Uncontrolled burning of biomass in the open air, as encountered in slash-and — burn agriculture, sugarcane trash burning, forest fires, and the clearing of forest land for development, emits large numbers of pollutants to the atmo­sphere. Some 345 chemicals have been identified as being released to the atmosphere from such fires (Khalil and Rasmussen, 1995; Graedel, Hawkins, and Claxton, 1986). Many of these compounds are greenhouse gases and ozone-depleting compounds that are believed to affect the earth’s climate. Some national, state, and local governments have enacted legislation or regulations to attempt to limit large-scale burning of biomass in open air, but most of what occurs is beyond human control. Some local governments have prohibited or limited even small-scale biomass burning such as leaf burning.

In contrast to uncontrolled biomass burning, regulations exist or are being developed in industrialized countries that apply to controlled burning of bio­mass and other fuels. Many different types of emissions — and ash-removal systems have been installed and operated in large-scale commercial biomass combustion plants to meet national and regional regulations and emissions limits. Worldwide, regulations that govern the use of virgin and waste biomass combustion processes range from nonexistent to very detailed and complex. The United States is in the latter category. The state programs complicate matters further because they are often at odds with federal programs and are sometimes more severe than the national requirements. Some regulatory systems in the United States are designed to meet California’s South Coast regulations, which are more constrictive than those of the U. S. Environmental Protection Agency and are believed to be among the most stringent in the world (с/. Moore and Cooper, 1990). Some of the U. S. federal requirements are reviewed here to illustrate how national mandates can affect commercial biomass combustion technologies. Basically, federal regulations that affect biomass combustion are divided into two groups—one concerned with emis­sions and one concerned with residue or ash disposal.

The development of U. S. federal pollution policies began in 1970 when the U. S. Environmental Protection Agency was established. The details of pollution legislation have been undergoing continual revision since then. Included in this legislation is the Clean Air Act of 1970, the New Source Performance Standards of 1971, the Resource Conservation and Recovery Act of 1976, the Pollution Prevention Act of 1990, the Clean Air Act Amendments of 1990, and the Revised New Source Performance Standards of 1991. The Clean Air Act Amendments of 1990 have been called the most complicated and far — reaching environmental legislation ever enacted. Titles II, III, IV, and VI of this Act deal with mobile sources, hazardous air pollutants, acid deposition control, and stratospheric ozone protection, respectively. Title V of this legisla­tion established a program for issuing operating permits to all major sources of air pollution in the United States. A “major source” was defined in Part D of Title I of the Act as having “the potential to emit.” As originally proposed, the standards applied only to “furnaces and boilers” with a heat input of 250 million Btu/h used in the process of burning fossil fuels for the primary purpose of producing steam by heat transfer.

Subsequent revisions expanded the source category to cover some steam generators firing nonfossil fuels and those used in commercial and institutional applications, and quantified emission limits for sulfur and nitrogen oxides and particulates (Dykes, 1989). An air pollution program was formulated for the entire country. The Hazardous and Solid Waste Amendments of 1984 amended the Resource Conservation and Recovery Act of 1976 and set as national policy that “wherever feasible, the generation of hazardous waste is to be reduced or eliminated as expeditiously as possible” (U. S. General Accounting Office, 1994). The Pollution Prevention Act of 1990 broadened the scope of the policy by stating that pollution “should be prevented or reduced at the source whenever feasible” for all environmental media—air, land, and water. A com­prehensive pollution prevention strategy has thus become a national policy of the United States. This strategy affects a multitude of sources of emissions and residues, including virgin and waste biomass combustion systems.

A potential barrier to the adoption of solid biomass fuels is the problem posed by the disposal and utilization of the ash (cf. McGinnis e£ ah, 1995). Users must have a means to dispose of the ash that is cost effective and does not degrade the environment or become a regulatory burden for the ash producer. Soil amendment with the ash would avoid the problems created by landfill disposal, but there is uncertainty regarding the variability of ash composition and the fate of the heavy metals and toxic organic compounds present in the ash. At the federal level, wood ash disposal may be regulated, but certain exemptions may apply, depending on the nature of the facility, the fuel, and the ash (Rughani et ah, 1995). Determination of whether wood ash is to be treated as a hazardous waste is based on the corrosiveness of the ash and whether it exceeds certain threshold concentrations for toxic contaminants as measured by a Toxic Characteristic Leaching Procedure (TCLP). The regulatory concentration limits in the leachate by the TCLP are shown in Table 7.6. The standard for corrosiveness of aqueous ash leachate is a pH of 2 or less or 12.5 or more. Ash exceeding these thresholds must be handled and disposed of in a designated hazardous waste facility (cf. McGinnis et al., 1995). If the ash is not so designated, soil amendment is a preferred disposal option among large producers, whereas landfilling is the preferred option for small producers. Most wood-fueled facilities that burn “clean wood,” or wood that has not been treated with chemicals, and “treated wood,” or wood that has been treated with chemicals, have not experienced problems with ash being designated as hazardous materials. The federal regulations for ash disposal from municipal waste combustion facilities are more complicated (cf. Malloy and McAdams, 1994; Clearwater and Hill, 1989). Because of the composition of municipal wastes and the large variations, strict application of the threshold values in Table 7.6 to the ashes from MSW and RDF combustion facilities would appear to make it more difficult for them to be classified as nonhazardous materials.

