liquid fuel production without substantial improvement. The most attractive approach to such improvement is believed to be a combination of solvolysis with a pyrolysis or reduction step. However, the oxygen content of the resulting complex liquid mix­ture is still high (6-10 wt%), and considerable processing is necessary to upgrade this material (http://journeytoforever. org/biofuel_library/ liquefaction. html).

Direct liquefaction has some similarity with pyrolysis in terms of the target products (liquid products). However, they are different in terms of operational conditions. Specifi­cally, direct liquefaction requires lower reaction temperatures but higher pressures than pyrolysis (0.5-2 atm for liquefaction vs. 0.01-0.05 atm for pyrolysis). In addition, drying of the feedstock is not a necessary step for direct liquefaction, but it is crucial for pyrolysis. Moreover, catalysts are always essential for liquefaction, whereas they are not as critical for pyrolysis. At the beginning of the liquefaction process, biomass undergoes depolymeriza­tion and is decomposed into monomer units. These monomer units, however, may be repolymerized or condensed into solid chars, which are undesirable (Zhang et al., 2010).

4.1 Syngas Composition

Gasification of biomass produces a gas mixture consisting of CO, CO2, H2, N2, CH4, trace amounts of NOx and SOx, tar, char, particulate matter, and higher hydrocarbons such as C2H2, C2H4, and C2H6. This gas mixture is sometimes referred to as a producer gas (Datar et al., 2004). The gas composition greatly depends on the type of feedstocks and the operating conditions of gasifier (Klasson et al., 1992). Table 1 summarizes the gas constituents from various types of gasifiers and feedstocks. Datar et al. (2004) used a pro­ducer gas consisting of N2 (56.8%), CO (14.7%), CO2 (16.5%), H2 (4.4%), CH4 (4.2%), C2H4 (2.4%), and C2H6 (0.8%) to determine the effects of the gas composition on cell growth, H2 uptake, and acid and alcohol production. The authors found that there was a process alter­ation due to some trace species in the producer gas. Further, they suggested that the producer gas without purification could inhibit hydrogenase enzyme leading to low prod­uct formation.

Ablative pyrolysis, in which much larger particle sizes can be employed than in other systems, as the heat is transferred from a hot surface to the biomass particle and the process, is limited by the rate of heat supply to the reactor rather than the rate of heat absorption by the pyrolyzing biomass. Ablative pyrolysis is fundamentally different from fluid bed processes from the mode of heat transfer through a molten layer at the hot reactor surface, use of large particles, and absence of a fluidizing gas.