“Adapted from McGinnis et al. (1995).

Energy facilities that fire virgin and waste biomass, or that cofire these fuels with fossil fuels, may be subject to some of the emission standards under subpart D of the Code of Federal Regulations, Part 60 (Dykes, 1989). In 1984, the U. S. Environmental Protection Agency expanded the standards, which initially pertained only to fossil-fuel-fired systems, to include all steam genera­tors firing nonfossil fuels in industrial-commercial-institutional steam generat­ing units. Biomass-fired or cofired boilers became subject to selected emission limits under Subpart D. Fuels designated under the standards included fossil fuels, wood, municipal-type solid waste, and chemical by-product fuels.

Because of their complexity, the U. S. environmental laws and revisions of various performance standards for industrial-commercial-institutional steam generating units must be studied in detail to ascertain how they apply to existing
and new systems. With regard to biomass combustion systems specifically, in addition to municipal waste combustors and some biomass-fueled systems which are already subject to pollution control regulations (cf. U. S. Environmen­tal Protection Agency, 1986, 1987, 1995), it appears that both virgin and waste biomass-fueled plants will eventually be covered in depth (cf. U. S. Environmen­tal Protection Agency, 1996). Pollutants to be considered for regulation include CO, S02, NOx, HC1, lead, cadmium, mercury, particulate matter, and dioxins and furans.

A potential route to synthesis gas from biomass is gasification under conditions in which water is in the liquid or fluid phase at elevated temperature and hydrostatic pressure. Exploratory research has been done in a laboratory-scale, plug-flow reactor with solutions of glucose, the monomeric unit of cellulose, in pure water without addition of any potential catalyst (Klass, Kroenke, and Landahl, 1981; Ng, 1979). Some of the results are summarized in Table 9.7. Gasification experiments carried out below the critical temperature for water (374°C) indicated little or no gasification. At temperatures above 374°C, con­version to a relatively clean synthesis gas began to occur, as shown in this table. Char was not observed. Hydrogen yield and concentration in the product gas and the molar ratio of H2: CO exhibited significant increases with increasing temperature. Biomass gasification under these conditions might be expected to offer unique opportunities for homogeneous catalysis at lower capital and operating costs than heterogeneously catalyzed systems.

Heterogeneous catalysts have been found to be effective for the low — temperature, elevated-pressure gasification of 2 to 10% aqueous biomass slur­ries or solutions that range from dilute organics in wastewater to waste sludges

from food processing (Elliott et ah, 1991, 1993). Continuous, fixed-bed cata­lytic reactor systems have been operated on three scales ranging from 0.03 to 33 L/h. The residence time in the supported metallic catalyst bed is less than 10 min at 360°C and 20,365 kPa at liquid hourly space velocities of 1.8 to 4.6 L of feedstock/L of catalyst/h depending on the feedstock. Aqueous effluents with low residual COD (chemical oxygen demand) and a product gas of medium-energy quality have been produced. Catalysts have been demonstrated to have reasonable stability for up to six weeks. Ruthenium appears to be a more stable catalyst than nickel. The product gas contains 25 to 50 mol % carbon dioxide, 45 to 70 mol % methane, and less than 5 mol % hydrogen with as much as 2 mol % ethane. The by-product water stream carries residual organics and has a COD of 40 to 500 ppm. The medium-energy product gas is produced directly in contrast to medium-energy, gas-phase processes that require either oxygen in place of air or the dual reactor system to keep the nitrogen in air separated from the product.

A. The Greenhouse Effect

Since the early 1960s, climate change and air quality have become major and often controversial issues in many countries and among groups from

governments to various scientific communities. Prominent among these issues is the greenhouse effect, in which the gradually increasing tropospheric concen­trations of carbon dioxide (C02), methane (CH4), and nitrous oxide (N20) are believed to trap an excessive amount of solar radiation reflected from the earth. The trapped radiation is predicted to cause significant ambient temperature increases. Other issues include ozone (03) formation over popu­lated areas due to photochemical interactions of hydrocarbon, carbon monox­ide (CO), and nitrogen oxide (NOx) emissions, primarily from motor vehicles; natural ozone layer destruction in the stratosphere by photochemical reactions of organic chlorofluorocarbon compounds (CFCs) resulting in increased pene­tration to the earth’s surface of shorter-wavelength ultraviolet light that can cause skin cancers; and acid rain, which has harmful effects on buildings and the growth of biomass and is caused by sulfur oxide (SOx) emissions from the combustion of sulfur-containing fossil fuels. The predictions of some of the resulting environmental effects are quite dramatic. In the U. S. National Re­search Council’s first assessment of the greenhouse effect in 1979, one of the primary conclusions was that if the C02 content of the atmosphere is doubled and thermal equilibrium is achieved, a global surface warming of between 2 and 3.5°C can occur, with greater increases occurring at higher latitudes (National Research Council, 1979). Some of the earlier predictions indicated that this increase is sufficient to cause warming of the upper layers of the oceans and a substantial rise in sea level, a pronounced shift of the agricultural zones, and major but unknown changes in the polar ice caps.

There has by no means been universal acceptance among the experts of many of the predictions that have been made, and there are many who have opposing views of the causes of some of the phenomena that have been observed and experimentally measured. However, several detailed reports were issued in the 1990s in which the consensus of large groups of experts is that human activities, largely the burning of fossil fuels, are affecting global climate. At any one location, annual variations can be large, but analyses of meteorologi­cal and other data over decades for large areas provide evidence of important systemic changes.

One of the first comprehensive estimates of global mean, near-surface tem­perature over the earth’s lands and oceans was reported in 1986 (Jones et ah, 1986). The data showed a long-timescale warming trend. The three warmest years were 1980, 1981, and 1983, and five of the nine warmest years in the entire 124-year record up to 1984 were found to have occurred after 1978. It was apparent from this study that over this period, annual mean temperature increased by about 0.6 to 0.7°C, and that about 40 to 50% of this increase occurred since about 1975. According to many analysts, the warmest year on record up to 1995 is 1995, and recent years have been the warmest since 1860 despite the cooling effect of the volcanic eruption of Mt. Pinatubo in 1991

(с/. Intergovernmental Panel on Climate Change, 1991 and 1995). Nighttime temperatures over land have generally increased more than daytime tempera­tures, and regional changes are also evident. Warming has been the greatest over the mid-latitude continents in winter and spring, with a few areas of cooling such as the North Atlantic Ocean. Precipitation has increased over land in the high latitudes of the Northern Hemisphere, especially during the cold season. Global mean surface temperature has increased by between 0.3 and 0.6°C since the late nineteenth century and average global surface temperature increases of 1 to 3.5°C, somewhat lower than originally predicted, are expected to occur by the middle of the twenty-first century. Global sea level has risen by between 10 and 25 cm over the past 100 years, and much of the rise may be related to the increase in global mean temperature.

Since preindustrial times, ambient concentrations of the greenhouse gases have exhibited substantial increases, inter alia C02 by 30% to about 360 parts per million (ppm), CH4 by 145% to more than 1700 parts per billion (ppb), and N20 by 15% to more than 300 ppb. The growth rates in the concentrations of these gases in the early 1990s were lower than predicted, while subsequent data indicate that the growth rates are comparable to those averaged over the 1980s. If C02 emissions were maintained near mid-1990 levels, analysts have predicted that this would lead to a nearly constant increase in atmospheric concentrations for at least two centuries, reaching about 500 ppm by the end of the twenty-first century, and that stabilization of atmospheric C02 concentrations at 450 ppm could only be achieved if global anthropogenic emissions drop to 1990 levels by about 2035, and subsequently drop substan­tially below 1990 levels (Intergovernmental Panel on Climate Change, 1995). It is estimated that the corresponding atmospheric lifetimes of C02, CH4, and N20 are about 50 to 200, 12, and 120 years, respectively, and that together with increasing emissions to the atmosphere, they account for the steadily rising ambient concentrations of the greenhouse gases.

These gases are called greenhouse gases because they selectively allow more of the shorter wavelengths of solar radiation to reach the earth’s surface, but absorb more of the reflected longer wavelength infrared radiation than that allowed to leave the atmosphere. The result is the greenhouse effect on reradia­tion of the absorbed energy. An example of the change in atmospheric concen­tration of C02 at one measuring site is shown in Fig. 1.11 (Whorf, 1996). These data were accumulated from 1958 to 1995 by experimental measurement at Mauna Loa, Hawaii and show how the concentration increased from about 315 to 360 ppm over the measurement period and how it varies during the biomass growing season. The data show an approximate proportionality be­tween the rising atmospheric concentrations and industrial C02 emissions (Keeling et ah, 1995). The distribution and a few properties of selected atmo­spheric gases that have infrared absorption in the atmospheric window (7 to

13 fxm) are listed in Table 1.7. Carbon dioxide is by far the most abundant and is indicated in this table as the relative infrared standard. The gas-to — carbon dioxide infrared absorption ratios in the atmospheric window of CH4, N20, and the CFCs are much greater than 1.0. The effect of doubling the concentration of N20, CO, CH4, and C02 on the earth’s surface temperature is estimated to be 0.25, 0.6 to 0.9, 0.95, and 2 to 3°C, respectively.

Methane is present at much lower concentrations than C02, but is estimated to increase the surface temperature by almost 1°C on doubling of its concentra­tion. This is predicted to occur because the methane-to-carbon dioxide infrared ratio in the atmospheric infrared window is about 25, and hence CH4 is a much stronger absorber of infrared radiation than C02. Presuming the current rates of increase in ambient concentrations of the greenhouse gases continue, the doubling times can be estimated at which the surface temperature effects in Table 1.7 can be expected. For C02, various studies indicate that its concen­tration will double by the latter part of the twenty-first century. Although there is disagreement as to the exact time of doubling, there is virtually no

dispute among scientists that the concentrations of atmospheric C02 have increased about 30% since 1850.

A note of caution is necessary regarding the predictions that have been made regarding global temperature increases. The predictions made in the mid-1990s by the Intergovernmental Panel on Climate Change rely heavily on the use of computerized climate models. There is much uncertainty inherent in this technique because few models can reliably simulate even the present climate without “flux adjustments” (с/. Kerr, 1997). Consequently, there is considerable disagreement about the specific effects on global temperature of the greenhouse gases, and even clouds and pollutant hazes, and whether global warming can be correlated with human activities or is a natural phenomenon. Application of improved computer models that do not use flux adjustments indicates that global warming is occurring at the lower end of the many predictions that have been made.

A. Terrestrial Biomass

Much effort to evaluate terrestrial biomass for energy applications has been expended (for the United States, see Hohenstein and Wright, 1994; Ferrell, Wright, and Tuskan, 1995). In general, this work has been aimed at selecting high-yield biomass species, characterizing their physical and chemical proper­ties, defining their growth requirements, and rating their energy use potential. Several species have been proposed specifically for energy usage, whereas others have been recommended for multiple uses, one of which is as an energy resource. The latter case is exemplified by sugarcane; bagasse, the fibrous material remaining after sugar extraction, is used in several sugar factories as a boiler fuel. It is probable that most land-based biomass plantations operated for energy production or synfuel manufacture will also yield products for nonenergy markets. Large-scale biomass energy plantations that produce single energy products will probably be the exception rather than the rule. Land-based biomass for energy production can be divided into forest biomass, grasses, and cultivated plants.

Dry shredders are commercially used for reducing the size of biomass. The two most common types of machines are vertical and horizontal shaft hammermills. Metal hammers on rotating shafts or drums reduce particle size by impacting the feed material until the particles are small enough to drop through grate openings. Hammermills are commonly used in MSW-processing systems to reduce the size of the components before separation of RDF (i. e., refuse — derived fuel or the combustible fraction of MSW), and other materials. Ham­mermills are also used as agricultural choppers and tree chippers. Rotating cutters equipped with knife blades that reduce particle size by a cutting or shearing action are used for the same applications, although they usually have smaller capacities than hammermills.

Early biomass grinders were used to produce wood pulps from roundwood for the manufacture of paper (Riegel, 1933). Logs are positioned with the sides against a rotating grindstone so that damage to the fibers is minimized. Water is passed over the stone to wash the wood meal into a storage tank. In the original designs, the logs were held against the grindstone by springs. Later versions were hydraulic magazine grinders that automatically replaced the ground logs with fresh logs. In the 1950s, attrition mills were introduced to produce mechanical pulps from wood chips. Rotating, opposing discs, either one stationary or both moving in opposite directions, are used. The chips are fed to the mill near the central, rotating shaft and move outward to the periphery of the discs through a series of successively smaller channels that progressively reduce the feed to pulp-size particles.

Hydropulpers are wet shredders in which a high-speed cutting blade pulver­izes a water suspension of the feed over a perforated plate. The pulped material passes through the plate and the nonpulping materials are ejected. The action is similar to that of a kitchen waste disposal unit. Hydropulpers can also be used for the simultaneous size reduction and separation of the combustible fraction of MSW from the inorganic materials. But since the product is a water slurry of small particles, which can have the consistency of a heavy cream, the hydropulper is quite suitable for the preparation of RDF for microbial conversion after passage through a liquid cyclone to remove gritty, mostly inert material. Fiber recovery operations where long fibers are removed for resale can be performed before microbial processing. Experimental studies have shown that hydropulpers can also supply good feedstocks for microbial processing from other biomass. Maintenance costs for wet shredders are lower than those for dry shredders.

Agricultural choppers that are operated as stationary cutters and as moving choppers in the field separately from the harvesters or that are part of forage

harvesters that chop the crop during the harvesting process are commonly employed for preparing hay and other forage crops for ensiling. When a chopper is used in the field, more stems pass through the chopper lengthwise and a shorter average cut is obtained than when the same setting is used on a stationary chopper. Forage chopping in the field usually supplies material 25 mm or less in length. Silage systems offer several advantages for the size reduction of herbaceous biomass energy crops. The major disadvantage for high-yielding biomass species is the inability of the forage chopper to effectively harvest severely lodged (fallen) crops or plants (Coble and Egg, 1989). Sugar­cane harvesters, which are designed for harvesting high-yield, thick-stemmed, lodged biomass, have been used to harvest such crops. A separate size-reduction step is needed prior to storage or conversion.

In contrast to mechanical pulps, chemical wood pulps are often made from bark-free wood chips. One form of chipper has four knives fastened at 90° from each other on a rotating disc in such a way that only the edge of the blade projects beyond the disc. Each knife cuts a thin strip about 10 mm thick from the logs, which are fed along their long axis to the chipper at about a 38° angle to the disc. The mechanical action is similar to that of a sausage slicer. The denser hardwoods produce thinner chips than softwoods, but both types of chips are suitable feedstocks for most fixed — and fluid-bed gasifiers and most other thermal conversion processes.

Chipping has been the traditional mechanical method of size reduction to prepare wood fuels for direct combustion. It is an energy-intensive operation, but it does improve bulk density, handling, and transportation costs. Disc chipping and hogging are two preferred means of preparing wood fuels (Suadi — cani and Heding, 1992). Hammer hogs with free-swing hammers break the feed into small pieces, whereas knife hogs cut the feed with blades. The least desirable option seems to be chipping in the field at the time of harvest, which requires that a power chipper accompany the harvester through the field. Whole-tree chips are also reported to lose approximately 10% of their oven dry weight after storage for 6 months (с/. Curtin and Barnett, 1986). A variety of machines are available for producing wood chips in the field. One of the notable developments in North America is a swath harvester called “Jaws” that produces fuel chips while clearing 2.4-m wide paths through young, crowded stands of pine (Ranney et al., 1985). Another is the large mobile wood chipper, “Chiparvestor,” that can handle trees up to 0.76 m in diameter, and the medium — size unit that chips trees 0.56 to 0.69 m in diameter (Biomass Energy Research Association, 1990). The latter unit can produce 544 t of chips in 8 h, while another model for small-diameter trees can produce 91 t of chips in 8 h. Commercial wood chippers, both mobile and stationary, and chip harvesting methods are far advanced in Finland, where the usage of wood chip fuels has increased greatly (Seppanen, 1988).

Among the other options that can be considered for producing wood are chunking, billeting, and crushing. For smaller trees, chunking and billeting are similar processes that cut stems and limbs into 6- to 20-cm long pieces. Machines have been designed that chop small-diameter stems into smaller chunks or slightly longer billets. Chunkwood is approximately fist-sized. The production of chunkwood requires less energy than chip production, but it is not certain that the cost is competitive (Suadicani and Heding, 1992). Crushing is carried out by passing the stems between two or more metal rolls of varying size, rotational speed, and surface. Tests have shown that crushing rates of approximately 15 linear m/min can be achieved on stems up to 21 cm in diameter using only 11.2 kW (15 HP) of power (с/. Ranney et ai, 1987). The crushing and bunching of wood may offer significant advantages over chipping. This technique is flexible and is able to process large stems and stem lengths to yield bolts of crushed wood that exhibit relatively rapid drying. For reactor feeding purposes, however, further size reduction would be necessary. The feedstock characteristics required for the combustion or conversion process used determine which of these methods of size reduction may be applicable.

The dry distillation of hardwood was commercial technology and was quite common up until the early 1900s (Riegel, 1933). In the older industrial plants,

“Adapted from Grover (1989). Thermograms were obtained with powdered samples. The heating rate was 4°C/min in an atmosphere of nitrogen flowing at a rate of 0.3 L/min. The devolatilization range is the temperature at the beginning and end of devolatilization. The volatiles emitted at 320-500°C are the wt % of the original sample.

hardwood logs were placed on steel buggies and pushed into large, horizontal steel retorts. The doors were closed and external heating was supplied with gas or oil, often within a brick enclosure. Openings were provided for removal of volatiles. Heating was rapid for a few hours until the process became exothermic, and then the external heating was reduced and increased again when needed to complete the distillation. The distillation cycle was about 24 h including a 2-h cooling period. The buggies were then cooled for a few days in special airtight cooling chambers. The dry distillation of softwoods to obtain turpentine was carried out in smaller horizontal retorts in which wood chips were placed without buggies. The average product yields per cord of seasoned hardwood from typical commercial pyrolysis processes were about 1025 kg (950 L) of pyroligneous acid containing 7% acetic acid or equivalent, 4% crude methanol and acetone, 9% tar and oik, and 80% wa­ter; 454 kg of charcoal; and 212 m3 of fuel gas having a heating value of 9.3 to 11.2 MJ/m3 (Lowenheim and Moran, 1975). The pyroligneous acid is allowed to settle to remove insoluble tar and the clear decanted liquor is subjected to extraction or distillation or both to separate acetic acid from methanol and tar. For each tonne of acetic acid produced, about 625 to 800 L of crude methanol is recovered.

Until about the 1930s, when they were displaced by other processes that utilized fossil feedstocks, wood pyrolysis processes were used in industrialized countries for the manufacture of several chemicals and products. One example of this practice is the distillation plant operated by the Ford Motor Company using feedstock of hogged scrapwood from the automobile body plant (Riegel, 1933). A flow schematic for this plant is shown in Fig. 8.3. Vertical, cylindrical, steel retorts 3 m wide by 12 m high with an inside refractory wall 0.46 m thick were used. The hogged wood was dried to a moisture content of 0.5 wt % and consisted of 70% maple, 25% birch, and 5% ash, elm, and oak. Plant capacity was 363 t/day of scrap wood. The retorts were operated continuously for 2-week periods and the heat was supplied entirely by the exothermic pyrolysis reactions or the pyrolysis gas. At startup, the gas was employed to raise the temperature to 540°C, and then external heat was not needed. The average temperatures were 515°C in the center of the retort and 255°C near the bottom. The charcoal was discharged at the bottom of the retorts, cooled, screened, and briquetted. The pyroligneous acid was recovered from the overhead and the pyrolysis gas was used as boiler fuel except for the fuel used on startup of the retorts. The pyroligneous acid was distilled in batch units to remove dissolved tar, and the overhead was then fractionated in other distillation units. The product yields from this plant are shown in Table 8.8. The char, tar, and pitch yields are considerably higher than the yields of chemicals. It is somewhat surprising to note, since wood was a primary feed­stock for the manufacture of methanol (wood alcohol) before its displacement

Product Yield per tonne of dry wood

aAdapted from Riegei (1933). The average heating value of the dry gas was 11.39 MJ/m3 (n). The average composition of the gas in mol % from the retorts was H2> 2.2; CO, 23.4; C02, 37.9; CH4) 16.8; C„Hm, 1.2; 02, 2.4; N2) 16.0. The esters were produced from intermediate acid and product methanol or external ethanol.

by natural gas, that the methanol yield, including that used to esterify the acids formed in the process, is much less than the yields of the other pyrolysis products. In the Ford plant, the acetic acid was converted to esters, since they and not the free acid were needed in other automobile manufacturing operations. Flowever, other companies produced the free acid from the pyrolig­neous acid by direct solvent extraction. The Brewster process used isopropyl ether as the solvent and the Suida process used a high-boiling wood oil fraction from the pyrolysis plant as the solvent (Riegel, 1933). Eventually, these and other pyrolysis processes were phased out, as they were replaced by synthetic methods based on fossil feedstocks.

An updraft, wood-chip gasifier was built by Applied Engineering Company in 1980 in Georgia (Jackson, 1982). At that time, the unit was the largest of its kind. It was sized at 26.4 GJ/h, fed with 2.8 t/h of wood chips, and supplied a hospital with steam. A similar unit was built in late 1981 for the Florida Power Corporation. The unit fired one of six boilers in a 30-MW power system. The gasifier was cylindrical in shape, insulated with firebrick, and enclosed in a carbon steel shell. Air was injected at the bottom, and green tree chips having a heating value of about 9.3 to 10.5 MJ/kg and 40 to 50% moisture content were charged at the top. Ash was removed from the bottom. In the design used, oxidation of the wood char occurs at the top of the grate, which is located just above the ash hoppers, and produces temperatures of about 1370°C. Pyrolysis and cracking occur in the middle of the gasifier, and incom­ing wood is dried by the exiting hot gases. Typical dry gas analyses were 26 to 30 mol % carbon monoxide, 2 to 3 mol % hydrocarbons, 10 to 12 mol % hydrogen, and 58 to 59 mol % carbon dioxide and nitrogen. The heating value of the gas was 5.9 to 6.5 MJ/m3 (n). The gasifiers were operated quite success­fully for an extended period of time. It is noteworthy that the carbon monoxide concentration was so high. This may have been caused by the use of green wood with high moisture contents and operation at relatively high temperature in the gasification zone.

A. Fundamentals

The historical development of our understanding of the photosynthesis of biomass began in 1772 when the English scientist Joseph Priestley discovered that green plants expire a life-sustaining substance (oxygen) to the atmosphere, while a live mouse or a burning candle removes this same substance from the atmosphere. A variety of suggestions were offered by the scientific community during the ensuing 30 years to explain these observations until in 1804, the Swiss scientist Nicolas Theodore de Sausseure showed that the amount of C02 absorbed by green plants is the molecular equivalent of the oxygen expired. From that point on, the stoichiometry of the process was developed and major advancements were made to detail the chemistry of photosynthesis and how the assimilation of C02 takes place. Much of this work paralleled the develop­ment of research done to understand the biochemical pathways of the cellular metabolism of foodstuffs. Indeed, there is much overlap in the chemistry of both processes.

About 75% of the energy in solar radiation, after passage through the atmosphere where much of the shorter wavelength, high-energy radiation is filtered out, is contained in light of wavelengths between the visible and near­infrared portions of the electromagnetic spectrum, 400 to 1100 nm. The light­absorbing pigments effective in photosynthesis have absorption bands in this range. Chlorophyll a and chlorophyll b, which strongly absorb wavelengths in the red and blue regions of the spectrum, and accessory carotenoid and phycobilin pigments participate in the process. Numerous investigations have established many of the parameters in the complex photosynthetic reactions
occurring in biomass membrane systems which contain the necessary pigments and electron carriers. Tracer studies to establish the chemical structures of the intermediates in photosynthesis were initiated in the 1940s by the U. S. investigators Melvin Calvin, J. A. Bassham, and Andrew A. Benson. These studies showed that the oxygen evolved in photosynthesis comes exclusively from water and not C02. With green algae (Chlorella), reducing power is accumulated during illumination in the absence of C02 and can later be used for the reduction of C02 in the absence of light (Calvin and Benson, 1947). After a short 30- to 90-second exposure to light, the main portion of the newly reduced, labeled C02 was found to be distributed in a dozen or more organic compounds. By progressively shortening the light exposure to 2 seconds before killing the cells, almost all of the 14C in the labeled C02 was found to be incorporated in 3-phosphoglyceric acid, a compound that occurs in practically all plant and animal cells. Such experiments led to elaboration of the biochemi­cal pathways and the essential compounds required for photosynthesis. It was found that the pentose ribulose-1,5-diphosphate is a key intermediate in the process. It reacts with C02 to yield 3-phosphoglyceric acid. Equimolar amounts of ribulose-1,5-diphosphate and C02 react to form 2 mol of 3-phosphoglyceric acid, and in the process, inorganic carbon is transformed into organic carbon.

CH2OP03H2

C=0

1

H-C-OH + COo

I

H-C-OH

I

ch2opo3h2

Ribulose-1,5-diphosphate 3-Phosphoglyceric acid

These reactions of course occur in the presence of the proper enzyme catalysts and cofactors. Glucose is the primary photosynthetic product. As will be shown later in the discussion of the structures of the organic intermediates in the various pathways, the dark reactions take place in such a manner that ribulose- 1,5-diphosphate is regenerated.

Considerable research has been conducted to screen and select herbaceous plants as potential biomass candidates that are mainly unexplored in the Continental United States. Other research has concentrated on cash crops such as sugarcane and sweet sorghum, and still other research has emphasized tropical grasses. In the late 1970s, a comprehensive screening study of plants grown in the United States generated a list of 280 promising candidates from which up to 20 species were recommended for field experiments in each region of the country (Saterson and Luppold, 1979). The four highest-yielding species recommended for further tests in each region are listed in Table 4.16. Since many of the plants in the original list of 280 species had not been grown for commercial use, the production costs were estimated as shown in Table 4.17 for the various classes of herbaceous species. The results were used in conjunc­tion with yield and other data to develop the recommendations in Table 4.16.

A large number of research projects directed to small-scale field tests of potential herbaceous energy crops have since been carried out. The productivity ranges for some of the promising species for the U. S. Midwest and Southeast are shown in Table 4.11. The results of this research helped to establish a strategy that herbaceous biomass energy crops should be primarily grasses and legumes produced by use of management systems similar to those used for conventional forage crops. It was concluded that the ideal selection of herba­ceous energy crops for these areas would consist of at least one annual species, one warm-season perennial species, one cool-season perennial species, and one legume. Production rates, cost estimates, and environmental considerations indicate that perennial species are preferred to annual species on many sites, but annuals may be more important in crop rotations.

In greenhouse, small-plot, and field-scale research tests conducted to screen tropical grasses as energy crops, three categories emerged, based on the time required to maximize dry-matter yields: short-rotation species (2 to 3 months), intermediate-rotation species (4 to 6 months), and long-rotation species (12 to 18 months) (Alexander, 1991). A sorghum-sudan grass hybrid (Sordan 70A), the forage grass napier grass, and sugarcane were outstanding candidates in these categories. Minimum-tillage grasses that produced moderate yields with little attention were wild Saccharum clones and Johnson grass in a fourth category. The maximum yield observed was 61.6 dry t/ha-year for sugarcane propagated at narrow row centers over 12 months. The estimated maximum yield is of the order of 112 dry t/ha-year using new generations of sugarcane and the propagation of ratoon (regrowth) plants for several years after a given crop is planted.

Overall, the research that has been completed in the United States on the development of herbaceous biomass energy crops shows that a wide range of suitable species exist from which good candidates can be chosen for each area.

aSatterson and Luppold (1979).

bAs defined by U. S. Dept, of Agriculture (1972); excludes Alaska and Hawaii.

An extremely large amount of basic research has been carried out on the combustion of solid fuels, including fossil, biomass, and inorganic solid fuels, to ascertain the mechanisms and kinetics of the process. Each category of fuel combusts under different conditions which are determined by a variety of intensive chemical and physical properties of the solids and external ambient factors. An empirical view of biomass combustion involves the evaporation of the high-energy volatiles such as the terpenes, which burn in the gas phase with flaming combustion (cf. Shafizadeh and DeGroot, 1976). The lignocellulosics in the solid biomass, under the influence of high temperature or a sufficiently

FLAME TEMPERATURE (°С)

FIGURE 7.1 Theoretical flame temperature vs wood moisture content and excess air. Adapted from Tewksbury (1991).

strong energy source, decompose to form pyrolysis products, which also burn in the gas phase with flaming combustion. The residual char burns at a lower rate by surface oxidation or glowing combustion. The cellulosics are converted mainly to combustible and noncombustible volatiles, including water and C02, while the lignins contribute mainly to the char fraction.

The temperature within the flame is a function of reaction time, combustion intensity, flame velocity, and energy transferred to the surroundings. The flame temperatures measured on combustion of acetylene, gasoline, hydrogen, and natural gas in air under controlled conditions are 2319, 2310, 2045, and 1880°C, respectively (Reed, 1983). The combustion of biomass does not reach these temperature levels because of the lower energy density of biomass and the mechanism of biomass combustion. Simplistically, the mechanism involved in the combustion of solid biomass can be viewed as a stepwise process, although all steps occur simultaneously in the combustion chamber. First, the increasing temperature dries the incoming, fresh biomass. The physically contained moisture in the biomass is vaporized. At about 150 to 200°C, thermal decomposition and devolatilization of the solid biomass then begins on the biomass surface and volatile organic compounds are evolved as a gas, which burns in the combustion chamber. (Note that the discussion of biomass pyroly­sis in Chapter 8 addresses some of the important research on the mechanisms and kinetics of biomass devolatilization which also apply to biomass combus­tion.) The remaining fuel components in the carbonaceous residue are com­busted by diffusion of oxygen to the solid surface at temperatures of about 400 to 800°C and greater. This temperature range is attained by absorption of radiant energy from the hot combustion products and the combustion chamber surfaces. Temperatures as high as 1500°C have been recorded when the incoming fresh fuel is dry and the combustion process is carefully con­trolled. The use of preheated air permits similar temperatures in some systems even with biomass that contains some moisture.